From 1c014420583564ac09e3b67006f2e7050861e66b Mon Sep 17 00:00:00 2001
From: Ivo van Doorn <ivdoorn@gmail.com>
Date: Thu, 17 Apr 2008 19:41:02 +0200
Subject: mac80211: Replace ieee80211_tx_control->key_idx with
 ieee80211_key_conf

The hw_key_idx inside the ieee80211_key_conf structure does
not provide all the information drivers might need to perform
hardware encryption.

This is in particular true for rt2x00 who needs to know the
key algorithm and whether it is a shared or pairwise key.

By passing the ieee80211_key_conf pointer it assures us that
drivers can make full use of all information that it should know
about a particular key.

Additionally this patch updates all drivers to grab the hw_key_idx from
the ieee80211_key_conf structure.

v2: Removed bogus u16 cast
v3: Add warning about ieee80211_tx_control pointers
v4: Update warning about ieee80211_tx_control pointers

Signed-off-by: Ivo van Doorn <IvDoorn@gmail.com>
Acked-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/net/mac80211.h | 19 +++++++++++++++----
 1 file changed, 15 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index 4a80d74975e8..27ef9f761ac5 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -286,8 +286,17 @@ enum mac80211_tx_control_flags {
 
 /* Transmit control fields. This data structure is passed to low-level driver
  * with each TX frame. The low-level driver is responsible for configuring
- * the hardware to use given values (depending on what is supported). */
-
+ * the hardware to use given values (depending on what is supported).
+ *
+ * NOTE: Be careful with using the pointers outside of the ieee80211_ops->tx()
+ * context (i.e. when defering the work to a workqueue).
+ * The vif pointer is valid until the it has been removed with the
+ * ieee80211_ops->remove_interface() callback funtion.
+ * The hw_key pointer is valid until it has been removed with the
+ * ieee80211_ops->set_key() callback function.
+ * The tx_rate and alt_retry_rate pointers are valid until the phy is
+ * deregistered.
+ */
 struct ieee80211_tx_control {
 	struct ieee80211_vif *vif;
 	struct ieee80211_rate *tx_rate;
@@ -298,9 +307,11 @@ struct ieee80211_tx_control {
 	/* retry rate for the last retries */
 	struct ieee80211_rate *alt_retry_rate;
 
+	/* Key used for hardware encryption
+	 * NULL if IEEE80211_TXCTL_DO_NOT_ENCRYPT is set */
+	struct ieee80211_key_conf *hw_key;
+
 	u32 flags;		/* tx control flags defined above */
-	u8 key_idx;		/* keyidx from hw->set_key(), undefined if
-				 * IEEE80211_TXCTL_DO_NOT_ENCRYPT is set */
 	u8 retry_limit;		/* 1 = only first attempt, 2 = one retry, ..
 				 * This could be used when set_retry_limit
 				 * is not implemented by the driver */
-- 
cgit v1.2.3


From c6adbd2158fee972adcc6232de5e2ef375f1f782 Mon Sep 17 00:00:00 2001
From: Ivo van Doorn <ivdoorn@gmail.com>
Date: Thu, 17 Apr 2008 21:11:18 +0200
Subject: mac80211: Add IEEE80211_KEY_FLAG_PAIRWISE

This adds a new flag to the ieee80211_key_conf structure.
This flag will inform the driver the key is pairwise rather then
a shared key.

This is important for drivers who support both types of keys,
and need to be informed which type of key this is. Alternative
would be drivers checking the address argument of set_key(),
but it will be safer when mac80211 is more explicit.

Signed-off-by: Ivo van Doorn <IvDoorn@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/net/mac80211.h | 3 +++
 net/mac80211/key.c     | 7 +++++++
 2 files changed, 10 insertions(+)

(limited to 'include')

diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index 27ef9f761ac5..740c11ca066c 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -621,11 +621,14 @@ enum ieee80211_key_alg {
  * @IEEE80211_KEY_FLAG_GENERATE_MMIC: This flag should be set by
  *	the driver for a TKIP key if it requires Michael MIC
  *	generation in software.
+ * @IEEE80211_KEY_FLAG_PAIRWISE: Set by mac80211, this flag indicates
+ *	that the key is pairwise rather then a shared key.
  */
 enum ieee80211_key_flags {
 	IEEE80211_KEY_FLAG_WMM_STA	= 1<<0,
 	IEEE80211_KEY_FLAG_GENERATE_IV	= 1<<1,
 	IEEE80211_KEY_FLAG_GENERATE_MMIC= 1<<2,
+	IEEE80211_KEY_FLAG_PAIRWISE	= 1<<3,
 };
 
 /**
diff --git a/net/mac80211/key.c b/net/mac80211/key.c
index 150d66dbda9d..88b211af7c1f 100644
--- a/net/mac80211/key.c
+++ b/net/mac80211/key.c
@@ -323,6 +323,13 @@ void ieee80211_key_link(struct ieee80211_key *key,
 		 */
 		if (sta->flags & WLAN_STA_WME)
 			key->conf.flags |= IEEE80211_KEY_FLAG_WMM_STA;
+
+		/*
+		 * This key is for a specific sta interface,
+		 * inform the driver that it should try to store
+		 * this key as pairwise key.
+		 */
+		key->conf.flags |= IEEE80211_KEY_FLAG_PAIRWISE;
 	} else {
 		if (sdata->vif.type == IEEE80211_IF_TYPE_STA) {
 			struct sta_info *ap;
-- 
cgit v1.2.3


From c12cf2109702c052688391f4171f239effb241bf Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Mon, 21 Apr 2008 11:48:28 +0300
Subject: remove ieee80211_tx_frame()

After the softmac removal ieee80211_tx_frame() was no longer used.

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/net/ieee80211.h      |  3 --
 net/ieee80211/ieee80211_tx.c | 86 --------------------------------------------
 2 files changed, 89 deletions(-)

(limited to 'include')

diff --git a/include/net/ieee80211.h b/include/net/ieee80211.h
index 529816bfbc52..facd0ffedc6e 100644
--- a/include/net/ieee80211.h
+++ b/include/net/ieee80211.h
@@ -1262,9 +1262,6 @@ extern int ieee80211_set_encryption(struct ieee80211_device *ieee);
 /* ieee80211_tx.c */
 extern int ieee80211_xmit(struct sk_buff *skb, struct net_device *dev);
 extern void ieee80211_txb_free(struct ieee80211_txb *);
-extern int ieee80211_tx_frame(struct ieee80211_device *ieee,
-			      struct ieee80211_hdr *frame, int hdr_len,
-			      int total_len, int encrypt_mpdu);
 
 /* ieee80211_rx.c */
 extern void ieee80211_rx_any(struct ieee80211_device *ieee,
diff --git a/net/ieee80211/ieee80211_tx.c b/net/ieee80211/ieee80211_tx.c
index d8b02603cbe5..d996547f7a62 100644
--- a/net/ieee80211/ieee80211_tx.c
+++ b/net/ieee80211/ieee80211_tx.c
@@ -542,90 +542,4 @@ int ieee80211_xmit(struct sk_buff *skb, struct net_device *dev)
 	return 1;
 }
 
-/* Incoming 802.11 strucure is converted to a TXB
- * a block of 802.11 fragment packets (stored as skbs) */
-int ieee80211_tx_frame(struct ieee80211_device *ieee,
-		       struct ieee80211_hdr *frame, int hdr_len, int total_len,
-		       int encrypt_mpdu)
-{
-	struct ieee80211_txb *txb = NULL;
-	unsigned long flags;
-	struct net_device_stats *stats = &ieee->stats;
-	struct sk_buff *skb_frag;
-	int priority = -1;
-	int fraglen = total_len;
-	int headroom = ieee->tx_headroom;
-	struct ieee80211_crypt_data *crypt = ieee->crypt[ieee->tx_keyidx];
-
-	spin_lock_irqsave(&ieee->lock, flags);
-
-	if (encrypt_mpdu && (!ieee->sec.encrypt || !crypt))
-		encrypt_mpdu = 0;
-
-	/* If there is no driver handler to take the TXB, dont' bother
-	 * creating it... */
-	if (!ieee->hard_start_xmit) {
-		printk(KERN_WARNING "%s: No xmit handler.\n", ieee->dev->name);
-		goto success;
-	}
-
-	if (unlikely(total_len < 24)) {
-		printk(KERN_WARNING "%s: skb too small (%d).\n",
-		       ieee->dev->name, total_len);
-		goto success;
-	}
-
-	if (encrypt_mpdu) {
-		frame->frame_ctl |= cpu_to_le16(IEEE80211_FCTL_PROTECTED);
-		fraglen += crypt->ops->extra_mpdu_prefix_len +
-			   crypt->ops->extra_mpdu_postfix_len;
-		headroom += crypt->ops->extra_mpdu_prefix_len;
-	}
-
-	/* When we allocate the TXB we allocate enough space for the reserve
-	 * and full fragment bytes (bytes_per_frag doesn't include prefix,
-	 * postfix, header, FCS, etc.) */
-	txb = ieee80211_alloc_txb(1, fraglen, headroom, GFP_ATOMIC);
-	if (unlikely(!txb)) {
-		printk(KERN_WARNING "%s: Could not allocate TXB\n",
-		       ieee->dev->name);
-		goto failed;
-	}
-	txb->encrypted = 0;
-	txb->payload_size = fraglen;
-
-	skb_frag = txb->fragments[0];
-
-	memcpy(skb_put(skb_frag, total_len), frame, total_len);
-
-	if (ieee->config &
-	    (CFG_IEEE80211_COMPUTE_FCS | CFG_IEEE80211_RESERVE_FCS))
-		skb_put(skb_frag, 4);
-
-	/* To avoid overcomplicating things, we do the corner-case frame
-	 * encryption in software. The only real situation where encryption is
-	 * needed here is during software-based shared key authentication. */
-	if (encrypt_mpdu)
-		ieee80211_encrypt_fragment(ieee, skb_frag, hdr_len);
-
-      success:
-	spin_unlock_irqrestore(&ieee->lock, flags);
-
-	if (txb) {
-		if ((*ieee->hard_start_xmit) (txb, ieee->dev, priority) == 0) {
-			stats->tx_packets++;
-			stats->tx_bytes += txb->payload_size;
-			return 0;
-		}
-		ieee80211_txb_free(txb);
-	}
-	return 0;
-
-      failed:
-	spin_unlock_irqrestore(&ieee->lock, flags);
-	stats->tx_errors++;
-	return 1;
-}
-
-EXPORT_SYMBOL(ieee80211_tx_frame);
 EXPORT_SYMBOL(ieee80211_txb_free);
-- 
cgit v1.2.3


From 7eafd25d9559bd0f652449c222d38d63412e3d4a Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Mon, 21 Apr 2008 11:48:34 +0300
Subject: remove ieee80211_wx_{get,set}_auth()

After the bcm43xx removal ieee80211_wx_{get,set}_auth() were no longer
used.

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/net/ieee80211.h      |  8 ----
 net/ieee80211/ieee80211_wx.c | 89 --------------------------------------------
 2 files changed, 97 deletions(-)

(limited to 'include')

diff --git a/include/net/ieee80211.h b/include/net/ieee80211.h
index facd0ffedc6e..b31399e1fd83 100644
--- a/include/net/ieee80211.h
+++ b/include/net/ieee80211.h
@@ -1309,14 +1309,6 @@ extern int ieee80211_wx_set_encodeext(struct ieee80211_device *ieee,
 extern int ieee80211_wx_get_encodeext(struct ieee80211_device *ieee,
 				      struct iw_request_info *info,
 				      union iwreq_data *wrqu, char *extra);
-extern int ieee80211_wx_set_auth(struct net_device *dev,
-				 struct iw_request_info *info,
-				 union iwreq_data *wrqu,
-				 char *extra);
-extern int ieee80211_wx_get_auth(struct net_device *dev,
-				 struct iw_request_info *info,
-				 union iwreq_data *wrqu,
-				 char *extra);
 
 static inline void ieee80211_increment_scans(struct ieee80211_device *ieee)
 {
diff --git a/net/ieee80211/ieee80211_wx.c b/net/ieee80211/ieee80211_wx.c
index 623489afa62c..822606b615ca 100644
--- a/net/ieee80211/ieee80211_wx.c
+++ b/net/ieee80211/ieee80211_wx.c
@@ -744,98 +744,9 @@ int ieee80211_wx_get_encodeext(struct ieee80211_device *ieee,
 	return 0;
 }
 
-int ieee80211_wx_set_auth(struct net_device *dev,
-			  struct iw_request_info *info,
-			  union iwreq_data *wrqu,
-			  char *extra)
-{
-	struct ieee80211_device *ieee = netdev_priv(dev);
-	unsigned long flags;
-	int err = 0;
-
-	spin_lock_irqsave(&ieee->lock, flags);
-
-	switch (wrqu->param.flags & IW_AUTH_INDEX) {
-	case IW_AUTH_WPA_VERSION:
-	case IW_AUTH_CIPHER_PAIRWISE:
-	case IW_AUTH_CIPHER_GROUP:
-	case IW_AUTH_KEY_MGMT:
-		/*
-		 * Host AP driver does not use these parameters and allows
-		 * wpa_supplicant to control them internally.
-		 */
-		break;
-	case IW_AUTH_TKIP_COUNTERMEASURES:
-		break;		/* FIXME */
-	case IW_AUTH_DROP_UNENCRYPTED:
-		ieee->drop_unencrypted = !!wrqu->param.value;
-		break;
-	case IW_AUTH_80211_AUTH_ALG:
-		break;		/* FIXME */
-	case IW_AUTH_WPA_ENABLED:
-		ieee->privacy_invoked = ieee->wpa_enabled = !!wrqu->param.value;
-		break;
-	case IW_AUTH_RX_UNENCRYPTED_EAPOL:
-		ieee->ieee802_1x = !!wrqu->param.value;
-		break;
-	case IW_AUTH_PRIVACY_INVOKED:
-		ieee->privacy_invoked = !!wrqu->param.value;
-		break;
-	default:
-		err = -EOPNOTSUPP;
-		break;
-	}
-	spin_unlock_irqrestore(&ieee->lock, flags);
-	return err;
-}
-
-int ieee80211_wx_get_auth(struct net_device *dev,
-			  struct iw_request_info *info,
-			  union iwreq_data *wrqu,
-			  char *extra)
-{
-	struct ieee80211_device *ieee = netdev_priv(dev);
-	unsigned long flags;
-	int err = 0;
-
-	spin_lock_irqsave(&ieee->lock, flags);
-
-	switch (wrqu->param.flags & IW_AUTH_INDEX) {
-	case IW_AUTH_WPA_VERSION:
-	case IW_AUTH_CIPHER_PAIRWISE:
-	case IW_AUTH_CIPHER_GROUP:
-	case IW_AUTH_KEY_MGMT:
-	case IW_AUTH_TKIP_COUNTERMEASURES:		/* FIXME */
-	case IW_AUTH_80211_AUTH_ALG:			/* FIXME */
-		/*
-		 * Host AP driver does not use these parameters and allows
-		 * wpa_supplicant to control them internally.
-		 */
-		err = -EOPNOTSUPP;
-		break;
-	case IW_AUTH_DROP_UNENCRYPTED:
-		wrqu->param.value = ieee->drop_unencrypted;
-		break;
-	case IW_AUTH_WPA_ENABLED:
-		wrqu->param.value = ieee->wpa_enabled;
-		break;
-	case IW_AUTH_RX_UNENCRYPTED_EAPOL:
-		wrqu->param.value = ieee->ieee802_1x;
-		break;
-	default:
-		err = -EOPNOTSUPP;
-		break;
-	}
-	spin_unlock_irqrestore(&ieee->lock, flags);
-	return err;
-}
-
 EXPORT_SYMBOL(ieee80211_wx_set_encodeext);
 EXPORT_SYMBOL(ieee80211_wx_get_encodeext);
 
 EXPORT_SYMBOL(ieee80211_wx_get_scan);
 EXPORT_SYMBOL(ieee80211_wx_set_encode);
 EXPORT_SYMBOL(ieee80211_wx_get_encode);
-
-EXPORT_SYMBOL_GPL(ieee80211_wx_set_auth);
-EXPORT_SYMBOL_GPL(ieee80211_wx_get_auth);
-- 
cgit v1.2.3


From 57ffc589a92424f9def74fe0d49b2f7763ff07fd Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes@sipsolutions.net>
Date: Tue, 29 Apr 2008 17:18:59 +0200
Subject: mac80211: clean up get_tx_stats callback

The callback takes a ieee80211_tx_queue_stats with a contained
array of ieee80211_tx_queue_stats_data, remove the former, rename
the latter to ieee80211_tx_queue_stats and make tx_stats() take
the array directly.

Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/net/wireless/adm8211.c              |  7 +++----
 drivers/net/wireless/ath5k/base.c           |  8 ++++----
 drivers/net/wireless/ath5k/base.h           |  3 ++-
 drivers/net/wireless/b43/dma.c              |  8 +++-----
 drivers/net/wireless/b43/pio.c              |  8 +++-----
 drivers/net/wireless/b43legacy/dma.c        |  8 +++-----
 drivers/net/wireless/b43legacy/pio.c        |  8 +++-----
 drivers/net/wireless/iwlwifi/iwl3945-base.c |  6 +++---
 drivers/net/wireless/iwlwifi/iwl4965-base.c |  6 +++---
 drivers/net/wireless/p54/p54.h              |  2 +-
 drivers/net/wireless/p54/p54common.c        | 24 ++++++++++--------------
 drivers/net/wireless/rt2x00/rt2x00mac.c     |  6 +++---
 include/net/mac80211.h                      | 13 ++++---------
 13 files changed, 45 insertions(+), 62 deletions(-)

(limited to 'include')

diff --git a/drivers/net/wireless/adm8211.c b/drivers/net/wireless/adm8211.c
index 5c0d2b082750..d93a1de77eb0 100644
--- a/drivers/net/wireless/adm8211.c
+++ b/drivers/net/wireless/adm8211.c
@@ -306,11 +306,10 @@ static int adm8211_get_tx_stats(struct ieee80211_hw *dev,
 				struct ieee80211_tx_queue_stats *stats)
 {
 	struct adm8211_priv *priv = dev->priv;
-	struct ieee80211_tx_queue_stats_data *data = &stats->data[0];
 
-	data->len = priv->cur_tx - priv->dirty_tx;
-	data->limit = priv->tx_ring_size - 2;
-	data->count = priv->dirty_tx;
+	stats[0].len = priv->cur_tx - priv->dirty_tx;
+	stats[0].limit = priv->tx_ring_size - 2;
+	stats[0].count = priv->dirty_tx;
 
 	return 0;
 }
diff --git a/drivers/net/wireless/ath5k/base.c b/drivers/net/wireless/ath5k/base.c
index 3854619f3514..3201c1604340 100644
--- a/drivers/net/wireless/ath5k/base.c
+++ b/drivers/net/wireless/ath5k/base.c
@@ -1335,7 +1335,7 @@ ath5k_txbuf_setup(struct ath5k_softc *sc, struct ath5k_buf *bf,
 
 	spin_lock_bh(&txq->lock);
 	list_add_tail(&bf->list, &txq->q);
-	sc->tx_stats.data[txq->qnum].len++;
+	sc->tx_stats[txq->qnum].len++;
 	if (txq->link == NULL) /* is this first packet? */
 		ath5k_hw_put_tx_buf(ah, txq->qnum, bf->daddr);
 	else /* no, so only link it */
@@ -1566,7 +1566,7 @@ ath5k_txq_drainq(struct ath5k_softc *sc, struct ath5k_txq *txq)
 		ath5k_txbuf_free(sc, bf);
 
 		spin_lock_bh(&sc->txbuflock);
-		sc->tx_stats.data[txq->qnum].len--;
+		sc->tx_stats[txq->qnum].len--;
 		list_move_tail(&bf->list, &sc->txbuf);
 		sc->txbuf_len++;
 		spin_unlock_bh(&sc->txbuflock);
@@ -1979,10 +1979,10 @@ ath5k_tx_processq(struct ath5k_softc *sc, struct ath5k_txq *txq)
 		}
 
 		ieee80211_tx_status(sc->hw, skb, &txs);
-		sc->tx_stats.data[txq->qnum].count++;
+		sc->tx_stats[txq->qnum].count++;
 
 		spin_lock(&sc->txbuflock);
-		sc->tx_stats.data[txq->qnum].len--;
+		sc->tx_stats[txq->qnum].len--;
 		list_move_tail(&bf->list, &sc->txbuf);
 		sc->txbuf_len++;
 		spin_unlock(&sc->txbuflock);
diff --git a/drivers/net/wireless/ath5k/base.h b/drivers/net/wireless/ath5k/base.h
index 3a9755893018..ecb17495488c 100644
--- a/drivers/net/wireless/ath5k/base.h
+++ b/drivers/net/wireless/ath5k/base.h
@@ -92,7 +92,8 @@ struct ath5k_softc {
 	struct pci_dev		*pdev;		/* for dma mapping */
 	void __iomem		*iobase;	/* address of the device */
 	struct mutex		lock;		/* dev-level lock */
-	struct ieee80211_tx_queue_stats tx_stats;
+	/* FIXME: how many does it really need? */
+	struct ieee80211_tx_queue_stats tx_stats[16];
 	struct ieee80211_low_level_stats ll_stats;
 	struct ieee80211_hw	*hw;		/* IEEE 802.11 common */
 	struct ieee80211_supported_band sbands[IEEE80211_NUM_BANDS];
diff --git a/drivers/net/wireless/b43/dma.c b/drivers/net/wireless/b43/dma.c
index 6dcbb3c87e72..f50e2014ffbe 100644
--- a/drivers/net/wireless/b43/dma.c
+++ b/drivers/net/wireless/b43/dma.c
@@ -1427,18 +1427,16 @@ void b43_dma_get_tx_stats(struct b43_wldev *dev,
 {
 	const int nr_queues = dev->wl->hw->queues;
 	struct b43_dmaring *ring;
-	struct ieee80211_tx_queue_stats_data *data;
 	unsigned long flags;
 	int i;
 
 	for (i = 0; i < nr_queues; i++) {
-		data = &(stats->data[i]);
 		ring = select_ring_by_priority(dev, i);
 
 		spin_lock_irqsave(&ring->lock, flags);
-		data->len = ring->used_slots / SLOTS_PER_PACKET;
-		data->limit = ring->nr_slots / SLOTS_PER_PACKET;
-		data->count = ring->nr_tx_packets;
+		stats[i].len = ring->used_slots / SLOTS_PER_PACKET;
+		stats[i].limit = ring->nr_slots / SLOTS_PER_PACKET;
+		stats[i].count = ring->nr_tx_packets;
 		spin_unlock_irqrestore(&ring->lock, flags);
 	}
 }
diff --git a/drivers/net/wireless/b43/pio.c b/drivers/net/wireless/b43/pio.c
index fcacafb04346..08c8a087f30e 100644
--- a/drivers/net/wireless/b43/pio.c
+++ b/drivers/net/wireless/b43/pio.c
@@ -611,18 +611,16 @@ void b43_pio_get_tx_stats(struct b43_wldev *dev,
 {
 	const int nr_queues = dev->wl->hw->queues;
 	struct b43_pio_txqueue *q;
-	struct ieee80211_tx_queue_stats_data *data;
 	unsigned long flags;
 	int i;
 
 	for (i = 0; i < nr_queues; i++) {
-		data = &(stats->data[i]);
 		q = select_queue_by_priority(dev, i);
 
 		spin_lock_irqsave(&q->lock, flags);
-		data->len = B43_PIO_MAX_NR_TXPACKETS - q->free_packet_slots;
-		data->limit = B43_PIO_MAX_NR_TXPACKETS;
-		data->count = q->nr_tx_packets;
+		stats[i].len = B43_PIO_MAX_NR_TXPACKETS - q->free_packet_slots;
+		stats[i].limit = B43_PIO_MAX_NR_TXPACKETS;
+		stats[i].count = q->nr_tx_packets;
 		spin_unlock_irqrestore(&q->lock, flags);
 	}
 }
diff --git a/drivers/net/wireless/b43legacy/dma.c b/drivers/net/wireless/b43legacy/dma.c
index c990f87b107a..d6686f713b6d 100644
--- a/drivers/net/wireless/b43legacy/dma.c
+++ b/drivers/net/wireless/b43legacy/dma.c
@@ -1455,18 +1455,16 @@ void b43legacy_dma_get_tx_stats(struct b43legacy_wldev *dev,
 {
 	const int nr_queues = dev->wl->hw->queues;
 	struct b43legacy_dmaring *ring;
-	struct ieee80211_tx_queue_stats_data *data;
 	unsigned long flags;
 	int i;
 
 	for (i = 0; i < nr_queues; i++) {
-		data = &(stats->data[i]);
 		ring = priority_to_txring(dev, i);
 
 		spin_lock_irqsave(&ring->lock, flags);
-		data->len = ring->used_slots / SLOTS_PER_PACKET;
-		data->limit = ring->nr_slots / SLOTS_PER_PACKET;
-		data->count = ring->nr_tx_packets;
+		stats[i].len = ring->used_slots / SLOTS_PER_PACKET;
+		stats[i].limit = ring->nr_slots / SLOTS_PER_PACKET;
+		stats[i].count = ring->nr_tx_packets;
 		spin_unlock_irqrestore(&ring->lock, flags);
 	}
 }
diff --git a/drivers/net/wireless/b43legacy/pio.c b/drivers/net/wireless/b43legacy/pio.c
index bcdd54eb2edb..8d3d27d3cd67 100644
--- a/drivers/net/wireless/b43legacy/pio.c
+++ b/drivers/net/wireless/b43legacy/pio.c
@@ -525,13 +525,11 @@ void b43legacy_pio_get_tx_stats(struct b43legacy_wldev *dev,
 {
 	struct b43legacy_pio *pio = &dev->pio;
 	struct b43legacy_pioqueue *queue;
-	struct ieee80211_tx_queue_stats_data *data;
 
 	queue = pio->queue1;
-	data = &(stats->data[0]);
-	data->len = B43legacy_PIO_MAXTXPACKETS - queue->nr_txfree;
-	data->limit = B43legacy_PIO_MAXTXPACKETS;
-	data->count = queue->nr_tx_packets;
+	stats[0].len = B43legacy_PIO_MAXTXPACKETS - queue->nr_txfree;
+	stats[0].limit = B43legacy_PIO_MAXTXPACKETS;
+	stats[0].count = queue->nr_tx_packets;
 }
 
 static void pio_rx_error(struct b43legacy_pioqueue *queue,
diff --git a/drivers/net/wireless/iwlwifi/iwl3945-base.c b/drivers/net/wireless/iwlwifi/iwl3945-base.c
index 67fd267c99ca..bad367cfbee3 100644
--- a/drivers/net/wireless/iwlwifi/iwl3945-base.c
+++ b/drivers/net/wireless/iwlwifi/iwl3945-base.c
@@ -7217,9 +7217,9 @@ static int iwl3945_mac_get_tx_stats(struct ieee80211_hw *hw,
 		q = &txq->q;
 		avail = iwl3945_queue_space(q);
 
-		stats->data[i].len = q->n_window - avail;
-		stats->data[i].limit = q->n_window - q->high_mark;
-		stats->data[i].count = q->n_window;
+		stats[i].len = q->n_window - avail;
+		stats[i].limit = q->n_window - q->high_mark;
+		stats[i].count = q->n_window;
 
 	}
 	spin_unlock_irqrestore(&priv->lock, flags);
diff --git a/drivers/net/wireless/iwlwifi/iwl4965-base.c b/drivers/net/wireless/iwlwifi/iwl4965-base.c
index f887a4e54ba2..64cb905cdaed 100644
--- a/drivers/net/wireless/iwlwifi/iwl4965-base.c
+++ b/drivers/net/wireless/iwlwifi/iwl4965-base.c
@@ -6420,9 +6420,9 @@ static int iwl4965_mac_get_tx_stats(struct ieee80211_hw *hw,
 		q = &txq->q;
 		avail = iwl4965_queue_space(q);
 
-		stats->data[i].len = q->n_window - avail;
-		stats->data[i].limit = q->n_window - q->high_mark;
-		stats->data[i].count = q->n_window;
+		stats[i].len = q->n_window - avail;
+		stats[i].limit = q->n_window - q->high_mark;
+		stats[i].count = q->n_window;
 
 	}
 	spin_unlock_irqrestore(&priv->lock, flags);
diff --git a/drivers/net/wireless/p54/p54.h b/drivers/net/wireless/p54/p54.h
index 06d2c67f4c81..c6f27b9022f9 100644
--- a/drivers/net/wireless/p54/p54.h
+++ b/drivers/net/wireless/p54/p54.h
@@ -64,7 +64,7 @@ struct p54_common {
 	unsigned int tx_hdr_len;
 	void *cached_vdcf;
 	unsigned int fw_var;
-	struct ieee80211_tx_queue_stats tx_stats;
+	struct ieee80211_tx_queue_stats tx_stats[4];
 };
 
 int p54_rx(struct ieee80211_hw *dev, struct sk_buff *skb);
diff --git a/drivers/net/wireless/p54/p54common.c b/drivers/net/wireless/p54/p54common.c
index 63f9badf3f52..59a26978f877 100644
--- a/drivers/net/wireless/p54/p54common.c
+++ b/drivers/net/wireless/p54/p54common.c
@@ -146,10 +146,10 @@ void p54_parse_firmware(struct ieee80211_hw *dev, const struct firmware *fw)
 
 	if (priv->fw_var >= 0x300) {
 		/* Firmware supports QoS, use it! */
-		priv->tx_stats.data[0].limit = 3;
-		priv->tx_stats.data[1].limit = 4;
-		priv->tx_stats.data[2].limit = 3;
-		priv->tx_stats.data[3].limit = 1;
+		priv->tx_stats[0].limit = 3;
+		priv->tx_stats[1].limit = 4;
+		priv->tx_stats[2].limit = 3;
+		priv->tx_stats[3].limit = 1;
 		dev->queues = 4;
 	}
 }
@@ -379,7 +379,7 @@ static void inline p54_wake_free_queues(struct ieee80211_hw *dev)
 	 * But, what if some are full? */
 
 	for (i = 0; i < dev->queues; i++)
-		if (priv->tx_stats.data[i].len < priv->tx_stats.data[i].limit)
+		if (priv->tx_stats[i].len < priv->tx_stats[i].limit)
 			ieee80211_wake_queue(dev, i);
 }
 
@@ -417,8 +417,7 @@ static void p54_rx_frame_sent(struct ieee80211_hw *dev, struct sk_buff *skb)
 			memcpy(&status.control, range->control,
 			       sizeof(status.control));
 			kfree(range->control);
-			priv->tx_stats.data[status.control.queue].len--;
-
+			priv->tx_stats[status.control.queue].len--;
 			entry_hdr = (struct p54_control_hdr *) entry->data;
 			entry_data = (struct p54_tx_control_allocdata *) entry_hdr->data;
 			if ((entry_hdr->magic1 & cpu_to_le16(0x4000)) != 0)
@@ -555,7 +554,7 @@ static void p54_assign_address(struct ieee80211_hw *dev, struct sk_buff *skb,
 static int p54_tx(struct ieee80211_hw *dev, struct sk_buff *skb,
 		  struct ieee80211_tx_control *control)
 {
-	struct ieee80211_tx_queue_stats_data *current_queue;
+	struct ieee80211_tx_queue_stats *current_queue;
 	struct p54_common *priv = dev->priv;
 	struct p54_control_hdr *hdr;
 	struct p54_tx_control_allocdata *txhdr;
@@ -563,7 +562,7 @@ static int p54_tx(struct ieee80211_hw *dev, struct sk_buff *skb,
 	size_t padding, len;
 	u8 rate;
 
-	current_queue = &priv->tx_stats.data[control->queue];
+	current_queue = &priv->tx_stats[control->queue];
 	if (unlikely(current_queue->len > current_queue->limit))
 		return NETDEV_TX_BUSY;
 	current_queue->len++;
@@ -967,11 +966,8 @@ static int p54_get_tx_stats(struct ieee80211_hw *dev,
 			    struct ieee80211_tx_queue_stats *stats)
 {
 	struct p54_common *priv = dev->priv;
-	unsigned int i;
 
-	for (i = 0; i < dev->queues; i++)
-		memcpy(&stats->data[i], &priv->tx_stats.data[i],
-			sizeof(stats->data[i]));
+	memcpy(stats, &priv->tx_stats, sizeof(stats[0]) * dev->queues);
 
 	return 0;
 }
@@ -1008,7 +1004,7 @@ struct ieee80211_hw *p54_init_common(size_t priv_data_len)
 	dev->channel_change_time = 1000;	/* TODO: find actual value */
 	dev->max_rssi = 127;
 
-	priv->tx_stats.data[0].limit = 5;
+	priv->tx_stats[0].limit = 5;
 	dev->queues = 1;
 
 	dev->extra_tx_headroom = sizeof(struct p54_control_hdr) + 4 +
diff --git a/drivers/net/wireless/rt2x00/rt2x00mac.c b/drivers/net/wireless/rt2x00/rt2x00mac.c
index d4ceab7646e7..7bc5129484f2 100644
--- a/drivers/net/wireless/rt2x00/rt2x00mac.c
+++ b/drivers/net/wireless/rt2x00/rt2x00mac.c
@@ -457,9 +457,9 @@ int rt2x00mac_get_tx_stats(struct ieee80211_hw *hw,
 	unsigned int i;
 
 	for (i = 0; i < hw->queues; i++) {
-		stats->data[i].len = rt2x00dev->tx[i].length;
-		stats->data[i].limit = rt2x00dev->tx[i].limit;
-		stats->data[i].count = rt2x00dev->tx[i].count;
+		stats[i].len = rt2x00dev->tx[i].length;
+		stats[i].limit = rt2x00dev->tx[i].limit;
+		stats[i].count = rt2x00dev->tx[i].count;
 	}
 
 	return 0;
diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index 740c11ca066c..346a7563ef02 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -117,13 +117,13 @@ struct ieee80211_tx_queue_params {
 };
 
 /**
- * struct ieee80211_tx_queue_stats_data - transmit queue statistics
+ * struct ieee80211_tx_queue_stats - transmit queue statistics
  *
  * @len: number of packets in queue
  * @limit: queue length limit
  * @count: number of frames sent
  */
-struct ieee80211_tx_queue_stats_data {
+struct ieee80211_tx_queue_stats {
 	unsigned int len;
 	unsigned int limit;
 	unsigned int count;
@@ -165,10 +165,6 @@ enum ieee80211_tx_queue {
 	NUM_TX_DATA_QUEUES_AMPDU = 16
 };
 
-struct ieee80211_tx_queue_stats {
-	struct ieee80211_tx_queue_stats_data data[NUM_TX_DATA_QUEUES_AMPDU];
-};
-
 struct ieee80211_low_level_stats {
 	unsigned int dot11ACKFailureCount;
 	unsigned int dot11RTSFailureCount;
@@ -1083,9 +1079,8 @@ enum ieee80211_ampdu_mlme_action {
  * @get_tx_stats: Get statistics of the current TX queue status. This is used
  *	to get number of currently queued packets (queue length), maximum queue
  *	size (limit), and total number of packets sent using each TX queue
- *	(count). This information is used for WMM to find out which TX
- *	queues have room for more packets and by hostapd to provide
- *	statistics about the current queueing state to external programs.
+ *	(count). The 'stats' pointer points to an array that has hw->queues +
+ *	hw->ampdu_queues items.
  *
  * @get_tsf: Get the current TSF timer value from firmware/hardware. Currently,
  *	this is only used for IBSS mode debugging and, as such, is not a
-- 
cgit v1.2.3


From 36fc6757fe711def63ea3686bf6ed475d714e114 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes@sipsolutions.net>
Date: Tue, 29 Apr 2008 17:19:00 +0200
Subject: mac80211: remove queue info from ieee80211_tx_status

The queue info in struct ieee80211_tx_status is never used.

Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/net/wireless/iwlwifi/iwl4965-base.c | 7 -------
 drivers/net/wireless/rt2x00/rt2x00dev.c     | 3 ---
 include/net/mac80211.h                      | 4 ----
 3 files changed, 14 deletions(-)

(limited to 'include')

diff --git a/drivers/net/wireless/iwlwifi/iwl4965-base.c b/drivers/net/wireless/iwlwifi/iwl4965-base.c
index 64cb905cdaed..5363be7a305e 100644
--- a/drivers/net/wireless/iwlwifi/iwl4965-base.c
+++ b/drivers/net/wireless/iwlwifi/iwl4965-base.c
@@ -2522,8 +2522,6 @@ static void iwl4965_txstatus_to_ieee(struct iwl_priv *priv,
 
 	tx_sta->status.ack_signal = 0;
 	tx_sta->status.excessive_retries = 0;
-	tx_sta->status.queue_length = 0;
-	tx_sta->status.queue_number = 0;
 
 	if (in_interrupt())
 		ieee80211_tx_status_irqsafe(priv->hw,
@@ -2659,8 +2657,6 @@ static int iwl4965_tx_status_reply_tx(struct iwl_priv *priv,
 
 		tx_status = &(priv->txq[txq_id].txb[idx].status);
 		tx_status->retry_count = tx_resp->failure_frame;
-		tx_status->queue_number = status & 0xff;
-		tx_status->queue_length = tx_resp->failure_rts;
 		tx_status->control.flags &= ~IEEE80211_TXCTL_AMPDU;
 		tx_status->flags = iwl4965_is_tx_success(status)?
 			IEEE80211_TX_STATUS_ACK : 0;
@@ -2820,9 +2816,6 @@ static void iwl4965_rx_reply_tx(struct iwl_priv *priv,
 	tx_status = &(txq->txb[txq->q.read_ptr].status);
 
 	tx_status->retry_count = tx_resp->failure_frame;
-	tx_status->queue_number = status;
-	tx_status->queue_length = tx_resp->bt_kill_count;
-	tx_status->queue_length |= tx_resp->failure_rts;
 	tx_status->flags =
 	    iwl4965_is_tx_success(status) ? IEEE80211_TX_STATUS_ACK : 0;
 	iwl4965_hwrate_to_tx_control(priv, le32_to_cpu(tx_resp->rate_n_flags),
diff --git a/drivers/net/wireless/rt2x00/rt2x00dev.c b/drivers/net/wireless/rt2x00/rt2x00dev.c
index b1a324dead5f..9929b15e28ba 100644
--- a/drivers/net/wireless/rt2x00/rt2x00dev.c
+++ b/drivers/net/wireless/rt2x00/rt2x00dev.c
@@ -525,9 +525,6 @@ void rt2x00lib_txdone(struct queue_entry *entry,
 			rt2x00dev->low_level_stats.dot11ACKFailureCount++;
 	}
 
-	tx_status.queue_length = entry->queue->limit;
-	tx_status.queue_number = tx_status.control.queue;
-
 	if (tx_status.control.flags & IEEE80211_TXCTL_USE_RTS_CTS) {
 		if (success)
 			rt2x00dev->low_level_stats.dot11RTSSuccessCount++;
diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index 346a7563ef02..ef701d6fd66a 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -417,8 +417,6 @@ enum ieee80211_tx_status_flags {
  * @ampdu_ack_map: block ack bit map for the aggregation.
  * 	relevant only if IEEE80211_TX_STATUS_AMPDU was set.
  * @ack_signal: signal strength of the ACK frame
- * @queue_length: ?? REMOVE
- * @queue_number: ?? REMOVE
  */
 struct ieee80211_tx_status {
 	struct ieee80211_tx_control control;
@@ -428,8 +426,6 @@ struct ieee80211_tx_status {
 	u8 ampdu_ack_len;
 	u64 ampdu_ack_map;
 	int ack_signal;
-	int queue_length;
-	int queue_number;
 };
 
 /**
-- 
cgit v1.2.3


From e100bb64bf7cdeae7f742a65ee1985649a7fd1b4 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes@sipsolutions.net>
Date: Wed, 30 Apr 2008 18:51:21 +0200
Subject: mac80211: QoS related cleanups

This
 * makes the queue number passed to drivers a u16
   (as it will be with skb_get_queue_mapping)
 * removes the useless queue number defines
 * splits hw->queues into hw->queues/ampdu_queues
 * removes the debugfs files for per-queue counters
 * removes some dead QoS code
 * removes the beacon queue configuration for IBSS
   so that the drivers now never get a queue number
   bigger than (hw->queues + hw->ampdu_queues - 1)
   for tx and only in the range 0..hw->queues-1 for
   conf_tx.

Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/net/wireless/b43/main.c             |   3 +-
 drivers/net/wireless/b43legacy/main.c       |   3 +-
 drivers/net/wireless/iwlwifi/iwl-core.c     |   2 +-
 drivers/net/wireless/iwlwifi/iwl3945-base.c |   2 +-
 drivers/net/wireless/iwlwifi/iwl4965-base.c |   2 +-
 drivers/net/wireless/p54/p54common.c        |   2 +-
 drivers/net/wireless/rt2x00/rt2400pci.c     |   5 +-
 drivers/net/wireless/rt2x00/rt2x00.h        |   2 +-
 drivers/net/wireless/rt2x00/rt2x00mac.c     |   2 +-
 drivers/net/wireless/rt2x00/rt2x00queue.h   |   7 +-
 include/net/mac80211.h                      |  66 ++++++---------
 net/mac80211/debugfs.c                      |  43 ----------
 net/mac80211/debugfs_sta.c                  |  38 ---------
 net/mac80211/ieee80211_i.h                  |   8 +-
 net/mac80211/main.c                         |   5 ++
 net/mac80211/mlme.c                         |  23 ++----
 net/mac80211/rx.c                           |   5 --
 net/mac80211/sta_info.c                     |   2 +-
 net/mac80211/wme.c                          | 120 ++++++++++++----------------
 19 files changed, 100 insertions(+), 240 deletions(-)

(limited to 'include')

diff --git a/drivers/net/wireless/b43/main.c b/drivers/net/wireless/b43/main.c
index cf546e367d3e..c6e79a15d3cb 100644
--- a/drivers/net/wireless/b43/main.c
+++ b/drivers/net/wireless/b43/main.c
@@ -2996,8 +2996,7 @@ static void b43_qos_update_work(struct work_struct *work)
 	mutex_unlock(&wl->mutex);
 }
 
-static int b43_op_conf_tx(struct ieee80211_hw *hw,
-			  int _queue,
+static int b43_op_conf_tx(struct ieee80211_hw *hw, u16 _queue,
 			  const struct ieee80211_tx_queue_params *params)
 {
 	struct b43_wl *wl = hw_to_b43_wl(hw);
diff --git a/drivers/net/wireless/b43legacy/main.c b/drivers/net/wireless/b43legacy/main.c
index 14a5eea2573e..1ccc51e90a8e 100644
--- a/drivers/net/wireless/b43legacy/main.c
+++ b/drivers/net/wireless/b43legacy/main.c
@@ -2383,8 +2383,7 @@ out:
 	return NETDEV_TX_OK;
 }
 
-static int b43legacy_op_conf_tx(struct ieee80211_hw *hw,
-				int queue,
+static int b43legacy_op_conf_tx(struct ieee80211_hw *hw, u16 queue,
 				const struct ieee80211_tx_queue_params *params)
 {
 	return 0;
diff --git a/drivers/net/wireless/iwlwifi/iwl-core.c b/drivers/net/wireless/iwlwifi/iwl-core.c
index 27e5f496bc12..c4b5c1a100f2 100644
--- a/drivers/net/wireless/iwlwifi/iwl-core.c
+++ b/drivers/net/wireless/iwlwifi/iwl-core.c
@@ -568,7 +568,7 @@ static void iwlcore_init_hw(struct iwl_priv *priv)
 	hw->queues = 4;
 #ifdef CONFIG_IWL4965_HT
 	/* Enhanced value; more queues, to support 11n aggregation */
-	hw->queues = 16;
+	hw->ampdu_queues = 12;
 #endif /* CONFIG_IWL4965_HT */
 }
 
diff --git a/drivers/net/wireless/iwlwifi/iwl3945-base.c b/drivers/net/wireless/iwlwifi/iwl3945-base.c
index bad367cfbee3..7040cde8bc28 100644
--- a/drivers/net/wireless/iwlwifi/iwl3945-base.c
+++ b/drivers/net/wireless/iwlwifi/iwl3945-base.c
@@ -7143,7 +7143,7 @@ static int iwl3945_mac_set_key(struct ieee80211_hw *hw, enum set_key_cmd cmd,
 	return rc;
 }
 
-static int iwl3945_mac_conf_tx(struct ieee80211_hw *hw, int queue,
+static int iwl3945_mac_conf_tx(struct ieee80211_hw *hw, u16 queue,
 			   const struct ieee80211_tx_queue_params *params)
 {
 	struct iwl3945_priv *priv = hw->priv;
diff --git a/drivers/net/wireless/iwlwifi/iwl4965-base.c b/drivers/net/wireless/iwlwifi/iwl4965-base.c
index 5363be7a305e..4406fc72d881 100644
--- a/drivers/net/wireless/iwlwifi/iwl4965-base.c
+++ b/drivers/net/wireless/iwlwifi/iwl4965-base.c
@@ -6339,7 +6339,7 @@ static int iwl4965_mac_set_key(struct ieee80211_hw *hw, enum set_key_cmd cmd,
 	return ret;
 }
 
-static int iwl4965_mac_conf_tx(struct ieee80211_hw *hw, int queue,
+static int iwl4965_mac_conf_tx(struct ieee80211_hw *hw, u16 queue,
 			   const struct ieee80211_tx_queue_params *params)
 {
 	struct iwl_priv *priv = hw->priv;
diff --git a/drivers/net/wireless/p54/p54common.c b/drivers/net/wireless/p54/p54common.c
index 59a26978f877..34b91ccd8aec 100644
--- a/drivers/net/wireless/p54/p54common.c
+++ b/drivers/net/wireless/p54/p54common.c
@@ -935,7 +935,7 @@ static void p54_configure_filter(struct ieee80211_hw *dev,
 	}
 }
 
-static int p54_conf_tx(struct ieee80211_hw *dev, int queue,
+static int p54_conf_tx(struct ieee80211_hw *dev, u16 queue,
 		       const struct ieee80211_tx_queue_params *params)
 {
 	struct p54_common *priv = dev->priv;
diff --git a/drivers/net/wireless/rt2x00/rt2400pci.c b/drivers/net/wireless/rt2x00/rt2400pci.c
index 705bc2d41dc1..247fbbfb0f2b 100644
--- a/drivers/net/wireless/rt2x00/rt2400pci.c
+++ b/drivers/net/wireless/rt2x00/rt2400pci.c
@@ -1442,8 +1442,7 @@ static int rt2400pci_set_retry_limit(struct ieee80211_hw *hw,
 	return 0;
 }
 
-static int rt2400pci_conf_tx(struct ieee80211_hw *hw,
-			     int queue,
+static int rt2400pci_conf_tx(struct ieee80211_hw *hw, u16 queue,
 			     const struct ieee80211_tx_queue_params *params)
 {
 	struct rt2x00_dev *rt2x00dev = hw->priv;
@@ -1453,7 +1452,7 @@ static int rt2400pci_conf_tx(struct ieee80211_hw *hw,
 	 * per queue. So by default we only configure the TX queue,
 	 * and ignore all other configurations.
 	 */
-	if (queue != IEEE80211_TX_QUEUE_DATA0)
+	if (queue != 0)
 		return -EINVAL;
 
 	if (rt2x00mac_conf_tx(hw, queue, params))
diff --git a/drivers/net/wireless/rt2x00/rt2x00.h b/drivers/net/wireless/rt2x00/rt2x00.h
index 48a9af47921e..79bd9c9f8963 100644
--- a/drivers/net/wireless/rt2x00/rt2x00.h
+++ b/drivers/net/wireless/rt2x00/rt2x00.h
@@ -997,7 +997,7 @@ void rt2x00mac_bss_info_changed(struct ieee80211_hw *hw,
 				struct ieee80211_vif *vif,
 				struct ieee80211_bss_conf *bss_conf,
 				u32 changes);
-int rt2x00mac_conf_tx(struct ieee80211_hw *hw, int queue,
+int rt2x00mac_conf_tx(struct ieee80211_hw *hw, u16 queue,
 		      const struct ieee80211_tx_queue_params *params);
 
 /*
diff --git a/drivers/net/wireless/rt2x00/rt2x00mac.c b/drivers/net/wireless/rt2x00/rt2x00mac.c
index 7bc5129484f2..767e0ffce04e 100644
--- a/drivers/net/wireless/rt2x00/rt2x00mac.c
+++ b/drivers/net/wireless/rt2x00/rt2x00mac.c
@@ -517,7 +517,7 @@ void rt2x00mac_bss_info_changed(struct ieee80211_hw *hw,
 }
 EXPORT_SYMBOL_GPL(rt2x00mac_bss_info_changed);
 
-int rt2x00mac_conf_tx(struct ieee80211_hw *hw, int queue_idx,
+int rt2x00mac_conf_tx(struct ieee80211_hw *hw, u16 queue_idx,
 		      const struct ieee80211_tx_queue_params *params)
 {
 	struct rt2x00_dev *rt2x00dev = hw->priv;
diff --git a/drivers/net/wireless/rt2x00/rt2x00queue.h b/drivers/net/wireless/rt2x00/rt2x00queue.h
index 29fe26525550..d1707a7ca41f 100644
--- a/drivers/net/wireless/rt2x00/rt2x00queue.h
+++ b/drivers/net/wireless/rt2x00/rt2x00queue.h
@@ -86,12 +86,9 @@ enum data_queue_qid {
 static inline enum data_queue_qid mac80211_queue_to_qid(unsigned int queue)
 {
 	/* Regular TX queues are mapped directly */
-	if (queue < NUM_TX_DATA_QUEUES)
+	if (queue < 4)
 		return queue;
-	else if (queue == IEEE80211_TX_QUEUE_BEACON)
-		return QID_BEACON;
-	else if (queue == IEEE80211_TX_QUEUE_AFTER_BEACON)
-		return QID_ATIM;
+	WARN_ON(1);
 	return QID_OTHER;
 }
 
diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index ef701d6fd66a..75a34609eed7 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -97,6 +97,18 @@ struct ieee80211_ht_bss_info {
 	u8 bss_op_mode; /* use IEEE80211_HT_IE_ */
 };
 
+/**
+ * enum ieee80211_max_queues - maximum number of queues
+ *
+ * @IEEE80211_MAX_QUEUES: Maximum number of regular device queues.
+ * @IEEE80211_MAX_AMPDU_QUEUES: Maximum number of queues usable
+ *	for A-MPDU operation.
+ */
+enum ieee80211_max_queues {
+	IEEE80211_MAX_QUEUES =		16,
+	IEEE80211_MAX_AMPDU_QUEUES =	16,
+};
+
 /**
  * struct ieee80211_tx_queue_params - transmit queue configuration
  *
@@ -129,42 +141,6 @@ struct ieee80211_tx_queue_stats {
 	unsigned int count;
 };
 
-/**
- * enum ieee80211_tx_queue - transmit queue number
- *
- * These constants are used with some callbacks that take a
- * queue number to set parameters for a queue.
- *
- * @IEEE80211_TX_QUEUE_DATA0: data queue 0
- * @IEEE80211_TX_QUEUE_DATA1: data queue 1
- * @IEEE80211_TX_QUEUE_DATA2: data queue 2
- * @IEEE80211_TX_QUEUE_DATA3: data queue 3
- * @IEEE80211_TX_QUEUE_DATA4: data queue 4
- * @IEEE80211_TX_QUEUE_SVP: ??
- * @NUM_TX_DATA_QUEUES: number of data queues
- * @IEEE80211_TX_QUEUE_AFTER_BEACON: transmit queue for frames to be
- *	sent after a beacon
- * @IEEE80211_TX_QUEUE_BEACON: transmit queue for beacon frames
- * @NUM_TX_DATA_QUEUES_AMPDU: adding more queues for A-MPDU
- */
-enum ieee80211_tx_queue {
-	IEEE80211_TX_QUEUE_DATA0,
-	IEEE80211_TX_QUEUE_DATA1,
-	IEEE80211_TX_QUEUE_DATA2,
-	IEEE80211_TX_QUEUE_DATA3,
-	IEEE80211_TX_QUEUE_DATA4,
-	IEEE80211_TX_QUEUE_SVP,
-
-	NUM_TX_DATA_QUEUES,
-
-/* due to stupidity in the sub-ioctl userspace interface, the items in
- * this struct need to have fixed values. As soon as it is removed, we can
- * fix these entries. */
-	IEEE80211_TX_QUEUE_AFTER_BEACON = 6,
-	IEEE80211_TX_QUEUE_BEACON = 7,
-	NUM_TX_DATA_QUEUES_AMPDU = 16
-};
-
 struct ieee80211_low_level_stats {
 	unsigned int dot11ACKFailureCount;
 	unsigned int dot11RTSFailureCount;
@@ -315,7 +291,7 @@ struct ieee80211_tx_control {
 				 * position represents antenna number used */
 	u8 icv_len;		/* length of the ICV/MIC field in octets */
 	u8 iv_len;		/* length of the IV field in octets */
-	u8 queue;		/* hardware queue to use for this frame;
+	u16 queue;		/* hardware queue to use for this frame;
 				 * 0 = highest, hw->queues-1 = lowest */
 	u16 aid;		/* Station AID */
 	int type;	/* internal */
@@ -772,7 +748,14 @@ enum ieee80211_hw_flags {
  * @max_noise: like @max_rssi, but for the noise value.
  *
  * @queues: number of available hardware transmit queues for
- *	data packets. WMM/QoS requires at least four.
+ *	data packets. WMM/QoS requires at least four, these
+ *	queues need to have configurable access parameters.
+ *
+ * @ampdu_queues: number of available hardware transmit queues
+ *	for A-MPDU packets, these have no access parameters
+ *	because they're used only for A-MPDU frames. Note that
+ *	mac80211 will not currently use any of the regular queues
+ *	for aggregation.
  *
  * @rate_control_algorithm: rate control algorithm for this hardware.
  *	If unset (NULL), the default algorithm will be used. Must be
@@ -791,7 +774,7 @@ struct ieee80211_hw {
 	unsigned int extra_tx_headroom;
 	int channel_change_time;
 	int vif_data_size;
-	u8 queues;
+	u16 queues, ampdu_queues;
 	s8 max_rssi;
 	s8 max_signal;
 	s8 max_noise;
@@ -1069,8 +1052,7 @@ enum ieee80211_ampdu_mlme_action {
  *	of assocaited station or AP.
  *
  * @conf_tx: Configure TX queue parameters (EDCF (aifs, cw_min, cw_max),
- *	bursting) for a hardware TX queue. The @queue parameter uses the
- *	%IEEE80211_TX_QUEUE_* constants. Must be atomic.
+ *	bursting) for a hardware TX queue. Must be atomic.
  *
  * @get_tx_stats: Get statistics of the current TX queue status. This is used
  *	to get number of currently queued packets (queue length), maximum queue
@@ -1150,7 +1132,7 @@ struct ieee80211_ops {
 			       u32 short_retry, u32 long_retr);
 	void (*sta_notify)(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 			enum sta_notify_cmd, const u8 *addr);
-	int (*conf_tx)(struct ieee80211_hw *hw, int queue,
+	int (*conf_tx)(struct ieee80211_hw *hw, u16 queue,
 		       const struct ieee80211_tx_queue_params *params);
 	int (*get_tx_stats)(struct ieee80211_hw *hw,
 			    struct ieee80211_tx_queue_stats *stats);
diff --git a/net/mac80211/debugfs.c b/net/mac80211/debugfs.c
index 1cccbfd781f6..d20d90eead1f 100644
--- a/net/mac80211/debugfs.c
+++ b/net/mac80211/debugfs.c
@@ -197,45 +197,6 @@ DEBUGFS_STATS_FILE(rx_handlers_fragments, 20, "%u",
 DEBUGFS_STATS_FILE(tx_status_drop, 20, "%u",
 		   local->tx_status_drop);
 
-static ssize_t stats_wme_rx_queue_read(struct file *file,
-				       char __user *userbuf,
-				       size_t count, loff_t *ppos)
-{
-	struct ieee80211_local *local = file->private_data;
-	char buf[NUM_RX_DATA_QUEUES*15], *p = buf;
-	int i;
-
-	for (i = 0; i < NUM_RX_DATA_QUEUES; i++)
-		p += scnprintf(p, sizeof(buf)+buf-p,
-			       "%u\n", local->wme_rx_queue[i]);
-
-	return simple_read_from_buffer(userbuf, count, ppos, buf, p-buf);
-}
-
-static const struct file_operations stats_wme_rx_queue_ops = {
-	.read = stats_wme_rx_queue_read,
-	.open = mac80211_open_file_generic,
-};
-
-static ssize_t stats_wme_tx_queue_read(struct file *file,
-				       char __user *userbuf,
-				       size_t count, loff_t *ppos)
-{
-	struct ieee80211_local *local = file->private_data;
-	char buf[NUM_TX_DATA_QUEUES*15], *p = buf;
-	int i;
-
-	for (i = 0; i < NUM_TX_DATA_QUEUES; i++)
-		p += scnprintf(p, sizeof(buf)+buf-p,
-			       "%u\n", local->wme_tx_queue[i]);
-
-	return simple_read_from_buffer(userbuf, count, ppos, buf, p-buf);
-}
-
-static const struct file_operations stats_wme_tx_queue_ops = {
-	.read = stats_wme_tx_queue_read,
-	.open = mac80211_open_file_generic,
-};
 #endif
 
 DEBUGFS_DEVSTATS_FILE(dot11ACKFailureCount);
@@ -303,8 +264,6 @@ void debugfs_hw_add(struct ieee80211_local *local)
 	DEBUGFS_STATS_ADD(rx_expand_skb_head2);
 	DEBUGFS_STATS_ADD(rx_handlers_fragments);
 	DEBUGFS_STATS_ADD(tx_status_drop);
-	DEBUGFS_STATS_ADD(wme_tx_queue);
-	DEBUGFS_STATS_ADD(wme_rx_queue);
 #endif
 	DEBUGFS_STATS_ADD(dot11ACKFailureCount);
 	DEBUGFS_STATS_ADD(dot11RTSFailureCount);
@@ -356,8 +315,6 @@ void debugfs_hw_del(struct ieee80211_local *local)
 	DEBUGFS_STATS_DEL(rx_expand_skb_head2);
 	DEBUGFS_STATS_DEL(rx_handlers_fragments);
 	DEBUGFS_STATS_DEL(tx_status_drop);
-	DEBUGFS_STATS_DEL(wme_tx_queue);
-	DEBUGFS_STATS_DEL(wme_rx_queue);
 #endif
 	DEBUGFS_STATS_DEL(dot11ACKFailureCount);
 	DEBUGFS_STATS_DEL(dot11RTSFailureCount);
diff --git a/net/mac80211/debugfs_sta.c b/net/mac80211/debugfs_sta.c
index 6d47a1d31b37..676a93202ff9 100644
--- a/net/mac80211/debugfs_sta.c
+++ b/net/mac80211/debugfs_sta.c
@@ -123,36 +123,6 @@ static ssize_t sta_last_seq_ctrl_read(struct file *file, char __user *userbuf,
 }
 STA_OPS(last_seq_ctrl);
 
-#ifdef CONFIG_MAC80211_DEBUG_COUNTERS
-static ssize_t sta_wme_rx_queue_read(struct file *file, char __user *userbuf,
-				     size_t count, loff_t *ppos)
-{
-	char buf[15*NUM_RX_DATA_QUEUES], *p = buf;
-	int i;
-	struct sta_info *sta = file->private_data;
-	for (i = 0; i < NUM_RX_DATA_QUEUES; i++)
-		p += scnprintf(p, sizeof(buf)+buf-p, "%u ",
-			       sta->wme_rx_queue[i]);
-	p += scnprintf(p, sizeof(buf)+buf-p, "\n");
-	return simple_read_from_buffer(userbuf, count, ppos, buf, p - buf);
-}
-STA_OPS(wme_rx_queue);
-
-static ssize_t sta_wme_tx_queue_read(struct file *file, char __user *userbuf,
-				     size_t count, loff_t *ppos)
-{
-	char buf[15*NUM_TX_DATA_QUEUES], *p = buf;
-	int i;
-	struct sta_info *sta = file->private_data;
-	for (i = 0; i < NUM_TX_DATA_QUEUES; i++)
-		p += scnprintf(p, sizeof(buf)+buf-p, "%u ",
-			       sta->wme_tx_queue[i]);
-	p += scnprintf(p, sizeof(buf)+buf-p, "\n");
-	return simple_read_from_buffer(userbuf, count, ppos, buf, p - buf);
-}
-STA_OPS(wme_tx_queue);
-#endif
-
 static ssize_t sta_agg_status_read(struct file *file, char __user *userbuf,
 					size_t count, loff_t *ppos)
 {
@@ -293,10 +263,6 @@ void ieee80211_sta_debugfs_add(struct sta_info *sta)
 	DEBUGFS_ADD(num_ps_buf_frames);
 	DEBUGFS_ADD(inactive_ms);
 	DEBUGFS_ADD(last_seq_ctrl);
-#ifdef CONFIG_MAC80211_DEBUG_COUNTERS
-	DEBUGFS_ADD(wme_rx_queue);
-	DEBUGFS_ADD(wme_tx_queue);
-#endif
 	DEBUGFS_ADD(agg_status);
 }
 
@@ -306,10 +272,6 @@ void ieee80211_sta_debugfs_remove(struct sta_info *sta)
 	DEBUGFS_DEL(num_ps_buf_frames);
 	DEBUGFS_DEL(inactive_ms);
 	DEBUGFS_DEL(last_seq_ctrl);
-#ifdef CONFIG_MAC80211_DEBUG_COUNTERS
-	DEBUGFS_DEL(wme_rx_queue);
-	DEBUGFS_DEL(wme_tx_queue);
-#endif
 	DEBUGFS_DEL(agg_status);
 
 	debugfs_remove(sta->debugfs.dir);
diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h
index cabdc1439d57..d82ed20a344c 100644
--- a/net/mac80211/ieee80211_i.h
+++ b/net/mac80211/ieee80211_i.h
@@ -610,8 +610,8 @@ struct ieee80211_local {
 	struct sta_info *sta_hash[STA_HASH_SIZE];
 	struct timer_list sta_cleanup;
 
-	unsigned long state[NUM_TX_DATA_QUEUES_AMPDU];
-	struct ieee80211_tx_stored_packet pending_packet[NUM_TX_DATA_QUEUES_AMPDU];
+	unsigned long state[IEEE80211_MAX_AMPDU_QUEUES + IEEE80211_MAX_AMPDU_QUEUES];
+	struct ieee80211_tx_stored_packet pending_packet[IEEE80211_MAX_AMPDU_QUEUES + IEEE80211_MAX_AMPDU_QUEUES];
 	struct tasklet_struct tx_pending_tasklet;
 
 	/* number of interfaces with corresponding IFF_ flags */
@@ -705,8 +705,6 @@ struct ieee80211_local {
 	unsigned int rx_expand_skb_head2;
 	unsigned int rx_handlers_fragments;
 	unsigned int tx_status_drop;
-	unsigned int wme_rx_queue[NUM_RX_DATA_QUEUES];
-	unsigned int wme_tx_queue[NUM_RX_DATA_QUEUES];
 #define I802_DEBUG_INC(c) (c)++
 #else /* CONFIG_MAC80211_DEBUG_COUNTERS */
 #define I802_DEBUG_INC(c) do { } while (0)
@@ -764,8 +762,6 @@ struct ieee80211_local {
 			struct dentry *rx_expand_skb_head2;
 			struct dentry *rx_handlers_fragments;
 			struct dentry *tx_status_drop;
-			struct dentry *wme_tx_queue;
-			struct dentry *wme_rx_queue;
 #endif
 			struct dentry *dot11ACKFailureCount;
 			struct dentry *dot11RTSFailureCount;
diff --git a/net/mac80211/main.c b/net/mac80211/main.c
index e19be27a3def..55e76117da9e 100644
--- a/net/mac80211/main.c
+++ b/net/mac80211/main.c
@@ -1745,6 +1745,11 @@ int ieee80211_register_hw(struct ieee80211_hw *hw)
 		goto fail_wep;
 	}
 
+	if (hw->queues > IEEE80211_MAX_QUEUES)
+		hw->queues = IEEE80211_MAX_QUEUES;
+	if (hw->ampdu_queues > IEEE80211_MAX_AMPDU_QUEUES)
+		hw->ampdu_queues = IEEE80211_MAX_AMPDU_QUEUES;
+
 	ieee80211_install_qdisc(local->mdev);
 
 	/* add one default STA interface */
diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c
index e860d0bacea9..55b85ae5bc11 100644
--- a/net/mac80211/mlme.c
+++ b/net/mac80211/mlme.c
@@ -257,19 +257,8 @@ static void ieee80211_sta_def_wmm_params(struct net_device *dev,
 		qparam.cw_max = 1023;
 		qparam.txop = 0;
 
-		for (i = IEEE80211_TX_QUEUE_DATA0; i < NUM_TX_DATA_QUEUES; i++)
-			local->ops->conf_tx(local_to_hw(local),
-					   i + IEEE80211_TX_QUEUE_DATA0,
-					   &qparam);
-
-		if (ibss) {
-			/* IBSS uses different parameters for Beacon sending */
-			qparam.cw_min++;
-			qparam.cw_min *= 2;
-			qparam.cw_min--;
-			local->ops->conf_tx(local_to_hw(local),
-					   IEEE80211_TX_QUEUE_BEACON, &qparam);
-		}
+		for (i = 0; i < local_to_hw(local)->queues; i++)
+			local->ops->conf_tx(local_to_hw(local), i, &qparam);
 	}
 }
 
@@ -306,23 +295,23 @@ static void ieee80211_sta_wmm_params(struct net_device *dev,
 
 		switch (aci) {
 		case 1:
-			queue = IEEE80211_TX_QUEUE_DATA3;
+			queue = 3;
 			if (acm)
 				local->wmm_acm |= BIT(0) | BIT(3);
 			break;
 		case 2:
-			queue = IEEE80211_TX_QUEUE_DATA1;
+			queue = 1;
 			if (acm)
 				local->wmm_acm |= BIT(4) | BIT(5);
 			break;
 		case 3:
-			queue = IEEE80211_TX_QUEUE_DATA0;
+			queue = 0;
 			if (acm)
 				local->wmm_acm |= BIT(6) | BIT(7);
 			break;
 		case 0:
 		default:
-			queue = IEEE80211_TX_QUEUE_DATA2;
+			queue = 2;
 			if (acm)
 				local->wmm_acm |= BIT(1) | BIT(2);
 			break;
diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c
index 02f436a86061..e8b89c89e875 100644
--- a/net/mac80211/rx.c
+++ b/net/mac80211/rx.c
@@ -275,11 +275,6 @@ static void ieee80211_parse_qos(struct ieee80211_rx_data *rx)
 		}
 	}
 
-	I802_DEBUG_INC(rx->local->wme_rx_queue[tid]);
-	/* only a debug counter, sta might not be assigned properly yet */
-	if (rx->sta)
-		I802_DEBUG_INC(rx->sta->wme_rx_queue[tid]);
-
 	rx->queue = tid;
 	/* Set skb->priority to 1d tag if highest order bit of TID is not set.
 	 * For now, set skb->priority to 0 for other cases. */
diff --git a/net/mac80211/sta_info.c b/net/mac80211/sta_info.c
index 7d4fe4a52929..631943e8af8b 100644
--- a/net/mac80211/sta_info.c
+++ b/net/mac80211/sta_info.c
@@ -257,7 +257,7 @@ struct sta_info *sta_info_alloc(struct ieee80211_sub_if_data *sdata,
 		 * sta_rx_agg_session_timer_expired for useage */
 		sta->timer_to_tid[i] = i;
 		/* tid to tx queue: initialize according to HW (0 is valid) */
-		sta->tid_to_tx_q[i] = local->hw.queues;
+		sta->tid_to_tx_q[i] = local->hw.queues + local->hw.ampdu_queues;
 		/* rx */
 		sta->ampdu_mlme.tid_state_rx[i] = HT_AGG_STATE_IDLE;
 		sta->ampdu_mlme.tid_rx[i] = NULL;
diff --git a/net/mac80211/wme.c b/net/mac80211/wme.c
index 64faa3dc488f..5eddf1f32ed9 100644
--- a/net/mac80211/wme.c
+++ b/net/mac80211/wme.c
@@ -19,16 +19,22 @@
 #include "wme.h"
 
 /* maximum number of hardware queues we support. */
-#define TC_80211_MAX_QUEUES 16
+#define QD_MAX_QUEUES (IEEE80211_MAX_AMPDU_QUEUES + IEEE80211_MAX_QUEUES)
+/* current number of hardware queues we support. */
+#define QD_NUM(hw) ((hw)->queues + (hw)->ampdu_queues)
 
+/*
+ * Default mapping in classifier to work with default
+ * queue setup.
+ */
 const int ieee802_1d_to_ac[8] = { 2, 3, 3, 2, 1, 1, 0, 0 };
 
 struct ieee80211_sched_data
 {
-	unsigned long qdisc_pool[BITS_TO_LONGS(TC_80211_MAX_QUEUES)];
+	unsigned long qdisc_pool[BITS_TO_LONGS(QD_MAX_QUEUES)];
 	struct tcf_proto *filter_list;
-	struct Qdisc *queues[TC_80211_MAX_QUEUES];
-	struct sk_buff_head requeued[TC_80211_MAX_QUEUES];
+	struct Qdisc *queues[QD_MAX_QUEUES];
+	struct sk_buff_head requeued[QD_MAX_QUEUES];
 };
 
 static const char llc_ip_hdr[8] = {0xAA, 0xAA, 0x3, 0, 0, 0, 0x08, 0};
@@ -95,7 +101,7 @@ static inline int wme_downgrade_ac(struct sk_buff *skb)
 
 /* positive return value indicates which queue to use
  * negative return value indicates to drop the frame */
-static inline int classify80211(struct sk_buff *skb, struct Qdisc *qd)
+static int classify80211(struct sk_buff *skb, struct Qdisc *qd)
 {
 	struct ieee80211_local *local = wdev_priv(qd->dev->ieee80211_ptr);
 	struct ieee80211_hdr *hdr = (struct ieee80211_hdr *) skb->data;
@@ -106,7 +112,7 @@ static inline int classify80211(struct sk_buff *skb, struct Qdisc *qd)
 	if (unlikely((fc & IEEE80211_FCTL_FTYPE) != IEEE80211_FTYPE_DATA)) {
 		/* management frames go on AC_VO queue, but are sent
 		* without QoS control fields */
-		return IEEE80211_TX_QUEUE_DATA0;
+		return 0;
 	}
 
 	if (0 /* injected */) {
@@ -141,14 +147,16 @@ static inline int classify80211(struct sk_buff *skb, struct Qdisc *qd)
 static int wme_qdiscop_enqueue(struct sk_buff *skb, struct Qdisc* qd)
 {
 	struct ieee80211_local *local = wdev_priv(qd->dev->ieee80211_ptr);
+	struct ieee80211_hw *hw = &local->hw;
 	struct ieee80211_sched_data *q = qdisc_priv(qd);
 	struct ieee80211_tx_packet_data *pkt_data =
 		(struct ieee80211_tx_packet_data *) skb->cb;
 	struct ieee80211_hdr *hdr = (struct ieee80211_hdr *) skb->data;
 	unsigned short fc = le16_to_cpu(hdr->frame_control);
 	struct Qdisc *qdisc;
-	int err, queue;
 	struct sta_info *sta;
+	int err;
+	u16 queue;
 	u8 tid;
 
 	if (pkt_data->flags & IEEE80211_TXPD_REQUEUE) {
@@ -158,7 +166,7 @@ static int wme_qdiscop_enqueue(struct sk_buff *skb, struct Qdisc* qd)
 		tid = skb->priority & QOS_CONTROL_TAG1D_MASK;
 		if (sta) {
 			int ampdu_queue = sta->tid_to_tx_q[tid];
-			if ((ampdu_queue < local->hw.queues) &&
+			if ((ampdu_queue < QD_NUM(hw)) &&
 			    test_bit(ampdu_queue, q->qdisc_pool)) {
 				queue = ampdu_queue;
 				pkt_data->flags |= IEEE80211_TXPD_AMPDU;
@@ -174,6 +182,9 @@ static int wme_qdiscop_enqueue(struct sk_buff *skb, struct Qdisc* qd)
 
 	queue = classify80211(skb, qd);
 
+	if (unlikely(queue >= local->hw.queues))
+		queue = local->hw.queues - 1;
+
 	/* now we know the 1d priority, fill in the QoS header if there is one
 	 */
 	if (WLAN_FC_IS_QOS_DATA(fc)) {
@@ -193,8 +204,8 @@ static int wme_qdiscop_enqueue(struct sk_buff *skb, struct Qdisc* qd)
 		sta = sta_info_get(local, hdr->addr1);
 		if (sta) {
 			int ampdu_queue = sta->tid_to_tx_q[tid];
-			if ((ampdu_queue < local->hw.queues) &&
-				test_bit(ampdu_queue, q->qdisc_pool)) {
+			if ((ampdu_queue < QD_NUM(hw)) &&
+			    test_bit(ampdu_queue, q->qdisc_pool)) {
 				queue = ampdu_queue;
 				pkt_data->flags |= IEEE80211_TXPD_AMPDU;
 			} else {
@@ -205,17 +216,6 @@ static int wme_qdiscop_enqueue(struct sk_buff *skb, struct Qdisc* qd)
 		rcu_read_unlock();
 	}
 
-	if (unlikely(queue >= local->hw.queues)) {
-#if 0
-		if (net_ratelimit()) {
-			printk(KERN_DEBUG "%s - queue=%d (hw does not "
-			       "support) -> %d\n",
-			       __func__, queue, local->hw.queues - 1);
-		}
-#endif
-		queue = local->hw.queues - 1;
-	}
-
 	if (unlikely(queue < 0)) {
 			kfree_skb(skb);
 			err = NET_XMIT_DROP;
@@ -270,7 +270,7 @@ static struct sk_buff *wme_qdiscop_dequeue(struct Qdisc* qd)
 	int queue;
 
 	/* check all the h/w queues in numeric/priority order */
-	for (queue = 0; queue < hw->queues; queue++) {
+	for (queue = 0; queue < QD_NUM(hw); queue++) {
 		/* see if there is room in this hardware queue */
 		if ((test_bit(IEEE80211_LINK_STATE_XOFF,
 				&local->state[queue])) ||
@@ -308,7 +308,7 @@ static void wme_qdiscop_reset(struct Qdisc* qd)
 
 	/* QUESTION: should we have some hardware flush functionality here? */
 
-	for (queue = 0; queue < hw->queues; queue++) {
+	for (queue = 0; queue < QD_NUM(hw); queue++) {
 		skb_queue_purge(&q->requeued[queue]);
 		qdisc_reset(q->queues[queue]);
 	}
@@ -326,7 +326,7 @@ static void wme_qdiscop_destroy(struct Qdisc* qd)
 	tcf_destroy_chain(q->filter_list);
 	q->filter_list = NULL;
 
-	for (queue=0; queue < hw->queues; queue++) {
+	for (queue = 0; queue < QD_NUM(hw); queue++) {
 		skb_queue_purge(&q->requeued[queue]);
 		qdisc_destroy(q->queues[queue]);
 		q->queues[queue] = &noop_qdisc;
@@ -337,17 +337,6 @@ static void wme_qdiscop_destroy(struct Qdisc* qd)
 /* called whenever parameters are updated on existing qdisc */
 static int wme_qdiscop_tune(struct Qdisc *qd, struct nlattr *opt)
 {
-/*	struct ieee80211_sched_data *q = qdisc_priv(qd);
-*/
-	/* check our options block is the right size */
-	/* copy any options to our local structure */
-/*	Ignore options block for now - always use static mapping
-	struct tc_ieee80211_qopt *qopt = nla_data(opt);
-
-	if (opt->nla_len < nla_attr_size(sizeof(*qopt)))
-		return -EINVAL;
-	memcpy(q->tag2queue, qopt->tag2queue, sizeof(qopt->tag2queue));
-*/
 	return 0;
 }
 
@@ -358,7 +347,7 @@ static int wme_qdiscop_init(struct Qdisc *qd, struct nlattr *opt)
 	struct ieee80211_sched_data *q = qdisc_priv(qd);
 	struct net_device *dev = qd->dev;
 	struct ieee80211_local *local;
-	int queues;
+	struct ieee80211_hw *hw;
 	int err = 0, i;
 
 	/* check that device is a mac80211 device */
@@ -366,29 +355,26 @@ static int wme_qdiscop_init(struct Qdisc *qd, struct nlattr *opt)
 	    dev->ieee80211_ptr->wiphy->privid != mac80211_wiphy_privid)
 		return -EINVAL;
 
-	/* check this device is an ieee80211 master type device */
-	if (dev->type != ARPHRD_IEEE80211)
+	local = wdev_priv(dev->ieee80211_ptr);
+	hw = &local->hw;
+
+	/* only allow on master dev */
+	if (dev != local->mdev)
 		return -EINVAL;
 
-	/* check that there is no qdisc currently attached to device
-	 * this ensures that we will be the root qdisc. (I can't find a better
-	 * way to test this explicitly) */
-	if (dev->qdisc_sleeping != &noop_qdisc)
+	/* ensure that we are root qdisc */
+	if (qd->parent != TC_H_ROOT)
 		return -EINVAL;
 
 	if (qd->flags & TCQ_F_INGRESS)
 		return -EINVAL;
 
-	local = wdev_priv(dev->ieee80211_ptr);
-	queues = local->hw.queues;
-
 	/* if options were passed in, set them */
-	if (opt) {
+	if (opt)
 		err = wme_qdiscop_tune(qd, opt);
-	}
 
 	/* create child queues */
-	for (i = 0; i < queues; i++) {
+	for (i = 0; i < QD_NUM(hw); i++) {
 		skb_queue_head_init(&q->requeued[i]);
 		q->queues[i] = qdisc_create_dflt(qd->dev, &pfifo_qdisc_ops,
 						 qd->handle);
@@ -398,8 +384,8 @@ static int wme_qdiscop_init(struct Qdisc *qd, struct nlattr *opt)
 		}
 	}
 
-	/* reserve all legacy QoS queues */
-	for (i = 0; i < min(IEEE80211_TX_QUEUE_DATA4, queues); i++)
+	/* non-aggregation queues: reserve/mark as used */
+	for (i = 0; i < local->hw.queues; i++)
 		set_bit(i, q->qdisc_pool);
 
 	return err;
@@ -407,16 +393,6 @@ static int wme_qdiscop_init(struct Qdisc *qd, struct nlattr *opt)
 
 static int wme_qdiscop_dump(struct Qdisc *qd, struct sk_buff *skb)
 {
-/*	struct ieee80211_sched_data *q = qdisc_priv(qd);
-	unsigned char *p = skb->tail;
-	struct tc_ieee80211_qopt opt;
-
-	memcpy(&opt.tag2queue, q->tag2queue, TC_80211_MAX_TAG + 1);
-	NLA_PUT(skb, TCA_OPTIONS, sizeof(opt), &opt);
-*/	return skb->len;
-/*
-nla_put_failure:
-	skb_trim(skb, p - skb->data);*/
 	return -1;
 }
 
@@ -429,7 +405,7 @@ static int wme_classop_graft(struct Qdisc *qd, unsigned long arg,
 	struct ieee80211_hw *hw = &local->hw;
 	unsigned long queue = arg - 1;
 
-	if (queue >= hw->queues)
+	if (queue >= QD_NUM(hw))
 		return -EINVAL;
 
 	if (!new)
@@ -453,7 +429,7 @@ wme_classop_leaf(struct Qdisc *qd, unsigned long arg)
 	struct ieee80211_hw *hw = &local->hw;
 	unsigned long queue = arg - 1;
 
-	if (queue >= hw->queues)
+	if (queue >= QD_NUM(hw))
 		return NULL;
 
 	return q->queues[queue];
@@ -466,7 +442,7 @@ static unsigned long wme_classop_get(struct Qdisc *qd, u32 classid)
 	struct ieee80211_hw *hw = &local->hw;
 	unsigned long queue = TC_H_MIN(classid);
 
-	if (queue - 1 >= hw->queues)
+	if (queue - 1 >= QD_NUM(hw))
 		return 0;
 
 	return queue;
@@ -492,7 +468,7 @@ static int wme_classop_change(struct Qdisc *qd, u32 handle, u32 parent,
 	struct ieee80211_local *local = wdev_priv(qd->dev->ieee80211_ptr);
 	struct ieee80211_hw *hw = &local->hw;
 
-	if (cl - 1 > hw->queues)
+	if (cl - 1 > QD_NUM(hw))
 		return -ENOENT;
 
 	/* TODO: put code to program hardware queue parameters here,
@@ -509,7 +485,7 @@ static int wme_classop_delete(struct Qdisc *qd, unsigned long cl)
 	struct ieee80211_local *local = wdev_priv(qd->dev->ieee80211_ptr);
 	struct ieee80211_hw *hw = &local->hw;
 
-	if (cl - 1 > hw->queues)
+	if (cl - 1 > QD_NUM(hw))
 		return -ENOENT;
 	return 0;
 }
@@ -522,7 +498,7 @@ static int wme_classop_dump_class(struct Qdisc *qd, unsigned long cl,
 	struct ieee80211_local *local = wdev_priv(qd->dev->ieee80211_ptr);
 	struct ieee80211_hw *hw = &local->hw;
 
-	if (cl - 1 > hw->queues)
+	if (cl - 1 > QD_NUM(hw))
 		return -ENOENT;
 	tcm->tcm_handle = TC_H_MIN(cl);
 	tcm->tcm_parent = qd->handle;
@@ -540,7 +516,7 @@ static void wme_classop_walk(struct Qdisc *qd, struct qdisc_walker *arg)
 	if (arg->stop)
 		return;
 
-	for (queue = 0; queue < hw->queues; queue++) {
+	for (queue = 0; queue < QD_NUM(hw); queue++) {
 		if (arg->count < arg->skip) {
 			arg->count++;
 			continue;
@@ -657,10 +633,13 @@ int ieee80211_ht_agg_queue_add(struct ieee80211_local *local,
 	DECLARE_MAC_BUF(mac);
 
 	/* prepare the filter and save it for the SW queue
-	 * matching the recieved HW queue */
+	 * matching the received HW queue */
+
+	if (!local->hw.ampdu_queues)
+		return -EPERM;
 
 	/* try to get a Qdisc from the pool */
-	for (i = IEEE80211_TX_QUEUE_BEACON; i < local->hw.queues; i++)
+	for (i = local->hw.queues; i < QD_NUM(&local->hw); i++)
 		if (!test_and_set_bit(i, q->qdisc_pool)) {
 			ieee80211_stop_queue(local_to_hw(local), i);
 			sta->tid_to_tx_q[tid] = i;
@@ -689,13 +668,14 @@ void ieee80211_ht_agg_queue_remove(struct ieee80211_local *local,
 				   struct sta_info *sta, u16 tid,
 				   u8 requeue)
 {
+	struct ieee80211_hw *hw = &local->hw;
 	struct ieee80211_sched_data *q =
 		qdisc_priv(local->mdev->qdisc_sleeping);
 	int agg_queue = sta->tid_to_tx_q[tid];
 
 	/* return the qdisc to the pool */
 	clear_bit(agg_queue, q->qdisc_pool);
-	sta->tid_to_tx_q[tid] = local->hw.queues;
+	sta->tid_to_tx_q[tid] = QD_NUM(hw);
 
 	if (requeue)
 		ieee80211_requeue(local, agg_queue);
-- 
cgit v1.2.3


From d364d9276b54af16fcb4db83f1315b620daec102 Mon Sep 17 00:00:00 2001
From: Wei Yongjun <yjwei@cn.fujitsu.com>
Date: Fri, 9 May 2008 15:13:26 -0700
Subject: sctp: Bring SCTP_DELAYED_ACK socket option into API compliance

Brings delayed_ack socket option set/get into line with the latest ietf
socket extensions API draft, while maintaining backwards compatibility.

Signed-off-by: Wei Yongjun <yjwei@cn.fujitsu.com>
Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/sctp/structs.h |   4 +
 include/net/sctp/user.h    |  28 ++++--
 net/sctp/associola.c       |   3 +
 net/sctp/sm_sideeffect.c   |  17 +++-
 net/sctp/socket.c          | 242 ++++++++++++++++++++++++++++-----------------
 5 files changed, 194 insertions(+), 100 deletions(-)

(limited to 'include')

diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h
index 0ce0443c5b79..e11151702bee 100644
--- a/include/net/sctp/structs.h
+++ b/include/net/sctp/structs.h
@@ -300,6 +300,7 @@ struct sctp_sock {
 
 	/* The default SACK delay timeout for new associations. */
 	__u32 sackdelay;
+	__u32 sackfreq;
 
 	/* Flags controlling Heartbeat, SACK delay, and Path MTU Discovery. */
 	__u32 param_flags;
@@ -938,6 +939,7 @@ struct sctp_transport {
 
 	/* SACK delay timeout */
 	unsigned long sackdelay;
+	__u32 sackfreq;
 
 	/* When was the last time (in jiffies) that we heard from this
 	 * transport?  We use this to pick new active and retran paths.
@@ -1542,6 +1544,7 @@ struct sctp_association {
 		 *             : SACK's are not delayed (see Section 6).
 		 */
 		__u8    sack_needed;     /* Do we need to sack the peer? */
+		__u32	sack_cnt;
 
 		/* These are capabilities which our peer advertised.  */
 		__u8	ecn_capable;	 /* Can peer do ECN? */
@@ -1651,6 +1654,7 @@ struct sctp_association {
 
 	/* SACK delay timeout */
 	unsigned long sackdelay;
+	__u32 sackfreq;
 
 
 	unsigned long timeouts[SCTP_NUM_TIMEOUT_TYPES];
diff --git a/include/net/sctp/user.h b/include/net/sctp/user.h
index 9619b9d35c9e..31371d2553e8 100644
--- a/include/net/sctp/user.h
+++ b/include/net/sctp/user.h
@@ -93,8 +93,9 @@ enum sctp_optname {
 #define SCTP_STATUS SCTP_STATUS
 	SCTP_GET_PEER_ADDR_INFO,
 #define SCTP_GET_PEER_ADDR_INFO SCTP_GET_PEER_ADDR_INFO
-	SCTP_DELAYED_ACK_TIME,
-#define SCTP_DELAYED_ACK_TIME SCTP_DELAYED_ACK_TIME
+	SCTP_DELAYED_ACK,
+#define SCTP_DELAYED_ACK_TIME SCTP_DELAYED_ACK
+#define SCTP_DELAYED_ACK SCTP_DELAYED_ACK
 	SCTP_CONTEXT,	/* Receive Context */
 #define SCTP_CONTEXT SCTP_CONTEXT
 	SCTP_FRAGMENT_INTERLEAVE,
@@ -618,13 +619,26 @@ struct sctp_authkeyid {
 };
 
 
-/* 7.1.23. Delayed Ack Timer (SCTP_DELAYED_ACK_TIME)
+/*
+ * 7.1.23.  Get or set delayed ack timer (SCTP_DELAYED_SACK)
  *
- *   This options will get or set the delayed ack timer.  The time is set
- *   in milliseconds.  If the assoc_id is 0, then this sets or gets the
- *   endpoints default delayed ack timer value.  If the assoc_id field is
- *   non-zero, then the set or get effects the specified association.
+ * This option will effect the way delayed acks are performed.  This
+ * option allows you to get or set the delayed ack time, in
+ * milliseconds.  It also allows changing the delayed ack frequency.
+ * Changing the frequency to 1 disables the delayed sack algorithm.  If
+ * the assoc_id is 0, then this sets or gets the endpoints default
+ * values.  If the assoc_id field is non-zero, then the set or get
+ * effects the specified association for the one to many model (the
+ * assoc_id field is ignored by the one to one model).  Note that if
+ * sack_delay or sack_freq are 0 when setting this option, then the
+ * current values will remain unchanged.
  */
+struct sctp_sack_info {
+	sctp_assoc_t	sack_assoc_id;
+	uint32_t	sack_delay;
+	uint32_t	sack_freq;
+};
+
 struct sctp_assoc_value {
     sctp_assoc_t            assoc_id;
     uint32_t                assoc_value;
diff --git a/net/sctp/associola.c b/net/sctp/associola.c
index b4cd2b71953f..7b79d1e781a8 100644
--- a/net/sctp/associola.c
+++ b/net/sctp/associola.c
@@ -136,6 +136,7 @@ static struct sctp_association *sctp_association_init(struct sctp_association *a
 
 	/* Set association default SACK delay */
 	asoc->sackdelay = msecs_to_jiffies(sp->sackdelay);
+	asoc->sackfreq = sp->sackfreq;
 
 	/* Set the association default flags controlling
 	 * Heartbeat, SACK delay, and Path MTU Discovery.
@@ -261,6 +262,7 @@ static struct sctp_association *sctp_association_init(struct sctp_association *a
 	 * already received one packet.]
 	 */
 	asoc->peer.sack_needed = 1;
+	asoc->peer.sack_cnt = 0;
 
 	/* Assume that the peer will tell us if he recognizes ASCONF
 	 * as part of INIT exchange.
@@ -615,6 +617,7 @@ struct sctp_transport *sctp_assoc_add_peer(struct sctp_association *asoc,
 	 * association configured value.
 	 */
 	peer->sackdelay = asoc->sackdelay;
+	peer->sackfreq = asoc->sackfreq;
 
 	/* Enable/disable heartbeat, SACK delay, and path MTU discovery
 	 * based on association setting.
diff --git a/net/sctp/sm_sideeffect.c b/net/sctp/sm_sideeffect.c
index 23a9f1a95b7d..b083312c725a 100644
--- a/net/sctp/sm_sideeffect.c
+++ b/net/sctp/sm_sideeffect.c
@@ -190,20 +190,28 @@ static int sctp_gen_sack(struct sctp_association *asoc, int force,
 	 * unacknowledged DATA chunk. ...
 	 */
 	if (!asoc->peer.sack_needed) {
-		/* We will need a SACK for the next packet.  */
-		asoc->peer.sack_needed = 1;
+		asoc->peer.sack_cnt++;
 
 		/* Set the SACK delay timeout based on the
 		 * SACK delay for the last transport
 		 * data was received from, or the default
 		 * for the association.
 		 */
-		if (trans)
+		if (trans) {
+			/* We will need a SACK for the next packet.  */
+			if (asoc->peer.sack_cnt >= trans->sackfreq - 1)
+				asoc->peer.sack_needed = 1;
+
 			asoc->timeouts[SCTP_EVENT_TIMEOUT_SACK] =
 				trans->sackdelay;
-		else
+		} else {
+			/* We will need a SACK for the next packet.  */
+			if (asoc->peer.sack_cnt >= asoc->sackfreq - 1)
+				asoc->peer.sack_needed = 1;
+
 			asoc->timeouts[SCTP_EVENT_TIMEOUT_SACK] =
 				asoc->sackdelay;
+		}
 
 		/* Restart the SACK timer. */
 		sctp_add_cmd_sf(commands, SCTP_CMD_TIMER_RESTART,
@@ -216,6 +224,7 @@ static int sctp_gen_sack(struct sctp_association *asoc, int force,
 			goto nomem;
 
 		asoc->peer.sack_needed = 0;
+		asoc->peer.sack_cnt = 0;
 
 		sctp_add_cmd_sf(commands, SCTP_CMD_REPLY, SCTP_CHUNK(sack));
 
diff --git a/net/sctp/socket.c b/net/sctp/socket.c
index e7e3baf7009e..66985871401b 100644
--- a/net/sctp/socket.c
+++ b/net/sctp/socket.c
@@ -2305,74 +2305,98 @@ static int sctp_setsockopt_peer_addr_params(struct sock *sk,
 	return 0;
 }
 
-/* 7.1.23. Delayed Ack Timer (SCTP_DELAYED_ACK_TIME)
- *
- *   This options will get or set the delayed ack timer.  The time is set
- *   in milliseconds.  If the assoc_id is 0, then this sets or gets the
- *   endpoints default delayed ack timer value.  If the assoc_id field is
- *   non-zero, then the set or get effects the specified association.
- *
- *   struct sctp_assoc_value {
- *       sctp_assoc_t            assoc_id;
- *       uint32_t                assoc_value;
- *   };
- *
- *     assoc_id    - This parameter, indicates which association the
- *                   user is preforming an action upon. Note that if
- *                   this field's value is zero then the endpoints
- *                   default value is changed (effecting future
- *                   associations only).
- *
- *     assoc_value - This parameter contains the number of milliseconds
- *                   that the user is requesting the delayed ACK timer
- *                   be set to. Note that this value is defined in
- *                   the standard to be between 200 and 500 milliseconds.
- *
- *                   Note: a value of zero will leave the value alone,
- *                   but disable SACK delay. A non-zero value will also
- *                   enable SACK delay.
+/*
+ * 7.1.23.  Get or set delayed ack timer (SCTP_DELAYED_SACK)
+ *
+ * This option will effect the way delayed acks are performed.  This
+ * option allows you to get or set the delayed ack time, in
+ * milliseconds.  It also allows changing the delayed ack frequency.
+ * Changing the frequency to 1 disables the delayed sack algorithm.  If
+ * the assoc_id is 0, then this sets or gets the endpoints default
+ * values.  If the assoc_id field is non-zero, then the set or get
+ * effects the specified association for the one to many model (the
+ * assoc_id field is ignored by the one to one model).  Note that if
+ * sack_delay or sack_freq are 0 when setting this option, then the
+ * current values will remain unchanged.
+ *
+ * struct sctp_sack_info {
+ *     sctp_assoc_t            sack_assoc_id;
+ *     uint32_t                sack_delay;
+ *     uint32_t                sack_freq;
+ * };
+ *
+ * sack_assoc_id -  This parameter, indicates which association the user
+ *    is performing an action upon.  Note that if this field's value is
+ *    zero then the endpoints default value is changed (effecting future
+ *    associations only).
+ *
+ * sack_delay -  This parameter contains the number of milliseconds that
+ *    the user is requesting the delayed ACK timer be set to.  Note that
+ *    this value is defined in the standard to be between 200 and 500
+ *    milliseconds.
+ *
+ * sack_freq -  This parameter contains the number of packets that must
+ *    be received before a sack is sent without waiting for the delay
+ *    timer to expire.  The default value for this is 2, setting this
+ *    value to 1 will disable the delayed sack algorithm.
  */
 
-static int sctp_setsockopt_delayed_ack_time(struct sock *sk,
+static int sctp_setsockopt_delayed_ack(struct sock *sk,
 					    char __user *optval, int optlen)
 {
-	struct sctp_assoc_value  params;
+	struct sctp_sack_info    params;
 	struct sctp_transport   *trans = NULL;
 	struct sctp_association *asoc = NULL;
 	struct sctp_sock        *sp = sctp_sk(sk);
 
-	if (optlen != sizeof(struct sctp_assoc_value))
-		return - EINVAL;
+	if (optlen == sizeof(struct sctp_sack_info)) {
+		if (copy_from_user(&params, optval, optlen))
+			return -EFAULT;
 
-	if (copy_from_user(&params, optval, optlen))
-		return -EFAULT;
+		if (params.sack_delay == 0 && params.sack_freq == 0)
+			return 0;
+	} else if (optlen == sizeof(struct sctp_assoc_value)) {
+		printk(KERN_WARNING "SCTP: Use of struct sctp_sack_info "
+		       "in delayed_ack socket option deprecated\n");
+		printk(KERN_WARNING "SCTP: struct sctp_sack_info instead\n");
+		if (copy_from_user(&params, optval, optlen))
+			return -EFAULT;
+
+		if (params.sack_delay == 0)
+			params.sack_freq = 1;
+		else
+			params.sack_freq = 0;
+	} else
+		return - EINVAL;
 
 	/* Validate value parameter. */
-	if (params.assoc_value > 500)
+	if (params.sack_delay > 500)
 		return -EINVAL;
 
-	/* Get association, if assoc_id != 0 and the socket is a one
+	/* Get association, if sack_assoc_id != 0 and the socket is a one
 	 * to many style socket, and an association was not found, then
 	 * the id was invalid.
 	 */
-	asoc = sctp_id2assoc(sk, params.assoc_id);
-	if (!asoc && params.assoc_id && sctp_style(sk, UDP))
+	asoc = sctp_id2assoc(sk, params.sack_assoc_id);
+	if (!asoc && params.sack_assoc_id && sctp_style(sk, UDP))
 		return -EINVAL;
 
-	if (params.assoc_value) {
+	if (params.sack_delay) {
 		if (asoc) {
 			asoc->sackdelay =
-				msecs_to_jiffies(params.assoc_value);
+				msecs_to_jiffies(params.sack_delay);
 			asoc->param_flags =
 				(asoc->param_flags & ~SPP_SACKDELAY) |
 				SPP_SACKDELAY_ENABLE;
 		} else {
-			sp->sackdelay = params.assoc_value;
+			sp->sackdelay = params.sack_delay;
 			sp->param_flags =
 				(sp->param_flags & ~SPP_SACKDELAY) |
 				SPP_SACKDELAY_ENABLE;
 		}
-	} else {
+	}
+
+	if (params.sack_freq == 1) {
 		if (asoc) {
 			asoc->param_flags =
 				(asoc->param_flags & ~SPP_SACKDELAY) |
@@ -2382,22 +2406,40 @@ static int sctp_setsockopt_delayed_ack_time(struct sock *sk,
 				(sp->param_flags & ~SPP_SACKDELAY) |
 				SPP_SACKDELAY_DISABLE;
 		}
+	} else if (params.sack_freq > 1) {
+		if (asoc) {
+			asoc->sackfreq = params.sack_freq;
+			asoc->param_flags =
+				(asoc->param_flags & ~SPP_SACKDELAY) |
+				SPP_SACKDELAY_ENABLE;
+		} else {
+			sp->sackfreq = params.sack_freq;
+			sp->param_flags =
+				(sp->param_flags & ~SPP_SACKDELAY) |
+				SPP_SACKDELAY_ENABLE;
+		}
 	}
 
 	/* If change is for association, also apply to each transport. */
 	if (asoc) {
 		list_for_each_entry(trans, &asoc->peer.transport_addr_list,
 				transports) {
-			if (params.assoc_value) {
+			if (params.sack_delay) {
 				trans->sackdelay =
-					msecs_to_jiffies(params.assoc_value);
+					msecs_to_jiffies(params.sack_delay);
 				trans->param_flags =
 					(trans->param_flags & ~SPP_SACKDELAY) |
 					SPP_SACKDELAY_ENABLE;
-			} else {
+			}
+			if (params.sack_delay == 1) {
 				trans->param_flags =
 					(trans->param_flags & ~SPP_SACKDELAY) |
 					SPP_SACKDELAY_DISABLE;
+			} else if (params.sack_freq > 1) {
+				trans->sackfreq = params.sack_freq;
+				trans->param_flags =
+					(trans->param_flags & ~SPP_SACKDELAY) |
+					SPP_SACKDELAY_ENABLE;
 			}
 		}
 	}
@@ -3186,8 +3228,8 @@ SCTP_STATIC int sctp_setsockopt(struct sock *sk, int level, int optname,
 		retval = sctp_setsockopt_peer_addr_params(sk, optval, optlen);
 		break;
 
-	case SCTP_DELAYED_ACK_TIME:
-		retval = sctp_setsockopt_delayed_ack_time(sk, optval, optlen);
+	case SCTP_DELAYED_ACK:
+		retval = sctp_setsockopt_delayed_ack(sk, optval, optlen);
 		break;
 	case SCTP_PARTIAL_DELIVERY_POINT:
 		retval = sctp_setsockopt_partial_delivery_point(sk, optval, optlen);
@@ -3446,6 +3488,7 @@ SCTP_STATIC int sctp_init_sock(struct sock *sk)
 	sp->pathmaxrxt  = sctp_max_retrans_path;
 	sp->pathmtu     = 0; // allow default discovery
 	sp->sackdelay   = sctp_sack_timeout;
+	sp->sackfreq	= 3;
 	sp->param_flags = SPP_HB_ENABLE |
 			  SPP_PMTUD_ENABLE |
 			  SPP_SACKDELAY_ENABLE;
@@ -3999,70 +4042,91 @@ static int sctp_getsockopt_peer_addr_params(struct sock *sk, int len,
 	return 0;
 }
 
-/* 7.1.23. Delayed Ack Timer (SCTP_DELAYED_ACK_TIME)
- *
- *   This options will get or set the delayed ack timer.  The time is set
- *   in milliseconds.  If the assoc_id is 0, then this sets or gets the
- *   endpoints default delayed ack timer value.  If the assoc_id field is
- *   non-zero, then the set or get effects the specified association.
- *
- *   struct sctp_assoc_value {
- *       sctp_assoc_t            assoc_id;
- *       uint32_t                assoc_value;
- *   };
+/*
+ * 7.1.23.  Get or set delayed ack timer (SCTP_DELAYED_SACK)
+ *
+ * This option will effect the way delayed acks are performed.  This
+ * option allows you to get or set the delayed ack time, in
+ * milliseconds.  It also allows changing the delayed ack frequency.
+ * Changing the frequency to 1 disables the delayed sack algorithm.  If
+ * the assoc_id is 0, then this sets or gets the endpoints default
+ * values.  If the assoc_id field is non-zero, then the set or get
+ * effects the specified association for the one to many model (the
+ * assoc_id field is ignored by the one to one model).  Note that if
+ * sack_delay or sack_freq are 0 when setting this option, then the
+ * current values will remain unchanged.
+ *
+ * struct sctp_sack_info {
+ *     sctp_assoc_t            sack_assoc_id;
+ *     uint32_t                sack_delay;
+ *     uint32_t                sack_freq;
+ * };
  *
- *     assoc_id    - This parameter, indicates which association the
- *                   user is preforming an action upon. Note that if
- *                   this field's value is zero then the endpoints
- *                   default value is changed (effecting future
- *                   associations only).
+ * sack_assoc_id -  This parameter, indicates which association the user
+ *    is performing an action upon.  Note that if this field's value is
+ *    zero then the endpoints default value is changed (effecting future
+ *    associations only).
  *
- *     assoc_value - This parameter contains the number of milliseconds
- *                   that the user is requesting the delayed ACK timer
- *                   be set to. Note that this value is defined in
- *                   the standard to be between 200 and 500 milliseconds.
+ * sack_delay -  This parameter contains the number of milliseconds that
+ *    the user is requesting the delayed ACK timer be set to.  Note that
+ *    this value is defined in the standard to be between 200 and 500
+ *    milliseconds.
  *
- *                   Note: a value of zero will leave the value alone,
- *                   but disable SACK delay. A non-zero value will also
- *                   enable SACK delay.
+ * sack_freq -  This parameter contains the number of packets that must
+ *    be received before a sack is sent without waiting for the delay
+ *    timer to expire.  The default value for this is 2, setting this
+ *    value to 1 will disable the delayed sack algorithm.
  */
-static int sctp_getsockopt_delayed_ack_time(struct sock *sk, int len,
+static int sctp_getsockopt_delayed_ack(struct sock *sk, int len,
 					    char __user *optval,
 					    int __user *optlen)
 {
-	struct sctp_assoc_value  params;
+	struct sctp_sack_info    params;
 	struct sctp_association *asoc = NULL;
 	struct sctp_sock        *sp = sctp_sk(sk);
 
-	if (len < sizeof(struct sctp_assoc_value))
-		return - EINVAL;
+	if (len >= sizeof(struct sctp_sack_info)) {
+		len = sizeof(struct sctp_sack_info);
 
-	len = sizeof(struct sctp_assoc_value);
-
-	if (copy_from_user(&params, optval, len))
-		return -EFAULT;
+		if (copy_from_user(&params, optval, len))
+			return -EFAULT;
+	} else if (len == sizeof(struct sctp_assoc_value)) {
+		printk(KERN_WARNING "SCTP: Use of struct sctp_sack_info "
+		       "in delayed_ack socket option deprecated\n");
+		printk(KERN_WARNING "SCTP: struct sctp_sack_info instead\n");
+		if (copy_from_user(&params, optval, len))
+			return -EFAULT;
+	} else
+		return - EINVAL;
 
-	/* Get association, if assoc_id != 0 and the socket is a one
+	/* Get association, if sack_assoc_id != 0 and the socket is a one
 	 * to many style socket, and an association was not found, then
 	 * the id was invalid.
 	 */
-	asoc = sctp_id2assoc(sk, params.assoc_id);
-	if (!asoc && params.assoc_id && sctp_style(sk, UDP))
+	asoc = sctp_id2assoc(sk, params.sack_assoc_id);
+	if (!asoc && params.sack_assoc_id && sctp_style(sk, UDP))
 		return -EINVAL;
 
 	if (asoc) {
 		/* Fetch association values. */
-		if (asoc->param_flags & SPP_SACKDELAY_ENABLE)
-			params.assoc_value = jiffies_to_msecs(
+		if (asoc->param_flags & SPP_SACKDELAY_ENABLE) {
+			params.sack_delay = jiffies_to_msecs(
 				asoc->sackdelay);
-		else
-			params.assoc_value = 0;
+			params.sack_freq = asoc->sackfreq;
+
+		} else {
+			params.sack_delay = 0;
+			params.sack_freq = 1;
+		}
 	} else {
 		/* Fetch socket values. */
-		if (sp->param_flags & SPP_SACKDELAY_ENABLE)
-			params.assoc_value  = sp->sackdelay;
-		else
-			params.assoc_value  = 0;
+		if (sp->param_flags & SPP_SACKDELAY_ENABLE) {
+			params.sack_delay  = sp->sackdelay;
+			params.sack_freq = sp->sackfreq;
+		} else {
+			params.sack_delay  = 0;
+			params.sack_freq = 1;
+		}
 	}
 
 	if (copy_to_user(optval, &params, len))
@@ -5218,8 +5282,8 @@ SCTP_STATIC int sctp_getsockopt(struct sock *sk, int level, int optname,
 		retval = sctp_getsockopt_peer_addr_params(sk, len, optval,
 							  optlen);
 		break;
-	case SCTP_DELAYED_ACK_TIME:
-		retval = sctp_getsockopt_delayed_ack_time(sk, len, optval,
+	case SCTP_DELAYED_ACK:
+		retval = sctp_getsockopt_delayed_ack(sk, len, optval,
 							  optlen);
 		break;
 	case SCTP_INITMSG:
-- 
cgit v1.2.3


From 88a0a948e752bb9b617a8c55417a9fd9b0257199 Mon Sep 17 00:00:00 2001
From: Vlad Yasevich <vladislav.yasevich@hp.com>
Date: Fri, 9 May 2008 15:14:11 -0700
Subject: sctp: Support the new specification of sctp_connectx()

The specification of sctp_connectx() has been changed to return
an association id.  We've added a new socket option that will
return the association id as the return value from the setsockopt()
call.  The library that implements sctp_connectx() interface will
implement both socket options.

Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/sctp/user.h |  6 +++--
 net/sctp/socket.c       | 70 +++++++++++++++++++++++++++++++++++++++++--------
 2 files changed, 63 insertions(+), 13 deletions(-)

(limited to 'include')

diff --git a/include/net/sctp/user.h b/include/net/sctp/user.h
index 31371d2553e8..f205b10f0ab9 100644
--- a/include/net/sctp/user.h
+++ b/include/net/sctp/user.h
@@ -137,12 +137,14 @@ enum sctp_optname {
 #define SCTP_GET_LOCAL_ADDRS_NUM_OLD	SCTP_GET_LOCAL_ADDRS_NUM_OLD
 	SCTP_GET_LOCAL_ADDRS_OLD, 	/* Get all local addresss. */
 #define SCTP_GET_LOCAL_ADDRS_OLD	SCTP_GET_LOCAL_ADDRS_OLD
-	SCTP_SOCKOPT_CONNECTX, /* CONNECTX requests. */
-#define SCTP_SOCKOPT_CONNECTX	SCTP_SOCKOPT_CONNECTX
+	SCTP_SOCKOPT_CONNECTX_OLD, /* CONNECTX old requests. */
+#define SCTP_SOCKOPT_CONNECTX_OLD	SCTP_SOCKOPT_CONNECTX_OLD
 	SCTP_GET_PEER_ADDRS, 	/* Get all peer addresss. */
 #define SCTP_GET_PEER_ADDRS	SCTP_GET_PEER_ADDRS
 	SCTP_GET_LOCAL_ADDRS, 	/* Get all local addresss. */
 #define SCTP_GET_LOCAL_ADDRS	SCTP_GET_LOCAL_ADDRS
+	SCTP_SOCKOPT_CONNECTX, /* CONNECTX requests. */
+#define SCTP_SOCKOPT_CONNECTX	SCTP_SOCKOPT_CONNECTX
 };
 
 /*
diff --git a/net/sctp/socket.c b/net/sctp/socket.c
index 66985871401b..81600eea05d2 100644
--- a/net/sctp/socket.c
+++ b/net/sctp/socket.c
@@ -956,7 +956,8 @@ out:
  */
 static int __sctp_connect(struct sock* sk,
 			  struct sockaddr *kaddrs,
-			  int addrs_size)
+			  int addrs_size,
+			  sctp_assoc_t *assoc_id)
 {
 	struct sctp_sock *sp;
 	struct sctp_endpoint *ep;
@@ -1111,6 +1112,8 @@ static int __sctp_connect(struct sock* sk,
 	timeo = sock_sndtimeo(sk, f_flags & O_NONBLOCK);
 
 	err = sctp_wait_for_connect(asoc, &timeo);
+	if (!err && assoc_id)
+		*assoc_id = asoc->assoc_id;
 
 	/* Don't free association on exit. */
 	asoc = NULL;
@@ -1128,7 +1131,8 @@ out_free:
 /* Helper for tunneling sctp_connectx() requests through sctp_setsockopt()
  *
  * API 8.9
- * int sctp_connectx(int sd, struct sockaddr *addrs, int addrcnt);
+ * int sctp_connectx(int sd, struct sockaddr *addrs, int addrcnt,
+ * 			sctp_assoc_t *asoc);
  *
  * If sd is an IPv4 socket, the addresses passed must be IPv4 addresses.
  * If the sd is an IPv6 socket, the addresses passed can either be IPv4
@@ -1144,8 +1148,10 @@ out_free:
  * representation is termed a "packed array" of addresses). The caller
  * specifies the number of addresses in the array with addrcnt.
  *
- * On success, sctp_connectx() returns 0. On failure, sctp_connectx() returns
- * -1, and sets errno to the appropriate error code.
+ * On success, sctp_connectx() returns 0. It also sets the assoc_id to
+ * the association id of the new association.  On failure, sctp_connectx()
+ * returns -1, and sets errno to the appropriate error code.  The assoc_id
+ * is not touched by the kernel.
  *
  * For SCTP, the port given in each socket address must be the same, or
  * sctp_connectx() will fail, setting errno to EINVAL.
@@ -1182,11 +1188,12 @@ out_free:
  * addrs     The pointer to the addresses in user land
  * addrssize Size of the addrs buffer
  *
- * Returns 0 if ok, <0 errno code on error.
+ * Returns >=0 if ok, <0 errno code on error.
  */
-SCTP_STATIC int sctp_setsockopt_connectx(struct sock* sk,
+SCTP_STATIC int __sctp_setsockopt_connectx(struct sock* sk,
 				      struct sockaddr __user *addrs,
-				      int addrs_size)
+				      int addrs_size,
+				      sctp_assoc_t *assoc_id)
 {
 	int err = 0;
 	struct sockaddr *kaddrs;
@@ -1209,13 +1216,46 @@ SCTP_STATIC int sctp_setsockopt_connectx(struct sock* sk,
 	if (__copy_from_user(kaddrs, addrs, addrs_size)) {
 		err = -EFAULT;
 	} else {
-		err = __sctp_connect(sk, kaddrs, addrs_size);
+		err = __sctp_connect(sk, kaddrs, addrs_size, assoc_id);
 	}
 
 	kfree(kaddrs);
+
 	return err;
 }
 
+/*
+ * This is an older interface.  It's kept for backward compatibility
+ * to the option that doesn't provide association id.
+ */
+SCTP_STATIC int sctp_setsockopt_connectx_old(struct sock* sk,
+				      struct sockaddr __user *addrs,
+				      int addrs_size)
+{
+	return __sctp_setsockopt_connectx(sk, addrs, addrs_size, NULL);
+}
+
+/*
+ * New interface for the API.  The since the API is done with a socket
+ * option, to make it simple we feed back the association id is as a return
+ * indication to the call.  Error is always negative and association id is
+ * always positive.
+ */
+SCTP_STATIC int sctp_setsockopt_connectx(struct sock* sk,
+				      struct sockaddr __user *addrs,
+				      int addrs_size)
+{
+	sctp_assoc_t assoc_id = 0;
+	int err = 0;
+
+	err = __sctp_setsockopt_connectx(sk, addrs, addrs_size, &assoc_id);
+
+	if (err)
+		return err;
+	else
+		return assoc_id;
+}
+
 /* API 3.1.4 close() - UDP Style Syntax
  * Applications use close() to perform graceful shutdown (as described in
  * Section 10.1 of [SCTP]) on ALL the associations currently represented
@@ -3206,10 +3246,18 @@ SCTP_STATIC int sctp_setsockopt(struct sock *sk, int level, int optname,
 					       optlen, SCTP_BINDX_REM_ADDR);
 		break;
 
+	case SCTP_SOCKOPT_CONNECTX_OLD:
+		/* 'optlen' is the size of the addresses buffer. */
+		retval = sctp_setsockopt_connectx_old(sk,
+					    (struct sockaddr __user *)optval,
+					    optlen);
+		break;
+
 	case SCTP_SOCKOPT_CONNECTX:
 		/* 'optlen' is the size of the addresses buffer. */
-		retval = sctp_setsockopt_connectx(sk, (struct sockaddr __user *)optval,
-					       optlen);
+		retval = sctp_setsockopt_connectx(sk,
+					    (struct sockaddr __user *)optval,
+					    optlen);
 		break;
 
 	case SCTP_DISABLE_FRAGMENTS:
@@ -3336,7 +3384,7 @@ SCTP_STATIC int sctp_connect(struct sock *sk, struct sockaddr *addr,
 		/* Pass correct addr len to common routine (so it knows there
 		 * is only one address being passed.
 		 */
-		err = __sctp_connect(sk, addr, af->sockaddr_len);
+		err = __sctp_connect(sk, addr, af->sockaddr_len, NULL);
 	}
 
 	sctp_release_sock(sk);
-- 
cgit v1.2.3


From 20c2c1fd6c842caf70dcb1d94b9d58861949fd3d Mon Sep 17 00:00:00 2001
From: Neil Horman <nhorman@tuxdriver.com>
Date: Fri, 9 May 2008 15:14:50 -0700
Subject: sctp: add sctp/remaddr table to complete RFC remote address table OID

Add support for RFC3873 remote address table OID.

      +--(5) sctpAssocRemAddrTable
      |   |
      |   |--(-) sctpAssocId (shared index)
      |   |
      |   +--(1) sctpAssocRemAddrType (index)
      .   |
      .   +--(2) sctpAssocRemAddr (index)
      .   |
          +--(3) sctpAssocRemAddrActive
          |
          +--(4) sctpAssocRemAddrHBActive
          |
          +--(5) sctpAssocRemAddrRTO
          |
          +--(6) sctpAssocRemAddrMaxPathRtx
          |
          +--(7) sctpAssocRemAddrRtx
          |
          +--(8) sctpAssocRemAddrStartTime

This patch places all the requsite data in /proc/net/sctp/remaddr.

Signed-off-by: Neil Horman <nhorman@tuxdriver.com>
Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/sctp/sctp.h |   2 +
 net/sctp/proc.c         | 141 ++++++++++++++++++++++++++++++++++++++++++++++++
 net/sctp/protocol.c     |   3 ++
 3 files changed, 146 insertions(+)

(limited to 'include')

diff --git a/include/net/sctp/sctp.h b/include/net/sctp/sctp.h
index 90b1e8d23b16..5672d489e924 100644
--- a/include/net/sctp/sctp.h
+++ b/include/net/sctp/sctp.h
@@ -179,6 +179,8 @@ int sctp_eps_proc_init(void);
 void sctp_eps_proc_exit(void);
 int sctp_assocs_proc_init(void);
 void sctp_assocs_proc_exit(void);
+int sctp_remaddr_proc_init(void);
+void sctp_remaddr_proc_exit(void);
 
 
 /*
diff --git a/net/sctp/proc.c b/net/sctp/proc.c
index 0aba759cb9b7..5dd89831eceb 100644
--- a/net/sctp/proc.c
+++ b/net/sctp/proc.c
@@ -383,3 +383,144 @@ void sctp_assocs_proc_exit(void)
 {
 	remove_proc_entry("assocs", proc_net_sctp);
 }
+
+static void *sctp_remaddr_seq_start(struct seq_file *seq, loff_t *pos)
+{
+	if (*pos >= sctp_assoc_hashsize)
+		return NULL;
+
+	if (*pos < 0)
+		*pos = 0;
+
+	if (*pos == 0)
+		seq_printf(seq, "ADDR ASSOC_ID HB_ACT RTO MAX_PATH_RTX "
+				"REM_ADDR_RTX  START\n");
+
+	return (void *)pos;
+}
+
+static void *sctp_remaddr_seq_next(struct seq_file *seq, void *v, loff_t *pos)
+{
+	if (++*pos >= sctp_assoc_hashsize)
+		return NULL;
+
+	return pos;
+}
+
+static void sctp_remaddr_seq_stop(struct seq_file *seq, void *v)
+{
+	return;
+}
+
+static int sctp_remaddr_seq_show(struct seq_file *seq, void *v)
+{
+	struct sctp_hashbucket *head;
+	struct sctp_ep_common *epb;
+	struct sctp_association *assoc;
+	struct hlist_node *node;
+	struct sctp_transport *tsp;
+	int    hash = *(loff_t *)v;
+
+	if (hash >= sctp_assoc_hashsize)
+		return -ENOMEM;
+
+	head = &sctp_assoc_hashtable[hash];
+	sctp_local_bh_disable();
+	read_lock(&head->lock);
+	sctp_for_each_hentry(epb, node, &head->chain) {
+		assoc = sctp_assoc(epb);
+		list_for_each_entry(tsp, &assoc->peer.transport_addr_list,
+					transports) {
+			/*
+			 * The remote address (ADDR)
+			 */
+			tsp->af_specific->seq_dump_addr(seq, &tsp->ipaddr);
+			seq_printf(seq, " ");
+
+			/*
+			 * The association ID (ASSOC_ID)
+			 */
+			seq_printf(seq, "%d ", tsp->asoc->assoc_id);
+
+			/*
+			 * If the Heartbeat is active (HB_ACT)
+			 * Note: 1 = Active, 0 = Inactive
+			 */
+			seq_printf(seq, "%d ", timer_pending(&tsp->hb_timer));
+
+			/*
+			 * Retransmit time out (RTO)
+			 */
+			seq_printf(seq, "%lu ", tsp->rto);
+
+			/*
+			 * Maximum path retransmit count (PATH_MAX_RTX)
+			 */
+			seq_printf(seq, "%d ", tsp->pathmaxrxt);
+
+			/*
+			 * remote address retransmit count (REM_ADDR_RTX)
+			 * Note: We don't have a way to tally this at the moment
+			 * so lets just leave it as zero for the moment
+			 */
+			seq_printf(seq, "0 ");
+
+			/*
+			 * remote address start time (START).  This is also not
+			 * currently implemented, but we can record it with a
+			 * jiffies marker in a subsequent patch
+			 */
+			seq_printf(seq, "0");
+
+			seq_printf(seq, "\n");
+		}
+	}
+
+	read_unlock(&head->lock);
+	sctp_local_bh_enable();
+
+	return 0;
+
+}
+
+static const struct seq_operations sctp_remaddr_ops = {
+	.start = sctp_remaddr_seq_start,
+	.next  = sctp_remaddr_seq_next,
+	.stop  = sctp_remaddr_seq_stop,
+	.show  = sctp_remaddr_seq_show,
+};
+
+/* Cleanup the proc fs entry for 'remaddr' object. */
+void sctp_remaddr_proc_exit(void)
+{
+	remove_proc_entry("remaddr", proc_net_sctp);
+}
+
+static int sctp_remaddr_seq_open(struct inode *inode, struct file *file)
+{
+	return seq_open(file, &sctp_remaddr_ops);
+}
+
+static const struct file_operations sctp_remaddr_seq_fops = {
+	.open = sctp_remaddr_seq_open,
+	.read = seq_read,
+	.llseek = seq_lseek,
+	.release = seq_release,
+};
+
+int __init sctp_remaddr_proc_init(void)
+{
+	struct proc_dir_entry *p;
+
+	p = create_proc_entry("remaddr", S_IRUGO, proc_net_sctp);
+	if (!p)
+		return -ENOMEM;
+	p->proc_fops = &sctp_remaddr_seq_fops;
+
+	return 0;
+}
+
+void sctp_assoc_proc_exit(void)
+{
+	remove_proc_entry("remaddr", proc_net_sctp);
+}
diff --git a/net/sctp/protocol.c b/net/sctp/protocol.c
index 0ec234b762c2..b8bd9e014498 100644
--- a/net/sctp/protocol.c
+++ b/net/sctp/protocol.c
@@ -113,6 +113,8 @@ static __init int sctp_proc_init(void)
 		goto out_nomem;
 	if (sctp_assocs_proc_init())
 		goto out_nomem;
+	if (sctp_remaddr_proc_init())
+		goto out_nomem;
 
 	return 0;
 
@@ -129,6 +131,7 @@ static void sctp_proc_exit(void)
 	sctp_snmp_proc_exit();
 	sctp_eps_proc_exit();
 	sctp_assocs_proc_exit();
+	sctp_remaddr_proc_exit();
 
 	if (proc_net_sctp) {
 		proc_net_sctp = NULL;
-- 
cgit v1.2.3


From 7ef43ebaa538e0cc9063cbf84593a05091bcace2 Mon Sep 17 00:00:00 2001
From: Allan Stephens <allan.stephens@windriver.com>
Date: Mon, 12 May 2008 15:42:28 -0700
Subject: tipc: Fix race condition when creating socket or native port

This patch eliminates the (very remote) chance of a crash resulting
from a partially initialized socket or native port unexpectedly
receiving a message.  Now, during the creation of a socket or native
port, the underlying generic port's lock is not released until all
initialization required to handle incoming messages has been done.

Signed-off-by: Allan Stephens <allan.stephens@windriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/tipc/tipc_port.h |  3 ++-
 net/tipc/port.c              | 20 ++++++++++++--------
 net/tipc/ref.c               | 12 ++++++++----
 net/tipc/socket.c            |  5 ++++-
 net/tipc/subscr.c            |  1 +
 5 files changed, 27 insertions(+), 14 deletions(-)

(limited to 'include')

diff --git a/include/net/tipc/tipc_port.h b/include/net/tipc/tipc_port.h
index 11105bcc4457..9923e41a8215 100644
--- a/include/net/tipc/tipc_port.h
+++ b/include/net/tipc/tipc_port.h
@@ -84,7 +84,8 @@ struct tipc_port {
 u32 tipc_createport_raw(void *usr_handle,
 			u32 (*dispatcher)(struct tipc_port *, struct sk_buff *),
 			void (*wakeup)(struct tipc_port *),
-			const u32 importance);
+			const u32 importance,
+			struct tipc_port **tp_ptr);
 
 int tipc_reject_msg(struct sk_buff *buf, u32 err);
 
diff --git a/net/tipc/port.c b/net/tipc/port.c
index 2f5806410c64..757de38fe6af 100644
--- a/net/tipc/port.c
+++ b/net/tipc/port.c
@@ -211,15 +211,18 @@ exit:
 }
 
 /**
- * tipc_createport_raw - create a native TIPC port
+ * tipc_createport_raw - create a generic TIPC port
  *
- * Returns local port reference
+ * Returns port reference, or 0 if unable to create it
+ *
+ * Note: The newly created port is returned in the locked state.
  */
 
 u32 tipc_createport_raw(void *usr_handle,
 			u32 (*dispatcher)(struct tipc_port *, struct sk_buff *),
 			void (*wakeup)(struct tipc_port *),
-			const u32 importance)
+			const u32 importance,
+			struct tipc_port **tp_ptr)
 {
 	struct port *p_ptr;
 	struct tipc_msg *msg;
@@ -237,7 +240,6 @@ u32 tipc_createport_raw(void *usr_handle,
 		return 0;
 	}
 
-	tipc_port_lock(ref);
 	p_ptr->publ.usr_handle = usr_handle;
 	p_ptr->publ.max_pkt = MAX_PKT_DEFAULT;
 	p_ptr->publ.ref = ref;
@@ -262,7 +264,7 @@ u32 tipc_createport_raw(void *usr_handle,
 	INIT_LIST_HEAD(&p_ptr->port_list);
 	list_add_tail(&p_ptr->port_list, &ports);
 	spin_unlock_bh(&tipc_port_list_lock);
-	tipc_port_unlock(p_ptr);
+	*tp_ptr = &p_ptr->publ;
 	return ref;
 }
 
@@ -1053,6 +1055,7 @@ int tipc_createport(u32 user_ref,
 {
 	struct user_port *up_ptr;
 	struct port *p_ptr;
+	struct tipc_port *tp_ptr;
 	u32 ref;
 
 	up_ptr = kmalloc(sizeof(*up_ptr), GFP_ATOMIC);
@@ -1060,12 +1063,13 @@ int tipc_createport(u32 user_ref,
 		warn("Port creation failed, no memory\n");
 		return -ENOMEM;
 	}
-	ref = tipc_createport_raw(NULL, port_dispatcher, port_wakeup, importance);
-	p_ptr = tipc_port_lock(ref);
-	if (!p_ptr) {
+	ref = tipc_createport_raw(NULL, port_dispatcher, port_wakeup,
+				  importance, &tp_ptr);
+	if (ref == 0) {
 		kfree(up_ptr);
 		return -ENOMEM;
 	}
+	p_ptr = (struct port *)tp_ptr;
 
 	p_ptr->user_port = up_ptr;
 	up_ptr->user_ref = user_ref;
diff --git a/net/tipc/ref.c b/net/tipc/ref.c
index 89cbab24d08f..a101de86824d 100644
--- a/net/tipc/ref.c
+++ b/net/tipc/ref.c
@@ -142,9 +142,13 @@ void tipc_ref_table_stop(void)
 /**
  * tipc_ref_acquire - create reference to an object
  *
- * Return a unique reference value which can be translated back to the pointer
- * 'object' at a later time.  Also, pass back a pointer to the lock protecting
- * the object, but without locking it.
+ * Register an object pointer in reference table and lock the object.
+ * Returns a unique reference value that is used from then on to retrieve the
+ * object pointer, or to determine that the object has been deregistered.
+ *
+ * Note: The object is returned in the locked state so that the caller can
+ * register a partially initialized object, without running the risk that
+ * the object will be accessed before initialization is complete.
  */
 
 u32 tipc_ref_acquire(void *object, spinlock_t **lock)
@@ -178,13 +182,13 @@ u32 tipc_ref_acquire(void *object, spinlock_t **lock)
 		ref = (next_plus_upper & ~index_mask) + index;
 		entry->ref = ref;
 		entry->object = object;
-		spin_unlock_bh(&entry->lock);
 		*lock = &entry->lock;
 	}
 	else if (tipc_ref_table.init_point < tipc_ref_table.capacity) {
 		index = tipc_ref_table.init_point++;
 		entry = &(tipc_ref_table.entries[index]);
 		spin_lock_init(&entry->lock);
+		spin_lock_bh(&entry->lock);
 		ref = tipc_ref_table.start_mask + index;
 		entry->ref = ref;
 		entry->object = object;
diff --git a/net/tipc/socket.c b/net/tipc/socket.c
index 230f9ca2ad6b..38f48795b40e 100644
--- a/net/tipc/socket.c
+++ b/net/tipc/socket.c
@@ -188,6 +188,7 @@ static int tipc_create(struct net *net, struct socket *sock, int protocol)
 	const struct proto_ops *ops;
 	socket_state state;
 	struct sock *sk;
+	struct tipc_port *tp_ptr;
 	u32 portref;
 
 	/* Validate arguments */
@@ -225,7 +226,7 @@ static int tipc_create(struct net *net, struct socket *sock, int protocol)
 	/* Allocate TIPC port for socket to use */
 
 	portref = tipc_createport_raw(sk, &dispatch, &wakeupdispatch,
-				      TIPC_LOW_IMPORTANCE);
+				      TIPC_LOW_IMPORTANCE, &tp_ptr);
 	if (unlikely(portref == 0)) {
 		sk_free(sk);
 		return -ENOMEM;
@@ -241,6 +242,8 @@ static int tipc_create(struct net *net, struct socket *sock, int protocol)
 	sk->sk_backlog_rcv = backlog_rcv;
 	tipc_sk(sk)->p = tipc_get_port(portref);
 
+	spin_unlock_bh(tp_ptr->lock);
+
 	if (sock->state == SS_READY) {
 		tipc_set_portunreturnable(portref, 1);
 		if (sock->type == SOCK_DGRAM)
diff --git a/net/tipc/subscr.c b/net/tipc/subscr.c
index 8c01ccd3626c..8f8d0a6c1c16 100644
--- a/net/tipc/subscr.c
+++ b/net/tipc/subscr.c
@@ -474,6 +474,7 @@ static void subscr_named_msg_event(void *usr_handle,
 		kfree(subscriber);
 		return;
 	}
+	spin_unlock_bh(subscriber->lock);
 
 	/* Establish a connection to subscriber */
 
-- 
cgit v1.2.3


From 5a433b3ad497120d80f1045d37bd1a9ce897388f Mon Sep 17 00:00:00 2001
From: Harvey Harrison <harvey.harrison@gmail.com>
Date: Mon, 21 Apr 2008 10:41:10 -0700
Subject: mac80211: remove unnecessary byteshifts in frame control testing

Byteswap the constants rather than the frame_control member.

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/linux/ieee80211.h | 32 +++++++++++++++++---------------
 1 file changed, 17 insertions(+), 15 deletions(-)

(limited to 'include')

diff --git a/include/linux/ieee80211.h b/include/linux/ieee80211.h
index 0b5e03eae6d2..a9102bc78b61 100644
--- a/include/linux/ieee80211.h
+++ b/include/linux/ieee80211.h
@@ -552,16 +552,17 @@ enum ieee80211_back_parties {
  */
 static inline u8 *ieee80211_get_SA(struct ieee80211_hdr *hdr)
 {
-	u8 *raw = (u8 *) hdr;
-	u8 tofrom = (*(raw+1)) & 3; /* get the TODS and FROMDS bits */
-
-	switch (tofrom) {
-		case 2:
-			return hdr->addr3;
-		case 3:
-			return hdr->addr4;
+	__le16 fc = hdr->frame_control;
+	fc &= cpu_to_le16(IEEE80211_FCTL_TODS | IEEE80211_FCTL_FROMDS);
+
+	switch (fc) {
+	case __constant_cpu_to_le16(IEEE80211_FCTL_FROMDS):
+		return hdr->addr3;
+	case __constant_cpu_to_le16(IEEE80211_FCTL_TODS|IEEE80211_FCTL_FROMDS):
+		return hdr->addr4;
+	default:
+		return hdr->addr2;
 	}
-	return hdr->addr2;
 }
 
 /**
@@ -577,12 +578,13 @@ static inline u8 *ieee80211_get_SA(struct ieee80211_hdr *hdr)
  */
 static inline u8 *ieee80211_get_DA(struct ieee80211_hdr *hdr)
 {
-	u8 *raw = (u8 *) hdr;
-	u8 to_ds = (*(raw+1)) & 1; /* get the TODS bit */
+	__le16 fc = hdr->frame_control;
+	fc &= cpu_to_le16(IEEE80211_FCTL_TODS);
 
-	if (to_ds)
+	if (fc)
 		return hdr->addr3;
-	return hdr->addr1;
+	else
+		return hdr->addr1;
 }
 
 /**
@@ -595,8 +597,8 @@ static inline u8 *ieee80211_get_DA(struct ieee80211_hdr *hdr)
  */
 static inline int ieee80211_get_morefrag(struct ieee80211_hdr *hdr)
 {
-	return (le16_to_cpu(hdr->frame_control) &
-		IEEE80211_FCTL_MOREFRAGS) != 0;
+	__le16 fc = hdr->frame_control;
+	return !!(fc & cpu_to_le16(IEEE80211_FCTL_MOREFRAGS));
 }
 
 #endif /* IEEE80211_H */
-- 
cgit v1.2.3


From 566bfe5a8bcde13188a356f77666f8115813cf31 Mon Sep 17 00:00:00 2001
From: Bruno Randolf <br1@einfach.org>
Date: Thu, 8 May 2008 19:15:40 +0200
Subject: mac80211: use hardware flags for signal/noise units

trying to clean up the signal/noise code. the previous code in mac80211 had
confusing names for the related variables, did not have much definition of
what units of signal and noise were provided and used implicit mechanisms from
the wireless extensions.

this patch introduces hardware capability flags to let the hardware specify
clearly if it can provide signal and noise level values and which units it can
provide. this also anticipates possible new units like RCPI in the future.

for signal:

  IEEE80211_HW_SIGNAL_UNSPEC - unspecified, unknown, hw specific
  IEEE80211_HW_SIGNAL_DB     - dB difference to unspecified reference point
  IEEE80211_HW_SIGNAL_DBM    - dBm, difference to 1mW

for noise we currently only have dBm:

  IEEE80211_HW_NOISE_DBM     - dBm, difference to 1mW

if IEEE80211_HW_SIGNAL_UNSPEC or IEEE80211_HW_SIGNAL_DB is used the driver has
to provide the maximum value (max_signal) it reports in order for applications
to make sense of the signal values.

i tried my best to find out for each driver what it can provide and update it
but i'm not sure (?) for some of them and used the more conservative guess in
doubt. this can be fixed easily after this patch has been merged by changing
the hardware flags of the driver.

DRIVER          SIGNAL    MAX	NOISE   QUAL
-----------------------------------------------------------------
adm8211         unspec(?) 100   n/a     missing
at76_usb        unspec(?) (?)   unused  missing
ath5k           dBm             dBm     percent rssi
b43legacy       dBm             dBm     percent jssi(?)
b43             dBm             dBm     percent jssi(?)
iwl-3945        dBm             dBm     percent snr+more
iwl-4965        dBm             dBm     percent snr+more
p54             unspec    127   n/a     missing
rt2x00          dBm	        n/a     percent rssi+tx/rx frame success
  rt2400        dBm             n/a
  rt2500pci     dBm             n/a
  rt2500usb     dBm             n/a
  rt61pci       dBm             n/a
  rt73usb       dBm             n/a
rtl8180         unspec(?) 65    n/a     (?)
rtl8187         unspec(?) 65    (?)     noise(?)
zd1211          dB(?)     100   n/a     percent

drivers/net/wireless/ath5k/base.c:      Changes-licensed-under: 3-Clause-BSD

Signed-off-by: Bruno Randolf <br1@einfach.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/net/wireless/adm8211.c              |  7 +++--
 drivers/net/wireless/ath5k/base.c           | 23 +++-----------
 drivers/net/wireless/b43/main.c             |  8 ++---
 drivers/net/wireless/b43/xmit.c             |  5 ++-
 drivers/net/wireless/b43legacy/main.c       |  7 ++---
 drivers/net/wireless/b43legacy/xmit.c       |  4 +--
 drivers/net/wireless/iwlwifi/iwl-3945.c     | 18 +++++------
 drivers/net/wireless/iwlwifi/iwl-4965.c     | 12 ++++----
 drivers/net/wireless/iwlwifi/iwl-core.c     | 16 +++-------
 drivers/net/wireless/iwlwifi/iwl3945-base.c | 15 +++------
 drivers/net/wireless/p54/p54common.c        |  7 +++--
 drivers/net/wireless/rt2x00/rt2400pci.c     |  5 ++-
 drivers/net/wireless/rt2x00/rt2500pci.c     |  6 ++--
 drivers/net/wireless/rt2x00/rt2500usb.c     |  6 ++--
 drivers/net/wireless/rt2x00/rt2x00dev.c     |  4 +--
 drivers/net/wireless/rt2x00/rt61pci.c       |  5 ++-
 drivers/net/wireless/rt2x00/rt73usb.c       |  5 ++-
 drivers/net/wireless/rtl8180_dev.c          |  9 +++---
 drivers/net/wireless/rtl8187_dev.c          | 10 +++---
 drivers/net/wireless/zd1211rw/zd_mac.c      | 12 ++++----
 include/net/mac80211.h                      | 47 +++++++++++++++++++++--------
 net/ieee80211/ieee80211_rx.c                |  2 +-
 net/mac80211/debugfs_sta.c                  |  2 +-
 net/mac80211/ieee80211_i.h                  |  2 +-
 net/mac80211/main.c                         | 10 +++---
 net/mac80211/mlme.c                         | 12 ++++----
 net/mac80211/rx.c                           |  4 +--
 net/mac80211/sta_info.h                     |  2 +-
 net/mac80211/wext.c                         | 28 ++++++++++++-----
 29 files changed, 148 insertions(+), 145 deletions(-)

(limited to 'include')

diff --git a/drivers/net/wireless/adm8211.c b/drivers/net/wireless/adm8211.c
index d93a1de77eb0..7af5d8851f67 100644
--- a/drivers/net/wireless/adm8211.c
+++ b/drivers/net/wireless/adm8211.c
@@ -445,9 +445,9 @@ static void adm8211_interrupt_rci(struct ieee80211_hw *dev)
 			struct ieee80211_rx_status rx_status = {0};
 
 			if (priv->pdev->revision < ADM8211_REV_CA)
-				rx_status.ssi = rssi;
+				rx_status.signal = rssi;
 			else
-				rx_status.ssi = 100 - rssi;
+				rx_status.signal = 100 - rssi;
 
 			rx_status.rate_idx = rate;
 
@@ -1893,9 +1893,10 @@ static int __devinit adm8211_probe(struct pci_dev *pdev,
 
 	dev->extra_tx_headroom = sizeof(struct adm8211_tx_hdr);
 	/* dev->flags = IEEE80211_HW_RX_INCLUDES_FCS in promisc mode */
+	dev->flags = IEEE80211_HW_SIGNAL_UNSPEC;
 
 	dev->channel_change_time = 1000;
-	dev->max_rssi = 100;	/* FIXME: find better value */
+	dev->max_signal = 100;    /* FIXME: find better value */
 
 	dev->queues = 1; /* ADM8211C supports more, maybe ADM8211B too */
 
diff --git a/drivers/net/wireless/ath5k/base.c b/drivers/net/wireless/ath5k/base.c
index 3201c1604340..dd8adaddd7a0 100644
--- a/drivers/net/wireless/ath5k/base.c
+++ b/drivers/net/wireless/ath5k/base.c
@@ -458,13 +458,11 @@ ath5k_pci_probe(struct pci_dev *pdev,
 
 	/* Initialize driver private data */
 	SET_IEEE80211_DEV(hw, &pdev->dev);
-	hw->flags = IEEE80211_HW_RX_INCLUDES_FCS;
+	hw->flags = IEEE80211_HW_RX_INCLUDES_FCS |
+		    IEEE80211_HW_SIGNAL_DBM |
+		    IEEE80211_HW_NOISE_DBM;
 	hw->extra_tx_headroom = 2;
 	hw->channel_change_time = 5000;
-	/* these names are misleading */
-	hw->max_rssi = -110; /* signal in dBm */
-	hw->max_noise = -110; /* noise in dBm */
-	hw->max_signal = 100; /* we will provide a percentage based on rssi */
 	sc = hw->priv;
 	sc->hw = hw;
 	sc->pdev = pdev;
@@ -1893,20 +1891,9 @@ accept:
 		rxs.freq = sc->curchan->center_freq;
 		rxs.band = sc->curband->band;
 
-		/*
-		 * signal quality:
-		 * the names here are misleading and the usage of these
-		 * values by iwconfig makes it even worse
-		 */
-		/* noise floor in dBm, from the last noise calibration */
 		rxs.noise = sc->ah->ah_noise_floor;
-		/* signal level in dBm */
-		rxs.ssi = rxs.noise + rs.rs_rssi;
-		/*
-		 * "signal" is actually displayed as Link Quality by iwconfig
-		 * we provide a percentage based on rssi (assuming max rssi 64)
-		 */
-		rxs.signal = rs.rs_rssi * 100 / 64;
+		rxs.signal = rxs.noise + rs.rs_rssi;
+		rxs.qual = rs.rs_rssi * 100 / 64;
 
 		rxs.antenna = rs.rs_antenna;
 		rxs.rate_idx = ath5k_hw_to_driver_rix(sc, rs.rs_rate);
diff --git a/drivers/net/wireless/b43/main.c b/drivers/net/wireless/b43/main.c
index 7c23aa8705c2..fc23ba5309bd 100644
--- a/drivers/net/wireless/b43/main.c
+++ b/drivers/net/wireless/b43/main.c
@@ -4466,10 +4466,10 @@ static int b43_wireless_init(struct ssb_device *dev)
 
 	/* fill hw info */
 	hw->flags = IEEE80211_HW_HOST_GEN_BEACON_TEMPLATE |
-		    IEEE80211_HW_RX_INCLUDES_FCS;
-	hw->max_signal = 100;
-	hw->max_rssi = -110;
-	hw->max_noise = -110;
+		    IEEE80211_HW_RX_INCLUDES_FCS |
+		    IEEE80211_HW_SIGNAL_DBM |
+		    IEEE80211_HW_NOISE_DBM;
+
 	hw->queues = b43_modparam_qos ? 4 : 1;
 	SET_IEEE80211_DEV(hw, dev->dev);
 	if (is_valid_ether_addr(sprom->et1mac))
diff --git a/drivers/net/wireless/b43/xmit.c b/drivers/net/wireless/b43/xmit.c
index 88491947a209..afce9338d83a 100644
--- a/drivers/net/wireless/b43/xmit.c
+++ b/drivers/net/wireless/b43/xmit.c
@@ -581,12 +581,11 @@ void b43_rx(struct b43_wldev *dev, struct sk_buff *skb, const void *_rxhdr)
 		//      and also find out what the maximum possible value is.
 		//      Fill status.ssi and status.signal fields.
 	} else {
-		status.ssi = b43_rssi_postprocess(dev, rxhdr->jssi,
+		status.signal = b43_rssi_postprocess(dev, rxhdr->jssi,
 						  (phystat0 & B43_RX_PHYST0_OFDM),
 						  (phystat0 & B43_RX_PHYST0_GAINCTL),
 						  (phystat3 & B43_RX_PHYST3_TRSTATE));
-		/* the next line looks wrong, but is what mac80211 wants */
-		status.signal = (rxhdr->jssi * 100) / B43_RX_MAX_SSI;
+		status.qual = (rxhdr->jssi * 100) / B43_RX_MAX_SSI;
 	}
 
 	if (phystat0 & B43_RX_PHYST0_OFDM)
diff --git a/drivers/net/wireless/b43legacy/main.c b/drivers/net/wireless/b43legacy/main.c
index d3820dce3c43..7755c59e0803 100644
--- a/drivers/net/wireless/b43legacy/main.c
+++ b/drivers/net/wireless/b43legacy/main.c
@@ -3718,10 +3718,9 @@ static int b43legacy_wireless_init(struct ssb_device *dev)
 
 	/* fill hw info */
 	hw->flags = IEEE80211_HW_HOST_GEN_BEACON_TEMPLATE |
-		    IEEE80211_HW_RX_INCLUDES_FCS;
-	hw->max_signal = 100;
-	hw->max_rssi = -110;
-	hw->max_noise = -110;
+		    IEEE80211_HW_RX_INCLUDES_FCS |
+		    IEEE80211_HW_SIGNAL_DBM |
+		    IEEE80211_HW_NOISE_DBM;
 	hw->queues = 1; /* FIXME: hardware has more queues */
 	SET_IEEE80211_DEV(hw, dev->dev);
 	if (is_valid_ether_addr(sprom->et1mac))
diff --git a/drivers/net/wireless/b43legacy/xmit.c b/drivers/net/wireless/b43legacy/xmit.c
index fc83dab6e2c7..bed9e041d6c5 100644
--- a/drivers/net/wireless/b43legacy/xmit.c
+++ b/drivers/net/wireless/b43legacy/xmit.c
@@ -532,12 +532,12 @@ void b43legacy_rx(struct b43legacy_wldev *dev,
 		}
 	}
 
-	status.ssi = b43legacy_rssi_postprocess(dev, jssi,
+	status.signal = b43legacy_rssi_postprocess(dev, jssi,
 				      (phystat0 & B43legacy_RX_PHYST0_OFDM),
 				      (phystat0 & B43legacy_RX_PHYST0_GAINCTL),
 				      (phystat3 & B43legacy_RX_PHYST3_TRSTATE));
 	status.noise = dev->stats.link_noise;
-	status.signal = (jssi * 100) / B43legacy_RX_MAX_SSI;
+	status.qual = (jssi * 100) / B43legacy_RX_MAX_SSI;
 	/* change to support A PHY */
 	if (phystat0 & B43legacy_RX_PHYST0_OFDM)
 		status.rate_idx = b43legacy_plcp_get_bitrate_idx_ofdm(plcp, false);
diff --git a/drivers/net/wireless/iwlwifi/iwl-3945.c b/drivers/net/wireless/iwlwifi/iwl-3945.c
index 8464397f7816..d7beeff2642c 100644
--- a/drivers/net/wireless/iwlwifi/iwl-3945.c
+++ b/drivers/net/wireless/iwlwifi/iwl-3945.c
@@ -520,7 +520,7 @@ static void iwl3945_add_radiotap(struct iwl3945_priv *priv,
 {
 	/* First cache any information we need before we overwrite
 	 * the information provided in the skb from the hardware */
-	s8 signal = stats->ssi;
+	s8 signal = stats->signal;
 	s8 noise = 0;
 	int rate = stats->rate_idx;
 	u64 tsf = stats->mactime;
@@ -693,7 +693,7 @@ static void iwl3945_rx_reply_rx(struct iwl3945_priv *priv,
 	}
 
 	/* Convert 3945's rssi indicator to dBm */
-	rx_status.ssi = rx_stats->rssi - IWL_RSSI_OFFSET;
+	rx_status.signal = rx_stats->rssi - IWL_RSSI_OFFSET;
 
 	/* Set default noise value to -127 */
 	if (priv->last_rx_noise == 0)
@@ -712,21 +712,21 @@ static void iwl3945_rx_reply_rx(struct iwl3945_priv *priv,
 	 * Calculate rx_status.signal (quality indicator in %) based on SNR. */
 	if (rx_stats_noise_diff) {
 		snr = rx_stats_sig_avg / rx_stats_noise_diff;
-		rx_status.noise = rx_status.ssi -
+		rx_status.noise = rx_status.signal -
 					iwl3945_calc_db_from_ratio(snr);
-		rx_status.signal = iwl3945_calc_sig_qual(rx_status.ssi,
+		rx_status.qual = iwl3945_calc_sig_qual(rx_status.signal,
 							 rx_status.noise);
 
 	/* If noise info not available, calculate signal quality indicator (%)
 	 *   using just the dBm signal level. */
 	} else {
 		rx_status.noise = priv->last_rx_noise;
-		rx_status.signal = iwl3945_calc_sig_qual(rx_status.ssi, 0);
+		rx_status.qual = iwl3945_calc_sig_qual(rx_status.signal, 0);
 	}
 
 
 	IWL_DEBUG_STATS("Rssi %d noise %d qual %d sig_avg %d noise_diff %d\n",
-			rx_status.ssi, rx_status.noise, rx_status.signal,
+			rx_status.signal, rx_status.noise, rx_status.qual,
 			rx_stats_sig_avg, rx_stats_noise_diff);
 
 	header = (struct ieee80211_hdr *)IWL_RX_DATA(pkt);
@@ -736,8 +736,8 @@ static void iwl3945_rx_reply_rx(struct iwl3945_priv *priv,
 	IWL_DEBUG_STATS_LIMIT("[%c] %d RSSI:%d Signal:%u, Noise:%u, Rate:%u\n",
 			      network_packet ? '*' : ' ',
 			      le16_to_cpu(rx_hdr->channel),
-			      rx_status.ssi, rx_status.ssi,
-			      rx_status.ssi, rx_status.rate_idx);
+			      rx_status.signal, rx_status.signal,
+			      rx_status.noise, rx_status.rate_idx);
 
 #ifdef CONFIG_IWL3945_DEBUG
 	if (iwl3945_debug_level & (IWL_DL_RX))
@@ -748,7 +748,7 @@ static void iwl3945_rx_reply_rx(struct iwl3945_priv *priv,
 	if (network_packet) {
 		priv->last_beacon_time = le32_to_cpu(rx_end->beacon_timestamp);
 		priv->last_tsf = le64_to_cpu(rx_end->timestamp);
-		priv->last_rx_rssi = rx_status.ssi;
+		priv->last_rx_rssi = rx_status.signal;
 		priv->last_rx_noise = rx_status.noise;
 	}
 
diff --git a/drivers/net/wireless/iwlwifi/iwl-4965.c b/drivers/net/wireless/iwlwifi/iwl-4965.c
index b7036e011457..45cbd7789bd0 100644
--- a/drivers/net/wireless/iwlwifi/iwl-4965.c
+++ b/drivers/net/wireless/iwlwifi/iwl-4965.c
@@ -2362,7 +2362,7 @@ static void iwl4965_add_radiotap(struct iwl_priv *priv,
 				 struct ieee80211_rx_status *stats,
 				 u32 ampdu_status)
 {
-	s8 signal = stats->ssi;
+	s8 signal = stats->signal;
 	s8 noise = 0;
 	int rate = stats->rate_idx;
 	u64 tsf = stats->mactime;
@@ -2963,7 +2963,7 @@ static void iwl4965_rx_reply_rx(struct iwl_priv *priv,
 	priv->ucode_beacon_time = le32_to_cpu(rx_start->beacon_time_stamp);
 
 	/* Find max signal strength (dBm) among 3 antenna/receiver chains */
-	rx_status.ssi = iwl4965_calc_rssi(priv, rx_start);
+	rx_status.signal = iwl4965_calc_rssi(priv, rx_start);
 
 	/* Meaningful noise values are available only from beacon statistics,
 	 *   which are gathered only when associated, and indicate noise
@@ -2972,11 +2972,11 @@ static void iwl4965_rx_reply_rx(struct iwl_priv *priv,
 	if (iwl_is_associated(priv) &&
 	    !test_bit(STATUS_SCANNING, &priv->status)) {
 		rx_status.noise = priv->last_rx_noise;
-		rx_status.signal = iwl4965_calc_sig_qual(rx_status.ssi,
+		rx_status.qual = iwl4965_calc_sig_qual(rx_status.signal,
 							 rx_status.noise);
 	} else {
 		rx_status.noise = IWL_NOISE_MEAS_NOT_AVAILABLE;
-		rx_status.signal = iwl4965_calc_sig_qual(rx_status.ssi, 0);
+		rx_status.qual = iwl4965_calc_sig_qual(rx_status.signal, 0);
 	}
 
 	/* Reset beacon noise level if not associated. */
@@ -2988,7 +2988,7 @@ static void iwl4965_rx_reply_rx(struct iwl_priv *priv,
 	iwl4965_dbg_report_frame(priv, pkt, header, 1);
 
 	IWL_DEBUG_STATS_LIMIT("Rssi %d, noise %d, qual %d, TSF %llu\n",
-			      rx_status.ssi, rx_status.noise, rx_status.signal,
+			      rx_status.signal, rx_status.noise, rx_status.signal,
 			      (unsigned long long)rx_status.mactime);
 
 
@@ -3000,7 +3000,7 @@ static void iwl4965_rx_reply_rx(struct iwl_priv *priv,
 
 	network_packet = iwl4965_is_network_packet(priv, header);
 	if (network_packet) {
-		priv->last_rx_rssi = rx_status.ssi;
+		priv->last_rx_rssi = rx_status.signal;
 		priv->last_beacon_time =  priv->ucode_beacon_time;
 		priv->last_tsf = le64_to_cpu(rx_start->timestamp);
 	}
diff --git a/drivers/net/wireless/iwlwifi/iwl-core.c b/drivers/net/wireless/iwlwifi/iwl-core.c
index 27e56b89d940..d3cbad2bf877 100644
--- a/drivers/net/wireless/iwlwifi/iwl-core.c
+++ b/drivers/net/wireless/iwlwifi/iwl-core.c
@@ -746,18 +746,10 @@ static void iwlcore_init_hw(struct iwl_priv *priv)
 	struct ieee80211_hw *hw = priv->hw;
 	hw->rate_control_algorithm = "iwl-4965-rs";
 
-	/* Tell mac80211 and its clients (e.g. Wireless Extensions)
-	 *	 the range of signal quality values that we'll provide.
-	 * Negative values for level/noise indicate that we'll provide dBm.
-	 * For WE, at least, non-0 values here *enable* display of values
-	 *	 in app (iwconfig). */
-	hw->max_rssi = -20; /* signal level, negative indicates dBm */
-	hw->max_noise = -20;	/* noise level, negative indicates dBm */
-	hw->max_signal = 100;	/* link quality indication (%) */
-
-	/* Tell mac80211 our Tx characteristics */
-	hw->flags = IEEE80211_HW_HOST_GEN_BEACON_TEMPLATE;
-
+	/* Tell mac80211 our characteristics */
+	hw->flags = IEEE80211_HW_HOST_GEN_BEACON_TEMPLATE |
+		    IEEE80211_HW_SIGNAL_DBM |
+		    IEEE80211_HW_NOISE_DBM;
 	/* Default value; 4 EDCA QOS priorities */
 	hw->queues = 4;
 #ifdef CONFIG_IWL4965_HT
diff --git a/drivers/net/wireless/iwlwifi/iwl3945-base.c b/drivers/net/wireless/iwlwifi/iwl3945-base.c
index 4baa185ba500..c1234ff4fc98 100644
--- a/drivers/net/wireless/iwlwifi/iwl3945-base.c
+++ b/drivers/net/wireless/iwlwifi/iwl3945-base.c
@@ -8029,17 +8029,10 @@ static int iwl3945_pci_probe(struct pci_dev *pdev, const struct pci_device_id *e
 
 	priv->ibss_beacon = NULL;
 
-	/* Tell mac80211 and its clients (e.g. Wireless Extensions)
-	 *   the range of signal quality values that we'll provide.
-	 * Negative values for level/noise indicate that we'll provide dBm.
-	 * For WE, at least, non-0 values here *enable* display of values
-	 *   in app (iwconfig). */
-	hw->max_rssi = -20;	/* signal level, negative indicates dBm */
-	hw->max_noise = -20;	/* noise level, negative indicates dBm */
-	hw->max_signal = 100;	/* link quality indication (%) */
-
-	/* Tell mac80211 our Tx characteristics */
-	hw->flags = IEEE80211_HW_HOST_GEN_BEACON_TEMPLATE;
+	/* Tell mac80211 our characteristics */
+	hw->flags = IEEE80211_HW_HOST_GEN_BEACON_TEMPLATE |
+		    IEEE80211_HW_SIGNAL_DBM |
+		    IEEE80211_HW_NOISE_DBM;
 
 	/* 4 EDCA QOS priorities */
 	hw->queues = 4;
diff --git a/drivers/net/wireless/p54/p54common.c b/drivers/net/wireless/p54/p54common.c
index 33d608a60d79..9cbef5bce0f6 100644
--- a/drivers/net/wireless/p54/p54common.c
+++ b/drivers/net/wireless/p54/p54common.c
@@ -355,7 +355,7 @@ static void p54_rx_data(struct ieee80211_hw *dev, struct sk_buff *skb)
 	struct ieee80211_rx_status rx_status = {0};
 	u16 freq = le16_to_cpu(hdr->freq);
 
-	rx_status.ssi = hdr->rssi;
+	rx_status.signal = hdr->rssi;
 	/* XX correct? */
 	rx_status.rate_idx = hdr->rate & 0xf;
 	rx_status.freq = freq;
@@ -1000,9 +1000,10 @@ struct ieee80211_hw *p54_init_common(size_t priv_data_len)
 	skb_queue_head_init(&priv->tx_queue);
 	dev->wiphy->bands[IEEE80211_BAND_2GHZ] = &band_2GHz;
 	dev->flags = IEEE80211_HW_HOST_BROADCAST_PS_BUFFERING | /* not sure */
-		    IEEE80211_HW_RX_INCLUDES_FCS;
+		     IEEE80211_HW_RX_INCLUDES_FCS |
+		     IEEE80211_HW_SIGNAL_UNSPEC;
 	dev->channel_change_time = 1000;	/* TODO: find actual value */
-	dev->max_rssi = 127;
+	dev->max_signal = 127;
 
 	priv->tx_stats[0].limit = 5;
 	dev->queues = 1;
diff --git a/drivers/net/wireless/rt2x00/rt2400pci.c b/drivers/net/wireless/rt2x00/rt2400pci.c
index 247fbbfb0f2b..afa565c63621 100644
--- a/drivers/net/wireless/rt2x00/rt2400pci.c
+++ b/drivers/net/wireless/rt2x00/rt2400pci.c
@@ -1361,10 +1361,9 @@ static void rt2400pci_probe_hw_mode(struct rt2x00_dev *rt2x00dev)
 	/*
 	 * Initialize all hw fields.
 	 */
-	rt2x00dev->hw->flags = IEEE80211_HW_HOST_BROADCAST_PS_BUFFERING;
+	rt2x00dev->hw->flags = IEEE80211_HW_HOST_BROADCAST_PS_BUFFERING |
+			       IEEE80211_HW_SIGNAL_DBM;
 	rt2x00dev->hw->extra_tx_headroom = 0;
-	rt2x00dev->hw->max_signal = MAX_SIGNAL;
-	rt2x00dev->hw->max_rssi = MAX_RX_SSI;
 	rt2x00dev->hw->queues = 2;
 
 	SET_IEEE80211_DEV(rt2x00dev->hw, &rt2x00dev_pci(rt2x00dev)->dev);
diff --git a/drivers/net/wireless/rt2x00/rt2500pci.c b/drivers/net/wireless/rt2x00/rt2500pci.c
index 0d53c75d55dd..c06f1b5e5887 100644
--- a/drivers/net/wireless/rt2x00/rt2500pci.c
+++ b/drivers/net/wireless/rt2x00/rt2500pci.c
@@ -1680,10 +1680,10 @@ static void rt2500pci_probe_hw_mode(struct rt2x00_dev *rt2x00dev)
 	/*
 	 * Initialize all hw fields.
 	 */
-	rt2x00dev->hw->flags = IEEE80211_HW_HOST_BROADCAST_PS_BUFFERING;
+	rt2x00dev->hw->flags = IEEE80211_HW_HOST_BROADCAST_PS_BUFFERING |
+			       IEEE80211_HW_SIGNAL_DBM;
+
 	rt2x00dev->hw->extra_tx_headroom = 0;
-	rt2x00dev->hw->max_signal = MAX_SIGNAL;
-	rt2x00dev->hw->max_rssi = MAX_RX_SSI;
 	rt2x00dev->hw->queues = 2;
 
 	SET_IEEE80211_DEV(rt2x00dev->hw, &rt2x00dev_pci(rt2x00dev)->dev);
diff --git a/drivers/net/wireless/rt2x00/rt2500usb.c b/drivers/net/wireless/rt2x00/rt2500usb.c
index 80b34d47114a..88bafdf8f0fa 100644
--- a/drivers/net/wireless/rt2x00/rt2500usb.c
+++ b/drivers/net/wireless/rt2x00/rt2500usb.c
@@ -1587,10 +1587,10 @@ static void rt2500usb_probe_hw_mode(struct rt2x00_dev *rt2x00dev)
 	rt2x00dev->hw->flags =
 	    IEEE80211_HW_HOST_GEN_BEACON_TEMPLATE |
 	    IEEE80211_HW_RX_INCLUDES_FCS |
-	    IEEE80211_HW_HOST_BROADCAST_PS_BUFFERING;
+	    IEEE80211_HW_HOST_BROADCAST_PS_BUFFERING |
+	    IEEE80211_HW_SIGNAL_DBM;
+
 	rt2x00dev->hw->extra_tx_headroom = TXD_DESC_SIZE;
-	rt2x00dev->hw->max_signal = MAX_SIGNAL;
-	rt2x00dev->hw->max_rssi = MAX_RX_SSI;
 	rt2x00dev->hw->queues = 2;
 
 	SET_IEEE80211_DEV(rt2x00dev->hw, &rt2x00dev_usb(rt2x00dev)->dev);
diff --git a/drivers/net/wireless/rt2x00/rt2x00dev.c b/drivers/net/wireless/rt2x00/rt2x00dev.c
index 9929b15e28ba..f51a50b31e16 100644
--- a/drivers/net/wireless/rt2x00/rt2x00dev.c
+++ b/drivers/net/wireless/rt2x00/rt2x00dev.c
@@ -600,9 +600,9 @@ void rt2x00lib_rxdone(struct queue_entry *entry,
 	rt2x00dev->link.qual.rx_success++;
 
 	rx_status->rate_idx = idx;
-	rx_status->signal =
+	rx_status->qual =
 	    rt2x00lib_calculate_link_signal(rt2x00dev, rxdesc->rssi);
-	rx_status->ssi = rxdesc->rssi;
+	rx_status->signal = rxdesc->rssi;
 	rx_status->flag = rxdesc->flags;
 	rx_status->antenna = rt2x00dev->link.ant.active.rx;
 
diff --git a/drivers/net/wireless/rt2x00/rt61pci.c b/drivers/net/wireless/rt2x00/rt61pci.c
index 98af4d26583d..eaf23b908772 100644
--- a/drivers/net/wireless/rt2x00/rt61pci.c
+++ b/drivers/net/wireless/rt2x00/rt61pci.c
@@ -2245,10 +2245,9 @@ static void rt61pci_probe_hw_mode(struct rt2x00_dev *rt2x00dev)
 	 */
 	rt2x00dev->hw->flags =
 	    IEEE80211_HW_HOST_GEN_BEACON_TEMPLATE |
-	    IEEE80211_HW_HOST_BROADCAST_PS_BUFFERING;
+	    IEEE80211_HW_HOST_BROADCAST_PS_BUFFERING |
+	    IEEE80211_HW_SIGNAL_DBM;
 	rt2x00dev->hw->extra_tx_headroom = 0;
-	rt2x00dev->hw->max_signal = MAX_SIGNAL;
-	rt2x00dev->hw->max_rssi = MAX_RX_SSI;
 	rt2x00dev->hw->queues = 4;
 
 	SET_IEEE80211_DEV(rt2x00dev->hw, &rt2x00dev_pci(rt2x00dev)->dev);
diff --git a/drivers/net/wireless/rt2x00/rt73usb.c b/drivers/net/wireless/rt2x00/rt73usb.c
index 351d95c4f50d..51c5575ed02f 100644
--- a/drivers/net/wireless/rt2x00/rt73usb.c
+++ b/drivers/net/wireless/rt2x00/rt73usb.c
@@ -1830,10 +1830,9 @@ static void rt73usb_probe_hw_mode(struct rt2x00_dev *rt2x00dev)
 	 */
 	rt2x00dev->hw->flags =
 	    IEEE80211_HW_HOST_GEN_BEACON_TEMPLATE |
-	    IEEE80211_HW_HOST_BROADCAST_PS_BUFFERING;
+	    IEEE80211_HW_HOST_BROADCAST_PS_BUFFERING |
+	    IEEE80211_HW_SIGNAL_DBM;
 	rt2x00dev->hw->extra_tx_headroom = TXD_DESC_SIZE;
-	rt2x00dev->hw->max_signal = MAX_SIGNAL;
-	rt2x00dev->hw->max_rssi = MAX_RX_SSI;
 	rt2x00dev->hw->queues = 4;
 
 	SET_IEEE80211_DEV(rt2x00dev->hw, &rt2x00dev_usb(rt2x00dev)->dev);
diff --git a/drivers/net/wireless/rtl8180_dev.c b/drivers/net/wireless/rtl8180_dev.c
index c181f23e930d..c220998cee65 100644
--- a/drivers/net/wireless/rtl8180_dev.c
+++ b/drivers/net/wireless/rtl8180_dev.c
@@ -132,8 +132,8 @@ static void rtl8180_handle_rx(struct ieee80211_hw *dev)
 
 			rx_status.antenna = (flags2 >> 15) & 1;
 			/* TODO: improve signal/rssi reporting */
-			rx_status.signal = flags2 & 0xFF;
-			rx_status.ssi = (flags2 >> 8) & 0x7F;
+			rx_status.qual = flags2 & 0xFF;
+			rx_status.signal = (flags2 >> 8) & 0x7F;
 			/* XXX: is this correct? */
 			rx_status.rate_idx = (flags >> 20) & 0xF;
 			rx_status.freq = dev->conf.channel->center_freq;
@@ -894,9 +894,10 @@ static int __devinit rtl8180_probe(struct pci_dev *pdev,
 	dev->wiphy->bands[IEEE80211_BAND_2GHZ] = &priv->band;
 
 	dev->flags = IEEE80211_HW_HOST_BROADCAST_PS_BUFFERING |
-		     IEEE80211_HW_RX_INCLUDES_FCS;
+		     IEEE80211_HW_RX_INCLUDES_FCS |
+		     IEEE80211_HW_SIGNAL_UNSPEC;
 	dev->queues = 1;
-	dev->max_rssi = 65;
+	dev->max_signal = 65;
 
 	reg = rtl818x_ioread32(priv, &priv->map->TX_CONF);
 	reg &= RTL818X_TX_CONF_HWVER_MASK;
diff --git a/drivers/net/wireless/rtl8187_dev.c b/drivers/net/wireless/rtl8187_dev.c
index d5787b37e1fb..e14c84248686 100644
--- a/drivers/net/wireless/rtl8187_dev.c
+++ b/drivers/net/wireless/rtl8187_dev.c
@@ -261,8 +261,8 @@ static void rtl8187_rx_cb(struct urb *urb)
 	}
 
 	rx_status.antenna = (hdr->signal >> 7) & 1;
-	rx_status.signal = 64 - min(hdr->noise, (u8)64);
-	rx_status.ssi = signal;
+	rx_status.qual = 64 - min(hdr->noise, (u8)64);
+	rx_status.signal = signal;
 	rx_status.rate_idx = rate;
 	rx_status.freq = dev->conf.channel->center_freq;
 	rx_status.band = dev->conf.channel->band;
@@ -740,11 +740,11 @@ static int __devinit rtl8187_probe(struct usb_interface *intf,
 
 	priv->mode = IEEE80211_IF_TYPE_MNTR;
 	dev->flags = IEEE80211_HW_HOST_BROADCAST_PS_BUFFERING |
-		     IEEE80211_HW_RX_INCLUDES_FCS;
+		     IEEE80211_HW_RX_INCLUDES_FCS |
+		     IEEE80211_HW_SIGNAL_UNSPEC;
 	dev->extra_tx_headroom = sizeof(struct rtl8187_tx_hdr);
 	dev->queues = 1;
-	dev->max_rssi = 65;
-	dev->max_signal = 64;
+	dev->max_signal = 65;
 
 	eeprom.data = dev;
 	eeprom.register_read = rtl8187_eeprom_register_read;
diff --git a/drivers/net/wireless/zd1211rw/zd_mac.c b/drivers/net/wireless/zd1211rw/zd_mac.c
index ee4331229f1f..0c736735e217 100644
--- a/drivers/net/wireless/zd1211rw/zd_mac.c
+++ b/drivers/net/wireless/zd1211rw/zd_mac.c
@@ -638,7 +638,7 @@ static int filter_ack(struct ieee80211_hw *hw, struct ieee80211_hdr *rx_hdr,
 
 			memset(&status, 0, sizeof(status));
 			status.flags = IEEE80211_TX_STATUS_ACK;
-			status.ack_signal = stats->ssi;
+			status.ack_signal = stats->signal;
 			__skb_unlink(skb, q);
 			tx_status(hw, skb, &status, 1);
 			goto out;
@@ -691,8 +691,8 @@ int zd_mac_rx(struct ieee80211_hw *hw, const u8 *buffer, unsigned int length)
 
 	stats.freq = zd_channels[_zd_chip_get_channel(&mac->chip) - 1].center_freq;
 	stats.band = IEEE80211_BAND_2GHZ;
-	stats.ssi = status->signal_strength;
-	stats.signal = zd_rx_qual_percent(buffer,
+	stats.signal = status->signal_strength;
+	stats.qual = zd_rx_qual_percent(buffer,
 		                          length - sizeof(struct rx_status),
 		                          status);
 
@@ -997,10 +997,10 @@ struct ieee80211_hw *zd_mac_alloc_hw(struct usb_interface *intf)
 	hw->wiphy->bands[IEEE80211_BAND_2GHZ] = &mac->band;
 
 	hw->flags = IEEE80211_HW_RX_INCLUDES_FCS |
-		    IEEE80211_HW_HOST_GEN_BEACON_TEMPLATE;
-	hw->max_rssi = 100;
-	hw->max_signal = 100;
+		    IEEE80211_HW_HOST_GEN_BEACON_TEMPLATE |
+		    IEEE80211_HW_SIGNAL_DB;
 
+	hw->max_signal = 100;
 	hw->queues = 1;
 	hw->extra_tx_headroom = sizeof(struct zd_ctrlset);
 
diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index 75a34609eed7..909956c97c44 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -336,13 +336,16 @@ enum mac80211_rx_flags {
  * The low-level driver should provide this information (the subset
  * supported by hardware) to the 802.11 code with each received
  * frame.
+ *
  * @mactime: value in microseconds of the 64-bit Time Synchronization Function
  * 	(TSF) timer when the first data symbol (MPDU) arrived at the hardware.
  * @band: the active band when this frame was received
  * @freq: frequency the radio was tuned to when receiving this frame, in MHz
- * @ssi: signal strength when receiving this frame
- * @signal: used as 'qual' in statistics reporting
- * @noise: PHY noise when receiving this frame
+ * @signal: signal strength when receiving this frame, either in dBm, in dB or
+ *	unspecified depending on the hardware capabilities flags
+ *	@IEEE80211_HW_SIGNAL_*
+ * @noise: noise when receiving this frame, in dBm.
+ * @qual: overall signal quality indication, in percent (0-100).
  * @antenna: antenna used
  * @rate_idx: index of data rate into band's supported rates
  * @flag: %RX_FLAG_*
@@ -351,9 +354,9 @@ struct ieee80211_rx_status {
 	u64 mactime;
 	enum ieee80211_band band;
 	int freq;
-	int ssi;
 	int signal;
 	int noise;
+	int qual;
 	int antenna;
 	int rate_idx;
 	int flag;
@@ -392,7 +395,8 @@ enum ieee80211_tx_status_flags {
  * 	relevant only if IEEE80211_TX_STATUS_AMPDU was set.
  * @ampdu_ack_map: block ack bit map for the aggregation.
  * 	relevant only if IEEE80211_TX_STATUS_AMPDU was set.
- * @ack_signal: signal strength of the ACK frame
+ * @ack_signal: signal strength of the ACK frame either in dBm, dB or unspec
+ *	depending on hardware capabilites flags @IEEE80211_HW_SIGNAL_*
  */
 struct ieee80211_tx_status {
 	struct ieee80211_tx_control control;
@@ -703,6 +707,25 @@ enum ieee80211_tkip_key_type {
  * @IEEE80211_HW_2GHZ_SHORT_PREAMBLE_INCAPABLE:
  *	Hardware is not capable of receiving frames with short preamble on
  *	the 2.4 GHz band.
+ *
+ * @IEEE80211_HW_SIGNAL_UNSPEC:
+ *	Hardware can provide signal values but we don't know its units. We
+ *	expect values between 0 and @max_signal.
+ *	If possible please provide dB or dBm instead.
+ *
+ * @IEEE80211_HW_SIGNAL_DB:
+ *	Hardware gives signal values in dB, decibel difference from an
+ *	arbitrary, fixed reference. We expect values between 0 and @max_signal.
+ *	If possible please provide dBm instead.
+ *
+ * @IEEE80211_HW_SIGNAL_DBM:
+ *	Hardware gives signal values in dBm, decibel difference from
+ *	one milliwatt. This is the preferred method since it is standardized
+ *	between different devices. @max_signal does not need to be set.
+ *
+ * @IEEE80211_HW_NOISE_DBM:
+ *	Hardware can provide noise (radio interference) values in units dBm,
+ *      decibel difference from one milliwatt.
  */
 enum ieee80211_hw_flags {
 	IEEE80211_HW_HOST_GEN_BEACON_TEMPLATE		= 1<<0,
@@ -710,6 +733,10 @@ enum ieee80211_hw_flags {
 	IEEE80211_HW_HOST_BROADCAST_PS_BUFFERING	= 1<<2,
 	IEEE80211_HW_2GHZ_SHORT_SLOT_INCAPABLE		= 1<<3,
 	IEEE80211_HW_2GHZ_SHORT_PREAMBLE_INCAPABLE	= 1<<4,
+	IEEE80211_HW_SIGNAL_UNSPEC			= 1<<5,
+	IEEE80211_HW_SIGNAL_DB				= 1<<6,
+	IEEE80211_HW_SIGNAL_DBM				= 1<<7,
+	IEEE80211_HW_NOISE_DBM				= 1<<8,
 };
 
 /**
@@ -740,12 +767,8 @@ enum ieee80211_hw_flags {
  *
  * @channel_change_time: time (in microseconds) it takes to change channels.
  *
- * @max_rssi: Maximum value for ssi in RX information, use
- *	negative numbers for dBm and 0 to indicate no support.
- *
- * @max_signal: like @max_rssi, but for the signal value.
- *
- * @max_noise: like @max_rssi, but for the noise value.
+ * @max_signal: Maximum value for signal (rssi) in RX information, used
+ *     only when @IEEE80211_HW_SIGNAL_UNSPEC or @IEEE80211_HW_SIGNAL_DB
  *
  * @queues: number of available hardware transmit queues for
  *	data packets. WMM/QoS requires at least four, these
@@ -775,9 +798,7 @@ struct ieee80211_hw {
 	int channel_change_time;
 	int vif_data_size;
 	u16 queues, ampdu_queues;
-	s8 max_rssi;
 	s8 max_signal;
-	s8 max_noise;
 };
 
 /**
diff --git a/net/ieee80211/ieee80211_rx.c b/net/ieee80211/ieee80211_rx.c
index 200ee1e63728..69dbc342a464 100644
--- a/net/ieee80211/ieee80211_rx.c
+++ b/net/ieee80211/ieee80211_rx.c
@@ -391,7 +391,7 @@ int ieee80211_rx(struct ieee80211_device *ieee, struct sk_buff *skb,
 
 		wstats.updated = 0;
 		if (rx_stats->mask & IEEE80211_STATMASK_RSSI) {
-			wstats.level = rx_stats->rssi;
+			wstats.level = rx_stats->signal;
 			wstats.updated |= IW_QUAL_LEVEL_UPDATED;
 		} else
 			wstats.updated |= IW_QUAL_LEVEL_INVALID;
diff --git a/net/mac80211/debugfs_sta.c b/net/mac80211/debugfs_sta.c
index 1f00273d99c2..a2cc0284c9d0 100644
--- a/net/mac80211/debugfs_sta.c
+++ b/net/mac80211/debugfs_sta.c
@@ -63,8 +63,8 @@ STA_FILE(tx_fragments, tx_fragments, LU);
 STA_FILE(tx_filtered, tx_filtered_count, LU);
 STA_FILE(tx_retry_failed, tx_retry_failed, LU);
 STA_FILE(tx_retry_count, tx_retry_count, LU);
-STA_FILE(last_rssi, last_rssi, D);
 STA_FILE(last_signal, last_signal, D);
+STA_FILE(last_qual, last_qual, D);
 STA_FILE(last_noise, last_noise, D);
 STA_FILE(channel_use, channel_use, D);
 STA_FILE(wep_weak_iv_count, wep_weak_iv_count, LU);
diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h
index 173b7fcb3029..ed0d9b35ae6f 100644
--- a/net/mac80211/ieee80211_i.h
+++ b/net/mac80211/ieee80211_i.h
@@ -82,7 +82,7 @@ struct ieee80211_sta_bss {
 	u16 capability; /* host byte order */
 	enum ieee80211_band band;
 	int freq;
-	int rssi, signal, noise;
+	int signal, noise, qual;
 	u8 *wpa_ie;
 	size_t wpa_ie_len;
 	u8 *rsn_ie;
diff --git a/net/mac80211/main.c b/net/mac80211/main.c
index f277407f0f5a..390df48dfd40 100644
--- a/net/mac80211/main.c
+++ b/net/mac80211/main.c
@@ -1702,13 +1702,13 @@ int ieee80211_register_hw(struct ieee80211_hw *hw)
 
 	local->hw.conf.beacon_int = 1000;
 
-	local->wstats_flags |= local->hw.max_rssi ?
-			       IW_QUAL_LEVEL_UPDATED : IW_QUAL_LEVEL_INVALID;
-	local->wstats_flags |= local->hw.max_signal ?
+	local->wstats_flags |= local->hw.flags & (IEEE80211_HW_SIGNAL_UNSPEC |
+						  IEEE80211_HW_SIGNAL_DB |
+						  IEEE80211_HW_SIGNAL_DBM) ?
 			       IW_QUAL_QUAL_UPDATED : IW_QUAL_QUAL_INVALID;
-	local->wstats_flags |= local->hw.max_noise ?
+	local->wstats_flags |= local->hw.flags & IEEE80211_HW_NOISE_DBM ?
 			       IW_QUAL_NOISE_UPDATED : IW_QUAL_NOISE_INVALID;
-	if (local->hw.max_rssi < 0 || local->hw.max_noise < 0)
+	if (local->hw.flags & IEEE80211_HW_SIGNAL_DBM)
 		local->wstats_flags |= IW_QUAL_DBM;
 
 	result = sta_info_start(local);
diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c
index 6c4b27be35da..dda30e98aa1e 100644
--- a/net/mac80211/mlme.c
+++ b/net/mac80211/mlme.c
@@ -1959,8 +1959,8 @@ static void ieee80211_rx_mgmt_assoc_resp(struct ieee80211_sub_if_data *sdata,
 					   local->hw.conf.channel->center_freq,
 					   ifsta->ssid, ifsta->ssid_len);
 		if (bss) {
-			sta->last_rssi = bss->rssi;
 			sta->last_signal = bss->signal;
+			sta->last_qual = bss->qual;
 			sta->last_noise = bss->noise;
 			ieee80211_rx_bss_put(dev, bss);
 		}
@@ -2629,9 +2629,9 @@ static void ieee80211_rx_bss_info(struct net_device *dev,
 
 	bss->timestamp = beacon_timestamp;
 	bss->last_update = jiffies;
-	bss->rssi = rx_status->ssi;
 	bss->signal = rx_status->signal;
 	bss->noise = rx_status->noise;
+	bss->qual = rx_status->qual;
 	if (!beacon && !bss->probe_resp)
 		bss->probe_resp = true;
 
@@ -3427,9 +3427,9 @@ static int ieee80211_sta_config_auth(struct net_device *dev,
 		    !ieee80211_sta_match_ssid(ifsta, bss->ssid, bss->ssid_len))
 			continue;
 
-		if (!selected || top_rssi < bss->rssi) {
+		if (!selected || top_rssi < bss->signal) {
 			selected = bss;
-			top_rssi = bss->rssi;
+			top_rssi = bss->signal;
 		}
 	}
 	if (selected)
@@ -4060,8 +4060,8 @@ ieee80211_sta_scan_result(struct net_device *dev,
 
 	memset(&iwe, 0, sizeof(iwe));
 	iwe.cmd = IWEVQUAL;
-	iwe.u.qual.qual = bss->signal;
-	iwe.u.qual.level = bss->rssi;
+	iwe.u.qual.qual = bss->qual;
+	iwe.u.qual.level = bss->signal;
 	iwe.u.qual.noise = bss->noise;
 	iwe.u.qual.updated = local->wstats_flags;
 	current_ev = iwe_stream_add_event(current_ev, end_buf, &iwe,
diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c
index b399e09ec7aa..474f13662c84 100644
--- a/net/mac80211/rx.c
+++ b/net/mac80211/rx.c
@@ -209,7 +209,7 @@ ieee80211_rx_monitor(struct ieee80211_local *local, struct sk_buff *origskb,
 				cpu_to_le16(IEEE80211_CHAN_DYN |
 					    IEEE80211_CHAN_2GHZ);
 
-		rtfixed->antsignal = status->ssi;
+		rtfixed->antsignal = status->signal;
 		rthdr->it_len = cpu_to_le16(rtap_len);
 	}
 
@@ -719,8 +719,8 @@ ieee80211_rx_h_sta_process(struct ieee80211_rx_data *rx)
 
 	sta->rx_fragments++;
 	sta->rx_bytes += rx->skb->len;
-	sta->last_rssi = rx->status->ssi;
 	sta->last_signal = rx->status->signal;
+	sta->last_qual = rx->status->qual;
 	sta->last_noise = rx->status->noise;
 
 	if (!(rx->fc & IEEE80211_FCTL_MOREFRAGS)) {
diff --git a/net/mac80211/sta_info.h b/net/mac80211/sta_info.h
index 97a61c39ad90..e89cc1655547 100644
--- a/net/mac80211/sta_info.h
+++ b/net/mac80211/sta_info.h
@@ -216,8 +216,8 @@ struct sta_info {
 				       * from this STA */
 	unsigned long rx_fragments; /* number of received MPDUs */
 	unsigned long rx_dropped; /* number of dropped MPDUs from this STA */
-	int last_rssi; /* RSSI of last received frame from this STA */
 	int last_signal; /* signal of last received frame from this STA */
+	int last_qual;  /* qual of last received frame from this STA */
 	int last_noise; /* noise of last received frame from this STA */
 	/* last received seq/frag number from this STA (per RX queue) */
 	__le16 last_seq_ctrl[NUM_RX_DATA_QUEUES];
diff --git a/net/mac80211/wext.c b/net/mac80211/wext.c
index 76e1de1dc735..6a342a9a40cd 100644
--- a/net/mac80211/wext.c
+++ b/net/mac80211/wext.c
@@ -169,14 +169,26 @@ static int ieee80211_ioctl_giwrange(struct net_device *dev,
 	range->num_encoding_sizes = 2;
 	range->max_encoding_tokens = NUM_DEFAULT_KEYS;
 
-	range->max_qual.qual = local->hw.max_signal;
-	range->max_qual.level = local->hw.max_rssi;
-	range->max_qual.noise = local->hw.max_noise;
+	if (local->hw.flags & IEEE80211_HW_SIGNAL_UNSPEC ||
+	    local->hw.flags & IEEE80211_HW_SIGNAL_DB)
+		range->max_qual.level = local->hw.max_signal;
+	else if  (local->hw.flags & IEEE80211_HW_SIGNAL_DBM)
+		range->max_qual.level = -110;
+	else
+		range->max_qual.level = 0;
+
+	if (local->hw.flags & IEEE80211_HW_NOISE_DBM)
+		range->max_qual.noise = -110;
+	else
+		range->max_qual.noise = 0;
+
+	range->max_qual.qual = 100;
 	range->max_qual.updated = local->wstats_flags;
 
-	range->avg_qual.qual = local->hw.max_signal/2;
-	range->avg_qual.level = 0;
-	range->avg_qual.noise = 0;
+	range->avg_qual.qual = 50;
+	/* not always true but better than nothing */
+	range->avg_qual.level = range->max_qual.level / 2;
+	range->avg_qual.noise = range->max_qual.noise / 2;
 	range->avg_qual.updated = local->wstats_flags;
 
 	range->enc_capa = IW_ENC_CAPA_WPA | IW_ENC_CAPA_WPA2 |
@@ -996,8 +1008,8 @@ static struct iw_statistics *ieee80211_get_wireless_stats(struct net_device *dev
 		wstats->qual.noise = 0;
 		wstats->qual.updated = IW_QUAL_ALL_INVALID;
 	} else {
-		wstats->qual.level = sta->last_rssi;
-		wstats->qual.qual = sta->last_signal;
+		wstats->qual.level = sta->last_signal;
+		wstats->qual.qual = sta->last_qual;
 		wstats->qual.noise = sta->last_noise;
 		wstats->qual.updated = local->wstats_flags;
 	}
-- 
cgit v1.2.3


From d62c612ef8a66be534a3ada598cfa28d40cd0b3c Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Mon, 19 May 2008 13:45:33 -0700
Subject: netns: Introduce sysctl root for read-only net sysctls.

This one stores all ctl-heads in one list and restricts the
permissions not give write access to non-init net namespaces.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/net_namespace.h |  3 +++
 net/sysctl_net.c            | 30 ++++++++++++++++++++++++++++++
 2 files changed, 33 insertions(+)

(limited to 'include')

diff --git a/include/net/net_namespace.h b/include/net/net_namespace.h
index aa540e6be502..8df751b3be55 100644
--- a/include/net/net_namespace.h
+++ b/include/net/net_namespace.h
@@ -201,8 +201,11 @@ extern void unregister_pernet_gen_device(int id, struct pernet_operations *);
 struct ctl_path;
 struct ctl_table;
 struct ctl_table_header;
+
 extern struct ctl_table_header *register_net_sysctl_table(struct net *net,
 	const struct ctl_path *path, struct ctl_table *table);
+extern struct ctl_table_header *register_net_sysctl_rotable(
+	const struct ctl_path *path, struct ctl_table *table);
 extern void unregister_net_sysctl_table(struct ctl_table_header *header);
 
 #endif /* __NET_NET_NAMESPACE_H */
diff --git a/net/sysctl_net.c b/net/sysctl_net.c
index b4f0525f91af..d8e79162724c 100644
--- a/net/sysctl_net.c
+++ b/net/sysctl_net.c
@@ -40,6 +40,27 @@ static struct ctl_table_root net_sysctl_root = {
 	.lookup = net_ctl_header_lookup,
 };
 
+static LIST_HEAD(net_sysctl_ro_tables);
+static struct list_head *net_ctl_ro_header_lookup(struct ctl_table_root *root,
+		struct nsproxy *namespaces)
+{
+	return &net_sysctl_ro_tables;
+}
+
+static int net_ctl_ro_header_perms(struct ctl_table_root *root,
+		struct nsproxy *namespaces, struct ctl_table *table)
+{
+	if (namespaces->net_ns == &init_net)
+		return table->mode;
+	else
+		return table->mode & ~0222;
+}
+
+static struct ctl_table_root net_sysctl_ro_root = {
+	.lookup = net_ctl_ro_header_lookup,
+	.permissions = net_ctl_ro_header_perms,
+};
+
 static int sysctl_net_init(struct net *net)
 {
 	INIT_LIST_HEAD(&net->sysctl_table_headers);
@@ -64,6 +85,7 @@ static __init int sysctl_init(void)
 	if (ret)
 		goto out;
 	register_sysctl_root(&net_sysctl_root);
+	register_sysctl_root(&net_sysctl_ro_root);
 out:
 	return ret;
 }
@@ -80,6 +102,14 @@ struct ctl_table_header *register_net_sysctl_table(struct net *net,
 }
 EXPORT_SYMBOL_GPL(register_net_sysctl_table);
 
+struct ctl_table_header *register_net_sysctl_rotable(const
+		struct ctl_path *path, struct ctl_table *table)
+{
+	return __register_sysctl_paths(&net_sysctl_ro_root,
+			&init_nsproxy, path, table);
+}
+EXPORT_SYMBOL_GPL(register_net_sysctl_rotable);
+
 void unregister_net_sysctl_table(struct ctl_table_header *header)
 {
 	unregister_sysctl_table(header);
-- 
cgit v1.2.3


From f56dd017c32e759bbdc88ee59d3f76235d720788 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Wed, 21 May 2008 14:16:36 -0700
Subject: tunnels: Remove stat member from ip_tunnel struct.

All users already use on-device statistics, so this field can be
safely removed.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ipip.h | 1 -
 1 file changed, 1 deletion(-)

(limited to 'include')

diff --git a/include/net/ipip.h b/include/net/ipip.h
index 633ed4def8e3..a85bda64b852 100644
--- a/include/net/ipip.h
+++ b/include/net/ipip.h
@@ -11,7 +11,6 @@ struct ip_tunnel
 {
 	struct ip_tunnel	*next;
 	struct net_device	*dev;
-	struct net_device_stats	stat;
 
 	int			recursion;	/* Depth of hard_start_xmit recursion */
 	int			err_count;	/* Number of arrived ICMP errors */
-- 
cgit v1.2.3


From 3dca02af38c11a970160387ab36ae6043feb03cd Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Wed, 21 May 2008 14:17:05 -0700
Subject: ip6tnl: Use on-device stats instead of private ones.

This tunnel uses its own private structure and requires separate
patch to switch from private stats to on-device ones.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip6_tunnel.h |  1 -
 net/ipv6/ip6_tunnel.c    | 26 ++++++--------------------
 2 files changed, 6 insertions(+), 21 deletions(-)

(limited to 'include')

diff --git a/include/net/ip6_tunnel.h b/include/net/ip6_tunnel.h
index 6512d85f11b3..3780592ebe8e 100644
--- a/include/net/ip6_tunnel.h
+++ b/include/net/ip6_tunnel.h
@@ -19,7 +19,6 @@
 struct ip6_tnl {
 	struct ip6_tnl *next;	/* next tunnel in list */
 	struct net_device *dev;	/* virtual device associated with tunnel */
-	struct net_device_stats stat;	/* statistics for tunnel device */
 	int recursion;		/* depth of hard_start_xmit recursion */
 	struct ip6_tnl_parm parms;	/* tunnel configuration parameters */
 	struct flowi fl;	/* flowi template for xmit */
diff --git a/net/ipv6/ip6_tunnel.c b/net/ipv6/ip6_tunnel.c
index 2bda3ba100b1..37814810ac49 100644
--- a/net/ipv6/ip6_tunnel.c
+++ b/net/ipv6/ip6_tunnel.c
@@ -711,7 +711,7 @@ static int ip6_tnl_rcv(struct sk_buff *skb, __u16 protocol,
 		}
 
 		if (!ip6_tnl_rcv_ctl(t)) {
-			t->stat.rx_dropped++;
+			t->dev->stats.rx_dropped++;
 			read_unlock(&ip6_tnl_lock);
 			goto discard;
 		}
@@ -728,8 +728,8 @@ static int ip6_tnl_rcv(struct sk_buff *skb, __u16 protocol,
 
 		dscp_ecn_decapsulate(t, ipv6h, skb);
 
-		t->stat.rx_packets++;
-		t->stat.rx_bytes += skb->len;
+		t->dev->stats.rx_packets++;
+		t->dev->stats.rx_bytes += skb->len;
 		netif_rx(skb);
 		read_unlock(&ip6_tnl_lock);
 		return 0;
@@ -849,7 +849,7 @@ static int ip6_tnl_xmit2(struct sk_buff *skb,
 			 __u32 *pmtu)
 {
 	struct ip6_tnl *t = netdev_priv(dev);
-	struct net_device_stats *stats = &t->stat;
+	struct net_device_stats *stats = &t->dev->stats;
 	struct ipv6hdr *ipv6h = ipv6_hdr(skb);
 	struct ipv6_tel_txoption opt;
 	struct dst_entry *dst;
@@ -1043,11 +1043,11 @@ static int
 ip6_tnl_xmit(struct sk_buff *skb, struct net_device *dev)
 {
 	struct ip6_tnl *t = netdev_priv(dev);
-	struct net_device_stats *stats = &t->stat;
+	struct net_device_stats *stats = &t->dev->stats;
 	int ret;
 
 	if (t->recursion++) {
-		t->stat.collisions++;
+		stats->collisions++;
 		goto tx_err;
 	}
 
@@ -1288,19 +1288,6 @@ ip6_tnl_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
 	return err;
 }
 
-/**
- * ip6_tnl_get_stats - return the stats for tunnel device
- *   @dev: virtual device associated with tunnel
- *
- * Return: stats for device
- **/
-
-static struct net_device_stats *
-ip6_tnl_get_stats(struct net_device *dev)
-{
-	return &(((struct ip6_tnl *)netdev_priv(dev))->stat);
-}
-
 /**
  * ip6_tnl_change_mtu - change mtu manually for tunnel device
  *   @dev: virtual device associated with tunnel
@@ -1334,7 +1321,6 @@ static void ip6_tnl_dev_setup(struct net_device *dev)
 	dev->uninit = ip6_tnl_dev_uninit;
 	dev->destructor = free_netdev;
 	dev->hard_start_xmit = ip6_tnl_xmit;
-	dev->get_stats = ip6_tnl_get_stats;
 	dev->do_ioctl = ip6_tnl_ioctl;
 	dev->change_mtu = ip6_tnl_change_mtu;
 
-- 
cgit v1.2.3


From 59f0c4523fdea865fab7d69d878269992a9d08dd Mon Sep 17 00:00:00 2001
From: Allan Stephens <allan.stephens@windriver.com>
Date: Wed, 21 May 2008 14:52:30 -0700
Subject: tipc: Fix skb_under_panic when configuring TIPC without privileges

This patch prevents a TIPC configuration command requiring network
administrator privileges from triggering an skbuff underrun if it
is issued by a process lacking those privileges.  The revised error
handling code avoids the use of a potentially uninitialized global
variable by transforming the unauthorized command into a new command,
then following the standard command processing path to generate the
required error message.

Signed-off-by: Allan Stephens <allan.stephens@windriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/tipc_config.h | 10 +++++++++-
 net/tipc/config.c           |  6 +++++-
 net/tipc/netlink.c          | 16 +++++++++-------
 3 files changed, 23 insertions(+), 9 deletions(-)

(limited to 'include')

diff --git a/include/linux/tipc_config.h b/include/linux/tipc_config.h
index b0c916d1f375..2bc6fa4adeb5 100644
--- a/include/linux/tipc_config.h
+++ b/include/linux/tipc_config.h
@@ -2,7 +2,7 @@
  * include/linux/tipc_config.h: Include file for TIPC configuration interface
  * 
  * Copyright (c) 2003-2006, Ericsson AB
- * Copyright (c) 2005, Wind River Systems
+ * Copyright (c) 2005-2007, Wind River Systems
  * All rights reserved.
  *
  * Redistribution and use in source and binary forms, with or without
@@ -135,6 +135,14 @@
 #define  TIPC_CMD_SET_MAX_SLAVES    0x800A    /* tx unsigned, rx none */
 #define  TIPC_CMD_SET_NETID         0x800B    /* tx unsigned, rx none */
 
+/*
+ * Reserved commands:
+ * May not be issued by any process.
+ * Used internally by TIPC.
+ */
+
+#define  TIPC_CMD_NOT_NET_ADMIN     0xC001    /* tx none, rx none */
+
 /*
  * TLV types defined for TIPC
  */
diff --git a/net/tipc/config.c b/net/tipc/config.c
index 91d56f8fee9f..16e7cb74969b 100644
--- a/net/tipc/config.c
+++ b/net/tipc/config.c
@@ -2,7 +2,7 @@
  * net/tipc/config.c: TIPC configuration management code
  *
  * Copyright (c) 2002-2006, Ericsson AB
- * Copyright (c) 2004-2006, Wind River Systems
+ * Copyright (c) 2004-2007, Wind River Systems
  * All rights reserved.
  *
  * Redistribution and use in source and binary forms, with or without
@@ -602,6 +602,10 @@ struct sk_buff *tipc_cfg_do_cmd(u32 orig_node, u16 cmd, const void *request_area
 	case TIPC_CMD_GET_NETID:
 		rep_tlv_buf = tipc_cfg_reply_unsigned(tipc_net_id);
 		break;
+	case TIPC_CMD_NOT_NET_ADMIN:
+		rep_tlv_buf =
+			tipc_cfg_reply_error_string(TIPC_CFG_NOT_NET_ADMIN);
+		break;
 	default:
 		rep_tlv_buf = tipc_cfg_reply_error_string(TIPC_CFG_NOT_SUPPORTED
 							  " (unknown command)");
diff --git a/net/tipc/netlink.c b/net/tipc/netlink.c
index 6a7f7b4c2595..c387217bb230 100644
--- a/net/tipc/netlink.c
+++ b/net/tipc/netlink.c
@@ -2,7 +2,7 @@
  * net/tipc/netlink.c: TIPC configuration handling
  *
  * Copyright (c) 2005-2006, Ericsson AB
- * Copyright (c) 2005, Wind River Systems
+ * Copyright (c) 2005-2007, Wind River Systems
  * All rights reserved.
  *
  * Redistribution and use in source and binary forms, with or without
@@ -45,15 +45,17 @@ static int handle_cmd(struct sk_buff *skb, struct genl_info *info)
 	struct nlmsghdr *req_nlh = info->nlhdr;
 	struct tipc_genlmsghdr *req_userhdr = info->userhdr;
 	int hdr_space = NLMSG_SPACE(GENL_HDRLEN + TIPC_GENL_HDRLEN);
+	u16 cmd;
 
 	if ((req_userhdr->cmd & 0xC000) && (!capable(CAP_NET_ADMIN)))
-		rep_buf = tipc_cfg_reply_error_string(TIPC_CFG_NOT_NET_ADMIN);
+		cmd = TIPC_CMD_NOT_NET_ADMIN;
 	else
-		rep_buf = tipc_cfg_do_cmd(req_userhdr->dest,
-					  req_userhdr->cmd,
-					  NLMSG_DATA(req_nlh) + GENL_HDRLEN + TIPC_GENL_HDRLEN,
-					  NLMSG_PAYLOAD(req_nlh, GENL_HDRLEN + TIPC_GENL_HDRLEN),
-					  hdr_space);
+		cmd = req_userhdr->cmd;
+
+	rep_buf = tipc_cfg_do_cmd(req_userhdr->dest, cmd,
+			NLMSG_DATA(req_nlh) + GENL_HDRLEN + TIPC_GENL_HDRLEN,
+			NLMSG_PAYLOAD(req_nlh, GENL_HDRLEN + TIPC_GENL_HDRLEN),
+			hdr_space);
 
 	if (rep_buf) {
 		skb_push(rep_buf, hdr_space);
-- 
cgit v1.2.3


From edcdf8b21ac920e06b4180246123fe43b022e020 Mon Sep 17 00:00:00 2001
From: Ron Rindjunsky <ron.rindjunsky@intel.com>
Date: Thu, 15 May 2008 13:53:55 +0800
Subject: mac80211: separate Tx and Rx MCS when configuring HT

This patch follows the 11n spec in separation between Tx and Rx MCS
capabilities. Up until now, when configuring the HT possible set of Tx
MCS only Rx MCS were considered, assuming they are the same as the Tx MCS.
This patch fixed this by looking at low level driver Tx capabilities.

Signed-off-by: Ron Rindjunsky <ron.rindjunsky@intel.com>
Signed-off-by: Tomas Winkler <tomas.winkler@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/net/wireless/iwlwifi/iwl4965-base.c | 14 +++--
 include/linux/ieee80211.h                   | 18 ++++--
 net/mac80211/main.c                         | 96 ++++++++++++++++++-----------
 3 files changed, 83 insertions(+), 45 deletions(-)

(limited to 'include')

diff --git a/drivers/net/wireless/iwlwifi/iwl4965-base.c b/drivers/net/wireless/iwlwifi/iwl4965-base.c
index 1a2d3768867d..c713b27ec653 100644
--- a/drivers/net/wireless/iwlwifi/iwl4965-base.c
+++ b/drivers/net/wireless/iwlwifi/iwl4965-base.c
@@ -3958,6 +3958,13 @@ static void iwl4965_dealloc_ucode_pci(struct iwl_priv *priv)
 	iwl_free_fw_desc(priv->pci_dev, &priv->ucode_boot);
 }
 
+static void iwl4965_nic_start(struct iwl_priv *priv)
+{
+	/* Remove all resets to allow NIC to operate */
+	iwl_write32(priv, CSR_RESET, 0);
+}
+
+
 /**
  * iwl4965_read_ucode - Read uCode images from disk file.
  *
@@ -4447,8 +4454,7 @@ static int __iwl4965_up(struct iwl_priv *priv)
 		}
 
 		/* start card; "initialize" will load runtime ucode */
-		/* Remove all resets to allow NIC to operate */
-		iwl_write32(priv, CSR_RESET, 0);
+		iwl4965_nic_start(priv);
 
 		IWL_DEBUG_INFO(DRV_NAME " is coming up\n");
 
@@ -6842,10 +6848,6 @@ static int __init iwl4965_init(void)
 
 	return ret;
 
-
-#ifdef CONFIG_IWLWIFI_DEBUG
-	pci_unregister_driver(&iwl_driver);
-#endif
 error_register:
 	iwl4965_rate_control_unregister();
 	return ret;
diff --git a/include/linux/ieee80211.h b/include/linux/ieee80211.h
index a9102bc78b61..3c2ac0c37aa8 100644
--- a/include/linux/ieee80211.h
+++ b/include/linux/ieee80211.h
@@ -306,8 +306,18 @@ struct ieee80211_ht_addt_info {
 #define IEEE80211_HT_CAP_SGI_40			0x0040
 #define IEEE80211_HT_CAP_DELAY_BA		0x0400
 #define IEEE80211_HT_CAP_MAX_AMSDU		0x0800
+/* 802.11n HT capability AMPDU settings */
 #define IEEE80211_HT_CAP_AMPDU_FACTOR		0x03
 #define IEEE80211_HT_CAP_AMPDU_DENSITY		0x1C
+/* 802.11n HT capability MSC set */
+#define IEEE80211_SUPP_MCS_SET_UEQM		4
+#define IEEE80211_HT_CAP_MAX_STREAMS		4
+#define IEEE80211_SUPP_MCS_SET_LEN		10
+/* maximum streams the spec allows */
+#define IEEE80211_HT_CAP_MCS_TX_DEFINED		0x01
+#define IEEE80211_HT_CAP_MCS_TX_RX_DIFF		0x02
+#define IEEE80211_HT_CAP_MCS_TX_STREAMS		0x0C
+#define IEEE80211_HT_CAP_MCS_TX_UEQM		0x10
 /* 802.11n HT IE masks */
 #define IEEE80211_HT_IE_CHA_SEC_OFFSET		0x03
 #define IEEE80211_HT_IE_CHA_WIDTH		0x04
@@ -316,10 +326,10 @@ struct ieee80211_ht_addt_info {
 #define IEEE80211_HT_IE_NON_HT_STA_PRSNT	0x0010
 
 /* MIMO Power Save Modes */
-#define WLAN_HT_CAP_MIMO_PS_STATIC         0
-#define WLAN_HT_CAP_MIMO_PS_DYNAMIC        1
-#define WLAN_HT_CAP_MIMO_PS_INVALID        2
-#define WLAN_HT_CAP_MIMO_PS_DISABLED       3
+#define WLAN_HT_CAP_MIMO_PS_STATIC	0
+#define WLAN_HT_CAP_MIMO_PS_DYNAMIC	1
+#define WLAN_HT_CAP_MIMO_PS_INVALID	2
+#define WLAN_HT_CAP_MIMO_PS_DISABLED	3
 
 /* Authentication algorithms */
 #define WLAN_AUTH_OPEN 0
diff --git a/net/mac80211/main.c b/net/mac80211/main.c
index 36016363d225..b0fddb7de549 100644
--- a/net/mac80211/main.c
+++ b/net/mac80211/main.c
@@ -35,8 +35,6 @@
 #include "debugfs.h"
 #include "debugfs_netdev.h"
 
-#define SUPP_MCS_SET_LEN 16
-
 /*
  * For seeing transmitted packets on monitor interfaces
  * we have a radiotap header too.
@@ -1068,56 +1066,84 @@ u32 ieee80211_handle_ht(struct ieee80211_local *local, int enable_ht,
 	struct ieee80211_supported_band *sband;
 	struct ieee80211_ht_info ht_conf;
 	struct ieee80211_ht_bss_info ht_bss_conf;
-	int i;
 	u32 changed = 0;
+	int i;
+	u8 max_tx_streams = IEEE80211_HT_CAP_MAX_STREAMS;
+	u8 tx_mcs_set_cap;
 
 	sband = local->hw.wiphy->bands[conf->channel->band];
 
+	memset(&ht_conf, 0, sizeof(struct ieee80211_ht_info));
+	memset(&ht_bss_conf, 0, sizeof(struct ieee80211_ht_bss_info));
+
 	/* HT is not supported */
 	if (!sband->ht_info.ht_supported) {
 		conf->flags &= ~IEEE80211_CONF_SUPPORT_HT_MODE;
-		return 0;
+		goto out;
 	}
 
-	memset(&ht_conf, 0, sizeof(struct ieee80211_ht_info));
-	memset(&ht_bss_conf, 0, sizeof(struct ieee80211_ht_bss_info));
-
-	if (enable_ht) {
-		if (!(conf->flags & IEEE80211_CONF_SUPPORT_HT_MODE))
+	/* disable HT */
+	if (!enable_ht) {
+		if (conf->flags & IEEE80211_CONF_SUPPORT_HT_MODE)
 			changed |= BSS_CHANGED_HT;
+		conf->flags &= ~IEEE80211_CONF_SUPPORT_HT_MODE;
+		conf->ht_conf.ht_supported = 0;
+		goto out;
+	}
 
-		conf->flags |= IEEE80211_CONF_SUPPORT_HT_MODE;
-		ht_conf.ht_supported = 1;
 
-		ht_conf.cap = req_ht_cap->cap & sband->ht_info.cap;
-		ht_conf.cap &= ~(IEEE80211_HT_CAP_MIMO_PS);
-		ht_conf.cap |= sband->ht_info.cap & IEEE80211_HT_CAP_MIMO_PS;
+	if (!(conf->flags & IEEE80211_CONF_SUPPORT_HT_MODE))
+		changed |= BSS_CHANGED_HT;
 
-		for (i = 0; i < SUPP_MCS_SET_LEN; i++)
-			ht_conf.supp_mcs_set[i] =
-					sband->ht_info.supp_mcs_set[i] &
-					req_ht_cap->supp_mcs_set[i];
+	conf->flags |= IEEE80211_CONF_SUPPORT_HT_MODE;
+	ht_conf.ht_supported = 1;
 
-		ht_bss_conf.primary_channel = req_bss_cap->primary_channel;
-		ht_bss_conf.bss_cap = req_bss_cap->bss_cap;
-		ht_bss_conf.bss_op_mode = req_bss_cap->bss_op_mode;
+	ht_conf.cap = req_ht_cap->cap & sband->ht_info.cap;
+	ht_conf.cap &= ~(IEEE80211_HT_CAP_MIMO_PS);
+	ht_conf.cap |= sband->ht_info.cap & IEEE80211_HT_CAP_MIMO_PS;
+	ht_bss_conf.primary_channel = req_bss_cap->primary_channel;
+	ht_bss_conf.bss_cap = req_bss_cap->bss_cap;
+	ht_bss_conf.bss_op_mode = req_bss_cap->bss_op_mode;
 
-		ht_conf.ampdu_factor = req_ht_cap->ampdu_factor;
-		ht_conf.ampdu_density = req_ht_cap->ampdu_density;
+	ht_conf.ampdu_factor = req_ht_cap->ampdu_factor;
+	ht_conf.ampdu_density = req_ht_cap->ampdu_density;
 
-		/* if bss configuration changed store the new one */
-		if (memcmp(&conf->ht_conf, &ht_conf, sizeof(ht_conf)) ||
-		    memcmp(&conf->ht_bss_conf, &ht_bss_conf, sizeof(ht_bss_conf))) {
-			changed |= BSS_CHANGED_HT;
-			memcpy(&conf->ht_conf, &ht_conf, sizeof(ht_conf));
-			memcpy(&conf->ht_bss_conf, &ht_bss_conf, sizeof(ht_bss_conf));
-		}
-	} else {
-		if (conf->flags & IEEE80211_CONF_SUPPORT_HT_MODE)
-			changed |= BSS_CHANGED_HT;
-		conf->flags &= ~IEEE80211_CONF_SUPPORT_HT_MODE;
-	}
+	/* Bits 96-100 */
+	tx_mcs_set_cap = sband->ht_info.supp_mcs_set[12];
 
+	/* configure suppoerted Tx MCS according to requested MCS
+	 * (based in most cases on Rx capabilities of peer) and self
+	 * Tx MCS capabilities (as defined by low level driver HW
+	 * Tx capabilities) */
+	if (!(tx_mcs_set_cap & IEEE80211_HT_CAP_MCS_TX_DEFINED))
+		goto check_changed;
+
+	/* Counting from 0 therfore + 1 */
+	if (tx_mcs_set_cap & IEEE80211_HT_CAP_MCS_TX_RX_DIFF)
+		max_tx_streams = ((tx_mcs_set_cap &
+				IEEE80211_HT_CAP_MCS_TX_STREAMS) >> 2) + 1;
+
+	for (i = 0; i < max_tx_streams; i++)
+		ht_conf.supp_mcs_set[i] =
+			sband->ht_info.supp_mcs_set[i] &
+					req_ht_cap->supp_mcs_set[i];
+
+	if (tx_mcs_set_cap & IEEE80211_HT_CAP_MCS_TX_UEQM)
+		for (i = IEEE80211_SUPP_MCS_SET_UEQM;
+		     i < IEEE80211_SUPP_MCS_SET_LEN; i++)
+			ht_conf.supp_mcs_set[i] =
+				sband->ht_info.supp_mcs_set[i] &
+					req_ht_cap->supp_mcs_set[i];
+
+check_changed:
+	/* if bss configuration changed store the new one */
+	if (memcmp(&conf->ht_conf, &ht_conf, sizeof(ht_conf)) ||
+	    memcmp(&conf->ht_bss_conf, &ht_bss_conf, sizeof(ht_bss_conf))) {
+		changed |= BSS_CHANGED_HT;
+		memcpy(&conf->ht_conf, &ht_conf, sizeof(ht_conf));
+		memcpy(&conf->ht_bss_conf, &ht_bss_conf, sizeof(ht_bss_conf));
+	}
+out:
 	return changed;
 }
 
-- 
cgit v1.2.3


From 36d6825b91bc492b65b6333c369cd96a2fc8c903 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes@sipsolutions.net>
Date: Thu, 15 May 2008 12:55:26 +0200
Subject: mac80211: let drivers wake but not start queues

Having drivers start queues is just confusing, their ->start()
callback can block and do whatever is necessary, so let mac80211
start queues and have drivers wake queues when necessary (to get
packets flowing again right away.)

Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
Acked-by: David S. Miller <davem@davemloft.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/net/wireless/adm8211.c              |  2 +-
 drivers/net/wireless/ath5k/base.c           |  2 +-
 drivers/net/wireless/b43/main.c             |  1 -
 drivers/net/wireless/b43legacy/main.c       |  1 -
 drivers/net/wireless/iwlwifi/iwl3945-base.c |  2 +-
 drivers/net/wireless/iwlwifi/iwl4965-base.c |  2 +-
 drivers/net/wireless/p54/p54common.c        |  3 ---
 drivers/net/wireless/p54/p54pci.c           |  2 +-
 drivers/net/wireless/rt2x00/rt2x00dev.c     |  4 ++--
 include/net/mac80211.h                      |  8 --------
 net/mac80211/main.c                         |  8 +++++++-
 net/mac80211/util.c                         | 12 ------------
 12 files changed, 14 insertions(+), 33 deletions(-)

(limited to 'include')

diff --git a/drivers/net/wireless/adm8211.c b/drivers/net/wireless/adm8211.c
index 7af5d8851f67..79dfca546c89 100644
--- a/drivers/net/wireless/adm8211.c
+++ b/drivers/net/wireless/adm8211.c
@@ -2015,7 +2015,7 @@ static int adm8211_resume(struct pci_dev *pdev)
 
 	if (priv->mode != IEEE80211_IF_TYPE_INVALID) {
 		adm8211_start(dev);
-		ieee80211_start_queues(dev);
+		ieee80211_wake_queues(dev);
 	}
 
 	return 0;
diff --git a/drivers/net/wireless/ath5k/base.c b/drivers/net/wireless/ath5k/base.c
index c76ada178781..3f16ad66bdb5 100644
--- a/drivers/net/wireless/ath5k/base.c
+++ b/drivers/net/wireless/ath5k/base.c
@@ -1599,7 +1599,7 @@ ath5k_txq_cleanup(struct ath5k_softc *sc)
 					sc->txqs[i].link);
 			}
 	}
-	ieee80211_start_queues(sc->hw); /* XXX move to callers */
+	ieee80211_wake_queues(sc->hw); /* XXX move to callers */
 
 	for (i = 0; i < ARRAY_SIZE(sc->txqs); i++)
 		if (sc->txqs[i].setup)
diff --git a/drivers/net/wireless/b43/main.c b/drivers/net/wireless/b43/main.c
index fc23ba5309bd..9445a604a966 100644
--- a/drivers/net/wireless/b43/main.c
+++ b/drivers/net/wireless/b43/main.c
@@ -3497,7 +3497,6 @@ static int b43_wireless_core_start(struct b43_wldev *dev)
 	/* Start data flow (TX/RX). */
 	b43_mac_enable(dev);
 	b43_interrupt_enable(dev, dev->irq_savedstate);
-	ieee80211_start_queues(dev->wl->hw);
 
 	/* Start maintainance work */
 	b43_periodic_tasks_setup(dev);
diff --git a/drivers/net/wireless/b43legacy/main.c b/drivers/net/wireless/b43legacy/main.c
index 7755c59e0803..b05a507ed44d 100644
--- a/drivers/net/wireless/b43legacy/main.c
+++ b/drivers/net/wireless/b43legacy/main.c
@@ -2794,7 +2794,6 @@ static int b43legacy_wireless_core_start(struct b43legacy_wldev *dev)
 	/* Start data flow (TX/RX) */
 	b43legacy_mac_enable(dev);
 	b43legacy_interrupt_enable(dev, dev->irq_savedstate);
-	ieee80211_start_queues(dev->wl->hw);
 
 	/* Start maintenance work */
 	b43legacy_periodic_tasks_setup(dev);
diff --git a/drivers/net/wireless/iwlwifi/iwl3945-base.c b/drivers/net/wireless/iwlwifi/iwl3945-base.c
index b8fb8d8d95ff..54cde8a7b5fa 100644
--- a/drivers/net/wireless/iwlwifi/iwl3945-base.c
+++ b/drivers/net/wireless/iwlwifi/iwl3945-base.c
@@ -5823,7 +5823,7 @@ static void iwl3945_alive_start(struct iwl3945_priv *priv)
 	if (iwl3945_is_rfkill(priv))
 		return;
 
-	ieee80211_start_queues(priv->hw);
+	ieee80211_wake_queues(priv->hw);
 
 	priv->active_rate = priv->rates_mask;
 	priv->active_rate_basic = priv->rates_mask & IWL_BASIC_RATES_MASK;
diff --git a/drivers/net/wireless/iwlwifi/iwl4965-base.c b/drivers/net/wireless/iwlwifi/iwl4965-base.c
index 48c59cbefb4a..db4f606bad50 100644
--- a/drivers/net/wireless/iwlwifi/iwl4965-base.c
+++ b/drivers/net/wireless/iwlwifi/iwl4965-base.c
@@ -3367,7 +3367,7 @@ static void iwl4965_alive_start(struct iwl_priv *priv)
 	if (iwl_is_rfkill(priv))
 		return;
 
-	ieee80211_start_queues(priv->hw);
+	ieee80211_wake_queues(priv->hw);
 
 	priv->active_rate = priv->rates_mask;
 	priv->active_rate_basic = priv->rates_mask & IWL_BASIC_RATES_MASK;
diff --git a/drivers/net/wireless/p54/p54common.c b/drivers/net/wireless/p54/p54common.c
index 9cbef5bce0f6..3d35fe6a8f5f 100644
--- a/drivers/net/wireless/p54/p54common.c
+++ b/drivers/net/wireless/p54/p54common.c
@@ -375,9 +375,6 @@ static void inline p54_wake_free_queues(struct ieee80211_hw *dev)
 	struct p54_common *priv = dev->priv;
 	int i;
 
-	/* ieee80211_start_queues is great if all queues are really empty.
-	 * But, what if some are full? */
-
 	for (i = 0; i < dev->queues; i++)
 		if (priv->tx_stats[i].len < priv->tx_stats[i].limit)
 			ieee80211_wake_queue(dev, i);
diff --git a/drivers/net/wireless/p54/p54pci.c b/drivers/net/wireless/p54/p54pci.c
index fa527723fbe0..7dd4add4bf4e 100644
--- a/drivers/net/wireless/p54/p54pci.c
+++ b/drivers/net/wireless/p54/p54pci.c
@@ -665,7 +665,7 @@ static int p54p_resume(struct pci_dev *pdev)
 
 	if (priv->common.mode != IEEE80211_IF_TYPE_INVALID) {
 		p54p_open(dev);
-		ieee80211_start_queues(dev);
+		ieee80211_wake_queues(dev);
 	}
 
 	return 0;
diff --git a/drivers/net/wireless/rt2x00/rt2x00dev.c b/drivers/net/wireless/rt2x00/rt2x00dev.c
index 171f445962db..d341764e1b24 100644
--- a/drivers/net/wireless/rt2x00/rt2x00dev.c
+++ b/drivers/net/wireless/rt2x00/rt2x00dev.c
@@ -125,7 +125,7 @@ int rt2x00lib_enable_radio(struct rt2x00_dev *rt2x00dev)
 	/*
 	 * Start the TX queues.
 	 */
-	ieee80211_start_queues(rt2x00dev->hw);
+	ieee80211_wake_queues(rt2x00dev->hw);
 
 	return 0;
 }
@@ -1186,7 +1186,7 @@ int rt2x00lib_resume(struct rt2x00_dev *rt2x00dev)
 	 * In that case we have disabled the TX queue and should
 	 * now enable it again
 	 */
-	ieee80211_start_queues(rt2x00dev->hw);
+	ieee80211_wake_queues(rt2x00dev->hw);
 
 	/*
 	 * During interface iteration we might have changed the
diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index 909956c97c44..f00fc76a7344 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -1566,14 +1566,6 @@ void ieee80211_wake_queue(struct ieee80211_hw *hw, int queue);
  */
 void ieee80211_stop_queue(struct ieee80211_hw *hw, int queue);
 
-/**
- * ieee80211_start_queues - start all queues
- * @hw: pointer to as obtained from ieee80211_alloc_hw().
- *
- * Drivers should use this function instead of netif_start_queue.
- */
-void ieee80211_start_queues(struct ieee80211_hw *hw);
-
 /**
  * ieee80211_stop_queues - stop all queues
  * @hw: pointer as obtained from ieee80211_alloc_hw().
diff --git a/net/mac80211/main.c b/net/mac80211/main.c
index b0fddb7de549..9761d9bd5a79 100644
--- a/net/mac80211/main.c
+++ b/net/mac80211/main.c
@@ -110,7 +110,13 @@ static int ieee80211_master_open(struct net_device *dev)
 			break;
 		}
 	}
-	return res;
+
+	if (res)
+		return res;
+
+	netif_start_queue(local->mdev);
+
+	return 0;
 }
 
 static int ieee80211_master_stop(struct net_device *dev)
diff --git a/net/mac80211/util.c b/net/mac80211/util.c
index 9cd07e1031af..800c15aff6e7 100644
--- a/net/mac80211/util.c
+++ b/net/mac80211/util.c
@@ -350,18 +350,6 @@ void ieee80211_stop_queue(struct ieee80211_hw *hw, int queue)
 }
 EXPORT_SYMBOL(ieee80211_stop_queue);
 
-void ieee80211_start_queues(struct ieee80211_hw *hw)
-{
-	struct ieee80211_local *local = hw_to_local(hw);
-	int i;
-
-	for (i = 0; i < hw->queues + hw->ampdu_queues; i++)
-		clear_bit(IEEE80211_LINK_STATE_XOFF, &local->state[i]);
-	if (!ieee80211_qdisc_installed(local->mdev))
-		netif_start_queue(local->mdev);
-}
-EXPORT_SYMBOL(ieee80211_start_queues);
-
 void ieee80211_stop_queues(struct ieee80211_hw *hw)
 {
 	int i;
-- 
cgit v1.2.3


From 2e92e6f2c50b4baf85cca968f0e6f1b5c0df7d39 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes@sipsolutions.net>
Date: Thu, 15 May 2008 12:55:27 +0200
Subject: mac80211: use rate index in TX control

This patch modifies struct ieee80211_tx_control to give band
info and the rate index (instead of rate pointers) to drivers.
This mostly serves to reduce the TX control structure size to
make it fit into skb->cb so that the fragmentation code can
put it there and we can think about passing it to drivers that
way in the future.

The rt2x00 driver update was done by Ivo, thanks.

Signed-off-by: Ivo van Doorn <IvDoorn@gmail.com>
Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/net/wireless/adm8211.c              |   6 +-
 drivers/net/wireless/ath5k/base.c           |   8 ++-
 drivers/net/wireless/b43/main.c             |   2 +-
 drivers/net/wireless/b43/xmit.c             |  13 ++--
 drivers/net/wireless/b43legacy/xmit.c       |   9 ++-
 drivers/net/wireless/iwlwifi/iwl-3945-rs.c  |   6 +-
 drivers/net/wireless/iwlwifi/iwl-3945.c     |   7 +-
 drivers/net/wireless/iwlwifi/iwl-4965-rs.c  |  12 ++--
 drivers/net/wireless/iwlwifi/iwl-4965.c     |  10 +--
 drivers/net/wireless/iwlwifi/iwl-tx.c       |   8 ++-
 drivers/net/wireless/iwlwifi/iwl3945-base.c |   4 +-
 drivers/net/wireless/iwlwifi/iwl4965-base.c |   2 +-
 drivers/net/wireless/p54/p54common.c        |   2 +-
 drivers/net/wireless/rt2x00/rt2x00queue.c   |   9 ++-
 drivers/net/wireless/rtl8180_dev.c          |  19 +++--
 drivers/net/wireless/rtl8187_dev.c          |  10 +--
 drivers/net/wireless/zd1211rw/zd_mac.c      |   7 +-
 include/net/mac80211.h                      |  51 ++++++++++----
 net/mac80211/ieee80211_i.h                  |   8 +--
 net/mac80211/mlme.c                         |   6 +-
 net/mac80211/rate.c                         |  12 ++--
 net/mac80211/rate.h                         |  25 +++----
 net/mac80211/rc80211_pid_algo.c             |   6 +-
 net/mac80211/tx.c                           | 104 ++++++++++++++++------------
 net/mac80211/util.c                         |  10 ++-
 25 files changed, 205 insertions(+), 151 deletions(-)

(limited to 'include')

diff --git a/drivers/net/wireless/adm8211.c b/drivers/net/wireless/adm8211.c
index 79dfca546c89..22db664a58d9 100644
--- a/drivers/net/wireless/adm8211.c
+++ b/drivers/net/wireless/adm8211.c
@@ -1693,10 +1693,10 @@ static int adm8211_tx(struct ieee80211_hw *dev, struct sk_buff *skb,
 	size_t payload_len, hdrlen;
 	int plcp, dur, len, plcp_signal, short_preamble;
 	struct ieee80211_hdr *hdr;
+	struct ieee80211_rate *txrate = ieee80211_get_tx_rate(dev, control);
 
-	short_preamble = !!(control->tx_rate->flags &
-					IEEE80211_TXCTL_SHORT_PREAMBLE);
-	plcp_signal = control->tx_rate->bitrate;
+	short_preamble = !!(txrate->flags & IEEE80211_TXCTL_SHORT_PREAMBLE);
+	plcp_signal = txrate->bitrate;
 
 	hdr = (struct ieee80211_hdr *)skb->data;
 	fc = le16_to_cpu(hdr->frame_control) & ~IEEE80211_FCTL_PROTECTED;
diff --git a/drivers/net/wireless/ath5k/base.c b/drivers/net/wireless/ath5k/base.c
index 3f16ad66bdb5..32ee351a7650 100644
--- a/drivers/net/wireless/ath5k/base.c
+++ b/drivers/net/wireless/ath5k/base.c
@@ -1323,7 +1323,8 @@ ath5k_txbuf_setup(struct ath5k_softc *sc, struct ath5k_buf *bf,
 
 	ret = ah->ah_setup_tx_desc(ah, ds, pktlen,
 		ieee80211_get_hdrlen_from_skb(skb), AR5K_PKT_TYPE_NORMAL,
-		(sc->power_level * 2), ctl->tx_rate->hw_value,
+		(sc->power_level * 2),
+		ieee80211_get_tx_rate(sc->hw, ctl)->hw_value,
 		ctl->retry_limit, keyidx, 0, flags, 0, 0);
 	if (ret)
 		goto err_unmap;
@@ -2046,7 +2047,8 @@ ath5k_beacon_setup(struct ath5k_softc *sc, struct ath5k_buf *bf,
 	ret = ah->ah_setup_tx_desc(ah, ds, skb->len,
 			ieee80211_get_hdrlen_from_skb(skb),
 			AR5K_PKT_TYPE_BEACON, (sc->power_level * 2),
-			ctl->tx_rate->hw_value, 1, AR5K_TXKEYIX_INVALID,
+			ieee80211_get_tx_rate(sc->hw, ctl)->hw_value,
+			1, AR5K_TXKEYIX_INVALID,
 			antenna, flags, 0, 0);
 	if (ret)
 		goto err_unmap;
@@ -2654,7 +2656,7 @@ ath5k_tx(struct ieee80211_hw *hw, struct sk_buff *skb,
 		memmove(skb->data, skb->data+pad, hdrlen);
 	}
 
-	sc->led_txrate = ctl->tx_rate->hw_value;
+	sc->led_txrate = ieee80211_get_tx_rate(hw, ctl)->hw_value;
 
 	spin_lock_irqsave(&sc->txbuflock, flags);
 	if (list_empty(&sc->txbuf)) {
diff --git a/drivers/net/wireless/b43/main.c b/drivers/net/wireless/b43/main.c
index 9445a604a966..e428645352b4 100644
--- a/drivers/net/wireless/b43/main.c
+++ b/drivers/net/wireless/b43/main.c
@@ -1372,7 +1372,7 @@ static void b43_write_beacon_template(struct b43_wldev *dev,
 	bcn = (const struct ieee80211_mgmt *)(dev->wl->current_beacon->data);
 	len = min((size_t) dev->wl->current_beacon->len,
 		  0x200 - sizeof(struct b43_plcp_hdr6));
-	rate = dev->wl->beacon_txctl.tx_rate->hw_value;
+	rate = ieee80211_get_tx_rate(dev->wl->hw, &dev->wl->beacon_txctl)->hw_value;
 
 	b43_write_template_common(dev, (const u8 *)bcn,
 				  len, ram_offset, shm_size_offset, rate);
diff --git a/drivers/net/wireless/b43/xmit.c b/drivers/net/wireless/b43/xmit.c
index afce9338d83a..9b682a3cf5e4 100644
--- a/drivers/net/wireless/b43/xmit.c
+++ b/drivers/net/wireless/b43/xmit.c
@@ -201,13 +201,14 @@ int b43_generate_txhdr(struct b43_wldev *dev,
 	u32 mac_ctl = 0;
 	u16 phy_ctl = 0;
 	u8 extra_ft = 0;
+	struct ieee80211_rate *txrate;
 
 	memset(txhdr, 0, sizeof(*txhdr));
 
-	WARN_ON(!txctl->tx_rate);
-	rate = txctl->tx_rate ? txctl->tx_rate->hw_value : B43_CCK_RATE_1MB;
+	txrate = ieee80211_get_tx_rate(dev->wl->hw, txctl);
+	rate = txrate ? txrate->hw_value : B43_CCK_RATE_1MB;
 	rate_ofdm = b43_is_ofdm_rate(rate);
-	fbrate = txctl->alt_retry_rate ? : txctl->tx_rate;
+	fbrate = ieee80211_get_alt_retry_rate(dev->wl->hw, txctl) ? : txrate;
 	rate_fb = fbrate->hw_value;
 	rate_fb_ofdm = b43_is_ofdm_rate(rate_fb);
 
@@ -336,9 +337,11 @@ int b43_generate_txhdr(struct b43_wldev *dev,
 		int rts_rate, rts_rate_fb;
 		int rts_rate_ofdm, rts_rate_fb_ofdm;
 		struct b43_plcp_hdr6 *plcp;
+		struct ieee80211_rate *rts_cts_rate;
 
-		WARN_ON(!txctl->rts_cts_rate);
-		rts_rate = txctl->rts_cts_rate ? txctl->rts_cts_rate->hw_value : B43_CCK_RATE_1MB;
+		rts_cts_rate = ieee80211_get_rts_cts_rate(dev->wl->hw, txctl);
+
+		rts_rate = rts_cts_rate ? rts_cts_rate->hw_value : B43_CCK_RATE_1MB;
 		rts_rate_ofdm = b43_is_ofdm_rate(rts_rate);
 		rts_rate_fb = b43_calc_fallback_rate(rts_rate);
 		rts_rate_fb_ofdm = b43_is_ofdm_rate(rts_rate_fb);
diff --git a/drivers/net/wireless/b43legacy/xmit.c b/drivers/net/wireless/b43legacy/xmit.c
index bed9e041d6c5..55dc251bf510 100644
--- a/drivers/net/wireless/b43legacy/xmit.c
+++ b/drivers/net/wireless/b43legacy/xmit.c
@@ -201,15 +201,18 @@ static int generate_txhdr_fw3(struct b43legacy_wldev *dev,
 	unsigned int plcp_fragment_len;
 	u32 mac_ctl = 0;
 	u16 phy_ctl = 0;
+	struct ieee80211_rate *tx_rate;
 
 	wlhdr = (const struct ieee80211_hdr *)fragment_data;
 	fctl = le16_to_cpu(wlhdr->frame_control);
 
 	memset(txhdr, 0, sizeof(*txhdr));
 
-	rate = txctl->tx_rate->hw_value;
+	tx_rate = ieee80211_get_tx_rate(dev->wl->hw, txctl);
+
+	rate = tx_rate->hw_value;
 	rate_ofdm = b43legacy_is_ofdm_rate(rate);
-	rate_fb = txctl->alt_retry_rate ? : txctl->tx_rate;
+	rate_fb = ieee80211_get_alt_retry_rate(dev->wl->hw, txctl) ? : tx_rate;
 	rate_fb_ofdm = b43legacy_is_ofdm_rate(rate_fb->hw_value);
 
 	txhdr->mac_frame_ctl = wlhdr->frame_control;
@@ -312,7 +315,7 @@ static int generate_txhdr_fw3(struct b43legacy_wldev *dev,
 		int rts_rate_ofdm;
 		int rts_rate_fb_ofdm;
 
-		rts_rate = txctl->rts_cts_rate->hw_value;
+		rts_rate = ieee80211_get_rts_cts_rate(dev->wl->hw, txctl)->hw_value;
 		rts_rate_ofdm = b43legacy_is_ofdm_rate(rts_rate);
 		rts_rate_fb = b43legacy_calc_fallback_rate(rts_rate);
 		rts_rate_fb_ofdm = b43legacy_is_ofdm_rate(rts_rate_fb);
diff --git a/drivers/net/wireless/iwlwifi/iwl-3945-rs.c b/drivers/net/wireless/iwlwifi/iwl-3945-rs.c
index e51eeeff6992..f3ca02fe9619 100644
--- a/drivers/net/wireless/iwlwifi/iwl-3945-rs.c
+++ b/drivers/net/wireless/iwlwifi/iwl-3945-rs.c
@@ -464,7 +464,7 @@ static void rs_tx_status(void *priv_rate,
 
 
 	retries = tx_resp->retry_count;
-	first_index = tx_resp->control.tx_rate->hw_value;
+	first_index = sband->bitrates[tx_resp->control.tx_rate_idx].hw_value;
 	if ((first_index < 0) || (first_index >= IWL_RATE_COUNT)) {
 		IWL_DEBUG_RATE("leave: Rate out of bounds: %d\n", first_index);
 		return;
@@ -669,7 +669,7 @@ static void rs_get_rate(void *priv_rate, struct net_device *dev,
 	    is_multicast_ether_addr(hdr->addr1) ||
 	    !sta || !sta->rate_ctrl_priv) {
 		IWL_DEBUG_RATE("leave: No STA priv data to update!\n");
-		sel->rate = rate_lowest(local, sband, sta);
+		sel->rate_idx = rate_lowest_index(local, sband, sta);
 		rcu_read_unlock();
 		return;
 	}
@@ -813,7 +813,7 @@ static void rs_get_rate(void *priv_rate, struct net_device *dev,
 
 	IWL_DEBUG_RATE("leave: %d\n", index);
 
-	sel->rate = &sband->bitrates[sta->txrate_idx];
+	sel->rate_idx = sta->txrate_idx;
 }
 
 static struct rate_control_ops rs_ops = {
diff --git a/drivers/net/wireless/iwlwifi/iwl-3945.c b/drivers/net/wireless/iwlwifi/iwl-3945.c
index ad4e7b74ca24..f8e691f88ab3 100644
--- a/drivers/net/wireless/iwlwifi/iwl-3945.c
+++ b/drivers/net/wireless/iwlwifi/iwl-3945.c
@@ -331,7 +331,9 @@ static void iwl3945_rx_reply_tx(struct iwl3945_priv *priv,
 			tx_resp->rate, tx_resp->failure_frame);
 
 	rate_idx = iwl3945_hwrate_to_plcp_idx(tx_resp->rate);
-	tx_status->control.tx_rate = &priv->ieee_rates[rate_idx];
+	if (tx_status->control.band == IEEE80211_BAND_5GHZ)
+		rate_idx -= IWL_FIRST_OFDM_RATE;
+	tx_status->control.tx_rate_idx = rate_idx;
 	IWL_DEBUG_TX_REPLY("Tx queue reclaim %d\n", index);
 	iwl3945_tx_queue_reclaim(priv, txq_id, index);
 
@@ -962,7 +964,8 @@ void iwl3945_hw_build_tx_cmd_rate(struct iwl3945_priv *priv,
 			      struct ieee80211_hdr *hdr, int sta_id, int tx_id)
 {
 	unsigned long flags;
-	u16 rate_index = min(ctrl->tx_rate->hw_value & 0xffff, IWL_RATE_COUNT - 1);
+	u16 hw_value = ieee80211_get_tx_rate(priv->hw, ctrl)->hw_value;
+	u16 rate_index = min(hw_value & 0xffff, IWL_RATE_COUNT - 1);
 	u16 rate_mask;
 	int rate;
 	u8 rts_retry_limit;
diff --git a/drivers/net/wireless/iwlwifi/iwl-4965-rs.c b/drivers/net/wireless/iwlwifi/iwl-4965-rs.c
index 2adc2281c77c..7993a1d83025 100644
--- a/drivers/net/wireless/iwlwifi/iwl-4965-rs.c
+++ b/drivers/net/wireless/iwlwifi/iwl-4965-rs.c
@@ -862,7 +862,7 @@ static void rs_tx_status(void *priv_rate, struct net_device *dev,
 	if (priv->band == IEEE80211_BAND_5GHZ)
 		rs_index -= IWL_FIRST_OFDM_RATE;
 
-	if ((tx_resp->control.tx_rate == NULL) ||
+	if ((tx_resp->control.tx_rate_idx < 0) ||
 	    (tbl_type.is_SGI ^
 		!!(tx_resp->control.flags & IEEE80211_TXCTL_SHORT_GI)) ||
 	    (tbl_type.is_fat ^
@@ -875,7 +875,7 @@ static void rs_tx_status(void *priv_rate, struct net_device *dev,
 	    (!!(tx_rate & RATE_MCS_GF_MSK) ^
 		!!(tx_resp->control.flags & IEEE80211_TXCTL_GREEN_FIELD)) ||
 	    (hw->wiphy->bands[priv->band]->bitrates[rs_index].bitrate !=
-		tx_resp->control.tx_rate->bitrate)) {
+	     hw->wiphy->bands[tx_resp->control.band]->bitrates[tx_resp->control.tx_rate_idx].bitrate)) {
 		IWL_DEBUG_RATE("initial rate does not match 0x%x\n", tx_rate);
 		goto out;
 	}
@@ -2154,7 +2154,7 @@ static void rs_get_rate(void *priv_rate, struct net_device *dev,
 	fc = le16_to_cpu(hdr->frame_control);
 	if (!ieee80211_is_data(fc) || is_multicast_ether_addr(hdr->addr1) ||
 	    !sta || !sta->rate_ctrl_priv) {
-		sel->rate = rate_lowest(local, sband, sta);
+		sel->rate_idx = rate_lowest_index(local, sband, sta);
 		goto out;
 	}
 
@@ -2184,11 +2184,13 @@ static void rs_get_rate(void *priv_rate, struct net_device *dev,
 
 done:
 	if ((i < 0) || (i > IWL_RATE_COUNT)) {
-		sel->rate = rate_lowest(local, sband, sta);
+		sel->rate_idx = rate_lowest_index(local, sband, sta);
 		goto out;
 	}
 
-	sel->rate = &priv->ieee_rates[i];
+	if (sband->band == IEEE80211_BAND_5GHZ)
+		i -= IWL_FIRST_OFDM_RATE;
+	sel->rate_idx = i;
 out:
 	rcu_read_unlock();
 }
diff --git a/drivers/net/wireless/iwlwifi/iwl-4965.c b/drivers/net/wireless/iwlwifi/iwl-4965.c
index f848a5b0f622..fb670b5cfebb 100644
--- a/drivers/net/wireless/iwlwifi/iwl-4965.c
+++ b/drivers/net/wireless/iwlwifi/iwl-4965.c
@@ -373,14 +373,10 @@ void iwl4965_hwrate_to_tx_control(struct iwl_priv *priv, u32 rate_n_flags,
 		control->flags |= IEEE80211_TXCTL_DUP_DATA;
 	if (rate_n_flags & RATE_MCS_SGI_MSK)
 		control->flags |= IEEE80211_TXCTL_SHORT_GI;
-	/* since iwl4965_hwrate_to_plcp_idx is band indifferent, we always use
-	 * IEEE80211_BAND_2GHZ band as it contains all the rates */
 	rate_index = iwl4965_hwrate_to_plcp_idx(rate_n_flags);
-	if (rate_index == -1)
-		control->tx_rate = NULL;
-	else
-		control->tx_rate =
-			&priv->bands[IEEE80211_BAND_2GHZ].bitrates[rate_index];
+	if (control->band == IEEE80211_BAND_5GHZ)
+		rate_index -= IWL_FIRST_OFDM_RATE;
+	control->tx_rate_idx = rate_index;
 }
 
 int iwl4965_hw_rxq_stop(struct iwl_priv *priv)
diff --git a/drivers/net/wireless/iwlwifi/iwl-tx.c b/drivers/net/wireless/iwlwifi/iwl-tx.c
index f32cddabdf64..4b5149c8c32e 100644
--- a/drivers/net/wireless/iwlwifi/iwl-tx.c
+++ b/drivers/net/wireless/iwlwifi/iwl-tx.c
@@ -578,7 +578,10 @@ static void iwl_tx_cmd_build_rate(struct iwl_priv *priv,
 	u8 data_retry_limit = 0;
 	u8 rate_plcp;
 	u16 rate_flags = 0;
-	int rate_idx = min(ctrl->tx_rate->hw_value & 0xffff, IWL_RATE_COUNT - 1);
+	int rate_idx;
+
+	rate_idx = min(ieee80211_get_tx_rate(priv->hw, ctrl)->hw_value & 0xffff,
+			IWL_RATE_COUNT - 1);
 
 	rate_plcp = iwl_rates[rate_idx].plcp;
 
@@ -723,7 +726,8 @@ int iwl_tx_skb(struct iwl_priv *priv,
 		goto drop_unlock;
 	}
 
-	if ((ctl->tx_rate->hw_value & 0xFF) == IWL_INVALID_RATE) {
+	if ((ieee80211_get_tx_rate(priv->hw, ctl)->hw_value & 0xFF) ==
+	     IWL_INVALID_RATE) {
 		IWL_ERROR("ERROR: No TX rate available.\n");
 		goto drop_unlock;
 	}
diff --git a/drivers/net/wireless/iwlwifi/iwl3945-base.c b/drivers/net/wireless/iwlwifi/iwl3945-base.c
index 54cde8a7b5fa..a28b4c9f6524 100644
--- a/drivers/net/wireless/iwlwifi/iwl3945-base.c
+++ b/drivers/net/wireless/iwlwifi/iwl3945-base.c
@@ -2581,7 +2581,7 @@ static int iwl3945_tx_skb(struct iwl3945_priv *priv,
 		goto drop_unlock;
 	}
 
-	if ((ctl->tx_rate->hw_value & 0xFF) == IWL_INVALID_RATE) {
+	if ((ieee80211_get_tx_rate(priv->hw, ctl)->hw_value & 0xFF) == IWL_INVALID_RATE) {
 		IWL_ERROR("ERROR: No TX rate available.\n");
 		goto drop_unlock;
 	}
@@ -6694,7 +6694,7 @@ static int iwl3945_mac_tx(struct ieee80211_hw *hw, struct sk_buff *skb,
 	}
 
 	IWL_DEBUG_TX("dev->xmit(%d bytes) at rate 0x%02x\n", skb->len,
-		     ctl->tx_rate->bitrate);
+		     ieee80211_get_tx_rate(hw, ctl)->bitrate);
 
 	if (iwl3945_tx_skb(priv, skb, ctl))
 		dev_kfree_skb_any(skb);
diff --git a/drivers/net/wireless/iwlwifi/iwl4965-base.c b/drivers/net/wireless/iwlwifi/iwl4965-base.c
index db4f606bad50..1fad6227aa51 100644
--- a/drivers/net/wireless/iwlwifi/iwl4965-base.c
+++ b/drivers/net/wireless/iwlwifi/iwl4965-base.c
@@ -4281,7 +4281,7 @@ static int iwl4965_mac_tx(struct ieee80211_hw *hw, struct sk_buff *skb,
 	}
 
 	IWL_DEBUG_TX("dev->xmit(%d bytes) at rate 0x%02x\n", skb->len,
-		     ctl->tx_rate->bitrate);
+		     ieee80211_get_tx_rate(hw, ctl)->bitrate);
 
 	if (iwl_tx_skb(priv, skb, ctl))
 		dev_kfree_skb_any(skb);
diff --git a/drivers/net/wireless/p54/p54common.c b/drivers/net/wireless/p54/p54common.c
index 3d35fe6a8f5f..3ca9386561ff 100644
--- a/drivers/net/wireless/p54/p54common.c
+++ b/drivers/net/wireless/p54/p54common.c
@@ -592,7 +592,7 @@ static int p54_tx(struct ieee80211_hw *dev, struct sk_buff *skb,
 	txhdr->padding2 = 0;
 
 	/* TODO: add support for alternate retry TX rates */
-	rate = control->tx_rate->hw_value;
+	rate = ieee80211_get_tx_rate(dev, control)->hw_value;
 	if (control->flags & IEEE80211_TXCTL_SHORT_PREAMBLE)
 		rate |= 0x10;
 	if (control->flags & IEEE80211_TXCTL_USE_RTS_CTS)
diff --git a/drivers/net/wireless/rt2x00/rt2x00queue.c b/drivers/net/wireless/rt2x00/rt2x00queue.c
index 19c10629c767..5cf4c2f59260 100644
--- a/drivers/net/wireless/rt2x00/rt2x00queue.c
+++ b/drivers/net/wireless/rt2x00/rt2x00queue.c
@@ -33,8 +33,10 @@ void rt2x00queue_create_tx_descriptor(struct queue_entry *entry,
 				      struct txentry_desc *txdesc,
 				      struct ieee80211_tx_control *control)
 {
+	struct rt2x00_dev *rt2x00dev = entry->queue->rt2x00dev;
 	struct ieee80211_hdr *hdr = (struct ieee80211_hdr *)entry->skb->data;
-	struct ieee80211_rate *rate = control->tx_rate;
+	struct ieee80211_rate *rate =
+	    ieee80211_get_tx_rate(rt2x00dev->hw, control);
 	const struct rt2x00_rate *hwrate;
 	unsigned int data_length;
 	unsigned int duration;
@@ -77,8 +79,9 @@ void rt2x00queue_create_tx_descriptor(struct queue_entry *entry,
 			__set_bit(ENTRY_TXD_CTS_FRAME, &txdesc->flags);
 			__clear_bit(ENTRY_TXD_ACK, &txdesc->flags);
 		}
-		if (control->rts_cts_rate)
-			rate = control->rts_cts_rate;
+		if (control->rts_cts_rate_idx >= 0)
+			rate =
+			    ieee80211_get_rts_cts_rate(rt2x00dev->hw, control);
 	}
 
 	/*
diff --git a/drivers/net/wireless/rtl8180_dev.c b/drivers/net/wireless/rtl8180_dev.c
index c220998cee65..6263209b889e 100644
--- a/drivers/net/wireless/rtl8180_dev.c
+++ b/drivers/net/wireless/rtl8180_dev.c
@@ -257,24 +257,21 @@ static int rtl8180_tx(struct ieee80211_hw *dev, struct sk_buff *skb,
 	mapping = pci_map_single(priv->pdev, skb->data,
 				 skb->len, PCI_DMA_TODEVICE);
 
-	BUG_ON(!control->tx_rate);
-
 	tx_flags = RTL8180_TX_DESC_FLAG_OWN | RTL8180_TX_DESC_FLAG_FS |
 		   RTL8180_TX_DESC_FLAG_LS |
-		   (control->tx_rate->hw_value << 24) | skb->len;
+		   (ieee80211_get_tx_rate(dev, control)->hw_value << 24) |
+		   skb->len;
 
 	if (priv->r8185)
 		tx_flags |= RTL8180_TX_DESC_FLAG_DMA |
 			    RTL8180_TX_DESC_FLAG_NO_ENC;
 
 	if (control->flags & IEEE80211_TXCTL_USE_RTS_CTS) {
-		BUG_ON(!control->rts_cts_rate);
 		tx_flags |= RTL8180_TX_DESC_FLAG_RTS;
-		tx_flags |= control->rts_cts_rate->hw_value << 19;
+		tx_flags |= ieee80211_get_rts_cts_rate(dev, control)->hw_value << 19;
 	} else if (control->flags & IEEE80211_TXCTL_USE_CTS_PROTECT) {
-		BUG_ON(!control->rts_cts_rate);
 		tx_flags |= RTL8180_TX_DESC_FLAG_CTS;
-		tx_flags |= control->rts_cts_rate->hw_value << 19;
+		tx_flags |= ieee80211_get_rts_cts_rate(dev, control)->hw_value << 19;
 	}
 
 	*((struct ieee80211_tx_control **) skb->cb) =
@@ -288,9 +285,9 @@ static int rtl8180_tx(struct ieee80211_hw *dev, struct sk_buff *skb,
 		unsigned int remainder;
 
 		plcp_len = DIV_ROUND_UP(16 * (skb->len + 4),
-					(control->tx_rate->bitrate * 2) / 10);
+				(ieee80211_get_tx_rate(dev, control)->bitrate * 2) / 10);
 		remainder = (16 * (skb->len + 4)) %
-			    ((control->tx_rate->bitrate * 2) / 10);
+			    ((ieee80211_get_tx_rate(dev, control)->bitrate * 2) / 10);
 		if (remainder > 0 && remainder <= 6)
 			plcp_len |= 1 << 15;
 	}
@@ -303,8 +300,8 @@ static int rtl8180_tx(struct ieee80211_hw *dev, struct sk_buff *skb,
 	entry->plcp_len = cpu_to_le16(plcp_len);
 	entry->tx_buf = cpu_to_le32(mapping);
 	entry->frame_len = cpu_to_le32(skb->len);
-	entry->flags2 = control->alt_retry_rate != NULL ?
-			control->alt_retry_rate->bitrate << 4 : 0;
+	entry->flags2 = control->alt_retry_rate_idx >= 0 ?
+		ieee80211_get_alt_retry_rate(dev, control)->bitrate << 4 : 0;
 	entry->retry_limit = control->retry_limit;
 	entry->flags = cpu_to_le32(tx_flags);
 	__skb_queue_tail(&ring->queue, skb);
diff --git a/drivers/net/wireless/rtl8187_dev.c b/drivers/net/wireless/rtl8187_dev.c
index e14c84248686..86a09b49681c 100644
--- a/drivers/net/wireless/rtl8187_dev.c
+++ b/drivers/net/wireless/rtl8187_dev.c
@@ -179,21 +179,17 @@ static int rtl8187_tx(struct ieee80211_hw *dev, struct sk_buff *skb,
 	flags = skb->len;
 	flags |= RTL8187_TX_FLAG_NO_ENCRYPT;
 
-	BUG_ON(!control->tx_rate);
-
-	flags |= control->tx_rate->hw_value << 24;
+	flags |= ieee80211_get_tx_rate(dev, control)->hw_value << 24;
 	if (ieee80211_get_morefrag((struct ieee80211_hdr *)skb->data))
 		flags |= RTL8187_TX_FLAG_MORE_FRAG;
 	if (control->flags & IEEE80211_TXCTL_USE_RTS_CTS) {
-		BUG_ON(!control->rts_cts_rate);
 		flags |= RTL8187_TX_FLAG_RTS;
-		flags |= control->rts_cts_rate->hw_value << 19;
+		flags |= ieee80211_get_rts_cts_rate(dev, control)->hw_value << 19;
 		rts_dur = ieee80211_rts_duration(dev, priv->vif,
 						 skb->len, control);
 	} else if (control->flags & IEEE80211_TXCTL_USE_CTS_PROTECT) {
-		BUG_ON(!control->rts_cts_rate);
 		flags |= RTL8187_TX_FLAG_CTS;
-		flags |= control->rts_cts_rate->hw_value << 19;
+		flags |= ieee80211_get_rts_cts_rate(dev, control)->hw_value << 19;
 	}
 
 	hdr = (struct rtl8187_tx_hdr *)skb_push(skb, sizeof(*hdr));
diff --git a/drivers/net/wireless/zd1211rw/zd_mac.c b/drivers/net/wireless/zd1211rw/zd_mac.c
index 0c736735e217..99c508c09e5b 100644
--- a/drivers/net/wireless/zd1211rw/zd_mac.c
+++ b/drivers/net/wireless/zd1211rw/zd_mac.c
@@ -523,14 +523,17 @@ static int fill_ctrlset(struct zd_mac *mac,
 	struct ieee80211_hdr *hdr = (struct ieee80211_hdr *) skb->data;
 	unsigned int frag_len = skb->len + FCS_LEN;
 	unsigned int packet_length;
+	struct ieee80211_rate *txrate;
 	struct zd_ctrlset *cs = (struct zd_ctrlset *)
 		skb_push(skb, sizeof(struct zd_ctrlset));
 
 	ZD_ASSERT(frag_len <= 0xffff);
 
-	cs->modulation = control->tx_rate->hw_value;
+	txrate = ieee80211_get_tx_rate(mac->hw, control);
+
+	cs->modulation = txrate->hw_value;
 	if (control->flags & IEEE80211_TXCTL_SHORT_PREAMBLE)
-		cs->modulation = control->tx_rate->hw_value_short;
+		cs->modulation = txrate->hw_value_short;
 
 	cs->tx_length = cpu_to_le16(frag_len);
 
diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index f00fc76a7344..0df91bea6c1f 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -266,27 +266,26 @@ enum mac80211_tx_control_flags {
  * ieee80211_ops->remove_interface() callback funtion.
  * The hw_key pointer is valid until it has been removed with the
  * ieee80211_ops->set_key() callback function.
- * The tx_rate and alt_retry_rate pointers are valid until the phy is
- * deregistered.
  */
 struct ieee80211_tx_control {
-	struct ieee80211_vif *vif;
-	struct ieee80211_rate *tx_rate;
+	u32 flags;		/* tx control flags defined above */
+
+	s8 tx_rate_idx,		/* Transmit rate (indexes registered rates) */
+	   rts_cts_rate_idx,	/* Transmit rate for RTS/CTS frame */
+	   alt_retry_rate_idx;	/* retry rate for the last retries */
 
-	/* Transmit rate for RTS/CTS frame */
-	struct ieee80211_rate *rts_cts_rate;
+	s8 retry_limit;		/* 1 = only first attempt, 2 = one retry, ..
+				 * This could be used when set_retry_limit
+				 * is not implemented by the driver */
 
-	/* retry rate for the last retries */
-	struct ieee80211_rate *alt_retry_rate;
+	struct ieee80211_vif *vif;
 
 	/* Key used for hardware encryption
 	 * NULL if IEEE80211_TXCTL_DO_NOT_ENCRYPT is set */
 	struct ieee80211_key_conf *hw_key;
 
-	u32 flags;		/* tx control flags defined above */
-	u8 retry_limit;		/* 1 = only first attempt, 2 = one retry, ..
-				 * This could be used when set_retry_limit
-				 * is not implemented by the driver */
+	enum ieee80211_band band;
+
 	u8 antenna_sel_tx; 	/* 0 = default/diversity, otherwise bit
 				 * position represents antenna number used */
 	u8 icv_len;		/* length of the ICV/MIC field in octets */
@@ -298,6 +297,7 @@ struct ieee80211_tx_control {
 };
 
 
+
 /**
  * enum mac80211_rx_flags - receive flags
  *
@@ -823,6 +823,33 @@ static inline void SET_IEEE80211_PERM_ADDR(struct ieee80211_hw *hw, u8 *addr)
 	memcpy(hw->wiphy->perm_addr, addr, ETH_ALEN);
 }
 
+static inline struct ieee80211_rate *
+ieee80211_get_tx_rate(const struct ieee80211_hw *hw,
+		      const struct ieee80211_tx_control *c)
+{
+	if (WARN_ON(c->tx_rate_idx < 0))
+		return NULL;
+	return &hw->wiphy->bands[c->band]->bitrates[c->tx_rate_idx];
+}
+
+static inline struct ieee80211_rate *
+ieee80211_get_rts_cts_rate(const struct ieee80211_hw *hw,
+			   const struct ieee80211_tx_control *c)
+{
+	if (c->rts_cts_rate_idx < 0)
+		return NULL;
+	return &hw->wiphy->bands[c->band]->bitrates[c->rts_cts_rate_idx];
+}
+
+static inline struct ieee80211_rate *
+ieee80211_get_alt_retry_rate(const struct ieee80211_hw *hw,
+			     const struct ieee80211_tx_control *c)
+{
+	if (c->alt_retry_rate_idx < 0)
+		return NULL;
+	return &hw->wiphy->bands[c->band]->bitrates[c->alt_retry_rate_idx];
+}
+
 /**
  * DOC: Hardware crypto acceleration
  *
diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h
index ed0d9b35ae6f..a4cccd1b7d53 100644
--- a/net/mac80211/ieee80211_i.h
+++ b/net/mac80211/ieee80211_i.h
@@ -159,11 +159,11 @@ struct ieee80211_tx_data {
 
 	struct ieee80211_tx_control *control;
 	struct ieee80211_channel *channel;
-	struct ieee80211_rate *rate;
+	s8 rate_idx;
 	/* use this rate (if set) for last fragment; rate can
 	 * be set to lower rate for the first fragments, e.g.,
 	 * when using CTS protection with IEEE 802.11g. */
-	struct ieee80211_rate *last_frag_rate;
+	s8 last_frag_rate_idx;
 
 	/* Extra fragments (in addition to the first fragment
 	 * in skb) */
@@ -225,9 +225,9 @@ struct ieee80211_tx_stored_packet {
 	struct ieee80211_tx_control control;
 	struct sk_buff *skb;
 	struct sk_buff **extra_frag;
-	struct ieee80211_rate *last_frag_rate;
+	s8 last_frag_rate_idx;
 	int num_extra_frag;
-	unsigned int last_frag_rate_ctrl_probe;
+	bool last_frag_rate_ctrl_probe;
 };
 
 struct beacon_data {
diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c
index 7877d3b3f4cb..604149369dc9 100644
--- a/net/mac80211/mlme.c
+++ b/net/mac80211/mlme.c
@@ -2406,15 +2406,15 @@ static int ieee80211_sta_join_ibss(struct net_device *dev,
 
 		memset(&control, 0, sizeof(control));
 		rate_control_get_rate(dev, sband, skb, &ratesel);
-		if (!ratesel.rate) {
+		if (ratesel.rate_idx < 0) {
 			printk(KERN_DEBUG "%s: Failed to determine TX rate "
 			       "for IBSS beacon\n", dev->name);
 			break;
 		}
 		control.vif = &sdata->vif;
-		control.tx_rate = ratesel.rate;
+		control.tx_rate_idx = ratesel.rate_idx;
 		if (sdata->bss_conf.use_short_preamble &&
-		    ratesel.rate->flags & IEEE80211_RATE_SHORT_PREAMBLE)
+		    sband->bitrates[ratesel.rate_idx].flags & IEEE80211_RATE_SHORT_PREAMBLE)
 			control.flags |= IEEE80211_TXCTL_SHORT_PREAMBLE;
 		control.antenna_sel_tx = local->hw.conf.antenna_sel_tx;
 		control.flags |= IEEE80211_TXCTL_NO_ACK;
diff --git a/net/mac80211/rate.c b/net/mac80211/rate.c
index 841df93807fc..0388c090dfe9 100644
--- a/net/mac80211/rate.c
+++ b/net/mac80211/rate.c
@@ -176,20 +176,24 @@ void rate_control_get_rate(struct net_device *dev,
 	rcu_read_lock();
 	sta = sta_info_get(local, hdr->addr1);
 
-	memset(sel, 0, sizeof(struct rate_selection));
+	sel->rate_idx = -1;
+	sel->nonerp_idx = -1;
+	sel->probe_idx = -1;
 
 	ref->ops->get_rate(ref->priv, dev, sband, skb, sel);
 
+	BUG_ON(sel->rate_idx < 0);
+
 	/* Select a non-ERP backup rate. */
-	if (!sel->nonerp) {
+	if (sel->nonerp_idx < 0) {
 		for (i = 0; i < sband->n_bitrates; i++) {
 			struct ieee80211_rate *rate = &sband->bitrates[i];
-			if (sel->rate->bitrate < rate->bitrate)
+			if (sband->bitrates[sel->rate_idx].bitrate < rate->bitrate)
 				break;
 
 			if (rate_supported(sta, sband->band, i) &&
 			    !(rate->flags & IEEE80211_RATE_ERP_G))
-				sel->nonerp = rate;
+				sel->nonerp_idx = i;
 		}
 	}
 
diff --git a/net/mac80211/rate.h b/net/mac80211/rate.h
index 5b45f33cb766..a29148dcca99 100644
--- a/net/mac80211/rate.h
+++ b/net/mac80211/rate.h
@@ -19,14 +19,15 @@
 #include "ieee80211_i.h"
 #include "sta_info.h"
 
-/* TODO: kdoc */
+/**
+ * struct rate_selection - rate selection for rate control algos
+ * @rate: selected transmission rate index
+ * @nonerp: Non-ERP rate to use instead if ERP cannot be used
+ * @probe: rate for probing (or -1)
+ *
+ */
 struct rate_selection {
-	/* Selected transmission rate */
-	struct ieee80211_rate *rate;
-	/* Non-ERP rate to use if mac80211 decides it cannot use an ERP rate */
-	struct ieee80211_rate *nonerp;
-	/* probe with this rate, or NULL for no probing */
-	struct ieee80211_rate *probe;
+	s8 rate_idx, nonerp_idx, probe_idx;
 };
 
 struct rate_control_ops {
@@ -138,7 +139,7 @@ static inline int rate_supported(struct sta_info *sta,
 	return (sta == NULL || sta->supp_rates[band] & BIT(index));
 }
 
-static inline int
+static inline s8
 rate_lowest_index(struct ieee80211_local *local,
 		  struct ieee80211_supported_band *sband,
 		  struct sta_info *sta)
@@ -155,14 +156,6 @@ rate_lowest_index(struct ieee80211_local *local,
 	return 0;
 }
 
-static inline struct ieee80211_rate *
-rate_lowest(struct ieee80211_local *local,
-	    struct ieee80211_supported_band *sband,
-	    struct sta_info *sta)
-{
-	return &sband->bitrates[rate_lowest_index(local, sband, sta)];
-}
-
 
 /* functions for rate control related to a device */
 int ieee80211_init_rate_ctrl_alg(struct ieee80211_local *local,
diff --git a/net/mac80211/rc80211_pid_algo.c b/net/mac80211/rc80211_pid_algo.c
index a849b745bdb5..14cde36f5070 100644
--- a/net/mac80211/rc80211_pid_algo.c
+++ b/net/mac80211/rc80211_pid_algo.c
@@ -266,7 +266,7 @@ static void rate_control_pid_tx_status(void *priv, struct net_device *dev,
 
 	/* Ignore all frames that were sent with a different rate than the rate
 	 * we currently advise mac80211 to use. */
-	if (status->control.tx_rate != &sband->bitrates[sta->txrate_idx])
+	if (status->control.tx_rate_idx != sta->txrate_idx)
 		goto unlock;
 
 	spinfo = sta->rate_ctrl_priv;
@@ -330,7 +330,7 @@ static void rate_control_pid_get_rate(void *priv, struct net_device *dev,
 	fc = le16_to_cpu(hdr->frame_control);
 	if ((fc & IEEE80211_FCTL_FTYPE) != IEEE80211_FTYPE_DATA ||
 	    is_multicast_ether_addr(hdr->addr1) || !sta) {
-		sel->rate = rate_lowest(local, sband, sta);
+		sel->rate_idx = rate_lowest_index(local, sband, sta);
 		rcu_read_unlock();
 		return;
 	}
@@ -349,7 +349,7 @@ static void rate_control_pid_get_rate(void *priv, struct net_device *dev,
 
 	rcu_read_unlock();
 
-	sel->rate = &sband->bitrates[rateidx];
+	sel->rate_idx = rateidx;
 
 #ifdef CONFIG_MAC80211_DEBUGFS
 	rate_control_pid_event_tx_rate(
diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c
index aecec2a72b08..99c3860bc0e2 100644
--- a/net/mac80211/tx.c
+++ b/net/mac80211/tx.c
@@ -91,11 +91,12 @@ static u16 ieee80211_duration(struct ieee80211_tx_data *tx, int group_addr,
 			      int next_frag_len)
 {
 	int rate, mrate, erp, dur, i;
-	struct ieee80211_rate *txrate = tx->rate;
+	struct ieee80211_rate *txrate;
 	struct ieee80211_local *local = tx->local;
 	struct ieee80211_supported_band *sband;
 
-	sband = local->hw.wiphy->bands[local->hw.conf.channel->band];
+	sband = local->hw.wiphy->bands[tx->channel->band];
+	txrate = &sband->bitrates[tx->rate_idx];
 
 	erp = 0;
 	if (tx->sdata->flags & IEEE80211_SDATA_OPERATING_GMODE)
@@ -610,40 +611,40 @@ ieee80211_tx_h_rate_ctrl(struct ieee80211_tx_data *tx)
 	struct rate_selection rsel;
 	struct ieee80211_supported_band *sband;
 
-	sband = tx->local->hw.wiphy->bands[tx->local->hw.conf.channel->band];
+	sband = tx->local->hw.wiphy->bands[tx->channel->band];
 
-	if (likely(!tx->rate)) {
+	if (likely(tx->rate_idx < 0)) {
 		rate_control_get_rate(tx->dev, sband, tx->skb, &rsel);
-		tx->rate = rsel.rate;
-		if (unlikely(rsel.probe)) {
+		tx->rate_idx = rsel.rate_idx;
+		if (unlikely(rsel.probe_idx >= 0)) {
 			tx->control->flags |=
 				IEEE80211_TXCTL_RATE_CTRL_PROBE;
 			tx->flags |= IEEE80211_TX_PROBE_LAST_FRAG;
-			tx->control->alt_retry_rate = tx->rate;
-			tx->rate = rsel.probe;
+			tx->control->alt_retry_rate_idx = tx->rate_idx;
+			tx->rate_idx = rsel.probe_idx;
 		} else
-			tx->control->alt_retry_rate = NULL;
+			tx->control->alt_retry_rate_idx = -1;
 
-		if (!tx->rate)
+		if (unlikely(tx->rate_idx < 0))
 			return TX_DROP;
 	} else
-		tx->control->alt_retry_rate = NULL;
+		tx->control->alt_retry_rate_idx = -1;
 
 	if (tx->sdata->bss_conf.use_cts_prot &&
-	    (tx->flags & IEEE80211_TX_FRAGMENTED) && rsel.nonerp) {
-		tx->last_frag_rate = tx->rate;
-		if (rsel.probe)
+	    (tx->flags & IEEE80211_TX_FRAGMENTED) && (rsel.nonerp_idx >= 0)) {
+		tx->last_frag_rate_idx = tx->rate_idx;
+		if (rsel.probe_idx >= 0)
 			tx->flags &= ~IEEE80211_TX_PROBE_LAST_FRAG;
 		else
 			tx->flags |= IEEE80211_TX_PROBE_LAST_FRAG;
-		tx->rate = rsel.nonerp;
-		tx->control->tx_rate = rsel.nonerp;
+		tx->rate_idx = rsel.nonerp_idx;
+		tx->control->tx_rate_idx = rsel.nonerp_idx;
 		tx->control->flags &= ~IEEE80211_TXCTL_RATE_CTRL_PROBE;
 	} else {
-		tx->last_frag_rate = tx->rate;
-		tx->control->tx_rate = tx->rate;
+		tx->last_frag_rate_idx = tx->rate_idx;
+		tx->control->tx_rate_idx = tx->rate_idx;
 	}
-	tx->control->tx_rate = tx->rate;
+	tx->control->tx_rate_idx = tx->rate_idx;
 
 	return TX_CONTINUE;
 }
@@ -655,6 +656,9 @@ ieee80211_tx_h_misc(struct ieee80211_tx_data *tx)
 	u16 fc = le16_to_cpu(hdr->frame_control);
 	u16 dur;
 	struct ieee80211_tx_control *control = tx->control;
+	struct ieee80211_supported_band *sband;
+
+	sband = tx->local->hw.wiphy->bands[tx->channel->band];
 
 	if (!control->retry_limit) {
 		if (!is_multicast_ether_addr(hdr->addr1)) {
@@ -681,14 +685,14 @@ ieee80211_tx_h_misc(struct ieee80211_tx_data *tx)
 		 * frames.
 		 * TODO: The last fragment could still use multiple retry
 		 * rates. */
-		control->alt_retry_rate = NULL;
+		control->alt_retry_rate_idx = -1;
 	}
 
 	/* Use CTS protection for unicast frames sent using extended rates if
 	 * there are associated non-ERP stations and RTS/CTS is not configured
 	 * for the frame. */
 	if ((tx->sdata->flags & IEEE80211_SDATA_OPERATING_GMODE) &&
-	    (tx->rate->flags & IEEE80211_RATE_ERP_G) &&
+	    (sband->bitrates[tx->rate_idx].flags & IEEE80211_RATE_ERP_G) &&
 	    (tx->flags & IEEE80211_TX_UNICAST) &&
 	    tx->sdata->bss_conf.use_cts_prot &&
 	    !(control->flags & IEEE80211_TXCTL_USE_RTS_CTS))
@@ -698,7 +702,7 @@ ieee80211_tx_h_misc(struct ieee80211_tx_data *tx)
 	 * short preambles at the selected rate and short preambles are
 	 * available on the network at the current point in time. */
 	if (((fc & IEEE80211_FCTL_FTYPE) == IEEE80211_FTYPE_DATA) &&
-	    (tx->rate->flags & IEEE80211_RATE_SHORT_PREAMBLE) &&
+	    (sband->bitrates[tx->rate_idx].flags & IEEE80211_RATE_SHORT_PREAMBLE) &&
 	    tx->sdata->bss_conf.use_short_preamble &&
 	    (!tx->sta || test_sta_flags(tx->sta, WLAN_STA_SHORT_PREAMBLE))) {
 		tx->control->flags |= IEEE80211_TXCTL_SHORT_PREAMBLE;
@@ -715,32 +719,32 @@ ieee80211_tx_h_misc(struct ieee80211_tx_data *tx)
 	if ((control->flags & IEEE80211_TXCTL_USE_RTS_CTS) ||
 	    (control->flags & IEEE80211_TXCTL_USE_CTS_PROTECT)) {
 		struct ieee80211_supported_band *sband;
-		struct ieee80211_rate *rate, *baserate;
+		struct ieee80211_rate *rate;
+		s8 baserate = -1;
 		int idx;
 
-		sband = tx->local->hw.wiphy->bands[
-				tx->local->hw.conf.channel->band];
+		sband = tx->local->hw.wiphy->bands[tx->channel->band];
 
 		/* Do not use multiple retry rates when using RTS/CTS */
-		control->alt_retry_rate = NULL;
+		control->alt_retry_rate_idx = -1;
 
 		/* Use min(data rate, max base rate) as CTS/RTS rate */
-		rate = tx->rate;
-		baserate = NULL;
+		rate = &sband->bitrates[tx->rate_idx];
 
 		for (idx = 0; idx < sband->n_bitrates; idx++) {
 			if (sband->bitrates[idx].bitrate > rate->bitrate)
 				continue;
 			if (tx->sdata->basic_rates & BIT(idx) &&
-			    (!baserate ||
-			     (baserate->bitrate < sband->bitrates[idx].bitrate)))
-				baserate = &sband->bitrates[idx];
+			    (baserate < 0 ||
+			     (sband->bitrates[baserate].bitrate
+			      < sband->bitrates[idx].bitrate)))
+				baserate = idx;
 		}
 
-		if (baserate)
-			control->rts_cts_rate = baserate;
+		if (baserate >= 0)
+			control->rts_cts_rate_idx = baserate;
 		else
-			control->rts_cts_rate = &sband->bitrates[0];
+			control->rts_cts_rate_idx = 0;
 	}
 
 	if (tx->sta) {
@@ -768,7 +772,11 @@ ieee80211_tx_h_load_stats(struct ieee80211_tx_data *tx)
 	struct sk_buff *skb = tx->skb;
 	struct ieee80211_hdr *hdr = (struct ieee80211_hdr *) skb->data;
 	u32 load = 0, hdrtime;
-	struct ieee80211_rate *rate = tx->rate;
+	struct ieee80211_rate *rate;
+	struct ieee80211_supported_band *sband;
+
+	sband = tx->local->hw.wiphy->bands[tx->channel->band];
+	rate = &sband->bitrates[tx->rate_idx];
 
 	/* TODO: this could be part of tx_status handling, so that the number
 	 * of retries would be known; TX rate should in that case be stored
@@ -803,7 +811,7 @@ ieee80211_tx_h_load_stats(struct ieee80211_tx_data *tx)
 		for (i = 0; i < tx->num_extra_frag; i++) {
 			load += 2 * hdrtime;
 			load += tx->extra_frag[i]->len *
-				tx->rate->bitrate;
+				rate->bitrate;
 		}
 	}
 
@@ -859,7 +867,7 @@ __ieee80211_parse_tx_radiotap(struct ieee80211_tx_data *tx,
 	int ret = ieee80211_radiotap_iterator_init(&iterator, rthdr, skb->len);
 	struct ieee80211_tx_control *control = tx->control;
 
-	sband = tx->local->hw.wiphy->bands[tx->local->hw.conf.channel->band];
+	sband = tx->local->hw.wiphy->bands[tx->channel->band];
 
 	control->flags |= IEEE80211_TXCTL_DO_NOT_ENCRYPT;
 	tx->flags |= IEEE80211_TX_INJECTED;
@@ -899,7 +907,7 @@ __ieee80211_parse_tx_radiotap(struct ieee80211_tx_data *tx,
 				r = &sband->bitrates[i];
 
 				if (r->bitrate == target_rate) {
-					tx->rate = r;
+					tx->rate_idx = i;
 					break;
 				}
 			}
@@ -1097,7 +1105,7 @@ static int __ieee80211_tx(struct ieee80211_local *local, struct sk_buff *skb,
 			if (__ieee80211_queue_stopped(local, control->queue))
 				return IEEE80211_TX_FRAG_AGAIN;
 			if (i == tx->num_extra_frag) {
-				control->tx_rate = tx->last_frag_rate;
+				control->tx_rate_idx = tx->last_frag_rate_idx;
 
 				if (tx->flags & IEEE80211_TX_PROBE_LAST_FRAG)
 					control->flags |=
@@ -1155,6 +1163,7 @@ static int ieee80211_tx(struct net_device *dev, struct sk_buff *skb,
 
 	sta = tx.sta;
 	tx.channel = local->hw.conf.channel;
+	control->band = tx.channel->band;
 
 	for (handler = ieee80211_tx_handlers; *handler != NULL;
 	     handler++) {
@@ -1187,7 +1196,7 @@ static int ieee80211_tx(struct net_device *dev, struct sk_buff *skb,
 				next_len = tx.extra_frag[i + 1]->len;
 			} else {
 				next_len = 0;
-				tx.rate = tx.last_frag_rate;
+				tx.rate_idx = tx.last_frag_rate_idx;
 			}
 			dur = ieee80211_duration(&tx, 0, next_len);
 			hdr->duration_id = cpu_to_le16(dur);
@@ -1224,7 +1233,7 @@ retry:
 		store->skb = skb;
 		store->extra_frag = tx.extra_frag;
 		store->num_extra_frag = tx.num_extra_frag;
-		store->last_frag_rate = tx.last_frag_rate;
+		store->last_frag_rate_idx = tx.last_frag_rate_idx;
 		store->last_frag_rate_ctrl_probe =
 			!!(tx.flags & IEEE80211_TX_PROBE_LAST_FRAG);
 	}
@@ -1685,7 +1694,7 @@ void ieee80211_tx_pending(unsigned long data)
 		tx.control = &store->control;
 		tx.extra_frag = store->extra_frag;
 		tx.num_extra_frag = store->num_extra_frag;
-		tx.last_frag_rate = store->last_frag_rate;
+		tx.last_frag_rate_idx = store->last_frag_rate_idx;
 		tx.flags = 0;
 		if (store->last_frag_rate_ctrl_probe)
 			tx.flags |= IEEE80211_TX_PROBE_LAST_FRAG;
@@ -1789,9 +1798,10 @@ struct sk_buff *ieee80211_beacon_get(struct ieee80211_hw *hw,
 	struct ieee80211_mgmt *mgmt;
 	int *num_beacons;
 	bool err = true;
+	enum ieee80211_band band = local->hw.conf.channel->band;
 	u8 *pos;
 
-	sband = local->hw.wiphy->bands[local->hw.conf.channel->band];
+	sband = local->hw.wiphy->bands[band];
 
 	rcu_read_lock();
 
@@ -1885,8 +1895,9 @@ struct sk_buff *ieee80211_beacon_get(struct ieee80211_hw *hw,
 	}
 
 	if (control) {
+		control->band = band;
 		rate_control_get_rate(local->mdev, sband, skb, &rsel);
-		if (!rsel.rate) {
+		if (unlikely(rsel.rate_idx < 0)) {
 			if (net_ratelimit()) {
 				printk(KERN_DEBUG "%s: ieee80211_beacon_get: "
 				       "no rate found\n",
@@ -1898,9 +1909,9 @@ struct sk_buff *ieee80211_beacon_get(struct ieee80211_hw *hw,
 		}
 
 		control->vif = vif;
-		control->tx_rate = rsel.rate;
+		control->tx_rate_idx = rsel.rate_idx;
 		if (sdata->bss_conf.use_short_preamble &&
-		    rsel.rate->flags & IEEE80211_RATE_SHORT_PREAMBLE)
+		    sband->bitrates[rsel.rate_idx].flags & IEEE80211_RATE_SHORT_PREAMBLE)
 			control->flags |= IEEE80211_TXCTL_SHORT_PREAMBLE;
 		control->antenna_sel_tx = local->hw.conf.antenna_sel_tx;
 		control->flags |= IEEE80211_TXCTL_NO_ACK;
@@ -2006,6 +2017,7 @@ ieee80211_get_buffered_bc(struct ieee80211_hw *hw,
 	sta = tx.sta;
 	tx.flags |= IEEE80211_TX_PS_BUFFERED;
 	tx.channel = local->hw.conf.channel;
+	control->band = tx.channel->band;
 
 	for (handler = ieee80211_tx_handlers; *handler != NULL; handler++) {
 		res = (*handler)(&tx);
diff --git a/net/mac80211/util.c b/net/mac80211/util.c
index 800c15aff6e7..65a34fddeb00 100644
--- a/net/mac80211/util.c
+++ b/net/mac80211/util.c
@@ -266,10 +266,13 @@ __le16 ieee80211_rts_duration(struct ieee80211_hw *hw,
 	bool short_preamble;
 	int erp;
 	u16 dur;
+	struct ieee80211_supported_band *sband;
+
+	sband = local->hw.wiphy->bands[local->hw.conf.channel->band];
 
 	short_preamble = sdata->bss_conf.use_short_preamble;
 
-	rate = frame_txctl->rts_cts_rate;
+	rate = &sband->bitrates[frame_txctl->rts_cts_rate_idx];
 
 	erp = 0;
 	if (sdata->flags & IEEE80211_SDATA_OPERATING_GMODE)
@@ -300,10 +303,13 @@ __le16 ieee80211_ctstoself_duration(struct ieee80211_hw *hw,
 	bool short_preamble;
 	int erp;
 	u16 dur;
+	struct ieee80211_supported_band *sband;
+
+	sband = local->hw.wiphy->bands[local->hw.conf.channel->band];
 
 	short_preamble = sdata->bss_conf.use_short_preamble;
 
-	rate = frame_txctl->rts_cts_rate;
+	rate = &sband->bitrates[frame_txctl->rts_cts_rate_idx];
 	erp = 0;
 	if (sdata->flags & IEEE80211_SDATA_OPERATING_GMODE)
 		erp = rate->flags & IEEE80211_RATE_ERP_G;
-- 
cgit v1.2.3


From e039fa4a4195ac4ee895e6f3d1334beed63256fe Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes@sipsolutions.net>
Date: Thu, 15 May 2008 12:55:29 +0200
Subject: mac80211: move TX info into skb->cb

This patch converts mac80211 and all drivers to have transmit
information and status in skb->cb rather than allocating extra
memory for it and copying all the data around. To make it fit,
a union is used where only data that is necessary for all steps
is kept outside of the union.

A number of fixes were done by Ivo, as well as the rt2x00 part
of this patch.

Signed-off-by: Ivo van Doorn <IvDoorn@gmail.com>
Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
Acked-by: David S. Miller <davem@davemloft.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/net/wireless/adm8211.c              |  31 ++-
 drivers/net/wireless/adm8211.h              |   1 -
 drivers/net/wireless/ath5k/base.c           |  70 +++----
 drivers/net/wireless/ath5k/base.h           |   1 -
 drivers/net/wireless/b43/b43.h              |   1 -
 drivers/net/wireless/b43/dma.c              |  45 ++--
 drivers/net/wireless/b43/dma.h              |   3 +-
 drivers/net/wireless/b43/main.c             |  33 ++-
 drivers/net/wireless/b43/pio.c              |  36 ++--
 drivers/net/wireless/b43/pio.h              |   8 +-
 drivers/net/wireless/b43/xmit.c             |  58 +++---
 drivers/net/wireless/b43/xmit.h             |   4 +-
 drivers/net/wireless/b43legacy/dma.c        |  36 ++--
 drivers/net/wireless/b43legacy/dma.h        |   4 +-
 drivers/net/wireless/b43legacy/main.c       |  12 +-
 drivers/net/wireless/b43legacy/pio.c        |  19 +-
 drivers/net/wireless/b43legacy/pio.h        |   4 +-
 drivers/net/wireless/b43legacy/xmit.c       |  44 ++--
 drivers/net/wireless/b43legacy/xmit.h       |   2 +-
 drivers/net/wireless/iwlwifi/iwl-3945-rs.c  |  12 +-
 drivers/net/wireless/iwlwifi/iwl-3945.c     |  24 +--
 drivers/net/wireless/iwlwifi/iwl-3945.h     |   3 +-
 drivers/net/wireless/iwlwifi/iwl-4965-rs.c  |  49 +++--
 drivers/net/wireless/iwlwifi/iwl-4965.c     |  28 +--
 drivers/net/wireless/iwlwifi/iwl-core.h     |   3 +-
 drivers/net/wireless/iwlwifi/iwl-dev.h      |   5 +-
 drivers/net/wireless/iwlwifi/iwl-tx.c       |  40 ++--
 drivers/net/wireless/iwlwifi/iwl3945-base.c |  48 ++---
 drivers/net/wireless/iwlwifi/iwl4965-base.c |  52 ++---
 drivers/net/wireless/p54/p54common.c        |  90 ++++----
 drivers/net/wireless/p54/p54common.h        |   1 -
 drivers/net/wireless/rt2x00/rt2400pci.c     |   8 +-
 drivers/net/wireless/rt2x00/rt2500pci.c     |   8 +-
 drivers/net/wireless/rt2x00/rt2500usb.c     |   9 +-
 drivers/net/wireless/rt2x00/rt2x00.h        |  10 +-
 drivers/net/wireless/rt2x00/rt2x00dev.c     |  49 +++--
 drivers/net/wireless/rt2x00/rt2x00mac.c     |  74 ++++---
 drivers/net/wireless/rt2x00/rt2x00pci.c     |  11 +-
 drivers/net/wireless/rt2x00/rt2x00pci.h     |   6 +-
 drivers/net/wireless/rt2x00/rt2x00queue.c   |  31 ++-
 drivers/net/wireless/rt2x00/rt2x00queue.h   |  15 +-
 drivers/net/wireless/rt2x00/rt2x00usb.c     |  12 +-
 drivers/net/wireless/rt2x00/rt2x00usb.h     |   9 +-
 drivers/net/wireless/rt2x00/rt61pci.c       |   8 +-
 drivers/net/wireless/rt2x00/rt73usb.c       |   8 +-
 drivers/net/wireless/rtl8180_dev.c          |  57 +++--
 drivers/net/wireless/rtl8187.h              |   6 -
 drivers/net/wireless/rtl8187_dev.c          |  41 ++--
 drivers/net/wireless/zd1211rw/zd_mac.c      | 156 ++++----------
 drivers/net/wireless/zd1211rw/zd_mac.h      |  16 --
 drivers/net/wireless/zd1211rw/zd_usb.c      |   6 +-
 include/net/mac80211.h                      | 299 ++++++++++++---------------
 net/mac80211/ieee80211_i.h                  |  18 +-
 net/mac80211/main.c                         | 128 ++++--------
 net/mac80211/mlme.c                         |  30 +--
 net/mac80211/rate.h                         |   8 +-
 net/mac80211/rc80211_pid.h                  |   4 +-
 net/mac80211/rc80211_pid_algo.c             |  18 +-
 net/mac80211/rc80211_pid_debugfs.c          |   8 +-
 net/mac80211/rx.c                           |  10 +-
 net/mac80211/sta_info.c                     |   6 +-
 net/mac80211/sta_info.h                     |   2 +-
 net/mac80211/tx.c                           | 310 ++++++++++++++--------------
 net/mac80211/util.c                         |  10 +-
 net/mac80211/wep.c                          |   9 +-
 net/mac80211/wme.c                          |  22 +-
 net/mac80211/wpa.c                          |  65 +++---
 67 files changed, 980 insertions(+), 1274 deletions(-)

(limited to 'include')

diff --git a/drivers/net/wireless/adm8211.c b/drivers/net/wireless/adm8211.c
index 22db664a58d9..0ba55ba93958 100644
--- a/drivers/net/wireless/adm8211.c
+++ b/drivers/net/wireless/adm8211.c
@@ -324,7 +324,7 @@ static void adm8211_interrupt_tci(struct ieee80211_hw *dev)
 	for (dirty_tx = priv->dirty_tx; priv->cur_tx - dirty_tx; dirty_tx++) {
 		unsigned int entry = dirty_tx % priv->tx_ring_size;
 		u32 status = le32_to_cpu(priv->tx_ring[entry].status);
-		struct ieee80211_tx_status tx_status;
+		struct ieee80211_tx_info *txi;
 		struct adm8211_tx_ring_info *info;
 		struct sk_buff *skb;
 
@@ -334,24 +334,23 @@ static void adm8211_interrupt_tci(struct ieee80211_hw *dev)
 
 		info = &priv->tx_buffers[entry];
 		skb = info->skb;
+		txi = IEEE80211_SKB_CB(skb);
 
 		/* TODO: check TDES0_STATUS_TUF and TDES0_STATUS_TRO */
 
 		pci_unmap_single(priv->pdev, info->mapping,
 				 info->skb->len, PCI_DMA_TODEVICE);
 
-		memset(&tx_status, 0, sizeof(tx_status));
+		memset(&txi->status, 0, sizeof(txi->status));
 		skb_pull(skb, sizeof(struct adm8211_tx_hdr));
 		memcpy(skb_push(skb, info->hdrlen), skb->cb, info->hdrlen);
-		memcpy(&tx_status.control, &info->tx_control,
-		       sizeof(tx_status.control));
-		if (!(tx_status.control.flags & IEEE80211_TXCTL_NO_ACK)) {
+		if (!(txi->flags & IEEE80211_TX_CTL_NO_ACK)) {
 			if (status & TDES0_STATUS_ES)
-				tx_status.excessive_retries = 1;
+				txi->status.excessive_retries = 1;
 			else
-				tx_status.flags |= IEEE80211_TX_STATUS_ACK;
+				txi->flags |= IEEE80211_TX_STAT_ACK;
 		}
-		ieee80211_tx_status_irqsafe(dev, skb, &tx_status);
+		ieee80211_tx_status_irqsafe(dev, skb);
 
 		info->skb = NULL;
 	}
@@ -1638,7 +1637,6 @@ static void adm8211_calc_durations(int *dur, int *plcp, size_t payload_len, int
 /* Transmit skb w/adm8211_tx_hdr (802.11 header created by hardware) */
 static void adm8211_tx_raw(struct ieee80211_hw *dev, struct sk_buff *skb,
 			   u16 plcp_signal,
-			   struct ieee80211_tx_control *control,
 			   size_t hdrlen)
 {
 	struct adm8211_priv *priv = dev->priv;
@@ -1664,7 +1662,6 @@ static void adm8211_tx_raw(struct ieee80211_hw *dev, struct sk_buff *skb,
 
 	priv->tx_buffers[entry].skb = skb;
 	priv->tx_buffers[entry].mapping = mapping;
-	memcpy(&priv->tx_buffers[entry].tx_control, control, sizeof(*control));
 	priv->tx_buffers[entry].hdrlen = hdrlen;
 	priv->tx_ring[entry].buffer1 = cpu_to_le32(mapping);
 
@@ -1685,17 +1682,17 @@ static void adm8211_tx_raw(struct ieee80211_hw *dev, struct sk_buff *skb,
 }
 
 /* Put adm8211_tx_hdr on skb and transmit */
-static int adm8211_tx(struct ieee80211_hw *dev, struct sk_buff *skb,
-		      struct ieee80211_tx_control *control)
+static int adm8211_tx(struct ieee80211_hw *dev, struct sk_buff *skb)
 {
 	struct adm8211_tx_hdr *txhdr;
 	u16 fc;
 	size_t payload_len, hdrlen;
 	int plcp, dur, len, plcp_signal, short_preamble;
 	struct ieee80211_hdr *hdr;
-	struct ieee80211_rate *txrate = ieee80211_get_tx_rate(dev, control);
+	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
+	struct ieee80211_rate *txrate = ieee80211_get_tx_rate(dev, info);
 
-	short_preamble = !!(txrate->flags & IEEE80211_TXCTL_SHORT_PREAMBLE);
+	short_preamble = !!(txrate->flags & IEEE80211_TX_CTL_SHORT_PREAMBLE);
 	plcp_signal = txrate->bitrate;
 
 	hdr = (struct ieee80211_hdr *)skb->data;
@@ -1730,15 +1727,15 @@ static int adm8211_tx(struct ieee80211_hw *dev, struct sk_buff *skb,
 	if (short_preamble)
 		txhdr->header_control |= cpu_to_le16(ADM8211_TXHDRCTL_SHORT_PREAMBLE);
 
-	if (control->flags & IEEE80211_TXCTL_USE_RTS_CTS)
+	if (info->flags & IEEE80211_TX_CTL_USE_RTS_CTS)
 		txhdr->header_control |= cpu_to_le16(ADM8211_TXHDRCTL_ENABLE_RTS);
 
 	if (fc & IEEE80211_FCTL_PROTECTED)
 		txhdr->header_control |= cpu_to_le16(ADM8211_TXHDRCTL_ENABLE_WEP_ENGINE);
 
-	txhdr->retry_limit = control->retry_limit;
+	txhdr->retry_limit = info->control.retry_limit;
 
-	adm8211_tx_raw(dev, skb, plcp_signal, control, hdrlen);
+	adm8211_tx_raw(dev, skb, plcp_signal, hdrlen);
 
 	return NETDEV_TX_OK;
 }
diff --git a/drivers/net/wireless/adm8211.h b/drivers/net/wireless/adm8211.h
index 8d7c564b3b04..9b190ee26e90 100644
--- a/drivers/net/wireless/adm8211.h
+++ b/drivers/net/wireless/adm8211.h
@@ -443,7 +443,6 @@ struct adm8211_rx_ring_info {
 struct adm8211_tx_ring_info {
 	struct sk_buff *skb;
 	dma_addr_t mapping;
-	struct ieee80211_tx_control tx_control;
 	size_t hdrlen;
 };
 
diff --git a/drivers/net/wireless/ath5k/base.c b/drivers/net/wireless/ath5k/base.c
index 32ee351a7650..7d97934265db 100644
--- a/drivers/net/wireless/ath5k/base.c
+++ b/drivers/net/wireless/ath5k/base.c
@@ -167,8 +167,7 @@ static struct pci_driver ath5k_pci_driver = {
 /*
  * Prototypes - MAC 802.11 stack related functions
  */
-static int ath5k_tx(struct ieee80211_hw *hw, struct sk_buff *skb,
-		struct ieee80211_tx_control *ctl);
+static int ath5k_tx(struct ieee80211_hw *hw, struct sk_buff *skb);
 static int ath5k_reset(struct ieee80211_hw *hw);
 static int ath5k_start(struct ieee80211_hw *hw);
 static void ath5k_stop(struct ieee80211_hw *hw);
@@ -196,8 +195,7 @@ static int ath5k_get_tx_stats(struct ieee80211_hw *hw,
 static u64 ath5k_get_tsf(struct ieee80211_hw *hw);
 static void ath5k_reset_tsf(struct ieee80211_hw *hw);
 static int ath5k_beacon_update(struct ieee80211_hw *hw,
-		struct sk_buff *skb,
-		struct ieee80211_tx_control *ctl);
+		struct sk_buff *skb);
 
 static struct ieee80211_ops ath5k_hw_ops = {
 	.tx 		= ath5k_tx,
@@ -251,9 +249,7 @@ static void	ath5k_desc_free(struct ath5k_softc *sc,
 static int 	ath5k_rxbuf_setup(struct ath5k_softc *sc,
 				struct ath5k_buf *bf);
 static int 	ath5k_txbuf_setup(struct ath5k_softc *sc,
-				struct ath5k_buf *bf,
-				struct ieee80211_tx_control *ctl);
-
+				struct ath5k_buf *bf);
 static inline void ath5k_txbuf_free(struct ath5k_softc *sc,
 				struct ath5k_buf *bf)
 {
@@ -289,8 +285,7 @@ static void 	ath5k_tx_processq(struct ath5k_softc *sc,
 static void 	ath5k_tasklet_tx(unsigned long data);
 /* Beacon handling */
 static int 	ath5k_beacon_setup(struct ath5k_softc *sc,
-				struct ath5k_buf *bf,
-				struct ieee80211_tx_control *ctl);
+					struct ath5k_buf *bf);
 static void 	ath5k_beacon_send(struct ath5k_softc *sc);
 static void 	ath5k_beacon_config(struct ath5k_softc *sc);
 static void	ath5k_beacon_update_timers(struct ath5k_softc *sc, u64 bc_tsf);
@@ -1295,37 +1290,36 @@ ath5k_rxbuf_setup(struct ath5k_softc *sc, struct ath5k_buf *bf)
 }
 
 static int
-ath5k_txbuf_setup(struct ath5k_softc *sc, struct ath5k_buf *bf,
-		struct ieee80211_tx_control *ctl)
+ath5k_txbuf_setup(struct ath5k_softc *sc, struct ath5k_buf *bf)
 {
 	struct ath5k_hw *ah = sc->ah;
 	struct ath5k_txq *txq = sc->txq;
 	struct ath5k_desc *ds = bf->desc;
 	struct sk_buff *skb = bf->skb;
+	struct ieee80211_tx_info *info = (void*) skb->cb;
 	unsigned int pktlen, flags, keyidx = AR5K_TXKEYIX_INVALID;
 	int ret;
 
 	flags = AR5K_TXDESC_INTREQ | AR5K_TXDESC_CLRDMASK;
-	bf->ctl = *ctl;
+
 	/* XXX endianness */
 	bf->skbaddr = pci_map_single(sc->pdev, skb->data, skb->len,
 			PCI_DMA_TODEVICE);
 
-	if (ctl->flags & IEEE80211_TXCTL_NO_ACK)
+	if (info->flags & IEEE80211_TX_CTL_NO_ACK)
 		flags |= AR5K_TXDESC_NOACK;
 
 	pktlen = skb->len;
 
-	if (!(ctl->flags & IEEE80211_TXCTL_DO_NOT_ENCRYPT)) {
-		keyidx = ctl->hw_key->hw_key_idx;
-		pktlen += ctl->icv_len;
+	if (!(info->flags & IEEE80211_TX_CTL_DO_NOT_ENCRYPT)) {
+		keyidx = info->control.hw_key->hw_key_idx;
+		pktlen += info->control.icv_len;
 	}
-
 	ret = ah->ah_setup_tx_desc(ah, ds, pktlen,
 		ieee80211_get_hdrlen_from_skb(skb), AR5K_PKT_TYPE_NORMAL,
 		(sc->power_level * 2),
-		ieee80211_get_tx_rate(sc->hw, ctl)->hw_value,
-		ctl->retry_limit, keyidx, 0, flags, 0, 0);
+		ieee80211_get_tx_rate(sc->hw, info)->hw_value,
+		info->control.retry_limit, keyidx, 0, flags, 0, 0);
 	if (ret)
 		goto err_unmap;
 
@@ -1927,11 +1921,11 @@ next:
 static void
 ath5k_tx_processq(struct ath5k_softc *sc, struct ath5k_txq *txq)
 {
-	struct ieee80211_tx_status txs = {};
 	struct ath5k_tx_status ts = {};
 	struct ath5k_buf *bf, *bf0;
 	struct ath5k_desc *ds;
 	struct sk_buff *skb;
+	struct ieee80211_tx_info *info;
 	int ret;
 
 	spin_lock(&txq->lock);
@@ -1951,24 +1945,25 @@ ath5k_tx_processq(struct ath5k_softc *sc, struct ath5k_txq *txq)
 		}
 
 		skb = bf->skb;
+		info = (void*) skb->cb;
 		bf->skb = NULL;
+
 		pci_unmap_single(sc->pdev, bf->skbaddr, skb->len,
 				PCI_DMA_TODEVICE);
 
-		txs.control = bf->ctl;
-		txs.retry_count = ts.ts_shortretry + ts.ts_longretry / 6;
+		info->status.retry_count = ts.ts_shortretry + ts.ts_longretry / 6;
 		if (unlikely(ts.ts_status)) {
 			sc->ll_stats.dot11ACKFailureCount++;
 			if (ts.ts_status & AR5K_TXERR_XRETRY)
-				txs.excessive_retries = 1;
+				info->status.excessive_retries = 1;
 			else if (ts.ts_status & AR5K_TXERR_FILT)
-				txs.flags |= IEEE80211_TX_STATUS_TX_FILTERED;
+				info->flags |= IEEE80211_TX_STAT_TX_FILTERED;
 		} else {
-			txs.flags |= IEEE80211_TX_STATUS_ACK;
-			txs.ack_signal = ts.ts_rssi;
+			info->flags |= IEEE80211_TX_STAT_ACK;
+			info->status.ack_signal = ts.ts_rssi;
 		}
 
-		ieee80211_tx_status(sc->hw, skb, &txs);
+		ieee80211_tx_status(sc->hw, skb);
 		sc->tx_stats[txq->qnum].count++;
 
 		spin_lock(&sc->txbuflock);
@@ -2005,10 +2000,10 @@ ath5k_tasklet_tx(unsigned long data)
  * Setup the beacon frame for transmit.
  */
 static int
-ath5k_beacon_setup(struct ath5k_softc *sc, struct ath5k_buf *bf,
-		struct ieee80211_tx_control *ctl)
+ath5k_beacon_setup(struct ath5k_softc *sc, struct ath5k_buf *bf)
 {
 	struct sk_buff *skb = bf->skb;
+	struct	ieee80211_tx_info *info = (void*) skb->cb;
 	struct ath5k_hw *ah = sc->ah;
 	struct ath5k_desc *ds;
 	int ret, antenna = 0;
@@ -2047,7 +2042,7 @@ ath5k_beacon_setup(struct ath5k_softc *sc, struct ath5k_buf *bf,
 	ret = ah->ah_setup_tx_desc(ah, ds, skb->len,
 			ieee80211_get_hdrlen_from_skb(skb),
 			AR5K_PKT_TYPE_BEACON, (sc->power_level * 2),
-			ieee80211_get_tx_rate(sc->hw, ctl)->hw_value,
+			ieee80211_get_tx_rate(sc->hw, info)->hw_value,
 			1, AR5K_TXKEYIX_INVALID,
 			antenna, flags, 0, 0);
 	if (ret)
@@ -2626,11 +2621,11 @@ ath5k_led_event(struct ath5k_softc *sc, int event)
 \********************/
 
 static int
-ath5k_tx(struct ieee80211_hw *hw, struct sk_buff *skb,
-			struct ieee80211_tx_control *ctl)
+ath5k_tx(struct ieee80211_hw *hw, struct sk_buff *skb)
 {
 	struct ath5k_softc *sc = hw->priv;
 	struct ath5k_buf *bf;
+	struct ieee80211_tx_info *info = (void*) skb->cb;
 	unsigned long flags;
 	int hdrlen;
 	int pad;
@@ -2656,13 +2651,13 @@ ath5k_tx(struct ieee80211_hw *hw, struct sk_buff *skb,
 		memmove(skb->data, skb->data+pad, hdrlen);
 	}
 
-	sc->led_txrate = ieee80211_get_tx_rate(hw, ctl)->hw_value;
+	sc->led_txrate = ieee80211_get_tx_rate(hw, info)->hw_value;
 
 	spin_lock_irqsave(&sc->txbuflock, flags);
 	if (list_empty(&sc->txbuf)) {
 		ATH5K_ERR(sc, "no further txbuf available, dropping packet\n");
 		spin_unlock_irqrestore(&sc->txbuflock, flags);
-		ieee80211_stop_queue(hw, ctl->queue);
+		ieee80211_stop_queue(hw, info->queue);
 		return -1;
 	}
 	bf = list_first_entry(&sc->txbuf, struct ath5k_buf, list);
@@ -2674,7 +2669,7 @@ ath5k_tx(struct ieee80211_hw *hw, struct sk_buff *skb,
 
 	bf->skb = skb;
 
-	if (ath5k_txbuf_setup(sc, bf, ctl)) {
+	if (ath5k_txbuf_setup(sc, bf)) {
 		bf->skb = NULL;
 		spin_lock_irqsave(&sc->txbuflock, flags);
 		list_add_tail(&bf->list, &sc->txbuf);
@@ -3052,8 +3047,7 @@ ath5k_reset_tsf(struct ieee80211_hw *hw)
 }
 
 static int
-ath5k_beacon_update(struct ieee80211_hw *hw, struct sk_buff *skb,
-			struct ieee80211_tx_control *ctl)
+ath5k_beacon_update(struct ieee80211_hw *hw, struct sk_buff *skb)
 {
 	struct ath5k_softc *sc = hw->priv;
 	int ret;
@@ -3069,7 +3063,7 @@ ath5k_beacon_update(struct ieee80211_hw *hw, struct sk_buff *skb,
 
 	ath5k_txbuf_free(sc, sc->bbuf);
 	sc->bbuf->skb = skb;
-	ret = ath5k_beacon_setup(sc, sc->bbuf, ctl);
+	ret = ath5k_beacon_setup(sc, sc->bbuf);
 	if (ret)
 		sc->bbuf->skb = NULL;
 	else
diff --git a/drivers/net/wireless/ath5k/base.h b/drivers/net/wireless/ath5k/base.h
index ecb17495488c..bb4b26d523ab 100644
--- a/drivers/net/wireless/ath5k/base.h
+++ b/drivers/net/wireless/ath5k/base.h
@@ -60,7 +60,6 @@ struct ath5k_buf {
 	dma_addr_t		daddr;	/* physical addr of desc */
 	struct sk_buff		*skb;	/* skbuff for buf */
 	dma_addr_t		skbaddr;/* physical addr of skb data */
-	struct ieee80211_tx_control ctl;
 };
 
 /*
diff --git a/drivers/net/wireless/b43/b43.h b/drivers/net/wireless/b43/b43.h
index 0c2bc061e8f3..aa493830a82d 100644
--- a/drivers/net/wireless/b43/b43.h
+++ b/drivers/net/wireless/b43/b43.h
@@ -733,7 +733,6 @@ struct b43_wl {
 	/* The beacon we are currently using (AP or IBSS mode).
 	 * This beacon stuff is protected by the irq_lock. */
 	struct sk_buff *current_beacon;
-	struct ieee80211_tx_control beacon_txctl;
 	bool beacon0_uploaded;
 	bool beacon1_uploaded;
 	struct work_struct beacon_update_trigger;
diff --git a/drivers/net/wireless/b43/dma.c b/drivers/net/wireless/b43/dma.c
index f50e2014ffbe..aced9866d815 100644
--- a/drivers/net/wireless/b43/dma.c
+++ b/drivers/net/wireless/b43/dma.c
@@ -1131,10 +1131,10 @@ struct b43_dmaring *parse_cookie(struct b43_wldev *dev, u16 cookie, int *slot)
 }
 
 static int dma_tx_fragment(struct b43_dmaring *ring,
-			   struct sk_buff *skb,
-			   struct ieee80211_tx_control *ctl)
+			   struct sk_buff *skb)
 {
 	const struct b43_dma_ops *ops = ring->ops;
+	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
 	u8 *header;
 	int slot, old_top_slot, old_used_slots;
 	int err;
@@ -1158,7 +1158,7 @@ static int dma_tx_fragment(struct b43_dmaring *ring,
 	header = &(ring->txhdr_cache[slot * hdrsize]);
 	cookie = generate_cookie(ring, slot);
 	err = b43_generate_txhdr(ring->dev, header,
-				 skb->data, skb->len, ctl, cookie);
+				 skb->data, skb->len, info, cookie);
 	if (unlikely(err)) {
 		ring->current_slot = old_top_slot;
 		ring->used_slots = old_used_slots;
@@ -1180,7 +1180,6 @@ static int dma_tx_fragment(struct b43_dmaring *ring,
 	desc = ops->idx2desc(ring, slot, &meta);
 	memset(meta, 0, sizeof(*meta));
 
-	memcpy(&meta->txstat.control, ctl, sizeof(*ctl));
 	meta->skb = skb;
 	meta->is_last_fragment = 1;
 
@@ -1210,7 +1209,7 @@ static int dma_tx_fragment(struct b43_dmaring *ring,
 
 	ops->fill_descriptor(ring, desc, meta->dmaaddr, skb->len, 0, 1, 1);
 
-	if (ctl->flags & IEEE80211_TXCTL_SEND_AFTER_DTIM) {
+	if (info->flags & IEEE80211_TX_CTL_SEND_AFTER_DTIM) {
 		/* Tell the firmware about the cookie of the last
 		 * mcast frame, so it can clear the more-data bit in it. */
 		b43_shm_write16(ring->dev, B43_SHM_SHARED,
@@ -1281,16 +1280,16 @@ static struct b43_dmaring * select_ring_by_priority(struct b43_wldev *dev,
 	return ring;
 }
 
-int b43_dma_tx(struct b43_wldev *dev,
-	       struct sk_buff *skb, struct ieee80211_tx_control *ctl)
+int b43_dma_tx(struct b43_wldev *dev, struct sk_buff *skb)
 {
 	struct b43_dmaring *ring;
 	struct ieee80211_hdr *hdr;
 	int err = 0;
 	unsigned long flags;
+	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
 
 	hdr = (struct ieee80211_hdr *)skb->data;
-	if (ctl->flags & IEEE80211_TXCTL_SEND_AFTER_DTIM) {
+	if (info->flags & IEEE80211_TX_CTL_SEND_AFTER_DTIM) {
 		/* The multicast ring will be sent after the DTIM */
 		ring = dev->dma.tx_ring_mcast;
 		/* Set the more-data bit. Ucode will clear it on
@@ -1298,7 +1297,7 @@ int b43_dma_tx(struct b43_wldev *dev,
 		hdr->frame_control |= cpu_to_le16(IEEE80211_FCTL_MOREDATA);
 	} else {
 		/* Decide by priority where to put this frame. */
-		ring = select_ring_by_priority(dev, ctl->queue);
+		ring = select_ring_by_priority(dev, info->queue);
 	}
 
 	spin_lock_irqsave(&ring->lock, flags);
@@ -1316,9 +1315,9 @@ int b43_dma_tx(struct b43_wldev *dev,
 	/* Assign the queue number to the ring (if not already done before)
 	 * so TX status handling can use it. The queue to ring mapping is
 	 * static, so we don't need to store it per frame. */
-	ring->queue_prio = ctl->queue;
+	ring->queue_prio = info->queue;
 
-	err = dma_tx_fragment(ring, skb, ctl);
+	err = dma_tx_fragment(ring, skb);
 	if (unlikely(err == -ENOKEY)) {
 		/* Drop this packet, as we don't have the encryption key
 		 * anymore and must not transmit it unencrypted. */
@@ -1334,7 +1333,7 @@ int b43_dma_tx(struct b43_wldev *dev,
 	if ((free_slots(ring) < SLOTS_PER_PACKET) ||
 	    should_inject_overflow(ring)) {
 		/* This TX ring is full. */
-		ieee80211_stop_queue(dev->wl->hw, ctl->queue);
+		ieee80211_stop_queue(dev->wl->hw, info->queue);
 		ring->stopped = 1;
 		if (b43_debug(dev, B43_DBG_DMAVERBOSE)) {
 			b43dbg(dev->wl, "Stopped TX ring %d\n", ring->index);
@@ -1377,13 +1376,19 @@ void b43_dma_handle_txstatus(struct b43_wldev *dev,
 					 b43_txhdr_size(dev), 1);
 
 		if (meta->is_last_fragment) {
-			B43_WARN_ON(!meta->skb);
-			/* Call back to inform the ieee80211 subsystem about the
-			 * status of the transmission.
-			 * Some fields of txstat are already filled in dma_tx().
+			struct ieee80211_tx_info *info;
+
+			BUG_ON(!meta->skb);
+
+			info = IEEE80211_SKB_CB(meta->skb);
+
+			memset(&info->status, 0, sizeof(info->status));
+
+			/*
+			 * Call back to inform the ieee80211 subsystem about
+			 * the status of the transmission.
 			 */
-			frame_succeed = b43_fill_txstatus_report(
-						&(meta->txstat), status);
+			frame_succeed = b43_fill_txstatus_report(info, status);
 #ifdef CONFIG_B43_DEBUG
 			if (frame_succeed)
 				ring->nr_succeed_tx_packets++;
@@ -1391,8 +1396,8 @@ void b43_dma_handle_txstatus(struct b43_wldev *dev,
 				ring->nr_failed_tx_packets++;
 			ring->nr_total_packet_tries += status->frame_count;
 #endif /* DEBUG */
-			ieee80211_tx_status_irqsafe(dev->wl->hw, meta->skb,
-						    &(meta->txstat));
+			ieee80211_tx_status_irqsafe(dev->wl->hw, meta->skb);
+
 			/* skb is freed by ieee80211_tx_status_irqsafe() */
 			meta->skb = NULL;
 		} else {
diff --git a/drivers/net/wireless/b43/dma.h b/drivers/net/wireless/b43/dma.h
index 20acf885dba5..d1eb5c0848a5 100644
--- a/drivers/net/wireless/b43/dma.h
+++ b/drivers/net/wireless/b43/dma.h
@@ -181,7 +181,6 @@ struct b43_dmadesc_meta {
 	dma_addr_t dmaaddr;
 	/* ieee80211 TX status. Only used once per 802.11 frag. */
 	bool is_last_fragment;
-	struct ieee80211_tx_status txstat;
 };
 
 struct b43_dmaring;
@@ -285,7 +284,7 @@ void b43_dma_get_tx_stats(struct b43_wldev *dev,
 			  struct ieee80211_tx_queue_stats *stats);
 
 int b43_dma_tx(struct b43_wldev *dev,
-	       struct sk_buff *skb, struct ieee80211_tx_control *ctl);
+	       struct sk_buff *skb);
 void b43_dma_handle_txstatus(struct b43_wldev *dev,
 			     const struct b43_txstatus *status);
 
diff --git a/drivers/net/wireless/b43/main.c b/drivers/net/wireless/b43/main.c
index e428645352b4..3622d76de1e1 100644
--- a/drivers/net/wireless/b43/main.c
+++ b/drivers/net/wireless/b43/main.c
@@ -1368,18 +1368,18 @@ static void b43_write_beacon_template(struct b43_wldev *dev,
 	unsigned int rate;
 	u16 ctl;
 	int antenna;
+	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(dev->wl->current_beacon);
 
 	bcn = (const struct ieee80211_mgmt *)(dev->wl->current_beacon->data);
 	len = min((size_t) dev->wl->current_beacon->len,
 		  0x200 - sizeof(struct b43_plcp_hdr6));
-	rate = ieee80211_get_tx_rate(dev->wl->hw, &dev->wl->beacon_txctl)->hw_value;
+	rate = ieee80211_get_tx_rate(dev->wl->hw, info)->hw_value;
 
 	b43_write_template_common(dev, (const u8 *)bcn,
 				  len, ram_offset, shm_size_offset, rate);
 
 	/* Write the PHY TX control parameters. */
-	antenna = b43_antenna_from_ieee80211(dev,
-			dev->wl->beacon_txctl.antenna_sel_tx);
+	antenna = b43_antenna_from_ieee80211(dev, info->antenna_sel_tx);
 	antenna = b43_antenna_to_phyctl(antenna);
 	ctl = b43_shm_read16(dev, B43_SHM_SHARED, B43_SHM_SH_BEACPHYCTL);
 	/* We can't send beacons with short preamble. Would get PHY errors. */
@@ -1613,8 +1613,7 @@ static void b43_beacon_update_trigger_work(struct work_struct *work)
 
 /* Asynchronously update the packet templates in template RAM.
  * Locking: Requires wl->irq_lock to be locked. */
-static void b43_update_templates(struct b43_wl *wl, struct sk_buff *beacon,
-				 const struct ieee80211_tx_control *txctl)
+static void b43_update_templates(struct b43_wl *wl, struct sk_buff *beacon)
 {
 	/* This is the top half of the ansynchronous beacon update.
 	 * The bottom half is the beacon IRQ.
@@ -1625,7 +1624,6 @@ static void b43_update_templates(struct b43_wl *wl, struct sk_buff *beacon,
 	if (wl->current_beacon)
 		dev_kfree_skb_any(wl->current_beacon);
 	wl->current_beacon = beacon;
-	memcpy(&wl->beacon_txctl, txctl, sizeof(wl->beacon_txctl));
 	wl->beacon0_uploaded = 0;
 	wl->beacon1_uploaded = 0;
 	queue_work(wl->hw->workqueue, &wl->beacon_update_trigger);
@@ -2813,8 +2811,7 @@ static int b43_rng_init(struct b43_wl *wl)
 }
 
 static int b43_op_tx(struct ieee80211_hw *hw,
-		     struct sk_buff *skb,
-		     struct ieee80211_tx_control *ctl)
+		     struct sk_buff *skb)
 {
 	struct b43_wl *wl = hw_to_b43_wl(hw);
 	struct b43_wldev *dev = wl->current_dev;
@@ -2836,9 +2833,9 @@ static int b43_op_tx(struct ieee80211_hw *hw,
 	err = -ENODEV;
 	if (likely(b43_status(dev) >= B43_STAT_STARTED)) {
 		if (b43_using_pio_transfers(dev))
-			err = b43_pio_tx(dev, skb, ctl);
+			err = b43_pio_tx(dev, skb);
 		else
-			err = b43_dma_tx(dev, skb, ctl);
+			err = b43_dma_tx(dev, skb);
 	}
 
 	read_unlock_irqrestore(&wl->tx_lock, flags);
@@ -3429,10 +3426,8 @@ static int b43_op_config_interface(struct ieee80211_hw *hw,
 		if (b43_is_mode(wl, IEEE80211_IF_TYPE_AP)) {
 			B43_WARN_ON(conf->type != IEEE80211_IF_TYPE_AP);
 			b43_set_ssid(dev, conf->ssid, conf->ssid_len);
-			if (conf->beacon) {
-				b43_update_templates(wl, conf->beacon,
-						     conf->beacon_control);
-			}
+			if (conf->beacon)
+				b43_update_templates(wl, conf->beacon);
 		}
 		b43_write_mac_bssid_templates(dev);
 	}
@@ -4118,31 +4113,29 @@ static int b43_op_beacon_set_tim(struct ieee80211_hw *hw, int aid, int set)
 	struct b43_wl *wl = hw_to_b43_wl(hw);
 	struct sk_buff *beacon;
 	unsigned long flags;
-	struct ieee80211_tx_control txctl;
 
 	/* We could modify the existing beacon and set the aid bit in
 	 * the TIM field, but that would probably require resizing and
 	 * moving of data within the beacon template.
 	 * Simply request a new beacon and let mac80211 do the hard work. */
-	beacon = ieee80211_beacon_get(hw, wl->vif, &txctl);
+	beacon = ieee80211_beacon_get(hw, wl->vif);
 	if (unlikely(!beacon))
 		return -ENOMEM;
 	spin_lock_irqsave(&wl->irq_lock, flags);
-	b43_update_templates(wl, beacon, &txctl);
+	b43_update_templates(wl, beacon);
 	spin_unlock_irqrestore(&wl->irq_lock, flags);
 
 	return 0;
 }
 
 static int b43_op_ibss_beacon_update(struct ieee80211_hw *hw,
-				     struct sk_buff *beacon,
-				     struct ieee80211_tx_control *ctl)
+				     struct sk_buff *beacon)
 {
 	struct b43_wl *wl = hw_to_b43_wl(hw);
 	unsigned long flags;
 
 	spin_lock_irqsave(&wl->irq_lock, flags);
-	b43_update_templates(wl, beacon, ctl);
+	b43_update_templates(wl, beacon);
 	spin_unlock_irqrestore(&wl->irq_lock, flags);
 
 	return 0;
diff --git a/drivers/net/wireless/b43/pio.c b/drivers/net/wireless/b43/pio.c
index 08c8a087f30e..284786a94e7d 100644
--- a/drivers/net/wireless/b43/pio.c
+++ b/drivers/net/wireless/b43/pio.c
@@ -446,29 +446,27 @@ static void pio_tx_frame_4byte_queue(struct b43_pio_txpacket *pack,
 }
 
 static int pio_tx_frame(struct b43_pio_txqueue *q,
-			struct sk_buff *skb,
-			struct ieee80211_tx_control *ctl)
+			struct sk_buff *skb)
 {
 	struct b43_pio_txpacket *pack;
 	struct b43_txhdr txhdr;
 	u16 cookie;
 	int err;
 	unsigned int hdrlen;
+	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
 
 	B43_WARN_ON(list_empty(&q->packets_list));
 	pack = list_entry(q->packets_list.next,
 			  struct b43_pio_txpacket, list);
-	memset(&pack->txstat, 0, sizeof(pack->txstat));
-	memcpy(&pack->txstat.control, ctl, sizeof(*ctl));
 
 	cookie = generate_cookie(q, pack);
 	hdrlen = b43_txhdr_size(q->dev);
 	err = b43_generate_txhdr(q->dev, (u8 *)&txhdr, skb->data,
-				 skb->len, ctl, cookie);
+				 skb->len, info, cookie);
 	if (err)
 		return err;
 
-	if (ctl->flags & IEEE80211_TXCTL_SEND_AFTER_DTIM) {
+	if (info->flags & IEEE80211_TX_CTL_SEND_AFTER_DTIM) {
 		/* Tell the firmware about the cookie of the last
 		 * mcast frame, so it can clear the more-data bit in it. */
 		b43_shm_write16(q->dev, B43_SHM_SHARED,
@@ -492,17 +490,18 @@ static int pio_tx_frame(struct b43_pio_txqueue *q,
 	return 0;
 }
 
-int b43_pio_tx(struct b43_wldev *dev,
-	       struct sk_buff *skb, struct ieee80211_tx_control *ctl)
+int b43_pio_tx(struct b43_wldev *dev, struct sk_buff *skb)
 {
 	struct b43_pio_txqueue *q;
 	struct ieee80211_hdr *hdr;
 	unsigned long flags;
 	unsigned int hdrlen, total_len;
 	int err = 0;
+	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
 
 	hdr = (struct ieee80211_hdr *)skb->data;
-	if (ctl->flags & IEEE80211_TXCTL_SEND_AFTER_DTIM) {
+
+	if (info->flags & IEEE80211_TX_CTL_SEND_AFTER_DTIM) {
 		/* The multicast queue will be sent after the DTIM. */
 		q = dev->pio.tx_queue_mcast;
 		/* Set the frame More-Data bit. Ucode will clear it
@@ -510,7 +509,7 @@ int b43_pio_tx(struct b43_wldev *dev,
 		hdr->frame_control |= cpu_to_le16(IEEE80211_FCTL_MOREDATA);
 	} else {
 		/* Decide by priority where to put this frame. */
-		q = select_queue_by_priority(dev, ctl->queue);
+		q = select_queue_by_priority(dev, info->queue);
 	}
 
 	spin_lock_irqsave(&q->lock, flags);
@@ -533,7 +532,7 @@ int b43_pio_tx(struct b43_wldev *dev,
 	if (total_len > (q->buffer_size - q->buffer_used)) {
 		/* Not enough memory on the queue. */
 		err = -EBUSY;
-		ieee80211_stop_queue(dev->wl->hw, ctl->queue);
+		ieee80211_stop_queue(dev->wl->hw, info->queue);
 		q->stopped = 1;
 		goto out_unlock;
 	}
@@ -541,9 +540,9 @@ int b43_pio_tx(struct b43_wldev *dev,
 	/* Assign the queue number to the ring (if not already done before)
 	 * so TX status handling can use it. The mac80211-queue to b43-queue
 	 * mapping is static, so we don't need to store it per frame. */
-	q->queue_prio = ctl->queue;
+	q->queue_prio = info->queue;
 
-	err = pio_tx_frame(q, skb, ctl);
+	err = pio_tx_frame(q, skb);
 	if (unlikely(err == -ENOKEY)) {
 		/* Drop this packet, as we don't have the encryption key
 		 * anymore and must not transmit it unencrypted. */
@@ -561,7 +560,7 @@ int b43_pio_tx(struct b43_wldev *dev,
 	if (((q->buffer_size - q->buffer_used) < roundup(2 + 2 + 6, 4)) ||
 	    (q->free_packet_slots == 0)) {
 		/* The queue is full. */
-		ieee80211_stop_queue(dev->wl->hw, ctl->queue);
+		ieee80211_stop_queue(dev->wl->hw, info->queue);
 		q->stopped = 1;
 	}
 
@@ -578,6 +577,7 @@ void b43_pio_handle_txstatus(struct b43_wldev *dev,
 	struct b43_pio_txqueue *q;
 	struct b43_pio_txpacket *pack = NULL;
 	unsigned int total_len;
+	struct ieee80211_tx_info *info;
 
 	q = parse_cookie(dev, status->cookie, &pack);
 	if (unlikely(!q))
@@ -586,15 +586,17 @@ void b43_pio_handle_txstatus(struct b43_wldev *dev,
 
 	spin_lock(&q->lock); /* IRQs are already disabled. */
 
-	b43_fill_txstatus_report(&(pack->txstat), status);
+	info = (void *)pack->skb;
+	memset(&info->status, 0, sizeof(info->status));
+
+	b43_fill_txstatus_report(info, status);
 
 	total_len = pack->skb->len + b43_txhdr_size(dev);
 	total_len = roundup(total_len, 4);
 	q->buffer_used -= total_len;
 	q->free_packet_slots += 1;
 
-	ieee80211_tx_status_irqsafe(dev->wl->hw, pack->skb,
-				    &(pack->txstat));
+	ieee80211_tx_status_irqsafe(dev->wl->hw, pack->skb);
 	pack->skb = NULL;
 	list_add(&pack->list, &q->packets_list);
 
diff --git a/drivers/net/wireless/b43/pio.h b/drivers/net/wireless/b43/pio.h
index e2ec676cc9e4..6c174c91ca20 100644
--- a/drivers/net/wireless/b43/pio.h
+++ b/drivers/net/wireless/b43/pio.h
@@ -62,8 +62,6 @@ struct b43_pio_txpacket {
 	struct b43_pio_txqueue *queue;
 	/* The TX data packet. */
 	struct sk_buff *skb;
-	/* The status meta data. */
-	struct ieee80211_tx_status txstat;
 	/* Index in the (struct b43_pio_txqueue)->packets array. */
 	u8 index;
 
@@ -167,8 +165,7 @@ int b43_pio_init(struct b43_wldev *dev);
 void b43_pio_stop(struct b43_wldev *dev);
 void b43_pio_free(struct b43_wldev *dev);
 
-int b43_pio_tx(struct b43_wldev *dev,
-	       struct sk_buff *skb, struct ieee80211_tx_control *ctl);
+int b43_pio_tx(struct b43_wldev *dev, struct sk_buff *skb);
 void b43_pio_handle_txstatus(struct b43_wldev *dev,
 			     const struct b43_txstatus *status);
 void b43_pio_get_tx_stats(struct b43_wldev *dev,
@@ -193,8 +190,7 @@ static inline void b43_pio_stop(struct b43_wldev *dev)
 {
 }
 static inline int b43_pio_tx(struct b43_wldev *dev,
-			     struct sk_buff *skb,
-			     struct ieee80211_tx_control *ctl)
+			     struct sk_buff *skb)
 {
 	return 0;
 }
diff --git a/drivers/net/wireless/b43/xmit.c b/drivers/net/wireless/b43/xmit.c
index 9b682a3cf5e4..f9e1cff2aecb 100644
--- a/drivers/net/wireless/b43/xmit.c
+++ b/drivers/net/wireless/b43/xmit.c
@@ -185,14 +185,14 @@ int b43_generate_txhdr(struct b43_wldev *dev,
 		       u8 *_txhdr,
 		       const unsigned char *fragment_data,
 		       unsigned int fragment_len,
-		       const struct ieee80211_tx_control *txctl,
+		       const struct ieee80211_tx_info *info,
 		       u16 cookie)
 {
 	struct b43_txhdr *txhdr = (struct b43_txhdr *)_txhdr;
 	const struct b43_phy *phy = &dev->phy;
 	const struct ieee80211_hdr *wlhdr =
 	    (const struct ieee80211_hdr *)fragment_data;
-	int use_encryption = (!(txctl->flags & IEEE80211_TXCTL_DO_NOT_ENCRYPT));
+	int use_encryption = (!(info->flags & IEEE80211_TX_CTL_DO_NOT_ENCRYPT));
 	u16 fctl = le16_to_cpu(wlhdr->frame_control);
 	struct ieee80211_rate *fbrate;
 	u8 rate, rate_fb;
@@ -205,10 +205,10 @@ int b43_generate_txhdr(struct b43_wldev *dev,
 
 	memset(txhdr, 0, sizeof(*txhdr));
 
-	txrate = ieee80211_get_tx_rate(dev->wl->hw, txctl);
+	txrate = ieee80211_get_tx_rate(dev->wl->hw, info);
 	rate = txrate ? txrate->hw_value : B43_CCK_RATE_1MB;
 	rate_ofdm = b43_is_ofdm_rate(rate);
-	fbrate = ieee80211_get_alt_retry_rate(dev->wl->hw, txctl) ? : txrate;
+	fbrate = ieee80211_get_alt_retry_rate(dev->wl->hw, info) ? : txrate;
 	rate_fb = fbrate->hw_value;
 	rate_fb_ofdm = b43_is_ofdm_rate(rate_fb);
 
@@ -228,15 +228,13 @@ int b43_generate_txhdr(struct b43_wldev *dev,
 		 * use the original dur_id field. */
 		txhdr->dur_fb = wlhdr->duration_id;
 	} else {
-		txhdr->dur_fb = ieee80211_generic_frame_duration(dev->wl->hw,
-								 txctl->vif,
-								 fragment_len,
-								 fbrate);
+		txhdr->dur_fb = ieee80211_generic_frame_duration(
+			dev->wl->hw, info->control.vif, fragment_len, fbrate);
 	}
 
 	plcp_fragment_len = fragment_len + FCS_LEN;
 	if (use_encryption) {
-		u8 key_idx = txctl->hw_key->hw_key_idx;
+		u8 key_idx = info->control.hw_key->hw_key_idx;
 		struct b43_key *key;
 		int wlhdr_len;
 		size_t iv_len;
@@ -254,7 +252,7 @@ int b43_generate_txhdr(struct b43_wldev *dev,
 		}
 
 		/* Hardware appends ICV. */
-		plcp_fragment_len += txctl->icv_len;
+		plcp_fragment_len += info->control.icv_len;
 
 		key_idx = b43_kidx_to_fw(dev, key_idx);
 		mac_ctl |= (key_idx << B43_TXH_MAC_KEYIDX_SHIFT) &
@@ -262,7 +260,7 @@ int b43_generate_txhdr(struct b43_wldev *dev,
 		mac_ctl |= (key->algorithm << B43_TXH_MAC_KEYALG_SHIFT) &
 			   B43_TXH_MAC_KEYALG;
 		wlhdr_len = ieee80211_get_hdrlen(fctl);
-		iv_len = min((size_t) txctl->iv_len,
+		iv_len = min((size_t) info->control.iv_len,
 			     ARRAY_SIZE(txhdr->iv));
 		memcpy(txhdr->iv, ((u8 *) wlhdr) + wlhdr_len, iv_len);
 	}
@@ -293,10 +291,10 @@ int b43_generate_txhdr(struct b43_wldev *dev,
 		phy_ctl |= B43_TXH_PHY_ENC_OFDM;
 	else
 		phy_ctl |= B43_TXH_PHY_ENC_CCK;
-	if (txctl->flags & IEEE80211_TXCTL_SHORT_PREAMBLE)
+	if (info->flags & IEEE80211_TX_CTL_SHORT_PREAMBLE)
 		phy_ctl |= B43_TXH_PHY_SHORTPRMBL;
 
-	switch (b43_ieee80211_antenna_sanitize(dev, txctl->antenna_sel_tx)) {
+	switch (b43_ieee80211_antenna_sanitize(dev, info->antenna_sel_tx)) {
 	case 0: /* Default */
 		phy_ctl |= B43_TXH_PHY_ANT01AUTO;
 		break;
@@ -317,21 +315,21 @@ int b43_generate_txhdr(struct b43_wldev *dev,
 	}
 
 	/* MAC control */
-	if (!(txctl->flags & IEEE80211_TXCTL_NO_ACK))
+	if (!(info->flags & IEEE80211_TX_CTL_NO_ACK))
 		mac_ctl |= B43_TXH_MAC_ACK;
 	if (!(((fctl & IEEE80211_FCTL_FTYPE) == IEEE80211_FTYPE_CTL) &&
 	      ((fctl & IEEE80211_FCTL_STYPE) == IEEE80211_STYPE_PSPOLL)))
 		mac_ctl |= B43_TXH_MAC_HWSEQ;
-	if (txctl->flags & IEEE80211_TXCTL_FIRST_FRAGMENT)
+	if (info->flags & IEEE80211_TX_CTL_FIRST_FRAGMENT)
 		mac_ctl |= B43_TXH_MAC_STMSDU;
 	if (phy->type == B43_PHYTYPE_A)
 		mac_ctl |= B43_TXH_MAC_5GHZ;
-	if (txctl->flags & IEEE80211_TXCTL_LONG_RETRY_LIMIT)
+	if (info->flags & IEEE80211_TX_CTL_LONG_RETRY_LIMIT)
 		mac_ctl |= B43_TXH_MAC_LONGFRAME;
 
 	/* Generate the RTS or CTS-to-self frame */
-	if ((txctl->flags & IEEE80211_TXCTL_USE_RTS_CTS) ||
-	    (txctl->flags & IEEE80211_TXCTL_USE_CTS_PROTECT)) {
+	if ((info->flags & IEEE80211_TX_CTL_USE_RTS_CTS) ||
+	    (info->flags & IEEE80211_TX_CTL_USE_CTS_PROTECT)) {
 		unsigned int len;
 		struct ieee80211_hdr *hdr;
 		int rts_rate, rts_rate_fb;
@@ -339,14 +337,14 @@ int b43_generate_txhdr(struct b43_wldev *dev,
 		struct b43_plcp_hdr6 *plcp;
 		struct ieee80211_rate *rts_cts_rate;
 
-		rts_cts_rate = ieee80211_get_rts_cts_rate(dev->wl->hw, txctl);
+		rts_cts_rate = ieee80211_get_rts_cts_rate(dev->wl->hw, info);
 
 		rts_rate = rts_cts_rate ? rts_cts_rate->hw_value : B43_CCK_RATE_1MB;
 		rts_rate_ofdm = b43_is_ofdm_rate(rts_rate);
 		rts_rate_fb = b43_calc_fallback_rate(rts_rate);
 		rts_rate_fb_ofdm = b43_is_ofdm_rate(rts_rate_fb);
 
-		if (txctl->flags & IEEE80211_TXCTL_USE_CTS_PROTECT) {
+		if (info->flags & IEEE80211_TX_CTL_USE_CTS_PROTECT) {
 			struct ieee80211_cts *cts;
 
 			if (b43_is_old_txhdr_format(dev)) {
@@ -356,9 +354,9 @@ int b43_generate_txhdr(struct b43_wldev *dev,
 				cts = (struct ieee80211_cts *)
 					(txhdr->new_format.rts_frame);
 			}
-			ieee80211_ctstoself_get(dev->wl->hw, txctl->vif,
+			ieee80211_ctstoself_get(dev->wl->hw, info->control.vif,
 						fragment_data, fragment_len,
-						txctl, cts);
+						info, cts);
 			mac_ctl |= B43_TXH_MAC_SENDCTS;
 			len = sizeof(struct ieee80211_cts);
 		} else {
@@ -371,9 +369,9 @@ int b43_generate_txhdr(struct b43_wldev *dev,
 				rts = (struct ieee80211_rts *)
 					(txhdr->new_format.rts_frame);
 			}
-			ieee80211_rts_get(dev->wl->hw, txctl->vif,
+			ieee80211_rts_get(dev->wl->hw, info->control.vif,
 					  fragment_data, fragment_len,
-					  txctl, rts);
+					  info, rts);
 			mac_ctl |= B43_TXH_MAC_SENDRTS;
 			len = sizeof(struct ieee80211_rts);
 		}
@@ -687,27 +685,27 @@ void b43_handle_txstatus(struct b43_wldev *dev,
 /* Fill out the mac80211 TXstatus report based on the b43-specific
  * txstatus report data. This returns a boolean whether the frame was
  * successfully transmitted. */
-bool b43_fill_txstatus_report(struct ieee80211_tx_status *report,
+bool b43_fill_txstatus_report(struct ieee80211_tx_info *report,
 			      const struct b43_txstatus *status)
 {
 	bool frame_success = 1;
 
 	if (status->acked) {
 		/* The frame was ACKed. */
-		report->flags |= IEEE80211_TX_STATUS_ACK;
+		report->flags |= IEEE80211_TX_STAT_ACK;
 	} else {
 		/* The frame was not ACKed... */
-		if (!(report->control.flags & IEEE80211_TXCTL_NO_ACK)) {
+		if (!(report->flags & IEEE80211_TX_CTL_NO_ACK)) {
 			/* ...but we expected an ACK. */
 			frame_success = 0;
-			report->excessive_retries = 1;
+			report->status.excessive_retries = 1;
 		}
 	}
 	if (status->frame_count == 0) {
 		/* The frame was not transmitted at all. */
-		report->retry_count = 0;
+		report->status.retry_count = 0;
 	} else
-		report->retry_count = status->frame_count - 1;
+		report->status.retry_count = status->frame_count - 1;
 
 	return frame_success;
 }
diff --git a/drivers/net/wireless/b43/xmit.h b/drivers/net/wireless/b43/xmit.h
index b05f44e0d626..0215faf47541 100644
--- a/drivers/net/wireless/b43/xmit.h
+++ b/drivers/net/wireless/b43/xmit.h
@@ -178,7 +178,7 @@ int b43_generate_txhdr(struct b43_wldev *dev,
 		       u8 * txhdr,
 		       const unsigned char *fragment_data,
 		       unsigned int fragment_len,
-		       const struct ieee80211_tx_control *txctl, u16 cookie);
+		       const struct ieee80211_tx_info *txctl, u16 cookie);
 
 /* Transmit Status */
 struct b43_txstatus {
@@ -294,7 +294,7 @@ void b43_rx(struct b43_wldev *dev, struct sk_buff *skb, const void *_rxhdr);
 
 void b43_handle_txstatus(struct b43_wldev *dev,
 			 const struct b43_txstatus *status);
-bool b43_fill_txstatus_report(struct ieee80211_tx_status *report,
+bool b43_fill_txstatus_report(struct ieee80211_tx_info *report,
 			      const struct b43_txstatus *status);
 
 void b43_tx_suspend(struct b43_wldev *dev);
diff --git a/drivers/net/wireless/b43legacy/dma.c b/drivers/net/wireless/b43legacy/dma.c
index d6686f713b6d..c1c501d963bc 100644
--- a/drivers/net/wireless/b43legacy/dma.c
+++ b/drivers/net/wireless/b43legacy/dma.c
@@ -1205,10 +1205,10 @@ struct b43legacy_dmaring *parse_cookie(struct b43legacy_wldev *dev,
 }
 
 static int dma_tx_fragment(struct b43legacy_dmaring *ring,
-			    struct sk_buff *skb,
-			    struct ieee80211_tx_control *ctl)
+			    struct sk_buff *skb)
 {
 	const struct b43legacy_dma_ops *ops = ring->ops;
+	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
 	u8 *header;
 	int slot, old_top_slot, old_used_slots;
 	int err;
@@ -1231,7 +1231,7 @@ static int dma_tx_fragment(struct b43legacy_dmaring *ring,
 	header = &(ring->txhdr_cache[slot * sizeof(
 			       struct b43legacy_txhdr_fw3)]);
 	err = b43legacy_generate_txhdr(ring->dev, header,
-				 skb->data, skb->len, ctl,
+				 skb->data, skb->len, info,
 				 generate_cookie(ring, slot));
 	if (unlikely(err)) {
 		ring->current_slot = old_top_slot;
@@ -1255,7 +1255,6 @@ static int dma_tx_fragment(struct b43legacy_dmaring *ring,
 	desc = ops->idx2desc(ring, slot, &meta);
 	memset(meta, 0, sizeof(*meta));
 
-	memcpy(&meta->txstat.control, ctl, sizeof(*ctl));
 	meta->skb = skb;
 	meta->is_last_fragment = 1;
 
@@ -1323,14 +1322,14 @@ int should_inject_overflow(struct b43legacy_dmaring *ring)
 }
 
 int b43legacy_dma_tx(struct b43legacy_wldev *dev,
-		     struct sk_buff *skb,
-		     struct ieee80211_tx_control *ctl)
+		     struct sk_buff *skb)
 {
 	struct b43legacy_dmaring *ring;
+	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
 	int err = 0;
 	unsigned long flags;
 
-	ring = priority_to_txring(dev, ctl->queue);
+	ring = priority_to_txring(dev, info->queue);
 	spin_lock_irqsave(&ring->lock, flags);
 	B43legacy_WARN_ON(!ring->tx);
 	if (unlikely(free_slots(ring) < SLOTS_PER_PACKET)) {
@@ -1343,7 +1342,7 @@ int b43legacy_dma_tx(struct b43legacy_wldev *dev,
 	 * That would be a mac80211 bug. */
 	B43legacy_BUG_ON(ring->stopped);
 
-	err = dma_tx_fragment(ring, skb, ctl);
+	err = dma_tx_fragment(ring, skb);
 	if (unlikely(err == -ENOKEY)) {
 		/* Drop this packet, as we don't have the encryption key
 		 * anymore and must not transmit it unencrypted. */
@@ -1401,26 +1400,29 @@ void b43legacy_dma_handle_txstatus(struct b43legacy_wldev *dev,
 					 1);
 
 		if (meta->is_last_fragment) {
-			B43legacy_WARN_ON(!meta->skb);
+			struct ieee80211_tx_info *info;
+			BUG_ON(!meta->skb);
+			info = IEEE80211_SKB_CB(meta->skb);
 			/* Call back to inform the ieee80211 subsystem about the
 			 * status of the transmission.
 			 * Some fields of txstat are already filled in dma_tx().
 			 */
+
+			memset(&info->status, 0, sizeof(info->status));
+
 			if (status->acked) {
-				meta->txstat.flags |= IEEE80211_TX_STATUS_ACK;
+				info->flags |= IEEE80211_TX_STAT_ACK;
 			} else {
-				if (!(meta->txstat.control.flags
-				      & IEEE80211_TXCTL_NO_ACK))
-					 meta->txstat.excessive_retries = 1;
+				if (!(info->flags & IEEE80211_TX_CTL_NO_ACK))
+					 info->status.excessive_retries = 1;
 			}
 			if (status->frame_count == 0) {
 				/* The frame was not transmitted at all. */
-				meta->txstat.retry_count = 0;
+				info->status.retry_count = 0;
 			} else
-				meta->txstat.retry_count = status->frame_count
+				info->status.retry_count = status->frame_count
 							   - 1;
-			ieee80211_tx_status_irqsafe(dev->wl->hw, meta->skb,
-						    &(meta->txstat));
+			ieee80211_tx_status_irqsafe(dev->wl->hw, meta->skb);
 			/* skb is freed by ieee80211_tx_status_irqsafe() */
 			meta->skb = NULL;
 		} else {
diff --git a/drivers/net/wireless/b43legacy/dma.h b/drivers/net/wireless/b43legacy/dma.h
index 2dd488c5be2d..67537a7495ff 100644
--- a/drivers/net/wireless/b43legacy/dma.h
+++ b/drivers/net/wireless/b43legacy/dma.h
@@ -195,7 +195,6 @@ struct b43legacy_dmadesc_meta {
 	dma_addr_t dmaaddr;
 	/* ieee80211 TX status. Only used once per 802.11 frag. */
 	bool is_last_fragment;
-	struct ieee80211_tx_status txstat;
 };
 
 struct b43legacy_dmaring;
@@ -297,8 +296,7 @@ void b43legacy_dma_get_tx_stats(struct b43legacy_wldev *dev,
 				struct ieee80211_tx_queue_stats *stats);
 
 int b43legacy_dma_tx(struct b43legacy_wldev *dev,
-		     struct sk_buff *skb,
-		     struct ieee80211_tx_control *ctl);
+		     struct sk_buff *skb);
 void b43legacy_dma_handle_txstatus(struct b43legacy_wldev *dev,
 				   const struct b43legacy_txstatus *status);
 
diff --git a/drivers/net/wireless/b43legacy/main.c b/drivers/net/wireless/b43legacy/main.c
index b05a507ed44d..f706ca65f159 100644
--- a/drivers/net/wireless/b43legacy/main.c
+++ b/drivers/net/wireless/b43legacy/main.c
@@ -2358,8 +2358,7 @@ static int b43legacy_rng_init(struct b43legacy_wl *wl)
 }
 
 static int b43legacy_op_tx(struct ieee80211_hw *hw,
-			   struct sk_buff *skb,
-			   struct ieee80211_tx_control *ctl)
+			   struct sk_buff *skb)
 {
 	struct b43legacy_wl *wl = hw_to_b43legacy_wl(hw);
 	struct b43legacy_wldev *dev = wl->current_dev;
@@ -2373,10 +2372,10 @@ static int b43legacy_op_tx(struct ieee80211_hw *hw,
 	/* DMA-TX is done without a global lock. */
 	if (b43legacy_using_pio(dev)) {
 		spin_lock_irqsave(&wl->irq_lock, flags);
-		err = b43legacy_pio_tx(dev, skb, ctl);
+		err = b43legacy_pio_tx(dev, skb);
 		spin_unlock_irqrestore(&wl->irq_lock, flags);
 	} else
-		err = b43legacy_dma_tx(dev, skb, ctl);
+		err = b43legacy_dma_tx(dev, skb);
 out:
 	if (unlikely(err))
 		return NETDEV_TX_BUSY;
@@ -3409,7 +3408,7 @@ static int b43legacy_op_beacon_set_tim(struct ieee80211_hw *hw,
 	 * field, but that would probably require resizing and moving of data
 	 * within the beacon template. Simply request a new beacon and let
 	 * mac80211 do the hard work. */
-	beacon = ieee80211_beacon_get(hw, wl->vif, NULL);
+	beacon = ieee80211_beacon_get(hw, wl->vif);
 	if (unlikely(!beacon))
 		return -ENOMEM;
 	spin_lock_irqsave(&wl->irq_lock, flags);
@@ -3420,8 +3419,7 @@ static int b43legacy_op_beacon_set_tim(struct ieee80211_hw *hw,
 }
 
 static int b43legacy_op_ibss_beacon_update(struct ieee80211_hw *hw,
-					   struct sk_buff *beacon,
-					   struct ieee80211_tx_control *ctl)
+					   struct sk_buff *beacon)
 {
 	struct b43legacy_wl *wl = hw_to_b43legacy_wl(hw);
 	unsigned long flags;
diff --git a/drivers/net/wireless/b43legacy/pio.c b/drivers/net/wireless/b43legacy/pio.c
index 8d3d27d3cd67..a86c7647fa2d 100644
--- a/drivers/net/wireless/b43legacy/pio.c
+++ b/drivers/net/wireless/b43legacy/pio.c
@@ -196,7 +196,7 @@ static int pio_tx_write_fragment(struct b43legacy_pioqueue *queue,
 	B43legacy_WARN_ON(skb_shinfo(skb)->nr_frags != 0);
 	err = b43legacy_generate_txhdr(queue->dev,
 				 txhdr, skb->data, skb->len,
-				 &packet->txstat.control,
+				 IEEE80211_SKB_CB(skb),
 				 generate_cookie(queue, packet));
 	if (err)
 		return err;
@@ -463,8 +463,7 @@ err_destroy0:
 }
 
 int b43legacy_pio_tx(struct b43legacy_wldev *dev,
-		     struct sk_buff *skb,
-		     struct ieee80211_tx_control *ctl)
+		     struct sk_buff *skb)
 {
 	struct b43legacy_pioqueue *queue = dev->pio.queue1;
 	struct b43legacy_pio_txpacket *packet;
@@ -476,9 +475,6 @@ int b43legacy_pio_tx(struct b43legacy_wldev *dev,
 			    list);
 	packet->skb = skb;
 
-	memset(&packet->txstat, 0, sizeof(packet->txstat));
-	memcpy(&packet->txstat.control, ctl, sizeof(*ctl));
-
 	list_move_tail(&packet->list, &queue->txqueue);
 	queue->nr_txfree--;
 	queue->nr_tx_packets++;
@@ -494,6 +490,7 @@ void b43legacy_pio_handle_txstatus(struct b43legacy_wldev *dev,
 {
 	struct b43legacy_pioqueue *queue;
 	struct b43legacy_pio_txpacket *packet;
+	struct ieee80211_tx_info *info;
 
 	queue = parse_cookie(dev, status->cookie, &packet);
 	B43legacy_WARN_ON(!queue);
@@ -505,11 +502,13 @@ void b43legacy_pio_handle_txstatus(struct b43legacy_wldev *dev,
 	queue->tx_devq_used -= (packet->skb->len +
 				sizeof(struct b43legacy_txhdr_fw3));
 
+	info = IEEE80211_SKB_CB(packet->skb);
+	memset(&info->status, 0, sizeof(info->status));
+
 	if (status->acked)
-		packet->txstat.flags |= IEEE80211_TX_STATUS_ACK;
-	packet->txstat.retry_count = status->frame_count - 1;
-	ieee80211_tx_status_irqsafe(dev->wl->hw, packet->skb,
-				    &(packet->txstat));
+		info->flags |= IEEE80211_TX_STAT_ACK;
+	info->status.retry_count = status->frame_count - 1;
+	ieee80211_tx_status_irqsafe(dev->wl->hw, packet->skb);
 	packet->skb = NULL;
 
 	free_txpacket(packet, 1);
diff --git a/drivers/net/wireless/b43legacy/pio.h b/drivers/net/wireless/b43legacy/pio.h
index 5bfed0c40030..49310dfaf2d1 100644
--- a/drivers/net/wireless/b43legacy/pio.h
+++ b/drivers/net/wireless/b43legacy/pio.h
@@ -41,7 +41,6 @@ struct b43legacy_xmitstatus;
 struct b43legacy_pio_txpacket {
 	struct b43legacy_pioqueue *queue;
 	struct sk_buff *skb;
-	struct ieee80211_tx_status txstat;
 	struct list_head list;
 };
 
@@ -104,8 +103,7 @@ int b43legacy_pio_init(struct b43legacy_wldev *dev);
 void b43legacy_pio_free(struct b43legacy_wldev *dev);
 
 int b43legacy_pio_tx(struct b43legacy_wldev *dev,
-		   struct sk_buff *skb,
-		   struct ieee80211_tx_control *ctl);
+		   struct sk_buff *skb);
 void b43legacy_pio_handle_txstatus(struct b43legacy_wldev *dev,
 				 const struct b43legacy_txstatus *status);
 void b43legacy_pio_get_tx_stats(struct b43legacy_wldev *dev,
diff --git a/drivers/net/wireless/b43legacy/xmit.c b/drivers/net/wireless/b43legacy/xmit.c
index 55dc251bf510..82dc04d59446 100644
--- a/drivers/net/wireless/b43legacy/xmit.c
+++ b/drivers/net/wireless/b43legacy/xmit.c
@@ -188,11 +188,11 @@ static int generate_txhdr_fw3(struct b43legacy_wldev *dev,
 			       struct b43legacy_txhdr_fw3 *txhdr,
 			       const unsigned char *fragment_data,
 			       unsigned int fragment_len,
-			       const struct ieee80211_tx_control *txctl,
+			       const struct ieee80211_tx_info *info,
 			       u16 cookie)
 {
 	const struct ieee80211_hdr *wlhdr;
-	int use_encryption = (!(txctl->flags & IEEE80211_TXCTL_DO_NOT_ENCRYPT));
+	int use_encryption = (!(info->flags & IEEE80211_TX_CTL_DO_NOT_ENCRYPT));
 	u16 fctl;
 	u8 rate;
 	struct ieee80211_rate *rate_fb;
@@ -208,11 +208,11 @@ static int generate_txhdr_fw3(struct b43legacy_wldev *dev,
 
 	memset(txhdr, 0, sizeof(*txhdr));
 
-	tx_rate = ieee80211_get_tx_rate(dev->wl->hw, txctl);
+	tx_rate = ieee80211_get_tx_rate(dev->wl->hw, info);
 
 	rate = tx_rate->hw_value;
 	rate_ofdm = b43legacy_is_ofdm_rate(rate);
-	rate_fb = ieee80211_get_alt_retry_rate(dev->wl->hw, txctl) ? : tx_rate;
+	rate_fb = ieee80211_get_alt_retry_rate(dev->wl->hw, info) ? : tx_rate;
 	rate_fb_ofdm = b43legacy_is_ofdm_rate(rate_fb->hw_value);
 
 	txhdr->mac_frame_ctl = wlhdr->frame_control;
@@ -228,14 +228,14 @@ static int generate_txhdr_fw3(struct b43legacy_wldev *dev,
 		txhdr->dur_fb = wlhdr->duration_id;
 	} else {
 		txhdr->dur_fb = ieee80211_generic_frame_duration(dev->wl->hw,
-							 txctl->vif,
+							 info->control.vif,
 							 fragment_len,
 							 rate_fb);
 	}
 
 	plcp_fragment_len = fragment_len + FCS_LEN;
 	if (use_encryption) {
-		u8 key_idx = txctl->hw_key->hw_key_idx;
+		u8 key_idx = info->control.hw_key->hw_key_idx;
 		struct b43legacy_key *key;
 		int wlhdr_len;
 		size_t iv_len;
@@ -245,7 +245,7 @@ static int generate_txhdr_fw3(struct b43legacy_wldev *dev,
 
 		if (key->enabled) {
 			/* Hardware appends ICV. */
-			plcp_fragment_len += txctl->icv_len;
+			plcp_fragment_len += info->control.icv_len;
 
 			key_idx = b43legacy_kidx_to_fw(dev, key_idx);
 			mac_ctl |= (key_idx << B43legacy_TX4_MAC_KEYIDX_SHIFT) &
@@ -254,7 +254,7 @@ static int generate_txhdr_fw3(struct b43legacy_wldev *dev,
 				   B43legacy_TX4_MAC_KEYALG_SHIFT) &
 				   B43legacy_TX4_MAC_KEYALG;
 			wlhdr_len = ieee80211_get_hdrlen(fctl);
-			iv_len = min((size_t)txctl->iv_len,
+			iv_len = min((size_t)info->control.iv_len,
 				     ARRAY_SIZE(txhdr->iv));
 			memcpy(txhdr->iv, ((u8 *)wlhdr) + wlhdr_len, iv_len);
 		} else {
@@ -278,7 +278,7 @@ static int generate_txhdr_fw3(struct b43legacy_wldev *dev,
 		phy_ctl |= B43legacy_TX4_PHY_OFDM;
 	if (dev->short_preamble)
 		phy_ctl |= B43legacy_TX4_PHY_SHORTPRMBL;
-	switch (txctl->antenna_sel_tx) {
+	switch (info->antenna_sel_tx) {
 	case 0:
 		phy_ctl |= B43legacy_TX4_PHY_ANTLAST;
 		break;
@@ -293,21 +293,21 @@ static int generate_txhdr_fw3(struct b43legacy_wldev *dev,
 	}
 
 	/* MAC control */
-	if (!(txctl->flags & IEEE80211_TXCTL_NO_ACK))
+	if (!(info->flags & IEEE80211_TX_CTL_NO_ACK))
 		mac_ctl |= B43legacy_TX4_MAC_ACK;
 	if (!(((fctl & IEEE80211_FCTL_FTYPE) == IEEE80211_FTYPE_CTL) &&
 	      ((fctl & IEEE80211_FCTL_STYPE) == IEEE80211_STYPE_PSPOLL)))
 		mac_ctl |= B43legacy_TX4_MAC_HWSEQ;
-	if (txctl->flags & IEEE80211_TXCTL_FIRST_FRAGMENT)
+	if (info->flags & IEEE80211_TX_CTL_FIRST_FRAGMENT)
 		mac_ctl |= B43legacy_TX4_MAC_STMSDU;
 	if (rate_fb_ofdm)
 		mac_ctl |= B43legacy_TX4_MAC_FALLBACKOFDM;
-	if (txctl->flags & IEEE80211_TXCTL_LONG_RETRY_LIMIT)
+	if (info->flags & IEEE80211_TX_CTL_LONG_RETRY_LIMIT)
 		mac_ctl |= B43legacy_TX4_MAC_LONGFRAME;
 
 	/* Generate the RTS or CTS-to-self frame */
-	if ((txctl->flags & IEEE80211_TXCTL_USE_RTS_CTS) ||
-	    (txctl->flags & IEEE80211_TXCTL_USE_CTS_PROTECT)) {
+	if ((info->flags & IEEE80211_TX_CTL_USE_RTS_CTS) ||
+	    (info->flags & IEEE80211_TX_CTL_USE_CTS_PROTECT)) {
 		unsigned int len;
 		struct ieee80211_hdr *hdr;
 		int rts_rate;
@@ -315,26 +315,26 @@ static int generate_txhdr_fw3(struct b43legacy_wldev *dev,
 		int rts_rate_ofdm;
 		int rts_rate_fb_ofdm;
 
-		rts_rate = ieee80211_get_rts_cts_rate(dev->wl->hw, txctl)->hw_value;
+		rts_rate = ieee80211_get_rts_cts_rate(dev->wl->hw, info)->hw_value;
 		rts_rate_ofdm = b43legacy_is_ofdm_rate(rts_rate);
 		rts_rate_fb = b43legacy_calc_fallback_rate(rts_rate);
 		rts_rate_fb_ofdm = b43legacy_is_ofdm_rate(rts_rate_fb);
 		if (rts_rate_fb_ofdm)
 			mac_ctl |= B43legacy_TX4_MAC_CTSFALLBACKOFDM;
 
-		if (txctl->flags & IEEE80211_TXCTL_USE_CTS_PROTECT) {
+		if (info->flags & IEEE80211_TX_CTL_USE_CTS_PROTECT) {
 			ieee80211_ctstoself_get(dev->wl->hw,
-						txctl->vif,
+						info->control.vif,
 						fragment_data,
-						fragment_len, txctl,
+						fragment_len, info,
 						(struct ieee80211_cts *)
 						(txhdr->rts_frame));
 			mac_ctl |= B43legacy_TX4_MAC_SENDCTS;
 			len = sizeof(struct ieee80211_cts);
 		} else {
 			ieee80211_rts_get(dev->wl->hw,
-					  txctl->vif,
-					  fragment_data, fragment_len, txctl,
+					  info->control.vif,
+					  fragment_data, fragment_len, info,
 					  (struct ieee80211_rts *)
 					  (txhdr->rts_frame));
 			mac_ctl |= B43legacy_TX4_MAC_SENDRTS;
@@ -365,12 +365,12 @@ int b43legacy_generate_txhdr(struct b43legacy_wldev *dev,
 			      u8 *txhdr,
 			      const unsigned char *fragment_data,
 			      unsigned int fragment_len,
-			      const struct ieee80211_tx_control *txctl,
+			      const struct ieee80211_tx_info *info,
 			      u16 cookie)
 {
 	return generate_txhdr_fw3(dev, (struct b43legacy_txhdr_fw3 *)txhdr,
 			   fragment_data, fragment_len,
-			   txctl, cookie);
+			   info, cookie);
 }
 
 static s8 b43legacy_rssi_postprocess(struct b43legacy_wldev *dev,
diff --git a/drivers/net/wireless/b43legacy/xmit.h b/drivers/net/wireless/b43legacy/xmit.h
index bab47928a0c9..e56777e0feab 100644
--- a/drivers/net/wireless/b43legacy/xmit.h
+++ b/drivers/net/wireless/b43legacy/xmit.h
@@ -80,7 +80,7 @@ int b43legacy_generate_txhdr(struct b43legacy_wldev *dev,
 			      u8 *txhdr,
 			      const unsigned char *fragment_data,
 			      unsigned int fragment_len,
-			      const struct ieee80211_tx_control *txctl,
+			      const struct ieee80211_tx_info *info,
 			      u16 cookie);
 
 
diff --git a/drivers/net/wireless/iwlwifi/iwl-3945-rs.c b/drivers/net/wireless/iwlwifi/iwl-3945-rs.c
index f3ca02fe9619..10c64bdb314c 100644
--- a/drivers/net/wireless/iwlwifi/iwl-3945-rs.c
+++ b/drivers/net/wireless/iwlwifi/iwl-3945-rs.c
@@ -445,8 +445,7 @@ static int rs_adjust_next_rate(struct iwl3945_priv *priv, int rate)
  */
 static void rs_tx_status(void *priv_rate,
 			 struct net_device *dev,
-			 struct sk_buff *skb,
-			 struct ieee80211_tx_status *tx_resp)
+			 struct sk_buff *skb)
 {
 	u8 retries, current_count;
 	int scale_rate_index, first_index, last_index;
@@ -457,14 +456,15 @@ static void rs_tx_status(void *priv_rate,
 	struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr);
 	struct iwl3945_rs_sta *rs_sta;
 	struct ieee80211_supported_band *sband;
+	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
 
 	IWL_DEBUG_RATE("enter\n");
 
 	sband = local->hw.wiphy->bands[local->hw.conf.channel->band];
 
 
-	retries = tx_resp->retry_count;
-	first_index = sband->bitrates[tx_resp->control.tx_rate_idx].hw_value;
+	retries = info->status.retry_count;
+	first_index = sband->bitrates[info->tx_rate_idx].hw_value;
 	if ((first_index < 0) || (first_index >= IWL_RATE_COUNT)) {
 		IWL_DEBUG_RATE("leave: Rate out of bounds: %d\n", first_index);
 		return;
@@ -525,11 +525,11 @@ static void rs_tx_status(void *priv_rate,
 	/* Update the last index window with success/failure based on ACK */
 	IWL_DEBUG_RATE("Update rate %d with %s.\n",
 		       last_index,
-		       (tx_resp->flags & IEEE80211_TX_STATUS_ACK) ?
+		       (info->flags & IEEE80211_TX_STAT_ACK) ?
 		       "success" : "failure");
 	iwl3945_collect_tx_data(rs_sta,
 			    &rs_sta->win[last_index],
-			    tx_resp->flags & IEEE80211_TX_STATUS_ACK, 1);
+			    info->flags & IEEE80211_TX_STAT_ACK, 1);
 
 	/* We updated the rate scale window -- if its been more than
 	 * flush_time since the last run, schedule the flush
diff --git a/drivers/net/wireless/iwlwifi/iwl-3945.c b/drivers/net/wireless/iwlwifi/iwl-3945.c
index f8e691f88ab3..0ba6889dfd41 100644
--- a/drivers/net/wireless/iwlwifi/iwl-3945.c
+++ b/drivers/net/wireless/iwlwifi/iwl-3945.c
@@ -283,8 +283,7 @@ static void iwl3945_tx_queue_reclaim(struct iwl3945_priv *priv,
 		q->read_ptr = iwl_queue_inc_wrap(q->read_ptr, q->n_bd)) {
 
 		tx_info = &txq->txb[txq->q.read_ptr];
-		ieee80211_tx_status_irqsafe(priv->hw, tx_info->skb[0],
-					    &tx_info->status);
+		ieee80211_tx_status_irqsafe(priv->hw, tx_info->skb[0]);
 		tx_info->skb[0] = NULL;
 		iwl3945_hw_txq_free_tfd(priv, txq);
 	}
@@ -306,7 +305,7 @@ static void iwl3945_rx_reply_tx(struct iwl3945_priv *priv,
 	int txq_id = SEQ_TO_QUEUE(sequence);
 	int index = SEQ_TO_INDEX(sequence);
 	struct iwl3945_tx_queue *txq = &priv->txq[txq_id];
-	struct ieee80211_tx_status *tx_status;
+	struct ieee80211_tx_info *info;
 	struct iwl3945_tx_resp *tx_resp = (void *)&pkt->u.raw[0];
 	u32  status = le32_to_cpu(tx_resp->status);
 	int rate_idx;
@@ -319,21 +318,22 @@ static void iwl3945_rx_reply_tx(struct iwl3945_priv *priv,
 		return;
 	}
 
-	tx_status = &(txq->txb[txq->q.read_ptr].status);
+	info = IEEE80211_SKB_CB(txq->txb[txq->q.read_ptr].skb[0]);
+	memset(&info->status, 0, sizeof(info->status));
 
-	tx_status->retry_count = tx_resp->failure_frame;
+	info->status.retry_count = tx_resp->failure_frame;
 	/* tx_status->rts_retry_count = tx_resp->failure_rts; */
-	tx_status->flags = ((status & TX_STATUS_MSK) == TX_STATUS_SUCCESS) ?
-				IEEE80211_TX_STATUS_ACK : 0;
+	info->flags |= ((status & TX_STATUS_MSK) == TX_STATUS_SUCCESS) ?
+				IEEE80211_TX_STAT_ACK : 0;
 
 	IWL_DEBUG_TX("Tx queue %d Status %s (0x%08x) plcp rate %d retries %d\n",
 			txq_id, iwl3945_get_tx_fail_reason(status), status,
 			tx_resp->rate, tx_resp->failure_frame);
 
 	rate_idx = iwl3945_hwrate_to_plcp_idx(tx_resp->rate);
-	if (tx_status->control.band == IEEE80211_BAND_5GHZ)
+	if (info->band == IEEE80211_BAND_5GHZ)
 		rate_idx -= IWL_FIRST_OFDM_RATE;
-	tx_status->control.tx_rate_idx = rate_idx;
+	info->tx_rate_idx = rate_idx;
 	IWL_DEBUG_TX_REPLY("Tx queue reclaim %d\n", index);
 	iwl3945_tx_queue_reclaim(priv, txq_id, index);
 
@@ -960,11 +960,11 @@ u8 iwl3945_hw_find_station(struct iwl3945_priv *priv, const u8 *addr)
 */
 void iwl3945_hw_build_tx_cmd_rate(struct iwl3945_priv *priv,
 			      struct iwl3945_cmd *cmd,
-			      struct ieee80211_tx_control *ctrl,
+			      struct ieee80211_tx_info *info,
 			      struct ieee80211_hdr *hdr, int sta_id, int tx_id)
 {
 	unsigned long flags;
-	u16 hw_value = ieee80211_get_tx_rate(priv->hw, ctrl)->hw_value;
+	u16 hw_value = ieee80211_get_tx_rate(priv->hw, info)->hw_value;
 	u16 rate_index = min(hw_value & 0xffff, IWL_RATE_COUNT - 1);
 	u16 rate_mask;
 	int rate;
@@ -977,7 +977,7 @@ void iwl3945_hw_build_tx_cmd_rate(struct iwl3945_priv *priv,
 	tx_flags = cmd->cmd.tx.tx_flags;
 
 	/* We need to figure out how to get the sta->supp_rates while
-	 * in this running context; perhaps encoding into ctrl->tx_rate? */
+	 * in this running context */
 	rate_mask = IWL_RATES_MASK;
 
 	spin_lock_irqsave(&priv->sta_lock, flags);
diff --git a/drivers/net/wireless/iwlwifi/iwl-3945.h b/drivers/net/wireless/iwlwifi/iwl-3945.h
index 9fdc1405e853..835c5b4320e9 100644
--- a/drivers/net/wireless/iwlwifi/iwl-3945.h
+++ b/drivers/net/wireless/iwlwifi/iwl-3945.h
@@ -124,7 +124,6 @@ int iwl3945_x2_queue_used(const struct iwl3945_queue *q, int i);
 
 /* One for each TFD */
 struct iwl3945_tx_info {
-	struct ieee80211_tx_status status;
 	struct sk_buff *skb[MAX_NUM_OF_TBS];
 };
 
@@ -645,7 +644,7 @@ extern unsigned int iwl3945_hw_get_beacon_cmd(struct iwl3945_priv *priv,
 extern int iwl3945_hw_get_rx_read(struct iwl3945_priv *priv);
 extern void iwl3945_hw_build_tx_cmd_rate(struct iwl3945_priv *priv,
 				     struct iwl3945_cmd *cmd,
-				     struct ieee80211_tx_control *ctrl,
+				     struct ieee80211_tx_info *info,
 				     struct ieee80211_hdr *hdr,
 				     int sta_id, int tx_id);
 extern int iwl3945_hw_reg_send_txpower(struct iwl3945_priv *priv);
diff --git a/drivers/net/wireless/iwlwifi/iwl-4965-rs.c b/drivers/net/wireless/iwlwifi/iwl-4965-rs.c
index 7993a1d83025..f28b3cc272df 100644
--- a/drivers/net/wireless/iwlwifi/iwl-4965-rs.c
+++ b/drivers/net/wireless/iwlwifi/iwl-4965-rs.c
@@ -785,8 +785,7 @@ out:
  * mac80211 sends us Tx status
  */
 static void rs_tx_status(void *priv_rate, struct net_device *dev,
-			 struct sk_buff *skb,
-			 struct ieee80211_tx_status *tx_resp)
+			 struct sk_buff *skb)
 {
 	int status;
 	u8 retries;
@@ -798,6 +797,7 @@ static void rs_tx_status(void *priv_rate, struct net_device *dev,
 	struct iwl_priv *priv = (struct iwl_priv *)priv_rate;
 	struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr);
 	struct ieee80211_hw *hw = local_to_hw(local);
+	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
 	struct iwl4965_rate_scale_data *window = NULL;
 	struct iwl4965_rate_scale_data *search_win = NULL;
 	u32 tx_rate;
@@ -813,11 +813,11 @@ static void rs_tx_status(void *priv_rate, struct net_device *dev,
 		return;
 
 	/* This packet was aggregated but doesn't carry rate scale info */
-	if ((tx_resp->control.flags & IEEE80211_TXCTL_AMPDU) &&
-	    !(tx_resp->flags & IEEE80211_TX_STATUS_AMPDU))
+	if ((info->flags & IEEE80211_TX_CTL_AMPDU) &&
+	    !(info->flags & IEEE80211_TX_STAT_AMPDU))
 		return;
 
-	retries = tx_resp->retry_count;
+	retries = info->status.retry_count;
 
 	if (retries > 15)
 		retries = 15;
@@ -862,20 +862,20 @@ static void rs_tx_status(void *priv_rate, struct net_device *dev,
 	if (priv->band == IEEE80211_BAND_5GHZ)
 		rs_index -= IWL_FIRST_OFDM_RATE;
 
-	if ((tx_resp->control.tx_rate_idx < 0) ||
+	if ((info->tx_rate_idx < 0) ||
 	    (tbl_type.is_SGI ^
-		!!(tx_resp->control.flags & IEEE80211_TXCTL_SHORT_GI)) ||
+		!!(info->flags & IEEE80211_TX_CTL_SHORT_GI)) ||
 	    (tbl_type.is_fat ^
-		!!(tx_resp->control.flags & IEEE80211_TXCTL_40_MHZ_WIDTH)) ||
+		!!(info->flags & IEEE80211_TX_CTL_40_MHZ_WIDTH)) ||
 	    (tbl_type.is_dup ^
-		!!(tx_resp->control.flags & IEEE80211_TXCTL_DUP_DATA)) ||
-	    (tbl_type.ant_type ^ tx_resp->control.antenna_sel_tx) ||
+		!!(info->flags & IEEE80211_TX_CTL_DUP_DATA)) ||
+	    (tbl_type.ant_type ^ info->antenna_sel_tx) ||
 	    (!!(tx_rate & RATE_MCS_HT_MSK) ^
-		!!(tx_resp->control.flags & IEEE80211_TXCTL_OFDM_HT)) ||
+		!!(info->flags & IEEE80211_TX_CTL_OFDM_HT)) ||
 	    (!!(tx_rate & RATE_MCS_GF_MSK) ^
-		!!(tx_resp->control.flags & IEEE80211_TXCTL_GREEN_FIELD)) ||
+		!!(info->flags & IEEE80211_TX_CTL_GREEN_FIELD)) ||
 	    (hw->wiphy->bands[priv->band]->bitrates[rs_index].bitrate !=
-	     hw->wiphy->bands[tx_resp->control.band]->bitrates[tx_resp->control.tx_rate_idx].bitrate)) {
+	     hw->wiphy->bands[info->band]->bitrates[info->tx_rate_idx].bitrate)) {
 		IWL_DEBUG_RATE("initial rate does not match 0x%x\n", tx_rate);
 		goto out;
 	}
@@ -929,10 +929,7 @@ static void rs_tx_status(void *priv_rate, struct net_device *dev,
 	rs_get_tbl_info_from_mcs(tx_rate, priv->band, &tbl_type, &rs_index);
 
 	/* Update frame history window with "success" if Tx got ACKed ... */
-	if (tx_resp->flags & IEEE80211_TX_STATUS_ACK)
-		status = 1;
-	else
-		status = 0;
+	status = !!(info->flags & IEEE80211_TX_STAT_ACK);
 
 	/* If type matches "search" table,
 	 * add final tx status to "search" history */
@@ -943,10 +940,10 @@ static void rs_tx_status(void *priv_rate, struct net_device *dev,
 			tpt = search_tbl->expected_tpt[rs_index];
 		else
 			tpt = 0;
-		if (tx_resp->control.flags & IEEE80211_TXCTL_AMPDU)
+		if (info->flags & IEEE80211_TX_CTL_AMPDU)
 			rs_collect_tx_data(search_win, rs_index, tpt,
-					   tx_resp->ampdu_ack_len,
-					   tx_resp->ampdu_ack_map);
+					   info->status.ampdu_ack_len,
+					   info->status.ampdu_ack_map);
 		else
 			rs_collect_tx_data(search_win, rs_index, tpt,
 					   1, status);
@@ -959,10 +956,10 @@ static void rs_tx_status(void *priv_rate, struct net_device *dev,
 			tpt = curr_tbl->expected_tpt[rs_index];
 		else
 			tpt = 0;
-		if (tx_resp->control.flags & IEEE80211_TXCTL_AMPDU)
+		if (info->flags & IEEE80211_TX_CTL_AMPDU)
 			rs_collect_tx_data(window, rs_index, tpt,
-					   tx_resp->ampdu_ack_len,
-					   tx_resp->ampdu_ack_map);
+					   info->status.ampdu_ack_len,
+					   info->status.ampdu_ack_map);
 		else
 			rs_collect_tx_data(window, rs_index, tpt,
 					   1, status);
@@ -971,10 +968,10 @@ static void rs_tx_status(void *priv_rate, struct net_device *dev,
 	/* If not searching for new mode, increment success/failed counter
 	 * ... these help determine when to start searching again */
 	if (lq_sta->stay_in_tbl) {
-		if (tx_resp->control.flags & IEEE80211_TXCTL_AMPDU) {
-			lq_sta->total_success += tx_resp->ampdu_ack_map;
+		if (info->flags & IEEE80211_TX_CTL_AMPDU) {
+			lq_sta->total_success += info->status.ampdu_ack_map;
 			lq_sta->total_failed +=
-			     (tx_resp->ampdu_ack_len - tx_resp->ampdu_ack_map);
+			     (info->status.ampdu_ack_len - info->status.ampdu_ack_map);
 		} else {
 			if (status)
 				lq_sta->total_success++;
diff --git a/drivers/net/wireless/iwlwifi/iwl-4965.c b/drivers/net/wireless/iwlwifi/iwl-4965.c
index fb670b5cfebb..ca9ca92bb7fd 100644
--- a/drivers/net/wireless/iwlwifi/iwl-4965.c
+++ b/drivers/net/wireless/iwlwifi/iwl-4965.c
@@ -357,22 +357,22 @@ int iwl4965_hwrate_to_plcp_idx(u32 rate_n_flags)
  * translate ucode response to mac80211 tx status control values
  */
 void iwl4965_hwrate_to_tx_control(struct iwl_priv *priv, u32 rate_n_flags,
-				  struct ieee80211_tx_control *control)
+				  struct ieee80211_tx_info *control)
 {
 	int rate_index;
 
 	control->antenna_sel_tx =
 		((rate_n_flags & RATE_MCS_ANT_ABC_MSK) >> RATE_MCS_ANT_POS);
 	if (rate_n_flags & RATE_MCS_HT_MSK)
-		control->flags |= IEEE80211_TXCTL_OFDM_HT;
+		control->flags |= IEEE80211_TX_CTL_OFDM_HT;
 	if (rate_n_flags & RATE_MCS_GF_MSK)
-		control->flags |= IEEE80211_TXCTL_GREEN_FIELD;
+		control->flags |= IEEE80211_TX_CTL_GREEN_FIELD;
 	if (rate_n_flags & RATE_MCS_FAT_MSK)
-		control->flags |= IEEE80211_TXCTL_40_MHZ_WIDTH;
+		control->flags |= IEEE80211_TX_CTL_40_MHZ_WIDTH;
 	if (rate_n_flags & RATE_MCS_DUP_MSK)
-		control->flags |= IEEE80211_TXCTL_DUP_DATA;
+		control->flags |= IEEE80211_TX_CTL_DUP_DATA;
 	if (rate_n_flags & RATE_MCS_SGI_MSK)
-		control->flags |= IEEE80211_TXCTL_SHORT_GI;
+		control->flags |= IEEE80211_TX_CTL_SHORT_GI;
 	rate_index = iwl4965_hwrate_to_plcp_idx(rate_n_flags);
 	if (control->band == IEEE80211_BAND_5GHZ)
 		rate_index -= IWL_FIRST_OFDM_RATE;
@@ -3007,7 +3007,7 @@ static int iwl4965_tx_status_reply_compressed_ba(struct iwl_priv *priv,
 	u16 scd_flow = le16_to_cpu(ba_resp->scd_flow);
 	u64 bitmap;
 	int successes = 0;
-	struct ieee80211_tx_status *tx_status;
+	struct ieee80211_tx_info *info;
 
 	if (unlikely(!agg->wait_for_ba))  {
 		IWL_ERROR("Received BA when not expected\n");
@@ -3045,13 +3045,13 @@ static int iwl4965_tx_status_reply_compressed_ba(struct iwl_priv *priv,
 			agg->start_idx + i);
 	}
 
-	tx_status = &priv->txq[scd_flow].txb[agg->start_idx].status;
-	tx_status->flags = IEEE80211_TX_STATUS_ACK;
-	tx_status->flags |= IEEE80211_TX_STATUS_AMPDU;
-	tx_status->ampdu_ack_map = successes;
-	tx_status->ampdu_ack_len = agg->frame_count;
-	iwl4965_hwrate_to_tx_control(priv, agg->rate_n_flags,
-				     &tx_status->control);
+	info = IEEE80211_SKB_CB(priv->txq[scd_flow].txb[agg->start_idx].skb[0]);
+	memset(&info->status, 0, sizeof(info->status));
+	info->flags = IEEE80211_TX_STAT_ACK;
+	info->flags |= IEEE80211_TX_STAT_AMPDU;
+	info->status.ampdu_ack_map = successes;
+	info->status.ampdu_ack_len = agg->frame_count;
+	iwl4965_hwrate_to_tx_control(priv, agg->rate_n_flags, info);
 
 	IWL_DEBUG_TX_REPLY("Bitmap %llx\n", (unsigned long long)bitmap);
 
diff --git a/drivers/net/wireless/iwlwifi/iwl-core.h b/drivers/net/wireless/iwlwifi/iwl-core.h
index a8d062f7b87a..ad7422eadab2 100644
--- a/drivers/net/wireless/iwlwifi/iwl-core.h
+++ b/drivers/net/wireless/iwlwifi/iwl-core.h
@@ -207,8 +207,7 @@ void iwl_rx_allocate(struct iwl_priv *priv);
 * TX
 ******************************************************/
 int iwl_txq_ctx_reset(struct iwl_priv *priv);
-int iwl_tx_skb(struct iwl_priv *priv,
-		struct sk_buff *skb, struct ieee80211_tx_control *ctl);
+int iwl_tx_skb(struct iwl_priv *priv, struct sk_buff *skb);
 /* FIXME: remove when free Tx is fully merged into iwlcore */
 int iwl_hw_txq_free_tfd(struct iwl_priv *priv, struct iwl_tx_queue *txq);
 void iwl_hw_txq_ctx_free(struct iwl_priv *priv);
diff --git a/drivers/net/wireless/iwlwifi/iwl-dev.h b/drivers/net/wireless/iwlwifi/iwl-dev.h
index 820542bac443..f7fd8ea61779 100644
--- a/drivers/net/wireless/iwlwifi/iwl-dev.h
+++ b/drivers/net/wireless/iwlwifi/iwl-dev.h
@@ -119,7 +119,6 @@ struct iwl_queue {
 
 /* One for each TFD */
 struct iwl_tx_info {
-	struct ieee80211_tx_status status;
 	struct sk_buff *skb[MAX_NUM_OF_TBS];
 };
 
@@ -693,7 +692,7 @@ extern unsigned int iwl4965_hw_get_beacon_cmd(struct iwl_priv *priv,
 				 struct iwl_frame *frame, u8 rate);
 extern void iwl4965_hw_build_tx_cmd_rate(struct iwl_priv *priv,
 				     struct iwl_cmd *cmd,
-				     struct ieee80211_tx_control *ctrl,
+				     struct ieee80211_tx_info *info,
 				     struct ieee80211_hdr *hdr,
 				     int sta_id, int tx_id);
 extern int iwl4965_hw_reg_send_txpower(struct iwl_priv *priv);
@@ -749,7 +748,7 @@ extern void iwl4965_update_rate_scaling(struct iwl_priv *priv, u8 mode);
 extern void iwl4965_rf_kill_ct_config(struct iwl_priv *priv);
 extern void iwl4965_hwrate_to_tx_control(struct iwl_priv *priv,
 					 u32 rate_n_flags,
-					 struct ieee80211_tx_control *control);
+					 struct ieee80211_tx_info *info);
 
 #ifdef CONFIG_IWL4965_HT
 extern void iwl4965_init_ht_hw_capab(const struct iwl_priv *priv,
diff --git a/drivers/net/wireless/iwlwifi/iwl-tx.c b/drivers/net/wireless/iwlwifi/iwl-tx.c
index 4b5149c8c32e..a61293ba3f6b 100644
--- a/drivers/net/wireless/iwlwifi/iwl-tx.c
+++ b/drivers/net/wireless/iwlwifi/iwl-tx.c
@@ -502,7 +502,7 @@ int iwl_txq_ctx_reset(struct iwl_priv *priv)
  */
 static void iwl_tx_cmd_build_basic(struct iwl_priv *priv,
 				  struct iwl_tx_cmd *tx_cmd,
-				  struct ieee80211_tx_control *ctrl,
+				  struct ieee80211_tx_info *info,
 				  struct ieee80211_hdr *hdr,
 				  int is_unicast, u8 std_id)
 {
@@ -510,7 +510,7 @@ static void iwl_tx_cmd_build_basic(struct iwl_priv *priv,
 	__le32 tx_flags = tx_cmd->tx_flags;
 
 	tx_cmd->stop_time.life_time = TX_CMD_LIFE_TIME_INFINITE;
-	if (!(ctrl->flags & IEEE80211_TXCTL_NO_ACK)) {
+	if (!(info->flags & IEEE80211_TX_CTL_NO_ACK)) {
 		tx_flags |= TX_CMD_FLG_ACK_MSK;
 		if ((fc & IEEE80211_FCTL_FTYPE) == IEEE80211_FTYPE_MGMT)
 			tx_flags |= TX_CMD_FLG_SEQ_CTL_MSK;
@@ -538,10 +538,10 @@ static void iwl_tx_cmd_build_basic(struct iwl_priv *priv,
 		tx_flags |= TX_CMD_FLG_SEQ_CTL_MSK;
 	}
 
-	if (ctrl->flags & IEEE80211_TXCTL_USE_RTS_CTS) {
+	if (info->flags & IEEE80211_TX_CTL_USE_RTS_CTS) {
 		tx_flags |= TX_CMD_FLG_RTS_MSK;
 		tx_flags &= ~TX_CMD_FLG_CTS_MSK;
-	} else if (ctrl->flags & IEEE80211_TXCTL_USE_CTS_PROTECT) {
+	} else if (info->flags & IEEE80211_TX_CTL_USE_CTS_PROTECT) {
 		tx_flags &= ~TX_CMD_FLG_RTS_MSK;
 		tx_flags |= TX_CMD_FLG_CTS_MSK;
 	}
@@ -570,7 +570,7 @@ static void iwl_tx_cmd_build_basic(struct iwl_priv *priv,
 
 static void iwl_tx_cmd_build_rate(struct iwl_priv *priv,
 			      struct iwl_tx_cmd *tx_cmd,
-			      struct ieee80211_tx_control *ctrl,
+			      struct ieee80211_tx_info *info,
 			      u16 fc, int sta_id,
 			      int is_hcca)
 {
@@ -580,7 +580,7 @@ static void iwl_tx_cmd_build_rate(struct iwl_priv *priv,
 	u16 rate_flags = 0;
 	int rate_idx;
 
-	rate_idx = min(ieee80211_get_tx_rate(priv->hw, ctrl)->hw_value & 0xffff,
+	rate_idx = min(ieee80211_get_tx_rate(priv->hw, info)->hw_value & 0xffff,
 			IWL_RATE_COUNT - 1);
 
 	rate_plcp = iwl_rates[rate_idx].plcp;
@@ -637,18 +637,18 @@ static void iwl_tx_cmd_build_rate(struct iwl_priv *priv,
 }
 
 static void iwl_tx_cmd_build_hwcrypto(struct iwl_priv *priv,
-				      struct ieee80211_tx_control *ctl,
+				      struct ieee80211_tx_info *info,
 				      struct iwl_tx_cmd *tx_cmd,
 				      struct sk_buff *skb_frag,
 				      int sta_id)
 {
-	struct ieee80211_key_conf *keyconf = ctl->hw_key;
+	struct ieee80211_key_conf *keyconf = info->control.hw_key;
 
 	switch (keyconf->alg) {
 	case ALG_CCMP:
 		tx_cmd->sec_ctl = TX_CMD_SEC_CCM;
 		memcpy(tx_cmd->key, keyconf->key, keyconf->keylen);
-		if (ctl->flags & IEEE80211_TXCTL_AMPDU)
+		if (info->flags & IEEE80211_TX_CTL_AMPDU)
 			tx_cmd->tx_flags |= TX_CMD_FLG_AGG_CCMP_MSK;
 		IWL_DEBUG_TX("tx_cmd with aes hwcrypto\n");
 		break;
@@ -690,13 +690,13 @@ static void iwl_update_tx_stats(struct iwl_priv *priv, u16 fc, u16 len)
 /*
  * start REPLY_TX command process
  */
-int iwl_tx_skb(struct iwl_priv *priv,
-		struct sk_buff *skb, struct ieee80211_tx_control *ctl)
+int iwl_tx_skb(struct iwl_priv *priv, struct sk_buff *skb)
 {
 	struct ieee80211_hdr *hdr = (struct ieee80211_hdr *)skb->data;
+	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
 	struct iwl_tfd_frame *tfd;
 	u32 *control_flags;
-	int txq_id = ctl->queue;
+	int txq_id = info->queue;
 	struct iwl_tx_queue *txq = NULL;
 	struct iwl_queue *q = NULL;
 	dma_addr_t phys_addr;
@@ -726,7 +726,7 @@ int iwl_tx_skb(struct iwl_priv *priv,
 		goto drop_unlock;
 	}
 
-	if ((ieee80211_get_tx_rate(priv->hw, ctl)->hw_value & 0xFF) ==
+	if ((ieee80211_get_tx_rate(priv->hw, info)->hw_value & 0xFF) ==
 	     IWL_INVALID_RATE) {
 		IWL_ERROR("ERROR: No TX rate available.\n");
 		goto drop_unlock;
@@ -782,7 +782,7 @@ int iwl_tx_skb(struct iwl_priv *priv,
 		seq_number += 0x10;
 #ifdef CONFIG_IWL4965_HT
 		/* aggregation is on for this <sta,tid> */
-		if (ctl->flags & IEEE80211_TXCTL_AMPDU)
+		if (info->flags & IEEE80211_TX_CTL_AMPDU)
 			txq_id = priv->stations[sta_id].tid[tid].agg.txq_id;
 		priv->stations[sta_id].tid[tid].tfds_in_queue++;
 #endif /* CONFIG_IWL4965_HT */
@@ -803,8 +803,6 @@ int iwl_tx_skb(struct iwl_priv *priv,
 	/* Set up driver data for this TFD */
 	memset(&(txq->txb[q->write_ptr]), 0, sizeof(struct iwl_tx_info));
 	txq->txb[q->write_ptr].skb[0] = skb;
-	memcpy(&(txq->txb[q->write_ptr].status.control),
-	       ctl, sizeof(struct ieee80211_tx_control));
 
 	/* Set up first empty entry in queue's array of Tx/cmd buffers */
 	out_cmd = &txq->cmd[idx];
@@ -854,8 +852,8 @@ int iwl_tx_skb(struct iwl_priv *priv,
 	 * first entry */
 	iwl_hw_txq_attach_buf_to_tfd(priv, tfd, txcmd_phys, len);
 
-	if (!(ctl->flags & IEEE80211_TXCTL_DO_NOT_ENCRYPT))
-		iwl_tx_cmd_build_hwcrypto(priv, ctl, tx_cmd, skb, sta_id);
+	if (!(info->flags & IEEE80211_TX_CTL_DO_NOT_ENCRYPT))
+		iwl_tx_cmd_build_hwcrypto(priv, info, tx_cmd, skb, sta_id);
 
 	/* Set up TFD's 2nd entry to point directly to remainder of skb,
 	 * if any (802.11 null frames have no payload). */
@@ -874,10 +872,10 @@ int iwl_tx_skb(struct iwl_priv *priv,
 	len = (u16)skb->len;
 	tx_cmd->len = cpu_to_le16(len);
 	/* TODO need this for burst mode later on */
-	iwl_tx_cmd_build_basic(priv, tx_cmd, ctl, hdr, unicast, sta_id);
+	iwl_tx_cmd_build_basic(priv, tx_cmd, info, hdr, unicast, sta_id);
 
 	/* set is_hcca to 0; it probably will never be implemented */
-	iwl_tx_cmd_build_rate(priv, tx_cmd, ctl, fc, sta_id, 0);
+	iwl_tx_cmd_build_rate(priv, tx_cmd, info, fc, sta_id, 0);
 
 	iwl_update_tx_stats(priv, fc, len);
 
@@ -919,7 +917,7 @@ int iwl_tx_skb(struct iwl_priv *priv,
 			spin_unlock_irqrestore(&priv->lock, flags);
 		}
 
-		ieee80211_stop_queue(priv->hw, ctl->queue);
+		ieee80211_stop_queue(priv->hw, info->queue);
 	}
 
 	return 0;
diff --git a/drivers/net/wireless/iwlwifi/iwl3945-base.c b/drivers/net/wireless/iwlwifi/iwl3945-base.c
index a28b4c9f6524..a740a1817d16 100644
--- a/drivers/net/wireless/iwlwifi/iwl3945-base.c
+++ b/drivers/net/wireless/iwlwifi/iwl3945-base.c
@@ -2376,13 +2376,13 @@ static int iwl3945_set_mode(struct iwl3945_priv *priv, int mode)
 }
 
 static void iwl3945_build_tx_cmd_hwcrypto(struct iwl3945_priv *priv,
-				      struct ieee80211_tx_control *ctl,
+				      struct ieee80211_tx_info *info,
 				      struct iwl3945_cmd *cmd,
 				      struct sk_buff *skb_frag,
 				      int last_frag)
 {
 	struct iwl3945_hw_key *keyinfo =
-	    &priv->stations[ctl->hw_key->hw_key_idx].keyinfo;
+	    &priv->stations[info->control.hw_key->hw_key_idx].keyinfo;
 
 	switch (keyinfo->alg) {
 	case ALG_CCMP:
@@ -2405,7 +2405,7 @@ static void iwl3945_build_tx_cmd_hwcrypto(struct iwl3945_priv *priv,
 
 	case ALG_WEP:
 		cmd->cmd.tx.sec_ctl = TX_CMD_SEC_WEP |
-		    (ctl->hw_key->hw_key_idx & TX_CMD_SEC_MSK) << TX_CMD_SEC_SHIFT;
+		    (info->control.hw_key->hw_key_idx & TX_CMD_SEC_MSK) << TX_CMD_SEC_SHIFT;
 
 		if (keyinfo->keylen == 13)
 			cmd->cmd.tx.sec_ctl |= TX_CMD_SEC_KEY128;
@@ -2413,7 +2413,7 @@ static void iwl3945_build_tx_cmd_hwcrypto(struct iwl3945_priv *priv,
 		memcpy(&cmd->cmd.tx.key[3], keyinfo->key, keyinfo->keylen);
 
 		IWL_DEBUG_TX("Configuring packet for WEP encryption "
-			     "with key %d\n", ctl->hw_key->hw_key_idx);
+			     "with key %d\n", info->control.hw_key->hw_key_idx);
 		break;
 
 	default:
@@ -2427,7 +2427,7 @@ static void iwl3945_build_tx_cmd_hwcrypto(struct iwl3945_priv *priv,
  */
 static void iwl3945_build_tx_cmd_basic(struct iwl3945_priv *priv,
 				  struct iwl3945_cmd *cmd,
-				  struct ieee80211_tx_control *ctrl,
+				  struct ieee80211_tx_info *info,
 				  struct ieee80211_hdr *hdr,
 				  int is_unicast, u8 std_id)
 {
@@ -2435,7 +2435,7 @@ static void iwl3945_build_tx_cmd_basic(struct iwl3945_priv *priv,
 	__le32 tx_flags = cmd->cmd.tx.tx_flags;
 
 	cmd->cmd.tx.stop_time.life_time = TX_CMD_LIFE_TIME_INFINITE;
-	if (!(ctrl->flags & IEEE80211_TXCTL_NO_ACK)) {
+	if (!(info->flags & IEEE80211_TX_CTL_NO_ACK)) {
 		tx_flags |= TX_CMD_FLG_ACK_MSK;
 		if ((fc & IEEE80211_FCTL_FTYPE) == IEEE80211_FTYPE_MGMT)
 			tx_flags |= TX_CMD_FLG_SEQ_CTL_MSK;
@@ -2459,10 +2459,10 @@ static void iwl3945_build_tx_cmd_basic(struct iwl3945_priv *priv,
 		tx_flags |= TX_CMD_FLG_SEQ_CTL_MSK;
 	}
 
-	if (ctrl->flags & IEEE80211_TXCTL_USE_RTS_CTS) {
+	if (info->flags & IEEE80211_TX_CTL_USE_RTS_CTS) {
 		tx_flags |= TX_CMD_FLG_RTS_MSK;
 		tx_flags &= ~TX_CMD_FLG_CTS_MSK;
-	} else if (ctrl->flags & IEEE80211_TXCTL_USE_CTS_PROTECT) {
+	} else if (info->flags & IEEE80211_TX_CTL_USE_CTS_PROTECT) {
 		tx_flags &= ~TX_CMD_FLG_RTS_MSK;
 		tx_flags |= TX_CMD_FLG_CTS_MSK;
 	}
@@ -2546,13 +2546,13 @@ static int iwl3945_get_sta_id(struct iwl3945_priv *priv, struct ieee80211_hdr *h
 /*
  * start REPLY_TX command process
  */
-static int iwl3945_tx_skb(struct iwl3945_priv *priv,
-		      struct sk_buff *skb, struct ieee80211_tx_control *ctl)
+static int iwl3945_tx_skb(struct iwl3945_priv *priv, struct sk_buff *skb)
 {
 	struct ieee80211_hdr *hdr = (struct ieee80211_hdr *)skb->data;
+	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
 	struct iwl3945_tfd_frame *tfd;
 	u32 *control_flags;
-	int txq_id = ctl->queue;
+	int txq_id = info->queue;
 	struct iwl3945_tx_queue *txq = NULL;
 	struct iwl3945_queue *q = NULL;
 	dma_addr_t phys_addr;
@@ -2581,7 +2581,7 @@ static int iwl3945_tx_skb(struct iwl3945_priv *priv,
 		goto drop_unlock;
 	}
 
-	if ((ieee80211_get_tx_rate(priv->hw, ctl)->hw_value & 0xFF) == IWL_INVALID_RATE) {
+	if ((ieee80211_get_tx_rate(priv->hw, info)->hw_value & 0xFF) == IWL_INVALID_RATE) {
 		IWL_ERROR("ERROR: No TX rate available.\n");
 		goto drop_unlock;
 	}
@@ -2650,8 +2650,6 @@ static int iwl3945_tx_skb(struct iwl3945_priv *priv,
 	/* Set up driver data for this TFD */
 	memset(&(txq->txb[q->write_ptr]), 0, sizeof(struct iwl3945_tx_info));
 	txq->txb[q->write_ptr].skb[0] = skb;
-	memcpy(&(txq->txb[q->write_ptr].status.control),
-	       ctl, sizeof(struct ieee80211_tx_control));
 
 	/* Init first empty entry in queue's array of Tx/cmd buffers */
 	out_cmd = &txq->cmd[idx];
@@ -2700,8 +2698,8 @@ static int iwl3945_tx_skb(struct iwl3945_priv *priv,
 	 * first entry */
 	iwl3945_hw_txq_attach_buf_to_tfd(priv, tfd, txcmd_phys, len);
 
-	if (!(ctl->flags & IEEE80211_TXCTL_DO_NOT_ENCRYPT))
-		iwl3945_build_tx_cmd_hwcrypto(priv, ctl, out_cmd, skb, 0);
+	if (!(info->flags & IEEE80211_TX_CTL_DO_NOT_ENCRYPT))
+		iwl3945_build_tx_cmd_hwcrypto(priv, info, out_cmd, skb, 0);
 
 	/* Set up TFD's 2nd entry to point directly to remainder of skb,
 	 * if any (802.11 null frames have no payload). */
@@ -2726,10 +2724,10 @@ static int iwl3945_tx_skb(struct iwl3945_priv *priv,
 	out_cmd->cmd.tx.len = cpu_to_le16(len);
 
 	/* TODO need this for burst mode later on */
-	iwl3945_build_tx_cmd_basic(priv, out_cmd, ctl, hdr, unicast, sta_id);
+	iwl3945_build_tx_cmd_basic(priv, out_cmd, info, hdr, unicast, sta_id);
 
 	/* set is_hcca to 0; it probably will never be implemented */
-	iwl3945_hw_build_tx_cmd_rate(priv, out_cmd, ctl, hdr, sta_id, 0);
+	iwl3945_hw_build_tx_cmd_rate(priv, out_cmd, info, hdr, sta_id, 0);
 
 	out_cmd->cmd.tx.tx_flags &= ~TX_CMD_FLG_ANT_A_MSK;
 	out_cmd->cmd.tx.tx_flags &= ~TX_CMD_FLG_ANT_B_MSK;
@@ -2767,7 +2765,7 @@ static int iwl3945_tx_skb(struct iwl3945_priv *priv,
 			spin_unlock_irqrestore(&priv->lock, flags);
 		}
 
-		ieee80211_stop_queue(priv->hw, ctl->queue);
+		ieee80211_stop_queue(priv->hw, info->queue);
 	}
 
 	return 0;
@@ -3230,7 +3228,7 @@ static void iwl3945_bg_beacon_update(struct work_struct *work)
 	struct sk_buff *beacon;
 
 	/* Pull updated AP beacon from mac80211. will fail if not in AP mode */
-	beacon = ieee80211_beacon_get(priv->hw, priv->vif, NULL);
+	beacon = ieee80211_beacon_get(priv->hw, priv->vif);
 
 	if (!beacon) {
 		IWL_ERROR("update beacon failed\n");
@@ -6681,8 +6679,7 @@ static void iwl3945_mac_stop(struct ieee80211_hw *hw)
 	IWL_DEBUG_MAC80211("leave\n");
 }
 
-static int iwl3945_mac_tx(struct ieee80211_hw *hw, struct sk_buff *skb,
-		      struct ieee80211_tx_control *ctl)
+static int iwl3945_mac_tx(struct ieee80211_hw *hw, struct sk_buff *skb)
 {
 	struct iwl3945_priv *priv = hw->priv;
 
@@ -6694,9 +6691,9 @@ static int iwl3945_mac_tx(struct ieee80211_hw *hw, struct sk_buff *skb,
 	}
 
 	IWL_DEBUG_TX("dev->xmit(%d bytes) at rate 0x%02x\n", skb->len,
-		     ieee80211_get_tx_rate(hw, ctl)->bitrate);
+		     ieee80211_get_tx_rate(hw, IEEE80211_SKB_CB(skb))->bitrate);
 
-	if (iwl3945_tx_skb(priv, skb, ctl))
+	if (iwl3945_tx_skb(priv, skb))
 		dev_kfree_skb_any(skb);
 
 	IWL_DEBUG_MAC80211("leave\n");
@@ -7333,8 +7330,7 @@ static void iwl3945_mac_reset_tsf(struct ieee80211_hw *hw)
 
 }
 
-static int iwl3945_mac_beacon_update(struct ieee80211_hw *hw, struct sk_buff *skb,
-				 struct ieee80211_tx_control *control)
+static int iwl3945_mac_beacon_update(struct ieee80211_hw *hw, struct sk_buff *skb)
 {
 	struct iwl3945_priv *priv = hw->priv;
 	unsigned long flags;
diff --git a/drivers/net/wireless/iwlwifi/iwl4965-base.c b/drivers/net/wireless/iwlwifi/iwl4965-base.c
index 1fad6227aa51..07c52b69d1e2 100644
--- a/drivers/net/wireless/iwlwifi/iwl4965-base.c
+++ b/drivers/net/wireless/iwlwifi/iwl4965-base.c
@@ -1606,17 +1606,7 @@ static int iwl4965_get_measurement(struct iwl_priv *priv,
 static void iwl4965_txstatus_to_ieee(struct iwl_priv *priv,
 				     struct iwl_tx_info *tx_sta)
 {
-
-	tx_sta->status.ack_signal = 0;
-	tx_sta->status.excessive_retries = 0;
-
-	if (in_interrupt())
-		ieee80211_tx_status_irqsafe(priv->hw,
-					    tx_sta->skb[0], &(tx_sta->status));
-	else
-		ieee80211_tx_status(priv->hw,
-				    tx_sta->skb[0], &(tx_sta->status));
-
+	ieee80211_tx_status_irqsafe(priv->hw, tx_sta->skb[0]);
 	tx_sta->skb[0] = NULL;
 }
 
@@ -1710,7 +1700,7 @@ static int iwl4965_tx_status_reply_tx(struct iwl_priv *priv,
 {
 	u16 status;
 	struct agg_tx_status *frame_status = &tx_resp->status;
-	struct ieee80211_tx_status *tx_status = NULL;
+	struct ieee80211_tx_info *info = NULL;
 	struct ieee80211_hdr *hdr = NULL;
 	int i, sh;
 	int txq_id, idx;
@@ -1736,14 +1726,14 @@ static int iwl4965_tx_status_reply_tx(struct iwl_priv *priv,
 		IWL_DEBUG_TX_REPLY("FrameCnt = %d, StartIdx=%d idx=%d\n",
 				   agg->frame_count, agg->start_idx, idx);
 
-		tx_status = &(priv->txq[txq_id].txb[idx].status);
-		tx_status->retry_count = tx_resp->failure_frame;
-		tx_status->control.flags &= ~IEEE80211_TXCTL_AMPDU;
-		tx_status->flags = iwl4965_is_tx_success(status)?
-			IEEE80211_TX_STATUS_ACK : 0;
+		info = IEEE80211_SKB_CB(priv->txq[txq_id].txb[idx].skb[0]);
+		info->status.retry_count = tx_resp->failure_frame;
+		info->flags &= ~IEEE80211_TX_CTL_AMPDU;
+		info->flags |= iwl4965_is_tx_success(status)?
+			IEEE80211_TX_STAT_ACK : 0;
 		iwl4965_hwrate_to_tx_control(priv,
 					     le32_to_cpu(tx_resp->rate_n_flags),
-					     &tx_status->control);
+					     info);
 		/* FIXME: code repetition end */
 
 		IWL_DEBUG_TX_REPLY("1 Frame 0x%x failure :%d\n",
@@ -1830,7 +1820,7 @@ static void iwl4965_rx_reply_tx(struct iwl_priv *priv,
 	int txq_id = SEQ_TO_QUEUE(sequence);
 	int index = SEQ_TO_INDEX(sequence);
 	struct iwl_tx_queue *txq = &priv->txq[txq_id];
-	struct ieee80211_tx_status *tx_status;
+	struct ieee80211_tx_info *info;
 	struct iwl4965_tx_resp *tx_resp = (void *)&pkt->u.raw[0];
 	u32  status = le32_to_cpu(tx_resp->status);
 #ifdef CONFIG_IWL4965_HT
@@ -1848,6 +1838,9 @@ static void iwl4965_rx_reply_tx(struct iwl_priv *priv,
 		return;
 	}
 
+	info = IEEE80211_SKB_CB(txq->txb[txq->q.read_ptr].skb[0]);
+	memset(&info->status, 0, sizeof(info->status));
+
 #ifdef CONFIG_IWL4965_HT
 	hdr = iwl4965_tx_queue_get_hdr(priv, txq_id, index);
 	fc = le16_to_cpu(hdr->frame_control);
@@ -1902,13 +1895,12 @@ static void iwl4965_rx_reply_tx(struct iwl_priv *priv,
 		}
 	} else {
 #endif /* CONFIG_IWL4965_HT */
-	tx_status = &(txq->txb[txq->q.read_ptr].status);
 
-	tx_status->retry_count = tx_resp->failure_frame;
-	tx_status->flags =
-	    iwl4965_is_tx_success(status) ? IEEE80211_TX_STATUS_ACK : 0;
+	info->status.retry_count = tx_resp->failure_frame;
+	info->flags |=
+	    iwl4965_is_tx_success(status) ? IEEE80211_TX_STAT_ACK : 0;
 	iwl4965_hwrate_to_tx_control(priv, le32_to_cpu(tx_resp->rate_n_flags),
-				     &tx_status->control);
+				     info);
 
 	IWL_DEBUG_TX("Tx queue %d Status %s (0x%08x) rate_n_flags 0x%x "
 		     "retries %d\n", txq_id, iwl4965_get_tx_fail_reason(status),
@@ -2053,7 +2045,7 @@ static void iwl4965_bg_beacon_update(struct work_struct *work)
 	struct sk_buff *beacon;
 
 	/* Pull updated AP beacon from mac80211. will fail if not in AP mode */
-	beacon = ieee80211_beacon_get(priv->hw, priv->vif, NULL);
+	beacon = ieee80211_beacon_get(priv->hw, priv->vif);
 
 	if (!beacon) {
 		IWL_ERROR("update beacon failed\n");
@@ -4268,8 +4260,7 @@ static void iwl4965_mac_stop(struct ieee80211_hw *hw)
 	IWL_DEBUG_MAC80211("leave\n");
 }
 
-static int iwl4965_mac_tx(struct ieee80211_hw *hw, struct sk_buff *skb,
-		      struct ieee80211_tx_control *ctl)
+static int iwl4965_mac_tx(struct ieee80211_hw *hw, struct sk_buff *skb)
 {
 	struct iwl_priv *priv = hw->priv;
 
@@ -4281,9 +4272,9 @@ static int iwl4965_mac_tx(struct ieee80211_hw *hw, struct sk_buff *skb,
 	}
 
 	IWL_DEBUG_TX("dev->xmit(%d bytes) at rate 0x%02x\n", skb->len,
-		     ieee80211_get_tx_rate(hw, ctl)->bitrate);
+		     ieee80211_get_tx_rate(hw, IEEE80211_SKB_CB(skb))->bitrate);
 
-	if (iwl_tx_skb(priv, skb, ctl))
+	if (iwl_tx_skb(priv, skb))
 		dev_kfree_skb_any(skb);
 
 	IWL_DEBUG_MAC80211("leave\n");
@@ -5065,8 +5056,7 @@ static void iwl4965_mac_reset_tsf(struct ieee80211_hw *hw)
 	IWL_DEBUG_MAC80211("leave\n");
 }
 
-static int iwl4965_mac_beacon_update(struct ieee80211_hw *hw, struct sk_buff *skb,
-				 struct ieee80211_tx_control *control)
+static int iwl4965_mac_beacon_update(struct ieee80211_hw *hw, struct sk_buff *skb)
 {
 	struct iwl_priv *priv = hw->priv;
 	unsigned long flags;
diff --git a/drivers/net/wireless/p54/p54common.c b/drivers/net/wireless/p54/p54common.c
index 3ca9386561ff..850857932e29 100644
--- a/drivers/net/wireless/p54/p54common.c
+++ b/drivers/net/wireless/p54/p54common.c
@@ -394,7 +394,7 @@ static void p54_rx_frame_sent(struct ieee80211_hw *dev, struct sk_buff *skb)
 	while (entry != (struct sk_buff *)&priv->tx_queue) {
 		range = (struct memrecord *)&entry->cb;
 		if (range->start_addr == addr) {
-			struct ieee80211_tx_status status;
+			struct ieee80211_tx_info *info = IEEE80211_SKB_CB(entry);
 			struct p54_control_hdr *entry_hdr;
 			struct p54_tx_control_allocdata *entry_data;
 			int pad = 0;
@@ -406,30 +406,23 @@ static void p54_rx_frame_sent(struct ieee80211_hw *dev, struct sk_buff *skb)
 
 			last_addr = range->end_addr;
 			__skb_unlink(entry, &priv->tx_queue);
-			if (!range->control) {
-				kfree_skb(entry);
-				break;
-			}
-			memset(&status, 0, sizeof(status));
-			memcpy(&status.control, range->control,
-			       sizeof(status.control));
-			kfree(range->control);
-			priv->tx_stats[status.control.queue].len--;
+			memset(&info->status, 0, sizeof(info->status));
+			priv->tx_stats[info->queue].len--;
 			entry_hdr = (struct p54_control_hdr *) entry->data;
 			entry_data = (struct p54_tx_control_allocdata *) entry_hdr->data;
 			if ((entry_hdr->magic1 & cpu_to_le16(0x4000)) != 0)
 				pad = entry_data->align[0];
 
-			if (!(status.control.flags & IEEE80211_TXCTL_NO_ACK)) {
+			if (!(info->flags & IEEE80211_TX_CTL_NO_ACK)) {
 				if (!(payload->status & 0x01))
-					status.flags |= IEEE80211_TX_STATUS_ACK;
+					info->flags |= IEEE80211_TX_STAT_ACK;
 				else
-					status.excessive_retries = 1;
+					info->status.excessive_retries = 1;
 			}
-			status.retry_count = payload->retries - 1;
-			status.ack_signal = le16_to_cpu(payload->ack_rssi);
+			info->status.retry_count = payload->retries - 1;
+			info->status.ack_signal = le16_to_cpu(payload->ack_rssi);
 			skb_pull(entry, sizeof(*hdr) + pad + sizeof(*entry_data));
-			ieee80211_tx_status_irqsafe(dev, entry, &status);
+			ieee80211_tx_status_irqsafe(dev, entry);
 			break;
 		} else
 			last_addr = range->end_addr;
@@ -494,13 +487,11 @@ EXPORT_SYMBOL_GPL(p54_rx);
  * allocated areas.
  */
 static void p54_assign_address(struct ieee80211_hw *dev, struct sk_buff *skb,
-			       struct p54_control_hdr *data, u32 len,
-			       struct ieee80211_tx_control *control)
+			       struct p54_control_hdr *data, u32 len)
 {
 	struct p54_common *priv = dev->priv;
 	struct sk_buff *entry = priv->tx_queue.next;
 	struct sk_buff *target_skb = NULL;
-	struct memrecord *range;
 	u32 last_addr = priv->rx_start;
 	u32 largest_hole = 0;
 	u32 target_addr = priv->rx_start;
@@ -512,7 +503,8 @@ static void p54_assign_address(struct ieee80211_hw *dev, struct sk_buff *skb,
 	left = skb_queue_len(&priv->tx_queue);
 	while (left--) {
 		u32 hole_size;
-		range = (struct memrecord *)&entry->cb;
+		struct ieee80211_tx_info *info = IEEE80211_SKB_CB(entry);
+		struct memrecord *range = (void *)info->driver_data;
 		hole_size = range->start_addr - last_addr;
 		if (!target_skb && hole_size >= len) {
 			target_skb = entry->prev;
@@ -527,17 +519,18 @@ static void p54_assign_address(struct ieee80211_hw *dev, struct sk_buff *skb,
 		target_skb = priv->tx_queue.prev;
 		largest_hole = max(largest_hole, priv->rx_end - last_addr - len);
 		if (!skb_queue_empty(&priv->tx_queue)) {
-			range = (struct memrecord *)&target_skb->cb;
+			struct ieee80211_tx_info *info = IEEE80211_SKB_CB(target_skb);
+			struct memrecord *range = (void *)info->driver_data;
 			target_addr = range->end_addr;
 		}
 	} else
 		largest_hole = max(largest_hole, priv->rx_end - last_addr);
 
 	if (skb) {
-		range = (struct memrecord *)&skb->cb;
+		struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
+		struct memrecord *range = (void *)info->driver_data;
 		range->start_addr = target_addr;
 		range->end_addr = target_addr + len;
-		range->control = control;
 		__skb_queue_after(&priv->tx_queue, target_skb, skb);
 		if (largest_hole < IEEE80211_MAX_RTS_THRESHOLD + 0x170 +
 				   sizeof(struct p54_control_hdr))
@@ -548,32 +541,27 @@ static void p54_assign_address(struct ieee80211_hw *dev, struct sk_buff *skb,
 	data->req_id = cpu_to_le32(target_addr + 0x70);
 }
 
-static int p54_tx(struct ieee80211_hw *dev, struct sk_buff *skb,
-		  struct ieee80211_tx_control *control)
+static int p54_tx(struct ieee80211_hw *dev, struct sk_buff *skb)
 {
+	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
 	struct ieee80211_tx_queue_stats *current_queue;
 	struct p54_common *priv = dev->priv;
 	struct p54_control_hdr *hdr;
 	struct p54_tx_control_allocdata *txhdr;
-	struct ieee80211_tx_control *control_copy;
 	size_t padding, len;
 	u8 rate;
 
-	current_queue = &priv->tx_stats[control->queue];
+	current_queue = &priv->tx_stats[info->queue];
 	if (unlikely(current_queue->len > current_queue->limit))
 		return NETDEV_TX_BUSY;
 	current_queue->len++;
 	current_queue->count++;
 	if (current_queue->len == current_queue->limit)
-		ieee80211_stop_queue(dev, control->queue);
+		ieee80211_stop_queue(dev, info->queue);
 
 	padding = (unsigned long)(skb->data - (sizeof(*hdr) + sizeof(*txhdr))) & 3;
 	len = skb->len;
 
-	control_copy = kmalloc(sizeof(*control), GFP_ATOMIC);
-	if (control_copy)
-		memcpy(control_copy, control, sizeof(*control));
-
 	txhdr = (struct p54_tx_control_allocdata *)
 			skb_push(skb, sizeof(*txhdr) + padding);
 	hdr = (struct p54_control_hdr *) skb_push(skb, sizeof(*hdr));
@@ -583,35 +571,37 @@ static int p54_tx(struct ieee80211_hw *dev, struct sk_buff *skb,
 	else
 		hdr->magic1 = cpu_to_le16(0x0010);
 	hdr->len = cpu_to_le16(len);
-	hdr->type = (control->flags & IEEE80211_TXCTL_NO_ACK) ? 0 : cpu_to_le16(1);
-	hdr->retry1 = hdr->retry2 = control->retry_limit;
-	p54_assign_address(dev, skb, hdr, skb->len, control_copy);
+	hdr->type = (info->flags & IEEE80211_TX_CTL_NO_ACK) ? 0 : cpu_to_le16(1);
+	hdr->retry1 = hdr->retry2 = info->control.retry_limit;
 
 	memset(txhdr->wep_key, 0x0, 16);
 	txhdr->padding = 0;
 	txhdr->padding2 = 0;
 
 	/* TODO: add support for alternate retry TX rates */
-	rate = ieee80211_get_tx_rate(dev, control)->hw_value;
-	if (control->flags & IEEE80211_TXCTL_SHORT_PREAMBLE)
+	rate = ieee80211_get_tx_rate(dev, info)->hw_value;
+	if (info->flags & IEEE80211_TX_CTL_SHORT_PREAMBLE)
 		rate |= 0x10;
-	if (control->flags & IEEE80211_TXCTL_USE_RTS_CTS)
+	if (info->flags & IEEE80211_TX_CTL_USE_RTS_CTS)
 		rate |= 0x40;
-	else if (control->flags & IEEE80211_TXCTL_USE_CTS_PROTECT)
+	else if (info->flags & IEEE80211_TX_CTL_USE_CTS_PROTECT)
 		rate |= 0x20;
 	memset(txhdr->rateset, rate, 8);
 	txhdr->wep_key_present = 0;
 	txhdr->wep_key_len = 0;
-	txhdr->frame_type = cpu_to_le32(control->queue + 4);
+	txhdr->frame_type = cpu_to_le32(info->queue + 4);
 	txhdr->magic4 = 0;
-	txhdr->antenna = (control->antenna_sel_tx == 0) ?
-		2 : control->antenna_sel_tx - 1;
+	txhdr->antenna = (info->antenna_sel_tx == 0) ?
+		2 : info->antenna_sel_tx - 1;
 	txhdr->output_power = 0x7f; // HW Maximum
-	txhdr->magic5 = (control->flags & IEEE80211_TXCTL_NO_ACK) ?
+	txhdr->magic5 = (info->flags & IEEE80211_TX_CTL_NO_ACK) ?
 		0 : ((rate > 0x3) ? cpu_to_le32(0x33) : cpu_to_le32(0x23));
 	if (padding)
 		txhdr->align[0] = padding;
 
+	/* modifies skb->cb and with it info, so must be last! */
+	p54_assign_address(dev, skb, hdr, skb->len);
+
 	priv->tx(dev, hdr, skb->len, 0);
 	return 0;
 }
@@ -634,7 +624,7 @@ static int p54_set_filter(struct ieee80211_hw *dev, u16 filter_type,
 	filter = (struct p54_tx_control_filter *) hdr->data;
 	hdr->magic1 = cpu_to_le16(0x8001);
 	hdr->len = cpu_to_le16(sizeof(*filter));
-	p54_assign_address(dev, NULL, hdr, sizeof(*hdr) + sizeof(*filter), NULL);
+	p54_assign_address(dev, NULL, hdr, sizeof(*hdr) + sizeof(*filter));
 	hdr->type = cpu_to_le16(P54_CONTROL_TYPE_FILTER_SET);
 
 	filter->filter_type = cpu_to_le16(filter_type);
@@ -678,7 +668,7 @@ static int p54_set_freq(struct ieee80211_hw *dev, __le16 freq)
 	hdr->magic1 = cpu_to_le16(0x8001);
 	hdr->len = cpu_to_le16(sizeof(*chan));
 	hdr->type = cpu_to_le16(P54_CONTROL_TYPE_CHANNEL_CHANGE);
-	p54_assign_address(dev, NULL, hdr, sizeof(*hdr) + payload_len, NULL);
+	p54_assign_address(dev, NULL, hdr, sizeof(*hdr) + payload_len);
 
 	chan->magic1 = cpu_to_le16(0x1);
 	chan->magic2 = cpu_to_le16(0x0);
@@ -751,7 +741,7 @@ static int p54_set_leds(struct ieee80211_hw *dev, int mode, int link, int act)
 	hdr->magic1 = cpu_to_le16(0x8001);
 	hdr->len = cpu_to_le16(sizeof(*led));
 	hdr->type = cpu_to_le16(P54_CONTROL_TYPE_LED);
-	p54_assign_address(dev, NULL, hdr, sizeof(*hdr) + sizeof(*led), NULL);
+	p54_assign_address(dev, NULL, hdr, sizeof(*hdr) + sizeof(*led));
 
 	led = (struct p54_tx_control_led *) hdr->data;
 	led->mode = cpu_to_le16(mode);
@@ -801,7 +791,7 @@ static void p54_set_vdcf(struct ieee80211_hw *dev)
 
 	hdr = (void *)priv->cached_vdcf + priv->tx_hdr_len;
 
-	p54_assign_address(dev, NULL, hdr, sizeof(*hdr) + sizeof(*vdcf), NULL);
+	p54_assign_address(dev, NULL, hdr, sizeof(*hdr) + sizeof(*vdcf));
 
 	vdcf = (struct p54_tx_control_vdcf *) hdr->data;
 
@@ -837,12 +827,8 @@ static void p54_stop(struct ieee80211_hw *dev)
 {
 	struct p54_common *priv = dev->priv;
 	struct sk_buff *skb;
-	while ((skb = skb_dequeue(&priv->tx_queue))) {
-		struct memrecord *range = (struct memrecord *)&skb->cb;
-		if (range->control)
-			kfree(range->control);
+	while ((skb = skb_dequeue(&priv->tx_queue)))
 		kfree_skb(skb);
-	}
 	priv->stop(dev);
 	priv->mode = IEEE80211_IF_TYPE_INVALID;
 }
diff --git a/drivers/net/wireless/p54/p54common.h b/drivers/net/wireless/p54/p54common.h
index c15b56e1d75e..2245fcce92dc 100644
--- a/drivers/net/wireless/p54/p54common.h
+++ b/drivers/net/wireless/p54/p54common.h
@@ -152,7 +152,6 @@ struct pda_pa_curve_data {
 struct memrecord {
 	u32 start_addr;
 	u32 end_addr;
-	struct ieee80211_tx_control *control;
 };
 
 struct p54_eeprom_lm86 {
diff --git a/drivers/net/wireless/rt2x00/rt2400pci.c b/drivers/net/wireless/rt2x00/rt2400pci.c
index 4fcba9b4635f..900140d3b304 100644
--- a/drivers/net/wireless/rt2x00/rt2400pci.c
+++ b/drivers/net/wireless/rt2x00/rt2400pci.c
@@ -1484,11 +1484,11 @@ static u64 rt2400pci_get_tsf(struct ieee80211_hw *hw)
 	return tsf;
 }
 
-static int rt2400pci_beacon_update(struct ieee80211_hw *hw, struct sk_buff *skb,
-				   struct ieee80211_tx_control *control)
+static int rt2400pci_beacon_update(struct ieee80211_hw *hw, struct sk_buff *skb)
 {
 	struct rt2x00_dev *rt2x00dev = hw->priv;
-	struct rt2x00_intf *intf = vif_to_intf(control->vif);
+	struct ieee80211_tx_info *tx_info = IEEE80211_SKB_CB(skb);
+	struct rt2x00_intf *intf = vif_to_intf(tx_info->control.vif);
 	struct queue_entry_priv_pci *entry_priv;
 	struct skb_frame_desc *skbdesc;
 	struct txentry_desc txdesc;
@@ -1504,7 +1504,7 @@ static int rt2400pci_beacon_update(struct ieee80211_hw *hw, struct sk_buff *skb,
 	 * for our information.
 	 */
 	intf->beacon->skb = skb;
-	rt2x00queue_create_tx_descriptor(intf->beacon, &txdesc, control);
+	rt2x00queue_create_tx_descriptor(intf->beacon, &txdesc);
 
 	/*
 	 * Fill in skb descriptor
diff --git a/drivers/net/wireless/rt2x00/rt2500pci.c b/drivers/net/wireless/rt2x00/rt2500pci.c
index 06e87cdff455..673350953b89 100644
--- a/drivers/net/wireless/rt2x00/rt2500pci.c
+++ b/drivers/net/wireless/rt2x00/rt2500pci.c
@@ -1799,11 +1799,11 @@ static u64 rt2500pci_get_tsf(struct ieee80211_hw *hw)
 	return tsf;
 }
 
-static int rt2500pci_beacon_update(struct ieee80211_hw *hw, struct sk_buff *skb,
-				   struct ieee80211_tx_control *control)
+static int rt2500pci_beacon_update(struct ieee80211_hw *hw, struct sk_buff *skb)
 {
 	struct rt2x00_dev *rt2x00dev = hw->priv;
-	struct rt2x00_intf *intf = vif_to_intf(control->vif);
+	struct ieee80211_tx_info *tx_info = IEEE80211_SKB_CB(skb);
+	struct rt2x00_intf *intf = vif_to_intf(tx_info->control.vif);
 	struct queue_entry_priv_pci *entry_priv;
 	struct skb_frame_desc *skbdesc;
 	struct txentry_desc txdesc;
@@ -1820,7 +1820,7 @@ static int rt2500pci_beacon_update(struct ieee80211_hw *hw, struct sk_buff *skb,
 	 * for our information.
 	 */
 	intf->beacon->skb = skb;
-	rt2x00queue_create_tx_descriptor(intf->beacon, &txdesc, control);
+	rt2x00queue_create_tx_descriptor(intf->beacon, &txdesc);
 
 	/*
 	 * Fill in skb descriptor
diff --git a/drivers/net/wireless/rt2x00/rt2500usb.c b/drivers/net/wireless/rt2x00/rt2500usb.c
index 4122c5ebe7c3..cca1504550dc 100644
--- a/drivers/net/wireless/rt2x00/rt2500usb.c
+++ b/drivers/net/wireless/rt2x00/rt2500usb.c
@@ -1666,13 +1666,12 @@ static int rt2500usb_probe_hw(struct rt2x00_dev *rt2x00dev)
 /*
  * IEEE80211 stack callback functions.
  */
-static int rt2500usb_beacon_update(struct ieee80211_hw *hw,
-				   struct sk_buff *skb,
-				   struct ieee80211_tx_control *control)
+static int rt2500usb_beacon_update(struct ieee80211_hw *hw, struct sk_buff *skb)
 {
 	struct rt2x00_dev *rt2x00dev = hw->priv;
 	struct usb_device *usb_dev = rt2x00dev_usb_dev(rt2x00dev);
-	struct rt2x00_intf *intf = vif_to_intf(control->vif);
+	struct ieee80211_tx_info *tx_info = IEEE80211_SKB_CB(skb);
+	struct rt2x00_intf *intf = vif_to_intf(tx_info->control.vif);
 	struct queue_entry_priv_usb_bcn *bcn_priv;
 	struct skb_frame_desc *skbdesc;
 	struct txentry_desc txdesc;
@@ -1691,7 +1690,7 @@ static int rt2500usb_beacon_update(struct ieee80211_hw *hw,
 	 * for our information.
 	 */
 	intf->beacon->skb = skb;
-	rt2x00queue_create_tx_descriptor(intf->beacon, &txdesc, control);
+	rt2x00queue_create_tx_descriptor(intf->beacon, &txdesc);
 
 	/*
 	 * Add the descriptor in front of the skb.
diff --git a/drivers/net/wireless/rt2x00/rt2x00.h b/drivers/net/wireless/rt2x00/rt2x00.h
index 1900d4c0e846..5c7220ea46e6 100644
--- a/drivers/net/wireless/rt2x00/rt2x00.h
+++ b/drivers/net/wireless/rt2x00/rt2x00.h
@@ -542,8 +542,7 @@ struct rt2x00lib_ops {
 			       struct sk_buff *skb,
 			       struct txentry_desc *txdesc);
 	int (*write_tx_data) (struct rt2x00_dev *rt2x00dev,
-			      struct data_queue *queue, struct sk_buff *skb,
-			      struct ieee80211_tx_control *control);
+			      struct data_queue *queue, struct sk_buff *skb);
 	int (*get_tx_data_len) (struct rt2x00_dev *rt2x00dev,
 				struct sk_buff *skb);
 	void (*kick_tx_queue) (struct rt2x00_dev *rt2x00dev,
@@ -930,7 +929,6 @@ static inline u16 get_duration_res(const unsigned int size, const u8 rate)
  * rt2x00queue_create_tx_descriptor - Create TX descriptor from mac80211 input
  * @entry: The entry which will be used to transfer the TX frame.
  * @txdesc: rt2x00 TX descriptor which will be initialized by this function.
- * @control: mac80211 TX control structure from where we read the information.
  *
  * This function will initialize the &struct txentry_desc based on information
  * from mac80211. This descriptor can then be used by rt2x00lib and the drivers
@@ -943,8 +941,7 @@ static inline u16 get_duration_res(const unsigned int size, const u8 rate)
  * the &struct txentry_desc structure.
  */
 void rt2x00queue_create_tx_descriptor(struct queue_entry *entry,
-				      struct txentry_desc *txdesc,
-				      struct ieee80211_tx_control *control);
+				      struct txentry_desc *txdesc);
 
 /**
  * rt2x00queue_write_tx_descriptor - Write TX descriptor to hardware
@@ -1001,8 +998,7 @@ void rt2x00lib_rxdone(struct queue_entry *entry,
 /*
  * mac80211 handlers.
  */
-int rt2x00mac_tx(struct ieee80211_hw *hw, struct sk_buff *skb,
-		 struct ieee80211_tx_control *control);
+int rt2x00mac_tx(struct ieee80211_hw *hw, struct sk_buff *skb);
 int rt2x00mac_start(struct ieee80211_hw *hw);
 void rt2x00mac_stop(struct ieee80211_hw *hw);
 int rt2x00mac_add_interface(struct ieee80211_hw *hw,
diff --git a/drivers/net/wireless/rt2x00/rt2x00dev.c b/drivers/net/wireless/rt2x00/rt2x00dev.c
index d341764e1b24..69e233610c94 100644
--- a/drivers/net/wireless/rt2x00/rt2x00dev.c
+++ b/drivers/net/wireless/rt2x00/rt2x00dev.c
@@ -415,7 +415,6 @@ static void rt2x00lib_intf_scheduled_iter(void *data, u8 *mac,
 	struct rt2x00_dev *rt2x00dev = data;
 	struct rt2x00_intf *intf = vif_to_intf(vif);
 	struct sk_buff *skb;
-	struct ieee80211_tx_control control;
 	struct ieee80211_bss_conf conf;
 	int delayed_flags;
 
@@ -433,9 +432,9 @@ static void rt2x00lib_intf_scheduled_iter(void *data, u8 *mac,
 	spin_unlock(&intf->lock);
 
 	if (delayed_flags & DELAYED_UPDATE_BEACON) {
-		skb = ieee80211_beacon_get(rt2x00dev->hw, vif, &control);
-		if (skb && rt2x00dev->ops->hw->beacon_update(rt2x00dev->hw,
-							     skb, &control))
+		skb = ieee80211_beacon_get(rt2x00dev->hw, vif);
+		if (skb &&
+		    rt2x00dev->ops->hw->beacon_update(rt2x00dev->hw, skb))
 			dev_kfree_skb(skb);
 	}
 
@@ -494,8 +493,13 @@ void rt2x00lib_txdone(struct queue_entry *entry,
 		      struct txdone_entry_desc *txdesc)
 {
 	struct rt2x00_dev *rt2x00dev = entry->queue->rt2x00dev;
-	struct skb_frame_desc *skbdesc;
-	struct ieee80211_tx_status tx_status;
+	struct ieee80211_tx_info *tx_info = IEEE80211_SKB_CB(entry->skb);
+
+	/*
+	 * Send frame to debugfs immediately, after this call is completed
+	 * we are going to overwrite the skb->cb array.
+	 */
+	rt2x00debug_dump_frame(rt2x00dev, DUMP_FRAME_TXDONE, entry->skb);
 
 	/*
 	 * Update TX statistics.
@@ -508,21 +512,20 @@ void rt2x00lib_txdone(struct queue_entry *entry,
 	/*
 	 * Initialize TX status
 	 */
-	tx_status.flags = 0;
-	tx_status.ack_signal = 0;
-	tx_status.excessive_retries =
+	memset(&tx_info->status, 0, sizeof(tx_info->status));
+	tx_info->status.ack_signal = 0;
+	tx_info->status.excessive_retries =
 	    test_bit(TXDONE_EXCESSIVE_RETRY, &txdesc->flags);
-	tx_status.retry_count = txdesc->retry;
-	memcpy(&tx_status.control, txdesc->control, sizeof(*txdesc->control));
+	tx_info->status.retry_count = txdesc->retry;
 
-	if (!(tx_status.control.flags & IEEE80211_TXCTL_NO_ACK)) {
+	if (!(tx_info->flags & IEEE80211_TX_CTL_NO_ACK)) {
 		if (test_bit(TXDONE_SUCCESS, &txdesc->flags))
-			tx_status.flags |= IEEE80211_TX_STATUS_ACK;
+			tx_info->flags |= IEEE80211_TX_STAT_ACK;
 		else if (test_bit(TXDONE_FAILURE, &txdesc->flags))
 			rt2x00dev->low_level_stats.dot11ACKFailureCount++;
 	}
 
-	if (tx_status.control.flags & IEEE80211_TXCTL_USE_RTS_CTS) {
+	if (tx_info->flags & IEEE80211_TX_CTL_USE_RTS_CTS) {
 		if (test_bit(TXDONE_SUCCESS, &txdesc->flags))
 			rt2x00dev->low_level_stats.dot11RTSSuccessCount++;
 		else if (test_bit(TXDONE_FAILURE, &txdesc->flags))
@@ -530,19 +533,13 @@ void rt2x00lib_txdone(struct queue_entry *entry,
 	}
 
 	/*
-	 * Send the tx_status to debugfs. Only send the status report
-	 * to mac80211 when the frame originated from there. If this was
-	 * a extra frame coming through a mac80211 library call (RTS/CTS)
-	 * then we should not send the status report back.
-	 * If send to mac80211, mac80211 will clean up the skb structure,
-	 * otherwise we have to do it ourself.
+	 * Only send the status report to mac80211 when TX status was
+	 * requested by it. If this was a extra frame coming through
+	 * a mac80211 library call (RTS/CTS) then we should not send the
+	 * status report back.
 	 */
-	rt2x00debug_dump_frame(rt2x00dev, DUMP_FRAME_TXDONE, entry->skb);
-
-	skbdesc = get_skb_frame_desc(entry->skb);
-	if (!(skbdesc->flags & FRAME_DESC_DRIVER_GENERATED))
-		ieee80211_tx_status_irqsafe(rt2x00dev->hw,
-					    entry->skb, &tx_status);
+	if (tx_info->flags & IEEE80211_TX_CTL_REQ_TX_STATUS)
+		ieee80211_tx_status_irqsafe(rt2x00dev->hw, entry->skb);
 	else
 		dev_kfree_skb_irq(entry->skb);
 	entry->skb = NULL;
diff --git a/drivers/net/wireless/rt2x00/rt2x00mac.c b/drivers/net/wireless/rt2x00/rt2x00mac.c
index c5cedb29b87d..b5379b027b18 100644
--- a/drivers/net/wireless/rt2x00/rt2x00mac.c
+++ b/drivers/net/wireless/rt2x00/rt2x00mac.c
@@ -31,14 +31,15 @@
 
 static int rt2x00mac_tx_rts_cts(struct rt2x00_dev *rt2x00dev,
 				struct data_queue *queue,
-				struct sk_buff *frag_skb,
-				struct ieee80211_tx_control *control)
+				struct sk_buff *frag_skb)
 {
+	struct ieee80211_tx_info *tx_info = IEEE80211_SKB_CB(frag_skb);
 	struct skb_frame_desc *skbdesc;
+	struct ieee80211_tx_info *rts_info;
 	struct sk_buff *skb;
 	int size;
 
-	if (control->flags & IEEE80211_TXCTL_USE_CTS_PROTECT)
+	if (tx_info->flags & IEEE80211_TX_CTL_USE_CTS_PROTECT)
 		size = sizeof(struct ieee80211_cts);
 	else
 		size = sizeof(struct ieee80211_rts);
@@ -52,13 +53,33 @@ static int rt2x00mac_tx_rts_cts(struct rt2x00_dev *rt2x00dev,
 	skb_reserve(skb, rt2x00dev->hw->extra_tx_headroom);
 	skb_put(skb, size);
 
-	if (control->flags & IEEE80211_TXCTL_USE_CTS_PROTECT)
-		ieee80211_ctstoself_get(rt2x00dev->hw, control->vif,
-					frag_skb->data, frag_skb->len, control,
+	/*
+	 * Copy TX information over from original frame to
+	 * RTS/CTS frame. Note that we set the no encryption flag
+	 * since we don't want this frame to be encrypted.
+	 * RTS frames should be acked, while CTS-to-self frames
+	 * should not. The ready for TX flag is cleared to prevent
+	 * it being automatically send when the descriptor is
+	 * written to the hardware.
+	 */
+	memcpy(skb->cb, frag_skb->cb, sizeof(skb->cb));
+	rts_info = IEEE80211_SKB_CB(skb);
+	rts_info->flags |= IEEE80211_TX_CTL_DO_NOT_ENCRYPT;
+	rts_info->flags &= ~IEEE80211_TX_CTL_USE_CTS_PROTECT;
+	rts_info->flags &= ~IEEE80211_TX_CTL_REQ_TX_STATUS;
+
+	if (tx_info->flags & IEEE80211_TX_CTL_USE_CTS_PROTECT)
+		rts_info->flags |= IEEE80211_TX_CTL_NO_ACK;
+	else
+		rts_info->flags &= ~IEEE80211_TX_CTL_NO_ACK;
+
+	if (tx_info->flags & IEEE80211_TX_CTL_USE_CTS_PROTECT)
+		ieee80211_ctstoself_get(rt2x00dev->hw, tx_info->control.vif,
+					frag_skb->data, size, tx_info,
 					(struct ieee80211_cts *)(skb->data));
 	else
-		ieee80211_rts_get(rt2x00dev->hw, control->vif,
-				  frag_skb->data, frag_skb->len, control,
+		ieee80211_rts_get(rt2x00dev->hw, tx_info->control.vif,
+				  frag_skb->data, size, tx_info,
 				  (struct ieee80211_rts *)(skb->data));
 
 	/*
@@ -68,7 +89,7 @@ static int rt2x00mac_tx_rts_cts(struct rt2x00_dev *rt2x00dev,
 	memset(skbdesc, 0, sizeof(*skbdesc));
 	skbdesc->flags |= FRAME_DESC_DRIVER_GENERATED;
 
-	if (rt2x00dev->ops->lib->write_tx_data(rt2x00dev, queue, skb, control)) {
+	if (rt2x00dev->ops->lib->write_tx_data(rt2x00dev, queue, skb)) {
 		WARNING(rt2x00dev, "Failed to send RTS/CTS frame.\n");
 		return NETDEV_TX_BUSY;
 	}
@@ -76,14 +97,13 @@ static int rt2x00mac_tx_rts_cts(struct rt2x00_dev *rt2x00dev,
 	return NETDEV_TX_OK;
 }
 
-int rt2x00mac_tx(struct ieee80211_hw *hw, struct sk_buff *skb,
-		 struct ieee80211_tx_control *control)
+int rt2x00mac_tx(struct ieee80211_hw *hw, struct sk_buff *skb)
 {
 	struct rt2x00_dev *rt2x00dev = hw->priv;
+	struct ieee80211_tx_info *tx_info = IEEE80211_SKB_CB(skb);
 	struct ieee80211_hdr *ieee80211hdr = (struct ieee80211_hdr *)skb->data;
-	enum data_queue_qid qid = mac80211_queue_to_qid(control->queue);
+	enum data_queue_qid qid = mac80211_queue_to_qid(tx_info->queue);
 	struct data_queue *queue;
-	struct skb_frame_desc *skbdesc;
 	u16 frame_control;
 
 	/*
@@ -100,7 +120,7 @@ int rt2x00mac_tx(struct ieee80211_hw *hw, struct sk_buff *skb,
 	/*
 	 * Determine which queue to put packet on.
 	 */
-	if (control->flags & IEEE80211_TXCTL_SEND_AFTER_DTIM &&
+	if (tx_info->flags & IEEE80211_TX_CTL_SEND_AFTER_DTIM &&
 	    test_bit(DRIVER_REQUIRE_ATIM_QUEUE, &rt2x00dev->flags))
 		queue = rt2x00queue_get_queue(rt2x00dev, QID_ATIM);
 	else
@@ -125,33 +145,27 @@ int rt2x00mac_tx(struct ieee80211_hw *hw, struct sk_buff *skb,
 	 */
 	frame_control = le16_to_cpu(ieee80211hdr->frame_control);
 	if (!is_rts_frame(frame_control) && !is_cts_frame(frame_control) &&
-	    (control->flags & (IEEE80211_TXCTL_USE_RTS_CTS |
-			       IEEE80211_TXCTL_USE_CTS_PROTECT)) &&
+	    (tx_info->flags & (IEEE80211_TX_CTL_USE_RTS_CTS |
+			       IEEE80211_TX_CTL_USE_CTS_PROTECT)) &&
 	    !rt2x00dev->ops->hw->set_rts_threshold) {
 		if (rt2x00queue_available(queue) <= 1) {
-			ieee80211_stop_queue(rt2x00dev->hw, control->queue);
+			ieee80211_stop_queue(rt2x00dev->hw, tx_info->queue);
 			return NETDEV_TX_BUSY;
 		}
 
-		if (rt2x00mac_tx_rts_cts(rt2x00dev, queue, skb, control)) {
-			ieee80211_stop_queue(rt2x00dev->hw, control->queue);
+		if (rt2x00mac_tx_rts_cts(rt2x00dev, queue, skb)) {
+			ieee80211_stop_queue(rt2x00dev->hw, tx_info->queue);
 			return NETDEV_TX_BUSY;
 		}
 	}
 
-	/*
-	 * Initialize skb descriptor
-	 */
-	skbdesc = get_skb_frame_desc(skb);
-	memset(skbdesc, 0, sizeof(*skbdesc));
-
-	if (rt2x00dev->ops->lib->write_tx_data(rt2x00dev, queue, skb, control)) {
-		ieee80211_stop_queue(rt2x00dev->hw, control->queue);
+	if (rt2x00dev->ops->lib->write_tx_data(rt2x00dev, queue, skb)) {
+		ieee80211_stop_queue(rt2x00dev->hw, tx_info->queue);
 		return NETDEV_TX_BUSY;
 	}
 
 	if (rt2x00queue_full(queue))
-		ieee80211_stop_queue(rt2x00dev->hw, control->queue);
+		ieee80211_stop_queue(rt2x00dev->hw, tx_info->queue);
 
 	if (rt2x00dev->ops->lib->kick_tx_queue)
 		rt2x00dev->ops->lib->kick_tx_queue(rt2x00dev, qid);
@@ -380,9 +394,7 @@ int rt2x00mac_config_interface(struct ieee80211_hw *hw,
 	if (conf->type != IEEE80211_IF_TYPE_AP || !conf->beacon)
 		return 0;
 
-	status = rt2x00dev->ops->hw->beacon_update(rt2x00dev->hw,
-						   conf->beacon,
-						   conf->beacon_control);
+	status = rt2x00dev->ops->hw->beacon_update(rt2x00dev->hw, conf->beacon);
 	if (status)
 		dev_kfree_skb(conf->beacon);
 
diff --git a/drivers/net/wireless/rt2x00/rt2x00pci.c b/drivers/net/wireless/rt2x00/rt2x00pci.c
index fa7de41be049..70a3d135f64e 100644
--- a/drivers/net/wireless/rt2x00/rt2x00pci.c
+++ b/drivers/net/wireless/rt2x00/rt2x00pci.c
@@ -35,8 +35,7 @@
  * TX data handlers.
  */
 int rt2x00pci_write_tx_data(struct rt2x00_dev *rt2x00dev,
-			    struct data_queue *queue, struct sk_buff *skb,
-			    struct ieee80211_tx_control *control)
+			    struct data_queue *queue, struct sk_buff *skb)
 {
 	struct queue_entry *entry = rt2x00queue_get_entry(queue, Q_INDEX);
 	struct queue_entry_priv_pci *entry_priv = entry->priv_data;
@@ -64,19 +63,19 @@ int rt2x00pci_write_tx_data(struct rt2x00_dev *rt2x00dev,
 	 * for our information.
 	 */
 	entry->skb = skb;
-	rt2x00queue_create_tx_descriptor(entry, &txdesc, control);
+	rt2x00queue_create_tx_descriptor(entry, &txdesc);
 
 	/*
 	 * Fill in skb descriptor
 	 */
 	skbdesc = get_skb_frame_desc(skb);
+	memset(skbdesc, 0, sizeof(*skbdesc));
 	skbdesc->data = skb->data;
 	skbdesc->data_len = skb->len;
 	skbdesc->desc = entry_priv->desc;
 	skbdesc->desc_len = queue->desc_size;
 	skbdesc->entry = entry;
 
-	memcpy(&entry_priv->control, control, sizeof(entry_priv->control));
 	memcpy(entry_priv->data, skb->data, skb->len);
 
 	rt2x00queue_write_tx_descriptor(entry, &txdesc);
@@ -164,9 +163,9 @@ void rt2x00pci_txdone(struct rt2x00_dev *rt2x00dev, struct queue_entry *entry,
 		      struct txdone_entry_desc *txdesc)
 {
 	struct queue_entry_priv_pci *entry_priv = entry->priv_data;
+	enum data_queue_qid qid = skb_get_queue_mapping(entry->skb);
 	u32 word;
 
-	txdesc->control = &entry_priv->control;
 	rt2x00lib_txdone(entry, txdesc);
 
 	/*
@@ -187,7 +186,7 @@ void rt2x00pci_txdone(struct rt2x00_dev *rt2x00dev, struct queue_entry *entry,
 	 * is reenabled when the txdone handler has finished.
 	 */
 	if (!rt2x00queue_full(entry->queue))
-		ieee80211_wake_queue(rt2x00dev->hw, entry_priv->control.queue);
+		ieee80211_wake_queue(rt2x00dev->hw, qid);
 
 }
 EXPORT_SYMBOL_GPL(rt2x00pci_txdone);
diff --git a/drivers/net/wireless/rt2x00/rt2x00pci.h b/drivers/net/wireless/rt2x00/rt2x00pci.h
index 557d15a888ab..37c851e442c1 100644
--- a/drivers/net/wireless/rt2x00/rt2x00pci.h
+++ b/drivers/net/wireless/rt2x00/rt2x00pci.h
@@ -91,8 +91,7 @@ rt2x00pci_register_multiwrite(struct rt2x00_dev *rt2x00dev,
  * TX data handlers.
  */
 int rt2x00pci_write_tx_data(struct rt2x00_dev *rt2x00dev,
-			    struct data_queue *queue, struct sk_buff *skb,
-			    struct ieee80211_tx_control *control);
+			    struct data_queue *queue, struct sk_buff *skb);
 
 /**
  * struct queue_entry_priv_pci: Per entry PCI specific information
@@ -101,7 +100,6 @@ int rt2x00pci_write_tx_data(struct rt2x00_dev *rt2x00dev,
  * @desc_dma: DMA pointer to &desc.
  * @data: Pointer to device's entry memory.
  * @data_dma: DMA pointer to &data.
- * @control: mac80211 control structure used to transmit data.
  */
 struct queue_entry_priv_pci {
 	__le32 *desc;
@@ -109,8 +107,6 @@ struct queue_entry_priv_pci {
 
 	void *data;
 	dma_addr_t data_dma;
-
-	struct ieee80211_tx_control control;
 };
 
 /**
diff --git a/drivers/net/wireless/rt2x00/rt2x00queue.c b/drivers/net/wireless/rt2x00/rt2x00queue.c
index 5cf4c2f59260..e69ef4b19239 100644
--- a/drivers/net/wireless/rt2x00/rt2x00queue.c
+++ b/drivers/net/wireless/rt2x00/rt2x00queue.c
@@ -30,13 +30,13 @@
 #include "rt2x00lib.h"
 
 void rt2x00queue_create_tx_descriptor(struct queue_entry *entry,
-				      struct txentry_desc *txdesc,
-				      struct ieee80211_tx_control *control)
+				      struct txentry_desc *txdesc)
 {
 	struct rt2x00_dev *rt2x00dev = entry->queue->rt2x00dev;
+	struct ieee80211_tx_info *tx_info = IEEE80211_SKB_CB(entry->skb);
 	struct ieee80211_hdr *hdr = (struct ieee80211_hdr *)entry->skb->data;
 	struct ieee80211_rate *rate =
-	    ieee80211_get_tx_rate(rt2x00dev->hw, control);
+	    ieee80211_get_tx_rate(rt2x00dev->hw, tx_info);
 	const struct rt2x00_rate *hwrate;
 	unsigned int data_length;
 	unsigned int duration;
@@ -64,7 +64,7 @@ void rt2x00queue_create_tx_descriptor(struct queue_entry *entry,
 	/*
 	 * Check whether this frame is to be acked.
 	 */
-	if (!(control->flags & IEEE80211_TXCTL_NO_ACK))
+	if (!(tx_info->flags & IEEE80211_TX_CTL_NO_ACK))
 		__set_bit(ENTRY_TXD_ACK, &txdesc->flags);
 
 	/*
@@ -72,23 +72,20 @@ void rt2x00queue_create_tx_descriptor(struct queue_entry *entry,
 	 */
 	if (is_rts_frame(frame_control) || is_cts_frame(frame_control)) {
 		__set_bit(ENTRY_TXD_BURST, &txdesc->flags);
-		if (is_rts_frame(frame_control)) {
+		if (is_rts_frame(frame_control))
 			__set_bit(ENTRY_TXD_RTS_FRAME, &txdesc->flags);
-			__set_bit(ENTRY_TXD_ACK, &txdesc->flags);
-		} else {
+		else
 			__set_bit(ENTRY_TXD_CTS_FRAME, &txdesc->flags);
-			__clear_bit(ENTRY_TXD_ACK, &txdesc->flags);
-		}
-		if (control->rts_cts_rate_idx >= 0)
+		if (tx_info->control.rts_cts_rate_idx >= 0)
 			rate =
-			    ieee80211_get_rts_cts_rate(rt2x00dev->hw, control);
+			    ieee80211_get_rts_cts_rate(rt2x00dev->hw, tx_info);
 	}
 
 	/*
 	 * Determine retry information.
 	 */
-	txdesc->retry_limit = control->retry_limit;
-	if (control->flags & IEEE80211_TXCTL_LONG_RETRY_LIMIT)
+	txdesc->retry_limit = tx_info->control.retry_limit;
+	if (tx_info->flags & IEEE80211_TX_CTL_LONG_RETRY_LIMIT)
 		__set_bit(ENTRY_TXD_RETRY_MODE, &txdesc->flags);
 
 	/*
@@ -113,7 +110,7 @@ void rt2x00queue_create_tx_descriptor(struct queue_entry *entry,
 	 */
 	if (test_bit(ENTRY_TXD_RTS_FRAME, &txdesc->flags)) {
 		txdesc->ifs = IFS_SIFS;
-	} else if (control->flags & IEEE80211_TXCTL_FIRST_FRAGMENT) {
+	} else if (tx_info->flags & IEEE80211_TX_CTL_FIRST_FRAGMENT) {
 		__set_bit(ENTRY_TXD_FIRST_FRAGMENT, &txdesc->flags);
 		txdesc->ifs = IFS_BACKOFF;
 	} else {
@@ -179,8 +176,10 @@ void rt2x00queue_write_tx_descriptor(struct queue_entry *entry,
 
 	/*
 	 * We are done writing the frame to the queue entry,
-	 * if this entry is a RTS of CTS-to-self frame we are done,
-	 * otherwise we need to kick the queue.
+	 * also kick the queue in case the correct flags are set,
+	 * note that this will automatically filter beacons and
+	 * RTS/CTS frames since those frames don't have this flag
+	 * set.
 	 */
 	if (rt2x00dev->ops->lib->kick_tx_queue &&
 	    !(skbdesc->flags & FRAME_DESC_DRIVER_GENERATED))
diff --git a/drivers/net/wireless/rt2x00/rt2x00queue.h b/drivers/net/wireless/rt2x00/rt2x00queue.h
index c6edc52873c4..f263fe422f87 100644
--- a/drivers/net/wireless/rt2x00/rt2x00queue.h
+++ b/drivers/net/wireless/rt2x00/rt2x00queue.h
@@ -105,8 +105,8 @@ enum skb_frame_desc_flags {
 /**
  * struct skb_frame_desc: Descriptor information for the skb buffer
  *
- * This structure is placed over the skb->cb array, this means that
- * this structure should not exceed the size of that array (48 bytes).
+ * This structure is placed over the driver_data array, this means that
+ * this structure should not exceed the size of that array (40 bytes).
  *
  * @flags: Frame flags, see &enum skb_frame_desc_flags.
  * @data: Pointer to data part of frame (Start of ieee80211 header).
@@ -129,10 +129,15 @@ struct skb_frame_desc {
 	struct queue_entry *entry;
 };
 
+/**
+ * get_skb_frame_desc - Obtain the rt2x00 frame descriptor from a sk_buff.
+ * @skb: &struct sk_buff from where we obtain the &struct skb_frame_desc
+ */
 static inline struct skb_frame_desc* get_skb_frame_desc(struct sk_buff *skb)
 {
-	BUILD_BUG_ON(sizeof(struct skb_frame_desc) > sizeof(skb->cb));
-	return (struct skb_frame_desc *)&skb->cb[0];
+	BUILD_BUG_ON(sizeof(struct skb_frame_desc) >
+		     IEEE80211_TX_INFO_DRIVER_DATA_SIZE);
+	return (struct skb_frame_desc *)&IEEE80211_SKB_CB(skb)->driver_data;
 }
 
 /**
@@ -189,12 +194,10 @@ enum txdone_entry_desc_flags {
  * Summary of information that has been read from the TX frame descriptor
  * after the device is done with transmission.
  *
- * @control: Control structure which was used to transmit the frame.
  * @flags: TX done flags (See &enum txdone_entry_desc_flags).
  * @retry: Retry count.
  */
 struct txdone_entry_desc {
-	struct ieee80211_tx_control *control;
 	unsigned long flags;
 	int retry;
 };
diff --git a/drivers/net/wireless/rt2x00/rt2x00usb.c b/drivers/net/wireless/rt2x00/rt2x00usb.c
index dcee1b4f152b..52d12fdc0ccf 100644
--- a/drivers/net/wireless/rt2x00/rt2x00usb.c
+++ b/drivers/net/wireless/rt2x00/rt2x00usb.c
@@ -129,9 +129,9 @@ static void rt2x00usb_interrupt_txdone(struct urb *urb)
 {
 	struct queue_entry *entry = (struct queue_entry *)urb->context;
 	struct rt2x00_dev *rt2x00dev = entry->queue->rt2x00dev;
-	struct queue_entry_priv_usb *entry_priv = entry->priv_data;
 	struct txdone_entry_desc txdesc;
 	__le32 *txd = (__le32 *)entry->skb->data;
+	enum data_queue_qid qid = skb_get_queue_mapping(entry->skb);
 	u32 word;
 
 	if (!test_bit(DEVICE_ENABLED_RADIO, &rt2x00dev->flags) ||
@@ -159,7 +159,6 @@ static void rt2x00usb_interrupt_txdone(struct urb *urb)
 	else
 		__set_bit(TXDONE_FAILURE, &txdesc.flags);
 	txdesc.retry = 0;
-	txdesc.control = &entry_priv->control;
 
 	rt2x00lib_txdone(entry, &txdesc);
 
@@ -175,12 +174,11 @@ static void rt2x00usb_interrupt_txdone(struct urb *urb)
 	 * is reenabled when the txdone handler has finished.
 	 */
 	if (!rt2x00queue_full(entry->queue))
-		ieee80211_wake_queue(rt2x00dev->hw, entry_priv->control.queue);
+		ieee80211_wake_queue(rt2x00dev->hw, qid);
 }
 
 int rt2x00usb_write_tx_data(struct rt2x00_dev *rt2x00dev,
-			    struct data_queue *queue, struct sk_buff *skb,
-			    struct ieee80211_tx_control *control)
+			    struct data_queue *queue, struct sk_buff *skb)
 {
 	struct usb_device *usb_dev = rt2x00dev_usb_dev(rt2x00dev);
 	struct queue_entry *entry = rt2x00queue_get_entry(queue, Q_INDEX);
@@ -206,7 +204,7 @@ int rt2x00usb_write_tx_data(struct rt2x00_dev *rt2x00dev,
 	 * for our information.
 	 */
 	entry->skb = skb;
-	rt2x00queue_create_tx_descriptor(entry, &txdesc, control);
+	rt2x00queue_create_tx_descriptor(entry, &txdesc);
 
 	/*
 	 * Add the descriptor in front of the skb.
@@ -218,13 +216,13 @@ int rt2x00usb_write_tx_data(struct rt2x00_dev *rt2x00dev,
 	 * Fill in skb descriptor
 	 */
 	skbdesc = get_skb_frame_desc(skb);
+	memset(skbdesc, 0, sizeof(*skbdesc));
 	skbdesc->data = skb->data + queue->desc_size;
 	skbdesc->data_len = skb->len - queue->desc_size;
 	skbdesc->desc = skb->data;
 	skbdesc->desc_len = queue->desc_size;
 	skbdesc->entry = entry;
 
-	memcpy(&entry_priv->control, control, sizeof(entry_priv->control));
 	rt2x00queue_write_tx_descriptor(entry, &txdesc);
 
 	/*
diff --git a/drivers/net/wireless/rt2x00/rt2x00usb.h b/drivers/net/wireless/rt2x00/rt2x00usb.h
index 15b404aa714d..26f53f868af6 100644
--- a/drivers/net/wireless/rt2x00/rt2x00usb.h
+++ b/drivers/net/wireless/rt2x00/rt2x00usb.h
@@ -216,19 +216,15 @@ void rt2x00usb_disable_radio(struct rt2x00_dev *rt2x00dev);
  * TX data handlers.
  */
 int rt2x00usb_write_tx_data(struct rt2x00_dev *rt2x00dev,
-			    struct data_queue *queue, struct sk_buff *skb,
-			    struct ieee80211_tx_control *control);
+			    struct data_queue *queue, struct sk_buff *skb);
 
 /**
  * struct queue_entry_priv_usb: Per entry USB specific information
  *
  * @urb: Urb structure used for device communication.
- * @control: mac80211 control structure used to transmit data.
  */
 struct queue_entry_priv_usb {
 	struct urb *urb;
-
-	struct ieee80211_tx_control control;
 };
 
 /**
@@ -239,15 +235,12 @@ struct queue_entry_priv_usb {
  * with beacons.
  *
  * @urb: Urb structure used for device communication.
- * @control: mac80211 control structure used to transmit data.
  * @guardian_data: Set to 0, used for sending the guardian data.
  * @guardian_urb: Urb structure used to send the guardian data.
  */
 struct queue_entry_priv_usb_bcn {
 	struct urb *urb;
 
-	struct ieee80211_tx_control control;
-
 	unsigned int guardian_data;
 	struct urb *guardian_urb;
 };
diff --git a/drivers/net/wireless/rt2x00/rt61pci.c b/drivers/net/wireless/rt2x00/rt61pci.c
index 7598b6e15784..e13ed5ced26e 100644
--- a/drivers/net/wireless/rt2x00/rt61pci.c
+++ b/drivers/net/wireless/rt2x00/rt61pci.c
@@ -2357,11 +2357,11 @@ static u64 rt61pci_get_tsf(struct ieee80211_hw *hw)
 	return tsf;
 }
 
-static int rt61pci_beacon_update(struct ieee80211_hw *hw, struct sk_buff *skb,
-				 struct ieee80211_tx_control *control)
+static int rt61pci_beacon_update(struct ieee80211_hw *hw, struct sk_buff *skb)
 {
 	struct rt2x00_dev *rt2x00dev = hw->priv;
-	struct rt2x00_intf *intf = vif_to_intf(control->vif);
+	struct ieee80211_tx_info *tx_info = IEEE80211_SKB_CB(skb);
+	struct rt2x00_intf *intf = vif_to_intf(tx_info->control.vif);
 	struct queue_entry_priv_pci *entry_priv;
 	struct skb_frame_desc *skbdesc;
 	struct txentry_desc txdesc;
@@ -2377,7 +2377,7 @@ static int rt61pci_beacon_update(struct ieee80211_hw *hw, struct sk_buff *skb,
 	 * for our information.
 	 */
 	intf->beacon->skb = skb;
-	rt2x00queue_create_tx_descriptor(intf->beacon, &txdesc, control);
+	rt2x00queue_create_tx_descriptor(intf->beacon, &txdesc);
 
 	entry_priv = intf->beacon->priv_data;
 	memset(entry_priv->desc, 0, intf->beacon->queue->desc_size);
diff --git a/drivers/net/wireless/rt2x00/rt73usb.c b/drivers/net/wireless/rt2x00/rt73usb.c
index e55bcbdfc1e1..26c2e0a1a308 100644
--- a/drivers/net/wireless/rt2x00/rt73usb.c
+++ b/drivers/net/wireless/rt2x00/rt73usb.c
@@ -1948,11 +1948,11 @@ static u64 rt73usb_get_tsf(struct ieee80211_hw *hw)
 #define rt73usb_get_tsf	NULL
 #endif
 
-static int rt73usb_beacon_update(struct ieee80211_hw *hw, struct sk_buff *skb,
-				 struct ieee80211_tx_control *control)
+static int rt73usb_beacon_update(struct ieee80211_hw *hw, struct sk_buff *skb)
 {
 	struct rt2x00_dev *rt2x00dev = hw->priv;
-	struct rt2x00_intf *intf = vif_to_intf(control->vif);
+	struct ieee80211_tx_info *tx_info = IEEE80211_SKB_CB(skb);
+	struct rt2x00_intf *intf = vif_to_intf(tx_info->control.vif);
 	struct skb_frame_desc *skbdesc;
 	struct txentry_desc txdesc;
 	unsigned int beacon_base;
@@ -1967,7 +1967,7 @@ static int rt73usb_beacon_update(struct ieee80211_hw *hw, struct sk_buff *skb,
 	 * for our information.
 	 */
 	intf->beacon->skb = skb;
-	rt2x00queue_create_tx_descriptor(intf->beacon, &txdesc, control);
+	rt2x00queue_create_tx_descriptor(intf->beacon, &txdesc);
 
 	/*
 	 * Add the descriptor in front of the skb.
diff --git a/drivers/net/wireless/rtl8180_dev.c b/drivers/net/wireless/rtl8180_dev.c
index 6263209b889e..4427bc9f78a9 100644
--- a/drivers/net/wireless/rtl8180_dev.c
+++ b/drivers/net/wireless/rtl8180_dev.c
@@ -170,34 +170,29 @@ static void rtl8180_handle_tx(struct ieee80211_hw *dev, unsigned int prio)
 	while (skb_queue_len(&ring->queue)) {
 		struct rtl8180_tx_desc *entry = &ring->desc[ring->idx];
 		struct sk_buff *skb;
-		struct ieee80211_tx_status status;
-		struct ieee80211_tx_control *control;
+		struct ieee80211_tx_info *info;
 		u32 flags = le32_to_cpu(entry->flags);
 
 		if (flags & RTL8180_TX_DESC_FLAG_OWN)
 			return;
 
-		memset(&status, 0, sizeof(status));
-
 		ring->idx = (ring->idx + 1) % ring->entries;
 		skb = __skb_dequeue(&ring->queue);
 		pci_unmap_single(priv->pdev, le32_to_cpu(entry->tx_buf),
 				 skb->len, PCI_DMA_TODEVICE);
 
-		control = *((struct ieee80211_tx_control **)skb->cb);
-		if (control)
-			memcpy(&status.control, control, sizeof(*control));
-		kfree(control);
+		info = IEEE80211_SKB_CB(skb);
+		memset(&info->status, 0, sizeof(info->status));
 
-		if (!(status.control.flags & IEEE80211_TXCTL_NO_ACK)) {
+		if (!(info->flags & IEEE80211_TX_CTL_NO_ACK)) {
 			if (flags & RTL8180_TX_DESC_FLAG_TX_OK)
-				status.flags = IEEE80211_TX_STATUS_ACK;
+				info->flags |= IEEE80211_TX_STAT_ACK;
 			else
-				status.excessive_retries = 1;
+				info->status.excessive_retries = 1;
 		}
-		status.retry_count = flags & 0xFF;
+		info->status.retry_count = flags & 0xFF;
 
-		ieee80211_tx_status_irqsafe(dev, skb, &status);
+		ieee80211_tx_status_irqsafe(dev, skb);
 		if (ring->entries - skb_queue_len(&ring->queue) == 2)
 			ieee80211_wake_queue(dev, prio);
 	}
@@ -238,9 +233,9 @@ static irqreturn_t rtl8180_interrupt(int irq, void *dev_id)
 	return IRQ_HANDLED;
 }
 
-static int rtl8180_tx(struct ieee80211_hw *dev, struct sk_buff *skb,
-		      struct ieee80211_tx_control *control)
+static int rtl8180_tx(struct ieee80211_hw *dev, struct sk_buff *skb)
 {
+	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
 	struct rtl8180_priv *priv = dev->priv;
 	struct rtl8180_tx_ring *ring;
 	struct rtl8180_tx_desc *entry;
@@ -251,7 +246,7 @@ static int rtl8180_tx(struct ieee80211_hw *dev, struct sk_buff *skb,
 	u16 plcp_len = 0;
 	__le16 rts_duration = 0;
 
-	prio = control->queue;
+	prio = info->queue;
 	ring = &priv->tx_ring[prio];
 
 	mapping = pci_map_single(priv->pdev, skb->data,
@@ -259,35 +254,32 @@ static int rtl8180_tx(struct ieee80211_hw *dev, struct sk_buff *skb,
 
 	tx_flags = RTL8180_TX_DESC_FLAG_OWN | RTL8180_TX_DESC_FLAG_FS |
 		   RTL8180_TX_DESC_FLAG_LS |
-		   (ieee80211_get_tx_rate(dev, control)->hw_value << 24) |
+		   (ieee80211_get_tx_rate(dev, info)->hw_value << 24) |
 		   skb->len;
 
 	if (priv->r8185)
 		tx_flags |= RTL8180_TX_DESC_FLAG_DMA |
 			    RTL8180_TX_DESC_FLAG_NO_ENC;
 
-	if (control->flags & IEEE80211_TXCTL_USE_RTS_CTS) {
+	if (info->flags & IEEE80211_TX_CTL_USE_RTS_CTS) {
 		tx_flags |= RTL8180_TX_DESC_FLAG_RTS;
-		tx_flags |= ieee80211_get_rts_cts_rate(dev, control)->hw_value << 19;
-	} else if (control->flags & IEEE80211_TXCTL_USE_CTS_PROTECT) {
+		tx_flags |= ieee80211_get_rts_cts_rate(dev, info)->hw_value << 19;
+	} else if (info->flags & IEEE80211_TX_CTL_USE_CTS_PROTECT) {
 		tx_flags |= RTL8180_TX_DESC_FLAG_CTS;
-		tx_flags |= ieee80211_get_rts_cts_rate(dev, control)->hw_value << 19;
+		tx_flags |= ieee80211_get_rts_cts_rate(dev, info)->hw_value << 19;
 	}
 
-	*((struct ieee80211_tx_control **) skb->cb) =
-		kmemdup(control, sizeof(*control), GFP_ATOMIC);
-
-	if (control->flags & IEEE80211_TXCTL_USE_RTS_CTS)
+	if (info->flags & IEEE80211_TX_CTL_USE_RTS_CTS)
 		rts_duration = ieee80211_rts_duration(dev, priv->vif, skb->len,
-						      control);
+						      info);
 
 	if (!priv->r8185) {
 		unsigned int remainder;
 
 		plcp_len = DIV_ROUND_UP(16 * (skb->len + 4),
-				(ieee80211_get_tx_rate(dev, control)->bitrate * 2) / 10);
+				(ieee80211_get_tx_rate(dev, info)->bitrate * 2) / 10);
 		remainder = (16 * (skb->len + 4)) %
-			    ((ieee80211_get_tx_rate(dev, control)->bitrate * 2) / 10);
+			    ((ieee80211_get_tx_rate(dev, info)->bitrate * 2) / 10);
 		if (remainder > 0 && remainder <= 6)
 			plcp_len |= 1 << 15;
 	}
@@ -300,13 +292,13 @@ static int rtl8180_tx(struct ieee80211_hw *dev, struct sk_buff *skb,
 	entry->plcp_len = cpu_to_le16(plcp_len);
 	entry->tx_buf = cpu_to_le32(mapping);
 	entry->frame_len = cpu_to_le32(skb->len);
-	entry->flags2 = control->alt_retry_rate_idx >= 0 ?
-		ieee80211_get_alt_retry_rate(dev, control)->bitrate << 4 : 0;
-	entry->retry_limit = control->retry_limit;
+	entry->flags2 = info->control.alt_retry_rate_idx >= 0 ?
+		ieee80211_get_alt_retry_rate(dev, info)->bitrate << 4 : 0;
+	entry->retry_limit = info->control.retry_limit;
 	entry->flags = cpu_to_le32(tx_flags);
 	__skb_queue_tail(&ring->queue, skb);
 	if (ring->entries - skb_queue_len(&ring->queue) < 2)
-		ieee80211_stop_queue(dev, control->queue);
+		ieee80211_stop_queue(dev, info->queue);
 	spin_unlock_irqrestore(&priv->lock, flags);
 
 	rtl818x_iowrite8(priv, &priv->map->TX_DMA_POLLING, (1 << (prio + 4)));
@@ -522,7 +514,6 @@ static void rtl8180_free_tx_ring(struct ieee80211_hw *dev, unsigned int prio)
 
 		pci_unmap_single(priv->pdev, le32_to_cpu(entry->tx_buf),
 				 skb->len, PCI_DMA_TODEVICE);
-		kfree(*((struct ieee80211_tx_control **) skb->cb));
 		kfree_skb(skb);
 		ring->idx = (ring->idx + 1) % ring->entries;
 	}
diff --git a/drivers/net/wireless/rtl8187.h b/drivers/net/wireless/rtl8187.h
index 076d88b6db0e..a0cfb666de0e 100644
--- a/drivers/net/wireless/rtl8187.h
+++ b/drivers/net/wireless/rtl8187.h
@@ -44,12 +44,6 @@ struct rtl8187_rx_hdr {
 	__le64 mac_time;
 } __attribute__((packed));
 
-struct rtl8187_tx_info {
-	struct ieee80211_tx_control *control;
-	struct urb *urb;
-	struct ieee80211_hw *dev;
-};
-
 struct rtl8187_tx_hdr {
 	__le32 flags;
 #define RTL8187_TX_FLAG_NO_ENCRYPT	(1 << 15)
diff --git a/drivers/net/wireless/rtl8187_dev.c b/drivers/net/wireless/rtl8187_dev.c
index 86a09b49681c..b581ef8a6377 100644
--- a/drivers/net/wireless/rtl8187_dev.c
+++ b/drivers/net/wireless/rtl8187_dev.c
@@ -145,27 +145,22 @@ void rtl8187_write_phy(struct ieee80211_hw *dev, u8 addr, u32 data)
 
 static void rtl8187_tx_cb(struct urb *urb)
 {
-	struct ieee80211_tx_status status;
 	struct sk_buff *skb = (struct sk_buff *)urb->context;
-	struct rtl8187_tx_info *info = (struct rtl8187_tx_info *)skb->cb;
+	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
+	struct ieee80211_hw *hw = info->driver_data[0];
 
-	memset(&status, 0, sizeof(status));
-
-	usb_free_urb(info->urb);
-	if (info->control)
-		memcpy(&status.control, info->control, sizeof(status.control));
-	kfree(info->control);
+	usb_free_urb(info->driver_data[1]);
 	skb_pull(skb, sizeof(struct rtl8187_tx_hdr));
-	status.flags |= IEEE80211_TX_STATUS_ACK;
-	ieee80211_tx_status_irqsafe(info->dev, skb, &status);
+	memset(&info->status, 0, sizeof(info->status));
+	info->flags |= IEEE80211_TX_STAT_ACK;
+	ieee80211_tx_status_irqsafe(hw, skb);
 }
 
-static int rtl8187_tx(struct ieee80211_hw *dev, struct sk_buff *skb,
-		      struct ieee80211_tx_control *control)
+static int rtl8187_tx(struct ieee80211_hw *dev, struct sk_buff *skb)
 {
 	struct rtl8187_priv *priv = dev->priv;
+	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
 	struct rtl8187_tx_hdr *hdr;
-	struct rtl8187_tx_info *info;
 	struct urb *urb;
 	__le16 rts_dur = 0;
 	u32 flags;
@@ -179,29 +174,27 @@ static int rtl8187_tx(struct ieee80211_hw *dev, struct sk_buff *skb,
 	flags = skb->len;
 	flags |= RTL8187_TX_FLAG_NO_ENCRYPT;
 
-	flags |= ieee80211_get_tx_rate(dev, control)->hw_value << 24;
+	flags |= ieee80211_get_tx_rate(dev, info)->hw_value << 24;
 	if (ieee80211_get_morefrag((struct ieee80211_hdr *)skb->data))
 		flags |= RTL8187_TX_FLAG_MORE_FRAG;
-	if (control->flags & IEEE80211_TXCTL_USE_RTS_CTS) {
+	if (info->flags & IEEE80211_TX_CTL_USE_RTS_CTS) {
 		flags |= RTL8187_TX_FLAG_RTS;
-		flags |= ieee80211_get_rts_cts_rate(dev, control)->hw_value << 19;
+		flags |= ieee80211_get_rts_cts_rate(dev, info)->hw_value << 19;
 		rts_dur = ieee80211_rts_duration(dev, priv->vif,
-						 skb->len, control);
-	} else if (control->flags & IEEE80211_TXCTL_USE_CTS_PROTECT) {
+						 skb->len, info);
+	} else if (info->flags & IEEE80211_TX_CTL_USE_CTS_PROTECT) {
 		flags |= RTL8187_TX_FLAG_CTS;
-		flags |= ieee80211_get_rts_cts_rate(dev, control)->hw_value << 19;
+		flags |= ieee80211_get_rts_cts_rate(dev, info)->hw_value << 19;
 	}
 
 	hdr = (struct rtl8187_tx_hdr *)skb_push(skb, sizeof(*hdr));
 	hdr->flags = cpu_to_le32(flags);
 	hdr->len = 0;
 	hdr->rts_duration = rts_dur;
-	hdr->retry = cpu_to_le32(control->retry_limit << 8);
+	hdr->retry = cpu_to_le32(info->control.retry_limit << 8);
 
-	info = (struct rtl8187_tx_info *)skb->cb;
-	info->control = kmemdup(control, sizeof(*control), GFP_ATOMIC);
-	info->urb = urb;
-	info->dev = dev;
+	info->driver_data[0] = dev;
+	info->driver_data[1] = urb;
 	usb_fill_bulk_urb(urb, priv->udev, usb_sndbulkpipe(priv->udev, 2),
 			  hdr, skb->len, rtl8187_tx_cb, skb);
 	usb_submit_urb(urb, GFP_ATOMIC);
diff --git a/drivers/net/wireless/zd1211rw/zd_mac.c b/drivers/net/wireless/zd1211rw/zd_mac.c
index 99c508c09e5b..edb1aefb4add 100644
--- a/drivers/net/wireless/zd1211rw/zd_mac.c
+++ b/drivers/net/wireless/zd1211rw/zd_mac.c
@@ -224,36 +224,6 @@ out:
 	return r;
 }
 
-/**
- * clear_tx_skb_control_block - clears the control block of tx skbuffs
- * @skb: a &struct sk_buff pointer
- *
- * This clears the control block of skbuff buffers, which were transmitted to
- * the device. Notify that the function is not thread-safe, so prevent
- * multiple calls.
- */
-static void clear_tx_skb_control_block(struct sk_buff *skb)
-{
-	struct zd_tx_skb_control_block *cb =
-		(struct zd_tx_skb_control_block *)skb->cb;
-
-	kfree(cb->control);
-	cb->control = NULL;
-}
-
-/**
- * kfree_tx_skb - frees a tx skbuff
- * @skb: a &struct sk_buff pointer
- *
- * Frees the tx skbuff. Frees also the allocated control structure in the
- * control block if necessary.
- */
-static void kfree_tx_skb(struct sk_buff *skb)
-{
-	clear_tx_skb_control_block(skb);
-	dev_kfree_skb_any(skb);
-}
-
 static void zd_op_stop(struct ieee80211_hw *hw)
 {
 	struct zd_mac *mac = zd_hw_mac(hw);
@@ -276,40 +246,15 @@ static void zd_op_stop(struct ieee80211_hw *hw)
 
 
 	while ((skb = skb_dequeue(ack_wait_queue)))
-		kfree_tx_skb(skb);
-}
-
-/**
- * init_tx_skb_control_block - initializes skb control block
- * @skb: a &sk_buff pointer
- * @dev: pointer to the mac80221 device
- * @control: mac80211 tx control applying for the frame in @skb
- *
- * Initializes the control block of the skbuff to be transmitted.
- */
-static int init_tx_skb_control_block(struct sk_buff *skb,
-				     struct ieee80211_hw *hw,
-	                             struct ieee80211_tx_control *control)
-{
-	struct zd_tx_skb_control_block *cb =
-		(struct zd_tx_skb_control_block *)skb->cb;
-
-	ZD_ASSERT(sizeof(*cb) <= sizeof(skb->cb));
-	memset(cb, 0, sizeof(*cb));
-	cb->hw= hw;
-	cb->control = kmalloc(sizeof(*control), GFP_ATOMIC);
-	if (cb->control == NULL)
-		return -ENOMEM;
-	memcpy(cb->control, control, sizeof(*control));
-
-	return 0;
+		dev_kfree_skb_any(skb);
 }
 
 /**
  * tx_status - reports tx status of a packet if required
  * @hw - a &struct ieee80211_hw pointer
  * @skb - a sk-buffer
- * @status - the tx status of the packet without control information
+ * @flags: extra flags to set in the TX status info
+ * @ackssi: ACK signal strength
  * @success - True for successfull transmission of the frame
  *
  * This information calls ieee80211_tx_status_irqsafe() if required by the
@@ -319,18 +264,17 @@ static int init_tx_skb_control_block(struct sk_buff *skb,
  * If no status information has been requested, the skb is freed.
  */
 static void tx_status(struct ieee80211_hw *hw, struct sk_buff *skb,
-	              struct ieee80211_tx_status *status,
-		      bool success)
+		      u32 flags, int ackssi, bool success)
 {
-	struct zd_tx_skb_control_block *cb = (struct zd_tx_skb_control_block *)
-		skb->cb;
+	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
+
+	memset(&info->status, 0, sizeof(info->status));
 
-	ZD_ASSERT(cb->control != NULL);
-	memcpy(&status->control, cb->control, sizeof(status->control));
 	if (!success)
-		status->excessive_retries = 1;
-	clear_tx_skb_control_block(skb);
-	ieee80211_tx_status_irqsafe(hw, skb, status);
+		info->status.excessive_retries = 1;
+	info->flags |= flags;
+	info->status.ack_signal = ackssi;
+	ieee80211_tx_status_irqsafe(hw, skb);
 }
 
 /**
@@ -345,15 +289,12 @@ void zd_mac_tx_failed(struct ieee80211_hw *hw)
 {
 	struct sk_buff_head *q = &zd_hw_mac(hw)->ack_wait_queue;
 	struct sk_buff *skb;
-	struct ieee80211_tx_status status;
 
 	skb = skb_dequeue(q);
 	if (skb == NULL)
 		return;
 
-	memset(&status, 0, sizeof(status));
-
-	tx_status(hw, skb, &status, 0);
+	tx_status(hw, skb, 0, 0, 0);
 }
 
 /**
@@ -368,28 +309,20 @@ void zd_mac_tx_failed(struct ieee80211_hw *hw)
  */
 void zd_mac_tx_to_dev(struct sk_buff *skb, int error)
 {
-	struct zd_tx_skb_control_block *cb =
-		(struct zd_tx_skb_control_block *)skb->cb;
-	struct ieee80211_hw *hw = cb->hw;
-
-	if (likely(cb->control)) {
-		skb_pull(skb, sizeof(struct zd_ctrlset));
-		if (unlikely(error ||
-		    (cb->control->flags & IEEE80211_TXCTL_NO_ACK)))
-		{
-			struct ieee80211_tx_status status;
-			memset(&status, 0, sizeof(status));
-			tx_status(hw, skb, &status, !error);
-		} else {
-			struct sk_buff_head *q =
-				&zd_hw_mac(hw)->ack_wait_queue;
+	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
+	struct ieee80211_hw *hw = info->driver_data[0];
 
-			skb_queue_tail(q, skb);
-			while (skb_queue_len(q) > ZD_MAC_MAX_ACK_WAITERS)
-				zd_mac_tx_failed(hw);
-		}
+	skb_pull(skb, sizeof(struct zd_ctrlset));
+	if (unlikely(error ||
+	    (info->flags & IEEE80211_TX_CTL_NO_ACK))) {
+		tx_status(hw, skb, 0, 0, !error);
 	} else {
-		kfree_tx_skb(skb);
+		struct sk_buff_head *q =
+			&zd_hw_mac(hw)->ack_wait_queue;
+
+		skb_queue_tail(q, skb);
+		while (skb_queue_len(q) > ZD_MAC_MAX_ACK_WAITERS)
+			zd_mac_tx_failed(hw);
 	}
 }
 
@@ -454,7 +387,7 @@ static void cs_set_control(struct zd_mac *mac, struct zd_ctrlset *cs,
 	cs->control = 0;
 
 	/* First fragment */
-	if (flags & IEEE80211_TXCTL_FIRST_FRAGMENT)
+	if (flags & IEEE80211_TX_CTL_FIRST_FRAGMENT)
 		cs->control |= ZD_CS_NEED_RANDOM_BACKOFF;
 
 	/* Multicast */
@@ -466,10 +399,10 @@ static void cs_set_control(struct zd_mac *mac, struct zd_ctrlset *cs,
 	    (IEEE80211_FTYPE_CTL|IEEE80211_STYPE_PSPOLL))
 		cs->control |= ZD_CS_PS_POLL_FRAME;
 
-	if (flags & IEEE80211_TXCTL_USE_RTS_CTS)
+	if (flags & IEEE80211_TX_CTL_USE_RTS_CTS)
 		cs->control |= ZD_CS_RTS;
 
-	if (flags & IEEE80211_TXCTL_USE_CTS_PROTECT)
+	if (flags & IEEE80211_TX_CTL_USE_CTS_PROTECT)
 		cs->control |= ZD_CS_SELF_CTS;
 
 	/* FIXME: Management frame? */
@@ -516,8 +449,7 @@ void zd_mac_config_beacon(struct ieee80211_hw *hw, struct sk_buff *beacon)
 }
 
 static int fill_ctrlset(struct zd_mac *mac,
-			struct sk_buff *skb,
-			struct ieee80211_tx_control *control)
+			struct sk_buff *skb)
 {
 	int r;
 	struct ieee80211_hdr *hdr = (struct ieee80211_hdr *) skb->data;
@@ -526,18 +458,19 @@ static int fill_ctrlset(struct zd_mac *mac,
 	struct ieee80211_rate *txrate;
 	struct zd_ctrlset *cs = (struct zd_ctrlset *)
 		skb_push(skb, sizeof(struct zd_ctrlset));
+	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
 
 	ZD_ASSERT(frag_len <= 0xffff);
 
-	txrate = ieee80211_get_tx_rate(mac->hw, control);
+	txrate = ieee80211_get_tx_rate(mac->hw, info);
 
 	cs->modulation = txrate->hw_value;
-	if (control->flags & IEEE80211_TXCTL_SHORT_PREAMBLE)
+	if (info->flags & IEEE80211_TX_CTL_SHORT_PREAMBLE)
 		cs->modulation = txrate->hw_value_short;
 
 	cs->tx_length = cpu_to_le16(frag_len);
 
-	cs_set_control(mac, cs, hdr, control->flags);
+	cs_set_control(mac, cs, hdr, info->flags);
 
 	packet_length = frag_len + sizeof(struct zd_ctrlset) + 10;
 	ZD_ASSERT(packet_length <= 0xffff);
@@ -582,24 +515,21 @@ static int fill_ctrlset(struct zd_mac *mac,
  * control block of the skbuff will be initialized. If necessary the incoming
  * mac80211 queues will be stopped.
  */
-static int zd_op_tx(struct ieee80211_hw *hw, struct sk_buff *skb,
-		     struct ieee80211_tx_control *control)
+static int zd_op_tx(struct ieee80211_hw *hw, struct sk_buff *skb)
 {
 	struct zd_mac *mac = zd_hw_mac(hw);
+	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
 	int r;
 
-	r = fill_ctrlset(mac, skb, control);
+	r = fill_ctrlset(mac, skb);
 	if (r)
 		return r;
 
-	r = init_tx_skb_control_block(skb, hw, control);
-	if (r)
-		return r;
+	info->driver_data[0] = hw;
+
 	r = zd_usb_tx(&mac->chip.usb, skb);
-	if (r) {
-		clear_tx_skb_control_block(skb);
+	if (r)
 		return r;
-	}
 	return 0;
 }
 
@@ -637,13 +567,8 @@ static int filter_ack(struct ieee80211_hw *hw, struct ieee80211_hdr *rx_hdr,
 		tx_hdr = (struct ieee80211_hdr *)skb->data;
 		if (likely(!compare_ether_addr(tx_hdr->addr2, rx_hdr->addr1)))
 		{
-			struct ieee80211_tx_status status;
-
-			memset(&status, 0, sizeof(status));
-			status.flags = IEEE80211_TX_STATUS_ACK;
-			status.ack_signal = stats->signal;
 			__skb_unlink(skb, q);
-			tx_status(hw, skb, &status, 1);
+			tx_status(hw, skb, IEEE80211_TX_STAT_ACK, stats->signal, 1);
 			goto out;
 		}
 	}
@@ -947,8 +872,7 @@ static void zd_op_bss_info_changed(struct ieee80211_hw *hw,
 }
 
 static int zd_op_beacon_update(struct ieee80211_hw *hw,
-			       struct sk_buff *skb,
-			       struct ieee80211_tx_control *ctl)
+			       struct sk_buff *skb)
 {
 	struct zd_mac *mac = zd_hw_mac(hw);
 	zd_mac_config_beacon(hw, skb);
diff --git a/drivers/net/wireless/zd1211rw/zd_mac.h b/drivers/net/wireless/zd1211rw/zd_mac.h
index 71170244d2c9..18c1d56d3dd7 100644
--- a/drivers/net/wireless/zd1211rw/zd_mac.h
+++ b/drivers/net/wireless/zd1211rw/zd_mac.h
@@ -149,22 +149,6 @@ struct housekeeping {
 	struct delayed_work link_led_work;
 };
 
-/**
- * struct zd_tx_skb_control_block - control block for tx skbuffs
- * @control: &struct ieee80211_tx_control pointer
- * @context: context pointer
- *
- * This structure is used to fill the cb field in an &sk_buff to transmit.
- * The control field is NULL, if there is no requirement from the mac80211
- * stack to report about the packet ACK. This is the case if the flag
- * IEEE80211_TXCTL_NO_ACK is not set in &struct ieee80211_tx_control.
- */
-struct zd_tx_skb_control_block {
-	struct ieee80211_tx_control *control;
-	struct ieee80211_hw *hw;
-	void *context;
-};
-
 #define ZD_MAC_STATS_BUFFER_SIZE 16
 
 #define ZD_MAC_MAX_ACK_WAITERS 10
diff --git a/drivers/net/wireless/zd1211rw/zd_usb.c b/drivers/net/wireless/zd1211rw/zd_usb.c
index 12e24f04dddf..c8a0b34aecc8 100644
--- a/drivers/net/wireless/zd1211rw/zd_usb.c
+++ b/drivers/net/wireless/zd1211rw/zd_usb.c
@@ -869,7 +869,7 @@ static void tx_urb_complete(struct urb *urb)
 {
 	int r;
 	struct sk_buff *skb;
-	struct zd_tx_skb_control_block *cb;
+	struct ieee80211_tx_info *info;
 	struct zd_usb *usb;
 
 	switch (urb->status) {
@@ -893,8 +893,8 @@ free_urb:
 	 * grab 'usb' pointer before handing off the skb (since
 	 * it might be freed by zd_mac_tx_to_dev or mac80211)
 	 */
-	cb = (struct zd_tx_skb_control_block *)skb->cb;
-	usb = &zd_hw_mac(cb->hw)->chip.usb;
+	info = IEEE80211_SKB_CB(skb);
+	usb = &zd_hw_mac(info->driver_data[0])->chip.usb;
 	zd_mac_tx_to_dev(skb, urb->status);
 	free_tx_urb(usb, urb);
 	tx_dec_submitted_urbs(usb);
diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index 0df91bea6c1f..54960b83db79 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -201,101 +201,127 @@ struct ieee80211_bss_conf {
 };
 
 /**
- * enum mac80211_tx_control_flags - flags to describe Tx configuration for
- * 				    the Tx frame
- *
- * These flags are used with the @flags member of &ieee80211_tx_control
- *
- * @IEEE80211_TXCTL_REQ_TX_STATUS: request TX status callback for this frame.
- * @IEEE80211_TXCTL_DO_NOT_ENCRYPT: send this frame without encryption;
- * 				    e.g., for EAPOL frame
- * @IEEE80211_TXCTL_USE_RTS_CTS: use RTS-CTS before sending frame
- * @IEEE80211_TXCTL_USE_CTS_PROTECT: use CTS protection for the frame (e.g.,
- * 				     for combined 802.11g / 802.11b networks)
- * @IEEE80211_TXCTL_NO_ACK: tell the low level not to wait for an ack
- * @IEEE80211_TXCTL_RATE_CTRL_PROBE
- * @EEE80211_TXCTL_CLEAR_PS_FILT: clear powersave filter
- *                                 for destination station
- * @IEEE80211_TXCTL_REQUEUE:
- * @IEEE80211_TXCTL_FIRST_FRAGMENT: this is a first fragment of the frame
- * @IEEE80211_TXCTL_LONG_RETRY_LIMIT: this frame should be send using the
- * 				      through set_retry_limit configured long
- * 				      retry value
- * @IEEE80211_TXCTL_EAPOL_FRAME: internal to mac80211
- * @IEEE80211_TXCTL_SEND_AFTER_DTIM: send this frame after DTIM beacon
- * @IEEE80211_TXCTL_AMPDU: this frame should be sent as part of an A-MPDU
- * @IEEE80211_TXCTL_OFDM_HT: this frame can be sent in HT OFDM rates. number
- * 			     of streams when this flag is on can be extracted
- *			     from antenna_sel_tx, so if 1 antenna is marked
- *			     use SISO, 2 antennas marked use MIMO, n antennas
- *			     marked use MIMO_n.
- * @IEEE80211_TXCTL_GREEN_FIELD: use green field protection for this frame
- * @IEEE80211_TXCTL_40_MHZ_WIDTH: send this frame using 40 Mhz channel width
- * @IEEE80211_TXCTL_DUP_DATA: duplicate data frame on both 20 Mhz channels
- * @IEEE80211_TXCTL_SHORT_GI: send this frame using short guard interval
+ * enum mac80211_tx_flags - flags to transmission information/status
+ *
+ * These flags are used with the @flags member of &ieee80211_tx_info
+ *
+ * @IEEE80211_TX_CTL_REQ_TX_STATUS: request TX status callback for this frame.
+ * @IEEE80211_TX_CTL_DO_NOT_ENCRYPT: send this frame without encryption;
+ *	e.g., for EAPOL frame
+ * @IEEE80211_TX_CTL_USE_RTS_CTS: use RTS-CTS before sending frame
+ * @IEEE80211_TX_CTL_USE_CTS_PROTECT: use CTS protection for the frame (e.g.,
+ *	for combined 802.11g / 802.11b networks)
+ * @IEEE80211_TX_CTL_NO_ACK: tell the low level not to wait for an ack
+ * @IEEE80211_TX_CTL_RATE_CTRL_PROBE
+ * @IEEE80211_TX_CTL_CLEAR_PS_FILT: clear powersave filter for destination
+ *	station
+ * @IEEE80211_TX_CTL_REQUEUE:
+ * @IEEE80211_TX_CTL_FIRST_FRAGMENT: this is a first fragment of the frame
+ * @IEEE80211_TX_CTL_LONG_RETRY_LIMIT: this frame should be send using the
+ *	through set_retry_limit configured long retry value
+ * @IEEE80211_TX_CTL_EAPOL_FRAME: internal to mac80211
+ * @IEEE80211_TX_CTL_SEND_AFTER_DTIM: send this frame after DTIM beacon
+ * @IEEE80211_TX_CTL_AMPDU: this frame should be sent as part of an A-MPDU
+ * @IEEE80211_TX_CTL_OFDM_HT: this frame can be sent in HT OFDM rates. number
+ *	of streams when this flag is on can be extracted from antenna_sel_tx,
+ *	so if 1 antenna is marked use SISO, 2 antennas marked use MIMO, n
+ *	antennas marked use MIMO_n.
+ * @IEEE80211_TX_CTL_GREEN_FIELD: use green field protection for this frame
+ * @IEEE80211_TX_CTL_40_MHZ_WIDTH: send this frame using 40 Mhz channel width
+ * @IEEE80211_TX_CTL_DUP_DATA: duplicate data frame on both 20 Mhz channels
+ * @IEEE80211_TX_CTL_SHORT_GI: send this frame using short guard interval
+ * @IEEE80211_TX_STAT_TX_FILTERED: The frame was not transmitted
+ *	because the destination STA was in powersave mode.
+ * @IEEE80211_TX_STAT_ACK: Frame was acknowledged
+ * @IEEE80211_TX_STAT_AMPDU: The frame was aggregated, so status
+ * 	is for the whole aggregation.
  */
 enum mac80211_tx_control_flags {
-	IEEE80211_TXCTL_REQ_TX_STATUS		= (1<<0),
-	IEEE80211_TXCTL_DO_NOT_ENCRYPT		= (1<<1),
-	IEEE80211_TXCTL_USE_RTS_CTS		= (1<<2),
-	IEEE80211_TXCTL_USE_CTS_PROTECT		= (1<<3),
-	IEEE80211_TXCTL_NO_ACK			= (1<<4),
-	IEEE80211_TXCTL_RATE_CTRL_PROBE		= (1<<5),
-	IEEE80211_TXCTL_CLEAR_PS_FILT		= (1<<6),
-	IEEE80211_TXCTL_REQUEUE			= (1<<7),
-	IEEE80211_TXCTL_FIRST_FRAGMENT		= (1<<8),
-	IEEE80211_TXCTL_SHORT_PREAMBLE		= (1<<9),
-	IEEE80211_TXCTL_LONG_RETRY_LIMIT	= (1<<10),
-	IEEE80211_TXCTL_EAPOL_FRAME		= (1<<11),
-	IEEE80211_TXCTL_SEND_AFTER_DTIM		= (1<<12),
-	IEEE80211_TXCTL_AMPDU			= (1<<13),
-	IEEE80211_TXCTL_OFDM_HT			= (1<<14),
-	IEEE80211_TXCTL_GREEN_FIELD		= (1<<15),
-	IEEE80211_TXCTL_40_MHZ_WIDTH		= (1<<16),
-	IEEE80211_TXCTL_DUP_DATA		= (1<<17),
-	IEEE80211_TXCTL_SHORT_GI		= (1<<18),
+	IEEE80211_TX_CTL_REQ_TX_STATUS		= BIT(0),
+	IEEE80211_TX_CTL_DO_NOT_ENCRYPT		= BIT(1),
+	IEEE80211_TX_CTL_USE_RTS_CTS		= BIT(2),
+	IEEE80211_TX_CTL_USE_CTS_PROTECT	= BIT(3),
+	IEEE80211_TX_CTL_NO_ACK			= BIT(4),
+	IEEE80211_TX_CTL_RATE_CTRL_PROBE	= BIT(5),
+	IEEE80211_TX_CTL_CLEAR_PS_FILT		= BIT(6),
+	IEEE80211_TX_CTL_REQUEUE		= BIT(7),
+	IEEE80211_TX_CTL_FIRST_FRAGMENT		= BIT(8),
+	IEEE80211_TX_CTL_SHORT_PREAMBLE		= BIT(9),
+	IEEE80211_TX_CTL_LONG_RETRY_LIMIT	= BIT(10),
+	IEEE80211_TX_CTL_EAPOL_FRAME		= BIT(11),
+	IEEE80211_TX_CTL_SEND_AFTER_DTIM	= BIT(12),
+	IEEE80211_TX_CTL_AMPDU			= BIT(13),
+	IEEE80211_TX_CTL_OFDM_HT		= BIT(14),
+	IEEE80211_TX_CTL_GREEN_FIELD		= BIT(15),
+	IEEE80211_TX_CTL_40_MHZ_WIDTH		= BIT(16),
+	IEEE80211_TX_CTL_DUP_DATA		= BIT(17),
+	IEEE80211_TX_CTL_SHORT_GI		= BIT(18),
+	IEEE80211_TX_CTL_INJECTED		= BIT(19),
+	IEEE80211_TX_STAT_TX_FILTERED		= BIT(20),
+	IEEE80211_TX_STAT_ACK			= BIT(21),
+	IEEE80211_TX_STAT_AMPDU			= BIT(22),
 };
 
-/* Transmit control fields. This data structure is passed to low-level driver
- * with each TX frame. The low-level driver is responsible for configuring
- * the hardware to use given values (depending on what is supported).
- *
- * NOTE: Be careful with using the pointers outside of the ieee80211_ops->tx()
- * context (i.e. when defering the work to a workqueue).
- * The vif pointer is valid until the it has been removed with the
- * ieee80211_ops->remove_interface() callback funtion.
- * The hw_key pointer is valid until it has been removed with the
- * ieee80211_ops->set_key() callback function.
- */
-struct ieee80211_tx_control {
-	u32 flags;		/* tx control flags defined above */
-
-	s8 tx_rate_idx,		/* Transmit rate (indexes registered rates) */
-	   rts_cts_rate_idx,	/* Transmit rate for RTS/CTS frame */
-	   alt_retry_rate_idx;	/* retry rate for the last retries */
-
-	s8 retry_limit;		/* 1 = only first attempt, 2 = one retry, ..
-				 * This could be used when set_retry_limit
-				 * is not implemented by the driver */
 
-	struct ieee80211_vif *vif;
-
-	/* Key used for hardware encryption
-	 * NULL if IEEE80211_TXCTL_DO_NOT_ENCRYPT is set */
-	struct ieee80211_key_conf *hw_key;
+#define IEEE80211_TX_INFO_DRIVER_DATA_SIZE \
+	(sizeof(((struct sk_buff *)0)->cb) - 8)
+#define IEEE80211_TX_INFO_DRIVER_DATA_PTRS \
+	(IEEE80211_TX_INFO_DRIVER_DATA_SIZE / sizeof(void *))
 
-	enum ieee80211_band band;
+/**
+ * struct ieee80211_tx_info - skb transmit information
+ *
+ * This structure is placed in skb->cb for three uses:
+ *  (1) mac80211 TX control - mac80211 tells the driver what to do
+ *  (2) driver internal use (if applicable)
+ *  (3) TX status information - driver tells mac80211 what happened
+ *
+ * @flags: transmit info flags, defined above
+ * @retry_count: number of retries
+ * @excessive_retries: set to 1 if the frame was retried many times
+ *	but not acknowledged
+ * @ampdu_ack_len: number of aggregated frames.
+ * 	relevant only if IEEE80211_TX_STATUS_AMPDU was set.
+ * @ampdu_ack_map: block ack bit map for the aggregation.
+ * 	relevant only if IEEE80211_TX_STATUS_AMPDU was set.
+ * @ack_signal: signal strength of the ACK frame
+ */
+struct ieee80211_tx_info {
+	/* common information */
+	u32 flags;
+	u8 band;
+	s8 tx_rate_idx;
+	u8 antenna_sel_tx;
 
-	u8 antenna_sel_tx; 	/* 0 = default/diversity, otherwise bit
-				 * position represents antenna number used */
-	u8 icv_len;		/* length of the ICV/MIC field in octets */
-	u8 iv_len;		/* length of the IV field in octets */
-	u16 queue;		/* hardware queue to use for this frame;
-				 * 0 = highest, hw->queues-1 = lowest */
-	u16 aid;		/* Station AID */
-	int type;	/* internal */
+	u8 queue; /* use skb_queue_mapping soon */
+
+	union {
+		struct {
+			struct ieee80211_vif *vif;
+			struct ieee80211_key_conf *hw_key;
+			unsigned long jiffies;
+			int ifindex;
+			u16 aid;
+			s8 rts_cts_rate_idx, alt_retry_rate_idx;
+			u8 retry_limit;
+			u8 icv_len;
+			u8 iv_len;
+		} control;
+		struct {
+			u64 ampdu_ack_map;
+			int ack_signal;
+			u8 retry_count;
+			bool excessive_retries;
+			u8 ampdu_ack_len;
+		} status;
+		void *driver_data[IEEE80211_TX_INFO_DRIVER_DATA_PTRS];
+	};
 };
 
+static inline struct ieee80211_tx_info *IEEE80211_SKB_CB(struct sk_buff *skb)
+{
+	return (struct ieee80211_tx_info *)skb->cb;
+}
 
 
 /**
@@ -362,52 +388,6 @@ struct ieee80211_rx_status {
 	int flag;
 };
 
-/**
- * enum ieee80211_tx_status_flags - transmit status flags
- *
- * Status flags to indicate various transmit conditions.
- *
- * @IEEE80211_TX_STATUS_TX_FILTERED: The frame was not transmitted
- *	because the destination STA was in powersave mode.
- * @IEEE80211_TX_STATUS_ACK: Frame was acknowledged
- * @IEEE80211_TX_STATUS_AMPDU: The frame was aggregated, so status
- * 	is for the whole aggregation.
- */
-enum ieee80211_tx_status_flags {
-	IEEE80211_TX_STATUS_TX_FILTERED	= 1<<0,
-	IEEE80211_TX_STATUS_ACK		= 1<<1,
-	IEEE80211_TX_STATUS_AMPDU	= 1<<2,
-};
-
-/**
- * struct ieee80211_tx_status - transmit status
- *
- * As much information as possible should be provided for each transmitted
- * frame with ieee80211_tx_status().
- *
- * @control: a copy of the &struct ieee80211_tx_control passed to the driver
- *	in the tx() callback.
- * @flags: transmit status flags, defined above
- * @retry_count: number of retries
- * @excessive_retries: set to 1 if the frame was retried many times
- *	but not acknowledged
- * @ampdu_ack_len: number of aggregated frames.
- * 	relevant only if IEEE80211_TX_STATUS_AMPDU was set.
- * @ampdu_ack_map: block ack bit map for the aggregation.
- * 	relevant only if IEEE80211_TX_STATUS_AMPDU was set.
- * @ack_signal: signal strength of the ACK frame either in dBm, dB or unspec
- *	depending on hardware capabilites flags @IEEE80211_HW_SIGNAL_*
- */
-struct ieee80211_tx_status {
-	struct ieee80211_tx_control control;
-	u8 flags;
-	u8 retry_count;
-	bool excessive_retries;
-	u8 ampdu_ack_len;
-	u64 ampdu_ack_map;
-	int ack_signal;
-};
-
 /**
  * enum ieee80211_conf_flags - configuration flags
  *
@@ -563,7 +543,6 @@ struct ieee80211_if_conf {
 	u8 *ssid;
 	size_t ssid_len;
 	struct sk_buff *beacon;
-	struct ieee80211_tx_control *beacon_control;
 };
 
 /**
@@ -825,7 +804,7 @@ static inline void SET_IEEE80211_PERM_ADDR(struct ieee80211_hw *hw, u8 *addr)
 
 static inline struct ieee80211_rate *
 ieee80211_get_tx_rate(const struct ieee80211_hw *hw,
-		      const struct ieee80211_tx_control *c)
+		      const struct ieee80211_tx_info *c)
 {
 	if (WARN_ON(c->tx_rate_idx < 0))
 		return NULL;
@@ -834,20 +813,20 @@ ieee80211_get_tx_rate(const struct ieee80211_hw *hw,
 
 static inline struct ieee80211_rate *
 ieee80211_get_rts_cts_rate(const struct ieee80211_hw *hw,
-			   const struct ieee80211_tx_control *c)
+			   const struct ieee80211_tx_info *c)
 {
-	if (c->rts_cts_rate_idx < 0)
+	if (c->control.rts_cts_rate_idx < 0)
 		return NULL;
-	return &hw->wiphy->bands[c->band]->bitrates[c->rts_cts_rate_idx];
+	return &hw->wiphy->bands[c->band]->bitrates[c->control.rts_cts_rate_idx];
 }
 
 static inline struct ieee80211_rate *
 ieee80211_get_alt_retry_rate(const struct ieee80211_hw *hw,
-			     const struct ieee80211_tx_control *c)
+			     const struct ieee80211_tx_info *c)
 {
-	if (c->alt_retry_rate_idx < 0)
+	if (c->control.alt_retry_rate_idx < 0)
 		return NULL;
-	return &hw->wiphy->bands[c->band]->bitrates[c->alt_retry_rate_idx];
+	return &hw->wiphy->bands[c->band]->bitrates[c->control.alt_retry_rate_idx];
 }
 
 /**
@@ -1142,8 +1121,7 @@ enum ieee80211_ampdu_mlme_action {
  * 	that TX/RX_STOP can pass NULL for this parameter.
  */
 struct ieee80211_ops {
-	int (*tx)(struct ieee80211_hw *hw, struct sk_buff *skb,
-		  struct ieee80211_tx_control *control);
+	int (*tx)(struct ieee80211_hw *hw, struct sk_buff *skb);
 	int (*start)(struct ieee80211_hw *hw);
 	void (*stop)(struct ieee80211_hw *hw);
 	int (*add_interface)(struct ieee80211_hw *hw,
@@ -1187,8 +1165,7 @@ struct ieee80211_ops {
 	u64 (*get_tsf)(struct ieee80211_hw *hw);
 	void (*reset_tsf)(struct ieee80211_hw *hw);
 	int (*beacon_update)(struct ieee80211_hw *hw,
-			     struct sk_buff *skb,
-			     struct ieee80211_tx_control *control);
+			     struct sk_buff *skb);
 	int (*tx_last_beacon)(struct ieee80211_hw *hw);
 	int (*ampdu_action)(struct ieee80211_hw *hw,
 			    enum ieee80211_ampdu_mlme_action action,
@@ -1384,13 +1361,9 @@ void ieee80211_rx_irqsafe(struct ieee80211_hw *hw,
  *
  * @hw: the hardware the frame was transmitted by
  * @skb: the frame that was transmitted, owned by mac80211 after this call
- * @status: status information for this frame; the status pointer need not
- *	be valid after this function returns and is not freed by mac80211,
- *	it is recommended that it points to a stack area
  */
 void ieee80211_tx_status(struct ieee80211_hw *hw,
-			 struct sk_buff *skb,
-			 struct ieee80211_tx_status *status);
+			 struct sk_buff *skb);
 
 /**
  * ieee80211_tx_status_irqsafe - irq-safe transmit status callback
@@ -1403,13 +1376,9 @@ void ieee80211_tx_status(struct ieee80211_hw *hw,
  *
  * @hw: the hardware the frame was transmitted by
  * @skb: the frame that was transmitted, owned by mac80211 after this call
- * @status: status information for this frame; the status pointer need not
- *	be valid after this function returns and is not freed by mac80211,
- *	it is recommended that it points to a stack area
  */
 void ieee80211_tx_status_irqsafe(struct ieee80211_hw *hw,
-				 struct sk_buff *skb,
-				 struct ieee80211_tx_status *status);
+				 struct sk_buff *skb);
 
 /**
  * ieee80211_beacon_get - beacon generation function
@@ -1425,8 +1394,7 @@ void ieee80211_tx_status_irqsafe(struct ieee80211_hw *hw,
  * is responsible of freeing it.
  */
 struct sk_buff *ieee80211_beacon_get(struct ieee80211_hw *hw,
-				     struct ieee80211_vif *vif,
-				     struct ieee80211_tx_control *control);
+				     struct ieee80211_vif *vif);
 
 /**
  * ieee80211_rts_get - RTS frame generation function
@@ -1434,7 +1402,7 @@ struct sk_buff *ieee80211_beacon_get(struct ieee80211_hw *hw,
  * @vif: &struct ieee80211_vif pointer from &struct ieee80211_if_init_conf.
  * @frame: pointer to the frame that is going to be protected by the RTS.
  * @frame_len: the frame length (in octets).
- * @frame_txctl: &struct ieee80211_tx_control of the frame.
+ * @frame_txctl: &struct ieee80211_tx_info of the frame.
  * @rts: The buffer where to store the RTS frame.
  *
  * If the RTS frames are generated by the host system (i.e., not in
@@ -1444,7 +1412,7 @@ struct sk_buff *ieee80211_beacon_get(struct ieee80211_hw *hw,
  */
 void ieee80211_rts_get(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 		       const void *frame, size_t frame_len,
-		       const struct ieee80211_tx_control *frame_txctl,
+		       const struct ieee80211_tx_info *frame_txctl,
 		       struct ieee80211_rts *rts);
 
 /**
@@ -1452,7 +1420,7 @@ void ieee80211_rts_get(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
  * @hw: pointer obtained from ieee80211_alloc_hw().
  * @vif: &struct ieee80211_vif pointer from &struct ieee80211_if_init_conf.
  * @frame_len: the length of the frame that is going to be protected by the RTS.
- * @frame_txctl: &struct ieee80211_tx_control of the frame.
+ * @frame_txctl: &struct ieee80211_tx_info of the frame.
  *
  * If the RTS is generated in firmware, but the host system must provide
  * the duration field, the low-level driver uses this function to receive
@@ -1460,7 +1428,7 @@ void ieee80211_rts_get(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
  */
 __le16 ieee80211_rts_duration(struct ieee80211_hw *hw,
 			      struct ieee80211_vif *vif, size_t frame_len,
-			      const struct ieee80211_tx_control *frame_txctl);
+			      const struct ieee80211_tx_info *frame_txctl);
 
 /**
  * ieee80211_ctstoself_get - CTS-to-self frame generation function
@@ -1468,7 +1436,7 @@ __le16 ieee80211_rts_duration(struct ieee80211_hw *hw,
  * @vif: &struct ieee80211_vif pointer from &struct ieee80211_if_init_conf.
  * @frame: pointer to the frame that is going to be protected by the CTS-to-self.
  * @frame_len: the frame length (in octets).
- * @frame_txctl: &struct ieee80211_tx_control of the frame.
+ * @frame_txctl: &struct ieee80211_tx_info of the frame.
  * @cts: The buffer where to store the CTS-to-self frame.
  *
  * If the CTS-to-self frames are generated by the host system (i.e., not in
@@ -1479,7 +1447,7 @@ __le16 ieee80211_rts_duration(struct ieee80211_hw *hw,
 void ieee80211_ctstoself_get(struct ieee80211_hw *hw,
 			     struct ieee80211_vif *vif,
 			     const void *frame, size_t frame_len,
-			     const struct ieee80211_tx_control *frame_txctl,
+			     const struct ieee80211_tx_info *frame_txctl,
 			     struct ieee80211_cts *cts);
 
 /**
@@ -1487,7 +1455,7 @@ void ieee80211_ctstoself_get(struct ieee80211_hw *hw,
  * @hw: pointer obtained from ieee80211_alloc_hw().
  * @vif: &struct ieee80211_vif pointer from &struct ieee80211_if_init_conf.
  * @frame_len: the length of the frame that is going to be protected by the CTS-to-self.
- * @frame_txctl: &struct ieee80211_tx_control of the frame.
+ * @frame_txctl: &struct ieee80211_tx_info of the frame.
  *
  * If the CTS-to-self is generated in firmware, but the host system must provide
  * the duration field, the low-level driver uses this function to receive
@@ -1496,7 +1464,7 @@ void ieee80211_ctstoself_get(struct ieee80211_hw *hw,
 __le16 ieee80211_ctstoself_duration(struct ieee80211_hw *hw,
 				    struct ieee80211_vif *vif,
 				    size_t frame_len,
-				    const struct ieee80211_tx_control *frame_txctl);
+				    const struct ieee80211_tx_info *frame_txctl);
 
 /**
  * ieee80211_generic_frame_duration - Calculate the duration field for a frame
@@ -1535,8 +1503,7 @@ __le16 ieee80211_generic_frame_duration(struct ieee80211_hw *hw,
  * use common code for all beacons.
  */
 struct sk_buff *
-ieee80211_get_buffered_bc(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
-			  struct ieee80211_tx_control *control);
+ieee80211_get_buffered_bc(struct ieee80211_hw *hw, struct ieee80211_vif *vif);
 
 /**
  * ieee80211_get_hdrlen_from_skb - get header length from data
diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h
index a4cccd1b7d53..79a65b3ee02b 100644
--- a/net/mac80211/ieee80211_i.h
+++ b/net/mac80211/ieee80211_i.h
@@ -2,6 +2,7 @@
  * Copyright 2002-2005, Instant802 Networks, Inc.
  * Copyright 2005, Devicescape Software, Inc.
  * Copyright 2006-2007	Jiri Benc <jbenc@suse.cz>
+ * Copyright 2007-2008	Johannes Berg <johannes@sipsolutions.net>
  *
  * 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
@@ -147,7 +148,6 @@ typedef unsigned __bitwise__ ieee80211_tx_result;
 #define IEEE80211_TX_UNICAST		BIT(1)
 #define IEEE80211_TX_PS_BUFFERED	BIT(2)
 #define IEEE80211_TX_PROBE_LAST_FRAG	BIT(3)
-#define IEEE80211_TX_INJECTED		BIT(4)
 
 struct ieee80211_tx_data {
 	struct sk_buff *skb;
@@ -157,7 +157,6 @@ struct ieee80211_tx_data {
 	struct sta_info *sta;
 	struct ieee80211_key *key;
 
-	struct ieee80211_tx_control *control;
 	struct ieee80211_channel *channel;
 	s8 rate_idx;
 	/* use this rate (if set) for last fragment; rate can
@@ -207,22 +206,7 @@ struct ieee80211_rx_data {
 	u16 tkip_iv16;
 };
 
-/* flags used in struct ieee80211_tx_packet_data.flags */
-#define IEEE80211_TXPD_REQ_TX_STATUS	BIT(0)
-#define IEEE80211_TXPD_DO_NOT_ENCRYPT	BIT(1)
-#define IEEE80211_TXPD_REQUEUE		BIT(2)
-#define IEEE80211_TXPD_EAPOL_FRAME	BIT(3)
-#define IEEE80211_TXPD_AMPDU		BIT(4)
-/* Stored in sk_buff->cb */
-struct ieee80211_tx_packet_data {
-	int ifindex;
-	unsigned long jiffies;
-	unsigned int flags;
-	u8 queue;
-};
-
 struct ieee80211_tx_stored_packet {
-	struct ieee80211_tx_control control;
 	struct sk_buff *skb;
 	struct sk_buff **extra_frag;
 	s8 last_frag_rate_idx;
diff --git a/net/mac80211/main.c b/net/mac80211/main.c
index 9761d9bd5a79..8f1ff7ef1667 100644
--- a/net/mac80211/main.c
+++ b/net/mac80211/main.c
@@ -971,8 +971,7 @@ void ieee80211_if_setup(struct net_device *dev)
 /* everything else */
 
 static int __ieee80211_if_config(struct net_device *dev,
-				 struct sk_buff *beacon,
-				 struct ieee80211_tx_control *control)
+				 struct sk_buff *beacon)
 {
 	struct ieee80211_sub_if_data *sdata = IEEE80211_DEV_TO_SUB_IF(dev);
 	struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr);
@@ -990,13 +989,11 @@ static int __ieee80211_if_config(struct net_device *dev,
 		conf.ssid_len = sdata->u.sta.ssid_len;
 	} else if (ieee80211_vif_is_mesh(&sdata->vif)) {
 		conf.beacon = beacon;
-		conf.beacon_control = control;
 		ieee80211_start_mesh(dev);
 	} else if (sdata->vif.type == IEEE80211_IF_TYPE_AP) {
 		conf.ssid = sdata->u.ap.ssid;
 		conf.ssid_len = sdata->u.ap.ssid_len;
 		conf.beacon = beacon;
-		conf.beacon_control = control;
 	}
 	return local->ops->config_interface(local_to_hw(local),
 					    &sdata->vif, &conf);
@@ -1009,23 +1006,21 @@ int ieee80211_if_config(struct net_device *dev)
 	if (sdata->vif.type == IEEE80211_IF_TYPE_MESH_POINT &&
 	    (local->hw.flags & IEEE80211_HW_HOST_GEN_BEACON_TEMPLATE))
 		return ieee80211_if_config_beacon(dev);
-	return __ieee80211_if_config(dev, NULL, NULL);
+	return __ieee80211_if_config(dev, NULL);
 }
 
 int ieee80211_if_config_beacon(struct net_device *dev)
 {
 	struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr);
-	struct ieee80211_tx_control control;
 	struct ieee80211_sub_if_data *sdata = IEEE80211_DEV_TO_SUB_IF(dev);
 	struct sk_buff *skb;
 
 	if (!(local->hw.flags & IEEE80211_HW_HOST_GEN_BEACON_TEMPLATE))
 		return 0;
-	skb = ieee80211_beacon_get(local_to_hw(local), &sdata->vif,
-				   &control);
+	skb = ieee80211_beacon_get(local_to_hw(local), &sdata->vif);
 	if (!skb)
 		return -ENOMEM;
-	return __ieee80211_if_config(dev, skb, &control);
+	return __ieee80211_if_config(dev, skb);
 }
 
 int ieee80211_hw_config(struct ieee80211_local *local)
@@ -1180,38 +1175,20 @@ void ieee80211_reset_erp_info(struct net_device *dev)
 }
 
 void ieee80211_tx_status_irqsafe(struct ieee80211_hw *hw,
-				 struct sk_buff *skb,
-				 struct ieee80211_tx_status *status)
+				 struct sk_buff *skb)
 {
 	struct ieee80211_local *local = hw_to_local(hw);
-	struct ieee80211_tx_status *saved;
+	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
 	int tmp;
 
 	skb->dev = local->mdev;
-	saved = kmalloc(sizeof(struct ieee80211_tx_status), GFP_ATOMIC);
-	if (unlikely(!saved)) {
-		if (net_ratelimit())
-			printk(KERN_WARNING "%s: Not enough memory, "
-			       "dropping tx status", skb->dev->name);
-		/* should be dev_kfree_skb_irq, but due to this function being
-		 * named _irqsafe instead of just _irq we can't be sure that
-		 * people won't call it from non-irq contexts */
-		dev_kfree_skb_any(skb);
-		return;
-	}
-	memcpy(saved, status, sizeof(struct ieee80211_tx_status));
-	/* copy pointer to saved status into skb->cb for use by tasklet */
-	memcpy(skb->cb, &saved, sizeof(saved));
-
 	skb->pkt_type = IEEE80211_TX_STATUS_MSG;
-	skb_queue_tail(status->control.flags & IEEE80211_TXCTL_REQ_TX_STATUS ?
+	skb_queue_tail(info->flags & IEEE80211_TX_CTL_REQ_TX_STATUS ?
 		       &local->skb_queue : &local->skb_queue_unreliable, skb);
 	tmp = skb_queue_len(&local->skb_queue) +
 		skb_queue_len(&local->skb_queue_unreliable);
 	while (tmp > IEEE80211_IRQSAFE_QUEUE_LIMIT &&
 	       (skb = skb_dequeue(&local->skb_queue_unreliable))) {
-		memcpy(&saved, skb->cb, sizeof(saved));
-		kfree(saved);
 		dev_kfree_skb_irq(skb);
 		tmp--;
 		I802_DEBUG_INC(local->tx_status_drop);
@@ -1225,7 +1202,6 @@ static void ieee80211_tasklet_handler(unsigned long data)
 	struct ieee80211_local *local = (struct ieee80211_local *) data;
 	struct sk_buff *skb;
 	struct ieee80211_rx_status rx_status;
-	struct ieee80211_tx_status *tx_status;
 	struct ieee80211_ra_tid *ra_tid;
 
 	while ((skb = skb_dequeue(&local->skb_queue)) ||
@@ -1240,12 +1216,8 @@ static void ieee80211_tasklet_handler(unsigned long data)
 			__ieee80211_rx(local_to_hw(local), skb, &rx_status);
 			break;
 		case IEEE80211_TX_STATUS_MSG:
-			/* get pointer to saved status out of skb->cb */
-			memcpy(&tx_status, skb->cb, sizeof(tx_status));
 			skb->pkt_type = 0;
-			ieee80211_tx_status(local_to_hw(local),
-					    skb, tx_status);
-			kfree(tx_status);
+			ieee80211_tx_status(local_to_hw(local), skb);
 			break;
 		case IEEE80211_DELBA_MSG:
 			ra_tid = (struct ieee80211_ra_tid *) &skb->cb;
@@ -1274,24 +1246,15 @@ static void ieee80211_tasklet_handler(unsigned long data)
  * Also, tx_packet_data in cb is restored from tx_control. */
 static void ieee80211_remove_tx_extra(struct ieee80211_local *local,
 				      struct ieee80211_key *key,
-				      struct sk_buff *skb,
-				      struct ieee80211_tx_control *control)
+				      struct sk_buff *skb)
 {
 	int hdrlen, iv_len, mic_len;
-	struct ieee80211_tx_packet_data *pkt_data;
-
-	pkt_data = (struct ieee80211_tx_packet_data *)skb->cb;
-	pkt_data->ifindex = vif_to_sdata(control->vif)->dev->ifindex;
-	pkt_data->flags = 0;
-	if (control->flags & IEEE80211_TXCTL_REQ_TX_STATUS)
-		pkt_data->flags |= IEEE80211_TXPD_REQ_TX_STATUS;
-	if (control->flags & IEEE80211_TXCTL_DO_NOT_ENCRYPT)
-		pkt_data->flags |= IEEE80211_TXPD_DO_NOT_ENCRYPT;
-	if (control->flags & IEEE80211_TXCTL_REQUEUE)
-		pkt_data->flags |= IEEE80211_TXPD_REQUEUE;
-	if (control->flags & IEEE80211_TXCTL_EAPOL_FRAME)
-		pkt_data->flags |= IEEE80211_TXPD_EAPOL_FRAME;
-	pkt_data->queue = control->queue;
+	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
+
+	info->flags &=	IEEE80211_TX_CTL_REQ_TX_STATUS |
+			IEEE80211_TX_CTL_DO_NOT_ENCRYPT |
+			IEEE80211_TX_CTL_REQUEUE |
+			IEEE80211_TX_CTL_EAPOL_FRAME;
 
 	hdrlen = ieee80211_get_hdrlen_from_skb(skb);
 
@@ -1338,9 +1301,10 @@ no_key:
 
 static void ieee80211_handle_filtered_frame(struct ieee80211_local *local,
 					    struct sta_info *sta,
-					    struct sk_buff *skb,
-					    struct ieee80211_tx_status *status)
+					    struct sk_buff *skb)
 {
+	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
+
 	sta->tx_filtered_count++;
 
 	/*
@@ -1382,18 +1346,16 @@ static void ieee80211_handle_filtered_frame(struct ieee80211_local *local,
 	 */
 	if (test_sta_flags(sta, WLAN_STA_PS) &&
 	    skb_queue_len(&sta->tx_filtered) < STA_MAX_TX_BUFFER) {
-		ieee80211_remove_tx_extra(local, sta->key, skb,
-					  &status->control);
+		ieee80211_remove_tx_extra(local, sta->key, skb);
 		skb_queue_tail(&sta->tx_filtered, skb);
 		return;
 	}
 
 	if (!test_sta_flags(sta, WLAN_STA_PS) &&
-	    !(status->control.flags & IEEE80211_TXCTL_REQUEUE)) {
+	    !(info->flags & IEEE80211_TX_CTL_REQUEUE)) {
 		/* Software retry the packet once */
-		status->control.flags |= IEEE80211_TXCTL_REQUEUE;
-		ieee80211_remove_tx_extra(local, sta->key, skb,
-					  &status->control);
+		info->flags |= IEEE80211_TX_CTL_REQUEUE;
+		ieee80211_remove_tx_extra(local, sta->key, skb);
 		dev_queue_xmit(skb);
 		return;
 	}
@@ -1407,28 +1369,20 @@ static void ieee80211_handle_filtered_frame(struct ieee80211_local *local,
 	dev_kfree_skb(skb);
 }
 
-void ieee80211_tx_status(struct ieee80211_hw *hw, struct sk_buff *skb,
-			 struct ieee80211_tx_status *status)
+void ieee80211_tx_status(struct ieee80211_hw *hw, struct sk_buff *skb)
 {
 	struct sk_buff *skb2;
 	struct ieee80211_hdr *hdr = (struct ieee80211_hdr *) skb->data;
 	struct ieee80211_local *local = hw_to_local(hw);
+	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
 	u16 frag, type;
 	struct ieee80211_tx_status_rtap_hdr *rthdr;
 	struct ieee80211_sub_if_data *sdata;
 	struct net_device *prev_dev = NULL;
 
-	if (!status) {
-		printk(KERN_ERR
-		       "%s: ieee80211_tx_status called with NULL status\n",
-		       wiphy_name(local->hw.wiphy));
-		dev_kfree_skb(skb);
-		return;
-	}
-
 	rcu_read_lock();
 
-	if (status->excessive_retries) {
+	if (info->status.excessive_retries) {
 		struct sta_info *sta;
 		sta = sta_info_get(local, hdr->addr1);
 		if (sta) {
@@ -1437,27 +1391,23 @@ void ieee80211_tx_status(struct ieee80211_hw *hw, struct sk_buff *skb,
 				 * The STA is in power save mode, so assume
 				 * that this TX packet failed because of that.
 				 */
-				status->excessive_retries = 0;
-				status->flags |= IEEE80211_TX_STATUS_TX_FILTERED;
-				ieee80211_handle_filtered_frame(local, sta,
-								skb, status);
+				ieee80211_handle_filtered_frame(local, sta, skb);
 				rcu_read_unlock();
 				return;
 			}
 		}
 	}
 
-	if (status->flags & IEEE80211_TX_STATUS_TX_FILTERED) {
+	if (info->flags & IEEE80211_TX_STAT_TX_FILTERED) {
 		struct sta_info *sta;
 		sta = sta_info_get(local, hdr->addr1);
 		if (sta) {
-			ieee80211_handle_filtered_frame(local, sta, skb,
-							status);
+			ieee80211_handle_filtered_frame(local, sta, skb);
 			rcu_read_unlock();
 			return;
 		}
 	} else
-		rate_control_tx_status(local->mdev, skb, status);
+		rate_control_tx_status(local->mdev, skb);
 
 	rcu_read_unlock();
 
@@ -1471,14 +1421,14 @@ void ieee80211_tx_status(struct ieee80211_hw *hw, struct sk_buff *skb,
 	frag = le16_to_cpu(hdr->seq_ctrl) & IEEE80211_SCTL_FRAG;
 	type = le16_to_cpu(hdr->frame_control) & IEEE80211_FCTL_FTYPE;
 
-	if (status->flags & IEEE80211_TX_STATUS_ACK) {
+	if (info->flags & IEEE80211_TX_STAT_ACK) {
 		if (frag == 0) {
 			local->dot11TransmittedFrameCount++;
 			if (is_multicast_ether_addr(hdr->addr1))
 				local->dot11MulticastTransmittedFrameCount++;
-			if (status->retry_count > 0)
+			if (info->status.retry_count > 0)
 				local->dot11RetryCount++;
-			if (status->retry_count > 1)
+			if (info->status.retry_count > 1)
 				local->dot11MultipleRetryCount++;
 		}
 
@@ -1524,17 +1474,17 @@ void ieee80211_tx_status(struct ieee80211_hw *hw, struct sk_buff *skb,
 		cpu_to_le32((1 << IEEE80211_RADIOTAP_TX_FLAGS) |
 			    (1 << IEEE80211_RADIOTAP_DATA_RETRIES));
 
-	if (!(status->flags & IEEE80211_TX_STATUS_ACK) &&
+	if (!(info->flags & IEEE80211_TX_STAT_ACK) &&
 	    !is_multicast_ether_addr(hdr->addr1))
 		rthdr->tx_flags |= cpu_to_le16(IEEE80211_RADIOTAP_F_TX_FAIL);
 
-	if ((status->control.flags & IEEE80211_TXCTL_USE_RTS_CTS) &&
-	    (status->control.flags & IEEE80211_TXCTL_USE_CTS_PROTECT))
+	if ((info->flags & IEEE80211_TX_CTL_USE_RTS_CTS) &&
+	    (info->flags & IEEE80211_TX_CTL_USE_CTS_PROTECT))
 		rthdr->tx_flags |= cpu_to_le16(IEEE80211_RADIOTAP_F_TX_CTS);
-	else if (status->control.flags & IEEE80211_TXCTL_USE_RTS_CTS)
+	else if (info->flags & IEEE80211_TX_CTL_USE_RTS_CTS)
 		rthdr->tx_flags |= cpu_to_le16(IEEE80211_RADIOTAP_F_TX_RTS);
 
-	rthdr->data_retries = status->retry_count;
+	rthdr->data_retries = info->status.retry_count;
 
 	/* XXX: is this sufficient for BPF? */
 	skb_set_mac_header(skb, 0);
@@ -1895,7 +1845,9 @@ static int __init ieee80211_init(void)
 	struct sk_buff *skb;
 	int ret;
 
-	BUILD_BUG_ON(sizeof(struct ieee80211_tx_packet_data) > sizeof(skb->cb));
+	BUILD_BUG_ON(sizeof(struct ieee80211_tx_info) > sizeof(skb->cb));
+	BUILD_BUG_ON(offsetof(struct ieee80211_tx_info, driver_data) +
+	             IEEE80211_TX_INFO_DRIVER_DATA_SIZE > sizeof(skb->cb));
 
 	ret = rc80211_pid_init();
 	if (ret)
diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c
index 604149369dc9..9a264379d7b1 100644
--- a/net/mac80211/mlme.c
+++ b/net/mac80211/mlme.c
@@ -578,7 +578,7 @@ void ieee80211_sta_tx(struct net_device *dev, struct sk_buff *skb,
 		      int encrypt)
 {
 	struct ieee80211_sub_if_data *sdata;
-	struct ieee80211_tx_packet_data *pkt_data;
+	struct ieee80211_tx_info *info;
 
 	sdata = IEEE80211_DEV_TO_SUB_IF(dev);
 	skb->dev = sdata->local->mdev;
@@ -586,11 +586,11 @@ void ieee80211_sta_tx(struct net_device *dev, struct sk_buff *skb,
 	skb_set_network_header(skb, 0);
 	skb_set_transport_header(skb, 0);
 
-	pkt_data = (struct ieee80211_tx_packet_data *) skb->cb;
-	memset(pkt_data, 0, sizeof(struct ieee80211_tx_packet_data));
-	pkt_data->ifindex = sdata->dev->ifindex;
+	info = IEEE80211_SKB_CB(skb);
+	memset(info, 0, sizeof(struct ieee80211_tx_info));
+	info->control.ifindex = sdata->dev->ifindex;
 	if (!encrypt)
-		pkt_data->flags |= IEEE80211_TXPD_DO_NOT_ENCRYPT;
+		info->flags |= IEEE80211_TX_CTL_DO_NOT_ENCRYPT;
 
 	dev_queue_xmit(skb);
 }
@@ -2314,7 +2314,7 @@ static int ieee80211_sta_join_ibss(struct net_device *dev,
 	int res, rates, i, j;
 	struct sk_buff *skb;
 	struct ieee80211_mgmt *mgmt;
-	struct ieee80211_tx_control control;
+	struct ieee80211_tx_info *control;
 	struct rate_selection ratesel;
 	u8 *pos;
 	struct ieee80211_sub_if_data *sdata;
@@ -2404,21 +2404,22 @@ static int ieee80211_sta_join_ibss(struct net_device *dev,
 			memcpy(pos, &bss->supp_rates[8], rates);
 		}
 
-		memset(&control, 0, sizeof(control));
+		control = IEEE80211_SKB_CB(skb);
+
 		rate_control_get_rate(dev, sband, skb, &ratesel);
 		if (ratesel.rate_idx < 0) {
 			printk(KERN_DEBUG "%s: Failed to determine TX rate "
 			       "for IBSS beacon\n", dev->name);
 			break;
 		}
-		control.vif = &sdata->vif;
-		control.tx_rate_idx = ratesel.rate_idx;
+		control->control.vif = &sdata->vif;
+		control->tx_rate_idx = ratesel.rate_idx;
 		if (sdata->bss_conf.use_short_preamble &&
 		    sband->bitrates[ratesel.rate_idx].flags & IEEE80211_RATE_SHORT_PREAMBLE)
-			control.flags |= IEEE80211_TXCTL_SHORT_PREAMBLE;
-		control.antenna_sel_tx = local->hw.conf.antenna_sel_tx;
-		control.flags |= IEEE80211_TXCTL_NO_ACK;
-		control.retry_limit = 1;
+			control->flags |= IEEE80211_TX_CTL_SHORT_PREAMBLE;
+		control->antenna_sel_tx = local->hw.conf.antenna_sel_tx;
+		control->flags |= IEEE80211_TX_CTL_NO_ACK;
+		control->control.retry_limit = 1;
 
 		ifsta->probe_resp = skb_copy(skb, GFP_ATOMIC);
 		if (ifsta->probe_resp) {
@@ -2433,8 +2434,7 @@ static int ieee80211_sta_join_ibss(struct net_device *dev,
 		}
 
 		if (local->ops->beacon_update &&
-		    local->ops->beacon_update(local_to_hw(local),
-					     skb, &control) == 0) {
+		    local->ops->beacon_update(local_to_hw(local), skb) == 0) {
 			printk(KERN_DEBUG "%s: Configured IBSS beacon "
 			       "template\n", dev->name);
 			skb = NULL;
diff --git a/net/mac80211/rate.h b/net/mac80211/rate.h
index a29148dcca99..0ed9c8a2f56f 100644
--- a/net/mac80211/rate.h
+++ b/net/mac80211/rate.h
@@ -34,8 +34,7 @@ struct rate_control_ops {
 	struct module *module;
 	const char *name;
 	void (*tx_status)(void *priv, struct net_device *dev,
-			  struct sk_buff *skb,
-			  struct ieee80211_tx_status *status);
+			  struct sk_buff *skb);
 	void (*get_rate)(void *priv, struct net_device *dev,
 			 struct ieee80211_supported_band *band,
 			 struct sk_buff *skb,
@@ -77,13 +76,12 @@ struct rate_control_ref *rate_control_get(struct rate_control_ref *ref);
 void rate_control_put(struct rate_control_ref *ref);
 
 static inline void rate_control_tx_status(struct net_device *dev,
-					  struct sk_buff *skb,
-					  struct ieee80211_tx_status *status)
+					  struct sk_buff *skb)
 {
 	struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr);
 	struct rate_control_ref *ref = local->rate_ctrl;
 
-	ref->ops->tx_status(ref->priv, dev, skb, status);
+	ref->ops->tx_status(ref->priv, dev, skb);
 }
 
 
diff --git a/net/mac80211/rc80211_pid.h b/net/mac80211/rc80211_pid.h
index 04afc13ed825..2078803d3581 100644
--- a/net/mac80211/rc80211_pid.h
+++ b/net/mac80211/rc80211_pid.h
@@ -61,7 +61,7 @@ enum rc_pid_event_type {
 union rc_pid_event_data {
 	/* RC_PID_EVENT_TX_STATUS */
 	struct {
-		struct ieee80211_tx_status tx_status;
+		struct ieee80211_tx_info tx_status;
 	};
 	/* RC_PID_EVENT_TYPE_RATE_CHANGE */
 	/* RC_PID_EVENT_TYPE_TX_RATE */
@@ -158,7 +158,7 @@ struct rc_pid_debugfs_entries {
 };
 
 void rate_control_pid_event_tx_status(struct rc_pid_event_buffer *buf,
-					     struct ieee80211_tx_status *stat);
+				      struct ieee80211_tx_info *stat);
 
 void rate_control_pid_event_rate_change(struct rc_pid_event_buffer *buf,
 					       int index, int rate);
diff --git a/net/mac80211/rc80211_pid_algo.c b/net/mac80211/rc80211_pid_algo.c
index 14cde36f5070..e8945413e4a2 100644
--- a/net/mac80211/rc80211_pid_algo.c
+++ b/net/mac80211/rc80211_pid_algo.c
@@ -237,8 +237,7 @@ static void rate_control_pid_sample(struct rc_pid_info *pinfo,
 }
 
 static void rate_control_pid_tx_status(void *priv, struct net_device *dev,
-				       struct sk_buff *skb,
-				       struct ieee80211_tx_status *status)
+				       struct sk_buff *skb)
 {
 	struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr);
 	struct ieee80211_hdr *hdr = (struct ieee80211_hdr *) skb->data;
@@ -248,6 +247,7 @@ static void rate_control_pid_tx_status(void *priv, struct net_device *dev,
 	struct rc_pid_sta_info *spinfo;
 	unsigned long period;
 	struct ieee80211_supported_band *sband;
+	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
 
 	rcu_read_lock();
 
@@ -266,28 +266,28 @@ static void rate_control_pid_tx_status(void *priv, struct net_device *dev,
 
 	/* Ignore all frames that were sent with a different rate than the rate
 	 * we currently advise mac80211 to use. */
-	if (status->control.tx_rate_idx != sta->txrate_idx)
+	if (info->tx_rate_idx != sta->txrate_idx)
 		goto unlock;
 
 	spinfo = sta->rate_ctrl_priv;
 	spinfo->tx_num_xmit++;
 
 #ifdef CONFIG_MAC80211_DEBUGFS
-	rate_control_pid_event_tx_status(&spinfo->events, status);
+	rate_control_pid_event_tx_status(&spinfo->events, info);
 #endif
 
 	/* We count frames that totally failed to be transmitted as two bad
 	 * frames, those that made it out but had some retries as one good and
 	 * one bad frame. */
-	if (status->excessive_retries) {
+	if (info->status.excessive_retries) {
 		spinfo->tx_num_failed += 2;
 		spinfo->tx_num_xmit++;
-	} else if (status->retry_count) {
+	} else if (info->status.retry_count) {
 		spinfo->tx_num_failed++;
 		spinfo->tx_num_xmit++;
 	}
 
-	if (status->excessive_retries) {
+	if (info->status.excessive_retries) {
 		sta->tx_retry_failed++;
 		sta->tx_num_consecutive_failures++;
 		sta->tx_num_mpdu_fail++;
@@ -295,8 +295,8 @@ static void rate_control_pid_tx_status(void *priv, struct net_device *dev,
 		sta->tx_num_consecutive_failures = 0;
 		sta->tx_num_mpdu_ok++;
 	}
-	sta->tx_retry_count += status->retry_count;
-	sta->tx_num_mpdu_fail += status->retry_count;
+	sta->tx_retry_count += info->status.retry_count;
+	sta->tx_num_mpdu_fail += info->status.retry_count;
 
 	/* Update PID controller state. */
 	period = (HZ * pinfo->sampling_period + 500) / 1000;
diff --git a/net/mac80211/rc80211_pid_debugfs.c b/net/mac80211/rc80211_pid_debugfs.c
index ff5c380f3c13..8121d3bc6835 100644
--- a/net/mac80211/rc80211_pid_debugfs.c
+++ b/net/mac80211/rc80211_pid_debugfs.c
@@ -39,11 +39,11 @@ static void rate_control_pid_event(struct rc_pid_event_buffer *buf,
 }
 
 void rate_control_pid_event_tx_status(struct rc_pid_event_buffer *buf,
-					     struct ieee80211_tx_status *stat)
+				      struct ieee80211_tx_info *stat)
 {
 	union rc_pid_event_data evd;
 
-	memcpy(&evd.tx_status, stat, sizeof(struct ieee80211_tx_status));
+	memcpy(&evd.tx_status, stat, sizeof(struct ieee80211_tx_info));
 	rate_control_pid_event(buf, RC_PID_EVENT_TYPE_TX_STATUS, &evd);
 }
 
@@ -167,8 +167,8 @@ static ssize_t rate_control_pid_events_read(struct file *file, char __user *buf,
 	switch (ev->type) {
 	case RC_PID_EVENT_TYPE_TX_STATUS:
 		p += snprintf(pb + p, length - p, "tx_status %u %u",
-			      ev->data.tx_status.excessive_retries,
-			      ev->data.tx_status.retry_count);
+			      ev->data.tx_status.status.excessive_retries,
+			      ev->data.tx_status.status.retry_count);
 		break;
 	case RC_PID_EVENT_TYPE_RATE_CHANGE:
 		p += snprintf(pb + p, length - p, "rate_change %d %d",
diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c
index fa68305fd59e..cf0de3b0fe24 100644
--- a/net/mac80211/rx.c
+++ b/net/mac80211/rx.c
@@ -714,7 +714,7 @@ static int ap_sta_ps_end(struct net_device *dev, struct sta_info *sta)
 	struct sk_buff *skb;
 	int sent = 0;
 	struct ieee80211_sub_if_data *sdata;
-	struct ieee80211_tx_packet_data *pkt_data;
+	struct ieee80211_tx_info *info;
 	DECLARE_MAC_BUF(mac);
 
 	sdata = sta->sdata;
@@ -734,13 +734,13 @@ static int ap_sta_ps_end(struct net_device *dev, struct sta_info *sta)
 
 	/* Send all buffered frames to the station */
 	while ((skb = skb_dequeue(&sta->tx_filtered)) != NULL) {
-		pkt_data = (struct ieee80211_tx_packet_data *) skb->cb;
+		info = IEEE80211_SKB_CB(skb);
 		sent++;
-		pkt_data->flags |= IEEE80211_TXPD_REQUEUE;
+		info->flags |= IEEE80211_TX_CTL_REQUEUE;
 		dev_queue_xmit(skb);
 	}
 	while ((skb = skb_dequeue(&sta->ps_tx_buf)) != NULL) {
-		pkt_data = (struct ieee80211_tx_packet_data *) skb->cb;
+		info = IEEE80211_SKB_CB(skb);
 		local->total_ps_buffered--;
 		sent++;
 #ifdef CONFIG_MAC80211_VERBOSE_PS_DEBUG
@@ -748,7 +748,7 @@ static int ap_sta_ps_end(struct net_device *dev, struct sta_info *sta)
 		       "since STA not sleeping anymore\n", dev->name,
 		       print_mac(mac, sta->addr), sta->aid);
 #endif /* CONFIG_MAC80211_VERBOSE_PS_DEBUG */
-		pkt_data->flags |= IEEE80211_TXPD_REQUEUE;
+		info->flags |= IEEE80211_TX_CTL_REQUEUE;
 		dev_queue_xmit(skb);
 	}
 
diff --git a/net/mac80211/sta_info.c b/net/mac80211/sta_info.c
index baf5e4746884..ef3149324d54 100644
--- a/net/mac80211/sta_info.c
+++ b/net/mac80211/sta_info.c
@@ -511,20 +511,20 @@ static inline int sta_info_buffer_expired(struct ieee80211_local *local,
 					  struct sta_info *sta,
 					  struct sk_buff *skb)
 {
-	struct ieee80211_tx_packet_data *pkt_data;
+	struct ieee80211_tx_info *info;
 	int timeout;
 
 	if (!skb)
 		return 0;
 
-	pkt_data = (struct ieee80211_tx_packet_data *) skb->cb;
+	info = IEEE80211_SKB_CB(skb);
 
 	/* Timeout: (2 * listen_interval * beacon_int * 1024 / 1000000) sec */
 	timeout = (sta->listen_interval * local->hw.conf.beacon_int * 32 /
 		   15625) * HZ;
 	if (timeout < STA_TX_BUFFER_EXPIRE)
 		timeout = STA_TX_BUFFER_EXPIRE;
-	return time_after(jiffies, pkt_data->jiffies + timeout);
+	return time_after(jiffies, info->control.jiffies + timeout);
 }
 
 
diff --git a/net/mac80211/sta_info.h b/net/mac80211/sta_info.h
index e89cc1655547..f592290e42b9 100644
--- a/net/mac80211/sta_info.h
+++ b/net/mac80211/sta_info.h
@@ -32,7 +32,7 @@
  * @WLAN_STA_WDS: Station is one of our WDS peers.
  * @WLAN_STA_PSPOLL: Station has just PS-polled us.
  * @WLAN_STA_CLEAR_PS_FILT: Clear PS filter in hardware (using the
- *	IEEE80211_TXCTL_CLEAR_PS_FILT control flag) when the next
+ *	IEEE80211_TX_CTL_CLEAR_PS_FILT control flag) when the next
  *	frame to this station is transmitted.
  */
 enum ieee80211_sta_info_flags {
diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c
index 666158f02a89..ac9a4af7ad42 100644
--- a/net/mac80211/tx.c
+++ b/net/mac80211/tx.c
@@ -238,12 +238,12 @@ static ieee80211_tx_result
 ieee80211_tx_h_check_assoc(struct ieee80211_tx_data *tx)
 {
 #ifdef CONFIG_MAC80211_VERBOSE_DEBUG
-	struct sk_buff *skb = tx->skb;
-	struct ieee80211_hdr *hdr = (struct ieee80211_hdr *) skb->data;
+	struct ieee80211_hdr *hdr = (struct ieee80211_hdr *)tx->skb->data;
 #endif /* CONFIG_MAC80211_VERBOSE_DEBUG */
+	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(tx->skb);
 	u32 sta_flags;
 
-	if (unlikely(tx->flags & IEEE80211_TX_INJECTED))
+	if (unlikely(info->flags & IEEE80211_TX_CTL_INJECTED))
 		return TX_CONTINUE;
 
 	if (unlikely(tx->local->sta_sw_scanning) &&
@@ -348,6 +348,8 @@ static void purge_old_ps_buffers(struct ieee80211_local *local)
 static ieee80211_tx_result
 ieee80211_tx_h_multicast_ps_buf(struct ieee80211_tx_data *tx)
 {
+	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(tx->skb);
+
 	/*
 	 * broadcast/multicast frame
 	 *
@@ -383,7 +385,7 @@ ieee80211_tx_h_multicast_ps_buf(struct ieee80211_tx_data *tx)
 	}
 
 	/* buffered in hardware */
-	tx->control->flags |= IEEE80211_TXCTL_SEND_AFTER_DTIM;
+	info->flags |= IEEE80211_TX_CTL_SEND_AFTER_DTIM;
 
 	return TX_CONTINUE;
 }
@@ -392,6 +394,7 @@ static ieee80211_tx_result
 ieee80211_tx_h_unicast_ps_buf(struct ieee80211_tx_data *tx)
 {
 	struct sta_info *sta = tx->sta;
+	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(tx->skb);
 	u32 staflags;
 	DECLARE_MAC_BUF(mac);
 
@@ -404,7 +407,6 @@ ieee80211_tx_h_unicast_ps_buf(struct ieee80211_tx_data *tx)
 
 	if (unlikely((staflags & WLAN_STA_PS) &&
 		     !(staflags & WLAN_STA_PSPOLL))) {
-		struct ieee80211_tx_packet_data *pkt_data;
 #ifdef CONFIG_MAC80211_VERBOSE_PS_DEBUG
 		printk(KERN_DEBUG "STA %s aid %d: PS buffer (entries "
 		       "before %d)\n",
@@ -428,8 +430,7 @@ ieee80211_tx_h_unicast_ps_buf(struct ieee80211_tx_data *tx)
 		if (skb_queue_empty(&sta->ps_tx_buf))
 			sta_info_set_tim_bit(sta);
 
-		pkt_data = (struct ieee80211_tx_packet_data *)tx->skb->cb;
-		pkt_data->jiffies = jiffies;
+		info->control.jiffies = jiffies;
 		skb_queue_tail(&sta->ps_tx_buf, tx->skb);
 		return TX_QUEUED;
 	}
@@ -461,17 +462,18 @@ static ieee80211_tx_result
 ieee80211_tx_h_select_key(struct ieee80211_tx_data *tx)
 {
 	struct ieee80211_key *key;
+	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(tx->skb);
 	u16 fc = tx->fc;
 
-	if (unlikely(tx->control->flags & IEEE80211_TXCTL_DO_NOT_ENCRYPT))
+	if (unlikely(info->flags & IEEE80211_TX_CTL_DO_NOT_ENCRYPT))
 		tx->key = NULL;
 	else if (tx->sta && (key = rcu_dereference(tx->sta->key)))
 		tx->key = key;
 	else if ((key = rcu_dereference(tx->sdata->default_key)))
 		tx->key = key;
 	else if (tx->sdata->drop_unencrypted &&
-		 !(tx->control->flags & IEEE80211_TXCTL_EAPOL_FRAME) &&
-		 !(tx->flags & IEEE80211_TX_INJECTED)) {
+		 !(info->flags & IEEE80211_TX_CTL_EAPOL_FRAME) &&
+		 !(info->flags & IEEE80211_TX_CTL_INJECTED)) {
 		I802_DEBUG_INC(tx->local->tx_handlers_drop_unencrypted);
 		return TX_DROP;
 	} else
@@ -500,7 +502,7 @@ ieee80211_tx_h_select_key(struct ieee80211_tx_data *tx)
 	}
 
 	if (!tx->key || !(tx->key->flags & KEY_FLAG_UPLOADED_TO_HARDWARE))
-		tx->control->flags |= IEEE80211_TXCTL_DO_NOT_ENCRYPT;
+		info->flags |= IEEE80211_TX_CTL_DO_NOT_ENCRYPT;
 
 	return TX_CONTINUE;
 }
@@ -510,6 +512,7 @@ ieee80211_tx_h_rate_ctrl(struct ieee80211_tx_data *tx)
 {
 	struct rate_selection rsel;
 	struct ieee80211_supported_band *sband;
+	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(tx->skb);
 
 	sband = tx->local->hw.wiphy->bands[tx->channel->band];
 
@@ -517,18 +520,17 @@ ieee80211_tx_h_rate_ctrl(struct ieee80211_tx_data *tx)
 		rate_control_get_rate(tx->dev, sband, tx->skb, &rsel);
 		tx->rate_idx = rsel.rate_idx;
 		if (unlikely(rsel.probe_idx >= 0)) {
-			tx->control->flags |=
-				IEEE80211_TXCTL_RATE_CTRL_PROBE;
+			info->flags |= IEEE80211_TX_CTL_RATE_CTRL_PROBE;
 			tx->flags |= IEEE80211_TX_PROBE_LAST_FRAG;
-			tx->control->alt_retry_rate_idx = tx->rate_idx;
+			info->control.alt_retry_rate_idx = tx->rate_idx;
 			tx->rate_idx = rsel.probe_idx;
 		} else
-			tx->control->alt_retry_rate_idx = -1;
+			info->control.alt_retry_rate_idx = -1;
 
 		if (unlikely(tx->rate_idx < 0))
 			return TX_DROP;
 	} else
-		tx->control->alt_retry_rate_idx = -1;
+		info->control.alt_retry_rate_idx = -1;
 
 	if (tx->sdata->bss_conf.use_cts_prot &&
 	    (tx->flags & IEEE80211_TX_FRAGMENTED) && (rsel.nonerp_idx >= 0)) {
@@ -538,13 +540,13 @@ ieee80211_tx_h_rate_ctrl(struct ieee80211_tx_data *tx)
 		else
 			tx->flags |= IEEE80211_TX_PROBE_LAST_FRAG;
 		tx->rate_idx = rsel.nonerp_idx;
-		tx->control->tx_rate_idx = rsel.nonerp_idx;
-		tx->control->flags &= ~IEEE80211_TXCTL_RATE_CTRL_PROBE;
+		info->tx_rate_idx = rsel.nonerp_idx;
+		info->flags &= ~IEEE80211_TX_CTL_RATE_CTRL_PROBE;
 	} else {
 		tx->last_frag_rate_idx = tx->rate_idx;
-		tx->control->tx_rate_idx = tx->rate_idx;
+		info->tx_rate_idx = tx->rate_idx;
 	}
-	tx->control->tx_rate_idx = tx->rate_idx;
+	info->tx_rate_idx = tx->rate_idx;
 
 	return TX_CONTINUE;
 }
@@ -555,28 +557,32 @@ ieee80211_tx_h_misc(struct ieee80211_tx_data *tx)
 	struct ieee80211_hdr *hdr = (struct ieee80211_hdr *) tx->skb->data;
 	u16 fc = le16_to_cpu(hdr->frame_control);
 	u16 dur;
-	struct ieee80211_tx_control *control = tx->control;
+	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(tx->skb);
 	struct ieee80211_supported_band *sband;
 
 	sband = tx->local->hw.wiphy->bands[tx->channel->band];
 
-	if (!control->retry_limit) {
+	if (tx->sta)
+		info->control.aid = tx->sta->aid;
+
+	if (!info->control.retry_limit) {
 		if (!is_multicast_ether_addr(hdr->addr1)) {
-			if (tx->skb->len + FCS_LEN > tx->local->rts_threshold
+			int len = min_t(int, tx->skb->len + FCS_LEN,
+					tx->local->fragmentation_threshold);
+			if (len > tx->local->rts_threshold
 			    && tx->local->rts_threshold <
-					IEEE80211_MAX_RTS_THRESHOLD) {
-				control->flags |=
-					IEEE80211_TXCTL_USE_RTS_CTS;
-				control->flags |=
-					IEEE80211_TXCTL_LONG_RETRY_LIMIT;
-				control->retry_limit =
+						IEEE80211_MAX_RTS_THRESHOLD) {
+				info->flags |= IEEE80211_TX_CTL_USE_RTS_CTS;
+				info->flags |=
+					IEEE80211_TX_CTL_LONG_RETRY_LIMIT;
+				info->control.retry_limit =
 					tx->local->long_retry_limit;
 			} else {
-				control->retry_limit =
+				info->control.retry_limit =
 					tx->local->short_retry_limit;
 			}
 		} else {
-			control->retry_limit = 1;
+			info->control.retry_limit = 1;
 		}
 	}
 
@@ -585,7 +591,7 @@ ieee80211_tx_h_misc(struct ieee80211_tx_data *tx)
 		 * frames.
 		 * TODO: The last fragment could still use multiple retry
 		 * rates. */
-		control->alt_retry_rate_idx = -1;
+		info->control.alt_retry_rate_idx = -1;
 	}
 
 	/* Use CTS protection for unicast frames sent using extended rates if
@@ -595,8 +601,8 @@ ieee80211_tx_h_misc(struct ieee80211_tx_data *tx)
 	    (sband->bitrates[tx->rate_idx].flags & IEEE80211_RATE_ERP_G) &&
 	    (tx->flags & IEEE80211_TX_UNICAST) &&
 	    tx->sdata->bss_conf.use_cts_prot &&
-	    !(control->flags & IEEE80211_TXCTL_USE_RTS_CTS))
-		control->flags |= IEEE80211_TXCTL_USE_CTS_PROTECT;
+	    !(info->flags & IEEE80211_TX_CTL_USE_RTS_CTS))
+		info->flags |= IEEE80211_TX_CTL_USE_CTS_PROTECT;
 
 	/* Transmit data frames using short preambles if the driver supports
 	 * short preambles at the selected rate and short preambles are
@@ -605,7 +611,7 @@ ieee80211_tx_h_misc(struct ieee80211_tx_data *tx)
 	    (sband->bitrates[tx->rate_idx].flags & IEEE80211_RATE_SHORT_PREAMBLE) &&
 	    tx->sdata->bss_conf.use_short_preamble &&
 	    (!tx->sta || test_sta_flags(tx->sta, WLAN_STA_SHORT_PREAMBLE))) {
-		tx->control->flags |= IEEE80211_TXCTL_SHORT_PREAMBLE;
+		info->flags |= IEEE80211_TX_CTL_SHORT_PREAMBLE;
 	}
 
 	/* Setup duration field for the first fragment of the frame. Duration
@@ -616,8 +622,8 @@ ieee80211_tx_h_misc(struct ieee80211_tx_data *tx)
 				 tx->extra_frag[0]->len : 0);
 	hdr->duration_id = cpu_to_le16(dur);
 
-	if ((control->flags & IEEE80211_TXCTL_USE_RTS_CTS) ||
-	    (control->flags & IEEE80211_TXCTL_USE_CTS_PROTECT)) {
+	if ((info->flags & IEEE80211_TX_CTL_USE_RTS_CTS) ||
+	    (info->flags & IEEE80211_TX_CTL_USE_CTS_PROTECT)) {
 		struct ieee80211_supported_band *sband;
 		struct ieee80211_rate *rate;
 		s8 baserate = -1;
@@ -626,7 +632,7 @@ ieee80211_tx_h_misc(struct ieee80211_tx_data *tx)
 		sband = tx->local->hw.wiphy->bands[tx->channel->band];
 
 		/* Do not use multiple retry rates when using RTS/CTS */
-		control->alt_retry_rate_idx = -1;
+		info->control.alt_retry_rate_idx = -1;
 
 		/* Use min(data rate, max base rate) as CTS/RTS rate */
 		rate = &sband->bitrates[tx->rate_idx];
@@ -642,13 +648,13 @@ ieee80211_tx_h_misc(struct ieee80211_tx_data *tx)
 		}
 
 		if (baserate >= 0)
-			control->rts_cts_rate_idx = baserate;
+			info->control.rts_cts_rate_idx = baserate;
 		else
-			control->rts_cts_rate_idx = 0;
+			info->control.rts_cts_rate_idx = 0;
 	}
 
 	if (tx->sta)
-		control->aid = tx->sta->aid;
+		info->control.aid = tx->sta->aid;
 
 	return TX_CONTINUE;
 }
@@ -762,6 +768,7 @@ ieee80211_tx_h_stats(struct ieee80211_tx_data *tx)
 	u32 load = 0, hdrtime;
 	struct ieee80211_rate *rate;
 	struct ieee80211_supported_band *sband;
+	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
 
 	sband = tx->local->hw.wiphy->bands[tx->channel->band];
 	rate = &sband->bitrates[tx->rate_idx];
@@ -786,9 +793,9 @@ ieee80211_tx_h_stats(struct ieee80211_tx_data *tx)
 	if (!is_multicast_ether_addr(hdr->addr1))
 		load += hdrtime;
 
-	if (tx->control->flags & IEEE80211_TXCTL_USE_RTS_CTS)
+	if (info->flags & IEEE80211_TX_CTL_USE_RTS_CTS)
 		load += 2 * hdrtime;
-	else if (tx->control->flags & IEEE80211_TXCTL_USE_CTS_PROTECT)
+	else if (info->flags & IEEE80211_TX_CTL_USE_CTS_PROTECT)
 		load += hdrtime;
 
 	/* TODO: optimise again */
@@ -839,6 +846,7 @@ static ieee80211_tx_handler ieee80211_tx_handlers[] =
 	ieee80211_tx_h_rate_ctrl,
 	ieee80211_tx_h_misc,
 	ieee80211_tx_h_fragment,
+	/* handlers after fragment must be aware of tx info fragmentation! */
 	ieee80211_tx_h_encrypt,
 	ieee80211_tx_h_stats,
 	NULL
@@ -867,12 +875,12 @@ __ieee80211_parse_tx_radiotap(struct ieee80211_tx_data *tx,
 		(struct ieee80211_radiotap_header *) skb->data;
 	struct ieee80211_supported_band *sband;
 	int ret = ieee80211_radiotap_iterator_init(&iterator, rthdr, skb->len);
-	struct ieee80211_tx_control *control = tx->control;
+	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
 
 	sband = tx->local->hw.wiphy->bands[tx->channel->band];
 
-	control->flags |= IEEE80211_TXCTL_DO_NOT_ENCRYPT;
-	tx->flags |= IEEE80211_TX_INJECTED;
+	info->flags |= IEEE80211_TX_CTL_DO_NOT_ENCRYPT;
+	info->flags |= IEEE80211_TX_CTL_INJECTED;
 	tx->flags &= ~IEEE80211_TX_FRAGMENTED;
 
 	/*
@@ -920,7 +928,7 @@ __ieee80211_parse_tx_radiotap(struct ieee80211_tx_data *tx,
 			 * radiotap uses 0 for 1st ant, mac80211 is 1 for
 			 * 1st ant
 			 */
-			control->antenna_sel_tx = (*iterator.this_arg) + 1;
+			info->antenna_sel_tx = (*iterator.this_arg) + 1;
 			break;
 
 #if 0
@@ -944,8 +952,8 @@ __ieee80211_parse_tx_radiotap(struct ieee80211_tx_data *tx,
 				skb_trim(skb, skb->len - FCS_LEN);
 			}
 			if (*iterator.this_arg & IEEE80211_RADIOTAP_F_WEP)
-				control->flags &=
-					~IEEE80211_TXCTL_DO_NOT_ENCRYPT;
+				info->flags &=
+					~IEEE80211_TX_CTL_DO_NOT_ENCRYPT;
 			if (*iterator.this_arg & IEEE80211_RADIOTAP_F_FRAG)
 				tx->flags |= IEEE80211_TX_FRAGMENTED;
 			break;
@@ -980,12 +988,12 @@ __ieee80211_parse_tx_radiotap(struct ieee80211_tx_data *tx,
 static ieee80211_tx_result
 __ieee80211_tx_prepare(struct ieee80211_tx_data *tx,
 		       struct sk_buff *skb,
-		       struct net_device *dev,
-		       struct ieee80211_tx_control *control)
+		       struct net_device *dev)
 {
 	struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr);
 	struct ieee80211_hdr *hdr;
 	struct ieee80211_sub_if_data *sdata;
+	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
 
 	int hdrlen;
 
@@ -994,7 +1002,7 @@ __ieee80211_tx_prepare(struct ieee80211_tx_data *tx,
 	tx->dev = dev; /* use original interface */
 	tx->local = local;
 	tx->sdata = IEEE80211_DEV_TO_SUB_IF(dev);
-	tx->control = control;
+	tx->channel = local->hw.conf.channel;
 	/*
 	 * Set this flag (used below to indicate "automatic fragmentation"),
 	 * it will be cleared/left by radiotap as desired.
@@ -1021,10 +1029,10 @@ __ieee80211_tx_prepare(struct ieee80211_tx_data *tx,
 
 	if (is_multicast_ether_addr(hdr->addr1)) {
 		tx->flags &= ~IEEE80211_TX_UNICAST;
-		control->flags |= IEEE80211_TXCTL_NO_ACK;
+		info->flags |= IEEE80211_TX_CTL_NO_ACK;
 	} else {
 		tx->flags |= IEEE80211_TX_UNICAST;
-		control->flags &= ~IEEE80211_TXCTL_NO_ACK;
+		info->flags &= ~IEEE80211_TX_CTL_NO_ACK;
 	}
 
 	if (tx->flags & IEEE80211_TX_FRAGMENTED) {
@@ -1037,16 +1045,16 @@ __ieee80211_tx_prepare(struct ieee80211_tx_data *tx,
 	}
 
 	if (!tx->sta)
-		control->flags |= IEEE80211_TXCTL_CLEAR_PS_FILT;
+		info->flags |= IEEE80211_TX_CTL_CLEAR_PS_FILT;
 	else if (test_and_clear_sta_flags(tx->sta, WLAN_STA_CLEAR_PS_FILT))
-		control->flags |= IEEE80211_TXCTL_CLEAR_PS_FILT;
+		info->flags |= IEEE80211_TX_CTL_CLEAR_PS_FILT;
 
 	hdrlen = ieee80211_get_hdrlen(tx->fc);
 	if (skb->len > hdrlen + sizeof(rfc1042_header) + 2) {
 		u8 *pos = &skb->data[hdrlen + sizeof(rfc1042_header)];
 		tx->ethertype = (pos[0] << 8) | pos[1];
 	}
-	control->flags |= IEEE80211_TXCTL_FIRST_FRAGMENT;
+	info->flags |= IEEE80211_TX_CTL_FIRST_FRAGMENT;
 
 	return TX_CONTINUE;
 }
@@ -1056,14 +1064,12 @@ __ieee80211_tx_prepare(struct ieee80211_tx_data *tx,
  */
 static int ieee80211_tx_prepare(struct ieee80211_tx_data *tx,
 				struct sk_buff *skb,
-				struct net_device *mdev,
-				struct ieee80211_tx_control *control)
+				struct net_device *mdev)
 {
-	struct ieee80211_tx_packet_data *pkt_data;
+	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
 	struct net_device *dev;
 
-	pkt_data = (struct ieee80211_tx_packet_data *)skb->cb;
-	dev = dev_get_by_index(&init_net, pkt_data->ifindex);
+	dev = dev_get_by_index(&init_net, info->control.ifindex);
 	if (unlikely(dev && !is_ieee80211_device(dev, mdev))) {
 		dev_put(dev);
 		dev = NULL;
@@ -1071,7 +1077,7 @@ static int ieee80211_tx_prepare(struct ieee80211_tx_data *tx,
 	if (unlikely(!dev))
 		return -ENODEV;
 	/* initialises tx with control */
-	__ieee80211_tx_prepare(tx, skb, dev, control);
+	__ieee80211_tx_prepare(tx, skb, dev);
 	dev_put(dev);
 	return 0;
 }
@@ -1079,7 +1085,7 @@ static int ieee80211_tx_prepare(struct ieee80211_tx_data *tx,
 static int __ieee80211_tx(struct ieee80211_local *local, struct sk_buff *skb,
 			  struct ieee80211_tx_data *tx)
 {
-	struct ieee80211_tx_control *control = tx->control;
+	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
 	int ret, i;
 
 	if (!ieee80211_qdisc_installed(local->mdev) &&
@@ -1090,39 +1096,39 @@ static int __ieee80211_tx(struct ieee80211_local *local, struct sk_buff *skb,
 	if (skb) {
 		ieee80211_dump_frame(wiphy_name(local->hw.wiphy),
 				     "TX to low-level driver", skb);
-		ret = local->ops->tx(local_to_hw(local), skb, control);
+		ret = local->ops->tx(local_to_hw(local), skb);
 		if (ret)
 			return IEEE80211_TX_AGAIN;
 		local->mdev->trans_start = jiffies;
 		ieee80211_led_tx(local, 1);
 	}
 	if (tx->extra_frag) {
-		control->flags &= ~(IEEE80211_TXCTL_USE_RTS_CTS |
-				    IEEE80211_TXCTL_USE_CTS_PROTECT |
-				    IEEE80211_TXCTL_CLEAR_PS_FILT |
-				    IEEE80211_TXCTL_FIRST_FRAGMENT);
 		for (i = 0; i < tx->num_extra_frag; i++) {
 			if (!tx->extra_frag[i])
 				continue;
-			if (__ieee80211_queue_stopped(local, control->queue))
+			info = IEEE80211_SKB_CB(tx->extra_frag[i]);
+			info->flags &= ~(IEEE80211_TX_CTL_USE_RTS_CTS |
+					 IEEE80211_TX_CTL_USE_CTS_PROTECT |
+					 IEEE80211_TX_CTL_CLEAR_PS_FILT |
+					 IEEE80211_TX_CTL_FIRST_FRAGMENT);
+			if (__ieee80211_queue_stopped(local, info->queue))
 				return IEEE80211_TX_FRAG_AGAIN;
 			if (i == tx->num_extra_frag) {
-				control->tx_rate_idx = tx->last_frag_rate_idx;
+				info->tx_rate_idx = tx->last_frag_rate_idx;
 
 				if (tx->flags & IEEE80211_TX_PROBE_LAST_FRAG)
-					control->flags |=
-						IEEE80211_TXCTL_RATE_CTRL_PROBE;
+					info->flags |=
+						IEEE80211_TX_CTL_RATE_CTRL_PROBE;
 				else
-					control->flags &=
-						~IEEE80211_TXCTL_RATE_CTRL_PROBE;
+					info->flags &=
+						~IEEE80211_TX_CTL_RATE_CTRL_PROBE;
 			}
 
 			ieee80211_dump_frame(wiphy_name(local->hw.wiphy),
 					     "TX to low-level driver",
 					     tx->extra_frag[i]);
 			ret = local->ops->tx(local_to_hw(local),
-					    tx->extra_frag[i],
-					    control);
+					    tx->extra_frag[i]);
 			if (ret)
 				return IEEE80211_TX_FRAG_AGAIN;
 			local->mdev->trans_start = jiffies;
@@ -1135,17 +1141,18 @@ static int __ieee80211_tx(struct ieee80211_local *local, struct sk_buff *skb,
 	return IEEE80211_TX_OK;
 }
 
-static int ieee80211_tx(struct net_device *dev, struct sk_buff *skb,
-			struct ieee80211_tx_control *control)
+static int ieee80211_tx(struct net_device *dev, struct sk_buff *skb)
 {
 	struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr);
 	struct sta_info *sta;
 	ieee80211_tx_handler *handler;
 	struct ieee80211_tx_data tx;
 	ieee80211_tx_result res = TX_DROP, res_prepare;
+	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
 	int ret, i;
+	int queue = info->queue;
 
-	WARN_ON(__ieee80211_queue_pending(local, control->queue));
+	WARN_ON(__ieee80211_queue_pending(local, queue));
 
 	if (unlikely(skb->len < 10)) {
 		dev_kfree_skb(skb);
@@ -1155,7 +1162,7 @@ static int ieee80211_tx(struct net_device *dev, struct sk_buff *skb,
 	rcu_read_lock();
 
 	/* initialises tx */
-	res_prepare = __ieee80211_tx_prepare(&tx, skb, dev, control);
+	res_prepare = __ieee80211_tx_prepare(&tx, skb, dev);
 
 	if (res_prepare == TX_DROP) {
 		dev_kfree_skb(skb);
@@ -1165,7 +1172,7 @@ static int ieee80211_tx(struct net_device *dev, struct sk_buff *skb,
 
 	sta = tx.sta;
 	tx.channel = local->hw.conf.channel;
-	control->band = tx.channel->band;
+	info->band = tx.channel->band;
 
 	for (handler = ieee80211_tx_handlers; *handler != NULL;
 	     handler++) {
@@ -1174,7 +1181,8 @@ static int ieee80211_tx(struct net_device *dev, struct sk_buff *skb,
 			break;
 	}
 
-	skb = tx.skb; /* handlers are allowed to change skb */
+	if (WARN_ON(tx.skb != skb))
+		goto drop;
 
 	if (unlikely(res == TX_DROP)) {
 		I802_DEBUG_INC(local->tx_handlers_drop);
@@ -1209,12 +1217,12 @@ retry:
 	ret = __ieee80211_tx(local, skb, &tx);
 	if (ret) {
 		struct ieee80211_tx_stored_packet *store =
-			&local->pending_packet[control->queue];
+			&local->pending_packet[info->queue];
 
 		if (ret == IEEE80211_TX_FRAG_AGAIN)
 			skb = NULL;
 		set_bit(IEEE80211_LINK_STATE_PENDING,
-			&local->state[control->queue]);
+			&local->state[queue]);
 		smp_mb();
 		/* When the driver gets out of buffers during sending of
 		 * fragments and calls ieee80211_stop_queue, there is
@@ -1225,13 +1233,11 @@ retry:
 		 * called with IEEE80211_LINK_STATE_PENDING. Prevent this by
 		 * continuing transmitting here when that situation is
 		 * possible to have happened. */
-		if (!__ieee80211_queue_stopped(local, control->queue)) {
+		if (!__ieee80211_queue_stopped(local, queue)) {
 			clear_bit(IEEE80211_LINK_STATE_PENDING,
-				  &local->state[control->queue]);
+				  &local->state[queue]);
 			goto retry;
 		}
-		memcpy(&store->control, control,
-		       sizeof(struct ieee80211_tx_control));
 		store->skb = skb;
 		store->extra_frag = tx.extra_frag;
 		store->num_extra_frag = tx.num_extra_frag;
@@ -1258,21 +1264,14 @@ retry:
 int ieee80211_master_start_xmit(struct sk_buff *skb,
 				struct net_device *dev)
 {
-	struct ieee80211_tx_control control;
-	struct ieee80211_tx_packet_data *pkt_data;
+	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
 	struct net_device *odev = NULL;
 	struct ieee80211_sub_if_data *osdata;
 	int headroom;
 	int ret;
 
-	/*
-	 * copy control out of the skb so other people can use skb->cb
-	 */
-	pkt_data = (struct ieee80211_tx_packet_data *)skb->cb;
-	memset(&control, 0, sizeof(struct ieee80211_tx_control));
-
-	if (pkt_data->ifindex)
-		odev = dev_get_by_index(&init_net, pkt_data->ifindex);
+	if (info->control.ifindex)
+		odev = dev_get_by_index(&init_net, info->control.ifindex);
 	if (unlikely(odev && !is_ieee80211_device(odev, dev))) {
 		dev_put(odev);
 		odev = NULL;
@@ -1285,6 +1284,7 @@ int ieee80211_master_start_xmit(struct sk_buff *skb,
 		dev_kfree_skb(skb);
 		return 0;
 	}
+
 	osdata = IEEE80211_DEV_TO_SUB_IF(odev);
 
 	headroom = osdata->local->tx_headroom + IEEE80211_ENCRYPT_HEADROOM;
@@ -1296,21 +1296,8 @@ int ieee80211_master_start_xmit(struct sk_buff *skb,
 		}
 	}
 
-	control.vif = &osdata->vif;
-	control.type = osdata->vif.type;
-	if (pkt_data->flags & IEEE80211_TXPD_REQ_TX_STATUS)
-		control.flags |= IEEE80211_TXCTL_REQ_TX_STATUS;
-	if (pkt_data->flags & IEEE80211_TXPD_DO_NOT_ENCRYPT)
-		control.flags |= IEEE80211_TXCTL_DO_NOT_ENCRYPT;
-	if (pkt_data->flags & IEEE80211_TXPD_REQUEUE)
-		control.flags |= IEEE80211_TXCTL_REQUEUE;
-	if (pkt_data->flags & IEEE80211_TXPD_EAPOL_FRAME)
-		control.flags |= IEEE80211_TXCTL_EAPOL_FRAME;
-	if (pkt_data->flags & IEEE80211_TXPD_AMPDU)
-		control.flags |= IEEE80211_TXCTL_AMPDU;
-	control.queue = pkt_data->queue;
-
-	ret = ieee80211_tx(odev, skb, &control);
+	info->control.vif = &osdata->vif;
+	ret = ieee80211_tx(odev, skb);
 	dev_put(odev);
 
 	return ret;
@@ -1320,7 +1307,7 @@ int ieee80211_monitor_start_xmit(struct sk_buff *skb,
 				 struct net_device *dev)
 {
 	struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr);
-	struct ieee80211_tx_packet_data *pkt_data;
+	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
 	struct ieee80211_radiotap_header *prthdr =
 		(struct ieee80211_radiotap_header *)skb->data;
 	u16 len_rthdr;
@@ -1342,14 +1329,12 @@ int ieee80211_monitor_start_xmit(struct sk_buff *skb,
 
 	skb->dev = local->mdev;
 
-	pkt_data = (struct ieee80211_tx_packet_data *)skb->cb;
-	memset(pkt_data, 0, sizeof(*pkt_data));
 	/* needed because we set skb device to master */
-	pkt_data->ifindex = dev->ifindex;
+	info->control.ifindex = dev->ifindex;
 
-	pkt_data->flags |= IEEE80211_TXPD_DO_NOT_ENCRYPT;
+	info->flags |= IEEE80211_TX_CTL_DO_NOT_ENCRYPT;
 	/* Interfaces should always request a status report */
-	pkt_data->flags |= IEEE80211_TXPD_REQ_TX_STATUS;
+	info->flags |= IEEE80211_TX_CTL_REQ_TX_STATUS;
 
 	/*
 	 * fix up the pointers accounting for the radiotap
@@ -1393,7 +1378,7 @@ int ieee80211_subif_start_xmit(struct sk_buff *skb,
 			       struct net_device *dev)
 {
 	struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr);
-	struct ieee80211_tx_packet_data *pkt_data;
+	struct ieee80211_tx_info *info;
 	struct ieee80211_sub_if_data *sdata;
 	int ret = 1, head_need;
 	u16 ethertype, hdrlen,  meshhdrlen = 0, fc;
@@ -1625,14 +1610,14 @@ int ieee80211_subif_start_xmit(struct sk_buff *skb,
 	nh_pos += hdrlen;
 	h_pos += hdrlen;
 
-	pkt_data = (struct ieee80211_tx_packet_data *)skb->cb;
-	memset(pkt_data, 0, sizeof(struct ieee80211_tx_packet_data));
-	pkt_data->ifindex = dev->ifindex;
+	info = IEEE80211_SKB_CB(skb);
+	memset(info, 0, sizeof(*info));
+	info->control.ifindex = dev->ifindex;
 	if (ethertype == ETH_P_PAE)
-		pkt_data->flags |= IEEE80211_TXPD_EAPOL_FRAME;
+		info->flags |= IEEE80211_TX_CTL_EAPOL_FRAME;
 
 	/* Interfaces should always request a status report */
-	pkt_data->flags |= IEEE80211_TXPD_REQ_TX_STATUS;
+	info->flags |= IEEE80211_TX_CTL_REQ_TX_STATUS;
 
 	skb->dev = local->mdev;
 	dev->stats.tx_packets++;
@@ -1693,7 +1678,6 @@ void ieee80211_tx_pending(unsigned long data)
 			continue;
 		}
 		store = &local->pending_packet[i];
-		tx.control = &store->control;
 		tx.extra_frag = store->extra_frag;
 		tx.num_extra_frag = store->num_extra_frag;
 		tx.last_frag_rate_idx = store->last_frag_rate_idx;
@@ -1786,11 +1770,11 @@ static void ieee80211_beacon_add_tim(struct ieee80211_local *local,
 }
 
 struct sk_buff *ieee80211_beacon_get(struct ieee80211_hw *hw,
-				     struct ieee80211_vif *vif,
-				     struct ieee80211_tx_control *control)
+				     struct ieee80211_vif *vif)
 {
 	struct ieee80211_local *local = hw_to_local(hw);
 	struct sk_buff *skb;
+	struct ieee80211_tx_info *info;
 	struct net_device *bdev;
 	struct ieee80211_sub_if_data *sdata = NULL;
 	struct ieee80211_if_ap *ap = NULL;
@@ -1896,31 +1880,32 @@ struct sk_buff *ieee80211_beacon_get(struct ieee80211_hw *hw,
 		goto out;
 	}
 
-	if (control) {
-		control->band = band;
-		rate_control_get_rate(local->mdev, sband, skb, &rsel);
-		if (unlikely(rsel.rate_idx < 0)) {
-			if (net_ratelimit()) {
-				printk(KERN_DEBUG "%s: ieee80211_beacon_get: "
-				       "no rate found\n",
-				       wiphy_name(local->hw.wiphy));
-			}
-			dev_kfree_skb(skb);
-			skb = NULL;
-			goto out;
-		}
+	info = IEEE80211_SKB_CB(skb);
+
+	info->band = band;
+	rate_control_get_rate(local->mdev, sband, skb, &rsel);
 
-		control->vif = vif;
-		control->tx_rate_idx = rsel.rate_idx;
-		if (sdata->bss_conf.use_short_preamble &&
-		    sband->bitrates[rsel.rate_idx].flags & IEEE80211_RATE_SHORT_PREAMBLE)
-			control->flags |= IEEE80211_TXCTL_SHORT_PREAMBLE;
-		control->antenna_sel_tx = local->hw.conf.antenna_sel_tx;
-		control->flags |= IEEE80211_TXCTL_NO_ACK;
-		control->flags |= IEEE80211_TXCTL_DO_NOT_ENCRYPT;
-		control->retry_limit = 1;
-		control->flags |= IEEE80211_TXCTL_CLEAR_PS_FILT;
+	if (unlikely(rsel.rate_idx < 0)) {
+		if (net_ratelimit()) {
+			printk(KERN_DEBUG "%s: ieee80211_beacon_get: "
+			       "no rate found\n",
+			       wiphy_name(local->hw.wiphy));
+		}
+		dev_kfree_skb(skb);
+		skb = NULL;
+		goto out;
 	}
+
+	info->control.vif = vif;
+	info->tx_rate_idx = rsel.rate_idx;
+	if (sdata->bss_conf.use_short_preamble &&
+	    sband->bitrates[rsel.rate_idx].flags & IEEE80211_RATE_SHORT_PREAMBLE)
+		info->flags |= IEEE80211_TX_CTL_SHORT_PREAMBLE;
+	info->antenna_sel_tx = local->hw.conf.antenna_sel_tx;
+	info->flags |= IEEE80211_TX_CTL_NO_ACK;
+	info->flags |= IEEE80211_TX_CTL_DO_NOT_ENCRYPT;
+	info->control.retry_limit = 1;
+	info->flags |= IEEE80211_TX_CTL_CLEAR_PS_FILT;
 	(*num_beacons)++;
 out:
 	rcu_read_unlock();
@@ -1930,7 +1915,7 @@ EXPORT_SYMBOL(ieee80211_beacon_get);
 
 void ieee80211_rts_get(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 		       const void *frame, size_t frame_len,
-		       const struct ieee80211_tx_control *frame_txctl,
+		       const struct ieee80211_tx_info *frame_txctl,
 		       struct ieee80211_rts *rts)
 {
 	const struct ieee80211_hdr *hdr = frame;
@@ -1947,7 +1932,7 @@ EXPORT_SYMBOL(ieee80211_rts_get);
 
 void ieee80211_ctstoself_get(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 			     const void *frame, size_t frame_len,
-			     const struct ieee80211_tx_control *frame_txctl,
+			     const struct ieee80211_tx_info *frame_txctl,
 			     struct ieee80211_cts *cts)
 {
 	const struct ieee80211_hdr *hdr = frame;
@@ -1963,8 +1948,7 @@ EXPORT_SYMBOL(ieee80211_ctstoself_get);
 
 struct sk_buff *
 ieee80211_get_buffered_bc(struct ieee80211_hw *hw,
-			  struct ieee80211_vif *vif,
-			  struct ieee80211_tx_control *control)
+			  struct ieee80211_vif *vif)
 {
 	struct ieee80211_local *local = hw_to_local(hw);
 	struct sk_buff *skb;
@@ -1976,6 +1960,7 @@ ieee80211_get_buffered_bc(struct ieee80211_hw *hw,
 	struct ieee80211_sub_if_data *sdata;
 	struct ieee80211_if_ap *bss = NULL;
 	struct beacon_data *beacon;
+	struct ieee80211_tx_info *info;
 
 	sdata = vif_to_sdata(vif);
 	bdev = sdata->dev;
@@ -1995,7 +1980,7 @@ ieee80211_get_buffered_bc(struct ieee80211_hw *hw,
 
 	if (bss->dtim_count != 0)
 		return NULL; /* send buffered bc/mc only after DTIM beacon */
-	memset(control, 0, sizeof(*control));
+
 	while (1) {
 		skb = skb_dequeue(&bss->ps_bc_buf);
 		if (!skb)
@@ -2012,21 +1997,26 @@ ieee80211_get_buffered_bc(struct ieee80211_hw *hw,
 				cpu_to_le16(IEEE80211_FCTL_MOREDATA);
 		}
 
-		if (!ieee80211_tx_prepare(&tx, skb, local->mdev, control))
+		if (!ieee80211_tx_prepare(&tx, skb, local->mdev))
 			break;
 		dev_kfree_skb_any(skb);
 	}
+
+	info = IEEE80211_SKB_CB(skb);
+
 	sta = tx.sta;
 	tx.flags |= IEEE80211_TX_PS_BUFFERED;
 	tx.channel = local->hw.conf.channel;
-	control->band = tx.channel->band;
+	info->band = tx.channel->band;
 
 	for (handler = ieee80211_tx_handlers; *handler != NULL; handler++) {
 		res = (*handler)(&tx);
 		if (res == TX_DROP || res == TX_QUEUED)
 			break;
 	}
-	skb = tx.skb; /* handlers are allowed to change skb */
+
+	if (WARN_ON(tx.skb != skb))
+		return NULL;
 
 	if (res == TX_DROP) {
 		I802_DEBUG_INC(local->tx_handlers_drop);
diff --git a/net/mac80211/util.c b/net/mac80211/util.c
index 65a34fddeb00..d9109dee461f 100644
--- a/net/mac80211/util.c
+++ b/net/mac80211/util.c
@@ -258,7 +258,7 @@ EXPORT_SYMBOL(ieee80211_generic_frame_duration);
 
 __le16 ieee80211_rts_duration(struct ieee80211_hw *hw,
 			      struct ieee80211_vif *vif, size_t frame_len,
-			      const struct ieee80211_tx_control *frame_txctl)
+			      const struct ieee80211_tx_info *frame_txctl)
 {
 	struct ieee80211_local *local = hw_to_local(hw);
 	struct ieee80211_rate *rate;
@@ -272,7 +272,7 @@ __le16 ieee80211_rts_duration(struct ieee80211_hw *hw,
 
 	short_preamble = sdata->bss_conf.use_short_preamble;
 
-	rate = &sband->bitrates[frame_txctl->rts_cts_rate_idx];
+	rate = &sband->bitrates[frame_txctl->control.rts_cts_rate_idx];
 
 	erp = 0;
 	if (sdata->flags & IEEE80211_SDATA_OPERATING_GMODE)
@@ -295,7 +295,7 @@ EXPORT_SYMBOL(ieee80211_rts_duration);
 __le16 ieee80211_ctstoself_duration(struct ieee80211_hw *hw,
 				    struct ieee80211_vif *vif,
 				    size_t frame_len,
-				    const struct ieee80211_tx_control *frame_txctl)
+				    const struct ieee80211_tx_info *frame_txctl)
 {
 	struct ieee80211_local *local = hw_to_local(hw);
 	struct ieee80211_rate *rate;
@@ -309,7 +309,7 @@ __le16 ieee80211_ctstoself_duration(struct ieee80211_hw *hw,
 
 	short_preamble = sdata->bss_conf.use_short_preamble;
 
-	rate = &sband->bitrates[frame_txctl->rts_cts_rate_idx];
+	rate = &sband->bitrates[frame_txctl->control.rts_cts_rate_idx];
 	erp = 0;
 	if (sdata->flags & IEEE80211_SDATA_OPERATING_GMODE)
 		erp = rate->flags & IEEE80211_RATE_ERP_G;
@@ -317,7 +317,7 @@ __le16 ieee80211_ctstoself_duration(struct ieee80211_hw *hw,
 	/* Data frame duration */
 	dur = ieee80211_frame_duration(local, frame_len, rate->bitrate,
 				       erp, short_preamble);
-	if (!(frame_txctl->flags & IEEE80211_TXCTL_NO_ACK)) {
+	if (!(frame_txctl->flags & IEEE80211_TX_CTL_NO_ACK)) {
 		/* ACK duration */
 		dur += ieee80211_frame_duration(local, 10, rate->bitrate,
 						erp, short_preamble);
diff --git a/net/mac80211/wep.c b/net/mac80211/wep.c
index 3cbae42ec504..1e7f03dd8f6b 100644
--- a/net/mac80211/wep.c
+++ b/net/mac80211/wep.c
@@ -333,11 +333,16 @@ ieee80211_crypto_wep_decrypt(struct ieee80211_rx_data *rx)
 
 static int wep_encrypt_skb(struct ieee80211_tx_data *tx, struct sk_buff *skb)
 {
+	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
+
+	info->control.iv_len = WEP_IV_LEN;
+	info->control.icv_len = WEP_ICV_LEN;
+
 	if (!(tx->key->flags & KEY_FLAG_UPLOADED_TO_HARDWARE)) {
+		info->control.hw_key = &tx->key->conf;
 		if (ieee80211_wep_encrypt(tx->local, skb, tx->key))
 			return -1;
 	} else {
-		tx->control->hw_key = &tx->key->conf;
 		if (tx->key->conf.flags & IEEE80211_KEY_FLAG_GENERATE_IV) {
 			if (!ieee80211_wep_add_iv(tx->local, skb, tx->key))
 				return -1;
@@ -349,8 +354,6 @@ static int wep_encrypt_skb(struct ieee80211_tx_data *tx, struct sk_buff *skb)
 ieee80211_tx_result
 ieee80211_crypto_wep_encrypt(struct ieee80211_tx_data *tx)
 {
-	tx->control->iv_len = WEP_IV_LEN;
-	tx->control->icv_len = WEP_ICV_LEN;
 	ieee80211_tx_set_protected(tx);
 
 	if (wep_encrypt_skb(tx, tx->skb) < 0) {
diff --git a/net/mac80211/wme.c b/net/mac80211/wme.c
index c87baf4ce979..477690f4dca7 100644
--- a/net/mac80211/wme.c
+++ b/net/mac80211/wme.c
@@ -149,8 +149,7 @@ static int wme_qdiscop_enqueue(struct sk_buff *skb, struct Qdisc* qd)
 	struct ieee80211_local *local = wdev_priv(qd->dev->ieee80211_ptr);
 	struct ieee80211_hw *hw = &local->hw;
 	struct ieee80211_sched_data *q = qdisc_priv(qd);
-	struct ieee80211_tx_packet_data *pkt_data =
-		(struct ieee80211_tx_packet_data *) skb->cb;
+	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
 	struct ieee80211_hdr *hdr = (struct ieee80211_hdr *) skb->data;
 	unsigned short fc = le16_to_cpu(hdr->frame_control);
 	struct Qdisc *qdisc;
@@ -158,8 +157,8 @@ static int wme_qdiscop_enqueue(struct sk_buff *skb, struct Qdisc* qd)
 	int err, queue;
 	u8 tid;
 
-	if (pkt_data->flags & IEEE80211_TXPD_REQUEUE) {
-		queue = pkt_data->queue;
+	if (info->flags & IEEE80211_TX_CTL_REQUEUE) {
+		queue = info->queue;
 		rcu_read_lock();
 		sta = sta_info_get(local, hdr->addr1);
 		tid = skb->priority & QOS_CONTROL_TAG1D_MASK;
@@ -168,9 +167,9 @@ static int wme_qdiscop_enqueue(struct sk_buff *skb, struct Qdisc* qd)
 			if ((ampdu_queue < QD_NUM(hw)) &&
 			    test_bit(ampdu_queue, q->qdisc_pool)) {
 				queue = ampdu_queue;
-				pkt_data->flags |= IEEE80211_TXPD_AMPDU;
+				info->flags |= IEEE80211_TX_CTL_AMPDU;
 			} else {
-				pkt_data->flags &= ~IEEE80211_TXPD_AMPDU;
+				info->flags &= ~IEEE80211_TX_CTL_AMPDU;
 			}
 		}
 		rcu_read_unlock();
@@ -206,9 +205,9 @@ static int wme_qdiscop_enqueue(struct sk_buff *skb, struct Qdisc* qd)
 			if ((ampdu_queue < QD_NUM(hw)) &&
 			    test_bit(ampdu_queue, q->qdisc_pool)) {
 				queue = ampdu_queue;
-				pkt_data->flags |= IEEE80211_TXPD_AMPDU;
+				info->flags |= IEEE80211_TX_CTL_AMPDU;
 			} else {
-				pkt_data->flags &= ~IEEE80211_TXPD_AMPDU;
+				info->flags &= ~IEEE80211_TX_CTL_AMPDU;
 			}
 		}
 
@@ -220,7 +219,7 @@ static int wme_qdiscop_enqueue(struct sk_buff *skb, struct Qdisc* qd)
 			err = NET_XMIT_DROP;
 	} else {
 		tid = skb->priority & QOS_CONTROL_TAG1D_MASK;
-		pkt_data->queue = (unsigned int) queue;
+		info->queue = (unsigned int) queue;
 		qdisc = q->queues[queue];
 		err = qdisc->enqueue(skb, qdisc);
 		if (err == NET_XMIT_SUCCESS) {
@@ -241,13 +240,12 @@ static int wme_qdiscop_enqueue(struct sk_buff *skb, struct Qdisc* qd)
 static int wme_qdiscop_requeue(struct sk_buff *skb, struct Qdisc* qd)
 {
 	struct ieee80211_sched_data *q = qdisc_priv(qd);
-	struct ieee80211_tx_packet_data *pkt_data =
-		(struct ieee80211_tx_packet_data *) skb->cb;
+	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
 	struct Qdisc *qdisc;
 	int err;
 
 	/* we recorded which queue to use earlier! */
-	qdisc = q->queues[pkt_data->queue];
+	qdisc = q->queues[info->queue];
 
 	if ((err = qdisc->ops->requeue(skb, qdisc)) == 0) {
 		qd->q.qlen++;
diff --git a/net/mac80211/wpa.c b/net/mac80211/wpa.c
index d7304490d2ec..d6635f6e5618 100644
--- a/net/mac80211/wpa.c
+++ b/net/mac80211/wpa.c
@@ -183,15 +183,25 @@ ieee80211_rx_h_michael_mic_verify(struct ieee80211_rx_data *rx)
 }
 
 
-static int tkip_encrypt_skb(struct ieee80211_tx_data *tx,
-			    struct sk_buff *skb, int test)
+static int tkip_encrypt_skb(struct ieee80211_tx_data *tx, struct sk_buff *skb)
 {
 	struct ieee80211_hdr *hdr = (struct ieee80211_hdr *) skb->data;
 	struct ieee80211_key *key = tx->key;
+	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
 	int hdrlen, len, tailneed;
 	u16 fc;
 	u8 *pos;
 
+	info->control.icv_len = TKIP_ICV_LEN;
+	info->control.iv_len = TKIP_IV_LEN;
+
+	if ((tx->key->flags & KEY_FLAG_UPLOADED_TO_HARDWARE) &&
+	    !(tx->key->conf.flags & IEEE80211_KEY_FLAG_GENERATE_IV)) {
+		/* hwaccel - with no need for preallocated room for IV/ICV */
+		info->control.hw_key = &tx->key->conf;
+		return TX_CONTINUE;
+	}
+
 	fc = le16_to_cpu(hdr->frame_control);
 	hdrlen = ieee80211_get_hdrlen(fc);
 	len = skb->len - hdrlen;
@@ -228,7 +238,7 @@ static int tkip_encrypt_skb(struct ieee80211_tx_data *tx,
 					    0x7f),
 				      (u8) key->u.tkip.tx.iv16);
 
-		tx->control->hw_key = &tx->key->conf;
+		info->control.hw_key = &tx->key->conf;
 		return 0;
 	}
 
@@ -246,28 +256,16 @@ ieee80211_tx_result
 ieee80211_crypto_tkip_encrypt(struct ieee80211_tx_data *tx)
 {
 	struct sk_buff *skb = tx->skb;
-	int wpa_test = 0, test = 0;
 
-	tx->control->icv_len = TKIP_ICV_LEN;
-	tx->control->iv_len = TKIP_IV_LEN;
 	ieee80211_tx_set_protected(tx);
 
-	if ((tx->key->flags & KEY_FLAG_UPLOADED_TO_HARDWARE) &&
-	    !(tx->key->conf.flags & IEEE80211_KEY_FLAG_GENERATE_IV) &&
-	    !wpa_test) {
-		/* hwaccel - with no need for preallocated room for IV/ICV */
-		tx->control->hw_key = &tx->key->conf;
-		return TX_CONTINUE;
-	}
-
-	if (tkip_encrypt_skb(tx, skb, test) < 0)
+	if (tkip_encrypt_skb(tx, skb) < 0)
 		return TX_DROP;
 
 	if (tx->extra_frag) {
 		int i;
 		for (i = 0; i < tx->num_extra_frag; i++) {
-			if (tkip_encrypt_skb(tx, tx->extra_frag[i], test)
-			    < 0)
+			if (tkip_encrypt_skb(tx, tx->extra_frag[i]) < 0)
 				return TX_DROP;
 		}
 	}
@@ -429,16 +427,27 @@ static inline int ccmp_hdr2pn(u8 *pn, u8 *hdr)
 }
 
 
-static int ccmp_encrypt_skb(struct ieee80211_tx_data *tx,
-			    struct sk_buff *skb, int test)
+static int ccmp_encrypt_skb(struct ieee80211_tx_data *tx, struct sk_buff *skb)
 {
 	struct ieee80211_hdr *hdr = (struct ieee80211_hdr *) skb->data;
 	struct ieee80211_key *key = tx->key;
+	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
 	int hdrlen, len, tailneed;
 	u16 fc;
 	u8 *pos, *pn, *b_0, *aad, *scratch;
 	int i;
 
+	info->control.icv_len = CCMP_MIC_LEN;
+	info->control.iv_len = CCMP_HDR_LEN;
+
+	if ((tx->key->flags & KEY_FLAG_UPLOADED_TO_HARDWARE) &&
+	    !(tx->key->conf.flags & IEEE80211_KEY_FLAG_GENERATE_IV)) {
+		/* hwaccel - with no need for preallocated room for CCMP "
+		 * header or MIC fields */
+		info->control.hw_key = &tx->key->conf;
+		return TX_CONTINUE;
+	}
+
 	scratch = key->u.ccmp.tx_crypto_buf;
 	b_0 = scratch + 3 * AES_BLOCK_LEN;
 	aad = scratch + 4 * AES_BLOCK_LEN;
@@ -478,7 +487,7 @@ static int ccmp_encrypt_skb(struct ieee80211_tx_data *tx,
 
 	if (key->flags & KEY_FLAG_UPLOADED_TO_HARDWARE) {
 		/* hwaccel - with preallocated room for CCMP header */
-		tx->control->hw_key = &tx->key->conf;
+		info->control.hw_key = &tx->key->conf;
 		return 0;
 	}
 
@@ -495,28 +504,16 @@ ieee80211_tx_result
 ieee80211_crypto_ccmp_encrypt(struct ieee80211_tx_data *tx)
 {
 	struct sk_buff *skb = tx->skb;
-	int test = 0;
 
-	tx->control->icv_len = CCMP_MIC_LEN;
-	tx->control->iv_len = CCMP_HDR_LEN;
 	ieee80211_tx_set_protected(tx);
 
-	if ((tx->key->flags & KEY_FLAG_UPLOADED_TO_HARDWARE) &&
-	    !(tx->key->conf.flags & IEEE80211_KEY_FLAG_GENERATE_IV)) {
-		/* hwaccel - with no need for preallocated room for CCMP "
-		 * header or MIC fields */
-		tx->control->hw_key = &tx->key->conf;
-		return TX_CONTINUE;
-	}
-
-	if (ccmp_encrypt_skb(tx, skb, test) < 0)
+	if (ccmp_encrypt_skb(tx, skb) < 0)
 		return TX_DROP;
 
 	if (tx->extra_frag) {
 		int i;
 		for (i = 0; i < tx->num_extra_frag; i++) {
-			if (ccmp_encrypt_skb(tx, tx->extra_frag[i], test)
-			    < 0)
+			if (ccmp_encrypt_skb(tx, tx->extra_frag[i]) < 0)
 				return TX_DROP;
 		}
 	}
-- 
cgit v1.2.3


From eefce91a384a64c7bbf913eb08c4adfb911c3639 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes@sipsolutions.net>
Date: Sat, 17 May 2008 00:57:13 +0200
Subject: mac80211: dont allow fragmentation and requeuing on A-MPDU queues

There really is no reason for a driver to reject a frame on
an A-MPDU queue when it can stop that queue for any period
of time and is given frames one by one. Hence, disallow it
with a big warning and reduce mac80211-internal state.

Also add a warning when we try to fragment a frame destined
for an A-MPDU queue and drop it, the actual bug needs to be
fixed elsewhere but I'm not exactly sure how to yet.

Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
Cc: Ron Rindjunsky <ron.rindjunsky@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/net/mac80211.h     |  6 ++++--
 net/mac80211/ieee80211_i.h |  2 +-
 net/mac80211/tx.c          | 23 +++++++++++++++++++++--
 3 files changed, 26 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index 54960b83db79..4df39eb9115f 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -986,8 +986,10 @@ enum ieee80211_ampdu_mlme_action {
  * @tx: Handler that 802.11 module calls for each transmitted frame.
  *	skb contains the buffer starting from the IEEE 802.11 header.
  *	The low-level driver should send the frame out based on
- *	configuration in the TX control data. Must be implemented and
- *	atomic.
+ *	configuration in the TX control data. This handler should,
+ *	preferably, never fail and stop queues appropriately, more
+ *	importantly, however, it must never fail for A-MPDU-queues.
+ *	Must be implemented and atomic.
  *
  * @start: Called before the first netdevice attached to the hardware
  *	is enabled. This should turn on the hardware and must turn on
diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h
index 79a65b3ee02b..86a861251e8c 100644
--- a/net/mac80211/ieee80211_i.h
+++ b/net/mac80211/ieee80211_i.h
@@ -595,7 +595,7 @@ struct ieee80211_local {
 	struct timer_list sta_cleanup;
 
 	unsigned long state[IEEE80211_MAX_QUEUES + IEEE80211_MAX_AMPDU_QUEUES];
-	struct ieee80211_tx_stored_packet pending_packet[IEEE80211_MAX_QUEUES + IEEE80211_MAX_AMPDU_QUEUES];
+	struct ieee80211_tx_stored_packet pending_packet[IEEE80211_MAX_QUEUES];
 	struct tasklet_struct tx_pending_tasklet;
 
 	/* number of interfaces with corresponding IFF_ flags */
diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c
index ac9a4af7ad42..6268bbca148e 100644
--- a/net/mac80211/tx.c
+++ b/net/mac80211/tx.c
@@ -673,6 +673,16 @@ ieee80211_tx_h_fragment(struct ieee80211_tx_data *tx)
 	if (!(tx->flags & IEEE80211_TX_FRAGMENTED))
 		return TX_CONTINUE;
 
+	/*
+	 * Warn when submitting a fragmented A-MPDU frame and drop it.
+	 * This is an error and needs to be fixed elsewhere, but when
+	 * done needs to take care of monitor interfaces (injection)
+	 * etc.
+	 */
+	if (WARN_ON(tx->flags & IEEE80211_TX_CTL_AMPDU ||
+		    IEEE80211_SKB_CB(tx->skb)->queue >= tx->local->hw.queues))
+		return TX_DROP;
+
 	first = tx->skb;
 
 	hdrlen = ieee80211_get_hdrlen(tx->fc);
@@ -1216,8 +1226,17 @@ static int ieee80211_tx(struct net_device *dev, struct sk_buff *skb)
 retry:
 	ret = __ieee80211_tx(local, skb, &tx);
 	if (ret) {
-		struct ieee80211_tx_stored_packet *store =
-			&local->pending_packet[info->queue];
+		struct ieee80211_tx_stored_packet *store;
+
+		/*
+		 * Since there are no fragmented frames on A-MPDU
+		 * queues, there's no reason for a driver to reject
+		 * a frame there, warn and drop it.
+		 */
+		if (WARN_ON(queue >= local->hw.queues))
+			goto drop;
+
+		store = &local->pending_packet[queue];
 
 		if (ret == IEEE80211_TX_FRAG_AGAIN)
 			skb = NULL;
-- 
cgit v1.2.3


From e2530083609148a7835b54c431f6b8956407c1f6 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes@sipsolutions.net>
Date: Sat, 17 May 2008 00:57:14 +0200
Subject: mac80211: use multi-queue master netdevice

This patch updates mac80211 and drivers to be multi-queue aware and
use that instead of the internal queue mapping. Also does a number
of cleanups in various pieces of the code that fall out and reduces
internal mac80211 state size.

Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/net/wireless/ath5k/base.c           |  2 +-
 drivers/net/wireless/b43/dma.c              |  7 ++-
 drivers/net/wireless/b43/pio.c              |  8 +--
 drivers/net/wireless/b43legacy/dma.c        |  3 +-
 drivers/net/wireless/iwlwifi/iwl-tx.c       |  4 +-
 drivers/net/wireless/iwlwifi/iwl3945-base.c |  4 +-
 drivers/net/wireless/p54/p54common.c        |  8 +--
 drivers/net/wireless/rt2x00/rt2x00mac.c     | 10 +--
 drivers/net/wireless/rt2x00/rt2x00queue.h   | 13 ----
 drivers/net/wireless/rtl8180_dev.c          |  4 +-
 include/net/mac80211.h                      | 20 +++++-
 net/mac80211/Kconfig                        | 14 ++++-
 net/mac80211/Makefile                       |  2 +-
 net/mac80211/ieee80211_i.h                  | 16 +++--
 net/mac80211/iface.c                        |  2 +-
 net/mac80211/main.c                         | 29 ++++++---
 net/mac80211/sta_info.c                     |  2 +-
 net/mac80211/tx.c                           | 97 +++++++++++++----------------
 net/mac80211/util.c                         | 29 ++++-----
 net/mac80211/wme.c                          | 14 ++---
 net/mac80211/wme.h                          |  2 +-
 21 files changed, 158 insertions(+), 132 deletions(-)

(limited to 'include')

diff --git a/drivers/net/wireless/ath5k/base.c b/drivers/net/wireless/ath5k/base.c
index 7d97934265db..18e9422d26dd 100644
--- a/drivers/net/wireless/ath5k/base.c
+++ b/drivers/net/wireless/ath5k/base.c
@@ -2657,7 +2657,7 @@ ath5k_tx(struct ieee80211_hw *hw, struct sk_buff *skb)
 	if (list_empty(&sc->txbuf)) {
 		ATH5K_ERR(sc, "no further txbuf available, dropping packet\n");
 		spin_unlock_irqrestore(&sc->txbuflock, flags);
-		ieee80211_stop_queue(hw, info->queue);
+		ieee80211_stop_queue(hw, skb_get_queue_mapping(skb));
 		return -1;
 	}
 	bf = list_first_entry(&sc->txbuf, struct ath5k_buf, list);
diff --git a/drivers/net/wireless/b43/dma.c b/drivers/net/wireless/b43/dma.c
index aced9866d815..b4eadd908bea 100644
--- a/drivers/net/wireless/b43/dma.c
+++ b/drivers/net/wireless/b43/dma.c
@@ -1297,7 +1297,8 @@ int b43_dma_tx(struct b43_wldev *dev, struct sk_buff *skb)
 		hdr->frame_control |= cpu_to_le16(IEEE80211_FCTL_MOREDATA);
 	} else {
 		/* Decide by priority where to put this frame. */
-		ring = select_ring_by_priority(dev, info->queue);
+		ring = select_ring_by_priority(
+			dev, skb_get_queue_mapping(skb));
 	}
 
 	spin_lock_irqsave(&ring->lock, flags);
@@ -1315,7 +1316,7 @@ int b43_dma_tx(struct b43_wldev *dev, struct sk_buff *skb)
 	/* Assign the queue number to the ring (if not already done before)
 	 * so TX status handling can use it. The queue to ring mapping is
 	 * static, so we don't need to store it per frame. */
-	ring->queue_prio = info->queue;
+	ring->queue_prio = skb_get_queue_mapping(skb);
 
 	err = dma_tx_fragment(ring, skb);
 	if (unlikely(err == -ENOKEY)) {
@@ -1333,7 +1334,7 @@ int b43_dma_tx(struct b43_wldev *dev, struct sk_buff *skb)
 	if ((free_slots(ring) < SLOTS_PER_PACKET) ||
 	    should_inject_overflow(ring)) {
 		/* This TX ring is full. */
-		ieee80211_stop_queue(dev->wl->hw, info->queue);
+		ieee80211_stop_queue(dev->wl->hw, skb_get_queue_mapping(skb));
 		ring->stopped = 1;
 		if (b43_debug(dev, B43_DBG_DMAVERBOSE)) {
 			b43dbg(dev->wl, "Stopped TX ring %d\n", ring->index);
diff --git a/drivers/net/wireless/b43/pio.c b/drivers/net/wireless/b43/pio.c
index 284786a94e7d..8b1555d95f1c 100644
--- a/drivers/net/wireless/b43/pio.c
+++ b/drivers/net/wireless/b43/pio.c
@@ -509,7 +509,7 @@ int b43_pio_tx(struct b43_wldev *dev, struct sk_buff *skb)
 		hdr->frame_control |= cpu_to_le16(IEEE80211_FCTL_MOREDATA);
 	} else {
 		/* Decide by priority where to put this frame. */
-		q = select_queue_by_priority(dev, info->queue);
+		q = select_queue_by_priority(dev, skb_get_queue_mapping(skb));
 	}
 
 	spin_lock_irqsave(&q->lock, flags);
@@ -532,7 +532,7 @@ int b43_pio_tx(struct b43_wldev *dev, struct sk_buff *skb)
 	if (total_len > (q->buffer_size - q->buffer_used)) {
 		/* Not enough memory on the queue. */
 		err = -EBUSY;
-		ieee80211_stop_queue(dev->wl->hw, info->queue);
+		ieee80211_stop_queue(dev->wl->hw, skb_get_queue_mapping(skb));
 		q->stopped = 1;
 		goto out_unlock;
 	}
@@ -540,7 +540,7 @@ int b43_pio_tx(struct b43_wldev *dev, struct sk_buff *skb)
 	/* Assign the queue number to the ring (if not already done before)
 	 * so TX status handling can use it. The mac80211-queue to b43-queue
 	 * mapping is static, so we don't need to store it per frame. */
-	q->queue_prio = info->queue;
+	q->queue_prio = skb_get_queue_mapping(skb);
 
 	err = pio_tx_frame(q, skb);
 	if (unlikely(err == -ENOKEY)) {
@@ -560,7 +560,7 @@ int b43_pio_tx(struct b43_wldev *dev, struct sk_buff *skb)
 	if (((q->buffer_size - q->buffer_used) < roundup(2 + 2 + 6, 4)) ||
 	    (q->free_packet_slots == 0)) {
 		/* The queue is full. */
-		ieee80211_stop_queue(dev->wl->hw, info->queue);
+		ieee80211_stop_queue(dev->wl->hw, skb_get_queue_mapping(skb));
 		q->stopped = 1;
 	}
 
diff --git a/drivers/net/wireless/b43legacy/dma.c b/drivers/net/wireless/b43legacy/dma.c
index c1c501d963bc..33cc256c5baf 100644
--- a/drivers/net/wireless/b43legacy/dma.c
+++ b/drivers/net/wireless/b43legacy/dma.c
@@ -1325,11 +1325,10 @@ int b43legacy_dma_tx(struct b43legacy_wldev *dev,
 		     struct sk_buff *skb)
 {
 	struct b43legacy_dmaring *ring;
-	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
 	int err = 0;
 	unsigned long flags;
 
-	ring = priority_to_txring(dev, info->queue);
+	ring = priority_to_txring(dev, skb_get_queue_mapping(skb));
 	spin_lock_irqsave(&ring->lock, flags);
 	B43legacy_WARN_ON(!ring->tx);
 	if (unlikely(free_slots(ring) < SLOTS_PER_PACKET)) {
diff --git a/drivers/net/wireless/iwlwifi/iwl-tx.c b/drivers/net/wireless/iwlwifi/iwl-tx.c
index a61293ba3f6b..0ebab967d5e7 100644
--- a/drivers/net/wireless/iwlwifi/iwl-tx.c
+++ b/drivers/net/wireless/iwlwifi/iwl-tx.c
@@ -696,7 +696,7 @@ int iwl_tx_skb(struct iwl_priv *priv, struct sk_buff *skb)
 	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
 	struct iwl_tfd_frame *tfd;
 	u32 *control_flags;
-	int txq_id = info->queue;
+	int txq_id = skb_get_queue_mapping(skb);
 	struct iwl_tx_queue *txq = NULL;
 	struct iwl_queue *q = NULL;
 	dma_addr_t phys_addr;
@@ -917,7 +917,7 @@ int iwl_tx_skb(struct iwl_priv *priv, struct sk_buff *skb)
 			spin_unlock_irqrestore(&priv->lock, flags);
 		}
 
-		ieee80211_stop_queue(priv->hw, info->queue);
+		ieee80211_stop_queue(priv->hw, skb_get_queue_mapping(skb));
 	}
 
 	return 0;
diff --git a/drivers/net/wireless/iwlwifi/iwl3945-base.c b/drivers/net/wireless/iwlwifi/iwl3945-base.c
index a740a1817d16..e0f52e264418 100644
--- a/drivers/net/wireless/iwlwifi/iwl3945-base.c
+++ b/drivers/net/wireless/iwlwifi/iwl3945-base.c
@@ -2552,7 +2552,7 @@ static int iwl3945_tx_skb(struct iwl3945_priv *priv, struct sk_buff *skb)
 	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
 	struct iwl3945_tfd_frame *tfd;
 	u32 *control_flags;
-	int txq_id = info->queue;
+	int txq_id = skb_get_queue_mapping(skb);
 	struct iwl3945_tx_queue *txq = NULL;
 	struct iwl3945_queue *q = NULL;
 	dma_addr_t phys_addr;
@@ -2765,7 +2765,7 @@ static int iwl3945_tx_skb(struct iwl3945_priv *priv, struct sk_buff *skb)
 			spin_unlock_irqrestore(&priv->lock, flags);
 		}
 
-		ieee80211_stop_queue(priv->hw, info->queue);
+		ieee80211_stop_queue(priv->hw, skb_get_queue_mapping(skb));
 	}
 
 	return 0;
diff --git a/drivers/net/wireless/p54/p54common.c b/drivers/net/wireless/p54/p54common.c
index 850857932e29..91ac9208b77d 100644
--- a/drivers/net/wireless/p54/p54common.c
+++ b/drivers/net/wireless/p54/p54common.c
@@ -407,7 +407,7 @@ static void p54_rx_frame_sent(struct ieee80211_hw *dev, struct sk_buff *skb)
 			last_addr = range->end_addr;
 			__skb_unlink(entry, &priv->tx_queue);
 			memset(&info->status, 0, sizeof(info->status));
-			priv->tx_stats[info->queue].len--;
+			priv->tx_stats[skb_get_queue_mapping(skb)].len--;
 			entry_hdr = (struct p54_control_hdr *) entry->data;
 			entry_data = (struct p54_tx_control_allocdata *) entry_hdr->data;
 			if ((entry_hdr->magic1 & cpu_to_le16(0x4000)) != 0)
@@ -551,13 +551,13 @@ static int p54_tx(struct ieee80211_hw *dev, struct sk_buff *skb)
 	size_t padding, len;
 	u8 rate;
 
-	current_queue = &priv->tx_stats[info->queue];
+	current_queue = &priv->tx_stats[skb_get_queue_mapping(skb)];
 	if (unlikely(current_queue->len > current_queue->limit))
 		return NETDEV_TX_BUSY;
 	current_queue->len++;
 	current_queue->count++;
 	if (current_queue->len == current_queue->limit)
-		ieee80211_stop_queue(dev, info->queue);
+		ieee80211_stop_queue(dev, skb_get_queue_mapping(skb));
 
 	padding = (unsigned long)(skb->data - (sizeof(*hdr) + sizeof(*txhdr))) & 3;
 	len = skb->len;
@@ -589,7 +589,7 @@ static int p54_tx(struct ieee80211_hw *dev, struct sk_buff *skb)
 	memset(txhdr->rateset, rate, 8);
 	txhdr->wep_key_present = 0;
 	txhdr->wep_key_len = 0;
-	txhdr->frame_type = cpu_to_le32(info->queue + 4);
+	txhdr->frame_type = cpu_to_le32(skb_get_queue_mapping(skb) + 4);
 	txhdr->magic4 = 0;
 	txhdr->antenna = (info->antenna_sel_tx == 0) ?
 		2 : info->antenna_sel_tx - 1;
diff --git a/drivers/net/wireless/rt2x00/rt2x00mac.c b/drivers/net/wireless/rt2x00/rt2x00mac.c
index b5379b027b18..c05e05b58887 100644
--- a/drivers/net/wireless/rt2x00/rt2x00mac.c
+++ b/drivers/net/wireless/rt2x00/rt2x00mac.c
@@ -102,7 +102,7 @@ int rt2x00mac_tx(struct ieee80211_hw *hw, struct sk_buff *skb)
 	struct rt2x00_dev *rt2x00dev = hw->priv;
 	struct ieee80211_tx_info *tx_info = IEEE80211_SKB_CB(skb);
 	struct ieee80211_hdr *ieee80211hdr = (struct ieee80211_hdr *)skb->data;
-	enum data_queue_qid qid = mac80211_queue_to_qid(tx_info->queue);
+	enum data_queue_qid qid = skb_get_queue_mapping(skb);
 	struct data_queue *queue;
 	u16 frame_control;
 
@@ -149,23 +149,23 @@ int rt2x00mac_tx(struct ieee80211_hw *hw, struct sk_buff *skb)
 			       IEEE80211_TX_CTL_USE_CTS_PROTECT)) &&
 	    !rt2x00dev->ops->hw->set_rts_threshold) {
 		if (rt2x00queue_available(queue) <= 1) {
-			ieee80211_stop_queue(rt2x00dev->hw, tx_info->queue);
+			ieee80211_stop_queue(rt2x00dev->hw, qid);
 			return NETDEV_TX_BUSY;
 		}
 
 		if (rt2x00mac_tx_rts_cts(rt2x00dev, queue, skb)) {
-			ieee80211_stop_queue(rt2x00dev->hw, tx_info->queue);
+			ieee80211_stop_queue(rt2x00dev->hw, qid);
 			return NETDEV_TX_BUSY;
 		}
 	}
 
 	if (rt2x00dev->ops->lib->write_tx_data(rt2x00dev, queue, skb)) {
-		ieee80211_stop_queue(rt2x00dev->hw, tx_info->queue);
+		ieee80211_stop_queue(rt2x00dev->hw, qid);
 		return NETDEV_TX_BUSY;
 	}
 
 	if (rt2x00queue_full(queue))
-		ieee80211_stop_queue(rt2x00dev->hw, tx_info->queue);
+		ieee80211_stop_queue(rt2x00dev->hw, qid);
 
 	if (rt2x00dev->ops->lib->kick_tx_queue)
 		rt2x00dev->ops->lib->kick_tx_queue(rt2x00dev, qid);
diff --git a/drivers/net/wireless/rt2x00/rt2x00queue.h b/drivers/net/wireless/rt2x00/rt2x00queue.h
index f263fe422f87..4d00ced14cc7 100644
--- a/drivers/net/wireless/rt2x00/rt2x00queue.h
+++ b/drivers/net/wireless/rt2x00/rt2x00queue.h
@@ -79,19 +79,6 @@ enum data_queue_qid {
 	QID_ATIM,
 };
 
-/**
- * mac80211_queue_to_qid - Convert mac80211 queue to rt2x00 qid
- * @queue: mac80211 queue.
- */
-static inline enum data_queue_qid mac80211_queue_to_qid(unsigned int queue)
-{
-	/* Regular TX queues are mapped directly */
-	if (queue < 4)
-		return queue;
-	WARN_ON(1);
-	return QID_OTHER;
-}
-
 /**
  * enum skb_frame_desc_flags: Flags for &struct skb_frame_desc
  *
diff --git a/drivers/net/wireless/rtl8180_dev.c b/drivers/net/wireless/rtl8180_dev.c
index 4427bc9f78a9..b7172a12c057 100644
--- a/drivers/net/wireless/rtl8180_dev.c
+++ b/drivers/net/wireless/rtl8180_dev.c
@@ -246,7 +246,7 @@ static int rtl8180_tx(struct ieee80211_hw *dev, struct sk_buff *skb)
 	u16 plcp_len = 0;
 	__le16 rts_duration = 0;
 
-	prio = info->queue;
+	prio = skb_get_queue_mapping(skb);
 	ring = &priv->tx_ring[prio];
 
 	mapping = pci_map_single(priv->pdev, skb->data,
@@ -298,7 +298,7 @@ static int rtl8180_tx(struct ieee80211_hw *dev, struct sk_buff *skb)
 	entry->flags = cpu_to_le32(tx_flags);
 	__skb_queue_tail(&ring->queue, skb);
 	if (ring->entries - skb_queue_len(&ring->queue) < 2)
-		ieee80211_stop_queue(dev, info->queue);
+		ieee80211_stop_queue(dev, skb_get_queue_mapping(skb));
 	spin_unlock_irqrestore(&priv->lock, flags);
 
 	rtl818x_iowrite8(priv, &priv->map->TX_DMA_POLLING, (1 << (prio + 4)));
diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index 4df39eb9115f..c80e3be8f79e 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -293,7 +293,7 @@ struct ieee80211_tx_info {
 	s8 tx_rate_idx;
 	u8 antenna_sel_tx;
 
-	u8 queue; /* use skb_queue_mapping soon */
+	/* 1 byte hole */
 
 	union {
 		struct {
@@ -802,6 +802,24 @@ static inline void SET_IEEE80211_PERM_ADDR(struct ieee80211_hw *hw, u8 *addr)
 	memcpy(hw->wiphy->perm_addr, addr, ETH_ALEN);
 }
 
+static inline int ieee80211_num_regular_queues(struct ieee80211_hw *hw)
+{
+#ifdef CONFIG_MAC80211_QOS
+	return hw->queues;
+#else
+	return 1;
+#endif
+}
+
+static inline int ieee80211_num_queues(struct ieee80211_hw *hw)
+{
+#ifdef CONFIG_MAC80211_QOS
+	return hw->queues + hw->ampdu_queues;
+#else
+	return 1;
+#endif
+}
+
 static inline struct ieee80211_rate *
 ieee80211_get_tx_rate(const struct ieee80211_hw *hw,
 		      const struct ieee80211_tx_info *c)
diff --git a/net/mac80211/Kconfig b/net/mac80211/Kconfig
index a24b459dd45a..590e00b2766c 100644
--- a/net/mac80211/Kconfig
+++ b/net/mac80211/Kconfig
@@ -7,11 +7,23 @@ config MAC80211
 	select CRC32
 	select WIRELESS_EXT
 	select CFG80211
-	select NET_SCH_FIFO
 	---help---
 	  This option enables the hardware independent IEEE 802.11
 	  networking stack.
 
+config MAC80211_QOS
+	def_bool y
+	depends on MAC80211
+	depends on NET_SCHED
+	depends on NETDEVICES_MULTIQUEUE
+
+comment "QoS/HT support disabled"
+	depends on MAC80211 && !MAC80211_QOS
+comment "QoS/HT support needs CONFIG_NET_SCHED"
+	depends on MAC80211 && !NET_SCHED
+comment "QoS/HT support needs CONFIG_NETDEVICES_MULTIQUEUE"
+	depends on MAC80211 && !NETDEVICES_MULTIQUEUE
+
 menu "Rate control algorithm selection"
 	depends on MAC80211 != n
 
diff --git a/net/mac80211/Makefile b/net/mac80211/Makefile
index 4e5847fd316c..1d2a4e010e5c 100644
--- a/net/mac80211/Makefile
+++ b/net/mac80211/Makefile
@@ -29,7 +29,7 @@ mac80211-y := \
 	event.o
 
 mac80211-$(CONFIG_MAC80211_LEDS) += led.o
-mac80211-$(CONFIG_NET_SCHED) += wme.o
+mac80211-$(CONFIG_MAC80211_QOS) += wme.o
 mac80211-$(CONFIG_MAC80211_DEBUGFS) += \
 	debugfs.o \
 	debugfs_sta.o \
diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h
index 86a861251e8c..7d614cdcefcb 100644
--- a/net/mac80211/ieee80211_i.h
+++ b/net/mac80211/ieee80211_i.h
@@ -594,7 +594,7 @@ struct ieee80211_local {
 	struct sta_info *sta_hash[STA_HASH_SIZE];
 	struct timer_list sta_cleanup;
 
-	unsigned long state[IEEE80211_MAX_QUEUES + IEEE80211_MAX_AMPDU_QUEUES];
+	unsigned long queues_pending[BITS_TO_LONGS(IEEE80211_MAX_QUEUES)];
 	struct ieee80211_tx_stored_packet pending_packet[IEEE80211_MAX_QUEUES];
 	struct tasklet_struct tx_pending_tasklet;
 
@@ -758,6 +758,15 @@ struct ieee80211_local {
 #endif
 };
 
+static inline int ieee80211_is_multiqueue(struct ieee80211_local *local)
+{
+#ifdef CONFIG_MAC80211_QOS
+	return netif_is_multiqueue(local->mdev);
+#else
+	return 0;
+#endif
+}
+
 /* this struct represents 802.11n's RA/TID combination */
 struct ieee80211_ra_tid {
 	u8 ra[ETH_ALEN];
@@ -827,11 +836,6 @@ static inline struct ieee80211_hw *local_to_hw(
 	return &local->hw;
 }
 
-enum ieee80211_link_state_t {
-	IEEE80211_LINK_STATE_XOFF = 0,
-	IEEE80211_LINK_STATE_PENDING,
-};
-
 struct sta_attribute {
 	struct attribute attr;
 	ssize_t (*show)(const struct sta_info *, char *buf);
diff --git a/net/mac80211/iface.c b/net/mac80211/iface.c
index 3c64e42eb12e..984472702381 100644
--- a/net/mac80211/iface.c
+++ b/net/mac80211/iface.c
@@ -168,7 +168,7 @@ void ieee80211_if_set_type(struct net_device *dev, int type)
 		ifsta->flags |= IEEE80211_STA_CREATE_IBSS |
 			IEEE80211_STA_AUTO_BSSID_SEL |
 			IEEE80211_STA_AUTO_CHANNEL_SEL;
-		if (sdata->local->hw.queues >= 4)
+		if (ieee80211_num_regular_queues(&sdata->local->hw) >= 4)
 			ifsta->flags |= IEEE80211_STA_WMM_ENABLED;
 
 		msdata = IEEE80211_DEV_TO_SUB_IF(sdata->local->mdev);
diff --git a/net/mac80211/main.c b/net/mac80211/main.c
index 8f1ff7ef1667..97d4a537ca2f 100644
--- a/net/mac80211/main.c
+++ b/net/mac80211/main.c
@@ -1634,12 +1634,32 @@ int ieee80211_register_hw(struct ieee80211_hw *hw)
 	if (result < 0)
 		return result;
 
+	/*
+	 * We use the number of queues for feature tests (QoS, HT) internally
+	 * so restrict them appropriately.
+	 */
+#ifdef CONFIG_MAC80211_QOS
+	if (hw->queues > IEEE80211_MAX_QUEUES)
+		hw->queues = IEEE80211_MAX_QUEUES;
+	if (hw->ampdu_queues > IEEE80211_MAX_AMPDU_QUEUES)
+		hw->ampdu_queues = IEEE80211_MAX_AMPDU_QUEUES;
+	if (hw->queues < 4)
+		hw->ampdu_queues = 0;
+#else
+	hw->queues = 1;
+	hw->ampdu_queues = 0;
+#endif
+
 	/* for now, mdev needs sub_if_data :/ */
-	mdev = alloc_netdev(sizeof(struct ieee80211_sub_if_data),
-			    "wmaster%d", ether_setup);
+	mdev = alloc_netdev_mq(sizeof(struct ieee80211_sub_if_data),
+			       "wmaster%d", ether_setup,
+			       ieee80211_num_queues(hw));
 	if (!mdev)
 		goto fail_mdev_alloc;
 
+	if (ieee80211_num_queues(hw) > 1)
+		mdev->features |= NETIF_F_MULTI_QUEUE;
+
 	sdata = IEEE80211_DEV_TO_SUB_IF(mdev);
 	mdev->ieee80211_ptr = &sdata->wdev;
 	sdata->wdev.wiphy = local->hw.wiphy;
@@ -1728,11 +1748,6 @@ int ieee80211_register_hw(struct ieee80211_hw *hw)
 		goto fail_wep;
 	}
 
-	if (hw->queues > IEEE80211_MAX_QUEUES)
-		hw->queues = IEEE80211_MAX_QUEUES;
-	if (hw->ampdu_queues > IEEE80211_MAX_AMPDU_QUEUES)
-		hw->ampdu_queues = IEEE80211_MAX_AMPDU_QUEUES;
-
 	ieee80211_install_qdisc(local->mdev);
 
 	/* add one default STA interface */
diff --git a/net/mac80211/sta_info.c b/net/mac80211/sta_info.c
index ef3149324d54..c24770cb02c5 100644
--- a/net/mac80211/sta_info.c
+++ b/net/mac80211/sta_info.c
@@ -255,7 +255,7 @@ struct sta_info *sta_info_alloc(struct ieee80211_sub_if_data *sdata,
 		 * sta_rx_agg_session_timer_expired for useage */
 		sta->timer_to_tid[i] = i;
 		/* tid to tx queue: initialize according to HW (0 is valid) */
-		sta->tid_to_tx_q[i] = local->hw.queues + local->hw.ampdu_queues;
+		sta->tid_to_tx_q[i] = ieee80211_num_queues(&local->hw);
 		/* rx */
 		sta->ampdu_mlme.tid_state_rx[i] = HT_AGG_STATE_IDLE;
 		sta->ampdu_mlme.tid_rx[i] = NULL;
diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c
index 6268bbca148e..9273651d3d7c 100644
--- a/net/mac80211/tx.c
+++ b/net/mac80211/tx.c
@@ -213,18 +213,6 @@ static u16 ieee80211_duration(struct ieee80211_tx_data *tx, int group_addr,
 	return dur;
 }
 
-static inline int __ieee80211_queue_stopped(const struct ieee80211_local *local,
-					    int queue)
-{
-	return test_bit(IEEE80211_LINK_STATE_XOFF, &local->state[queue]);
-}
-
-static inline int __ieee80211_queue_pending(const struct ieee80211_local *local,
-					    int queue)
-{
-	return test_bit(IEEE80211_LINK_STATE_PENDING, &local->state[queue]);
-}
-
 static int inline is_ieee80211_device(struct net_device *dev,
 				      struct net_device *master)
 {
@@ -680,7 +668,8 @@ ieee80211_tx_h_fragment(struct ieee80211_tx_data *tx)
 	 * etc.
 	 */
 	if (WARN_ON(tx->flags & IEEE80211_TX_CTL_AMPDU ||
-		    IEEE80211_SKB_CB(tx->skb)->queue >= tx->local->hw.queues))
+		    skb_get_queue_mapping(tx->skb) >=
+			ieee80211_num_regular_queues(&tx->local->hw)))
 		return TX_DROP;
 
 	first = tx->skb;
@@ -1098,11 +1087,9 @@ static int __ieee80211_tx(struct ieee80211_local *local, struct sk_buff *skb,
 	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
 	int ret, i;
 
-	if (!ieee80211_qdisc_installed(local->mdev) &&
-	    __ieee80211_queue_stopped(local, 0)) {
-		netif_stop_queue(local->mdev);
+	if (netif_subqueue_stopped(local->mdev, skb))
 		return IEEE80211_TX_AGAIN;
-	}
+
 	if (skb) {
 		ieee80211_dump_frame(wiphy_name(local->hw.wiphy),
 				     "TX to low-level driver", skb);
@@ -1121,7 +1108,8 @@ static int __ieee80211_tx(struct ieee80211_local *local, struct sk_buff *skb,
 					 IEEE80211_TX_CTL_USE_CTS_PROTECT |
 					 IEEE80211_TX_CTL_CLEAR_PS_FILT |
 					 IEEE80211_TX_CTL_FIRST_FRAGMENT);
-			if (__ieee80211_queue_stopped(local, info->queue))
+			if (netif_subqueue_stopped(local->mdev,
+						   tx->extra_frag[i]))
 				return IEEE80211_TX_FRAG_AGAIN;
 			if (i == tx->num_extra_frag) {
 				info->tx_rate_idx = tx->last_frag_rate_idx;
@@ -1160,9 +1148,11 @@ static int ieee80211_tx(struct net_device *dev, struct sk_buff *skb)
 	ieee80211_tx_result res = TX_DROP, res_prepare;
 	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
 	int ret, i;
-	int queue = info->queue;
+	u16 queue;
 
-	WARN_ON(__ieee80211_queue_pending(local, queue));
+	queue = skb_get_queue_mapping(skb);
+
+	WARN_ON(test_bit(queue, local->queues_pending));
 
 	if (unlikely(skb->len < 10)) {
 		dev_kfree_skb(skb);
@@ -1233,28 +1223,28 @@ retry:
 		 * queues, there's no reason for a driver to reject
 		 * a frame there, warn and drop it.
 		 */
-		if (WARN_ON(queue >= local->hw.queues))
+		if (WARN_ON(queue >= ieee80211_num_regular_queues(&local->hw)))
 			goto drop;
 
 		store = &local->pending_packet[queue];
 
 		if (ret == IEEE80211_TX_FRAG_AGAIN)
 			skb = NULL;
-		set_bit(IEEE80211_LINK_STATE_PENDING,
-			&local->state[queue]);
+		set_bit(queue, local->queues_pending);
 		smp_mb();
-		/* When the driver gets out of buffers during sending of
-		 * fragments and calls ieee80211_stop_queue, there is
-		 * a small window between IEEE80211_LINK_STATE_XOFF and
-		 * IEEE80211_LINK_STATE_PENDING flags are set. If a buffer
+		/*
+		 * When the driver gets out of buffers during sending of
+		 * fragments and calls ieee80211_stop_queue, the netif
+		 * subqueue is stopped. There is, however, a small window
+		 * in which the PENDING bit is not yet set. If a buffer
 		 * gets available in that window (i.e. driver calls
 		 * ieee80211_wake_queue), we would end up with ieee80211_tx
-		 * called with IEEE80211_LINK_STATE_PENDING. Prevent this by
+		 * called with the PENDING bit still set. Prevent this by
 		 * continuing transmitting here when that situation is
-		 * possible to have happened. */
-		if (!__ieee80211_queue_stopped(local, queue)) {
-			clear_bit(IEEE80211_LINK_STATE_PENDING,
-				  &local->state[queue]);
+		 * possible to have happened.
+		 */
+		if (!__netif_subqueue_stopped(local->mdev, queue)) {
+			clear_bit(queue, local->queues_pending);
 			goto retry;
 		}
 		store->skb = skb;
@@ -1509,7 +1499,8 @@ int ieee80211_subif_start_xmit(struct sk_buff *skb,
 	}
 
 	/* receiver and we are QoS enabled, use a QoS type frame */
-	if (sta_flags & WLAN_STA_WME && local->hw.queues >= 4) {
+	if (sta_flags & WLAN_STA_WME &&
+	    ieee80211_num_regular_queues(&local->hw) >= 4) {
 		fc |= IEEE80211_STYPE_QOS_DATA;
 		hdrlen += 2;
 	}
@@ -1661,41 +1652,51 @@ int ieee80211_subif_start_xmit(struct sk_buff *skb,
 	return ret;
 }
 
-/* helper functions for pending packets for when queues are stopped */
 
+/*
+ * ieee80211_clear_tx_pending may not be called in a context where
+ * it is possible that it packets could come in again.
+ */
 void ieee80211_clear_tx_pending(struct ieee80211_local *local)
 {
 	int i, j;
 	struct ieee80211_tx_stored_packet *store;
 
-	for (i = 0; i < local->hw.queues; i++) {
-		if (!__ieee80211_queue_pending(local, i))
+	for (i = 0; i < ieee80211_num_regular_queues(&local->hw); i++) {
+		if (!test_bit(i, local->queues_pending))
 			continue;
 		store = &local->pending_packet[i];
 		kfree_skb(store->skb);
 		for (j = 0; j < store->num_extra_frag; j++)
 			kfree_skb(store->extra_frag[j]);
 		kfree(store->extra_frag);
-		clear_bit(IEEE80211_LINK_STATE_PENDING, &local->state[i]);
+		clear_bit(i, local->queues_pending);
 	}
 }
 
+/*
+ * Transmit all pending packets. Called from tasklet, locks master device
+ * TX lock so that no new packets can come in.
+ */
 void ieee80211_tx_pending(unsigned long data)
 {
 	struct ieee80211_local *local = (struct ieee80211_local *)data;
 	struct net_device *dev = local->mdev;
 	struct ieee80211_tx_stored_packet *store;
 	struct ieee80211_tx_data tx;
-	int i, ret, reschedule = 0;
+	int i, ret;
 
 	netif_tx_lock_bh(dev);
-	for (i = 0; i < local->hw.queues; i++) {
-		if (__ieee80211_queue_stopped(local, i))
+	for (i = 0; i < ieee80211_num_regular_queues(&local->hw); i++) {
+		/* Check that this queue is ok */
+		if (__netif_subqueue_stopped(local->mdev, i))
 			continue;
-		if (!__ieee80211_queue_pending(local, i)) {
-			reschedule = 1;
+
+		if (!test_bit(i, local->queues_pending)) {
+			ieee80211_wake_queue(&local->hw, i);
 			continue;
 		}
+
 		store = &local->pending_packet[i];
 		tx.extra_frag = store->extra_frag;
 		tx.num_extra_frag = store->num_extra_frag;
@@ -1708,19 +1709,11 @@ void ieee80211_tx_pending(unsigned long data)
 			if (ret == IEEE80211_TX_FRAG_AGAIN)
 				store->skb = NULL;
 		} else {
-			clear_bit(IEEE80211_LINK_STATE_PENDING,
-				  &local->state[i]);
-			reschedule = 1;
+			clear_bit(i, local->queues_pending);
+			ieee80211_wake_queue(&local->hw, i);
 		}
 	}
 	netif_tx_unlock_bh(dev);
-	if (reschedule) {
-		if (!ieee80211_qdisc_installed(dev)) {
-			if (!__ieee80211_queue_stopped(local, 0))
-				netif_wake_queue(dev);
-		} else
-			netif_schedule(dev);
-	}
 }
 
 /* functions for drivers to get certain frames */
diff --git a/net/mac80211/util.c b/net/mac80211/util.c
index d9109dee461f..4f7180b287da 100644
--- a/net/mac80211/util.c
+++ b/net/mac80211/util.c
@@ -331,17 +331,15 @@ void ieee80211_wake_queue(struct ieee80211_hw *hw, int queue)
 {
 	struct ieee80211_local *local = hw_to_local(hw);
 
-	if (test_and_clear_bit(IEEE80211_LINK_STATE_XOFF,
-			       &local->state[queue])) {
-		if (test_bit(IEEE80211_LINK_STATE_PENDING,
-			     &local->state[queue]))
-			tasklet_schedule(&local->tx_pending_tasklet);
-		else
-			if (!ieee80211_qdisc_installed(local->mdev)) {
-				if (queue == 0)
-					netif_wake_queue(local->mdev);
-			} else
-				__netif_schedule(local->mdev);
+	if (test_bit(queue, local->queues_pending)) {
+		tasklet_schedule(&local->tx_pending_tasklet);
+	} else {
+		if (ieee80211_is_multiqueue(local)) {
+			netif_wake_subqueue(local->mdev, queue);
+		} else {
+			WARN_ON(queue != 0);
+			netif_wake_queue(local->mdev);
+		}
 	}
 }
 EXPORT_SYMBOL(ieee80211_wake_queue);
@@ -350,9 +348,12 @@ void ieee80211_stop_queue(struct ieee80211_hw *hw, int queue)
 {
 	struct ieee80211_local *local = hw_to_local(hw);
 
-	if (!ieee80211_qdisc_installed(local->mdev) && queue == 0)
+	if (ieee80211_is_multiqueue(local)) {
+		netif_stop_subqueue(local->mdev, queue);
+	} else {
+		WARN_ON(queue != 0);
 		netif_stop_queue(local->mdev);
-	set_bit(IEEE80211_LINK_STATE_XOFF, &local->state[queue]);
+	}
 }
 EXPORT_SYMBOL(ieee80211_stop_queue);
 
@@ -360,7 +361,7 @@ void ieee80211_stop_queues(struct ieee80211_hw *hw)
 {
 	int i;
 
-	for (i = 0; i < hw->queues + hw->ampdu_queues; i++)
+	for (i = 0; i < ieee80211_num_queues(hw); i++)
 		ieee80211_stop_queue(hw, i);
 }
 EXPORT_SYMBOL(ieee80211_stop_queues);
diff --git a/net/mac80211/wme.c b/net/mac80211/wme.c
index 477690f4dca7..14a9ff10a1e9 100644
--- a/net/mac80211/wme.c
+++ b/net/mac80211/wme.c
@@ -158,7 +158,7 @@ static int wme_qdiscop_enqueue(struct sk_buff *skb, struct Qdisc* qd)
 	u8 tid;
 
 	if (info->flags & IEEE80211_TX_CTL_REQUEUE) {
-		queue = info->queue;
+		queue = skb_get_queue_mapping(skb);
 		rcu_read_lock();
 		sta = sta_info_get(local, hdr->addr1);
 		tid = skb->priority & QOS_CONTROL_TAG1D_MASK;
@@ -219,7 +219,7 @@ static int wme_qdiscop_enqueue(struct sk_buff *skb, struct Qdisc* qd)
 			err = NET_XMIT_DROP;
 	} else {
 		tid = skb->priority & QOS_CONTROL_TAG1D_MASK;
-		info->queue = (unsigned int) queue;
+		skb_set_queue_mapping(skb, queue);
 		qdisc = q->queues[queue];
 		err = qdisc->enqueue(skb, qdisc);
 		if (err == NET_XMIT_SUCCESS) {
@@ -240,12 +240,11 @@ static int wme_qdiscop_enqueue(struct sk_buff *skb, struct Qdisc* qd)
 static int wme_qdiscop_requeue(struct sk_buff *skb, struct Qdisc* qd)
 {
 	struct ieee80211_sched_data *q = qdisc_priv(qd);
-	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
 	struct Qdisc *qdisc;
 	int err;
 
 	/* we recorded which queue to use earlier! */
-	qdisc = q->queues[info->queue];
+	qdisc = q->queues[skb_get_queue_mapping(skb)];
 
 	if ((err = qdisc->ops->requeue(skb, qdisc)) == 0) {
 		qd->q.qlen++;
@@ -269,11 +268,8 @@ static struct sk_buff *wme_qdiscop_dequeue(struct Qdisc* qd)
 	/* check all the h/w queues in numeric/priority order */
 	for (queue = 0; queue < QD_NUM(hw); queue++) {
 		/* see if there is room in this hardware queue */
-		if ((test_bit(IEEE80211_LINK_STATE_XOFF,
-				&local->state[queue])) ||
-		    (test_bit(IEEE80211_LINK_STATE_PENDING,
-				&local->state[queue])) ||
-			 (!test_bit(queue, q->qdisc_pool)))
+		if (__netif_subqueue_stopped(local->mdev, queue) ||
+		    !test_bit(queue, q->qdisc_pool))
 			continue;
 
 		/* there is space - try and get a frame */
diff --git a/net/mac80211/wme.h b/net/mac80211/wme.h
index fcc6b05508cc..bbdb53344817 100644
--- a/net/mac80211/wme.h
+++ b/net/mac80211/wme.h
@@ -31,7 +31,7 @@ static inline int WLAN_FC_IS_QOS_DATA(u16 fc)
 	return (fc & 0x8C) == 0x88;
 }
 
-#ifdef CONFIG_NET_SCHED
+#ifdef CONFIG_MAC80211_QOS
 void ieee80211_install_qdisc(struct net_device *dev);
 int ieee80211_qdisc_installed(struct net_device *dev);
 int ieee80211_ht_agg_queue_add(struct ieee80211_local *local,
-- 
cgit v1.2.3


From 57e6983cbde91b4569b4014b933f3a16e12b99fd Mon Sep 17 00:00:00 2001
From: Matt Carlson <mcarlson@broadcom.com>
Date: Sun, 25 May 2008 23:48:31 -0700
Subject: tg3: Add 5785 ASIC revision

This patch added the 5785 device ID and ASIC revision to the code.

Signed-off-by: Matt Carlson <mcarlson@broadcom.com>
Signed-off-by: Michael Chan <mchan@broadcom.com>
Signed-off-by: Benjamin Li <benli@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/tg3.c       | 58 ++++++++++++++++++++++++++++++++++++-------------
 drivers/net/tg3.h       |  1 +
 include/linux/pci_ids.h |  1 +
 3 files changed, 45 insertions(+), 15 deletions(-)

(limited to 'include')

diff --git a/drivers/net/tg3.c b/drivers/net/tg3.c
index 028276edd3bc..69ded90cd96e 100644
--- a/drivers/net/tg3.c
+++ b/drivers/net/tg3.c
@@ -204,6 +204,7 @@ static struct pci_device_id tg3_pci_tbl[] = {
 	{PCI_DEVICE(PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_TIGON3_5723)},
 	{PCI_DEVICE(PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_TIGON3_5761)},
 	{PCI_DEVICE(PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_TIGON3_5761E)},
+	{PCI_DEVICE(PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_TIGON3_5785)},
 	{PCI_DEVICE(PCI_VENDOR_ID_SYSKONNECT, PCI_DEVICE_ID_SYSKONNECT_9DXX)},
 	{PCI_DEVICE(PCI_VENDOR_ID_SYSKONNECT, PCI_DEVICE_ID_SYSKONNECT_9MXX)},
 	{PCI_DEVICE(PCI_VENDOR_ID_ALTIMA, PCI_DEVICE_ID_ALTIMA_AC1000)},
@@ -5710,7 +5711,8 @@ static int tg3_chip_reset(struct tg3 *tp)
 	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5755 ||
 	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5787 ||
 	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5784 ||
-	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5761)
+	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5761 ||
+	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5785)
 		tw32(GRC_FASTBOOT_PC, 0);
 
 	/*
@@ -7004,7 +7006,8 @@ static int tg3_reset_hw(struct tg3 *tp, int reset_phy)
 		return err;
 
 	if (GET_ASIC_REV(tp->pci_chip_rev_id) != ASIC_REV_5784 &&
-	    GET_ASIC_REV(tp->pci_chip_rev_id) != ASIC_REV_5761) {
+	    GET_ASIC_REV(tp->pci_chip_rev_id) != ASIC_REV_5761 &&
+	    GET_ASIC_REV(tp->pci_chip_rev_id) != ASIC_REV_5785) {
 		/* This value is determined during the probe time DMA
 		 * engine test, tg3_test_dma.
 		 */
@@ -7243,7 +7246,8 @@ static int tg3_reset_hw(struct tg3 *tp, int reset_phy)
 		      RDMAC_MODE_FIFOURUN_ENAB | RDMAC_MODE_FIFOOREAD_ENAB |
 		      RDMAC_MODE_LNGREAD_ENAB);
 
-	if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5784)
+	if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5784 ||
+	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5785)
 		rdmac_mode |= RDMAC_MODE_BD_SBD_CRPT_ENAB |
 			      RDMAC_MODE_MBUF_RBD_CRPT_ENAB |
 			      RDMAC_MODE_MBUF_SBD_CRPT_ENAB;
@@ -7411,7 +7415,8 @@ static int tg3_reset_hw(struct tg3 *tp, int reset_phy)
 	if ((GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5755) ||
 	    (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5787) ||
 	    (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5784) ||
-	    (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5761))
+	    (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5761) ||
+	    (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5785))
 		val |= WDMAC_MODE_STATUS_TAG_FIX;
 
 	tw32_f(WDMAC_MODE, val);
@@ -7473,7 +7478,9 @@ static int tg3_reset_hw(struct tg3 *tp, int reset_phy)
 
 	tp->rx_mode = RX_MODE_ENABLE;
 	if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5755 ||
-	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5761)
+	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5784 ||
+	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5761 ||
+	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5785)
 		tp->rx_mode |= RX_MODE_IPV6_CSUM_ENABLE;
 
 	tw32_f(MAC_RX_MODE, tp->rx_mode);
@@ -9028,7 +9035,10 @@ static int tg3_set_tso(struct net_device *dev, u32 value)
 	    (GET_ASIC_REV(tp->pci_chip_rev_id) != ASIC_REV_5906)) {
 		if (value) {
 			dev->features |= NETIF_F_TSO6;
-			if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5761)
+			if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5761 ||
+			    (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5784 &&
+			     GET_CHIP_REV(tp->pci_chip_rev_id) != CHIPREV_5784_AX) ||
+			    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5785)
 				dev->features |= NETIF_F_TSO_ECN;
 		} else
 			dev->features &= ~(NETIF_F_TSO6 | NETIF_F_TSO_ECN);
@@ -9286,7 +9296,8 @@ static int tg3_set_tx_csum(struct net_device *dev, u32 data)
 	if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5755 ||
 	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5787 ||
 	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5784 ||
-	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5761)
+	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5761 ||
+	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5785)
 		ethtool_op_set_tx_ipv6_csum(dev, data);
 	else
 		ethtool_op_set_tx_csum(dev, data);
@@ -9807,7 +9818,8 @@ static int tg3_test_memory(struct tg3 *tp)
 		if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5755 ||
 		    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5787 ||
 		    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5784 ||
-		    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5761)
+		    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5761 ||
+		    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5785)
 			mem_tbl = mem_tbl_5755;
 		else if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5906)
 			mem_tbl = mem_tbl_5906;
@@ -10014,7 +10026,8 @@ static int tg3_test_loopback(struct tg3 *tp)
 		return TG3_LOOPBACK_FAILED;
 
 	if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5784 ||
-	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5761) {
+	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5761 ||
+	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5785) {
 		int i;
 		u32 status;
 
@@ -10042,7 +10055,8 @@ static int tg3_test_loopback(struct tg3 *tp)
 		err |= TG3_MAC_LOOPBACK_FAILED;
 
 	if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5784 ||
-	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5761) {
+	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5761 ||
+	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5785) {
 		tw32(TG3_CPMU_CTRL, cpmuctrl);
 
 		/* Release the mutex */
@@ -10690,7 +10704,8 @@ static void __devinit tg3_nvram_init(struct tg3 *tp)
 		else if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5755)
 			tg3_get_5755_nvram_info(tp);
 		else if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5787 ||
-			 GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5784)
+			 GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5784 ||
+			 GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5785)
 			tg3_get_5787_nvram_info(tp);
 		else if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5761)
 			tg3_get_5761_nvram_info(tp);
@@ -11021,6 +11036,7 @@ static int tg3_nvram_write_block_buffered(struct tg3 *tp, u32 offset, u32 len,
 		    (GET_ASIC_REV(tp->pci_chip_rev_id) != ASIC_REV_5787) &&
 		    (GET_ASIC_REV(tp->pci_chip_rev_id) != ASIC_REV_5784) &&
 		    (GET_ASIC_REV(tp->pci_chip_rev_id) != ASIC_REV_5761) &&
+		    (GET_ASIC_REV(tp->pci_chip_rev_id) != ASIC_REV_5785) &&
 		    (tp->nvram_jedecnum == JEDEC_ST) &&
 		    (nvram_cmd & NVRAM_CMD_FIRST)) {
 
@@ -11924,6 +11940,7 @@ static int __devinit tg3_get_invariants(struct tg3 *tp)
 	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5787 ||
 	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5784 ||
 	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5761 ||
+	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5785 ||
 	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5906 ||
 	    (tp->tg3_flags2 & TG3_FLG2_5780_CLASS))
 		tp->tg3_flags2 |= TG3_FLG2_5750_PLUS;
@@ -11945,6 +11962,7 @@ static int __devinit tg3_get_invariants(struct tg3 *tp)
 		    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5787 ||
 		    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5784 ||
 		    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5761 ||
+		    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5785 ||
 		    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5906) {
 			tp->tg3_flags2 |= TG3_FLG2_HW_TSO_2;
 			tp->tg3_flags2 |= TG3_FLG2_1SHOT_MSI;
@@ -12147,7 +12165,8 @@ static int __devinit tg3_get_invariants(struct tg3 *tp)
 	}
 
 	if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5784 ||
-	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5761) {
+	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5761 ||
+	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5785) {
 		tp->tg3_flags |= TG3_FLAG_CPMU_PRESENT;
 
 		if (tp->pci_chip_rev_id == CHIPREV_ID_5784_A0 ||
@@ -12231,7 +12250,8 @@ static int __devinit tg3_get_invariants(struct tg3 *tp)
 				tp->tg3_flags2 |= TG3_FLG2_PHY_JITTER_BUG;
 			if (tp->pdev->device == PCI_DEVICE_ID_TIGON3_5755M)
 				tp->tg3_flags2 |= TG3_FLG2_PHY_ADJUST_TRIM;
-		} else if (GET_ASIC_REV(tp->pci_chip_rev_id) != ASIC_REV_5906)
+		} else if (GET_ASIC_REV(tp->pci_chip_rev_id) != ASIC_REV_5906 &&
+			   GET_ASIC_REV(tp->pci_chip_rev_id) != ASIC_REV_5785)
 			tp->tg3_flags2 |= TG3_FLG2_PHY_BER_BUG;
 	}
 
@@ -12252,6 +12272,9 @@ static int __devinit tg3_get_invariants(struct tg3 *tp)
 	    GET_CHIP_REV(tp->pci_chip_rev_id) != CHIPREV_5700_BX)
 		tp->coalesce_mode |= HOSTCC_MODE_32BYTE;
 
+	if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5785)
+		tp->tg3_flags3 |= TG3_FLG3_USE_PHYLIB;
+
 	err = tg3_mdio_init(tp);
 	if (err)
 		return err;
@@ -12383,6 +12406,7 @@ static int __devinit tg3_get_invariants(struct tg3 *tp)
 	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5787 ||
 	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5784 ||
 	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5761 ||
+	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5785 ||
 	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5906)
 		tp->dev->hard_start_xmit = tg3_start_xmit;
 	else
@@ -13307,7 +13331,10 @@ static int __devinit tg3_init_one(struct pci_dev *pdev,
 		if ((tp->tg3_flags2 & TG3_FLG2_HW_TSO_2) &&
 		    (GET_ASIC_REV(tp->pci_chip_rev_id) != ASIC_REV_5906))
 			dev->features |= NETIF_F_TSO6;
-		if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5761)
+		if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5761 ||
+		    (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5784 &&
+		     GET_CHIP_REV(tp->pci_chip_rev_id) != CHIPREV_5784_AX) ||
+			GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5785)
 			dev->features |= NETIF_F_TSO_ECN;
 	}
 
@@ -13373,7 +13400,8 @@ static int __devinit tg3_init_one(struct pci_dev *pdev,
 		if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5755 ||
 		    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5787 ||
 		    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5784 ||
-		    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5761)
+		    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5761 ||
+		    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5785)
 			dev->features |= NETIF_F_IPV6_CSUM;
 
 		tp->tg3_flags |= TG3_FLAG_RX_CHECKSUMS;
diff --git a/drivers/net/tg3.h b/drivers/net/tg3.h
index 48f45c17f60d..95ac4c9590a1 100644
--- a/drivers/net/tg3.h
+++ b/drivers/net/tg3.h
@@ -128,6 +128,7 @@
 #define   ASIC_REV_USE_PROD_ID_REG	 0x0f
 #define   ASIC_REV_5784			 0x5784
 #define   ASIC_REV_5761			 0x5761
+#define   ASIC_REV_5785			 0x5785
 #define  GET_CHIP_REV(CHIP_REV_ID)	((CHIP_REV_ID) >> 8)
 #define   CHIPREV_5700_AX		 0x70
 #define   CHIPREV_5700_BX		 0x71
diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h
index cf6dbd759395..72c038560e7d 100644
--- a/include/linux/pci_ids.h
+++ b/include/linux/pci_ids.h
@@ -1979,6 +1979,7 @@
 #define PCI_DEVICE_ID_TIGON3_5787M	0x1693
 #define PCI_DEVICE_ID_TIGON3_5782	0x1696
 #define PCI_DEVICE_ID_TIGON3_5784	0x1698
+#define PCI_DEVICE_ID_TIGON3_5785	0x1699
 #define PCI_DEVICE_ID_TIGON3_5786	0x169a
 #define PCI_DEVICE_ID_TIGON3_5787	0x169b
 #define PCI_DEVICE_ID_TIGON3_5788	0x169c
-- 
cgit v1.2.3


From a9daf36746b1fb5c2db8d164ca70c30c63a0d7b2 Mon Sep 17 00:00:00 2001
From: Matt Carlson <mcarlson@broadcom.com>
Date: Sun, 25 May 2008 23:49:44 -0700
Subject: tg3: Add shmem options.

This patch adds some options obtained through shared memory.

Signed-off-by: Matt Carlson <mcarlson@broadcom.com>
Signed-off-by: Michael Chan <mchan@broadcom.com>
Signed-off-by: Benjamin Li <benli@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/tg3.c       | 95 +++++++++++++++++++++++++++++++++++++++++++++----
 drivers/net/tg3.h       | 30 +++++++++++++++-
 include/linux/brcmphy.h |  6 ++++
 3 files changed, 123 insertions(+), 8 deletions(-)
 create mode 100644 include/linux/brcmphy.h

(limited to 'include')

diff --git a/drivers/net/tg3.c b/drivers/net/tg3.c
index 69ded90cd96e..365fcffc3144 100644
--- a/drivers/net/tg3.c
+++ b/drivers/net/tg3.c
@@ -33,6 +33,7 @@
 #include <linux/ethtool.h>
 #include <linux/mii.h>
 #include <linux/phy.h>
+#include <linux/brcmphy.h>
 #include <linux/if_vlan.h>
 #include <linux/ip.h>
 #include <linux/tcp.h>
@@ -869,6 +870,51 @@ static int tg3_mdio_reset(struct mii_bus *bp)
 	return 0;
 }
 
+static void tg3_mdio_config(struct tg3 *tp)
+{
+	u32 val;
+
+	if (tp->mdio_bus.phy_map[PHY_ADDR]->interface !=
+	    PHY_INTERFACE_MODE_RGMII)
+		return;
+
+	val = tr32(MAC_PHYCFG1) & ~(MAC_PHYCFG1_RGMII_EXT_RX_DEC |
+				    MAC_PHYCFG1_RGMII_SND_STAT_EN);
+	if (tp->tg3_flags3 & TG3_FLG3_RGMII_STD_IBND_DISABLE) {
+		if (tp->tg3_flags3 & TG3_FLG3_RGMII_EXT_IBND_RX_EN)
+			val |= MAC_PHYCFG1_RGMII_EXT_RX_DEC;
+		if (tp->tg3_flags3 & TG3_FLG3_RGMII_EXT_IBND_TX_EN)
+			val |= MAC_PHYCFG1_RGMII_SND_STAT_EN;
+	}
+	tw32(MAC_PHYCFG1, val | MAC_PHYCFG1_RGMII_INT | MAC_PHYCFG1_TXC_DRV);
+
+	val = tr32(MAC_PHYCFG2) & ~(MAC_PHYCFG2_INBAND_ENABLE);
+	if (!(tp->tg3_flags3 & TG3_FLG3_RGMII_STD_IBND_DISABLE))
+		val |= MAC_PHYCFG2_INBAND_ENABLE;
+	tw32(MAC_PHYCFG2, val);
+
+	val = tr32(MAC_EXT_RGMII_MODE);
+	val &= ~(MAC_RGMII_MODE_RX_INT_B |
+		 MAC_RGMII_MODE_RX_QUALITY |
+		 MAC_RGMII_MODE_RX_ACTIVITY |
+		 MAC_RGMII_MODE_RX_ENG_DET |
+		 MAC_RGMII_MODE_TX_ENABLE |
+		 MAC_RGMII_MODE_TX_LOWPWR |
+		 MAC_RGMII_MODE_TX_RESET);
+	if (tp->tg3_flags3 & TG3_FLG3_RGMII_STD_IBND_DISABLE) {
+		if (tp->tg3_flags3 & TG3_FLG3_RGMII_EXT_IBND_RX_EN)
+			val |= MAC_RGMII_MODE_RX_INT_B |
+			       MAC_RGMII_MODE_RX_QUALITY |
+			       MAC_RGMII_MODE_RX_ACTIVITY |
+			       MAC_RGMII_MODE_RX_ENG_DET;
+		if (tp->tg3_flags3 & TG3_FLG3_RGMII_EXT_IBND_TX_EN)
+			val |= MAC_RGMII_MODE_TX_ENABLE |
+			       MAC_RGMII_MODE_TX_LOWPWR |
+			       MAC_RGMII_MODE_TX_RESET;
+	}
+	tw32(MAC_EXT_RGMII_MODE, val);
+}
+
 static void tg3_mdio_start(struct tg3 *tp)
 {
 	if (tp->tg3_flags3 & TG3_FLG3_MDIOBUS_INITED) {
@@ -880,6 +926,9 @@ static void tg3_mdio_start(struct tg3 *tp)
 	tp->mi_mode &= ~MAC_MI_MODE_AUTO_POLL;
 	tw32_f(MAC_MI_MODE, tp->mi_mode);
 	udelay(80);
+
+	if (tp->tg3_flags3 & TG3_FLG3_MDIOBUS_INITED)
+		tg3_mdio_config(tp);
 }
 
 static void tg3_mdio_stop(struct tg3 *tp)
@@ -895,6 +944,7 @@ static int tg3_mdio_init(struct tg3 *tp)
 {
 	int i;
 	u32 reg;
+	struct phy_device *phydev;
 	struct mii_bus *mdio_bus = &tp->mdio_bus;
 
 	tg3_mdio_start(tp);
@@ -928,13 +978,34 @@ static int tg3_mdio_init(struct tg3 *tp)
 		tg3_bmcr_reset(tp);
 
 	i = mdiobus_register(mdio_bus);
-	if (!i)
-		tp->tg3_flags3 |= TG3_FLG3_MDIOBUS_INITED;
-	else
+	if (i) {
 		printk(KERN_WARNING "%s: mdiobus_reg failed (0x%x)\n",
 			tp->dev->name, i);
+		return i;
+	}
 
-	return i;
+	tp->tg3_flags3 |= TG3_FLG3_MDIOBUS_INITED;
+
+	phydev = tp->mdio_bus.phy_map[PHY_ADDR];
+
+	switch (phydev->phy_id) {
+	case TG3_PHY_ID_BCM50610:
+		phydev->interface = PHY_INTERFACE_MODE_RGMII;
+		if (tp->tg3_flags3 & TG3_FLG3_RGMII_STD_IBND_DISABLE)
+			phydev->dev_flags |= PHY_BRCM_STD_IBND_DISABLE;
+		if (tp->tg3_flags3 & TG3_FLG3_RGMII_EXT_IBND_RX_EN)
+			phydev->dev_flags |= PHY_BRCM_EXT_IBND_RX_ENABLE;
+		if (tp->tg3_flags3 & TG3_FLG3_RGMII_EXT_IBND_TX_EN)
+			phydev->dev_flags |= PHY_BRCM_EXT_IBND_TX_ENABLE;
+		break;
+	case TG3_PHY_ID_BCMAC131:
+		phydev->interface = PHY_INTERFACE_MODE_MII;
+		break;
+	}
+
+	tg3_mdio_config(tp);
+
+	return 0;
 }
 
 static void tg3_mdio_fini(struct tg3 *tp)
@@ -1238,8 +1309,8 @@ static int tg3_phy_init(struct tg3 *tp)
 	phydev = tp->mdio_bus.phy_map[PHY_ADDR];
 
 	/* Attach the MAC to the PHY. */
-	phydev = phy_connect(tp->dev, phydev->dev.bus_id,
-			     tg3_adjust_link, 0, phydev->interface);
+	phydev = phy_connect(tp->dev, phydev->dev.bus_id, tg3_adjust_link,
+			     phydev->dev_flags, phydev->interface);
 	if (IS_ERR(phydev)) {
 		printk(KERN_ERR "%s: Could not attach to PHY\n", tp->dev->name);
 		return PTR_ERR(phydev);
@@ -11219,7 +11290,7 @@ static void __devinit tg3_get_eeprom_hw_cfg(struct tg3 *tp)
 	tg3_read_mem(tp, NIC_SRAM_DATA_SIG, &val);
 	if (val == NIC_SRAM_DATA_SIG_MAGIC) {
 		u32 nic_cfg, led_cfg;
-		u32 nic_phy_id, ver, cfg2 = 0, eeprom_phy_id;
+		u32 nic_phy_id, ver, cfg2 = 0, cfg4 = 0, eeprom_phy_id;
 		int eeprom_phy_serdes = 0;
 
 		tg3_read_mem(tp, NIC_SRAM_DATA_CFG, &nic_cfg);
@@ -11233,6 +11304,9 @@ static void __devinit tg3_get_eeprom_hw_cfg(struct tg3 *tp)
 		    (ver > 0) && (ver < 0x100))
 			tg3_read_mem(tp, NIC_SRAM_DATA_CFG_2, &cfg2);
 
+		if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5785)
+			tg3_read_mem(tp, NIC_SRAM_DATA_CFG_4, &cfg4);
+
 		if ((nic_cfg & NIC_SRAM_DATA_CFG_PHY_TYPE_MASK) ==
 		    NIC_SRAM_DATA_CFG_PHY_TYPE_FIBER)
 			eeprom_phy_serdes = 1;
@@ -11357,6 +11431,13 @@ static void __devinit tg3_get_eeprom_hw_cfg(struct tg3 *tp)
 			if (cfg3 & NIC_SRAM_ASPM_DEBOUNCE)
 				tp->tg3_flags |= TG3_FLAG_ASPM_WORKAROUND;
 		}
+
+		if (cfg4 & NIC_SRAM_RGMII_STD_IBND_DISABLE)
+			tp->tg3_flags3 |= TG3_FLG3_RGMII_STD_IBND_DISABLE;
+		if (cfg4 & NIC_SRAM_RGMII_EXT_IBND_RX_EN)
+			tp->tg3_flags3 |= TG3_FLG3_RGMII_EXT_IBND_RX_EN;
+		if (cfg4 & NIC_SRAM_RGMII_EXT_IBND_TX_EN)
+			tp->tg3_flags3 |= TG3_FLG3_RGMII_EXT_IBND_TX_EN;
 	}
 }
 
diff --git a/drivers/net/tg3.h b/drivers/net/tg3.h
index 95ac4c9590a1..df07842172b7 100644
--- a/drivers/net/tg3.h
+++ b/drivers/net/tg3.h
@@ -529,7 +529,23 @@
 #define MAC_SERDES_CFG			0x00000590
 #define  MAC_SERDES_CFG_EDGE_SELECT	 0x00001000
 #define MAC_SERDES_STAT			0x00000594
-/* 0x598 --> 0x5b0 unused */
+/* 0x598 --> 0x5a0 unused */
+#define MAC_PHYCFG1			0x000005a0
+#define  MAC_PHYCFG1_RGMII_INT		 0x00000001
+#define  MAC_PHYCFG1_RGMII_EXT_RX_DEC	 0x02000000
+#define  MAC_PHYCFG1_RGMII_SND_STAT_EN	 0x04000000
+#define  MAC_PHYCFG1_TXC_DRV		 0x20000000
+#define MAC_PHYCFG2			0x000005a4
+#define  MAC_PHYCFG2_INBAND_ENABLE	 0x00000001
+#define MAC_EXT_RGMII_MODE		0x000005a8
+#define  MAC_RGMII_MODE_TX_ENABLE	 0x00000001
+#define  MAC_RGMII_MODE_TX_LOWPWR	 0x00000002
+#define  MAC_RGMII_MODE_TX_RESET	 0x00000004
+#define  MAC_RGMII_MODE_RX_INT_B	 0x00000100
+#define  MAC_RGMII_MODE_RX_QUALITY	 0x00000200
+#define  MAC_RGMII_MODE_RX_ACTIVITY	 0x00000400
+#define  MAC_RGMII_MODE_RX_ENG_DET	 0x00000800
+/* 0x5ac --> 0x5b0 unused */
 #define SERDES_RX_CTRL			0x000005b0	/* 5780/5714 only */
 #define  SERDES_RX_SIG_DETECT		 0x00000400
 #define SG_DIG_CTRL			0x000005b0
@@ -1715,6 +1731,12 @@
 #define NIC_SRAM_DATA_CFG_3		0x00000d3c
 #define  NIC_SRAM_ASPM_DEBOUNCE		 0x00000002
 
+#define NIC_SRAM_DATA_CFG_4		0x00000d60
+#define  NIC_SRAM_GMII_MODE		 0x00000002
+#define  NIC_SRAM_RGMII_STD_IBND_DISABLE 0x00000004
+#define  NIC_SRAM_RGMII_EXT_IBND_RX_EN	 0x00000008
+#define  NIC_SRAM_RGMII_EXT_IBND_TX_EN	 0x00000010
+
 #define NIC_SRAM_RX_MINI_BUFFER_DESC	0x00001000
 
 #define NIC_SRAM_DMA_DESC_POOL_BASE	0x00002000
@@ -2486,6 +2508,9 @@ struct tg3 {
 #define TG3_FLG3_MDIOBUS_INITED		0x00000020
 #define TG3_FLG3_MDIOBUS_PAUSED		0x00000040
 #define TG3_FLG3_PHY_CONNECTED		0x00000080
+#define TG3_FLG3_RGMII_STD_IBND_DISABLE	0x00000100
+#define TG3_FLG3_RGMII_EXT_IBND_RX_EN	0x00000200
+#define TG3_FLG3_RGMII_EXT_IBND_TX_EN	0x00000400
 
 	struct timer_list		timer;
 	u16				timer_counter;
@@ -2556,6 +2581,9 @@ struct tg3 {
 #define PHY_REV_BCM5401_B2		0x3
 #define PHY_REV_BCM5401_C0		0x6
 #define PHY_REV_BCM5411_X0		0x1 /* Found on Netgear GA302T */
+#define TG3_PHY_ID_BCM50610		0x143bd60
+#define TG3_PHY_ID_BCMAC131		0x143bc70
+
 
 	u32				led_ctrl;
 	u32				phy_otp;
diff --git a/include/linux/brcmphy.h b/include/linux/brcmphy.h
new file mode 100644
index 000000000000..9b64b6d67873
--- /dev/null
+++ b/include/linux/brcmphy.h
@@ -0,0 +1,6 @@
+#define PHY_BRCM_WIRESPEED_ENABLE	0x00000001
+#define PHY_BRCM_AUTO_PWRDWN_ENABLE	0x00000002
+#define PHY_BRCM_APD_CLK125_ENABLE	0x00000004
+#define PHY_BRCM_STD_IBND_DISABLE	0x00000008
+#define PHY_BRCM_EXT_IBND_RX_ENABLE	0x00000010
+#define PHY_BRCM_EXT_IBND_TX_ENABLE	0x00000020
-- 
cgit v1.2.3


From 75a9cd524c6bb441c61c85bae7020ce5e8b2e807 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Thu, 29 May 2008 01:43:48 -0700
Subject: wanrouter: Fix ioctl handler declaration.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/wanrouter.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/linux/wanrouter.h b/include/linux/wanrouter.h
index 3add87465b1f..e0aa39612eba 100644
--- a/include/linux/wanrouter.h
+++ b/include/linux/wanrouter.h
@@ -522,7 +522,7 @@ extern int wanrouter_proc_init(void);
 extern void wanrouter_proc_cleanup(void);
 extern int wanrouter_proc_add(struct wan_device *wandev);
 extern int wanrouter_proc_delete(struct wan_device *wandev);
-extern int wanrouter_ioctl( struct inode *inode, struct file *file, unsigned int cmd, unsigned long arg);
+extern long wanrouter_ioctl(struct file *file, unsigned int cmd, unsigned long arg);
 
 /* Public Data */
 /* list of registered devices */
-- 
cgit v1.2.3


From b79eeeb9e48457579cb742cd02e162fcd673c4a3 Mon Sep 17 00:00:00 2001
From: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Date: Thu, 29 May 2008 03:25:23 -0700
Subject: tcp: Reorganize tcp_sock to fill 64-bit holes & improve locality
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

I tried to group recovery related fields nearby (non-CA_Open related
variables, to be more accurate) so that one to three cachelines would
not be necessary in CA_Open. These are now contiguously deployed:

  struct sk_buff_head        out_of_order_queue;   /*  1968    80 */
  /* --- cacheline 32 boundary (2048 bytes) --- */
  struct tcp_sack_block      duplicate_sack[1];    /*  2048     8 */
  struct tcp_sack_block      selective_acks[4];    /*  2056    32 */
  struct tcp_sack_block      recv_sack_cache[4];   /*  2088    32 */
  /* --- cacheline 33 boundary (2112 bytes) was 8 bytes ago --- */
  struct sk_buff *           highest_sack;         /*  2120     8 */
  int                        lost_cnt_hint;        /*  2128     4 */
  int                        retransmit_cnt_hint;  /*  2132     4 */
  u32                        lost_retrans_low;     /*  2136     4 */
  u8                         reordering;           /*  2140     1 */
  u8                         keepalive_probes;     /*  2141     1 */

  /* XXX 2 bytes hole, try to pack */

  u32                        prior_ssthresh;       /*  2144     4 */
  u32                        high_seq;             /*  2148     4 */
  u32                        retrans_stamp;        /*  2152     4 */
  u32                        undo_marker;          /*  2156     4 */
  int                        undo_retrans;         /*  2160     4 */
  u32                        total_retrans;        /*  2164     4 */

...and they're then followed by URG slowpath & keepalive related
variables.

Head of the out_of_order_queue always needed for empty checks, if
that's empty (and TCP is in CA_Open), following ~200 bytes (in 64-bit)
shouldn't be necessary for anything. If only OFO queue exists but TCP
is in CA_Open, selective_acks (and possibly duplicate_sack) are
necessary besides the out_of_order_queue but the rest of the block
again shouldn't be (ie., the other direction had losses).

As the cacheline boundaries depend on many factors in the preceeding
stuff, trying to align considering them doesn't make too much sense.

Commented one ordering hazard.

There are number of low utilized u8/16s that could be combined get 2
bytes less in total so that the hole could be made to vanish (includes
at least ecn_flags, urg_data, urg_mode, frto_counter, nonagle).

Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Acked-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/tcp.h | 50 +++++++++++++++++++++++++-------------------------
 1 file changed, 25 insertions(+), 25 deletions(-)

(limited to 'include')

diff --git a/include/linux/tcp.h b/include/linux/tcp.h
index 18e62e3d406f..9881295f3857 100644
--- a/include/linux/tcp.h
+++ b/include/linux/tcp.h
@@ -296,10 +296,9 @@ struct tcp_sock {
 	u32	rcv_ssthresh;	/* Current window clamp			*/
 
 	u32	frto_highmark;	/* snd_nxt when RTO occurred */
-	u8	reordering;	/* Packet reordering metric.		*/
+	u16	advmss;		/* Advertised MSS			*/
 	u8	frto_counter;	/* Number of new acks after RTO */
 	u8	nonagle;	/* Disable Nagle algorithm?             */
-	u8	keepalive_probes; /* num of allowed keep alive probes	*/
 
 /* RTT measurement */
 	u32	srtt;		/* smoothed round trip time << 3	*/
@@ -310,6 +309,10 @@ struct tcp_sock {
 
 	u32	packets_out;	/* Packets which are "in flight"	*/
 	u32	retrans_out;	/* Retransmitted packets out		*/
+
+	u16	urg_data;	/* Saved octet of OOB data and control flags */
+	u8	urg_mode;	/* In urgent mode		*/
+	u8	ecn_flags;	/* ECN status bits.			*/
 /*
  *      Options received (usually on last packet, some only on SYN packets).
  */
@@ -325,13 +328,24 @@ struct tcp_sock {
 	u32	snd_cwnd_used;
 	u32	snd_cwnd_stamp;
 
-	struct sk_buff_head	out_of_order_queue; /* Out of order segments go here */
-
  	u32	rcv_wnd;	/* Current receiver window		*/
 	u32	write_seq;	/* Tail(+1) of data held in tcp send buffer */
 	u32	pushed_seq;	/* Last pushed seq, required to talk to windows */
+	u32	lost_out;	/* Lost packets			*/
+	u32	sacked_out;	/* SACK'd packets			*/
+	u32	fackets_out;	/* FACK'd packets			*/
+	u32	tso_deferred;
+	u32	bytes_acked;	/* Appropriate Byte Counting - RFC3465 */
 
-/*	SACKs data	*/
+	/* from STCP, retrans queue hinting */
+	struct sk_buff* lost_skb_hint;
+	struct sk_buff *scoreboard_skb_hint;
+	struct sk_buff *retransmit_skb_hint;
+	struct sk_buff *forward_skb_hint;
+
+	struct sk_buff_head	out_of_order_queue; /* Out of order segments go here */
+
+	/* SACKs data, these 2 need to be together (see tcp_build_and_update_options) */
 	struct tcp_sack_block duplicate_sack[1]; /* D-SACK block */
 	struct tcp_sack_block selective_acks[4]; /* The SACKS themselves*/
 
@@ -342,23 +356,14 @@ struct tcp_sock {
 					 * sacked_out > 0)
 					 */
 
-	/* from STCP, retrans queue hinting */
-	struct sk_buff* lost_skb_hint;
-
-	struct sk_buff *scoreboard_skb_hint;
-	struct sk_buff *retransmit_skb_hint;
-	struct sk_buff *forward_skb_hint;
-
 	int     lost_cnt_hint;
 	int     retransmit_cnt_hint;
 
 	u32	lost_retrans_low;	/* Sent seq after any rxmit (lowest) */
 
-	u16	advmss;		/* Advertised MSS			*/
+	u8	reordering;	/* Packet reordering metric.		*/
+	u8	keepalive_probes; /* num of allowed keep alive probes	*/
 	u32	prior_ssthresh; /* ssthresh saved at recovery start	*/
-	u32	lost_out;	/* Lost packets			*/
-	u32	sacked_out;	/* SACK'd packets			*/
-	u32	fackets_out;	/* FACK'd packets			*/
 	u32	high_seq;	/* snd_nxt at onset of congestion	*/
 
 	u32	retrans_stamp;	/* Timestamp of the last retransmit,
@@ -366,25 +371,18 @@ struct tcp_sock {
 				 * the first SYN. */
 	u32	undo_marker;	/* tracking retrans started here. */
 	int	undo_retrans;	/* number of undoable retransmissions. */
+	u32	total_retrans;	/* Total retransmits for entire connection */
+
 	u32	urg_seq;	/* Seq of received urgent pointer */
-	u16	urg_data;	/* Saved octet of OOB data and control flags */
-	u8	urg_mode;	/* In urgent mode		*/
-	u8	ecn_flags;	/* ECN status bits.			*/
 	u32	snd_up;		/* Urgent pointer		*/
 
-	u32	total_retrans;	/* Total retransmits for entire connection */
-	u32	bytes_acked;	/* Appropriate Byte Counting - RFC3465 */
-
 	unsigned int		keepalive_time;	  /* time before keep alive takes place */
 	unsigned int		keepalive_intvl;  /* time interval between keep alive probes */
-	int			linger2;
 
 	struct tcp_deferred_accept_info defer_tcp_accept;
 
 	unsigned long last_synq_overflow; 
 
-	u32	tso_deferred;
-
 /* Receiver side RTT estimation */
 	struct {
 		u32	rtt;
@@ -412,6 +410,8 @@ struct tcp_sock {
 /* TCP MD5 Signagure Option information */
 	struct tcp_md5sig_info	*md5sig_info;
 #endif
+
+	int			linger2;
 };
 
 static inline struct tcp_sock *tcp_sk(const struct sock *sk)
-- 
cgit v1.2.3


From 9306102ea5696a3815f8d24ac0c0fbd1e19be7d3 Mon Sep 17 00:00:00 2001
From: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Date: Thu, 29 May 2008 16:35:23 +0800
Subject: mac80211: allow disable FAT in specific configurations

This patch allows to disable FAT channel in specific configurations.

For example the configuration (8, +1), (primary channel 8, extension
channel 12) isn't permitted in U.S., but (8, -1), (primary channel 8,
extension channel 4) is. When FAT channel configuration is not
permitted, FAT channel should be reported as not supported in the
capabilities of the HT IE in association request. And sssociation is
performed on 20Mhz channel.

Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Tomas Winkler <tomas.winkler@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/linux/ieee80211.h  |  2 ++
 include/net/wireless.h     |  6 ++++++
 net/mac80211/ieee80211_i.h |  2 ++
 net/mac80211/mlme.c        | 46 ++++++++++++++++++++++++++++++++++++++++++++--
 4 files changed, 54 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/linux/ieee80211.h b/include/linux/ieee80211.h
index 3c2ac0c37aa8..9300f37cd7e8 100644
--- a/include/linux/ieee80211.h
+++ b/include/linux/ieee80211.h
@@ -320,6 +320,8 @@ struct ieee80211_ht_addt_info {
 #define IEEE80211_HT_CAP_MCS_TX_UEQM		0x10
 /* 802.11n HT IE masks */
 #define IEEE80211_HT_IE_CHA_SEC_OFFSET		0x03
+#define IEEE80211_HT_IE_CHA_SEC_ABOVE 		0x01
+#define IEEE80211_HT_IE_CHA_SEC_BELOW 		0x03
 #define IEEE80211_HT_IE_CHA_WIDTH		0x04
 #define IEEE80211_HT_IE_HT_PROTECTION		0x0003
 #define IEEE80211_HT_IE_NON_GF_STA_PRSNT	0x0004
diff --git a/include/net/wireless.h b/include/net/wireless.h
index 667b4080d30f..9324f8dd183e 100644
--- a/include/net/wireless.h
+++ b/include/net/wireless.h
@@ -39,12 +39,18 @@ enum ieee80211_band {
  *	on this channel.
  * @IEEE80211_CHAN_NO_IBSS: IBSS is not allowed on this channel.
  * @IEEE80211_CHAN_RADAR: Radar detection is required on this channel.
+ * @IEEE80211_CHAN_NO_FAT_ABOVE: extension channel above this channel
+ * 	is not permitted.
+ * @IEEE80211_CHAN_NO_FAT_BELOW: extension channel below this channel
+ * 	is not permitted.
  */
 enum ieee80211_channel_flags {
 	IEEE80211_CHAN_DISABLED		= 1<<0,
 	IEEE80211_CHAN_PASSIVE_SCAN	= 1<<1,
 	IEEE80211_CHAN_NO_IBSS		= 1<<2,
 	IEEE80211_CHAN_RADAR		= 1<<3,
+	IEEE80211_CHAN_NO_FAT_ABOVE	= 1<<4,
+	IEEE80211_CHAN_NO_FAT_BELOW	= 1<<5,
 };
 
 /**
diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h
index 3f8601cafffb..432011cd3647 100644
--- a/net/mac80211/ieee80211_i.h
+++ b/net/mac80211/ieee80211_i.h
@@ -92,6 +92,8 @@ struct ieee80211_sta_bss {
 	size_t wmm_ie_len;
 	u8 *ht_ie;
 	size_t ht_ie_len;
+	u8 *ht_add_ie;
+	size_t ht_add_ie_len;
 #ifdef CONFIG_MAC80211_MESH
 	u8 *mesh_id;
 	size_t mesh_id_len;
diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c
index 6faa7006681a..d30c11337b04 100644
--- a/net/mac80211/mlme.c
+++ b/net/mac80211/mlme.c
@@ -808,8 +808,29 @@ static void ieee80211_send_assoc(struct net_device *dev,
 
 	/* wmm support is a must to HT */
 	if (wmm && (ifsta->flags & IEEE80211_STA_WMM_ENABLED) &&
-	    sband->ht_info.ht_supported) {
-		__le16 tmp = cpu_to_le16(sband->ht_info.cap);
+	    sband->ht_info.ht_supported && bss->ht_add_ie) {
+		struct ieee80211_ht_addt_info *ht_add_info =
+			(struct ieee80211_ht_addt_info *)bss->ht_add_ie;
+		u16 cap = sband->ht_info.cap;
+		__le16 tmp;
+		u32 flags = local->hw.conf.channel->flags;
+
+		switch (ht_add_info->ht_param & IEEE80211_HT_IE_CHA_SEC_OFFSET) {
+		case IEEE80211_HT_IE_CHA_SEC_ABOVE:
+			if (flags & IEEE80211_CHAN_NO_FAT_ABOVE) {
+				cap &= ~IEEE80211_HT_CAP_SUP_WIDTH;
+				cap &= ~IEEE80211_HT_CAP_SGI_40;
+			}
+			break;
+		case IEEE80211_HT_IE_CHA_SEC_BELOW:
+			if (flags & IEEE80211_CHAN_NO_FAT_BELOW) {
+				cap &= ~IEEE80211_HT_CAP_SUP_WIDTH;
+				cap &= ~IEEE80211_HT_CAP_SGI_40;
+			}
+			break;
+		}
+
+		tmp = cpu_to_le16(cap);
 		pos = skb_put(skb, sizeof(struct ieee80211_ht_cap)+2);
 		*pos++ = WLAN_EID_HT_CAPABILITY;
 		*pos++ = sizeof(struct ieee80211_ht_cap);
@@ -2264,6 +2285,7 @@ static void ieee80211_rx_bss_free(struct ieee80211_sta_bss *bss)
 	kfree(bss->rsn_ie);
 	kfree(bss->wmm_ie);
 	kfree(bss->ht_ie);
+	kfree(bss->ht_add_ie);
 	kfree(bss_mesh_id(bss));
 	kfree(bss_mesh_cfg(bss));
 	kfree(bss);
@@ -2640,6 +2662,26 @@ static void ieee80211_rx_bss_info(struct net_device *dev,
 		bss->ht_ie_len = 0;
 	}
 
+	if (elems.ht_info_elem &&
+	     (!bss->ht_add_ie ||
+	     bss->ht_add_ie_len != elems.ht_info_elem_len ||
+	     memcmp(bss->ht_add_ie, elems.ht_info_elem,
+			elems.ht_info_elem_len))) {
+		kfree(bss->ht_add_ie);
+		bss->ht_add_ie =
+			kmalloc(elems.ht_info_elem_len + 2, GFP_ATOMIC);
+		if (bss->ht_add_ie) {
+			memcpy(bss->ht_add_ie, elems.ht_info_elem - 2,
+				elems.ht_info_elem_len + 2);
+			bss->ht_add_ie_len = elems.ht_info_elem_len + 2;
+		} else
+			bss->ht_add_ie_len = 0;
+	} else if (!elems.ht_info_elem && bss->ht_add_ie) {
+		kfree(bss->ht_add_ie);
+		bss->ht_add_ie = NULL;
+		bss->ht_add_ie_len = 0;
+	}
+
 	bss->beacon_int = le16_to_cpu(mgmt->u.beacon.beacon_int);
 	bss->capability = le16_to_cpu(mgmt->u.beacon.capab_info);
 
-- 
cgit v1.2.3


From ba569b4c68f11906da2996ee252bcff0df61cb90 Mon Sep 17 00:00:00 2001
From: Masakazu Mokuno <mokuno@sm.sony.co.jp>
Date: Fri, 30 May 2008 16:52:23 +0900
Subject: WEXT: Add support for passing PMK and capability flags to WEXT

This defines the flags for setting the PMK to the driver and the
capability flag for this so that the user space program can figure out
whether the target driver wants to do 4-way hand shake by itself and
pass the PMK which is needed before 4-way handshake to the driver.

Signed-off-by: Masakazu Mokuno <mokuno@sm.sony.co.jp>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/linux/wireless.h | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'include')

diff --git a/include/linux/wireless.h b/include/linux/wireless.h
index 0a9b5b41ed67..4a95a0e5eeca 100644
--- a/include/linux/wireless.h
+++ b/include/linux/wireless.h
@@ -611,6 +611,7 @@
 #define IW_ENCODE_ALG_WEP	1
 #define IW_ENCODE_ALG_TKIP	2
 #define IW_ENCODE_ALG_CCMP	3
+#define IW_ENCODE_ALG_PMK	4
 /* struct iw_encode_ext ->ext_flags */
 #define IW_ENCODE_EXT_TX_SEQ_VALID	0x00000001
 #define IW_ENCODE_EXT_RX_SEQ_VALID	0x00000002
@@ -630,6 +631,7 @@
 #define IW_ENC_CAPA_WPA2	0x00000002
 #define IW_ENC_CAPA_CIPHER_TKIP	0x00000004
 #define IW_ENC_CAPA_CIPHER_CCMP	0x00000008
+#define IW_ENC_CAPA_4WAY_HANDSHAKE	0x00000010
 
 /* Event capability macros - in (struct iw_range *)->event_capa
  * Because we have more than 32 possible events, we use an array of
-- 
cgit v1.2.3


From 9457afee85e0dfc2b5075a391d6f34463b4c2b90 Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Thu, 5 Jun 2008 11:23:39 -0700
Subject: netlink: Remove nonblock parameter from netlink_attachskb

Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netlink.h  | 2 +-
 ipc/mqueue.c             | 2 +-
 net/netlink/af_netlink.c | 4 ++--
 3 files changed, 4 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/include/linux/netlink.h b/include/linux/netlink.h
index bec1062a25a1..9ff1b54908f3 100644
--- a/include/linux/netlink.h
+++ b/include/linux/netlink.h
@@ -193,7 +193,7 @@ extern int netlink_unregister_notifier(struct notifier_block *nb);
 
 /* finegrained unicast helpers: */
 struct sock *netlink_getsockbyfilp(struct file *filp);
-int netlink_attachskb(struct sock *sk, struct sk_buff *skb, int nonblock,
+int netlink_attachskb(struct sock *sk, struct sk_buff *skb,
 		      long *timeo, struct sock *ssk);
 void netlink_detachskb(struct sock *sk, struct sk_buff *skb);
 int netlink_sendskb(struct sock *sk, struct sk_buff *skb);
diff --git a/ipc/mqueue.c b/ipc/mqueue.c
index b3b69fd51330..3e84b958186b 100644
--- a/ipc/mqueue.c
+++ b/ipc/mqueue.c
@@ -1054,7 +1054,7 @@ retry:
 			}
 
 			timeo = MAX_SCHEDULE_TIMEOUT;
-			ret = netlink_attachskb(sock, nc, 0, &timeo, NULL);
+			ret = netlink_attachskb(sock, nc, &timeo, NULL);
 			if (ret == 1)
 		       		goto retry;
 			if (ret) {
diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c
index 9b97f8006c9c..6507c02dbe0d 100644
--- a/net/netlink/af_netlink.c
+++ b/net/netlink/af_netlink.c
@@ -759,7 +759,7 @@ struct sock *netlink_getsockbyfilp(struct file *filp)
  * 0: continue
  * 1: repeat lookup - reference dropped while waiting for socket memory.
  */
-int netlink_attachskb(struct sock *sk, struct sk_buff *skb, int nonblock,
+int netlink_attachskb(struct sock *sk, struct sk_buff *skb,
 		      long *timeo, struct sock *ssk)
 {
 	struct netlink_sock *nlk;
@@ -892,7 +892,7 @@ retry:
 		return err;
 	}
 
-	err = netlink_attachskb(sk, skb, nonblock, &timeo, ssk);
+	err = netlink_attachskb(sk, skb, &timeo, ssk);
 	if (err == 1)
 		goto retry;
 	if (err)
-- 
cgit v1.2.3


From 93f65158723ceb7078ee9a0fd4830c0de00f4b9e Mon Sep 17 00:00:00 2001
From: Kuo-lang Tseng <kuo-lang.tseng@intel.com>
Date: Mon, 9 Jun 2008 15:55:45 -0700
Subject: netfilter: ebtables: add IPv6 support

It implements matching functions for IPv6 address & traffic class
(merged from the patch sent by Jan Engelhardt [jengelh@computergmbh.de]
http://marc.info/?l=netfilter-devel&m=120182168424052&w=2), protocol,
and layer-4 port id. Corresponding watcher logging function is also
added for IPv6.

Signed-off-by: Kuo-lang Tseng <kuo-lang.tseng@intel.com>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter_bridge/ebt_ip6.h |  40 +++++++++
 include/linux/netfilter_bridge/ebt_log.h |   3 +-
 net/bridge/netfilter/Kconfig             |   9 ++
 net/bridge/netfilter/Makefile            |   1 +
 net/bridge/netfilter/ebt_ip6.c           | 144 +++++++++++++++++++++++++++++++
 net/bridge/netfilter/ebt_log.c           |  64 ++++++++++----
 6 files changed, 244 insertions(+), 17 deletions(-)
 create mode 100644 include/linux/netfilter_bridge/ebt_ip6.h
 create mode 100644 net/bridge/netfilter/ebt_ip6.c

(limited to 'include')

diff --git a/include/linux/netfilter_bridge/ebt_ip6.h b/include/linux/netfilter_bridge/ebt_ip6.h
new file mode 100644
index 000000000000..2273c3ae33ca
--- /dev/null
+++ b/include/linux/netfilter_bridge/ebt_ip6.h
@@ -0,0 +1,40 @@
+/*
+ *  ebt_ip6
+ *
+ *	Authors:
+ * Kuo-Lang Tseng <kuo-lang.tseng@intel.com>
+ * Manohar Castelino <manohar.r.castelino@intel.com>
+ *
+ *  Jan 11, 2008
+ *
+ */
+
+#ifndef __LINUX_BRIDGE_EBT_IP6_H
+#define __LINUX_BRIDGE_EBT_IP6_H
+
+#define EBT_IP6_SOURCE 0x01
+#define EBT_IP6_DEST 0x02
+#define EBT_IP6_TCLASS 0x04
+#define EBT_IP6_PROTO 0x08
+#define EBT_IP6_SPORT 0x10
+#define EBT_IP6_DPORT 0x20
+#define EBT_IP6_MASK (EBT_IP6_SOURCE | EBT_IP6_DEST | EBT_IP6_TCLASS |\
+		      EBT_IP6_PROTO | EBT_IP6_SPORT | EBT_IP6_DPORT)
+#define EBT_IP6_MATCH "ip6"
+
+/* the same values are used for the invflags */
+struct ebt_ip6_info
+{
+	struct in6_addr saddr;
+	struct in6_addr daddr;
+	struct in6_addr smsk;
+	struct in6_addr dmsk;
+	uint8_t  tclass;
+	uint8_t  protocol;
+	uint8_t  bitmask;
+	uint8_t  invflags;
+	uint16_t sport[2];
+	uint16_t dport[2];
+};
+
+#endif
diff --git a/include/linux/netfilter_bridge/ebt_log.h b/include/linux/netfilter_bridge/ebt_log.h
index 96e231ae7554..b76e653157e5 100644
--- a/include/linux/netfilter_bridge/ebt_log.h
+++ b/include/linux/netfilter_bridge/ebt_log.h
@@ -4,7 +4,8 @@
 #define EBT_LOG_IP 0x01 /* if the frame is made by ip, log the ip information */
 #define EBT_LOG_ARP 0x02
 #define EBT_LOG_NFLOG 0x04
-#define EBT_LOG_MASK (EBT_LOG_IP | EBT_LOG_ARP)
+#define EBT_LOG_IP6 0x08
+#define EBT_LOG_MASK (EBT_LOG_IP | EBT_LOG_ARP | EBT_LOG_IP6)
 #define EBT_LOG_PREFIX_SIZE 30
 #define EBT_LOG_WATCHER "log"
 
diff --git a/net/bridge/netfilter/Kconfig b/net/bridge/netfilter/Kconfig
index 7beeefa0f9c0..fb684c2ff8b6 100644
--- a/net/bridge/netfilter/Kconfig
+++ b/net/bridge/netfilter/Kconfig
@@ -83,6 +83,15 @@ config BRIDGE_EBT_IP
 
 	  To compile it as a module, choose M here.  If unsure, say N.
 
+config BRIDGE_EBT_IP6
+	tristate "ebt: IP6 filter support"
+	depends on BRIDGE_NF_EBTABLES
+	help
+	  This option adds the IP6 match, which allows basic IPV6 header field
+	  filtering.
+
+	  To compile it as a module, choose M here.  If unsure, say N.
+
 config BRIDGE_EBT_LIMIT
 	tristate "ebt: limit match support"
 	depends on BRIDGE_NF_EBTABLES
diff --git a/net/bridge/netfilter/Makefile b/net/bridge/netfilter/Makefile
index 83715d73a503..dd960645b413 100644
--- a/net/bridge/netfilter/Makefile
+++ b/net/bridge/netfilter/Makefile
@@ -14,6 +14,7 @@ obj-$(CONFIG_BRIDGE_EBT_802_3) += ebt_802_3.o
 obj-$(CONFIG_BRIDGE_EBT_AMONG) += ebt_among.o
 obj-$(CONFIG_BRIDGE_EBT_ARP) += ebt_arp.o
 obj-$(CONFIG_BRIDGE_EBT_IP) += ebt_ip.o
+obj-$(CONFIG_BRIDGE_EBT_IP) += ebt_ip6.o
 obj-$(CONFIG_BRIDGE_EBT_LIMIT) += ebt_limit.o
 obj-$(CONFIG_BRIDGE_EBT_MARK) += ebt_mark_m.o
 obj-$(CONFIG_BRIDGE_EBT_PKTTYPE) += ebt_pkttype.o
diff --git a/net/bridge/netfilter/ebt_ip6.c b/net/bridge/netfilter/ebt_ip6.c
new file mode 100644
index 000000000000..36efb3a75249
--- /dev/null
+++ b/net/bridge/netfilter/ebt_ip6.c
@@ -0,0 +1,144 @@
+/*
+ *  ebt_ip6
+ *
+ *	Authors:
+ *	Manohar Castelino <manohar.r.castelino@intel.com>
+ *	Kuo-Lang Tseng <kuo-lang.tseng@intel.com>
+ *	Jan Engelhardt <jengelh@computergmbh.de>
+ *
+ * Summary:
+ * This is just a modification of the IPv4 code written by
+ * Bart De Schuymer <bdschuym@pandora.be>
+ * with the changes required to support IPv6
+ *
+ *  Jan, 2008
+ */
+
+#include <linux/netfilter_bridge/ebtables.h>
+#include <linux/netfilter_bridge/ebt_ip6.h>
+#include <linux/ipv6.h>
+#include <net/ipv6.h>
+#include <linux/in.h>
+#include <linux/module.h>
+#include <net/dsfield.h>
+
+struct tcpudphdr {
+	__be16 src;
+	__be16 dst;
+};
+
+static int ebt_filter_ip6(const struct sk_buff *skb,
+   const struct net_device *in,
+   const struct net_device *out, const void *data,
+   unsigned int datalen)
+{
+	const struct ebt_ip6_info *info = (struct ebt_ip6_info *)data;
+	const struct ipv6hdr *ih6;
+	struct ipv6hdr _ip6h;
+	const struct tcpudphdr *pptr;
+	struct tcpudphdr _ports;
+	struct in6_addr tmp_addr;
+	int i;
+
+	ih6 = skb_header_pointer(skb, 0, sizeof(_ip6h), &_ip6h);
+	if (ih6 == NULL)
+		return EBT_NOMATCH;
+	if (info->bitmask & EBT_IP6_TCLASS &&
+	   FWINV(info->tclass != ipv6_get_dsfield(ih6), EBT_IP6_TCLASS))
+		return EBT_NOMATCH;
+	for (i = 0; i < 4; i++)
+		tmp_addr.in6_u.u6_addr32[i] = ih6->saddr.in6_u.u6_addr32[i] &
+			info->smsk.in6_u.u6_addr32[i];
+	if (info->bitmask & EBT_IP6_SOURCE &&
+		FWINV((ipv6_addr_cmp(&tmp_addr, &info->saddr) != 0),
+			EBT_IP6_SOURCE))
+		return EBT_NOMATCH;
+	for (i = 0; i < 4; i++)
+		tmp_addr.in6_u.u6_addr32[i] = ih6->daddr.in6_u.u6_addr32[i] &
+			info->dmsk.in6_u.u6_addr32[i];
+	if (info->bitmask & EBT_IP6_DEST &&
+	   FWINV((ipv6_addr_cmp(&tmp_addr, &info->daddr) != 0), EBT_IP6_DEST))
+		return EBT_NOMATCH;
+	if (info->bitmask & EBT_IP6_PROTO) {
+		uint8_t nexthdr = ih6->nexthdr;
+		int offset_ph;
+
+		offset_ph = ipv6_skip_exthdr(skb, sizeof(_ip6h), &nexthdr);
+		if (offset_ph == -1)
+			return EBT_NOMATCH;
+		if (FWINV(info->protocol != nexthdr, EBT_IP6_PROTO))
+			return EBT_NOMATCH;
+		if (!(info->bitmask & EBT_IP6_DPORT) &&
+		    !(info->bitmask & EBT_IP6_SPORT))
+			return EBT_MATCH;
+		pptr = skb_header_pointer(skb, offset_ph, sizeof(_ports),
+					  &_ports);
+		if (pptr == NULL)
+			return EBT_NOMATCH;
+		if (info->bitmask & EBT_IP6_DPORT) {
+			u32 dst = ntohs(pptr->dst);
+			if (FWINV(dst < info->dport[0] ||
+				  dst > info->dport[1], EBT_IP6_DPORT))
+				return EBT_NOMATCH;
+		}
+		if (info->bitmask & EBT_IP6_SPORT) {
+			u32 src = ntohs(pptr->src);
+			if (FWINV(src < info->sport[0] ||
+				  src > info->sport[1], EBT_IP6_SPORT))
+			return EBT_NOMATCH;
+		}
+		return EBT_MATCH;
+	}
+	return EBT_MATCH;
+}
+
+static int ebt_ip6_check(const char *tablename, unsigned int hookmask,
+   const struct ebt_entry *e, void *data, unsigned int datalen)
+{
+	struct ebt_ip6_info *info = (struct ebt_ip6_info *)data;
+
+	if (datalen != EBT_ALIGN(sizeof(struct ebt_ip6_info)))
+		return -EINVAL;
+	if (e->ethproto != htons(ETH_P_IPV6) || e->invflags & EBT_IPROTO)
+		return -EINVAL;
+	if (info->bitmask & ~EBT_IP6_MASK || info->invflags & ~EBT_IP6_MASK)
+		return -EINVAL;
+	if (info->bitmask & (EBT_IP6_DPORT | EBT_IP6_SPORT)) {
+		if (info->invflags & EBT_IP6_PROTO)
+			return -EINVAL;
+		if (info->protocol != IPPROTO_TCP &&
+		    info->protocol != IPPROTO_UDP &&
+		    info->protocol != IPPROTO_UDPLITE &&
+		    info->protocol != IPPROTO_SCTP &&
+		    info->protocol != IPPROTO_DCCP)
+			 return -EINVAL;
+	}
+	if (info->bitmask & EBT_IP6_DPORT && info->dport[0] > info->dport[1])
+		return -EINVAL;
+	if (info->bitmask & EBT_IP6_SPORT && info->sport[0] > info->sport[1])
+		return -EINVAL;
+	return 0;
+}
+
+static struct ebt_match filter_ip6 =
+{
+	.name		= EBT_IP6_MATCH,
+	.match		= ebt_filter_ip6,
+	.check		= ebt_ip6_check,
+	.me		= THIS_MODULE,
+};
+
+static int __init ebt_ip6_init(void)
+{
+	return ebt_register_match(&filter_ip6);
+}
+
+static void __exit ebt_ip6_fini(void)
+{
+	ebt_unregister_match(&filter_ip6);
+}
+
+module_init(ebt_ip6_init);
+module_exit(ebt_ip6_fini);
+MODULE_DESCRIPTION("Ebtables: IPv6 protocol packet match");
+MODULE_LICENSE("GPL");
diff --git a/net/bridge/netfilter/ebt_log.c b/net/bridge/netfilter/ebt_log.c
index 0b209e4aad0a..c883ec8a28b4 100644
--- a/net/bridge/netfilter/ebt_log.c
+++ b/net/bridge/netfilter/ebt_log.c
@@ -18,6 +18,9 @@
 #include <linux/if_arp.h>
 #include <linux/spinlock.h>
 #include <net/netfilter/nf_log.h>
+#include <linux/ipv6.h>
+#include <net/ipv6.h>
+#include <linux/in6.h>
 
 static DEFINE_SPINLOCK(ebt_log_lock);
 
@@ -58,6 +61,27 @@ static void print_MAC(const unsigned char *p)
 		printk("%02x%c", *p, i == ETH_ALEN - 1 ? ' ':':');
 }
 
+static void
+print_ports(const struct sk_buff *skb, uint8_t protocol, int offset)
+{
+	if (protocol == IPPROTO_TCP ||
+	    protocol == IPPROTO_UDP ||
+	    protocol == IPPROTO_UDPLITE ||
+	    protocol == IPPROTO_SCTP ||
+	    protocol == IPPROTO_DCCP) {
+		const struct tcpudphdr *pptr;
+		struct tcpudphdr _ports;
+
+		pptr = skb_header_pointer(skb, offset,
+					  sizeof(_ports), &_ports);
+		if (pptr == NULL) {
+			printk(" INCOMPLETE TCP/UDP header");
+			return;
+		}
+		printk(" SPT=%u DPT=%u", ntohs(pptr->src), ntohs(pptr->dst));
+	}
+}
+
 #define myNIPQUAD(a) a[0], a[1], a[2], a[3]
 static void
 ebt_log_packet(unsigned int pf, unsigned int hooknum,
@@ -95,23 +119,31 @@ ebt_log_packet(unsigned int pf, unsigned int hooknum,
 		printk(" IP SRC=%u.%u.%u.%u IP DST=%u.%u.%u.%u, IP "
 		       "tos=0x%02X, IP proto=%d", NIPQUAD(ih->saddr),
 		       NIPQUAD(ih->daddr), ih->tos, ih->protocol);
-		if (ih->protocol == IPPROTO_TCP ||
-		    ih->protocol == IPPROTO_UDP ||
-		    ih->protocol == IPPROTO_UDPLITE ||
-		    ih->protocol == IPPROTO_SCTP ||
-		    ih->protocol == IPPROTO_DCCP) {
-			const struct tcpudphdr *pptr;
-			struct tcpudphdr _ports;
-
-			pptr = skb_header_pointer(skb, ih->ihl*4,
-						  sizeof(_ports), &_ports);
-			if (pptr == NULL) {
-				printk(" INCOMPLETE TCP/UDP header");
-				goto out;
-			}
-			printk(" SPT=%u DPT=%u", ntohs(pptr->src),
-			   ntohs(pptr->dst));
+		print_ports(skb, ih->protocol, ih->ihl*4);
+		goto out;
+	}
+
+	if ((bitmask & EBT_LOG_IP6) && eth_hdr(skb)->h_proto ==
+	   htons(ETH_P_IPV6)) {
+		const struct ipv6hdr *ih;
+		struct ipv6hdr _iph;
+		uint8_t nexthdr;
+		int offset_ph;
+
+		ih = skb_header_pointer(skb, 0, sizeof(_iph), &_iph);
+		if (ih == NULL) {
+			printk(" INCOMPLETE IPv6 header");
+			goto out;
 		}
+		printk(" IPv6 SRC=%x:%x:%x:%x:%x:%x:%x:%x "
+		       "IPv6 DST=%x:%x:%x:%x:%x:%x:%x:%x, IPv6 "
+		       "priority=0x%01X, Next Header=%d", NIP6(ih->saddr),
+		       NIP6(ih->daddr), ih->priority, ih->nexthdr);
+		nexthdr = ih->nexthdr;
+		offset_ph = ipv6_skip_exthdr(skb, sizeof(_iph), &nexthdr);
+		if (offset_ph == -1)
+			goto out;
+		print_ports(skb, nexthdr, offset_ph);
 		goto out;
 	}
 
-- 
cgit v1.2.3


From a258860e01b80e8f554a4ab1a6c95e6042eb8b73 Mon Sep 17 00:00:00 2001
From: Pablo Neira Ayuso <pablo@netfilter.org>
Date: Mon, 9 Jun 2008 15:56:39 -0700
Subject: netfilter: ctnetlink: add full support for SCTP to ctnetlink

This patch adds full support for SCTP to ctnetlink. This includes three
new attributes: state, original vtag and reply vtag.

Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter/nfnetlink_conntrack.h | 10 ++++
 net/netfilter/nf_conntrack_proto_sctp.c       | 80 +++++++++++++++++++++++++++
 2 files changed, 90 insertions(+)

(limited to 'include')

diff --git a/include/linux/netfilter/nfnetlink_conntrack.h b/include/linux/netfilter/nfnetlink_conntrack.h
index 0a383ac083cb..759bc043dc65 100644
--- a/include/linux/netfilter/nfnetlink_conntrack.h
+++ b/include/linux/netfilter/nfnetlink_conntrack.h
@@ -81,6 +81,7 @@ enum ctattr_protoinfo {
 	CTA_PROTOINFO_UNSPEC,
 	CTA_PROTOINFO_TCP,
 	CTA_PROTOINFO_DCCP,
+	CTA_PROTOINFO_SCTP,
 	__CTA_PROTOINFO_MAX
 };
 #define CTA_PROTOINFO_MAX (__CTA_PROTOINFO_MAX - 1)
@@ -103,6 +104,15 @@ enum ctattr_protoinfo_dccp {
 };
 #define CTA_PROTOINFO_DCCP_MAX (__CTA_PROTOINFO_DCCP_MAX - 1)
 
+enum ctattr_protoinfo_sctp {
+	CTA_PROTOINFO_SCTP_UNSPEC,
+	CTA_PROTOINFO_SCTP_STATE,
+	CTA_PROTOINFO_SCTP_VTAG_ORIGINAL,
+	CTA_PROTOINFO_SCTP_VTAG_REPLY,
+	__CTA_PROTOINFO_SCTP_MAX
+};
+#define CTA_PROTOINFO_SCTP_MAX (__CTA_PROTOINFO_SCTP_MAX - 1)
+
 enum ctattr_counters {
 	CTA_COUNTERS_UNSPEC,
 	CTA_COUNTERS_PACKETS,		/* old 64bit counters */
diff --git a/net/netfilter/nf_conntrack_proto_sctp.c b/net/netfilter/nf_conntrack_proto_sctp.c
index cbf2e27a22b2..41183a4d2d62 100644
--- a/net/netfilter/nf_conntrack_proto_sctp.c
+++ b/net/netfilter/nf_conntrack_proto_sctp.c
@@ -463,6 +463,82 @@ static bool sctp_new(struct nf_conn *ct, const struct sk_buff *skb,
 	return true;
 }
 
+#if defined(CONFIG_NF_CT_NETLINK) || defined(CONFIG_NF_CT_NETLINK_MODULE)
+
+#include <linux/netfilter/nfnetlink.h>
+#include <linux/netfilter/nfnetlink_conntrack.h>
+
+static int sctp_to_nlattr(struct sk_buff *skb, struct nlattr *nla,
+			  const struct nf_conn *ct)
+{
+	struct nlattr *nest_parms;
+
+	read_lock_bh(&sctp_lock);
+	nest_parms = nla_nest_start(skb, CTA_PROTOINFO_SCTP | NLA_F_NESTED);
+	if (!nest_parms)
+		goto nla_put_failure;
+
+	NLA_PUT_U8(skb, CTA_PROTOINFO_SCTP_STATE, ct->proto.sctp.state);
+
+	NLA_PUT_BE32(skb,
+		     CTA_PROTOINFO_SCTP_VTAG_ORIGINAL,
+		     htonl(ct->proto.sctp.vtag[IP_CT_DIR_ORIGINAL]));
+
+	NLA_PUT_BE32(skb,
+		     CTA_PROTOINFO_SCTP_VTAG_REPLY,
+		     htonl(ct->proto.sctp.vtag[IP_CT_DIR_REPLY]));
+
+	read_unlock_bh(&sctp_lock);
+
+	nla_nest_end(skb, nest_parms);
+
+	return 0;
+
+nla_put_failure:
+	read_unlock_bh(&sctp_lock);
+	return -1;
+}
+
+static const struct nla_policy sctp_nla_policy[CTA_PROTOINFO_SCTP_MAX+1] = {
+	[CTA_PROTOINFO_SCTP_STATE]	    = { .type = NLA_U8 },
+	[CTA_PROTOINFO_SCTP_VTAG_ORIGINAL]  = { .type = NLA_U32 },
+	[CTA_PROTOINFO_SCTP_VTAG_REPLY]     = { .type = NLA_U32 },
+};
+
+static int nlattr_to_sctp(struct nlattr *cda[], struct nf_conn *ct)
+{
+	struct nlattr *attr = cda[CTA_PROTOINFO_SCTP];
+	struct nlattr *tb[CTA_PROTOINFO_SCTP_MAX+1];
+	int err;
+
+	/* updates may not contain the internal protocol info, skip parsing */
+	if (!attr)
+		return 0;
+
+	err = nla_parse_nested(tb,
+			       CTA_PROTOINFO_SCTP_MAX,
+			       attr,
+			       sctp_nla_policy);
+	if (err < 0)
+		return err;
+
+	if (!tb[CTA_PROTOINFO_SCTP_STATE] ||
+	    !tb[CTA_PROTOINFO_SCTP_VTAG_ORIGINAL] ||
+	    !tb[CTA_PROTOINFO_SCTP_VTAG_REPLY])
+		return -EINVAL;
+
+	write_lock_bh(&sctp_lock);
+	ct->proto.sctp.state = nla_get_u8(tb[CTA_PROTOINFO_SCTP_STATE]);
+	ct->proto.sctp.vtag[IP_CT_DIR_ORIGINAL] =
+		ntohl(nla_get_be32(tb[CTA_PROTOINFO_SCTP_VTAG_ORIGINAL]));
+	ct->proto.sctp.vtag[IP_CT_DIR_REPLY] =
+		ntohl(nla_get_be32(tb[CTA_PROTOINFO_SCTP_VTAG_REPLY]));
+	write_unlock_bh(&sctp_lock);
+
+	return 0;
+}
+#endif
+
 #ifdef CONFIG_SYSCTL
 static unsigned int sctp_sysctl_table_users;
 static struct ctl_table_header *sctp_sysctl_header;
@@ -591,6 +667,8 @@ static struct nf_conntrack_l4proto nf_conntrack_l4proto_sctp4 __read_mostly = {
 	.new 			= sctp_new,
 	.me 			= THIS_MODULE,
 #if defined(CONFIG_NF_CT_NETLINK) || defined(CONFIG_NF_CT_NETLINK_MODULE)
+	.to_nlattr		= sctp_to_nlattr,
+	.from_nlattr		= nlattr_to_sctp,
 	.tuple_to_nlattr	= nf_ct_port_tuple_to_nlattr,
 	.nlattr_to_tuple	= nf_ct_port_nlattr_to_tuple,
 	.nla_policy		= nf_ct_port_nla_policy,
@@ -617,6 +695,8 @@ static struct nf_conntrack_l4proto nf_conntrack_l4proto_sctp6 __read_mostly = {
 	.new 			= sctp_new,
 	.me 			= THIS_MODULE,
 #if defined(CONFIG_NF_CT_NETLINK) || defined(CONFIG_NF_CT_NETLINK_MODULE)
+	.to_nlattr		= sctp_to_nlattr,
+	.from_nlattr		= nlattr_to_sctp,
 	.tuple_to_nlattr	= nf_ct_port_tuple_to_nlattr,
 	.nlattr_to_tuple	= nf_ct_port_nlattr_to_tuple,
 	.nla_policy		= nf_ct_port_nla_policy,
-- 
cgit v1.2.3


From 560ee653b67074b805f1b661988a72a0e58811a5 Mon Sep 17 00:00:00 2001
From: James Morris <jmorris@namei.org>
Date: Mon, 9 Jun 2008 15:57:24 -0700
Subject: netfilter: ip_tables: add iptables security table for mandatory
 access control rules

The following patch implements a new "security" table for iptables, so
that MAC (SELinux etc.) networking rules can be managed separately to
standard DAC rules.

This is to help with distro integration of the new secmark-based
network controls, per various previous discussions.

The need for a separate table arises from the fact that existing tools
and usage of iptables will likely clash with centralized MAC policy
management.

The SECMARK and CONNSECMARK targets will still be valid in the mangle
table to prevent breakage of existing users.

Signed-off-by: James Morris <jmorris@namei.org>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter_ipv4.h        |   1 +
 include/net/netns/ipv4.h              |   1 +
 net/ipv4/netfilter/Kconfig            |  12 +++
 net/ipv4/netfilter/Makefile           |   1 +
 net/ipv4/netfilter/iptable_security.c | 180 ++++++++++++++++++++++++++++++++++
 net/netfilter/xt_CONNSECMARK.c        |  10 +-
 net/netfilter/xt_SECMARK.c            |  10 +-
 7 files changed, 209 insertions(+), 6 deletions(-)
 create mode 100644 net/ipv4/netfilter/iptable_security.c

(limited to 'include')

diff --git a/include/linux/netfilter_ipv4.h b/include/linux/netfilter_ipv4.h
index 650318b0c405..29c7727ff0e8 100644
--- a/include/linux/netfilter_ipv4.h
+++ b/include/linux/netfilter_ipv4.h
@@ -60,6 +60,7 @@ enum nf_ip_hook_priorities {
 	NF_IP_PRI_MANGLE = -150,
 	NF_IP_PRI_NAT_DST = -100,
 	NF_IP_PRI_FILTER = 0,
+	NF_IP_PRI_SECURITY = 50,
 	NF_IP_PRI_NAT_SRC = 100,
 	NF_IP_PRI_SELINUX_LAST = 225,
 	NF_IP_PRI_CONNTRACK_CONFIRM = INT_MAX,
diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h
index 34ee348a2cf2..6ef90b5fafb3 100644
--- a/include/net/netns/ipv4.h
+++ b/include/net/netns/ipv4.h
@@ -36,6 +36,7 @@ struct netns_ipv4 {
 	struct xt_table		*iptable_mangle;
 	struct xt_table		*iptable_raw;
 	struct xt_table		*arptable_filter;
+	struct xt_table		*iptable_security;
 #endif
 
 	int sysctl_icmp_echo_ignore_all;
diff --git a/net/ipv4/netfilter/Kconfig b/net/ipv4/netfilter/Kconfig
index 2767841a8cef..6e251402506e 100644
--- a/net/ipv4/netfilter/Kconfig
+++ b/net/ipv4/netfilter/Kconfig
@@ -365,6 +365,18 @@ config IP_NF_RAW
 	  If you want to compile it as a module, say M here and read
 	  <file:Documentation/kbuild/modules.txt>.  If unsure, say `N'.
 
+# security table for MAC policy
+config IP_NF_SECURITY
+	tristate "Security table"
+	depends on IP_NF_IPTABLES
+	depends on SECURITY
+	default m if NETFILTER_ADVANCED=n
+	help
+	  This option adds a `security' table to iptables, for use
+	  with Mandatory Access Control (MAC) policy.
+	 
+	  If unsure, say N.
+
 # ARP tables
 config IP_NF_ARPTABLES
 	tristate "ARP tables support"
diff --git a/net/ipv4/netfilter/Makefile b/net/ipv4/netfilter/Makefile
index d9b92fbf5579..3f31291f37ce 100644
--- a/net/ipv4/netfilter/Makefile
+++ b/net/ipv4/netfilter/Makefile
@@ -42,6 +42,7 @@ obj-$(CONFIG_IP_NF_FILTER) += iptable_filter.o
 obj-$(CONFIG_IP_NF_MANGLE) += iptable_mangle.o
 obj-$(CONFIG_NF_NAT) += iptable_nat.o
 obj-$(CONFIG_IP_NF_RAW) += iptable_raw.o
+obj-$(CONFIG_IP_NF_SECURITY) += iptable_security.o
 
 # matches
 obj-$(CONFIG_IP_NF_MATCH_ADDRTYPE) += ipt_addrtype.o
diff --git a/net/ipv4/netfilter/iptable_security.c b/net/ipv4/netfilter/iptable_security.c
new file mode 100644
index 000000000000..2b472ac2263a
--- /dev/null
+++ b/net/ipv4/netfilter/iptable_security.c
@@ -0,0 +1,180 @@
+/*
+ * "security" table
+ *
+ * This is for use by Mandatory Access Control (MAC) security models,
+ * which need to be able to manage security policy in separate context
+ * to DAC.
+ *
+ * Based on iptable_mangle.c
+ *
+ * Copyright (C) 1999 Paul `Rusty' Russell & Michael J. Neuling
+ * Copyright (C) 2000-2004 Netfilter Core Team <coreteam <at> netfilter.org>
+ * Copyright (C) 2008 Red Hat, Inc., James Morris <jmorris <at> redhat.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/module.h>
+#include <linux/netfilter_ipv4/ip_tables.h>
+#include <net/ip.h>
+
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("James Morris <jmorris <at> redhat.com>");
+MODULE_DESCRIPTION("iptables security table, for MAC rules");
+
+#define SECURITY_VALID_HOOKS	(1 << NF_INET_LOCAL_IN) | \
+				(1 << NF_INET_FORWARD) | \
+				(1 << NF_INET_LOCAL_OUT)
+
+static struct
+{
+	struct ipt_replace repl;
+	struct ipt_standard entries[3];
+	struct ipt_error term;
+} initial_table __initdata = {
+	.repl = {
+		.name = "security",
+		.valid_hooks = SECURITY_VALID_HOOKS,
+		.num_entries = 4,
+		.size = sizeof(struct ipt_standard) * 3 + sizeof(struct ipt_error),
+		.hook_entry = {
+			[NF_INET_LOCAL_IN] 	= 0,
+			[NF_INET_FORWARD] 	= sizeof(struct ipt_standard),
+			[NF_INET_LOCAL_OUT] 	= sizeof(struct ipt_standard) * 2,
+		},
+		.underflow = {
+			[NF_INET_LOCAL_IN] 	= 0,
+			[NF_INET_FORWARD] 	= sizeof(struct ipt_standard),
+			[NF_INET_LOCAL_OUT] 	= sizeof(struct ipt_standard) * 2,
+		},
+	},
+	.entries = {
+		IPT_STANDARD_INIT(NF_ACCEPT),	/* LOCAL_IN */
+		IPT_STANDARD_INIT(NF_ACCEPT),	/* FORWARD */
+		IPT_STANDARD_INIT(NF_ACCEPT),	/* LOCAL_OUT */
+	},
+	.term = IPT_ERROR_INIT,			/* ERROR */
+};
+
+static struct xt_table security_table = {
+	.name		= "security",
+	.valid_hooks	= SECURITY_VALID_HOOKS,
+	.lock		= __RW_LOCK_UNLOCKED(security_table.lock),
+	.me		= THIS_MODULE,
+	.af		= AF_INET,
+};
+
+static unsigned int
+ipt_local_in_hook(unsigned int hook,
+		  struct sk_buff *skb,
+		  const struct net_device *in,
+		  const struct net_device *out,
+		  int (*okfn)(struct sk_buff *))
+{
+	return ipt_do_table(skb, hook, in, out,
+			    nf_local_in_net(in, out)->ipv4.iptable_security);
+}
+
+static unsigned int
+ipt_forward_hook(unsigned int hook,
+		 struct sk_buff *skb,
+		 const struct net_device *in,
+		 const struct net_device *out,
+		 int (*okfn)(struct sk_buff *))
+{
+	return ipt_do_table(skb, hook, in, out,
+			    nf_forward_net(in, out)->ipv4.iptable_security);
+}
+
+static unsigned int
+ipt_local_out_hook(unsigned int hook,
+		   struct sk_buff *skb,
+		   const struct net_device *in,
+		   const struct net_device *out,
+		   int (*okfn)(struct sk_buff *))
+{
+	/* Somebody is playing with raw sockets. */
+	if (skb->len < sizeof(struct iphdr)
+	    || ip_hdrlen(skb) < sizeof(struct iphdr)) {
+		if (net_ratelimit())
+			printk(KERN_INFO "iptable_security: ignoring short "
+			       "SOCK_RAW packet.\n");
+		return NF_ACCEPT;
+	}
+	return ipt_do_table(skb, hook, in, out,
+			    nf_local_out_net(in, out)->ipv4.iptable_security);
+}
+
+static struct nf_hook_ops ipt_ops[] __read_mostly = {
+	{
+		.hook		= ipt_local_in_hook,
+		.owner		= THIS_MODULE,
+		.pf		= PF_INET,
+		.hooknum	= NF_INET_LOCAL_IN,
+		.priority	= NF_IP_PRI_SECURITY,
+	},
+	{
+		.hook		= ipt_forward_hook,
+		.owner		= THIS_MODULE,
+		.pf		= PF_INET,
+		.hooknum	= NF_INET_FORWARD,
+		.priority	= NF_IP_PRI_SECURITY,
+	},
+	{
+		.hook		= ipt_local_out_hook,
+		.owner		= THIS_MODULE,
+		.pf		= PF_INET,
+		.hooknum	= NF_INET_LOCAL_OUT,
+		.priority	= NF_IP_PRI_SECURITY,
+	},
+};
+
+static int __net_init iptable_security_net_init(struct net *net)
+{
+	net->ipv4.iptable_security =
+		ipt_register_table(net, &security_table, &initial_table.repl);
+
+	if (IS_ERR(net->ipv4.iptable_security))
+		return PTR_ERR(net->ipv4.iptable_security);
+
+	return 0;
+}
+
+static void __net_exit iptable_security_net_exit(struct net *net)
+{
+	ipt_unregister_table(net->ipv4.iptable_security);
+}
+
+static struct pernet_operations iptable_security_net_ops = {
+	.init = iptable_security_net_init,
+	.exit = iptable_security_net_exit,
+};
+
+static int __init iptable_security_init(void)
+{
+	int ret;
+
+	ret = register_pernet_subsys(&iptable_security_net_ops);
+        if (ret < 0)
+		return ret;
+
+	ret = nf_register_hooks(ipt_ops, ARRAY_SIZE(ipt_ops));
+	if (ret < 0)
+		goto cleanup_table;
+
+	return ret;
+
+cleanup_table:
+	unregister_pernet_subsys(&iptable_security_net_ops);
+	return ret;
+}
+
+static void __exit iptable_security_fini(void)
+{
+	nf_unregister_hooks(ipt_ops, ARRAY_SIZE(ipt_ops));
+	unregister_pernet_subsys(&iptable_security_net_ops);
+}
+
+module_init(iptable_security_init);
+module_exit(iptable_security_fini);
diff --git a/net/netfilter/xt_CONNSECMARK.c b/net/netfilter/xt_CONNSECMARK.c
index 211189eb2b67..76ca1f2421eb 100644
--- a/net/netfilter/xt_CONNSECMARK.c
+++ b/net/netfilter/xt_CONNSECMARK.c
@@ -8,7 +8,7 @@
  *   Copyright (C) 2002,2004 MARA Systems AB <http://www.marasystems.com>
  *    by Henrik Nordstrom <hno@marasystems.com>
  *
- * (C) 2006 Red Hat, Inc., James Morris <jmorris@redhat.com>
+ * (C) 2006,2008 Red Hat, Inc., James Morris <jmorris@redhat.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
@@ -94,6 +94,12 @@ connsecmark_tg_check(const char *tablename, const void *entry,
 {
 	const struct xt_connsecmark_target_info *info = targinfo;
 
+	if (strcmp(tablename, "mangle") && strcmp(tablename, "security")) {
+		printk(KERN_INFO PFX "target only valid in the \'mangle\' "
+		       "or \'security\' tables, not \'%s\'.\n", tablename);
+		return false;
+	}
+
 	switch (info->mode) {
 	case CONNSECMARK_SAVE:
 	case CONNSECMARK_RESTORE:
@@ -126,7 +132,6 @@ static struct xt_target connsecmark_tg_reg[] __read_mostly = {
 		.destroy	= connsecmark_tg_destroy,
 		.target		= connsecmark_tg,
 		.targetsize	= sizeof(struct xt_connsecmark_target_info),
-		.table		= "mangle",
 		.me		= THIS_MODULE,
 	},
 	{
@@ -136,7 +141,6 @@ static struct xt_target connsecmark_tg_reg[] __read_mostly = {
 		.destroy	= connsecmark_tg_destroy,
 		.target		= connsecmark_tg,
 		.targetsize	= sizeof(struct xt_connsecmark_target_info),
-		.table		= "mangle",
 		.me		= THIS_MODULE,
 	},
 };
diff --git a/net/netfilter/xt_SECMARK.c b/net/netfilter/xt_SECMARK.c
index c0284856ccd4..94f87ee7552b 100644
--- a/net/netfilter/xt_SECMARK.c
+++ b/net/netfilter/xt_SECMARK.c
@@ -5,7 +5,7 @@
  * Based on the nfmark match by:
  * (C) 1999-2001 Marc Boucher <marc@mbsi.ca>
  *
- * (C) 2006 Red Hat, Inc., James Morris <jmorris@redhat.com>
+ * (C) 2006,2008 Red Hat, Inc., James Morris <jmorris@redhat.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
@@ -89,6 +89,12 @@ secmark_tg_check(const char *tablename, const void *entry,
 {
 	struct xt_secmark_target_info *info = targinfo;
 
+	if (strcmp(tablename, "mangle") && strcmp(tablename, "security")) {
+		printk(KERN_INFO PFX "target only valid in the \'mangle\' "
+		       "or \'security\' tables, not \'%s\'.\n", tablename);
+		return false;
+	}
+
 	if (mode && mode != info->mode) {
 		printk(KERN_INFO PFX "mode already set to %hu cannot mix with "
 		       "rules for mode %hu\n", mode, info->mode);
@@ -127,7 +133,6 @@ static struct xt_target secmark_tg_reg[] __read_mostly = {
 		.destroy	= secmark_tg_destroy,
 		.target		= secmark_tg,
 		.targetsize	= sizeof(struct xt_secmark_target_info),
-		.table		= "mangle",
 		.me		= THIS_MODULE,
 	},
 	{
@@ -137,7 +142,6 @@ static struct xt_target secmark_tg_reg[] __read_mostly = {
 		.destroy	= secmark_tg_destroy,
 		.target		= secmark_tg,
 		.targetsize	= sizeof(struct xt_secmark_target_info),
-		.table		= "mangle",
 		.me		= THIS_MODULE,
 	},
 };
-- 
cgit v1.2.3


From 17e6e59f0a1d7188d783c15dc3ccebd95a0840cd Mon Sep 17 00:00:00 2001
From: James Morris <jmorris@namei.org>
Date: Mon, 9 Jun 2008 15:58:05 -0700
Subject: netfilter: ip6_tables: add ip6tables security table

This is a port of the IPv4 security table for IPv6.

Signed-off-by: James Morris <jmorris@namei.org>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter_ipv6.h         |   1 +
 include/net/netns/ipv6.h               |   1 +
 net/ipv6/netfilter/Kconfig             |  12 +++
 net/ipv6/netfilter/Makefile            |   1 +
 net/ipv6/netfilter/ip6table_security.c | 172 +++++++++++++++++++++++++++++++++
 5 files changed, 187 insertions(+)
 create mode 100644 net/ipv6/netfilter/ip6table_security.c

(limited to 'include')

diff --git a/include/linux/netfilter_ipv6.h b/include/linux/netfilter_ipv6.h
index 3475a65dae9b..fd50988b83ec 100644
--- a/include/linux/netfilter_ipv6.h
+++ b/include/linux/netfilter_ipv6.h
@@ -64,6 +64,7 @@ enum nf_ip6_hook_priorities {
 	NF_IP6_PRI_MANGLE = -150,
 	NF_IP6_PRI_NAT_DST = -100,
 	NF_IP6_PRI_FILTER = 0,
+	NF_IP6_PRI_SECURITY = 50,
 	NF_IP6_PRI_NAT_SRC = 100,
 	NF_IP6_PRI_SELINUX_LAST = 225,
 	NF_IP6_PRI_LAST = INT_MAX,
diff --git a/include/net/netns/ipv6.h b/include/net/netns/ipv6.h
index ac053be6c256..5bacd838e88b 100644
--- a/include/net/netns/ipv6.h
+++ b/include/net/netns/ipv6.h
@@ -35,6 +35,7 @@ struct netns_ipv6 {
 	struct xt_table		*ip6table_filter;
 	struct xt_table		*ip6table_mangle;
 	struct xt_table		*ip6table_raw;
+	struct xt_table		*ip6table_security;
 #endif
 	struct rt6_info         *ip6_null_entry;
 	struct rt6_statistics   *rt6_stats;
diff --git a/net/ipv6/netfilter/Kconfig b/net/ipv6/netfilter/Kconfig
index 6cae5475737e..689dec899c57 100644
--- a/net/ipv6/netfilter/Kconfig
+++ b/net/ipv6/netfilter/Kconfig
@@ -208,5 +208,17 @@ config IP6_NF_RAW
 	  If you want to compile it as a module, say M here and read
 	  <file:Documentation/kbuild/modules.txt>.  If unsure, say `N'.
 
+# security table for MAC policy
+config IP6_NF_SECURITY
+       tristate "Security table"
+       depends on IP6_NF_IPTABLES
+       depends on SECURITY
+       default m if NETFILTER_ADVANCED=n
+       help
+         This option adds a `security' table to iptables, for use
+         with Mandatory Access Control (MAC) policy.
+        
+         If unsure, say N.
+
 endmenu
 
diff --git a/net/ipv6/netfilter/Makefile b/net/ipv6/netfilter/Makefile
index fbf2c14ed887..3f17c948eefb 100644
--- a/net/ipv6/netfilter/Makefile
+++ b/net/ipv6/netfilter/Makefile
@@ -8,6 +8,7 @@ obj-$(CONFIG_IP6_NF_FILTER) += ip6table_filter.o
 obj-$(CONFIG_IP6_NF_MANGLE) += ip6table_mangle.o
 obj-$(CONFIG_IP6_NF_QUEUE) += ip6_queue.o
 obj-$(CONFIG_IP6_NF_RAW) += ip6table_raw.o
+obj-$(CONFIG_IP6_NF_SECURITY) += ip6table_security.o
 
 # objects for l3 independent conntrack
 nf_conntrack_ipv6-objs  :=  nf_conntrack_l3proto_ipv6.o nf_conntrack_proto_icmpv6.o nf_conntrack_reasm.o
diff --git a/net/ipv6/netfilter/ip6table_security.c b/net/ipv6/netfilter/ip6table_security.c
new file mode 100644
index 000000000000..063a3d9c3c67
--- /dev/null
+++ b/net/ipv6/netfilter/ip6table_security.c
@@ -0,0 +1,172 @@
+/*
+ * "security" table for IPv6
+ *
+ * This is for use by Mandatory Access Control (MAC) security models,
+ * which need to be able to manage security policy in separate context
+ * to DAC.
+ *
+ * Based on iptable_mangle.c
+ *
+ * Copyright (C) 1999 Paul `Rusty' Russell & Michael J. Neuling
+ * Copyright (C) 2000-2004 Netfilter Core Team <coreteam <at> netfilter.org>
+ * Copyright (C) 2008 Red Hat, Inc., James Morris <jmorris <at> redhat.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/module.h>
+#include <linux/netfilter_ipv6/ip6_tables.h>
+
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("James Morris <jmorris <at> redhat.com>");
+MODULE_DESCRIPTION("ip6tables security table, for MAC rules");
+
+#define SECURITY_VALID_HOOKS	(1 << NF_INET_LOCAL_IN) | \
+				(1 << NF_INET_FORWARD) | \
+				(1 << NF_INET_LOCAL_OUT)
+
+static struct
+{
+	struct ip6t_replace repl;
+	struct ip6t_standard entries[3];
+	struct ip6t_error term;
+} initial_table __initdata = {
+	.repl = {
+		.name = "security",
+		.valid_hooks = SECURITY_VALID_HOOKS,
+		.num_entries = 4,
+		.size = sizeof(struct ip6t_standard) * 3 + sizeof(struct ip6t_error),
+		.hook_entry = {
+			[NF_INET_LOCAL_IN] 	= 0,
+			[NF_INET_FORWARD] 	= sizeof(struct ip6t_standard),
+			[NF_INET_LOCAL_OUT] 	= sizeof(struct ip6t_standard) * 2,
+		},
+		.underflow = {
+			[NF_INET_LOCAL_IN] 	= 0,
+			[NF_INET_FORWARD] 	= sizeof(struct ip6t_standard),
+			[NF_INET_LOCAL_OUT] 	= sizeof(struct ip6t_standard) * 2,
+		},
+	},
+	.entries = {
+		IP6T_STANDARD_INIT(NF_ACCEPT),	/* LOCAL_IN */
+		IP6T_STANDARD_INIT(NF_ACCEPT),	/* FORWARD */
+		IP6T_STANDARD_INIT(NF_ACCEPT),	/* LOCAL_OUT */
+	},
+	.term = IP6T_ERROR_INIT,		/* ERROR */
+};
+
+static struct xt_table security_table = {
+	.name		= "security",
+	.valid_hooks	= SECURITY_VALID_HOOKS,
+	.lock		= __RW_LOCK_UNLOCKED(security_table.lock),
+	.me		= THIS_MODULE,
+	.af		= AF_INET6,
+};
+
+static unsigned int
+ip6t_local_in_hook(unsigned int hook,
+		   struct sk_buff *skb,
+		   const struct net_device *in,
+		   const struct net_device *out,
+		   int (*okfn)(struct sk_buff *))
+{
+	return ip6t_do_table(skb, hook, in, out,
+			     init_net.ipv6.ip6table_security);
+}
+
+static unsigned int
+ip6t_forward_hook(unsigned int hook,
+		  struct sk_buff *skb,
+		  const struct net_device *in,
+		  const struct net_device *out,
+		  int (*okfn)(struct sk_buff *))
+{
+	return ip6t_do_table(skb, hook, in, out,
+			     init_net.ipv6.ip6table_security);
+}
+
+static unsigned int
+ip6t_local_out_hook(unsigned int hook,
+		    struct sk_buff *skb,
+		    const struct net_device *in,
+		    const struct net_device *out,
+		    int (*okfn)(struct sk_buff *))
+{
+	/* TBD: handle short packets via raw socket */
+	return ip6t_do_table(skb, hook, in, out,
+			     init_net.ipv6.ip6table_security);
+}
+
+static struct nf_hook_ops ip6t_ops[] __read_mostly = {
+	{
+		.hook		= ip6t_local_in_hook,
+		.owner		= THIS_MODULE,
+		.pf		= PF_INET6,
+		.hooknum	= NF_INET_LOCAL_IN,
+		.priority	= NF_IP6_PRI_SECURITY,
+	},
+	{
+		.hook		= ip6t_forward_hook,
+		.owner		= THIS_MODULE,
+		.pf		= PF_INET6,
+		.hooknum	= NF_INET_FORWARD,
+		.priority	= NF_IP6_PRI_SECURITY,
+	},
+	{
+		.hook		= ip6t_local_out_hook,
+		.owner		= THIS_MODULE,
+		.pf		= PF_INET6,
+		.hooknum	= NF_INET_LOCAL_OUT,
+		.priority	= NF_IP6_PRI_SECURITY,
+	},
+};
+
+static int __net_init ip6table_security_net_init(struct net *net)
+{
+	net->ipv6.ip6table_security =
+		ip6t_register_table(net, &security_table, &initial_table.repl);
+
+	if (IS_ERR(net->ipv6.ip6table_security))
+		return PTR_ERR(net->ipv6.ip6table_security);
+
+	return 0;
+}
+
+static void __net_exit ip6table_security_net_exit(struct net *net)
+{
+	ip6t_unregister_table(net->ipv6.ip6table_security);
+}
+
+static struct pernet_operations ip6table_security_net_ops = {
+	.init = ip6table_security_net_init,
+	.exit = ip6table_security_net_exit,
+};
+
+static int __init ip6table_security_init(void)
+{
+	int ret;
+
+	ret = register_pernet_subsys(&ip6table_security_net_ops);
+	if (ret < 0)
+		return ret;
+
+	ret = nf_register_hooks(ip6t_ops, ARRAY_SIZE(ip6t_ops));
+	if (ret < 0)
+		goto cleanup_table;
+
+	return ret;
+
+cleanup_table:
+	unregister_pernet_subsys(&ip6table_security_net_ops);
+	return ret;
+}
+
+static void __exit ip6table_security_fini(void)
+{
+	nf_unregister_hooks(ip6t_ops, ARRAY_SIZE(ip6t_ops));
+	unregister_pernet_subsys(&ip6table_security_net_ops);
+}
+
+module_init(ip6table_security_init);
+module_exit(ip6table_security_fini);
-- 
cgit v1.2.3


From 51091764f26ec36c02e35166f083193a30f426fc Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Mon, 9 Jun 2008 15:59:06 -0700
Subject: netfilter: nf_conntrack: add nf_ct_kill()

Encapsulate the common

	if (del_timer(&ct->timeout))
		ct->timeout.function((unsigned long)ct)

sequence in a new function.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/netfilter/nf_conntrack.h           | 2 ++
 net/ipv4/netfilter/nf_conntrack_proto_icmp.c   | 5 ++---
 net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c | 5 ++---
 net/netfilter/nf_conntrack_core.c              | 7 +++++++
 net/netfilter/nf_conntrack_netlink.c           | 3 +--
 net/netfilter/nf_conntrack_proto_dccp.c        | 3 +--
 net/netfilter/nf_conntrack_proto_tcp.c         | 9 +++------
 7 files changed, 18 insertions(+), 16 deletions(-)

(limited to 'include')

diff --git a/include/net/netfilter/nf_conntrack.h b/include/net/netfilter/nf_conntrack.h
index 2dbd6c015b94..fc19ab23eeaa 100644
--- a/include/net/netfilter/nf_conntrack.h
+++ b/include/net/netfilter/nf_conntrack.h
@@ -223,6 +223,8 @@ static inline void nf_ct_refresh(struct nf_conn *ct,
 	__nf_ct_refresh_acct(ct, 0, skb, extra_jiffies, 0);
 }
 
+extern void nf_ct_kill(struct nf_conn *ct);
+
 /* These are for NAT.  Icky. */
 /* Update TCP window tracking data when NAT mangles the packet */
 extern void nf_conntrack_tcp_update(const struct sk_buff *skb,
diff --git a/net/ipv4/netfilter/nf_conntrack_proto_icmp.c b/net/ipv4/netfilter/nf_conntrack_proto_icmp.c
index 78ab19accace..0e21a46184fb 100644
--- a/net/ipv4/netfilter/nf_conntrack_proto_icmp.c
+++ b/net/ipv4/netfilter/nf_conntrack_proto_icmp.c
@@ -87,9 +87,8 @@ static int icmp_packet(struct nf_conn *ct,
 	   means this will only run once even if count hits zero twice
 	   (theoretically possible with SMP) */
 	if (CTINFO2DIR(ctinfo) == IP_CT_DIR_REPLY) {
-		if (atomic_dec_and_test(&ct->proto.icmp.count)
-		    && del_timer(&ct->timeout))
-			ct->timeout.function((unsigned long)ct);
+		if (atomic_dec_and_test(&ct->proto.icmp.count))
+			nf_ct_kill(ct);
 	} else {
 		atomic_inc(&ct->proto.icmp.count);
 		nf_conntrack_event_cache(IPCT_PROTOINFO_VOLATILE, skb);
diff --git a/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c b/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c
index ee713b03e9ec..fe081b90e05e 100644
--- a/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c
+++ b/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c
@@ -89,9 +89,8 @@ static int icmpv6_packet(struct nf_conn *ct,
 	   means this will only run once even if count hits zero twice
 	   (theoretically possible with SMP) */
 	if (CTINFO2DIR(ctinfo) == IP_CT_DIR_REPLY) {
-		if (atomic_dec_and_test(&ct->proto.icmp.count)
-		    && del_timer(&ct->timeout))
-			ct->timeout.function((unsigned long)ct);
+		if (atomic_dec_and_test(&ct->proto.icmp.count))
+			nf_ct_kill(ct);
 	} else {
 		atomic_inc(&ct->proto.icmp.count);
 		nf_conntrack_event_cache(IPCT_PROTOINFO_VOLATILE, skb);
diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c
index c4b1799da5d7..79b07c35eb87 100644
--- a/net/netfilter/nf_conntrack_core.c
+++ b/net/netfilter/nf_conntrack_core.c
@@ -848,6 +848,13 @@ acct:
 }
 EXPORT_SYMBOL_GPL(__nf_ct_refresh_acct);
 
+void nf_ct_kill(struct nf_conn *ct)
+{
+	if (del_timer(&ct->timeout))
+		ct->timeout.function((unsigned long)ct);
+}
+EXPORT_SYMBOL_GPL(nf_ct_kill);
+
 #if defined(CONFIG_NF_CT_NETLINK) || defined(CONFIG_NF_CT_NETLINK_MODULE)
 
 #include <linux/netfilter/nfnetlink.h>
diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c
index 13918c1fbf66..ab655f660dfa 100644
--- a/net/netfilter/nf_conntrack_netlink.c
+++ b/net/netfilter/nf_conntrack_netlink.c
@@ -812,9 +812,8 @@ ctnetlink_del_conntrack(struct sock *ctnl, struct sk_buff *skb,
 			return -ENOENT;
 		}
 	}
-	if (del_timer(&ct->timeout))
-		ct->timeout.function((unsigned long)ct);
 
+	nf_ct_kill(ct);
 	nf_ct_put(ct);
 
 	return 0;
diff --git a/net/netfilter/nf_conntrack_proto_dccp.c b/net/netfilter/nf_conntrack_proto_dccp.c
index afb4a1861d2c..223742f371f9 100644
--- a/net/netfilter/nf_conntrack_proto_dccp.c
+++ b/net/netfilter/nf_conntrack_proto_dccp.c
@@ -475,8 +475,7 @@ static int dccp_packet(struct nf_conn *ct, const struct sk_buff *skb,
 	if (type == DCCP_PKT_RESET &&
 	    !test_bit(IPS_SEEN_REPLY_BIT, &ct->status)) {
 		/* Tear down connection immediately if only reply is a RESET */
-		if (del_timer(&ct->timeout))
-			ct->timeout.function((unsigned long)ct);
+		nf_ct_kill(ct);
 		return NF_ACCEPT;
 	}
 
diff --git a/net/netfilter/nf_conntrack_proto_tcp.c b/net/netfilter/nf_conntrack_proto_tcp.c
index ba94004fe323..c4aa11e01405 100644
--- a/net/netfilter/nf_conntrack_proto_tcp.c
+++ b/net/netfilter/nf_conntrack_proto_tcp.c
@@ -843,8 +843,7 @@ static int tcp_packet(struct nf_conn *ct,
 			/* Attempt to reopen a closed/aborted connection.
 			 * Delete this connection and look up again. */
 			write_unlock_bh(&tcp_lock);
-			if (del_timer(&ct->timeout))
-				ct->timeout.function((unsigned long)ct);
+			nf_ct_kill(ct);
 			return -NF_REPEAT;
 		}
 		/* Fall through */
@@ -877,8 +876,7 @@ static int tcp_packet(struct nf_conn *ct,
 			if (LOG_INVALID(IPPROTO_TCP))
 				nf_log_packet(pf, 0, skb, NULL, NULL, NULL,
 					  "nf_ct_tcp: killing out of sync session ");
-			if (del_timer(&ct->timeout))
-				ct->timeout.function((unsigned long)ct);
+			nf_ct_kill(ct);
 			return -NF_DROP;
 		}
 		ct->proto.tcp.last_index = index;
@@ -961,8 +959,7 @@ static int tcp_packet(struct nf_conn *ct,
 		   problem case, so we can delete the conntrack
 		   immediately.  --RR */
 		if (th->rst) {
-			if (del_timer(&ct->timeout))
-				ct->timeout.function((unsigned long)ct);
+			nf_ct_kill(ct);
 			return NF_ACCEPT;
 		}
 	} else if (!test_bit(IPS_ASSURED_BIT, &ct->status)
-- 
cgit v1.2.3


From 718d4ad98e272daebc258e49dc02f52a6a8de9d3 Mon Sep 17 00:00:00 2001
From: Fabian Hugelshofer <hugelshofer2006@gmx.ch>
Date: Mon, 9 Jun 2008 15:59:40 -0700
Subject: netfilter: nf_conntrack: properly account terminating packets

Currently the last packet of a connection isn't accounted when its causing
abnormal termination.

Introduces nf_ct_kill_acct() which increments the accounting counters on
conntrack kill. The new function was necessary, because there are calls
to nf_ct_kill() which don't need accounting:

nf_conntrack_proto_tcp.c line ~847:
Kills ct and returns NF_REPEAT. We don't want to count twice.

nf_conntrack_proto_tcp.c line ~880:
Kills ct and returns NF_DROP. I think we don't want to count dropped
packets.

nf_conntrack_netlink.c line ~824:
As far as I can see ctnetlink_del_conntrack() is used to destroy a
conntrack on behalf of the user. There is an sk_buff, but I don't think
this is an actual packet. Incrementing counters here is therefore not
desired.

Signed-off-by: Fabian Hugelshofer <hugelshofer2006@gmx.ch>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/netfilter/nf_conntrack.h           | 19 ++++++++++++++++++-
 net/ipv4/netfilter/nf_conntrack_proto_icmp.c   |  2 +-
 net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c |  2 +-
 net/netfilter/nf_conntrack_core.c              | 16 ++++++++++++++--
 net/netfilter/nf_conntrack_proto_dccp.c        |  2 +-
 net/netfilter/nf_conntrack_proto_tcp.c         |  2 +-
 6 files changed, 36 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/include/net/netfilter/nf_conntrack.h b/include/net/netfilter/nf_conntrack.h
index fc19ab23eeaa..d77dec768dc2 100644
--- a/include/net/netfilter/nf_conntrack.h
+++ b/include/net/netfilter/nf_conntrack.h
@@ -223,7 +223,24 @@ static inline void nf_ct_refresh(struct nf_conn *ct,
 	__nf_ct_refresh_acct(ct, 0, skb, extra_jiffies, 0);
 }
 
-extern void nf_ct_kill(struct nf_conn *ct);
+extern void __nf_ct_kill_acct(struct nf_conn *ct,
+				enum ip_conntrack_info ctinfo,
+				const struct sk_buff *skb,
+				int do_acct);
+
+/* kill conntrack and do accounting */
+static inline void nf_ct_kill_acct(struct nf_conn *ct,
+				enum ip_conntrack_info ctinfo,
+				const struct sk_buff *skb)
+{
+	__nf_ct_kill_acct(ct, ctinfo, skb, 1);
+}
+
+/* kill conntrack without accounting */
+static inline void nf_ct_kill(struct nf_conn *ct)
+{
+	__nf_ct_kill_acct(ct, 0, NULL, 0);
+}
 
 /* These are for NAT.  Icky. */
 /* Update TCP window tracking data when NAT mangles the packet */
diff --git a/net/ipv4/netfilter/nf_conntrack_proto_icmp.c b/net/ipv4/netfilter/nf_conntrack_proto_icmp.c
index 0e21a46184fb..97791048fa9b 100644
--- a/net/ipv4/netfilter/nf_conntrack_proto_icmp.c
+++ b/net/ipv4/netfilter/nf_conntrack_proto_icmp.c
@@ -88,7 +88,7 @@ static int icmp_packet(struct nf_conn *ct,
 	   (theoretically possible with SMP) */
 	if (CTINFO2DIR(ctinfo) == IP_CT_DIR_REPLY) {
 		if (atomic_dec_and_test(&ct->proto.icmp.count))
-			nf_ct_kill(ct);
+			nf_ct_kill_acct(ct, ctinfo, skb);
 	} else {
 		atomic_inc(&ct->proto.icmp.count);
 		nf_conntrack_event_cache(IPCT_PROTOINFO_VOLATILE, skb);
diff --git a/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c b/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c
index fe081b90e05e..14d47d833545 100644
--- a/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c
+++ b/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c
@@ -90,7 +90,7 @@ static int icmpv6_packet(struct nf_conn *ct,
 	   (theoretically possible with SMP) */
 	if (CTINFO2DIR(ctinfo) == IP_CT_DIR_REPLY) {
 		if (atomic_dec_and_test(&ct->proto.icmp.count))
-			nf_ct_kill(ct);
+			nf_ct_kill_acct(ct, ctinfo, skb);
 	} else {
 		atomic_inc(&ct->proto.icmp.count);
 		nf_conntrack_event_cache(IPCT_PROTOINFO_VOLATILE, skb);
diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c
index 79b07c35eb87..e6d645221d5c 100644
--- a/net/netfilter/nf_conntrack_core.c
+++ b/net/netfilter/nf_conntrack_core.c
@@ -848,12 +848,24 @@ acct:
 }
 EXPORT_SYMBOL_GPL(__nf_ct_refresh_acct);
 
-void nf_ct_kill(struct nf_conn *ct)
+void __nf_ct_kill_acct(struct nf_conn *ct,
+		enum ip_conntrack_info ctinfo,
+		const struct sk_buff *skb,
+		int do_acct)
 {
+#ifdef CONFIG_NF_CT_ACCT
+	if (do_acct) {
+		spin_lock_bh(&nf_conntrack_lock);
+		ct->counters[CTINFO2DIR(ctinfo)].packets++;
+		ct->counters[CTINFO2DIR(ctinfo)].bytes +=
+			skb->len - skb_network_offset(skb);
+		spin_unlock_bh(&nf_conntrack_lock);
+	}
+#endif
 	if (del_timer(&ct->timeout))
 		ct->timeout.function((unsigned long)ct);
 }
-EXPORT_SYMBOL_GPL(nf_ct_kill);
+EXPORT_SYMBOL_GPL(__nf_ct_kill_acct);
 
 #if defined(CONFIG_NF_CT_NETLINK) || defined(CONFIG_NF_CT_NETLINK_MODULE)
 
diff --git a/net/netfilter/nf_conntrack_proto_dccp.c b/net/netfilter/nf_conntrack_proto_dccp.c
index 223742f371f9..e7866dd3cde6 100644
--- a/net/netfilter/nf_conntrack_proto_dccp.c
+++ b/net/netfilter/nf_conntrack_proto_dccp.c
@@ -475,7 +475,7 @@ static int dccp_packet(struct nf_conn *ct, const struct sk_buff *skb,
 	if (type == DCCP_PKT_RESET &&
 	    !test_bit(IPS_SEEN_REPLY_BIT, &ct->status)) {
 		/* Tear down connection immediately if only reply is a RESET */
-		nf_ct_kill(ct);
+		nf_ct_kill_acct(ct, ctinfo, skb);
 		return NF_ACCEPT;
 	}
 
diff --git a/net/netfilter/nf_conntrack_proto_tcp.c b/net/netfilter/nf_conntrack_proto_tcp.c
index c4aa11e01405..8db13fba10bc 100644
--- a/net/netfilter/nf_conntrack_proto_tcp.c
+++ b/net/netfilter/nf_conntrack_proto_tcp.c
@@ -959,7 +959,7 @@ static int tcp_packet(struct nf_conn *ct,
 		   problem case, so we can delete the conntrack
 		   immediately.  --RR */
 		if (th->rst) {
-			nf_ct_kill(ct);
+			nf_ct_kill_acct(ct, ctinfo, skb);
 			return NF_ACCEPT;
 		}
 	} else if (!test_bit(IPS_ASSURED_BIT, &ct->status)
-- 
cgit v1.2.3


From 7bcd978e8cf2a1a9502d454cd2f80f9834e82610 Mon Sep 17 00:00:00 2001
From: Rami Rosen <ramirose@gmail.com>
Date: Mon, 9 Jun 2008 16:00:22 -0700
Subject: netfilter: nf_conntrack: remove unnecessary function declaration

This patch removes nf_ct_ipv4_ct_gather_frags() method declaration from
include/net/netfilter/ipv4/nf_conntrack_ipv4.h, since it is unused in
the Linux kernel.

Signed-off-by: Rami Rosen <ramirose@gmail.com>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/netfilter/ipv4/nf_conntrack_ipv4.h | 2 --
 1 file changed, 2 deletions(-)

(limited to 'include')

diff --git a/include/net/netfilter/ipv4/nf_conntrack_ipv4.h b/include/net/netfilter/ipv4/nf_conntrack_ipv4.h
index 9bf059817aec..7573d52a4346 100644
--- a/include/net/netfilter/ipv4/nf_conntrack_ipv4.h
+++ b/include/net/netfilter/ipv4/nf_conntrack_ipv4.h
@@ -9,8 +9,6 @@
 #ifndef _NF_CONNTRACK_IPV4_H
 #define _NF_CONNTRACK_IPV4_H
 
-/* Returns new sk_buff, or NULL */
-struct sk_buff *nf_ct_ipv4_ct_gather_frags(struct sk_buff *skb);
 
 extern struct nf_conntrack_l3proto nf_conntrack_l3proto_ipv4;
 
-- 
cgit v1.2.3


From 0399e5f07ad7024f70881d44f19d767d3f25a980 Mon Sep 17 00:00:00 2001
From: Rami Rosen <ramirose@gmail.com>
Date: Wed, 23 Apr 2008 14:36:58 +0300
Subject: ipv6 addrconf: Remove IFA_GLOBAL definition from
 include/net/if_inet6.h.

This patches removes IFA_GLOBAL definition from linux/include/net/if_inet6.h
as it is unused.

Signed-off-by: Rami Rosen <ramirose@gmail.com>
Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
---
 include/net/if_inet6.h | 1 -
 1 file changed, 1 deletion(-)

(limited to 'include')

diff --git a/include/net/if_inet6.h b/include/net/if_inet6.h
index b2cfc4927257..db66c7927743 100644
--- a/include/net/if_inet6.h
+++ b/include/net/if_inet6.h
@@ -148,7 +148,6 @@ struct ifacaddr6
 #define	IFA_HOST	IPV6_ADDR_LOOPBACK
 #define	IFA_LINK	IPV6_ADDR_LINKLOCAL
 #define	IFA_SITE	IPV6_ADDR_SITELOCAL
-#define	IFA_GLOBAL	0x0000U
 
 struct ipv6_devstat {
 	struct proc_dir_entry	*proc_dir_entry;
-- 
cgit v1.2.3


From 3de232554a91adc74e80dc15c304be806bd7e1f9 Mon Sep 17 00:00:00 2001
From: Benjamin Thery <benjamin.thery@bull.net>
Date: Wed, 28 May 2008 14:51:24 +0200
Subject: ipv6 netns: Address labels per namespace

This pacth makes IPv6 address labels per network namespace.
It keeps the global label tables, ip6addrlbl_table, but
adds a 'net' member to each ip6addrlbl_entry.
This new member is taken into account when matching labels.

Changelog
=========
* v1: Initial version
* v2:
  * Minize the penalty when network namespaces are not configured:
      *  the 'net' member is added only if CONFIG_NET_NS is
         defined. This saves space when network namespaces are not
         configured.
      * 'net' value is retrieved with the inlined function
         ip6addrlbl_net() that always return &init_net when
         CONFIG_NET_NS is not defined.
  * 'net' member in ip6addrlbl_entry renamed to the less generic
    'lbl_net' name (helps code search).

Signed-off-by: Benjamin Thery <benjamin.thery@bull.net>
Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
---
 include/net/addrconf.h |   3 +-
 net/ipv6/addrconf.c    |  12 +++---
 net/ipv6/addrlabel.c   | 106 ++++++++++++++++++++++++++++++++++---------------
 3 files changed, 84 insertions(+), 37 deletions(-)

(limited to 'include')

diff --git a/include/net/addrconf.h b/include/net/addrconf.h
index bbd3d583c6e6..06b28142b3ab 100644
--- a/include/net/addrconf.h
+++ b/include/net/addrconf.h
@@ -121,7 +121,8 @@ static inline int addrconf_finite_timeout(unsigned long timeout)
  */
 extern int			ipv6_addr_label_init(void);
 extern void			ipv6_addr_label_rtnl_register(void);
-extern u32			ipv6_addr_label(const struct in6_addr *addr,
+extern u32			ipv6_addr_label(struct net *net,
+						const struct in6_addr *addr,
 						int type, int ifindex);
 
 /*
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index 9ea4e62741ee..fa43374e85c5 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -964,7 +964,8 @@ static inline int ipv6_saddr_preferred(int type)
 	return 0;
 }
 
-static int ipv6_get_saddr_eval(struct ipv6_saddr_score *score,
+static int ipv6_get_saddr_eval(struct net *net,
+			       struct ipv6_saddr_score *score,
 			       struct ipv6_saddr_dst *dst,
 			       int i)
 {
@@ -1043,7 +1044,8 @@ static int ipv6_get_saddr_eval(struct ipv6_saddr_score *score,
 		break;
 	case IPV6_SADDR_RULE_LABEL:
 		/* Rule 6: Prefer matching label */
-		ret = ipv6_addr_label(&score->ifa->addr, score->addr_type,
+		ret = ipv6_addr_label(net,
+				      &score->ifa->addr, score->addr_type,
 				      score->ifa->idev->dev->ifindex) == dst->label;
 		break;
 #ifdef CONFIG_IPV6_PRIVACY
@@ -1097,7 +1099,7 @@ int ipv6_dev_get_saddr(struct net_device *dst_dev,
 	dst.addr = daddr;
 	dst.ifindex = dst_dev ? dst_dev->ifindex : 0;
 	dst.scope = __ipv6_addr_src_scope(dst_type);
-	dst.label = ipv6_addr_label(daddr, dst_type, dst.ifindex);
+	dst.label = ipv6_addr_label(net, daddr, dst_type, dst.ifindex);
 	dst.prefs = prefs;
 
 	hiscore->rule = -1;
@@ -1165,8 +1167,8 @@ int ipv6_dev_get_saddr(struct net_device *dst_dev,
 			for (i = 0; i < IPV6_SADDR_RULE_MAX; i++) {
 				int minihiscore, miniscore;
 
-				minihiscore = ipv6_get_saddr_eval(hiscore, &dst, i);
-				miniscore = ipv6_get_saddr_eval(score, &dst, i);
+				minihiscore = ipv6_get_saddr_eval(net, hiscore, &dst, i);
+				miniscore = ipv6_get_saddr_eval(net, score, &dst, i);
 
 				if (minihiscore > miniscore) {
 					if (i == IPV6_SADDR_RULE_SCOPE &&
diff --git a/net/ipv6/addrlabel.c b/net/ipv6/addrlabel.c
index 9bfa8846f262..08909039d87b 100644
--- a/net/ipv6/addrlabel.c
+++ b/net/ipv6/addrlabel.c
@@ -29,6 +29,9 @@
  */
 struct ip6addrlbl_entry
 {
+#ifdef CONFIG_NET_NS
+	struct net *lbl_net;
+#endif
 	struct in6_addr prefix;
 	int prefixlen;
 	int ifindex;
@@ -46,6 +49,16 @@ static struct ip6addrlbl_table
 	u32 seq;
 } ip6addrlbl_table;
 
+static inline
+struct net *ip6addrlbl_net(const struct ip6addrlbl_entry *lbl)
+{
+#ifdef CONFIG_NET_NS
+	return lbl->lbl_net;
+#else
+	return &init_net;
+#endif
+}
+
 /*
  * Default policy table (RFC3484 + extensions)
  *
@@ -65,7 +78,7 @@ static struct ip6addrlbl_table
 
 #define IPV6_ADDR_LABEL_DEFAULT	0xffffffffUL
 
-static const __initdata struct ip6addrlbl_init_table
+static const __net_initdata struct ip6addrlbl_init_table
 {
 	const struct in6_addr *prefix;
 	int prefixlen;
@@ -108,6 +121,9 @@ static const __initdata struct ip6addrlbl_init_table
 /* Object management */
 static inline void ip6addrlbl_free(struct ip6addrlbl_entry *p)
 {
+#ifdef CONFIG_NET_NS
+	release_net(p->lbl_net);
+#endif
 	kfree(p);
 }
 
@@ -128,10 +144,13 @@ static inline void ip6addrlbl_put(struct ip6addrlbl_entry *p)
 }
 
 /* Find label */
-static int __ip6addrlbl_match(struct ip6addrlbl_entry *p,
+static int __ip6addrlbl_match(struct net *net,
+			      struct ip6addrlbl_entry *p,
 			      const struct in6_addr *addr,
 			      int addrtype, int ifindex)
 {
+	if (!net_eq(ip6addrlbl_net(p), net))
+		return 0;
 	if (p->ifindex && p->ifindex != ifindex)
 		return 0;
 	if (p->addrtype && p->addrtype != addrtype)
@@ -141,19 +160,21 @@ static int __ip6addrlbl_match(struct ip6addrlbl_entry *p,
 	return 1;
 }
 
-static struct ip6addrlbl_entry *__ipv6_addr_label(const struct in6_addr *addr,
+static struct ip6addrlbl_entry *__ipv6_addr_label(struct net *net,
+						  const struct in6_addr *addr,
 						  int type, int ifindex)
 {
 	struct hlist_node *pos;
 	struct ip6addrlbl_entry *p;
 	hlist_for_each_entry_rcu(p, pos, &ip6addrlbl_table.head, list) {
-		if (__ip6addrlbl_match(p, addr, type, ifindex))
+		if (__ip6addrlbl_match(net, p, addr, type, ifindex))
 			return p;
 	}
 	return NULL;
 }
 
-u32 ipv6_addr_label(const struct in6_addr *addr, int type, int ifindex)
+u32 ipv6_addr_label(struct net *net,
+		    const struct in6_addr *addr, int type, int ifindex)
 {
 	u32 label;
 	struct ip6addrlbl_entry *p;
@@ -161,7 +182,7 @@ u32 ipv6_addr_label(const struct in6_addr *addr, int type, int ifindex)
 	type &= IPV6_ADDR_MAPPED | IPV6_ADDR_COMPATv4 | IPV6_ADDR_LOOPBACK;
 
 	rcu_read_lock();
-	p = __ipv6_addr_label(addr, type, ifindex);
+	p = __ipv6_addr_label(net, addr, type, ifindex);
 	label = p ? p->label : IPV6_ADDR_LABEL_DEFAULT;
 	rcu_read_unlock();
 
@@ -174,7 +195,8 @@ u32 ipv6_addr_label(const struct in6_addr *addr, int type, int ifindex)
 }
 
 /* allocate one entry */
-static struct ip6addrlbl_entry *ip6addrlbl_alloc(const struct in6_addr *prefix,
+static struct ip6addrlbl_entry *ip6addrlbl_alloc(struct net *net,
+						 const struct in6_addr *prefix,
 						 int prefixlen, int ifindex,
 						 u32 label)
 {
@@ -216,6 +238,9 @@ static struct ip6addrlbl_entry *ip6addrlbl_alloc(const struct in6_addr *prefix,
 	newp->addrtype = addrtype;
 	newp->label = label;
 	INIT_HLIST_NODE(&newp->list);
+#ifdef CONFIG_NET_NS
+	newp->lbl_net = hold_net(net);
+#endif
 	atomic_set(&newp->refcnt, 1);
 	return newp;
 }
@@ -237,6 +262,7 @@ static int __ip6addrlbl_add(struct ip6addrlbl_entry *newp, int replace)
 		hlist_for_each_entry_safe(p, pos, n,
 					  &ip6addrlbl_table.head, list) {
 			if (p->prefixlen == newp->prefixlen &&
+			    net_eq(ip6addrlbl_net(p), ip6addrlbl_net(newp)) &&
 			    p->ifindex == newp->ifindex &&
 			    ipv6_addr_equal(&p->prefix, &newp->prefix)) {
 				if (!replace) {
@@ -261,7 +287,8 @@ out:
 }
 
 /* add a label */
-static int ip6addrlbl_add(const struct in6_addr *prefix, int prefixlen,
+static int ip6addrlbl_add(struct net *net,
+			  const struct in6_addr *prefix, int prefixlen,
 			  int ifindex, u32 label, int replace)
 {
 	struct ip6addrlbl_entry *newp;
@@ -274,7 +301,7 @@ static int ip6addrlbl_add(const struct in6_addr *prefix, int prefixlen,
 			(unsigned int)label,
 			replace);
 
-	newp = ip6addrlbl_alloc(prefix, prefixlen, ifindex, label);
+	newp = ip6addrlbl_alloc(net, prefix, prefixlen, ifindex, label);
 	if (IS_ERR(newp))
 		return PTR_ERR(newp);
 	spin_lock(&ip6addrlbl_table.lock);
@@ -286,7 +313,8 @@ static int ip6addrlbl_add(const struct in6_addr *prefix, int prefixlen,
 }
 
 /* remove a label */
-static int __ip6addrlbl_del(const struct in6_addr *prefix, int prefixlen,
+static int __ip6addrlbl_del(struct net *net,
+			    const struct in6_addr *prefix, int prefixlen,
 			    int ifindex)
 {
 	struct ip6addrlbl_entry *p = NULL;
@@ -300,6 +328,7 @@ static int __ip6addrlbl_del(const struct in6_addr *prefix, int prefixlen,
 
 	hlist_for_each_entry_safe(p, pos, n, &ip6addrlbl_table.head, list) {
 		if (p->prefixlen == prefixlen &&
+		    net_eq(ip6addrlbl_net(p), net) &&
 		    p->ifindex == ifindex &&
 		    ipv6_addr_equal(&p->prefix, prefix)) {
 			hlist_del_rcu(&p->list);
@@ -311,7 +340,8 @@ static int __ip6addrlbl_del(const struct in6_addr *prefix, int prefixlen,
 	return ret;
 }
 
-static int ip6addrlbl_del(const struct in6_addr *prefix, int prefixlen,
+static int ip6addrlbl_del(struct net *net,
+			  const struct in6_addr *prefix, int prefixlen,
 			  int ifindex)
 {
 	struct in6_addr prefix_buf;
@@ -324,13 +354,13 @@ static int ip6addrlbl_del(const struct in6_addr *prefix, int prefixlen,
 
 	ipv6_addr_prefix(&prefix_buf, prefix, prefixlen);
 	spin_lock(&ip6addrlbl_table.lock);
-	ret = __ip6addrlbl_del(&prefix_buf, prefixlen, ifindex);
+	ret = __ip6addrlbl_del(net, &prefix_buf, prefixlen, ifindex);
 	spin_unlock(&ip6addrlbl_table.lock);
 	return ret;
 }
 
 /* add default label */
-static __init int ip6addrlbl_init(void)
+static int __net_init ip6addrlbl_net_init(struct net *net)
 {
 	int err = 0;
 	int i;
@@ -338,7 +368,8 @@ static __init int ip6addrlbl_init(void)
 	ADDRLABEL(KERN_DEBUG "%s()\n", __func__);
 
 	for (i = 0; i < ARRAY_SIZE(ip6addrlbl_init_table); i++) {
-		int ret = ip6addrlbl_add(ip6addrlbl_init_table[i].prefix,
+		int ret = ip6addrlbl_add(net,
+					 ip6addrlbl_init_table[i].prefix,
 					 ip6addrlbl_init_table[i].prefixlen,
 					 0,
 					 ip6addrlbl_init_table[i].label, 0);
@@ -349,11 +380,32 @@ static __init int ip6addrlbl_init(void)
 	return err;
 }
 
+static void __net_exit ip6addrlbl_net_exit(struct net *net)
+{
+	struct ip6addrlbl_entry *p = NULL;
+	struct hlist_node *pos, *n;
+
+	/* Remove all labels belonging to the exiting net */
+	spin_lock(&ip6addrlbl_table.lock);
+	hlist_for_each_entry_safe(p, pos, n, &ip6addrlbl_table.head, list) {
+		if (net_eq(ip6addrlbl_net(p), net)) {
+			hlist_del_rcu(&p->list);
+			ip6addrlbl_put(p);
+		}
+	}
+	spin_unlock(&ip6addrlbl_table.lock);
+}
+
+static struct pernet_operations ipv6_addr_label_ops = {
+	.init = ip6addrlbl_net_init,
+	.exit = ip6addrlbl_net_exit,
+};
+
 int __init ipv6_addr_label_init(void)
 {
 	spin_lock_init(&ip6addrlbl_table.lock);
 
-	return ip6addrlbl_init();
+	return register_pernet_subsys(&ipv6_addr_label_ops);
 }
 
 static const struct nla_policy ifal_policy[IFAL_MAX+1] = {
@@ -371,9 +423,6 @@ static int ip6addrlbl_newdel(struct sk_buff *skb, struct nlmsghdr *nlh,
 	u32 label;
 	int err = 0;
 
-	if (net != &init_net)
-		return 0;
-
 	err = nlmsg_parse(nlh, sizeof(*ifal), tb, IFAL_MAX, ifal_policy);
 	if (err < 0)
 		return err;
@@ -385,7 +434,7 @@ static int ip6addrlbl_newdel(struct sk_buff *skb, struct nlmsghdr *nlh,
 		return -EINVAL;
 
 	if (ifal->ifal_index &&
-	    !__dev_get_by_index(&init_net, ifal->ifal_index))
+	    !__dev_get_by_index(net, ifal->ifal_index))
 		return -EINVAL;
 
 	if (!tb[IFAL_ADDRESS])
@@ -403,12 +452,12 @@ static int ip6addrlbl_newdel(struct sk_buff *skb, struct nlmsghdr *nlh,
 
 	switch(nlh->nlmsg_type) {
 	case RTM_NEWADDRLABEL:
-		err = ip6addrlbl_add(pfx, ifal->ifal_prefixlen,
+		err = ip6addrlbl_add(net, pfx, ifal->ifal_prefixlen,
 				     ifal->ifal_index, label,
 				     nlh->nlmsg_flags & NLM_F_REPLACE);
 		break;
 	case RTM_DELADDRLABEL:
-		err = ip6addrlbl_del(pfx, ifal->ifal_prefixlen,
+		err = ip6addrlbl_del(net, pfx, ifal->ifal_prefixlen,
 				     ifal->ifal_index);
 		break;
 	default:
@@ -458,12 +507,10 @@ static int ip6addrlbl_dump(struct sk_buff *skb, struct netlink_callback *cb)
 	int idx = 0, s_idx = cb->args[0];
 	int err;
 
-	if (net != &init_net)
-		return 0;
-
 	rcu_read_lock();
 	hlist_for_each_entry_rcu(p, pos, &ip6addrlbl_table.head, list) {
-		if (idx >= s_idx) {
+		if (idx >= s_idx &&
+		    net_eq(ip6addrlbl_net(p), net)) {
 			if ((err = ip6addrlbl_fill(skb, p,
 						   ip6addrlbl_table.seq,
 						   NETLINK_CB(cb->skb).pid,
@@ -499,9 +546,6 @@ static int ip6addrlbl_get(struct sk_buff *in_skb, struct nlmsghdr* nlh,
 	struct ip6addrlbl_entry *p;
 	struct sk_buff *skb;
 
-	if (net != &init_net)
-		return 0;
-
 	err = nlmsg_parse(nlh, sizeof(*ifal), tb, IFAL_MAX, ifal_policy);
 	if (err < 0)
 		return err;
@@ -513,7 +557,7 @@ static int ip6addrlbl_get(struct sk_buff *in_skb, struct nlmsghdr* nlh,
 		return -EINVAL;
 
 	if (ifal->ifal_index &&
-	    !__dev_get_by_index(&init_net, ifal->ifal_index))
+	    !__dev_get_by_index(net, ifal->ifal_index))
 		return -EINVAL;
 
 	if (!tb[IFAL_ADDRESS])
@@ -524,7 +568,7 @@ static int ip6addrlbl_get(struct sk_buff *in_skb, struct nlmsghdr* nlh,
 		return -EINVAL;
 
 	rcu_read_lock();
-	p = __ipv6_addr_label(addr, ipv6_addr_type(addr), ifal->ifal_index);
+	p = __ipv6_addr_label(net, addr, ipv6_addr_type(addr), ifal->ifal_index);
 	if (p && ip6addrlbl_hold(p))
 		p = NULL;
 	lseq = ip6addrlbl_table.seq;
@@ -552,7 +596,7 @@ static int ip6addrlbl_get(struct sk_buff *in_skb, struct nlmsghdr* nlh,
 		goto out;
 	}
 
-	err = rtnl_unicast(skb, &init_net, NETLINK_CB(in_skb).pid);
+	err = rtnl_unicast(skb, net, NETLINK_CB(in_skb).pid);
 out:
 	return err;
 }
-- 
cgit v1.2.3


From 7d5d5525bd88313e6fd90c0659665aee5114bc2d Mon Sep 17 00:00:00 2001
From: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Date: Thu, 17 Apr 2008 12:29:53 +0900
Subject: tcp md5sig: Share MD5 Signature option parser between IPv4 and IPv6.

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
---
 include/net/tcp.h    |  2 ++
 net/ipv4/tcp_input.c | 40 ++++++++++++++++++++++++++++++++++++++++
 net/ipv4/tcp_ipv4.c  | 42 +-----------------------------------------
 net/ipv6/tcp_ipv6.c  | 33 +--------------------------------
 4 files changed, 44 insertions(+), 73 deletions(-)

(limited to 'include')

diff --git a/include/net/tcp.h b/include/net/tcp.h
index 633147cb6bbc..b1c4b78f1cb8 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -399,6 +399,8 @@ extern void			tcp_parse_options(struct sk_buff *skb,
 						  struct tcp_options_received *opt_rx,
 						  int estab);
 
+extern u8			*tcp_parse_md5sig_option(struct tcphdr *th);
+
 /*
  *	TCP v4 functions exported for the inet6 API
  */
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index eba873e9b560..f331e67f2328 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -3450,6 +3450,43 @@ static int tcp_fast_parse_options(struct sk_buff *skb, struct tcphdr *th,
 	return 1;
 }
 
+#ifdef CONFIG_TCP_MD5SIG
+/*
+ * Parse MD5 Signature option
+ */
+u8 *tcp_parse_md5sig_option(struct tcphdr *th)
+{
+	int length = (th->doff << 2) - sizeof (*th);
+	u8 *ptr = (u8*)(th + 1);
+
+	/* If the TCP option is too short, we can short cut */
+	if (length < TCPOLEN_MD5SIG)
+		return NULL;
+
+	while (length > 0) {
+		int opcode = *ptr++;
+		int opsize;
+
+		switch(opcode) {
+		case TCPOPT_EOL:
+			return NULL;
+		case TCPOPT_NOP:
+			length--;
+			continue;
+		default:
+			opsize = *ptr++;
+			if (opsize < 2 || opsize > length)
+				return NULL;
+			if (opcode == TCPOPT_MD5SIG)
+				return ptr;
+		}
+		ptr += opsize - 2;
+		length -= opsize;
+	}
+	return NULL;
+}
+#endif
+
 static inline void tcp_store_ts_recent(struct tcp_sock *tp)
 {
 	tp->rx_opt.ts_recent = tp->rx_opt.rcv_tsval;
@@ -5467,6 +5504,9 @@ EXPORT_SYMBOL(sysctl_tcp_ecn);
 EXPORT_SYMBOL(sysctl_tcp_reordering);
 EXPORT_SYMBOL(sysctl_tcp_adv_win_scale);
 EXPORT_SYMBOL(tcp_parse_options);
+#ifdef CONFIG_TCP_MD5SIG
+EXPORT_SYMBOL(tcp_parse_md5sig_option);
+#endif
 EXPORT_SYMBOL(tcp_rcv_established);
 EXPORT_SYMBOL(tcp_rcv_state_process);
 EXPORT_SYMBOL(tcp_initialize_rcv_mss);
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index cd601a866c2f..56f550933644 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -1134,52 +1134,12 @@ static int tcp_v4_inbound_md5_hash(struct sock *sk, struct sk_buff *skb)
 	struct tcp_md5sig_key *hash_expected;
 	const struct iphdr *iph = ip_hdr(skb);
 	struct tcphdr *th = tcp_hdr(skb);
-	int length = (th->doff << 2) - sizeof(struct tcphdr);
 	int genhash;
-	unsigned char *ptr;
 	unsigned char newhash[16];
 
 	hash_expected = tcp_v4_md5_do_lookup(sk, iph->saddr);
+	hash_location = tcp_parse_md5sig_option(th);
 
-	/*
-	 * If the TCP option length is less than the TCP_MD5SIG
-	 * option length, then we can shortcut
-	 */
-	if (length < TCPOLEN_MD5SIG) {
-		if (hash_expected)
-			return 1;
-		else
-			return 0;
-	}
-
-	/* Okay, we can't shortcut - we have to grub through the options */
-	ptr = (unsigned char *)(th + 1);
-	while (length > 0) {
-		int opcode = *ptr++;
-		int opsize;
-
-		switch (opcode) {
-		case TCPOPT_EOL:
-			goto done_opts;
-		case TCPOPT_NOP:
-			length--;
-			continue;
-		default:
-			opsize = *ptr++;
-			if (opsize < 2)
-				goto done_opts;
-			if (opsize > length)
-				goto done_opts;
-
-			if (opcode == TCPOPT_MD5SIG) {
-				hash_location = ptr;
-				goto done_opts;
-			}
-		}
-		ptr += opsize-2;
-		length -= opsize;
-	}
-done_opts:
 	/* We've parsed the options - do we have a hash? */
 	if (!hash_expected && !hash_location)
 		return 0;
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index 715965f0fac0..dd4ddb30a3a9 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -844,43 +844,12 @@ static int tcp_v6_inbound_md5_hash (struct sock *sk, struct sk_buff *skb)
 	struct tcp_md5sig_key *hash_expected;
 	struct ipv6hdr *ip6h = ipv6_hdr(skb);
 	struct tcphdr *th = tcp_hdr(skb);
-	int length = (th->doff << 2) - sizeof (*th);
 	int genhash;
-	u8 *ptr;
 	u8 newhash[16];
 
 	hash_expected = tcp_v6_md5_do_lookup(sk, &ip6h->saddr);
+	hash_location = tcp_parse_md5sig_option(th);
 
-	/* If the TCP option is too short, we can short cut */
-	if (length < TCPOLEN_MD5SIG)
-		return hash_expected ? 1 : 0;
-
-	/* parse options */
-	ptr = (u8*)(th + 1);
-	while (length > 0) {
-		int opcode = *ptr++;
-		int opsize;
-
-		switch(opcode) {
-		case TCPOPT_EOL:
-			goto done_opts;
-		case TCPOPT_NOP:
-			length--;
-			continue;
-		default:
-			opsize = *ptr++;
-			if (opsize < 2 || opsize > length)
-				goto done_opts;
-			if (opcode == TCPOPT_MD5SIG) {
-				hash_location = ptr;
-				goto done_opts;
-			}
-		}
-		ptr += opsize - 2;
-		length -= opsize;
-	}
-
-done_opts:
 	/* do we have a hash as expected? */
 	if (!hash_expected) {
 		if (!hash_location)
-- 
cgit v1.2.3


From 076fb7223357769c39f3ddf900bba6752369c76a Mon Sep 17 00:00:00 2001
From: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Date: Thu, 17 Apr 2008 12:48:12 +0900
Subject: tcp md5sig: Remove redundant protocol argument.

Protocol is always TCP, so remove useless protocol argument.

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
---
 include/net/tcp.h     |  2 --
 net/ipv4/tcp_ipv4.c   | 20 ++++++++------------
 net/ipv4/tcp_output.c |  3 +--
 net/ipv6/tcp_ipv6.c   | 17 +++++++----------
 4 files changed, 16 insertions(+), 26 deletions(-)

(limited to 'include')

diff --git a/include/net/tcp.h b/include/net/tcp.h
index b1c4b78f1cb8..ae2549c45fc7 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -1123,7 +1123,6 @@ extern int			tcp_v4_calc_md5_hash(char *md5_hash,
 						     struct dst_entry *dst,
 						     struct request_sock *req,
 						     struct tcphdr *th,
-						     int protocol,
 						     unsigned int tcplen);
 extern struct tcp_md5sig_key	*tcp_v4_md5_lookup(struct sock *sk,
 						   struct sock *addr_sk);
@@ -1373,7 +1372,6 @@ struct tcp_sock_af_ops {
 						  struct dst_entry *dst,
 						  struct request_sock *req,
 						  struct tcphdr *th,
-						  int protocol,
 						  unsigned int len);
 	int			(*md5_add) (struct sock *sk,
 					    struct sock *addr_sk,
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 56f550933644..f25445d21bdf 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -95,8 +95,7 @@ static struct tcp_md5sig_key *tcp_v4_md5_do_lookup(struct sock *sk,
 						   __be32 addr);
 static int tcp_v4_do_calc_md5_hash(char *md5_hash, struct tcp_md5sig_key *key,
 				   __be32 saddr, __be32 daddr,
-				   struct tcphdr *th, int protocol,
-				   unsigned int tcplen);
+				   struct tcphdr *th, unsigned int tcplen);
 #endif
 
 struct inet_hashinfo __cacheline_aligned tcp_hashinfo = {
@@ -586,8 +585,7 @@ static void tcp_v4_send_reset(struct sock *sk, struct sk_buff *skb)
 					key,
 					ip_hdr(skb)->daddr,
 					ip_hdr(skb)->saddr,
-					&rep.th, IPPROTO_TCP,
-					arg.iov[0].iov_len);
+					&rep.th, arg.iov[0].iov_len);
 	}
 #endif
 	arg.csum = csum_tcpudp_nofold(ip_hdr(skb)->daddr,
@@ -680,8 +678,7 @@ static void tcp_v4_send_ack(struct tcp_timewait_sock *twsk,
 					key,
 					ip_hdr(skb)->daddr,
 					ip_hdr(skb)->saddr,
-					&rep.th, IPPROTO_TCP,
-					arg.iov[0].iov_len);
+					&rep.th, arg.iov[0].iov_len);
 	}
 #endif
 	arg.csum = csum_tcpudp_nofold(ip_hdr(skb)->daddr,
@@ -1006,7 +1003,7 @@ static int tcp_v4_parse_md5_keys(struct sock *sk, char __user *optval,
 
 static int tcp_v4_do_calc_md5_hash(char *md5_hash, struct tcp_md5sig_key *key,
 				   __be32 saddr, __be32 daddr,
-				   struct tcphdr *th, int protocol,
+				   struct tcphdr *th,
 				   unsigned int tcplen)
 {
 	struct scatterlist sg[4];
@@ -1039,7 +1036,7 @@ static int tcp_v4_do_calc_md5_hash(char *md5_hash, struct tcp_md5sig_key *key,
 	bp->saddr = saddr;
 	bp->daddr = daddr;
 	bp->pad = 0;
-	bp->protocol = protocol;
+	bp->protocol = IPPROTO_TCP;
 	bp->len = htons(tcplen);
 
 	sg_init_table(sg, 4);
@@ -1099,7 +1096,7 @@ int tcp_v4_calc_md5_hash(char *md5_hash, struct tcp_md5sig_key *key,
 			 struct sock *sk,
 			 struct dst_entry *dst,
 			 struct request_sock *req,
-			 struct tcphdr *th, int protocol,
+			 struct tcphdr *th,
 			 unsigned int tcplen)
 {
 	__be32 saddr, daddr;
@@ -1115,7 +1112,7 @@ int tcp_v4_calc_md5_hash(char *md5_hash, struct tcp_md5sig_key *key,
 	}
 	return tcp_v4_do_calc_md5_hash(md5_hash, key,
 				       saddr, daddr,
-				       th, protocol, tcplen);
+				       th, tcplen);
 }
 
 EXPORT_SYMBOL(tcp_v4_calc_md5_hash);
@@ -1166,8 +1163,7 @@ static int tcp_v4_inbound_md5_hash(struct sock *sk, struct sk_buff *skb)
 	genhash = tcp_v4_do_calc_md5_hash(newhash,
 					  hash_expected,
 					  iph->saddr, iph->daddr,
-					  th, sk->sk_protocol,
-					  skb->len);
+					  th, skb->len);
 
 	if (genhash || memcmp(hash_location, newhash, 16) != 0) {
 		if (net_ratelimit()) {
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index ad993ecb4810..f3ffd674edde 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -607,7 +607,6 @@ static int tcp_transmit_skb(struct sock *sk, struct sk_buff *skb, int clone_it,
 					       md5,
 					       sk, NULL, NULL,
 					       tcp_hdr(skb),
-					       sk->sk_protocol,
 					       skb->len);
 	}
 #endif
@@ -2266,7 +2265,7 @@ struct sk_buff *tcp_make_synack(struct sock *sk, struct dst_entry *dst,
 		tp->af_specific->calc_md5_hash(md5_hash_location,
 					       md5,
 					       NULL, dst, req,
-					       tcp_hdr(skb), sk->sk_protocol,
+					       tcp_hdr(skb),
 					       skb->len);
 	}
 #endif
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index dd4ddb30a3a9..334d21c23da9 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -736,8 +736,7 @@ static int tcp_v6_parse_md5_keys (struct sock *sk, char __user *optval,
 static int tcp_v6_do_calc_md5_hash(char *md5_hash, struct tcp_md5sig_key *key,
 				   struct in6_addr *saddr,
 				   struct in6_addr *daddr,
-				   struct tcphdr *th, int protocol,
-				   unsigned int tcplen)
+				   struct tcphdr *th, unsigned int tcplen)
 {
 	struct scatterlist sg[4];
 	__u16 data_len;
@@ -761,7 +760,7 @@ static int tcp_v6_do_calc_md5_hash(char *md5_hash, struct tcp_md5sig_key *key,
 	ipv6_addr_copy(&bp->saddr, saddr);
 	ipv6_addr_copy(&bp->daddr, daddr);
 	bp->len = htonl(tcplen);
-	bp->protocol = htonl(protocol);
+	bp->protocol = htonl(IPPROTO_TCP);
 
 	sg_init_table(sg, 4);
 
@@ -821,8 +820,7 @@ static int tcp_v6_calc_md5_hash(char *md5_hash, struct tcp_md5sig_key *key,
 				struct sock *sk,
 				struct dst_entry *dst,
 				struct request_sock *req,
-				struct tcphdr *th, int protocol,
-				unsigned int tcplen)
+				struct tcphdr *th, unsigned int tcplen)
 {
 	struct in6_addr *saddr, *daddr;
 
@@ -835,7 +833,7 @@ static int tcp_v6_calc_md5_hash(char *md5_hash, struct tcp_md5sig_key *key,
 	}
 	return tcp_v6_do_calc_md5_hash(md5_hash, key,
 				       saddr, daddr,
-				       th, protocol, tcplen);
+				       th, tcplen);
 }
 
 static int tcp_v6_inbound_md5_hash (struct sock *sk, struct sk_buff *skb)
@@ -879,8 +877,7 @@ static int tcp_v6_inbound_md5_hash (struct sock *sk, struct sk_buff *skb)
 	genhash = tcp_v6_do_calc_md5_hash(newhash,
 					  hash_expected,
 					  &ip6h->saddr, &ip6h->daddr,
-					  th, sk->sk_protocol,
-					  skb->len);
+					  th, skb->len);
 	if (genhash || memcmp(hash_location, newhash, 16) != 0) {
 		if (net_ratelimit()) {
 			printk(KERN_INFO "MD5 Hash %s for "
@@ -1020,7 +1017,7 @@ static void tcp_v6_send_reset(struct sock *sk, struct sk_buff *skb)
 		tcp_v6_do_calc_md5_hash((__u8 *)&opt[1], key,
 					&ipv6_hdr(skb)->daddr,
 					&ipv6_hdr(skb)->saddr,
-					t1, IPPROTO_TCP, tot_len);
+					t1, tot_len);
 	}
 #endif
 
@@ -1126,7 +1123,7 @@ static void tcp_v6_send_ack(struct tcp_timewait_sock *tw,
 		tcp_v6_do_calc_md5_hash((__u8 *)topt, key,
 					&ipv6_hdr(skb)->daddr,
 					&ipv6_hdr(skb)->saddr,
-					t1, IPPROTO_TCP, tot_len);
+					t1, tot_len);
 	}
 #endif
 
-- 
cgit v1.2.3


From 8d26d76dd4a4c87ef037a44a42a0608ffc730199 Mon Sep 17 00:00:00 2001
From: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Date: Thu, 17 Apr 2008 13:19:16 +0900
Subject: tcp md5sig: Share most of hash calcucaltion bits between IPv4 and
 IPv6.

We can share most part of the hash calculation code because
the only difference between IPv4 and IPv6 is their pseudo headers.

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
---
 include/net/tcp.h   |  7 ++++++
 net/ipv4/tcp.c      | 70 +++++++++++++++++++++++++++++++++++++++++++++++++++++
 net/ipv4/tcp_ipv4.c | 52 +++------------------------------------
 net/ipv6/tcp_ipv6.c | 53 ++++------------------------------------
 4 files changed, 86 insertions(+), 96 deletions(-)

(limited to 'include')

diff --git a/include/net/tcp.h b/include/net/tcp.h
index ae2549c45fc7..eac26b73bcc3 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -1117,6 +1117,13 @@ struct tcp_md5sig_pool {
 #define TCP_MD5SIG_MAXKEYS	(~(u32)0)	/* really?! */
 
 /* - functions */
+extern int			tcp_calc_md5_hash(char *md5_hash,
+						  struct tcp_md5sig_key *key,
+						  int bplen,
+						  struct tcphdr *th,
+						  unsigned int tcplen,
+						  struct tcp_md5sig_pool *hp);
+
 extern int			tcp_v4_calc_md5_hash(char *md5_hash,
 						     struct tcp_md5sig_key *key,
 						     struct sock *sk,
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index ab66683b8043..6efbae0e5512 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -2459,6 +2459,76 @@ static unsigned long tcp_md5sig_users;
 static struct tcp_md5sig_pool **tcp_md5sig_pool;
 static DEFINE_SPINLOCK(tcp_md5sig_pool_lock);
 
+int tcp_calc_md5_hash(char *md5_hash, struct tcp_md5sig_key *key,
+		      int bplen,
+		      struct tcphdr *th, unsigned int tcplen,
+		      struct tcp_md5sig_pool *hp)
+{
+	struct scatterlist sg[4];
+	__u16 data_len;
+	int block = 0;
+	__sum16 cksum;
+	struct hash_desc *desc = &hp->md5_desc;
+	int err;
+	unsigned int nbytes = 0;
+
+	sg_init_table(sg, 4);
+
+	/* 1. The TCP pseudo-header */
+	sg_set_buf(&sg[block++], &hp->md5_blk, bplen);
+	nbytes += bplen;
+
+	/* 2. The TCP header, excluding options, and assuming a
+	 * checksum of zero
+	 */
+	cksum = th->check;
+	th->check = 0;
+	sg_set_buf(&sg[block++], th, sizeof(*th));
+	nbytes += sizeof(*th);
+
+	/* 3. The TCP segment data (if any) */
+	data_len = tcplen - (th->doff << 2);
+	if (data_len > 0) {
+		u8 *data = (u8 *)th + (th->doff << 2);
+		sg_set_buf(&sg[block++], data, data_len);
+		nbytes += data_len;
+	}
+
+	/* 4. an independently-specified key or password, known to both
+	 * TCPs and presumably connection-specific
+	 */
+	sg_set_buf(&sg[block++], key->key, key->keylen);
+	nbytes += key->keylen;
+
+	sg_mark_end(&sg[block - 1]);
+
+	/* Now store the hash into the packet */
+	err = crypto_hash_init(desc);
+	if (err) {
+		if (net_ratelimit())
+			printk(KERN_WARNING "%s(): hash_init failed\n", __func__);
+		return -1;
+	}
+	err = crypto_hash_update(desc, sg, nbytes);
+	if (err) {
+		if (net_ratelimit())
+			printk(KERN_WARNING "%s(): hash_update failed\n", __func__);
+		return -1;
+	}
+	err = crypto_hash_final(desc, md5_hash);
+	if (err) {
+		if (net_ratelimit())
+			printk(KERN_WARNING "%s(): hash_final failed\n", __func__);
+		return -1;
+	}
+
+	/* Reset header */
+	th->check = cksum;
+
+	return 0;
+}
+EXPORT_SYMBOL(tcp_calc_md5_hash);
+
 static void __tcp_free_md5sig_pool(struct tcp_md5sig_pool **pool)
 {
 	int cpu;
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index f25445d21bdf..e331cdbd0953 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -1006,15 +1006,9 @@ static int tcp_v4_do_calc_md5_hash(char *md5_hash, struct tcp_md5sig_key *key,
 				   struct tcphdr *th,
 				   unsigned int tcplen)
 {
-	struct scatterlist sg[4];
-	__u16 data_len;
-	int block = 0;
-	__sum16 old_checksum;
 	struct tcp_md5sig_pool *hp;
 	struct tcp4_pseudohdr *bp;
-	struct hash_desc *desc;
 	int err;
-	unsigned int nbytes = 0;
 
 	/*
 	 * Okay, so RFC2385 is turned on for this connection,
@@ -1026,10 +1020,9 @@ static int tcp_v4_do_calc_md5_hash(char *md5_hash, struct tcp_md5sig_key *key,
 		goto clear_hash_noput;
 
 	bp = &hp->md5_blk.ip4;
-	desc = &hp->md5_desc;
 
 	/*
-	 * 1. the TCP pseudo-header (in the order: source IP address,
+	 * The TCP pseudo-header (in the order: source IP address,
 	 * destination IP address, zero-padded protocol number, and
 	 * segment length)
 	 */
@@ -1039,50 +1032,13 @@ static int tcp_v4_do_calc_md5_hash(char *md5_hash, struct tcp_md5sig_key *key,
 	bp->protocol = IPPROTO_TCP;
 	bp->len = htons(tcplen);
 
-	sg_init_table(sg, 4);
-
-	sg_set_buf(&sg[block++], bp, sizeof(*bp));
-	nbytes += sizeof(*bp);
-
-	/* 2. the TCP header, excluding options, and assuming a
-	 * checksum of zero/
-	 */
-	old_checksum = th->check;
-	th->check = 0;
-	sg_set_buf(&sg[block++], th, sizeof(struct tcphdr));
-	nbytes += sizeof(struct tcphdr);
-
-	/* 3. the TCP segment data (if any) */
-	data_len = tcplen - (th->doff << 2);
-	if (data_len > 0) {
-		unsigned char *data = (unsigned char *)th + (th->doff << 2);
-		sg_set_buf(&sg[block++], data, data_len);
-		nbytes += data_len;
-	}
-
-	/* 4. an independently-specified key or password, known to both
-	 * TCPs and presumably connection-specific
-	 */
-	sg_set_buf(&sg[block++], key->key, key->keylen);
-	nbytes += key->keylen;
-
-	sg_mark_end(&sg[block - 1]);
-
-	/* Now store the Hash into the packet */
-	err = crypto_hash_init(desc);
-	if (err)
-		goto clear_hash;
-	err = crypto_hash_update(desc, sg, nbytes);
-	if (err)
-		goto clear_hash;
-	err = crypto_hash_final(desc, md5_hash);
+	err = tcp_calc_md5_hash(md5_hash, key, sizeof(*bp),
+				th, tcplen, hp);
 	if (err)
 		goto clear_hash;
 
-	/* Reset header, and free up the crypto */
+	/* Free up the crypto pool */
 	tcp_put_md5sig_pool();
-	th->check = old_checksum;
-
 out:
 	return 0;
 clear_hash:
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index 334d21c23da9..0ae0311082fb 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -738,23 +738,17 @@ static int tcp_v6_do_calc_md5_hash(char *md5_hash, struct tcp_md5sig_key *key,
 				   struct in6_addr *daddr,
 				   struct tcphdr *th, unsigned int tcplen)
 {
-	struct scatterlist sg[4];
-	__u16 data_len;
-	int block = 0;
-	__sum16 cksum;
 	struct tcp_md5sig_pool *hp;
 	struct tcp6_pseudohdr *bp;
-	struct hash_desc *desc;
 	int err;
-	unsigned int nbytes = 0;
 
 	hp = tcp_get_md5sig_pool();
 	if (!hp) {
 		printk(KERN_WARNING "%s(): hash pool not found...\n", __func__);
 		goto clear_hash_noput;
 	}
+
 	bp = &hp->md5_blk.ip6;
-	desc = &hp->md5_desc;
 
 	/* 1. TCP pseudo-header (RFC2460) */
 	ipv6_addr_copy(&bp->saddr, saddr);
@@ -762,51 +756,14 @@ static int tcp_v6_do_calc_md5_hash(char *md5_hash, struct tcp_md5sig_key *key,
 	bp->len = htonl(tcplen);
 	bp->protocol = htonl(IPPROTO_TCP);
 
-	sg_init_table(sg, 4);
-
-	sg_set_buf(&sg[block++], bp, sizeof(*bp));
-	nbytes += sizeof(*bp);
-
-	/* 2. TCP header, excluding options */
-	cksum = th->check;
-	th->check = 0;
-	sg_set_buf(&sg[block++], th, sizeof(*th));
-	nbytes += sizeof(*th);
-
-	/* 3. TCP segment data (if any) */
-	data_len = tcplen - (th->doff << 2);
-	if (data_len > 0) {
-		u8 *data = (u8 *)th + (th->doff << 2);
-		sg_set_buf(&sg[block++], data, data_len);
-		nbytes += data_len;
-	}
-
-	/* 4. shared key */
-	sg_set_buf(&sg[block++], key->key, key->keylen);
-	nbytes += key->keylen;
-
-	sg_mark_end(&sg[block - 1]);
+	err = tcp_calc_md5_hash(md5_hash, key, sizeof(*bp),
+				th, tcplen, hp);
 
-	/* Now store the hash into the packet */
-	err = crypto_hash_init(desc);
-	if (err) {
-		printk(KERN_WARNING "%s(): hash_init failed\n", __func__);
-		goto clear_hash;
-	}
-	err = crypto_hash_update(desc, sg, nbytes);
-	if (err) {
-		printk(KERN_WARNING "%s(): hash_update failed\n", __func__);
-		goto clear_hash;
-	}
-	err = crypto_hash_final(desc, md5_hash);
-	if (err) {
-		printk(KERN_WARNING "%s(): hash_final failed\n", __func__);
+	if (err)
 		goto clear_hash;
-	}
 
-	/* Reset header, and free up the crypto */
+	/* Free up the crypto pool */
 	tcp_put_md5sig_pool();
-	th->check = cksum;
 out:
 	return 0;
 clear_hash:
-- 
cgit v1.2.3


From 9501f9722922f2e80e1f9dc6682311d65c2b5690 Mon Sep 17 00:00:00 2001
From: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Date: Fri, 18 Apr 2008 12:45:16 +0900
Subject: tcp md5sig: Let the caller pass appropriate key for
 tcp_v{4,6}_do_calc_md5_hash().

As we do for other socket/timewait-socket specific parameters,
let the callers pass appropriate arguments to
tcp_v{4,6}_do_calc_md5_hash().

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
---
 include/net/tcp.h   | 10 ++++++++++
 net/ipv4/tcp_ipv4.c | 50 ++++++++++++++++++++------------------------------
 net/ipv6/tcp_ipv6.c | 33 ++++++++++++---------------------
 3 files changed, 42 insertions(+), 51 deletions(-)

(limited to 'include')

diff --git a/include/net/tcp.h b/include/net/tcp.h
index eac26b73bcc3..07005ebb47a7 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -1142,6 +1142,16 @@ extern int			tcp_v4_md5_do_add(struct sock *sk,
 extern int			tcp_v4_md5_do_del(struct sock *sk,
 						  __be32 addr);
 
+#ifdef CONFIG_TCP_MD5SIG
+#define tcp_twsk_md5_key(twsk)	((twsk)->tw_md5_keylen ? 		 \
+				 &(struct tcp_md5sig_key) {		 \
+					.key = (twsk)->tw_md5_key,	 \
+					.keylen = (twsk)->tw_md5_keylen, \
+				} : NULL)
+#else
+#define tcp_twsk_md5_key(twsk)	NULL
+#endif
+
 extern struct tcp_md5sig_pool	**tcp_alloc_md5sig_pool(void);
 extern void			tcp_free_md5sig_pool(void);
 
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index e331cdbd0953..f7ff2a64a7f0 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -96,6 +96,12 @@ static struct tcp_md5sig_key *tcp_v4_md5_do_lookup(struct sock *sk,
 static int tcp_v4_do_calc_md5_hash(char *md5_hash, struct tcp_md5sig_key *key,
 				   __be32 saddr, __be32 daddr,
 				   struct tcphdr *th, unsigned int tcplen);
+#else
+static inline
+struct tcp_md5sig_key *tcp_v4_md5_do_lookup(struct sock *sk, __be32 addr)
+{
+	return NULL;
+}
 #endif
 
 struct inet_hashinfo __cacheline_aligned tcp_hashinfo = {
@@ -604,9 +610,9 @@ static void tcp_v4_send_reset(struct sock *sk, struct sk_buff *skb)
    outside socket context is ugly, certainly. What can I do?
  */
 
-static void tcp_v4_send_ack(struct tcp_timewait_sock *twsk,
-			    struct sk_buff *skb, u32 seq, u32 ack,
-			    u32 win, u32 ts)
+static void tcp_v4_send_ack(struct sk_buff *skb, u32 seq, u32 ack,
+			    u32 win, u32 ts, int oif,
+			    struct tcp_md5sig_key *key)
 {
 	struct tcphdr *th = tcp_hdr(skb);
 	struct {
@@ -618,10 +624,6 @@ static void tcp_v4_send_ack(struct tcp_timewait_sock *twsk,
 			];
 	} rep;
 	struct ip_reply_arg arg;
-#ifdef CONFIG_TCP_MD5SIG
-	struct tcp_md5sig_key *key;
-	struct tcp_md5sig_key tw_key;
-#endif
 
 	memset(&rep.th, 0, sizeof(struct tcphdr));
 	memset(&arg, 0, sizeof(arg));
@@ -647,23 +649,6 @@ static void tcp_v4_send_ack(struct tcp_timewait_sock *twsk,
 	rep.th.window  = htons(win);
 
 #ifdef CONFIG_TCP_MD5SIG
-	/*
-	 * The SKB holds an imcoming packet, but may not have a valid ->sk
-	 * pointer. This is especially the case when we're dealing with a
-	 * TIME_WAIT ack, because the sk structure is long gone, and only
-	 * the tcp_timewait_sock remains. So the md5 key is stashed in that
-	 * structure, and we use it in preference.  I believe that (twsk ||
-	 * skb->sk) holds true, but we program defensively.
-	 */
-	if (!twsk && skb->sk) {
-		key = tcp_v4_md5_do_lookup(skb->sk, ip_hdr(skb)->daddr);
-	} else if (twsk && twsk->tw_md5_keylen) {
-		tw_key.key = twsk->tw_md5_key;
-		tw_key.keylen = twsk->tw_md5_keylen;
-		key = &tw_key;
-	} else
-		key = NULL;
-
 	if (key) {
 		int offset = (ts) ? 3 : 0;
 
@@ -685,8 +670,8 @@ static void tcp_v4_send_ack(struct tcp_timewait_sock *twsk,
 				      ip_hdr(skb)->saddr, /* XXX */
 				      arg.iov[0].iov_len, IPPROTO_TCP, 0);
 	arg.csumoffset = offsetof(struct tcphdr, check) / 2;
-	if (twsk)
-		arg.bound_dev_if = twsk->tw_sk.tw_bound_dev_if;
+	if (oif)
+		arg.bound_dev_if = oif;
 
 	ip_send_reply(dev_net(skb->dev)->ipv4.tcp_sock, skb,
 		      &arg, arg.iov[0].iov_len);
@@ -699,9 +684,12 @@ static void tcp_v4_timewait_ack(struct sock *sk, struct sk_buff *skb)
 	struct inet_timewait_sock *tw = inet_twsk(sk);
 	struct tcp_timewait_sock *tcptw = tcp_twsk(sk);
 
-	tcp_v4_send_ack(tcptw, skb, tcptw->tw_snd_nxt, tcptw->tw_rcv_nxt,
+	tcp_v4_send_ack(skb, tcptw->tw_snd_nxt, tcptw->tw_rcv_nxt,
 			tcptw->tw_rcv_wnd >> tw->tw_rcv_wscale,
-			tcptw->tw_ts_recent);
+			tcptw->tw_ts_recent,
+			tw->tw_bound_dev_if,
+			tcp_twsk_md5_key(tcptw)
+			);
 
 	inet_twsk_put(tw);
 }
@@ -709,9 +697,11 @@ static void tcp_v4_timewait_ack(struct sock *sk, struct sk_buff *skb)
 static void tcp_v4_reqsk_send_ack(struct sk_buff *skb,
 				  struct request_sock *req)
 {
-	tcp_v4_send_ack(NULL, skb, tcp_rsk(req)->snt_isn + 1,
+	tcp_v4_send_ack(skb, tcp_rsk(req)->snt_isn + 1,
 			tcp_rsk(req)->rcv_isn + 1, req->rcv_wnd,
-			req->ts_recent);
+			req->ts_recent,
+			0,
+			tcp_v4_md5_do_lookup(skb->sk, ip_hdr(skb)->daddr));
 }
 
 /*
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index 0ae0311082fb..ecdbb9f46541 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -82,6 +82,12 @@ static struct inet_connection_sock_af_ops ipv6_specific;
 #ifdef CONFIG_TCP_MD5SIG
 static struct tcp_sock_af_ops tcp_sock_ipv6_specific;
 static struct tcp_sock_af_ops tcp_sock_ipv6_mapped_specific;
+#else
+static struct tcp_md5sig_key *tcp_v6_md5_do_lookup(struct sock *sk,
+						   struct in6_addr *addr)
+{
+	return NULL;
+}
 #endif
 
 static void tcp_v6_hash(struct sock *sk)
@@ -1011,8 +1017,8 @@ static void tcp_v6_send_reset(struct sock *sk, struct sk_buff *skb)
 	kfree_skb(buff);
 }
 
-static void tcp_v6_send_ack(struct tcp_timewait_sock *tw,
-			    struct sk_buff *skb, u32 seq, u32 ack, u32 win, u32 ts)
+static void tcp_v6_send_ack(struct sk_buff *skb, u32 seq, u32 ack, u32 win, u32 ts,
+			    struct tcp_md5sig_key *key)
 {
 	struct tcphdr *th = tcp_hdr(skb), *t1;
 	struct sk_buff *buff;
@@ -1021,22 +1027,6 @@ static void tcp_v6_send_ack(struct tcp_timewait_sock *tw,
 	struct sock *ctl_sk = net->ipv6.tcp_sk;
 	unsigned int tot_len = sizeof(struct tcphdr);
 	__be32 *topt;
-#ifdef CONFIG_TCP_MD5SIG
-	struct tcp_md5sig_key *key;
-	struct tcp_md5sig_key tw_key;
-#endif
-
-#ifdef CONFIG_TCP_MD5SIG
-	if (!tw && skb->sk) {
-		key = tcp_v6_md5_do_lookup(skb->sk, &ipv6_hdr(skb)->daddr);
-	} else if (tw && tw->tw_md5_keylen) {
-		tw_key.key = tw->tw_md5_key;
-		tw_key.keylen = tw->tw_md5_keylen;
-		key = &tw_key;
-	} else {
-		key = NULL;
-	}
-#endif
 
 	if (ts)
 		tot_len += TCPOLEN_TSTAMP_ALIGNED;
@@ -1116,16 +1106,17 @@ static void tcp_v6_timewait_ack(struct sock *sk, struct sk_buff *skb)
 	struct inet_timewait_sock *tw = inet_twsk(sk);
 	struct tcp_timewait_sock *tcptw = tcp_twsk(sk);
 
-	tcp_v6_send_ack(tcptw, skb, tcptw->tw_snd_nxt, tcptw->tw_rcv_nxt,
+	tcp_v6_send_ack(skb, tcptw->tw_snd_nxt, tcptw->tw_rcv_nxt,
 			tcptw->tw_rcv_wnd >> tw->tw_rcv_wscale,
-			tcptw->tw_ts_recent);
+			tcptw->tw_ts_recent, tcp_twsk_md5_key(tcptw));
 
 	inet_twsk_put(tw);
 }
 
 static void tcp_v6_reqsk_send_ack(struct sk_buff *skb, struct request_sock *req)
 {
-	tcp_v6_send_ack(NULL, skb, tcp_rsk(req)->snt_isn + 1, tcp_rsk(req)->rcv_isn + 1, req->rcv_wnd, req->ts_recent);
+	tcp_v6_send_ack(skb, tcp_rsk(req)->snt_isn + 1, tcp_rsk(req)->rcv_isn + 1, req->rcv_wnd, req->ts_recent,
+			tcp_v6_md5_do_lookup(skb->sk, &ipv6_hdr(skb)->daddr));
 }
 
 
-- 
cgit v1.2.3


From 12c03f59c3909159010b87a926f5626d4380d441 Mon Sep 17 00:00:00 2001
From: Magnus Damm <magnus.damm@gmail.com>
Date: Mon, 9 Jun 2008 16:33:55 -0700
Subject: smc911x: introduce platform data flags

This patch adds a new header file for platform data information
together with code that adds run time bus width and irq flag support.

Signed-off-by: Magnus Damm <damm@igel.co.jp>
Cc: Jeff Garzik <jeff@garzik.org>
Cc: Paul Mundt <lethal@linux-sh.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
---
 drivers/net/smc911x.c   | 17 ++++++++++-
 drivers/net/smc911x.h   | 81 +++++++++++++++++++++++++++++++++++++++++++++++++
 include/linux/smc911x.h | 12 ++++++++
 3 files changed, 109 insertions(+), 1 deletion(-)
 create mode 100644 include/linux/smc911x.h

(limited to 'include')

diff --git a/drivers/net/smc911x.c b/drivers/net/smc911x.c
index 2ca4db85f938..fc605f276c00 100644
--- a/drivers/net/smc911x.c
+++ b/drivers/net/smc911x.c
@@ -1819,6 +1819,7 @@ static int __init smc911x_probe(struct net_device *dev)
 	int i, retval;
 	unsigned int val, chip_id, revision;
 	const char *version_string;
+	unsigned long irq_flags;
 
 	DBG(SMC_DEBUG_FUNC, "%s: --> %s\n", dev->name, __FUNCTION__);
 
@@ -1985,9 +1986,15 @@ static int __init smc911x_probe(struct net_device *dev)
 	lp->ctl_rfduplx = 1;
 	lp->ctl_rspeed = 100;
 
+#ifdef SMC_DYNAMIC_BUS_CONFIG
+	irq_flags = lp->cfg.irq_flags;
+#else
+	irq_flags = IRQF_SHARED | SMC_IRQ_SENSE;
+#endif
+
 	/* Grab the IRQ */
 	retval = request_irq(dev->irq, &smc911x_interrupt,
-			IRQF_SHARED | SMC_IRQ_SENSE, dev->name, dev);
+			     irq_flags, dev->name, dev);
 	if (retval)
 		goto err_out;
 
@@ -2057,6 +2064,7 @@ err_out:
  */
 static int smc911x_drv_probe(struct platform_device *pdev)
 {
+	struct smc91x_platdata *pd = pdev->dev.platform_data;
 	struct net_device *ndev;
 	struct resource *res;
 	struct smc911x_local *lp;
@@ -2090,6 +2098,13 @@ static int smc911x_drv_probe(struct platform_device *pdev)
 	ndev->irq = platform_get_irq(pdev, 0);
 	lp = netdev_priv(ndev);
 	lp->netdev = ndev;
+#ifdef SMC_DYNAMIC_BUS_CONFIG
+	if (!pd) {
+		ret = -EINVAL;
+		goto release_both;
+	}
+	memcpy(&lp->cfg, pd, sizeof(lp->cfg));
+#endif
 
 	addr = ioremap(res->start, SMC911X_IO_EXTENT);
 	if (!addr) {
diff --git a/drivers/net/smc911x.h b/drivers/net/smc911x.h
index 271a3e8cf683..266232271a72 100644
--- a/drivers/net/smc911x.h
+++ b/drivers/net/smc911x.h
@@ -29,6 +29,7 @@
 #ifndef _SMC911X_H_
 #define _SMC911X_H_
 
+#include <linux/smc911x.h>
 /*
  * Use the DMA feature on PXA chips
  */
@@ -42,6 +43,12 @@
   #define SMC_USE_16BIT		0
   #define SMC_USE_32BIT		1
   #define SMC_IRQ_SENSE		IRQF_TRIGGER_LOW
+#else
+/*
+ * Default configuration
+ */
+
+#define SMC_DYNAMIC_BUS_CONFIG
 #endif
 
 /* store this information for the driver.. */
@@ -92,12 +99,84 @@ struct smc911x_local {
 	struct device *dev;
 #endif
 	void __iomem *base;
+#ifdef SMC_DYNAMIC_BUS_CONFIG
+	struct smc911x_platdata cfg;
+#endif
 };
 
 /*
  * Define the bus width specific IO macros
  */
 
+#ifdef SMC_DYNAMIC_BUS_CONFIG
+static inline unsigned int SMC_inl(struct smc911x_local *lp, int reg)
+{
+	void __iomem *ioaddr = lp->base + reg;
+
+	if (lp->cfg.flags & SMC911X_USE_32BIT)
+		return readl(ioaddr);
+
+	if (lp->cfg.flags & SMC911X_USE_16BIT)
+		return readw(ioaddr) | (readw(ioaddr + 2) << 16);
+
+	BUG();
+}
+
+static inline void SMC_outl(unsigned int value, struct smc911x_local *lp,
+			    int reg)
+{
+	void __iomem *ioaddr = lp->base + reg;
+
+	if (lp->cfg.flags & SMC911X_USE_32BIT) {
+		writel(value, ioaddr);
+		return;
+	}
+
+	if (lp->cfg.flags & SMC911X_USE_16BIT) {
+		writew(value & 0xffff, ioaddr);
+		writew(value >> 16, ioaddr + 2);
+		return;
+	}
+
+	BUG();
+}
+
+static inline void SMC_insl(struct smc911x_local *lp, int reg,
+			      void *addr, unsigned int count)
+{
+	void __iomem *ioaddr = lp->base + reg;
+
+	if (lp->cfg.flags & SMC911X_USE_32BIT) {
+		readsl(ioaddr, addr, count);
+		return;
+	}
+
+	if (lp->cfg.flags & SMC911X_USE_16BIT) {
+		readsw(ioaddr, addr, count * 2);
+		return;
+	}
+
+	BUG();
+}
+
+static inline void SMC_outsl(struct smc911x_local *lp, int reg,
+			     void *addr, unsigned int count)
+{
+	void __iomem *ioaddr = lp->base + reg;
+
+	if (lp->cfg.flags & SMC911X_USE_32BIT) {
+		writesl(ioaddr, addr, count);
+		return;
+	}
+
+	if (lp->cfg.flags & SMC911X_USE_16BIT) {
+		writesw(ioaddr, addr, count * 2);
+		return;
+	}
+
+	BUG();
+}
+#else
 #if	SMC_USE_16BIT
 #define SMC_inl(lp, r)		 ((readw((lp)->base + (r)) & 0xFFFF) + (readw((lp)->base + (r) + 2) << 16))
 #define SMC_outl(v, lp, r) 			 \
@@ -115,6 +194,8 @@ struct smc911x_local {
 #define SMC_outsl(lp, r, p, l)	 writesl((int*)((lp)->base + (r)), p, l)
 
 #endif /* SMC_USE_16BIT */
+#endif /* SMC_DYNAMIC_BUS_CONFIG */
+
 
 #ifdef SMC_USE_PXA_DMA
 #define SMC_USE_DMA
diff --git a/include/linux/smc911x.h b/include/linux/smc911x.h
new file mode 100644
index 000000000000..b58f54c24183
--- /dev/null
+++ b/include/linux/smc911x.h
@@ -0,0 +1,12 @@
+#ifndef __SMC911X_H__
+#define __SMC911X_H__
+
+#define SMC911X_USE_16BIT (1 << 0)
+#define SMC911X_USE_32BIT (1 << 1)
+
+struct smc911x_platdata {
+	unsigned long flags;
+	unsigned long irq_flags; /* IRQF_... */
+};
+
+#endif /* __SMC911X_H__ */
-- 
cgit v1.2.3


From 0b040829952d84bf2a62526f0e24b624e0699447 Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Tue, 10 Jun 2008 22:46:50 -0700
Subject: net: remove CVS keywords

This patch removes CVS keywords that weren't updated for a long time
from comments.

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/if_bridge.h          | 2 --
 include/linux/if_ppp.h             | 2 --
 include/linux/if_tun.h             | 2 --
 include/linux/ip6_tunnel.h         | 4 ----
 include/linux/ppp-comp.h           | 2 --
 include/linux/ppp_defs.h           | 2 --
 include/linux/sunrpc/auth_gss.h    | 2 --
 include/linux/sunrpc/gss_api.h     | 2 --
 include/linux/sunrpc/svcauth_gss.h | 3 ---
 include/net/inetpeer.h             | 2 --
 include/net/ip6_tunnel.h           | 4 ----
 include/net/ipconfig.h             | 2 --
 include/net/ipv6.h                 | 2 --
 include/net/snmp.h                 | 2 --
 net/bluetooth/bnep/bnep.h          | 4 ----
 net/bluetooth/bnep/core.c          | 4 ----
 net/bluetooth/bnep/netdev.c        | 4 ----
 net/bluetooth/bnep/sock.c          | 4 ----
 net/bluetooth/rfcomm/core.c        | 2 --
 net/bluetooth/rfcomm/sock.c        | 2 --
 net/bluetooth/rfcomm/tty.c         | 2 --
 net/bridge/br.c                    | 2 --
 net/bridge/br_device.c             | 2 --
 net/bridge/br_fdb.c                | 2 --
 net/bridge/br_forward.c            | 2 --
 net/bridge/br_if.c                 | 2 --
 net/bridge/br_input.c              | 2 --
 net/bridge/br_ioctl.c              | 2 --
 net/bridge/br_notify.c             | 2 --
 net/bridge/br_private.h            | 2 --
 net/bridge/br_private_stp.h        | 2 --
 net/bridge/br_stp.c                | 2 --
 net/bridge/br_stp_bpdu.c           | 2 --
 net/bridge/br_stp_if.c             | 2 --
 net/bridge/br_stp_timer.c          | 2 --
 net/core/skbuff.c                  | 2 --
 net/core/sock.c                    | 2 --
 net/ipv4/af_inet.c                 | 2 --
 net/ipv4/arp.c                     | 2 --
 net/ipv4/devinet.c                 | 2 --
 net/ipv4/fib_frontend.c            | 2 --
 net/ipv4/fib_hash.c                | 2 --
 net/ipv4/fib_semantics.c           | 2 --
 net/ipv4/fib_trie.c                | 2 --
 net/ipv4/icmp.c                    | 2 --
 net/ipv4/igmp.c                    | 2 --
 net/ipv4/inet_diag.c               | 2 --
 net/ipv4/inetpeer.c                | 2 --
 net/ipv4/ip_forward.c              | 2 --
 net/ipv4/ip_fragment.c             | 2 --
 net/ipv4/ip_input.c                | 2 --
 net/ipv4/ip_options.c              | 2 --
 net/ipv4/ip_output.c               | 2 --
 net/ipv4/ip_sockglue.c             | 2 --
 net/ipv4/ipconfig.c                | 2 --
 net/ipv4/ipip.c                    | 2 --
 net/ipv4/ipmr.c                    | 2 --
 net/ipv4/ipvs/ip_vs_app.c          | 2 --
 net/ipv4/ipvs/ip_vs_conn.c         | 2 --
 net/ipv4/ipvs/ip_vs_core.c         | 2 --
 net/ipv4/ipvs/ip_vs_ctl.c          | 2 --
 net/ipv4/ipvs/ip_vs_dh.c           | 2 --
 net/ipv4/ipvs/ip_vs_est.c          | 2 --
 net/ipv4/ipvs/ip_vs_ftp.c          | 2 --
 net/ipv4/ipvs/ip_vs_lblc.c         | 2 --
 net/ipv4/ipvs/ip_vs_lblcr.c        | 2 --
 net/ipv4/ipvs/ip_vs_lc.c           | 2 --
 net/ipv4/ipvs/ip_vs_nq.c           | 2 --
 net/ipv4/ipvs/ip_vs_proto.c        | 2 --
 net/ipv4/ipvs/ip_vs_proto_ah.c     | 2 --
 net/ipv4/ipvs/ip_vs_proto_esp.c    | 2 --
 net/ipv4/ipvs/ip_vs_proto_tcp.c    | 2 --
 net/ipv4/ipvs/ip_vs_proto_udp.c    | 2 --
 net/ipv4/ipvs/ip_vs_rr.c           | 2 --
 net/ipv4/ipvs/ip_vs_sched.c        | 2 --
 net/ipv4/ipvs/ip_vs_sed.c          | 2 --
 net/ipv4/ipvs/ip_vs_sh.c           | 2 --
 net/ipv4/ipvs/ip_vs_sync.c         | 2 --
 net/ipv4/ipvs/ip_vs_wlc.c          | 2 --
 net/ipv4/ipvs/ip_vs_wrr.c          | 2 --
 net/ipv4/ipvs/ip_vs_xmit.c         | 2 --
 net/ipv4/proc.c                    | 2 --
 net/ipv4/protocol.c                | 2 --
 net/ipv4/raw.c                     | 2 --
 net/ipv4/route.c                   | 2 --
 net/ipv4/syncookies.c              | 2 --
 net/ipv4/sysctl_net_ipv4.c         | 2 --
 net/ipv4/tcp.c                     | 2 --
 net/ipv4/tcp_diag.c                | 2 --
 net/ipv4/tcp_input.c               | 2 --
 net/ipv4/tcp_ipv4.c                | 2 --
 net/ipv4/tcp_minisocks.c           | 2 --
 net/ipv4/tcp_output.c              | 2 --
 net/ipv4/tcp_timer.c               | 2 --
 net/ipv4/udp.c                     | 2 --
 net/ipv4/udplite.c                 | 2 --
 net/ipv6/addrconf.c                | 2 --
 net/ipv6/af_inet6.c                | 2 --
 net/ipv6/datagram.c                | 2 --
 net/ipv6/exthdrs.c                 | 2 --
 net/ipv6/icmp.c                    | 2 --
 net/ipv6/ip6_fib.c                 | 2 --
 net/ipv6/ip6_input.c               | 2 --
 net/ipv6/ip6_output.c              | 2 --
 net/ipv6/ip6_tunnel.c              | 2 --
 net/ipv6/ipv6_sockglue.c           | 2 --
 net/ipv6/mcast.c                   | 2 --
 net/ipv6/proc.c                    | 2 --
 net/ipv6/protocol.c                | 2 --
 net/ipv6/raw.c                     | 2 --
 net/ipv6/reassembly.c              | 2 --
 net/ipv6/route.c                   | 2 --
 net/ipv6/sit.c                     | 2 --
 net/ipv6/tcp_ipv6.c                | 2 --
 net/ipv6/udp.c                     | 2 --
 net/ipv6/udplite.c                 | 2 --
 net/packet/af_packet.c             | 2 --
 net/sched/sch_htb.c                | 2 --
 net/sunrpc/auth_gss/auth_gss.c     | 2 --
 net/sysctl_net.c                   | 1 -
 net/unix/af_unix.c                 | 2 --
 121 files changed, 254 deletions(-)

(limited to 'include')

diff --git a/include/linux/if_bridge.h b/include/linux/if_bridge.h
index 950e13d09e06..6badb3e2c4e4 100644
--- a/include/linux/if_bridge.h
+++ b/include/linux/if_bridge.h
@@ -4,8 +4,6 @@
  *	Authors:
  *	Lennert Buytenhek		<buytenh@gnu.org>
  *
- *	$Id: if_bridge.h,v 1.1 2000/02/18 16:47:01 davem Exp $
- *
  *	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
diff --git a/include/linux/if_ppp.h b/include/linux/if_ppp.h
index 0f2f70d4e48c..c3b1f8562709 100644
--- a/include/linux/if_ppp.h
+++ b/include/linux/if_ppp.h
@@ -1,5 +1,3 @@
-/*	$Id: if_ppp.h,v 1.21 2000/03/27 06:03:36 paulus Exp $	*/
-
 /*
  * if_ppp.h - Point-to-Point Protocol definitions.
  *
diff --git a/include/linux/if_tun.h b/include/linux/if_tun.h
index 8c71fe2fb1f5..18f31b6187a3 100644
--- a/include/linux/if_tun.h
+++ b/include/linux/if_tun.h
@@ -11,8 +11,6 @@
  *  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.
- *
- *  $Id: if_tun.h,v 1.2 2001/06/01 18:39:47 davem Exp $
  */
 
 #ifndef __IF_TUN_H
diff --git a/include/linux/ip6_tunnel.h b/include/linux/ip6_tunnel.h
index af3f4a70f3df..1e7cc4af40de 100644
--- a/include/linux/ip6_tunnel.h
+++ b/include/linux/ip6_tunnel.h
@@ -1,7 +1,3 @@
-/*
- * $Id$
- */
-
 #ifndef _IP6_TUNNEL_H
 #define _IP6_TUNNEL_H
 
diff --git a/include/linux/ppp-comp.h b/include/linux/ppp-comp.h
index e86a7a5cf355..b8d4ddd22736 100644
--- a/include/linux/ppp-comp.h
+++ b/include/linux/ppp-comp.h
@@ -23,8 +23,6 @@
  * ON AN "AS IS" BASIS, AND THE AUSTRALIAN NATIONAL UNIVERSITY HAS NO
  * OBLIGATION TO PROVIDE MAINTENANCE, SUPPORT, UPDATES, ENHANCEMENTS,
  * OR MODIFICATIONS.
- *
- * $Id: ppp-comp.h,v 1.6 1997/11/27 06:04:44 paulus Exp $
  */
 
 /*
diff --git a/include/linux/ppp_defs.h b/include/linux/ppp_defs.h
index c6b13ff85028..6e8adc77522c 100644
--- a/include/linux/ppp_defs.h
+++ b/include/linux/ppp_defs.h
@@ -1,5 +1,3 @@
-/*	$Id: ppp_defs.h,v 1.2 1994/09/21 01:31:06 paulus Exp $	*/
-
 /*
  * ppp_defs.h - PPP definitions.
  *
diff --git a/include/linux/sunrpc/auth_gss.h b/include/linux/sunrpc/auth_gss.h
index fec6899bf355..d48d4e605f74 100644
--- a/include/linux/sunrpc/auth_gss.h
+++ b/include/linux/sunrpc/auth_gss.h
@@ -7,8 +7,6 @@
  * Andy Adamson <andros@umich.edu>
  * Bruce Fields <bfields@umich.edu>
  * Copyright (c) 2000 The Regents of the University of Michigan
- *
- * $Id$
  */
 
 #ifndef _LINUX_SUNRPC_AUTH_GSS_H
diff --git a/include/linux/sunrpc/gss_api.h b/include/linux/sunrpc/gss_api.h
index 459c5fc11d51..03f33330ece2 100644
--- a/include/linux/sunrpc/gss_api.h
+++ b/include/linux/sunrpc/gss_api.h
@@ -7,8 +7,6 @@
  * Andy Adamson <andros@umich.edu>
  * Bruce Fields <bfields@umich.edu>
  * Copyright (c) 2000 The Regents of the University of Michigan
- *
- * $Id$
  */
 
 #ifndef _LINUX_SUNRPC_GSS_API_H
diff --git a/include/linux/sunrpc/svcauth_gss.h b/include/linux/sunrpc/svcauth_gss.h
index 417a1def56db..c9165d9771a8 100644
--- a/include/linux/sunrpc/svcauth_gss.h
+++ b/include/linux/sunrpc/svcauth_gss.h
@@ -3,9 +3,6 @@
  *
  * Bruce Fields <bfields@umich.edu>
  * Copyright (c) 2002 The Regents of the Unviersity of Michigan
- *
- * $Id$
- *
  */
 
 #ifndef _LINUX_SUNRPC_SVCAUTH_GSS_H
diff --git a/include/net/inetpeer.h b/include/net/inetpeer.h
index ad8404b56113..15e1f8fe4c1f 100644
--- a/include/net/inetpeer.h
+++ b/include/net/inetpeer.h
@@ -1,8 +1,6 @@
 /*
  *		INETPEER - A storage for permanent information about peers
  *
- *  Version:	$Id: inetpeer.h,v 1.2 2002/01/12 07:54:56 davem Exp $
- *
  *  Authors:	Andrey V. Savochkin <saw@msu.ru>
  */
 
diff --git a/include/net/ip6_tunnel.h b/include/net/ip6_tunnel.h
index 3780592ebe8e..83b4e008b16d 100644
--- a/include/net/ip6_tunnel.h
+++ b/include/net/ip6_tunnel.h
@@ -1,7 +1,3 @@
-/*
- * $Id$
- */
-
 #ifndef _NET_IP6_TUNNEL_H
 #define _NET_IP6_TUNNEL_H
 
diff --git a/include/net/ipconfig.h b/include/net/ipconfig.h
index 3924d7d2cb11..c74cc1bd5a02 100644
--- a/include/net/ipconfig.h
+++ b/include/net/ipconfig.h
@@ -1,6 +1,4 @@
 /*
- *  $Id: ipconfig.h,v 1.4 2001/04/30 04:51:46 davem Exp $
- *
  *  Copyright (C) 1997 Martin Mares
  *
  *  Automatic IP Layer Configuration
diff --git a/include/net/ipv6.h b/include/net/ipv6.h
index e0a612bc9c4e..7f7db8d57934 100644
--- a/include/net/ipv6.h
+++ b/include/net/ipv6.h
@@ -4,8 +4,6 @@
  *	Authors:
  *	Pedro Roque		<roque@di.fc.ul.pt>
  *
- *	$Id: ipv6.h,v 1.1 2002/05/20 15:13:07 jgrimm Exp $
- *
  *	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
diff --git a/include/net/snmp.h b/include/net/snmp.h
index ce2f48507510..57c93628695f 100644
--- a/include/net/snmp.h
+++ b/include/net/snmp.h
@@ -14,8 +14,6 @@
  *		as published by the Free Software Foundation; either version
  *		2 of the License, or (at your option) any later version.
  *
- *		$Id: snmp.h,v 1.19 2001/06/14 13:40:46 davem Exp $
- *
  */
  
 #ifndef _SNMP_H
diff --git a/net/bluetooth/bnep/bnep.h b/net/bluetooth/bnep/bnep.h
index e69244dd8de8..b69bf4e7c48b 100644
--- a/net/bluetooth/bnep/bnep.h
+++ b/net/bluetooth/bnep/bnep.h
@@ -16,10 +16,6 @@
   Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307 USA
 */
 
-/*
- * $Id: bnep.h,v 1.5 2002/08/04 21:23:58 maxk Exp $
- */
-
 #ifndef _BNEP_H
 #define _BNEP_H
 
diff --git a/net/bluetooth/bnep/core.c b/net/bluetooth/bnep/core.c
index f85d94643aaf..1d98a1b80da7 100644
--- a/net/bluetooth/bnep/core.c
+++ b/net/bluetooth/bnep/core.c
@@ -25,10 +25,6 @@
    SOFTWARE IS DISCLAIMED.
 */
 
-/*
- * $Id: core.c,v 1.20 2002/08/04 21:23:58 maxk Exp $
- */
-
 #include <linux/module.h>
 
 #include <linux/kernel.h>
diff --git a/net/bluetooth/bnep/netdev.c b/net/bluetooth/bnep/netdev.c
index 95e3837e4312..d9fa0ab2c87f 100644
--- a/net/bluetooth/bnep/netdev.c
+++ b/net/bluetooth/bnep/netdev.c
@@ -25,10 +25,6 @@
    SOFTWARE IS DISCLAIMED.
 */
 
-/*
- * $Id: netdev.c,v 1.8 2002/08/04 21:23:58 maxk Exp $
- */
-
 #include <linux/module.h>
 
 #include <linux/socket.h>
diff --git a/net/bluetooth/bnep/sock.c b/net/bluetooth/bnep/sock.c
index 201e5b1ce473..8ffb57f2303a 100644
--- a/net/bluetooth/bnep/sock.c
+++ b/net/bluetooth/bnep/sock.c
@@ -24,10 +24,6 @@
    SOFTWARE IS DISCLAIMED.
 */
 
-/*
- * $Id: sock.c,v 1.4 2002/08/04 21:23:58 maxk Exp $
- */
-
 #include <linux/module.h>
 
 #include <linux/types.h>
diff --git a/net/bluetooth/rfcomm/core.c b/net/bluetooth/rfcomm/core.c
index 0c2c93735e93..b4fb84e398e5 100644
--- a/net/bluetooth/rfcomm/core.c
+++ b/net/bluetooth/rfcomm/core.c
@@ -23,8 +23,6 @@
 
 /*
  * Bluetooth RFCOMM core.
- *
- * $Id: core.c,v 1.42 2002/10/01 23:26:25 maxk Exp $
  */
 
 #include <linux/module.h>
diff --git a/net/bluetooth/rfcomm/sock.c b/net/bluetooth/rfcomm/sock.c
index 5083adcbfae5..c9054487670a 100644
--- a/net/bluetooth/rfcomm/sock.c
+++ b/net/bluetooth/rfcomm/sock.c
@@ -23,8 +23,6 @@
 
 /*
  * RFCOMM sockets.
- *
- * $Id: sock.c,v 1.24 2002/10/03 01:00:34 maxk Exp $
  */
 
 #include <linux/module.h>
diff --git a/net/bluetooth/rfcomm/tty.c b/net/bluetooth/rfcomm/tty.c
index c9191871c1e0..be84f4fc1477 100644
--- a/net/bluetooth/rfcomm/tty.c
+++ b/net/bluetooth/rfcomm/tty.c
@@ -23,8 +23,6 @@
 
 /*
  * RFCOMM TTY.
- *
- * $Id: tty.c,v 1.24 2002/10/03 01:54:38 holtmann Exp $
  */
 
 #include <linux/module.h>
diff --git a/net/bridge/br.c b/net/bridge/br.c
index 8f3c58e5f7a5..cede010f4ddd 100644
--- a/net/bridge/br.c
+++ b/net/bridge/br.c
@@ -5,8 +5,6 @@
  *	Authors:
  *	Lennert Buytenhek		<buytenh@gnu.org>
  *
- *	$Id: br.c,v 1.47 2001/12/24 00:56:41 davem Exp $
- *
  *	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
diff --git a/net/bridge/br_device.c b/net/bridge/br_device.c
index 626c7795ae30..a6ffc6c2a69f 100644
--- a/net/bridge/br_device.c
+++ b/net/bridge/br_device.c
@@ -5,8 +5,6 @@
  *	Authors:
  *	Lennert Buytenhek		<buytenh@gnu.org>
  *
- *	$Id: br_device.c,v 1.6 2001/12/24 00:59:55 davem Exp $
- *
  *	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
diff --git a/net/bridge/br_fdb.c b/net/bridge/br_fdb.c
index 72c5976a5ce3..4de74cdd091d 100644
--- a/net/bridge/br_fdb.c
+++ b/net/bridge/br_fdb.c
@@ -5,8 +5,6 @@
  *	Authors:
  *	Lennert Buytenhek		<buytenh@gnu.org>
  *
- *	$Id: br_fdb.c,v 1.6 2002/01/17 00:57:07 davem Exp $
- *
  *	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
diff --git a/net/bridge/br_forward.c b/net/bridge/br_forward.c
index a4711674b3df..512645727f51 100644
--- a/net/bridge/br_forward.c
+++ b/net/bridge/br_forward.c
@@ -5,8 +5,6 @@
  *	Authors:
  *	Lennert Buytenhek		<buytenh@gnu.org>
  *
- *	$Id: br_forward.c,v 1.4 2001/08/14 22:05:57 davem Exp $
- *
  *	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
diff --git a/net/bridge/br_if.c b/net/bridge/br_if.c
index c2397f503b0f..143c954681b8 100644
--- a/net/bridge/br_if.c
+++ b/net/bridge/br_if.c
@@ -5,8 +5,6 @@
  *	Authors:
  *	Lennert Buytenhek		<buytenh@gnu.org>
  *
- *	$Id: br_if.c,v 1.7 2001/12/24 00:59:55 davem Exp $
- *
  *	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
diff --git a/net/bridge/br_input.c b/net/bridge/br_input.c
index fa0f5711a996..0145e9416714 100644
--- a/net/bridge/br_input.c
+++ b/net/bridge/br_input.c
@@ -5,8 +5,6 @@
  *	Authors:
  *	Lennert Buytenhek		<buytenh@gnu.org>
  *
- *	$Id: br_input.c,v 1.10 2001/12/24 04:50:20 davem Exp $
- *
  *	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
diff --git a/net/bridge/br_ioctl.c b/net/bridge/br_ioctl.c
index 0655a5f07f58..eeee218eed80 100644
--- a/net/bridge/br_ioctl.c
+++ b/net/bridge/br_ioctl.c
@@ -5,8 +5,6 @@
  *	Authors:
  *	Lennert Buytenhek		<buytenh@gnu.org>
  *
- *	$Id: br_ioctl.c,v 1.4 2000/11/08 05:16:40 davem Exp $
- *
  *	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
diff --git a/net/bridge/br_notify.c b/net/bridge/br_notify.c
index 00644a544e3c..88d8ec7b3142 100644
--- a/net/bridge/br_notify.c
+++ b/net/bridge/br_notify.c
@@ -5,8 +5,6 @@
  *	Authors:
  *	Lennert Buytenhek		<buytenh@gnu.org>
  *
- *	$Id: br_notify.c,v 1.2 2000/02/21 15:51:34 davem Exp $
- *
  *	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
diff --git a/net/bridge/br_private.h b/net/bridge/br_private.h
index 0243cb489edd..83ff5861c2d2 100644
--- a/net/bridge/br_private.h
+++ b/net/bridge/br_private.h
@@ -4,8 +4,6 @@
  *	Authors:
  *	Lennert Buytenhek		<buytenh@gnu.org>
  *
- *	$Id: br_private.h,v 1.7 2001/12/24 00:59:55 davem Exp $
- *
  *	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
diff --git a/net/bridge/br_private_stp.h b/net/bridge/br_private_stp.h
index e29f01ac1adf..8b650f7fbfa0 100644
--- a/net/bridge/br_private_stp.h
+++ b/net/bridge/br_private_stp.h
@@ -4,8 +4,6 @@
  *	Authors:
  *	Lennert Buytenhek		<buytenh@gnu.org>
  *
- *	$Id: br_private_stp.h,v 1.3 2001/02/05 06:03:47 davem Exp $
- *
  *	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
diff --git a/net/bridge/br_stp.c b/net/bridge/br_stp.c
index e38034aa56f5..284d1b2fa1ff 100644
--- a/net/bridge/br_stp.c
+++ b/net/bridge/br_stp.c
@@ -5,8 +5,6 @@
  *	Authors:
  *	Lennert Buytenhek		<buytenh@gnu.org>
  *
- *	$Id: br_stp.c,v 1.4 2000/06/19 10:13:35 davem Exp $
- *
  *	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
diff --git a/net/bridge/br_stp_bpdu.c b/net/bridge/br_stp_bpdu.c
index ddeb6e5d45d6..9dc2de656965 100644
--- a/net/bridge/br_stp_bpdu.c
+++ b/net/bridge/br_stp_bpdu.c
@@ -5,8 +5,6 @@
  *	Authors:
  *	Lennert Buytenhek		<buytenh@gnu.org>
  *
- *	$Id: br_stp_bpdu.c,v 1.3 2001/11/10 02:35:25 davem Exp $
- *
  *	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
diff --git a/net/bridge/br_stp_if.c b/net/bridge/br_stp_if.c
index 1a430eccec9b..1a4e5c37a0cf 100644
--- a/net/bridge/br_stp_if.c
+++ b/net/bridge/br_stp_if.c
@@ -5,8 +5,6 @@
  *	Authors:
  *	Lennert Buytenhek		<buytenh@gnu.org>
  *
- *	$Id: br_stp_if.c,v 1.4 2001/04/14 21:14:39 davem Exp $
- *
  *	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
diff --git a/net/bridge/br_stp_timer.c b/net/bridge/br_stp_timer.c
index 77f5255e6915..772a140bfdf0 100644
--- a/net/bridge/br_stp_timer.c
+++ b/net/bridge/br_stp_timer.c
@@ -5,8 +5,6 @@
  *	Authors:
  *	Lennert Buytenhek		<buytenh@gnu.org>
  *
- *	$Id: br_stp_timer.c,v 1.3 2000/05/05 02:17:17 davem Exp $
- *
  *	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
diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index 1e556d312117..3e18f8525e82 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -4,8 +4,6 @@
  *	Authors:	Alan Cox <iiitac@pyr.swan.ac.uk>
  *			Florian La Roche <rzsfl@rz.uni-sb.de>
  *
- *	Version:	$Id: skbuff.c,v 1.90 2001/11/07 05:56:19 davem Exp $
- *
  *	Fixes:
  *		Alan Cox	:	Fixed the worst of the load
  *					balancer bugs.
diff --git a/net/core/sock.c b/net/core/sock.c
index 88094cb09c06..3879bf65897e 100644
--- a/net/core/sock.c
+++ b/net/core/sock.c
@@ -7,8 +7,6 @@
  *		handler for protocols to use and generic option handler.
  *
  *
- * Version:	$Id: sock.c,v 1.117 2002/02/01 22:01:03 davem Exp $
- *
  * Authors:	Ross Biro
  *		Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG>
  *		Florian La Roche, <flla@stud.uni-sb.de>
diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c
index 24eca23c2db3..42bd24b64b57 100644
--- a/net/ipv4/af_inet.c
+++ b/net/ipv4/af_inet.c
@@ -5,8 +5,6 @@
  *
  *		PF_INET protocol family socket handler.
  *
- * Version:	$Id: af_inet.c,v 1.137 2002/02/01 22:01:03 davem Exp $
- *
  * Authors:	Ross Biro
  *		Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG>
  *		Florian La Roche, <flla@stud.uni-sb.de>
diff --git a/net/ipv4/arp.c b/net/ipv4/arp.c
index 9b539fa9fe18..20c515a1be28 100644
--- a/net/ipv4/arp.c
+++ b/net/ipv4/arp.c
@@ -1,6 +1,4 @@
 /* linux/net/ipv4/arp.c
- *
- * Version:	$Id: arp.c,v 1.99 2001/08/30 22:55:42 davem Exp $
  *
  * Copyright (C) 1994 by Florian  La Roche
  *
diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c
index 61011e1d580e..f8c0b0aea93a 100644
--- a/net/ipv4/devinet.c
+++ b/net/ipv4/devinet.c
@@ -1,8 +1,6 @@
 /*
  *	NET3	IP device support routines.
  *
- *	Version: $Id: devinet.c,v 1.44 2001/10/31 21:55:54 davem Exp $
- *
  *		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
diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c
index 0b2ac6a3d903..5ad01d63f83b 100644
--- a/net/ipv4/fib_frontend.c
+++ b/net/ipv4/fib_frontend.c
@@ -5,8 +5,6 @@
  *
  *		IPv4 Forwarding Information Base: FIB frontend.
  *
- * Version:	$Id: fib_frontend.c,v 1.26 2001/10/31 21:55:54 davem Exp $
- *
  * Authors:	Alexey Kuznetsov, <kuznet@ms2.inr.ac.ru>
  *
  *		This program is free software; you can redistribute it and/or
diff --git a/net/ipv4/fib_hash.c b/net/ipv4/fib_hash.c
index 2e2fc3376ac9..eeec4bf982b8 100644
--- a/net/ipv4/fib_hash.c
+++ b/net/ipv4/fib_hash.c
@@ -5,8 +5,6 @@
  *
  *		IPv4 FIB: lookup engine and maintenance routines.
  *
- * Version:	$Id: fib_hash.c,v 1.13 2001/10/31 21:55:54 davem Exp $
- *
  * Authors:	Alexey Kuznetsov, <kuznet@ms2.inr.ac.ru>
  *
  *		This program is free software; you can redistribute it and/or
diff --git a/net/ipv4/fib_semantics.c b/net/ipv4/fib_semantics.c
index 3b83c34019fc..9335eba683c3 100644
--- a/net/ipv4/fib_semantics.c
+++ b/net/ipv4/fib_semantics.c
@@ -5,8 +5,6 @@
  *
  *		IPv4 Forwarding Information Base: semantics.
  *
- * Version:	$Id: fib_semantics.c,v 1.19 2002/01/12 07:54:56 davem Exp $
- *
  * Authors:	Alexey Kuznetsov, <kuznet@ms2.inr.ac.ru>
  *
  *		This program is free software; you can redistribute it and/or
diff --git a/net/ipv4/fib_trie.c b/net/ipv4/fib_trie.c
index 4b02d14e7ab9..394db9c941a1 100644
--- a/net/ipv4/fib_trie.c
+++ b/net/ipv4/fib_trie.c
@@ -22,8 +22,6 @@
  * IP-address lookup using LC-tries. Stefan Nilsson and Gunnar Karlsson
  * IEEE Journal on Selected Areas in Communications, 17(6):1083-1092, June 1999
  *
- * Version:	$Id: fib_trie.c,v 1.3 2005/06/08 14:20:01 robert Exp $
- *
  *
  * Code from fib_hash has been reused which includes the following header:
  *
diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c
index 87397351ddac..aa7cf46853b7 100644
--- a/net/ipv4/icmp.c
+++ b/net/ipv4/icmp.c
@@ -3,8 +3,6 @@
  *
  *		Alan Cox, <alan@redhat.com>
  *
- *	Version: $Id: icmp.c,v 1.85 2002/02/01 22:01:03 davem Exp $
- *
  *	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
diff --git a/net/ipv4/igmp.c b/net/ipv4/igmp.c
index 2769dc4a4c84..68e84a933e90 100644
--- a/net/ipv4/igmp.c
+++ b/net/ipv4/igmp.c
@@ -8,8 +8,6 @@
  *	the older version didn't come out right using gcc 2.5.8, the newer one
  *	seems to fall out with gcc 2.6.2.
  *
- *	Version: $Id: igmp.c,v 1.47 2002/02/01 22:01:03 davem Exp $
- *
  *	Authors:
  *		Alan Cox <Alan.Cox@linux.org>
  *
diff --git a/net/ipv4/inet_diag.c b/net/ipv4/inet_diag.c
index da97695e7096..c10036e7a463 100644
--- a/net/ipv4/inet_diag.c
+++ b/net/ipv4/inet_diag.c
@@ -1,8 +1,6 @@
 /*
  * inet_diag.c	Module for monitoring INET transport protocols sockets.
  *
- * Version:	$Id: inet_diag.c,v 1.3 2002/02/01 22:01:04 davem Exp $
- *
  * Authors:	Alexey Kuznetsov, <kuznet@ms2.inr.ac.ru>
  *
  *	This program is free software; you can redistribute it and/or
diff --git a/net/ipv4/inetpeer.c b/net/ipv4/inetpeer.c
index af995198f643..a456ceeac3f2 100644
--- a/net/ipv4/inetpeer.c
+++ b/net/ipv4/inetpeer.c
@@ -3,8 +3,6 @@
  *
  *  This source is covered by the GNU GPL, the same as all kernel sources.
  *
- *  Version:	$Id: inetpeer.c,v 1.7 2001/09/20 21:22:50 davem Exp $
- *
  *  Authors:	Andrey V. Savochkin <saw@msu.ru>
  */
 
diff --git a/net/ipv4/ip_forward.c b/net/ipv4/ip_forward.c
index 4813c39b438b..37d36a3f33cd 100644
--- a/net/ipv4/ip_forward.c
+++ b/net/ipv4/ip_forward.c
@@ -5,8 +5,6 @@
  *
  *		The IP forwarding functionality.
  *
- * Version:	$Id: ip_forward.c,v 1.48 2000/12/13 18:31:48 davem Exp $
- *
  * Authors:	see ip.c
  *
  * Fixes:
diff --git a/net/ipv4/ip_fragment.c b/net/ipv4/ip_fragment.c
index be1cb89a8d5a..91e321407313 100644
--- a/net/ipv4/ip_fragment.c
+++ b/net/ipv4/ip_fragment.c
@@ -5,8 +5,6 @@
  *
  *		The IP fragmentation functionality.
  *
- * Version:	$Id: ip_fragment.c,v 1.59 2002/01/12 07:54:56 davem Exp $
- *
  * Authors:	Fred N. van Kempen <waltje@uWalt.NL.Mugnet.ORG>
  *		Alan Cox <Alan.Cox@linux.org>
  *
diff --git a/net/ipv4/ip_input.c b/net/ipv4/ip_input.c
index ff77a4a7f9ec..7c26428ea67b 100644
--- a/net/ipv4/ip_input.c
+++ b/net/ipv4/ip_input.c
@@ -5,8 +5,6 @@
  *
  *		The Internet Protocol (IP) module.
  *
- * Version:	$Id: ip_input.c,v 1.55 2002/01/12 07:39:45 davem Exp $
- *
  * Authors:	Ross Biro
  *		Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG>
  *		Donald Becker, <becker@super.org>
diff --git a/net/ipv4/ip_options.c b/net/ipv4/ip_options.c
index 33126ad2cfdc..be3f18a7a40e 100644
--- a/net/ipv4/ip_options.c
+++ b/net/ipv4/ip_options.c
@@ -5,8 +5,6 @@
  *
  *		The options processing module for ip.c
  *
- * Version:	$Id: ip_options.c,v 1.21 2001/09/01 00:31:50 davem Exp $
- *
  * Authors:	A.N.Kuznetsov
  *
  */
diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c
index e527628f56cf..f1278eecf56d 100644
--- a/net/ipv4/ip_output.c
+++ b/net/ipv4/ip_output.c
@@ -5,8 +5,6 @@
  *
  *		The Internet Protocol (IP) output module.
  *
- * Version:	$Id: ip_output.c,v 1.100 2002/02/01 22:01:03 davem Exp $
- *
  * Authors:	Ross Biro
  *		Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG>
  *		Donald Becker, <becker@super.org>
diff --git a/net/ipv4/ip_sockglue.c b/net/ipv4/ip_sockglue.c
index e0514e82308e..105d92a039b9 100644
--- a/net/ipv4/ip_sockglue.c
+++ b/net/ipv4/ip_sockglue.c
@@ -5,8 +5,6 @@
  *
  *		The IP to API glue.
  *
- * Version:	$Id: ip_sockglue.c,v 1.62 2002/02/01 22:01:04 davem Exp $
- *
  * Authors:	see ip.c
  *
  * Fixes:
diff --git a/net/ipv4/ipconfig.c b/net/ipv4/ipconfig.c
index ed45037ce9be..b88aa9afa42e 100644
--- a/net/ipv4/ipconfig.c
+++ b/net/ipv4/ipconfig.c
@@ -1,6 +1,4 @@
 /*
- *  $Id: ipconfig.c,v 1.46 2002/02/01 22:01:04 davem Exp $
- *
  *  Automatic Configuration of IP -- use DHCP, BOOTP, RARP, or
  *  user-supplied information to configure own IP address and routes.
  *
diff --git a/net/ipv4/ipip.c b/net/ipv4/ipip.c
index 86d8836551b9..4c6d2caf9203 100644
--- a/net/ipv4/ipip.c
+++ b/net/ipv4/ipip.c
@@ -1,8 +1,6 @@
 /*
  *	Linux NET3:	IP/IP protocol decoder.
  *
- *	Version: $Id: ipip.c,v 1.50 2001/10/02 02:22:36 davem Exp $
- *
  *	Authors:
  *		Sam Lantinga (slouken@cs.ucdavis.edu)  02/01/95
  *
diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c
index a34da4977c73..300ab0c2919e 100644
--- a/net/ipv4/ipmr.c
+++ b/net/ipv4/ipmr.c
@@ -9,8 +9,6 @@
  *	as published by the Free Software Foundation; either version
  *	2 of the License, or (at your option) any later version.
  *
- *	Version: $Id: ipmr.c,v 1.65 2001/10/31 21:55:54 davem Exp $
- *
  *	Fixes:
  *	Michael Chastain	:	Incorrect size of copying.
  *	Alan Cox		:	Added the cache manager code
diff --git a/net/ipv4/ipvs/ip_vs_app.c b/net/ipv4/ipvs/ip_vs_app.c
index 535abe0c45e7..1f1897a1a702 100644
--- a/net/ipv4/ipvs/ip_vs_app.c
+++ b/net/ipv4/ipvs/ip_vs_app.c
@@ -1,8 +1,6 @@
 /*
  * ip_vs_app.c: Application module support for IPVS
  *
- * Version:     $Id: ip_vs_app.c,v 1.17 2003/03/22 06:31:21 wensong Exp $
- *
  * Authors:     Wensong Zhang <wensong@linuxvirtualserver.org>
  *
  *              This program is free software; you can redistribute it and/or
diff --git a/net/ipv4/ipvs/ip_vs_conn.c b/net/ipv4/ipvs/ip_vs_conn.c
index 65f1ba112752..f8bdae47a77f 100644
--- a/net/ipv4/ipvs/ip_vs_conn.c
+++ b/net/ipv4/ipvs/ip_vs_conn.c
@@ -5,8 +5,6 @@
  *              high-performance and highly available server based on a
  *              cluster of servers.
  *
- * Version:     $Id: ip_vs_conn.c,v 1.31 2003/04/18 09:03:16 wensong Exp $
- *
  * Authors:     Wensong Zhang <wensong@linuxvirtualserver.org>
  *              Peter Kese <peter.kese@ijs.si>
  *              Julian Anastasov <ja@ssi.bg>
diff --git a/net/ipv4/ipvs/ip_vs_core.c b/net/ipv4/ipvs/ip_vs_core.c
index 963981a9d501..bcf6276ba4b2 100644
--- a/net/ipv4/ipvs/ip_vs_core.c
+++ b/net/ipv4/ipvs/ip_vs_core.c
@@ -5,8 +5,6 @@
  *              high-performance and highly available server based on a
  *              cluster of servers.
  *
- * Version:     $Id: ip_vs_core.c,v 1.34 2003/05/10 03:05:23 wensong Exp $
- *
  * Authors:     Wensong Zhang <wensong@linuxvirtualserver.org>
  *              Peter Kese <peter.kese@ijs.si>
  *              Julian Anastasov <ja@ssi.bg>
diff --git a/net/ipv4/ipvs/ip_vs_ctl.c b/net/ipv4/ipvs/ip_vs_ctl.c
index 94c5767c8e01..9a5ace0b4dd6 100644
--- a/net/ipv4/ipvs/ip_vs_ctl.c
+++ b/net/ipv4/ipvs/ip_vs_ctl.c
@@ -5,8 +5,6 @@
  *              high-performance and highly available server based on a
  *              cluster of servers.
  *
- * Version:     $Id: ip_vs_ctl.c,v 1.36 2003/06/08 09:31:19 wensong Exp $
- *
  * Authors:     Wensong Zhang <wensong@linuxvirtualserver.org>
  *              Peter Kese <peter.kese@ijs.si>
  *              Julian Anastasov <ja@ssi.bg>
diff --git a/net/ipv4/ipvs/ip_vs_dh.c b/net/ipv4/ipvs/ip_vs_dh.c
index dcf5d46aaa5e..8afc1503ed20 100644
--- a/net/ipv4/ipvs/ip_vs_dh.c
+++ b/net/ipv4/ipvs/ip_vs_dh.c
@@ -1,8 +1,6 @@
 /*
  * IPVS:        Destination Hashing scheduling module
  *
- * Version:     $Id: ip_vs_dh.c,v 1.5 2002/09/15 08:14:08 wensong Exp $
- *
  * Authors:     Wensong Zhang <wensong@gnuchina.org>
  *
  *              Inspired by the consistent hashing scheduler patch from
diff --git a/net/ipv4/ipvs/ip_vs_est.c b/net/ipv4/ipvs/ip_vs_est.c
index dfa0d713c801..bc04eedd6dbb 100644
--- a/net/ipv4/ipvs/ip_vs_est.c
+++ b/net/ipv4/ipvs/ip_vs_est.c
@@ -1,8 +1,6 @@
 /*
  * ip_vs_est.c: simple rate estimator for IPVS
  *
- * Version:     $Id: ip_vs_est.c,v 1.4 2002/11/30 01:50:35 wensong Exp $
- *
  * Authors:     Wensong Zhang <wensong@linuxvirtualserver.org>
  *
  *              This program is free software; you can redistribute it and/or
diff --git a/net/ipv4/ipvs/ip_vs_ftp.c b/net/ipv4/ipvs/ip_vs_ftp.c
index 59aa166b7678..c1c758e4f733 100644
--- a/net/ipv4/ipvs/ip_vs_ftp.c
+++ b/net/ipv4/ipvs/ip_vs_ftp.c
@@ -1,8 +1,6 @@
 /*
  * ip_vs_ftp.c: IPVS ftp application module
  *
- * Version:	$Id: ip_vs_ftp.c,v 1.13 2002/09/15 08:14:08 wensong Exp $
- *
  * Authors:	Wensong Zhang <wensong@linuxvirtualserver.org>
  *
  * Changes:
diff --git a/net/ipv4/ipvs/ip_vs_lblc.c b/net/ipv4/ipvs/ip_vs_lblc.c
index 3888642706ad..0efa3db4b180 100644
--- a/net/ipv4/ipvs/ip_vs_lblc.c
+++ b/net/ipv4/ipvs/ip_vs_lblc.c
@@ -1,8 +1,6 @@
 /*
  * IPVS:        Locality-Based Least-Connection scheduling module
  *
- * Version:     $Id: ip_vs_lblc.c,v 1.10 2002/09/15 08:14:08 wensong Exp $
- *
  * Authors:     Wensong Zhang <wensong@gnuchina.org>
  *
  *              This program is free software; you can redistribute it and/or
diff --git a/net/ipv4/ipvs/ip_vs_lblcr.c b/net/ipv4/ipvs/ip_vs_lblcr.c
index daa260eb21cf..8e3bbeb45138 100644
--- a/net/ipv4/ipvs/ip_vs_lblcr.c
+++ b/net/ipv4/ipvs/ip_vs_lblcr.c
@@ -1,8 +1,6 @@
 /*
  * IPVS:        Locality-Based Least-Connection with Replication scheduler
  *
- * Version:     $Id: ip_vs_lblcr.c,v 1.11 2002/09/15 08:14:08 wensong Exp $
- *
  * Authors:     Wensong Zhang <wensong@gnuchina.org>
  *
  *              This program is free software; you can redistribute it and/or
diff --git a/net/ipv4/ipvs/ip_vs_lc.c b/net/ipv4/ipvs/ip_vs_lc.c
index d88fef90a641..ac9f08e065d5 100644
--- a/net/ipv4/ipvs/ip_vs_lc.c
+++ b/net/ipv4/ipvs/ip_vs_lc.c
@@ -1,8 +1,6 @@
 /*
  * IPVS:        Least-Connection Scheduling module
  *
- * Version:     $Id: ip_vs_lc.c,v 1.10 2003/04/18 09:03:16 wensong Exp $
- *
  * Authors:     Wensong Zhang <wensong@linuxvirtualserver.org>
  *
  *              This program is free software; you can redistribute it and/or
diff --git a/net/ipv4/ipvs/ip_vs_nq.c b/net/ipv4/ipvs/ip_vs_nq.c
index bc2a9e5f2a7b..a46bf258d420 100644
--- a/net/ipv4/ipvs/ip_vs_nq.c
+++ b/net/ipv4/ipvs/ip_vs_nq.c
@@ -1,8 +1,6 @@
 /*
  * IPVS:        Never Queue scheduling module
  *
- * Version:     $Id: ip_vs_nq.c,v 1.2 2003/06/08 09:31:19 wensong Exp $
- *
  * Authors:     Wensong Zhang <wensong@linuxvirtualserver.org>
  *
  *              This program is free software; you can redistribute it and/or
diff --git a/net/ipv4/ipvs/ip_vs_proto.c b/net/ipv4/ipvs/ip_vs_proto.c
index 4b1c16cbb16b..876714f23d65 100644
--- a/net/ipv4/ipvs/ip_vs_proto.c
+++ b/net/ipv4/ipvs/ip_vs_proto.c
@@ -1,8 +1,6 @@
 /*
  * ip_vs_proto.c: transport protocol load balancing support for IPVS
  *
- * Version:     $Id: ip_vs_proto.c,v 1.2 2003/04/18 09:03:16 wensong Exp $
- *
  * Authors:     Wensong Zhang <wensong@linuxvirtualserver.org>
  *              Julian Anastasov <ja@ssi.bg>
  *
diff --git a/net/ipv4/ipvs/ip_vs_proto_ah.c b/net/ipv4/ipvs/ip_vs_proto_ah.c
index 4bf835e1d86d..73e0ea87c1f5 100644
--- a/net/ipv4/ipvs/ip_vs_proto_ah.c
+++ b/net/ipv4/ipvs/ip_vs_proto_ah.c
@@ -1,8 +1,6 @@
 /*
  * ip_vs_proto_ah.c:	AH IPSec load balancing support for IPVS
  *
- * Version:     $Id: ip_vs_proto_ah.c,v 1.1 2003/07/04 15:04:37 wensong Exp $
- *
  * Authors:	Julian Anastasov <ja@ssi.bg>, February 2002
  *		Wensong Zhang <wensong@linuxvirtualserver.org>
  *
diff --git a/net/ipv4/ipvs/ip_vs_proto_esp.c b/net/ipv4/ipvs/ip_vs_proto_esp.c
index db6a6b7b1a0b..21d70c8ffa54 100644
--- a/net/ipv4/ipvs/ip_vs_proto_esp.c
+++ b/net/ipv4/ipvs/ip_vs_proto_esp.c
@@ -1,8 +1,6 @@
 /*
  * ip_vs_proto_esp.c:	ESP IPSec load balancing support for IPVS
  *
- * Version:     $Id: ip_vs_proto_esp.c,v 1.1 2003/07/04 15:04:37 wensong Exp $
- *
  * Authors:	Julian Anastasov <ja@ssi.bg>, February 2002
  *		Wensong Zhang <wensong@linuxvirtualserver.org>
  *
diff --git a/net/ipv4/ipvs/ip_vs_proto_tcp.c b/net/ipv4/ipvs/ip_vs_proto_tcp.c
index b83dc14b0a4d..d0ea467986a0 100644
--- a/net/ipv4/ipvs/ip_vs_proto_tcp.c
+++ b/net/ipv4/ipvs/ip_vs_proto_tcp.c
@@ -1,8 +1,6 @@
 /*
  * ip_vs_proto_tcp.c:	TCP load balancing support for IPVS
  *
- * Version:     $Id: ip_vs_proto_tcp.c,v 1.3 2002/11/30 01:50:35 wensong Exp $
- *
  * Authors:     Wensong Zhang <wensong@linuxvirtualserver.org>
  *              Julian Anastasov <ja@ssi.bg>
  *
diff --git a/net/ipv4/ipvs/ip_vs_proto_udp.c b/net/ipv4/ipvs/ip_vs_proto_udp.c
index 75771cb3cd6f..c6be5d56823f 100644
--- a/net/ipv4/ipvs/ip_vs_proto_udp.c
+++ b/net/ipv4/ipvs/ip_vs_proto_udp.c
@@ -1,8 +1,6 @@
 /*
  * ip_vs_proto_udp.c:	UDP load balancing support for IPVS
  *
- * Version:     $Id: ip_vs_proto_udp.c,v 1.3 2002/11/30 01:50:35 wensong Exp $
- *
  * Authors:     Wensong Zhang <wensong@linuxvirtualserver.org>
  *              Julian Anastasov <ja@ssi.bg>
  *
diff --git a/net/ipv4/ipvs/ip_vs_rr.c b/net/ipv4/ipvs/ip_vs_rr.c
index 433f8a947924..c8db12d39e61 100644
--- a/net/ipv4/ipvs/ip_vs_rr.c
+++ b/net/ipv4/ipvs/ip_vs_rr.c
@@ -1,8 +1,6 @@
 /*
  * IPVS:        Round-Robin Scheduling module
  *
- * Version:     $Id: ip_vs_rr.c,v 1.9 2002/09/15 08:14:08 wensong Exp $
- *
  * Authors:     Wensong Zhang <wensong@linuxvirtualserver.org>
  *              Peter Kese <peter.kese@ijs.si>
  *
diff --git a/net/ipv4/ipvs/ip_vs_sched.c b/net/ipv4/ipvs/ip_vs_sched.c
index 121a32b1b756..b64767309855 100644
--- a/net/ipv4/ipvs/ip_vs_sched.c
+++ b/net/ipv4/ipvs/ip_vs_sched.c
@@ -5,8 +5,6 @@
  *              high-performance and highly available server based on a
  *              cluster of servers.
  *
- * Version:     $Id: ip_vs_sched.c,v 1.13 2003/05/10 03:05:23 wensong Exp $
- *
  * Authors:     Wensong Zhang <wensong@linuxvirtualserver.org>
  *              Peter Kese <peter.kese@ijs.si>
  *
diff --git a/net/ipv4/ipvs/ip_vs_sed.c b/net/ipv4/ipvs/ip_vs_sed.c
index dd7c128f9db3..2a7d31358181 100644
--- a/net/ipv4/ipvs/ip_vs_sed.c
+++ b/net/ipv4/ipvs/ip_vs_sed.c
@@ -1,8 +1,6 @@
 /*
  * IPVS:        Shortest Expected Delay scheduling module
  *
- * Version:     $Id: ip_vs_sed.c,v 1.1 2003/05/10 03:06:08 wensong Exp $
- *
  * Authors:     Wensong Zhang <wensong@linuxvirtualserver.org>
  *
  *              This program is free software; you can redistribute it and/or
diff --git a/net/ipv4/ipvs/ip_vs_sh.c b/net/ipv4/ipvs/ip_vs_sh.c
index 1b25b00ef1e1..b8fdfac65001 100644
--- a/net/ipv4/ipvs/ip_vs_sh.c
+++ b/net/ipv4/ipvs/ip_vs_sh.c
@@ -1,8 +1,6 @@
 /*
  * IPVS:        Source Hashing scheduling module
  *
- * Version:     $Id: ip_vs_sh.c,v 1.5 2002/09/15 08:14:08 wensong Exp $
- *
  * Authors:     Wensong Zhang <wensong@gnuchina.org>
  *
  *              This program is free software; you can redistribute it and/or
diff --git a/net/ipv4/ipvs/ip_vs_sync.c b/net/ipv4/ipvs/ip_vs_sync.c
index eff54efe0351..2d4a86f73325 100644
--- a/net/ipv4/ipvs/ip_vs_sync.c
+++ b/net/ipv4/ipvs/ip_vs_sync.c
@@ -5,8 +5,6 @@
  *              high-performance and highly available server based on a
  *              cluster of servers.
  *
- * Version:     $Id: ip_vs_sync.c,v 1.13 2003/06/08 09:31:19 wensong Exp $
- *
  * Authors:     Wensong Zhang <wensong@linuxvirtualserver.org>
  *
  * ip_vs_sync:  sync connection info from master load balancer to backups
diff --git a/net/ipv4/ipvs/ip_vs_wlc.c b/net/ipv4/ipvs/ip_vs_wlc.c
index 8a9d913261d8..772c3cb4eca1 100644
--- a/net/ipv4/ipvs/ip_vs_wlc.c
+++ b/net/ipv4/ipvs/ip_vs_wlc.c
@@ -1,8 +1,6 @@
 /*
  * IPVS:        Weighted Least-Connection Scheduling module
  *
- * Version:     $Id: ip_vs_wlc.c,v 1.13 2003/04/18 09:03:16 wensong Exp $
- *
  * Authors:     Wensong Zhang <wensong@linuxvirtualserver.org>
  *              Peter Kese <peter.kese@ijs.si>
  *
diff --git a/net/ipv4/ipvs/ip_vs_wrr.c b/net/ipv4/ipvs/ip_vs_wrr.c
index 85c680add6df..1d6932d7dc97 100644
--- a/net/ipv4/ipvs/ip_vs_wrr.c
+++ b/net/ipv4/ipvs/ip_vs_wrr.c
@@ -1,8 +1,6 @@
 /*
  * IPVS:        Weighted Round-Robin Scheduling module
  *
- * Version:     $Id: ip_vs_wrr.c,v 1.12 2002/09/15 08:14:08 wensong Exp $
- *
  * Authors:     Wensong Zhang <wensong@linuxvirtualserver.org>
  *
  *              This program is free software; you can redistribute it and/or
diff --git a/net/ipv4/ipvs/ip_vs_xmit.c b/net/ipv4/ipvs/ip_vs_xmit.c
index f63006caea03..9892d4aca42e 100644
--- a/net/ipv4/ipvs/ip_vs_xmit.c
+++ b/net/ipv4/ipvs/ip_vs_xmit.c
@@ -1,8 +1,6 @@
 /*
  * ip_vs_xmit.c: various packet transmitters for IPVS
  *
- * Version:     $Id: ip_vs_xmit.c,v 1.2 2002/11/30 01:50:35 wensong Exp $
- *
  * Authors:     Wensong Zhang <wensong@linuxvirtualserver.org>
  *              Julian Anastasov <ja@ssi.bg>
  *
diff --git a/net/ipv4/proc.c b/net/ipv4/proc.c
index 552169b41b16..eb5cee279c5f 100644
--- a/net/ipv4/proc.c
+++ b/net/ipv4/proc.c
@@ -7,8 +7,6 @@
  *		PROC file system.  It is mainly used for debugging and
  *		statistics.
  *
- * Version:	$Id: proc.c,v 1.45 2001/05/16 16:45:35 davem Exp $
- *
  * Authors:	Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG>
  *		Gerald J. Heim, <heim@peanuts.informatik.uni-tuebingen.de>
  *		Fred Baumgarten, <dc6iq@insu1.etec.uni-karlsruhe.de>
diff --git a/net/ipv4/protocol.c b/net/ipv4/protocol.c
index 971ab9356e51..ea50da0649fd 100644
--- a/net/ipv4/protocol.c
+++ b/net/ipv4/protocol.c
@@ -5,8 +5,6 @@
  *
  *		INET protocol dispatch tables.
  *
- * Version:	$Id: protocol.c,v 1.14 2001/05/18 02:25:49 davem Exp $
- *
  * Authors:	Ross Biro
  *		Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG>
  *
diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c
index e7e091d365ff..1d0c97c8712d 100644
--- a/net/ipv4/raw.c
+++ b/net/ipv4/raw.c
@@ -5,8 +5,6 @@
  *
  *		RAW - implementation of IP "raw" sockets.
  *
- * Version:	$Id: raw.c,v 1.64 2002/02/01 22:01:04 davem Exp $
- *
  * Authors:	Ross Biro
  *		Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG>
  *
diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index 96be336064fb..fe3a02237286 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -5,8 +5,6 @@
  *
  *		ROUTE - implementation of the IP router.
  *
- * Version:	$Id: route.c,v 1.103 2002/01/12 07:44:09 davem Exp $
- *
  * Authors:	Ross Biro
  *		Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG>
  *		Alan Cox, <gw4pts@gw4pts.ampr.org>
diff --git a/net/ipv4/syncookies.c b/net/ipv4/syncookies.c
index 73ba98921d64..6317d3c8dc0d 100644
--- a/net/ipv4/syncookies.c
+++ b/net/ipv4/syncookies.c
@@ -8,8 +8,6 @@
  *      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.
- *
- *  $Id: syncookies.c,v 1.18 2002/02/01 22:01:04 davem Exp $
  */
 
 #include <linux/tcp.h>
diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c
index c437f804ee38..901607003205 100644
--- a/net/ipv4/sysctl_net_ipv4.c
+++ b/net/ipv4/sysctl_net_ipv4.c
@@ -1,8 +1,6 @@
 /*
  * sysctl_net_ipv4.c: sysctl interface to net IPV4 subsystem.
  *
- * $Id: sysctl_net_ipv4.c,v 1.50 2001/10/20 00:00:11 davem Exp $
- *
  * Begun April 1, 1996, Mike Shaver.
  * Added /proc/sys/net/ipv4 directory entry (empty =) ). [MS]
  */
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index ab66683b8043..ad66b09e0bcd 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -5,8 +5,6 @@
  *
  *		Implementation of the Transmission Control Protocol(TCP).
  *
- * Version:	$Id: tcp.c,v 1.216 2002/02/01 22:01:04 davem Exp $
- *
  * Authors:	Ross Biro
  *		Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG>
  *		Mark Evans, <evansmp@uhura.aston.ac.uk>
diff --git a/net/ipv4/tcp_diag.c b/net/ipv4/tcp_diag.c
index 2fbcc7d1b1a0..838d491dfda7 100644
--- a/net/ipv4/tcp_diag.c
+++ b/net/ipv4/tcp_diag.c
@@ -1,8 +1,6 @@
 /*
  * tcp_diag.c	Module for monitoring TCP transport protocols sockets.
  *
- * Version:	$Id: tcp_diag.c,v 1.3 2002/02/01 22:01:04 davem Exp $
- *
  * Authors:	Alexey Kuznetsov, <kuznet@ms2.inr.ac.ru>
  *
  *	This program is free software; you can redistribute it and/or
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index eba873e9b560..b68c3c7d906b 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -5,8 +5,6 @@
  *
  *		Implementation of the Transmission Control Protocol(TCP).
  *
- * Version:	$Id: tcp_input.c,v 1.243 2002/02/01 22:01:04 davem Exp $
- *
  * Authors:	Ross Biro
  *		Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG>
  *		Mark Evans, <evansmp@uhura.aston.ac.uk>
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index cd601a866c2f..f2926ae1de57 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -5,8 +5,6 @@
  *
  *		Implementation of the Transmission Control Protocol(TCP).
  *
- * Version:	$Id: tcp_ipv4.c,v 1.240 2002/02/01 22:01:04 davem Exp $
- *
  *		IPv4 specific functions
  *
  *
diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c
index 019c8c16e5cc..1276cab85e3e 100644
--- a/net/ipv4/tcp_minisocks.c
+++ b/net/ipv4/tcp_minisocks.c
@@ -5,8 +5,6 @@
  *
  *		Implementation of the Transmission Control Protocol(TCP).
  *
- * Version:	$Id: tcp_minisocks.c,v 1.15 2002/02/01 22:01:04 davem Exp $
- *
  * Authors:	Ross Biro
  *		Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG>
  *		Mark Evans, <evansmp@uhura.aston.ac.uk>
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index ad993ecb4810..b171ac65ccab 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -5,8 +5,6 @@
  *
  *		Implementation of the Transmission Control Protocol(TCP).
  *
- * Version:	$Id: tcp_output.c,v 1.146 2002/02/01 22:01:04 davem Exp $
- *
  * Authors:	Ross Biro
  *		Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG>
  *		Mark Evans, <evansmp@uhura.aston.ac.uk>
diff --git a/net/ipv4/tcp_timer.c b/net/ipv4/tcp_timer.c
index 4de68cf5f2aa..e77e7ae0bf2c 100644
--- a/net/ipv4/tcp_timer.c
+++ b/net/ipv4/tcp_timer.c
@@ -5,8 +5,6 @@
  *
  *		Implementation of the Transmission Control Protocol(TCP).
  *
- * Version:	$Id: tcp_timer.c,v 1.88 2002/02/01 22:01:04 davem Exp $
- *
  * Authors:	Ross Biro
  *		Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG>
  *		Mark Evans, <evansmp@uhura.aston.ac.uk>
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index 56fcda3694ba..355e6d62d483 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -5,8 +5,6 @@
  *
  *		The User Datagram Protocol (UDP).
  *
- * Version:	$Id: udp.c,v 1.102 2002/02/01 22:01:04 davem Exp $
- *
  * Authors:	Ross Biro
  *		Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG>
  *		Arnt Gulbrandsen, <agulbra@nvg.unit.no>
diff --git a/net/ipv4/udplite.c b/net/ipv4/udplite.c
index 72ce26b6c4d3..4ad16b6d5138 100644
--- a/net/ipv4/udplite.c
+++ b/net/ipv4/udplite.c
@@ -1,8 +1,6 @@
 /*
  *  UDPLITE     An implementation of the UDP-Lite protocol (RFC 3828).
  *
- *  Version:    $Id: udplite.c,v 1.25 2006/10/19 07:22:36 gerrit Exp $
- *
  *  Authors:    Gerrit Renker       <gerrit@erg.abdn.ac.uk>
  *
  *  Changes:
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index 147588f4c7c0..deb38bf03376 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -6,8 +6,6 @@
  *	Pedro Roque		<roque@di.fc.ul.pt>
  *	Alexey Kuznetsov	<kuznet@ms2.inr.ac.ru>
  *
- *	$Id: addrconf.c,v 1.69 2001/10/31 21:55:54 davem Exp $
- *
  *	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
diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c
index e84b3fd17fb4..350457c761e6 100644
--- a/net/ipv6/af_inet6.c
+++ b/net/ipv6/af_inet6.c
@@ -7,8 +7,6 @@
  *
  *	Adapted from linux/net/ipv4/af_inet.c
  *
- *	$Id: af_inet6.c,v 1.66 2002/02/01 22:01:04 davem Exp $
- *
  * 	Fixes:
  *	piggy, Karl Knutson	:	Socket protocol table
  * 	Hideaki YOSHIFUJI	:	sin6_scope_id support
diff --git a/net/ipv6/datagram.c b/net/ipv6/datagram.c
index b9c2de84a8a2..8cdb6b65ee96 100644
--- a/net/ipv6/datagram.c
+++ b/net/ipv6/datagram.c
@@ -5,8 +5,6 @@
  *	Authors:
  *	Pedro Roque		<roque@di.fc.ul.pt>
  *
- *	$Id: datagram.c,v 1.24 2002/02/01 22:01:04 davem Exp $
- *
  *	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
diff --git a/net/ipv6/exthdrs.c b/net/ipv6/exthdrs.c
index 3cd1c993d52b..602ea826f0a5 100644
--- a/net/ipv6/exthdrs.c
+++ b/net/ipv6/exthdrs.c
@@ -7,8 +7,6 @@
  *	Andi Kleen		<ak@muc.de>
  *	Alexey Kuznetsov	<kuznet@ms2.inr.ac.ru>
  *
- *	$Id: exthdrs.c,v 1.13 2001/06/19 15:58:56 davem Exp $
- *
  *	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
diff --git a/net/ipv6/icmp.c b/net/ipv6/icmp.c
index d42dd16d3487..399d41f65437 100644
--- a/net/ipv6/icmp.c
+++ b/net/ipv6/icmp.c
@@ -5,8 +5,6 @@
  *	Authors:
  *	Pedro Roque		<roque@di.fc.ul.pt>
  *
- *	$Id: icmp.c,v 1.38 2002/02/08 03:57:19 davem Exp $
- *
  *	Based on net/ipv4/icmp.c
  *
  *	RFC 1885
diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c
index 1ee4fa17c129..4de2b9efcacb 100644
--- a/net/ipv6/ip6_fib.c
+++ b/net/ipv6/ip6_fib.c
@@ -5,8 +5,6 @@
  *	Authors:
  *	Pedro Roque		<roque@di.fc.ul.pt>
  *
- *	$Id: ip6_fib.c,v 1.25 2001/10/31 21:55:55 davem Exp $
- *
  *	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
diff --git a/net/ipv6/ip6_input.c b/net/ipv6/ip6_input.c
index 4e5c8615832c..f77a6011c302 100644
--- a/net/ipv6/ip6_input.c
+++ b/net/ipv6/ip6_input.c
@@ -6,8 +6,6 @@
  *	Pedro Roque		<roque@di.fc.ul.pt>
  *	Ian P. Morris		<I.P.Morris@soton.ac.uk>
  *
- *	$Id: ip6_input.c,v 1.19 2000/12/13 18:31:50 davem Exp $
- *
  *	Based in linux/net/ipv4/ip_input.c
  *
  *	This program is free software; you can redistribute it and/or
diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c
index 48cdce9c696c..40a2813a63d1 100644
--- a/net/ipv6/ip6_output.c
+++ b/net/ipv6/ip6_output.c
@@ -5,8 +5,6 @@
  *	Authors:
  *	Pedro Roque		<roque@di.fc.ul.pt>
  *
- *	$Id: ip6_output.c,v 1.34 2002/02/01 22:01:04 davem Exp $
- *
  *	Based on linux/net/ipv4/ip_output.c
  *
  *	This program is free software; you can redistribute it and/or
diff --git a/net/ipv6/ip6_tunnel.c b/net/ipv6/ip6_tunnel.c
index 37814810ac49..17c7b098cdb0 100644
--- a/net/ipv6/ip6_tunnel.c
+++ b/net/ipv6/ip6_tunnel.c
@@ -6,8 +6,6 @@
  *	Ville Nuorvala		<vnuorval@tcs.hut.fi>
  *	Yasuyuki Kozakai	<kozakai@linux-ipv6.org>
  *
- *	$Id$
- *
  *      Based on:
  *      linux/net/ipv6/sit.c and linux/net/ipv4/ipip.c
  *
diff --git a/net/ipv6/ipv6_sockglue.c b/net/ipv6/ipv6_sockglue.c
index 26b83e512a09..237ebbb9383b 100644
--- a/net/ipv6/ipv6_sockglue.c
+++ b/net/ipv6/ipv6_sockglue.c
@@ -7,8 +7,6 @@
  *
  *	Based on linux/net/ipv4/ip_sockglue.c
  *
- *	$Id: ipv6_sockglue.c,v 1.41 2002/02/01 22:01:04 davem Exp $
- *
  *	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
diff --git a/net/ipv6/mcast.c b/net/ipv6/mcast.c
index fd632dd7f98d..fbb2d12c41bc 100644
--- a/net/ipv6/mcast.c
+++ b/net/ipv6/mcast.c
@@ -5,8 +5,6 @@
  *	Authors:
  *	Pedro Roque		<roque@di.fc.ul.pt>
  *
- *	$Id: mcast.c,v 1.40 2002/02/08 03:57:19 davem Exp $
- *
  *	Based on linux/ipv4/igmp.c and linux/ipv4/ip_sockglue.c
  *
  *	This program is free software; you can redistribute it and/or
diff --git a/net/ipv6/proc.c b/net/ipv6/proc.c
index df0736a4cafa..cbc7e514d3ec 100644
--- a/net/ipv6/proc.c
+++ b/net/ipv6/proc.c
@@ -7,8 +7,6 @@
  *		PROC file system.  This is very similar to the IPv4 version,
  *		except it reports the sockets in the INET6 address family.
  *
- * Version:	$Id: proc.c,v 1.17 2002/02/01 22:01:04 davem Exp $
- *
  * Authors:	David S. Miller (davem@caip.rutgers.edu)
  * 		YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
  *
diff --git a/net/ipv6/protocol.c b/net/ipv6/protocol.c
index f929f47b925e..9ab789159913 100644
--- a/net/ipv6/protocol.c
+++ b/net/ipv6/protocol.c
@@ -5,8 +5,6 @@
  *
  *		PF_INET6 protocol dispatch tables.
  *
- * Version:	$Id: protocol.c,v 1.10 2001/05/18 02:25:49 davem Exp $
- *
  * Authors:	Pedro Roque	<roque@di.fc.ul.pt>
  *
  *		This program is free software; you can redistribute it and/or
diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c
index 8fee9a15b2d3..e03c1898ab2e 100644
--- a/net/ipv6/raw.c
+++ b/net/ipv6/raw.c
@@ -7,8 +7,6 @@
  *
  *	Adapted from linux/net/ipv4/raw.c
  *
- *	$Id: raw.c,v 1.51 2002/02/01 22:01:04 davem Exp $
- *
  *	Fixes:
  *	Hideaki YOSHIFUJI	:	sin6_scope_id support
  *	YOSHIFUJI,H.@USAGI	:	raw checksum (RFC2292(bis) compliance)
diff --git a/net/ipv6/reassembly.c b/net/ipv6/reassembly.c
index 9391a6949b96..13509f906d89 100644
--- a/net/ipv6/reassembly.c
+++ b/net/ipv6/reassembly.c
@@ -5,8 +5,6 @@
  *	Authors:
  *	Pedro Roque		<roque@di.fc.ul.pt>
  *
- *	$Id: reassembly.c,v 1.26 2001/03/07 22:00:57 davem Exp $
- *
  *	Based on: net/ipv4/ip_fragment.c
  *
  *	This program is free software; you can redistribute it and/or
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index 220cffe9e63b..edae81319b51 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -5,8 +5,6 @@
  *	Authors:
  *	Pedro Roque		<roque@di.fc.ul.pt>
  *
- *	$Id: route.c,v 1.56 2001/10/31 21:55:55 davem Exp $
- *
  *	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
diff --git a/net/ipv6/sit.c b/net/ipv6/sit.c
index 6b8f0583b637..b0c5080420a8 100644
--- a/net/ipv6/sit.c
+++ b/net/ipv6/sit.c
@@ -6,8 +6,6 @@
  *	Pedro Roque		<roque@di.fc.ul.pt>
  *	Alexey Kuznetsov	<kuznet@ms2.inr.ac.ru>
  *
- *	$Id: sit.c,v 1.53 2001/09/25 05:09:53 davem Exp $
- *
  *	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
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index 715965f0fac0..155499197fc5 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -5,8 +5,6 @@
  *	Authors:
  *	Pedro Roque		<roque@di.fc.ul.pt>
  *
- *	$Id: tcp_ipv6.c,v 1.144 2002/02/01 22:01:04 davem Exp $
- *
  *	Based on:
  *	linux/net/ipv4/tcp.c
  *	linux/net/ipv4/tcp_input.c
diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c
index dd309626ae9a..e0693fffc9bd 100644
--- a/net/ipv6/udp.c
+++ b/net/ipv6/udp.c
@@ -7,8 +7,6 @@
  *
  *	Based on linux/ipv4/udp.c
  *
- *	$Id: udp.c,v 1.65 2002/02/01 22:01:04 davem Exp $
- *
  *	Fixes:
  *	Hideaki YOSHIFUJI	:	sin6_scope_id support
  *	YOSHIFUJI Hideaki @USAGI and:	Support IPV6_V6ONLY socket option, which
diff --git a/net/ipv6/udplite.c b/net/ipv6/udplite.c
index 491efd00a866..f6cdcb348e05 100644
--- a/net/ipv6/udplite.c
+++ b/net/ipv6/udplite.c
@@ -2,8 +2,6 @@
  *  UDPLITEv6   An implementation of the UDP-Lite protocol over IPv6.
  *              See also net/ipv4/udplite.c
  *
- *  Version:    $Id: udplite.c,v 1.9 2006/10/19 08:28:10 gerrit Exp $
- *
  *  Authors:    Gerrit Renker       <gerrit@erg.abdn.ac.uk>
  *
  *  Changes:
diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c
index 2cee87da4441..beca6402f1cf 100644
--- a/net/packet/af_packet.c
+++ b/net/packet/af_packet.c
@@ -5,8 +5,6 @@
  *
  *		PACKET - implements raw packet sockets.
  *
- * Version:	$Id: af_packet.c,v 1.61 2002/02/08 03:57:19 davem Exp $
- *
  * Authors:	Ross Biro
  *		Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG>
  *		Alan Cox, <gw4pts@gw4pts.ampr.org>
diff --git a/net/sched/sch_htb.c b/net/sched/sch_htb.c
index 5bc1ed490180..213071859030 100644
--- a/net/sched/sch_htb.c
+++ b/net/sched/sch_htb.c
@@ -24,8 +24,6 @@
  *		Jiri Fojtasek
  *			fixed requeue routine
  *		and many others. thanks.
- *
- * $Id: sch_htb.c,v 1.25 2003/12/07 11:08:25 devik Exp devik $
  */
 #include <linux/module.h>
 #include <linux/types.h>
diff --git a/net/sunrpc/auth_gss/auth_gss.c b/net/sunrpc/auth_gss/auth_gss.c
index cc12d5f5d5da..019d4b4478c9 100644
--- a/net/sunrpc/auth_gss/auth_gss.c
+++ b/net/sunrpc/auth_gss/auth_gss.c
@@ -33,8 +33,6 @@
  *  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.
- *
- * $Id$
  */
 
 
diff --git a/net/sysctl_net.c b/net/sysctl_net.c
index d8e79162724c..007c1a6708ee 100644
--- a/net/sysctl_net.c
+++ b/net/sysctl_net.c
@@ -4,7 +4,6 @@
  * Begun April 1, 1996, Mike Shaver.
  * Added /proc/sys/net directories for each protocol family. [MS]
  *
- * $Log: sysctl_net.c,v $
  * Revision 1.2  1996/05/08  20:24:40  shaver
  * Added bits for NET_BRIDGE and the NET_IPV4_ARP stuff and
  * NET_IPV4_IP_FORWARD.
diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c
index e18cd3628db4..392e80e3268d 100644
--- a/net/unix/af_unix.c
+++ b/net/unix/af_unix.c
@@ -8,8 +8,6 @@
  *		as published by the Free Software Foundation; either version
  *		2 of the License, or (at your option) any later version.
  *
- * Version:	$Id: af_unix.c,v 1.133 2002/02/08 03:57:19 davem Exp $
- *
  * Fixes:
  *		Linus Torvalds	:	Assorted bug cures.
  *		Niibe Yutaka	:	async I/O support.
-- 
cgit v1.2.3


From fc32b0e28df6655a15b488aaddfc1339f82dc13a Mon Sep 17 00:00:00 2001
From: Lennert Buytenhek <buytenh@wantstofly.org>
Date: Mon, 2 Jun 2008 00:28:40 +0200
Subject: mv643xx_eth: general cleanup

General cleanup of the mv643xx_eth driver.  Mainly fixes coding
style / indentation issues, get rid of some useless 'volatile's,
kill some more superfluous comments, and such.

Signed-off-by: Lennert Buytenhek <buytenh@marvell.com>
Acked-by: Dale Farnsworth <dale@farnsworth.org>
---
 drivers/net/mv643xx_eth.c   | 949 ++++++++++++++++++++------------------------
 include/linux/mv643xx_eth.h |  59 ++-
 2 files changed, 479 insertions(+), 529 deletions(-)

(limited to 'include')

diff --git a/drivers/net/mv643xx_eth.c b/drivers/net/mv643xx_eth.c
index ff6460124307..cf18419f96ab 100644
--- a/drivers/net/mv643xx_eth.c
+++ b/drivers/net/mv643xx_eth.c
@@ -67,8 +67,6 @@ static char mv643xx_eth_driver_version[] = "1.0";
 #define MAX_DESCS_PER_SKB	1
 #endif
 
-#define ETH_HW_IP_ALIGN		2
-
 /*
  * Registers shared between all ports.
  */
@@ -158,12 +156,6 @@ static char mv643xx_eth_driver_version[] = "1.0";
 #define DEFAULT_RX_QUEUE_SIZE		400
 #define DEFAULT_TX_QUEUE_SIZE		800
 
-/* SMI reg */
-#define SMI_BUSY		0x10000000	/* 0 - Write, 1 - Read	*/
-#define SMI_READ_VALID		0x08000000	/* 0 - Write, 1 - Read	*/
-#define SMI_OPCODE_WRITE	0		/* Completion of Read	*/
-#define SMI_OPCODE_READ		0x04000000	/* Operation is in progress */
-
 
 /*
  * RX/TX descriptors.
@@ -231,13 +223,24 @@ struct tx_desc {
 
 /* global *******************************************************************/
 struct mv643xx_eth_shared_private {
+	/*
+	 * Ethernet controller base address.
+	 */
 	void __iomem *base;
 
-	/* used to protect SMI_REG, which is shared across ports */
+	/*
+	 * Protects access to SMI_REG, which is shared between ports.
+	 */
 	spinlock_t phy_lock;
 
+	/*
+	 * Per-port MBUS window access register value.
+	 */
 	u32 win_protect;
 
+	/*
+	 * Hardware-specific parameters.
+	 */
 	unsigned int t_clk;
 };
 
@@ -306,16 +309,17 @@ struct tx_queue {
 
 struct mv643xx_eth_private {
 	struct mv643xx_eth_shared_private *shared;
-	int port_num;			/* User Ethernet port number	*/
+	int port_num;
 
-	struct mv643xx_eth_shared_private *shared_smi;
+	struct net_device *dev;
 
-	struct work_struct tx_timeout_task;
+	struct mv643xx_eth_shared_private *shared_smi;
+	int phy_addr;
 
-	struct net_device *dev;
-	struct mib_counters mib_counters;
 	spinlock_t lock;
 
+	struct mib_counters mib_counters;
+	struct work_struct tx_timeout_task;
 	struct mii_if_info mii;
 
 	/*
@@ -450,7 +454,12 @@ static void rxq_refill(struct rx_queue *rxq)
 						RX_ENABLE_INTERRUPT;
 		wmb();
 
-		skb_reserve(skb, ETH_HW_IP_ALIGN);
+		/*
+		 * The hardware automatically prepends 2 bytes of
+		 * dummy data to each received packet, so that the
+		 * IP header ends up 16-byte aligned.
+		 */
+		skb_reserve(skb, 2);
 	}
 
 	if (rxq->rx_desc_count == 0) {
@@ -474,9 +483,9 @@ static int rxq_process(struct rx_queue *rxq, int budget)
 
 	rx = 0;
 	while (rx < budget) {
-		struct sk_buff *skb;
-		volatile struct rx_desc *rx_desc;
+		struct rx_desc *rx_desc;
 		unsigned int cmd_sts;
+		struct sk_buff *skb;
 		unsigned long flags;
 
 		spin_lock_irqsave(&mp->lock, flags);
@@ -497,34 +506,40 @@ static int rxq_process(struct rx_queue *rxq, int budget)
 
 		spin_unlock_irqrestore(&mp->lock, flags);
 
-		dma_unmap_single(NULL, rx_desc->buf_ptr + ETH_HW_IP_ALIGN,
-					mp->dev->mtu + 24, DMA_FROM_DEVICE);
+		dma_unmap_single(NULL, rx_desc->buf_ptr + 2,
+				 mp->dev->mtu + 24, DMA_FROM_DEVICE);
 		rxq->rx_desc_count--;
 		rx++;
 
 		/*
 		 * Update statistics.
-		 * Note byte count includes 4 byte CRC count
+		 *
+		 * Note that the descriptor byte count includes 2 dummy
+		 * bytes automatically inserted by the hardware at the
+		 * start of the packet (which we don't count), and a 4
+		 * byte CRC at the end of the packet (which we do count).
 		 */
 		stats->rx_packets++;
-		stats->rx_bytes += rx_desc->byte_cnt - ETH_HW_IP_ALIGN;
+		stats->rx_bytes += rx_desc->byte_cnt - 2;
 
 		/*
-		 * In case received a packet without first / last bits on OR
-		 * the error summary bit is on, the packets needs to be dropeed.
+		 * In case we received a packet without first / last bits
+		 * on, or the error summary bit is set, the packet needs
+		 * to be dropped.
 		 */
 		if (((cmd_sts & (RX_FIRST_DESC | RX_LAST_DESC)) !=
 					(RX_FIRST_DESC | RX_LAST_DESC))
 				|| (cmd_sts & ERROR_SUMMARY)) {
 			stats->rx_dropped++;
+
 			if ((cmd_sts & (RX_FIRST_DESC | RX_LAST_DESC)) !=
 				(RX_FIRST_DESC | RX_LAST_DESC)) {
 				if (net_ratelimit())
-					printk(KERN_ERR
-						"%s: Received packet spread "
-						"on multiple descriptors\n",
-						mp->dev->name);
+					dev_printk(KERN_ERR, &mp->dev->dev,
+						   "received packet spanning "
+						   "multiple descriptors\n");
 			}
+
 			if (cmd_sts & ERROR_SUMMARY)
 				stats->rx_errors++;
 
@@ -534,7 +549,7 @@ static int rxq_process(struct rx_queue *rxq, int budget)
 			 * The -4 is for the CRC in the trailer of the
 			 * received packet
 			 */
-			skb_put(skb, rx_desc->byte_cnt - ETH_HW_IP_ALIGN - 4);
+			skb_put(skb, rx_desc->byte_cnt - 2 - 4);
 
 			if (cmd_sts & LAYER_4_CHECKSUM_OK) {
 				skb->ip_summed = CHECKSUM_UNNECESSARY;
@@ -548,8 +563,10 @@ static int rxq_process(struct rx_queue *rxq, int budget)
 			netif_rx(skb);
 #endif
 		}
+
 		mp->dev->last_rx = jiffies;
 	}
+
 	rxq_refill(rxq);
 
 	return rx;
@@ -716,7 +733,7 @@ static void txq_submit_skb(struct tx_queue *txq, struct sk_buff *skb)
 	txq->tx_desc_count += nr_frags + 1;
 }
 
-static int mv643xx_eth_start_xmit(struct sk_buff *skb, struct net_device *dev)
+static int mv643xx_eth_xmit(struct sk_buff *skb, struct net_device *dev)
 {
 	struct mv643xx_eth_private *mp = netdev_priv(dev);
 	struct net_device_stats *stats = &dev->stats;
@@ -727,8 +744,9 @@ static int mv643xx_eth_start_xmit(struct sk_buff *skb, struct net_device *dev)
 
 	if (has_tiny_unaligned_frags(skb) && __skb_linearize(skb)) {
 		stats->tx_dropped++;
-		printk(KERN_DEBUG "%s: failed to linearize tiny "
-				"unaligned fragment\n", dev->name);
+		dev_printk(KERN_DEBUG, &dev->dev,
+			   "failed to linearize skb with tiny "
+			   "unaligned fragment\n");
 		return NETDEV_TX_BUSY;
 	}
 
@@ -758,13 +776,15 @@ static int mv643xx_eth_start_xmit(struct sk_buff *skb, struct net_device *dev)
 
 
 /* mii management interface *************************************************/
-static int phy_addr_get(struct mv643xx_eth_private *mp);
+#define SMI_BUSY		0x10000000
+#define SMI_READ_VALID		0x08000000
+#define SMI_OPCODE_READ		0x04000000
+#define SMI_OPCODE_WRITE	0x00000000
 
-static void read_smi_reg(struct mv643xx_eth_private *mp,
-				unsigned int phy_reg, unsigned int *value)
+static void smi_reg_read(struct mv643xx_eth_private *mp, unsigned int addr,
+			 unsigned int reg, unsigned int *value)
 {
 	void __iomem *smi_reg = mp->shared_smi->base + SMI_REG;
-	int phy_addr = phy_addr_get(mp);
 	unsigned long flags;
 	int i;
 
@@ -780,7 +800,7 @@ static void read_smi_reg(struct mv643xx_eth_private *mp,
 		udelay(10);
 	}
 
-	writel((phy_addr << 16) | (phy_reg << 21) | SMI_OPCODE_READ, smi_reg);
+	writel(SMI_OPCODE_READ | (reg << 21) | (addr << 16), smi_reg);
 
 	/* now wait for the data to be valid */
 	for (i = 0; !(readl(smi_reg) & SMI_READ_VALID); i++) {
@@ -796,11 +816,11 @@ out:
 	spin_unlock_irqrestore(&mp->shared_smi->phy_lock, flags);
 }
 
-static void write_smi_reg(struct mv643xx_eth_private *mp,
-				   unsigned int phy_reg, unsigned int value)
+static void smi_reg_write(struct mv643xx_eth_private *mp,
+			  unsigned int addr,
+			  unsigned int reg, unsigned int value)
 {
 	void __iomem *smi_reg = mp->shared_smi->base + SMI_REG;
-	int phy_addr = phy_addr_get(mp);
 	unsigned long flags;
 	int i;
 
@@ -816,65 +836,63 @@ static void write_smi_reg(struct mv643xx_eth_private *mp,
 		udelay(10);
 	}
 
-	writel((phy_addr << 16) | (phy_reg << 21) |
-		SMI_OPCODE_WRITE | (value & 0xffff), smi_reg);
+	writel(SMI_OPCODE_WRITE | (reg << 21) |
+		(addr << 16) | (value & 0xffff), smi_reg);
 out:
 	spin_unlock_irqrestore(&mp->shared_smi->phy_lock, flags);
 }
 
 
 /* mib counters *************************************************************/
-static void clear_mib_counters(struct mv643xx_eth_private *mp)
+static inline u32 mib_read(struct mv643xx_eth_private *mp, int offset)
 {
-	unsigned int port_num = mp->port_num;
-	int i;
-
-	/* Perform dummy reads from MIB counters */
-	for (i = 0; i < 0x80; i += 4)
-		rdl(mp, MIB_COUNTERS(port_num) + i);
+	return rdl(mp, MIB_COUNTERS(mp->port_num) + offset);
 }
 
-static inline u32 read_mib(struct mv643xx_eth_private *mp, int offset)
+static void mib_counters_clear(struct mv643xx_eth_private *mp)
 {
-	return rdl(mp, MIB_COUNTERS(mp->port_num) + offset);
+	int i;
+
+	for (i = 0; i < 0x80; i += 4)
+		mib_read(mp, i);
 }
 
-static void update_mib_counters(struct mv643xx_eth_private *mp)
+static void mib_counters_update(struct mv643xx_eth_private *mp)
 {
 	struct mib_counters *p = &mp->mib_counters;
 
-	p->good_octets_received += read_mib(mp, 0x00);
-	p->good_octets_received += (u64)read_mib(mp, 0x04) << 32;
-	p->bad_octets_received += read_mib(mp, 0x08);
-	p->internal_mac_transmit_err += read_mib(mp, 0x0c);
-	p->good_frames_received += read_mib(mp, 0x10);
-	p->bad_frames_received += read_mib(mp, 0x14);
-	p->broadcast_frames_received += read_mib(mp, 0x18);
-	p->multicast_frames_received += read_mib(mp, 0x1c);
-	p->frames_64_octets += read_mib(mp, 0x20);
-	p->frames_65_to_127_octets += read_mib(mp, 0x24);
-	p->frames_128_to_255_octets += read_mib(mp, 0x28);
-	p->frames_256_to_511_octets += read_mib(mp, 0x2c);
-	p->frames_512_to_1023_octets += read_mib(mp, 0x30);
-	p->frames_1024_to_max_octets += read_mib(mp, 0x34);
-	p->good_octets_sent += read_mib(mp, 0x38);
-	p->good_octets_sent += (u64)read_mib(mp, 0x3c) << 32;
-	p->good_frames_sent += read_mib(mp, 0x40);
-	p->excessive_collision += read_mib(mp, 0x44);
-	p->multicast_frames_sent += read_mib(mp, 0x48);
-	p->broadcast_frames_sent += read_mib(mp, 0x4c);
-	p->unrec_mac_control_received += read_mib(mp, 0x50);
-	p->fc_sent += read_mib(mp, 0x54);
-	p->good_fc_received += read_mib(mp, 0x58);
-	p->bad_fc_received += read_mib(mp, 0x5c);
-	p->undersize_received += read_mib(mp, 0x60);
-	p->fragments_received += read_mib(mp, 0x64);
-	p->oversize_received += read_mib(mp, 0x68);
-	p->jabber_received += read_mib(mp, 0x6c);
-	p->mac_receive_error += read_mib(mp, 0x70);
-	p->bad_crc_event += read_mib(mp, 0x74);
-	p->collision += read_mib(mp, 0x78);
-	p->late_collision += read_mib(mp, 0x7c);
+	p->good_octets_received += mib_read(mp, 0x00);
+	p->good_octets_received += (u64)mib_read(mp, 0x04) << 32;
+	p->bad_octets_received += mib_read(mp, 0x08);
+	p->internal_mac_transmit_err += mib_read(mp, 0x0c);
+	p->good_frames_received += mib_read(mp, 0x10);
+	p->bad_frames_received += mib_read(mp, 0x14);
+	p->broadcast_frames_received += mib_read(mp, 0x18);
+	p->multicast_frames_received += mib_read(mp, 0x1c);
+	p->frames_64_octets += mib_read(mp, 0x20);
+	p->frames_65_to_127_octets += mib_read(mp, 0x24);
+	p->frames_128_to_255_octets += mib_read(mp, 0x28);
+	p->frames_256_to_511_octets += mib_read(mp, 0x2c);
+	p->frames_512_to_1023_octets += mib_read(mp, 0x30);
+	p->frames_1024_to_max_octets += mib_read(mp, 0x34);
+	p->good_octets_sent += mib_read(mp, 0x38);
+	p->good_octets_sent += (u64)mib_read(mp, 0x3c) << 32;
+	p->good_frames_sent += mib_read(mp, 0x40);
+	p->excessive_collision += mib_read(mp, 0x44);
+	p->multicast_frames_sent += mib_read(mp, 0x48);
+	p->broadcast_frames_sent += mib_read(mp, 0x4c);
+	p->unrec_mac_control_received += mib_read(mp, 0x50);
+	p->fc_sent += mib_read(mp, 0x54);
+	p->good_fc_received += mib_read(mp, 0x58);
+	p->bad_fc_received += mib_read(mp, 0x5c);
+	p->undersize_received += mib_read(mp, 0x60);
+	p->fragments_received += mib_read(mp, 0x64);
+	p->oversize_received += mib_read(mp, 0x68);
+	p->jabber_received += mib_read(mp, 0x6c);
+	p->mac_receive_error += mib_read(mp, 0x70);
+	p->bad_crc_event += mib_read(mp, 0x74);
+	p->collision += mib_read(mp, 0x78);
+	p->late_collision += mib_read(mp, 0x7c);
 }
 
 
@@ -944,7 +962,9 @@ static int mv643xx_eth_get_settings(struct net_device *dev, struct ethtool_cmd *
 	err = mii_ethtool_gset(&mp->mii, cmd);
 	spin_unlock_irq(&mp->lock);
 
-	/* The PHY may support 1000baseT_Half, but the mv643xx does not */
+	/*
+	 * The MAC does not support 1000baseT_Half.
+	 */
 	cmd->supported &= ~SUPPORTED_1000baseT_Half;
 	cmd->advertising &= ~ADVERTISED_1000baseT_Half;
 
@@ -956,6 +976,11 @@ static int mv643xx_eth_set_settings(struct net_device *dev, struct ethtool_cmd *
 	struct mv643xx_eth_private *mp = netdev_priv(dev);
 	int err;
 
+	/*
+	 * The MAC does not support 1000baseT_Half.
+	 */
+	cmd->advertising &= ~ADVERTISED_1000baseT_Half;
+
 	spin_lock_irq(&mp->lock);
 	err = mii_ethtool_sset(&mp->mii, cmd);
 	spin_unlock_irq(&mp->lock);
@@ -963,17 +988,17 @@ static int mv643xx_eth_set_settings(struct net_device *dev, struct ethtool_cmd *
 	return err;
 }
 
-static void mv643xx_eth_get_drvinfo(struct net_device *netdev,
-				struct ethtool_drvinfo *drvinfo)
+static void mv643xx_eth_get_drvinfo(struct net_device *dev,
+				    struct ethtool_drvinfo *drvinfo)
 {
 	strncpy(drvinfo->driver,  mv643xx_eth_driver_name, 32);
 	strncpy(drvinfo->version, mv643xx_eth_driver_version, 32);
 	strncpy(drvinfo->fw_version, "N/A", 32);
-	strncpy(drvinfo->bus_info, "mv643xx", 32);
+	strncpy(drvinfo->bus_info, "platform", 32);
 	drvinfo->n_stats = ARRAY_SIZE(mv643xx_eth_stats);
 }
 
-static int mv643xx_eth_nway_restart(struct net_device *dev)
+static int mv643xx_eth_nway_reset(struct net_device *dev)
 {
 	struct mv643xx_eth_private *mp = netdev_priv(dev);
 
@@ -987,29 +1012,28 @@ static u32 mv643xx_eth_get_link(struct net_device *dev)
 	return mii_link_ok(&mp->mii);
 }
 
-static void mv643xx_eth_get_strings(struct net_device *netdev, uint32_t stringset,
-				uint8_t *data)
+static void mv643xx_eth_get_strings(struct net_device *dev,
+				    uint32_t stringset, uint8_t *data)
 {
 	int i;
 
-	switch(stringset) {
-	case ETH_SS_STATS:
-		for (i=0; i < ARRAY_SIZE(mv643xx_eth_stats); i++) {
+	if (stringset == ETH_SS_STATS) {
+		for (i = 0; i < ARRAY_SIZE(mv643xx_eth_stats); i++) {
 			memcpy(data + i * ETH_GSTRING_LEN,
 				mv643xx_eth_stats[i].stat_string,
 				ETH_GSTRING_LEN);
 		}
-		break;
 	}
 }
 
-static void mv643xx_eth_get_ethtool_stats(struct net_device *netdev,
-				struct ethtool_stats *stats, uint64_t *data)
+static void mv643xx_eth_get_ethtool_stats(struct net_device *dev,
+					  struct ethtool_stats *stats,
+					  uint64_t *data)
 {
-	struct mv643xx_eth_private *mp = netdev->priv;
+	struct mv643xx_eth_private *mp = dev->priv;
 	int i;
 
-	update_mib_counters(mp);
+	mib_counters_update(mp);
 
 	for (i = 0; i < ARRAY_SIZE(mv643xx_eth_stats); i++) {
 		const struct mv643xx_eth_stats *stat;
@@ -1027,38 +1051,35 @@ static void mv643xx_eth_get_ethtool_stats(struct net_device *netdev,
 	}
 }
 
-static int mv643xx_eth_get_sset_count(struct net_device *netdev, int sset)
+static int mv643xx_eth_get_sset_count(struct net_device *dev, int sset)
 {
-	switch (sset) {
-	case ETH_SS_STATS:
+	if (sset == ETH_SS_STATS)
 		return ARRAY_SIZE(mv643xx_eth_stats);
-	default:
-		return -EOPNOTSUPP;
-	}
+
+	return -EOPNOTSUPP;
 }
 
 static const struct ethtool_ops mv643xx_eth_ethtool_ops = {
-	.get_settings           = mv643xx_eth_get_settings,
-	.set_settings           = mv643xx_eth_set_settings,
-	.get_drvinfo            = mv643xx_eth_get_drvinfo,
-	.get_link               = mv643xx_eth_get_link,
+	.get_settings		= mv643xx_eth_get_settings,
+	.set_settings		= mv643xx_eth_set_settings,
+	.get_drvinfo		= mv643xx_eth_get_drvinfo,
+	.nway_reset		= mv643xx_eth_nway_reset,
+	.get_link		= mv643xx_eth_get_link,
 	.set_sg			= ethtool_op_set_sg,
+	.get_strings		= mv643xx_eth_get_strings,
+	.get_ethtool_stats	= mv643xx_eth_get_ethtool_stats,
 	.get_sset_count		= mv643xx_eth_get_sset_count,
-	.get_ethtool_stats      = mv643xx_eth_get_ethtool_stats,
-	.get_strings            = mv643xx_eth_get_strings,
-	.nway_reset		= mv643xx_eth_nway_restart,
 };
 
 
 /* address handling *********************************************************/
 static void uc_addr_get(struct mv643xx_eth_private *mp, unsigned char *addr)
 {
-	unsigned int port_num = mp->port_num;
 	unsigned int mac_h;
 	unsigned int mac_l;
 
-	mac_h = rdl(mp, MAC_ADDR_HIGH(port_num));
-	mac_l = rdl(mp, MAC_ADDR_LOW(port_num));
+	mac_h = rdl(mp, MAC_ADDR_HIGH(mp->port_num));
+	mac_l = rdl(mp, MAC_ADDR_LOW(mp->port_num));
 
 	addr[0] = (mac_h >> 24) & 0xff;
 	addr[1] = (mac_h >> 16) & 0xff;
@@ -1070,72 +1091,54 @@ static void uc_addr_get(struct mv643xx_eth_private *mp, unsigned char *addr)
 
 static void init_mac_tables(struct mv643xx_eth_private *mp)
 {
-	unsigned int port_num = mp->port_num;
-	int table_index;
-
-	/* Clear DA filter unicast table (Ex_dFUT) */
-	for (table_index = 0; table_index <= 0xC; table_index += 4)
-		wrl(mp, UNICAST_TABLE(port_num) + table_index, 0);
+	int i;
 
-	for (table_index = 0; table_index <= 0xFC; table_index += 4) {
-		/* Clear DA filter special multicast table (Ex_dFSMT) */
-		wrl(mp, SPECIAL_MCAST_TABLE(port_num) + table_index, 0);
-		/* Clear DA filter other multicast table (Ex_dFOMT) */
-		wrl(mp, OTHER_MCAST_TABLE(port_num) + table_index, 0);
+	for (i = 0; i < 0x100; i += 4) {
+		wrl(mp, SPECIAL_MCAST_TABLE(mp->port_num) + i, 0);
+		wrl(mp, OTHER_MCAST_TABLE(mp->port_num) + i, 0);
 	}
+
+	for (i = 0; i < 0x10; i += 4)
+		wrl(mp, UNICAST_TABLE(mp->port_num) + i, 0);
 }
 
 static void set_filter_table_entry(struct mv643xx_eth_private *mp,
-					    int table, unsigned char entry)
+				   int table, unsigned char entry)
 {
 	unsigned int table_reg;
-	unsigned int tbl_offset;
-	unsigned int reg_offset;
-
-	tbl_offset = (entry / 4) * 4;	/* Register offset of DA table entry */
-	reg_offset = entry % 4;		/* Entry offset within the register */
 
 	/* Set "accepts frame bit" at specified table entry */
-	table_reg = rdl(mp, table + tbl_offset);
-	table_reg |= 0x01 << (8 * reg_offset);
-	wrl(mp, table + tbl_offset, table_reg);
+	table_reg = rdl(mp, table + (entry & 0xfc));
+	table_reg |= 0x01 << (8 * (entry & 3));
+	wrl(mp, table + (entry & 0xfc), table_reg);
 }
 
 static void uc_addr_set(struct mv643xx_eth_private *mp, unsigned char *addr)
 {
-	unsigned int port_num = mp->port_num;
 	unsigned int mac_h;
 	unsigned int mac_l;
 	int table;
 
-	mac_l = (addr[4] << 8) | (addr[5]);
-	mac_h = (addr[0] << 24) | (addr[1] << 16) | (addr[2] << 8) |
-							(addr[3] << 0);
+	mac_l = (addr[4] << 8) | addr[5];
+	mac_h = (addr[0] << 24) | (addr[1] << 16) | (addr[2] << 8) | addr[3];
 
-	wrl(mp, MAC_ADDR_LOW(port_num), mac_l);
-	wrl(mp, MAC_ADDR_HIGH(port_num), mac_h);
+	wrl(mp, MAC_ADDR_LOW(mp->port_num), mac_l);
+	wrl(mp, MAC_ADDR_HIGH(mp->port_num), mac_h);
 
-	/* Accept frames with this address */
-	table = UNICAST_TABLE(port_num);
+	table = UNICAST_TABLE(mp->port_num);
 	set_filter_table_entry(mp, table, addr[5] & 0x0f);
 }
 
-static void mv643xx_eth_update_mac_address(struct net_device *dev)
+static int mv643xx_eth_set_mac_address(struct net_device *dev, void *addr)
 {
 	struct mv643xx_eth_private *mp = netdev_priv(dev);
 
+	/* +2 is for the offset of the HW addr type */
+	memcpy(dev->dev_addr, addr + 2, 6);
+
 	init_mac_tables(mp);
 	uc_addr_set(mp, dev->dev_addr);
-}
-
-static int mv643xx_eth_set_mac_address(struct net_device *dev, void *addr)
-{
-	int i;
 
-	for (i = 0; i < 6; i++)
-		/* +2 is for the offset of the HW addr type */
-		dev->dev_addr[i] = ((unsigned char *)addr)[i + 2];
-	mv643xx_eth_update_mac_address(dev);
 	return 0;
 }
 
@@ -1157,95 +1160,53 @@ static int addr_crc(unsigned char *addr)
 	return crc;
 }
 
-static void mc_addr(struct mv643xx_eth_private *mp, unsigned char *addr)
+static void mv643xx_eth_set_rx_mode(struct net_device *dev)
 {
-	unsigned int port_num = mp->port_num;
-	int table;
-	int crc;
-
-	if ((addr[0] == 0x01) && (addr[1] == 0x00) &&
-	    (addr[2] == 0x5E) && (addr[3] == 0x00) && (addr[4] == 0x00)) {
-		table = SPECIAL_MCAST_TABLE(port_num);
-		set_filter_table_entry(mp, table, addr[5]);
-		return;
-	}
-
-	crc = addr_crc(addr);
-
-	table = OTHER_MCAST_TABLE(port_num);
-	set_filter_table_entry(mp, table, crc);
-}
+	struct mv643xx_eth_private *mp = netdev_priv(dev);
+	u32 port_config;
+	struct dev_addr_list *addr;
+	int i;
 
-static void set_multicast_list(struct net_device *dev)
-{
+	port_config = rdl(mp, PORT_CONFIG(mp->port_num));
+	if (dev->flags & IFF_PROMISC)
+		port_config |= UNICAST_PROMISCUOUS_MODE;
+	else
+		port_config &= ~UNICAST_PROMISCUOUS_MODE;
+	wrl(mp, PORT_CONFIG(mp->port_num), port_config);
 
-	struct dev_mc_list	*mc_list;
-	int			i;
-	int			table_index;
-	struct mv643xx_eth_private	*mp = netdev_priv(dev);
-	unsigned int		port_num = mp->port_num;
+	if (dev->flags & (IFF_PROMISC | IFF_ALLMULTI)) {
+		int port_num = mp->port_num;
+		u32 accept = 0x01010101;
 
-	/* If the device is in promiscuous mode or in all multicast mode,
-	 * we will fully populate both multicast tables with accept.
-	 * This is guaranteed to yield a match on all multicast addresses...
-	 */
-	if ((dev->flags & IFF_PROMISC) || (dev->flags & IFF_ALLMULTI)) {
-		for (table_index = 0; table_index <= 0xFC; table_index += 4) {
-			/* Set all entries in DA filter special multicast
-			 * table (Ex_dFSMT)
-			 * Set for ETH_Q0 for now
-			 * Bits
-			 * 0	  Accept=1, Drop=0
-			 * 3-1  Queue	 ETH_Q0=0
-			 * 7-4  Reserved = 0;
-			 */
-			wrl(mp, SPECIAL_MCAST_TABLE(port_num) + table_index, 0x01010101);
-
-			/* Set all entries in DA filter other multicast
-			 * table (Ex_dFOMT)
-			 * Set for ETH_Q0 for now
-			 * Bits
-			 * 0	  Accept=1, Drop=0
-			 * 3-1  Queue	 ETH_Q0=0
-			 * 7-4  Reserved = 0;
-			 */
-			wrl(mp, OTHER_MCAST_TABLE(port_num) + table_index, 0x01010101);
+		for (i = 0; i < 0x100; i += 4) {
+			wrl(mp, SPECIAL_MCAST_TABLE(port_num) + i, accept);
+			wrl(mp, OTHER_MCAST_TABLE(port_num) + i, accept);
 		}
 		return;
 	}
 
-	/* We will clear out multicast tables every time we get the list.
-	 * Then add the entire new list...
-	 */
-	for (table_index = 0; table_index <= 0xFC; table_index += 4) {
-		/* Clear DA filter special multicast table (Ex_dFSMT) */
-		wrl(mp, SPECIAL_MCAST_TABLE(port_num) + table_index, 0);
-
-		/* Clear DA filter other multicast table (Ex_dFOMT) */
-		wrl(mp, OTHER_MCAST_TABLE(port_num) + table_index, 0);
+	for (i = 0; i < 0x100; i += 4) {
+		wrl(mp, SPECIAL_MCAST_TABLE(mp->port_num) + i, 0);
+		wrl(mp, OTHER_MCAST_TABLE(mp->port_num) + i, 0);
 	}
 
-	/* Get pointer to net_device multicast list and add each one... */
-	for (i = 0, mc_list = dev->mc_list;
-			(i < 256) && (mc_list != NULL) && (i < dev->mc_count);
-			i++, mc_list = mc_list->next)
-		if (mc_list->dmi_addrlen == 6)
-			mc_addr(mp, mc_list->dmi_addr);
-}
+	for (addr = dev->mc_list; addr != NULL; addr = addr->next) {
+		u8 *a = addr->da_addr;
+		int table;
 
-static void mv643xx_eth_set_rx_mode(struct net_device *dev)
-{
-	struct mv643xx_eth_private *mp = netdev_priv(dev);
-	u32 config_reg;
+		if (addr->da_addrlen != 6)
+			continue;
 
-	config_reg = rdl(mp, PORT_CONFIG(mp->port_num));
-	if (dev->flags & IFF_PROMISC)
-		config_reg |= UNICAST_PROMISCUOUS_MODE;
-	else
-		config_reg &= ~UNICAST_PROMISCUOUS_MODE;
-	wrl(mp, PORT_CONFIG(mp->port_num), config_reg);
+		if (memcmp(a, "\x01\x00\x5e\x00\x00", 5) == 0) {
+			table = SPECIAL_MCAST_TABLE(mp->port_num);
+			set_filter_table_entry(mp, table, a[5]);
+		} else {
+			int crc = addr_crc(a);
 
-	set_multicast_list(dev);
+			table = OTHER_MCAST_TABLE(mp->port_num);
+			set_filter_table_entry(mp, table, crc);
+		}
+	}
 }
 
 
@@ -1483,10 +1444,7 @@ static void txq_deinit(struct tx_queue *txq)
 
 
 /* netdev ops and related ***************************************************/
-static void port_reset(struct mv643xx_eth_private *mp);
-
-static void mv643xx_eth_update_pscr(struct mv643xx_eth_private *mp,
-				    struct ethtool_cmd *ecmd)
+static void update_pscr(struct mv643xx_eth_private *mp, int speed, int duplex)
 {
 	u32 pscr_o;
 	u32 pscr_n;
@@ -1499,15 +1457,15 @@ static void mv643xx_eth_update_pscr(struct mv643xx_eth_private *mp,
 			    SET_FULL_DUPLEX_MODE   |
 			    MAX_RX_PACKET_MASK);
 
-	if (ecmd->speed == SPEED_1000) {
+	if (speed == SPEED_1000) {
 		pscr_n |= SET_GMII_SPEED_TO_1000 | MAX_RX_PACKET_9700BYTE;
 	} else {
-		if (ecmd->speed == SPEED_100)
+		if (speed == SPEED_100)
 			pscr_n |= SET_MII_SPEED_TO_100;
 		pscr_n |= MAX_RX_PACKET_1522BYTE;
 	}
 
-	if (ecmd->duplex == DUPLEX_FULL)
+	if (duplex == DUPLEX_FULL)
 		pscr_n |= SET_FULL_DUPLEX_MODE;
 
 	if (pscr_n != pscr_o) {
@@ -1524,27 +1482,30 @@ static void mv643xx_eth_update_pscr(struct mv643xx_eth_private *mp,
 	}
 }
 
-static irqreturn_t mv643xx_eth_int_handler(int irq, void *dev_id)
+static irqreturn_t mv643xx_eth_irq(int irq, void *dev_id)
 {
 	struct net_device *dev = (struct net_device *)dev_id;
 	struct mv643xx_eth_private *mp = netdev_priv(dev);
-	u32 int_cause, int_cause_ext = 0;
+	u32 int_cause;
+	u32 int_cause_ext;
 
-	/* Read interrupt cause registers */
 	int_cause = rdl(mp, INT_CAUSE(mp->port_num)) & (INT_RX | INT_EXT);
+	if (int_cause == 0)
+		return IRQ_NONE;
+
+	int_cause_ext = 0;
 	if (int_cause & INT_EXT) {
 		int_cause_ext = rdl(mp, INT_CAUSE_EXT(mp->port_num))
 				& (INT_EXT_LINK | INT_EXT_PHY | INT_EXT_TX);
 		wrl(mp, INT_CAUSE_EXT(mp->port_num), ~int_cause_ext);
 	}
 
-	/* PHY status changed */
-	if (int_cause_ext & (INT_EXT_LINK | INT_EXT_PHY)) {
+	if (int_cause_ext & (INT_EXT_PHY | INT_EXT_LINK)) {
 		if (mii_link_ok(&mp->mii)) {
 			struct ethtool_cmd cmd;
 
 			mii_ethtool_gset(&mp->mii, &cmd);
-			mv643xx_eth_update_pscr(mp, &cmd);
+			update_pscr(mp, cmd.speed, cmd.duplex);
 			txq_enable(mp->txq);
 			if (!netif_carrier_ok(dev)) {
 				netif_carrier_on(dev);
@@ -1558,10 +1519,7 @@ static irqreturn_t mv643xx_eth_int_handler(int irq, void *dev_id)
 
 #ifdef MV643XX_ETH_NAPI
 	if (int_cause & INT_RX) {
-		/* schedule the NAPI poll routine to maintain port */
 		wrl(mp, INT_MASK(mp->port_num), 0x00000000);
-
-		/* wait for previous write to complete */
 		rdl(mp, INT_MASK(mp->port_num));
 
 		netif_rx_schedule(dev, &mp->napi);
@@ -1570,40 +1528,31 @@ static irqreturn_t mv643xx_eth_int_handler(int irq, void *dev_id)
 	if (int_cause & INT_RX)
 		rxq_process(mp->rxq, INT_MAX);
 #endif
+
 	if (int_cause_ext & INT_EXT_TX) {
 		txq_reclaim(mp->txq, 0);
 		__txq_maybe_wake(mp->txq);
 	}
 
-	/*
-	 * If no real interrupt occured, exit.
-	 * This can happen when using gigE interrupt coalescing mechanism.
-	 */
-	if ((int_cause == 0x0) && (int_cause_ext == 0x0))
-		return IRQ_NONE;
-
 	return IRQ_HANDLED;
 }
 
 static void phy_reset(struct mv643xx_eth_private *mp)
 {
-	unsigned int phy_reg_data;
+	unsigned int data;
 
-	/* Reset the PHY */
-	read_smi_reg(mp, 0, &phy_reg_data);
-	phy_reg_data |= 0x8000;	/* Set bit 15 to reset the PHY */
-	write_smi_reg(mp, 0, phy_reg_data);
+	smi_reg_read(mp, mp->phy_addr, 0, &data);
+	data |= 0x8000;
+	smi_reg_write(mp, mp->phy_addr, 0, data);
 
-	/* wait for PHY to come out of reset */
 	do {
 		udelay(1);
-		read_smi_reg(mp, 0, &phy_reg_data);
-	} while (phy_reg_data & 0x8000);
+		smi_reg_read(mp, mp->phy_addr, 0, &data);
+	} while (data & 0x8000);
 }
 
-static void port_start(struct net_device *dev)
+static void port_start(struct mv643xx_eth_private *mp)
 {
-	struct mv643xx_eth_private *mp = netdev_priv(dev);
 	u32 pscr;
 	struct ethtool_cmd ethtool_cmd;
 	int i;
@@ -1625,9 +1574,9 @@ static void port_start(struct net_device *dev)
 
 	wrl(mp, SDMA_CONFIG(mp->port_num), PORT_SDMA_CONFIG_DEFAULT_VALUE);
 
-	mv643xx_eth_get_settings(dev, &ethtool_cmd);
+	mv643xx_eth_get_settings(mp->dev, &ethtool_cmd);
 	phy_reset(mp);
-	mv643xx_eth_set_settings(dev, &ethtool_cmd);
+	mv643xx_eth_set_settings(mp->dev, &ethtool_cmd);
 
 	/*
 	 * Configure TX path and queues.
@@ -1643,8 +1592,10 @@ static void port_start(struct net_device *dev)
 		wrl(mp, off, addr);
 	}
 
-	/* Add the assigned Ethernet address to the port's address table */
-	uc_addr_set(mp, dev->dev_addr);
+	/*
+	 * Add configured unicast address to address filter table.
+	 */
+	uc_addr_set(mp, mp->dev->dev_addr);
 
 	/*
 	 * Receive all unmatched unicast, TCP, UDP, BPDU and broadcast
@@ -1675,13 +1626,14 @@ static void port_start(struct net_device *dev)
 
 static void set_rx_coal(struct mv643xx_eth_private *mp, unsigned int delay)
 {
-	unsigned int port_num = mp->port_num;
 	unsigned int coal = ((mp->shared->t_clk / 1000000) * delay) / 64;
 
-	/* Set RX Coalescing mechanism */
-	wrl(mp, SDMA_CONFIG(port_num),
+	if (coal > 0x3fff)
+		coal = 0x3fff;
+
+	wrl(mp, SDMA_CONFIG(mp->port_num),
 		((coal & 0x3fff) << 8) |
-		(rdl(mp, SDMA_CONFIG(port_num))
+		(rdl(mp, SDMA_CONFIG(mp->port_num))
 			& 0xffc000ff));
 }
 
@@ -1689,68 +1641,59 @@ static void set_tx_coal(struct mv643xx_eth_private *mp, unsigned int delay)
 {
 	unsigned int coal = ((mp->shared->t_clk / 1000000) * delay) / 64;
 
-	/* Set TX Coalescing mechanism */
-	wrl(mp, TX_FIFO_URGENT_THRESHOLD(mp->port_num), coal << 4);
-}
-
-static void port_init(struct mv643xx_eth_private *mp)
-{
-	port_reset(mp);
-
-	init_mac_tables(mp);
+	if (coal > 0x3fff)
+		coal = 0x3fff;
+	wrl(mp, TX_FIFO_URGENT_THRESHOLD(mp->port_num), (coal & 0x3fff) << 4);
 }
 
 static int mv643xx_eth_open(struct net_device *dev)
 {
 	struct mv643xx_eth_private *mp = netdev_priv(dev);
-	unsigned int port_num = mp->port_num;
 	int err;
 
-	/* Clear any pending ethernet port interrupts */
-	wrl(mp, INT_CAUSE(port_num), 0);
-	wrl(mp, INT_CAUSE_EXT(port_num), 0);
-	/* wait for previous write to complete */
-	rdl(mp, INT_CAUSE_EXT(port_num));
+	wrl(mp, INT_CAUSE(mp->port_num), 0);
+	wrl(mp, INT_CAUSE_EXT(mp->port_num), 0);
+	rdl(mp, INT_CAUSE_EXT(mp->port_num));
 
-	err = request_irq(dev->irq, mv643xx_eth_int_handler,
-			IRQF_SHARED | IRQF_SAMPLE_RANDOM, dev->name, dev);
+	err = request_irq(dev->irq, mv643xx_eth_irq,
+			  IRQF_SHARED | IRQF_SAMPLE_RANDOM,
+			  dev->name, dev);
 	if (err) {
-		printk(KERN_ERR "%s: Can not assign IRQ\n", dev->name);
+		dev_printk(KERN_ERR, &dev->dev, "can't assign irq\n");
 		return -EAGAIN;
 	}
 
-	port_init(mp);
+	init_mac_tables(mp);
 
 	err = rxq_init(mp);
 	if (err)
-		goto out_free_irq;
+		goto out;
 	rxq_refill(mp->rxq);
 
 	err = txq_init(mp);
 	if (err)
-		goto out_free_rx_skb;
+		goto out_free;
 
 #ifdef MV643XX_ETH_NAPI
 	napi_enable(&mp->napi);
 #endif
 
-	port_start(dev);
+	port_start(mp);
 
 	set_rx_coal(mp, 0);
 	set_tx_coal(mp, 0);
 
-	/* Unmask phy and link status changes interrupts */
-	wrl(mp, INT_MASK_EXT(port_num), INT_EXT_LINK | INT_EXT_PHY | INT_EXT_TX);
+	wrl(mp, INT_MASK_EXT(mp->port_num),
+	    INT_EXT_LINK | INT_EXT_PHY | INT_EXT_TX);
 
-	/* Unmask RX buffer and TX end interrupt */
-	wrl(mp, INT_MASK(port_num), INT_RX | INT_EXT);
+	wrl(mp, INT_MASK(mp->port_num), INT_RX | INT_EXT);
 
 	return 0;
 
 
-out_free_rx_skb:
+out_free:
 	rxq_deinit(mp->rxq);
-out_free_irq:
+out:
 	free_irq(dev->irq, dev);
 
 	return err;
@@ -1758,34 +1701,27 @@ out_free_irq:
 
 static void port_reset(struct mv643xx_eth_private *mp)
 {
-	unsigned int port_num = mp->port_num;
-	unsigned int reg_data;
+	unsigned int data;
 
 	txq_disable(mp->txq);
 	rxq_disable(mp->rxq);
 	while (!(rdl(mp, PORT_STATUS(mp->port_num)) & TX_FIFO_EMPTY))
 		udelay(10);
 
-	/* Clear all MIB counters */
-	clear_mib_counters(mp);
-
 	/* Reset the Enable bit in the Configuration Register */
-	reg_data = rdl(mp, PORT_SERIAL_CONTROL(port_num));
-	reg_data &= ~(SERIAL_PORT_ENABLE		|
-			DO_NOT_FORCE_LINK_FAIL	|
-			FORCE_LINK_PASS);
-	wrl(mp, PORT_SERIAL_CONTROL(port_num), reg_data);
+	data = rdl(mp, PORT_SERIAL_CONTROL(mp->port_num));
+	data &= ~(SERIAL_PORT_ENABLE		|
+		  DO_NOT_FORCE_LINK_FAIL	|
+		  FORCE_LINK_PASS);
+	wrl(mp, PORT_SERIAL_CONTROL(mp->port_num), data);
 }
 
 static int mv643xx_eth_stop(struct net_device *dev)
 {
 	struct mv643xx_eth_private *mp = netdev_priv(dev);
-	unsigned int port_num = mp->port_num;
 
-	/* Mask all interrupts on ethernet port */
-	wrl(mp, INT_MASK(port_num), 0x00000000);
-	/* wait for previous write to complete */
-	rdl(mp, INT_MASK(port_num));
+	wrl(mp, INT_MASK(mp->port_num), 0x00000000);
+	rdl(mp, INT_MASK(mp->port_num));
 
 #ifdef MV643XX_ETH_NAPI
 	napi_disable(&mp->napi);
@@ -1793,17 +1729,18 @@ static int mv643xx_eth_stop(struct net_device *dev)
 	netif_carrier_off(dev);
 	netif_stop_queue(dev);
 
+	free_irq(dev->irq, dev);
+
 	port_reset(mp);
+	mib_counters_update(mp);
 
 	txq_deinit(mp->txq);
 	rxq_deinit(mp->rxq);
 
-	free_irq(dev->irq, dev);
-
 	return 0;
 }
 
-static int mv643xx_eth_do_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
+static int mv643xx_eth_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
 {
 	struct mv643xx_eth_private *mp = netdev_priv(dev);
 
@@ -1812,7 +1749,7 @@ static int mv643xx_eth_do_ioctl(struct net_device *dev, struct ifreq *ifr, int c
 
 static int mv643xx_eth_change_mtu(struct net_device *dev, int new_mtu)
 {
-	if ((new_mtu > 9500) || (new_mtu < 64))
+	if (new_mtu < 64 || new_mtu > 9500)
 		return -EINVAL;
 
 	dev->mtu = new_mtu;
@@ -1823,73 +1760,70 @@ static int mv643xx_eth_change_mtu(struct net_device *dev, int new_mtu)
 	 * Stop and then re-open the interface. This will allocate RX
 	 * skbs of the new MTU.
 	 * There is a possible danger that the open will not succeed,
-	 * due to memory being full, which might fail the open function.
+	 * due to memory being full.
 	 */
 	mv643xx_eth_stop(dev);
 	if (mv643xx_eth_open(dev)) {
-		printk(KERN_ERR "%s: Fatal error on opening device\n",
-			dev->name);
+		dev_printk(KERN_ERR, &dev->dev,
+			   "fatal error on re-opening device after "
+			   "MTU change\n");
 	}
 
 	return 0;
 }
 
-static void mv643xx_eth_tx_timeout_task(struct work_struct *ugly)
+static void tx_timeout_task(struct work_struct *ugly)
 {
-	struct mv643xx_eth_private *mp = container_of(ugly, struct mv643xx_eth_private,
-						  tx_timeout_task);
-	struct net_device *dev = mp->dev;
-
-	if (!netif_running(dev))
-		return;
+	struct mv643xx_eth_private *mp;
 
-	netif_stop_queue(dev);
+	mp = container_of(ugly, struct mv643xx_eth_private, tx_timeout_task);
+	if (netif_running(mp->dev)) {
+		netif_stop_queue(mp->dev);
 
-	port_reset(mp);
-	port_start(dev);
+		port_reset(mp);
+		port_start(mp);
 
-	__txq_maybe_wake(mp->txq);
+		__txq_maybe_wake(mp->txq);
+	}
 }
 
 static void mv643xx_eth_tx_timeout(struct net_device *dev)
 {
 	struct mv643xx_eth_private *mp = netdev_priv(dev);
 
-	printk(KERN_INFO "%s: TX timeout  ", dev->name);
+	dev_printk(KERN_INFO, &dev->dev, "tx timeout\n");
 
-	/* Do the reset outside of interrupt context */
 	schedule_work(&mp->tx_timeout_task);
 }
 
 #ifdef CONFIG_NET_POLL_CONTROLLER
-static void mv643xx_eth_netpoll(struct net_device *netdev)
+static void mv643xx_eth_netpoll(struct net_device *dev)
 {
-	struct mv643xx_eth_private *mp = netdev_priv(netdev);
-	int port_num = mp->port_num;
+	struct mv643xx_eth_private *mp = netdev_priv(dev);
 
-	wrl(mp, INT_MASK(port_num), 0x00000000);
-	/* wait for previous write to complete */
-	rdl(mp, INT_MASK(port_num));
+	wrl(mp, INT_MASK(mp->port_num), 0x00000000);
+	rdl(mp, INT_MASK(mp->port_num));
 
-	mv643xx_eth_int_handler(netdev->irq, netdev);
+	mv643xx_eth_irq(dev->irq, dev);
 
-	wrl(mp, INT_MASK(port_num), INT_RX | INT_CAUSE_EXT);
+	wrl(mp, INT_MASK(mp->port_num), INT_RX | INT_CAUSE_EXT);
 }
 #endif
 
-static int mv643xx_eth_mdio_read(struct net_device *dev, int phy_id, int location)
+static int mv643xx_eth_mdio_read(struct net_device *dev, int addr, int reg)
 {
 	struct mv643xx_eth_private *mp = netdev_priv(dev);
 	int val;
 
-	read_smi_reg(mp, location, &val);
+	smi_reg_read(mp, addr, reg, &val);
+
 	return val;
 }
 
-static void mv643xx_eth_mdio_write(struct net_device *dev, int phy_id, int location, int val)
+static void mv643xx_eth_mdio_write(struct net_device *dev, int addr, int reg, int val)
 {
 	struct mv643xx_eth_private *mp = netdev_priv(dev);
-	write_smi_reg(mp, location, val);
+	smi_reg_write(mp, addr, reg, val);
 }
 
 
@@ -1956,9 +1890,6 @@ static int mv643xx_eth_shared_probe(struct platform_device *pdev)
 		goto out_free;
 
 	spin_lock_init(&msp->phy_lock);
-	msp->t_clk = (pd != NULL && pd->t_clk != 0) ? pd->t_clk : 133000000;
-
-	platform_set_drvdata(pdev, msp);
 
 	/*
 	 * (Re-)program MBUS remapping windows if we are asked to.
@@ -1966,6 +1897,13 @@ static int mv643xx_eth_shared_probe(struct platform_device *pdev)
 	if (pd != NULL && pd->dram != NULL)
 		mv643xx_eth_conf_mbus_windows(msp, pd->dram);
 
+	/*
+	 * Detect hardware parameters.
+	 */
+	msp->t_clk = (pd != NULL && pd->t_clk != 0) ? pd->t_clk : 133000000;
+
+	platform_set_drvdata(pdev, msp);
+
 	return 0;
 
 out_free:
@@ -1985,104 +1923,158 @@ static int mv643xx_eth_shared_remove(struct platform_device *pdev)
 }
 
 static struct platform_driver mv643xx_eth_shared_driver = {
-	.probe = mv643xx_eth_shared_probe,
-	.remove = mv643xx_eth_shared_remove,
+	.probe		= mv643xx_eth_shared_probe,
+	.remove		= mv643xx_eth_shared_remove,
 	.driver = {
-		.name = MV643XX_ETH_SHARED_NAME,
+		.name	= MV643XX_ETH_SHARED_NAME,
 		.owner	= THIS_MODULE,
 	},
 };
 
 static void phy_addr_set(struct mv643xx_eth_private *mp, int phy_addr)
 {
-	u32 reg_data;
 	int addr_shift = 5 * mp->port_num;
+	u32 data;
 
-	reg_data = rdl(mp, PHY_ADDR);
-	reg_data &= ~(0x1f << addr_shift);
-	reg_data |= (phy_addr & 0x1f) << addr_shift;
-	wrl(mp, PHY_ADDR, reg_data);
+	data = rdl(mp, PHY_ADDR);
+	data &= ~(0x1f << addr_shift);
+	data |= (phy_addr & 0x1f) << addr_shift;
+	wrl(mp, PHY_ADDR, data);
 }
 
 static int phy_addr_get(struct mv643xx_eth_private *mp)
 {
-	unsigned int reg_data;
+	unsigned int data;
+
+	data = rdl(mp, PHY_ADDR);
+
+	return (data >> (5 * mp->port_num)) & 0x1f;
+}
+
+static void set_params(struct mv643xx_eth_private *mp,
+		       struct mv643xx_eth_platform_data *pd)
+{
+	struct net_device *dev = mp->dev;
+
+	if (is_valid_ether_addr(pd->mac_addr))
+		memcpy(dev->dev_addr, pd->mac_addr, 6);
+	else
+		uc_addr_get(mp, dev->dev_addr);
+
+	if (pd->phy_addr == -1) {
+		mp->shared_smi = NULL;
+		mp->phy_addr = -1;
+	} else {
+		mp->shared_smi = mp->shared;
+		if (pd->shared_smi != NULL)
+			mp->shared_smi = platform_get_drvdata(pd->shared_smi);
+
+		if (pd->force_phy_addr || pd->phy_addr) {
+			mp->phy_addr = pd->phy_addr & 0x3f;
+			phy_addr_set(mp, mp->phy_addr);
+		} else {
+			mp->phy_addr = phy_addr_get(mp);
+		}
+	}
 
-	reg_data = rdl(mp, PHY_ADDR);
+	mp->default_rx_ring_size = DEFAULT_RX_QUEUE_SIZE;
+	if (pd->rx_queue_size)
+		mp->default_rx_ring_size = pd->rx_queue_size;
+	mp->rx_desc_sram_addr = pd->rx_sram_addr;
+	mp->rx_desc_sram_size = pd->rx_sram_size;
 
-	return ((reg_data >> (5 * mp->port_num)) & 0x1f);
+	mp->default_tx_ring_size = DEFAULT_TX_QUEUE_SIZE;
+	if (pd->tx_queue_size)
+		mp->default_tx_ring_size = pd->tx_queue_size;
+	mp->tx_desc_sram_addr = pd->tx_sram_addr;
+	mp->tx_desc_sram_size = pd->tx_sram_size;
 }
 
 static int phy_detect(struct mv643xx_eth_private *mp)
 {
-	unsigned int phy_reg_data0;
-	int auto_neg;
+	unsigned int data;
+	unsigned int data2;
+
+	smi_reg_read(mp, mp->phy_addr, 0, &data);
+	smi_reg_write(mp, mp->phy_addr, 0, data ^ 0x1000);
 
-	read_smi_reg(mp, 0, &phy_reg_data0);
-	auto_neg = phy_reg_data0 & 0x1000;
-	phy_reg_data0 ^= 0x1000;	/* invert auto_neg */
-	write_smi_reg(mp, 0, phy_reg_data0);
+	smi_reg_read(mp, mp->phy_addr, 0, &data2);
+	if (((data ^ data2) & 0x1000) == 0)
+		return -ENODEV;
 
-	read_smi_reg(mp, 0, &phy_reg_data0);
-	if ((phy_reg_data0 & 0x1000) == auto_neg)
-		return -ENODEV;				/* change didn't take */
+	smi_reg_write(mp, mp->phy_addr, 0, data);
 
-	phy_reg_data0 ^= 0x1000;
-	write_smi_reg(mp, 0, phy_reg_data0);
 	return 0;
 }
 
-static void mv643xx_init_ethtool_cmd(struct net_device *dev, int phy_address,
-				     int speed, int duplex,
-				     struct ethtool_cmd *cmd)
+static int phy_init(struct mv643xx_eth_private *mp,
+		    struct mv643xx_eth_platform_data *pd)
 {
-	struct mv643xx_eth_private *mp = netdev_priv(dev);
+	struct ethtool_cmd cmd;
+	int err;
 
-	memset(cmd, 0, sizeof(*cmd));
+	err = phy_detect(mp);
+	if (err) {
+		dev_printk(KERN_INFO, &mp->dev->dev,
+			   "no PHY detected at addr %d\n", mp->phy_addr);
+		return err;
+	}
+	phy_reset(mp);
+
+	mp->mii.phy_id = mp->phy_addr;
+	mp->mii.phy_id_mask = 0x3f;
+	mp->mii.reg_num_mask = 0x1f;
+	mp->mii.dev = mp->dev;
+	mp->mii.mdio_read = mv643xx_eth_mdio_read;
+	mp->mii.mdio_write = mv643xx_eth_mdio_write;
 
-	cmd->port = PORT_MII;
-	cmd->transceiver = XCVR_INTERNAL;
-	cmd->phy_address = phy_address;
+	mp->mii.supports_gmii = mii_check_gmii_support(&mp->mii);
 
-	if (speed == 0) {
-		cmd->autoneg = AUTONEG_ENABLE;
-		/* mii lib checks, but doesn't use speed on AUTONEG_ENABLE */
-		cmd->speed = SPEED_100;
-		cmd->advertising = ADVERTISED_10baseT_Half  |
-				   ADVERTISED_10baseT_Full  |
-				   ADVERTISED_100baseT_Half |
-				   ADVERTISED_100baseT_Full;
+	memset(&cmd, 0, sizeof(cmd));
+
+	cmd.port = PORT_MII;
+	cmd.transceiver = XCVR_INTERNAL;
+	cmd.phy_address = mp->phy_addr;
+	if (pd->speed == 0) {
+		cmd.autoneg = AUTONEG_ENABLE;
+		cmd.speed = SPEED_100;
+		cmd.advertising = ADVERTISED_10baseT_Half  |
+				  ADVERTISED_10baseT_Full  |
+				  ADVERTISED_100baseT_Half |
+				  ADVERTISED_100baseT_Full;
 		if (mp->mii.supports_gmii)
-			cmd->advertising |= ADVERTISED_1000baseT_Full;
+			cmd.advertising |= ADVERTISED_1000baseT_Full;
 	} else {
-		cmd->autoneg = AUTONEG_DISABLE;
-		cmd->speed = speed;
-		cmd->duplex = duplex;
+		cmd.autoneg = AUTONEG_DISABLE;
+		cmd.speed = pd->speed;
+		cmd.duplex = pd->duplex;
 	}
+
+	update_pscr(mp, cmd.speed, cmd.duplex);
+	mv643xx_eth_set_settings(mp->dev, &cmd);
+
+	return 0;
 }
 
 static int mv643xx_eth_probe(struct platform_device *pdev)
 {
 	struct mv643xx_eth_platform_data *pd;
-	int port_num;
 	struct mv643xx_eth_private *mp;
 	struct net_device *dev;
-	u8 *p;
 	struct resource *res;
-	int err;
-	struct ethtool_cmd cmd;
-	int duplex = DUPLEX_HALF;
-	int speed = 0;			/* default to auto-negotiation */
 	DECLARE_MAC_BUF(mac);
+	int err;
 
 	pd = pdev->dev.platform_data;
 	if (pd == NULL) {
-		printk(KERN_ERR "No mv643xx_eth_platform_data\n");
+		dev_printk(KERN_ERR, &pdev->dev,
+			   "no mv643xx_eth_platform_data\n");
 		return -ENODEV;
 	}
 
 	if (pd->shared == NULL) {
-		printk(KERN_ERR "No mv643xx_eth_platform_data->shared\n");
+		dev_printk(KERN_ERR, &pdev->dev,
+			   "no mv643xx_eth_platform_data->shared\n");
 		return -ENODEV;
 	}
 
@@ -2090,145 +2082,80 @@ static int mv643xx_eth_probe(struct platform_device *pdev)
 	if (!dev)
 		return -ENOMEM;
 
-	platform_set_drvdata(pdev, dev);
-
 	mp = netdev_priv(dev);
+	platform_set_drvdata(pdev, mp);
+
+	mp->shared = platform_get_drvdata(pd->shared);
+	mp->port_num = pd->port_number;
+
 	mp->dev = dev;
 #ifdef MV643XX_ETH_NAPI
 	netif_napi_add(dev, &mp->napi, mv643xx_eth_poll, 64);
 #endif
 
+	set_params(mp, pd);
+
+	spin_lock_init(&mp->lock);
+
+	mib_counters_clear(mp);
+	INIT_WORK(&mp->tx_timeout_task, tx_timeout_task);
+
+	err = phy_init(mp, pd);
+	if (err)
+		goto out;
+	SET_ETHTOOL_OPS(dev, &mv643xx_eth_ethtool_ops);
+
+
 	res = platform_get_resource(pdev, IORESOURCE_IRQ, 0);
 	BUG_ON(!res);
 	dev->irq = res->start;
 
+	dev->hard_start_xmit = mv643xx_eth_xmit;
 	dev->open = mv643xx_eth_open;
 	dev->stop = mv643xx_eth_stop;
-	dev->hard_start_xmit = mv643xx_eth_start_xmit;
-	dev->set_mac_address = mv643xx_eth_set_mac_address;
 	dev->set_multicast_list = mv643xx_eth_set_rx_mode;
-
-	/* No need to Tx Timeout */
+	dev->set_mac_address = mv643xx_eth_set_mac_address;
+	dev->do_ioctl = mv643xx_eth_ioctl;
+	dev->change_mtu = mv643xx_eth_change_mtu;
 	dev->tx_timeout = mv643xx_eth_tx_timeout;
-
 #ifdef CONFIG_NET_POLL_CONTROLLER
 	dev->poll_controller = mv643xx_eth_netpoll;
 #endif
-
 	dev->watchdog_timeo = 2 * HZ;
 	dev->base_addr = 0;
-	dev->change_mtu = mv643xx_eth_change_mtu;
-	dev->do_ioctl = mv643xx_eth_do_ioctl;
-	SET_ETHTOOL_OPS(dev, &mv643xx_eth_ethtool_ops);
 
 #ifdef MV643XX_ETH_CHECKSUM_OFFLOAD_TX
-#ifdef MAX_SKB_FRAGS
 	/*
 	 * Zero copy can only work if we use Discovery II memory. Else, we will
 	 * have to map the buffers to ISA memory which is only 16 MB
 	 */
 	dev->features = NETIF_F_SG | NETIF_F_IP_CSUM;
-#endif
 #endif
 
-	/* Configure the timeout task */
-	INIT_WORK(&mp->tx_timeout_task, mv643xx_eth_tx_timeout_task);
-
-	spin_lock_init(&mp->lock);
-
-	mp->shared = platform_get_drvdata(pd->shared);
-	port_num = mp->port_num = pd->port_number;
+	SET_NETDEV_DEV(dev, &pdev->dev);
 
 	if (mp->shared->win_protect)
-		wrl(mp, WINDOW_PROTECT(port_num), mp->shared->win_protect);
-
-	mp->shared_smi = mp->shared;
-	if (pd->shared_smi != NULL)
-		mp->shared_smi = platform_get_drvdata(pd->shared_smi);
-
-	/* set default config values */
-	uc_addr_get(mp, dev->dev_addr);
-
-	if (is_valid_ether_addr(pd->mac_addr))
-		memcpy(dev->dev_addr, pd->mac_addr, 6);
-
-	if (pd->phy_addr || pd->force_phy_addr)
-		phy_addr_set(mp, pd->phy_addr);
-
-	mp->default_rx_ring_size = DEFAULT_RX_QUEUE_SIZE;
-	if (pd->rx_queue_size)
-		mp->default_rx_ring_size = pd->rx_queue_size;
-
-	mp->default_tx_ring_size = DEFAULT_TX_QUEUE_SIZE;
-	if (pd->tx_queue_size)
-		mp->default_tx_ring_size = pd->tx_queue_size;
-
-	if (pd->tx_sram_size) {
-		mp->tx_desc_sram_size = pd->tx_sram_size;
-		mp->tx_desc_sram_addr = pd->tx_sram_addr;
-	}
-
-	if (pd->rx_sram_size) {
-		mp->rx_desc_sram_addr = pd->rx_sram_addr;
-		mp->rx_desc_sram_size = pd->rx_sram_size;
-	}
-
-	duplex = pd->duplex;
-	speed = pd->speed;
+		wrl(mp, WINDOW_PROTECT(mp->port_num), mp->shared->win_protect);
 
-	/* Hook up MII support for ethtool */
-	mp->mii.dev = dev;
-	mp->mii.mdio_read = mv643xx_eth_mdio_read;
-	mp->mii.mdio_write = mv643xx_eth_mdio_write;
-	mp->mii.phy_id = phy_addr_get(mp);
-	mp->mii.phy_id_mask = 0x3f;
-	mp->mii.reg_num_mask = 0x1f;
-
-	err = phy_detect(mp);
-	if (err) {
-		pr_debug("%s: No PHY detected at addr %d\n",
-				dev->name, phy_addr_get(mp));
-		goto out;
-	}
-
-	phy_reset(mp);
-	mp->mii.supports_gmii = mii_check_gmii_support(&mp->mii);
-	mv643xx_init_ethtool_cmd(dev, mp->mii.phy_id, speed, duplex, &cmd);
-	mv643xx_eth_update_pscr(mp, &cmd);
-	mv643xx_eth_set_settings(dev, &cmd);
-
-	SET_NETDEV_DEV(dev, &pdev->dev);
 	err = register_netdev(dev);
 	if (err)
 		goto out;
 
-	p = dev->dev_addr;
-	printk(KERN_NOTICE
-		"%s: port %d with MAC address %s\n",
-		dev->name, port_num, print_mac(mac, p));
+	dev_printk(KERN_NOTICE, &dev->dev, "port %d with MAC address %s\n",
+		   mp->port_num, print_mac(mac, dev->dev_addr));
 
 	if (dev->features & NETIF_F_SG)
-		printk(KERN_NOTICE "%s: Scatter Gather Enabled\n", dev->name);
+		dev_printk(KERN_NOTICE, &dev->dev, "scatter/gather enabled\n");
 
 	if (dev->features & NETIF_F_IP_CSUM)
-		printk(KERN_NOTICE "%s: TX TCP/IP Checksumming Supported\n",
-								dev->name);
-
-#ifdef MV643XX_ETH_CHECKSUM_OFFLOAD_TX
-	printk(KERN_NOTICE "%s: RX TCP/UDP Checksum Offload ON \n", dev->name);
-#endif
-
-#ifdef MV643XX_ETH_COAL
-	printk(KERN_NOTICE "%s: TX and RX Interrupt Coalescing ON \n",
-								dev->name);
-#endif
+		dev_printk(KERN_NOTICE, &dev->dev, "tx checksum offload\n");
 
 #ifdef MV643XX_ETH_NAPI
-	printk(KERN_NOTICE "%s: RX NAPI Enabled \n", dev->name);
+	dev_printk(KERN_NOTICE, &dev->dev, "napi enabled\n");
 #endif
 
 	if (mp->tx_desc_sram_size > 0)
-		printk(KERN_NOTICE "%s: Using SRAM\n", dev->name);
+		dev_printk(KERN_NOTICE, &dev->dev, "configured with sram\n");
 
 	return 0;
 
@@ -2240,35 +2167,35 @@ out:
 
 static int mv643xx_eth_remove(struct platform_device *pdev)
 {
-	struct net_device *dev = platform_get_drvdata(pdev);
+	struct mv643xx_eth_private *mp = platform_get_drvdata(pdev);
 
-	unregister_netdev(dev);
+	unregister_netdev(mp->dev);
 	flush_scheduled_work();
+	free_netdev(mp->dev);
 
-	free_netdev(dev);
 	platform_set_drvdata(pdev, NULL);
+
 	return 0;
 }
 
 static void mv643xx_eth_shutdown(struct platform_device *pdev)
 {
-	struct net_device *dev = platform_get_drvdata(pdev);
-	struct mv643xx_eth_private *mp = netdev_priv(dev);
-	unsigned int port_num = mp->port_num;
+	struct mv643xx_eth_private *mp = platform_get_drvdata(pdev);
 
 	/* Mask all interrupts on ethernet port */
-	wrl(mp, INT_MASK(port_num), 0);
-	rdl(mp, INT_MASK(port_num));
+	wrl(mp, INT_MASK(mp->port_num), 0);
+	rdl(mp, INT_MASK(mp->port_num));
 
-	port_reset(mp);
+	if (netif_running(mp->dev))
+		port_reset(mp);
 }
 
 static struct platform_driver mv643xx_eth_driver = {
-	.probe = mv643xx_eth_probe,
-	.remove = mv643xx_eth_remove,
-	.shutdown = mv643xx_eth_shutdown,
+	.probe		= mv643xx_eth_probe,
+	.remove		= mv643xx_eth_remove,
+	.shutdown	= mv643xx_eth_shutdown,
 	.driver = {
-		.name = MV643XX_ETH_NAME,
+		.name	= MV643XX_ETH_NAME,
 		.owner	= THIS_MODULE,
 	},
 };
@@ -2283,21 +2210,21 @@ static int __init mv643xx_eth_init_module(void)
 		if (rc)
 			platform_driver_unregister(&mv643xx_eth_shared_driver);
 	}
+
 	return rc;
 }
+module_init(mv643xx_eth_init_module);
 
 static void __exit mv643xx_eth_cleanup_module(void)
 {
 	platform_driver_unregister(&mv643xx_eth_driver);
 	platform_driver_unregister(&mv643xx_eth_shared_driver);
 }
-
-module_init(mv643xx_eth_init_module);
 module_exit(mv643xx_eth_cleanup_module);
 
-MODULE_LICENSE("GPL");
-MODULE_AUTHOR(	"Rabeeh Khoury, Assaf Hoffman, Matthew Dharm, Manish Lachwani"
-		" and Dale Farnsworth");
+MODULE_AUTHOR("Rabeeh Khoury, Assaf Hoffman, Matthew Dharm, Manish Lachwani "
+	      "and Dale Farnsworth");
 MODULE_DESCRIPTION("Ethernet driver for Marvell MV643XX");
-MODULE_ALIAS("platform:" MV643XX_ETH_NAME);
+MODULE_LICENSE("GPL");
 MODULE_ALIAS("platform:" MV643XX_ETH_SHARED_NAME);
+MODULE_ALIAS("platform:" MV643XX_ETH_NAME);
diff --git a/include/linux/mv643xx_eth.h b/include/linux/mv643xx_eth.h
index a15cdd4a8e58..646177660495 100644
--- a/include/linux/mv643xx_eth.h
+++ b/include/linux/mv643xx_eth.h
@@ -17,30 +17,53 @@
 
 struct mv643xx_eth_shared_platform_data {
 	struct mbus_dram_target_info	*dram;
-	unsigned int	t_clk;
+	unsigned int		t_clk;
 };
 
 struct mv643xx_eth_platform_data {
+	/*
+	 * Pointer back to our parent instance, and our port number.
+	 */
 	struct platform_device	*shared;
-	int		port_number;
+	int			port_number;
 
+	/*
+	 * Whether a PHY is present, and if yes, at which address.
+	 */
 	struct platform_device	*shared_smi;
+	int			force_phy_addr;
+	int			phy_addr;
 
-	u16		force_phy_addr;	/* force override if phy_addr == 0 */
-	u16		phy_addr;
-
-	/* If speed is 0, then speed and duplex are autonegotiated. */
-	int		speed;		/* 0, SPEED_10, SPEED_100, SPEED_1000 */
-	int		duplex;		/* DUPLEX_HALF or DUPLEX_FULL */
-
-	/* non-zero values of the following fields override defaults */
-	u32		tx_queue_size;
-	u32		rx_queue_size;
-	u32		tx_sram_addr;
-	u32		tx_sram_size;
-	u32		rx_sram_addr;
-	u32		rx_sram_size;
-	u8		mac_addr[6];	/* mac address if non-zero*/
+	/*
+	 * Use this MAC address if it is valid, overriding the
+	 * address that is already in the hardware.
+	 */
+	u8			mac_addr[6];
+
+	/*
+	 * If speed is 0, autonegotiation is enabled.
+	 *   Valid values for speed: 0, SPEED_10, SPEED_100, SPEED_1000.
+	 *   Valid values for duplex: DUPLEX_HALF, DUPLEX_FULL.
+	 */
+	int			speed;
+	int			duplex;
+
+	/*
+	 * Override default RX/TX queue sizes if nonzero.
+	 */
+	int			rx_queue_size;
+	int			tx_queue_size;
+
+	/*
+	 * Use on-chip SRAM for RX/TX descriptors if size is nonzero
+	 * and sufficient to contain all descriptors for the requested
+	 * ring sizes.
+	 */
+	unsigned long		rx_sram_addr;
+	int			rx_sram_size;
+	unsigned long		tx_sram_addr;
+	int			tx_sram_size;
 };
 
-#endif /* __LINUX_MV643XX_ETH_H */
+
+#endif
-- 
cgit v1.2.3


From 64da80a29c7455321a7df7b47e27d639e3944c1a Mon Sep 17 00:00:00 2001
From: Lennert Buytenhek <buytenh@wantstofly.org>
Date: Mon, 2 Jun 2008 01:01:26 +0200
Subject: mv643xx_eth: allow multiple RX queues

Allow the platform code to specify that we are running on hardware
that is capable of supporting multiple RX queues.  If this option
is used, initialise all of the given RX queues instead of just RX
queue zero.

Signed-off-by: Lennert Buytenhek <buytenh@marvell.com>
Acked-by: Dale Farnsworth <dale@farnsworth.org>
---
 drivers/net/mv643xx_eth.c   | 99 +++++++++++++++++++++++++++++++++------------
 include/linux/mv643xx_eth.h |  5 +++
 2 files changed, 79 insertions(+), 25 deletions(-)

(limited to 'include')

diff --git a/drivers/net/mv643xx_eth.c b/drivers/net/mv643xx_eth.c
index 9ce7be09e295..3c8591853999 100644
--- a/drivers/net/mv643xx_eth.c
+++ b/drivers/net/mv643xx_eth.c
@@ -96,7 +96,7 @@ static char mv643xx_eth_driver_version[] = "1.0";
 #define TX_BW_MTU(p)			(0x0458 + ((p) << 10))
 #define TX_BW_BURST(p)			(0x045c + ((p) << 10))
 #define INT_CAUSE(p)			(0x0460 + ((p) << 10))
-#define  INT_RX				0x00000804
+#define  INT_RX				0x0007fbfc
 #define  INT_EXT			0x00000002
 #define INT_CAUSE_EXT(p)		(0x0464 + ((p) << 10))
 #define  INT_EXT_LINK			0x00100000
@@ -107,7 +107,7 @@ static char mv643xx_eth_driver_version[] = "1.0";
 #define INT_MASK(p)			(0x0468 + ((p) << 10))
 #define INT_MASK_EXT(p)			(0x046c + ((p) << 10))
 #define TX_FIFO_URGENT_THRESHOLD(p)	(0x0474 + ((p) << 10))
-#define RXQ_CURRENT_DESC_PTR(p)		(0x060c + ((p) << 10))
+#define RXQ_CURRENT_DESC_PTR(p, q)	(0x060c + ((p) << 10) + ((q) << 4))
 #define RXQ_COMMAND(p)			(0x0680 + ((p) << 10))
 #define TXQ_CURRENT_DESC_PTR(p)		(0x06c0 + ((p) << 10))
 #define TXQ_BW_TOKENS(p)		(0x0700 + ((p) << 10))
@@ -286,6 +286,8 @@ struct mib_counters {
 };
 
 struct rx_queue {
+	int index;
+
 	int rx_ring_size;
 
 	int rx_desc_count;
@@ -334,8 +336,10 @@ struct mv643xx_eth_private {
 	int default_rx_ring_size;
 	unsigned long rx_desc_sram_addr;
 	int rx_desc_sram_size;
+	u8 rxq_mask;
+	int rxq_primary;
 	struct napi_struct napi;
-	struct rx_queue rxq[1];
+	struct rx_queue rxq[8];
 
 	/*
 	 * TX state.
@@ -365,7 +369,7 @@ static inline void wrl(struct mv643xx_eth_private *mp, int offset, u32 data)
 /* rxq/txq helper functions *************************************************/
 static struct mv643xx_eth_private *rxq_to_mp(struct rx_queue *rxq)
 {
-	return container_of(rxq, struct mv643xx_eth_private, rxq[0]);
+	return container_of(rxq, struct mv643xx_eth_private, rxq[rxq->index]);
 }
 
 static struct mv643xx_eth_private *txq_to_mp(struct tx_queue *txq)
@@ -376,13 +380,13 @@ static struct mv643xx_eth_private *txq_to_mp(struct tx_queue *txq)
 static void rxq_enable(struct rx_queue *rxq)
 {
 	struct mv643xx_eth_private *mp = rxq_to_mp(rxq);
-	wrl(mp, RXQ_COMMAND(mp->port_num), 1);
+	wrl(mp, RXQ_COMMAND(mp->port_num), 1 << rxq->index);
 }
 
 static void rxq_disable(struct rx_queue *rxq)
 {
 	struct mv643xx_eth_private *mp = rxq_to_mp(rxq);
-	u8 mask = 1;
+	u8 mask = 1 << rxq->index;
 
 	wrl(mp, RXQ_COMMAND(mp->port_num), mask << 8);
 	while (rdl(mp, RXQ_COMMAND(mp->port_num)) & mask)
@@ -583,6 +587,7 @@ static int mv643xx_eth_poll(struct napi_struct *napi, int budget)
 {
 	struct mv643xx_eth_private *mp;
 	int rx;
+	int i;
 
 	mp = container_of(napi, struct mv643xx_eth_private, napi);
 
@@ -593,7 +598,10 @@ static int mv643xx_eth_poll(struct napi_struct *napi, int budget)
 	}
 #endif
 
-	rx = rxq_process(mp->rxq, budget);
+	rx = 0;
+	for (i = 7; rx < budget && i >= 0; i--)
+		if (mp->rxq_mask & (1 << i))
+			rx += rxq_process(mp->rxq + i, budget - rx);
 
 	if (rx < budget) {
 		netif_rx_complete(mp->dev, napi);
@@ -1306,13 +1314,15 @@ static void mv643xx_eth_set_rx_mode(struct net_device *dev)
 
 
 /* rx/tx queue initialisation ***********************************************/
-static int rxq_init(struct mv643xx_eth_private *mp)
+static int rxq_init(struct mv643xx_eth_private *mp, int index)
 {
-	struct rx_queue *rxq = mp->rxq;
+	struct rx_queue *rxq = mp->rxq + index;
 	struct rx_desc *rx_desc;
 	int size;
 	int i;
 
+	rxq->index = index;
+
 	rxq->rx_ring_size = mp->default_rx_ring_size;
 
 	rxq->rx_desc_count = 0;
@@ -1321,7 +1331,7 @@ static int rxq_init(struct mv643xx_eth_private *mp)
 
 	size = rxq->rx_ring_size * sizeof(struct rx_desc);
 
-	if (size <= mp->rx_desc_sram_size) {
+	if (index == mp->rxq_primary && size <= mp->rx_desc_sram_size) {
 		rxq->rx_desc_area = ioremap(mp->rx_desc_sram_addr,
 						mp->rx_desc_sram_size);
 		rxq->rx_desc_dma = mp->rx_desc_sram_addr;
@@ -1362,7 +1372,7 @@ static int rxq_init(struct mv643xx_eth_private *mp)
 
 
 out_free:
-	if (size <= mp->rx_desc_sram_size)
+	if (index == mp->rxq_primary && size <= mp->rx_desc_sram_size)
 		iounmap(rxq->rx_desc_area);
 	else
 		dma_free_coherent(NULL, size,
@@ -1395,7 +1405,8 @@ static void rxq_deinit(struct rx_queue *rxq)
 			   rxq->rx_desc_count);
 	}
 
-	if (rxq->rx_desc_area_size <= mp->rx_desc_sram_size)
+	if (rxq->index == mp->rxq_primary &&
+	    rxq->rx_desc_area_size <= mp->rx_desc_sram_size)
 		iounmap(rxq->rx_desc_area);
 	else
 		dma_free_coherent(NULL, rxq->rx_desc_area_size,
@@ -1612,6 +1623,9 @@ static irqreturn_t mv643xx_eth_irq(int irq, void *dev_id)
 		}
 	}
 
+	/*
+	 * RxBuffer or RxError set for any of the 8 queues?
+	 */
 #ifdef MV643XX_ETH_NAPI
 	if (int_cause & INT_RX) {
 		wrl(mp, INT_MASK(mp->port_num), 0x00000000);
@@ -1620,8 +1634,13 @@ static irqreturn_t mv643xx_eth_irq(int irq, void *dev_id)
 		netif_rx_schedule(dev, &mp->napi);
 	}
 #else
-	if (int_cause & INT_RX)
-		rxq_process(mp->rxq, INT_MAX);
+	if (int_cause & INT_RX) {
+		int i;
+
+		for (i = 7; i >= 0; i--)
+			if (mp->rxq_mask & (1 << i))
+				rxq_process(mp->rxq + i, INT_MAX);
+	}
 #endif
 
 	if (int_cause_ext & INT_EXT_TX) {
@@ -1707,13 +1726,16 @@ static void port_start(struct mv643xx_eth_private *mp)
 	wrl(mp, PORT_CONFIG_EXT(mp->port_num), 0x00000000);
 
 	/*
-	 * Enable the receive queue.
+	 * Enable the receive queues.
 	 */
-	for (i = 0; i < 1; i++) {
-		struct rx_queue *rxq = mp->rxq;
-		int off = RXQ_CURRENT_DESC_PTR(mp->port_num);
+	for (i = 0; i < 8; i++) {
+		struct rx_queue *rxq = mp->rxq + i;
+		int off = RXQ_CURRENT_DESC_PTR(mp->port_num, i);
 		u32 addr;
 
+		if ((mp->rxq_mask & (1 << i)) == 0)
+			continue;
+
 		addr = (u32)rxq->rx_desc_dma;
 		addr += rxq->rx_curr_desc * sizeof(struct rx_desc);
 		wrl(mp, off, addr);
@@ -1748,6 +1770,7 @@ static int mv643xx_eth_open(struct net_device *dev)
 {
 	struct mv643xx_eth_private *mp = netdev_priv(dev);
 	int err;
+	int i;
 
 	wrl(mp, INT_CAUSE(mp->port_num), 0);
 	wrl(mp, INT_CAUSE_EXT(mp->port_num), 0);
@@ -1763,10 +1786,20 @@ static int mv643xx_eth_open(struct net_device *dev)
 
 	init_mac_tables(mp);
 
-	err = rxq_init(mp);
-	if (err)
-		goto out;
-	rxq_refill(mp->rxq);
+	for (i = 0; i < 8; i++) {
+		if ((mp->rxq_mask & (1 << i)) == 0)
+			continue;
+
+		err = rxq_init(mp, i);
+		if (err) {
+			while (--i >= 0)
+				if (mp->rxq_mask & (1 << i))
+					rxq_deinit(mp->rxq + i);
+			goto out;
+		}
+
+		rxq_refill(mp->rxq + i);
+	}
 
 	err = txq_init(mp);
 	if (err)
@@ -1790,7 +1823,9 @@ static int mv643xx_eth_open(struct net_device *dev)
 
 
 out_free:
-	rxq_deinit(mp->rxq);
+	for (i = 0; i < 8; i++)
+		if (mp->rxq_mask & (1 << i))
+			rxq_deinit(mp->rxq + i);
 out:
 	free_irq(dev->irq, dev);
 
@@ -1800,9 +1835,13 @@ out:
 static void port_reset(struct mv643xx_eth_private *mp)
 {
 	unsigned int data;
+	int i;
 
+	for (i = 0; i < 8; i++) {
+		if (mp->rxq_mask & (1 << i))
+			rxq_disable(mp->rxq + i);
+	}
 	txq_disable(mp->txq);
-	rxq_disable(mp->rxq);
 	while (!(rdl(mp, PORT_STATUS(mp->port_num)) & TX_FIFO_EMPTY))
 		udelay(10);
 
@@ -1817,6 +1856,7 @@ static void port_reset(struct mv643xx_eth_private *mp)
 static int mv643xx_eth_stop(struct net_device *dev)
 {
 	struct mv643xx_eth_private *mp = netdev_priv(dev);
+	int i;
 
 	wrl(mp, INT_MASK(mp->port_num), 0x00000000);
 	rdl(mp, INT_MASK(mp->port_num));
@@ -1832,8 +1872,11 @@ static int mv643xx_eth_stop(struct net_device *dev)
 	port_reset(mp);
 	mib_counters_update(mp);
 
+	for (i = 0; i < 8; i++) {
+		if (mp->rxq_mask & (1 << i))
+			rxq_deinit(mp->rxq + i);
+	}
 	txq_deinit(mp->txq);
-	rxq_deinit(mp->rxq);
 
 	return 0;
 }
@@ -2085,6 +2128,12 @@ static void set_params(struct mv643xx_eth_private *mp,
 	mp->rx_desc_sram_addr = pd->rx_sram_addr;
 	mp->rx_desc_sram_size = pd->rx_sram_size;
 
+	if (pd->rx_queue_mask)
+		mp->rxq_mask = pd->rx_queue_mask;
+	else
+		mp->rxq_mask = 0x01;
+	mp->rxq_primary = fls(mp->rxq_mask) - 1;
+
 	mp->default_tx_ring_size = DEFAULT_TX_QUEUE_SIZE;
 	if (pd->tx_queue_size)
 		mp->default_tx_ring_size = pd->tx_queue_size;
diff --git a/include/linux/mv643xx_eth.h b/include/linux/mv643xx_eth.h
index 646177660495..1afd7ba6d303 100644
--- a/include/linux/mv643xx_eth.h
+++ b/include/linux/mv643xx_eth.h
@@ -48,6 +48,11 @@ struct mv643xx_eth_platform_data {
 	int			speed;
 	int			duplex;
 
+	/*
+	 * Which RX queues to use.
+	 */
+	int			rx_queue_mask;
+
 	/*
 	 * Override default RX/TX queue sizes if nonzero.
 	 */
-- 
cgit v1.2.3


From 3d6b35bc5090cf8d8b7e62eca1f9c21ca56fc6c7 Mon Sep 17 00:00:00 2001
From: Lennert Buytenhek <buytenh@wantstofly.org>
Date: Mon, 2 Jun 2008 01:28:22 +0200
Subject: mv643xx_eth: allow multiple TX queues

As with the multiple RX queue support, allow the platform code to
specify that the hardware we are running on supports multiple TX
queues.  This patch only uses the highest-numbered enabled queue
to send packets to for now, this can be extended later to enable
QoS and such.

Signed-off-by: Lennert Buytenhek <buytenh@marvell.com>
Acked-by: Dale Farnsworth <dale@farnsworth.org>
---
 drivers/net/mv643xx_eth.c   | 146 +++++++++++++++++++++++++++++++-------------
 include/linux/mv643xx_eth.h |   3 +-
 2 files changed, 105 insertions(+), 44 deletions(-)

(limited to 'include')

diff --git a/drivers/net/mv643xx_eth.c b/drivers/net/mv643xx_eth.c
index 3c8591853999..287155ea9ce1 100644
--- a/drivers/net/mv643xx_eth.c
+++ b/drivers/net/mv643xx_eth.c
@@ -103,16 +103,16 @@ static char mv643xx_eth_driver_version[] = "1.0";
 #define  INT_EXT_PHY			0x00010000
 #define  INT_EXT_TX_ERROR_0		0x00000100
 #define  INT_EXT_TX_0			0x00000001
-#define  INT_EXT_TX			0x00000101
+#define  INT_EXT_TX			0x0000ffff
 #define INT_MASK(p)			(0x0468 + ((p) << 10))
 #define INT_MASK_EXT(p)			(0x046c + ((p) << 10))
 #define TX_FIFO_URGENT_THRESHOLD(p)	(0x0474 + ((p) << 10))
 #define RXQ_CURRENT_DESC_PTR(p, q)	(0x060c + ((p) << 10) + ((q) << 4))
 #define RXQ_COMMAND(p)			(0x0680 + ((p) << 10))
-#define TXQ_CURRENT_DESC_PTR(p)		(0x06c0 + ((p) << 10))
-#define TXQ_BW_TOKENS(p)		(0x0700 + ((p) << 10))
-#define TXQ_BW_CONF(p)			(0x0704 + ((p) << 10))
-#define TXQ_BW_WRR_CONF(p)		(0x0708 + ((p) << 10))
+#define TXQ_CURRENT_DESC_PTR(p, q)	(0x06c0 + ((p) << 10) + ((q) << 2))
+#define TXQ_BW_TOKENS(p, q)		(0x0700 + ((p) << 10) + ((q) << 4))
+#define TXQ_BW_CONF(p, q)		(0x0704 + ((p) << 10) + ((q) << 4))
+#define TXQ_BW_WRR_CONF(p, q)		(0x0708 + ((p) << 10) + ((q) << 4))
 #define MIB_COUNTERS(p)			(0x1000 + ((p) << 7))
 #define SPECIAL_MCAST_TABLE(p)		(0x1400 + ((p) << 10))
 #define OTHER_MCAST_TABLE(p)		(0x1500 + ((p) << 10))
@@ -303,6 +303,8 @@ struct rx_queue {
 };
 
 struct tx_queue {
+	int index;
+
 	int tx_ring_size;
 
 	int tx_desc_count;
@@ -347,7 +349,9 @@ struct mv643xx_eth_private {
 	int default_tx_ring_size;
 	unsigned long tx_desc_sram_addr;
 	int tx_desc_sram_size;
-	struct tx_queue txq[1];
+	u8 txq_mask;
+	int txq_primary;
+	struct tx_queue txq[8];
 #ifdef MV643XX_ETH_TX_FAST_REFILL
 	int tx_clean_threshold;
 #endif
@@ -374,7 +378,7 @@ static struct mv643xx_eth_private *rxq_to_mp(struct rx_queue *rxq)
 
 static struct mv643xx_eth_private *txq_to_mp(struct tx_queue *txq)
 {
-	return container_of(txq, struct mv643xx_eth_private, txq[0]);
+	return container_of(txq, struct mv643xx_eth_private, txq[txq->index]);
 }
 
 static void rxq_enable(struct rx_queue *rxq)
@@ -396,13 +400,13 @@ static void rxq_disable(struct rx_queue *rxq)
 static void txq_enable(struct tx_queue *txq)
 {
 	struct mv643xx_eth_private *mp = txq_to_mp(txq);
-	wrl(mp, TXQ_COMMAND(mp->port_num), 1);
+	wrl(mp, TXQ_COMMAND(mp->port_num), 1 << txq->index);
 }
 
 static void txq_disable(struct tx_queue *txq)
 {
 	struct mv643xx_eth_private *mp = txq_to_mp(txq);
-	u8 mask = 1;
+	u8 mask = 1 << txq->index;
 
 	wrl(mp, TXQ_COMMAND(mp->port_num), mask << 8);
 	while (rdl(mp, TXQ_COMMAND(mp->port_num)) & mask)
@@ -413,6 +417,12 @@ static void __txq_maybe_wake(struct tx_queue *txq)
 {
 	struct mv643xx_eth_private *mp = txq_to_mp(txq);
 
+	/*
+	 * netif_{stop,wake}_queue() flow control only applies to
+	 * the primary queue.
+	 */
+	BUG_ON(txq->index != mp->txq_primary);
+
 	if (txq->tx_ring_size - txq->tx_desc_count >= MAX_DESCS_PER_SKB)
 		netif_wake_queue(mp->dev);
 }
@@ -593,8 +603,10 @@ static int mv643xx_eth_poll(struct napi_struct *napi, int budget)
 
 #ifdef MV643XX_ETH_TX_FAST_REFILL
 	if (++mp->tx_clean_threshold > 5) {
-		txq_reclaim(mp->txq, 0);
 		mp->tx_clean_threshold = 0;
+		for (i = 0; i < 8; i++)
+			if (mp->txq_mask & (1 << i))
+				txq_reclaim(mp->txq + i, 0);
 	}
 #endif
 
@@ -754,8 +766,6 @@ static int mv643xx_eth_xmit(struct sk_buff *skb, struct net_device *dev)
 	struct tx_queue *txq;
 	unsigned long flags;
 
-	BUG_ON(netif_queue_stopped(dev));
-
 	if (has_tiny_unaligned_frags(skb) && __skb_linearize(skb)) {
 		stats->tx_dropped++;
 		dev_printk(KERN_DEBUG, &dev->dev,
@@ -766,13 +776,15 @@ static int mv643xx_eth_xmit(struct sk_buff *skb, struct net_device *dev)
 
 	spin_lock_irqsave(&mp->lock, flags);
 
-	txq = mp->txq;
+	txq = mp->txq + mp->txq_primary;
 
 	if (txq->tx_ring_size - txq->tx_desc_count < MAX_DESCS_PER_SKB) {
-		printk(KERN_ERR "%s: transmit with queue full\n", dev->name);
-		netif_stop_queue(dev);
 		spin_unlock_irqrestore(&mp->lock, flags);
-		return NETDEV_TX_BUSY;
+		if (txq->index == mp->txq_primary && net_ratelimit())
+			dev_printk(KERN_ERR, &dev->dev,
+				   "primary tx queue full?!\n");
+		kfree_skb(skb);
+		return NETDEV_TX_OK;
 	}
 
 	txq_submit_skb(txq, skb);
@@ -780,8 +792,13 @@ static int mv643xx_eth_xmit(struct sk_buff *skb, struct net_device *dev)
 	stats->tx_packets++;
 	dev->trans_start = jiffies;
 
-	if (txq->tx_ring_size - txq->tx_desc_count < MAX_DESCS_PER_SKB)
-		netif_stop_queue(dev);
+	if (txq->index == mp->txq_primary) {
+		int entries_left;
+
+		entries_left = txq->tx_ring_size - txq->tx_desc_count;
+		if (entries_left < MAX_DESCS_PER_SKB)
+			netif_stop_queue(dev);
+	}
 
 	spin_unlock_irqrestore(&mp->lock, flags);
 
@@ -831,8 +848,8 @@ static void txq_set_rate(struct tx_queue *txq, int rate, int burst)
 	if (bucket_size > 65535)
 		bucket_size = 65535;
 
-	wrl(mp, TXQ_BW_TOKENS(mp->port_num), token_rate << 14);
-	wrl(mp, TXQ_BW_CONF(mp->port_num),
+	wrl(mp, TXQ_BW_TOKENS(mp->port_num, txq->index), token_rate << 14);
+	wrl(mp, TXQ_BW_CONF(mp->port_num, txq->index),
 			(bucket_size << 10) | token_rate);
 }
 
@@ -848,7 +865,7 @@ static void txq_set_fixed_prio_mode(struct tx_queue *txq)
 	off = TXQ_FIX_PRIO_CONF(mp->port_num);
 
 	val = rdl(mp, off);
-	val |= 1;
+	val |= 1 << txq->index;
 	wrl(mp, off, val);
 }
 
@@ -864,13 +881,13 @@ static void txq_set_wrr(struct tx_queue *txq, int weight)
 	off = TXQ_FIX_PRIO_CONF(mp->port_num);
 
 	val = rdl(mp, off);
-	val &= ~1;
+	val &= ~(1 << txq->index);
 	wrl(mp, off, val);
 
 	/*
 	 * Configure WRR weight for this queue.
 	 */
-	off = TXQ_BW_WRR_CONF(mp->port_num);
+	off = TXQ_BW_WRR_CONF(mp->port_num, txq->index);
 
 	val = rdl(mp, off);
 	val = (val & ~0xff) | (weight & 0xff);
@@ -1415,13 +1432,15 @@ static void rxq_deinit(struct rx_queue *rxq)
 	kfree(rxq->rx_skb);
 }
 
-static int txq_init(struct mv643xx_eth_private *mp)
+static int txq_init(struct mv643xx_eth_private *mp, int index)
 {
-	struct tx_queue *txq = mp->txq;
+	struct tx_queue *txq = mp->txq + index;
 	struct tx_desc *tx_desc;
 	int size;
 	int i;
 
+	txq->index = index;
+
 	txq->tx_ring_size = mp->default_tx_ring_size;
 
 	txq->tx_desc_count = 0;
@@ -1430,7 +1449,7 @@ static int txq_init(struct mv643xx_eth_private *mp)
 
 	size = txq->tx_ring_size * sizeof(struct tx_desc);
 
-	if (size <= mp->tx_desc_sram_size) {
+	if (index == mp->txq_primary && size <= mp->tx_desc_sram_size) {
 		txq->tx_desc_area = ioremap(mp->tx_desc_sram_addr,
 						mp->tx_desc_sram_size);
 		txq->tx_desc_dma = mp->tx_desc_sram_addr;
@@ -1467,7 +1486,7 @@ static int txq_init(struct mv643xx_eth_private *mp)
 
 
 out_free:
-	if (size <= mp->tx_desc_sram_size)
+	if (index == mp->txq_primary && size <= mp->tx_desc_sram_size)
 		iounmap(txq->tx_desc_area);
 	else
 		dma_free_coherent(NULL, size,
@@ -1539,7 +1558,8 @@ static void txq_deinit(struct tx_queue *txq)
 
 	BUG_ON(txq->tx_used_desc != txq->tx_curr_desc);
 
-	if (txq->tx_desc_area_size <= mp->tx_desc_sram_size)
+	if (txq->index == mp->txq_primary &&
+	    txq->tx_desc_area_size <= mp->tx_desc_sram_size)
 		iounmap(txq->tx_desc_area);
 	else
 		dma_free_coherent(NULL, txq->tx_desc_area_size,
@@ -1578,12 +1598,20 @@ static void update_pscr(struct mv643xx_eth_private *mp, int speed, int duplex)
 		if ((pscr_o & SERIAL_PORT_ENABLE) == 0)
 			wrl(mp, PORT_SERIAL_CONTROL(mp->port_num), pscr_n);
 		else {
-			txq_disable(mp->txq);
+			int i;
+
+			for (i = 0; i < 8; i++)
+				if (mp->txq_mask & (1 << i))
+					txq_disable(mp->txq + i);
+
 			pscr_o &= ~SERIAL_PORT_ENABLE;
 			wrl(mp, PORT_SERIAL_CONTROL(mp->port_num), pscr_o);
 			wrl(mp, PORT_SERIAL_CONTROL(mp->port_num), pscr_n);
 			wrl(mp, PORT_SERIAL_CONTROL(mp->port_num), pscr_n);
-			txq_enable(mp->txq);
+
+			for (i = 0; i < 8; i++)
+				if (mp->txq_mask & (1 << i))
+					txq_enable(mp->txq + i);
 		}
 	}
 }
@@ -1609,13 +1637,17 @@ static irqreturn_t mv643xx_eth_irq(int irq, void *dev_id)
 	if (int_cause_ext & (INT_EXT_PHY | INT_EXT_LINK)) {
 		if (mii_link_ok(&mp->mii)) {
 			struct ethtool_cmd cmd;
+			int i;
 
 			mii_ethtool_gset(&mp->mii, &cmd);
 			update_pscr(mp, cmd.speed, cmd.duplex);
-			txq_enable(mp->txq);
+			for (i = 0; i < 8; i++)
+				if (mp->txq_mask & (1 << i))
+					txq_enable(mp->txq + i);
+
 			if (!netif_carrier_ok(dev)) {
 				netif_carrier_on(dev);
-				__txq_maybe_wake(mp->txq);
+				__txq_maybe_wake(mp->txq + mp->txq_primary);
 			}
 		} else if (netif_carrier_ok(dev)) {
 			netif_stop_queue(dev);
@@ -1643,9 +1675,17 @@ static irqreturn_t mv643xx_eth_irq(int irq, void *dev_id)
 	}
 #endif
 
+	/*
+	 * TxBuffer or TxError set for any of the 8 queues?
+	 */
 	if (int_cause_ext & INT_EXT_TX) {
-		txq_reclaim(mp->txq, 0);
-		__txq_maybe_wake(mp->txq);
+		int i;
+
+		for (i = 0; i < 8; i++)
+			if (mp->txq_mask & (1 << i))
+				txq_reclaim(mp->txq + i, 0);
+
+		__txq_maybe_wake(mp->txq + mp->txq_primary);
 	}
 
 	return IRQ_HANDLED;
@@ -1696,11 +1736,14 @@ static void port_start(struct mv643xx_eth_private *mp)
 	 * Configure TX path and queues.
 	 */
 	tx_set_rate(mp, 1000000000, 16777216);
-	for (i = 0; i < 1; i++) {
-		struct tx_queue *txq = mp->txq;
-		int off = TXQ_CURRENT_DESC_PTR(mp->port_num);
+	for (i = 0; i < 8; i++) {
+		struct tx_queue *txq = mp->txq + i;
+		int off = TXQ_CURRENT_DESC_PTR(mp->port_num, i);
 		u32 addr;
 
+		if ((mp->txq_mask & (1 << i)) == 0)
+			continue;
+
 		addr = (u32)txq->tx_desc_dma;
 		addr += txq->tx_curr_desc * sizeof(struct tx_desc);
 		wrl(mp, off, addr);
@@ -1801,9 +1844,18 @@ static int mv643xx_eth_open(struct net_device *dev)
 		rxq_refill(mp->rxq + i);
 	}
 
-	err = txq_init(mp);
-	if (err)
-		goto out_free;
+	for (i = 0; i < 8; i++) {
+		if ((mp->txq_mask & (1 << i)) == 0)
+			continue;
+
+		err = txq_init(mp, i);
+		if (err) {
+			while (--i >= 0)
+				if (mp->txq_mask & (1 << i))
+					txq_deinit(mp->txq + i);
+			goto out_free;
+		}
+	}
 
 #ifdef MV643XX_ETH_NAPI
 	napi_enable(&mp->napi);
@@ -1840,8 +1892,9 @@ static void port_reset(struct mv643xx_eth_private *mp)
 	for (i = 0; i < 8; i++) {
 		if (mp->rxq_mask & (1 << i))
 			rxq_disable(mp->rxq + i);
+		if (mp->txq_mask & (1 << i))
+			txq_disable(mp->txq + i);
 	}
-	txq_disable(mp->txq);
 	while (!(rdl(mp, PORT_STATUS(mp->port_num)) & TX_FIFO_EMPTY))
 		udelay(10);
 
@@ -1875,8 +1928,9 @@ static int mv643xx_eth_stop(struct net_device *dev)
 	for (i = 0; i < 8; i++) {
 		if (mp->rxq_mask & (1 << i))
 			rxq_deinit(mp->rxq + i);
+		if (mp->txq_mask & (1 << i))
+			txq_deinit(mp->txq + i);
 	}
-	txq_deinit(mp->txq);
 
 	return 0;
 }
@@ -1928,7 +1982,7 @@ static void tx_timeout_task(struct work_struct *ugly)
 		port_reset(mp);
 		port_start(mp);
 
-		__txq_maybe_wake(mp->txq);
+		__txq_maybe_wake(mp->txq + mp->txq_primary);
 	}
 }
 
@@ -2139,6 +2193,12 @@ static void set_params(struct mv643xx_eth_private *mp,
 		mp->default_tx_ring_size = pd->tx_queue_size;
 	mp->tx_desc_sram_addr = pd->tx_sram_addr;
 	mp->tx_desc_sram_size = pd->tx_sram_size;
+
+	if (pd->tx_queue_mask)
+		mp->txq_mask = pd->tx_queue_mask;
+	else
+		mp->txq_mask = 0x01;
+	mp->txq_primary = fls(mp->txq_mask) - 1;
 }
 
 static int phy_detect(struct mv643xx_eth_private *mp)
diff --git a/include/linux/mv643xx_eth.h b/include/linux/mv643xx_eth.h
index 1afd7ba6d303..12078577aef6 100644
--- a/include/linux/mv643xx_eth.h
+++ b/include/linux/mv643xx_eth.h
@@ -49,9 +49,10 @@ struct mv643xx_eth_platform_data {
 	int			duplex;
 
 	/*
-	 * Which RX queues to use.
+	 * Which RX/TX queues to use.
 	 */
 	int			rx_queue_mask;
+	int			tx_queue_mask;
 
 	/*
 	 * Override default RX/TX queue sizes if nonzero.
-- 
cgit v1.2.3


From 875ec4333b99144e2589e900a0bcd2c25c757b27 Mon Sep 17 00:00:00 2001
From: Richard Kennedy <richard@rsk.demon.co.uk>
Date: Fri, 13 Jun 2008 03:03:17 -0700
Subject: udp: reorder udp_iter_state to remove padding on 64bit builds

reorder udp_iter_state to remove padding on 64bit builds

shrinks from 24 to 16 bytes, moving to a smaller slab when
CONFIG_NET_NS is undefined & seq_net_private = {}

Signed-off-by: Richard Kennedy <richard@rsk.demon.co.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/udp.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/net/udp.h b/include/net/udp.h
index ccce83707046..7a8684855245 100644
--- a/include/net/udp.h
+++ b/include/net/udp.h
@@ -196,8 +196,8 @@ struct udp_seq_afinfo {
 struct udp_iter_state {
 	struct seq_net_private  p;
 	sa_family_t		family;
-	struct hlist_head	*hashtable;
 	int			bucket;
+	struct hlist_head	*hashtable;
 };
 
 #ifdef CONFIG_PROC_FS
-- 
cgit v1.2.3


From 564d9bdd89df4ceece5d126ff3b7db506ae06548 Mon Sep 17 00:00:00 2001
From: Michael Buesch <mb@bu3sch.de>
Date: Tue, 10 Jun 2008 14:04:14 +0200
Subject: nl80211: Fix comment merge error

The comments ended up in the wrong place due to a merge error.

Signed-off-by: Michael Buesch <mb@bu3sch.de>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/linux/nl80211.h | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/include/linux/nl80211.h b/include/linux/nl80211.h
index ea6517e58b04..aa8411e2a160 100644
--- a/include/linux/nl80211.h
+++ b/include/linux/nl80211.h
@@ -122,13 +122,13 @@ enum nl80211_commands {
 	NL80211_CMD_NEW_STATION,
 	NL80211_CMD_DEL_STATION,
 
-	/* add commands here */
-
 	NL80211_CMD_GET_MPATH,
 	NL80211_CMD_SET_MPATH,
 	NL80211_CMD_NEW_MPATH,
 	NL80211_CMD_DEL_MPATH,
 
+	/* add commands here */
+
 	/* used to define NL80211_CMD_MAX below */
 	__NL80211_CMD_AFTER_LAST,
 	NL80211_CMD_MAX = __NL80211_CMD_AFTER_LAST - 1
@@ -230,13 +230,13 @@ enum nl80211_attrs {
 
 	NL80211_ATTR_MNTR_FLAGS,
 
-	/* add attributes here, update the policy in nl80211.c */
-
 	NL80211_ATTR_MESH_ID,
 	NL80211_ATTR_STA_PLINK_ACTION,
 	NL80211_ATTR_MPATH_NEXT_HOP,
 	NL80211_ATTR_MPATH_INFO,
 
+	/* add attributes here, update the policy in nl80211.c */
+
 	__NL80211_ATTR_AFTER_LAST,
 	NL80211_ATTR_MAX = __NL80211_ATTR_AFTER_LAST - 1
 };
-- 
cgit v1.2.3


From 963f55178b25cb673ab438edaae4127b1d014bc1 Mon Sep 17 00:00:00 2001
From: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Date: Thu, 12 Jun 2008 09:47:00 +0800
Subject: iwlwifi: remove redundant flags regarding to FAT channel

This patch removes redundant flags regarding to FAT channel. Use
mac80211's flag instead.

Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Tomas Winkler <tomas.winkler@intel.com>
Signed-off-by: Zhu Yi <yi.zhu@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/net/wireless/iwlwifi/iwl-core.c     | 42 +++++++++--------------------
 drivers/net/wireless/iwlwifi/iwl-dev.h      | 14 ----------
 drivers/net/wireless/iwlwifi/iwl-eeprom.c   | 19 ++++++++-----
 drivers/net/wireless/iwlwifi/iwl4965-base.c |  6 +++--
 include/linux/ieee80211.h                   |  1 +
 5 files changed, 31 insertions(+), 51 deletions(-)

(limited to 'include')

diff --git a/drivers/net/wireless/iwlwifi/iwl-core.c b/drivers/net/wireless/iwlwifi/iwl-core.c
index 6c7617c1bb91..fad26f71d7ce 100644
--- a/drivers/net/wireless/iwlwifi/iwl-core.c
+++ b/drivers/net/wireless/iwlwifi/iwl-core.c
@@ -478,24 +478,7 @@ static int iwlcore_init_geos(struct iwl_priv *priv)
 			if (ch->flags & EEPROM_CHANNEL_RADAR)
 				geo_ch->flags |= IEEE80211_CHAN_RADAR;
 
-			switch (ch->fat_extension_channel) {
-			case HT_IE_EXT_CHANNEL_ABOVE:
-				/* only above is allowed, disable below */
-				geo_ch->flags |= IEEE80211_CHAN_NO_FAT_BELOW;
-				break;
-			case HT_IE_EXT_CHANNEL_BELOW:
-				/* only below is allowed, disable above */
-				geo_ch->flags |= IEEE80211_CHAN_NO_FAT_ABOVE;
-				break;
-			case HT_IE_EXT_CHANNEL_NONE:
-				/* fat not allowed: disable both*/
-				geo_ch->flags |= (IEEE80211_CHAN_NO_FAT_ABOVE |
-						  IEEE80211_CHAN_NO_FAT_BELOW);
-				break;
-			case HT_IE_EXT_CHANNEL_MAX:
-				/* both above and below are permitted */
-				break;
-			}
+			geo_ch->flags |= ch->fat_extension_channel;
 
 			if (ch->max_power_avg > priv->max_channel_txpower_limit)
 				priv->max_channel_txpower_limit =
@@ -507,7 +490,7 @@ static int iwlcore_init_geos(struct iwl_priv *priv)
 		/* Save flags for reg domain usage */
 		geo_ch->orig_flags = geo_ch->flags;
 
-		IWL_DEBUG_INFO("Channel %d Freq=%d[%sGHz] %s flag=0%X\n",
+		IWL_DEBUG_INFO("Channel %d Freq=%d[%sGHz] %s flag=0x%X\n",
 				ch->channel, geo_ch->center_freq,
 				is_channel_a_band(ch) ?  "5.2" : "2.4",
 				geo_ch->flags & IEEE80211_CHAN_DISABLED ?
@@ -552,6 +535,7 @@ static u8 is_single_rx_stream(struct iwl_priv *priv)
 		(priv->current_ht_config.supp_mcs_set[2] == 0)) ||
 	       priv->ps_mode == IWL_MIMO_PS_STATIC;
 }
+
 static u8 iwl_is_channel_extension(struct iwl_priv *priv,
 				   enum ieee80211_band band,
 				   u16 channel, u8 extension_chan_offset)
@@ -562,12 +546,12 @@ static u8 iwl_is_channel_extension(struct iwl_priv *priv,
 	if (!is_channel_valid(ch_info))
 		return 0;
 
-	if (extension_chan_offset == IWL_EXT_CHANNEL_OFFSET_NONE)
-		return 0;
-
-	if ((ch_info->fat_extension_channel == extension_chan_offset) ||
-	    (ch_info->fat_extension_channel == HT_IE_EXT_CHANNEL_MAX))
-		return 1;
+	if (extension_chan_offset == IEEE80211_HT_IE_CHA_SEC_ABOVE)
+		return !(ch_info->fat_extension_channel &
+					IEEE80211_CHAN_NO_FAT_ABOVE);
+	else if (extension_chan_offset == IEEE80211_HT_IE_CHA_SEC_BELOW)
+		return !(ch_info->fat_extension_channel &
+					IEEE80211_CHAN_NO_FAT_BELOW);
 
 	return 0;
 }
@@ -579,7 +563,7 @@ u8 iwl_is_fat_tx_allowed(struct iwl_priv *priv,
 
 	if ((!iwl_ht_conf->is_ht) ||
 	   (iwl_ht_conf->supported_chan_width != IWL_CHANNEL_WIDTH_40MHZ) ||
-	   (iwl_ht_conf->extension_chan_offset == IWL_EXT_CHANNEL_OFFSET_NONE))
+	   (iwl_ht_conf->extension_chan_offset == IEEE80211_HT_IE_CHA_SEC_NONE))
 		return 0;
 
 	if (sta_ht_inf) {
@@ -619,13 +603,13 @@ void iwl_set_rxon_ht(struct iwl_priv *priv, struct iwl_ht_info *ht_info)
 
 	/* Note: control channel is opposite of extension channel */
 	switch (ht_info->extension_chan_offset) {
-	case IWL_EXT_CHANNEL_OFFSET_ABOVE:
+	case IEEE80211_HT_IE_CHA_SEC_ABOVE:
 		rxon->flags &= ~(RXON_FLG_CTRL_CHANNEL_LOC_HI_MSK);
 		break;
-	case IWL_EXT_CHANNEL_OFFSET_BELOW:
+	case IEEE80211_HT_IE_CHA_SEC_BELOW:
 		rxon->flags |= RXON_FLG_CTRL_CHANNEL_LOC_HI_MSK;
 		break;
-	case IWL_EXT_CHANNEL_OFFSET_NONE:
+	case IEEE80211_HT_IE_CHA_SEC_NONE:
 	default:
 		rxon->flags &= ~RXON_FLG_CHANNEL_MODE_MIXED_MSK;
 		break;
diff --git a/drivers/net/wireless/iwlwifi/iwl-dev.h b/drivers/net/wireless/iwlwifi/iwl-dev.h
index fa35193e5a7a..daa14d9e5a80 100644
--- a/drivers/net/wireless/iwlwifi/iwl-dev.h
+++ b/drivers/net/wireless/iwlwifi/iwl-dev.h
@@ -179,15 +179,6 @@ struct iwl4965_scan_power_info {
 	s8 requested_power;	/* scan pwr (dBm) requested for chnl/rate */
 };
 
-/* For fat_extension_channel */
-enum {
-	HT_IE_EXT_CHANNEL_NONE = 0,
-	HT_IE_EXT_CHANNEL_ABOVE,
-	HT_IE_EXT_CHANNEL_INVALID,
-	HT_IE_EXT_CHANNEL_BELOW,
-	HT_IE_EXT_CHANNEL_MAX
-};
-
 /*
  * One for each channel, holds all channel setup data
  * Some of the fields (e.g. eeprom and flags/max_power_avg) are redundant
@@ -782,11 +773,6 @@ struct iwl_kw {
 #define IWL_OPERATION_MODE_MIXED    2
 #define IWL_OPERATION_MODE_20MHZ    3
 
-#define IWL_EXT_CHANNEL_OFFSET_NONE      0
-#define IWL_EXT_CHANNEL_OFFSET_ABOVE     1
-#define IWL_EXT_CHANNEL_OFFSET_RESERVE1  2
-#define IWL_EXT_CHANNEL_OFFSET_BELOW     3
-
 #define IWL_TX_CRC_SIZE 4
 #define IWL_TX_DELIMITER_SIZE 4
 
diff --git a/drivers/net/wireless/iwlwifi/iwl-eeprom.c b/drivers/net/wireless/iwlwifi/iwl-eeprom.c
index 11f9d9557a0e..cbb812f9f620 100644
--- a/drivers/net/wireless/iwlwifi/iwl-eeprom.c
+++ b/drivers/net/wireless/iwlwifi/iwl-eeprom.c
@@ -470,6 +470,11 @@ int iwl_init_channel_map(struct iwl_priv *priv)
 			/* Copy the run-time flags so they are there even on
 			 * invalid channels */
 			ch_info->flags = eeprom_ch_info[ch].flags;
+			/* First write that fat is not enabled, and then enable
+			 * one by one */
+			ch_info->fat_extension_channel =
+				(IEEE80211_CHAN_NO_FAT_ABOVE |
+				 IEEE80211_CHAN_NO_FAT_BELOW);
 
 			if (!(is_channel_valid(ch_info))) {
 				IWL_DEBUG_INFO("Ch. %d Flags %x [%sGHz] - "
@@ -534,12 +539,14 @@ int iwl_init_channel_map(struct iwl_priv *priv)
 		for (ch = 0; ch < eeprom_ch_count; ch++) {
 
 			if ((band == 6) &&
-			    ((eeprom_ch_index[ch] == 5) ||
-			    (eeprom_ch_index[ch] == 6) ||
-			    (eeprom_ch_index[ch] == 7)))
-			       fat_extension_chan = HT_IE_EXT_CHANNEL_MAX;
+				((eeprom_ch_index[ch] == 5) ||
+				 (eeprom_ch_index[ch] == 6) ||
+				 (eeprom_ch_index[ch] == 7)))
+				/* both are allowed: above and below */
+				fat_extension_chan = 0;
 			else
-				fat_extension_chan = HT_IE_EXT_CHANNEL_ABOVE;
+				fat_extension_chan =
+					IEEE80211_CHAN_NO_FAT_BELOW;
 
 			/* Set up driver's info for lower half */
 			iwl_set_fat_chan_info(priv, ieeeband,
@@ -551,7 +558,7 @@ int iwl_init_channel_map(struct iwl_priv *priv)
 			iwl_set_fat_chan_info(priv, ieeeband,
 						(eeprom_ch_index[ch] + 4),
 						&(eeprom_ch_info[ch]),
-						HT_IE_EXT_CHANNEL_BELOW);
+						IEEE80211_CHAN_NO_FAT_ABOVE);
 		}
 	}
 
diff --git a/drivers/net/wireless/iwlwifi/iwl4965-base.c b/drivers/net/wireless/iwlwifi/iwl4965-base.c
index 0acd42bd43c5..de40e893f09e 100644
--- a/drivers/net/wireless/iwlwifi/iwl4965-base.c
+++ b/drivers/net/wireless/iwlwifi/iwl4965-base.c
@@ -664,9 +664,11 @@ static void iwl4965_ht_conf(struct iwl_priv *priv,
 	iwl_conf->extension_chan_offset =
 		ht_bss_conf->bss_cap & IEEE80211_HT_IE_CHA_SEC_OFFSET;
 	/* If no above or below channel supplied disable FAT channel */
-	if (iwl_conf->extension_chan_offset != IWL_EXT_CHANNEL_OFFSET_ABOVE &&
-	    iwl_conf->extension_chan_offset != IWL_EXT_CHANNEL_OFFSET_BELOW)
+	if (iwl_conf->extension_chan_offset != IEEE80211_HT_IE_CHA_SEC_ABOVE &&
+	    iwl_conf->extension_chan_offset != IEEE80211_HT_IE_CHA_SEC_BELOW) {
+		iwl_conf->extension_chan_offset = IEEE80211_HT_IE_CHA_SEC_NONE;
 		iwl_conf->supported_chan_width = 0;
+	}
 
 	iwl_conf->tx_mimo_ps_mode =
 		(u8)((ht_conf->cap & IEEE80211_HT_CAP_MIMO_PS) >> 2);
diff --git a/include/linux/ieee80211.h b/include/linux/ieee80211.h
index 9300f37cd7e8..8f2c20b4a15d 100644
--- a/include/linux/ieee80211.h
+++ b/include/linux/ieee80211.h
@@ -320,6 +320,7 @@ struct ieee80211_ht_addt_info {
 #define IEEE80211_HT_CAP_MCS_TX_UEQM		0x10
 /* 802.11n HT IE masks */
 #define IEEE80211_HT_IE_CHA_SEC_OFFSET		0x03
+#define IEEE80211_HT_IE_CHA_SEC_NONE	 	0x00
 #define IEEE80211_HT_IE_CHA_SEC_ABOVE 		0x01
 #define IEEE80211_HT_IE_CHA_SEC_BELOW 		0x03
 #define IEEE80211_HT_IE_CHA_WIDTH		0x04
-- 
cgit v1.2.3


From fd7c8a40b2a63863f749e4d17f0d94d2e5ab1331 Mon Sep 17 00:00:00 2001
From: Harvey Harrison <harvey.harrison@gmail.com>
Date: Wed, 11 Jun 2008 14:21:56 -0700
Subject: mac80211: add helpers for frame control testing

A few general categories:

1) ieee80211_has_* tests if particular fctl bits are set, the helpers are de
in the same order as the fctl defines:

A combined _has_a4 was also added to test when both FROMDS and TODS are set.

2) ieee80211_is_* is meant to test whether the frame control is of a certain
ftype - data, mgmt, ctl, and two special helpers _is_data_qos, _is_data_pres
which also test a subset of the stype space.

When testing for a particular stype applicable only to one ftype, functions
like ieee80211_is_ack have been added.  Note that the ftype is also being
checked in these helpers.  They have been added for all mgmt and ctl stypes
in the same order as the STYPE defines.

3) ieee80211_get_* is meant to take a struct ieee80211_hdr * and returns a
pointer to somewhere in the struct, see get_SA, get_DA, get_qos_ctl.

The intel wireless drivers had helpers that used this namespace, convert the
all to use the new helpers and remove the byteshifting as they were defined
in cpu-order rather than little-endian.

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/net/wireless/iwlwifi/iwl-3945.c     |  36 +--
 drivers/net/wireless/iwlwifi/iwl-4965-rs.c  |  17 +-
 drivers/net/wireless/iwlwifi/iwl-4965.c     |  28 +-
 drivers/net/wireless/iwlwifi/iwl-5000.c     |   6 +-
 drivers/net/wireless/iwlwifi/iwl-helpers.h  | 109 --------
 drivers/net/wireless/iwlwifi/iwl-tx.c       |  49 ++--
 drivers/net/wireless/iwlwifi/iwl3945-base.c |  33 ++-
 include/linux/ieee80211.h                   | 392 ++++++++++++++++++++++++++--
 8 files changed, 457 insertions(+), 213 deletions(-)

(limited to 'include')

diff --git a/drivers/net/wireless/iwlwifi/iwl-3945.c b/drivers/net/wireless/iwlwifi/iwl-3945.c
index 0ba6889dfd41..63f20370032d 100644
--- a/drivers/net/wireless/iwlwifi/iwl-3945.c
+++ b/drivers/net/wireless/iwlwifi/iwl-3945.c
@@ -388,7 +388,7 @@ static void iwl3945_dbg_report_frame(struct iwl3945_priv *priv,
 	u32 print_dump = 0;	/* set to 1 to dump all frames' contents */
 	u32 hundred = 0;
 	u32 dataframe = 0;
-	u16 fc;
+	__le16 fc;
 	u16 seq_ctl;
 	u16 channel;
 	u16 phy_flags;
@@ -407,7 +407,7 @@ static void iwl3945_dbg_report_frame(struct iwl3945_priv *priv,
 	u8 *data = IWL_RX_DATA(pkt);
 
 	/* MAC header */
-	fc = le16_to_cpu(header->frame_control);
+	fc = header->frame_control;
 	seq_ctl = le16_to_cpu(header->seq_ctrl);
 
 	/* metadata */
@@ -431,8 +431,8 @@ static void iwl3945_dbg_report_frame(struct iwl3945_priv *priv,
 
 	/* if data frame is to us and all is good,
 	 *   (optionally) print summary for only 1 out of every 100 */
-	if (to_us && (fc & ~IEEE80211_FCTL_PROTECTED) ==
-	    (IEEE80211_FCTL_FROMDS | IEEE80211_FTYPE_DATA)) {
+	if (to_us && (fc & ~cpu_to_le16(IEEE80211_FCTL_PROTECTED)) ==
+	    cpu_to_le16(IEEE80211_FCTL_FROMDS | IEEE80211_FTYPE_DATA)) {
 		dataframe = 1;
 		if (!group100)
 			print_summary = 1;	/* print each frame */
@@ -455,13 +455,13 @@ static void iwl3945_dbg_report_frame(struct iwl3945_priv *priv,
 
 		if (hundred)
 			title = "100Frames";
-		else if (fc & IEEE80211_FCTL_RETRY)
+		else if (ieee80211_has_retry(fc))
 			title = "Retry";
-		else if (ieee80211_is_assoc_response(fc))
+		else if (ieee80211_is_assoc_resp(fc))
 			title = "AscRsp";
-		else if (ieee80211_is_reassoc_response(fc))
+		else if (ieee80211_is_reassoc_resp(fc))
 			title = "RasRsp";
-		else if (ieee80211_is_probe_response(fc)) {
+		else if (ieee80211_is_probe_resp(fc)) {
 			title = "PrbRsp";
 			print_dump = 1;	/* dump frame contents */
 		} else if (ieee80211_is_beacon(fc)) {
@@ -490,14 +490,14 @@ static void iwl3945_dbg_report_frame(struct iwl3945_priv *priv,
 		if (dataframe)
 			IWL_DEBUG_RX("%s: mhd=0x%04x, dst=0x%02x, "
 				     "len=%u, rssi=%d, chnl=%d, rate=%u, \n",
-				     title, fc, header->addr1[5],
+				     title, le16_to_cpu(fc), header->addr1[5],
 				     length, rssi, channel, rate);
 		else {
 			/* src/dst addresses assume managed mode */
 			IWL_DEBUG_RX("%s: 0x%04x, dst=0x%02x, "
 				     "src=0x%02x, rssi=%u, tim=%lu usec, "
 				     "phy=0x%02x, chnl=%d\n",
-				     title, fc, header->addr1[5],
+				     title, le16_to_cpu(fc), header->addr1[5],
 				     header->addr3[5], rssi,
 				     tsf_low - priv->scan_start_tsf,
 				     phy_flags, channel);
@@ -971,7 +971,7 @@ void iwl3945_hw_build_tx_cmd_rate(struct iwl3945_priv *priv,
 	u8 rts_retry_limit;
 	u8 data_retry_limit;
 	__le32 tx_flags;
-	u16 fc = le16_to_cpu(hdr->frame_control);
+	__le16 fc = hdr->frame_control;
 
 	rate = iwl3945_rates[rate_index].plcp;
 	tx_flags = cmd->cmd.tx.tx_flags;
@@ -996,7 +996,7 @@ void iwl3945_hw_build_tx_cmd_rate(struct iwl3945_priv *priv,
 	else
 		rts_retry_limit = 7;
 
-	if (ieee80211_is_probe_response(fc)) {
+	if (ieee80211_is_probe_resp(fc)) {
 		data_retry_limit = 3;
 		if (data_retry_limit < rts_retry_limit)
 			rts_retry_limit = data_retry_limit;
@@ -1006,12 +1006,12 @@ void iwl3945_hw_build_tx_cmd_rate(struct iwl3945_priv *priv,
 	if (priv->data_retry_limit != -1)
 		data_retry_limit = priv->data_retry_limit;
 
-	if ((fc & IEEE80211_FCTL_FTYPE) == IEEE80211_FTYPE_MGMT) {
-		switch (fc & IEEE80211_FCTL_STYPE) {
-		case IEEE80211_STYPE_AUTH:
-		case IEEE80211_STYPE_DEAUTH:
-		case IEEE80211_STYPE_ASSOC_REQ:
-		case IEEE80211_STYPE_REASSOC_REQ:
+	if (ieee80211_is_mgmt(fc)) {
+		switch (fc & cpu_to_le16(IEEE80211_FCTL_STYPE)) {
+		case cpu_to_le16(IEEE80211_STYPE_AUTH):
+		case cpu_to_le16(IEEE80211_STYPE_DEAUTH):
+		case cpu_to_le16(IEEE80211_STYPE_ASSOC_REQ):
+		case cpu_to_le16(IEEE80211_STYPE_REASSOC_REQ):
 			if (tx_flags & TX_CMD_FLG_RTS_MSK) {
 				tx_flags &= ~TX_CMD_FLG_RTS_MSK;
 				tx_flags |= TX_CMD_FLG_CTS_MSK;
diff --git a/drivers/net/wireless/iwlwifi/iwl-4965-rs.c b/drivers/net/wireless/iwlwifi/iwl-4965-rs.c
index d601d31e0de9..202303117285 100644
--- a/drivers/net/wireless/iwlwifi/iwl-4965-rs.c
+++ b/drivers/net/wireless/iwlwifi/iwl-4965-rs.c
@@ -281,11 +281,11 @@ static u8 rs_tl_add_packet(struct iwl4965_lq_sta *lq_data,
 	u32 time_diff;
 	s32 index;
 	struct iwl4965_traffic_load *tl = NULL;
-	u16 fc = le16_to_cpu(hdr->frame_control);
+	__le16 fc = hdr->frame_control;
 	u8 tid;
 
-	if (ieee80211_is_qos_data(fc)) {
-		u8 *qc = ieee80211_get_qos_ctrl(hdr, ieee80211_get_hdrlen(fc));
+	if (ieee80211_is_data_qos(fc)) {
+		u8 *qc = ieee80211_get_qos_ctl(hdr);
 		tid = qc[0] & 0xf;
 	} else
 		return MAX_TID_COUNT;
@@ -794,7 +794,7 @@ static void rs_tx_status(void *priv_rate, struct net_device *dev,
 	struct iwl4965_scale_tbl_info tbl_type;
 	struct iwl4965_scale_tbl_info *curr_tbl, *search_tbl;
 	u8 active_index = 0;
-	u16 fc = le16_to_cpu(hdr->frame_control);
+	__le16 fc = hdr->frame_control;
 	s32 tpt = 0;
 
 	IWL_DEBUG_RATE_LIMIT("get frame ack response, update rate scale window\n");
@@ -1651,7 +1651,8 @@ static void rs_rate_scale_perform(struct iwl_priv *priv,
 	int high_tpt = IWL_INVALID_VALUE;
 	u32 fail_count;
 	s8 scale_action = 0;
-	u16 fc, rate_mask;
+	__le16 fc;
+	u16 rate_mask;
 	u8 update_lq = 0;
 	struct iwl4965_lq_sta *lq_sta;
 	struct iwl4965_scale_tbl_info *tbl, *tbl1;
@@ -1666,7 +1667,7 @@ static void rs_rate_scale_perform(struct iwl_priv *priv,
 
 	IWL_DEBUG_RATE("rate scale calculate new rate for skb\n");
 
-	fc = le16_to_cpu(hdr->frame_control);
+	fc = hdr->frame_control;
 	if (!ieee80211_is_data(fc) || is_multicast_ether_addr(hdr->addr1)) {
 		/* Send management frames and broadcast/multicast data using
 		 * lowest rate. */
@@ -2100,7 +2101,7 @@ static void rs_get_rate(void *priv_rate, struct net_device *dev,
 	struct ieee80211_conf *conf = &local->hw.conf;
 	struct ieee80211_hdr *hdr = (struct ieee80211_hdr *)skb->data;
 	struct sta_info *sta;
-	u16 fc;
+	__le16 fc;
 	struct iwl_priv *priv = (struct iwl_priv *)priv_rate;
 	struct iwl4965_lq_sta *lq_sta;
 
@@ -2112,7 +2113,7 @@ static void rs_get_rate(void *priv_rate, struct net_device *dev,
 
 	/* Send management frames and broadcast/multicast data using lowest
 	 * rate. */
-	fc = le16_to_cpu(hdr->frame_control);
+	fc = hdr->frame_control;
 	if (!ieee80211_is_data(fc) || is_multicast_ether_addr(hdr->addr1) ||
 	    !sta || !sta->rate_ctrl_priv) {
 		sel->rate_idx = rate_lowest_index(local, sband, sta);
diff --git a/drivers/net/wireless/iwlwifi/iwl-4965.c b/drivers/net/wireless/iwlwifi/iwl-4965.c
index df345cb6fd7d..ab5027345a01 100644
--- a/drivers/net/wireless/iwlwifi/iwl-4965.c
+++ b/drivers/net/wireless/iwlwifi/iwl-4965.c
@@ -2388,7 +2388,7 @@ static void iwl4965_dbg_report_frame(struct iwl_priv *priv,
 	u32 print_dump = 0;	/* set to 1 to dump all frames' contents */
 	u32 hundred = 0;
 	u32 dataframe = 0;
-	u16 fc;
+	__le16 fc;
 	u16 seq_ctl;
 	u16 channel;
 	u16 phy_flags;
@@ -2411,7 +2411,7 @@ static void iwl4965_dbg_report_frame(struct iwl_priv *priv,
 		return;
 
 	/* MAC header */
-	fc = le16_to_cpu(header->frame_control);
+	fc = header->frame_control;
 	seq_ctl = le16_to_cpu(header->seq_ctrl);
 
 	/* metadata */
@@ -2436,8 +2436,8 @@ static void iwl4965_dbg_report_frame(struct iwl_priv *priv,
 
 	/* if data frame is to us and all is good,
 	 *   (optionally) print summary for only 1 out of every 100 */
-	if (to_us && (fc & ~IEEE80211_FCTL_PROTECTED) ==
-	    (IEEE80211_FCTL_FROMDS | IEEE80211_FTYPE_DATA)) {
+	if (to_us && (fc & ~cpu_to_le16(IEEE80211_FCTL_PROTECTED)) ==
+	    cpu_to_le16(IEEE80211_FCTL_FROMDS | IEEE80211_FTYPE_DATA)) {
 		dataframe = 1;
 		if (!group100)
 			print_summary = 1;	/* print each frame */
@@ -2461,13 +2461,13 @@ static void iwl4965_dbg_report_frame(struct iwl_priv *priv,
 
 		if (hundred)
 			title = "100Frames";
-		else if (fc & IEEE80211_FCTL_RETRY)
+		else if (ieee80211_has_retry(fc))
 			title = "Retry";
-		else if (ieee80211_is_assoc_response(fc))
+		else if (ieee80211_is_assoc_resp(fc))
 			title = "AscRsp";
-		else if (ieee80211_is_reassoc_response(fc))
+		else if (ieee80211_is_reassoc_resp(fc))
 			title = "RasRsp";
-		else if (ieee80211_is_probe_response(fc)) {
+		else if (ieee80211_is_probe_resp(fc)) {
 			title = "PrbRsp";
 			print_dump = 1;	/* dump frame contents */
 		} else if (ieee80211_is_beacon(fc)) {
@@ -2496,14 +2496,14 @@ static void iwl4965_dbg_report_frame(struct iwl_priv *priv,
 		if (dataframe)
 			IWL_DEBUG_RX("%s: mhd=0x%04x, dst=0x%02x, "
 				     "len=%u, rssi=%d, chnl=%d, rate=%u, \n",
-				     title, fc, header->addr1[5],
+				     title, le16_to_cpu(fc), header->addr1[5],
 				     length, rssi, channel, bitrate);
 		else {
 			/* src/dst addresses assume managed mode */
 			IWL_DEBUG_RX("%s: 0x%04x, dst=0x%02x, "
 				     "src=0x%02x, rssi=%u, tim=%lu usec, "
 				     "phy=0x%02x, chnl=%d\n",
-				     title, fc, header->addr1[5],
+				     title, le16_to_cpu(fc), header->addr1[5],
 				     header->addr3[5], rssi,
 				     tsf_low - priv->scan_start_tsf,
 				     phy_flags, channel);
@@ -3219,7 +3219,7 @@ static void iwl4965_rx_reply_tx(struct iwl_priv *priv,
 	struct iwl4965_tx_resp *tx_resp = (void *)&pkt->u.raw[0];
 	u32  status = le32_to_cpu(tx_resp->u.status);
 	int tid = MAX_TID_COUNT, sta_id = IWL_INVALID_STATION;
-	u16 fc;
+	__le16 fc;
 	struct ieee80211_hdr *hdr;
 	u8 *qc = NULL;
 
@@ -3235,9 +3235,9 @@ static void iwl4965_rx_reply_tx(struct iwl_priv *priv,
 	memset(&info->status, 0, sizeof(info->status));
 
 	hdr = iwl_tx_queue_get_hdr(priv, txq_id, index);
-	fc = le16_to_cpu(hdr->frame_control);
-	if (ieee80211_is_qos_data(fc)) {
-		qc = ieee80211_get_qos_ctrl(hdr, ieee80211_get_hdrlen(fc));
+	fc = hdr->frame_control;
+	if (ieee80211_is_data_qos(fc)) {
+		qc = ieee80211_get_qos_ctl(hdr);
 		tid = qc[0] & 0xf;
 	}
 
diff --git a/drivers/net/wireless/iwlwifi/iwl-5000.c b/drivers/net/wireless/iwlwifi/iwl-5000.c
index 8c466b02bdff..438c3812c390 100644
--- a/drivers/net/wireless/iwlwifi/iwl-5000.c
+++ b/drivers/net/wireless/iwlwifi/iwl-5000.c
@@ -1252,7 +1252,6 @@ static void iwl5000_rx_reply_tx(struct iwl_priv *priv,
 	struct iwl5000_tx_resp *tx_resp = (void *)&pkt->u.raw[0];
 	u32  status = le16_to_cpu(tx_resp->status.status);
 	int tid = MAX_TID_COUNT, sta_id = IWL_INVALID_STATION;
-	u16 fc;
 	struct ieee80211_hdr *hdr;
 	u8 *qc = NULL;
 
@@ -1268,9 +1267,8 @@ static void iwl5000_rx_reply_tx(struct iwl_priv *priv,
 	memset(&info->status, 0, sizeof(info->status));
 
 	hdr = iwl_tx_queue_get_hdr(priv, txq_id, index);
-	fc = le16_to_cpu(hdr->frame_control);
-	if (ieee80211_is_qos_data(fc)) {
-		qc = ieee80211_get_qos_ctrl(hdr, ieee80211_get_hdrlen(fc));
+	if (ieee80211_is_data_qos(hdr->frame_control)) {
+		qc = ieee80211_get_qos_ctl(hdr);
 		tid = qc[0] & 0xf;
 	}
 
diff --git a/drivers/net/wireless/iwlwifi/iwl-helpers.h b/drivers/net/wireless/iwlwifi/iwl-helpers.h
index eec423a98fe6..41eed6793328 100644
--- a/drivers/net/wireless/iwlwifi/iwl-helpers.h
+++ b/drivers/net/wireless/iwlwifi/iwl-helpers.h
@@ -145,115 +145,6 @@ static inline struct ieee80211_conf *ieee80211_get_hw_conf(
 	return &hw->conf;
 }
 
-#define QOS_CONTROL_LEN 2
-
-
-static inline int ieee80211_is_management(u16 fc)
-{
-	return (fc & IEEE80211_FCTL_FTYPE) == IEEE80211_FTYPE_MGMT;
-}
-
-static inline int ieee80211_is_control(u16 fc)
-{
-	return (fc & IEEE80211_FCTL_FTYPE) == IEEE80211_FTYPE_CTL;
-}
-
-static inline int ieee80211_is_data(u16 fc)
-{
-	return (fc & IEEE80211_FCTL_FTYPE) == IEEE80211_FTYPE_DATA;
-}
-
-static inline int ieee80211_is_back_request(u16 fc)
-{
-	return ((fc & IEEE80211_FCTL_FTYPE) == IEEE80211_FTYPE_CTL) &&
-	       ((fc & IEEE80211_FCTL_STYPE) == IEEE80211_STYPE_BACK_REQ);
-}
-
-static inline int ieee80211_is_probe_response(u16 fc)
-{
-	return ((fc & IEEE80211_FCTL_FTYPE) == IEEE80211_FTYPE_MGMT) &&
-	       ((fc & IEEE80211_FCTL_STYPE) == IEEE80211_STYPE_PROBE_RESP);
-}
-
-static inline int ieee80211_is_probe_request(u16 fc)
-{
-	return ((fc & IEEE80211_FCTL_FTYPE) == IEEE80211_FTYPE_MGMT) &&
-	       ((fc & IEEE80211_FCTL_STYPE) == IEEE80211_STYPE_PROBE_REQ);
-}
-
-static inline int ieee80211_is_beacon(u16 fc)
-{
-	return ((fc & IEEE80211_FCTL_FTYPE) == IEEE80211_FTYPE_MGMT) &&
-	       ((fc & IEEE80211_FCTL_STYPE) == IEEE80211_STYPE_BEACON);
-}
-
-static inline int ieee80211_is_atim(u16 fc)
-{
-	return ((fc & IEEE80211_FCTL_FTYPE) == IEEE80211_FTYPE_MGMT) &&
-	       ((fc & IEEE80211_FCTL_STYPE) == IEEE80211_STYPE_ATIM);
-}
-
-static inline int ieee80211_is_assoc_request(u16 fc)
-{
-	return ((fc & IEEE80211_FCTL_FTYPE) == IEEE80211_FTYPE_MGMT) &&
-	       ((fc & IEEE80211_FCTL_STYPE) == IEEE80211_STYPE_ASSOC_REQ);
-}
-
-static inline int ieee80211_is_assoc_response(u16 fc)
-{
-	return ((fc & IEEE80211_FCTL_FTYPE) == IEEE80211_FTYPE_MGMT) &&
-	       ((fc & IEEE80211_FCTL_STYPE) == IEEE80211_STYPE_ASSOC_RESP);
-}
-
-static inline int ieee80211_is_auth(u16 fc)
-{
-	return ((fc & IEEE80211_FCTL_FTYPE) == IEEE80211_FTYPE_MGMT) &&
-	       ((fc & IEEE80211_FCTL_STYPE) == IEEE80211_STYPE_ASSOC_REQ);
-}
-
-static inline int ieee80211_is_deauth(u16 fc)
-{
-	return ((fc & IEEE80211_FCTL_FTYPE) == IEEE80211_FTYPE_MGMT) &&
-	       ((fc & IEEE80211_FCTL_STYPE) == IEEE80211_STYPE_ASSOC_REQ);
-}
-
-static inline int ieee80211_is_disassoc(u16 fc)
-{
-	return ((fc & IEEE80211_FCTL_FTYPE) == IEEE80211_FTYPE_MGMT) &&
-	       ((fc & IEEE80211_FCTL_STYPE) == IEEE80211_STYPE_ASSOC_REQ);
-}
-
-static inline int ieee80211_is_reassoc_request(u16 fc)
-{
-	return ((fc & IEEE80211_FCTL_FTYPE) == IEEE80211_FTYPE_MGMT) &&
-	       ((fc & IEEE80211_FCTL_STYPE) == IEEE80211_STYPE_REASSOC_REQ);
-}
-
-static inline int ieee80211_is_reassoc_response(u16 fc)
-{
-	return ((fc & IEEE80211_FCTL_FTYPE) == IEEE80211_FTYPE_MGMT) &&
-	       ((fc & IEEE80211_FCTL_STYPE) == IEEE80211_STYPE_REASSOC_RESP);
-}
-
-static inline int ieee80211_is_qos_data(u16 fc)
-{
-	return ((fc & IEEE80211_FCTL_FTYPE) == IEEE80211_FTYPE_DATA) &&
-	       ((fc & IEEE80211_FCTL_STYPE) == IEEE80211_STYPE_QOS_DATA);
-}
-/**
- * ieee80211_get_qos_ctrl - get pointer to the QoS control field
- *
- * This function returns the pointer to 802.11 header QoS field (2 bytes)
- * This function doesn't check whether hdr is a QoS hdr, use with care
- * @hdr: struct ieee80211_hdr *hdr
- * @hdr_len: header length
- */
-
-static inline u8 *ieee80211_get_qos_ctrl(struct ieee80211_hdr *hdr, int hdr_len)
-{
-	return  ((u8 *) hdr + hdr_len - QOS_CONTROL_LEN);
-}
-
 static inline int iwl_check_bits(unsigned long field, unsigned long mask)
 {
 	return ((field & mask) == mask) ? 1 : 0;
diff --git a/drivers/net/wireless/iwlwifi/iwl-tx.c b/drivers/net/wireless/iwlwifi/iwl-tx.c
index e804bf8aea80..98c434c52a64 100644
--- a/drivers/net/wireless/iwlwifi/iwl-tx.c
+++ b/drivers/net/wireless/iwlwifi/iwl-tx.c
@@ -569,15 +569,15 @@ static void iwl_tx_cmd_build_basic(struct iwl_priv *priv,
 				  struct ieee80211_hdr *hdr,
 				  int is_unicast, u8 std_id)
 {
-	u16 fc = le16_to_cpu(hdr->frame_control);
+	__le16 fc = hdr->frame_control;
 	__le32 tx_flags = tx_cmd->tx_flags;
 
 	tx_cmd->stop_time.life_time = TX_CMD_LIFE_TIME_INFINITE;
 	if (!(info->flags & IEEE80211_TX_CTL_NO_ACK)) {
 		tx_flags |= TX_CMD_FLG_ACK_MSK;
-		if ((fc & IEEE80211_FCTL_FTYPE) == IEEE80211_FTYPE_MGMT)
+		if (ieee80211_is_mgmt(fc))
 			tx_flags |= TX_CMD_FLG_SEQ_CTL_MSK;
-		if (ieee80211_is_probe_response(fc) &&
+		if (ieee80211_is_probe_resp(fc) &&
 		    !(le16_to_cpu(hdr->seq_ctrl) & 0xf))
 			tx_flags |= TX_CMD_FLG_TSF_MSK;
 	} else {
@@ -585,7 +585,7 @@ static void iwl_tx_cmd_build_basic(struct iwl_priv *priv,
 		tx_flags |= TX_CMD_FLG_SEQ_CTL_MSK;
 	}
 
-	if (ieee80211_is_back_request(fc))
+	if (ieee80211_is_back_req(fc))
 		tx_flags |= TX_CMD_FLG_ACK_MSK | TX_CMD_FLG_IMM_BA_RSP_MASK;
 
 
@@ -593,8 +593,8 @@ static void iwl_tx_cmd_build_basic(struct iwl_priv *priv,
 	if (ieee80211_get_morefrag(hdr))
 		tx_flags |= TX_CMD_FLG_MORE_FRAG_MSK;
 
-	if (ieee80211_is_qos_data(fc)) {
-		u8 *qc = ieee80211_get_qos_ctrl(hdr, ieee80211_get_hdrlen(fc));
+	if (ieee80211_is_data_qos(fc)) {
+		u8 *qc = ieee80211_get_qos_ctl(hdr);
 		tx_cmd->tid_tspec = qc[0] & 0xf;
 		tx_flags &= ~TX_CMD_FLG_SEQ_CTL_MSK;
 	} else {
@@ -613,9 +613,8 @@ static void iwl_tx_cmd_build_basic(struct iwl_priv *priv,
 		tx_flags |= TX_CMD_FLG_FULL_TXOP_PROT_MSK;
 
 	tx_flags &= ~(TX_CMD_FLG_ANT_SEL_MSK);
-	if ((fc & IEEE80211_FCTL_FTYPE) == IEEE80211_FTYPE_MGMT) {
-		if ((fc & IEEE80211_FCTL_STYPE) == IEEE80211_STYPE_ASSOC_REQ ||
-		    (fc & IEEE80211_FCTL_STYPE) == IEEE80211_STYPE_REASSOC_REQ)
+	if (ieee80211_is_mgmt(fc)) {
+		if (ieee80211_is_assoc_req(fc) || ieee80211_is_reassoc_req(fc))
 			tx_cmd->timeout.pm_frame_timeout = cpu_to_le16(3);
 		else
 			tx_cmd->timeout.pm_frame_timeout = cpu_to_le16(2);
@@ -634,7 +633,7 @@ static void iwl_tx_cmd_build_basic(struct iwl_priv *priv,
 static void iwl_tx_cmd_build_rate(struct iwl_priv *priv,
 			      struct iwl_tx_cmd *tx_cmd,
 			      struct ieee80211_tx_info *info,
-			      u16 fc, int sta_id,
+			      __le16 fc, int sta_id,
 			      int is_hcca)
 {
 	u8 rts_retry_limit = 0;
@@ -655,7 +654,7 @@ static void iwl_tx_cmd_build_rate(struct iwl_priv *priv,
 		rate_flags |= RATE_MCS_CCK_MSK;
 
 
-	if (ieee80211_is_probe_response(fc)) {
+	if (ieee80211_is_probe_resp(fc)) {
 		data_retry_limit = 3;
 		if (data_retry_limit < rts_retry_limit)
 			rts_retry_limit = data_retry_limit;
@@ -670,11 +669,11 @@ static void iwl_tx_cmd_build_rate(struct iwl_priv *priv,
 		tx_cmd->initial_rate_index = 0;
 		tx_cmd->tx_flags |= TX_CMD_FLG_STA_RATE_MSK;
 	} else {
-		switch (fc & IEEE80211_FCTL_STYPE) {
-		case IEEE80211_STYPE_AUTH:
-		case IEEE80211_STYPE_DEAUTH:
-		case IEEE80211_STYPE_ASSOC_REQ:
-		case IEEE80211_STYPE_REASSOC_REQ:
+		switch (fc & cpu_to_le16(IEEE80211_FCTL_STYPE)) {
+		case cpu_to_le16(IEEE80211_STYPE_AUTH):
+		case cpu_to_le16(IEEE80211_STYPE_DEAUTH):
+		case cpu_to_le16(IEEE80211_STYPE_ASSOC_REQ):
+		case cpu_to_le16(IEEE80211_STYPE_REASSOC_REQ):
 			if (tx_cmd->tx_flags & TX_CMD_FLG_RTS_MSK) {
 				tx_cmd->tx_flags &= ~TX_CMD_FLG_RTS_MSK;
 				tx_cmd->tx_flags |= TX_CMD_FLG_CTS_MSK;
@@ -771,7 +770,7 @@ int iwl_tx_skb(struct iwl_priv *priv, struct sk_buff *skb)
 	u16 seq_number = 0;
 	u8 id, hdr_len, unicast;
 	u8 sta_id;
-	u16 fc;
+	__le16 fc;
 	u8 wait_write_ptr = 0;
 	u8 tid = 0;
 	u8 *qc = NULL;
@@ -798,19 +797,19 @@ int iwl_tx_skb(struct iwl_priv *priv, struct sk_buff *skb)
 	unicast = !is_multicast_ether_addr(hdr->addr1);
 	id = 0;
 
-	fc = le16_to_cpu(hdr->frame_control);
+	fc = hdr->frame_control;
 
 #ifdef CONFIG_IWLWIFI_DEBUG
 	if (ieee80211_is_auth(fc))
 		IWL_DEBUG_TX("Sending AUTH frame\n");
-	else if (ieee80211_is_assoc_request(fc))
+	else if (ieee80211_is_assoc_req(fc))
 		IWL_DEBUG_TX("Sending ASSOC frame\n");
-	else if (ieee80211_is_reassoc_request(fc))
+	else if (ieee80211_is_reassoc_req(fc))
 		IWL_DEBUG_TX("Sending REASSOC frame\n");
 #endif
 
 	/* drop all data frame if we are not associated */
-	if (((fc & IEEE80211_FCTL_FTYPE) == IEEE80211_FTYPE_DATA) &&
+	if (ieee80211_is_data(fc) &&
 	   (!iwl_is_associated(priv) ||
 	    ((priv->iw_mode == IEEE80211_IF_TYPE_STA) && !priv->assoc_id) ||
 	    !priv->assoc_station_added)) {
@@ -820,7 +819,7 @@ int iwl_tx_skb(struct iwl_priv *priv, struct sk_buff *skb)
 
 	spin_unlock_irqrestore(&priv->lock, flags);
 
-	hdr_len = ieee80211_get_hdrlen(fc);
+	hdr_len = ieee80211_get_hdrlen(le16_to_cpu(fc));
 
 	/* Find (or create) index into station table for destination station */
 	sta_id = iwl_get_sta_id(priv, hdr);
@@ -834,8 +833,8 @@ int iwl_tx_skb(struct iwl_priv *priv, struct sk_buff *skb)
 
 	IWL_DEBUG_TX("station Id %d\n", sta_id);
 
-	if (ieee80211_is_qos_data(fc)) {
-		qc = ieee80211_get_qos_ctrl(hdr, hdr_len);
+	if (ieee80211_is_data_qos(fc)) {
+		qc = ieee80211_get_qos_ctl(hdr);
 		tid = qc[0] & 0xf;
 		seq_number = priv->stations[sta_id].tid[tid].seq_number &
 				IEEE80211_SCTL_SEQ;
@@ -938,7 +937,7 @@ int iwl_tx_skb(struct iwl_priv *priv, struct sk_buff *skb)
 	/* set is_hcca to 0; it probably will never be implemented */
 	iwl_tx_cmd_build_rate(priv, tx_cmd, info, fc, sta_id, 0);
 
-	iwl_update_tx_stats(priv, fc, len);
+	iwl_update_tx_stats(priv, le16_to_cpu(fc), len);
 
 	scratch_phys = txcmd_phys + sizeof(struct iwl_cmd_header) +
 		offsetof(struct iwl_tx_cmd, scratch);
diff --git a/drivers/net/wireless/iwlwifi/iwl3945-base.c b/drivers/net/wireless/iwlwifi/iwl3945-base.c
index d0084bcb1eca..0a5bbe9ee938 100644
--- a/drivers/net/wireless/iwlwifi/iwl3945-base.c
+++ b/drivers/net/wireless/iwlwifi/iwl3945-base.c
@@ -2431,15 +2431,15 @@ static void iwl3945_build_tx_cmd_basic(struct iwl3945_priv *priv,
 				  struct ieee80211_hdr *hdr,
 				  int is_unicast, u8 std_id)
 {
-	u16 fc = le16_to_cpu(hdr->frame_control);
+	__le16 fc = hdr->frame_control;
 	__le32 tx_flags = cmd->cmd.tx.tx_flags;
 
 	cmd->cmd.tx.stop_time.life_time = TX_CMD_LIFE_TIME_INFINITE;
 	if (!(info->flags & IEEE80211_TX_CTL_NO_ACK)) {
 		tx_flags |= TX_CMD_FLG_ACK_MSK;
-		if ((fc & IEEE80211_FCTL_FTYPE) == IEEE80211_FTYPE_MGMT)
+		if (ieee80211_is_mgmt(fc))
 			tx_flags |= TX_CMD_FLG_SEQ_CTL_MSK;
-		if (ieee80211_is_probe_response(fc) &&
+		if (ieee80211_is_probe_resp(fc) &&
 		    !(le16_to_cpu(hdr->seq_ctrl) & 0xf))
 			tx_flags |= TX_CMD_FLG_TSF_MSK;
 	} else {
@@ -2451,8 +2451,8 @@ static void iwl3945_build_tx_cmd_basic(struct iwl3945_priv *priv,
 	if (ieee80211_get_morefrag(hdr))
 		tx_flags |= TX_CMD_FLG_MORE_FRAG_MSK;
 
-	if (ieee80211_is_qos_data(fc)) {
-		u8 *qc = ieee80211_get_qos_ctrl(hdr, ieee80211_get_hdrlen(fc));
+	if (ieee80211_is_data_qos(fc)) {
+		u8 *qc = ieee80211_get_qos_ctl(hdr);
 		cmd->cmd.tx.tid_tspec = qc[0] & 0xf;
 		tx_flags &= ~TX_CMD_FLG_SEQ_CTL_MSK;
 	} else {
@@ -2471,9 +2471,8 @@ static void iwl3945_build_tx_cmd_basic(struct iwl3945_priv *priv,
 		tx_flags |= TX_CMD_FLG_FULL_TXOP_PROT_MSK;
 
 	tx_flags &= ~(TX_CMD_FLG_ANT_SEL_MSK);
-	if ((fc & IEEE80211_FCTL_FTYPE) == IEEE80211_FTYPE_MGMT) {
-		if ((fc & IEEE80211_FCTL_STYPE) == IEEE80211_STYPE_ASSOC_REQ ||
-		    (fc & IEEE80211_FCTL_STYPE) == IEEE80211_STYPE_REASSOC_REQ)
+	if (ieee80211_is_mgmt(fc)) {
+		if (ieee80211_is_assoc_req(fc) || ieee80211_is_reassoc_req(fc))
 			cmd->cmd.tx.timeout.pm_frame_timeout = cpu_to_le16(3);
 		else
 			cmd->cmd.tx.timeout.pm_frame_timeout = cpu_to_le16(2);
@@ -2564,7 +2563,7 @@ static int iwl3945_tx_skb(struct iwl3945_priv *priv, struct sk_buff *skb)
 	u8 sta_id;
 	u8 tid = 0;
 	u16 seq_number = 0;
-	u16 fc;
+	__le16 fc;
 	u8 wait_write_ptr = 0;
 	u8 *qc = NULL;
 	unsigned long flags;
@@ -2589,28 +2588,28 @@ static int iwl3945_tx_skb(struct iwl3945_priv *priv, struct sk_buff *skb)
 	unicast = !is_multicast_ether_addr(hdr->addr1);
 	id = 0;
 
-	fc = le16_to_cpu(hdr->frame_control);
+	fc = hdr->frame_control;
 
 #ifdef CONFIG_IWL3945_DEBUG
 	if (ieee80211_is_auth(fc))
 		IWL_DEBUG_TX("Sending AUTH frame\n");
-	else if (ieee80211_is_assoc_request(fc))
+	else if (ieee80211_is_assoc_req(fc))
 		IWL_DEBUG_TX("Sending ASSOC frame\n");
-	else if (ieee80211_is_reassoc_request(fc))
+	else if (ieee80211_is_reassoc_req(fc))
 		IWL_DEBUG_TX("Sending REASSOC frame\n");
 #endif
 
 	/* drop all data frame if we are not associated */
 	if ((!iwl3945_is_associated(priv) ||
 	     ((priv->iw_mode == IEEE80211_IF_TYPE_STA) && !priv->assoc_id)) &&
-	    ((fc & IEEE80211_FCTL_FTYPE) == IEEE80211_FTYPE_DATA)) {
+	    ieee80211_is_data(fc)) {
 		IWL_DEBUG_DROP("Dropping - !iwl3945_is_associated\n");
 		goto drop_unlock;
 	}
 
 	spin_unlock_irqrestore(&priv->lock, flags);
 
-	hdr_len = ieee80211_get_hdrlen(fc);
+	hdr_len = ieee80211_get_hdrlen(le16_to_cpu(fc));
 
 	/* Find (or create) index into station table for destination station */
 	sta_id = iwl3945_get_sta_id(priv, hdr);
@@ -2624,8 +2623,8 @@ static int iwl3945_tx_skb(struct iwl3945_priv *priv, struct sk_buff *skb)
 
 	IWL_DEBUG_RATE("station Id %d\n", sta_id);
 
-	if (ieee80211_is_qos_data(fc)) {
-		qc = ieee80211_get_qos_ctrl(hdr, hdr_len);
+	if (ieee80211_is_data_qos(fc)) {
+		qc = ieee80211_get_qos_ctl(hdr);
 		tid = qc[0] & 0xf;
 		seq_number = priv->stations[sta_id].tid[tid].seq_number &
 				IEEE80211_SCTL_SEQ;
@@ -2746,7 +2745,7 @@ static int iwl3945_tx_skb(struct iwl3945_priv *priv, struct sk_buff *skb)
 			   sizeof(out_cmd->cmd.tx));
 
 	iwl3945_print_hex_dump(IWL_DL_TX, (u8 *)out_cmd->cmd.tx.hdr,
-			   ieee80211_get_hdrlen(fc));
+			   ieee80211_get_hdrlen(le16_to_cpu(fc)));
 
 	/* Tell device the write index *just past* this latest filled TFD */
 	q->write_ptr = iwl_queue_inc_wrap(q->write_ptr, q->n_bd);
diff --git a/include/linux/ieee80211.h b/include/linux/ieee80211.h
index 8f2c20b4a15d..371237b0d8b9 100644
--- a/include/linux/ieee80211.h
+++ b/include/linux/ieee80211.h
@@ -98,6 +98,7 @@
 
 #define IEEE80211_MAX_SSID_LEN		32
 #define IEEE80211_MAX_MESH_ID_LEN	32
+#define IEEE80211_QOS_CTL_LEN		2
 
 struct ieee80211_hdr {
 	__le16 frame_control;
@@ -109,6 +110,355 @@ struct ieee80211_hdr {
 	u8 addr4[6];
 } __attribute__ ((packed));
 
+/**
+ * ieee80211_has_tods - check if IEEE80211_FCTL_TODS is set
+ * @fc: frame control bytes in little-endian byteorder
+ */
+static inline int ieee80211_has_tods(__le16 fc)
+{
+	return (fc & cpu_to_le16(IEEE80211_FCTL_TODS)) != 0;
+}
+
+/**
+ * ieee80211_has_fromds - check if IEEE80211_FCTL_FROMDS is set
+ * @fc: frame control bytes in little-endian byteorder
+ */
+static inline int ieee80211_has_fromds(__le16 fc)
+{
+	return (fc & cpu_to_le16(IEEE80211_FCTL_FROMDS)) != 0;
+}
+
+/**
+ * ieee80211_has_a4 - check if IEEE80211_FCTL_TODS and IEEE80211_FCTL_FROMDS are set
+ * @fc: frame control bytes in little-endian byteorder
+ */
+static inline int ieee80211_has_a4(__le16 fc)
+{
+	__le16 tmp = cpu_to_le16(IEEE80211_FCTL_TODS | IEEE80211_FCTL_FROMDS);
+	return (fc & tmp) == tmp;
+}
+
+/**
+ * ieee80211_has_morefrags - check if IEEE80211_FCTL_MOREFRAGS is set
+ * @fc: frame control bytes in little-endian byteorder
+ */
+static inline int ieee80211_has_morefrags(__le16 fc)
+{
+	return (fc & cpu_to_le16(IEEE80211_FCTL_MOREFRAGS)) != 0;
+}
+
+/**
+ * ieee80211_has_retry - check if IEEE80211_FCTL_RETRY is set
+ * @fc: frame control bytes in little-endian byteorder
+ */
+static inline int ieee80211_has_retry(__le16 fc)
+{
+	return (fc & cpu_to_le16(IEEE80211_FCTL_RETRY)) != 0;
+}
+
+/**
+ * ieee80211_has_pm - check if IEEE80211_FCTL_PM is set
+ * @fc: frame control bytes in little-endian byteorder
+ */
+static inline int ieee80211_has_pm(__le16 fc)
+{
+	return (fc & cpu_to_le16(IEEE80211_FCTL_PM)) != 0;
+}
+
+/**
+ * ieee80211_has_moredata - check if IEEE80211_FCTL_MOREDATA is set
+ * @fc: frame control bytes in little-endian byteorder
+ */
+static inline int ieee80211_has_moredata(__le16 fc)
+{
+	return (fc & cpu_to_le16(IEEE80211_FCTL_MOREDATA)) != 0;
+}
+
+/**
+ * ieee80211_has_protected - check if IEEE80211_FCTL_PROTECTED is set
+ * @fc: frame control bytes in little-endian byteorder
+ */
+static inline int ieee80211_has_protected(__le16 fc)
+{
+	return (fc & cpu_to_le16(IEEE80211_FCTL_PROTECTED)) != 0;
+}
+
+/**
+ * ieee80211_has_order - check if IEEE80211_FCTL_ORDER is set
+ * @fc: frame control bytes in little-endian byteorder
+ */
+static inline int ieee80211_has_order(__le16 fc)
+{
+	return (fc & cpu_to_le16(IEEE80211_FCTL_ORDER)) != 0;
+}
+
+/**
+ * ieee80211_is_mgmt - check if type is IEEE80211_FTYPE_MGMT
+ * @fc: frame control bytes in little-endian byteorder
+ */
+static inline int ieee80211_is_mgmt(__le16 fc)
+{
+	return (fc & cpu_to_le16(IEEE80211_FCTL_FTYPE)) ==
+	       cpu_to_le16(IEEE80211_FTYPE_MGMT);
+}
+
+/**
+ * ieee80211_is_ctl - check if type is IEEE80211_FTYPE_CTL
+ * @fc: frame control bytes in little-endian byteorder
+ */
+static inline int ieee80211_is_ctl(__le16 fc)
+{
+	return (fc & cpu_to_le16(IEEE80211_FCTL_FTYPE)) ==
+	       cpu_to_le16(IEEE80211_FTYPE_CTL);
+}
+
+/**
+ * ieee80211_is_data - check if type is IEEE80211_FTYPE_DATA
+ * @fc: frame control bytes in little-endian byteorder
+ */
+static inline int ieee80211_is_data(__le16 fc)
+{
+	return (fc & cpu_to_le16(IEEE80211_FCTL_FTYPE)) ==
+	       cpu_to_le16(IEEE80211_FTYPE_DATA);
+}
+
+/**
+ * ieee80211_is_data_qos - check if type is IEEE80211_FTYPE_DATA and IEEE80211_STYPE_QOS_DATA is set
+ * @fc: frame control bytes in little-endian byteorder
+ */
+static inline int ieee80211_is_data_qos(__le16 fc)
+{
+	/*
+	 * mask with QOS_DATA rather than IEEE80211_FCTL_STYPE as we just need
+	 * to check the one bit
+	 */
+	return (fc & cpu_to_le16(IEEE80211_FCTL_FTYPE | IEEE80211_STYPE_QOS_DATA)) ==
+	       cpu_to_le16(IEEE80211_FTYPE_DATA | IEEE80211_STYPE_QOS_DATA);
+}
+
+/**
+ * ieee80211_is_data_present - check if type is IEEE80211_FTYPE_DATA and has data
+ * @fc: frame control bytes in little-endian byteorder
+ */
+static inline int ieee80211_is_data_present(__le16 fc)
+{
+	/*
+	 * mask with 0x40 and test that that bit is clear to only return true
+	 * for the data-containing substypes.
+	 */
+	return (fc & cpu_to_le16(IEEE80211_FCTL_FTYPE | 0x40)) ==
+	       cpu_to_le16(IEEE80211_FTYPE_DATA);
+}
+
+/**
+ * ieee80211_is_assoc_req - check if IEEE80211_FTYPE_MGMT && IEEE80211_STYPE_ASSOC_REQ
+ * @fc: frame control bytes in little-endian byteorder
+ */
+static inline int ieee80211_is_assoc_req(__le16 fc)
+{
+	return (fc & cpu_to_le16(IEEE80211_FCTL_FTYPE | IEEE80211_FCTL_STYPE)) ==
+	       cpu_to_le16(IEEE80211_FTYPE_MGMT | IEEE80211_STYPE_ASSOC_REQ);
+}
+
+/**
+ * ieee80211_is_assoc_resp - check if IEEE80211_FTYPE_MGMT && IEEE80211_STYPE_ASSOC_RESP
+ * @fc: frame control bytes in little-endian byteorder
+ */
+static inline int ieee80211_is_assoc_resp(__le16 fc)
+{
+	return (fc & cpu_to_le16(IEEE80211_FCTL_FTYPE | IEEE80211_FCTL_STYPE)) ==
+	       cpu_to_le16(IEEE80211_FTYPE_MGMT | IEEE80211_STYPE_ASSOC_RESP);
+}
+
+/**
+ * ieee80211_is_reassoc_req - check if IEEE80211_FTYPE_MGMT && IEEE80211_STYPE_REASSOC_REQ
+ * @fc: frame control bytes in little-endian byteorder
+ */
+static inline int ieee80211_is_reassoc_req(__le16 fc)
+{
+	return (fc & cpu_to_le16(IEEE80211_FCTL_FTYPE | IEEE80211_FCTL_STYPE)) ==
+	       cpu_to_le16(IEEE80211_FTYPE_MGMT | IEEE80211_STYPE_REASSOC_REQ);
+}
+
+/**
+ * ieee80211_is_reassoc_resp - check if IEEE80211_FTYPE_MGMT && IEEE80211_STYPE_REASSOC_RESP
+ * @fc: frame control bytes in little-endian byteorder
+ */
+static inline int ieee80211_is_reassoc_resp(__le16 fc)
+{
+	return (fc & cpu_to_le16(IEEE80211_FCTL_FTYPE | IEEE80211_FCTL_STYPE)) ==
+	       cpu_to_le16(IEEE80211_FTYPE_MGMT | IEEE80211_STYPE_REASSOC_RESP);
+}
+
+/**
+ * ieee80211_is_probe_req - check if IEEE80211_FTYPE_MGMT && IEEE80211_STYPE_PROBE_REQ
+ * @fc: frame control bytes in little-endian byteorder
+ */
+static inline int ieee80211_is_probe_req(__le16 fc)
+{
+	return (fc & cpu_to_le16(IEEE80211_FCTL_FTYPE | IEEE80211_FCTL_STYPE)) ==
+	       cpu_to_le16(IEEE80211_FTYPE_MGMT | IEEE80211_STYPE_PROBE_REQ);
+}
+
+/**
+ * ieee80211_is_probe_resp - check if IEEE80211_FTYPE_MGMT && IEEE80211_STYPE_PROBE_RESP
+ * @fc: frame control bytes in little-endian byteorder
+ */
+static inline int ieee80211_is_probe_resp(__le16 fc)
+{
+	return (fc & cpu_to_le16(IEEE80211_FCTL_FTYPE | IEEE80211_FCTL_STYPE)) ==
+	       cpu_to_le16(IEEE80211_FTYPE_MGMT | IEEE80211_STYPE_PROBE_RESP);
+}
+
+/**
+ * ieee80211_is_beacon - check if IEEE80211_FTYPE_MGMT && IEEE80211_STYPE_BEACON
+ * @fc: frame control bytes in little-endian byteorder
+ */
+static inline int ieee80211_is_beacon(__le16 fc)
+{
+	return (fc & cpu_to_le16(IEEE80211_FCTL_FTYPE | IEEE80211_FCTL_STYPE)) ==
+	       cpu_to_le16(IEEE80211_FTYPE_MGMT | IEEE80211_STYPE_BEACON);
+}
+
+/**
+ * ieee80211_is_atim - check if IEEE80211_FTYPE_MGMT && IEEE80211_STYPE_ATIM
+ * @fc: frame control bytes in little-endian byteorder
+ */
+static inline int ieee80211_is_atim(__le16 fc)
+{
+	return (fc & cpu_to_le16(IEEE80211_FCTL_FTYPE | IEEE80211_FCTL_STYPE)) ==
+	       cpu_to_le16(IEEE80211_FTYPE_MGMT | IEEE80211_STYPE_ATIM);
+}
+
+/**
+ * ieee80211_is_disassoc - check if IEEE80211_FTYPE_MGMT && IEEE80211_STYPE_DISASSOC
+ * @fc: frame control bytes in little-endian byteorder
+ */
+static inline int ieee80211_is_disassoc(__le16 fc)
+{
+	return (fc & cpu_to_le16(IEEE80211_FCTL_FTYPE | IEEE80211_FCTL_STYPE)) ==
+	       cpu_to_le16(IEEE80211_FTYPE_MGMT | IEEE80211_STYPE_DISASSOC);
+}
+
+/**
+ * ieee80211_is_auth - check if IEEE80211_FTYPE_MGMT && IEEE80211_STYPE_AUTH
+ * @fc: frame control bytes in little-endian byteorder
+ */
+static inline int ieee80211_is_auth(__le16 fc)
+{
+	return (fc & cpu_to_le16(IEEE80211_FCTL_FTYPE | IEEE80211_FCTL_STYPE)) ==
+	       cpu_to_le16(IEEE80211_FTYPE_MGMT | IEEE80211_STYPE_AUTH);
+}
+
+/**
+ * ieee80211_is_deauth - check if IEEE80211_FTYPE_MGMT && IEEE80211_STYPE_DEAUTH
+ * @fc: frame control bytes in little-endian byteorder
+ */
+static inline int ieee80211_is_deauth(__le16 fc)
+{
+	return (fc & cpu_to_le16(IEEE80211_FCTL_FTYPE | IEEE80211_FCTL_STYPE)) ==
+	       cpu_to_le16(IEEE80211_FTYPE_MGMT | IEEE80211_STYPE_DEAUTH);
+}
+
+/**
+ * ieee80211_is_action - check if IEEE80211_FTYPE_MGMT && IEEE80211_STYPE_ACTION
+ * @fc: frame control bytes in little-endian byteorder
+ */
+static inline int ieee80211_is_action(__le16 fc)
+{
+	return (fc & cpu_to_le16(IEEE80211_FCTL_FTYPE | IEEE80211_FCTL_STYPE)) ==
+	       cpu_to_le16(IEEE80211_FTYPE_MGMT | IEEE80211_STYPE_ACTION);
+}
+
+/**
+ * ieee80211_is_back_req - check if IEEE80211_FTYPE_CTL && IEEE80211_STYPE_BACK_REQ
+ * @fc: frame control bytes in little-endian byteorder
+ */
+static inline int ieee80211_is_back_req(__le16 fc)
+{
+	return (fc & cpu_to_le16(IEEE80211_FCTL_FTYPE | IEEE80211_FCTL_STYPE)) ==
+	       cpu_to_le16(IEEE80211_FTYPE_CTL | IEEE80211_STYPE_BACK_REQ);
+}
+
+/**
+ * ieee80211_is_back - check if IEEE80211_FTYPE_CTL && IEEE80211_STYPE_BACK
+ * @fc: frame control bytes in little-endian byteorder
+ */
+static inline int ieee80211_is_back(__le16 fc)
+{
+	return (fc & cpu_to_le16(IEEE80211_FCTL_FTYPE | IEEE80211_FCTL_STYPE)) ==
+	       cpu_to_le16(IEEE80211_FTYPE_CTL | IEEE80211_STYPE_BACK);
+}
+
+/**
+ * ieee80211_is_pspoll - check if IEEE80211_FTYPE_CTL && IEEE80211_STYPE_PSPOLL
+ * @fc: frame control bytes in little-endian byteorder
+ */
+static inline int ieee80211_is_pspoll(__le16 fc)
+{
+	return (fc & cpu_to_le16(IEEE80211_FCTL_FTYPE | IEEE80211_FCTL_STYPE)) ==
+	       cpu_to_le16(IEEE80211_FTYPE_CTL | IEEE80211_STYPE_PSPOLL);
+}
+
+/**
+ * ieee80211_is_rts - check if IEEE80211_FTYPE_CTL && IEEE80211_STYPE_RTS
+ * @fc: frame control bytes in little-endian byteorder
+ */
+static inline int ieee80211_is_rts(__le16 fc)
+{
+	return (fc & cpu_to_le16(IEEE80211_FCTL_FTYPE | IEEE80211_FCTL_STYPE)) ==
+	       cpu_to_le16(IEEE80211_FTYPE_CTL | IEEE80211_STYPE_RTS);
+}
+
+/**
+ * ieee80211_is_cts - check if IEEE80211_FTYPE_CTL && IEEE80211_STYPE_CTS
+ * @fc: frame control bytes in little-endian byteorder
+ */
+static inline int ieee80211_is_cts(__le16 fc)
+{
+	return (fc & cpu_to_le16(IEEE80211_FCTL_FTYPE | IEEE80211_FCTL_STYPE)) ==
+	       cpu_to_le16(IEEE80211_FTYPE_CTL | IEEE80211_STYPE_CTS);
+}
+
+/**
+ * ieee80211_is_ack - check if IEEE80211_FTYPE_CTL && IEEE80211_STYPE_ACK
+ * @fc: frame control bytes in little-endian byteorder
+ */
+static inline int ieee80211_is_ack(__le16 fc)
+{
+	return (fc & cpu_to_le16(IEEE80211_FCTL_FTYPE | IEEE80211_FCTL_STYPE)) ==
+	       cpu_to_le16(IEEE80211_FTYPE_CTL | IEEE80211_STYPE_ACK);
+}
+
+/**
+ * ieee80211_is_cfend - check if IEEE80211_FTYPE_CTL && IEEE80211_STYPE_CFEND
+ * @fc: frame control bytes in little-endian byteorder
+ */
+static inline int ieee80211_is_cfend(__le16 fc)
+{
+	return (fc & cpu_to_le16(IEEE80211_FCTL_FTYPE | IEEE80211_FCTL_STYPE)) ==
+	       cpu_to_le16(IEEE80211_FTYPE_CTL | IEEE80211_STYPE_CFEND);
+}
+
+/**
+ * ieee80211_is_cfendack - check if IEEE80211_FTYPE_CTL && IEEE80211_STYPE_CFENDACK
+ * @fc: frame control bytes in little-endian byteorder
+ */
+static inline int ieee80211_is_cfendack(__le16 fc)
+{
+	return (fc & cpu_to_le16(IEEE80211_FCTL_FTYPE | IEEE80211_FCTL_STYPE)) ==
+	       cpu_to_le16(IEEE80211_FTYPE_CTL | IEEE80211_STYPE_CFENDACK);
+}
+
+/**
+ * ieee80211_is_nullfunc - check if FTYPE=IEEE80211_FTYPE_DATA and STYPE=IEEE80211_STYPE_NULLFUNC
+ * @fc: frame control bytes in little-endian byteorder
+ */
+static inline int ieee80211_is_nullfunc(__le16 fc)
+{
+	return (fc & cpu_to_le16(IEEE80211_FCTL_FTYPE | IEEE80211_FCTL_STYPE)) ==
+	       cpu_to_le16(IEEE80211_FTYPE_DATA | IEEE80211_STYPE_NULLFUNC);
+}
 
 struct ieee80211s_hdr {
 	u8 flags;
@@ -552,49 +902,55 @@ enum ieee80211_back_parties {
 
 #define WLAN_MAX_KEY_LEN		32
 
+/**
+ * ieee80211_get_qos_ctl - get pointer to qos control bytes
+ * @hdr: the frame
+ *
+ * The qos ctrl bytes come after the frame_control, duration, seq_num
+ * and 3 or 4 addresses of length ETH_ALEN.
+ * 3 addr: 2 + 2 + 2 + 3*6 = 24
+ * 4 addr: 2 + 2 + 2 + 4*6 = 30
+ */
+static inline u8 *ieee80211_get_qos_ctl(struct ieee80211_hdr *hdr)
+{
+	if (ieee80211_has_a4(hdr->frame_control))
+		return (u8 *)hdr + 30;
+	else
+		return (u8 *)hdr + 24;
+}
+
 /**
  * ieee80211_get_SA - get pointer to SA
+ * @hdr: the frame
  *
  * Given an 802.11 frame, this function returns the offset
  * to the source address (SA). It does not verify that the
  * header is long enough to contain the address, and the
  * header must be long enough to contain the frame control
  * field.
- *
- * @hdr: the frame
  */
 static inline u8 *ieee80211_get_SA(struct ieee80211_hdr *hdr)
 {
-	__le16 fc = hdr->frame_control;
-	fc &= cpu_to_le16(IEEE80211_FCTL_TODS | IEEE80211_FCTL_FROMDS);
-
-	switch (fc) {
-	case __constant_cpu_to_le16(IEEE80211_FCTL_FROMDS):
-		return hdr->addr3;
-	case __constant_cpu_to_le16(IEEE80211_FCTL_TODS|IEEE80211_FCTL_FROMDS):
+	if (ieee80211_has_a4(hdr->frame_control))
 		return hdr->addr4;
-	default:
-		return hdr->addr2;
-	}
+	if (ieee80211_has_fromds(hdr->frame_control))
+		return hdr->addr3;
+	return hdr->addr2;
 }
 
 /**
  * ieee80211_get_DA - get pointer to DA
+ * @hdr: the frame
  *
  * Given an 802.11 frame, this function returns the offset
  * to the destination address (DA). It does not verify that
  * the header is long enough to contain the address, and the
  * header must be long enough to contain the frame control
  * field.
- *
- * @hdr: the frame
  */
 static inline u8 *ieee80211_get_DA(struct ieee80211_hdr *hdr)
 {
-	__le16 fc = hdr->frame_control;
-	fc &= cpu_to_le16(IEEE80211_FCTL_TODS);
-
-	if (fc)
+	if (ieee80211_has_tods(hdr->frame_control))
 		return hdr->addr3;
 	else
 		return hdr->addr1;
-- 
cgit v1.2.3


From 8b7b1e05b0454f232b8ae1e6ee134b7f0b38abfb Mon Sep 17 00:00:00 2001
From: Harvey Harrison <harvey.harrison@gmail.com>
Date: Wed, 11 Jun 2008 14:21:56 -0700
Subject: mac80211: remove ieee80211_get_morefrag

Replaced by the new helper ieee80211_has_morefrags which is
more consistent with the intent of the function.

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/net/wireless/iwlwifi/iwl-tx.c       |  4 ++--
 drivers/net/wireless/iwlwifi/iwl3945-base.c |  4 ++--
 drivers/net/wireless/rt2x00/rt2x00queue.c   |  2 +-
 drivers/net/wireless/rtl8187_dev.c          |  2 +-
 include/linux/ieee80211.h                   | 14 --------------
 5 files changed, 6 insertions(+), 20 deletions(-)

(limited to 'include')

diff --git a/drivers/net/wireless/iwlwifi/iwl-tx.c b/drivers/net/wireless/iwlwifi/iwl-tx.c
index 98c434c52a64..7296e2846ec3 100644
--- a/drivers/net/wireless/iwlwifi/iwl-tx.c
+++ b/drivers/net/wireless/iwlwifi/iwl-tx.c
@@ -590,7 +590,7 @@ static void iwl_tx_cmd_build_basic(struct iwl_priv *priv,
 
 
 	tx_cmd->sta_id = std_id;
-	if (ieee80211_get_morefrag(hdr))
+	if (ieee80211_has_morefrags(fc))
 		tx_flags |= TX_CMD_FLG_MORE_FRAG_MSK;
 
 	if (ieee80211_is_data_qos(fc)) {
@@ -944,7 +944,7 @@ int iwl_tx_skb(struct iwl_priv *priv, struct sk_buff *skb)
 	tx_cmd->dram_lsb_ptr = cpu_to_le32(scratch_phys);
 	tx_cmd->dram_msb_ptr = iwl_get_dma_hi_address(scratch_phys);
 
-	if (!ieee80211_get_morefrag(hdr)) {
+	if (!ieee80211_has_morefrags(hdr->frame_control)) {
 		txq->need_update = 1;
 		if (qc)
 			priv->stations[sta_id].tid[tid].seq_number = seq_number;
diff --git a/drivers/net/wireless/iwlwifi/iwl3945-base.c b/drivers/net/wireless/iwlwifi/iwl3945-base.c
index 0a5bbe9ee938..47cf4b997f50 100644
--- a/drivers/net/wireless/iwlwifi/iwl3945-base.c
+++ b/drivers/net/wireless/iwlwifi/iwl3945-base.c
@@ -2448,7 +2448,7 @@ static void iwl3945_build_tx_cmd_basic(struct iwl3945_priv *priv,
 	}
 
 	cmd->cmd.tx.sta_id = std_id;
-	if (ieee80211_get_morefrag(hdr))
+	if (ieee80211_has_morefrags(fc))
 		tx_flags |= TX_CMD_FLG_MORE_FRAG_MSK;
 
 	if (ieee80211_is_data_qos(fc)) {
@@ -2731,7 +2731,7 @@ static int iwl3945_tx_skb(struct iwl3945_priv *priv, struct sk_buff *skb)
 	out_cmd->cmd.tx.tx_flags &= ~TX_CMD_FLG_ANT_A_MSK;
 	out_cmd->cmd.tx.tx_flags &= ~TX_CMD_FLG_ANT_B_MSK;
 
-	if (!ieee80211_get_morefrag(hdr)) {
+	if (!ieee80211_has_morefrags(hdr->frame_control)) {
 		txq->need_update = 1;
 		if (qc) {
 			priv->stations[sta_id].tid[tid].seq_number = seq_number;
diff --git a/drivers/net/wireless/rt2x00/rt2x00queue.c b/drivers/net/wireless/rt2x00/rt2x00queue.c
index 2f3dd1d91a12..7b52039b01a6 100644
--- a/drivers/net/wireless/rt2x00/rt2x00queue.c
+++ b/drivers/net/wireless/rt2x00/rt2x00queue.c
@@ -130,7 +130,7 @@ void rt2x00queue_create_tx_descriptor(struct queue_entry *entry,
 	/*
 	 * Check if more fragments are pending
 	 */
-	if (ieee80211_get_morefrag(hdr)) {
+	if (ieee80211_has_morefrags(hdr->frame_control)) {
 		__set_bit(ENTRY_TXD_BURST, &txdesc->flags);
 		__set_bit(ENTRY_TXD_MORE_FRAG, &txdesc->flags);
 	}
diff --git a/drivers/net/wireless/rtl8187_dev.c b/drivers/net/wireless/rtl8187_dev.c
index 0078c7e9918c..bec96d762c6c 100644
--- a/drivers/net/wireless/rtl8187_dev.c
+++ b/drivers/net/wireless/rtl8187_dev.c
@@ -181,7 +181,7 @@ static int rtl8187_tx(struct ieee80211_hw *dev, struct sk_buff *skb)
 	flags |= RTL8187_TX_FLAG_NO_ENCRYPT;
 
 	flags |= ieee80211_get_tx_rate(dev, info)->hw_value << 24;
-	if (ieee80211_get_morefrag((struct ieee80211_hdr *)skb->data))
+	if (ieee80211_has_morefrags(((struct ieee80211_hdr *)skb->data)->frame_control))
 		flags |= RTL8187_TX_FLAG_MORE_FRAG;
 	if (info->flags & IEEE80211_TX_CTL_USE_RTS_CTS) {
 		flags |= RTL8187_TX_FLAG_RTS;
diff --git a/include/linux/ieee80211.h b/include/linux/ieee80211.h
index 371237b0d8b9..2998e3b5f166 100644
--- a/include/linux/ieee80211.h
+++ b/include/linux/ieee80211.h
@@ -956,18 +956,4 @@ static inline u8 *ieee80211_get_DA(struct ieee80211_hdr *hdr)
 		return hdr->addr1;
 }
 
-/**
- * ieee80211_get_morefrag - determine whether the MOREFRAGS bit is set
- *
- * This function determines whether the "more fragments" bit is set
- * in the frame.
- *
- * @hdr: the frame
- */
-static inline int ieee80211_get_morefrag(struct ieee80211_hdr *hdr)
-{
-	__le16 fc = hdr->frame_control;
-	return !!(fc & cpu_to_le16(IEEE80211_FCTL_MOREFRAGS));
-}
-
 #endif /* IEEE80211_H */
-- 
cgit v1.2.3


From c9c6950c14ffc0e30e592fec1ebcb203ad3dff10 Mon Sep 17 00:00:00 2001
From: Harvey Harrison <harvey.harrison@gmail.com>
Date: Wed, 11 Jun 2008 14:21:57 -0700
Subject: mac80211: make ieee80211_get_hdrlen_from_skb return unsigned

Many callers already expect it to.

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/net/mac80211.h | 2 +-
 net/mac80211/util.c    | 8 ++++----
 2 files changed, 5 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index 1196de85f8db..f9b391faaf76 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -1535,7 +1535,7 @@ ieee80211_get_buffered_bc(struct ieee80211_hw *hw, struct ieee80211_vif *vif);
  *
  * @skb: the frame
  */
-int ieee80211_get_hdrlen_from_skb(const struct sk_buff *skb);
+unsigned int ieee80211_get_hdrlen_from_skb(const struct sk_buff *skb);
 
 /**
  * ieee80211_get_hdrlen - get header length from frame control
diff --git a/net/mac80211/util.c b/net/mac80211/util.c
index 6513bc2d2707..f3c30d00b21e 100644
--- a/net/mac80211/util.c
+++ b/net/mac80211/util.c
@@ -133,14 +133,14 @@ int ieee80211_get_hdrlen(u16 fc)
 }
 EXPORT_SYMBOL(ieee80211_get_hdrlen);
 
-int ieee80211_get_hdrlen_from_skb(const struct sk_buff *skb)
+unsigned int ieee80211_get_hdrlen_from_skb(const struct sk_buff *skb)
 {
-	const struct ieee80211_hdr *hdr = (const struct ieee80211_hdr *) skb->data;
-	int hdrlen;
+	const struct ieee80211_hdr *hdr = (const struct ieee80211_hdr *)skb->data;
+	unsigned int hdrlen;
 
 	if (unlikely(skb->len < 10))
 		return 0;
-	hdrlen = ieee80211_get_hdrlen(le16_to_cpu(hdr->frame_control));
+	hdrlen = ieee80211_hdrlen(hdr->frame_control);
 	if (unlikely(hdrlen > skb->len))
 		return 0;
 	return hdrlen;
-- 
cgit v1.2.3


From 6693be7124cb8e4f15f0d80ed6e3e50678771737 Mon Sep 17 00:00:00 2001
From: Harvey Harrison <harvey.harrison@gmail.com>
Date: Wed, 11 Jun 2008 14:21:57 -0700
Subject: mac80211: add utility function to get header length

Take a __le16 directly rather than a host-endian value.

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/net/mac80211.h |  6 ++++++
 net/mac80211/util.c    | 32 ++++++++++++++++++++++++++++++++
 2 files changed, 38 insertions(+)

(limited to 'include')

diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index f9b391faaf76..7ab4ff6159a2 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -1547,6 +1547,12 @@ unsigned int ieee80211_get_hdrlen_from_skb(const struct sk_buff *skb);
  */
 int ieee80211_get_hdrlen(u16 fc);
 
+/**
+ * ieee80211_hdrlen - get header length in bytes from frame control
+ * @fc: frame control field in little-endian format
+ */
+unsigned int ieee80211_hdrlen(__le16 fc);
+
 /**
  * ieee80211_get_tkip_key - get a TKIP rc4 for skb
  *
diff --git a/net/mac80211/util.c b/net/mac80211/util.c
index f3c30d00b21e..9f365a3af96f 100644
--- a/net/mac80211/util.c
+++ b/net/mac80211/util.c
@@ -133,6 +133,38 @@ int ieee80211_get_hdrlen(u16 fc)
 }
 EXPORT_SYMBOL(ieee80211_get_hdrlen);
 
+unsigned int ieee80211_hdrlen(__le16 fc)
+{
+	unsigned int hdrlen = 24;
+
+	if (ieee80211_is_data(fc)) {
+		if (ieee80211_has_a4(fc))
+			hdrlen = 30;
+		if (ieee80211_is_data_qos(fc))
+			hdrlen += IEEE80211_QOS_CTL_LEN;
+		goto out;
+	}
+
+	if (ieee80211_is_ctl(fc)) {
+		/*
+		 * ACK and CTS are 10 bytes, all others 16. To see how
+		 * to get this condition consider
+		 *   subtype mask:   0b0000000011110000 (0x00F0)
+		 *   ACK subtype:    0b0000000011010000 (0x00D0)
+		 *   CTS subtype:    0b0000000011000000 (0x00C0)
+		 *   bits that matter:         ^^^      (0x00E0)
+		 *   value of those: 0b0000000011000000 (0x00C0)
+		 */
+		if ((fc & cpu_to_le16(0x00E0)) == cpu_to_le16(0x00C0))
+			hdrlen = 10;
+		else
+			hdrlen = 16;
+	}
+out:
+	return hdrlen;
+}
+EXPORT_SYMBOL(ieee80211_hdrlen);
+
 unsigned int ieee80211_get_hdrlen_from_skb(const struct sk_buff *skb)
 {
 	const struct ieee80211_hdr *hdr = (const struct ieee80211_hdr *)skb->data;
-- 
cgit v1.2.3


From 7d06b2e053d2d536348e3a0f6bb02982a41bea37 Mon Sep 17 00:00:00 2001
From: Brian Haley <brian.haley@hp.com>
Date: Sat, 14 Jun 2008 17:04:49 -0700
Subject: net: change proto destroy method to return void

Change struct proto destroy function pointer to return void.  Noticed
by Al Viro.

Signed-off-by: Brian Haley <brian.haley@hp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/sock.h      | 2 +-
 include/net/tcp.h       | 2 +-
 include/net/transp_v6.h | 2 +-
 net/dccp/dccp.h         | 2 +-
 net/dccp/ipv6.c         | 4 ++--
 net/dccp/proto.c        | 4 +---
 net/ipv4/raw.c          | 3 +--
 net/ipv4/tcp_ipv4.c     | 4 +---
 net/ipv4/udp.c          | 3 +--
 net/ipv4/udp_impl.h     | 2 +-
 net/ipv6/af_inet6.c     | 4 +---
 net/ipv6/raw.c          | 4 ++--
 net/ipv6/tcp_ipv6.c     | 4 ++--
 net/ipv6/udp.c          | 4 +---
 net/ipv6/udp_impl.h     | 2 +-
 net/sctp/socket.c       | 3 +--
 16 files changed, 19 insertions(+), 30 deletions(-)

(limited to 'include')

diff --git a/include/net/sock.h b/include/net/sock.h
index dc42b44c2aa1..0a80961a83c0 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -524,7 +524,7 @@ struct proto {
 	int			(*ioctl)(struct sock *sk, int cmd,
 					 unsigned long arg);
 	int			(*init)(struct sock *sk);
-	int			(*destroy)(struct sock *sk);
+	void			(*destroy)(struct sock *sk);
 	void			(*shutdown)(struct sock *sk, int how);
 	int			(*setsockopt)(struct sock *sk, int level, 
 					int optname, char __user *optval,
diff --git a/include/net/tcp.h b/include/net/tcp.h
index b5a1b9eb12e8..6b827cc6c52f 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -1366,7 +1366,7 @@ extern void tcp_proc_unregister(struct net *net, struct tcp_seq_afinfo *afinfo);
 extern struct request_sock_ops tcp_request_sock_ops;
 extern struct request_sock_ops tcp6_request_sock_ops;
 
-extern int tcp_v4_destroy_sock(struct sock *sk);
+extern void tcp_v4_destroy_sock(struct sock *sk);
 
 extern int tcp_v4_gso_send_check(struct sk_buff *skb);
 extern struct sk_buff *tcp_tso_segment(struct sk_buff *skb, int features);
diff --git a/include/net/transp_v6.h b/include/net/transp_v6.h
index 112934a3288d..876b6f2bb4fd 100644
--- a/include/net/transp_v6.h
+++ b/include/net/transp_v6.h
@@ -53,7 +53,7 @@ extern int			datagram_send_ctl(struct net *net,
  */
 extern struct inet_connection_sock_af_ops ipv4_specific;
 
-extern int inet6_destroy_sock(struct sock *sk);
+extern void inet6_destroy_sock(struct sock *sk);
 
 #endif
 
diff --git a/net/dccp/dccp.h b/net/dccp/dccp.h
index f44d492d3b74..1b2cea244e12 100644
--- a/net/dccp/dccp.h
+++ b/net/dccp/dccp.h
@@ -262,7 +262,7 @@ extern int dccp_rcv_established(struct sock *sk, struct sk_buff *skb,
 				const struct dccp_hdr *dh, const unsigned len);
 
 extern int dccp_init_sock(struct sock *sk, const __u8 ctl_sock_initialized);
-extern int dccp_destroy_sock(struct sock *sk);
+extern void dccp_destroy_sock(struct sock *sk);
 
 extern void		dccp_close(struct sock *sk, long timeout);
 extern struct sk_buff	*dccp_make_response(struct sock *sk,
diff --git a/net/dccp/ipv6.c b/net/dccp/ipv6.c
index f7fe2a572d7b..eec3c4717890 100644
--- a/net/dccp/ipv6.c
+++ b/net/dccp/ipv6.c
@@ -1091,10 +1091,10 @@ static int dccp_v6_init_sock(struct sock *sk)
 	return err;
 }
 
-static int dccp_v6_destroy_sock(struct sock *sk)
+static void dccp_v6_destroy_sock(struct sock *sk)
 {
 	dccp_destroy_sock(sk);
-	return inet6_destroy_sock(sk);
+	inet6_destroy_sock(sk);
 }
 
 static struct timewait_sock_ops dccp6_timewait_sock_ops = {
diff --git a/net/dccp/proto.c b/net/dccp/proto.c
index 9dfe2470962c..a0b56009611f 100644
--- a/net/dccp/proto.c
+++ b/net/dccp/proto.c
@@ -237,7 +237,7 @@ int dccp_init_sock(struct sock *sk, const __u8 ctl_sock_initialized)
 
 EXPORT_SYMBOL_GPL(dccp_init_sock);
 
-int dccp_destroy_sock(struct sock *sk)
+void dccp_destroy_sock(struct sock *sk)
 {
 	struct dccp_sock *dp = dccp_sk(sk);
 	struct dccp_minisock *dmsk = dccp_msk(sk);
@@ -268,8 +268,6 @@ int dccp_destroy_sock(struct sock *sk)
 
 	/* clean up feature negotiation state */
 	dccp_feat_clean(dmsk);
-
-	return 0;
 }
 
 EXPORT_SYMBOL_GPL(dccp_destroy_sock);
diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c
index 1d0c97c8712d..36035a0c6dc2 100644
--- a/net/ipv4/raw.c
+++ b/net/ipv4/raw.c
@@ -606,12 +606,11 @@ static void raw_close(struct sock *sk, long timeout)
 	sk_common_release(sk);
 }
 
-static int raw_destroy(struct sock *sk)
+static void raw_destroy(struct sock *sk)
 {
 	lock_sock(sk);
 	ip_flush_pending_frames(sk);
 	release_sock(sk);
-	return 0;
 }
 
 /* This gets rid of all the nasties in af_inet. -DaveM */
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index b219a7a7cd08..64b385f65930 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -1775,7 +1775,7 @@ static int tcp_v4_init_sock(struct sock *sk)
 	return 0;
 }
 
-int tcp_v4_destroy_sock(struct sock *sk)
+void tcp_v4_destroy_sock(struct sock *sk)
 {
 	struct tcp_sock *tp = tcp_sk(sk);
 
@@ -1819,8 +1819,6 @@ int tcp_v4_destroy_sock(struct sock *sk)
 	}
 
 	atomic_dec(&tcp_sockets_allocated);
-
-	return 0;
 }
 
 EXPORT_SYMBOL(tcp_v4_destroy_sock);
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index 355e6d62d483..eba790dcd16b 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -1253,12 +1253,11 @@ int udp_rcv(struct sk_buff *skb)
 	return __udp4_lib_rcv(skb, udp_hash, IPPROTO_UDP);
 }
 
-int udp_destroy_sock(struct sock *sk)
+void udp_destroy_sock(struct sock *sk)
 {
 	lock_sock(sk);
 	udp_flush_pending_frames(sk);
 	release_sock(sk);
-	return 0;
 }
 
 /*
diff --git a/net/ipv4/udp_impl.h b/net/ipv4/udp_impl.h
index 7288bf7977fb..2e9bad2fa1bc 100644
--- a/net/ipv4/udp_impl.h
+++ b/net/ipv4/udp_impl.h
@@ -26,7 +26,7 @@ extern int	udp_recvmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg,
 extern int	udp_sendpage(struct sock *sk, struct page *page, int offset,
 			     size_t size, int flags);
 extern int	udp_queue_rcv_skb(struct sock * sk, struct sk_buff *skb);
-extern int	udp_destroy_sock(struct sock *sk);
+extern void	udp_destroy_sock(struct sock *sk);
 
 #ifdef CONFIG_PROC_FS
 extern int	udp4_seq_show(struct seq_file *seq, void *v);
diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c
index 350457c761e6..3ce8d2f318c6 100644
--- a/net/ipv6/af_inet6.c
+++ b/net/ipv6/af_inet6.c
@@ -371,7 +371,7 @@ int inet6_release(struct socket *sock)
 
 EXPORT_SYMBOL(inet6_release);
 
-int inet6_destroy_sock(struct sock *sk)
+void inet6_destroy_sock(struct sock *sk)
 {
 	struct ipv6_pinfo *np = inet6_sk(sk);
 	struct sk_buff *skb;
@@ -389,8 +389,6 @@ int inet6_destroy_sock(struct sock *sk)
 
 	if ((opt = xchg(&np->opt, NULL)) != NULL)
 		sock_kfree_s(sk, opt, opt->tot_len);
-
-	return 0;
 }
 
 EXPORT_SYMBOL_GPL(inet6_destroy_sock);
diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c
index 70a57e45bf0e..456777d7a407 100644
--- a/net/ipv6/raw.c
+++ b/net/ipv6/raw.c
@@ -1162,13 +1162,13 @@ static void rawv6_close(struct sock *sk, long timeout)
 	sk_common_release(sk);
 }
 
-static int raw6_destroy(struct sock *sk)
+static void raw6_destroy(struct sock *sk)
 {
 	lock_sock(sk);
 	ip6_flush_pending_frames(sk);
 	release_sock(sk);
 
-	return inet6_destroy_sock(sk);
+	inet6_destroy_sock(sk);
 }
 
 static int rawv6_init_sk(struct sock *sk)
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index ebed5d3adb82..daefc18d50a0 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -1872,7 +1872,7 @@ static int tcp_v6_init_sock(struct sock *sk)
 	return 0;
 }
 
-static int tcp_v6_destroy_sock(struct sock *sk)
+static void tcp_v6_destroy_sock(struct sock *sk)
 {
 #ifdef CONFIG_TCP_MD5SIG
 	/* Clean up the MD5 key list */
@@ -1880,7 +1880,7 @@ static int tcp_v6_destroy_sock(struct sock *sk)
 		tcp_v6_clear_md5_list(sk);
 #endif
 	tcp_v4_destroy_sock(sk);
-	return inet6_destroy_sock(sk);
+	inet6_destroy_sock(sk);
 }
 
 #ifdef CONFIG_PROC_FS
diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c
index e0693fffc9bd..09687f7a8564 100644
--- a/net/ipv6/udp.c
+++ b/net/ipv6/udp.c
@@ -879,15 +879,13 @@ do_confirm:
 	goto out;
 }
 
-int udpv6_destroy_sock(struct sock *sk)
+void udpv6_destroy_sock(struct sock *sk)
 {
 	lock_sock(sk);
 	udp_v6_flush_pending_frames(sk);
 	release_sock(sk);
 
 	inet6_destroy_sock(sk);
-
-	return 0;
 }
 
 /*
diff --git a/net/ipv6/udp_impl.h b/net/ipv6/udp_impl.h
index 321b81a4d418..92dd7da766d8 100644
--- a/net/ipv6/udp_impl.h
+++ b/net/ipv6/udp_impl.h
@@ -29,7 +29,7 @@ extern int	udpv6_recvmsg(struct kiocb *iocb, struct sock *sk,
 			      struct msghdr *msg, size_t len,
 			      int noblock, int flags, int *addr_len);
 extern int	udpv6_queue_rcv_skb(struct sock * sk, struct sk_buff *skb);
-extern int	udpv6_destroy_sock(struct sock *sk);
+extern void	udpv6_destroy_sock(struct sock *sk);
 
 #ifdef CONFIG_PROC_FS
 extern int	udp6_seq_show(struct seq_file *seq, void *v);
diff --git a/net/sctp/socket.c b/net/sctp/socket.c
index 253e5ea7e1e8..f98650cc48d8 100644
--- a/net/sctp/socket.c
+++ b/net/sctp/socket.c
@@ -3588,7 +3588,7 @@ SCTP_STATIC int sctp_init_sock(struct sock *sk)
 }
 
 /* Cleanup any SCTP per socket resources.  */
-SCTP_STATIC int sctp_destroy_sock(struct sock *sk)
+SCTP_STATIC void sctp_destroy_sock(struct sock *sk)
 {
 	struct sctp_endpoint *ep;
 
@@ -3598,7 +3598,6 @@ SCTP_STATIC int sctp_destroy_sock(struct sock *sk)
 	ep = sctp_sk(sk)->ep;
 	sctp_endpoint_free(ep);
 	atomic_dec(&sctp_sockets_allocated);
-	return 0;
 }
 
 /* API 4.1.7 shutdown() - TCP Style Syntax
-- 
cgit v1.2.3


From d6266281f8175e3ad68c28b20a609b278b47ade5 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Mon, 16 Jun 2008 17:11:50 -0700
Subject: udp: introduce a udp_hashfn function

Currently the chain to store a UDP socket is calculated with
simple (x & (UDP_HTABLE_SIZE - 1)). But taking net into account
would make this calculation a bit more complex, so moving it into
a function would help.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/udp.h |  5 +++++
 net/ipv4/udp.c      | 12 ++++++------
 net/ipv6/udp.c      |  4 ++--
 3 files changed, 13 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/include/linux/udp.h b/include/linux/udp.h
index 581ca2c14c52..9c94312b2de5 100644
--- a/include/linux/udp.h
+++ b/include/linux/udp.h
@@ -46,6 +46,11 @@ static inline struct udphdr *udp_hdr(const struct sk_buff *skb)
 
 #define UDP_HTABLE_SIZE		128
 
+static inline int udp_hashfn(const unsigned num)
+{
+	return num & (UDP_HTABLE_SIZE - 1);
+}
+
 struct udp_sock {
 	/* inet_sock has to be the first member */
 	struct inet_sock inet;
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index eba790dcd16b..d8f527d15701 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -134,7 +134,7 @@ static inline int __udp_lib_lport_inuse(struct net *net, __u16 num,
 	struct sock *sk;
 	struct hlist_node *node;
 
-	sk_for_each(sk, node, &udptable[num & (UDP_HTABLE_SIZE - 1)])
+	sk_for_each(sk, node, &udptable[udp_hashfn(num)])
 		if (net_eq(sock_net(sk), net) && sk->sk_hash == num)
 			return 1;
 	return 0;
@@ -174,7 +174,7 @@ int udp_lib_get_port(struct sock *sk, unsigned short snum,
 		for (i = 0; i < UDP_HTABLE_SIZE; i++) {
 			int size = 0;
 
-			head = &udptable[rover & (UDP_HTABLE_SIZE - 1)];
+			head = &udptable[udp_hashfn(rover)];
 			if (hlist_empty(head))
 				goto gotit;
 
@@ -211,7 +211,7 @@ int udp_lib_get_port(struct sock *sk, unsigned short snum,
 gotit:
 		snum = rover;
 	} else {
-		head = &udptable[snum & (UDP_HTABLE_SIZE - 1)];
+		head = &udptable[udp_hashfn(snum)];
 
 		sk_for_each(sk2, node, head)
 			if (sk2->sk_hash == snum                             &&
@@ -227,7 +227,7 @@ gotit:
 	inet_sk(sk)->num = snum;
 	sk->sk_hash = snum;
 	if (sk_unhashed(sk)) {
-		head = &udptable[snum & (UDP_HTABLE_SIZE - 1)];
+		head = &udptable[udp_hashfn(snum)];
 		sk_add_node(sk, head);
 		sock_prot_inuse_add(sock_net(sk), sk->sk_prot, 1);
 	}
@@ -264,7 +264,7 @@ static struct sock *__udp4_lib_lookup(struct net *net, __be32 saddr,
 	int badness = -1;
 
 	read_lock(&udp_hash_lock);
-	sk_for_each(sk, node, &udptable[hnum & (UDP_HTABLE_SIZE - 1)]) {
+	sk_for_each(sk, node, &udptable[udp_hashfn(hnum)]) {
 		struct inet_sock *inet = inet_sk(sk);
 
 		if (net_eq(sock_net(sk), net) && sk->sk_hash == hnum &&
@@ -1068,7 +1068,7 @@ static int __udp4_lib_mcast_deliver(struct sk_buff *skb,
 	int dif;
 
 	read_lock(&udp_hash_lock);
-	sk = sk_head(&udptable[ntohs(uh->dest) & (UDP_HTABLE_SIZE - 1)]);
+	sk = sk_head(&udptable[udp_hashfn(ntohs(uh->dest))]);
 	dif = skb->dev->ifindex;
 	sk = udp_v4_mcast_next(sk, uh->dest, daddr, uh->source, saddr, dif);
 	if (sk) {
diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c
index 09687f7a8564..6e4a822ba651 100644
--- a/net/ipv6/udp.c
+++ b/net/ipv6/udp.c
@@ -65,7 +65,7 @@ static struct sock *__udp6_lib_lookup(struct net *net,
 	int badness = -1;
 
 	read_lock(&udp_hash_lock);
-	sk_for_each(sk, node, &udptable[hnum & (UDP_HTABLE_SIZE - 1)]) {
+	sk_for_each(sk, node, &udptable[udp_hashfn(hnum)]) {
 		struct inet_sock *inet = inet_sk(sk);
 
 		if (net_eq(sock_net(sk), net) && sk->sk_hash == hnum &&
@@ -361,7 +361,7 @@ static int __udp6_lib_mcast_deliver(struct sk_buff *skb, struct in6_addr *saddr,
 	int dif;
 
 	read_lock(&udp_hash_lock);
-	sk = sk_head(&udptable[ntohs(uh->dest) & (UDP_HTABLE_SIZE - 1)]);
+	sk = sk_head(&udptable[udp_hashfn(ntohs(uh->dest))]);
 	dif = inet6_iif(skb);
 	sk = udp_v6_mcast_next(sk, uh->dest, daddr, uh->source, saddr, dif);
 	if (!sk) {
-- 
cgit v1.2.3


From 19c7578fb22b0aef103222cae9b522f03ae489d6 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Mon, 16 Jun 2008 17:12:29 -0700
Subject: udp: add struct net argument to udp_hashfn

Every caller already has this one. The new argument is currently
unused, but this will be fixed shortly.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/udp.h |  2 +-
 net/ipv4/udp.c      | 12 ++++++------
 net/ipv6/udp.c      |  4 ++--
 3 files changed, 9 insertions(+), 9 deletions(-)

(limited to 'include')

diff --git a/include/linux/udp.h b/include/linux/udp.h
index 9c94312b2de5..3deccac2e815 100644
--- a/include/linux/udp.h
+++ b/include/linux/udp.h
@@ -46,7 +46,7 @@ static inline struct udphdr *udp_hdr(const struct sk_buff *skb)
 
 #define UDP_HTABLE_SIZE		128
 
-static inline int udp_hashfn(const unsigned num)
+static inline int udp_hashfn(struct net *net, const unsigned num)
 {
 	return num & (UDP_HTABLE_SIZE - 1);
 }
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index 6b0acb438f34..11eabf136144 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -134,7 +134,7 @@ static inline int __udp_lib_lport_inuse(struct net *net, __u16 num,
 	struct sock *sk;
 	struct hlist_node *node;
 
-	sk_for_each(sk, node, &udptable[udp_hashfn(num)])
+	sk_for_each(sk, node, &udptable[udp_hashfn(net, num)])
 		if (net_eq(sock_net(sk), net) && sk->sk_hash == num)
 			return 1;
 	return 0;
@@ -174,7 +174,7 @@ int udp_lib_get_port(struct sock *sk, unsigned short snum,
 		for (i = 0; i < UDP_HTABLE_SIZE; i++) {
 			int size = 0;
 
-			head = &udptable[udp_hashfn(rover)];
+			head = &udptable[udp_hashfn(net, rover)];
 			if (hlist_empty(head))
 				goto gotit;
 
@@ -211,7 +211,7 @@ int udp_lib_get_port(struct sock *sk, unsigned short snum,
 gotit:
 		snum = rover;
 	} else {
-		head = &udptable[udp_hashfn(snum)];
+		head = &udptable[udp_hashfn(net, snum)];
 
 		sk_for_each(sk2, node, head)
 			if (sk2->sk_hash == snum                             &&
@@ -227,7 +227,7 @@ gotit:
 	inet_sk(sk)->num = snum;
 	sk->sk_hash = snum;
 	if (sk_unhashed(sk)) {
-		head = &udptable[udp_hashfn(snum)];
+		head = &udptable[udp_hashfn(net, snum)];
 		sk_add_node(sk, head);
 		sock_prot_inuse_add(sock_net(sk), sk->sk_prot, 1);
 	}
@@ -264,7 +264,7 @@ static struct sock *__udp4_lib_lookup(struct net *net, __be32 saddr,
 	int badness = -1;
 
 	read_lock(&udp_hash_lock);
-	sk_for_each(sk, node, &udptable[udp_hashfn(hnum)]) {
+	sk_for_each(sk, node, &udptable[udp_hashfn(net, hnum)]) {
 		struct inet_sock *inet = inet_sk(sk);
 
 		if (net_eq(sock_net(sk), net) && sk->sk_hash == hnum &&
@@ -1068,7 +1068,7 @@ static int __udp4_lib_mcast_deliver(struct net *net, struct sk_buff *skb,
 	int dif;
 
 	read_lock(&udp_hash_lock);
-	sk = sk_head(&udptable[udp_hashfn(ntohs(uh->dest))]);
+	sk = sk_head(&udptable[udp_hashfn(net, ntohs(uh->dest))]);
 	dif = skb->dev->ifindex;
 	sk = udp_v4_mcast_next(sk, uh->dest, daddr, uh->source, saddr, dif);
 	if (sk) {
diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c
index 80fb72c48976..432edaa882f6 100644
--- a/net/ipv6/udp.c
+++ b/net/ipv6/udp.c
@@ -65,7 +65,7 @@ static struct sock *__udp6_lib_lookup(struct net *net,
 	int badness = -1;
 
 	read_lock(&udp_hash_lock);
-	sk_for_each(sk, node, &udptable[udp_hashfn(hnum)]) {
+	sk_for_each(sk, node, &udptable[udp_hashfn(net, hnum)]) {
 		struct inet_sock *inet = inet_sk(sk);
 
 		if (net_eq(sock_net(sk), net) && sk->sk_hash == hnum &&
@@ -362,7 +362,7 @@ static int __udp6_lib_mcast_deliver(struct net *net, struct sk_buff *skb,
 	int dif;
 
 	read_lock(&udp_hash_lock);
-	sk = sk_head(&udptable[udp_hashfn(ntohs(uh->dest))]);
+	sk = sk_head(&udptable[udp_hashfn(net, ntohs(uh->dest))]);
 	dif = inet6_iif(skb);
 	sk = udp_v6_mcast_next(sk, uh->dest, daddr, uh->source, saddr, dif);
 	if (!sk) {
-- 
cgit v1.2.3


From 7f635ab71eef8da012320c0092b662d6af8c1e69 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Mon, 16 Jun 2008 17:12:49 -0700
Subject: inet: add struct net argument to inet_bhashfn

Binding to some port in many namespaces may create too long
chains in bhash-es, so prepare the hashfn to take struct net
into account.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/inet_hashtables.h   |  3 ++-
 net/ipv4/inet_connection_sock.c |  6 ++++--
 net/ipv4/inet_hashtables.c      | 11 +++++++----
 net/ipv4/inet_timewait_sock.c   |  6 ++++--
 4 files changed, 17 insertions(+), 9 deletions(-)

(limited to 'include')

diff --git a/include/net/inet_hashtables.h b/include/net/inet_hashtables.h
index 735b926a3497..61dd3317089c 100644
--- a/include/net/inet_hashtables.h
+++ b/include/net/inet_hashtables.h
@@ -201,7 +201,8 @@ extern struct inet_bind_bucket *
 extern void inet_bind_bucket_destroy(struct kmem_cache *cachep,
 				     struct inet_bind_bucket *tb);
 
-static inline int inet_bhashfn(const __u16 lport, const int bhash_size)
+static inline int inet_bhashfn(struct net *net,
+		const __u16 lport, const int bhash_size)
 {
 	return lport & (bhash_size - 1);
 }
diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c
index 045e799d3e1d..4c804b3c287b 100644
--- a/net/ipv4/inet_connection_sock.c
+++ b/net/ipv4/inet_connection_sock.c
@@ -103,7 +103,8 @@ int inet_csk_get_port(struct sock *sk, unsigned short snum)
 		rover = net_random() % remaining + low;
 
 		do {
-			head = &hashinfo->bhash[inet_bhashfn(rover, hashinfo->bhash_size)];
+			head = &hashinfo->bhash[inet_bhashfn(net, rover,
+					hashinfo->bhash_size)];
 			spin_lock(&head->lock);
 			inet_bind_bucket_for_each(tb, node, &head->chain)
 				if (tb->ib_net == net && tb->port == rover)
@@ -130,7 +131,8 @@ int inet_csk_get_port(struct sock *sk, unsigned short snum)
 		 */
 		snum = rover;
 	} else {
-		head = &hashinfo->bhash[inet_bhashfn(snum, hashinfo->bhash_size)];
+		head = &hashinfo->bhash[inet_bhashfn(net, snum,
+				hashinfo->bhash_size)];
 		spin_lock(&head->lock);
 		inet_bind_bucket_for_each(tb, node, &head->chain)
 			if (tb->ib_net == net && tb->port == snum)
diff --git a/net/ipv4/inet_hashtables.c b/net/ipv4/inet_hashtables.c
index 2023d37b2708..dc1b78de8990 100644
--- a/net/ipv4/inet_hashtables.c
+++ b/net/ipv4/inet_hashtables.c
@@ -70,7 +70,8 @@ void inet_bind_hash(struct sock *sk, struct inet_bind_bucket *tb,
 static void __inet_put_port(struct sock *sk)
 {
 	struct inet_hashinfo *hashinfo = sk->sk_prot->h.hashinfo;
-	const int bhash = inet_bhashfn(inet_sk(sk)->num, hashinfo->bhash_size);
+	const int bhash = inet_bhashfn(sock_net(sk), inet_sk(sk)->num,
+			hashinfo->bhash_size);
 	struct inet_bind_hashbucket *head = &hashinfo->bhash[bhash];
 	struct inet_bind_bucket *tb;
 
@@ -95,7 +96,8 @@ EXPORT_SYMBOL(inet_put_port);
 void __inet_inherit_port(struct sock *sk, struct sock *child)
 {
 	struct inet_hashinfo *table = sk->sk_prot->h.hashinfo;
-	const int bhash = inet_bhashfn(inet_sk(child)->num, table->bhash_size);
+	const int bhash = inet_bhashfn(sock_net(sk), inet_sk(child)->num,
+			table->bhash_size);
 	struct inet_bind_hashbucket *head = &table->bhash[bhash];
 	struct inet_bind_bucket *tb;
 
@@ -438,7 +440,8 @@ int __inet_hash_connect(struct inet_timewait_death_row *death_row,
 		local_bh_disable();
 		for (i = 1; i <= remaining; i++) {
 			port = low + (i + offset) % remaining;
-			head = &hinfo->bhash[inet_bhashfn(port, hinfo->bhash_size)];
+			head = &hinfo->bhash[inet_bhashfn(net, port,
+					hinfo->bhash_size)];
 			spin_lock(&head->lock);
 
 			/* Does not bother with rcv_saddr checks,
@@ -493,7 +496,7 @@ ok:
 		goto out;
 	}
 
-	head = &hinfo->bhash[inet_bhashfn(snum, hinfo->bhash_size)];
+	head = &hinfo->bhash[inet_bhashfn(net, snum, hinfo->bhash_size)];
 	tb  = inet_csk(sk)->icsk_bind_hash;
 	spin_lock_bh(&head->lock);
 	if (sk_head(&tb->owners) == sk && !sk->sk_bind_node.next) {
diff --git a/net/ipv4/inet_timewait_sock.c b/net/ipv4/inet_timewait_sock.c
index ce16e9ac24c1..06006a5ac8b9 100644
--- a/net/ipv4/inet_timewait_sock.c
+++ b/net/ipv4/inet_timewait_sock.c
@@ -32,7 +32,8 @@ static void __inet_twsk_kill(struct inet_timewait_sock *tw,
 	write_unlock(lock);
 
 	/* Disassociate with bind bucket. */
-	bhead = &hashinfo->bhash[inet_bhashfn(tw->tw_num, hashinfo->bhash_size)];
+	bhead = &hashinfo->bhash[inet_bhashfn(twsk_net(tw), tw->tw_num,
+			hashinfo->bhash_size)];
 	spin_lock(&bhead->lock);
 	tb = tw->tw_tb;
 	__hlist_del(&tw->tw_bind_node);
@@ -81,7 +82,8 @@ void __inet_twsk_hashdance(struct inet_timewait_sock *tw, struct sock *sk,
 	   Note, that any socket with inet->num != 0 MUST be bound in
 	   binding cache, even if it is closed.
 	 */
-	bhead = &hashinfo->bhash[inet_bhashfn(inet->num, hashinfo->bhash_size)];
+	bhead = &hashinfo->bhash[inet_bhashfn(twsk_net(tw), inet->num,
+			hashinfo->bhash_size)];
 	spin_lock(&bhead->lock);
 	tw->tw_tb = icsk->icsk_bind_hash;
 	BUG_TRAP(icsk->icsk_bind_hash);
-- 
cgit v1.2.3


From 2086a65078bd24682bdcf413d9c91d81988b8359 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Mon, 16 Jun 2008 17:13:08 -0700
Subject: inet: add struct net argument to inet_lhashfn

Listening-on-one-port sockets in many namespaces produce long
chains in the listening_hash-es, so prepare the inet_lhashfn to
take struct net into account.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/inet_hashtables.h | 4 ++--
 net/ipv4/inet_hashtables.c    | 2 +-
 net/ipv6/inet6_hashtables.c   | 3 ++-
 3 files changed, 5 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/include/net/inet_hashtables.h b/include/net/inet_hashtables.h
index 61dd3317089c..26336cdcdc11 100644
--- a/include/net/inet_hashtables.h
+++ b/include/net/inet_hashtables.h
@@ -211,14 +211,14 @@ extern void inet_bind_hash(struct sock *sk, struct inet_bind_bucket *tb,
 			   const unsigned short snum);
 
 /* These can have wildcards, don't try too hard. */
-static inline int inet_lhashfn(const unsigned short num)
+static inline int inet_lhashfn(struct net *net, const unsigned short num)
 {
 	return num & (INET_LHTABLE_SIZE - 1);
 }
 
 static inline int inet_sk_listen_hashfn(const struct sock *sk)
 {
-	return inet_lhashfn(inet_sk(sk)->num);
+	return inet_lhashfn(sock_net(sk), inet_sk(sk)->num);
 }
 
 /* Caller must disable local BH processing. */
diff --git a/net/ipv4/inet_hashtables.c b/net/ipv4/inet_hashtables.c
index dc1b78de8990..4f597b3175e7 100644
--- a/net/ipv4/inet_hashtables.c
+++ b/net/ipv4/inet_hashtables.c
@@ -194,7 +194,7 @@ struct sock *__inet_lookup_listener(struct net *net,
 	const struct hlist_head *head;
 
 	read_lock(&hashinfo->lhash_lock);
-	head = &hashinfo->listening_hash[inet_lhashfn(hnum)];
+	head = &hashinfo->listening_hash[inet_lhashfn(net, hnum)];
 	if (!hlist_empty(head)) {
 		const struct inet_sock *inet = inet_sk((sk = __sk_head(head)));
 
diff --git a/net/ipv6/inet6_hashtables.c b/net/ipv6/inet6_hashtables.c
index 580014aea4d6..b940156ca4f5 100644
--- a/net/ipv6/inet6_hashtables.c
+++ b/net/ipv6/inet6_hashtables.c
@@ -104,7 +104,8 @@ struct sock *inet6_lookup_listener(struct net *net,
 	int score, hiscore = 0;
 
 	read_lock(&hashinfo->lhash_lock);
-	sk_for_each(sk, node, &hashinfo->listening_hash[inet_lhashfn(hnum)]) {
+	sk_for_each(sk, node,
+			&hashinfo->listening_hash[inet_lhashfn(net, hnum)]) {
 		if (net_eq(sock_net(sk), net) && inet_sk(sk)->num == hnum &&
 				sk->sk_family == PF_INET6) {
 			const struct ipv6_pinfo *np = inet6_sk(sk);
-- 
cgit v1.2.3


From 9f26b3add3783c0af869ea2207871da5dafefffa Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Mon, 16 Jun 2008 17:13:27 -0700
Subject: inet: add struct net argument to inet_ehashfn

Although this hash takes addresses into account, the ehash chains
can also be too long when, for instance, communications via lo occur.
So, prepare the inet_hashfn to take struct net into account.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/inet_sock.h    | 6 ++++--
 net/ipv4/inet_hashtables.c | 6 +++---
 2 files changed, 7 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/include/net/inet_sock.h b/include/net/inet_sock.h
index 9fabe5b38912..85bb420c5d86 100644
--- a/include/net/inet_sock.h
+++ b/include/net/inet_sock.h
@@ -171,7 +171,8 @@ extern int inet_sk_rebuild_header(struct sock *sk);
 extern u32 inet_ehash_secret;
 extern void build_ehash_secret(void);
 
-static inline unsigned int inet_ehashfn(const __be32 laddr, const __u16 lport,
+static inline unsigned int inet_ehashfn(struct net *net,
+					const __be32 laddr, const __u16 lport,
 					const __be32 faddr, const __be16 fport)
 {
 	return jhash_3words((__force __u32) laddr,
@@ -187,8 +188,9 @@ static inline int inet_sk_ehashfn(const struct sock *sk)
 	const __u16 lport = inet->num;
 	const __be32 faddr = inet->daddr;
 	const __be16 fport = inet->dport;
+	struct net *net = sock_net(sk);
 
-	return inet_ehashfn(laddr, lport, faddr, fport);
+	return inet_ehashfn(net, laddr, lport, faddr, fport);
 }
 
 
diff --git a/net/ipv4/inet_hashtables.c b/net/ipv4/inet_hashtables.c
index 4f597b3175e7..eca5899729e3 100644
--- a/net/ipv4/inet_hashtables.c
+++ b/net/ipv4/inet_hashtables.c
@@ -227,7 +227,7 @@ struct sock * __inet_lookup_established(struct net *net,
 	/* Optimize here for direct hit, only listening connections can
 	 * have wildcards anyways.
 	 */
-	unsigned int hash = inet_ehashfn(daddr, hnum, saddr, sport);
+	unsigned int hash = inet_ehashfn(net, daddr, hnum, saddr, sport);
 	struct inet_ehash_bucket *head = inet_ehash_bucket(hashinfo, hash);
 	rwlock_t *lock = inet_ehash_lockp(hashinfo, hash);
 
@@ -267,13 +267,13 @@ static int __inet_check_established(struct inet_timewait_death_row *death_row,
 	int dif = sk->sk_bound_dev_if;
 	INET_ADDR_COOKIE(acookie, saddr, daddr)
 	const __portpair ports = INET_COMBINED_PORTS(inet->dport, lport);
-	unsigned int hash = inet_ehashfn(daddr, lport, saddr, inet->dport);
+	struct net *net = sock_net(sk);
+	unsigned int hash = inet_ehashfn(net, daddr, lport, saddr, inet->dport);
 	struct inet_ehash_bucket *head = inet_ehash_bucket(hinfo, hash);
 	rwlock_t *lock = inet_ehash_lockp(hinfo, hash);
 	struct sock *sk2;
 	const struct hlist_node *node;
 	struct inet_timewait_sock *tw;
-	struct net *net = sock_net(sk);
 
 	prefetch(head->chain.first);
 	write_lock(lock);
-- 
cgit v1.2.3


From 33de014c63646f69f36f3673e3b4676f931dc878 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Mon, 16 Jun 2008 17:13:48 -0700
Subject: inet6: add struct net argument to inet6_ehashfn

Same as for inet_hashfn, prepare its ipv6 incarnation.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/inet6_hashtables.h | 7 +++++--
 net/ipv6/inet6_hashtables.c    | 6 +++---
 2 files changed, 8 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/include/net/inet6_hashtables.h b/include/net/inet6_hashtables.h
index 62a5b691858e..72f13a9928e4 100644
--- a/include/net/inet6_hashtables.h
+++ b/include/net/inet6_hashtables.h
@@ -28,7 +28,8 @@
 struct inet_hashinfo;
 
 /* I have no idea if this is a good hash for v6 or not. -DaveM */
-static inline unsigned int inet6_ehashfn(const struct in6_addr *laddr, const u16 lport,
+static inline unsigned int inet6_ehashfn(struct net *net,
+				const struct in6_addr *laddr, const u16 lport,
 				const struct in6_addr *faddr, const __be16 fport)
 {
 	u32 ports = (lport ^ (__force u16)fport);
@@ -46,7 +47,9 @@ static inline int inet6_sk_ehashfn(const struct sock *sk)
 	const struct in6_addr *faddr = &np->daddr;
 	const __u16 lport = inet->num;
 	const __be16 fport = inet->dport;
-	return inet6_ehashfn(laddr, lport, faddr, fport);
+	struct net *net = sock_net(sk);
+
+	return inet6_ehashfn(net, laddr, lport, faddr, fport);
 }
 
 extern void __inet6_hash(struct sock *sk);
diff --git a/net/ipv6/inet6_hashtables.c b/net/ipv6/inet6_hashtables.c
index b940156ca4f5..a9cc8ab33a49 100644
--- a/net/ipv6/inet6_hashtables.c
+++ b/net/ipv6/inet6_hashtables.c
@@ -68,7 +68,7 @@ struct sock *__inet6_lookup_established(struct net *net,
 	/* Optimize here for direct hit, only listening connections can
 	 * have wildcards anyways.
 	 */
-	unsigned int hash = inet6_ehashfn(daddr, hnum, saddr, sport);
+	unsigned int hash = inet6_ehashfn(net, daddr, hnum, saddr, sport);
 	struct inet_ehash_bucket *head = inet_ehash_bucket(hashinfo, hash);
 	rwlock_t *lock = inet_ehash_lockp(hashinfo, hash);
 
@@ -166,14 +166,14 @@ static int __inet6_check_established(struct inet_timewait_death_row *death_row,
 	const struct in6_addr *saddr = &np->daddr;
 	const int dif = sk->sk_bound_dev_if;
 	const __portpair ports = INET_COMBINED_PORTS(inet->dport, lport);
-	const unsigned int hash = inet6_ehashfn(daddr, lport, saddr,
+	struct net *net = sock_net(sk);
+	const unsigned int hash = inet6_ehashfn(net, daddr, lport, saddr,
 						inet->dport);
 	struct inet_ehash_bucket *head = inet_ehash_bucket(hinfo, hash);
 	rwlock_t *lock = inet_ehash_lockp(hinfo, hash);
 	struct sock *sk2;
 	const struct hlist_node *node;
 	struct inet_timewait_sock *tw;
-	struct net *net = sock_net(sk);
 
 	prefetch(head->chain.first);
 	write_lock(lock);
-- 
cgit v1.2.3


From 0b4419162aa6c4204843f3a13b48d9ab821d3167 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Mon, 16 Jun 2008 17:14:11 -0700
Subject: netns: introduce the net_hash_mix "salt" for hashes

There are many possible ways to add this "salt", thus I made this
patch to be the last in the series to change it if required.

Currently I propose to use the struct net pointer itself as this
salt, but since this pointer is most often cache-line aligned, shift
this right to eliminate the bits, that are most often zeroed.

After this, simply add this mix to prepared hashfn-s.

For CONFIG_NET_NS=n case this salt is 0 and no changes in hashfn
appear.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/udp.h            |  3 ++-
 include/net/inet6_hashtables.h |  3 ++-
 include/net/inet_hashtables.h  |  5 +++--
 include/net/inet_sock.h        |  3 ++-
 include/net/netns/hash.h       | 21 +++++++++++++++++++++
 5 files changed, 30 insertions(+), 5 deletions(-)
 create mode 100644 include/net/netns/hash.h

(limited to 'include')

diff --git a/include/linux/udp.h b/include/linux/udp.h
index 3deccac2e815..0cf5c4c0ec81 100644
--- a/include/linux/udp.h
+++ b/include/linux/udp.h
@@ -38,6 +38,7 @@ struct udphdr {
 #ifdef __KERNEL__
 #include <net/inet_sock.h>
 #include <linux/skbuff.h>
+#include <net/netns/hash.h>
 
 static inline struct udphdr *udp_hdr(const struct sk_buff *skb)
 {
@@ -48,7 +49,7 @@ static inline struct udphdr *udp_hdr(const struct sk_buff *skb)
 
 static inline int udp_hashfn(struct net *net, const unsigned num)
 {
-	return num & (UDP_HTABLE_SIZE - 1);
+	return (num + net_hash_mix(net)) & (UDP_HTABLE_SIZE - 1);
 }
 
 struct udp_sock {
diff --git a/include/net/inet6_hashtables.h b/include/net/inet6_hashtables.h
index 72f13a9928e4..e48989f04c24 100644
--- a/include/net/inet6_hashtables.h
+++ b/include/net/inet6_hashtables.h
@@ -24,6 +24,7 @@
 #include <net/inet_sock.h>
 
 #include <net/ipv6.h>
+#include <net/netns/hash.h>
 
 struct inet_hashinfo;
 
@@ -36,7 +37,7 @@ static inline unsigned int inet6_ehashfn(struct net *net,
 
 	return jhash_3words((__force u32)laddr->s6_addr32[3],
 			    (__force u32)faddr->s6_addr32[3],
-			    ports, inet_ehash_secret);
+			    ports, inet_ehash_secret + net_hash_mix(net));
 }
 
 static inline int inet6_sk_ehashfn(const struct sock *sk)
diff --git a/include/net/inet_hashtables.h b/include/net/inet_hashtables.h
index 26336cdcdc11..bb619d80f2e2 100644
--- a/include/net/inet_hashtables.h
+++ b/include/net/inet_hashtables.h
@@ -29,6 +29,7 @@
 #include <net/inet_sock.h>
 #include <net/sock.h>
 #include <net/tcp_states.h>
+#include <net/netns/hash.h>
 
 #include <asm/atomic.h>
 #include <asm/byteorder.h>
@@ -204,7 +205,7 @@ extern void inet_bind_bucket_destroy(struct kmem_cache *cachep,
 static inline int inet_bhashfn(struct net *net,
 		const __u16 lport, const int bhash_size)
 {
-	return lport & (bhash_size - 1);
+	return (lport + net_hash_mix(net)) & (bhash_size - 1);
 }
 
 extern void inet_bind_hash(struct sock *sk, struct inet_bind_bucket *tb,
@@ -213,7 +214,7 @@ extern void inet_bind_hash(struct sock *sk, struct inet_bind_bucket *tb,
 /* These can have wildcards, don't try too hard. */
 static inline int inet_lhashfn(struct net *net, const unsigned short num)
 {
-	return num & (INET_LHTABLE_SIZE - 1);
+	return (num + net_hash_mix(net)) & (INET_LHTABLE_SIZE - 1);
 }
 
 static inline int inet_sk_listen_hashfn(const struct sock *sk)
diff --git a/include/net/inet_sock.h b/include/net/inet_sock.h
index 85bb420c5d86..643e26be058e 100644
--- a/include/net/inet_sock.h
+++ b/include/net/inet_sock.h
@@ -25,6 +25,7 @@
 #include <net/sock.h>
 #include <net/request_sock.h>
 #include <net/route.h>
+#include <net/netns/hash.h>
 
 /** struct ip_options - IP Options
  *
@@ -178,7 +179,7 @@ static inline unsigned int inet_ehashfn(struct net *net,
 	return jhash_3words((__force __u32) laddr,
 			    (__force __u32) faddr,
 			    ((__u32) lport) << 16 | (__force __u32)fport,
-			    inet_ehash_secret);
+			    inet_ehash_secret + net_hash_mix(net));
 }
 
 static inline int inet_sk_ehashfn(const struct sock *sk)
diff --git a/include/net/netns/hash.h b/include/net/netns/hash.h
new file mode 100644
index 000000000000..548d78f2cc47
--- /dev/null
+++ b/include/net/netns/hash.h
@@ -0,0 +1,21 @@
+#ifndef __NET_NS_HASH_H__
+#define __NET_NS_HASH_H__
+
+#include <asm/cache.h>
+
+struct net;
+
+static inline unsigned net_hash_mix(struct net *net)
+{
+#ifdef CONFIG_NET_NS
+	/*
+	 * shift this right to eliminate bits, that are
+	 * always zeroed
+	 */
+
+	return (unsigned)(((unsigned long)net) >> L1_CACHE_SHIFT);
+#else
+	return 0;
+#endif
+}
+#endif
-- 
cgit v1.2.3


From 87de87d5e47f94b4ea647a5bd1bc8dc1f7930db4 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Tue, 3 Jun 2008 09:14:03 -0700
Subject: wext: Dispatch and handle compat ioctls entirely in
 net/wireless/wext.c

Next we can kill the hacks in fs/compat_ioctl.c and also
dispatch compat ioctls down into the driver and 80211 protocol
helper layers in order to handle iw_point objects embedded in
stream replies which need to be translated.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 fs/compat_ioctl.c        |   6 ---
 include/linux/wireless.h |  13 ++++++
 include/net/wext.h       |   7 ++++
 net/socket.c             |  10 +++++
 net/wireless/wext.c      | 104 +++++++++++++++++++++++++++++++++++++++++++++++
 5 files changed, 134 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/fs/compat_ioctl.c b/fs/compat_ioctl.c
index 97dba0d92348..8ab850bf2eee 100644
--- a/fs/compat_ioctl.c
+++ b/fs/compat_ioctl.c
@@ -1757,12 +1757,6 @@ static int do_i2c_smbus_ioctl(unsigned int fd, unsigned int cmd, unsigned long a
 	return sys_ioctl(fd, cmd, (unsigned long)tdata);
 }
 
-struct compat_iw_point {
-	compat_caddr_t pointer;
-	__u16 length;
-	__u16 flags;
-};
-
 static int do_wireless_ioctl(unsigned int fd, unsigned int cmd, unsigned long arg)
 {
 	struct iwreq __user *iwr;
diff --git a/include/linux/wireless.h b/include/linux/wireless.h
index 4a95a0e5eeca..79d846875825 100644
--- a/include/linux/wireless.h
+++ b/include/linux/wireless.h
@@ -677,6 +677,19 @@ struct	iw_point
   __u16		flags;		/* Optional params */
 };
 
+#ifdef __KERNEL__
+#ifdef CONFIG_COMPAT
+
+#include <linux/compat.h>
+
+struct compat_iw_point {
+	compat_caddr_t pointer;
+	__u16 length;
+	__u16 flags;
+};
+#endif
+#endif
+
 /*
  *	A frequency
  *	For numbers lower than 10^9, we encode the number in 'm' and
diff --git a/include/net/wext.h b/include/net/wext.h
index 80b31d826b7a..6d76a39a9c5b 100644
--- a/include/net/wext.h
+++ b/include/net/wext.h
@@ -12,6 +12,8 @@ extern int wext_proc_init(struct net *net);
 extern void wext_proc_exit(struct net *net);
 extern int wext_handle_ioctl(struct net *net, struct ifreq *ifr, unsigned int cmd,
 			     void __user *arg);
+extern int compat_wext_handle_ioctl(struct net *net, unsigned int cmd,
+				    unsigned long arg);
 #else
 static inline int wext_proc_init(struct net *net)
 {
@@ -26,6 +28,11 @@ static inline int wext_handle_ioctl(struct net *net, struct ifreq *ifr, unsigned
 {
 	return -EINVAL;
 }
+static inline int compat_wext_handle_ioctl(struct net *net, unsigned int cmd,
+					   unsigned long arg)
+{
+	return -EINVAL;
+}
 #endif
 
 #endif /* __NET_WEXT_H */
diff --git a/net/socket.c b/net/socket.c
index 66c4a8cf6db9..81fe82513046 100644
--- a/net/socket.c
+++ b/net/socket.c
@@ -90,6 +90,7 @@
 #include <asm/unistd.h>
 
 #include <net/compat.h>
+#include <net/wext.h>
 
 #include <net/sock.h>
 #include <linux/netfilter.h>
@@ -2210,10 +2211,19 @@ static long compat_sock_ioctl(struct file *file, unsigned cmd,
 {
 	struct socket *sock = file->private_data;
 	int ret = -ENOIOCTLCMD;
+	struct sock *sk;
+	struct net *net;
+
+	sk = sock->sk;
+	net = sock_net(sk);
 
 	if (sock->ops->compat_ioctl)
 		ret = sock->ops->compat_ioctl(sock, cmd, arg);
 
+	if (ret == -ENOIOCTLCMD &&
+	    (cmd >= SIOCIWFIRST && cmd <= SIOCIWLAST))
+		ret = compat_wext_handle_ioctl(net, cmd, arg);
+
 	return ret;
 }
 #endif
diff --git a/net/wireless/wext.c b/net/wireless/wext.c
index 09022cbb58ba..1a4636a9fcde 100644
--- a/net/wireless/wext.c
+++ b/net/wireless/wext.c
@@ -1112,6 +1112,110 @@ int wext_handle_ioctl(struct net *net, struct ifreq *ifr, unsigned int cmd,
 	return ret;
 }
 
+#ifdef CONFIG_COMPAT
+static int compat_standard_call(struct net_device	*dev,
+				struct iwreq		*iwr,
+				unsigned int		cmd,
+				iw_handler		handler)
+{
+	const struct iw_ioctl_description *descr;
+	struct compat_iw_point *iwp_compat;
+	struct iw_request_info info;
+	struct iw_point iwp;
+	int err;
+
+	descr = standard_ioctl + (cmd - SIOCIWFIRST);
+
+	if (descr->header_type != IW_HEADER_TYPE_POINT)
+		return ioctl_standard_call(dev, iwr, cmd, handler);
+
+	iwp_compat = (struct compat_iw_point *) &iwr->u.data;
+	iwp.pointer = compat_ptr(iwp_compat->pointer);
+	iwp.length = iwp_compat->length;
+	iwp.flags = iwp_compat->flags;
+
+	info.cmd = cmd;
+	info.flags = 0;
+
+	err = ioctl_standard_iw_point(&iwp, cmd, descr, handler, dev, &info);
+
+	iwp_compat->pointer = ptr_to_compat(iwp.pointer);
+	iwp_compat->length = iwp.length;
+	iwp_compat->flags = iwp.flags;
+
+	return err;
+}
+
+static int compat_private_call(struct net_device *dev, struct iwreq *iwr,
+			       unsigned int cmd, iw_handler handler)
+{
+	const struct iw_priv_args *descr;
+	struct iw_request_info info;
+	int ret, extra_size;
+
+	extra_size = get_priv_descr_and_size(dev, cmd, &descr);
+
+	/* Prepare the call */
+	info.cmd = cmd;
+	info.flags = 0;
+
+	/* Check if we have a pointer to user space data or not. */
+	if (extra_size == 0) {
+		/* No extra arguments. Trivial to handle */
+		ret = handler(dev, &info, &(iwr->u), (char *) &(iwr->u));
+	} else {
+		struct compat_iw_point *iwp_compat;
+		struct iw_point iwp;
+
+		iwp_compat = (struct compat_iw_point *) &iwr->u.data;
+		iwp.pointer = compat_ptr(iwp_compat->pointer);
+		iwp.length = iwp_compat->length;
+		iwp.flags = iwp_compat->flags;
+
+		ret = ioctl_private_iw_point(&iwp, cmd, descr,
+					     handler, dev, &info, extra_size);
+
+		iwp_compat->pointer = ptr_to_compat(iwp.pointer);
+		iwp_compat->length = iwp.length;
+		iwp_compat->flags = iwp.flags;
+	}
+
+	/* Call commit handler if needed and defined */
+	if (ret == -EIWCOMMIT)
+		ret = call_commit_handler(dev);
+
+	return ret;
+}
+
+int compat_wext_handle_ioctl(struct net *net, unsigned int cmd,
+			     unsigned long arg)
+{
+	void __user *argp = (void __user *)arg;
+	struct iwreq iwr;
+	char *colon;
+	int ret;
+
+	if (copy_from_user(&iwr, argp, sizeof(struct iwreq)))
+		return -EFAULT;
+
+	iwr.ifr_name[IFNAMSIZ-1] = 0;
+	colon = strchr(iwr.ifr_name, ':');
+	if (colon)
+		*colon = 0;
+
+	ret = wext_ioctl_dispatch(net, (struct ifreq *) &iwr, cmd,
+				  compat_standard_call,
+				  compat_private_call);
+
+	if (ret >= 0 &&
+	    IW_IS_GET(cmd) &&
+	    copy_to_user(argp, &iwr, sizeof(struct iwreq)))
+		return -EFAULT;
+
+	return ret;
+}
+#endif
+
 /************************* EVENT PROCESSING *************************/
 /*
  * Process events generated by the wireless layer or the driver.
-- 
cgit v1.2.3


From 0f5cabba49021d36e9f76bd97d7fa0f4a408063f Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Tue, 3 Jun 2008 07:39:16 -0700
Subject: wext: Create IW_REQUEST_FLAG_COMPAT and set it as needed.

Now low-level WEXT ioctl handlers can do compat handling
when necessary.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/iw_handler.h |  2 +-
 net/wireless/wext.c      | 73 ++++++++++++++++++++++--------------------------
 2 files changed, 34 insertions(+), 41 deletions(-)

(limited to 'include')

diff --git a/include/net/iw_handler.h b/include/net/iw_handler.h
index 369d50e08b99..c99a8eec84e7 100644
--- a/include/net/iw_handler.h
+++ b/include/net/iw_handler.h
@@ -256,7 +256,7 @@
 #define EIWCOMMIT	EINPROGRESS
 
 /* Flags available in struct iw_request_info */
-#define IW_REQUEST_FLAG_NONE	0x0000	/* No flag so far */
+#define IW_REQUEST_FLAG_COMPAT	0x0001	/* Compat ioctl call */
 
 /* Type of headers we know about (basically union iwreq_data) */
 #define IW_HEADER_TYPE_NULL	0	/* Not available */
diff --git a/net/wireless/wext.c b/net/wireless/wext.c
index 1a4636a9fcde..273a84359998 100644
--- a/net/wireless/wext.c
+++ b/net/wireless/wext.c
@@ -834,10 +834,10 @@ out:
 static int ioctl_standard_call(struct net_device *	dev,
 			       struct iwreq		*iwr,
 			       unsigned int		cmd,
+			       struct iw_request_info	*info,
 			       iw_handler		handler)
 {
 	const struct iw_ioctl_description *	descr;
-	struct iw_request_info			info;
 	int					ret = -EINVAL;
 
 	/* Get the description of the IOCTL */
@@ -845,15 +845,11 @@ static int ioctl_standard_call(struct net_device *	dev,
 		return -EOPNOTSUPP;
 	descr = &(standard_ioctl[cmd - SIOCIWFIRST]);
 
-	/* Prepare the call */
-	info.cmd = cmd;
-	info.flags = 0;
-
 	/* Check if we have a pointer to user space data or not */
 	if (descr->header_type != IW_HEADER_TYPE_POINT) {
 
 		/* No extra arguments. Trivial to handle */
-		ret = handler(dev, &info, &(iwr->u), NULL);
+		ret = handler(dev, info, &(iwr->u), NULL);
 
 		/* Generate an event to notify listeners of the change */
 		if ((descr->flags & IW_DESCR_FLAG_EVENT) &&
@@ -861,7 +857,7 @@ static int ioctl_standard_call(struct net_device *	dev,
 			wireless_send_event(dev, cmd, &(iwr->u), NULL);
 	} else {
 		ret = ioctl_standard_iw_point(&iwr->u.data, cmd, descr,
-					      handler, dev, &info);
+					      handler, dev, info);
 	}
 
 	/* Call commit handler if needed and defined */
@@ -984,25 +980,21 @@ out:
 }
 
 static int ioctl_private_call(struct net_device *dev, struct iwreq *iwr,
-			      unsigned int cmd, iw_handler handler)
+			      unsigned int cmd, struct iw_request_info *info,
+			      iw_handler handler)
 {
 	int extra_size = 0, ret = -EINVAL;
 	const struct iw_priv_args *descr;
-	struct iw_request_info info;
 
 	extra_size = get_priv_descr_and_size(dev, cmd, &descr);
 
-	/* Prepare the call */
-	info.cmd = cmd;
-	info.flags = 0;
-
 	/* Check if we have a pointer to user space data or not. */
 	if (extra_size == 0) {
 		/* No extra arguments. Trivial to handle */
-		ret = handler(dev, &info, &(iwr->u), (char *) &(iwr->u));
+		ret = handler(dev, info, &(iwr->u), (char *) &(iwr->u));
 	} else {
 		ret = ioctl_private_iw_point(&iwr->u.data, cmd, descr,
-					     handler, dev, &info, extra_size);
+					     handler, dev, info, extra_size);
 	}
 
 	/* Call commit handler if needed and defined */
@@ -1014,7 +1006,8 @@ static int ioctl_private_call(struct net_device *dev, struct iwreq *iwr,
 
 /* ---------------------------------------------------------------- */
 typedef int (*wext_ioctl_func)(struct net_device *, struct iwreq *,
-			       unsigned int, iw_handler);
+			       unsigned int, struct iw_request_info *,
+			       iw_handler);
 
 /*
  * Main IOCTl dispatcher.
@@ -1022,6 +1015,7 @@ typedef int (*wext_ioctl_func)(struct net_device *, struct iwreq *,
  */
 static int wireless_process_ioctl(struct net *net, struct ifreq *ifr,
 				  unsigned int cmd,
+				  struct iw_request_info *info,
 				  wext_ioctl_func standard,
 				  wext_ioctl_func private)
 {
@@ -1040,11 +1034,11 @@ static int wireless_process_ioctl(struct net *net, struct ifreq *ifr,
 	 * Note that 'cmd' is already filtered in dev_ioctl() with
 	 * (cmd >= SIOCIWFIRST && cmd <= SIOCIWLAST) */
 	if (cmd == SIOCGIWSTATS)
-		return standard(dev, iwr, cmd,
+		return standard(dev, iwr, cmd, info,
 				&iw_handler_get_iwstats);
 
 	if (cmd == SIOCGIWPRIV && dev->wireless_handlers)
-		return standard(dev, iwr, cmd,
+		return standard(dev, iwr, cmd, info,
 				&iw_handler_get_private);
 
 	/* Basic check */
@@ -1056,9 +1050,9 @@ static int wireless_process_ioctl(struct net *net, struct ifreq *ifr,
 	if (handler) {
 		/* Standard and private are not the same */
 		if (cmd < SIOCIWFIRSTPRIV)
-			return standard(dev, iwr, cmd, handler);
+			return standard(dev, iwr, cmd, info, handler);
 		else
-			return private(dev, iwr, cmd, handler);
+			return private(dev, iwr, cmd, info, handler);
 	}
 	/* Old driver API : call driver ioctl handler */
 	if (dev->do_ioctl)
@@ -1080,7 +1074,7 @@ static int wext_permission_check(unsigned int cmd)
 
 /* entry point from dev ioctl */
 static int wext_ioctl_dispatch(struct net *net, struct ifreq *ifr,
-			       unsigned int cmd,
+			       unsigned int cmd, struct iw_request_info *info,
 			       wext_ioctl_func standard,
 			       wext_ioctl_func private)
 {
@@ -1091,7 +1085,7 @@ static int wext_ioctl_dispatch(struct net *net, struct ifreq *ifr,
 
 	dev_load(net, ifr->ifr_name);
 	rtnl_lock();
-	ret = wireless_process_ioctl(net, ifr, cmd, standard, private);
+	ret = wireless_process_ioctl(net, ifr, cmd, info, standard, private);
 	rtnl_unlock();
 
 	return ret;
@@ -1100,10 +1094,12 @@ static int wext_ioctl_dispatch(struct net *net, struct ifreq *ifr,
 int wext_handle_ioctl(struct net *net, struct ifreq *ifr, unsigned int cmd,
 		      void __user *arg)
 {
-	int ret = wext_ioctl_dispatch(net, ifr, cmd,
-				      ioctl_standard_call,
-				      ioctl_private_call);
+	struct iw_request_info info = { .cmd = cmd, .flags = 0 };
+	int ret;
 
+	ret = wext_ioctl_dispatch(net, ifr, cmd, &info,
+				  ioctl_standard_call,
+				  ioctl_private_call);
 	if (ret >= 0 &&
 	    IW_IS_GET(cmd) &&
 	    copy_to_user(arg, ifr, sizeof(struct iwreq)))
@@ -1116,28 +1112,25 @@ int wext_handle_ioctl(struct net *net, struct ifreq *ifr, unsigned int cmd,
 static int compat_standard_call(struct net_device	*dev,
 				struct iwreq		*iwr,
 				unsigned int		cmd,
+				struct iw_request_info	*info,
 				iw_handler		handler)
 {
 	const struct iw_ioctl_description *descr;
 	struct compat_iw_point *iwp_compat;
-	struct iw_request_info info;
 	struct iw_point iwp;
 	int err;
 
 	descr = standard_ioctl + (cmd - SIOCIWFIRST);
 
 	if (descr->header_type != IW_HEADER_TYPE_POINT)
-		return ioctl_standard_call(dev, iwr, cmd, handler);
+		return ioctl_standard_call(dev, iwr, cmd, info, handler);
 
 	iwp_compat = (struct compat_iw_point *) &iwr->u.data;
 	iwp.pointer = compat_ptr(iwp_compat->pointer);
 	iwp.length = iwp_compat->length;
 	iwp.flags = iwp_compat->flags;
 
-	info.cmd = cmd;
-	info.flags = 0;
-
-	err = ioctl_standard_iw_point(&iwp, cmd, descr, handler, dev, &info);
+	err = ioctl_standard_iw_point(&iwp, cmd, descr, handler, dev, info);
 
 	iwp_compat->pointer = ptr_to_compat(iwp.pointer);
 	iwp_compat->length = iwp.length;
@@ -1147,22 +1140,18 @@ static int compat_standard_call(struct net_device	*dev,
 }
 
 static int compat_private_call(struct net_device *dev, struct iwreq *iwr,
-			       unsigned int cmd, iw_handler handler)
+			       unsigned int cmd, struct iw_request_info *info,
+			       iw_handler handler)
 {
 	const struct iw_priv_args *descr;
-	struct iw_request_info info;
 	int ret, extra_size;
 
 	extra_size = get_priv_descr_and_size(dev, cmd, &descr);
 
-	/* Prepare the call */
-	info.cmd = cmd;
-	info.flags = 0;
-
 	/* Check if we have a pointer to user space data or not. */
 	if (extra_size == 0) {
 		/* No extra arguments. Trivial to handle */
-		ret = handler(dev, &info, &(iwr->u), (char *) &(iwr->u));
+		ret = handler(dev, info, &(iwr->u), (char *) &(iwr->u));
 	} else {
 		struct compat_iw_point *iwp_compat;
 		struct iw_point iwp;
@@ -1173,7 +1162,7 @@ static int compat_private_call(struct net_device *dev, struct iwreq *iwr,
 		iwp.flags = iwp_compat->flags;
 
 		ret = ioctl_private_iw_point(&iwp, cmd, descr,
-					     handler, dev, &info, extra_size);
+					     handler, dev, info, extra_size);
 
 		iwp_compat->pointer = ptr_to_compat(iwp.pointer);
 		iwp_compat->length = iwp.length;
@@ -1191,6 +1180,7 @@ int compat_wext_handle_ioctl(struct net *net, unsigned int cmd,
 			     unsigned long arg)
 {
 	void __user *argp = (void __user *)arg;
+	struct iw_request_info info;
 	struct iwreq iwr;
 	char *colon;
 	int ret;
@@ -1203,7 +1193,10 @@ int compat_wext_handle_ioctl(struct net *net, unsigned int cmd,
 	if (colon)
 		*colon = 0;
 
-	ret = wext_ioctl_dispatch(net, (struct ifreq *) &iwr, cmd,
+	info.cmd = cmd;
+	info.flags = IW_REQUEST_FLAG_COMPAT;
+
+	ret = wext_ioctl_dispatch(net, (struct ifreq *) &iwr, cmd, &info,
 				  compat_standard_call,
 				  compat_private_call);
 
-- 
cgit v1.2.3


From ccc580571cf0799d0460a085a7632b77753f083e Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Mon, 16 Jun 2008 18:50:49 -0700
Subject: wext: Emit event stream entries correctly when compat.

Three major portions to this change:

1) Add IW_EV_COMPAT_LCP_LEN, IW_EV_COMPAT_POINT_OFF,
   and IW_EV_COMPAT_POINT_LEN helper defines.

2) Delete iw_stream_check_add_*(), they are unused.

3) Add iw_request_info argument to iwe_stream_add_*(), and use it to
   size the event and pointer lengths correctly depending upon whether
   IW_REQUEST_FLAG_COMPAT is set or not.

4) The mechanical transformations to the drivers and wireless stack
   bits to get the iw_request_info passed down into the routines
   modified in #3.  Also, explicit references to IW_EV_LCP_LEN are
   replaced with iwe_stream_lcp_len(info).

With a lot of help and bug fixes from Masakazu Mokuno.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ps3_gelic_wireless.c           |  30 +++---
 drivers/net/wireless/airo.c                |  43 ++++++---
 drivers/net/wireless/atmel.c               |  24 +++--
 drivers/net/wireless/hostap/hostap.h       |   3 +-
 drivers/net/wireless/hostap/hostap_ap.c    |  32 +++----
 drivers/net/wireless/hostap/hostap_ioctl.c |  58 +++++------
 drivers/net/wireless/libertas/scan.c       |  36 +++----
 drivers/net/wireless/orinoco.c             |  30 +++---
 drivers/net/wireless/prism54/isl_ioctl.c   |  49 +++++-----
 drivers/net/wireless/rndis_wlan.c          |  32 ++++---
 drivers/net/wireless/wl3501_cs.c           |  10 +-
 drivers/net/wireless/zd1201.c              |  21 ++--
 include/linux/wireless.h                   |  15 +++
 include/net/iw_handler.h                   | 149 ++++++++++-------------------
 net/ieee80211/ieee80211_wx.c               |  48 ++++++----
 net/mac80211/ieee80211_i.h                 |   5 +-
 net/mac80211/mlme.c                        |  66 +++++++------
 net/mac80211/wext.c                        |   2 +-
 18 files changed, 345 insertions(+), 308 deletions(-)

(limited to 'include')

diff --git a/drivers/net/ps3_gelic_wireless.c b/drivers/net/ps3_gelic_wireless.c
index aa963ac1e37b..6b2dee0cf3a9 100644
--- a/drivers/net/ps3_gelic_wireless.c
+++ b/drivers/net/ps3_gelic_wireless.c
@@ -571,6 +571,7 @@ static void gelic_wl_parse_ie(u8 *data, size_t len,
  * independent format
  */
 static char *gelic_wl_translate_scan(struct net_device *netdev,
+				     struct iw_request_info *info,
 				     char *ev,
 				     char *stop,
 				     struct gelic_wl_scan_info *network)
@@ -588,26 +589,26 @@ static char *gelic_wl_translate_scan(struct net_device *netdev,
 	iwe.cmd = SIOCGIWAP;
 	iwe.u.ap_addr.sa_family = ARPHRD_ETHER;
 	memcpy(iwe.u.ap_addr.sa_data, &scan->bssid[2], ETH_ALEN);
-	ev = iwe_stream_add_event(ev, stop, &iwe, IW_EV_ADDR_LEN);
+	ev = iwe_stream_add_event(info, ev, stop, &iwe, IW_EV_ADDR_LEN);
 
 	/* ESSID */
 	iwe.cmd = SIOCGIWESSID;
 	iwe.u.data.flags = 1;
 	iwe.u.data.length = strnlen(scan->essid, 32);
-	ev = iwe_stream_add_point(ev, stop, &iwe, scan->essid);
+	ev = iwe_stream_add_point(info, ev, stop, &iwe, scan->essid);
 
 	/* FREQUENCY */
 	iwe.cmd = SIOCGIWFREQ;
 	iwe.u.freq.m = be16_to_cpu(scan->channel);
 	iwe.u.freq.e = 0; /* table value in MHz */
 	iwe.u.freq.i = 0;
-	ev = iwe_stream_add_event(ev, stop, &iwe, IW_EV_FREQ_LEN);
+	ev = iwe_stream_add_event(info, ev, stop, &iwe, IW_EV_FREQ_LEN);
 
 	/* RATES */
 	iwe.cmd = SIOCGIWRATE;
 	iwe.u.bitrate.fixed = iwe.u.bitrate.disabled = 0;
 	/* to stuff multiple values in one event */
-	tmp = ev + IW_EV_LCP_LEN;
+	tmp = ev + iwe_stream_lcp_len(info);
 	/* put them in ascendant order (older is first) */
 	i = 0;
 	j = 0;
@@ -620,16 +621,16 @@ static char *gelic_wl_translate_scan(struct net_device *netdev,
 		else
 		    rate = scan->rate[i++] & 0x7f;
 		iwe.u.bitrate.value = rate * 500000; /* 500kbps unit */
-		tmp = iwe_stream_add_value(ev, tmp, stop, &iwe,
+		tmp = iwe_stream_add_value(info, ev, tmp, stop, &iwe,
 					   IW_EV_PARAM_LEN);
 	}
 	while (j < network->rate_ext_len) {
 		iwe.u.bitrate.value = (scan->ext_rate[j++] & 0x7f) * 500000;
-		tmp = iwe_stream_add_value(ev, tmp, stop, &iwe,
+		tmp = iwe_stream_add_value(info, ev, tmp, stop, &iwe,
 					   IW_EV_PARAM_LEN);
 	}
 	/* Check if we added any rate */
-	if (IW_EV_LCP_LEN < (tmp - ev))
+	if (iwe_stream_lcp_len(info) < (tmp - ev))
 		ev = tmp;
 
 	/* ENCODE */
@@ -639,7 +640,7 @@ static char *gelic_wl_translate_scan(struct net_device *netdev,
 	else
 		iwe.u.data.flags = IW_ENCODE_DISABLED;
 	iwe.u.data.length = 0;
-	ev = iwe_stream_add_point(ev, stop, &iwe, scan->essid);
+	ev = iwe_stream_add_point(info, ev, stop, &iwe, scan->essid);
 
 	/* MODE */
 	iwe.cmd = SIOCGIWMODE;
@@ -649,7 +650,7 @@ static char *gelic_wl_translate_scan(struct net_device *netdev,
 			iwe.u.mode = IW_MODE_MASTER;
 		else
 			iwe.u.mode = IW_MODE_ADHOC;
-		ev = iwe_stream_add_event(ev, stop, &iwe, IW_EV_UINT_LEN);
+		ev = iwe_stream_add_event(info, ev, stop, &iwe, IW_EV_UINT_LEN);
 	}
 
 	/* QUAL */
@@ -659,7 +660,7 @@ static char *gelic_wl_translate_scan(struct net_device *netdev,
 	iwe.u.qual.level = be16_to_cpu(scan->rssi);
 	iwe.u.qual.qual = be16_to_cpu(scan->rssi);
 	iwe.u.qual.noise = 0;
-	ev  = iwe_stream_add_event(ev, stop, &iwe, IW_EV_QUAL_LEN);
+	ev  = iwe_stream_add_event(info, ev, stop, &iwe, IW_EV_QUAL_LEN);
 
 	/* RSN */
 	memset(&iwe, 0, sizeof(iwe));
@@ -669,7 +670,7 @@ static char *gelic_wl_translate_scan(struct net_device *netdev,
 		if (len) {
 			iwe.cmd = IWEVGENIE;
 			iwe.u.data.length = len;
-			ev = iwe_stream_add_point(ev, stop, &iwe, buf);
+			ev = iwe_stream_add_point(info, ev, stop, &iwe, buf);
 		}
 	} else {
 		/* this scan info has IE data */
@@ -684,7 +685,7 @@ static char *gelic_wl_translate_scan(struct net_device *netdev,
 			memcpy(buf, ie_info.wpa.data, ie_info.wpa.len);
 			iwe.cmd = IWEVGENIE;
 			iwe.u.data.length = ie_info.wpa.len;
-			ev = iwe_stream_add_point(ev, stop, &iwe, buf);
+			ev = iwe_stream_add_point(info, ev, stop, &iwe, buf);
 		}
 
 		if (ie_info.rsn.len && (ie_info.rsn.len <= sizeof(buf))) {
@@ -692,7 +693,7 @@ static char *gelic_wl_translate_scan(struct net_device *netdev,
 			memcpy(buf, ie_info.rsn.data, ie_info.rsn.len);
 			iwe.cmd = IWEVGENIE;
 			iwe.u.data.length = ie_info.rsn.len;
-			ev = iwe_stream_add_point(ev, stop, &iwe, buf);
+			ev = iwe_stream_add_point(info, ev, stop, &iwe, buf);
 		}
 	}
 
@@ -737,7 +738,8 @@ static int gelic_wl_get_scan(struct net_device *netdev,
 		if (wl->scan_age == 0 ||
 		    time_after(scan_info->last_scanned + wl->scan_age,
 			       this_time))
-			ev = gelic_wl_translate_scan(netdev, ev, stop,
+			ev = gelic_wl_translate_scan(netdev, info,
+						     ev, stop,
 						     scan_info);
 		else
 			pr_debug("%s:entry too old\n", __func__);
diff --git a/drivers/net/wireless/airo.c b/drivers/net/wireless/airo.c
index e30f8b79ea89..73d66a80c4a3 100644
--- a/drivers/net/wireless/airo.c
+++ b/drivers/net/wireless/airo.c
@@ -7156,6 +7156,7 @@ out:
  * format that the Wireless Tools will understand - Jean II
  */
 static inline char *airo_translate_scan(struct net_device *dev,
+					struct iw_request_info *info,
 					char *current_ev,
 					char *end_buf,
 					BSSListRid *bss)
@@ -7172,7 +7173,8 @@ static inline char *airo_translate_scan(struct net_device *dev,
 	iwe.cmd = SIOCGIWAP;
 	iwe.u.ap_addr.sa_family = ARPHRD_ETHER;
 	memcpy(iwe.u.ap_addr.sa_data, bss->bssid, ETH_ALEN);
-	current_ev = iwe_stream_add_event(current_ev, end_buf, &iwe, IW_EV_ADDR_LEN);
+	current_ev = iwe_stream_add_event(info, current_ev, end_buf,
+					  &iwe, IW_EV_ADDR_LEN);
 
 	/* Other entries will be displayed in the order we give them */
 
@@ -7182,7 +7184,8 @@ static inline char *airo_translate_scan(struct net_device *dev,
 		iwe.u.data.length = 32;
 	iwe.cmd = SIOCGIWESSID;
 	iwe.u.data.flags = 1;
-	current_ev = iwe_stream_add_point(current_ev, end_buf, &iwe, bss->ssid);
+	current_ev = iwe_stream_add_point(info, current_ev, end_buf,
+					  &iwe, bss->ssid);
 
 	/* Add mode */
 	iwe.cmd = SIOCGIWMODE;
@@ -7192,7 +7195,8 @@ static inline char *airo_translate_scan(struct net_device *dev,
 			iwe.u.mode = IW_MODE_MASTER;
 		else
 			iwe.u.mode = IW_MODE_ADHOC;
-		current_ev = iwe_stream_add_event(current_ev, end_buf, &iwe, IW_EV_UINT_LEN);
+		current_ev = iwe_stream_add_event(info, current_ev, end_buf,
+						  &iwe, IW_EV_UINT_LEN);
 	}
 
 	/* Add frequency */
@@ -7203,7 +7207,8 @@ static inline char *airo_translate_scan(struct net_device *dev,
 	 */
 	iwe.u.freq.m = frequency_list[iwe.u.freq.m - 1] * 100000;
 	iwe.u.freq.e = 1;
-	current_ev = iwe_stream_add_event(current_ev, end_buf, &iwe, IW_EV_FREQ_LEN);
+	current_ev = iwe_stream_add_event(info, current_ev, end_buf,
+					  &iwe, IW_EV_FREQ_LEN);
 
 	dBm = le16_to_cpu(bss->dBm);
 
@@ -7223,7 +7228,8 @@ static inline char *airo_translate_scan(struct net_device *dev,
 				| IW_QUAL_DBM;
 	}
 	iwe.u.qual.noise = ai->wstats.qual.noise;
-	current_ev = iwe_stream_add_event(current_ev, end_buf, &iwe, IW_EV_QUAL_LEN);
+	current_ev = iwe_stream_add_event(info, current_ev, end_buf,
+					  &iwe, IW_EV_QUAL_LEN);
 
 	/* Add encryption capability */
 	iwe.cmd = SIOCGIWENCODE;
@@ -7232,11 +7238,12 @@ static inline char *airo_translate_scan(struct net_device *dev,
 	else
 		iwe.u.data.flags = IW_ENCODE_DISABLED;
 	iwe.u.data.length = 0;
-	current_ev = iwe_stream_add_point(current_ev, end_buf, &iwe, bss->ssid);
+	current_ev = iwe_stream_add_point(info, current_ev, end_buf,
+					  &iwe, bss->ssid);
 
 	/* Rate : stuffing multiple values in a single event require a bit
 	 * more of magic - Jean II */
-	current_val = current_ev + IW_EV_LCP_LEN;
+	current_val = current_ev + iwe_stream_lcp_len(info);
 
 	iwe.cmd = SIOCGIWRATE;
 	/* Those two flags are ignored... */
@@ -7249,10 +7256,12 @@ static inline char *airo_translate_scan(struct net_device *dev,
 		/* Bit rate given in 500 kb/s units (+ 0x80) */
 		iwe.u.bitrate.value = ((bss->rates[i] & 0x7f) * 500000);
 		/* Add new value to event */
-		current_val = iwe_stream_add_value(current_ev, current_val, end_buf, &iwe, IW_EV_PARAM_LEN);
+		current_val = iwe_stream_add_value(info, current_ev,
+						   current_val, end_buf,
+						   &iwe, IW_EV_PARAM_LEN);
 	}
 	/* Check if we added any event */
-	if((current_val - current_ev) > IW_EV_LCP_LEN)
+	if ((current_val - current_ev) > iwe_stream_lcp_len(info))
 		current_ev = current_val;
 
 	/* Beacon interval */
@@ -7261,7 +7270,8 @@ static inline char *airo_translate_scan(struct net_device *dev,
 		iwe.cmd = IWEVCUSTOM;
 		sprintf(buf, "bcn_int=%d", bss->beaconInterval);
 		iwe.u.data.length = strlen(buf);
-		current_ev = iwe_stream_add_point(current_ev, end_buf, &iwe, buf);
+		current_ev = iwe_stream_add_point(info, current_ev, end_buf,
+						  &iwe, buf);
 		kfree(buf);
 	}
 
@@ -7295,8 +7305,10 @@ static inline char *airo_translate_scan(struct net_device *dev,
 					iwe.cmd = IWEVGENIE;
 					iwe.u.data.length = min(info_element->len + 2,
 								  MAX_WPA_IE_LEN);
-					current_ev = iwe_stream_add_point(current_ev, end_buf,
-							&iwe, (char *) info_element);
+					current_ev = iwe_stream_add_point(
+							info, current_ev,
+							end_buf, &iwe,
+							(char *) info_element);
 				}
 				break;
 
@@ -7304,8 +7316,9 @@ static inline char *airo_translate_scan(struct net_device *dev,
 				iwe.cmd = IWEVGENIE;
 				iwe.u.data.length = min(info_element->len + 2,
 							  MAX_WPA_IE_LEN);
-				current_ev = iwe_stream_add_point(current_ev, end_buf,
-						&iwe, (char *) info_element);
+				current_ev = iwe_stream_add_point(
+					info, current_ev, end_buf,
+					&iwe, (char *) info_element);
 				break;
 
 			default:
@@ -7344,7 +7357,7 @@ static int airo_get_scan(struct net_device *dev,
 
 	list_for_each_entry (net, &ai->network_list, list) {
 		/* Translate to WE format this entry */
-		current_ev = airo_translate_scan(dev, current_ev,
+		current_ev = airo_translate_scan(dev, info, current_ev,
 						 extra + dwrq->length,
 						 &net->bss);
 
diff --git a/drivers/net/wireless/atmel.c b/drivers/net/wireless/atmel.c
index 7bb2646ae0ef..28b6ff3eaa37 100644
--- a/drivers/net/wireless/atmel.c
+++ b/drivers/net/wireless/atmel.c
@@ -2310,30 +2310,40 @@ static int atmel_get_scan(struct net_device *dev,
 		iwe.cmd = SIOCGIWAP;
 		iwe.u.ap_addr.sa_family = ARPHRD_ETHER;
 		memcpy(iwe.u.ap_addr.sa_data, priv->BSSinfo[i].BSSID, 6);
-		current_ev = iwe_stream_add_event(current_ev, extra + IW_SCAN_MAX_DATA, &iwe, IW_EV_ADDR_LEN);
+		current_ev = iwe_stream_add_event(info, current_ev,
+						  extra + IW_SCAN_MAX_DATA,
+						  &iwe, IW_EV_ADDR_LEN);
 
 		iwe.u.data.length =  priv->BSSinfo[i].SSIDsize;
 		if (iwe.u.data.length > 32)
 			iwe.u.data.length = 32;
 		iwe.cmd = SIOCGIWESSID;
 		iwe.u.data.flags = 1;
-		current_ev = iwe_stream_add_point(current_ev, extra + IW_SCAN_MAX_DATA, &iwe, priv->BSSinfo[i].SSID);
+		current_ev = iwe_stream_add_point(info, current_ev,
+						  extra + IW_SCAN_MAX_DATA,
+						  &iwe, priv->BSSinfo[i].SSID);
 
 		iwe.cmd = SIOCGIWMODE;
 		iwe.u.mode = priv->BSSinfo[i].BSStype;
-		current_ev = iwe_stream_add_event(current_ev, extra + IW_SCAN_MAX_DATA, &iwe, IW_EV_UINT_LEN);
+		current_ev = iwe_stream_add_event(info, current_ev,
+						  extra + IW_SCAN_MAX_DATA,
+						  &iwe, IW_EV_UINT_LEN);
 
 		iwe.cmd = SIOCGIWFREQ;
 		iwe.u.freq.m = priv->BSSinfo[i].channel;
 		iwe.u.freq.e = 0;
-		current_ev = iwe_stream_add_event(current_ev, extra + IW_SCAN_MAX_DATA, &iwe, IW_EV_FREQ_LEN);
+		current_ev = iwe_stream_add_event(info, current_ev,
+						  extra + IW_SCAN_MAX_DATA,
+						  &iwe, IW_EV_FREQ_LEN);
 
 		/* Add quality statistics */
 		iwe.cmd = IWEVQUAL;
 		iwe.u.qual.level = priv->BSSinfo[i].RSSI;
 		iwe.u.qual.qual  = iwe.u.qual.level;
 		/* iwe.u.qual.noise  = SOMETHING */
-		current_ev = iwe_stream_add_event(current_ev, extra + IW_SCAN_MAX_DATA , &iwe, IW_EV_QUAL_LEN);
+		current_ev = iwe_stream_add_event(info, current_ev,
+						  extra + IW_SCAN_MAX_DATA,
+						  &iwe, IW_EV_QUAL_LEN);
 
 
 		iwe.cmd = SIOCGIWENCODE;
@@ -2342,7 +2352,9 @@ static int atmel_get_scan(struct net_device *dev,
 		else
 			iwe.u.data.flags = IW_ENCODE_DISABLED;
 		iwe.u.data.length = 0;
-		current_ev = iwe_stream_add_point(current_ev, extra + IW_SCAN_MAX_DATA, &iwe, NULL);
+		current_ev = iwe_stream_add_point(info, current_ev,
+						  extra + IW_SCAN_MAX_DATA,
+						  &iwe, NULL);
 	}
 
 	/* Length of data */
diff --git a/drivers/net/wireless/hostap/hostap.h b/drivers/net/wireless/hostap/hostap.h
index 547ba84dc797..3a386a636cca 100644
--- a/drivers/net/wireless/hostap/hostap.h
+++ b/drivers/net/wireless/hostap/hostap.h
@@ -67,7 +67,8 @@ void * ap_crypt_get_ptrs(struct ap_data *ap, u8 *addr, int permanent,
 int prism2_ap_get_sta_qual(local_info_t *local, struct sockaddr addr[],
 			   struct iw_quality qual[], int buf_size,
 			   int aplist);
-int prism2_ap_translate_scan(struct net_device *dev, char *buffer);
+int prism2_ap_translate_scan(struct net_device *dev,
+			     struct iw_request_info *info, char *buffer);
 int prism2_hostapd(struct ap_data *ap, struct prism2_hostapd_param *param);
 
 
diff --git a/drivers/net/wireless/hostap/hostap_ap.c b/drivers/net/wireless/hostap/hostap_ap.c
index 0acd9589c48c..06b23df8f69b 100644
--- a/drivers/net/wireless/hostap/hostap_ap.c
+++ b/drivers/net/wireless/hostap/hostap_ap.c
@@ -2420,7 +2420,8 @@ int prism2_ap_get_sta_qual(local_info_t *local, struct sockaddr addr[],
 
 /* Translate our list of Access Points & Stations to a card independant
  * format that the Wireless Tools will understand - Jean II */
-int prism2_ap_translate_scan(struct net_device *dev, char *buffer)
+int prism2_ap_translate_scan(struct net_device *dev,
+			     struct iw_request_info *info, char *buffer)
 {
 	struct hostap_interface *iface;
 	local_info_t *local;
@@ -2449,8 +2450,8 @@ int prism2_ap_translate_scan(struct net_device *dev, char *buffer)
 		iwe.u.ap_addr.sa_family = ARPHRD_ETHER;
 		memcpy(iwe.u.ap_addr.sa_data, sta->addr, ETH_ALEN);
 		iwe.len = IW_EV_ADDR_LEN;
-		current_ev = iwe_stream_add_event(current_ev, end_buf, &iwe,
-						  IW_EV_ADDR_LEN);
+		current_ev = iwe_stream_add_event(info, current_ev, end_buf,
+						  &iwe, IW_EV_ADDR_LEN);
 
 		/* Use the mode to indicate if it's a station or
 		 * an Access Point */
@@ -2461,8 +2462,8 @@ int prism2_ap_translate_scan(struct net_device *dev, char *buffer)
 		else
 			iwe.u.mode = IW_MODE_INFRA;
 		iwe.len = IW_EV_UINT_LEN;
-		current_ev = iwe_stream_add_event(current_ev, end_buf, &iwe,
-						  IW_EV_UINT_LEN);
+		current_ev = iwe_stream_add_event(info, current_ev, end_buf,
+						  &iwe, IW_EV_UINT_LEN);
 
 		/* Some quality */
 		memset(&iwe, 0, sizeof(iwe));
@@ -2477,8 +2478,8 @@ int prism2_ap_translate_scan(struct net_device *dev, char *buffer)
 		iwe.u.qual.noise = HFA384X_LEVEL_TO_dBm(sta->last_rx_silence);
 		iwe.u.qual.updated = sta->last_rx_updated;
 		iwe.len = IW_EV_QUAL_LEN;
-		current_ev = iwe_stream_add_event(current_ev, end_buf, &iwe,
-						  IW_EV_QUAL_LEN);
+		current_ev = iwe_stream_add_event(info, current_ev, end_buf,
+						  &iwe, IW_EV_QUAL_LEN);
 
 #ifndef PRISM2_NO_KERNEL_IEEE80211_MGMT
 		if (sta->ap) {
@@ -2486,8 +2487,8 @@ int prism2_ap_translate_scan(struct net_device *dev, char *buffer)
 			iwe.cmd = SIOCGIWESSID;
 			iwe.u.data.length = sta->u.ap.ssid_len;
 			iwe.u.data.flags = 1;
-			current_ev = iwe_stream_add_point(current_ev, end_buf,
-							  &iwe,
+			current_ev = iwe_stream_add_point(info, current_ev,
+							  end_buf, &iwe,
 							  sta->u.ap.ssid);
 
 			memset(&iwe, 0, sizeof(iwe));
@@ -2497,10 +2498,9 @@ int prism2_ap_translate_scan(struct net_device *dev, char *buffer)
 					IW_ENCODE_ENABLED | IW_ENCODE_NOKEY;
 			else
 				iwe.u.data.flags = IW_ENCODE_DISABLED;
-			current_ev = iwe_stream_add_point(current_ev, end_buf,
-							  &iwe,
-							  sta->u.ap.ssid
-							  /* 0 byte memcpy */);
+			current_ev = iwe_stream_add_point(info, current_ev,
+							  end_buf, &iwe,
+							  sta->u.ap.ssid);
 
 			if (sta->u.ap.channel > 0 &&
 			    sta->u.ap.channel <= FREQ_COUNT) {
@@ -2510,7 +2510,7 @@ int prism2_ap_translate_scan(struct net_device *dev, char *buffer)
 					* 100000;
 				iwe.u.freq.e = 1;
 				current_ev = iwe_stream_add_event(
-					current_ev, end_buf, &iwe,
+					info, current_ev, end_buf, &iwe,
 					IW_EV_FREQ_LEN);
 			}
 
@@ -2519,8 +2519,8 @@ int prism2_ap_translate_scan(struct net_device *dev, char *buffer)
 			sprintf(buf, "beacon_interval=%d",
 				sta->listen_interval);
 			iwe.u.data.length = strlen(buf);
-			current_ev = iwe_stream_add_point(current_ev, end_buf,
-							  &iwe, buf);
+			current_ev = iwe_stream_add_point(info, current_ev,
+							  end_buf, &iwe, buf);
 		}
 #endif /* PRISM2_NO_KERNEL_IEEE80211_MGMT */
 
diff --git a/drivers/net/wireless/hostap/hostap_ioctl.c b/drivers/net/wireless/hostap/hostap_ioctl.c
index 0ca0bfeb0ada..ed52d98317cd 100644
--- a/drivers/net/wireless/hostap/hostap_ioctl.c
+++ b/drivers/net/wireless/hostap/hostap_ioctl.c
@@ -1793,6 +1793,7 @@ static int prism2_ioctl_siwscan(struct net_device *dev,
 
 #ifndef PRISM2_NO_STATION_MODES
 static char * __prism2_translate_scan(local_info_t *local,
+				      struct iw_request_info *info,
 				      struct hfa384x_hostscan_result *scan,
 				      struct hostap_bss_info *bss,
 				      char *current_ev, char *end_buf)
@@ -1823,7 +1824,7 @@ static char * __prism2_translate_scan(local_info_t *local,
 	iwe.cmd = SIOCGIWAP;
 	iwe.u.ap_addr.sa_family = ARPHRD_ETHER;
 	memcpy(iwe.u.ap_addr.sa_data, bssid, ETH_ALEN);
-	current_ev = iwe_stream_add_event(current_ev, end_buf, &iwe,
+	current_ev = iwe_stream_add_event(info, current_ev, end_buf, &iwe,
 					  IW_EV_ADDR_LEN);
 
 	/* Other entries will be displayed in the order we give them */
@@ -1832,7 +1833,8 @@ static char * __prism2_translate_scan(local_info_t *local,
 	iwe.cmd = SIOCGIWESSID;
 	iwe.u.data.length = ssid_len;
 	iwe.u.data.flags = 1;
-	current_ev = iwe_stream_add_point(current_ev, end_buf, &iwe, ssid);
+	current_ev = iwe_stream_add_point(info, current_ev, end_buf,
+					  &iwe, ssid);
 
 	memset(&iwe, 0, sizeof(iwe));
 	iwe.cmd = SIOCGIWMODE;
@@ -1847,8 +1849,8 @@ static char * __prism2_translate_scan(local_info_t *local,
 			iwe.u.mode = IW_MODE_MASTER;
 		else
 			iwe.u.mode = IW_MODE_ADHOC;
-		current_ev = iwe_stream_add_event(current_ev, end_buf, &iwe,
-						  IW_EV_UINT_LEN);
+		current_ev = iwe_stream_add_event(info, current_ev, end_buf,
+						  &iwe, IW_EV_UINT_LEN);
 	}
 
 	memset(&iwe, 0, sizeof(iwe));
@@ -1864,8 +1866,8 @@ static char * __prism2_translate_scan(local_info_t *local,
 	if (chan > 0) {
 		iwe.u.freq.m = freq_list[chan - 1] * 100000;
 		iwe.u.freq.e = 1;
-		current_ev = iwe_stream_add_event(current_ev, end_buf, &iwe,
-						  IW_EV_FREQ_LEN);
+		current_ev = iwe_stream_add_event(info, current_ev, end_buf,
+						  &iwe, IW_EV_FREQ_LEN);
 	}
 
 	if (scan) {
@@ -1884,8 +1886,8 @@ static char * __prism2_translate_scan(local_info_t *local,
 			| IW_QUAL_NOISE_UPDATED
 			| IW_QUAL_QUAL_INVALID
 			| IW_QUAL_DBM;
-		current_ev = iwe_stream_add_event(current_ev, end_buf, &iwe,
-						  IW_EV_QUAL_LEN);
+		current_ev = iwe_stream_add_event(info, current_ev, end_buf,
+						  &iwe, IW_EV_QUAL_LEN);
 	}
 
 	memset(&iwe, 0, sizeof(iwe));
@@ -1895,13 +1897,13 @@ static char * __prism2_translate_scan(local_info_t *local,
 	else
 		iwe.u.data.flags = IW_ENCODE_DISABLED;
 	iwe.u.data.length = 0;
-	current_ev = iwe_stream_add_point(current_ev, end_buf, &iwe, "");
+	current_ev = iwe_stream_add_point(info, current_ev, end_buf, &iwe, "");
 
 	/* TODO: add SuppRates into BSS table */
 	if (scan) {
 		memset(&iwe, 0, sizeof(iwe));
 		iwe.cmd = SIOCGIWRATE;
-		current_val = current_ev + IW_EV_LCP_LEN;
+		current_val = current_ev + iwe_stream_lcp_len(info);
 		pos = scan->sup_rates;
 		for (i = 0; i < sizeof(scan->sup_rates); i++) {
 			if (pos[i] == 0)
@@ -1909,11 +1911,11 @@ static char * __prism2_translate_scan(local_info_t *local,
 			/* Bit rate given in 500 kb/s units (+ 0x80) */
 			iwe.u.bitrate.value = ((pos[i] & 0x7f) * 500000);
 			current_val = iwe_stream_add_value(
-				current_ev, current_val, end_buf, &iwe,
+				info, current_ev, current_val, end_buf, &iwe,
 				IW_EV_PARAM_LEN);
 		}
 		/* Check if we added any event */
-		if ((current_val - current_ev) > IW_EV_LCP_LEN)
+		if ((current_val - current_ev) > iwe_stream_lcp_len(info))
 			current_ev = current_val;
 	}
 
@@ -1924,15 +1926,15 @@ static char * __prism2_translate_scan(local_info_t *local,
 		iwe.cmd = IWEVCUSTOM;
 		sprintf(buf, "bcn_int=%d", le16_to_cpu(scan->beacon_interval));
 		iwe.u.data.length = strlen(buf);
-		current_ev = iwe_stream_add_point(current_ev, end_buf, &iwe,
-						  buf);
+		current_ev = iwe_stream_add_point(info, current_ev, end_buf,
+						  &iwe, buf);
 
 		memset(&iwe, 0, sizeof(iwe));
 		iwe.cmd = IWEVCUSTOM;
 		sprintf(buf, "resp_rate=%d", le16_to_cpu(scan->rate));
 		iwe.u.data.length = strlen(buf);
-		current_ev = iwe_stream_add_point(current_ev, end_buf, &iwe,
-						  buf);
+		current_ev = iwe_stream_add_point(info, current_ev, end_buf,
+						  &iwe, buf);
 
 		if (local->last_scan_type == PRISM2_HOSTSCAN &&
 		    (capabilities & WLAN_CAPABILITY_IBSS)) {
@@ -1940,8 +1942,8 @@ static char * __prism2_translate_scan(local_info_t *local,
 			iwe.cmd = IWEVCUSTOM;
 			sprintf(buf, "atim=%d", le16_to_cpu(scan->atim));
 			iwe.u.data.length = strlen(buf);
-			current_ev = iwe_stream_add_point(current_ev, end_buf,
-							  &iwe, buf);
+			current_ev = iwe_stream_add_point(info, current_ev,
+							  end_buf, &iwe, buf);
 		}
 	}
 	kfree(buf);
@@ -1950,16 +1952,16 @@ static char * __prism2_translate_scan(local_info_t *local,
 		memset(&iwe, 0, sizeof(iwe));
 		iwe.cmd = IWEVGENIE;
 		iwe.u.data.length = bss->wpa_ie_len;
-		current_ev = iwe_stream_add_point(
-			current_ev, end_buf, &iwe, bss->wpa_ie);
+		current_ev = iwe_stream_add_point(info, current_ev, end_buf,
+						  &iwe, bss->wpa_ie);
 	}
 
 	if (bss && bss->rsn_ie_len > 0 && bss->rsn_ie_len <= MAX_WPA_IE_LEN) {
 		memset(&iwe, 0, sizeof(iwe));
 		iwe.cmd = IWEVGENIE;
 		iwe.u.data.length = bss->rsn_ie_len;
-		current_ev = iwe_stream_add_point(
-			current_ev, end_buf, &iwe, bss->rsn_ie);
+		current_ev = iwe_stream_add_point(info, current_ev, end_buf,
+						  &iwe, bss->rsn_ie);
 	}
 
 	return current_ev;
@@ -1969,6 +1971,7 @@ static char * __prism2_translate_scan(local_info_t *local,
 /* Translate scan data returned from the card to a card independant
  * format that the Wireless Tools will understand - Jean II */
 static inline int prism2_translate_scan(local_info_t *local,
+					struct iw_request_info *info,
 					char *buffer, int buflen)
 {
 	struct hfa384x_hostscan_result *scan;
@@ -1999,13 +2002,14 @@ static inline int prism2_translate_scan(local_info_t *local,
 			if (memcmp(bss->bssid, scan->bssid, ETH_ALEN) == 0) {
 				bss->included = 1;
 				current_ev = __prism2_translate_scan(
-					local, scan, bss, current_ev, end_buf);
+					local, info, scan, bss, current_ev,
+					end_buf);
 				found++;
 			}
 		}
 		if (!found) {
 			current_ev = __prism2_translate_scan(
-				local, scan, NULL, current_ev, end_buf);
+				local, info, scan, NULL, current_ev, end_buf);
 		}
 		/* Check if there is space for one more entry */
 		if ((end_buf - current_ev) <= IW_EV_ADDR_LEN) {
@@ -2023,7 +2027,7 @@ static inline int prism2_translate_scan(local_info_t *local,
 		bss = list_entry(ptr, struct hostap_bss_info, list);
 		if (bss->included)
 			continue;
-		current_ev = __prism2_translate_scan(local, NULL, bss,
+		current_ev = __prism2_translate_scan(local, info, NULL, bss,
 						     current_ev, end_buf);
 		/* Check if there is space for one more entry */
 		if ((end_buf - current_ev) <= IW_EV_ADDR_LEN) {
@@ -2070,7 +2074,7 @@ static inline int prism2_ioctl_giwscan_sta(struct net_device *dev,
 	}
 	local->scan_timestamp = 0;
 
-	res = prism2_translate_scan(local, extra, data->length);
+	res = prism2_translate_scan(local, info, extra, data->length);
 
 	if (res >= 0) {
 		data->length = res;
@@ -2103,7 +2107,7 @@ static int prism2_ioctl_giwscan(struct net_device *dev,
 		 * Jean II */
 
 		/* Translate to WE format */
-		res = prism2_ap_translate_scan(dev, extra);
+		res = prism2_ap_translate_scan(dev, info, extra);
 		if (res >= 0) {
 			printk(KERN_DEBUG "Scan result translation succeeded "
 			       "(length=%d)\n", res);
diff --git a/drivers/net/wireless/libertas/scan.c b/drivers/net/wireless/libertas/scan.c
index d448c9702a0f..343ed38f772d 100644
--- a/drivers/net/wireless/libertas/scan.c
+++ b/drivers/net/wireless/libertas/scan.c
@@ -776,8 +776,9 @@ out:
 #define MAX_CUSTOM_LEN 64
 
 static inline char *lbs_translate_scan(struct lbs_private *priv,
-				       char *start, char *stop,
-				       struct bss_descriptor *bss)
+					    struct iw_request_info *info,
+					    char *start, char *stop,
+					    struct bss_descriptor *bss)
 {
 	struct chan_freq_power *cfp;
 	char *current_val;	/* For rates */
@@ -801,24 +802,24 @@ static inline char *lbs_translate_scan(struct lbs_private *priv,
 	iwe.cmd = SIOCGIWAP;
 	iwe.u.ap_addr.sa_family = ARPHRD_ETHER;
 	memcpy(iwe.u.ap_addr.sa_data, &bss->bssid, ETH_ALEN);
-	start = iwe_stream_add_event(start, stop, &iwe, IW_EV_ADDR_LEN);
+	start = iwe_stream_add_event(info, start, stop, &iwe, IW_EV_ADDR_LEN);
 
 	/* SSID */
 	iwe.cmd = SIOCGIWESSID;
 	iwe.u.data.flags = 1;
 	iwe.u.data.length = min((uint32_t) bss->ssid_len, (uint32_t) IW_ESSID_MAX_SIZE);
-	start = iwe_stream_add_point(start, stop, &iwe, bss->ssid);
+	start = iwe_stream_add_point(info, start, stop, &iwe, bss->ssid);
 
 	/* Mode */
 	iwe.cmd = SIOCGIWMODE;
 	iwe.u.mode = bss->mode;
-	start = iwe_stream_add_event(start, stop, &iwe, IW_EV_UINT_LEN);
+	start = iwe_stream_add_event(info, start, stop, &iwe, IW_EV_UINT_LEN);
 
 	/* Frequency */
 	iwe.cmd = SIOCGIWFREQ;
 	iwe.u.freq.m = (long)cfp->freq * 100000;
 	iwe.u.freq.e = 1;
-	start = iwe_stream_add_event(start, stop, &iwe, IW_EV_FREQ_LEN);
+	start = iwe_stream_add_event(info, start, stop, &iwe, IW_EV_FREQ_LEN);
 
 	/* Add quality statistics */
 	iwe.cmd = IWEVQUAL;
@@ -852,7 +853,7 @@ static inline char *lbs_translate_scan(struct lbs_private *priv,
 		nf = priv->NF[TYPE_RXPD][TYPE_AVG] / AVG_SCALE;
 		iwe.u.qual.level = CAL_RSSI(snr, nf);
 	}
-	start = iwe_stream_add_event(start, stop, &iwe, IW_EV_QUAL_LEN);
+	start = iwe_stream_add_event(info, start, stop, &iwe, IW_EV_QUAL_LEN);
 
 	/* Add encryption capability */
 	iwe.cmd = SIOCGIWENCODE;
@@ -862,9 +863,9 @@ static inline char *lbs_translate_scan(struct lbs_private *priv,
 		iwe.u.data.flags = IW_ENCODE_DISABLED;
 	}
 	iwe.u.data.length = 0;
-	start = iwe_stream_add_point(start, stop, &iwe, bss->ssid);
+	start = iwe_stream_add_point(info, start, stop, &iwe, bss->ssid);
 
-	current_val = start + IW_EV_LCP_LEN;
+	current_val = start + iwe_stream_lcp_len(info);
 
 	iwe.cmd = SIOCGIWRATE;
 	iwe.u.bitrate.fixed = 0;
@@ -874,19 +875,19 @@ static inline char *lbs_translate_scan(struct lbs_private *priv,
 	for (j = 0; bss->rates[j] && (j < sizeof(bss->rates)); j++) {
 		/* Bit rate given in 500 kb/s units */
 		iwe.u.bitrate.value = bss->rates[j] * 500000;
-		current_val = iwe_stream_add_value(start, current_val,
-					 stop, &iwe, IW_EV_PARAM_LEN);
+		current_val = iwe_stream_add_value(info, start, current_val,
+						   stop, &iwe, IW_EV_PARAM_LEN);
 	}
 	if ((bss->mode == IW_MODE_ADHOC) && priv->adhoccreate
 	    && !lbs_ssid_cmp(priv->curbssparams.ssid,
 			     priv->curbssparams.ssid_len,
 			     bss->ssid, bss->ssid_len)) {
 		iwe.u.bitrate.value = 22 * 500000;
-		current_val = iwe_stream_add_value(start, current_val,
+		current_val = iwe_stream_add_value(info, start, current_val,
 						   stop, &iwe, IW_EV_PARAM_LEN);
 	}
 	/* Check if we added any event */
-	if((current_val - start) > IW_EV_LCP_LEN)
+	if ((current_val - start) > iwe_stream_lcp_len(info))
 		start = current_val;
 
 	memset(&iwe, 0, sizeof(iwe));
@@ -895,7 +896,7 @@ static inline char *lbs_translate_scan(struct lbs_private *priv,
 		memcpy(buf, bss->wpa_ie, bss->wpa_ie_len);
 		iwe.cmd = IWEVGENIE;
 		iwe.u.data.length = bss->wpa_ie_len;
-		start = iwe_stream_add_point(start, stop, &iwe, buf);
+		start = iwe_stream_add_point(info, start, stop, &iwe, buf);
 	}
 
 	memset(&iwe, 0, sizeof(iwe));
@@ -904,7 +905,7 @@ static inline char *lbs_translate_scan(struct lbs_private *priv,
 		memcpy(buf, bss->rsn_ie, bss->rsn_ie_len);
 		iwe.cmd = IWEVGENIE;
 		iwe.u.data.length = bss->rsn_ie_len;
-		start = iwe_stream_add_point(start, stop, &iwe, buf);
+		start = iwe_stream_add_point(info, start, stop, &iwe, buf);
 	}
 
 	if (bss->mesh) {
@@ -915,7 +916,8 @@ static inline char *lbs_translate_scan(struct lbs_private *priv,
 		p += snprintf(p, MAX_CUSTOM_LEN, "mesh-type: olpc");
 		iwe.u.data.length = p - custom;
 		if (iwe.u.data.length)
-			start = iwe_stream_add_point(start, stop, &iwe, custom);
+			start = iwe_stream_add_point(info, start, stop,
+						     &iwe, custom);
 	}
 
 out:
@@ -1036,7 +1038,7 @@ int lbs_get_scan(struct net_device *dev, struct iw_request_info *info,
 		}
 
 		/* Translate to WE format this entry */
-		next_ev = lbs_translate_scan(priv, ev, stop, iter_bss);
+		next_ev = lbs_translate_scan(priv, info, ev, stop, iter_bss);
 		if (next_ev == NULL)
 			continue;
 		ev = next_ev;
diff --git a/drivers/net/wireless/orinoco.c b/drivers/net/wireless/orinoco.c
index 6d13a0d15a0c..b047306bf386 100644
--- a/drivers/net/wireless/orinoco.c
+++ b/drivers/net/wireless/orinoco.c
@@ -4046,6 +4046,7 @@ static int orinoco_ioctl_setscan(struct net_device *dev,
  * format that the Wireless Tools will understand - Jean II
  * Return message length or -errno for fatal errors */
 static inline char *orinoco_translate_scan(struct net_device *dev,
+					   struct iw_request_info *info,
 					   char *current_ev,
 					   char *end_buf,
 					   union hermes_scan_info *bss,
@@ -4062,7 +4063,8 @@ static inline char *orinoco_translate_scan(struct net_device *dev,
 	iwe.cmd = SIOCGIWAP;
 	iwe.u.ap_addr.sa_family = ARPHRD_ETHER;
 	memcpy(iwe.u.ap_addr.sa_data, bss->a.bssid, ETH_ALEN);
-	current_ev = iwe_stream_add_event(current_ev, end_buf, &iwe, IW_EV_ADDR_LEN);
+	current_ev = iwe_stream_add_event(info, current_ev, end_buf,
+					  &iwe, IW_EV_ADDR_LEN);
 
 	/* Other entries will be displayed in the order we give them */
 
@@ -4072,7 +4074,8 @@ static inline char *orinoco_translate_scan(struct net_device *dev,
 		iwe.u.data.length = 32;
 	iwe.cmd = SIOCGIWESSID;
 	iwe.u.data.flags = 1;
-	current_ev = iwe_stream_add_point(current_ev, end_buf, &iwe, bss->a.essid);
+	current_ev = iwe_stream_add_point(info, current_ev, end_buf,
+					  &iwe, bss->a.essid);
 
 	/* Add mode */
 	iwe.cmd = SIOCGIWMODE;
@@ -4082,7 +4085,8 @@ static inline char *orinoco_translate_scan(struct net_device *dev,
 			iwe.u.mode = IW_MODE_MASTER;
 		else
 			iwe.u.mode = IW_MODE_ADHOC;
-		current_ev = iwe_stream_add_event(current_ev, end_buf, &iwe, IW_EV_UINT_LEN);
+		current_ev = iwe_stream_add_event(info, current_ev, end_buf,
+						  &iwe, IW_EV_UINT_LEN);
 	}
 
 	channel = bss->s.channel;
@@ -4091,7 +4095,7 @@ static inline char *orinoco_translate_scan(struct net_device *dev,
 		iwe.cmd = SIOCGIWFREQ;
 		iwe.u.freq.m = channel_frequency[channel-1] * 100000;
 		iwe.u.freq.e = 1;
-		current_ev = iwe_stream_add_event(current_ev, end_buf,
+		current_ev = iwe_stream_add_event(info, current_ev, end_buf,
 						  &iwe, IW_EV_FREQ_LEN);
 	}
 
@@ -4106,7 +4110,8 @@ static inline char *orinoco_translate_scan(struct net_device *dev,
 		iwe.u.qual.qual = iwe.u.qual.level - iwe.u.qual.noise;
 	else
 		iwe.u.qual.qual = 0;
-	current_ev = iwe_stream_add_event(current_ev, end_buf, &iwe, IW_EV_QUAL_LEN);
+	current_ev = iwe_stream_add_event(info, current_ev, end_buf,
+					  &iwe, IW_EV_QUAL_LEN);
 
 	/* Add encryption capability */
 	iwe.cmd = SIOCGIWENCODE;
@@ -4115,7 +4120,8 @@ static inline char *orinoco_translate_scan(struct net_device *dev,
 	else
 		iwe.u.data.flags = IW_ENCODE_DISABLED;
 	iwe.u.data.length = 0;
-	current_ev = iwe_stream_add_point(current_ev, end_buf, &iwe, bss->a.essid);
+	current_ev = iwe_stream_add_point(info, current_ev, end_buf,
+					  &iwe, bss->a.essid);
 
 	/* Add EXTRA: Age to display seconds since last beacon/probe response
 	 * for given network. */
@@ -4126,11 +4132,12 @@ static inline char *orinoco_translate_scan(struct net_device *dev,
 		      jiffies_to_msecs(jiffies - last_scanned));
 	iwe.u.data.length = p - custom;
 	if (iwe.u.data.length)
-		current_ev = iwe_stream_add_point(current_ev, end_buf, &iwe, custom);
+		current_ev = iwe_stream_add_point(info, current_ev, end_buf,
+						  &iwe, custom);
 
 	/* Bit rate is not available in Lucent/Agere firmwares */
 	if (priv->firmware_type != FIRMWARE_TYPE_AGERE) {
-		char *current_val = current_ev + IW_EV_LCP_LEN;
+		char *current_val = current_ev + iwe_stream_lcp_len(info);
 		int i;
 		int step;
 
@@ -4149,12 +4156,13 @@ static inline char *orinoco_translate_scan(struct net_device *dev,
 				break;
 			/* Bit rate given in 500 kb/s units (+ 0x80) */
 			iwe.u.bitrate.value = ((bss->p.rates[i] & 0x7f) * 500000);
-			current_val = iwe_stream_add_value(current_ev, current_val,
+			current_val = iwe_stream_add_value(info, current_ev,
+							   current_val,
 							   end_buf, &iwe,
 							   IW_EV_PARAM_LEN);
 		}
 		/* Check if we added any event */
-		if ((current_val - current_ev) > IW_EV_LCP_LEN)
+		if ((current_val - current_ev) > iwe_stream_lcp_len(info))
 			current_ev = current_val;
 	}
 
@@ -4190,7 +4198,7 @@ static int orinoco_ioctl_getscan(struct net_device *dev,
 
 	list_for_each_entry(bss, &priv->bss_list, list) {
 		/* Translate to WE format this entry */
-		current_ev = orinoco_translate_scan(dev, current_ev,
+		current_ev = orinoco_translate_scan(dev, info, current_ev,
 						    extra + srq->length,
 						    &bss->bss,
 						    bss->last_scanned);
diff --git a/drivers/net/wireless/prism54/isl_ioctl.c b/drivers/net/wireless/prism54/isl_ioctl.c
index 5b375b289036..97fa14e0a479 100644
--- a/drivers/net/wireless/prism54/isl_ioctl.c
+++ b/drivers/net/wireless/prism54/isl_ioctl.c
@@ -571,8 +571,9 @@ prism54_set_scan(struct net_device *dev, struct iw_request_info *info,
  */
 
 static char *
-prism54_translate_bss(struct net_device *ndev, char *current_ev,
-		      char *end_buf, struct obj_bss *bss, char noise)
+prism54_translate_bss(struct net_device *ndev, struct iw_request_info *info,
+		      char *current_ev, char *end_buf, struct obj_bss *bss,
+		      char noise)
 {
 	struct iw_event iwe;	/* Temporary buffer */
 	short cap;
@@ -584,8 +585,8 @@ prism54_translate_bss(struct net_device *ndev, char *current_ev,
 	memcpy(iwe.u.ap_addr.sa_data, bss->address, 6);
 	iwe.u.ap_addr.sa_family = ARPHRD_ETHER;
 	iwe.cmd = SIOCGIWAP;
-	current_ev =
-	    iwe_stream_add_event(current_ev, end_buf, &iwe, IW_EV_ADDR_LEN);
+	current_ev = iwe_stream_add_event(info, current_ev, end_buf,
+					  &iwe, IW_EV_ADDR_LEN);
 
 	/* The following entries will be displayed in the same order we give them */
 
@@ -593,7 +594,7 @@ prism54_translate_bss(struct net_device *ndev, char *current_ev,
 	iwe.u.data.length = bss->ssid.length;
 	iwe.u.data.flags = 1;
 	iwe.cmd = SIOCGIWESSID;
-	current_ev = iwe_stream_add_point(current_ev, end_buf,
+	current_ev = iwe_stream_add_point(info, current_ev, end_buf,
 					  &iwe, bss->ssid.octets);
 
 	/* Capabilities */
@@ -610,9 +611,8 @@ prism54_translate_bss(struct net_device *ndev, char *current_ev,
 		iwe.u.mode = IW_MODE_ADHOC;
 	iwe.cmd = SIOCGIWMODE;
 	if (iwe.u.mode)
-		current_ev =
-		    iwe_stream_add_event(current_ev, end_buf, &iwe,
-					 IW_EV_UINT_LEN);
+		current_ev = iwe_stream_add_event(info, current_ev, end_buf,
+						  &iwe, IW_EV_UINT_LEN);
 
 	/* Encryption capability */
 	if (cap & CAP_CRYPT)
@@ -621,14 +621,15 @@ prism54_translate_bss(struct net_device *ndev, char *current_ev,
 		iwe.u.data.flags = IW_ENCODE_DISABLED;
 	iwe.u.data.length = 0;
 	iwe.cmd = SIOCGIWENCODE;
-	current_ev = iwe_stream_add_point(current_ev, end_buf, &iwe, NULL);
+	current_ev = iwe_stream_add_point(info, current_ev, end_buf,
+					  &iwe, NULL);
 
 	/* Add frequency. (short) bss->channel is the frequency in MHz */
 	iwe.u.freq.m = bss->channel;
 	iwe.u.freq.e = 6;
 	iwe.cmd = SIOCGIWFREQ;
-	current_ev =
-	    iwe_stream_add_event(current_ev, end_buf, &iwe, IW_EV_FREQ_LEN);
+	current_ev = iwe_stream_add_event(info, current_ev, end_buf,
+					  &iwe, IW_EV_FREQ_LEN);
 
 	/* Add quality statistics */
 	iwe.u.qual.level = bss->rssi;
@@ -636,20 +637,20 @@ prism54_translate_bss(struct net_device *ndev, char *current_ev,
 	/* do a simple SNR for quality */
 	iwe.u.qual.qual = bss->rssi - noise;
 	iwe.cmd = IWEVQUAL;
-	current_ev =
-	    iwe_stream_add_event(current_ev, end_buf, &iwe, IW_EV_QUAL_LEN);
+	current_ev = iwe_stream_add_event(info, current_ev, end_buf,
+					  &iwe, IW_EV_QUAL_LEN);
 
 	/* Add WPA/RSN Information Element, if any */
 	wpa_ie_len = prism54_wpa_bss_ie_get(priv, bss->address, wpa_ie);
 	if (wpa_ie_len > 0) {
 		iwe.cmd = IWEVGENIE;
 		iwe.u.data.length = min(wpa_ie_len, (size_t)MAX_WPA_IE_LEN);
-		current_ev = iwe_stream_add_point(current_ev, end_buf,
-				&iwe, wpa_ie);
+		current_ev = iwe_stream_add_point(info, current_ev, end_buf,
+						  &iwe, wpa_ie);
 	}
 	/* Do the bitrates */
 	{
-		char *	current_val = current_ev + IW_EV_LCP_LEN;
+		char *current_val = current_ev + iwe_stream_lcp_len(info);
 		int i;
 		int mask;
 
@@ -662,14 +663,14 @@ prism54_translate_bss(struct net_device *ndev, char *current_ev,
 		for(i = 0; i < sizeof(scan_rate_list); i++) {
 			if(bss->rates & mask) {
 				iwe.u.bitrate.value = (scan_rate_list[i] * 500000);
-				current_val = iwe_stream_add_value(current_ev, current_val,
-								   end_buf, &iwe,
-								   IW_EV_PARAM_LEN);
+				current_val = iwe_stream_add_value(
+					info, current_ev, current_val,
+					end_buf, &iwe, IW_EV_PARAM_LEN);
 			}
 			mask <<= 1;
 		}
 		/* Check if we added any event */
-		if ((current_val - current_ev) > IW_EV_LCP_LEN)
+		if ((current_val - current_ev) > iwe_stream_lcp_len(info))
 			current_ev = current_val;
 	}
 
@@ -710,7 +711,7 @@ prism54_get_scan(struct net_device *ndev, struct iw_request_info *info,
 
 	/* ok now, scan the list and translate its info */
 	for (i = 0; i < (int) bsslist->nr; i++) {
-		current_ev = prism54_translate_bss(ndev, current_ev,
+		current_ev = prism54_translate_bss(ndev, info, current_ev,
 						   extra + dwrq->length,
 						   &(bsslist->bsslist[i]),
 						   noise);
@@ -2704,6 +2705,7 @@ prism2_ioctl_scan_req(struct net_device *ndev,
                      struct prism2_hostapd_param *param)
 {
 	islpci_private *priv = netdev_priv(ndev);
+	struct iw_request_info info;
 	int i, rvalue;
 	struct obj_bsslist *bsslist;
 	u32 noise = 0;
@@ -2727,9 +2729,12 @@ prism2_ioctl_scan_req(struct net_device *ndev,
 	rvalue |= mgt_get_request(priv, DOT11_OID_BSSLIST, 0, NULL, &r);
 	bsslist = r.ptr;
 
+	info.cmd = PRISM54_HOSTAPD;
+	info.flags = 0;
+
 	/* ok now, scan the list and translate its info */
 	for (i = 0; i < min(IW_MAX_AP, (int) bsslist->nr); i++)
-		current_ev = prism54_translate_bss(ndev, current_ev,
+		current_ev = prism54_translate_bss(ndev, &info, current_ev,
 						   extra + IW_SCAN_MAX_DATA,
 						   &(bsslist->bsslist[i]),
 						   noise);
diff --git a/drivers/net/wireless/rndis_wlan.c b/drivers/net/wireless/rndis_wlan.c
index a36d2c85e26e..65c50025c88f 100644
--- a/drivers/net/wireless/rndis_wlan.c
+++ b/drivers/net/wireless/rndis_wlan.c
@@ -1648,7 +1648,9 @@ static int rndis_iw_set_scan(struct net_device *dev,
 
 
 static char *rndis_translate_scan(struct net_device *dev,
-    char *cev, char *end_buf, struct ndis_80211_bssid_ex *bssid)
+				  struct iw_request_info *info, char *cev,
+				  char *end_buf,
+				  struct ndis_80211_bssid_ex *bssid)
 {
 #ifdef DEBUG
 	struct usbnet *usbdev = dev->priv;
@@ -1667,14 +1669,14 @@ static char *rndis_translate_scan(struct net_device *dev,
 	iwe.cmd = SIOCGIWAP;
 	iwe.u.ap_addr.sa_family = ARPHRD_ETHER;
 	memcpy(iwe.u.ap_addr.sa_data, bssid->mac, ETH_ALEN);
-	cev = iwe_stream_add_event(cev, end_buf, &iwe, IW_EV_ADDR_LEN);
+	cev = iwe_stream_add_event(info, cev, end_buf, &iwe, IW_EV_ADDR_LEN);
 
 	devdbg(usbdev, "SSID(%d) %s", le32_to_cpu(bssid->ssid.length),
 						bssid->ssid.essid);
 	iwe.cmd = SIOCGIWESSID;
 	iwe.u.essid.length = le32_to_cpu(bssid->ssid.length);
 	iwe.u.essid.flags = 1;
-	cev = iwe_stream_add_point(cev, end_buf, &iwe, bssid->ssid.essid);
+	cev = iwe_stream_add_point(info, cev, end_buf, &iwe, bssid->ssid.essid);
 
 	devdbg(usbdev, "MODE %d", le32_to_cpu(bssid->net_infra));
 	iwe.cmd = SIOCGIWMODE;
@@ -1690,12 +1692,12 @@ static char *rndis_translate_scan(struct net_device *dev,
 		iwe.u.mode = IW_MODE_AUTO;
 		break;
 	}
-	cev = iwe_stream_add_event(cev, end_buf, &iwe, IW_EV_UINT_LEN);
+	cev = iwe_stream_add_event(info, cev, end_buf, &iwe, IW_EV_UINT_LEN);
 
 	devdbg(usbdev, "FREQ %d kHz", le32_to_cpu(bssid->config.ds_config));
 	iwe.cmd = SIOCGIWFREQ;
 	dsconfig_to_freq(le32_to_cpu(bssid->config.ds_config), &iwe.u.freq);
-	cev = iwe_stream_add_event(cev, end_buf, &iwe, IW_EV_FREQ_LEN);
+	cev = iwe_stream_add_event(info, cev, end_buf, &iwe, IW_EV_FREQ_LEN);
 
 	devdbg(usbdev, "QUAL %d", le32_to_cpu(bssid->rssi));
 	iwe.cmd = IWEVQUAL;
@@ -1704,7 +1706,7 @@ static char *rndis_translate_scan(struct net_device *dev,
 	iwe.u.qual.updated = IW_QUAL_QUAL_UPDATED
 			| IW_QUAL_LEVEL_UPDATED
 			| IW_QUAL_NOISE_INVALID;
-	cev = iwe_stream_add_event(cev, end_buf, &iwe, IW_EV_QUAL_LEN);
+	cev = iwe_stream_add_event(info, cev, end_buf, &iwe, IW_EV_QUAL_LEN);
 
 	devdbg(usbdev, "ENCODE %d", le32_to_cpu(bssid->privacy));
 	iwe.cmd = SIOCGIWENCODE;
@@ -1714,10 +1716,10 @@ static char *rndis_translate_scan(struct net_device *dev,
 	else
 		iwe.u.data.flags = IW_ENCODE_ENABLED | IW_ENCODE_NOKEY;
 
-	cev = iwe_stream_add_point(cev, end_buf, &iwe, NULL);
+	cev = iwe_stream_add_point(info, cev, end_buf, &iwe, NULL);
 
 	devdbg(usbdev, "RATES:");
-	current_val = cev + IW_EV_LCP_LEN;
+	current_val = cev + iwe_stream_lcp_len(info);
 	iwe.cmd = SIOCGIWRATE;
 	for (i = 0; i < sizeof(bssid->rates); i++) {
 		if (bssid->rates[i] & 0x7f) {
@@ -1725,13 +1727,13 @@ static char *rndis_translate_scan(struct net_device *dev,
 				((bssid->rates[i] & 0x7f) *
 				500000);
 			devdbg(usbdev, " %d", iwe.u.bitrate.value);
-			current_val = iwe_stream_add_value(cev,
+			current_val = iwe_stream_add_value(info, cev,
 				current_val, end_buf, &iwe,
 				IW_EV_PARAM_LEN);
 		}
 	}
 
-	if ((current_val - cev) > IW_EV_LCP_LEN)
+	if ((current_val - cev) > iwe_stream_lcp_len(info))
 		cev = current_val;
 
 	beacon = le32_to_cpu(bssid->config.beacon_period);
@@ -1739,14 +1741,14 @@ static char *rndis_translate_scan(struct net_device *dev,
 	iwe.cmd = IWEVCUSTOM;
 	snprintf(sbuf, sizeof(sbuf), "bcn_int=%d", beacon);
 	iwe.u.data.length = strlen(sbuf);
-	cev = iwe_stream_add_point(cev, end_buf, &iwe, sbuf);
+	cev = iwe_stream_add_point(info, cev, end_buf, &iwe, sbuf);
 
 	atim = le32_to_cpu(bssid->config.atim_window);
 	devdbg(usbdev, "ATIM %d", atim);
 	iwe.cmd = IWEVCUSTOM;
 	snprintf(sbuf, sizeof(sbuf), "atim=%u", atim);
 	iwe.u.data.length = strlen(sbuf);
-	cev = iwe_stream_add_point(cev, end_buf, &iwe, sbuf);
+	cev = iwe_stream_add_point(info, cev, end_buf, &iwe, sbuf);
 
 	ie = (void *)(bssid->ies + sizeof(struct ndis_80211_fixed_ies));
 	ie_len = min(bssid_len - (int)sizeof(*bssid),
@@ -1760,7 +1762,7 @@ static char *rndis_translate_scan(struct net_device *dev,
 					(ie->id == MFIE_TYPE_RSN) ? 2 : 1);
 			iwe.cmd = IWEVGENIE;
 			iwe.u.data.length = min(ie->len + 2, MAX_WPA_IE_LEN);
-			cev = iwe_stream_add_point(cev, end_buf, &iwe,
+			cev = iwe_stream_add_point(info, cev, end_buf, &iwe,
 								(u8 *)ie);
 		}
 
@@ -1803,8 +1805,8 @@ static int rndis_iw_get_scan(struct net_device *dev,
 	devdbg(usbdev, "SIOCGIWSCAN: %d BSSIDs found", count);
 
 	while (count && ((void *)bssid + bssid_len) <= (buf + len)) {
-		cev = rndis_translate_scan(dev, cev, extra + IW_SCAN_MAX_DATA,
-									bssid);
+		cev = rndis_translate_scan(dev, info, cev,
+					   extra + IW_SCAN_MAX_DATA, bssid);
 		bssid = (void *)bssid + bssid_len;
 		bssid_len = le32_to_cpu(bssid->length);
 		count--;
diff --git a/drivers/net/wireless/wl3501_cs.c b/drivers/net/wireless/wl3501_cs.c
index 42a36b3f3ff7..377141995e36 100644
--- a/drivers/net/wireless/wl3501_cs.c
+++ b/drivers/net/wireless/wl3501_cs.c
@@ -1624,25 +1624,25 @@ static int wl3501_get_scan(struct net_device *dev, struct iw_request_info *info,
 		iwe.cmd			= SIOCGIWAP;
 		iwe.u.ap_addr.sa_family = ARPHRD_ETHER;
 		memcpy(iwe.u.ap_addr.sa_data, this->bss_set[i].bssid, ETH_ALEN);
-		current_ev = iwe_stream_add_event(current_ev,
+		current_ev = iwe_stream_add_event(info, current_ev,
 						  extra + IW_SCAN_MAX_DATA,
 						  &iwe, IW_EV_ADDR_LEN);
 		iwe.cmd		  = SIOCGIWESSID;
 		iwe.u.data.flags  = 1;
 		iwe.u.data.length = this->bss_set[i].ssid.el.len;
-		current_ev = iwe_stream_add_point(current_ev,
+		current_ev = iwe_stream_add_point(info, current_ev,
 						  extra + IW_SCAN_MAX_DATA,
 						  &iwe,
 						  this->bss_set[i].ssid.essid);
 		iwe.cmd	   = SIOCGIWMODE;
 		iwe.u.mode = this->bss_set[i].bss_type;
-		current_ev = iwe_stream_add_event(current_ev,
+		current_ev = iwe_stream_add_event(info, current_ev,
 						  extra + IW_SCAN_MAX_DATA,
 						  &iwe, IW_EV_UINT_LEN);
 		iwe.cmd = SIOCGIWFREQ;
 		iwe.u.freq.m = this->bss_set[i].ds_pset.chan;
 		iwe.u.freq.e = 0;
-		current_ev = iwe_stream_add_event(current_ev,
+		current_ev = iwe_stream_add_event(info, current_ev,
 						  extra + IW_SCAN_MAX_DATA,
 						  &iwe, IW_EV_FREQ_LEN);
 		iwe.cmd = SIOCGIWENCODE;
@@ -1651,7 +1651,7 @@ static int wl3501_get_scan(struct net_device *dev, struct iw_request_info *info,
 		else
 			iwe.u.data.flags = IW_ENCODE_DISABLED;
 		iwe.u.data.length = 0;
-		current_ev = iwe_stream_add_point(current_ev,
+		current_ev = iwe_stream_add_point(info, current_ev,
 						  extra + IW_SCAN_MAX_DATA,
 						  &iwe, NULL);
 	}
diff --git a/drivers/net/wireless/zd1201.c b/drivers/net/wireless/zd1201.c
index d5c0c66188ca..07e4d1f73207 100644
--- a/drivers/net/wireless/zd1201.c
+++ b/drivers/net/wireless/zd1201.c
@@ -1152,32 +1152,36 @@ static int zd1201_get_scan(struct net_device *dev,
 		iwe.cmd = SIOCGIWAP;
 		iwe.u.ap_addr.sa_family = ARPHRD_ETHER;
 		memcpy(iwe.u.ap_addr.sa_data, zd->rxdata+i+6, 6);
-		cev = iwe_stream_add_event(cev, end_buf, &iwe, IW_EV_ADDR_LEN);
+		cev = iwe_stream_add_event(info, cev, end_buf,
+					   &iwe, IW_EV_ADDR_LEN);
 
 		iwe.cmd = SIOCGIWESSID;
 		iwe.u.data.length = zd->rxdata[i+16];
 		iwe.u.data.flags = 1;
-		cev = iwe_stream_add_point(cev, end_buf, &iwe, zd->rxdata+i+18);
+		cev = iwe_stream_add_point(info, cev, end_buf,
+					   &iwe, zd->rxdata+i+18);
 
 		iwe.cmd = SIOCGIWMODE;
 		if (zd->rxdata[i+14]&0x01)
 			iwe.u.mode = IW_MODE_MASTER;
 		else
 			iwe.u.mode = IW_MODE_ADHOC;
-		cev = iwe_stream_add_event(cev, end_buf, &iwe, IW_EV_UINT_LEN);
+		cev = iwe_stream_add_event(info, cev, end_buf,
+					   &iwe, IW_EV_UINT_LEN);
 		
 		iwe.cmd = SIOCGIWFREQ;
 		iwe.u.freq.m = zd->rxdata[i+0];
 		iwe.u.freq.e = 0;
-		cev = iwe_stream_add_event(cev, end_buf, &iwe, IW_EV_FREQ_LEN);
+		cev = iwe_stream_add_event(info, cev, end_buf,
+					   &iwe, IW_EV_FREQ_LEN);
 		
 		iwe.cmd = SIOCGIWRATE;
 		iwe.u.bitrate.fixed = 0;
 		iwe.u.bitrate.disabled = 0;
 		for (j=0; j<10; j++) if (zd->rxdata[i+50+j]) {
 			iwe.u.bitrate.value = (zd->rxdata[i+50+j]&0x7f)*500000;
-			cev=iwe_stream_add_event(cev, end_buf, &iwe,
-			    IW_EV_PARAM_LEN);
+			cev = iwe_stream_add_event(info, cev, end_buf,
+						   &iwe, IW_EV_PARAM_LEN);
 		}
 		
 		iwe.cmd = SIOCGIWENCODE;
@@ -1186,14 +1190,15 @@ static int zd1201_get_scan(struct net_device *dev,
 			iwe.u.data.flags = IW_ENCODE_ENABLED;
 		else
 			iwe.u.data.flags = IW_ENCODE_DISABLED;
-		cev = iwe_stream_add_point(cev, end_buf, &iwe, NULL);
+		cev = iwe_stream_add_point(info, cev, end_buf, &iwe, NULL);
 		
 		iwe.cmd = IWEVQUAL;
 		iwe.u.qual.qual = zd->rxdata[i+4];
 		iwe.u.qual.noise= zd->rxdata[i+2]/10-100;
 		iwe.u.qual.level = (256+zd->rxdata[i+4]*100)/255-100;
 		iwe.u.qual.updated = 7;
-		cev = iwe_stream_add_event(cev, end_buf, &iwe, IW_EV_QUAL_LEN);
+		cev = iwe_stream_add_event(info, cev, end_buf,
+					   &iwe, IW_EV_QUAL_LEN);
 	}
 
 	if (!enabled_save)
diff --git a/include/linux/wireless.h b/include/linux/wireless.h
index 79d846875825..d7958f9b52cb 100644
--- a/include/linux/wireless.h
+++ b/include/linux/wireless.h
@@ -1113,6 +1113,21 @@ struct iw_event
 #define IW_EV_POINT_LEN	(IW_EV_LCP_LEN + sizeof(struct iw_point) - \
 			 IW_EV_POINT_OFF)
 
+#ifdef __KERNEL__
+#ifdef CONFIG_COMPAT
+struct __compat_iw_event {
+	__u16		len;			/* Real length of this stuff */
+	__u16		cmd;			/* Wireless IOCTL */
+	compat_caddr_t	pointer;
+};
+#define IW_EV_COMPAT_LCP_LEN offsetof(struct __compat_iw_event, pointer)
+#define IW_EV_COMPAT_POINT_OFF offsetof(struct compat_iw_point, length)
+#define IW_EV_COMPAT_POINT_LEN	\
+	(IW_EV_COMPAT_LCP_LEN + sizeof(struct compat_iw_point) - \
+	 IW_EV_COMPAT_POINT_OFF)
+#endif
+#endif
+
 /* Size of the Event prefix when packed in stream */
 #define IW_EV_LCP_PK_LEN	(4)
 /* Size of the various events when packed in stream */
diff --git a/include/net/iw_handler.h b/include/net/iw_handler.h
index c99a8eec84e7..51b9a37de991 100644
--- a/include/net/iw_handler.h
+++ b/include/net/iw_handler.h
@@ -478,105 +478,58 @@ extern void wireless_spy_update(struct net_device *	dev,
  * Function that are so simple that it's more efficient inlining them
  */
 
-/*------------------------------------------------------------------*/
-/*
- * Wrapper to add an Wireless Event to a stream of events.
- */
-static inline char *
-iwe_stream_add_event(char *	stream,		/* Stream of events */
-		     char *	ends,		/* End of stream */
-		     struct iw_event *iwe,	/* Payload */
-		     int	event_len)	/* Real size of payload */
+static inline int iwe_stream_lcp_len(struct iw_request_info *info)
 {
-	/* Check if it's possible */
-	if(likely((stream + event_len) < ends)) {
-		iwe->len = event_len;
-		/* Beware of alignement issues on 64 bits */
-		memcpy(stream, (char *) iwe, IW_EV_LCP_PK_LEN);
-		memcpy(stream + IW_EV_LCP_LEN,
-		       ((char *) iwe) + IW_EV_LCP_LEN,
-		       event_len - IW_EV_LCP_LEN);
-		stream += event_len;
-	}
-	return stream;
+#ifdef CONFIG_COMPAT
+	if (info->flags & IW_REQUEST_FLAG_COMPAT)
+		return IW_EV_COMPAT_LCP_LEN;
+#endif
+	return IW_EV_LCP_LEN;
 }
 
-/*------------------------------------------------------------------*/
-/*
- * Wrapper to add an short Wireless Event containing a pointer to a
- * stream of events.
- */
-static inline char *
-iwe_stream_add_point(char *	stream,		/* Stream of events */
-		     char *	ends,		/* End of stream */
-		     struct iw_event *iwe,	/* Payload length + flags */
-		     char *	extra)		/* More payload */
+static inline int iwe_stream_point_len(struct iw_request_info *info)
 {
-	int	event_len = IW_EV_POINT_LEN + iwe->u.data.length;
-	/* Check if it's possible */
-	if(likely((stream + event_len) < ends)) {
-		iwe->len = event_len;
-		memcpy(stream, (char *) iwe, IW_EV_LCP_PK_LEN);
-		memcpy(stream + IW_EV_LCP_LEN,
-		       ((char *) iwe) + IW_EV_LCP_LEN + IW_EV_POINT_OFF,
-		       IW_EV_POINT_PK_LEN - IW_EV_LCP_PK_LEN);
-		memcpy(stream + IW_EV_POINT_LEN, extra, iwe->u.data.length);
-		stream += event_len;
-	}
-	return stream;
+#ifdef CONFIG_COMPAT
+	if (info->flags & IW_REQUEST_FLAG_COMPAT)
+		return IW_EV_COMPAT_POINT_LEN;
+#endif
+	return IW_EV_POINT_LEN;
 }
 
-/*------------------------------------------------------------------*/
-/*
- * Wrapper to add a value to a Wireless Event in a stream of events.
- * Be careful, this one is tricky to use properly :
- * At the first run, you need to have (value = event + IW_EV_LCP_LEN).
- */
-static inline char *
-iwe_stream_add_value(char *	event,		/* Event in the stream */
-		     char *	value,		/* Value in event */
-		     char *	ends,		/* End of stream */
-		     struct iw_event *iwe,	/* Payload */
-		     int	event_len)	/* Real size of payload */
+static inline int iwe_stream_event_len_adjust(struct iw_request_info *info,
+					      int event_len)
 {
-	/* Don't duplicate LCP */
-	event_len -= IW_EV_LCP_LEN;
-
-	/* Check if it's possible */
-	if(likely((value + event_len) < ends)) {
-		/* Add new value */
-		memcpy(value, (char *) iwe + IW_EV_LCP_LEN, event_len);
-		value += event_len;
-		/* Patch LCP */
-		iwe->len = value - event;
-		memcpy(event, (char *) iwe, IW_EV_LCP_LEN);
+#ifdef CONFIG_COMPAT
+	if (info->flags & IW_REQUEST_FLAG_COMPAT) {
+		event_len -= IW_EV_LCP_LEN;
+		event_len += IW_EV_COMPAT_LCP_LEN;
 	}
-	return value;
+#endif
+
+	return event_len;
 }
 
 /*------------------------------------------------------------------*/
 /*
  * Wrapper to add an Wireless Event to a stream of events.
- * Same as above, with explicit error check...
  */
 static inline char *
-iwe_stream_check_add_event(char *	stream,		/* Stream of events */
-			   char *	ends,		/* End of stream */
-			   struct iw_event *iwe,	/* Payload */
-			   int		event_len,	/* Size of payload */
-			   int *	perr)		/* Error report */
+iwe_stream_add_event(struct iw_request_info *info, char *stream, char *ends,
+		     struct iw_event *iwe, int event_len)
 {
-	/* Check if it's possible, set error if not */
+	int lcp_len = iwe_stream_lcp_len(info);
+
+	event_len = iwe_stream_event_len_adjust(info, event_len);
+
+	/* Check if it's possible */
 	if(likely((stream + event_len) < ends)) {
 		iwe->len = event_len;
 		/* Beware of alignement issues on 64 bits */
 		memcpy(stream, (char *) iwe, IW_EV_LCP_PK_LEN);
-		memcpy(stream + IW_EV_LCP_LEN,
-		       ((char *) iwe) + IW_EV_LCP_LEN,
-		       event_len - IW_EV_LCP_LEN);
+		memcpy(stream + lcp_len, &iwe->u,
+		       event_len - lcp_len);
 		stream += event_len;
-	} else
-		*perr = -E2BIG;
+	}
 	return stream;
 }
 
@@ -584,27 +537,25 @@ iwe_stream_check_add_event(char *	stream,		/* Stream of events */
 /*
  * Wrapper to add an short Wireless Event containing a pointer to a
  * stream of events.
- * Same as above, with explicit error check...
  */
 static inline char *
-iwe_stream_check_add_point(char *	stream,		/* Stream of events */
-			   char *	ends,		/* End of stream */
-			   struct iw_event *iwe,	/* Payload length + flags */
-			   char *	extra,		/* More payload */
-			   int *	perr)		/* Error report */
+iwe_stream_add_point(struct iw_request_info *info, char *stream, char *ends,
+		     struct iw_event *iwe, char *extra)
 {
-	int	event_len = IW_EV_POINT_LEN + iwe->u.data.length;
+	int event_len = iwe_stream_point_len(info) + iwe->u.data.length;
+	int point_len = iwe_stream_point_len(info);
+	int lcp_len   = iwe_stream_lcp_len(info);
+
 	/* Check if it's possible */
 	if(likely((stream + event_len) < ends)) {
 		iwe->len = event_len;
 		memcpy(stream, (char *) iwe, IW_EV_LCP_PK_LEN);
-		memcpy(stream + IW_EV_LCP_LEN,
-		       ((char *) iwe) + IW_EV_LCP_LEN + IW_EV_POINT_OFF,
+		memcpy(stream + lcp_len,
+		       ((char *) &iwe->u) + IW_EV_POINT_OFF,
 		       IW_EV_POINT_PK_LEN - IW_EV_LCP_PK_LEN);
-		memcpy(stream + IW_EV_POINT_LEN, extra, iwe->u.data.length);
+		memcpy(stream + point_len, extra, iwe->u.data.length);
 		stream += event_len;
-	} else
-		*perr = -E2BIG;
+	}
 	return stream;
 }
 
@@ -613,29 +564,25 @@ iwe_stream_check_add_point(char *	stream,		/* Stream of events */
  * Wrapper to add a value to a Wireless Event in a stream of events.
  * Be careful, this one is tricky to use properly :
  * At the first run, you need to have (value = event + IW_EV_LCP_LEN).
- * Same as above, with explicit error check...
  */
 static inline char *
-iwe_stream_check_add_value(char *	event,		/* Event in the stream */
-			   char *	value,		/* Value in event */
-			   char *	ends,		/* End of stream */
-			   struct iw_event *iwe,	/* Payload */
-			   int		event_len,	/* Size of payload */
-			   int *	perr)		/* Error report */
+iwe_stream_add_value(struct iw_request_info *info, char *event, char *value,
+		     char *ends, struct iw_event *iwe, int event_len)
 {
+	int lcp_len = iwe_stream_lcp_len(info);
+
 	/* Don't duplicate LCP */
 	event_len -= IW_EV_LCP_LEN;
 
 	/* Check if it's possible */
 	if(likely((value + event_len) < ends)) {
 		/* Add new value */
-		memcpy(value, (char *) iwe + IW_EV_LCP_LEN, event_len);
+		memcpy(value, &iwe->u, event_len);
 		value += event_len;
 		/* Patch LCP */
 		iwe->len = value - event;
-		memcpy(event, (char *) iwe, IW_EV_LCP_LEN);
-	} else
-		*perr = -E2BIG;
+		memcpy(event, (char *) iwe, lcp_len);
+	}
 	return value;
 }
 
diff --git a/net/ieee80211/ieee80211_wx.c b/net/ieee80211/ieee80211_wx.c
index 822606b615ca..973832dd7faf 100644
--- a/net/ieee80211/ieee80211_wx.c
+++ b/net/ieee80211/ieee80211_wx.c
@@ -43,8 +43,9 @@ static const char *ieee80211_modes[] = {
 
 #define MAX_CUSTOM_LEN 64
 static char *ieee80211_translate_scan(struct ieee80211_device *ieee,
-					   char *start, char *stop,
-					   struct ieee80211_network *network)
+				      char *start, char *stop,
+				      struct ieee80211_network *network,
+				      struct iw_request_info *info)
 {
 	char custom[MAX_CUSTOM_LEN];
 	char *p;
@@ -57,7 +58,7 @@ static char *ieee80211_translate_scan(struct ieee80211_device *ieee,
 	iwe.cmd = SIOCGIWAP;
 	iwe.u.ap_addr.sa_family = ARPHRD_ETHER;
 	memcpy(iwe.u.ap_addr.sa_data, network->bssid, ETH_ALEN);
-	start = iwe_stream_add_event(start, stop, &iwe, IW_EV_ADDR_LEN);
+	start = iwe_stream_add_event(info, start, stop, &iwe, IW_EV_ADDR_LEN);
 
 	/* Remaining entries will be displayed in the order we provide them */
 
@@ -66,17 +67,19 @@ static char *ieee80211_translate_scan(struct ieee80211_device *ieee,
 	iwe.u.data.flags = 1;
 	if (network->flags & NETWORK_EMPTY_ESSID) {
 		iwe.u.data.length = sizeof("<hidden>");
-		start = iwe_stream_add_point(start, stop, &iwe, "<hidden>");
+		start = iwe_stream_add_point(info, start, stop,
+					     &iwe, "<hidden>");
 	} else {
 		iwe.u.data.length = min(network->ssid_len, (u8) 32);
-		start = iwe_stream_add_point(start, stop, &iwe, network->ssid);
+		start = iwe_stream_add_point(info, start, stop,
+					     &iwe, network->ssid);
 	}
 
 	/* Add the protocol name */
 	iwe.cmd = SIOCGIWNAME;
 	snprintf(iwe.u.name, IFNAMSIZ, "IEEE 802.11%s",
 		 ieee80211_modes[network->mode]);
-	start = iwe_stream_add_event(start, stop, &iwe, IW_EV_CHAR_LEN);
+	start = iwe_stream_add_event(info, start, stop, &iwe, IW_EV_CHAR_LEN);
 
 	/* Add mode */
 	iwe.cmd = SIOCGIWMODE;
@@ -86,7 +89,8 @@ static char *ieee80211_translate_scan(struct ieee80211_device *ieee,
 		else
 			iwe.u.mode = IW_MODE_ADHOC;
 
-		start = iwe_stream_add_event(start, stop, &iwe, IW_EV_UINT_LEN);
+		start = iwe_stream_add_event(info, start, stop,
+					     &iwe, IW_EV_UINT_LEN);
 	}
 
 	/* Add channel and frequency */
@@ -95,7 +99,7 @@ static char *ieee80211_translate_scan(struct ieee80211_device *ieee,
 	iwe.u.freq.m = ieee80211_channel_to_freq(ieee, network->channel);
 	iwe.u.freq.e = 6;
 	iwe.u.freq.i = 0;
-	start = iwe_stream_add_event(start, stop, &iwe, IW_EV_FREQ_LEN);
+	start = iwe_stream_add_event(info, start, stop, &iwe, IW_EV_FREQ_LEN);
 
 	/* Add encryption capability */
 	iwe.cmd = SIOCGIWENCODE;
@@ -104,12 +108,13 @@ static char *ieee80211_translate_scan(struct ieee80211_device *ieee,
 	else
 		iwe.u.data.flags = IW_ENCODE_DISABLED;
 	iwe.u.data.length = 0;
-	start = iwe_stream_add_point(start, stop, &iwe, network->ssid);
+	start = iwe_stream_add_point(info, start, stop,
+				     &iwe, network->ssid);
 
 	/* Add basic and extended rates */
 	/* Rate : stuffing multiple values in a single event require a bit
 	 * more of magic - Jean II */
-	current_val = start + IW_EV_LCP_LEN;
+	current_val = start + iwe_stream_lcp_len(info);
 	iwe.cmd = SIOCGIWRATE;
 	/* Those two flags are ignored... */
 	iwe.u.bitrate.fixed = iwe.u.bitrate.disabled = 0;
@@ -124,17 +129,19 @@ static char *ieee80211_translate_scan(struct ieee80211_device *ieee,
 		/* Bit rate given in 500 kb/s units (+ 0x80) */
 		iwe.u.bitrate.value = ((rate & 0x7f) * 500000);
 		/* Add new value to event */
-		current_val = iwe_stream_add_value(start, current_val, stop, &iwe, IW_EV_PARAM_LEN);
+		current_val = iwe_stream_add_value(info, start, current_val,
+						   stop, &iwe, IW_EV_PARAM_LEN);
 	}
 	for (; j < network->rates_ex_len; j++) {
 		rate = network->rates_ex[j] & 0x7F;
 		/* Bit rate given in 500 kb/s units (+ 0x80) */
 		iwe.u.bitrate.value = ((rate & 0x7f) * 500000);
 		/* Add new value to event */
-		current_val = iwe_stream_add_value(start, current_val, stop, &iwe, IW_EV_PARAM_LEN);
+		current_val = iwe_stream_add_value(info, start, current_val,
+						   stop, &iwe, IW_EV_PARAM_LEN);
 	}
 	/* Check if we added any rate */
-	if((current_val - start) > IW_EV_LCP_LEN)
+	if ((current_val - start) > iwe_stream_lcp_len(info))
 		start = current_val;
 
 	/* Add quality statistics */
@@ -181,14 +188,14 @@ static char *ieee80211_translate_scan(struct ieee80211_device *ieee,
 		iwe.u.qual.level = network->stats.signal;
 	}
 
-	start = iwe_stream_add_event(start, stop, &iwe, IW_EV_QUAL_LEN);
+	start = iwe_stream_add_event(info, start, stop, &iwe, IW_EV_QUAL_LEN);
 
 	iwe.cmd = IWEVCUSTOM;
 	p = custom;
 
 	iwe.u.data.length = p - custom;
 	if (iwe.u.data.length)
-		start = iwe_stream_add_point(start, stop, &iwe, custom);
+		start = iwe_stream_add_point(info, start, stop, &iwe, custom);
 
 	memset(&iwe, 0, sizeof(iwe));
 	if (network->wpa_ie_len) {
@@ -196,7 +203,7 @@ static char *ieee80211_translate_scan(struct ieee80211_device *ieee,
 		memcpy(buf, network->wpa_ie, network->wpa_ie_len);
 		iwe.cmd = IWEVGENIE;
 		iwe.u.data.length = network->wpa_ie_len;
-		start = iwe_stream_add_point(start, stop, &iwe, buf);
+		start = iwe_stream_add_point(info, start, stop, &iwe, buf);
 	}
 
 	memset(&iwe, 0, sizeof(iwe));
@@ -205,7 +212,7 @@ static char *ieee80211_translate_scan(struct ieee80211_device *ieee,
 		memcpy(buf, network->rsn_ie, network->rsn_ie_len);
 		iwe.cmd = IWEVGENIE;
 		iwe.u.data.length = network->rsn_ie_len;
-		start = iwe_stream_add_point(start, stop, &iwe, buf);
+		start = iwe_stream_add_point(info, start, stop, &iwe, buf);
 	}
 
 	/* Add EXTRA: Age to display seconds since last beacon/probe response
@@ -217,7 +224,7 @@ static char *ieee80211_translate_scan(struct ieee80211_device *ieee,
 		      jiffies_to_msecs(jiffies - network->last_scanned));
 	iwe.u.data.length = p - custom;
 	if (iwe.u.data.length)
-		start = iwe_stream_add_point(start, stop, &iwe, custom);
+		start = iwe_stream_add_point(info, start, stop, &iwe, custom);
 
 	/* Add spectrum management information */
 	iwe.cmd = -1;
@@ -238,7 +245,7 @@ static char *ieee80211_translate_scan(struct ieee80211_device *ieee,
 
 	if (iwe.cmd == IWEVCUSTOM) {
 		iwe.u.data.length = p - custom;
-		start = iwe_stream_add_point(start, stop, &iwe, custom);
+		start = iwe_stream_add_point(info, start, stop, &iwe, custom);
 	}
 
 	return start;
@@ -272,7 +279,8 @@ int ieee80211_wx_get_scan(struct ieee80211_device *ieee,
 
 		if (ieee->scan_age == 0 ||
 		    time_after(network->last_scanned + ieee->scan_age, jiffies))
-			ev = ieee80211_translate_scan(ieee, ev, stop, network);
+			ev = ieee80211_translate_scan(ieee, ev, stop, network,
+						      info);
 		else
 			IEEE80211_DEBUG_SCAN("Not showing network '%s ("
 					     "%s)' due to age (%dms).\n",
diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h
index 14fccf16b80f..80a9e7c07b47 100644
--- a/net/mac80211/ieee80211_i.h
+++ b/net/mac80211/ieee80211_i.h
@@ -24,6 +24,7 @@
 #include <linux/spinlock.h>
 #include <linux/etherdevice.h>
 #include <net/wireless.h>
+#include <net/iw_handler.h>
 #include "key.h"
 #include "sta_info.h"
 
@@ -867,7 +868,9 @@ int ieee80211_sta_set_bssid(struct net_device *dev, u8 *bssid);
 int ieee80211_sta_req_scan(struct net_device *dev, u8 *ssid, size_t ssid_len);
 void ieee80211_sta_req_auth(struct net_device *dev,
 			    struct ieee80211_if_sta *ifsta);
-int ieee80211_sta_scan_results(struct net_device *dev, char *buf, size_t len);
+int ieee80211_sta_scan_results(struct net_device *dev,
+			       struct iw_request_info *info,
+			       char *buf, size_t len);
 ieee80211_rx_result ieee80211_sta_rx_scan(
 	struct net_device *dev, struct sk_buff *skb,
 	struct ieee80211_rx_status *rx_status);
diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c
index 55659a730dc1..e06d6450f215 100644
--- a/net/mac80211/mlme.c
+++ b/net/mac80211/mlme.c
@@ -4087,6 +4087,7 @@ int ieee80211_sta_req_scan(struct net_device *dev, u8 *ssid, size_t ssid_len)
 
 static char *
 ieee80211_sta_scan_result(struct net_device *dev,
+			  struct iw_request_info *info,
 			  struct ieee80211_sta_bss *bss,
 			  char *current_ev, char *end_buf)
 {
@@ -4101,7 +4102,7 @@ ieee80211_sta_scan_result(struct net_device *dev,
 	iwe.cmd = SIOCGIWAP;
 	iwe.u.ap_addr.sa_family = ARPHRD_ETHER;
 	memcpy(iwe.u.ap_addr.sa_data, bss->bssid, ETH_ALEN);
-	current_ev = iwe_stream_add_event(current_ev, end_buf, &iwe,
+	current_ev = iwe_stream_add_event(info, current_ev, end_buf, &iwe,
 					  IW_EV_ADDR_LEN);
 
 	memset(&iwe, 0, sizeof(iwe));
@@ -4109,13 +4110,13 @@ ieee80211_sta_scan_result(struct net_device *dev,
 	if (bss_mesh_cfg(bss)) {
 		iwe.u.data.length = bss_mesh_id_len(bss);
 		iwe.u.data.flags = 1;
-		current_ev = iwe_stream_add_point(current_ev, end_buf, &iwe,
-						  bss_mesh_id(bss));
+		current_ev = iwe_stream_add_point(info, current_ev, end_buf,
+						  &iwe, bss_mesh_id(bss));
 	} else {
 		iwe.u.data.length = bss->ssid_len;
 		iwe.u.data.flags = 1;
-		current_ev = iwe_stream_add_point(current_ev, end_buf, &iwe,
-						  bss->ssid);
+		current_ev = iwe_stream_add_point(info, current_ev, end_buf,
+						  &iwe, bss->ssid);
 	}
 
 	if (bss->capability & (WLAN_CAPABILITY_ESS | WLAN_CAPABILITY_IBSS)
@@ -4128,22 +4129,22 @@ ieee80211_sta_scan_result(struct net_device *dev,
 			iwe.u.mode = IW_MODE_MASTER;
 		else
 			iwe.u.mode = IW_MODE_ADHOC;
-		current_ev = iwe_stream_add_event(current_ev, end_buf, &iwe,
-						  IW_EV_UINT_LEN);
+		current_ev = iwe_stream_add_event(info, current_ev, end_buf,
+						  &iwe, IW_EV_UINT_LEN);
 	}
 
 	memset(&iwe, 0, sizeof(iwe));
 	iwe.cmd = SIOCGIWFREQ;
 	iwe.u.freq.m = ieee80211_frequency_to_channel(bss->freq);
 	iwe.u.freq.e = 0;
-	current_ev = iwe_stream_add_event(current_ev, end_buf, &iwe,
+	current_ev = iwe_stream_add_event(info, current_ev, end_buf, &iwe,
 					  IW_EV_FREQ_LEN);
 
 	memset(&iwe, 0, sizeof(iwe));
 	iwe.cmd = SIOCGIWFREQ;
 	iwe.u.freq.m = bss->freq;
 	iwe.u.freq.e = 6;
-	current_ev = iwe_stream_add_event(current_ev, end_buf, &iwe,
+	current_ev = iwe_stream_add_event(info, current_ev, end_buf, &iwe,
 					  IW_EV_FREQ_LEN);
 	memset(&iwe, 0, sizeof(iwe));
 	iwe.cmd = IWEVQUAL;
@@ -4151,7 +4152,7 @@ ieee80211_sta_scan_result(struct net_device *dev,
 	iwe.u.qual.level = bss->signal;
 	iwe.u.qual.noise = bss->noise;
 	iwe.u.qual.updated = local->wstats_flags;
-	current_ev = iwe_stream_add_event(current_ev, end_buf, &iwe,
+	current_ev = iwe_stream_add_event(info, current_ev, end_buf, &iwe,
 					  IW_EV_QUAL_LEN);
 
 	memset(&iwe, 0, sizeof(iwe));
@@ -4161,35 +4162,36 @@ ieee80211_sta_scan_result(struct net_device *dev,
 	else
 		iwe.u.data.flags = IW_ENCODE_DISABLED;
 	iwe.u.data.length = 0;
-	current_ev = iwe_stream_add_point(current_ev, end_buf, &iwe, "");
+	current_ev = iwe_stream_add_point(info, current_ev, end_buf,
+					  &iwe, "");
 
 	if (bss && bss->wpa_ie) {
 		memset(&iwe, 0, sizeof(iwe));
 		iwe.cmd = IWEVGENIE;
 		iwe.u.data.length = bss->wpa_ie_len;
-		current_ev = iwe_stream_add_point(current_ev, end_buf, &iwe,
-						  bss->wpa_ie);
+		current_ev = iwe_stream_add_point(info, current_ev, end_buf,
+						  &iwe, bss->wpa_ie);
 	}
 
 	if (bss && bss->rsn_ie) {
 		memset(&iwe, 0, sizeof(iwe));
 		iwe.cmd = IWEVGENIE;
 		iwe.u.data.length = bss->rsn_ie_len;
-		current_ev = iwe_stream_add_point(current_ev, end_buf, &iwe,
-						  bss->rsn_ie);
+		current_ev = iwe_stream_add_point(info, current_ev, end_buf,
+						  &iwe, bss->rsn_ie);
 	}
 
 	if (bss && bss->ht_ie) {
 		memset(&iwe, 0, sizeof(iwe));
 		iwe.cmd = IWEVGENIE;
 		iwe.u.data.length = bss->ht_ie_len;
-		current_ev = iwe_stream_add_point(current_ev, end_buf, &iwe,
-						  bss->ht_ie);
+		current_ev = iwe_stream_add_point(info, current_ev, end_buf,
+						  &iwe, bss->ht_ie);
 	}
 
 	if (bss && bss->supp_rates_len > 0) {
 		/* display all supported rates in readable format */
-		char *p = current_ev + IW_EV_LCP_LEN;
+		char *p = current_ev + iwe_stream_lcp_len(info);
 		int i;
 
 		memset(&iwe, 0, sizeof(iwe));
@@ -4200,7 +4202,7 @@ ieee80211_sta_scan_result(struct net_device *dev,
 		for (i = 0; i < bss->supp_rates_len; i++) {
 			iwe.u.bitrate.value = ((bss->supp_rates[i] &
 							0x7f) * 500000);
-			p = iwe_stream_add_value(current_ev, p,
+			p = iwe_stream_add_value(info, current_ev, p,
 					end_buf, &iwe, IW_EV_PARAM_LEN);
 		}
 		current_ev = p;
@@ -4214,7 +4216,8 @@ ieee80211_sta_scan_result(struct net_device *dev,
 			iwe.cmd = IWEVCUSTOM;
 			sprintf(buf, "tsf=%016llx", (unsigned long long)(bss->timestamp));
 			iwe.u.data.length = strlen(buf);
-			current_ev = iwe_stream_add_point(current_ev, end_buf,
+			current_ev = iwe_stream_add_point(info, current_ev,
+							  end_buf,
 							  &iwe, buf);
 			kfree(buf);
 		}
@@ -4229,31 +4232,36 @@ ieee80211_sta_scan_result(struct net_device *dev,
 			iwe.cmd = IWEVCUSTOM;
 			sprintf(buf, "Mesh network (version %d)", cfg[0]);
 			iwe.u.data.length = strlen(buf);
-			current_ev = iwe_stream_add_point(current_ev, end_buf,
+			current_ev = iwe_stream_add_point(info, current_ev,
+							  end_buf,
 							  &iwe, buf);
 			sprintf(buf, "Path Selection Protocol ID: "
 				"0x%02X%02X%02X%02X", cfg[1], cfg[2], cfg[3],
 							cfg[4]);
 			iwe.u.data.length = strlen(buf);
-			current_ev = iwe_stream_add_point(current_ev, end_buf,
+			current_ev = iwe_stream_add_point(info, current_ev,
+							  end_buf,
 							  &iwe, buf);
 			sprintf(buf, "Path Selection Metric ID: "
 				"0x%02X%02X%02X%02X", cfg[5], cfg[6], cfg[7],
 							cfg[8]);
 			iwe.u.data.length = strlen(buf);
-			current_ev = iwe_stream_add_point(current_ev, end_buf,
+			current_ev = iwe_stream_add_point(info, current_ev,
+							  end_buf,
 							  &iwe, buf);
 			sprintf(buf, "Congestion Control Mode ID: "
 				"0x%02X%02X%02X%02X", cfg[9], cfg[10],
 							cfg[11], cfg[12]);
 			iwe.u.data.length = strlen(buf);
-			current_ev = iwe_stream_add_point(current_ev, end_buf,
+			current_ev = iwe_stream_add_point(info, current_ev,
+							  end_buf,
 							  &iwe, buf);
 			sprintf(buf, "Channel Precedence: "
 				"0x%02X%02X%02X%02X", cfg[13], cfg[14],
 							cfg[15], cfg[16]);
 			iwe.u.data.length = strlen(buf);
-			current_ev = iwe_stream_add_point(current_ev, end_buf,
+			current_ev = iwe_stream_add_point(info, current_ev,
+							  end_buf,
 							  &iwe, buf);
 			kfree(buf);
 		}
@@ -4263,7 +4271,9 @@ ieee80211_sta_scan_result(struct net_device *dev,
 }
 
 
-int ieee80211_sta_scan_results(struct net_device *dev, char *buf, size_t len)
+int ieee80211_sta_scan_results(struct net_device *dev,
+			       struct iw_request_info *info,
+			       char *buf, size_t len)
 {
 	struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr);
 	char *current_ev = buf;
@@ -4276,8 +4286,8 @@ int ieee80211_sta_scan_results(struct net_device *dev, char *buf, size_t len)
 			spin_unlock_bh(&local->sta_bss_lock);
 			return -E2BIG;
 		}
-		current_ev = ieee80211_sta_scan_result(dev, bss, current_ev,
-						       end_buf);
+		current_ev = ieee80211_sta_scan_result(dev, info, bss,
+						       current_ev, end_buf);
 	}
 	spin_unlock_bh(&local->sta_bss_lock);
 	return current_ev - buf;
diff --git a/net/mac80211/wext.c b/net/mac80211/wext.c
index 5af3862e7191..f47d13bdf7f7 100644
--- a/net/mac80211/wext.c
+++ b/net/mac80211/wext.c
@@ -567,7 +567,7 @@ static int ieee80211_ioctl_giwscan(struct net_device *dev,
 	if (local->sta_sw_scanning || local->sta_hw_scanning)
 		return -EAGAIN;
 
-	res = ieee80211_sta_scan_results(dev, extra, data->length);
+	res = ieee80211_sta_scan_results(dev, info, extra, data->length);
 	if (res >= 0) {
 		data->length = res;
 		return 0;
-- 
cgit v1.2.3


From 8ce9c6ede1504d29eead67862edc96c03dd4d0a2 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Tue, 17 Jun 2008 00:40:36 -0700
Subject: sctp: Kill SCTP_SOCK_SLEEP_{PRE,POST}, unused.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/sctp/sctp.h | 2 --
 1 file changed, 2 deletions(-)

(limited to 'include')

diff --git a/include/net/sctp/sctp.h b/include/net/sctp/sctp.h
index 5672d489e924..17b932b8a55a 100644
--- a/include/net/sctp/sctp.h
+++ b/include/net/sctp/sctp.h
@@ -220,8 +220,6 @@ extern struct kmem_cache *sctp_bucket_cachep __read_mostly;
 #define sctp_release_sock(sk)    release_sock(sk)
 #define sctp_bh_lock_sock(sk)    bh_lock_sock(sk)
 #define sctp_bh_unlock_sock(sk)  bh_unlock_sock(sk)
-#define SCTP_SOCK_SLEEP_PRE(sk)  SOCK_SLEEP_PRE(sk)
-#define SCTP_SOCK_SLEEP_POST(sk) SOCK_SLEEP_POST(sk)
 
 /* SCTP SNMP MIB stats handlers */
 DECLARE_SNMP_STAT(struct sctp_mib, sctp_statistics);
-- 
cgit v1.2.3


From 338db085518a8436cdecd33f7b52a06ec16d9ec1 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Tue, 17 Jun 2008 01:09:00 -0700
Subject: net: Kill SOCK_SLEEP_PRE and SOCK_SLEEP_POST, no users.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/sock.h | 24 ------------------------
 1 file changed, 24 deletions(-)

(limited to 'include')

diff --git a/include/net/sock.h b/include/net/sock.h
index 0a80961a83c0..83f74b11d09a 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -1331,30 +1331,6 @@ extern int net_msg_warn;
 #define LIMIT_NETDEBUG(fmt, args...) \
 	do { if (net_msg_warn && net_ratelimit()) printk(fmt,##args); } while(0)
 
-/*
- * Macros for sleeping on a socket. Use them like this:
- *
- * SOCK_SLEEP_PRE(sk)
- * if (condition)
- * 	schedule();
- * SOCK_SLEEP_POST(sk)
- *
- * N.B. These are now obsolete and were, afaik, only ever used in DECnet
- * and when the last use of them in DECnet has gone, I'm intending to
- * remove them.
- */
-
-#define SOCK_SLEEP_PRE(sk) 	{ struct task_struct *tsk = current; \
-				DECLARE_WAITQUEUE(wait, tsk); \
-				tsk->state = TASK_INTERRUPTIBLE; \
-				add_wait_queue((sk)->sk_sleep, &wait); \
-				release_sock(sk);
-
-#define SOCK_SLEEP_POST(sk)	tsk->state = TASK_RUNNING; \
-				remove_wait_queue((sk)->sk_sleep, &wait); \
-				lock_sock(sk); \
-				}
-
 extern __u32 sysctl_wmem_max;
 extern __u32 sysctl_rmem_max;
 
-- 
cgit v1.2.3


From df3bc8bd8f8fd17e9b22859d82af38fa702e75b7 Mon Sep 17 00:00:00 2001
From: Chas Williams <chas@cmf.nrl.navy.mil>
Date: Tue, 17 Jun 2008 16:19:24 -0700
Subject: atm: [suni] add support for setting loopback and framing modes

Signed-off-by: Chas Williams <chas@cmf.nrl.navy.mil>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/atm/suni.c    | 121 +++++++++++++++++++++++++++++++++++++++++++-------
 drivers/atm/suni.h    |  31 +++++++++++--
 include/linux/sonet.h |   2 +-
 3 files changed, 133 insertions(+), 21 deletions(-)

(limited to 'include')

diff --git a/drivers/atm/suni.c b/drivers/atm/suni.c
index 6097f6877a44..6dd3f5919968 100644
--- a/drivers/atm/suni.c
+++ b/drivers/atm/suni.c
@@ -1,8 +1,14 @@
-/* drivers/atm/suni.c - PMC PM5346 SUNI (PHY) driver */
+/*
+ * drivers/atm/suni.c - S/UNI PHY driver
+ *
+ * Supports the following:
+ * 	PMC PM5346 S/UNI LITE
+ * 	PMC PM5350 S/UNI 155 ULTRA
+ * 	PMC PM5355 S/UNI 622
+ */
  
 /* Written 1995-2000 by Werner Almesberger, EPFL LRC/ICA */
 
-
 #include <linux/module.h>
 #include <linux/jiffies.h>
 #include <linux/kernel.h>
@@ -146,25 +152,105 @@ static int get_diag(struct atm_dev *dev,void __user *arg)
 static int set_loopback(struct atm_dev *dev,int mode)
 {
 	unsigned char control;
+	int reg, dle, lle;
+
+	if (PRIV(dev)->type == SUNI_MRI_TYPE_PM5355) {
+		reg = SUNI_MCM;
+		dle = SUNI_MCM_DLE;
+		lle = SUNI_MCM_LLE;
+	} else {
+		reg = SUNI_MCT;
+		dle = SUNI_MCT_DLE;
+		lle = SUNI_MCT_LLE;
+	}
 
-	control = GET(MCT) & ~(SUNI_MCT_DLE | SUNI_MCT_LLE);
+	control = dev->ops->phy_get(dev, reg) & ~(dle | lle);
 	switch (mode) {
 		case ATM_LM_NONE:
 			break;
 		case ATM_LM_LOC_PHY:
-			control |= SUNI_MCT_DLE;
+			control |= dle;
 			break;
 		case ATM_LM_RMT_PHY:
-			control |= SUNI_MCT_LLE;
+			control |= lle;
 			break;
 		default:
 			return -EINVAL;
 	}
-	PUT(control,MCT);
+	 dev->ops->phy_put(dev, control, reg);
 	PRIV(dev)->loop_mode = mode;
 	return 0;
 }
 
+/*
+ * SONET vs. SDH Configuration
+ *
+ * Z0INS (register 0x06): 0 for SONET, 1 for SDH
+ * ENSS (register 0x3D): 0 for SONET, 1 for SDH
+ * LEN16 (register 0x28): 0 for SONET, 1 for SDH (n/a for S/UNI 155 QUAD)
+ * LEN16 (register 0x50): 0 for SONET, 1 for SDH (n/a for S/UNI 155 QUAD)
+ * S[1:0] (register 0x46): 00 for SONET, 10 for SDH
+ */
+
+static int set_sonet(struct atm_dev *dev)
+{
+	if (PRIV(dev)->type == SUNI_MRI_TYPE_PM5355) {
+		PUT(GET(RPOP_RC) & ~SUNI_RPOP_RC_ENSS, RPOP_RC);
+		PUT(GET(SSTB_CTRL) & ~SUNI_SSTB_CTRL_LEN16, SSTB_CTRL);
+		PUT(GET(SPTB_CTRL) & ~SUNI_SPTB_CTRL_LEN16, SPTB_CTRL);
+	}
+
+	REG_CHANGE(SUNI_TPOP_APM_S, SUNI_TPOP_APM_S_SHIFT,
+		   SUNI_TPOP_S_SONET, TPOP_APM);
+
+	return 0;
+}
+
+static int set_sdh(struct atm_dev *dev)
+{
+	if (PRIV(dev)->type == SUNI_MRI_TYPE_PM5355) {
+		PUT(GET(RPOP_RC) | SUNI_RPOP_RC_ENSS, RPOP_RC);
+		PUT(GET(SSTB_CTRL) | SUNI_SSTB_CTRL_LEN16, SSTB_CTRL);
+		PUT(GET(SPTB_CTRL) | SUNI_SPTB_CTRL_LEN16, SPTB_CTRL);
+	}
+
+	REG_CHANGE(SUNI_TPOP_APM_S, SUNI_TPOP_APM_S_SHIFT,
+		   SUNI_TPOP_S_SDH, TPOP_APM);
+
+	return 0;
+}
+
+
+static int get_framing(struct atm_dev *dev, void __user *arg)
+{
+	int framing;
+	unsigned char s;
+
+
+	s = (GET(TPOP_APM) & SUNI_TPOP_APM_S) >> SUNI_TPOP_APM_S_SHIFT;
+	if (s == SUNI_TPOP_S_SONET)
+		framing = SONET_FRAME_SONET;
+	else
+		framing = SONET_FRAME_SDH;
+
+	return put_user(framing, (int __user *) arg) ? -EFAULT : 0;
+}
+
+static int set_framing(struct atm_dev *dev, void __user *arg)
+{
+	int mode;
+
+	if (get_user(mode, (int __user *) arg))
+		return -EFAULT;
+
+	if (mode == SONET_FRAME_SONET)
+		return set_sonet(dev);
+	else if (mode == SONET_FRAME_SDH)
+		return set_sdh(dev);
+
+	return -EINVAL;
+}
+
 
 static int suni_ioctl(struct atm_dev *dev,unsigned int cmd,void __user *arg)
 {
@@ -179,14 +265,16 @@ static int suni_ioctl(struct atm_dev *dev,unsigned int cmd,void __user *arg)
 		case SONET_GETDIAG:
 			return get_diag(dev,arg);
 		case SONET_SETFRAMING:
-			if ((int)(unsigned long)arg != SONET_FRAME_SONET) return -EINVAL;
-			return 0;
+			if (!capable(CAP_NET_ADMIN))
+				return -EPERM;
+			return set_framing(dev, arg);
 		case SONET_GETFRAMING:
-			return put_user(SONET_FRAME_SONET,(int __user *)arg) ?
-			    -EFAULT : 0;
+			return get_framing(dev, arg);
 		case SONET_GETFRSENSE:
 			return -EINVAL;
 		case ATM_SETLOOP:
+			if (!capable(CAP_NET_ADMIN))
+				return -EPERM;
 			return set_loopback(dev,(int)(unsigned long)arg);
 		case ATM_GETLOOP:
 			return put_user(PRIV(dev)->loop_mode,(int __user *)arg) ?
@@ -220,10 +308,6 @@ static int suni_start(struct atm_dev *dev)
 	unsigned long flags;
 	int first;
 
-	if (!(dev->phy_data = kmalloc(sizeof(struct suni_priv),GFP_KERNEL)))
-		return -ENOMEM;
-
-	PRIV(dev)->dev = dev;
 	spin_lock_irqsave(&sunis_lock,flags);
 	first = !sunis;
 	PRIV(dev)->next = sunis;
@@ -284,16 +368,21 @@ int suni_init(struct atm_dev *dev)
 {
 	unsigned char mri;
 
+	if (!(dev->phy_data = kmalloc(sizeof(struct suni_priv),GFP_KERNEL)))
+		return -ENOMEM;
+	PRIV(dev)->dev = dev;
+
 	mri = GET(MRI); /* reset SUNI */
+	PRIV(dev)->type = (mri & SUNI_MRI_TYPE) >> SUNI_MRI_TYPE_SHIFT;
 	PUT(mri | SUNI_MRI_RESET,MRI);
 	PUT(mri,MRI);
 	PUT((GET(MT) & SUNI_MT_DS27_53),MT); /* disable all tests */
-	REG_CHANGE(SUNI_TPOP_APM_S,SUNI_TPOP_APM_S_SHIFT,SUNI_TPOP_S_SONET,
-	    TPOP_APM); /* use SONET */
+        set_sonet(dev);
 	REG_CHANGE(SUNI_TACP_IUCHP_CLP,0,SUNI_TACP_IUCHP_CLP,
 	    TACP_IUCHP); /* idle cells */
 	PUT(SUNI_IDLE_PATTERN,TACP_IUCPOP);
 	dev->phy = &suni_ops;
+
 	return 0;
 }
 
diff --git a/drivers/atm/suni.h b/drivers/atm/suni.h
index efa79bfae75b..7e3e656b3993 100644
--- a/drivers/atm/suni.h
+++ b/drivers/atm/suni.h
@@ -1,7 +1,8 @@
-/* drivers/atm/suni.h - PMC PM5346 SUNI (PHY) declarations */
+/*
+ * drivers/atm/suni.h - S/UNI PHY driver
+ */
  
 /* Written 1995-2000 by Werner Almesberger, EPFL LRC/ICA */
- 
 
 #ifndef DRIVER_ATM_SUNI_H
 #define DRIVER_ATM_SUNI_H
@@ -39,7 +40,8 @@
 #define SUNI_RLOP_LFM		0x1F	/* RLOP Line FEBE MSB */
 #define SUNI_TLOP_CTRL		0x20	/* TLOP Control */
 #define SUNI_TLOP_DIAG		0x21	/* TLOP Diagnostic */
-			     /* 0x22-0x2F reserved */
+			     /* 0x22-0x27 reserved */
+#define SUNI_SSTB_CTRL		0x28
 #define SUNI_RPOP_SC		0x30	/* RPOP Status/Control */
 #define SUNI_RPOP_IS		0x31	/* RPOP Interrupt Status */
 			     /* 0x32 reserved */
@@ -52,6 +54,7 @@
 #define SUNI_RPOP_PFM		0x3B	/* RPOP Path FEBE MSB */
 			     /* 0x3C reserved */
 #define SUNI_RPOP_PBC		0x3D	/* RPOP Path BIP-8 Configuration */
+#define SUNI_RPOP_RC		0x3D	/* RPOP Ring Control (PM5355) */
 			     /* 0x3E-0x3F reserved */
 #define SUNI_TPOP_CD		0x40	/* TPOP Control/Diagnostic */
 #define SUNI_TPOP_PC		0x41	/* TPOP Pointer Control */
@@ -82,7 +85,8 @@
 #define SUNI_TACP_TCC		0x65	/* TACP Transmit Cell Counter */
 #define SUNI_TACP_TCCM		0x66	/* TACP Transmit Cell Counter MSB */
 #define SUNI_TACP_CFG		0x67	/* TACP Configuration */
-			     /* 0x68-0x7F reserved */
+#define SUNI_SPTB_CTRL		0x68	/* SPTB Control */
+			     /* 0x69-0x7F reserved */
 #define	SUNI_MT			0x80	/* Master Test */
 			     /* 0x81-0xFF reserved */
 
@@ -94,9 +98,18 @@
 #define SUNI_MRI_ID_SHIFT 	0
 #define SUNI_MRI_TYPE		0x70	/* R, SUNI type (lite is 011) */
 #define SUNI_MRI_TYPE_SHIFT 	4
+#define SUNI_MRI_TYPE_PM5346	0x3	/* S/UNI 155 LITE */
+#define SUNI_MRI_TYPE_PM5347	0x4	/* S/UNI 155 PLUS */
+#define SUNI_MRI_TYPE_PM5350	0x7	/* S/UNI 155 ULTRA */
+#define SUNI_MRI_TYPE_PM5355	0x1	/* S/UNI 622 */
 #define SUNI_MRI_RESET		0x80	/* RW, reset & power down chip
 					   0: normal operation
 					   1: reset & low power */
+
+/* MCM is reg 0x4 */
+#define SUNI_MCM_LLE		0x20	/* line loopback (PM5355) */
+#define SUNI_MCM_DLE		0x10	/* diagnostic loopback (PM5355) */
+
 /* MCT is reg 5 */
 #define SUNI_MCT_LOOPT		0x01	/* RW, timing source, 0: from
 					   TRCLK+/- */
@@ -144,6 +157,12 @@
 /* TLOP_DIAG is reg 0x21 */
 #define SUNI_TLOP_DIAG_DBIP	0x01	/* insert line BIP err (continuously) */
 
+/* SSTB_CTRL is reg 0x28 */
+#define SUNI_SSTB_CTRL_LEN16	0x01	/* path trace message length bit */
+
+/* RPOP_RC is reg 0x3D (PM5355) */
+#define SUNI_RPOP_RC_ENSS	0x40	/* enable size bit */
+
 /* TPOP_DIAG is reg 0x40 */
 #define SUNI_TPOP_DIAG_PAIS	0x01	/* insert STS path alarm ind (cont) */
 #define SUNI_TPOP_DIAG_DB3	0x02	/* insert path BIP err (continuously) */
@@ -191,6 +210,9 @@
 					   pattern */
 #define SUNI_TACP_IUCHP_GFC_SHIFT 4
 
+/* SPTB_CTRL is reg 0x68 */
+#define SUNI_SPTB_CTRL_LEN16	0x01	/* path trace message length */
+
 /* MT is reg 0x80 */
 #define SUNI_MT_HIZIO		0x01	/* RW, all but data bus & MP interface
 					   tri-state */
@@ -208,6 +230,7 @@
 struct suni_priv {
 	struct k_sonet_stats sonet_stats;	/* link diagnostics */
 	int loop_mode;				/* loopback mode */
+	int type;				/* phy type */
 	struct atm_dev *dev;			/* device back-pointer */
 	struct suni_priv *next;			/* next SUNI */
 };
diff --git a/include/linux/sonet.h b/include/linux/sonet.h
index 753680296e17..67ad11fcf88b 100644
--- a/include/linux/sonet.h
+++ b/include/linux/sonet.h
@@ -34,7 +34,7 @@ struct sonet_stats {
 					/* clear error insertion */
 #define SONET_GETDIAG	_IOR('a',ATMIOC_PHYTYP+4,int)
 					/* query error insertion */
-#define SONET_SETFRAMING _IO('a',ATMIOC_PHYTYP+5)
+#define SONET_SETFRAMING _IOW('a',ATMIOC_PHYTYP+5,int)
 					/* set framing mode (SONET/SDH) */
 #define SONET_GETFRAMING _IOR('a',ATMIOC_PHYTYP+6,int)
 					/* get framing mode */
-- 
cgit v1.2.3


From fe2c802ab62aa63d276deafa905875f3455f2621 Mon Sep 17 00:00:00 2001
From: Bernard Pidoux <f6bvp@amsat.org>
Date: Tue, 17 Jun 2008 17:08:32 -0700
Subject: rose: improving AX25 routing frames via ROSE network

ROSE network is organized through nodes connected via hamradio or Internet.
AX25 packet radio frames sent to a remote ROSE address destination are routed
through these nodes.

Without the present patch, automatic routing mechanism did not work optimally
due to an improper parameter checking.

rose_get_neigh() function is called either by rose_connect() or by
rose_route_frame().

In the case of a call from rose_connect(), f0 timer is checked to find if a connection
is already pending. In that case it returns the address of the neighbour, or returns a NULL otherwise.

When called by rose_route_frame() the purpose was to route a packet AX25 frame
through an adjacent node given a destination rose address.
However, in that case, t0 timer checked does not indicate if the adjacent node
is actually connected even if the timer is not null. Thus, for each frame sent, the
function often tried to start a new connexion even if the adjacent node was already connected.

The patch adds a "new" parameter that is true when the function is called by
rose route_frame().
This instructs rose_get_neigh() to check node parameter "restarted".
If restarted is true it means that the route to the destination address is opened via a neighbour
node already connected.
If "restarted" is false the function returns a NULL.
In that case the calling function will initiate a new connection as before.

This results in a fast routing of frames, from nodes to nodes, until
destination is reached, as originaly specified by ROSE protocole.

Signed-off-by: Bernard Pidoux <f6bvp@amsat.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/rose.h    |  2 +-
 net/rose/af_rose.c    |  4 ++--
 net/rose/rose_route.c | 29 ++++++++++++++++++-----------
 3 files changed, 21 insertions(+), 14 deletions(-)

(limited to 'include')

diff --git a/include/net/rose.h b/include/net/rose.h
index e5bb084d8754..cbd5364b2c8a 100644
--- a/include/net/rose.h
+++ b/include/net/rose.h
@@ -201,7 +201,7 @@ extern void rose_link_device_down(struct net_device *);
 extern struct net_device *rose_dev_first(void);
 extern struct net_device *rose_dev_get(rose_address *);
 extern struct rose_route *rose_route_free_lci(unsigned int, struct rose_neigh *);
-extern struct rose_neigh *rose_get_neigh(rose_address *, unsigned char *, unsigned char *);
+extern struct rose_neigh *rose_get_neigh(rose_address *, unsigned char *, unsigned char *, int);
 extern int  rose_rt_ioctl(unsigned int, void __user *);
 extern void rose_link_failed(ax25_cb *, int);
 extern int  rose_route_frame(struct sk_buff *, ax25_cb *);
diff --git a/net/rose/af_rose.c b/net/rose/af_rose.c
index af86bcb604e5..46461a69cd0f 100644
--- a/net/rose/af_rose.c
+++ b/net/rose/af_rose.c
@@ -757,7 +757,7 @@ static int rose_connect(struct socket *sock, struct sockaddr *uaddr, int addr_le
 	sock->state = SS_UNCONNECTED;
 
 	rose->neighbour = rose_get_neigh(&addr->srose_addr, &cause,
-					 &diagnostic);
+					 &diagnostic, 0);
 	if (!rose->neighbour) {
 		err = -ENETUNREACH;
 		goto out_release;
@@ -853,7 +853,7 @@ rose_try_next_neigh:
 
 	if (sk->sk_state != TCP_ESTABLISHED) {
 	/* Try next neighbour */
-		rose->neighbour = rose_get_neigh(&addr->srose_addr, &cause, &diagnostic);
+		rose->neighbour = rose_get_neigh(&addr->srose_addr, &cause, &diagnostic, 0);
 		if (rose->neighbour)
 			goto rose_try_next_neigh;
 
diff --git a/net/rose/rose_route.c b/net/rose/rose_route.c
index bd593871c81e..a81066a1010a 100644
--- a/net/rose/rose_route.c
+++ b/net/rose/rose_route.c
@@ -662,27 +662,34 @@ struct rose_route *rose_route_free_lci(unsigned int lci, struct rose_neigh *neig
 }
 
 /*
- *	Find a neighbour given a ROSE address.
+ *	Find a neighbour or a route given a ROSE address.
  */
 struct rose_neigh *rose_get_neigh(rose_address *addr, unsigned char *cause,
-	unsigned char *diagnostic)
+	unsigned char *diagnostic, int new)
 {
 	struct rose_neigh *res = NULL;
 	struct rose_node *node;
 	int failed = 0;
 	int i;
 
-	spin_lock_bh(&rose_node_list_lock);
+	if (!new) spin_lock_bh(&rose_node_list_lock);
 	for (node = rose_node_list; node != NULL; node = node->next) {
 		if (rosecmpm(addr, &node->address, node->mask) == 0) {
 			for (i = 0; i < node->count; i++) {
-				if (!rose_ftimer_running(node->neighbour[i])) {
-					res = node->neighbour[i];
-					goto out;
-				} else
-					failed = 1;
+				if (new) {
+					if (node->neighbour[i]->restarted) {
+						res = node->neighbour[i];
+						goto out;
+					}
+				}
+				else {
+					if (!rose_ftimer_running(node->neighbour[i])) {
+						res = node->neighbour[i];
+						goto out;
+					} else
+						failed = 1;
+				}
 			}
-			break;
 		}
 	}
 
@@ -695,7 +702,7 @@ struct rose_neigh *rose_get_neigh(rose_address *addr, unsigned char *cause,
 	}
 
 out:
-	spin_unlock_bh(&rose_node_list_lock);
+	if (!new) spin_unlock_bh(&rose_node_list_lock);
 
 	return res;
 }
@@ -1018,7 +1025,7 @@ int rose_route_frame(struct sk_buff *skb, ax25_cb *ax25)
 		rose_route = rose_route->next;
 	}
 
-	if ((new_neigh = rose_get_neigh(dest_addr, &cause, &diagnostic)) == NULL) {
+	if ((new_neigh = rose_get_neigh(dest_addr, &cause, &diagnostic, 1)) == NULL) {
 		rose_transmit_clear_request(rose_neigh, lci, cause, diagnostic);
 		goto out;
 	}
-- 
cgit v1.2.3


From c1da4ac752b8b0411791d26c678fcf23d2eed242 Mon Sep 17 00:00:00 2001
From: Or Gerlitz <ogerlitz@voltaire.com>
Date: Fri, 13 Jun 2008 18:12:00 -0700
Subject: net/core: add NETDEV_BONDING_FAILOVER event

Add NETDEV_BONDING_FAILOVER event to be used in a successive patch
by bonding to announce fail-over for the active-backup mode through the
netdev events notifier chain mechanism. Such an event can be of use for the
RDMA CM (communication manager) to let native RDMA ULPs (eg NFS-RDMA, iSER)
always be aligned with the IP stack, in the sense that they use the same
ports/links as the stack does. More usages can be done to allow monitoring
tools based on netlink events being aware to bonding fail-over.

Signed-off-by: Or Gerlitz <ogerlitz@voltaire.com>
Signed-off-by: Jay Vosburgh <fubar@us.ibm.com>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
---
 include/linux/netdevice.h | 1 +
 include/linux/notifier.h  | 1 +
 net/core/dev.c            | 6 ++++++
 3 files changed, 8 insertions(+)

(limited to 'include')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index f27fd2009334..e92fc839ab1d 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -1479,6 +1479,7 @@ extern void		__dev_addr_unsync(struct dev_addr_list **to, int *to_count, struct
 extern void		dev_set_promiscuity(struct net_device *dev, int inc);
 extern void		dev_set_allmulti(struct net_device *dev, int inc);
 extern void		netdev_state_change(struct net_device *dev);
+extern void		netdev_bonding_change(struct net_device *dev);
 extern void		netdev_features_change(struct net_device *dev);
 /* Load a device via the kmod */
 extern void		dev_load(struct net *net, const char *name);
diff --git a/include/linux/notifier.h b/include/linux/notifier.h
index 0ff6224d172a..bd3d72ddf333 100644
--- a/include/linux/notifier.h
+++ b/include/linux/notifier.h
@@ -197,6 +197,7 @@ static inline int notifier_to_errno(int ret)
 #define NETDEV_GOING_DOWN	0x0009
 #define NETDEV_CHANGENAME	0x000A
 #define NETDEV_FEAT_CHANGE	0x000B
+#define NETDEV_BONDING_FAILOVER 0x000C
 
 #define SYS_DOWN	0x0001	/* Notify of system down */
 #define SYS_RESTART	SYS_DOWN
diff --git a/net/core/dev.c b/net/core/dev.c
index 68d8df0992ab..0e45742e7158 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -961,6 +961,12 @@ void netdev_state_change(struct net_device *dev)
 	}
 }
 
+void netdev_bonding_change(struct net_device *dev)
+{
+	call_netdevice_notifiers(NETDEV_BONDING_FAILOVER, dev);
+}
+EXPORT_SYMBOL(netdev_bonding_change);
+
 /**
  *	dev_load 	- load a network module
  *	@net: the applicable net namespace
-- 
cgit v1.2.3


From b8a9787eddb0e4665f31dd1d64584732b2b5d051 Mon Sep 17 00:00:00 2001
From: Jay Vosburgh <fubar@us.ibm.com>
Date: Fri, 13 Jun 2008 18:12:04 -0700
Subject: bonding: Allow setting max_bonds to zero

	Permit bonding to function rationally if max_bonds is set to
zero.  This will load the module, but create no master devices (which can
be created via sysfs).

	Requires some change to bond_create_sysfs; currently, the
netdev sysfs directory is determined from the first bonding device created,
but this is no longer possible.  Instead, an interface from net/core is
created to create and destroy files in net_class.

	Based on a patch submitted by Phil Oester <kernel@linuxaces.com>.
Modified by Jay Vosburgh to fix the sysfs issue mentioned above and to
update the documentation.

Signed-off-by: Phil Oester <kernel@linuxace.com>
Signed-off-by: Jay Vosburgh <fubar@us.ibm.com>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
---
 Documentation/networking/bonding.txt |  3 ++-
 drivers/net/bonding/bond_main.c      |  6 +++---
 drivers/net/bonding/bond_sysfs.c     | 22 +++-------------------
 include/linux/netdevice.h            |  3 +++
 net/core/net-sysfs.c                 | 13 +++++++++++++
 5 files changed, 24 insertions(+), 23 deletions(-)

(limited to 'include')

diff --git a/Documentation/networking/bonding.txt b/Documentation/networking/bonding.txt
index 370b7da73ab4..7fa7fe71d7a8 100644
--- a/Documentation/networking/bonding.txt
+++ b/Documentation/networking/bonding.txt
@@ -376,7 +376,8 @@ max_bonds
 	Specifies the number of bonding devices to create for this
 	instance of the bonding driver.  E.g., if max_bonds is 3, and
 	the bonding driver is not already loaded, then bond0, bond1
-	and bond2 will be created.  The default value is 1.
+	and bond2 will be created.  The default value is 1.  Specifying
+	a value of 0 will load bonding, but will not create any devices.
 
 miimon
 
diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
index 3b6d66a8ab98..d57b65dc2c72 100644
--- a/drivers/net/bonding/bond_main.c
+++ b/drivers/net/bonding/bond_main.c
@@ -4750,11 +4750,11 @@ static int bond_check_params(struct bond_params *params)
 		}
 	}
 
-	if (max_bonds < 1 || max_bonds > INT_MAX) {
+	if (max_bonds < 0 || max_bonds > INT_MAX) {
 		printk(KERN_WARNING DRV_NAME
 		       ": Warning: max_bonds (%d) not in range %d-%d, so it "
 		       "was reset to BOND_DEFAULT_MAX_BONDS (%d)\n",
-		       max_bonds, 1, INT_MAX, BOND_DEFAULT_MAX_BONDS);
+		       max_bonds, 0, INT_MAX, BOND_DEFAULT_MAX_BONDS);
 		max_bonds = BOND_DEFAULT_MAX_BONDS;
 	}
 
@@ -4953,7 +4953,7 @@ static int bond_check_params(struct bond_params *params)
 
 		printk("\n");
 
-	} else {
+	} else if (max_bonds) {
 		/* miimon and arp_interval not set, we need one so things
 		 * work as expected, see bonding.txt for details
 		 */
diff --git a/drivers/net/bonding/bond_sysfs.c b/drivers/net/bonding/bond_sysfs.c
index dd265c69b0df..6caac0ffb2f2 100644
--- a/drivers/net/bonding/bond_sysfs.c
+++ b/drivers/net/bonding/bond_sysfs.c
@@ -53,7 +53,6 @@ extern struct bond_parm_tbl arp_validate_tbl[];
 extern struct bond_parm_tbl fail_over_mac_tbl[];
 
 static int expected_refcount = -1;
-static struct class *netdev_class;
 /*--------------------------- Data Structures -----------------------------*/
 
 /* Bonding sysfs lock.  Why can't we just use the subsystem lock?
@@ -1447,19 +1446,9 @@ static struct attribute_group bonding_group = {
  */
 int bond_create_sysfs(void)
 {
-	int ret = 0;
-	struct bonding *firstbond;
-
-	/* get the netdev class pointer */
-	firstbond = container_of(bond_dev_list.next, struct bonding, bond_list);
-	if (!firstbond)
-		return -ENODEV;
-
-	netdev_class = firstbond->dev->dev.class;
-	if (!netdev_class)
-		return -ENODEV;
+	int ret;
 
-	ret = class_create_file(netdev_class, &class_attr_bonding_masters);
+	ret = netdev_class_create_file(&class_attr_bonding_masters);
 	/*
 	 * Permit multiple loads of the module by ignoring failures to
 	 * create the bonding_masters sysfs file.  Bonding devices
@@ -1478,10 +1467,6 @@ int bond_create_sysfs(void)
 			printk(KERN_ERR
 			       "network device named %s already exists in sysfs",
 			       class_attr_bonding_masters.attr.name);
-		else {
-			netdev_class = NULL;
-			return 0;
-		}
 	}
 
 	return ret;
@@ -1493,8 +1478,7 @@ int bond_create_sysfs(void)
  */
 void bond_destroy_sysfs(void)
 {
-	if (netdev_class)
-		class_remove_file(netdev_class, &class_attr_bonding_masters);
+	netdev_class_remove_file(&class_attr_bonding_masters);
 }
 
 /*
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index e92fc839ab1d..9ccbfac3fd95 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -1506,6 +1506,9 @@ extern void *dev_seq_next(struct seq_file *seq, void *v, loff_t *pos);
 extern void dev_seq_stop(struct seq_file *seq, void *v);
 #endif
 
+extern int netdev_class_create_file(struct class_attribute *class_attr);
+extern void netdev_class_remove_file(struct class_attribute *class_attr);
+
 extern void linkwatch_run_queue(void);
 
 extern int netdev_compute_features(unsigned long all, unsigned long one);
diff --git a/net/core/net-sysfs.c b/net/core/net-sysfs.c
index dccd737ea2e3..3f7941319217 100644
--- a/net/core/net-sysfs.c
+++ b/net/core/net-sysfs.c
@@ -468,6 +468,19 @@ int netdev_register_kobject(struct net_device *net)
 	return device_add(dev);
 }
 
+int netdev_class_create_file(struct class_attribute *class_attr)
+{
+	return class_create_file(&net_class, class_attr);
+}
+
+void netdev_class_remove_file(struct class_attribute *class_attr)
+{
+	class_remove_file(&net_class, class_attr);
+}
+
+EXPORT_SYMBOL(netdev_class_create_file);
+EXPORT_SYMBOL(netdev_class_remove_file);
+
 void netdev_initialize_kobject(struct net_device *net)
 {
 	struct device *device = &(net->dev);
-- 
cgit v1.2.3


From cb61cb9b8b5ef6c2697d84e5015e314626eb2fba Mon Sep 17 00:00:00 2001
From: Eric Dumazet <dada1@cosmosbay.com>
Date: Tue, 17 Jun 2008 21:04:56 -0700
Subject: udp: sk_drops handling

In commits 33c732c36169d7022ad7d6eb474b0c9be43a2dc1 ([IPV4]: Add raw
drops counter) and a92aa318b4b369091fd80433c80e62838db8bc1c ([IPV6]:
Add raw drops counter), Wang Chen added raw drops counter for
/proc/net/raw & /proc/net/raw6

This patch adds this capability to UDP sockets too (/proc/net/udp &
/proc/net/udp6).

This means that 'RcvbufErrors' errors found in /proc/net/snmp can be also
be examined for each udp socket.

# grep Udp: /proc/net/snmp
Udp: InDatagrams NoPorts InErrors OutDatagrams RcvbufErrors SndbufErrors
Udp: 23971006 75 899420 16390693 146348 0

# cat /proc/net/udp
 sl  local_address rem_address   st tx_queue rx_queue tr tm->when retrnsmt  ---
uid  timeout inode ref pointer drops
 75: 00000000:02CB 00000000:0000 07 00000000:00000000 00:00000000 00000000  ---
  0        0 2358 2 ffff81082a538c80 0
111: 00000000:006F 00000000:0000 07 00000000:00000000 00:00000000 00000000  ---
  0        0 2286 2 ffff81042dd35c80 146348

In this example, only port 111 (0x006F) was flooded by messages that
user program could not read fast enough. 146348 messages were lost.

Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/sock.h |  2 +-
 net/ipv4/raw.c     |  2 +-
 net/ipv4/udp.c     | 11 +++++++----
 net/ipv6/raw.c     |  2 +-
 net/ipv6/udp.c     | 11 +++++++----
 5 files changed, 17 insertions(+), 11 deletions(-)

(limited to 'include')

diff --git a/include/net/sock.h b/include/net/sock.h
index 83f74b11d09a..a7c30412de66 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -166,7 +166,7 @@ struct sock_common {
   *	@sk_err: last error
   *	@sk_err_soft: errors that don't cause failure but are the cause of a
   *		      persistent failure not just 'timed out'
-  *	@sk_drops: raw drops counter
+  *	@sk_drops: raw/udp drops counter
   *	@sk_ack_backlog: current listen backlog
   *	@sk_max_ack_backlog: listen backlog set in listen()
   *	@sk_priority: %SO_PRIORITY setting
diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c
index 7d449468409e..925fdf18cf92 100644
--- a/net/ipv4/raw.c
+++ b/net/ipv4/raw.c
@@ -944,7 +944,7 @@ static int raw_seq_show(struct seq_file *seq, void *v)
 	if (v == SEQ_START_TOKEN)
 		seq_printf(seq, "  sl  local_address rem_address   st tx_queue "
 				"rx_queue tr tm->when retrnsmt   uid  timeout "
-				"inode  drops\n");
+				"inode ref pointer drops\n");
 	else
 		raw_sock_seq_show(seq, v, raw_seq_private(seq)->bucket);
 	return 0;
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index 11eabf136144..3bbf6fb6e4f5 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -1040,8 +1040,10 @@ int udp_queue_rcv_skb(struct sock * sk, struct sk_buff *skb)
 
 	if ((rc = sock_queue_rcv_skb(sk,skb)) < 0) {
 		/* Note that an ENOMEM error is charged twice */
-		if (rc == -ENOMEM)
+		if (rc == -ENOMEM) {
 			UDP_INC_STATS_BH(UDP_MIB_RCVBUFERRORS, is_udplite);
+			atomic_inc(&sk->sk_drops);
+		}
 		goto drop;
 	}
 
@@ -1629,12 +1631,13 @@ static void udp4_format_sock(struct sock *sp, struct seq_file *f,
 	__u16 srcp	  = ntohs(inet->sport);
 
 	seq_printf(f, "%4d: %08X:%04X %08X:%04X"
-		" %02X %08X:%08X %02X:%08lX %08X %5d %8d %lu %d %p%n",
+		" %02X %08X:%08X %02X:%08lX %08X %5d %8d %lu %d %p %d%n",
 		bucket, src, srcp, dest, destp, sp->sk_state,
 		atomic_read(&sp->sk_wmem_alloc),
 		atomic_read(&sp->sk_rmem_alloc),
 		0, 0L, 0, sock_i_uid(sp), 0, sock_i_ino(sp),
-		atomic_read(&sp->sk_refcnt), sp, len);
+		atomic_read(&sp->sk_refcnt), sp,
+		atomic_read(&sp->sk_drops), len);
 }
 
 int udp4_seq_show(struct seq_file *seq, void *v)
@@ -1643,7 +1646,7 @@ int udp4_seq_show(struct seq_file *seq, void *v)
 		seq_printf(seq, "%-127s\n",
 			   "  sl  local_address rem_address   st tx_queue "
 			   "rx_queue tr tm->when retrnsmt   uid  timeout "
-			   "inode");
+			   "inode ref pointer drops");
 	else {
 		struct udp_iter_state *state = seq->private;
 		int len;
diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c
index 456777d7a407..34cfb3f41c2c 100644
--- a/net/ipv6/raw.c
+++ b/net/ipv6/raw.c
@@ -1251,7 +1251,7 @@ static int raw6_seq_show(struct seq_file *seq, void *v)
 			   "local_address                         "
 			   "remote_address                        "
 			   "st tx_queue rx_queue tr tm->when retrnsmt"
-			   "   uid  timeout inode  drops\n");
+			   "   uid  timeout inode ref pointer drops\n");
 	else
 		raw6_sock_seq_show(seq, v, raw_seq_private(seq)->bucket);
 	return 0;
diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c
index 432edaa882f6..f91e1df0d25e 100644
--- a/net/ipv6/udp.c
+++ b/net/ipv6/udp.c
@@ -297,8 +297,10 @@ int udpv6_queue_rcv_skb(struct sock * sk, struct sk_buff *skb)
 
 	if ((rc = sock_queue_rcv_skb(sk,skb)) < 0) {
 		/* Note that an ENOMEM error is charged twice */
-		if (rc == -ENOMEM)
+		if (rc == -ENOMEM) {
 			UDP6_INC_STATS_BH(UDP_MIB_RCVBUFERRORS, is_udplite);
+			atomic_inc(&sk->sk_drops);
+		}
 		goto drop;
 	}
 
@@ -955,7 +957,7 @@ static void udp6_sock_seq_show(struct seq_file *seq, struct sock *sp, int bucket
 	srcp  = ntohs(inet->sport);
 	seq_printf(seq,
 		   "%4d: %08X%08X%08X%08X:%04X %08X%08X%08X%08X:%04X "
-		   "%02X %08X:%08X %02X:%08lX %08X %5d %8d %lu %d %p\n",
+		   "%02X %08X:%08X %02X:%08lX %08X %5d %8d %lu %d %p %d\n",
 		   bucket,
 		   src->s6_addr32[0], src->s6_addr32[1],
 		   src->s6_addr32[2], src->s6_addr32[3], srcp,
@@ -967,7 +969,8 @@ static void udp6_sock_seq_show(struct seq_file *seq, struct sock *sp, int bucket
 		   0, 0L, 0,
 		   sock_i_uid(sp), 0,
 		   sock_i_ino(sp),
-		   atomic_read(&sp->sk_refcnt), sp);
+		   atomic_read(&sp->sk_refcnt), sp,
+		   atomic_read(&sp->sk_drops));
 }
 
 int udp6_seq_show(struct seq_file *seq, void *v)
@@ -978,7 +981,7 @@ int udp6_seq_show(struct seq_file *seq, void *v)
 			   "local_address                         "
 			   "remote_address                        "
 			   "st tx_queue rx_queue tr tm->when retrnsmt"
-			   "   uid  timeout inode\n");
+			   "   uid  timeout inode ref pointer drops\n");
 	else
 		udp6_sock_seq_show(seq, v, ((struct udp_iter_state *)seq->private)->bucket);
 	return 0;
-- 
cgit v1.2.3


From 9d45abe1c2949183e5d9cb25721bf1c42c7b5e3b Mon Sep 17 00:00:00 2001
From: Wang Chen <wangchen@cn.fujitsu.com>
Date: Tue, 17 Jun 2008 21:12:48 -0700
Subject: netdevice: change net_device->promiscuity/allmulti to unsigned int

The comments of dev_set_allmulti/promiscuity() is that "While the count in
the device remains above zero...". So negative count is useless.
Fix the type of the counter from "int" to "unsigned int".

Signed-off-by: Wang Chen <wangchen@cn.fujitsu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index f27fd2009334..06d8ea5992df 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -595,8 +595,8 @@ struct net_device
 	int			uc_promisc;
 	struct dev_addr_list	*mc_list;	/* Multicast mac addresses	*/
 	int			mc_count;	/* Number of installed mcasts	*/
-	int			promiscuity;
-	int			allmulti;
+	unsigned int		promiscuity;
+	unsigned int		allmulti;
 
 
 	/* Protocol specific pointers */
-- 
cgit v1.2.3


From 972692e0db9b0a62329ca394062b58917ddbd03c Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Tue, 17 Jun 2008 22:41:38 -0700
Subject: net: Add sk_set_socket() helper.

In order to more easily grep for all things that set
sk->sk_socket, add sk_set_socket() helper inline function.

Suggested (although only half-seriously) by Evgeniy Polyakov.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/sock.h | 9 +++++++--
 net/core/sock.c    | 4 ++--
 2 files changed, 9 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/include/net/sock.h b/include/net/sock.h
index a7c30412de66..3f4897ab432e 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -990,6 +990,11 @@ static inline void sock_put(struct sock *sk)
 extern int sk_receive_skb(struct sock *sk, struct sk_buff *skb,
 			  const int nested);
 
+static inline void sk_set_socket(struct sock *sk, struct socket *sock)
+{
+	sk->sk_socket = sock;
+}
+
 /* Detach socket from process context.
  * Announce socket dead, detach it from wait queue and inode.
  * Note that parent inode held reference count on this struct sock,
@@ -1001,7 +1006,7 @@ static inline void sock_orphan(struct sock *sk)
 {
 	write_lock_bh(&sk->sk_callback_lock);
 	sock_set_flag(sk, SOCK_DEAD);
-	sk->sk_socket = NULL;
+	sk_set_socket(sk, NULL);
 	sk->sk_sleep  = NULL;
 	write_unlock_bh(&sk->sk_callback_lock);
 }
@@ -1011,7 +1016,7 @@ static inline void sock_graft(struct sock *sk, struct socket *parent)
 	write_lock_bh(&sk->sk_callback_lock);
 	sk->sk_sleep = &parent->wait;
 	parent->sk = sk;
-	sk->sk_socket = parent;
+	sk_set_socket(sk, parent);
 	security_sock_graft(sk, parent);
 	write_unlock_bh(&sk->sk_callback_lock);
 }
diff --git a/net/core/sock.c b/net/core/sock.c
index 3879bf65897e..2c0ba52e5303 100644
--- a/net/core/sock.c
+++ b/net/core/sock.c
@@ -1066,7 +1066,7 @@ struct sock *sk_clone(const struct sock *sk, const gfp_t priority)
 		 * to be taken into account in all callers. -acme
 		 */
 		sk_refcnt_debug_inc(newsk);
-		newsk->sk_socket = NULL;
+		sk_set_socket(newsk, NULL);
 		newsk->sk_sleep	 = NULL;
 
 		if (newsk->sk_prot->sockets_allocated)
@@ -1702,7 +1702,7 @@ void sock_init_data(struct socket *sock, struct sock *sk)
 	sk->sk_rcvbuf		=	sysctl_rmem_default;
 	sk->sk_sndbuf		=	sysctl_wmem_default;
 	sk->sk_state		=	TCP_CLOSE;
-	sk->sk_socket		=	sock;
+	sk_set_socket(sk, sock);
 
 	sock_set_flag(sk, SOCK_ZAPPED);
 
-- 
cgit v1.2.3


From dad9b335c6940de2746a9788eb456d09cf102f81 Mon Sep 17 00:00:00 2001
From: Wang Chen <wangchen@cn.fujitsu.com>
Date: Wed, 18 Jun 2008 01:48:28 -0700
Subject: netdevice: Fix promiscuity and allmulti overflow

Max of promiscuity and allmulti plus positive @inc can cause overflow.
Fox example: when allmulti=0xFFFFFFFF, any caller give dev_set_allmulti() a
positive @inc will cause allmulti be off.
This is not what we want, though it's rare case.
The fix is that only negative @inc will cause allmulti or promiscuity be off
and when any caller makes the counters touch the roof, we return error.

Change of v2:
Change void function dev_set_promiscuity/allmulti to return int.
So callers can get the overflow error.
Caller's fix will be done later.

Change of v3:
1. Since we return error to caller, we don't need to print KERN_ERROR,
KERN_WARNING is enough.
2. In dev_set_promiscuity(), if __dev_set_promiscuity() failed, we
return at once.

Signed-off-by: Wang Chen <wangchen@cn.fujitsu.com>
Acked-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h |  4 ++--
 net/core/dev.c            | 55 ++++++++++++++++++++++++++++++++++++++---------
 2 files changed, 47 insertions(+), 12 deletions(-)

(limited to 'include')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 4bf613cd9e2d..45dce2b58d4c 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -1476,8 +1476,8 @@ extern int 		__dev_addr_delete(struct dev_addr_list **list, int *count, void *ad
 extern int		__dev_addr_add(struct dev_addr_list **list, int *count, void *addr, int alen, int newonly);
 extern int		__dev_addr_sync(struct dev_addr_list **to, int *to_count, struct dev_addr_list **from, int *from_count);
 extern void		__dev_addr_unsync(struct dev_addr_list **to, int *to_count, struct dev_addr_list **from, int *from_count);
-extern void		dev_set_promiscuity(struct net_device *dev, int inc);
-extern void		dev_set_allmulti(struct net_device *dev, int inc);
+extern int		dev_set_promiscuity(struct net_device *dev, int inc);
+extern int		dev_set_allmulti(struct net_device *dev, int inc);
 extern void		netdev_state_change(struct net_device *dev);
 extern void		netdev_bonding_change(struct net_device *dev);
 extern void		netdev_features_change(struct net_device *dev);
diff --git a/net/core/dev.c b/net/core/dev.c
index 0e45742e7158..a495f712d38c 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -2771,16 +2771,29 @@ int netdev_set_master(struct net_device *slave, struct net_device *master)
 	return 0;
 }
 
-static void __dev_set_promiscuity(struct net_device *dev, int inc)
+static int __dev_set_promiscuity(struct net_device *dev, int inc)
 {
 	unsigned short old_flags = dev->flags;
 
 	ASSERT_RTNL();
 
-	if ((dev->promiscuity += inc) == 0)
-		dev->flags &= ~IFF_PROMISC;
-	else
-		dev->flags |= IFF_PROMISC;
+	dev->flags |= IFF_PROMISC;
+	dev->promiscuity += inc;
+	if (dev->promiscuity == 0) {
+		/*
+		 * Avoid overflow.
+		 * If inc causes overflow, untouch promisc and return error.
+		 */
+		if (inc < 0)
+			dev->flags &= ~IFF_PROMISC;
+		else {
+			dev->promiscuity -= inc;
+			printk(KERN_WARNING "%s: promiscuity touches roof, "
+				"set promiscuity failed, promiscuity feature "
+				"of device might be broken.\n", dev->name);
+			return -EOVERFLOW;
+		}
+	}
 	if (dev->flags != old_flags) {
 		printk(KERN_INFO "device %s %s promiscuous mode\n",
 		       dev->name, (dev->flags & IFF_PROMISC) ? "entered" :
@@ -2798,6 +2811,7 @@ static void __dev_set_promiscuity(struct net_device *dev, int inc)
 		if (dev->change_rx_flags)
 			dev->change_rx_flags(dev, IFF_PROMISC);
 	}
+	return 0;
 }
 
 /**
@@ -2809,14 +2823,19 @@ static void __dev_set_promiscuity(struct net_device *dev, int inc)
  *	remains above zero the interface remains promiscuous. Once it hits zero
  *	the device reverts back to normal filtering operation. A negative inc
  *	value is used to drop promiscuity on the device.
+ *	Return 0 if successful or a negative errno code on error.
  */
-void dev_set_promiscuity(struct net_device *dev, int inc)
+int dev_set_promiscuity(struct net_device *dev, int inc)
 {
 	unsigned short old_flags = dev->flags;
+	int err;
 
-	__dev_set_promiscuity(dev, inc);
+	err = __dev_set_promiscuity(dev, inc);
+	if (!err)
+		return err;
 	if (dev->flags != old_flags)
 		dev_set_rx_mode(dev);
+	return err;
 }
 
 /**
@@ -2829,22 +2848,38 @@ void dev_set_promiscuity(struct net_device *dev, int inc)
  *	to all interfaces. Once it hits zero the device reverts back to normal
  *	filtering operation. A negative @inc value is used to drop the counter
  *	when releasing a resource needing all multicasts.
+ *	Return 0 if successful or a negative errno code on error.
  */
 
-void dev_set_allmulti(struct net_device *dev, int inc)
+int dev_set_allmulti(struct net_device *dev, int inc)
 {
 	unsigned short old_flags = dev->flags;
 
 	ASSERT_RTNL();
 
 	dev->flags |= IFF_ALLMULTI;
-	if ((dev->allmulti += inc) == 0)
-		dev->flags &= ~IFF_ALLMULTI;
+	dev->allmulti += inc;
+	if (dev->allmulti == 0) {
+		/*
+		 * Avoid overflow.
+		 * If inc causes overflow, untouch allmulti and return error.
+		 */
+		if (inc < 0)
+			dev->flags &= ~IFF_ALLMULTI;
+		else {
+			dev->allmulti -= inc;
+			printk(KERN_WARNING "%s: allmulti touches roof, "
+				"set allmulti failed, allmulti feature of "
+				"device might be broken.\n", dev->name);
+			return -EOVERFLOW;
+		}
+	}
 	if (dev->flags ^ old_flags) {
 		if (dev->change_rx_flags)
 			dev->change_rx_flags(dev, IFF_ALLMULTI);
 		dev_set_rx_mode(dev);
 	}
+	return 0;
 }
 
 /*
-- 
cgit v1.2.3


From 2e3216cd54b142ba605e87522e15f42e0c4e3996 Mon Sep 17 00:00:00 2001
From: Vlad Yasevich <vladislav.yasevich@hp.com>
Date: Thu, 19 Jun 2008 16:08:18 -0700
Subject: sctp: Follow security requirement of responding with 1 packet

RFC 4960, Section 11.4. Protection of Non-SCTP-Capable Hosts

When an SCTP stack receives a packet containing multiple control or
DATA chunks and the processing of the packet requires the sending of
multiple chunks in response, the sender of the response chunk(s) MUST
NOT send more than one packet.  If bundling is supported, multiple
response chunks that fit into a single packet MAY be bundled together
into one single response packet.  If bundling is not supported, then
the sender MUST NOT send more than one response chunk and MUST
discard all other responses.  Note that this rule does NOT apply to a
SACK chunk, since a SACK chunk is, in itself, a response to DATA and
a SACK does not require a response of more DATA.

We implement this by not servicing our outqueue until we reach the end
of the packet.  This enables maximum bundling.  We also identify
'response' chunks and make sure that we only send 1 packet when sending
such chunks.

Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/sctp/structs.h |  2 +-
 net/sctp/associola.c       |  1 +
 net/sctp/output.c          |  7 +++++--
 net/sctp/outqueue.c        | 34 ++++++++++++++++++++++++----------
 net/sctp/sm_sideeffect.c   | 27 ++++++++++++++++++---------
 net/sctp/sm_statefuns.c    | 16 +++++-----------
 6 files changed, 54 insertions(+), 33 deletions(-)

(limited to 'include')

diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h
index fbc27ac8a09e..82116e84ee34 100644
--- a/include/net/sctp/structs.h
+++ b/include/net/sctp/structs.h
@@ -827,7 +827,7 @@ struct sctp_packet *sctp_packet_init(struct sctp_packet *,
 				     __u16 sport, __u16 dport);
 struct sctp_packet *sctp_packet_config(struct sctp_packet *, __u32 vtag, int);
 sctp_xmit_t sctp_packet_transmit_chunk(struct sctp_packet *,
-                                       struct sctp_chunk *);
+                                       struct sctp_chunk *, int);
 sctp_xmit_t sctp_packet_append_chunk(struct sctp_packet *,
                                      struct sctp_chunk *);
 int sctp_packet_transmit(struct sctp_packet *);
diff --git a/net/sctp/associola.c b/net/sctp/associola.c
index 35b6a023a6d0..ff1dc5bd936c 100644
--- a/net/sctp/associola.c
+++ b/net/sctp/associola.c
@@ -1025,6 +1025,7 @@ static void sctp_assoc_bh_rcv(struct work_struct *work)
 	struct sctp_chunk *chunk;
 	struct sock *sk;
 	struct sctp_inq *inqueue;
+	struct sctp_outq *outq;
 	int state;
 	sctp_subtype_t subtype;
 	int error = 0;
diff --git a/net/sctp/output.c b/net/sctp/output.c
index 6d45bae93b46..abcd00dc05eb 100644
--- a/net/sctp/output.c
+++ b/net/sctp/output.c
@@ -157,7 +157,8 @@ void sctp_packet_free(struct sctp_packet *packet)
  * packet can be sent only after receiving the COOKIE_ACK.
  */
 sctp_xmit_t sctp_packet_transmit_chunk(struct sctp_packet *packet,
-				       struct sctp_chunk *chunk)
+				       struct sctp_chunk *chunk,
+				       int one_packet)
 {
 	sctp_xmit_t retval;
 	int error = 0;
@@ -175,7 +176,9 @@ sctp_xmit_t sctp_packet_transmit_chunk(struct sctp_packet *packet,
 			/* If we have an empty packet, then we can NOT ever
 			 * return PMTU_FULL.
 			 */
-			retval = sctp_packet_append_chunk(packet, chunk);
+			if (!one_packet)
+				retval = sctp_packet_append_chunk(packet,
+								  chunk);
 		}
 		break;
 
diff --git a/net/sctp/outqueue.c b/net/sctp/outqueue.c
index ace6770e9048..70ead8dc3485 100644
--- a/net/sctp/outqueue.c
+++ b/net/sctp/outqueue.c
@@ -702,6 +702,7 @@ int sctp_outq_uncork(struct sctp_outq *q)
 	return error;
 }
 
+
 /*
  * Try to flush an outqueue.
  *
@@ -725,6 +726,7 @@ int sctp_outq_flush(struct sctp_outq *q, int rtx_timeout)
 	sctp_xmit_t status;
 	int error = 0;
 	int start_timer = 0;
+	int one_packet = 0;
 
 	/* These transports have chunks to send. */
 	struct list_head transport_list;
@@ -830,20 +832,33 @@ int sctp_outq_flush(struct sctp_outq *q, int rtx_timeout)
 			if (sctp_test_T_bit(chunk)) {
 				packet->vtag = asoc->c.my_vtag;
 			}
-		case SCTP_CID_SACK:
-		case SCTP_CID_HEARTBEAT:
+		/* The following chunks are "response" chunks, i.e.
+		 * they are generated in response to something we
+		 * received.  If we are sending these, then we can
+		 * send only 1 packet containing these chunks.
+		 */
 		case SCTP_CID_HEARTBEAT_ACK:
-		case SCTP_CID_SHUTDOWN:
 		case SCTP_CID_SHUTDOWN_ACK:
-		case SCTP_CID_ERROR:
-		case SCTP_CID_COOKIE_ECHO:
 		case SCTP_CID_COOKIE_ACK:
-		case SCTP_CID_ECN_ECNE:
+		case SCTP_CID_COOKIE_ECHO:
+		case SCTP_CID_ERROR:
 		case SCTP_CID_ECN_CWR:
-		case SCTP_CID_ASCONF:
 		case SCTP_CID_ASCONF_ACK:
+			one_packet = 1;
+			/* Fall throught */
+
+		case SCTP_CID_SACK:
+		case SCTP_CID_HEARTBEAT:
+		case SCTP_CID_SHUTDOWN:
+		case SCTP_CID_ECN_ECNE:
+		case SCTP_CID_ASCONF:
 		case SCTP_CID_FWD_TSN:
-			sctp_packet_transmit_chunk(packet, chunk);
+			status = sctp_packet_transmit_chunk(packet, chunk,
+							    one_packet);
+			if (status  != SCTP_XMIT_OK) {
+				/* put the chunk back */
+				list_add(&chunk->list, &q->control_chunk_list);
+			}
 			break;
 
 		default:
@@ -974,7 +989,7 @@ int sctp_outq_flush(struct sctp_outq *q, int rtx_timeout)
 					atomic_read(&chunk->skb->users) : -1);
 
 			/* Add the chunk to the packet.  */
-			status = sctp_packet_transmit_chunk(packet, chunk);
+			status = sctp_packet_transmit_chunk(packet, chunk, 0);
 
 			switch (status) {
 			case SCTP_XMIT_PMTU_FULL:
@@ -1239,7 +1254,6 @@ int sctp_outq_sack(struct sctp_outq *q, struct sctp_sackhdr *sack)
 	 * Make sure the empty queue handler will get run later.
 	 */
 	q->empty = (list_empty(&q->out_chunk_list) &&
-		    list_empty(&q->control_chunk_list) &&
 		    list_empty(&q->retransmit));
 	if (!q->empty)
 		goto finish;
diff --git a/net/sctp/sm_sideeffect.c b/net/sctp/sm_sideeffect.c
index b083312c725a..9732c797e8ed 100644
--- a/net/sctp/sm_sideeffect.c
+++ b/net/sctp/sm_sideeffect.c
@@ -664,7 +664,7 @@ static int sctp_cmd_process_sack(sctp_cmd_seq_t *cmds,
 				 struct sctp_association *asoc,
 				 struct sctp_sackhdr *sackh)
 {
-	int err;
+	int err = 0;
 
 	if (sctp_outq_sack(&asoc->outqueue, sackh)) {
 		/* There are no more TSNs awaiting SACK.  */
@@ -672,11 +672,6 @@ static int sctp_cmd_process_sack(sctp_cmd_seq_t *cmds,
 				 SCTP_ST_OTHER(SCTP_EVENT_NO_PENDING_TSN),
 				 asoc->state, asoc->ep, asoc, NULL,
 				 GFP_ATOMIC);
-	} else {
-		/* Windows may have opened, so we need
-		 * to check if we have DATA to transmit
-		 */
-		err = sctp_outq_flush(&asoc->outqueue, 0);
 	}
 
 	return err;
@@ -1481,8 +1476,15 @@ static int sctp_cmd_interpreter(sctp_event_t event_type,
 			break;
 
 		case SCTP_CMD_DISCARD_PACKET:
-			/* We need to discard the whole packet.  */
+			/* We need to discard the whole packet.
+			 * Uncork the queue since there might be
+			 * responses pending
+			 */
 			chunk->pdiscard = 1;
+			if (asoc) {
+				sctp_outq_uncork(&asoc->outqueue);
+				local_cork = 0;
+			}
 			break;
 
 		case SCTP_CMD_RTO_PENDING:
@@ -1553,8 +1555,15 @@ static int sctp_cmd_interpreter(sctp_event_t event_type,
 	}
 
 out:
-	if (local_cork)
-		sctp_outq_uncork(&asoc->outqueue);
+	/* If this is in response to a received chunk, wait until
+	 * we are done with the packet to open the queue so that we don't
+	 * send multiple packets in response to a single request.
+	 */
+	if (asoc && SCTP_EVENT_T_CHUNK == event_type && chunk) {
+		if (chunk->end_of_packet || chunk->singleton)
+			sctp_outq_uncork(&asoc->outqueue);
+	} else if (local_cork)
+			sctp_outq_uncork(&asoc->outqueue);
 	return error;
 nomem:
 	error = -ENOMEM;
diff --git a/net/sctp/sm_statefuns.c b/net/sctp/sm_statefuns.c
index 0c9d5a6950fe..b66a41d03c0d 100644
--- a/net/sctp/sm_statefuns.c
+++ b/net/sctp/sm_statefuns.c
@@ -795,8 +795,6 @@ sctp_disposition_t sctp_sf_do_5_1D_ce(const struct sctp_endpoint *ep,
 		sctp_add_cmd_sf(commands, SCTP_CMD_TIMER_START,
 				SCTP_TO(SCTP_EVENT_TIMEOUT_AUTOCLOSE));
 
-	sctp_add_cmd_sf(commands, SCTP_CMD_TRANSMIT, SCTP_NULL());
-
 	/* This will send the COOKIE ACK */
 	sctp_add_cmd_sf(commands, SCTP_CMD_REPLY, SCTP_CHUNK(repl));
 
@@ -883,7 +881,6 @@ sctp_disposition_t sctp_sf_do_5_1E_ca(const struct sctp_endpoint *ep,
 	if (asoc->autoclose)
 		sctp_add_cmd_sf(commands, SCTP_CMD_TIMER_START,
 				SCTP_TO(SCTP_EVENT_TIMEOUT_AUTOCLOSE));
-	sctp_add_cmd_sf(commands, SCTP_CMD_TRANSMIT, SCTP_NULL());
 
 	/* It may also notify its ULP about the successful
 	 * establishment of the association with a Communication Up
@@ -1781,7 +1778,6 @@ static sctp_disposition_t sctp_sf_do_dupcook_b(const struct sctp_endpoint *ep,
 		goto nomem;
 
 	sctp_add_cmd_sf(commands, SCTP_CMD_REPLY, SCTP_CHUNK(repl));
-	sctp_add_cmd_sf(commands, SCTP_CMD_TRANSMIT, SCTP_NULL());
 
 	/* RFC 2960 5.1 Normal Establishment of an Association
 	 *
@@ -1898,12 +1894,13 @@ static sctp_disposition_t sctp_sf_do_dupcook_d(const struct sctp_endpoint *ep,
 
 		}
 	}
-	sctp_add_cmd_sf(commands, SCTP_CMD_TRANSMIT, SCTP_NULL());
 
 	repl = sctp_make_cookie_ack(new_asoc, chunk);
 	if (!repl)
 		goto nomem;
 
+	sctp_add_cmd_sf(commands, SCTP_CMD_REPLY, SCTP_CHUNK(repl));
+
 	if (ev)
 		sctp_add_cmd_sf(commands, SCTP_CMD_EVENT_ULP,
 				SCTP_ULPEVENT(ev));
@@ -1911,9 +1908,6 @@ static sctp_disposition_t sctp_sf_do_dupcook_d(const struct sctp_endpoint *ep,
 		sctp_add_cmd_sf(commands, SCTP_CMD_EVENT_ULP,
 					SCTP_ULPEVENT(ai_ev));
 
-	sctp_add_cmd_sf(commands, SCTP_CMD_REPLY, SCTP_CHUNK(repl));
-	sctp_add_cmd_sf(commands, SCTP_CMD_TRANSMIT, SCTP_NULL());
-
 	return SCTP_DISPOSITION_CONSUME;
 
 nomem:
@@ -3970,9 +3964,6 @@ sctp_disposition_t sctp_sf_unk_chunk(const struct sctp_endpoint *ep,
 		return sctp_sf_pdiscard(ep, asoc, type, arg, commands);
 		break;
 	case SCTP_CID_ACTION_DISCARD_ERR:
-		/* Discard the packet.  */
-		sctp_sf_pdiscard(ep, asoc, type, arg, commands);
-
 		/* Generate an ERROR chunk as response. */
 		hdr = unk_chunk->chunk_hdr;
 		err_chunk = sctp_make_op_error(asoc, unk_chunk,
@@ -3982,6 +3973,9 @@ sctp_disposition_t sctp_sf_unk_chunk(const struct sctp_endpoint *ep,
 			sctp_add_cmd_sf(commands, SCTP_CMD_REPLY,
 					SCTP_CHUNK(err_chunk));
 		}
+
+		/* Discard the packet.  */
+		sctp_sf_pdiscard(ep, asoc, type, arg, commands);
 		return SCTP_DISPOSITION_CONSUME;
 		break;
 	case SCTP_CID_ACTION_SKIP:
-- 
cgit v1.2.3


From 0187bdfb05674147774ca79a79942537f3ad54bd Mon Sep 17 00:00:00 2001
From: Ben Hutchings <bhutchings@solarflare.com>
Date: Thu, 19 Jun 2008 16:15:47 -0700
Subject: net: Disable LRO on devices that are forwarding

Large Receive Offload (LRO) is only appropriate for packets that are
destined for the host, and should be disabled if received packets may be
forwarded.  It can also confuse the GSO on output.

Add dev_disable_lro() function which uses the appropriate ethtool ops to
disable LRO if enabled.

Add calls to dev_disable_lro() in br_add_if() and functions that enable
IPv4 and IPv6 forwarding.

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h |  1 +
 net/bridge/br_if.c        |  1 +
 net/core/dev.c            | 24 ++++++++++++++++++++++++
 net/ipv4/devinet.c        | 21 ++++++++++++++++-----
 net/ipv6/addrconf.c       |  6 ++++++
 5 files changed, 48 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 45dce2b58d4c..1304ad2d7105 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -886,6 +886,7 @@ extern struct net_device	*__dev_get_by_name(struct net *net, const char *name);
 extern int		dev_alloc_name(struct net_device *dev, const char *name);
 extern int		dev_open(struct net_device *dev);
 extern int		dev_close(struct net_device *dev);
+extern void		dev_disable_lro(struct net_device *dev);
 extern int		dev_queue_xmit(struct sk_buff *skb);
 extern int		register_netdevice(struct net_device *dev);
 extern void		unregister_netdevice(struct net_device *dev);
diff --git a/net/bridge/br_if.c b/net/bridge/br_if.c
index 143c954681b8..832a561500d9 100644
--- a/net/bridge/br_if.c
+++ b/net/bridge/br_if.c
@@ -387,6 +387,7 @@ int br_add_if(struct net_bridge *br, struct net_device *dev)
 		goto err2;
 
 	rcu_assign_pointer(dev->br_port, p);
+	dev_disable_lro(dev);
 	dev_set_promiscuity(dev, 1);
 
 	list_add_rcu(&p->list, &br->port_list);
diff --git a/net/core/dev.c b/net/core/dev.c
index a495f712d38c..f6944ecd5b2e 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -90,6 +90,7 @@
 #include <linux/if_ether.h>
 #include <linux/netdevice.h>
 #include <linux/etherdevice.h>
+#include <linux/ethtool.h>
 #include <linux/notifier.h>
 #include <linux/skbuff.h>
 #include <net/net_namespace.h>
@@ -1123,6 +1124,29 @@ int dev_close(struct net_device *dev)
 }
 
 
+/**
+ *	dev_disable_lro - disable Large Receive Offload on a device
+ *	@dev: device
+ *
+ *	Disable Large Receive Offload (LRO) on a net device.  Must be
+ *	called under RTNL.  This is needed if received packets may be
+ *	forwarded to another interface.
+ */
+void dev_disable_lro(struct net_device *dev)
+{
+	if (dev->ethtool_ops && dev->ethtool_ops->get_flags &&
+	    dev->ethtool_ops->set_flags) {
+		u32 flags = dev->ethtool_ops->get_flags(dev);
+		if (flags & ETH_FLAG_LRO) {
+			flags &= ~ETH_FLAG_LRO;
+			dev->ethtool_ops->set_flags(dev, flags);
+		}
+	}
+	WARN_ON(dev->features & NETIF_F_LRO);
+}
+EXPORT_SYMBOL(dev_disable_lro);
+
+
 static int dev_boot_phase = 1;
 
 /*
diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c
index f8c0b0aea93a..9de2514946ca 100644
--- a/net/ipv4/devinet.c
+++ b/net/ipv4/devinet.c
@@ -168,6 +168,8 @@ static struct in_device *inetdev_init(struct net_device *dev)
 	in_dev->dev = dev;
 	if ((in_dev->arp_parms = neigh_parms_alloc(dev, &arp_tbl)) == NULL)
 		goto out_kfree;
+	if (IPV4_DEVCONF(in_dev->cnf, FORWARDING))
+		dev_disable_lro(dev);
 	/* Reference in_dev->dev */
 	dev_hold(dev);
 	/* Account for reference dev->ip_ptr (below) */
@@ -1241,6 +1243,8 @@ static void inet_forward_change(struct net *net)
 	read_lock(&dev_base_lock);
 	for_each_netdev(net, dev) {
 		struct in_device *in_dev;
+		if (on)
+			dev_disable_lro(dev);
 		rcu_read_lock();
 		in_dev = __in_dev_get_rcu(dev);
 		if (in_dev)
@@ -1248,8 +1252,6 @@ static void inet_forward_change(struct net *net)
 		rcu_read_unlock();
 	}
 	read_unlock(&dev_base_lock);
-
-	rt_cache_flush(0);
 }
 
 static int devinet_conf_proc(ctl_table *ctl, int write,
@@ -1335,10 +1337,19 @@ static int devinet_sysctl_forward(ctl_table *ctl, int write,
 	if (write && *valp != val) {
 		struct net *net = ctl->extra2;
 
-		if (valp == &IPV4_DEVCONF_ALL(net, FORWARDING))
-			inet_forward_change(net);
-		else if (valp != &IPV4_DEVCONF_DFLT(net, FORWARDING))
+		if (valp != &IPV4_DEVCONF_DFLT(net, FORWARDING)) {
+			rtnl_lock();
+			if (valp == &IPV4_DEVCONF_ALL(net, FORWARDING)) {
+				inet_forward_change(net);
+			} else if (*valp) {
+				struct ipv4_devconf *cnf = ctl->extra1;
+				struct in_device *idev =
+					container_of(cnf, struct in_device, cnf);
+				dev_disable_lro(idev->dev);
+			}
+			rtnl_unlock();
 			rt_cache_flush(0);
+		}
 	}
 
 	return ret;
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index 9be6be3a7ff3..84127d854cfc 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -348,6 +348,8 @@ static struct inet6_dev * ipv6_add_dev(struct net_device *dev)
 		kfree(ndev);
 		return NULL;
 	}
+	if (ndev->cnf.forwarding)
+		dev_disable_lro(dev);
 	/* We refer to the device */
 	dev_hold(dev);
 
@@ -442,6 +444,8 @@ static void dev_forward_change(struct inet6_dev *idev)
 	if (!idev)
 		return;
 	dev = idev->dev;
+	if (idev->cnf.forwarding)
+		dev_disable_lro(dev);
 	if (dev && (dev->flags & IFF_MULTICAST)) {
 		if (idev->cnf.forwarding)
 			ipv6_dev_mc_inc(dev, &in6addr_linklocal_allrouters);
@@ -487,12 +491,14 @@ static void addrconf_fixup_forwarding(struct ctl_table *table, int *p, int old)
 	if (p == &net->ipv6.devconf_dflt->forwarding)
 		return;
 
+	rtnl_lock();
 	if (p == &net->ipv6.devconf_all->forwarding) {
 		__s32 newf = net->ipv6.devconf_all->forwarding;
 		net->ipv6.devconf_dflt->forwarding = newf;
 		addrconf_forward_change(net, newf);
 	} else if ((!*p) ^ (!old))
 		dev_forward_change((struct inet6_dev *)table->extra1);
+	rtnl_unlock();
 
 	if (*p)
 		rt6_purge_dflt_routers(net);
-- 
cgit v1.2.3


From 4497b0763cb1afae463f5e144c28b5d806e28b60 Mon Sep 17 00:00:00 2001
From: Ben Hutchings <bhutchings@solarflare.com>
Date: Thu, 19 Jun 2008 16:22:28 -0700
Subject: net: Discard and warn about LRO'd skbs received for forwarding

Add skb_warn_if_lro() to test whether an skb was received with LRO and
warn if so.

Change br_forward(), ip_forward() and ip6_forward() to call it) and
discard the skb if it returns true.

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/skbuff.h  | 14 ++++++++++++++
 net/bridge/br_forward.c |  2 +-
 net/core/skbuff.c       |  8 ++++++++
 net/ipv4/ip_forward.c   |  3 +++
 net/ipv6/ip6_output.c   |  3 +++
 5 files changed, 29 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index 299ec4b31412..2220b9e2dab0 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -1702,6 +1702,20 @@ static inline int skb_is_gso_v6(const struct sk_buff *skb)
 	return skb_shinfo(skb)->gso_type & SKB_GSO_TCPV6;
 }
 
+extern void __skb_warn_lro_forwarding(const struct sk_buff *skb);
+
+static inline bool skb_warn_if_lro(const struct sk_buff *skb)
+{
+	/* LRO sets gso_size but not gso_type, whereas if GSO is really
+	 * wanted then gso_type will be set. */
+	struct skb_shared_info *shinfo = skb_shinfo(skb);
+	if (shinfo->gso_size != 0 && unlikely(shinfo->gso_type == 0)) {
+		__skb_warn_lro_forwarding(skb);
+		return true;
+	}
+	return false;
+}
+
 static inline void skb_forward_csum(struct sk_buff *skb)
 {
 	/* Unfortunately we don't support this one.  Any brave souls? */
diff --git a/net/bridge/br_forward.c b/net/bridge/br_forward.c
index 512645727f51..bdd9ccea17ce 100644
--- a/net/bridge/br_forward.c
+++ b/net/bridge/br_forward.c
@@ -89,7 +89,7 @@ void br_deliver(const struct net_bridge_port *to, struct sk_buff *skb)
 /* called with rcu_read_lock */
 void br_forward(const struct net_bridge_port *to, struct sk_buff *skb)
 {
-	if (should_deliver(to, skb)) {
+	if (!skb_warn_if_lro(skb) && should_deliver(to, skb)) {
 		__br_forward(to, skb);
 		return;
 	}
diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index 3e18f8525e82..2df012be973d 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -2583,6 +2583,13 @@ bool skb_partial_csum_set(struct sk_buff *skb, u16 start, u16 off)
 	return true;
 }
 
+void __skb_warn_lro_forwarding(const struct sk_buff *skb)
+{
+	if (net_ratelimit())
+		pr_warning("%s: received packets cannot be forwarded"
+			   " while LRO is enabled\n", skb->dev->name);
+}
+
 EXPORT_SYMBOL(___pskb_trim);
 EXPORT_SYMBOL(__kfree_skb);
 EXPORT_SYMBOL(kfree_skb);
@@ -2616,6 +2623,7 @@ EXPORT_SYMBOL(skb_seq_read);
 EXPORT_SYMBOL(skb_abort_seq_read);
 EXPORT_SYMBOL(skb_find_text);
 EXPORT_SYMBOL(skb_append_datato_frags);
+EXPORT_SYMBOL(__skb_warn_lro_forwarding);
 
 EXPORT_SYMBOL_GPL(skb_to_sgvec);
 EXPORT_SYMBOL_GPL(skb_cow_data);
diff --git a/net/ipv4/ip_forward.c b/net/ipv4/ip_forward.c
index 37d36a3f33cd..da14725916d3 100644
--- a/net/ipv4/ip_forward.c
+++ b/net/ipv4/ip_forward.c
@@ -56,6 +56,9 @@ int ip_forward(struct sk_buff *skb)
 	struct rtable *rt;	/* Route we use */
 	struct ip_options * opt	= &(IPCB(skb)->opt);
 
+	if (skb_warn_if_lro(skb))
+		goto drop;
+
 	if (!xfrm4_policy_check(NULL, XFRM_POLICY_FWD, skb))
 		goto drop;
 
diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c
index 40a2813a63d1..fd7cd1bfe151 100644
--- a/net/ipv6/ip6_output.c
+++ b/net/ipv6/ip6_output.c
@@ -407,6 +407,9 @@ int ip6_forward(struct sk_buff *skb)
 	if (ipv6_devconf.forwarding == 0)
 		goto error;
 
+	if (skb_warn_if_lro(skb))
+		goto drop;
+
 	if (!xfrm6_policy_check(NULL, XFRM_POLICY_FWD, skb)) {
 		IP6_INC_STATS(ip6_dst_idev(dst), IPSTATS_MIB_INDISCARDS);
 		goto drop;
-- 
cgit v1.2.3


From 34f80b04f325078ff21123579343d99756ad8d0e Mon Sep 17 00:00:00 2001
From: Eilon Greenstein <eilong@broadcom.com>
Date: Mon, 23 Jun 2008 20:33:01 -0700
Subject: bnx2x: Add support for BCM57711 HW

Supporting the 57711 and 57711E - refers to in the code as E1H. The
57710 is referred to as E1.

To support the new members in the family, the bnx2x structure was
divided to 3 parts: common, port and function. These changes caused some
rearrangement in the bnx2x.h file.

A set of accessories macros were added to make access to the bnx2x
structure more readable

Signed-off-by: Eilon Greenstein <eilong@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/Kconfig         |    1 +
 drivers/net/bnx2x.h         |  739 ++++----
 drivers/net/bnx2x_fw_defs.h |  483 ++++--
 drivers/net/bnx2x_hsi.h     |  708 +++++---
 drivers/net/bnx2x_init.h    |   20 +-
 drivers/net/bnx2x_link.c    |   11 +-
 drivers/net/bnx2x_main.c    | 3919 +++++++++++++++++++++++++++----------------
 drivers/net/bnx2x_reg.h     |   66 +-
 include/linux/pci_ids.h     |    2 +
 9 files changed, 3774 insertions(+), 2175 deletions(-)

(limited to 'include')

diff --git a/drivers/net/Kconfig b/drivers/net/Kconfig
index 0e0f6696ccac..287d0873c60d 100644
--- a/drivers/net/Kconfig
+++ b/drivers/net/Kconfig
@@ -2600,6 +2600,7 @@ config BNX2X
 	tristate "Broadcom NetXtremeII 10Gb support"
 	depends on PCI
 	select ZLIB_INFLATE
+	select LIBCRC32C
 	help
 	  This driver supports Broadcom NetXtremeII 10 gigabit Ethernet cards.
 	  To compile this driver as a module, choose M here: the module
diff --git a/drivers/net/bnx2x.h b/drivers/net/bnx2x.h
index 0979ca0ae408..e08b9439a933 100644
--- a/drivers/net/bnx2x.h
+++ b/drivers/net/bnx2x.h
@@ -14,39 +14,46 @@
 #ifndef BNX2X_H
 #define BNX2X_H
 
+/* compilation time flags */
+
+/* define this to make the driver freeze on error to allow getting debug info
+ * (you will need to reboot afterwards) */
+/* #define BNX2X_STOP_ON_ERROR */
+
 /* error/debug prints */
 
-#define DRV_MODULE_NAME 	"bnx2x"
-#define PFX DRV_MODULE_NAME     ": "
+#define DRV_MODULE_NAME		"bnx2x"
+#define PFX DRV_MODULE_NAME	": "
 
 /* for messages that are currently off */
-#define BNX2X_MSG_OFF   		0
-#define BNX2X_MSG_MCP   		0x10000 /* was: NETIF_MSG_HW */
-#define BNX2X_MSG_STATS 		0x20000 /* was: NETIF_MSG_TIMER */
-#define NETIF_MSG_NVM   		0x40000 /* was: NETIF_MSG_HW */
-#define NETIF_MSG_DMAE  		0x80000 /* was: NETIF_MSG_HW */
+#define BNX2X_MSG_OFF			0
+#define BNX2X_MSG_MCP			0x010000 /* was: NETIF_MSG_HW */
+#define BNX2X_MSG_STATS			0x020000 /* was: NETIF_MSG_TIMER */
+#define BNX2X_MSG_NVM			0x040000 /* was: NETIF_MSG_HW */
+#define BNX2X_MSG_DMAE			0x080000 /* was: NETIF_MSG_HW */
 #define BNX2X_MSG_SP			0x100000 /* was: NETIF_MSG_INTR */
 #define BNX2X_MSG_FP			0x200000 /* was: NETIF_MSG_INTR */
 
-#define DP_LEVEL			KERN_NOTICE     /* was: KERN_DEBUG */
+#define DP_LEVEL			KERN_NOTICE	/* was: KERN_DEBUG */
 
 /* regular debug print */
 #define DP(__mask, __fmt, __args...) do { \
 	if (bp->msglevel & (__mask)) \
-		printk(DP_LEVEL "[%s:%d(%s)]" __fmt, __FUNCTION__, \
-		__LINE__, bp->dev?(bp->dev->name):"?", ##__args); \
+		printk(DP_LEVEL "[%s:%d(%s)]" __fmt, __func__, __LINE__, \
+			bp->dev?(bp->dev->name):"?", ##__args); \
 	} while (0)
 
-/* for errors (never masked) */
-#define BNX2X_ERR(__fmt, __args...) do { \
-	printk(KERN_ERR "[%s:%d(%s)]" __fmt, __FUNCTION__, \
-		__LINE__, bp->dev?(bp->dev->name):"?", ##__args); \
+/* errors debug print */
+#define BNX2X_DBG_ERR(__fmt, __args...) do { \
+	if (bp->msglevel & NETIF_MSG_PROBE) \
+		printk(KERN_ERR "[%s:%d(%s)]" __fmt, __func__, __LINE__, \
+			bp->dev?(bp->dev->name):"?", ##__args); \
 	} while (0)
 
-/* for logging (never masked) */
-#define BNX2X_LOG(__fmt, __args...) do { \
-	printk(KERN_NOTICE "[%s:%d(%s)]" __fmt, __FUNCTION__, \
-		__LINE__, bp->dev?(bp->dev->name):"?", ##__args); \
+/* for errors (never masked) */
+#define BNX2X_ERR(__fmt, __args...) do { \
+	printk(KERN_ERR "[%s:%d(%s)]" __fmt, __func__, __LINE__, \
+		bp->dev?(bp->dev->name):"?", ##__args); \
 	} while (0)
 
 /* before we have a dev->name use dev_info() */
@@ -60,7 +67,7 @@
 #define bnx2x_panic() do { \
 		bp->panic = 1; \
 		BNX2X_ERR("driver assert\n"); \
-		bnx2x_disable_int(bp); \
+		bnx2x_int_disable(bp); \
 		bnx2x_panic_dump(bp); \
 	} while (0)
 #else
@@ -71,24 +78,29 @@
 #endif
 
 
-#define U64_LO(x)       		(((u64)x) & 0xffffffff)
-#define U64_HI(x)       		(((u64)x) >> 32)
-#define HILO_U64(hi, lo)		(((u64)hi << 32) + lo)
+#ifdef NETIF_F_HW_VLAN_TX
+#define BCM_VLAN			1
+#endif
+
 
+#define U64_LO(x)			(u32)(((u64)(x)) & 0xffffffff)
+#define U64_HI(x)			(u32)(((u64)(x)) >> 32)
+#define HILO_U64(hi, lo)		((((u64)(hi)) << 32) + (lo))
 
-#define REG_ADDR(bp, offset)    	(bp->regview + offset)
 
-#define REG_RD(bp, offset)      	readl(REG_ADDR(bp, offset))
-#define REG_RD8(bp, offset)     	readb(REG_ADDR(bp, offset))
-#define REG_RD64(bp, offset)    	readq(REG_ADDR(bp, offset))
+#define REG_ADDR(bp, offset)		(bp->regview + offset)
 
-#define REG_WR(bp, offset, val) 	writel((u32)val, REG_ADDR(bp, offset))
+#define REG_RD(bp, offset)		readl(REG_ADDR(bp, offset))
+#define REG_RD8(bp, offset)		readb(REG_ADDR(bp, offset))
+#define REG_RD64(bp, offset)		readq(REG_ADDR(bp, offset))
+
+#define REG_WR(bp, offset, val)		writel((u32)val, REG_ADDR(bp, offset))
 #define REG_WR8(bp, offset, val)	writeb((u8)val, REG_ADDR(bp, offset))
-#define REG_WR16(bp, offset, val)       writew((u16)val, REG_ADDR(bp, offset))
-#define REG_WR32(bp, offset, val)       REG_WR(bp, offset, val)
+#define REG_WR16(bp, offset, val)	writew((u16)val, REG_ADDR(bp, offset))
+#define REG_WR32(bp, offset, val)	REG_WR(bp, offset, val)
 
-#define REG_RD_IND(bp, offset)  	bnx2x_reg_rd_ind(bp, offset)
-#define REG_WR_IND(bp, offset, val)     bnx2x_reg_wr_ind(bp, offset, val)
+#define REG_RD_IND(bp, offset)		bnx2x_reg_rd_ind(bp, offset)
+#define REG_WR_IND(bp, offset, val)	bnx2x_reg_wr_ind(bp, offset, val)
 
 #define REG_RD_DMAE(bp, offset, valp, len32) \
 	do { \
@@ -96,28 +108,28 @@
 		memcpy(valp, bnx2x_sp(bp, wb_data[0]), len32 * 4); \
 	} while (0)
 
-#define REG_WR_DMAE(bp, offset, val, len32) \
+#define REG_WR_DMAE(bp, offset, valp, len32) \
 	do { \
-		memcpy(bnx2x_sp(bp, wb_data[0]), val, len32 * 4); \
+		memcpy(bnx2x_sp(bp, wb_data[0]), valp, len32 * 4); \
 		bnx2x_write_dmae(bp, bnx2x_sp_mapping(bp, wb_data), \
 				 offset, len32); \
 	} while (0)
 
-#define SHMEM_RD(bp, type) \
-	REG_RD(bp, bp->shmem_base + offsetof(struct shmem_region, type))
-#define SHMEM_WR(bp, type, val) \
-	REG_WR(bp, bp->shmem_base + offsetof(struct shmem_region, type), val)
+#define SHMEM_ADDR(bp, field)		(bp->common.shmem_base + \
+					 offsetof(struct shmem_region, field))
+#define SHMEM_RD(bp, field)		REG_RD(bp, SHMEM_ADDR(bp, field))
+#define SHMEM_WR(bp, field, val)	REG_WR(bp, SHMEM_ADDR(bp, field), val)
 
 #define NIG_WR(reg, val)	REG_WR(bp, reg, val)
-#define EMAC_WR(reg, val)       REG_WR(bp, emac_base + reg, val)
-#define BMAC_WR(reg, val)       REG_WR(bp, GRCBASE_NIG + bmac_addr + reg, val)
+#define EMAC_WR(reg, val)	REG_WR(bp, emac_base + reg, val)
+#define BMAC_WR(reg, val)	REG_WR(bp, GRCBASE_NIG + bmac_addr + reg, val)
 
 
-#define for_each_queue(bp, var) for (var = 0; var < bp->num_queues; var++)
+#define for_each_queue(bp, var)	for (var = 0; var < bp->num_queues; var++)
 
 #define for_each_nondefault_queue(bp, var) \
 				for (var = 1; var < bp->num_queues; var++)
-#define is_multi(bp)    	(bp->num_queues > 1)
+#define is_multi(bp)		(bp->num_queues > 1)
 
 
 struct regp {
@@ -358,210 +370,122 @@ struct bnx2x_eth_stats {
 	u32 number_of_bugs_found_in_stats_spec; /* just kidding */
 };
 
-#define MAC_STX_NA      		0xffffffff
-
-#ifdef BNX2X_MULTI
-#define MAX_CONTEXT     		16
-#else
-#define MAX_CONTEXT     		1
-#endif
-
-union cdu_context {
-	struct eth_context eth;
-	char pad[1024];
-};
-
-#define MAX_DMAE_C      		5
-
-/* DMA memory not used in fastpath */
-struct bnx2x_slowpath {
-	union cdu_context       	context[MAX_CONTEXT];
-	struct eth_stats_query  	fw_stats;
-	struct mac_configuration_cmd    mac_config;
-	struct mac_configuration_cmd    mcast_config;
-
-	/* used by dmae command executer */
-	struct dmae_command     	dmae[MAX_DMAE_C];
-
-	union mac_stats 		mac_stats;
-	struct nig_stats		nig;
-	struct bnx2x_eth_stats  	eth_stats;
-
-	u32     			wb_comp;
-#define BNX2X_WB_COMP_VAL       	0xe0d0d0ae
-	u32     			wb_data[4];
-};
-
-#define bnx2x_sp(bp, var)       	(&bp->slowpath->var)
 #define bnx2x_sp_check(bp, var) ((bp->slowpath) ? (&bp->slowpath->var) : NULL)
-#define bnx2x_sp_mapping(bp, var) \
-		(bp->slowpath_mapping + offsetof(struct bnx2x_slowpath, var))
-
-
 struct sw_rx_bd {
-	struct sk_buff  *skb;
+	struct sk_buff	*skb;
 	DECLARE_PCI_UNMAP_ADDR(mapping)
 };
 
 struct sw_tx_bd {
-	struct sk_buff  *skb;
-	u16     	first_bd;
+	struct sk_buff	*skb;
+	u16		first_bd;
 };
 
 struct bnx2x_fastpath {
 
-	struct napi_struct      napi;
+	struct napi_struct	napi;
 
 	struct host_status_block *status_blk;
-	dma_addr_t      	status_blk_mapping;
+	dma_addr_t		status_blk_mapping;
 
-	struct eth_tx_db_data   *hw_tx_prods;
-	dma_addr_t      	tx_prods_mapping;
+	struct eth_tx_db_data	*hw_tx_prods;
+	dma_addr_t		tx_prods_mapping;
 
-	struct sw_tx_bd 	*tx_buf_ring;
+	struct sw_tx_bd		*tx_buf_ring;
 
 	struct eth_tx_bd	*tx_desc_ring;
-	dma_addr_t      	tx_desc_mapping;
+	dma_addr_t		tx_desc_mapping;
 
 	struct sw_rx_bd 	*rx_buf_ring;
 
 	struct eth_rx_bd	*rx_desc_ring;
-	dma_addr_t      	rx_desc_mapping;
+	dma_addr_t		rx_desc_mapping;
 
 	union eth_rx_cqe	*rx_comp_ring;
-	dma_addr_t      	rx_comp_mapping;
-
-	int     		state;
-#define BNX2X_FP_STATE_CLOSED   	0
-#define BNX2X_FP_STATE_IRQ      	0x80000
-#define BNX2X_FP_STATE_OPENING  	0x90000
-#define BNX2X_FP_STATE_OPEN     	0xa0000
-#define BNX2X_FP_STATE_HALTING  	0xb0000
-#define BNX2X_FP_STATE_HALTED   	0xc0000
-
-	int     		index;
-
-	u16     		tx_pkt_prod;
-	u16     		tx_pkt_cons;
-	u16     		tx_bd_prod;
-	u16     		tx_bd_cons;
-	u16     		*tx_cons_sb;
-
-	u16     		fp_c_idx;
-	u16     		fp_u_idx;
-
-	u16     		rx_bd_prod;
-	u16     		rx_bd_cons;
-	u16     		rx_comp_prod;
-	u16     		rx_comp_cons;
-	u16     		*rx_cons_sb;
-
-	unsigned long   	tx_pkt,
+	dma_addr_t		rx_comp_mapping;
+
+	int			state;
+#define BNX2X_FP_STATE_CLOSED		0
+#define BNX2X_FP_STATE_IRQ		0x80000
+#define BNX2X_FP_STATE_OPENING		0x90000
+#define BNX2X_FP_STATE_OPEN		0xa0000
+#define BNX2X_FP_STATE_HALTING		0xb0000
+#define BNX2X_FP_STATE_HALTED		0xc0000
+
+	u8			index;	/* number in fp array */
+	u8			cl_id;	/* eth client id */
+	u8			sb_id;	/* status block number in HW */
+#define FP_IDX(fp)			(fp->index)
+#define FP_CL_ID(fp)			(fp->cl_id)
+#define BP_CL_ID(bp)			(bp->fp[0].cl_id)
+#define FP_SB_ID(fp)			(fp->sb_id)
+#define CNIC_SB_ID			0
+
+	u16			tx_pkt_prod;
+	u16			tx_pkt_cons;
+	u16			tx_bd_prod;
+	u16			tx_bd_cons;
+	u16			*tx_cons_sb;
+
+	u16			fp_c_idx;
+	u16			fp_u_idx;
+
+	u16			rx_bd_prod;
+	u16			rx_bd_cons;
+	u16			rx_comp_prod;
+	u16			rx_comp_cons;
+	u16			*rx_cons_sb;
+
+	unsigned long		tx_pkt,
 				rx_pkt,
 				rx_calls;
 
-	struct bnx2x    	*bp; /* parent */
-};
-
-#define bnx2x_fp(bp, nr, var)   	(bp->fp[nr].var)
-
-
-/* attn group wiring */
-#define MAX_DYNAMIC_ATTN_GRPS   	8
-
-struct attn_route {
-	u32     sig[4];
+	struct bnx2x		*bp; /* parent */
 };
 
-struct bnx2x {
-	/* Fields used in the tx and intr/napi performance paths
-	 * are grouped together in the beginning of the structure
-	 */
-	struct bnx2x_fastpath   *fp;
-	void __iomem    	*regview;
-	void __iomem    	*doorbells;
-
-	struct net_device       *dev;
-	struct pci_dev  	*pdev;
-
-	atomic_t		intr_sem;
-	struct msix_entry       msix_table[MAX_CONTEXT+1];
-
-	int     		tx_ring_size;
+#define bnx2x_fp(bp, nr, var)		(bp->fp[nr].var)
+/* This is needed for determening of last_max */
+#define SUB_S16(a, b)			(s16)((s16)(a) - (s16)(b))
 
-#ifdef BCM_VLAN
-	struct vlan_group       *vlgrp;
-#endif
-
-	u32     		rx_csum;
-	u32     		rx_offset;
-	u32     		rx_buf_use_size;	/* useable size */
-	u32     		rx_buf_size;    	/* with alignment */
-#define ETH_OVREHEAD    		(ETH_HLEN + 8)  /* 8 for CRC + VLAN */
-#define ETH_MIN_PACKET_SIZE     	60
-#define ETH_MAX_PACKET_SIZE     	1500
-#define ETH_MAX_JUMBO_PACKET_SIZE       9600
+/* stuff added to make the code fit 80Col */
 
-	struct host_def_status_block *def_status_blk;
-#define DEF_SB_ID       	16
-	u16     		def_c_idx;
-	u16     		def_u_idx;
-	u16     		def_t_idx;
-	u16     		def_x_idx;
-	u16     		def_att_idx;
-	u32     		attn_state;
-	struct attn_route       attn_group[MAX_DYNAMIC_ATTN_GRPS];
-	u32     		aeu_mask;
-	u32     		nig_mask;
+#define CQE_TYPE(cqe_fp_flags)	((cqe_fp_flags) & ETH_FAST_PATH_RX_CQE_TYPE)
 
-	/* slow path ring */
-	struct eth_spe  	*spq;
-	dma_addr_t      	spq_mapping;
-	u16     		spq_prod_idx;
-	struct eth_spe  	*spq_prod_bd;
-	struct eth_spe  	*spq_last_bd;
-	u16     		*dsb_sp_prod;
-	u16     		spq_left; /* serialize spq */
-	spinlock_t      	spq_lock;
-
-	/* Flag for marking that there is either
-	 * STAT_QUERY or CFC DELETE ramrod pending
-	 */
-	u8      		stat_pending;
+#define ETH_RX_ERROR_FALGS	(ETH_FAST_PATH_RX_CQE_PHY_DECODE_ERR_FLG | \
+				 ETH_FAST_PATH_RX_CQE_IP_BAD_XSUM_FLG | \
+				 ETH_FAST_PATH_RX_CQE_L4_BAD_XSUM_FLG)
 
-	/* End of fields used in the performance code paths */
 
-	int     		panic;
-	int     		msglevel;
+#define U_SB_ETH_RX_CQ_INDEX		HC_INDEX_U_ETH_RX_CQ_CONS
+#define U_SB_ETH_RX_BD_INDEX		HC_INDEX_U_ETH_RX_BD_CONS
+#define C_SB_ETH_TX_CQ_INDEX		HC_INDEX_C_ETH_TX_CQ_CONS
 
-	u32     		flags;
-#define PCIX_FLAG       		1
-#define PCI_32BIT_FLAG  		2
-#define ONE_TDMA_FLAG   		4       /* no longer used */
-#define NO_WOL_FLAG     		8
-#define USING_DAC_FLAG  		0x10
-#define USING_MSIX_FLAG 		0x20
-#define ASF_ENABLE_FLAG 		0x40
+#define BNX2X_RX_SB_INDEX \
+	(&fp->status_blk->u_status_block.index_values[U_SB_ETH_RX_CQ_INDEX])
 
-	int     		port;
+#define BNX2X_RX_SB_BD_INDEX \
+	(&fp->status_blk->u_status_block.index_values[U_SB_ETH_RX_BD_INDEX])
 
-	int     		pm_cap;
-	int     		pcie_cap;
+#define BNX2X_RX_SB_INDEX_NUM \
+		(((U_SB_ETH_RX_CQ_INDEX << \
+		   USTORM_ETH_ST_CONTEXT_CONFIG_CQE_SB_INDEX_NUMBER_SHIFT) & \
+		  USTORM_ETH_ST_CONTEXT_CONFIG_CQE_SB_INDEX_NUMBER) | \
+		 ((U_SB_ETH_RX_BD_INDEX << \
+		   USTORM_ETH_ST_CONTEXT_CONFIG_BD_SB_INDEX_NUMBER_SHIFT) & \
+		  USTORM_ETH_ST_CONTEXT_CONFIG_BD_SB_INDEX_NUMBER))
 
-	struct work_struct	sp_task;
-	struct work_struct	reset_task;
+#define BNX2X_TX_SB_INDEX \
+	(&fp->status_blk->c_status_block.index_values[C_SB_ETH_TX_CQ_INDEX])
 
-	struct timer_list       timer;
-	int     		timer_interval;
-	int     		current_interval;
+/* common */
 
-	u32     		shmem_base;
+struct bnx2x_common {
 
 	u32			chip_id;
 /* chip num:16-31, rev:12-15, metal:4-11, bond_id:0-3 */
-#define CHIP_ID(bp)			(bp->chip_id & 0xfffffff0)
+#define CHIP_ID(bp)			(bp->common.chip_id & 0xfffffff0)
 
-#define CHIP_NUM(bp)			(bp->chip_id >> 16)
+#define CHIP_NUM(bp)			(bp->common.chip_id >> 16)
 #define CHIP_NUM_57710			0x164e
 #define CHIP_NUM_57711			0x164f
 #define CHIP_NUM_57711E			0x1650
@@ -572,7 +496,7 @@ struct bnx2x {
 					 CHIP_IS_57711E(bp))
 #define IS_E1H_OFFSET			CHIP_IS_E1H(bp)
 
-#define CHIP_REV(bp)			(bp->chip_id & 0x0000f000)
+#define CHIP_REV(bp)			(bp->common.chip_id & 0x0000f000)
 #define CHIP_REV_Ax			0x00000000
 /* assume maximum 5 revisions */
 #define CHIP_REV_IS_SLOW(bp)		(CHIP_REV(bp) > 0x00005000)
@@ -586,86 +510,250 @@ struct bnx2x {
 #define CHIP_TIME(bp)			((CHIP_REV_IS_EMUL(bp)) ? 2000 : \
 					((CHIP_REV_IS_FPGA(bp)) ? 200 : 1))
 
-#define CHIP_METAL(bp)			(bp->chip_id & 0x00000ff0)
-#define CHIP_BOND_ID(bp)		(bp->chip_id & 0x0000000f)
+#define CHIP_METAL(bp)			(bp->common.chip_id & 0x00000ff0)
+#define CHIP_BOND_ID(bp)		(bp->common.chip_id & 0x0000000f)
 
-	u16     		fw_seq;
-	u16     		fw_drv_pulse_wr_seq;
-	u32     		fw_mb;
+	int			flash_size;
+#define NVRAM_1MB_SIZE			0x20000	/* 1M bit in bytes */
+#define NVRAM_TIMEOUT_COUNT		30000
+#define NVRAM_PAGE_SIZE			256
 
-	u32     		hw_config;
+	u32			shmem_base;
+
+	u32			hw_config;
 	u32			board;
 
-	struct link_params	link_params;
+	u32			bc_ver;
+
+	char			*name;
+};
 
-	struct link_vars	link_vars;
+
+/* end of common */
+
+/* port */
+
+struct bnx2x_port {
+	u32			pmf;
 
 	u32			link_config;
 
-	u32     		supported;
+	u32			supported;
+/* link settings - missing defines */
+#define SUPPORTED_2500baseX_Full	(1 << 15)
+
+	u32			advertising;
 /* link settings - missing defines */
-#define SUPPORTED_2500baseT_Full	(1 << 15)
+#define ADVERTISED_2500baseX_Full	(1 << 15)
 
-	u32     		phy_addr;
+	u32			phy_addr;
 
 	/* used to synchronize phy accesses */
 	struct mutex		phy_mutex;
 
-	u32     		phy_id;
+	u32			port_stx;
 
+	struct nig_stats	old_nig_stats;
+};
 
-	u32     		advertising;
-/* link settings - missing defines */
-#define ADVERTISED_2500baseT_Full       (1 << 15)
+/* end of port */
+
+#define MAC_STX_NA      		0xffffffff
+
+#ifdef BNX2X_MULTI
+#define MAX_CONTEXT			16
+#else
+#define MAX_CONTEXT			1
+#endif
+
+union cdu_context {
+	struct eth_context eth;
+	char pad[1024];
+};
+
+#define MAX_DMAE_C			6
+
+/* DMA memory not used in fastpath */
+struct bnx2x_slowpath {
+	union cdu_context		context[MAX_CONTEXT];
+	struct eth_stats_query		fw_stats;
+	struct mac_configuration_cmd	mac_config;
+	struct mac_configuration_cmd	mcast_config;
+
+	/* used by dmae command executer */
+	struct dmae_command		dmae[MAX_DMAE_C];
+
+	union mac_stats 		mac_stats;
+	struct nig_stats		nig;
+	struct bnx2x_eth_stats  	eth_stats;
+
+	u32				wb_comp;
+#define BNX2X_WB_COMP_VAL       	0xe0d0d0ae
+	u32				wb_data[4];
+};
+
+#define bnx2x_sp(bp, var)		(&bp->slowpath->var)
+#define bnx2x_sp_mapping(bp, var) \
+		(bp->slowpath_mapping + offsetof(struct bnx2x_slowpath, var))
+
+
+/* attn group wiring */
+#define MAX_DYNAMIC_ATTN_GRPS		8
+
+struct attn_route {
+	u32	sig[4];
+};
+
+struct bnx2x {
+	/* Fields used in the tx and intr/napi performance paths
+	 * are grouped together in the beginning of the structure
+	 */
+	struct bnx2x_fastpath	fp[MAX_CONTEXT];
+	void __iomem		*regview;
+	void __iomem		*doorbells;
+#define BNX2X_DB_SIZE		(16*2048)
+
+	struct net_device	*dev;
+	struct pci_dev		*pdev;
+
+	atomic_t		intr_sem;
+	struct msix_entry       msix_table[MAX_CONTEXT+1];
+
+	int			tx_ring_size;
+
+#ifdef BCM_VLAN
+	struct vlan_group	*vlgrp;
+#endif
 
+	u32			rx_csum;
+	u32			rx_offset;
+	u32			rx_buf_use_size;	/* useable size */
+	u32			rx_buf_size;		/* with alignment */
+#define ETH_OVREHEAD			(ETH_HLEN + 8)	/* 8 for CRC + VLAN */
+#define ETH_MIN_PACKET_SIZE		60
+#define ETH_MAX_PACKET_SIZE		1500
+#define ETH_MAX_JUMBO_PACKET_SIZE	9600
 
-	u32     		bc_ver;
+	struct host_def_status_block *def_status_blk;
+#define DEF_SB_ID			16
+	u16			def_c_idx;
+	u16			def_u_idx;
+	u16			def_x_idx;
+	u16			def_t_idx;
+	u16			def_att_idx;
+	u32			attn_state;
+	struct attn_route	attn_group[MAX_DYNAMIC_ATTN_GRPS];
+	u32			aeu_mask;
+	u32			nig_mask;
+
+	/* slow path ring */
+	struct eth_spe		*spq;
+	dma_addr_t		spq_mapping;
+	u16			spq_prod_idx;
+	struct eth_spe		*spq_prod_bd;
+	struct eth_spe		*spq_last_bd;
+	u16			*dsb_sp_prod;
+	u16			spq_left; /* serialize spq */
+	/* used to synchronize spq accesses */
+	spinlock_t		spq_lock;
+
+	/* Flag for marking that there is either
+	 * STAT_QUERY or CFC DELETE ramrod pending
+	 */
+	u8      		stat_pending;
+
+	/* End of fileds used in the performance code paths */
+
+	int			panic;
+	int			msglevel;
+
+	u32			flags;
+#define PCIX_FLAG			1
+#define PCI_32BIT_FLAG			2
+#define ONE_TDMA_FLAG			4	/* no longer used */
+#define NO_WOL_FLAG			8
+#define USING_DAC_FLAG			0x10
+#define USING_MSIX_FLAG			0x20
+#define ASF_ENABLE_FLAG			0x40
+#define NO_MCP_FLAG			0x100
+#define BP_NOMCP(bp)			(bp->flags & NO_MCP_FLAG)
+
+	int			func;
+#define BP_PORT(bp)			(bp->func % PORT_MAX)
+#define BP_FUNC(bp)			(bp->func)
+#define BP_E1HVN(bp)			(bp->func >> 1)
+#define BP_L_ID(bp)			(BP_E1HVN(bp) << 2)
+/* assorted E1HVN */
+#define IS_E1HMF(bp)			(bp->e1hmf != 0)
+#define BP_MAX_QUEUES(bp)		(IS_E1HMF(bp) ? 4 : 16)
+
+	int			pm_cap;
+	int			pcie_cap;
+
+	struct work_struct	sp_task;
+	struct work_struct	reset_task;
+
+	struct timer_list	timer;
+	int			timer_interval;
+	int			current_interval;
+
+	u16			fw_seq;
+	u16			fw_drv_pulse_wr_seq;
+	u32			func_stx;
+
+	struct link_params	link_params;
+	struct link_vars	link_vars;
 
-	int     		flash_size;
-#define NVRAM_1MB_SIZE  		0x20000 /* 1M bit in bytes */
-#define NVRAM_TIMEOUT_COUNT     	30000
-#define NVRAM_PAGE_SIZE 		256
+	struct bnx2x_common	common;
+	struct bnx2x_port	port;
+
+	u32			mf_config;
+	u16			e1hov;
+	u8			e1hmf;
 
 	u8			wol;
 
-	int     		rx_ring_size;
+	int			rx_ring_size;
 
-	u16     		tx_quick_cons_trip_int;
-	u16     		tx_quick_cons_trip;
-	u16     		tx_ticks_int;
-	u16     		tx_ticks;
+	u16			tx_quick_cons_trip_int;
+	u16			tx_quick_cons_trip;
+	u16			tx_ticks_int;
+	u16			tx_ticks;
 
-	u16     		rx_quick_cons_trip_int;
-	u16     		rx_quick_cons_trip;
-	u16     		rx_ticks_int;
-	u16     		rx_ticks;
+	u16			rx_quick_cons_trip_int;
+	u16			rx_quick_cons_trip;
+	u16			rx_ticks_int;
+	u16			rx_ticks;
 
-	u32     		stats_ticks;
+	u32			stats_ticks;
+	u32			lin_cnt;
 
-	int     		state;
-#define BNX2X_STATE_CLOSED      	0x0
-#define BNX2X_STATE_OPENING_WAIT4_LOAD  0x1000
-#define BNX2X_STATE_OPENING_WAIT4_PORT  0x2000
+	int			state;
+#define BNX2X_STATE_CLOSED		0x0
+#define BNX2X_STATE_OPENING_WAIT4_LOAD	0x1000
+#define BNX2X_STATE_OPENING_WAIT4_PORT	0x2000
 #define BNX2X_STATE_OPEN		0x3000
-#define BNX2X_STATE_CLOSING_WAIT4_HALT  0x4000
+#define BNX2X_STATE_CLOSING_WAIT4_HALT	0x4000
 #define BNX2X_STATE_CLOSING_WAIT4_DELETE 0x5000
 #define BNX2X_STATE_CLOSING_WAIT4_UNLOAD 0x6000
-#define BNX2X_STATE_ERROR       	0xF000
+#define BNX2X_STATE_DISABLED		0xd000
+#define BNX2X_STATE_DIAG		0xe000
+#define BNX2X_STATE_ERROR		0xf000
 
-	int     		num_queues;
+	int			num_queues;
 
-	u32     		rx_mode;
-#define BNX2X_RX_MODE_NONE      	0
-#define BNX2X_RX_MODE_NORMAL    	1
-#define BNX2X_RX_MODE_ALLMULTI  	2
-#define BNX2X_RX_MODE_PROMISC   	3
-#define BNX2X_MAX_MULTICAST     	64
-#define BNX2X_MAX_EMUL_MULTI    	16
+	u32			rx_mode;
+#define BNX2X_RX_MODE_NONE		0
+#define BNX2X_RX_MODE_NORMAL		1
+#define BNX2X_RX_MODE_ALLMULTI		2
+#define BNX2X_RX_MODE_PROMISC		3
+#define BNX2X_MAX_MULTICAST		64
+#define BNX2X_MAX_EMUL_MULTI		16
 
-	dma_addr_t      	def_status_blk_mapping;
+	dma_addr_t		def_status_blk_mapping;
 
-	struct bnx2x_slowpath   *slowpath;
-	dma_addr_t      	slowpath_mapping;
+	struct bnx2x_slowpath	*slowpath;
+	dma_addr_t		slowpath_mapping;
 
 #ifdef BCM_ISCSI
 	void    		*t1;
@@ -742,8 +830,10 @@ int bnx2x_set_gpio(struct bnx2x *bp, int gpio_num, u32 mode);
 
 /* MC hsi */
 #define RX_COPY_THRESH  		92
-#define BCM_PAGE_BITS   		12
-#define BCM_PAGE_SIZE   		(1 << BCM_PAGE_BITS)
+#define BCM_PAGE_SHIFT			12
+#define BCM_PAGE_SIZE			(1 << BCM_PAGE_SHIFT)
+#define BCM_PAGE_MASK			(~(BCM_PAGE_SIZE - 1))
+#define BCM_PAGE_ALIGN(addr)	(((addr) + BCM_PAGE_SIZE - 1) & BCM_PAGE_MASK)
 
 #define NUM_TX_RINGS    		16
 #define TX_DESC_CNT     	(BCM_PAGE_SIZE / sizeof(struct eth_tx_bd))
@@ -795,26 +885,11 @@ int bnx2x_set_gpio(struct bnx2x *bp, int gpio_num, u32 mode);
 
 
 /* must be used on a CID before placing it on a HW ring */
-#define HW_CID(bp, x)   		(x | (bp->port << 23))
+#define HW_CID(bp, x)		((BP_PORT(bp) << 23) | (BP_E1HVN(bp) << 17) | x)
 
 #define SP_DESC_CNT     	(BCM_PAGE_SIZE / sizeof(struct eth_spe))
 #define MAX_SP_DESC_CNT 		(SP_DESC_CNT - 1)
 
-#define ATTN_NIG_FOR_FUNC       	(1L << 8)
-#define ATTN_SW_TIMER_4_FUNC    	(1L << 9)
-#define GPIO_2_FUNC     		(1L << 10)
-#define GPIO_3_FUNC     		(1L << 11)
-#define GPIO_4_FUNC     		(1L << 12)
-#define ATTN_GENERAL_ATTN_1     	(1L << 13)
-#define ATTN_GENERAL_ATTN_2     	(1L << 14)
-#define ATTN_GENERAL_ATTN_3     	(1L << 15)
-#define ATTN_GENERAL_ATTN_4     	(1L << 13)
-#define ATTN_GENERAL_ATTN_5     	(1L << 14)
-#define ATTN_GENERAL_ATTN_6     	(1L << 15)
-
-#define ATTN_HARD_WIRED_MASK    	0xff00
-#define ATTENTION_ID    		4
-
 
 #define BNX2X_BTR       		3
 #define MAX_SPQ_PENDING 		8
@@ -831,6 +906,31 @@ int bnx2x_set_gpio(struct bnx2x *bp, int gpio_num, u32 mode);
 		       DPM_TRIGER_TYPE); \
 	} while (0)
 
+static inline u32 reg_poll(struct bnx2x *bp, u32 reg, u32 expected, int ms,
+			   int wait)
+{
+	u32 val;
+
+	do {
+		val = REG_RD(bp, reg);
+		if (val == expected)
+			break;
+		ms -= wait;
+		msleep(wait);
+
+	} while (ms > 0);
+
+	return val;
+}
+
+
+/* load/unload mode */
+#define LOAD_NORMAL			0
+#define LOAD_OPEN			1
+#define LOAD_DIAG			2
+#define UNLOAD_NORMAL			0
+#define UNLOAD_CLOSE			1
+
 /* DMAE command defines */
 #define DMAE_CMD_SRC_PCI		0
 #define DMAE_CMD_SRC_GRC		DMAE_COMMAND_SRC
@@ -877,23 +977,48 @@ int bnx2x_set_gpio(struct bnx2x *bp, int gpio_num, u32 mode);
 
 #define pbd_tcp_flags(skb)  	(ntohl(tcp_flag_word(tcp_hdr(skb)))>>16 & 0xff)
 
-/* stuff added to make the code fit 80Col */
+/* must be used on a CID before placing it on a HW ring */
 
-#define TPA_TYPE_START  		ETH_FAST_PATH_RX_CQE_START_FLG
-#define TPA_TYPE_END    		ETH_FAST_PATH_RX_CQE_END_FLG
-#define TPA_TYPE(cqe)   	(cqe->fast_path_cqe.error_type_flags & \
-				 (TPA_TYPE_START | TPA_TYPE_END))
 #define BNX2X_RX_SUM_OK(cqe) \
 			(!(cqe->fast_path_cqe.status_flags & \
 			 (ETH_FAST_PATH_RX_CQE_IP_XSUM_NO_VALIDATION_FLG | \
 			  ETH_FAST_PATH_RX_CQE_L4_XSUM_NO_VALIDATION_FLG)))
 
-#define BNX2X_RX_SUM_FIX(cqe) \
-			((le16_to_cpu(cqe->fast_path_cqe.pars_flags.flags) & \
-			  PARSING_FLAGS_OVER_ETHERNET_PROTOCOL) == \
-			 (1 << PARSING_FLAGS_OVER_ETHERNET_PROTOCOL_SHIFT))
+/* CMNG constants
+   derived from lab experiments, and not from system spec calculations !!! */
+#define DEF_MIN_RATE			100
+/* resolution of the rate shaping timer - 100 usec */
+#define RS_PERIODIC_TIMEOUT_USEC	100
+/* resolution of fairness algorithm in usecs -
+   coefficient for clauclating the actuall t fair */
+#define T_FAIR_COEF			10000000
+/* number of bytes in single QM arbitration cycle -
+   coeffiecnt for calculating the fairness timer */
+#define QM_ARB_BYTES			40000
+#define FAIR_MEM			2
+
+
+#define ATTN_NIG_FOR_FUNC		(1L << 8)
+#define ATTN_SW_TIMER_4_FUNC		(1L << 9)
+#define GPIO_2_FUNC			(1L << 10)
+#define GPIO_3_FUNC			(1L << 11)
+#define GPIO_4_FUNC			(1L << 12)
+#define ATTN_GENERAL_ATTN_1		(1L << 13)
+#define ATTN_GENERAL_ATTN_2		(1L << 14)
+#define ATTN_GENERAL_ATTN_3		(1L << 15)
+#define ATTN_GENERAL_ATTN_4		(1L << 13)
+#define ATTN_GENERAL_ATTN_5		(1L << 14)
+#define ATTN_GENERAL_ATTN_6		(1L << 15)
+
+#define ATTN_HARD_WIRED_MASK		0xff00
+#define ATTENTION_ID			4
 
 
+/* stuff added to make the code fit 80Col */
+
+#define BNX2X_PMF_LINK_ASSERT \
+	GENERAL_ATTEN_OFFSET(LINK_SYNC_ATTENTION_BIT_FUNC_0 + BP_FUNC(bp))
+
 #define BNX2X_MC_ASSERT_BITS \
 	(GENERAL_ATTEN_OFFSET(TSTORM_FATAL_ASSERT_ATTENTION_BIT) | \
 	 GENERAL_ATTEN_OFFSET(USTORM_FATAL_ASSERT_ATTENTION_BIT) | \
@@ -906,12 +1031,20 @@ int bnx2x_set_gpio(struct bnx2x *bp, int gpio_num, u32 mode);
 #define BNX2X_DOORQ_ASSERT \
 	AEU_INPUTS_ATTN_BITS_DOORBELLQ_HW_INTERRUPT
 
+#define BNX2X_GRC_TIMEOUT	GENERAL_ATTEN_OFFSET(LATCHED_ATTN_TIMEOUT_GRC)
+#define BNX2X_GRC_RSV		(GENERAL_ATTEN_OFFSET(LATCHED_ATTN_RBCR) | \
+				 GENERAL_ATTEN_OFFSET(LATCHED_ATTN_RBCT) | \
+				 GENERAL_ATTEN_OFFSET(LATCHED_ATTN_RBCN) | \
+				 GENERAL_ATTEN_OFFSET(LATCHED_ATTN_RBCU) | \
+				 GENERAL_ATTEN_OFFSET(LATCHED_ATTN_RBCP) | \
+				 GENERAL_ATTEN_OFFSET(LATCHED_ATTN_RSVD_GRC))
+
 #define HW_INTERRUT_ASSERT_SET_0 \
 				(AEU_INPUTS_ATTN_BITS_TSDM_HW_INTERRUPT | \
 				 AEU_INPUTS_ATTN_BITS_TCM_HW_INTERRUPT | \
 				 AEU_INPUTS_ATTN_BITS_TSEMI_HW_INTERRUPT | \
 				 AEU_INPUTS_ATTN_BITS_PBF_HW_INTERRUPT)
-#define HW_PRTY_ASSERT_SET_0    (AEU_INPUTS_ATTN_BITS_BRB_PARITY_ERROR | \
+#define HW_PRTY_ASSERT_SET_0	(AEU_INPUTS_ATTN_BITS_BRB_PARITY_ERROR | \
 				 AEU_INPUTS_ATTN_BITS_PARSER_PARITY_ERROR | \
 				 AEU_INPUTS_ATTN_BITS_TSDM_PARITY_ERROR | \
 				 AEU_INPUTS_ATTN_BITS_SEARCHER_PARITY_ERROR |\
@@ -928,7 +1061,7 @@ int bnx2x_set_gpio(struct bnx2x *bp, int gpio_num, u32 mode);
 				 AEU_INPUTS_ATTN_BITS_UPB_HW_INTERRUPT | \
 				 AEU_INPUTS_ATTN_BITS_CSDM_HW_INTERRUPT | \
 				 AEU_INPUTS_ATTN_BITS_CCM_HW_INTERRUPT)
-#define HW_PRTY_ASSERT_SET_1    (AEU_INPUTS_ATTN_BITS_PBCLIENT_PARITY_ERROR |\
+#define HW_PRTY_ASSERT_SET_1	(AEU_INPUTS_ATTN_BITS_PBCLIENT_PARITY_ERROR |\
 				 AEU_INPUTS_ATTN_BITS_QM_PARITY_ERROR | \
 				 AEU_INPUTS_ATTN_BITS_XSDM_PARITY_ERROR | \
 				 AEU_INPUTS_ATTN_BITS_XSEMI_PARITY_ERROR | \
@@ -945,7 +1078,7 @@ int bnx2x_set_gpio(struct bnx2x *bp, int gpio_num, u32 mode);
 				 AEU_INPUTS_ATTN_BITS_DMAE_HW_INTERRUPT | \
 			AEU_INPUTS_ATTN_BITS_PXPPCICLOCKCLIENT_HW_INTERRUPT |\
 				 AEU_INPUTS_ATTN_BITS_MISC_HW_INTERRUPT)
-#define HW_PRTY_ASSERT_SET_2    (AEU_INPUTS_ATTN_BITS_CSEMI_PARITY_ERROR | \
+#define HW_PRTY_ASSERT_SET_2	(AEU_INPUTS_ATTN_BITS_CSEMI_PARITY_ERROR | \
 				 AEU_INPUTS_ATTN_BITS_PXP_PARITY_ERROR | \
 			AEU_INPUTS_ATTN_BITS_PXPPCICLOCKCLIENT_PARITY_ERROR |\
 				 AEU_INPUTS_ATTN_BITS_CFC_PARITY_ERROR | \
@@ -954,42 +1087,44 @@ int bnx2x_set_gpio(struct bnx2x *bp, int gpio_num, u32 mode);
 				 AEU_INPUTS_ATTN_BITS_MISC_PARITY_ERROR)
 
 
-#define ETH_RX_ERROR_FALGS      (ETH_FAST_PATH_RX_CQE_PHY_DECODE_ERR_FLG | \
-				 ETH_FAST_PATH_RX_CQE_IP_BAD_XSUM_FLG | \
-				 ETH_FAST_PATH_RX_CQE_L4_BAD_XSUM_FLG)
-
-
 #define MULTI_FLAGS \
-	(TSTORM_ETH_FUNCTION_COMMON_CONFIG_RSS_IPV4_CAPABILITY     | \
-	 TSTORM_ETH_FUNCTION_COMMON_CONFIG_RSS_IPV4_TCP_CAPABILITY | \
-	 TSTORM_ETH_FUNCTION_COMMON_CONFIG_RSS_IPV6_CAPABILITY     | \
-	 TSTORM_ETH_FUNCTION_COMMON_CONFIG_RSS_IPV6_TCP_CAPABILITY | \
-	 TSTORM_ETH_FUNCTION_COMMON_CONFIG_RSS_ENABLE)
+		(TSTORM_ETH_FUNCTION_COMMON_CONFIG_RSS_IPV4_CAPABILITY | \
+		 TSTORM_ETH_FUNCTION_COMMON_CONFIG_RSS_IPV4_TCP_CAPABILITY | \
+		 TSTORM_ETH_FUNCTION_COMMON_CONFIG_RSS_IPV6_CAPABILITY | \
+		 TSTORM_ETH_FUNCTION_COMMON_CONFIG_RSS_IPV6_TCP_CAPABILITY | \
+		 TSTORM_ETH_FUNCTION_COMMON_CONFIG_RSS_ENABLE)
 
-#define MULTI_MASK      0x7f
+#define MULTI_MASK			0x7f
 
 
-#define U_SB_ETH_RX_CQ_INDEX    	HC_INDEX_U_ETH_RX_CQ_CONS
-#define C_SB_ETH_TX_CQ_INDEX    	HC_INDEX_C_ETH_TX_CQ_CONS
-#define C_DEF_SB_SP_INDEX       	HC_INDEX_DEF_C_ETH_SLOW_PATH
-
-#define BNX2X_RX_SB_INDEX \
-	&fp->status_blk->u_status_block.index_values[U_SB_ETH_RX_CQ_INDEX]
+#define DEF_USB_FUNC_OFF		(2 + 2*HC_USTORM_DEF_SB_NUM_INDICES)
+#define DEF_CSB_FUNC_OFF		(2 + 2*HC_CSTORM_DEF_SB_NUM_INDICES)
+#define DEF_XSB_FUNC_OFF		(2 + 2*HC_XSTORM_DEF_SB_NUM_INDICES)
+#define DEF_TSB_FUNC_OFF		(2 + 2*HC_TSTORM_DEF_SB_NUM_INDICES)
 
-#define BNX2X_TX_SB_INDEX \
-	&fp->status_blk->c_status_block.index_values[C_SB_ETH_TX_CQ_INDEX]
+#define C_DEF_SB_SP_INDEX		HC_INDEX_DEF_C_ETH_SLOW_PATH
 
 #define BNX2X_SP_DSB_INDEX \
-&bp->def_status_blk->c_def_status_block.index_values[C_DEF_SB_SP_INDEX]
+(&bp->def_status_blk->c_def_status_block.index_values[C_DEF_SB_SP_INDEX])
 
 
 #define CAM_IS_INVALID(x) \
 (x.target_table_entry.flags == TSTORM_CAM_TARGET_TABLE_ENTRY_ACTION_TYPE)
 
 #define CAM_INVALIDATE(x) \
-x.target_table_entry.flags = TSTORM_CAM_TARGET_TABLE_ENTRY_ACTION_TYPE
+	(x.target_table_entry.flags = TSTORM_CAM_TARGET_TABLE_ENTRY_ACTION_TYPE)
+
+
+/* Number of u32 elements in MC hash array */
+#define MC_HASH_SIZE			8
+#define MC_HASH_OFFSET(bp, i)		(BAR_TSTRORM_INTMEM + \
+	TSTORM_APPROXIMATE_MATCH_MULTICAST_FILTERING_OFFSET(BP_FUNC(bp)) + i*4)
 
 
+#ifndef PXP2_REG_PXP2_INT_STS
+#define PXP2_REG_PXP2_INT_STS		PXP2_REG_PXP2_INT_STS_0
+#endif
+
 /* MISC_REG_RESET_REG - this is here for the hsi to work don't touch */
 
 #endif /* bnx2x.h */
diff --git a/drivers/net/bnx2x_fw_defs.h b/drivers/net/bnx2x_fw_defs.h
index 3b968904ca65..e3da7f69d27b 100644
--- a/drivers/net/bnx2x_fw_defs.h
+++ b/drivers/net/bnx2x_fw_defs.h
@@ -8,191 +8,390 @@
  */
 
 
-#define CSTORM_DEF_SB_HC_DISABLE_OFFSET(port, index)\
-	(0x1922 + (port * 0x40) + (index * 0x4))
-#define CSTORM_DEF_SB_HOST_SB_ADDR_OFFSET(port)\
-	(0x1900 + (port * 0x40))
-#define CSTORM_HC_BTR_OFFSET(port)\
-	(0x1984 + (port * 0xc0))
-#define CSTORM_SB_HC_DISABLE_OFFSET(port, cpu_id, index)\
-	(0x141a + (port * 0x280) + (cpu_id * 0x28) + (index * 0x4))
-#define CSTORM_SB_HC_TIMEOUT_OFFSET(port, cpu_id, index)\
-	(0x1418 + (port * 0x280) + (cpu_id * 0x28) + (index * 0x4))
-#define CSTORM_SB_HOST_SB_ADDR_OFFSET(port, cpu_id)\
-	(0x1400 + (port * 0x280) + (cpu_id * 0x28))
-#define CSTORM_STATS_FLAGS_OFFSET(port) 		(0x5108 + (port * 0x8))
-#define TSTORM_CLIENT_CONFIG_OFFSET(port, client_id)\
-	(0x1510 + (port * 0x240) + (client_id * 0x20))
-#define TSTORM_DEF_SB_HC_DISABLE_OFFSET(port, index)\
-	(0x138a + (port * 0x28) + (index * 0x4))
-#define TSTORM_DEF_SB_HOST_SB_ADDR_OFFSET(port)\
-	(0x1370 + (port * 0x28))
-#define TSTORM_ETH_STATS_QUERY_ADDR_OFFSET(port)\
-	(0x4b70 + (port * 0x8))
-#define TSTORM_FUNCTION_COMMON_CONFIG_OFFSET(function)\
-	(0x1418 + (function * 0x30))
-#define TSTORM_HC_BTR_OFFSET(port)\
-	(0x13c4 + (port * 0x18))
-#define TSTORM_INDIRECTION_TABLE_OFFSET(port)\
-	(0x22c8 + (port * 0x80))
-#define TSTORM_INDIRECTION_TABLE_SIZE			0x80
-#define TSTORM_MAC_FILTER_CONFIG_OFFSET(port)\
-	(0x1420 + (port * 0x30))
-#define TSTORM_RCQ_PROD_OFFSET(port, client_id)\
-	(0x1508 + (port * 0x240) + (client_id * 0x20))
-#define TSTORM_STATS_FLAGS_OFFSET(port) 		(0x4b90 + (port * 0x8))
-#define USTORM_DEF_SB_HC_DISABLE_OFFSET(port, index)\
-	(0x191a + (port * 0x28) + (index * 0x4))
-#define USTORM_DEF_SB_HOST_SB_ADDR_OFFSET(port)\
-	(0x1900 + (port * 0x28))
-#define USTORM_HC_BTR_OFFSET(port)\
-	(0x1954 + (port * 0xb8))
-#define USTORM_MEM_WORKAROUND_ADDRESS_OFFSET(port)\
-	(0x5408 + (port * 0x8))
-#define USTORM_SB_HC_DISABLE_OFFSET(port, cpu_id, index)\
-	(0x141a + (port * 0x280) + (cpu_id * 0x28) + (index * 0x4))
-#define USTORM_SB_HC_TIMEOUT_OFFSET(port, cpu_id, index)\
-	(0x1418 + (port * 0x280) + (cpu_id * 0x28) + (index * 0x4))
-#define USTORM_SB_HOST_SB_ADDR_OFFSET(port, cpu_id)\
-	(0x1400 + (port * 0x280) + (cpu_id * 0x28))
-#define XSTORM_ASSERT_LIST_INDEX_OFFSET 		0x1000
-#define XSTORM_ASSERT_LIST_OFFSET(idx)			(0x1020 + (idx * 0x10))
-#define XSTORM_DEF_SB_HC_DISABLE_OFFSET(port, index)\
-	(0x141a + (port * 0x28) + (index * 0x4))
-#define XSTORM_DEF_SB_HOST_SB_ADDR_OFFSET(port)\
-	(0x1400 + (port * 0x28))
-#define XSTORM_ETH_STATS_QUERY_ADDR_OFFSET(port)\
-	(0x5408 + (port * 0x8))
-#define XSTORM_HC_BTR_OFFSET(port)\
-	(0x1454 + (port * 0x18))
-#define XSTORM_SPQ_PAGE_BASE_OFFSET(port)\
-	(0x5328 + (port * 0x18))
-#define XSTORM_SPQ_PROD_OFFSET(port)\
-	(0x5330 + (port * 0x18))
-#define XSTORM_STATS_FLAGS_OFFSET(port) 		(0x53f8 + (port * 0x8))
+#define CSTORM_ASSERT_LIST_INDEX_OFFSET \
+	(IS_E1H_OFFSET? 0x7000 : 0x1000)
+#define CSTORM_ASSERT_LIST_OFFSET(idx) \
+	(IS_E1H_OFFSET? (0x7020 + (idx * 0x10)) : (0x1020 + (idx * 0x10)))
+#define CSTORM_DEF_SB_HC_DISABLE_OFFSET(function, index) \
+	(IS_E1H_OFFSET? (0x8522 + ((function>>1) * 0x40) + ((function&1) \
+	* 0x100) + (index * 0x4)) : (0x1922 + (function * 0x40) + (index \
+	* 0x4)))
+#define CSTORM_DEF_SB_HOST_SB_ADDR_OFFSET(function) \
+	(IS_E1H_OFFSET? (0x8500 + ((function>>1) * 0x40) + ((function&1) \
+	* 0x100)) : (0x1900 + (function * 0x40)))
+#define CSTORM_DEF_SB_HOST_STATUS_BLOCK_OFFSET(function) \
+	(IS_E1H_OFFSET? (0x8508 + ((function>>1) * 0x40) + ((function&1) \
+	* 0x100)) : (0x1908 + (function * 0x40)))
+#define CSTORM_FUNCTION_MODE_OFFSET \
+	(IS_E1H_OFFSET? 0x11e8 : 0xffffffff)
+#define CSTORM_HC_BTR_OFFSET(port) \
+	(IS_E1H_OFFSET? (0x8704 + (port * 0xf0)) : (0x1984 + (port * 0xc0)))
+#define CSTORM_SB_HC_DISABLE_OFFSET(port, cpu_id, index) \
+	(IS_E1H_OFFSET? (0x801a + (port * 0x280) + (cpu_id * 0x28) + \
+	(index * 0x4)) : (0x141a + (port * 0x280) + (cpu_id * 0x28) + \
+	(index * 0x4)))
+#define CSTORM_SB_HC_TIMEOUT_OFFSET(port, cpu_id, index) \
+	(IS_E1H_OFFSET? (0x8018 + (port * 0x280) + (cpu_id * 0x28) + \
+	(index * 0x4)) : (0x1418 + (port * 0x280) + (cpu_id * 0x28) + \
+	(index * 0x4)))
+#define CSTORM_SB_HOST_SB_ADDR_OFFSET(port, cpu_id) \
+	(IS_E1H_OFFSET? (0x8000 + (port * 0x280) + (cpu_id * 0x28)) : \
+	(0x1400 + (port * 0x280) + (cpu_id * 0x28)))
+#define CSTORM_SB_HOST_STATUS_BLOCK_OFFSET(port, cpu_id) \
+	(IS_E1H_OFFSET? (0x8008 + (port * 0x280) + (cpu_id * 0x28)) : \
+	(0x1408 + (port * 0x280) + (cpu_id * 0x28)))
+#define CSTORM_STATS_FLAGS_OFFSET(function) \
+	(IS_E1H_OFFSET? (0x1108 + (function * 0x8)) : (0x5108 + \
+	(function * 0x8)))
+#define TSTORM_APPROXIMATE_MATCH_MULTICAST_FILTERING_OFFSET(function) \
+	(IS_E1H_OFFSET? (0x31c0 + (function * 0x20)) : 0xffffffff)
+#define TSTORM_ASSERT_LIST_INDEX_OFFSET \
+	(IS_E1H_OFFSET? 0xa000 : 0x1000)
+#define TSTORM_ASSERT_LIST_OFFSET(idx) \
+	(IS_E1H_OFFSET? (0xa020 + (idx * 0x10)) : (0x1020 + (idx * 0x10)))
+#define TSTORM_CLIENT_CONFIG_OFFSET(port, client_id) \
+	(IS_E1H_OFFSET? (0x3358 + (port * 0x3e8) + (client_id * 0x28)) : \
+	(0x9c8 + (port * 0x2f8) + (client_id * 0x28)))
+#define TSTORM_DEF_SB_HC_DISABLE_OFFSET(function, index) \
+	(IS_E1H_OFFSET? (0xb01a + ((function>>1) * 0x28) + ((function&1) \
+	* 0xa0) + (index * 0x4)) : (0x141a + (function * 0x28) + (index * \
+	0x4)))
+#define TSTORM_DEF_SB_HOST_SB_ADDR_OFFSET(function) \
+	(IS_E1H_OFFSET? (0xb000 + ((function>>1) * 0x28) + ((function&1) \
+	* 0xa0)) : (0x1400 + (function * 0x28)))
+#define TSTORM_DEF_SB_HOST_STATUS_BLOCK_OFFSET(function) \
+	(IS_E1H_OFFSET? (0xb008 + ((function>>1) * 0x28) + ((function&1) \
+	* 0xa0)) : (0x1408 + (function * 0x28)))
+#define TSTORM_ETH_STATS_QUERY_ADDR_OFFSET(function) \
+	(IS_E1H_OFFSET? (0x2b80 + (function * 0x8)) : (0x4b68 + \
+	(function * 0x8)))
+#define TSTORM_FUNCTION_COMMON_CONFIG_OFFSET(function) \
+	(IS_E1H_OFFSET? (0x3000 + (function * 0x38)) : (0x1500 + \
+	(function * 0x38)))
+#define TSTORM_FUNCTION_MODE_OFFSET \
+	(IS_E1H_OFFSET? 0x1ad0 : 0xffffffff)
+#define TSTORM_HC_BTR_OFFSET(port) \
+	(IS_E1H_OFFSET? (0xb144 + (port * 0x30)) : (0x1454 + (port * 0x18)))
+#define TSTORM_INDIRECTION_TABLE_OFFSET(function) \
+	(IS_E1H_OFFSET? (0x12c8 + (function * 0x80)) : (0x22c8 + \
+	(function * 0x80)))
+#define TSTORM_INDIRECTION_TABLE_SIZE 0x80
+#define TSTORM_MAC_FILTER_CONFIG_OFFSET(function) \
+	(IS_E1H_OFFSET? (0x3008 + (function * 0x38)) : (0x1508 + \
+	(function * 0x38)))
+#define TSTORM_RX_PRODS_OFFSET(port, client_id) \
+	(IS_E1H_OFFSET? (0x3350 + (port * 0x3e8) + (client_id * 0x28)) : \
+	(0x9c0 + (port * 0x2f8) + (client_id * 0x28)))
+#define TSTORM_STATS_FLAGS_OFFSET(function) \
+	(IS_E1H_OFFSET? (0x2c00 + (function * 0x8)) : (0x4b88 + \
+	(function * 0x8)))
+#define TSTORM_TPA_EXIST_OFFSET (IS_E1H_OFFSET? 0x3b30 : 0x1c20)
+#define USTORM_AGG_DATA_OFFSET (IS_E1H_OFFSET? 0xa040 : 0x2c10)
+#define USTORM_AGG_DATA_SIZE (IS_E1H_OFFSET? 0x2440 : 0x1200)
+#define USTORM_ASSERT_LIST_INDEX_OFFSET \
+	(IS_E1H_OFFSET? 0x8000 : 0x1000)
+#define USTORM_ASSERT_LIST_OFFSET(idx) \
+	(IS_E1H_OFFSET? (0x8020 + (idx * 0x10)) : (0x1020 + (idx * 0x10)))
+#define USTORM_CQE_PAGE_BASE_OFFSET(port, clientId) \
+	(IS_E1H_OFFSET? (0x3298 + (port * 0x258) + (clientId * 0x18)) : \
+	(0x5450 + (port * 0x1c8) + (clientId * 0x18)))
+#define USTORM_DEF_SB_HC_DISABLE_OFFSET(function, index) \
+	(IS_E1H_OFFSET? (0x951a + ((function>>1) * 0x28) + ((function&1) \
+	* 0xa0) + (index * 0x4)) : (0x191a + (function * 0x28) + (index * \
+	0x4)))
+#define USTORM_DEF_SB_HOST_SB_ADDR_OFFSET(function) \
+	(IS_E1H_OFFSET? (0x9500 + ((function>>1) * 0x28) + ((function&1) \
+	* 0xa0)) : (0x1900 + (function * 0x28)))
+#define USTORM_DEF_SB_HOST_STATUS_BLOCK_OFFSET(function) \
+	(IS_E1H_OFFSET? (0x9508 + ((function>>1) * 0x28) + ((function&1) \
+	* 0xa0)) : (0x1908 + (function * 0x28)))
+#define USTORM_FUNCTION_MODE_OFFSET \
+	(IS_E1H_OFFSET? 0x2448 : 0xffffffff)
+#define USTORM_HC_BTR_OFFSET(port) \
+	(IS_E1H_OFFSET? (0x9644 + (port * 0xd0)) : (0x1954 + (port * 0xb8)))
+#define USTORM_MAX_AGG_SIZE_OFFSET(port, clientId) \
+	(IS_E1H_OFFSET? (0x3290 + (port * 0x258) + (clientId * 0x18)) : \
+	(0x5448 + (port * 0x1c8) + (clientId * 0x18)))
+#define USTORM_MEM_WORKAROUND_ADDRESS_OFFSET(function) \
+	(IS_E1H_OFFSET? (0x2408 + (function * 0x8)) : (0x5408 + \
+	(function * 0x8)))
+#define USTORM_SB_HC_DISABLE_OFFSET(port, cpu_id, index) \
+	(IS_E1H_OFFSET? (0x901a + (port * 0x280) + (cpu_id * 0x28) + \
+	(index * 0x4)) : (0x141a + (port * 0x280) + (cpu_id * 0x28) + \
+	(index * 0x4)))
+#define USTORM_SB_HC_TIMEOUT_OFFSET(port, cpu_id, index) \
+	(IS_E1H_OFFSET? (0x9018 + (port * 0x280) + (cpu_id * 0x28) + \
+	(index * 0x4)) : (0x1418 + (port * 0x280) + (cpu_id * 0x28) + \
+	(index * 0x4)))
+#define USTORM_SB_HOST_SB_ADDR_OFFSET(port, cpu_id) \
+	(IS_E1H_OFFSET? (0x9000 + (port * 0x280) + (cpu_id * 0x28)) : \
+	(0x1400 + (port * 0x280) + (cpu_id * 0x28)))
+#define USTORM_SB_HOST_STATUS_BLOCK_OFFSET(port, cpu_id) \
+	(IS_E1H_OFFSET? (0x9008 + (port * 0x280) + (cpu_id * 0x28)) : \
+	(0x1408 + (port * 0x280) + (cpu_id * 0x28)))
+#define XSTORM_ASSERT_LIST_INDEX_OFFSET \
+	(IS_E1H_OFFSET? 0x9000 : 0x1000)
+#define XSTORM_ASSERT_LIST_OFFSET(idx) \
+	(IS_E1H_OFFSET? (0x9020 + (idx * 0x10)) : (0x1020 + (idx * 0x10)))
+#define XSTORM_CMNG_PER_PORT_VARS_OFFSET(port) \
+	(IS_E1H_OFFSET? (0x24a8 + (port * 0x40)) : (0x3ba0 + (port * 0x40)))
+#define XSTORM_DEF_SB_HC_DISABLE_OFFSET(function, index) \
+	(IS_E1H_OFFSET? (0xa01a + ((function>>1) * 0x28) + ((function&1) \
+	* 0xa0) + (index * 0x4)) : (0x141a + (function * 0x28) + (index * \
+	0x4)))
+#define XSTORM_DEF_SB_HOST_SB_ADDR_OFFSET(function) \
+	(IS_E1H_OFFSET? (0xa000 + ((function>>1) * 0x28) + ((function&1) \
+	* 0xa0)) : (0x1400 + (function * 0x28)))
+#define XSTORM_DEF_SB_HOST_STATUS_BLOCK_OFFSET(function) \
+	(IS_E1H_OFFSET? (0xa008 + ((function>>1) * 0x28) + ((function&1) \
+	* 0xa0)) : (0x1408 + (function * 0x28)))
+#define XSTORM_E1HOV_OFFSET(function) \
+	(IS_E1H_OFFSET? (0x2ab8 + (function * 0x2)) : 0xffffffff)
+#define XSTORM_ETH_STATS_QUERY_ADDR_OFFSET(function) \
+	(IS_E1H_OFFSET? (0x2418 + (function * 0x8)) : (0x3b70 + \
+	(function * 0x8)))
+#define XSTORM_FAIRNESS_PER_VN_VARS_OFFSET(function) \
+	(IS_E1H_OFFSET? (0x2568 + (function * 0x70)) : (0x3c60 + \
+	(function * 0x70)))
+#define XSTORM_FUNCTION_MODE_OFFSET \
+	(IS_E1H_OFFSET? 0x2ac8 : 0xffffffff)
+#define XSTORM_HC_BTR_OFFSET(port) \
+	(IS_E1H_OFFSET? (0xa144 + (port * 0x30)) : (0x1454 + (port * 0x18)))
+#define XSTORM_RATE_SHAPING_PER_VN_VARS_OFFSET(function) \
+	(IS_E1H_OFFSET? (0x2528 + (function * 0x70)) : (0x3c20 + \
+	(function * 0x70)))
+#define XSTORM_SPQ_PAGE_BASE_OFFSET(function) \
+	(IS_E1H_OFFSET? (0x2000 + (function * 0x10)) : (0x3328 + \
+	(function * 0x10)))
+#define XSTORM_SPQ_PROD_OFFSET(function) \
+	(IS_E1H_OFFSET? (0x2008 + (function * 0x10)) : (0x3330 + \
+	(function * 0x10)))
+#define XSTORM_STATS_FLAGS_OFFSET(function) \
+	(IS_E1H_OFFSET? (0x23d8 + (function * 0x8)) : (0x3b60 + \
+	(function * 0x8)))
 #define COMMON_ASM_INVALID_ASSERT_OPCODE 0x0
 
 /**
 * This file defines HSI constatnts for the ETH flow
 */
-
-/* hash types */
-#define DEFAULT_HASH_TYPE			0
-#define IPV4_HASH_TYPE				1
-#define TCP_IPV4_HASH_TYPE			2
-#define IPV6_HASH_TYPE				3
-#define TCP_IPV6_HASH_TYPE			4
+#ifdef _EVEREST_MICROCODE
+#include "microcode_constants.h"
+#include "eth_rx_bd.h"
+#include "eth_tx_bd.h"
+#include "eth_rx_cqe.h"
+#include "eth_rx_sge.h"
+#include "eth_rx_cqe_next_page.h"
+#endif
+
+/* RSS hash types */
+#define DEFAULT_HASH_TYPE 0
+#define IPV4_HASH_TYPE 1
+#define TCP_IPV4_HASH_TYPE 2
+#define IPV6_HASH_TYPE 3
+#define TCP_IPV6_HASH_TYPE 4
+
+/* Ethernet Ring parmaters */
+#define X_ETH_LOCAL_RING_SIZE 13
+#define FIRST_BD_IN_PKT 0
+#define PARSE_BD_INDEX 1
+#define NUM_OF_ETH_BDS_IN_PAGE \
+	((PAGE_SIZE) / (STRUCT_SIZE(eth_tx_bd)/8))
+
+
+/* Rx ring params */
+#define U_ETH_LOCAL_BD_RING_SIZE (16)
+#define U_ETH_LOCAL_SGE_RING_SIZE (12)
+#define U_ETH_SGL_SIZE (8)
+
+
+#define U_ETH_BDS_PER_PAGE_MASK \
+	((PAGE_SIZE/(STRUCT_SIZE(eth_rx_bd)/8))-1)
+#define U_ETH_CQE_PER_PAGE_MASK \
+	((PAGE_SIZE/(STRUCT_SIZE(eth_rx_cqe)/8))-1)
+#define U_ETH_SGES_PER_PAGE_MASK \
+	((PAGE_SIZE/(STRUCT_SIZE(eth_rx_sge)/8))-1)
+
+#define U_ETH_SGES_PER_PAGE_INVERSE_MASK \
+	(0xFFFF - ((PAGE_SIZE/((STRUCT_SIZE(eth_rx_sge))/8))-1))
+
+
+#define TU_ETH_CQES_PER_PAGE \
+	(PAGE_SIZE/(STRUCT_SIZE(eth_rx_cqe_next_page)/8))
+#define U_ETH_BDS_PER_PAGE (PAGE_SIZE/(STRUCT_SIZE(eth_rx_bd)/8))
+#define U_ETH_SGES_PER_PAGE (PAGE_SIZE/(STRUCT_SIZE(eth_rx_sge)/8))
+
+#define U_ETH_UNDEFINED_Q 0xFF
 
 /* values of command IDs in the ramrod message */
-#define RAMROD_CMD_ID_ETH_PORT_SETUP			(80)
-#define RAMROD_CMD_ID_ETH_CLIENT_SETUP			(85)
-#define RAMROD_CMD_ID_ETH_STAT_QUERY			(90)
-#define RAMROD_CMD_ID_ETH_UPDATE			(100)
-#define RAMROD_CMD_ID_ETH_HALT				(105)
-#define RAMROD_CMD_ID_ETH_SET_MAC			(110)
-#define RAMROD_CMD_ID_ETH_CFC_DEL			(115)
-#define RAMROD_CMD_ID_ETH_PORT_DEL			(120)
-#define RAMROD_CMD_ID_ETH_FORWARD_SETUP 		(125)
+#define RAMROD_CMD_ID_ETH_PORT_SETUP (80)
+#define RAMROD_CMD_ID_ETH_CLIENT_SETUP (85)
+#define RAMROD_CMD_ID_ETH_STAT_QUERY (90)
+#define RAMROD_CMD_ID_ETH_UPDATE (100)
+#define RAMROD_CMD_ID_ETH_HALT (105)
+#define RAMROD_CMD_ID_ETH_SET_MAC (110)
+#define RAMROD_CMD_ID_ETH_CFC_DEL (115)
+#define RAMROD_CMD_ID_ETH_PORT_DEL (120)
+#define RAMROD_CMD_ID_ETH_FORWARD_SETUP (125)
 
 
 /* command values for set mac command */
-#define T_ETH_MAC_COMMAND_SET				0
-#define T_ETH_MAC_COMMAND_INVALIDATE			1
+#define T_ETH_MAC_COMMAND_SET 0
+#define T_ETH_MAC_COMMAND_INVALIDATE 1
+
+#define T_ETH_INDIRECTION_TABLE_SIZE 128
 
-#define T_ETH_INDIRECTION_TABLE_SIZE			128
+/*The CRC32 seed, that is used for the hash(reduction) multicast address */
+#define T_ETH_CRC32_HASH_SEED 0x00000000
 
 /* Maximal L2 clients supported */
-#define ETH_MAX_RX_CLIENTS				(18)
+#define ETH_MAX_RX_CLIENTS_E1 19
+#define ETH_MAX_RX_CLIENTS_E1H 25
+
+/* Maximal aggregation queues supported */
+#define ETH_MAX_AGGREGATION_QUEUES_E1 (32)
+#define ETH_MAX_AGGREGATION_QUEUES_E1H (64)
+
 
 /**
 * This file defines HSI constatnts common to all microcode flows
 */
 
 /* Connection types */
-#define ETH_CONNECTION_TYPE			0
+#define ETH_CONNECTION_TYPE 0
+#define TOE_CONNECTION_TYPE 1
+#define RDMA_CONNECTION_TYPE 2
+#define ISCSI_CONNECTION_TYPE 3
+#define FCOE_CONNECTION_TYPE 4
+#define RESERVED_CONNECTION_TYPE_0 5
+#define RESERVED_CONNECTION_TYPE_1 6
+#define RESERVED_CONNECTION_TYPE_2 7
+
 
-#define PROTOCOL_STATE_BIT_OFFSET		6
+#define PROTOCOL_STATE_BIT_OFFSET 6
 
-#define ETH_STATE	(ETH_CONNECTION_TYPE << PROTOCOL_STATE_BIT_OFFSET)
+#define ETH_STATE (ETH_CONNECTION_TYPE << PROTOCOL_STATE_BIT_OFFSET)
+#define TOE_STATE (TOE_CONNECTION_TYPE << PROTOCOL_STATE_BIT_OFFSET)
+#define RDMA_STATE (RDMA_CONNECTION_TYPE << PROTOCOL_STATE_BIT_OFFSET)
+#define ISCSI_STATE \
+	(ISCSI_CONNECTION_TYPE << PROTOCOL_STATE_BIT_OFFSET)
+#define FCOE_STATE (FCOE_CONNECTION_TYPE << PROTOCOL_STATE_BIT_OFFSET)
 
 /* microcode fixed page page size 4K (chains and ring segments) */
-#define MC_PAGE_SIZE						(4096)
+#define MC_PAGE_SIZE (4096)
 
-/* Host coalescing constants */
 
-/* IGU constants */
-#define IGU_PORT_BASE				0x0400
-
-#define IGU_ADDR_MSIX				0x0000
-#define IGU_ADDR_INT_ACK			0x0200
-#define IGU_ADDR_PROD_UPD			0x0201
-#define IGU_ADDR_ATTN_BITS_UPD			0x0202
-#define IGU_ADDR_ATTN_BITS_SET			0x0203
-#define IGU_ADDR_ATTN_BITS_CLR			0x0204
-#define IGU_ADDR_COALESCE_NOW			0x0205
-#define IGU_ADDR_SIMD_MASK			0x0206
-#define IGU_ADDR_SIMD_NOMASK			0x0207
-#define IGU_ADDR_MSI_CTL			0x0210
-#define IGU_ADDR_MSI_ADDR_LO			0x0211
-#define IGU_ADDR_MSI_ADDR_HI			0x0212
-#define IGU_ADDR_MSI_DATA			0x0213
-
-#define IGU_INT_ENABLE				0
-#define IGU_INT_DISABLE 			1
-#define IGU_INT_NOP				2
-#define IGU_INT_NOP2				3
+/* Host coalescing constants */
 
 /* index numbers */
-#define HC_USTORM_DEF_SB_NUM_INDICES		4
-#define HC_CSTORM_DEF_SB_NUM_INDICES		8
-#define HC_XSTORM_DEF_SB_NUM_INDICES		4
-#define HC_TSTORM_DEF_SB_NUM_INDICES		4
-#define HC_USTORM_SB_NUM_INDICES		4
-#define HC_CSTORM_SB_NUM_INDICES		4
+#define HC_USTORM_DEF_SB_NUM_INDICES 4
+#define HC_CSTORM_DEF_SB_NUM_INDICES 8
+#define HC_XSTORM_DEF_SB_NUM_INDICES 4
+#define HC_TSTORM_DEF_SB_NUM_INDICES 4
+#define HC_USTORM_SB_NUM_INDICES 4
+#define HC_CSTORM_SB_NUM_INDICES 4
 
 /* index values - which counterto update */
 
-#define HC_INDEX_U_ETH_RX_CQ_CONS		1
+#define HC_INDEX_U_TOE_RX_CQ_CONS 0
+#define HC_INDEX_U_ETH_RX_CQ_CONS 1
+#define HC_INDEX_U_ETH_RX_BD_CONS 2
+#define HC_INDEX_U_FCOE_EQ_CONS 3
+
+#define HC_INDEX_C_TOE_TX_CQ_CONS 0
+#define HC_INDEX_C_ETH_TX_CQ_CONS 1
+#define HC_INDEX_C_ISCSI_EQ_CONS 2
+
+#define HC_INDEX_DEF_X_SPQ_CONS 0
 
-#define HC_INDEX_C_ETH_TX_CQ_CONS		1
+#define HC_INDEX_DEF_C_RDMA_EQ_CONS 0
+#define HC_INDEX_DEF_C_RDMA_NAL_PROD 1
+#define HC_INDEX_DEF_C_ETH_FW_TX_CQ_CONS 2
+#define HC_INDEX_DEF_C_ETH_SLOW_PATH 3
+#define HC_INDEX_DEF_C_ETH_RDMA_CQ_CONS 4
+#define HC_INDEX_DEF_C_ETH_ISCSI_CQ_CONS 5
 
-#define HC_INDEX_DEF_X_SPQ_CONS 		0
+#define HC_INDEX_DEF_U_ETH_RDMA_RX_CQ_CONS 0
+#define HC_INDEX_DEF_U_ETH_ISCSI_RX_CQ_CONS 1
+#define HC_INDEX_DEF_U_ETH_RDMA_RX_BD_CONS 2
+#define HC_INDEX_DEF_U_ETH_ISCSI_RX_BD_CONS 3
 
-#define HC_INDEX_DEF_C_ETH_FW_TX_CQ_CONS	2
-#define HC_INDEX_DEF_C_ETH_SLOW_PATH		3
 
 /* used by the driver to get the SB offset */
-#define USTORM_ID			0
-#define CSTORM_ID			1
-#define XSTORM_ID			2
-#define TSTORM_ID			3
-#define ATTENTION_ID			4
+#define USTORM_ID 0
+#define CSTORM_ID 1
+#define XSTORM_ID 2
+#define TSTORM_ID 3
+#define ATTENTION_ID 4
 
 /* max number of slow path commands per port */
-#define MAX_RAMRODS_PER_PORT		(8)
+#define MAX_RAMRODS_PER_PORT (8)
 
 /* values for RX ETH CQE type field */
-#define RX_ETH_CQE_TYPE_ETH_FASTPATH	(0)
-#define RX_ETH_CQE_TYPE_ETH_RAMROD		(1)
-
-/* MAC address list size */
-#define T_MAC_ADDRESS_LIST_SIZE 	(96)
-
+#define RX_ETH_CQE_TYPE_ETH_FASTPATH (0)
+#define RX_ETH_CQE_TYPE_ETH_RAMROD (1)
+
+
+/**** DEFINES FOR TIMERS/CLOCKS RESOLUTIONS ****/
+#define EMULATION_FREQUENCY_FACTOR (1600)
+#define FPGA_FREQUENCY_FACTOR (100)
+
+#define TIMERS_TICK_SIZE_CHIP (1e-3)
+#define TIMERS_TICK_SIZE_EMUL \
+ ((TIMERS_TICK_SIZE_CHIP)/((EMULATION_FREQUENCY_FACTOR)))
+#define TIMERS_TICK_SIZE_FPGA \
+ ((TIMERS_TICK_SIZE_CHIP)/((FPGA_FREQUENCY_FACTOR)))
+
+#define TSEMI_CLK1_RESUL_CHIP (1e-3)
+#define TSEMI_CLK1_RESUL_EMUL \
+ ((TSEMI_CLK1_RESUL_CHIP)/(EMULATION_FREQUENCY_FACTOR))
+#define TSEMI_CLK1_RESUL_FPGA \
+ ((TSEMI_CLK1_RESUL_CHIP)/(FPGA_FREQUENCY_FACTOR))
+
+#define USEMI_CLK1_RESUL_CHIP \
+ (TIMERS_TICK_SIZE_CHIP)
+#define USEMI_CLK1_RESUL_EMUL \
+ (TIMERS_TICK_SIZE_EMUL)
+#define USEMI_CLK1_RESUL_FPGA \
+ (TIMERS_TICK_SIZE_FPGA)
+
+#define XSEMI_CLK1_RESUL_CHIP (1e-3)
+#define XSEMI_CLK1_RESUL_EMUL \
+ ((XSEMI_CLK1_RESUL_CHIP)/(EMULATION_FREQUENCY_FACTOR))
+#define XSEMI_CLK1_RESUL_FPGA \
+ ((XSEMI_CLK1_RESUL_CHIP)/(FPGA_FREQUENCY_FACTOR))
+
+#define XSEMI_CLK2_RESUL_CHIP (1e-6)
+#define XSEMI_CLK2_RESUL_EMUL \
+ ((XSEMI_CLK2_RESUL_CHIP)/(EMULATION_FREQUENCY_FACTOR))
+#define XSEMI_CLK2_RESUL_FPGA \
+ ((XSEMI_CLK2_RESUL_CHIP)/(FPGA_FREQUENCY_FACTOR))
+
+#define SDM_TIMER_TICK_RESUL_CHIP (4*(1e-6))
+#define SDM_TIMER_TICK_RESUL_EMUL \
+ ((SDM_TIMER_TICK_RESUL_CHIP)/(EMULATION_FREQUENCY_FACTOR))
+#define SDM_TIMER_TICK_RESUL_FPGA \
+ ((SDM_TIMER_TICK_RESUL_CHIP)/(FPGA_FREQUENCY_FACTOR))
+
+
+/**** END DEFINES FOR TIMERS/CLOCKS RESOLUTIONS ****/
 #define XSTORM_IP_ID_ROLL_HALF 0x8000
 #define XSTORM_IP_ID_ROLL_ALL 0
 
-#define FW_LOG_LIST_SIZE	(50)
+#define FW_LOG_LIST_SIZE (50)
+
+#define NUM_OF_PROTOCOLS 4
+#define MAX_COS_NUMBER 16
+#define MAX_T_STAT_COUNTER_ID 18
+#define MAX_X_STAT_COUNTER_ID 18
 
-#define NUM_OF_PROTOCOLS		4
-#define MAX_COS_NUMBER			16
-#define MAX_T_STAT_COUNTER_ID	18
+#define UNKNOWN_ADDRESS 0
+#define UNICAST_ADDRESS 1
+#define MULTICAST_ADDRESS 2
+#define BROADCAST_ADDRESS 3
 
-#define T_FAIR							1
-#define FAIR_MEM						2
-#define RS_PERIODIC_TIMEOUT_IN_SDM_TICS 25
+#define SINGLE_FUNCTION 0
+#define MULTI_FUNCTION 1
 
-#define UNKNOWN_ADDRESS 	0
-#define UNICAST_ADDRESS 	1
-#define MULTICAST_ADDRESS	2
-#define BROADCAST_ADDRESS	3
+#define IP_V4 0
+#define IP_V6 1
 
diff --git a/drivers/net/bnx2x_hsi.h b/drivers/net/bnx2x_hsi.h
index 96208ace1466..e515d68ea20f 100644
--- a/drivers/net/bnx2x_hsi.h
+++ b/drivers/net/bnx2x_hsi.h
@@ -132,6 +132,12 @@ struct shared_hw_cfg {					 /* NVRAM Offset */
 #define SHARED_HW_CFG_BOARD_TYPE_BCM957710T1003G    0x00000008
 #define SHARED_HW_CFG_BOARD_TYPE_BCM957710A1022G    0x00000009
 #define SHARED_HW_CFG_BOARD_TYPE_BCM957710A1021G    0x0000000a
+#define SHARED_HW_CFG_BOARD_TYPE_BCM957710A1023G    0x0000000b
+#define SHARED_HW_CFG_BOARD_TYPE_BCM957710A1033G    0x0000000c
+#define SHARED_HW_CFG_BOARD_TYPE_BCM957711T1101     0x0000000d
+#define SHARED_HW_CFG_BOARD_TYPE_BCM957711ET1201    0x0000000e
+#define SHARED_HW_CFG_BOARD_TYPE_BCM957711A1133G    0x0000000f
+#define SHARED_HW_CFG_BOARD_TYPE_BCM957711EA1233G   0x00000010
 
 #define SHARED_HW_CFG_BOARD_VER_MASK		    0xffff0000
 #define SHARED_HW_CFG_BOARD_VER_SHIFT		    16
@@ -313,6 +319,7 @@ struct shared_feat_cfg {				 /* NVRAM Offset */
 
 	u32 config;						/* 0x450 */
 #define SHARED_FEATURE_BMC_ECHO_MODE_EN 	    0x00000001
+#define SHARED_FEATURE_MF_MODE_DISABLED 	    0x00000100
 
 };
 
@@ -502,20 +509,20 @@ struct port_feat_cfg {			    /* port 0: 0x454  port 1: 0x4c8 */
 };
 
 
-/*****************************************************************************
- * Device Information							     *
- *****************************************************************************/
-struct dev_info {						     /* size */
+/****************************************************************************
+ * Device Information							    *
+ ****************************************************************************/
+struct dev_info {						    /* size */
 
-	u32    bc_rev; /* 8 bits each: major, minor, build */	        /* 4 */
+	u32    bc_rev; /* 8 bits each: major, minor, build */	       /* 4 */
 
-	struct shared_hw_cfg	 shared_hw_config;		       /* 40 */
+	struct shared_hw_cfg	 shared_hw_config;		      /* 40 */
 
-	struct port_hw_cfg	 port_hw_config[PORT_MAX];      /* 400*2=800 */
+	struct port_hw_cfg	 port_hw_config[PORT_MAX];     /* 400*2=800 */
 
-	struct shared_feat_cfg	 shared_feature_config; 	        /* 4 */
+	struct shared_feat_cfg	 shared_feature_config; 	       /* 4 */
 
-	struct port_feat_cfg	 port_feature_config[PORT_MAX]; /* 116*2=232 */
+	struct port_feat_cfg	 port_feature_config[PORT_MAX];/* 116*2=232 */
 
 };
 
@@ -632,7 +639,9 @@ struct drv_port_mb {
 #define LINK_STATUS_LINK_PARTNER_15GXFD_CAPABLE 	0x08000000
 #define LINK_STATUS_LINK_PARTNER_16GXFD_CAPABLE 	0x10000000
 
-	u32 reserved[3];
+	u32 port_stx;
+
+	u32 reserved[2];
 
 };
 
@@ -655,6 +664,11 @@ struct drv_func_mb {
 #define DRV_MSG_CODE_GET_MANUF_KEY			0x82000000
 #define DRV_MSG_CODE_LOAD_L2B_PRAM			0x90000000
 
+#define BIOS_MSG_CODE_LIC_CHALLENGE			0xff010000
+#define BIOS_MSG_CODE_LIC_RESPONSE			0xff020000
+#define BIOS_MSG_CODE_VIRT_MAC_PRIM			0xff030000
+#define BIOS_MSG_CODE_VIRT_MAC_ISCSI			0xff040000
+
 #define DRV_MSG_SEQ_NUMBER_MASK 			0x0000ffff
 
 	u32 drv_mb_param;
@@ -684,6 +698,11 @@ struct drv_func_mb {
 #define FW_MSG_CODE_L2B_PRAM_X_LOAD_FAILURE		0x90230000
 #define FW_MSG_CODE_L2B_PRAM_U_LOAD_FAILURE		0x90240000
 
+#define FW_MSG_CODE_LIC_CHALLENGE			0xff010000
+#define FW_MSG_CODE_LIC_RESPONSE			0xff020000
+#define FW_MSG_CODE_VIRT_MAC_PRIM			0xff030000
+#define FW_MSG_CODE_VIRT_MAC_ISCSI			0xff040000
+
 #define FW_MSG_SEQ_NUMBER_MASK				0x0000ffff
 
 	u32 fw_mb_param;
@@ -709,7 +728,13 @@ struct drv_func_mb {
 	u32 iscsi_boot_signature;
 	u32 iscsi_boot_block_offset;
 
-	u32 reserved[3];
+	u32 drv_status;
+#define DRV_STATUS_PMF					0x00000001
+
+	u32 virt_mac_upper;
+#define VIRT_MAC_SIGN_MASK				0xffff0000
+#define VIRT_MAC_SIGNATURE				0x564d0000
+	u32 virt_mac_lower;
 
 };
 
@@ -725,6 +750,92 @@ struct mgmtfw_state {
 };
 
 
+/****************************************************************************
+ * Multi-Function configuration 					    *
+ ****************************************************************************/
+struct shared_mf_cfg {
+
+	u32 clp_mb;
+#define SHARED_MF_CLP_SET_DEFAULT		    0x00000000
+	/* set by CLP */
+#define SHARED_MF_CLP_EXIT			    0x00000001
+	/* set by MCP */
+#define SHARED_MF_CLP_EXIT_DONE 		    0x00010000
+
+};
+
+struct port_mf_cfg {
+
+	u32 dynamic_cfg;	/* device control channel */
+#define PORT_MF_CFG_OUTER_VLAN_TAG_MASK 	    0x0000ffff
+#define PORT_MF_CFG_OUTER_VLAN_TAG_SHIFT	    0
+#define PORT_MF_CFG_DYNAMIC_CFG_ENABLED 	    0x00010000
+#define PORT_MF_CFG_DYNAMIC_CFG_DEFAULT 	    0x00000000
+
+	u32 reserved[3];
+
+};
+
+struct func_mf_cfg {
+
+	u32 config;
+	/* E/R/I/D */
+	/* function 0 of each port cannot be hidden */
+#define FUNC_MF_CFG_FUNC_HIDE			    0x00000001
+
+#define FUNC_MF_CFG_PROTOCOL_MASK		    0x00000007
+#define FUNC_MF_CFG_PROTOCOL_ETHERNET		    0x00000002
+#define FUNC_MF_CFG_PROTOCOL_ETHERNET_WITH_RDMA     0x00000004
+#define FUNC_MF_CFG_PROTOCOL_ISCSI		    0x00000006
+#define FUNC_MF_CFG_PROTOCOL_DEFAULT\
+	FUNC_MF_CFG_PROTOCOL_ETHERNET_WITH_RDMA
+
+#define FUNC_MF_CFG_FUNC_DISABLED		    0x00000008
+
+	/* PRI */
+	/* 0 - low priority, 3 - high priority */
+#define FUNC_MF_CFG_TRANSMIT_PRIORITY_MASK	    0x00000300
+#define FUNC_MF_CFG_TRANSMIT_PRIORITY_SHIFT	    8
+#define FUNC_MF_CFG_TRANSMIT_PRIORITY_DEFAULT	    0x00000000
+
+	/* MINBW, MAXBW */
+	/* value range - 0..100, increments in 100Mbps */
+#define FUNC_MF_CFG_MIN_BW_MASK 		    0x00ff0000
+#define FUNC_MF_CFG_MIN_BW_SHIFT		    16
+#define FUNC_MF_CFG_MIN_BW_DEFAULT		    0x00000000
+#define FUNC_MF_CFG_MAX_BW_MASK 		    0xff000000
+#define FUNC_MF_CFG_MAX_BW_SHIFT		    24
+#define FUNC_MF_CFG_MAX_BW_DEFAULT		    0x64000000
+
+	u32 mac_upper;		/* MAC */
+#define FUNC_MF_CFG_UPPERMAC_MASK		    0x0000ffff
+#define FUNC_MF_CFG_UPPERMAC_SHIFT		    0
+#define FUNC_MF_CFG_UPPERMAC_DEFAULT		    FUNC_MF_CFG_UPPERMAC_MASK
+	u32 mac_lower;
+#define FUNC_MF_CFG_LOWERMAC_DEFAULT		    0xffffffff
+
+	u32 e1hov_tag;	/* VNI */
+#define FUNC_MF_CFG_E1HOV_TAG_MASK		    0x0000ffff
+#define FUNC_MF_CFG_E1HOV_TAG_SHIFT		    0
+#define FUNC_MF_CFG_E1HOV_TAG_DEFAULT		    FUNC_MF_CFG_E1HOV_TAG_MASK
+
+	u32 reserved[2];
+
+};
+
+struct mf_cfg {
+
+	struct shared_mf_cfg	shared_mf_config;
+	struct port_mf_cfg	port_mf_config[PORT_MAX];
+#if defined(b710)
+	struct func_mf_cfg	func_mf_config[E1_FUNC_MAX];
+#else
+	struct func_mf_cfg	func_mf_config[E1H_FUNC_MAX];
+#endif
+
+};
+
+
 /****************************************************************************
  * Shared Memory Region 						    *
  ****************************************************************************/
@@ -760,18 +871,18 @@ struct shmem_region {			       /*   SharedMem Offset (size) */
 	struct mgmtfw_state	mgmtfw_state;	       /* 0x4ac     (0x1b8) */
 
 	struct drv_port_mb	port_mb[PORT_MAX];     /* 0x664 (16*2=0x20) */
-#if defined(b710)
-	struct drv_func_mb	func_mb[E1_FUNC_MAX];  /* 0x684 (44*2=0x58) */
-#else
 	struct drv_func_mb	func_mb[E1H_FUNC_MAX];
-#endif
+
+	struct mf_cfg		mf_cfg;
 
 };						       /* 0x6dc */
 
 
+
+
 #define BCM_5710_FW_MAJOR_VERSION			4
-#define BCM_5710_FW_MINOR_VERSION			0
-#define BCM_5710_FW_REVISION_VERSION			14
+#define BCM_5710_FW_MINOR_VERSION			5
+#define BCM_5710_FW_REVISION_VERSION			1
 #define BCM_5710_FW_COMPILE_FLAGS			1
 
 
@@ -810,7 +921,7 @@ struct doorbell_hdr {
 };
 
 /*
- * doorbell message send to the chip
+ * doorbell message sent to the chip
  */
 struct doorbell {
 #if defined(__BIG_ENDIAN)
@@ -866,8 +977,10 @@ struct parsing_flags {
 	u16 flags;
 #define PARSING_FLAGS_ETHERNET_ADDRESS_TYPE (0x1<<0)
 #define PARSING_FLAGS_ETHERNET_ADDRESS_TYPE_SHIFT 0
-#define PARSING_FLAGS_NUMBER_OF_NESTED_VLANS (0x3<<1)
-#define PARSING_FLAGS_NUMBER_OF_NESTED_VLANS_SHIFT 1
+#define PARSING_FLAGS_VLAN (0x1<<1)
+#define PARSING_FLAGS_VLAN_SHIFT 1
+#define PARSING_FLAGS_EXTRA_VLAN (0x1<<2)
+#define PARSING_FLAGS_EXTRA_VLAN_SHIFT 2
 #define PARSING_FLAGS_OVER_ETHERNET_PROTOCOL (0x3<<3)
 #define PARSING_FLAGS_OVER_ETHERNET_PROTOCOL_SHIFT 3
 #define PARSING_FLAGS_IP_OPTIONS (0x1<<5)
@@ -891,6 +1004,12 @@ struct parsing_flags {
 };
 
 
+struct regpair {
+	u32 lo;
+	u32 hi;
+};
+
+
 /*
  * dmae command structure
  */
@@ -971,72 +1090,107 @@ struct double_regpair {
 
 
 /*
- * The eth Rx Buffer Descriptor
+ * The eth storm context of Ustorm (configuration part)
  */
-struct eth_rx_bd {
-	u32 addr_lo;
-	u32 addr_hi;
-};
-
-/*
- * The eth storm context of Ustorm
- */
-struct ustorm_eth_st_context {
+struct ustorm_eth_st_context_config {
 #if defined(__BIG_ENDIAN)
-	u8 sb_index_number;
+	u8 flags;
+#define USTORM_ETH_ST_CONTEXT_CONFIG_ENABLE_MC_ALIGNMENT (0x1<<0)
+#define USTORM_ETH_ST_CONTEXT_CONFIG_ENABLE_MC_ALIGNMENT_SHIFT 0
+#define USTORM_ETH_ST_CONTEXT_CONFIG_ENABLE_DYNAMIC_HC (0x1<<1)
+#define USTORM_ETH_ST_CONTEXT_CONFIG_ENABLE_DYNAMIC_HC_SHIFT 1
+#define USTORM_ETH_ST_CONTEXT_CONFIG_ENABLE_TPA (0x1<<2)
+#define USTORM_ETH_ST_CONTEXT_CONFIG_ENABLE_TPA_SHIFT 2
+#define USTORM_ETH_ST_CONTEXT_CONFIG_ENABLE_SGE_RING (0x1<<3)
+#define USTORM_ETH_ST_CONTEXT_CONFIG_ENABLE_SGE_RING_SHIFT 3
+#define __USTORM_ETH_ST_CONTEXT_CONFIG_RESERVED0 (0xF<<4)
+#define __USTORM_ETH_ST_CONTEXT_CONFIG_RESERVED0_SHIFT 4
 	u8 status_block_id;
-	u8 __local_rx_bd_cons;
-	u8 __local_rx_bd_prod;
+	u8 clientId;
+	u8 sb_index_numbers;
+#define USTORM_ETH_ST_CONTEXT_CONFIG_CQE_SB_INDEX_NUMBER (0xF<<0)
+#define USTORM_ETH_ST_CONTEXT_CONFIG_CQE_SB_INDEX_NUMBER_SHIFT 0
+#define USTORM_ETH_ST_CONTEXT_CONFIG_BD_SB_INDEX_NUMBER (0xF<<4)
+#define USTORM_ETH_ST_CONTEXT_CONFIG_BD_SB_INDEX_NUMBER_SHIFT 4
 #elif defined(__LITTLE_ENDIAN)
-	u8 __local_rx_bd_prod;
-	u8 __local_rx_bd_cons;
+	u8 sb_index_numbers;
+#define USTORM_ETH_ST_CONTEXT_CONFIG_CQE_SB_INDEX_NUMBER (0xF<<0)
+#define USTORM_ETH_ST_CONTEXT_CONFIG_CQE_SB_INDEX_NUMBER_SHIFT 0
+#define USTORM_ETH_ST_CONTEXT_CONFIG_BD_SB_INDEX_NUMBER (0xF<<4)
+#define USTORM_ETH_ST_CONTEXT_CONFIG_BD_SB_INDEX_NUMBER_SHIFT 4
+	u8 clientId;
 	u8 status_block_id;
-	u8 sb_index_number;
+	u8 flags;
+#define USTORM_ETH_ST_CONTEXT_CONFIG_ENABLE_MC_ALIGNMENT (0x1<<0)
+#define USTORM_ETH_ST_CONTEXT_CONFIG_ENABLE_MC_ALIGNMENT_SHIFT 0
+#define USTORM_ETH_ST_CONTEXT_CONFIG_ENABLE_DYNAMIC_HC (0x1<<1)
+#define USTORM_ETH_ST_CONTEXT_CONFIG_ENABLE_DYNAMIC_HC_SHIFT 1
+#define USTORM_ETH_ST_CONTEXT_CONFIG_ENABLE_TPA (0x1<<2)
+#define USTORM_ETH_ST_CONTEXT_CONFIG_ENABLE_TPA_SHIFT 2
+#define USTORM_ETH_ST_CONTEXT_CONFIG_ENABLE_SGE_RING (0x1<<3)
+#define USTORM_ETH_ST_CONTEXT_CONFIG_ENABLE_SGE_RING_SHIFT 3
+#define __USTORM_ETH_ST_CONTEXT_CONFIG_RESERVED0 (0xF<<4)
+#define __USTORM_ETH_ST_CONTEXT_CONFIG_RESERVED0_SHIFT 4
 #endif
 #if defined(__BIG_ENDIAN)
-	u16 rcq_cons;
-	u16 rx_bd_cons;
+	u16 bd_buff_size;
+	u16 mc_alignment_size;
 #elif defined(__LITTLE_ENDIAN)
-	u16 rx_bd_cons;
-	u16 rcq_cons;
+	u16 mc_alignment_size;
+	u16 bd_buff_size;
 #endif
-	u32 rx_bd_page_base_lo;
-	u32 rx_bd_page_base_hi;
-	u32 rcq_base_address_lo;
-	u32 rcq_base_address_hi;
 #if defined(__BIG_ENDIAN)
-	u16 __num_of_returned_cqes;
-	u8 num_rss;
-	u8 flags;
-#define USTORM_ETH_ST_CONTEXT_ENABLE_MC_ALIGNMENT (0x1<<0)
-#define USTORM_ETH_ST_CONTEXT_ENABLE_MC_ALIGNMENT_SHIFT 0
-#define USTORM_ETH_ST_CONTEXT_ENABLE_DYNAMIC_HC (0x1<<1)
-#define USTORM_ETH_ST_CONTEXT_ENABLE_DYNAMIC_HC_SHIFT 1
-#define USTORM_ETH_ST_CONTEXT_ENABLE_TPA (0x1<<2)
-#define USTORM_ETH_ST_CONTEXT_ENABLE_TPA_SHIFT 2
-#define __USTORM_ETH_ST_CONTEXT_RESERVED0 (0x1F<<3)
-#define __USTORM_ETH_ST_CONTEXT_RESERVED0_SHIFT 3
+	u8 __local_sge_prod;
+	u8 __local_bd_prod;
+	u16 sge_buff_size;
 #elif defined(__LITTLE_ENDIAN)
-	u8 flags;
-#define USTORM_ETH_ST_CONTEXT_ENABLE_MC_ALIGNMENT (0x1<<0)
-#define USTORM_ETH_ST_CONTEXT_ENABLE_MC_ALIGNMENT_SHIFT 0
-#define USTORM_ETH_ST_CONTEXT_ENABLE_DYNAMIC_HC (0x1<<1)
-#define USTORM_ETH_ST_CONTEXT_ENABLE_DYNAMIC_HC_SHIFT 1
-#define USTORM_ETH_ST_CONTEXT_ENABLE_TPA (0x1<<2)
-#define USTORM_ETH_ST_CONTEXT_ENABLE_TPA_SHIFT 2
-#define __USTORM_ETH_ST_CONTEXT_RESERVED0 (0x1F<<3)
-#define __USTORM_ETH_ST_CONTEXT_RESERVED0_SHIFT 3
-	u8 num_rss;
-	u16 __num_of_returned_cqes;
+	u16 sge_buff_size;
+	u8 __local_bd_prod;
+	u8 __local_sge_prod;
 #endif
 #if defined(__BIG_ENDIAN)
-	u16 mc_alignment_size;
-	u16 agg_threshold;
+	u16 __bd_cons;
+	u16 __sge_cons;
 #elif defined(__LITTLE_ENDIAN)
-	u16 agg_threshold;
-	u16 mc_alignment_size;
+	u16 __sge_cons;
+	u16 __bd_cons;
 #endif
+	u32 bd_page_base_lo;
+	u32 bd_page_base_hi;
+	u32 sge_page_base_lo;
+	u32 sge_page_base_hi;
+};
+
+/*
+ * The eth Rx Buffer Descriptor
+ */
+struct eth_rx_bd {
+	u32 addr_lo;
+	u32 addr_hi;
+};
+
+/*
+ * The eth Rx SGE Descriptor
+ */
+struct eth_rx_sge {
+	u32 addr_lo;
+	u32 addr_hi;
+};
+
+/*
+ * Local BDs and SGEs rings (in ETH)
+ */
+struct eth_local_rx_rings {
 	struct eth_rx_bd __local_bd_ring[16];
+	struct eth_rx_sge __local_sge_ring[12];
+};
+
+/*
+ * The eth storm context of Ustorm
+ */
+struct ustorm_eth_st_context {
+	struct ustorm_eth_st_context_config common;
+	struct eth_local_rx_rings __rings;
 };
 
 /*
@@ -1107,9 +1261,9 @@ struct xstorm_eth_extra_ag_context_section {
 #if defined(__BIG_ENDIAN)
 	u16 __reserved3;
 	u8 __reserved2;
-	u8 __agg_misc7;
+	u8 __da_only_cnt;
 #elif defined(__LITTLE_ENDIAN)
-	u8 __agg_misc7;
+	u8 __da_only_cnt;
 	u8 __reserved2;
 	u16 __reserved3;
 #endif
@@ -1387,7 +1541,13 @@ struct timers_block_context {
 	u32 __reserved_0;
 	u32 __reserved_1;
 	u32 __reserved_2;
-	u32 __reserved_flags;
+	u32 flags;
+#define __TIMERS_BLOCK_CONTEXT_NUM_OF_ACTIVE_TIMERS (0x3<<0)
+#define __TIMERS_BLOCK_CONTEXT_NUM_OF_ACTIVE_TIMERS_SHIFT 0
+#define TIMERS_BLOCK_CONTEXT_CONN_VALID_FLG (0x1<<2)
+#define TIMERS_BLOCK_CONTEXT_CONN_VALID_FLG_SHIFT 2
+#define __TIMERS_BLOCK_CONTEXT_RESERVED0 (0x1FFFFFFF<<3)
+#define __TIMERS_BLOCK_CONTEXT_RESERVED0_SHIFT 3
 };
 
 /*
@@ -1497,11 +1657,19 @@ struct xstorm_eth_st_context {
 	u32 tx_bd_page_base_hi;
 #if defined(__BIG_ENDIAN)
 	u16 tx_bd_cons;
-	u8 __reserved0;
+	u8 statistics_data;
+#define XSTORM_ETH_ST_CONTEXT_STATISTICS_COUNTER_ID (0x7F<<0)
+#define XSTORM_ETH_ST_CONTEXT_STATISTICS_COUNTER_ID_SHIFT 0
+#define XSTORM_ETH_ST_CONTEXT_STATISTICS_ENABLE (0x1<<7)
+#define XSTORM_ETH_ST_CONTEXT_STATISTICS_ENABLE_SHIFT 7
 	u8 __local_tx_bd_prod;
 #elif defined(__LITTLE_ENDIAN)
 	u8 __local_tx_bd_prod;
-	u8 __reserved0;
+	u8 statistics_data;
+#define XSTORM_ETH_ST_CONTEXT_STATISTICS_COUNTER_ID (0x7F<<0)
+#define XSTORM_ETH_ST_CONTEXT_STATISTICS_COUNTER_ID_SHIFT 0
+#define XSTORM_ETH_ST_CONTEXT_STATISTICS_ENABLE (0x1<<7)
+#define XSTORM_ETH_ST_CONTEXT_STATISTICS_ENABLE_SHIFT 7
 	u16 tx_bd_cons;
 #endif
 	u32 db_data_addr_lo;
@@ -1578,7 +1746,7 @@ struct eth_tx_doorbell {
 struct ustorm_def_status_block {
 	u16 index_values[HC_USTORM_DEF_SB_NUM_INDICES];
 	u16 status_block_index;
-	u8 reserved0;
+	u8 func;
 	u8 status_block_id;
 	u32 __flags;
 };
@@ -1589,7 +1757,7 @@ struct ustorm_def_status_block {
 struct cstorm_def_status_block {
 	u16 index_values[HC_CSTORM_DEF_SB_NUM_INDICES];
 	u16 status_block_index;
-	u8 reserved0;
+	u8 func;
 	u8 status_block_id;
 	u32 __flags;
 };
@@ -1600,7 +1768,7 @@ struct cstorm_def_status_block {
 struct xstorm_def_status_block {
 	u16 index_values[HC_XSTORM_DEF_SB_NUM_INDICES];
 	u16 status_block_index;
-	u8 reserved0;
+	u8 func;
 	u8 status_block_id;
 	u32 __flags;
 };
@@ -1611,7 +1779,7 @@ struct xstorm_def_status_block {
 struct tstorm_def_status_block {
 	u16 index_values[HC_TSTORM_DEF_SB_NUM_INDICES];
 	u16 status_block_index;
-	u8 reserved0;
+	u8 func;
 	u8 status_block_id;
 	u32 __flags;
 };
@@ -1634,7 +1802,7 @@ struct host_def_status_block {
 struct ustorm_status_block {
 	u16 index_values[HC_USTORM_SB_NUM_INDICES];
 	u16 status_block_index;
-	u8 reserved0;
+	u8 func;
 	u8 status_block_id;
 	u32 __flags;
 };
@@ -1645,7 +1813,7 @@ struct ustorm_status_block {
 struct cstorm_status_block {
 	u16 index_values[HC_CSTORM_SB_NUM_INDICES];
 	u16 status_block_index;
-	u8 reserved0;
+	u8 func;
 	u8 status_block_id;
 	u32 __flags;
 };
@@ -1683,20 +1851,21 @@ struct eth_dynamic_hc_config {
  * regular eth FP CQE parameters struct
  */
 struct eth_fast_path_rx_cqe {
-	u8 type;
-	u8 error_type_flags;
-#define ETH_FAST_PATH_RX_CQE_PHY_DECODE_ERR_FLG (0x1<<0)
-#define ETH_FAST_PATH_RX_CQE_PHY_DECODE_ERR_FLG_SHIFT 0
-#define ETH_FAST_PATH_RX_CQE_IP_BAD_XSUM_FLG (0x1<<1)
-#define ETH_FAST_PATH_RX_CQE_IP_BAD_XSUM_FLG_SHIFT 1
-#define ETH_FAST_PATH_RX_CQE_L4_BAD_XSUM_FLG (0x1<<2)
-#define ETH_FAST_PATH_RX_CQE_L4_BAD_XSUM_FLG_SHIFT 2
-#define ETH_FAST_PATH_RX_CQE_START_FLG (0x1<<3)
-#define ETH_FAST_PATH_RX_CQE_START_FLG_SHIFT 3
-#define ETH_FAST_PATH_RX_CQE_END_FLG (0x1<<4)
-#define ETH_FAST_PATH_RX_CQE_END_FLG_SHIFT 4
-#define ETH_FAST_PATH_RX_CQE_RESERVED0 (0x7<<5)
-#define ETH_FAST_PATH_RX_CQE_RESERVED0_SHIFT 5
+	u8 type_error_flags;
+#define ETH_FAST_PATH_RX_CQE_TYPE (0x1<<0)
+#define ETH_FAST_PATH_RX_CQE_TYPE_SHIFT 0
+#define ETH_FAST_PATH_RX_CQE_PHY_DECODE_ERR_FLG (0x1<<1)
+#define ETH_FAST_PATH_RX_CQE_PHY_DECODE_ERR_FLG_SHIFT 1
+#define ETH_FAST_PATH_RX_CQE_IP_BAD_XSUM_FLG (0x1<<2)
+#define ETH_FAST_PATH_RX_CQE_IP_BAD_XSUM_FLG_SHIFT 2
+#define ETH_FAST_PATH_RX_CQE_L4_BAD_XSUM_FLG (0x1<<3)
+#define ETH_FAST_PATH_RX_CQE_L4_BAD_XSUM_FLG_SHIFT 3
+#define ETH_FAST_PATH_RX_CQE_START_FLG (0x1<<4)
+#define ETH_FAST_PATH_RX_CQE_START_FLG_SHIFT 4
+#define ETH_FAST_PATH_RX_CQE_END_FLG (0x1<<5)
+#define ETH_FAST_PATH_RX_CQE_END_FLG_SHIFT 5
+#define ETH_FAST_PATH_RX_CQE_RESERVED0 (0x3<<6)
+#define ETH_FAST_PATH_RX_CQE_RESERVED0_SHIFT 6
 	u8 status_flags;
 #define ETH_FAST_PATH_RX_CQE_RSS_HASH_TYPE (0x7<<0)
 #define ETH_FAST_PATH_RX_CQE_RSS_HASH_TYPE_SHIFT 0
@@ -1711,11 +1880,13 @@ struct eth_fast_path_rx_cqe {
 #define ETH_FAST_PATH_RX_CQE_L4_XSUM_NO_VALIDATION_FLG (0x1<<7)
 #define ETH_FAST_PATH_RX_CQE_L4_XSUM_NO_VALIDATION_FLG_SHIFT 7
 	u8 placement_offset;
+	u8 queue_index;
 	u32 rss_hash_result;
 	u16 vlan_tag;
 	u16 pkt_len;
-	u16 queue_index;
+	u16 len_on_bd;
 	struct parsing_flags pars_flags;
+	u16 sgl[8];
 };
 
 
@@ -1728,6 +1899,23 @@ struct eth_halt_ramrod_data {
 };
 
 
+/*
+ * The data for statistics query ramrod
+ */
+struct eth_query_ramrod_data {
+#if defined(__BIG_ENDIAN)
+	u8 reserved0;
+	u8 collect_port_1b;
+	u16 drv_counter;
+#elif defined(__LITTLE_ENDIAN)
+	u16 drv_counter;
+	u8 collect_port_1b;
+	u8 reserved0;
+#endif
+	u32 ctr_id_vector;
+};
+
+
 /*
  * Place holder for ramrods protocol specific data
  */
@@ -1758,15 +1946,20 @@ struct eth_rx_bd_next_page {
  * Eth Rx Cqe structure- general structure for ramrods
  */
 struct common_ramrod_eth_rx_cqe {
-	u8 type;
+	u8 ramrod_type;
+#define COMMON_RAMROD_ETH_RX_CQE_TYPE (0x1<<0)
+#define COMMON_RAMROD_ETH_RX_CQE_TYPE_SHIFT 0
+#define COMMON_RAMROD_ETH_RX_CQE_RESERVED0 (0x7F<<1)
+#define COMMON_RAMROD_ETH_RX_CQE_RESERVED0_SHIFT 1
 	u8 conn_type_3b;
-	u16 reserved;
+	u16 reserved1;
 	u32 conn_and_cmd_data;
 #define COMMON_RAMROD_ETH_RX_CQE_CID (0xFFFFFF<<0)
 #define COMMON_RAMROD_ETH_RX_CQE_CID_SHIFT 0
 #define COMMON_RAMROD_ETH_RX_CQE_CMD_ID (0xFF<<24)
 #define COMMON_RAMROD_ETH_RX_CQE_CMD_ID_SHIFT 24
 	struct ramrod_data protocol_data;
+	u32 reserved2[4];
 };
 
 /*
@@ -1775,8 +1968,7 @@ struct common_ramrod_eth_rx_cqe {
 struct eth_rx_cqe_next_page {
 	u32 addr_lo;
 	u32 addr_hi;
-	u32 reserved0;
-	u32 reserved1;
+	u32 reserved[6];
 };
 
 /*
@@ -1806,11 +1998,6 @@ struct spe_hdr {
 	u16 reserved;
 };
 
-struct regpair {
-	u32 lo;
-	u32 hi;
-};
-
 /*
  * ethernet slow path element
  */
@@ -1821,6 +2008,7 @@ union eth_specific_data {
 	struct eth_halt_ramrod_data halt_ramrod_data;
 	struct regpair leading_cqe_addr;
 	struct regpair update_data_addr;
+	struct eth_query_ramrod_data query_ramrod_data;
 };
 
 /*
@@ -1843,10 +2031,13 @@ struct eth_tx_db_data {
 
 
 /*
- * Common configuration parameters per port in Tstorm
+ * Common configuration parameters per function in Tstorm
  */
 struct tstorm_eth_function_common_config {
-	u32 config_flags;
+#if defined(__BIG_ENDIAN)
+	u8 leading_client_id;
+	u8 rss_result_mask;
+	u16 config_flags;
 #define TSTORM_ETH_FUNCTION_COMMON_CONFIG_RSS_IPV4_CAPABILITY (0x1<<0)
 #define TSTORM_ETH_FUNCTION_COMMON_CONFIG_RSS_IPV4_CAPABILITY_SHIFT 0
 #define TSTORM_ETH_FUNCTION_COMMON_CONFIG_RSS_IPV4_TCP_CAPABILITY (0x1<<1)
@@ -1859,17 +2050,32 @@ struct tstorm_eth_function_common_config {
 #define TSTORM_ETH_FUNCTION_COMMON_CONFIG_RSS_ENABLE_SHIFT 4
 #define TSTORM_ETH_FUNCTION_COMMON_CONFIG_DEFAULT_ENABLE (0x1<<5)
 #define TSTORM_ETH_FUNCTION_COMMON_CONFIG_DEFAULT_ENABLE_SHIFT 5
-#define __TSTORM_ETH_FUNCTION_COMMON_CONFIG_RESERVED0 (0x3FFFFFF<<6)
-#define __TSTORM_ETH_FUNCTION_COMMON_CONFIG_RESERVED0_SHIFT 6
-#if defined(__BIG_ENDIAN)
-	u16 __secondary_vlan_id;
-	u8 leading_client_id;
-	u8 rss_result_mask;
+#define TSTORM_ETH_FUNCTION_COMMON_CONFIG_VLAN_IN_CAM (0x1<<6)
+#define TSTORM_ETH_FUNCTION_COMMON_CONFIG_VLAN_IN_CAM_SHIFT 6
+#define __TSTORM_ETH_FUNCTION_COMMON_CONFIG_RESERVED0 (0x1FF<<7)
+#define __TSTORM_ETH_FUNCTION_COMMON_CONFIG_RESERVED0_SHIFT 7
 #elif defined(__LITTLE_ENDIAN)
+	u16 config_flags;
+#define TSTORM_ETH_FUNCTION_COMMON_CONFIG_RSS_IPV4_CAPABILITY (0x1<<0)
+#define TSTORM_ETH_FUNCTION_COMMON_CONFIG_RSS_IPV4_CAPABILITY_SHIFT 0
+#define TSTORM_ETH_FUNCTION_COMMON_CONFIG_RSS_IPV4_TCP_CAPABILITY (0x1<<1)
+#define TSTORM_ETH_FUNCTION_COMMON_CONFIG_RSS_IPV4_TCP_CAPABILITY_SHIFT 1
+#define TSTORM_ETH_FUNCTION_COMMON_CONFIG_RSS_IPV6_CAPABILITY (0x1<<2)
+#define TSTORM_ETH_FUNCTION_COMMON_CONFIG_RSS_IPV6_CAPABILITY_SHIFT 2
+#define TSTORM_ETH_FUNCTION_COMMON_CONFIG_RSS_IPV6_TCP_CAPABILITY (0x1<<3)
+#define TSTORM_ETH_FUNCTION_COMMON_CONFIG_RSS_IPV6_TCP_CAPABILITY_SHIFT 3
+#define TSTORM_ETH_FUNCTION_COMMON_CONFIG_RSS_ENABLE (0x1<<4)
+#define TSTORM_ETH_FUNCTION_COMMON_CONFIG_RSS_ENABLE_SHIFT 4
+#define TSTORM_ETH_FUNCTION_COMMON_CONFIG_DEFAULT_ENABLE (0x1<<5)
+#define TSTORM_ETH_FUNCTION_COMMON_CONFIG_DEFAULT_ENABLE_SHIFT 5
+#define TSTORM_ETH_FUNCTION_COMMON_CONFIG_VLAN_IN_CAM (0x1<<6)
+#define TSTORM_ETH_FUNCTION_COMMON_CONFIG_VLAN_IN_CAM_SHIFT 6
+#define __TSTORM_ETH_FUNCTION_COMMON_CONFIG_RESERVED0 (0x1FF<<7)
+#define __TSTORM_ETH_FUNCTION_COMMON_CONFIG_RESERVED0_SHIFT 7
 	u8 rss_result_mask;
 	u8 leading_client_id;
-	u16 __secondary_vlan_id;
 #endif
+	u16 vlan_id[2];
 };
 
 /*
@@ -1887,7 +2093,7 @@ struct eth_update_ramrod_data {
 struct mac_configuration_hdr {
 	u8 length_6b;
 	u8 offset;
-	u16 reserved0;
+	u16 client_id;
 	u32 reserved1;
 };
 
@@ -1943,16 +2149,56 @@ struct mac_configuration_cmd {
 };
 
 
+/*
+ * MAC address in list for ramrod
+ */
+struct mac_configuration_entry_e1h {
+	u16 lsb_mac_addr;
+	u16 middle_mac_addr;
+	u16 msb_mac_addr;
+	u16 vlan_id;
+	u16 e1hov_id;
+	u8 client_id;
+	u8 flags;
+#define MAC_CONFIGURATION_ENTRY_E1H_PORT (0x1<<0)
+#define MAC_CONFIGURATION_ENTRY_E1H_PORT_SHIFT 0
+#define MAC_CONFIGURATION_ENTRY_E1H_ACTION_TYPE (0x1<<1)
+#define MAC_CONFIGURATION_ENTRY_E1H_ACTION_TYPE_SHIFT 1
+#define MAC_CONFIGURATION_ENTRY_E1H_RDMA_MAC (0x1<<2)
+#define MAC_CONFIGURATION_ENTRY_E1H_RDMA_MAC_SHIFT 2
+#define MAC_CONFIGURATION_ENTRY_E1H_RESERVED0 (0x1F<<3)
+#define MAC_CONFIGURATION_ENTRY_E1H_RESERVED0_SHIFT 3
+};
+
+/*
+ * MAC filtering configuration command
+ */
+struct mac_configuration_cmd_e1h {
+	struct mac_configuration_hdr hdr;
+	struct mac_configuration_entry_e1h config_table[32];
+};
+
+
+/*
+ * approximate-match multicast filtering for E1H per function in Tstorm
+ */
+struct tstorm_eth_approximate_match_multicast_filtering {
+	u32 mcast_add_hash_bit_array[8];
+};
+
+
 /*
  * Configuration parameters per client in Tstorm
  */
 struct tstorm_eth_client_config {
 #if defined(__BIG_ENDIAN)
-	u16 statistics_counter_id;
+	u8 max_sges_for_packet;
+	u8 statistics_counter_id;
 	u16 mtu;
 #elif defined(__LITTLE_ENDIAN)
 	u16 mtu;
-	u16 statistics_counter_id;
+	u8 statistics_counter_id;
+	u8 max_sges_for_packet;
 #endif
 #if defined(__BIG_ENDIAN)
 	u16 drop_flags;
@@ -1960,42 +2206,42 @@ struct tstorm_eth_client_config {
 #define TSTORM_ETH_CLIENT_CONFIG_DROP_IP_CS_ERR_SHIFT 0
 #define TSTORM_ETH_CLIENT_CONFIG_DROP_TCP_CS_ERR (0x1<<1)
 #define TSTORM_ETH_CLIENT_CONFIG_DROP_TCP_CS_ERR_SHIFT 1
-#define TSTORM_ETH_CLIENT_CONFIG_DROP_MAC_ERR (0x1<<2)
-#define TSTORM_ETH_CLIENT_CONFIG_DROP_MAC_ERR_SHIFT 2
-#define TSTORM_ETH_CLIENT_CONFIG_DROP_TTL0 (0x1<<3)
-#define TSTORM_ETH_CLIENT_CONFIG_DROP_TTL0_SHIFT 3
-#define TSTORM_ETH_CLIENT_CONFIG_DROP_UDP_CS_ERR (0x1<<4)
-#define TSTORM_ETH_CLIENT_CONFIG_DROP_UDP_CS_ERR_SHIFT 4
-#define __TSTORM_ETH_CLIENT_CONFIG_RESERVED1 (0x7FF<<5)
-#define __TSTORM_ETH_CLIENT_CONFIG_RESERVED1_SHIFT 5
+#define TSTORM_ETH_CLIENT_CONFIG_DROP_TTL0 (0x1<<2)
+#define TSTORM_ETH_CLIENT_CONFIG_DROP_TTL0_SHIFT 2
+#define TSTORM_ETH_CLIENT_CONFIG_DROP_UDP_CS_ERR (0x1<<3)
+#define TSTORM_ETH_CLIENT_CONFIG_DROP_UDP_CS_ERR_SHIFT 3
+#define __TSTORM_ETH_CLIENT_CONFIG_RESERVED1 (0xFFF<<4)
+#define __TSTORM_ETH_CLIENT_CONFIG_RESERVED1_SHIFT 4
 	u16 config_flags;
 #define TSTORM_ETH_CLIENT_CONFIG_VLAN_REMOVAL_ENABLE (0x1<<0)
 #define TSTORM_ETH_CLIENT_CONFIG_VLAN_REMOVAL_ENABLE_SHIFT 0
 #define TSTORM_ETH_CLIENT_CONFIG_STATSITICS_ENABLE (0x1<<1)
 #define TSTORM_ETH_CLIENT_CONFIG_STATSITICS_ENABLE_SHIFT 1
-#define __TSTORM_ETH_CLIENT_CONFIG_RESERVED0 (0x3FFF<<2)
-#define __TSTORM_ETH_CLIENT_CONFIG_RESERVED0_SHIFT 2
+#define TSTORM_ETH_CLIENT_CONFIG_ENABLE_SGE_RING (0x1<<2)
+#define TSTORM_ETH_CLIENT_CONFIG_ENABLE_SGE_RING_SHIFT 2
+#define __TSTORM_ETH_CLIENT_CONFIG_RESERVED0 (0x1FFF<<3)
+#define __TSTORM_ETH_CLIENT_CONFIG_RESERVED0_SHIFT 3
 #elif defined(__LITTLE_ENDIAN)
 	u16 config_flags;
 #define TSTORM_ETH_CLIENT_CONFIG_VLAN_REMOVAL_ENABLE (0x1<<0)
 #define TSTORM_ETH_CLIENT_CONFIG_VLAN_REMOVAL_ENABLE_SHIFT 0
 #define TSTORM_ETH_CLIENT_CONFIG_STATSITICS_ENABLE (0x1<<1)
 #define TSTORM_ETH_CLIENT_CONFIG_STATSITICS_ENABLE_SHIFT 1
-#define __TSTORM_ETH_CLIENT_CONFIG_RESERVED0 (0x3FFF<<2)
-#define __TSTORM_ETH_CLIENT_CONFIG_RESERVED0_SHIFT 2
+#define TSTORM_ETH_CLIENT_CONFIG_ENABLE_SGE_RING (0x1<<2)
+#define TSTORM_ETH_CLIENT_CONFIG_ENABLE_SGE_RING_SHIFT 2
+#define __TSTORM_ETH_CLIENT_CONFIG_RESERVED0 (0x1FFF<<3)
+#define __TSTORM_ETH_CLIENT_CONFIG_RESERVED0_SHIFT 3
 	u16 drop_flags;
 #define TSTORM_ETH_CLIENT_CONFIG_DROP_IP_CS_ERR (0x1<<0)
 #define TSTORM_ETH_CLIENT_CONFIG_DROP_IP_CS_ERR_SHIFT 0
 #define TSTORM_ETH_CLIENT_CONFIG_DROP_TCP_CS_ERR (0x1<<1)
 #define TSTORM_ETH_CLIENT_CONFIG_DROP_TCP_CS_ERR_SHIFT 1
-#define TSTORM_ETH_CLIENT_CONFIG_DROP_MAC_ERR (0x1<<2)
-#define TSTORM_ETH_CLIENT_CONFIG_DROP_MAC_ERR_SHIFT 2
-#define TSTORM_ETH_CLIENT_CONFIG_DROP_TTL0 (0x1<<3)
-#define TSTORM_ETH_CLIENT_CONFIG_DROP_TTL0_SHIFT 3
-#define TSTORM_ETH_CLIENT_CONFIG_DROP_UDP_CS_ERR (0x1<<4)
-#define TSTORM_ETH_CLIENT_CONFIG_DROP_UDP_CS_ERR_SHIFT 4
-#define __TSTORM_ETH_CLIENT_CONFIG_RESERVED1 (0x7FF<<5)
-#define __TSTORM_ETH_CLIENT_CONFIG_RESERVED1_SHIFT 5
+#define TSTORM_ETH_CLIENT_CONFIG_DROP_TTL0 (0x1<<2)
+#define TSTORM_ETH_CLIENT_CONFIG_DROP_TTL0_SHIFT 2
+#define TSTORM_ETH_CLIENT_CONFIG_DROP_UDP_CS_ERR (0x1<<3)
+#define TSTORM_ETH_CLIENT_CONFIG_DROP_UDP_CS_ERR_SHIFT 3
+#define __TSTORM_ETH_CLIENT_CONFIG_RESERVED1 (0xFFF<<4)
+#define __TSTORM_ETH_CLIENT_CONFIG_RESERVED1_SHIFT 4
 #endif
 };
 
@@ -2011,96 +2257,112 @@ struct tstorm_eth_mac_filter_config {
 	u32 bcast_drop_all;
 	u32 bcast_accept_all;
 	u32 strict_vlan;
-	u32 __secondary_vlan_clients;
+	u32 vlan_filter[2];
+	u32 reserved;
 };
 
 
-struct rate_shaping_per_protocol {
+/*
+ * Three RX producers for ETH
+ */
+struct tstorm_eth_rx_producers {
 #if defined(__BIG_ENDIAN)
-	u16 reserved0;
-	u16 protocol_rate;
+	u16 bd_prod;
+	u16 cqe_prod;
 #elif defined(__LITTLE_ENDIAN)
-	u16 protocol_rate;
-	u16 reserved0;
+	u16 cqe_prod;
+	u16 bd_prod;
 #endif
-	u32 protocol_quota;
-	s32 current_credit;
-	u32 reserved;
-};
-
-struct rate_shaping_vars {
-	struct rate_shaping_per_protocol protocol_vars[NUM_OF_PROTOCOLS];
-	u32 pause_mask;
-	u32 periodic_stop;
-	u32 rs_periodic_timeout;
-	u32 rs_threshold;
-	u32 last_periodic_time;
-	u32 reserved;
-};
-
-struct fairness_per_protocol {
-	u32 credit_delta;
-	s32 fair_credit;
 #if defined(__BIG_ENDIAN)
-	u16 reserved0;
-	u8 state;
-	u8 weight;
+	u16 reserved;
+	u16 sge_prod;
 #elif defined(__LITTLE_ENDIAN)
-	u8 weight;
-	u8 state;
-	u16 reserved0;
+	u16 sge_prod;
+	u16 reserved;
 #endif
-	u32 reserved1;
 };
 
-struct fairness_vars {
-	struct fairness_per_protocol protocol_vars[NUM_OF_PROTOCOLS];
-	u32 upper_bound;
-	u32 port_rate;
-	u32 pause_mask;
-	u32 fair_threshold;
-};
 
-struct safc_struct {
-	u32 cur_pause_mask;
-	u32 expire_time;
+/*
+ * common flag to indicate existance of TPA.
+ */
+struct tstorm_eth_tpa_exist {
 #if defined(__BIG_ENDIAN)
-	u16 reserved0;
-	u8 cur_cos_types;
-	u8 safc_timeout_usec;
+	u16 reserved1;
+	u8 reserved0;
+	u8 tpa_exist;
 #elif defined(__LITTLE_ENDIAN)
-	u8 safc_timeout_usec;
-	u8 cur_cos_types;
-	u16 reserved0;
+	u8 tpa_exist;
+	u8 reserved0;
+	u16 reserved1;
 #endif
-	u32 reserved1;
+	u32 reserved2;
 };
 
-struct demo_struct {
+
+/*
+ * per-port SAFC demo variables
+ */
+struct cmng_flags_per_port {
 	u8 con_number[NUM_OF_PROTOCOLS];
 #if defined(__BIG_ENDIAN)
-	u8 reserved1;
 	u8 fairness_enable;
 	u8 rate_shaping_enable;
-	u8 cmng_enable;
+	u8 cmng_protocol_enable;
+	u8 cmng_vn_enable;
 #elif defined(__LITTLE_ENDIAN)
-	u8 cmng_enable;
+	u8 cmng_vn_enable;
+	u8 cmng_protocol_enable;
 	u8 rate_shaping_enable;
 	u8 fairness_enable;
-	u8 reserved1;
 #endif
 };
 
-struct cmng_struct {
-	struct rate_shaping_vars rs_vars;
-	struct fairness_vars fair_vars;
-	struct safc_struct safc_vars;
-	struct demo_struct demo_vars;
+
+/*
+ * per-port rate shaping variables
+ */
+struct rate_shaping_vars_per_port {
+	u32 rs_periodic_timeout;
+	u32 rs_threshold;
+};
+
+
+/*
+ * per-port fairness variables
+ */
+struct fairness_vars_per_port {
+	u32 upper_bound;
+	u32 fair_threshold;
+	u32 fairness_timeout;
+};
+
+
+/*
+ * per-port SAFC variables
+ */
+struct safc_struct_per_port {
+#if defined(__BIG_ENDIAN)
+	u16 __reserved0;
+	u8 cur_cos_types;
+	u8 safc_timeout_usec;
+#elif defined(__LITTLE_ENDIAN)
+	u8 safc_timeout_usec;
+	u8 cur_cos_types;
+	u16 __reserved0;
+#endif
+	u8 cos_to_protocol[MAX_COS_NUMBER];
 };
 
 
-struct cos_to_protocol {
-	u8 mask[MAX_COS_NUMBER];
+/*
+ * Per-port congestion management variables
+ */
+struct cmng_struct_per_port {
+	struct rate_shaping_vars_per_port rs_vars;
+	struct fairness_vars_per_port fair_vars;
+	struct safc_struct_per_port safc_vars;
+	struct cmng_flags_per_port flags;
 };
 
 
@@ -2161,6 +2423,16 @@ struct eth_stats_query {
 };
 
 
+/*
+ * per-vnic fairness variables
+ */
+struct fairness_vars_per_vn {
+	u32 protocol_credit_delta[NUM_OF_PROTOCOLS];
+	u32 vn_credit_delta;
+	u32 __reserved0;
+};
+
+
 /*
  * FW version stored in the Xstorm RAM
  */
@@ -2179,8 +2451,10 @@ struct fw_version {
 #define FW_VERSION_OPTIMIZED_SHIFT 0
 #define FW_VERSION_BIG_ENDIEN (0x1<<1)
 #define FW_VERSION_BIG_ENDIEN_SHIFT 1
-#define __FW_VERSION_RESERVED (0x3FFFFFFF<<2)
-#define __FW_VERSION_RESERVED_SHIFT 2
+#define FW_VERSION_CHIP_VERSION (0x3<<2)
+#define FW_VERSION_CHIP_VERSION_SHIFT 2
+#define __FW_VERSION_RESERVED (0xFFFFFFF<<4)
+#define __FW_VERSION_RESERVED_SHIFT 4
 };
 
 
@@ -2188,15 +2462,9 @@ struct fw_version {
  * FW version stored in first line of pram
  */
 struct pram_fw_version {
-#if defined(__BIG_ENDIAN)
-	u16 patch;
-	u8 primary;
-	u8 client;
-#elif defined(__LITTLE_ENDIAN)
 	u8 client;
 	u8 primary;
 	u16 patch;
-#endif
 	u8 flags;
 #define PRAM_FW_VERSION_OPTIMIZED (0x1<<0)
 #define PRAM_FW_VERSION_OPTIMIZED_SHIFT 0
@@ -2204,8 +2472,34 @@ struct pram_fw_version {
 #define PRAM_FW_VERSION_STORM_ID_SHIFT 1
 #define PRAM_FW_VERSION_BIG_ENDIEN (0x1<<3)
 #define PRAM_FW_VERSION_BIG_ENDIEN_SHIFT 3
-#define __PRAM_FW_VERSION_RESERVED0 (0xF<<4)
-#define __PRAM_FW_VERSION_RESERVED0_SHIFT 4
+#define PRAM_FW_VERSION_CHIP_VERSION (0x3<<4)
+#define PRAM_FW_VERSION_CHIP_VERSION_SHIFT 4
+#define __PRAM_FW_VERSION_RESERVED0 (0x3<<6)
+#define __PRAM_FW_VERSION_RESERVED0_SHIFT 6
+};
+
+
+/*
+ * a single rate shaping counter. can be used as protocol or vnic counter
+ */
+struct rate_shaping_counter {
+	u32 quota;
+#if defined(__BIG_ENDIAN)
+	u16 __reserved0;
+	u16 rate;
+#elif defined(__LITTLE_ENDIAN)
+	u16 rate;
+	u16 __reserved0;
+#endif
+};
+
+
+/*
+ * per-vnic rate shaping variables
+ */
+struct rate_shaping_vars_per_vn {
+	struct rate_shaping_counter protocol_counters[NUM_OF_PROTOCOLS];
+	struct rate_shaping_counter vn_counter;
 };
 
 
diff --git a/drivers/net/bnx2x_init.h b/drivers/net/bnx2x_init.h
index 5a4e82b9e7bf..4c7750789b62 100644
--- a/drivers/net/bnx2x_init.h
+++ b/drivers/net/bnx2x_init.h
@@ -226,28 +226,28 @@ static const u32 *bnx2x_sel_blob(u32 addr, const u32 *data, int is_e1)
 			       tsem_int_table_data_e1h;
 	else
 		IF_IS_INT_TABLE_ADDR(CSEM_REG_INT_TABLE, addr)
-		data = is_e1 ? csem_int_table_data_e1 :
-			       csem_int_table_data_e1h;
+			data = is_e1 ? csem_int_table_data_e1 :
+				       csem_int_table_data_e1h;
 	else
 		IF_IS_INT_TABLE_ADDR(USEM_REG_INT_TABLE, addr)
-		data = is_e1 ? usem_int_table_data_e1 :
-			       usem_int_table_data_e1h;
+			data = is_e1 ? usem_int_table_data_e1 :
+				       usem_int_table_data_e1h;
 	else
 		IF_IS_INT_TABLE_ADDR(XSEM_REG_INT_TABLE, addr)
-		data = is_e1 ? xsem_int_table_data_e1 :
-			       xsem_int_table_data_e1h;
+			data = is_e1 ? xsem_int_table_data_e1 :
+				       xsem_int_table_data_e1h;
 	else
 		IF_IS_PRAM_ADDR(TSEM_REG_PRAM, addr)
-		data = is_e1 ? tsem_pram_data_e1 : tsem_pram_data_e1h;
+			data = is_e1 ? tsem_pram_data_e1 : tsem_pram_data_e1h;
 	else
 		IF_IS_PRAM_ADDR(CSEM_REG_PRAM, addr)
-		data = is_e1 ? csem_pram_data_e1 : csem_pram_data_e1h;
+			data = is_e1 ? csem_pram_data_e1 : csem_pram_data_e1h;
 	else
 		IF_IS_PRAM_ADDR(USEM_REG_PRAM, addr)
-		data = is_e1 ? usem_pram_data_e1 : usem_pram_data_e1h;
+			data = is_e1 ? usem_pram_data_e1 : usem_pram_data_e1h;
 	else
 		IF_IS_PRAM_ADDR(XSEM_REG_PRAM, addr)
-		data = is_e1 ? xsem_pram_data_e1 : xsem_pram_data_e1h;
+			data = is_e1 ? xsem_pram_data_e1 : xsem_pram_data_e1h;
 
 	return data;
 }
diff --git a/drivers/net/bnx2x_link.c b/drivers/net/bnx2x_link.c
index 22586ebd7b1e..ff2743db10d9 100644
--- a/drivers/net/bnx2x_link.c
+++ b/drivers/net/bnx2x_link.c
@@ -3572,7 +3572,8 @@ u8 bnx2x_set_led(struct bnx2x *bp, u8 port, u8 mode, u32 speed,
 			   LED_BLINK_RATE_VAL);
 		REG_WR(bp, NIG_REG_LED_CONTROL_BLINK_RATE_ENA_P0 +
 			   port*4, 1);
-		if (((speed == SPEED_2500) ||
+		if (!CHIP_IS_E1H(bp) &&
+		    ((speed == SPEED_2500) ||
 		     (speed == SPEED_1000) ||
 		     (speed == SPEED_100) ||
 		     (speed == SPEED_10))) {
@@ -3753,6 +3754,14 @@ u8 bnx2x_phy_init(struct link_params *params, struct link_vars *vars)
 		vars->duplex = DUPLEX_FULL;
 		vars->flow_ctrl = FLOW_CTRL_NONE;
 		vars->link_status = (LINK_STATUS_LINK_UP | LINK_10GTFD);
+		/* enable on E1.5 FPGA */
+		if (CHIP_IS_E1H(bp)) {
+			vars->flow_ctrl |=
+				(FLOW_CTRL_TX | FLOW_CTRL_RX);
+			vars->link_status |=
+					(LINK_STATUS_TX_FLOW_CONTROL_ENABLED |
+					 LINK_STATUS_RX_FLOW_CONTROL_ENABLED);
+		}
 
 		bnx2x_emac_enable(params, vars, 0);
 		bnx2x_pbf_update(params, vars->flow_ctrl, vars->line_speed);
diff --git a/drivers/net/bnx2x_main.c b/drivers/net/bnx2x_main.c
index efa942688f84..90b54e4c5c3b 100644
--- a/drivers/net/bnx2x_main.c
+++ b/drivers/net/bnx2x_main.c
@@ -1,4 +1,4 @@
-/* bnx2x.c: Broadcom Everest network driver.
+/* bnx2x_main.c: Broadcom Everest network driver.
  *
  * Copyright (c) 2007-2008 Broadcom Corporation
  *
@@ -15,12 +15,6 @@
  *
  */
 
-/* define this to make the driver freeze on error
- * to allow getting debug info
- * (you will need to reboot afterwards)
- */
-/*#define BNX2X_STOP_ON_ERROR*/
-
 #include <linux/module.h>
 #include <linux/moduleparam.h>
 #include <linux/kernel.h>
@@ -46,16 +40,17 @@
 #include <linux/mii.h>
 #ifdef NETIF_F_HW_VLAN_TX
 	#include <linux/if_vlan.h>
-	#define BCM_VLAN 1
 #endif
 #include <net/ip.h>
 #include <net/tcp.h>
 #include <net/checksum.h>
+#include <linux/version.h>
+#include <net/ip6_checksum.h>
 #include <linux/workqueue.h>
 #include <linux/crc32.h>
+#include <linux/crc32c.h>
 #include <linux/prefetch.h>
 #include <linux/zlib.h>
-#include <linux/version.h>
 #include <linux/io.h>
 
 #include "bnx2x_reg.h"
@@ -67,13 +62,13 @@
 
 #define DRV_MODULE_VERSION      "1.42.4"
 #define DRV_MODULE_RELDATE      "2008/4/9"
-#define BNX2X_BC_VER    	0x040200
+#define BNX2X_BC_VER		0x040200
 
-/* Time in jiffies before concluding the transmitter is hung. */
-#define TX_TIMEOUT      	(5*HZ)
+/* Time in jiffies before concluding the transmitter is hung */
+#define TX_TIMEOUT		(5*HZ)
 
 static char version[] __devinitdata =
-	"Broadcom NetXtreme II 5771X 10Gigabit Ethernet Driver "
+	"Broadcom NetXtreme II 5771x 10Gigabit Ethernet Driver "
 	DRV_MODULE_NAME " " DRV_MODULE_VERSION " (" DRV_MODULE_RELDATE ")\n";
 
 MODULE_AUTHOR("Eliezer Tamir");
@@ -83,20 +78,19 @@ MODULE_VERSION(DRV_MODULE_VERSION);
 
 static int use_inta;
 static int poll;
-static int onefunc;
-static int nomcp;
 static int debug;
+static int nomcp;
+static int load_count[3]; /* 0-common, 1-port0, 2-port1 */
 static int use_multi;
 
 module_param(use_inta, int, 0);
 module_param(poll, int, 0);
-module_param(onefunc, int, 0);
 module_param(debug, int, 0);
+module_param(nomcp, int, 0);
 MODULE_PARM_DESC(use_inta, "use INT#A instead of MSI-X");
 MODULE_PARM_DESC(poll, "use polling (for debug)");
-MODULE_PARM_DESC(onefunc, "enable only first function");
-MODULE_PARM_DESC(nomcp, "ignore management CPU (Implies onefunc)");
 MODULE_PARM_DESC(debug, "default debug msglevel");
+MODULE_PARM_DESC(nomcp, "ignore management CPU");
 
 #ifdef BNX2X_MULTI
 module_param(use_multi, int, 0);
@@ -105,18 +99,27 @@ MODULE_PARM_DESC(use_multi, "use per-CPU queues");
 
 enum bnx2x_board_type {
 	BCM57710 = 0,
+	BCM57711 = 1,
+	BCM57711E = 2,
 };
 
-/* indexed by board_t, above */
+/* indexed by board_type, above */
 static struct {
 	char *name;
 } board_info[] __devinitdata = {
-	{ "Broadcom NetXtreme II BCM57710 XGb" }
+	{ "Broadcom NetXtreme II BCM57710 XGb" },
+	{ "Broadcom NetXtreme II BCM57711 XGb" },
+	{ "Broadcom NetXtreme II BCM57711E XGb" }
 };
 
+
 static const struct pci_device_id bnx2x_pci_tbl[] = {
 	{ PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_NX2_57710,
 		PCI_ANY_ID, PCI_ANY_ID, 0, 0, BCM57710 },
+	{ PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_NX2_57711,
+		PCI_ANY_ID, PCI_ANY_ID, 0, 0, BCM57711 },
+	{ PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_NX2_57711E,
+		PCI_ANY_ID, PCI_ANY_ID, 0, 0, BCM57711E },
 	{ 0 }
 };
 
@@ -201,7 +204,8 @@ void bnx2x_write_dmae(struct bnx2x *bp, dma_addr_t dma_addr, u32 dst_addr,
 #else
 			DMAE_CMD_ENDIANITY_DW_SWAP |
 #endif
-			(bp->port ? DMAE_CMD_PORT_1 : DMAE_CMD_PORT_0));
+			(BP_PORT(bp) ? DMAE_CMD_PORT_1 : DMAE_CMD_PORT_0) |
+			(BP_E1HVN(bp) << DMAE_CMD_E1HVN_SHIFT));
 	dmae->src_addr_lo = U64_LO(dma_addr);
 	dmae->src_addr_hi = U64_HI(dma_addr);
 	dmae->dst_addr_lo = dst_addr >> 2;
@@ -224,7 +228,7 @@ void bnx2x_write_dmae(struct bnx2x *bp, dma_addr_t dma_addr, u32 dst_addr,
 
 	*wb_comp = 0;
 
-	bnx2x_post_dmae(bp, dmae, (bp->port)*MAX_DMAE_C_PER_PORT);
+	bnx2x_post_dmae(bp, dmae, INIT_DMAE_C(bp));
 
 	udelay(5);
 
@@ -277,7 +281,8 @@ void bnx2x_read_dmae(struct bnx2x *bp, u32 src_addr, u32 len32)
 #else
 			DMAE_CMD_ENDIANITY_DW_SWAP |
 #endif
-			(bp->port ? DMAE_CMD_PORT_1 : DMAE_CMD_PORT_0));
+			(BP_PORT(bp) ? DMAE_CMD_PORT_1 : DMAE_CMD_PORT_0) |
+			(BP_E1HVN(bp) << DMAE_CMD_E1HVN_SHIFT));
 	dmae->src_addr_lo = src_addr >> 2;
 	dmae->src_addr_hi = 0;
 	dmae->dst_addr_lo = U64_LO(bnx2x_sp_mapping(bp, wb_data));
@@ -297,7 +302,7 @@ void bnx2x_read_dmae(struct bnx2x *bp, u32 src_addr, u32 len32)
 
 	*wb_comp = 0;
 
-	bnx2x_post_dmae(bp, dmae, (bp->port)*MAX_DMAE_C_PER_PORT);
+	bnx2x_post_dmae(bp, dmae, INIT_DMAE_C(bp));
 
 	udelay(5);
 
@@ -345,47 +350,122 @@ static u64 bnx2x_wb_rd(struct bnx2x *bp, int reg)
 
 static int bnx2x_mc_assert(struct bnx2x *bp)
 {
-	int i, j, rc = 0;
 	char last_idx;
-	const char storm[] = {"XTCU"};
-	const u32 intmem_base[] = {
-		BAR_XSTRORM_INTMEM,
-		BAR_TSTRORM_INTMEM,
-		BAR_CSTRORM_INTMEM,
-		BAR_USTRORM_INTMEM
-	};
-
-	/* Go through all instances of all SEMIs */
-	for (i = 0; i < 4; i++) {
-		last_idx = REG_RD8(bp, XSTORM_ASSERT_LIST_INDEX_OFFSET +
-				   intmem_base[i]);
-		if (last_idx)
-			BNX2X_LOG("DATA %cSTORM_ASSERT_LIST_INDEX 0x%x\n",
-				  storm[i], last_idx);
-
-		/* print the asserts */
-		for (j = 0; j < STROM_ASSERT_ARRAY_SIZE; j++) {
-			u32 row0, row1, row2, row3;
-
-			row0 = REG_RD(bp, XSTORM_ASSERT_LIST_OFFSET(j) +
-				      intmem_base[i]);
-			row1 = REG_RD(bp, XSTORM_ASSERT_LIST_OFFSET(j) + 4 +
-				      intmem_base[i]);
-			row2 = REG_RD(bp, XSTORM_ASSERT_LIST_OFFSET(j) + 8 +
-				      intmem_base[i]);
-			row3 = REG_RD(bp, XSTORM_ASSERT_LIST_OFFSET(j) + 12 +
-				      intmem_base[i]);
-
-			if (row0 != COMMON_ASM_INVALID_ASSERT_OPCODE) {
-				BNX2X_LOG("DATA %cSTORM_ASSERT_INDEX 0x%x ="
-					  " 0x%08x 0x%08x 0x%08x 0x%08x\n",
-					  storm[i], j, row3, row2, row1, row0);
-				rc++;
-			} else {
-				break;
-			}
+	int i, rc = 0;
+	u32 row0, row1, row2, row3;
+
+	/* XSTORM */
+	last_idx = REG_RD8(bp, BAR_XSTRORM_INTMEM +
+			   XSTORM_ASSERT_LIST_INDEX_OFFSET);
+	if (last_idx)
+		BNX2X_ERR("XSTORM_ASSERT_LIST_INDEX 0x%x\n", last_idx);
+
+	/* print the asserts */
+	for (i = 0; i < STROM_ASSERT_ARRAY_SIZE; i++) {
+
+		row0 = REG_RD(bp, BAR_XSTRORM_INTMEM +
+			      XSTORM_ASSERT_LIST_OFFSET(i));
+		row1 = REG_RD(bp, BAR_XSTRORM_INTMEM +
+			      XSTORM_ASSERT_LIST_OFFSET(i) + 4);
+		row2 = REG_RD(bp, BAR_XSTRORM_INTMEM +
+			      XSTORM_ASSERT_LIST_OFFSET(i) + 8);
+		row3 = REG_RD(bp, BAR_XSTRORM_INTMEM +
+			      XSTORM_ASSERT_LIST_OFFSET(i) + 12);
+
+		if (row0 != COMMON_ASM_INVALID_ASSERT_OPCODE) {
+			BNX2X_ERR("XSTORM_ASSERT_INDEX 0x%x = 0x%08x"
+				  " 0x%08x 0x%08x 0x%08x\n",
+				  i, row3, row2, row1, row0);
+			rc++;
+		} else {
+			break;
+		}
+	}
+
+	/* TSTORM */
+	last_idx = REG_RD8(bp, BAR_TSTRORM_INTMEM +
+			   TSTORM_ASSERT_LIST_INDEX_OFFSET);
+	if (last_idx)
+		BNX2X_ERR("TSTORM_ASSERT_LIST_INDEX 0x%x\n", last_idx);
+
+	/* print the asserts */
+	for (i = 0; i < STROM_ASSERT_ARRAY_SIZE; i++) {
+
+		row0 = REG_RD(bp, BAR_TSTRORM_INTMEM +
+			      TSTORM_ASSERT_LIST_OFFSET(i));
+		row1 = REG_RD(bp, BAR_TSTRORM_INTMEM +
+			      TSTORM_ASSERT_LIST_OFFSET(i) + 4);
+		row2 = REG_RD(bp, BAR_TSTRORM_INTMEM +
+			      TSTORM_ASSERT_LIST_OFFSET(i) + 8);
+		row3 = REG_RD(bp, BAR_TSTRORM_INTMEM +
+			      TSTORM_ASSERT_LIST_OFFSET(i) + 12);
+
+		if (row0 != COMMON_ASM_INVALID_ASSERT_OPCODE) {
+			BNX2X_ERR("TSTORM_ASSERT_INDEX 0x%x = 0x%08x"
+				  " 0x%08x 0x%08x 0x%08x\n",
+				  i, row3, row2, row1, row0);
+			rc++;
+		} else {
+			break;
+		}
+	}
+
+	/* CSTORM */
+	last_idx = REG_RD8(bp, BAR_CSTRORM_INTMEM +
+			   CSTORM_ASSERT_LIST_INDEX_OFFSET);
+	if (last_idx)
+		BNX2X_ERR("CSTORM_ASSERT_LIST_INDEX 0x%x\n", last_idx);
+
+	/* print the asserts */
+	for (i = 0; i < STROM_ASSERT_ARRAY_SIZE; i++) {
+
+		row0 = REG_RD(bp, BAR_CSTRORM_INTMEM +
+			      CSTORM_ASSERT_LIST_OFFSET(i));
+		row1 = REG_RD(bp, BAR_CSTRORM_INTMEM +
+			      CSTORM_ASSERT_LIST_OFFSET(i) + 4);
+		row2 = REG_RD(bp, BAR_CSTRORM_INTMEM +
+			      CSTORM_ASSERT_LIST_OFFSET(i) + 8);
+		row3 = REG_RD(bp, BAR_CSTRORM_INTMEM +
+			      CSTORM_ASSERT_LIST_OFFSET(i) + 12);
+
+		if (row0 != COMMON_ASM_INVALID_ASSERT_OPCODE) {
+			BNX2X_ERR("CSTORM_ASSERT_INDEX 0x%x = 0x%08x"
+				  " 0x%08x 0x%08x 0x%08x\n",
+				  i, row3, row2, row1, row0);
+			rc++;
+		} else {
+			break;
+		}
+	}
+
+	/* USTORM */
+	last_idx = REG_RD8(bp, BAR_USTRORM_INTMEM +
+			   USTORM_ASSERT_LIST_INDEX_OFFSET);
+	if (last_idx)
+		BNX2X_ERR("USTORM_ASSERT_LIST_INDEX 0x%x\n", last_idx);
+
+	/* print the asserts */
+	for (i = 0; i < STROM_ASSERT_ARRAY_SIZE; i++) {
+
+		row0 = REG_RD(bp, BAR_USTRORM_INTMEM +
+			      USTORM_ASSERT_LIST_OFFSET(i));
+		row1 = REG_RD(bp, BAR_USTRORM_INTMEM +
+			      USTORM_ASSERT_LIST_OFFSET(i) + 4);
+		row2 = REG_RD(bp, BAR_USTRORM_INTMEM +
+			      USTORM_ASSERT_LIST_OFFSET(i) + 8);
+		row3 = REG_RD(bp, BAR_USTRORM_INTMEM +
+			      USTORM_ASSERT_LIST_OFFSET(i) + 12);
+
+		if (row0 != COMMON_ASM_INVALID_ASSERT_OPCODE) {
+			BNX2X_ERR("USTORM_ASSERT_INDEX 0x%x = 0x%08x"
+				  " 0x%08x 0x%08x 0x%08x\n",
+				  i, row3, row2, row1, row0);
+			rc++;
+		} else {
+			break;
 		}
 	}
+
 	return rc;
 }
 
@@ -428,14 +508,16 @@ static void bnx2x_panic_dump(struct bnx2x *bp)
 		struct eth_tx_db_data *hw_prods = fp->hw_tx_prods;
 
 		BNX2X_ERR("queue[%d]: tx_pkt_prod(%x)  tx_pkt_cons(%x)"
-			  "  tx_bd_prod(%x)  tx_bd_cons(%x)  *tx_cons_sb(%x)"
-			  "  *rx_cons_sb(%x)  rx_comp_prod(%x)"
-			  "  rx_comp_cons(%x)  fp_c_idx(%x)  fp_u_idx(%x)"
-			  "  bd data(%x,%x)\n",
+			  "  tx_bd_prod(%x)  tx_bd_cons(%x)  *tx_cons_sb(%x)\n",
 			  i, fp->tx_pkt_prod, fp->tx_pkt_cons, fp->tx_bd_prod,
-			  fp->tx_bd_cons, *fp->tx_cons_sb, *fp->rx_cons_sb,
-			  fp->rx_comp_prod, fp->rx_comp_cons, fp->fp_c_idx,
-			  fp->fp_u_idx, hw_prods->packets_prod,
+			  fp->tx_bd_cons, le16_to_cpu(*fp->tx_cons_sb));
+		BNX2X_ERR("          rx_comp_prod(%x)  rx_comp_cons(%x)"
+			  "  *rx_cons_sb(%x)\n",
+			  fp->rx_comp_prod, fp->rx_comp_cons,
+			  le16_to_cpu(*fp->rx_cons_sb));
+		BNX2X_ERR("          fp_c_idx(%x)  fp_u_idx(%x)"
+			  "  bd data(%x,%x)\n",
+			  fp->fp_c_idx, fp->fp_u_idx, hw_prods->packets_prod,
 			  hw_prods->bds_prod);
 
 		start = TX_BD(le16_to_cpu(*fp->tx_cons_sb) - 10);
@@ -463,7 +545,7 @@ static void bnx2x_panic_dump(struct bnx2x *bp)
 			struct sw_rx_bd *sw_bd = &fp->rx_buf_ring[j];
 
 			BNX2X_ERR("rx_bd[%x]=[%x:%x]  sw_bd=[%p]\n",
-				  j, rx_bd[0], rx_bd[1], sw_bd->skb);
+				  j, rx_bd[1], rx_bd[0], sw_bd->skb);
 		}
 
 		start = RCQ_BD(fp->rx_comp_cons - 10);
@@ -482,7 +564,7 @@ static void bnx2x_panic_dump(struct bnx2x *bp)
 		  bp->def_c_idx, bp->def_u_idx, bp->def_x_idx, bp->def_t_idx,
 		  bp->def_att_idx, bp->attn_state, bp->spq_prod_idx);
 
-
+	bnx2x_fw_dump(bp);
 	bnx2x_mc_assert(bp);
 	BNX2X_ERR("end crash dump -----------------\n");
 
@@ -492,7 +574,7 @@ static void bnx2x_panic_dump(struct bnx2x *bp)
 
 static void bnx2x_int_enable(struct bnx2x *bp)
 {
-	int port = bp->port;
+	int port = BP_PORT(bp);
 	u32 addr = port ? HC_REG_CONFIG_1 : HC_REG_CONFIG_0;
 	u32 val = REG_RD(bp, addr);
 	int msix = (bp->flags & USING_MSIX_FLAG) ? 1 : 0;
@@ -507,7 +589,6 @@ static void bnx2x_int_enable(struct bnx2x *bp)
 			HC_CONFIG_0_REG_INT_LINE_EN_0 |
 			HC_CONFIG_0_REG_ATTN_BIT_EN_0);
 
-		/* Errata A0.158 workaround */
 		DP(NETIF_MSG_INTR, "write %x to HC %d (addr 0x%x)  MSI-X %d\n",
 		   val, port, addr, msix);
 
@@ -520,11 +601,25 @@ static void bnx2x_int_enable(struct bnx2x *bp)
 	   val, port, addr, msix);
 
 	REG_WR(bp, addr, val);
+
+	if (CHIP_IS_E1H(bp)) {
+		/* init leading/trailing edge */
+		if (IS_E1HMF(bp)) {
+			val = (0xfe0f | (1 << (BP_E1HVN(bp) + 4)));
+			if (bp->port.pmf)
+				/* enable nig attention */
+				val |= 0x0100;
+		} else
+			val = 0xffff;
+
+		REG_WR(bp, HC_REG_TRAILING_EDGE_0 + port*8, val);
+		REG_WR(bp, HC_REG_LEADING_EDGE_0 + port*8, val);
+	}
 }
 
 static void bnx2x_int_disable(struct bnx2x *bp)
 {
-	int port = bp->port;
+	int port = BP_PORT(bp);
 	u32 addr = port ? HC_REG_CONFIG_1 : HC_REG_CONFIG_0;
 	u32 val = REG_RD(bp, addr);
 
@@ -543,10 +638,10 @@ static void bnx2x_int_disable(struct bnx2x *bp)
 
 static void bnx2x_int_disable_sync(struct bnx2x *bp)
 {
-
 	int msix = (bp->flags & USING_MSIX_FLAG) ? 1 : 0;
 	int i;
 
+	/* disable interrupt handling */
 	atomic_inc(&bp->intr_sem);
 	/* prevent the HW from sending interrupts */
 	bnx2x_int_disable(bp);
@@ -563,30 +658,29 @@ static void bnx2x_int_disable_sync(struct bnx2x *bp)
 
 	/* make sure sp_task is not running */
 	cancel_work_sync(&bp->sp_task);
-
 }
 
-/* fast path code */
+/* fast path */
 
 /*
- * general service functions
+ * General service functions
  */
 
-static inline void bnx2x_ack_sb(struct bnx2x *bp, u8 id,
+static inline void bnx2x_ack_sb(struct bnx2x *bp, u8 sb_id,
 				u8 storm, u16 index, u8 op, u8 update)
 {
-	u32 igu_addr = (IGU_ADDR_INT_ACK + IGU_PORT_BASE * bp->port) * 8;
+	u32 igu_addr = (IGU_ADDR_INT_ACK + IGU_FUNC_BASE * BP_FUNC(bp)) * 8;
 	struct igu_ack_register igu_ack;
 
 	igu_ack.status_block_index = index;
 	igu_ack.sb_id_and_flags =
-			((id << IGU_ACK_REGISTER_STATUS_BLOCK_ID_SHIFT) |
+			((sb_id << IGU_ACK_REGISTER_STATUS_BLOCK_ID_SHIFT) |
 			 (storm << IGU_ACK_REGISTER_STORM_ID_SHIFT) |
 			 (update << IGU_ACK_REGISTER_UPDATE_INDEX_SHIFT) |
 			 (op << IGU_ACK_REGISTER_INTERRUPT_MODE_SHIFT));
 
-/*      DP(NETIF_MSG_INTR, "write 0x%08x to IGU addr 0x%x\n",
-	   (*(u32 *)&igu_ack), BAR_IGU_INTMEM + igu_addr); */
+	DP(BNX2X_MSG_OFF, "write 0x%08x to IGU addr 0x%x\n",
+	   (*(u32 *)&igu_ack), BAR_IGU_INTMEM + igu_addr);
 	REG_WR(bp, BAR_IGU_INTMEM + igu_addr, (*(u32 *)&igu_ack));
 }
 
@@ -614,8 +708,9 @@ static inline int bnx2x_has_work(struct bnx2x_fastpath *fp)
 	if ((rx_cons_sb & MAX_RCQ_DESC_CNT) == MAX_RCQ_DESC_CNT)
 		rx_cons_sb++;
 
-	if ((rx_cons_sb != fp->rx_comp_cons) ||
-	    (le16_to_cpu(*fp->tx_cons_sb) != fp->tx_pkt_cons))
+	if ((fp->rx_comp_cons != rx_cons_sb) ||
+	    (fp->tx_pkt_prod != le16_to_cpu(*fp->tx_cons_sb)) ||
+	    (fp->tx_pkt_prod != fp->tx_pkt_cons))
 		return 1;
 
 	return 0;
@@ -623,11 +718,11 @@ static inline int bnx2x_has_work(struct bnx2x_fastpath *fp)
 
 static u16 bnx2x_ack_int(struct bnx2x *bp)
 {
-	u32 igu_addr = (IGU_ADDR_SIMD_MASK + IGU_PORT_BASE * bp->port) * 8;
+	u32 igu_addr = (IGU_ADDR_SIMD_MASK + IGU_FUNC_BASE * BP_FUNC(bp)) * 8;
 	u32 result = REG_RD(bp, BAR_IGU_INTMEM + igu_addr);
 
-/*      DP(NETIF_MSG_INTR, "read 0x%08x from IGU addr 0x%x\n",
-	   result, BAR_IGU_INTMEM + igu_addr); */
+	DP(BNX2X_MSG_OFF, "read 0x%08x from IGU addr 0x%x\n",
+	   result, BAR_IGU_INTMEM + igu_addr);
 
 #ifdef IGU_DEBUG
 #warning IGU_DEBUG active
@@ -653,7 +748,7 @@ static u16 bnx2x_free_tx_pkt(struct bnx2x *bp, struct bnx2x_fastpath *fp,
 	struct sw_tx_bd *tx_buf = &fp->tx_buf_ring[idx];
 	struct eth_tx_bd *tx_bd;
 	struct sk_buff *skb = tx_buf->skb;
-	u16 bd_idx = tx_buf->first_bd;
+	u16 bd_idx = TX_BD(tx_buf->first_bd), new_cons;
 	int nbd;
 
 	DP(BNX2X_MSG_OFF, "pkt_idx %d  buff @(%p)->skb %p\n",
@@ -666,9 +761,10 @@ static u16 bnx2x_free_tx_pkt(struct bnx2x *bp, struct bnx2x_fastpath *fp,
 			 BD_UNMAP_LEN(tx_bd), PCI_DMA_TODEVICE);
 
 	nbd = le16_to_cpu(tx_bd->nbd) - 1;
+	new_cons = nbd + tx_buf->first_bd;
 #ifdef BNX2X_STOP_ON_ERROR
 	if (nbd > (MAX_SKB_FRAGS + 2)) {
-		BNX2X_ERR("bad nbd!\n");
+		BNX2X_ERR("BAD nbd!\n");
 		bnx2x_panic();
 	}
 #endif
@@ -708,32 +804,30 @@ static u16 bnx2x_free_tx_pkt(struct bnx2x *bp, struct bnx2x_fastpath *fp,
 	tx_buf->first_bd = 0;
 	tx_buf->skb = NULL;
 
-	return bd_idx;
+	return new_cons;
 }
 
-static inline u32 bnx2x_tx_avail(struct bnx2x_fastpath *fp)
+static inline u16 bnx2x_tx_avail(struct bnx2x_fastpath *fp)
 {
-	u16 used;
-	u32 prod;
-	u32 cons;
+	s16 used;
+	u16 prod;
+	u16 cons;
 
-	/* Tell compiler that prod and cons can change */
-	barrier();
+	barrier(); /* Tell compiler that prod and cons can change */
 	prod = fp->tx_bd_prod;
 	cons = fp->tx_bd_cons;
 
-	used = (NUM_TX_BD - NUM_TX_RINGS + prod - cons +
-		(cons / TX_DESC_CNT) - (prod / TX_DESC_CNT));
-
-	if (prod >= cons) {
-		/* used = prod - cons - prod/size + cons/size */
-		used -= NUM_TX_BD - NUM_TX_RINGS;
-	}
+	/* NUM_TX_RINGS = number of "next-page" entries
+	   It will be used as a threshold */
+	used = SUB_S16(prod, cons) + (s16)NUM_TX_RINGS;
 
+#ifdef BNX2X_STOP_ON_ERROR
+	BUG_TRAP(used >= 0);
 	BUG_TRAP(used <= fp->bp->tx_ring_size);
 	BUG_TRAP((fp->bp->tx_ring_size - used) <= MAX_TX_AVAIL);
+#endif
 
-	return (fp->bp->tx_ring_size - used);
+	return (s16)(fp->bp->tx_ring_size) - used;
 }
 
 static void bnx2x_tx_int(struct bnx2x_fastpath *fp, int work)
@@ -757,10 +851,10 @@ static void bnx2x_tx_int(struct bnx2x_fastpath *fp, int work)
 
 		/* prefetch(bp->tx_buf_ring[pkt_cons].skb); */
 
-		DP(NETIF_MSG_TX_DONE, "hw_cons %u  sw_cons %u  pkt_cons %d\n",
+		DP(NETIF_MSG_TX_DONE, "hw_cons %u  sw_cons %u  pkt_cons %u\n",
 		   hw_cons, sw_cons, pkt_cons);
 
-/*      	if (NEXT_TX_IDX(sw_cons) != hw_cons) {
+/*		if (NEXT_TX_IDX(sw_cons) != hw_cons) {
 			rmb();
 			prefetch(fp->tx_buf_ring[NEXT_TX_IDX(sw_cons)].skb);
 		}
@@ -793,7 +887,6 @@ static void bnx2x_tx_int(struct bnx2x_fastpath *fp, int work)
 			netif_wake_queue(bp->dev);
 
 		netif_tx_unlock(bp->dev);
-
 	}
 }
 
@@ -804,13 +897,14 @@ static void bnx2x_sp_event(struct bnx2x_fastpath *fp,
 	int cid = SW_CID(rr_cqe->ramrod_cqe.conn_and_cmd_data);
 	int command = CQE_CMD(rr_cqe->ramrod_cqe.conn_and_cmd_data);
 
-	DP(NETIF_MSG_RX_STATUS,
+	DP(BNX2X_MSG_SP,
 	   "fp %d  cid %d  got ramrod #%d  state is %x  type is %d\n",
-	   fp->index, cid, command, bp->state, rr_cqe->ramrod_cqe.type);
+	   FP_IDX(fp), cid, command, bp->state,
+	   rr_cqe->ramrod_cqe.ramrod_type);
 
 	bp->spq_left++;
 
-	if (fp->index) {
+	if (FP_IDX(fp)) {
 		switch (command | fp->state) {
 		case (RAMROD_CMD_ID_ETH_CLIENT_SETUP |
 						BNX2X_FP_STATE_OPENING):
@@ -826,10 +920,11 @@ static void bnx2x_sp_event(struct bnx2x_fastpath *fp,
 			break;
 
 		default:
-			BNX2X_ERR("unexpected MC reply(%d)  state is %x\n",
-				  command, fp->state);
+			BNX2X_ERR("unexpected MC reply (%d)  "
+				  "fp->state is %x\n", command, fp->state);
+			break;
 		}
-		mb(); /* force bnx2x_wait_ramrod to see the change */
+		mb(); /* force bnx2x_wait_ramrod() to see the change */
 		return;
 	}
 
@@ -846,25 +941,25 @@ static void bnx2x_sp_event(struct bnx2x_fastpath *fp,
 		break;
 
 	case (RAMROD_CMD_ID_ETH_CFC_DEL | BNX2X_STATE_CLOSING_WAIT4_HALT):
-		DP(NETIF_MSG_IFDOWN, "got delete ramrod for MULTI[%d]\n",
-		   cid);
+		DP(NETIF_MSG_IFDOWN, "got delete ramrod for MULTI[%d]\n", cid);
 		bnx2x_fp(bp, cid, state) = BNX2X_FP_STATE_CLOSED;
 		break;
 
 	case (RAMROD_CMD_ID_ETH_SET_MAC | BNX2X_STATE_OPEN):
+	case (RAMROD_CMD_ID_ETH_SET_MAC | BNX2X_STATE_DIAG):
 		DP(NETIF_MSG_IFUP, "got set mac ramrod\n");
 		break;
 
 	case (RAMROD_CMD_ID_ETH_SET_MAC | BNX2X_STATE_CLOSING_WAIT4_HALT):
-		DP(NETIF_MSG_IFUP, "got (un)set mac ramrod\n");
+		DP(NETIF_MSG_IFDOWN, "got (un)set mac ramrod\n");
 		break;
 
 	default:
-		BNX2X_ERR("unexpected ramrod (%d)  state is %x\n",
+		BNX2X_ERR("unexpected MC reply (%d)  bp->state is %x\n",
 			  command, bp->state);
+		break;
 	}
-
-	mb(); /* force bnx2x_wait_ramrod to see the change */
+	mb(); /* force bnx2x_wait_ramrod() to see the change */
 }
 
 static inline int bnx2x_alloc_rx_skb(struct bnx2x *bp,
@@ -882,7 +977,6 @@ static inline int bnx2x_alloc_rx_skb(struct bnx2x *bp,
 	mapping = pci_map_single(bp->pdev, skb->data, bp->rx_buf_use_size,
 				 PCI_DMA_FROMDEVICE);
 	if (unlikely(dma_mapping_error(mapping))) {
-
 		dev_kfree_skb(skb);
 		return -ENOMEM;
 	}
@@ -924,7 +1018,7 @@ static void bnx2x_reuse_rx_skb(struct bnx2x_fastpath *fp,
 static int bnx2x_rx_int(struct bnx2x_fastpath *fp, int budget)
 {
 	struct bnx2x *bp = fp->bp;
-	u16 bd_cons, bd_prod, comp_ring_cons;
+	u16 bd_cons, bd_prod, bd_prod_fw, comp_ring_cons;
 	u16 hw_comp_cons, sw_comp_cons, sw_comp_prod;
 	int rx_pkt = 0;
 
@@ -933,12 +1027,15 @@ static int bnx2x_rx_int(struct bnx2x_fastpath *fp, int budget)
 		return 0;
 #endif
 
+	/* CQ "next element" is of the size of the regular element,
+	   that's why it's ok here */
 	hw_comp_cons = le16_to_cpu(*fp->rx_cons_sb);
 	if ((hw_comp_cons & MAX_RCQ_DESC_CNT) == MAX_RCQ_DESC_CNT)
 		hw_comp_cons++;
 
 	bd_cons = fp->rx_bd_cons;
 	bd_prod = fp->rx_bd_prod;
+	bd_prod_fw = bd_prod;
 	sw_comp_cons = fp->rx_comp_cons;
 	sw_comp_prod = fp->rx_comp_prod;
 
@@ -949,34 +1046,31 @@ static int bnx2x_rx_int(struct bnx2x_fastpath *fp, int budget)
 
 	DP(NETIF_MSG_RX_STATUS,
 	   "queue[%d]:  hw_comp_cons %u  sw_comp_cons %u\n",
-	   fp->index, hw_comp_cons, sw_comp_cons);
+	   FP_IDX(fp), hw_comp_cons, sw_comp_cons);
 
 	while (sw_comp_cons != hw_comp_cons) {
-		unsigned int len, pad;
-		struct sw_rx_bd *rx_buf;
+		struct sw_rx_bd *rx_buf = NULL;
 		struct sk_buff *skb;
 		union eth_rx_cqe *cqe;
+		u8 cqe_fp_flags;
+		u16 len, pad;
 
 		comp_ring_cons = RCQ_BD(sw_comp_cons);
 		bd_prod = RX_BD(bd_prod);
 		bd_cons = RX_BD(bd_cons);
 
 		cqe = &fp->rx_comp_ring[comp_ring_cons];
+		cqe_fp_flags = cqe->fast_path_cqe.type_error_flags;
 
-		DP(NETIF_MSG_RX_STATUS, "hw_comp_cons %u  sw_comp_cons %u"
-		   "  comp_ring (%u)  bd_ring (%u,%u)\n",
-		   hw_comp_cons, sw_comp_cons,
-		   comp_ring_cons, bd_prod, bd_cons);
 		DP(NETIF_MSG_RX_STATUS, "CQE type %x  err %x  status %x"
-		   "  queue %x  vlan %x  len %x\n",
-		   cqe->fast_path_cqe.type,
-		   cqe->fast_path_cqe.error_type_flags,
-		   cqe->fast_path_cqe.status_flags,
+		   "  queue %x  vlan %x  len %u\n", CQE_TYPE(cqe_fp_flags),
+		   cqe_fp_flags, cqe->fast_path_cqe.status_flags,
 		   cqe->fast_path_cqe.rss_hash_result,
-		   cqe->fast_path_cqe.vlan_tag, cqe->fast_path_cqe.pkt_len);
+		   le16_to_cpu(cqe->fast_path_cqe.vlan_tag),
+		   le16_to_cpu(cqe->fast_path_cqe.pkt_len));
 
 		/* is this a slowpath msg? */
-		if (unlikely(cqe->fast_path_cqe.type)) {
+		if (unlikely(CQE_TYPE(cqe_fp_flags))) {
 			bnx2x_sp_event(fp, cqe);
 			goto next_cqe;
 
@@ -984,7 +1078,6 @@ static int bnx2x_rx_int(struct bnx2x_fastpath *fp, int budget)
 		} else {
 			rx_buf = &fp->rx_buf_ring[bd_cons];
 			skb = rx_buf->skb;
-
 			len = le16_to_cpu(cqe->fast_path_cqe.pkt_len);
 			pad = cqe->fast_path_cqe.placement_offset;
 
@@ -996,13 +1089,11 @@ static int bnx2x_rx_int(struct bnx2x_fastpath *fp, int budget)
 			prefetch(((char *)(skb)) + 128);
 
 			/* is this an error packet? */
-			if (unlikely(cqe->fast_path_cqe.error_type_flags &
-							ETH_RX_ERROR_FALGS)) {
+			if (unlikely(cqe_fp_flags & ETH_RX_ERROR_FALGS)) {
 			/* do we sometimes forward error packets anyway? */
 				DP(NETIF_MSG_RX_ERR,
-				   "ERROR flags(%u) Rx packet(%u)\n",
-				   cqe->fast_path_cqe.error_type_flags,
-				   sw_comp_cons);
+				   "ERROR  flags %x  rx packet %u\n",
+				   cqe_fp_flags, sw_comp_cons);
 				/* TBD make sure MC counts this as a drop */
 				goto reuse_rx;
 			}
@@ -1018,7 +1109,7 @@ static int bnx2x_rx_int(struct bnx2x_fastpath *fp, int budget)
 							   len + pad);
 				if (new_skb == NULL) {
 					DP(NETIF_MSG_RX_ERR,
-					   "ERROR packet dropped "
+					   "ERROR  packet dropped "
 					   "because of alloc failure\n");
 					/* TBD count this as a drop? */
 					goto reuse_rx;
@@ -1044,7 +1135,7 @@ static int bnx2x_rx_int(struct bnx2x_fastpath *fp, int budget)
 
 			} else {
 				DP(NETIF_MSG_RX_ERR,
-				   "ERROR packet dropped because "
+				   "ERROR  packet dropped because "
 				   "of alloc failure\n");
 reuse_rx:
 				bnx2x_reuse_rx_skb(fp, skb, bd_cons, bd_prod);
@@ -1061,14 +1152,14 @@ reuse_rx:
 		}
 
 #ifdef BCM_VLAN
-		if ((le16_to_cpu(cqe->fast_path_cqe.pars_flags.flags)
-				& PARSING_FLAGS_NUMBER_OF_NESTED_VLANS)
-		    && (bp->vlgrp != NULL))
+		if ((bp->vlgrp != NULL) &&
+		    (le16_to_cpu(cqe->fast_path_cqe.pars_flags.flags) &
+		     PARSING_FLAGS_VLAN))
 			vlan_hwaccel_receive_skb(skb, bp->vlgrp,
 				le16_to_cpu(cqe->fast_path_cqe.vlan_tag));
 		else
 #endif
-		netif_receive_skb(skb);
+			netif_receive_skb(skb);
 
 		bp->dev->last_rx = jiffies;
 
@@ -1077,22 +1168,25 @@ next_rx:
 
 		bd_cons = NEXT_RX_IDX(bd_cons);
 		bd_prod = NEXT_RX_IDX(bd_prod);
+		bd_prod_fw = NEXT_RX_IDX(bd_prod_fw);
+		rx_pkt++;
 next_cqe:
 		sw_comp_prod = NEXT_RCQ_IDX(sw_comp_prod);
 		sw_comp_cons = NEXT_RCQ_IDX(sw_comp_cons);
-		rx_pkt++;
 
-		if ((rx_pkt == budget))
+		if (rx_pkt == budget)
 			break;
 	} /* while */
 
 	fp->rx_bd_cons = bd_cons;
-	fp->rx_bd_prod = bd_prod;
+	fp->rx_bd_prod = bd_prod_fw;
 	fp->rx_comp_cons = sw_comp_cons;
 	fp->rx_comp_prod = sw_comp_prod;
 
 	REG_WR(bp, BAR_TSTRORM_INTMEM +
-	       TSTORM_RCQ_PROD_OFFSET(bp->port, fp->index), sw_comp_prod);
+		TSTORM_RX_PRODS_OFFSET(BP_PORT(bp), FP_CL_ID(fp)),
+		sw_comp_prod);
+
 
 	mmiowb(); /* keep prod updates ordered */
 
@@ -1107,10 +1201,11 @@ static irqreturn_t bnx2x_msix_fp_int(int irq, void *fp_cookie)
 	struct bnx2x_fastpath *fp = fp_cookie;
 	struct bnx2x *bp = fp->bp;
 	struct net_device *dev = bp->dev;
-	int index = fp->index;
+	int index = FP_IDX(fp);
 
-	DP(NETIF_MSG_INTR, "got an msix interrupt on [%d]\n", index);
-	bnx2x_ack_sb(bp, index, USTORM_ID, 0, IGU_INT_DISABLE, 0);
+	DP(BNX2X_MSG_FP, "got an MSI-X interrupt on IDX:SB [%d:%d]\n",
+	   index, FP_SB_ID(fp));
+	bnx2x_ack_sb(bp, FP_SB_ID(fp), USTORM_ID, 0, IGU_INT_DISABLE, 0);
 
 #ifdef BNX2X_STOP_ON_ERROR
 	if (unlikely(bp->panic))
@@ -1123,6 +1218,7 @@ static irqreturn_t bnx2x_msix_fp_int(int irq, void *fp_cookie)
 	prefetch(&fp->status_blk->u_status_block.status_block_index);
 
 	netif_rx_schedule(dev, &bnx2x_fp(bp, index, napi));
+
 	return IRQ_HANDLED;
 }
 
@@ -1131,26 +1227,28 @@ static irqreturn_t bnx2x_interrupt(int irq, void *dev_instance)
 	struct net_device *dev = dev_instance;
 	struct bnx2x *bp = netdev_priv(dev);
 	u16 status = bnx2x_ack_int(bp);
+	u16 mask;
 
+	/* Return here if interrupt is shared and it's not for us */
 	if (unlikely(status == 0)) {
 		DP(NETIF_MSG_INTR, "not our interrupt!\n");
 		return IRQ_NONE;
 	}
-
-	DP(NETIF_MSG_INTR, "got an interrupt status is %u\n", status);
+	DP(NETIF_MSG_INTR, "got an interrupt  status %u\n", status);
 
 #ifdef BNX2X_STOP_ON_ERROR
 	if (unlikely(bp->panic))
 		return IRQ_HANDLED;
 #endif
 
-	/* Return here if interrupt is shared and is disabled */
+	/* Return here if interrupt is disabled */
 	if (unlikely(atomic_read(&bp->intr_sem) != 0)) {
 		DP(NETIF_MSG_INTR, "called but intr_sem not 0, returning\n");
 		return IRQ_HANDLED;
 	}
 
-	if (status & 0x2) {
+	mask = 0x2 << bp->fp[0].sb_id;
+	if (status & mask) {
 		struct bnx2x_fastpath *fp = &bp->fp[0];
 
 		prefetch(fp->rx_cons_sb);
@@ -1160,13 +1258,11 @@ static irqreturn_t bnx2x_interrupt(int irq, void *dev_instance)
 
 		netif_rx_schedule(dev, &bnx2x_fp(bp, 0, napi));
 
-		status &= ~0x2;
-		if (!status)
-			return IRQ_HANDLED;
+		status &= ~mask;
 	}
 
-	if (unlikely(status & 0x1)) {
 
+	if (unlikely(status & 0x1)) {
 		schedule_work(&bp->sp_task);
 
 		status &= ~0x1;
@@ -1174,8 +1270,9 @@ static irqreturn_t bnx2x_interrupt(int irq, void *dev_instance)
 			return IRQ_HANDLED;
 	}
 
-	DP(NETIF_MSG_INTR, "got an unknown interrupt! (status is %u)\n",
-	   status);
+	if (status)
+		DP(NETIF_MSG_INTR, "got an unknown interrupt! (status %u)\n",
+		   status);
 
 	return IRQ_HANDLED;
 }
@@ -1193,7 +1290,7 @@ static int bnx2x_hw_lock(struct bnx2x *bp, u32 resource)
 {
 	u32 lock_status;
 	u32 resource_bit = (1 << resource);
-	u8 port = bp->port;
+	u8 port = BP_PORT(bp);
 	int cnt;
 
 	/* Validating that the resource is within range */
@@ -1231,7 +1328,7 @@ static int bnx2x_hw_unlock(struct bnx2x *bp, u32 resource)
 {
 	u32 lock_status;
 	u32 resource_bit = (1 << resource);
-	u8 port = bp->port;
+	u8 port = BP_PORT(bp);
 
 	/* Validating that the resource is within range */
 	if (resource > HW_LOCK_MAX_RESOURCE_VALUE) {
@@ -1258,7 +1355,7 @@ static void bnx2x_phy_hw_lock(struct bnx2x *bp)
 {
 	u32 ext_phy_type = XGXS_EXT_PHY_TYPE(bp->link_params.ext_phy_config);
 
-	mutex_lock(&bp->phy_mutex);
+	mutex_lock(&bp->port.phy_mutex);
 
 	if ((ext_phy_type == PORT_HW_CFG_XGXS_EXT_PHY_TYPE_BCM8072) ||
 	    (ext_phy_type == PORT_HW_CFG_XGXS_EXT_PHY_TYPE_BCM8073))
@@ -1273,14 +1370,14 @@ static void bnx2x_phy_hw_unlock(struct bnx2x *bp)
 	    (ext_phy_type == PORT_HW_CFG_XGXS_EXT_PHY_TYPE_BCM8073))
 		bnx2x_hw_unlock(bp, HW_LOCK_RESOURCE_8072_MDIO);
 
-	mutex_unlock(&bp->phy_mutex);
+	mutex_unlock(&bp->port.phy_mutex);
 }
 
 int bnx2x_set_gpio(struct bnx2x *bp, int gpio_num, u32 mode)
 {
 	/* The GPIO should be swapped if swap register is set and active */
 	int gpio_port = (REG_RD(bp, NIG_REG_PORT_SWAP) &&
-			 REG_RD(bp, NIG_REG_STRAP_OVERRIDE)) ^ bp->port;
+			 REG_RD(bp, NIG_REG_STRAP_OVERRIDE)) ^ BP_PORT(bp);
 	int gpio_shift = gpio_num +
 			(gpio_port ? MISC_REGISTERS_GPIO_PORT_SHIFT : 0);
 	u32 gpio_mask = (1 << gpio_shift);
@@ -1379,18 +1476,18 @@ static void bnx2x_calc_fc_adv(struct bnx2x *bp)
 {
 	switch (bp->link_vars.ieee_fc) {
 	case MDIO_COMBO_IEEE0_AUTO_NEG_ADV_PAUSE_NONE:
-		bp->advertising &= ~(ADVERTISED_Asym_Pause |
+		bp->port.advertising &= ~(ADVERTISED_Asym_Pause |
 					  ADVERTISED_Pause);
 		break;
 	case MDIO_COMBO_IEEE0_AUTO_NEG_ADV_PAUSE_BOTH:
-		bp->advertising |= (ADVERTISED_Asym_Pause |
+		bp->port.advertising |= (ADVERTISED_Asym_Pause |
 					 ADVERTISED_Pause);
 		break;
 	case MDIO_COMBO_IEEE0_AUTO_NEG_ADV_PAUSE_ASYMMETRIC:
-		bp->advertising |= ADVERTISED_Asym_Pause;
+		bp->port.advertising |= ADVERTISED_Asym_Pause;
 		break;
 	default:
-		bp->advertising &= ~(ADVERTISED_Asym_Pause |
+		bp->port.advertising &= ~(ADVERTISED_Asym_Pause |
 					  ADVERTISED_Pause);
 		break;
 	}
@@ -1443,6 +1540,7 @@ static u8 bnx2x_initial_phy_init(struct bnx2x *bp)
 		bnx2x_link_report(bp);
 
 	bnx2x_calc_fc_adv(bp);
+
 	return rc;
 }
 
@@ -1473,15 +1571,261 @@ static u8 bnx2x_link_test(struct bnx2x *bp)
 	return rc;
 }
 
+/* Calculates the sum of vn_min_rates.
+   It's needed for further normalizing of the min_rates.
+
+   Returns:
+     sum of vn_min_rates
+       or
+     0 - if all the min_rates are 0.
+     In the later case fainess algorithm should be deactivated.
+     If not all min_rates are zero then those that are zeroes will
+     be set to 1.
+ */
+static u32 bnx2x_calc_vn_wsum(struct bnx2x *bp)
+{
+	int i, port = BP_PORT(bp);
+	u32 wsum = 0;
+	int all_zero = 1;
+
+	for (i = 0; i < E1HVN_MAX; i++) {
+		u32 vn_cfg =
+			SHMEM_RD(bp, mf_cfg.func_mf_config[2*i + port].config);
+		u32 vn_min_rate = ((vn_cfg & FUNC_MF_CFG_MIN_BW_MASK) >>
+				     FUNC_MF_CFG_MIN_BW_SHIFT) * 100;
+		if (!(vn_cfg & FUNC_MF_CFG_FUNC_HIDE)) {
+			/* If min rate is zero - set it to 1 */
+			if (!vn_min_rate)
+				vn_min_rate = DEF_MIN_RATE;
+			else
+				all_zero = 0;
+
+			wsum += vn_min_rate;
+		}
+	}
+
+	/* ... only if all min rates are zeros - disable FAIRNESS */
+	if (all_zero)
+		return 0;
+
+	return wsum;
+}
+
+static void bnx2x_init_port_minmax(struct bnx2x *bp,
+				   int en_fness,
+				   u16 port_rate,
+				   struct cmng_struct_per_port *m_cmng_port)
+{
+	u32 r_param = port_rate / 8;
+	int port = BP_PORT(bp);
+	int i;
+
+	memset(m_cmng_port, 0, sizeof(struct cmng_struct_per_port));
+
+	/* Enable minmax only if we are in e1hmf mode */
+	if (IS_E1HMF(bp)) {
+		u32 fair_periodic_timeout_usec;
+		u32 t_fair;
+
+		/* Enable rate shaping and fairness */
+		m_cmng_port->flags.cmng_vn_enable = 1;
+		m_cmng_port->flags.fairness_enable = en_fness ? 1 : 0;
+		m_cmng_port->flags.rate_shaping_enable = 1;
+
+		if (!en_fness)
+			DP(NETIF_MSG_IFUP, "All MIN values are zeroes"
+			   "  fairness will be disabled\n");
+
+		/* 100 usec in SDM ticks = 25 since each tick is 4 usec */
+		m_cmng_port->rs_vars.rs_periodic_timeout =
+						RS_PERIODIC_TIMEOUT_USEC / 4;
+
+		/* this is the threshold below which no timer arming will occur
+		   1.25 coefficient is for the threshold to be a little bigger
+		   than the real time, to compensate for timer in-accuracy */
+		m_cmng_port->rs_vars.rs_threshold =
+				(RS_PERIODIC_TIMEOUT_USEC * r_param * 5) / 4;
+
+		/* resolution of fairness timer */
+		fair_periodic_timeout_usec = QM_ARB_BYTES / r_param;
+		/* for 10G it is 1000usec. for 1G it is 10000usec. */
+		t_fair = T_FAIR_COEF / port_rate;
+
+		/* this is the threshold below which we won't arm
+		   the timer anymore */
+		m_cmng_port->fair_vars.fair_threshold = QM_ARB_BYTES;
+
+		/* we multiply by 1e3/8 to get bytes/msec.
+		   We don't want the credits to pass a credit
+		   of the T_FAIR*FAIR_MEM (algorithm resolution) */
+		m_cmng_port->fair_vars.upper_bound =
+						r_param * t_fair * FAIR_MEM;
+		/* since each tick is 4 usec */
+		m_cmng_port->fair_vars.fairness_timeout =
+						fair_periodic_timeout_usec / 4;
+
+	} else {
+		/* Disable rate shaping and fairness */
+		m_cmng_port->flags.cmng_vn_enable = 0;
+		m_cmng_port->flags.fairness_enable = 0;
+		m_cmng_port->flags.rate_shaping_enable = 0;
+
+		DP(NETIF_MSG_IFUP,
+		   "Single function mode  minmax will be disabled\n");
+	}
+
+	/* Store it to internal memory */
+	for (i = 0; i < sizeof(struct cmng_struct_per_port) / 4; i++)
+		REG_WR(bp, BAR_XSTRORM_INTMEM +
+		       XSTORM_CMNG_PER_PORT_VARS_OFFSET(port) + i * 4,
+		       ((u32 *)(m_cmng_port))[i]);
+}
+
+static void bnx2x_init_vn_minmax(struct bnx2x *bp, int func,
+				   u32 wsum, u16 port_rate,
+				 struct cmng_struct_per_port *m_cmng_port)
+{
+	struct rate_shaping_vars_per_vn m_rs_vn;
+	struct fairness_vars_per_vn m_fair_vn;
+	u32 vn_cfg = SHMEM_RD(bp, mf_cfg.func_mf_config[func].config);
+	u16 vn_min_rate, vn_max_rate;
+	int i;
+
+	/* If function is hidden - set min and max to zeroes */
+	if (vn_cfg & FUNC_MF_CFG_FUNC_HIDE) {
+		vn_min_rate = 0;
+		vn_max_rate = 0;
+
+	} else {
+		vn_min_rate = ((vn_cfg & FUNC_MF_CFG_MIN_BW_MASK) >>
+				FUNC_MF_CFG_MIN_BW_SHIFT) * 100;
+		/* If FAIRNESS is enabled (not all min rates are zeroes) and
+		   if current min rate is zero - set it to 1.
+		   This is a requirment of the algorithm. */
+		if ((vn_min_rate == 0) && wsum)
+			vn_min_rate = DEF_MIN_RATE;
+		vn_max_rate = ((vn_cfg & FUNC_MF_CFG_MAX_BW_MASK) >>
+				FUNC_MF_CFG_MAX_BW_SHIFT) * 100;
+	}
+
+	DP(NETIF_MSG_IFUP, "func %d: vn_min_rate=%d  vn_max_rate=%d  "
+	   "wsum=%d\n", func, vn_min_rate, vn_max_rate, wsum);
+
+	memset(&m_rs_vn, 0, sizeof(struct rate_shaping_vars_per_vn));
+	memset(&m_fair_vn, 0, sizeof(struct fairness_vars_per_vn));
+
+	/* global vn counter - maximal Mbps for this vn */
+	m_rs_vn.vn_counter.rate = vn_max_rate;
+
+	/* quota - number of bytes transmitted in this period */
+	m_rs_vn.vn_counter.quota =
+				(vn_max_rate * RS_PERIODIC_TIMEOUT_USEC) / 8;
+
+#ifdef BNX2X_PER_PROT_QOS
+	/* per protocol counter */
+	for (protocol = 0; protocol < NUM_OF_PROTOCOLS; protocol++) {
+		/* maximal Mbps for this protocol */
+		m_rs_vn.protocol_counters[protocol].rate =
+						protocol_max_rate[protocol];
+		/* the quota in each timer period -
+		   number of bytes transmitted in this period */
+		m_rs_vn.protocol_counters[protocol].quota =
+			(u32)(rs_periodic_timeout_usec *
+			  ((double)m_rs_vn.
+				   protocol_counters[protocol].rate/8));
+	}
+#endif
+
+	if (wsum) {
+		/* credit for each period of the fairness algorithm:
+		   number of bytes in T_FAIR (the vn share the port rate).
+		   wsum should not be larger than 10000, thus
+		   T_FAIR_COEF / (8 * wsum) will always be grater than zero */
+		m_fair_vn.vn_credit_delta =
+			max((u64)(vn_min_rate * (T_FAIR_COEF / (8 * wsum))),
+			    (u64)(m_cmng_port->fair_vars.fair_threshold * 2));
+		DP(NETIF_MSG_IFUP, "m_fair_vn.vn_credit_delta=%d\n",
+		   m_fair_vn.vn_credit_delta);
+	}
+
+#ifdef BNX2X_PER_PROT_QOS
+	do {
+		u32 protocolWeightSum = 0;
+
+		for (protocol = 0; protocol < NUM_OF_PROTOCOLS; protocol++)
+			protocolWeightSum +=
+					drvInit.protocol_min_rate[protocol];
+		/* per protocol counter -
+		   NOT NEEDED IF NO PER-PROTOCOL CONGESTION MANAGEMENT */
+		if (protocolWeightSum > 0) {
+			for (protocol = 0;
+			     protocol < NUM_OF_PROTOCOLS; protocol++)
+				/* credit for each period of the
+				   fairness algorithm - number of bytes in
+				   T_FAIR (the protocol share the vn rate) */
+				m_fair_vn.protocol_credit_delta[protocol] =
+					(u32)((vn_min_rate / 8) * t_fair *
+					protocol_min_rate / protocolWeightSum);
+		}
+	} while (0);
+#endif
+
+	/* Store it to internal memory */
+	for (i = 0; i < sizeof(struct rate_shaping_vars_per_vn)/4; i++)
+		REG_WR(bp, BAR_XSTRORM_INTMEM +
+		       XSTORM_RATE_SHAPING_PER_VN_VARS_OFFSET(func) + i * 4,
+		       ((u32 *)(&m_rs_vn))[i]);
+
+	for (i = 0; i < sizeof(struct fairness_vars_per_vn)/4; i++)
+		REG_WR(bp, BAR_XSTRORM_INTMEM +
+		       XSTORM_FAIRNESS_PER_VN_VARS_OFFSET(func) + i * 4,
+		       ((u32 *)(&m_fair_vn))[i]);
+}
+
 /* This function is called upon link interrupt */
 static void bnx2x_link_attn(struct bnx2x *bp)
 {
+	int vn;
+
 	bnx2x_phy_hw_lock(bp);
 	bnx2x_link_update(&bp->link_params, &bp->link_vars);
 	bnx2x_phy_hw_unlock(bp);
 
 	/* indicate link status */
 	bnx2x_link_report(bp);
+
+	if (IS_E1HMF(bp)) {
+		int func;
+
+		for (vn = VN_0; vn < E1HVN_MAX; vn++) {
+			if (vn == BP_E1HVN(bp))
+				continue;
+
+			func = ((vn << 1) | BP_PORT(bp));
+
+			/* Set the attention towards other drivers
+			   on the same port */
+			REG_WR(bp, MISC_REG_AEU_GENERAL_ATTN_0 +
+			       (LINK_SYNC_ATTENTION_BIT_FUNC_0 + func)*4, 1);
+		}
+	}
+
+	if (CHIP_IS_E1H(bp) && (bp->link_vars.line_speed > 0)) {
+		struct cmng_struct_per_port m_cmng_port;
+		u32 wsum;
+		int port = BP_PORT(bp);
+
+		/* Init RATE SHAPING and FAIRNESS contexts */
+		wsum = bnx2x_calc_vn_wsum(bp);
+		bnx2x_init_port_minmax(bp, (int)wsum,
+					bp->link_vars.line_speed,
+					&m_cmng_port);
+		if (IS_E1HMF(bp))
+			for (vn = VN_0; vn < E1HVN_MAX; vn++)
+				bnx2x_init_vn_minmax(bp, 2*vn + port,
+					wsum, bp->link_vars.line_speed,
+						     &m_cmng_port);
+	}
 }
 
 static void bnx2x__link_status_update(struct bnx2x *bp)
@@ -1495,6 +1839,20 @@ static void bnx2x__link_status_update(struct bnx2x *bp)
 	bnx2x_link_report(bp);
 }
 
+static void bnx2x_pmf_update(struct bnx2x *bp)
+{
+	int port = BP_PORT(bp);
+	u32 val;
+
+	bp->port.pmf = 1;
+	DP(NETIF_MSG_LINK, "pmf %d\n", bp->port.pmf);
+
+	/* enable nig attention */
+	val = (0xff0f | (1 << (BP_E1HVN(bp) + 4)));
+	REG_WR(bp, HC_REG_TRAILING_EDGE_0 + port*8, val);
+	REG_WR(bp, HC_REG_LEADING_EDGE_0 + port*8, val);
+}
+
 /* end of Link */
 
 /* slow path */
@@ -1507,10 +1865,10 @@ static void bnx2x__link_status_update(struct bnx2x *bp)
 static int bnx2x_sp_post(struct bnx2x *bp, int command, int cid,
 			 u32 data_hi, u32 data_lo, int common)
 {
-	int port = bp->port;
+	int func = BP_FUNC(bp);
 
-	DP(NETIF_MSG_TIMER,
-	   "spe (%x:%x)  command %d  hw_cid %x  data (%x:%x)  left %x\n",
+	DP(BNX2X_MSG_SP/*NETIF_MSG_TIMER*/,
+	   "SPQE (%x:%x)  command %d  hw_cid %x  data (%x:%x)  left %x\n",
 	   (u32)U64_HI(bp->spq_mapping), (u32)(U64_LO(bp->spq_mapping) +
 	   (void *)bp->spq_prod_bd - (void *)bp->spq), command,
 	   HW_CID(bp, cid), data_hi, data_lo, bp->spq_left);
@@ -1520,11 +1878,11 @@ static int bnx2x_sp_post(struct bnx2x *bp, int command, int cid,
 		return -EIO;
 #endif
 
-	spin_lock(&bp->spq_lock);
+	spin_lock_bh(&bp->spq_lock);
 
 	if (!bp->spq_left) {
 		BNX2X_ERR("BUG! SPQ ring full!\n");
-		spin_unlock(&bp->spq_lock);
+		spin_unlock_bh(&bp->spq_lock);
 		bnx2x_panic();
 		return -EBUSY;
 	}
@@ -1553,18 +1911,18 @@ static int bnx2x_sp_post(struct bnx2x *bp, int command, int cid,
 		bp->spq_prod_idx++;
 	}
 
-	REG_WR(bp, BAR_XSTRORM_INTMEM + XSTORM_SPQ_PROD_OFFSET(port),
+	REG_WR(bp, BAR_XSTRORM_INTMEM + XSTORM_SPQ_PROD_OFFSET(func),
 	       bp->spq_prod_idx);
 
-	spin_unlock(&bp->spq_lock);
+	spin_unlock_bh(&bp->spq_lock);
 	return 0;
 }
 
 /* acquire split MCP access lock register */
 static int bnx2x_lock_alr(struct bnx2x *bp)
 {
-	int rc = 0;
 	u32 i, j, val;
+	int rc = 0;
 
 	might_sleep();
 	i = 100;
@@ -1577,10 +1935,8 @@ static int bnx2x_lock_alr(struct bnx2x *bp)
 
 		msleep(5);
 	}
-
 	if (!(val & (1L << 31))) {
 		BNX2X_ERR("Cannot acquire nvram interface\n");
-
 		rc = -EBUSY;
 	}
 
@@ -1631,8 +1987,9 @@ static inline u16 bnx2x_update_dsb_idx(struct bnx2x *bp)
 
 static void bnx2x_attn_int_asserted(struct bnx2x *bp, u32 asserted)
 {
-	int port = bp->port;
-	u32 igu_addr = (IGU_ADDR_ATTN_BITS_SET + IGU_PORT_BASE * port) * 8;
+	int port = BP_PORT(bp);
+	int func = BP_FUNC(bp);
+	u32 igu_addr = (IGU_ADDR_ATTN_BITS_SET + IGU_FUNC_BASE * func) * 8;
 	u32 aeu_addr = port ? MISC_REG_AEU_MASK_ATTN_FUNC_1 :
 			      MISC_REG_AEU_MASK_ATTN_FUNC_0;
 	u32 nig_int_mask_addr = port ? NIG_REG_MASK_INTERRUPT_PORT1 :
@@ -1716,14 +2073,14 @@ static void bnx2x_attn_int_asserted(struct bnx2x *bp, u32 asserted)
 
 static inline void bnx2x_attn_int_deasserted0(struct bnx2x *bp, u32 attn)
 {
-	int port = bp->port;
+	int port = BP_PORT(bp);
 	int reg_offset;
 	u32 val;
 
-	if (attn & AEU_INPUTS_ATTN_BITS_SPIO5) {
+	reg_offset = (port ? MISC_REG_AEU_ENABLE1_FUNC_1_OUT_0 :
+			     MISC_REG_AEU_ENABLE1_FUNC_0_OUT_0);
 
-		reg_offset = (port ? MISC_REG_AEU_ENABLE1_FUNC_1_OUT_0 :
-				     MISC_REG_AEU_ENABLE1_FUNC_0_OUT_0);
+	if (attn & AEU_INPUTS_ATTN_BITS_SPIO5) {
 
 		val = REG_RD(bp, reg_offset);
 		val &= ~AEU_INPUTS_ATTN_BITS_SPIO5;
@@ -1731,7 +2088,7 @@ static inline void bnx2x_attn_int_deasserted0(struct bnx2x *bp, u32 attn)
 
 		BNX2X_ERR("SPIO5 hw attention\n");
 
-		switch (bp->board & SHARED_HW_CFG_BOARD_TYPE_MASK) {
+		switch (bp->common.board & SHARED_HW_CFG_BOARD_TYPE_MASK) {
 		case SHARED_HW_CFG_BOARD_TYPE_BCM957710A1022G:
 			/* Fan failure attention */
 
@@ -1762,6 +2119,17 @@ static inline void bnx2x_attn_int_deasserted0(struct bnx2x *bp, u32 attn)
 			break;
 		}
 	}
+
+	if (attn & HW_INTERRUT_ASSERT_SET_0) {
+
+		val = REG_RD(bp, reg_offset);
+		val &= ~(attn & HW_INTERRUT_ASSERT_SET_0);
+		REG_WR(bp, reg_offset, val);
+
+		BNX2X_ERR("FATAL HW block attention set0 0x%x\n",
+			  (attn & HW_INTERRUT_ASSERT_SET_0));
+		bnx2x_panic();
+	}
 }
 
 static inline void bnx2x_attn_int_deasserted1(struct bnx2x *bp, u32 attn)
@@ -1776,6 +2144,23 @@ static inline void bnx2x_attn_int_deasserted1(struct bnx2x *bp, u32 attn)
 		if (val & 0x2)
 			BNX2X_ERR("FATAL error from DORQ\n");
 	}
+
+	if (attn & HW_INTERRUT_ASSERT_SET_1) {
+
+		int port = BP_PORT(bp);
+		int reg_offset;
+
+		reg_offset = (port ? MISC_REG_AEU_ENABLE1_FUNC_1_OUT_1 :
+				     MISC_REG_AEU_ENABLE1_FUNC_0_OUT_1);
+
+		val = REG_RD(bp, reg_offset);
+		val &= ~(attn & HW_INTERRUT_ASSERT_SET_1);
+		REG_WR(bp, reg_offset, val);
+
+		BNX2X_ERR("FATAL HW block attention set1 0x%x\n",
+			  (attn & HW_INTERRUT_ASSERT_SET_1));
+		bnx2x_panic();
+	}
 }
 
 static inline void bnx2x_attn_int_deasserted2(struct bnx2x *bp, u32 attn)
@@ -1799,13 +2184,41 @@ static inline void bnx2x_attn_int_deasserted2(struct bnx2x *bp, u32 attn)
 		if (val & 0x18000)
 			BNX2X_ERR("FATAL error from PXP\n");
 	}
+
+	if (attn & HW_INTERRUT_ASSERT_SET_2) {
+
+		int port = BP_PORT(bp);
+		int reg_offset;
+
+		reg_offset = (port ? MISC_REG_AEU_ENABLE1_FUNC_1_OUT_2 :
+				     MISC_REG_AEU_ENABLE1_FUNC_0_OUT_2);
+
+		val = REG_RD(bp, reg_offset);
+		val &= ~(attn & HW_INTERRUT_ASSERT_SET_2);
+		REG_WR(bp, reg_offset, val);
+
+		BNX2X_ERR("FATAL HW block attention set2 0x%x\n",
+			  (attn & HW_INTERRUT_ASSERT_SET_2));
+		bnx2x_panic();
+	}
 }
 
 static inline void bnx2x_attn_int_deasserted3(struct bnx2x *bp, u32 attn)
 {
+	u32 val;
+
 	if (attn & EVEREST_GEN_ATTN_IN_USE_MASK) {
 
-		if (attn & BNX2X_MC_ASSERT_BITS) {
+		if (attn & BNX2X_PMF_LINK_ASSERT) {
+			int func = BP_FUNC(bp);
+
+			REG_WR(bp, MISC_REG_AEU_GENERAL_ATTN_12 + func*4, 0);
+			bnx2x__link_status_update(bp);
+			if (SHMEM_RD(bp, func_mb[func].drv_status) &
+							DRV_STATUS_PMF)
+				bnx2x_pmf_update(bp);
+
+		} else if (attn & BNX2X_MC_ASSERT_BITS) {
 
 			BNX2X_ERR("MC assert!\n");
 			REG_WR(bp, MISC_REG_AEU_GENERAL_ATTN_10, 0);
@@ -1818,16 +2231,25 @@ static inline void bnx2x_attn_int_deasserted3(struct bnx2x *bp, u32 attn)
 
 			BNX2X_ERR("MCP assert!\n");
 			REG_WR(bp, MISC_REG_AEU_GENERAL_ATTN_11, 0);
-			bnx2x_mc_assert(bp);
+			bnx2x_fw_dump(bp);
 
 		} else
 			BNX2X_ERR("Unknown HW assert! (attn 0x%x)\n", attn);
 	}
 
 	if (attn & EVEREST_LATCHED_ATTN_IN_USE_MASK) {
-
+		BNX2X_ERR("LATCHED attention 0x%08x (masked)\n", attn);
+		if (attn & BNX2X_GRC_TIMEOUT) {
+			val = CHIP_IS_E1H(bp) ?
+				REG_RD(bp, MISC_REG_GRC_TIMEOUT_ATTN) : 0;
+			BNX2X_ERR("GRC time-out 0x%08x\n", val);
+		}
+		if (attn & BNX2X_GRC_RSV) {
+			val = CHIP_IS_E1H(bp) ?
+				REG_RD(bp, MISC_REG_GRC_RSV_ATTN) : 0;
+			BNX2X_ERR("GRC reserved 0x%08x\n", val);
+		}
 		REG_WR(bp, MISC_REG_AEU_CLR_LATCH_SIGNAL, 0x7ff);
-		BNX2X_ERR("LATCHED attention 0x%x (masked)\n", attn);
 	}
 }
 
@@ -1835,7 +2257,7 @@ static void bnx2x_attn_int_deasserted(struct bnx2x *bp, u32 deasserted)
 {
 	struct attn_route attn;
 	struct attn_route group_mask;
-	int port = bp->port;
+	int port = BP_PORT(bp);
 	int index;
 	u32 reg_addr;
 	u32 val;
@@ -1848,14 +2270,16 @@ static void bnx2x_attn_int_deasserted(struct bnx2x *bp, u32 deasserted)
 	attn.sig[1] = REG_RD(bp, MISC_REG_AEU_AFTER_INVERT_2_FUNC_0 + port*4);
 	attn.sig[2] = REG_RD(bp, MISC_REG_AEU_AFTER_INVERT_3_FUNC_0 + port*4);
 	attn.sig[3] = REG_RD(bp, MISC_REG_AEU_AFTER_INVERT_4_FUNC_0 + port*4);
-	DP(NETIF_MSG_HW, "attn %llx\n", (unsigned long long)attn.sig[0]);
+	DP(NETIF_MSG_HW, "attn: %08x %08x %08x %08x\n",
+	   attn.sig[0], attn.sig[1], attn.sig[2], attn.sig[3]);
 
 	for (index = 0; index < MAX_DYNAMIC_ATTN_GRPS; index++) {
 		if (deasserted & (1 << index)) {
 			group_mask = bp->attn_group[index];
 
-			DP(NETIF_MSG_HW, "group[%d]: %llx\n", index,
-			   (unsigned long long)group_mask.sig[0]);
+			DP(NETIF_MSG_HW, "group[%d]: %08x %08x %08x %08x\n",
+			   index, group_mask.sig[0], group_mask.sig[1],
+			   group_mask.sig[2], group_mask.sig[3]);
 
 			bnx2x_attn_int_deasserted3(bp,
 					attn.sig[3] & group_mask.sig[3]);
@@ -1866,22 +2290,6 @@ static void bnx2x_attn_int_deasserted(struct bnx2x *bp, u32 deasserted)
 			bnx2x_attn_int_deasserted0(bp,
 					attn.sig[0] & group_mask.sig[0]);
 
-			if ((attn.sig[0] & group_mask.sig[0] &
-						HW_INTERRUT_ASSERT_SET_0) ||
-			    (attn.sig[1] & group_mask.sig[1] &
-						HW_INTERRUT_ASSERT_SET_1) ||
-			    (attn.sig[2] & group_mask.sig[2] &
-						HW_INTERRUT_ASSERT_SET_2))
-				BNX2X_ERR("FATAL HW block attention"
-					  "  set0 0x%x  set1 0x%x"
-					  "  set2 0x%x\n",
-					  (attn.sig[0] & group_mask.sig[0] &
-					   HW_INTERRUT_ASSERT_SET_0),
-					  (attn.sig[1] & group_mask.sig[1] &
-					   HW_INTERRUT_ASSERT_SET_1),
-					  (attn.sig[2] & group_mask.sig[2] &
-					   HW_INTERRUT_ASSERT_SET_2));
-
 			if ((attn.sig[0] & group_mask.sig[0] &
 						HW_PRTY_ASSERT_SET_0) ||
 			    (attn.sig[1] & group_mask.sig[1] &
@@ -1894,17 +2302,17 @@ static void bnx2x_attn_int_deasserted(struct bnx2x *bp, u32 deasserted)
 
 	bnx2x_unlock_alr(bp);
 
-	reg_addr = (IGU_ADDR_ATTN_BITS_CLR + IGU_PORT_BASE * port) * 8;
+	reg_addr = (IGU_ADDR_ATTN_BITS_CLR + IGU_FUNC_BASE * BP_FUNC(bp)) * 8;
 
 	val = ~deasserted;
-/*      DP(NETIF_MSG_INTR, "write 0x%08x to IGU addr 0x%x\n",
+/*	DP(NETIF_MSG_INTR, "write 0x%08x to IGU addr 0x%x\n",
 	   val, BAR_IGU_INTMEM + reg_addr); */
 	REG_WR(bp, BAR_IGU_INTMEM + reg_addr, val);
 
 	if (bp->aeu_mask & (deasserted & 0xff))
-		BNX2X_ERR("IGU BUG\n");
+		BNX2X_ERR("IGU BUG!\n");
 	if (~bp->attn_state & deasserted)
-		BNX2X_ERR("IGU BUG\n");
+		BNX2X_ERR("IGU BUG!\n");
 
 	reg_addr = port ? MISC_REG_AEU_MASK_ATTN_FUNC_1 :
 			  MISC_REG_AEU_MASK_ATTN_FUNC_0;
@@ -1936,7 +2344,7 @@ static void bnx2x_attn_int(struct bnx2x *bp)
 	   attn_bits, attn_ack, asserted, deasserted);
 
 	if (~(attn_bits ^ attn_ack) & (attn_bits ^ attn_state))
-		BNX2X_ERR("bad attention state\n");
+		BNX2X_ERR("BAD attention state\n");
 
 	/* handle bits that were raised */
 	if (asserted)
@@ -1951,6 +2359,7 @@ static void bnx2x_sp_task(struct work_struct *work)
 	struct bnx2x *bp = container_of(work, struct bnx2x, sp_task);
 	u16 status;
 
+
 	/* Return here if interrupt is disabled */
 	if (unlikely(atomic_read(&bp->intr_sem) != 0)) {
 		DP(BNX2X_MSG_SP, "called but intr_sem not 0, returning\n");
@@ -1958,19 +2367,15 @@ static void bnx2x_sp_task(struct work_struct *work)
 	}
 
 	status = bnx2x_update_dsb_idx(bp);
-	if (status == 0)
-		BNX2X_ERR("spurious slowpath interrupt!\n");
+/*	if (status == 0)				     */
+/*		BNX2X_ERR("spurious slowpath interrupt!\n"); */
 
-	DP(NETIF_MSG_INTR, "got a slowpath interrupt (updated %x)\n", status);
+	DP(BNX2X_MSG_SP, "got a slowpath interrupt (updated %x)\n", status);
 
 	/* HW attentions */
 	if (status & 0x1)
 		bnx2x_attn_int(bp);
 
-	/* CStorm events: query_stats, port delete ramrod */
-	if (status & 0x2)
-		bp->stat_pending = 0;
-
 	bnx2x_ack_sb(bp, DEF_SB_ID, ATTENTION_ID, bp->def_att_idx,
 		     IGU_INT_NOP, 1);
 	bnx2x_ack_sb(bp, DEF_SB_ID, USTORM_ID, le16_to_cpu(bp->def_u_idx),
@@ -2109,13 +2514,13 @@ static inline long bnx2x_hilo(u32 *hiref)
 static void bnx2x_init_mac_stats(struct bnx2x *bp)
 {
 	struct dmae_command *dmae;
-	int port = bp->port;
+	int port = BP_PORT(bp);
 	int loader_idx = port * 8;
 	u32 opcode;
 	u32 mac_addr;
 
 	bp->executer_idx = 0;
-	if (bp->fw_mb) {
+	if (bp->func_stx) {
 		/* MCP */
 		opcode = (DMAE_CMD_SRC_PCI | DMAE_CMD_DST_GRC |
 			  DMAE_CMD_SRC_RESET | DMAE_CMD_DST_RESET |
@@ -2135,7 +2540,7 @@ static void bnx2x_init_mac_stats(struct bnx2x *bp)
 					   sizeof(u32));
 		dmae->src_addr_hi = U64_HI(bnx2x_sp_mapping(bp, eth_stats) +
 					   sizeof(u32));
-		dmae->dst_addr_lo = bp->fw_mb >> 2;
+		dmae->dst_addr_lo = bp->func_stx >> 2;
 		dmae->dst_addr_hi = 0;
 		dmae->len = (offsetof(struct bnx2x_eth_stats, mac_stx_end) -
 			     sizeof(u32)) >> 2;
@@ -2280,7 +2685,7 @@ static void bnx2x_init_mac_stats(struct bnx2x *bp)
 
 static void bnx2x_init_stats(struct bnx2x *bp)
 {
-	int port = bp->port;
+	int port = BP_PORT(bp);
 
 	bp->stats_state = STATS_STATE_DISABLE;
 	bp->executer_idx = 0;
@@ -2641,8 +3046,6 @@ static void bnx2x_update_net_stats(struct bnx2x *bp)
 
 static void bnx2x_update_stats(struct bnx2x *bp)
 {
-	int i;
-
 	if (!bnx2x_update_storm_stats(bp)) {
 
 		if (bp->link_vars.mac_type == MAC_TYPE_BMAC) {
@@ -2662,6 +3065,7 @@ static void bnx2x_update_stats(struct bnx2x *bp)
 	if (bp->msglevel & NETIF_MSG_TIMER) {
 		struct bnx2x_eth_stats *estats = bnx2x_sp(bp, eth_stats);
 		struct net_device_stats *nstats = &bp->dev->stats;
+		int i;
 
 		printk(KERN_DEBUG "%s:\n", bp->dev->name);
 		printk(KERN_DEBUG "  tx avail (%4x)  tx hc idx (%x)"
@@ -2707,7 +3111,7 @@ static void bnx2x_update_stats(struct bnx2x *bp)
 	/* loader */
 	if (bp->executer_idx) {
 		struct dmae_command *dmae = &bp->dmae;
-		int port = bp->port;
+		int port = BP_PORT(bp);
 		int loader_idx = port * 8;
 
 		memset(dmae, 0, sizeof(struct dmae_command));
@@ -2766,8 +3170,8 @@ static void bnx2x_timer(unsigned long data)
 		rc = bnx2x_rx_int(fp, 1000);
 	}
 
-	if (!nomcp) {
-		int port = bp->port;
+	if (!BP_NOMCP(bp)) {
+		int func = BP_FUNC(bp);
 		u32 drv_pulse;
 		u32 mcp_pulse;
 
@@ -2775,9 +3179,9 @@ static void bnx2x_timer(unsigned long data)
 		bp->fw_drv_pulse_wr_seq &= DRV_PULSE_SEQ_MASK;
 		/* TBD - add SYSTEM_TIME */
 		drv_pulse = bp->fw_drv_pulse_wr_seq;
-		SHMEM_WR(bp, func_mb[port].drv_pulse_mb, drv_pulse);
+		SHMEM_WR(bp, func_mb[func].drv_pulse_mb, drv_pulse);
 
-		mcp_pulse = (SHMEM_RD(bp, func_mb[port].mcp_pulse_mb) &
+		mcp_pulse = (SHMEM_RD(bp, func_mb[func].mcp_pulse_mb) &
 			     MCP_PULSE_SEQ_MASK);
 		/* The delta between driver pulse and mcp response
 		 * should be 1 (before mcp response) or 0 (after mcp response)
@@ -2807,58 +3211,89 @@ timer_restart:
  * nic init service functions
  */
 
-static void bnx2x_init_sb(struct bnx2x *bp, struct host_status_block *sb,
-			  dma_addr_t mapping, int id)
+static void bnx2x_zero_sb(struct bnx2x *bp, int sb_id)
 {
-	int port = bp->port;
-	u64 section;
+	int port = BP_PORT(bp);
+
+	bnx2x_init_fill(bp, BAR_USTRORM_INTMEM +
+			USTORM_SB_HOST_STATUS_BLOCK_OFFSET(port, sb_id), 0,
+			sizeof(struct ustorm_def_status_block)/4);
+	bnx2x_init_fill(bp, BAR_CSTRORM_INTMEM +
+			CSTORM_SB_HOST_STATUS_BLOCK_OFFSET(port, sb_id), 0,
+			sizeof(struct cstorm_def_status_block)/4);
+}
+
+static void bnx2x_init_sb(struct bnx2x *bp, int sb_id,
+			  struct host_status_block *sb,	dma_addr_t mapping)
+{
+	int port = BP_PORT(bp);
 	int index;
+	u64 section;
 
 	/* USTORM */
 	section = ((u64)mapping) + offsetof(struct host_status_block,
 					    u_status_block);
-	sb->u_status_block.status_block_id = id;
+	sb->u_status_block.status_block_id = sb_id;
 
 	REG_WR(bp, BAR_USTRORM_INTMEM +
-	       USTORM_SB_HOST_SB_ADDR_OFFSET(port, id), U64_LO(section));
+	       USTORM_SB_HOST_SB_ADDR_OFFSET(port, sb_id), U64_LO(section));
 	REG_WR(bp, BAR_USTRORM_INTMEM +
-	       ((USTORM_SB_HOST_SB_ADDR_OFFSET(port, id)) + 4),
+	       ((USTORM_SB_HOST_SB_ADDR_OFFSET(port, sb_id)) + 4),
 	       U64_HI(section));
 
 	for (index = 0; index < HC_USTORM_SB_NUM_INDICES; index++)
 		REG_WR16(bp, BAR_USTRORM_INTMEM +
-			 USTORM_SB_HC_DISABLE_OFFSET(port, id, index), 0x1);
+			 USTORM_SB_HC_DISABLE_OFFSET(port, sb_id, index), 1);
 
 	/* CSTORM */
 	section = ((u64)mapping) + offsetof(struct host_status_block,
 					    c_status_block);
-	sb->c_status_block.status_block_id = id;
+	sb->c_status_block.status_block_id = sb_id;
 
 	REG_WR(bp, BAR_CSTRORM_INTMEM +
-	       CSTORM_SB_HOST_SB_ADDR_OFFSET(port, id), U64_LO(section));
+	       CSTORM_SB_HOST_SB_ADDR_OFFSET(port, sb_id), U64_LO(section));
 	REG_WR(bp, BAR_CSTRORM_INTMEM +
-	       ((CSTORM_SB_HOST_SB_ADDR_OFFSET(port, id)) + 4),
+	       ((CSTORM_SB_HOST_SB_ADDR_OFFSET(port, sb_id)) + 4),
 	       U64_HI(section));
 
 	for (index = 0; index < HC_CSTORM_SB_NUM_INDICES; index++)
 		REG_WR16(bp, BAR_CSTRORM_INTMEM +
-			 CSTORM_SB_HC_DISABLE_OFFSET(port, id, index), 0x1);
+			 CSTORM_SB_HC_DISABLE_OFFSET(port, sb_id, index), 1);
+
+	bnx2x_ack_sb(bp, sb_id, CSTORM_ID, 0, IGU_INT_ENABLE, 0);
+}
+
+static void bnx2x_zero_def_sb(struct bnx2x *bp)
+{
+	int func = BP_FUNC(bp);
 
-	bnx2x_ack_sb(bp, id, CSTORM_ID, 0, IGU_INT_ENABLE, 0);
+	bnx2x_init_fill(bp, BAR_USTRORM_INTMEM +
+			USTORM_DEF_SB_HOST_STATUS_BLOCK_OFFSET(func), 0,
+			sizeof(struct ustorm_def_status_block)/4);
+	bnx2x_init_fill(bp, BAR_CSTRORM_INTMEM +
+			CSTORM_DEF_SB_HOST_STATUS_BLOCK_OFFSET(func), 0,
+			sizeof(struct cstorm_def_status_block)/4);
+	bnx2x_init_fill(bp, BAR_XSTRORM_INTMEM +
+			XSTORM_DEF_SB_HOST_STATUS_BLOCK_OFFSET(func), 0,
+			sizeof(struct xstorm_def_status_block)/4);
+	bnx2x_init_fill(bp, BAR_TSTRORM_INTMEM +
+			TSTORM_DEF_SB_HOST_STATUS_BLOCK_OFFSET(func), 0,
+			sizeof(struct tstorm_def_status_block)/4);
 }
 
 static void bnx2x_init_def_sb(struct bnx2x *bp,
 			      struct host_def_status_block *def_sb,
-			      dma_addr_t mapping, int id)
+			      dma_addr_t mapping, int sb_id)
 {
-	int port = bp->port;
+	int port = BP_PORT(bp);
+	int func = BP_FUNC(bp);
 	int index, val, reg_offset;
 	u64 section;
 
 	/* ATTN */
 	section = ((u64)mapping) + offsetof(struct host_def_status_block,
 					    atten_status_block);
-	def_sb->atten_status_block.status_block_id = id;
+	def_sb->atten_status_block.status_block_id = sb_id;
 
 	bp->def_att_idx = 0;
 	bp->attn_state = 0;
@@ -2866,7 +3301,7 @@ static void bnx2x_init_def_sb(struct bnx2x *bp,
 	reg_offset = (port ? MISC_REG_AEU_ENABLE1_FUNC_1_OUT_0 :
 			     MISC_REG_AEU_ENABLE1_FUNC_0_OUT_0);
 
-	for (index = 0; index < 3; index++) {
+	for (index = 0; index < MAX_DYNAMIC_ATTN_GRPS; index++) {
 		bp->attn_group[index].sig[0] = REG_RD(bp,
 						     reg_offset + 0x10*index);
 		bp->attn_group[index].sig[1] = REG_RD(bp,
@@ -2889,116 +3324,123 @@ static void bnx2x_init_def_sb(struct bnx2x *bp,
 	reg_offset = (port ? HC_REG_ATTN_NUM_P1 : HC_REG_ATTN_NUM_P0);
 
 	val = REG_RD(bp, reg_offset);
-	val |= id;
+	val |= sb_id;
 	REG_WR(bp, reg_offset, val);
 
 	/* USTORM */
 	section = ((u64)mapping) + offsetof(struct host_def_status_block,
 					    u_def_status_block);
-	def_sb->u_def_status_block.status_block_id = id;
+	def_sb->u_def_status_block.status_block_id = sb_id;
 
 	bp->def_u_idx = 0;
 
 	REG_WR(bp, BAR_USTRORM_INTMEM +
-	       USTORM_DEF_SB_HOST_SB_ADDR_OFFSET(port), U64_LO(section));
+	       USTORM_DEF_SB_HOST_SB_ADDR_OFFSET(func), U64_LO(section));
 	REG_WR(bp, BAR_USTRORM_INTMEM +
-	       ((USTORM_DEF_SB_HOST_SB_ADDR_OFFSET(port)) + 4),
+	       ((USTORM_DEF_SB_HOST_SB_ADDR_OFFSET(func)) + 4),
 	       U64_HI(section));
-	REG_WR(bp, BAR_USTRORM_INTMEM + USTORM_HC_BTR_OFFSET(port),
+	REG_WR8(bp, BAR_USTRORM_INTMEM +  DEF_USB_FUNC_OFF +
+		USTORM_DEF_SB_HOST_STATUS_BLOCK_OFFSET(func), func);
+	REG_WR(bp, BAR_USTRORM_INTMEM + USTORM_HC_BTR_OFFSET(func),
 	       BNX2X_BTR);
 
 	for (index = 0; index < HC_USTORM_DEF_SB_NUM_INDICES; index++)
 		REG_WR16(bp, BAR_USTRORM_INTMEM +
-			 USTORM_DEF_SB_HC_DISABLE_OFFSET(port, index), 0x1);
+			 USTORM_DEF_SB_HC_DISABLE_OFFSET(func, index), 1);
 
 	/* CSTORM */
 	section = ((u64)mapping) + offsetof(struct host_def_status_block,
 					    c_def_status_block);
-	def_sb->c_def_status_block.status_block_id = id;
+	def_sb->c_def_status_block.status_block_id = sb_id;
 
 	bp->def_c_idx = 0;
 
 	REG_WR(bp, BAR_CSTRORM_INTMEM +
-	       CSTORM_DEF_SB_HOST_SB_ADDR_OFFSET(port), U64_LO(section));
+	       CSTORM_DEF_SB_HOST_SB_ADDR_OFFSET(func), U64_LO(section));
 	REG_WR(bp, BAR_CSTRORM_INTMEM +
-	       ((CSTORM_DEF_SB_HOST_SB_ADDR_OFFSET(port)) + 4),
+	       ((CSTORM_DEF_SB_HOST_SB_ADDR_OFFSET(func)) + 4),
 	       U64_HI(section));
-	REG_WR(bp, BAR_CSTRORM_INTMEM + CSTORM_HC_BTR_OFFSET(port),
+	REG_WR8(bp, BAR_CSTRORM_INTMEM +  DEF_CSB_FUNC_OFF +
+		CSTORM_DEF_SB_HOST_STATUS_BLOCK_OFFSET(func), func);
+	REG_WR(bp, BAR_CSTRORM_INTMEM + CSTORM_HC_BTR_OFFSET(func),
 	       BNX2X_BTR);
 
 	for (index = 0; index < HC_CSTORM_DEF_SB_NUM_INDICES; index++)
 		REG_WR16(bp, BAR_CSTRORM_INTMEM +
-			 CSTORM_DEF_SB_HC_DISABLE_OFFSET(port, index), 0x1);
+			 CSTORM_DEF_SB_HC_DISABLE_OFFSET(func, index), 1);
 
 	/* TSTORM */
 	section = ((u64)mapping) + offsetof(struct host_def_status_block,
 					    t_def_status_block);
-	def_sb->t_def_status_block.status_block_id = id;
+	def_sb->t_def_status_block.status_block_id = sb_id;
 
 	bp->def_t_idx = 0;
 
 	REG_WR(bp, BAR_TSTRORM_INTMEM +
-	       TSTORM_DEF_SB_HOST_SB_ADDR_OFFSET(port), U64_LO(section));
+	       TSTORM_DEF_SB_HOST_SB_ADDR_OFFSET(func), U64_LO(section));
 	REG_WR(bp, BAR_TSTRORM_INTMEM +
-	       ((TSTORM_DEF_SB_HOST_SB_ADDR_OFFSET(port)) + 4),
+	       ((TSTORM_DEF_SB_HOST_SB_ADDR_OFFSET(func)) + 4),
 	       U64_HI(section));
-	REG_WR(bp, BAR_TSTRORM_INTMEM + TSTORM_HC_BTR_OFFSET(port),
+	REG_WR8(bp, BAR_TSTRORM_INTMEM +  DEF_TSB_FUNC_OFF +
+		TSTORM_DEF_SB_HOST_STATUS_BLOCK_OFFSET(func), func);
+	REG_WR(bp, BAR_TSTRORM_INTMEM + TSTORM_HC_BTR_OFFSET(func),
 	       BNX2X_BTR);
 
 	for (index = 0; index < HC_TSTORM_DEF_SB_NUM_INDICES; index++)
 		REG_WR16(bp, BAR_TSTRORM_INTMEM +
-			 TSTORM_DEF_SB_HC_DISABLE_OFFSET(port, index), 0x1);
+			 TSTORM_DEF_SB_HC_DISABLE_OFFSET(func, index), 1);
 
 	/* XSTORM */
 	section = ((u64)mapping) + offsetof(struct host_def_status_block,
 					    x_def_status_block);
-	def_sb->x_def_status_block.status_block_id = id;
+	def_sb->x_def_status_block.status_block_id = sb_id;
 
 	bp->def_x_idx = 0;
 
 	REG_WR(bp, BAR_XSTRORM_INTMEM +
-	       XSTORM_DEF_SB_HOST_SB_ADDR_OFFSET(port), U64_LO(section));
+	       XSTORM_DEF_SB_HOST_SB_ADDR_OFFSET(func), U64_LO(section));
 	REG_WR(bp, BAR_XSTRORM_INTMEM +
-	       ((XSTORM_DEF_SB_HOST_SB_ADDR_OFFSET(port)) + 4),
+	       ((XSTORM_DEF_SB_HOST_SB_ADDR_OFFSET(func)) + 4),
 	       U64_HI(section));
-	REG_WR(bp, BAR_XSTRORM_INTMEM + XSTORM_HC_BTR_OFFSET(port),
+	REG_WR8(bp, BAR_XSTRORM_INTMEM +  DEF_XSB_FUNC_OFF +
+		XSTORM_DEF_SB_HOST_STATUS_BLOCK_OFFSET(func), func);
+	REG_WR(bp, BAR_XSTRORM_INTMEM + XSTORM_HC_BTR_OFFSET(func),
 	       BNX2X_BTR);
 
 	for (index = 0; index < HC_XSTORM_DEF_SB_NUM_INDICES; index++)
 		REG_WR16(bp, BAR_XSTRORM_INTMEM +
-			 XSTORM_DEF_SB_HC_DISABLE_OFFSET(port, index), 0x1);
-
-	bp->stat_pending = 0;
+			 XSTORM_DEF_SB_HC_DISABLE_OFFSET(func, index), 1);
 
-	bnx2x_ack_sb(bp, id, CSTORM_ID, 0, IGU_INT_ENABLE, 0);
+	bnx2x_ack_sb(bp, sb_id, CSTORM_ID, 0, IGU_INT_ENABLE, 0);
 }
 
 static void bnx2x_update_coalesce(struct bnx2x *bp)
 {
-	int port = bp->port;
+	int port = BP_PORT(bp);
 	int i;
 
 	for_each_queue(bp, i) {
+		int sb_id = bp->fp[i].sb_id;
 
 		/* HC_INDEX_U_ETH_RX_CQ_CONS */
 		REG_WR8(bp, BAR_USTRORM_INTMEM +
-			USTORM_SB_HC_TIMEOUT_OFFSET(port, i,
+			USTORM_SB_HC_TIMEOUT_OFFSET(port, sb_id,
 						   HC_INDEX_U_ETH_RX_CQ_CONS),
-			bp->rx_ticks_int/12);
+			bp->rx_ticks/12);
 		REG_WR16(bp, BAR_USTRORM_INTMEM +
-			 USTORM_SB_HC_DISABLE_OFFSET(port, i,
+			 USTORM_SB_HC_DISABLE_OFFSET(port, sb_id,
 						   HC_INDEX_U_ETH_RX_CQ_CONS),
-			 bp->rx_ticks_int ? 0 : 1);
+			 bp->rx_ticks ? 0 : 1);
 
 		/* HC_INDEX_C_ETH_TX_CQ_CONS */
 		REG_WR8(bp, BAR_CSTRORM_INTMEM +
-			CSTORM_SB_HC_TIMEOUT_OFFSET(port, i,
+			CSTORM_SB_HC_TIMEOUT_OFFSET(port, sb_id,
 						   HC_INDEX_C_ETH_TX_CQ_CONS),
-			bp->tx_ticks_int/12);
+			bp->tx_ticks/12);
 		REG_WR16(bp, BAR_CSTRORM_INTMEM +
-			 CSTORM_SB_HC_DISABLE_OFFSET(port, i,
+			 CSTORM_SB_HC_DISABLE_OFFSET(port, sb_id,
 						   HC_INDEX_C_ETH_TX_CQ_CONS),
-			 bp->tx_ticks_int ? 0 : 1);
+			 bp->tx_ticks ? 0 : 1);
 	}
 }
 
@@ -3006,7 +3448,6 @@ static void bnx2x_init_rx_rings(struct bnx2x *bp)
 {
 	u16 ring_prod;
 	int i, j;
-	int port = bp->port;
 
 	bp->rx_buf_use_size = bp->dev->mtu;
 
@@ -3025,13 +3466,13 @@ static void bnx2x_init_rx_rings(struct bnx2x *bp)
 			rx_bd = &fp->rx_desc_ring[RX_DESC_CNT * i - 2];
 			rx_bd->addr_hi =
 				cpu_to_le32(U64_HI(fp->rx_desc_mapping +
-					   BCM_PAGE_SIZE*(i % NUM_RX_RINGS)));
+					    BCM_PAGE_SIZE*(i % NUM_RX_RINGS)));
 			rx_bd->addr_lo =
 				cpu_to_le32(U64_LO(fp->rx_desc_mapping +
-					   BCM_PAGE_SIZE*(i % NUM_RX_RINGS)));
-
+					    BCM_PAGE_SIZE*(i % NUM_RX_RINGS)));
 		}
 
+		/* CQ ring */
 		for (i = 1; i <= NUM_RCQ_RINGS; i++) {
 			struct eth_rx_cqe_next_page *nextpg;
 
@@ -3039,10 +3480,10 @@ static void bnx2x_init_rx_rings(struct bnx2x *bp)
 				&fp->rx_comp_ring[RCQ_DESC_CNT * i - 1];
 			nextpg->addr_hi =
 				cpu_to_le32(U64_HI(fp->rx_comp_mapping +
-					  BCM_PAGE_SIZE*(i % NUM_RCQ_RINGS)));
+					   BCM_PAGE_SIZE*(i % NUM_RCQ_RINGS)));
 			nextpg->addr_lo =
 				cpu_to_le32(U64_LO(fp->rx_comp_mapping +
-					  BCM_PAGE_SIZE*(i % NUM_RCQ_RINGS)));
+					   BCM_PAGE_SIZE*(i % NUM_RCQ_RINGS)));
 		}
 
 		/* rx completion queue */
@@ -3064,15 +3505,16 @@ static void bnx2x_init_rx_rings(struct bnx2x *bp)
 		/* Warning! this will generate an interrupt (to the TSTORM) */
 		/* must only be done when chip is initialized */
 		REG_WR(bp, BAR_TSTRORM_INTMEM +
-		       TSTORM_RCQ_PROD_OFFSET(port, j), ring_prod);
+		       TSTORM_RX_PRODS_OFFSET(BP_PORT(bp), FP_CL_ID(fp)),
+			ring_prod);
 		if (j != 0)
 			continue;
 
 		REG_WR(bp, BAR_USTRORM_INTMEM +
-		       USTORM_MEM_WORKAROUND_ADDRESS_OFFSET(port),
+		       USTORM_MEM_WORKAROUND_ADDRESS_OFFSET(BP_PORT(bp)),
 		       U64_LO(fp->rx_comp_mapping));
 		REG_WR(bp, BAR_USTRORM_INTMEM +
-		       USTORM_MEM_WORKAROUND_ADDRESS_OFFSET(port) + 4,
+		       USTORM_MEM_WORKAROUND_ADDRESS_OFFSET(BP_PORT(bp)) + 4,
 		       U64_HI(fp->rx_comp_mapping));
 	}
 }
@@ -3090,10 +3532,10 @@ static void bnx2x_init_tx_ring(struct bnx2x *bp)
 
 			tx_bd->addr_hi =
 				cpu_to_le32(U64_HI(fp->tx_desc_mapping +
-					   BCM_PAGE_SIZE*(i % NUM_TX_RINGS)));
+					    BCM_PAGE_SIZE*(i % NUM_TX_RINGS)));
 			tx_bd->addr_lo =
 				cpu_to_le32(U64_LO(fp->tx_desc_mapping +
-					   BCM_PAGE_SIZE*(i % NUM_TX_RINGS)));
+					    BCM_PAGE_SIZE*(i % NUM_TX_RINGS)));
 		}
 
 		fp->tx_pkt_prod = 0;
@@ -3107,7 +3549,7 @@ static void bnx2x_init_tx_ring(struct bnx2x *bp)
 
 static void bnx2x_init_sp_ring(struct bnx2x *bp)
 {
-	int port = bp->port;
+	int func = BP_FUNC(bp);
 
 	spin_lock_init(&bp->spq_lock);
 
@@ -3117,12 +3559,13 @@ static void bnx2x_init_sp_ring(struct bnx2x *bp)
 	bp->spq_prod_bd = bp->spq;
 	bp->spq_last_bd = bp->spq_prod_bd + MAX_SP_DESC_CNT;
 
-	REG_WR(bp, BAR_XSTRORM_INTMEM + XSTORM_SPQ_PAGE_BASE_OFFSET(port),
+	REG_WR(bp, XSEM_REG_FAST_MEMORY + XSTORM_SPQ_PAGE_BASE_OFFSET(func),
 	       U64_LO(bp->spq_mapping));
-	REG_WR(bp, BAR_XSTRORM_INTMEM + XSTORM_SPQ_PAGE_BASE_OFFSET(port) + 4,
+	REG_WR(bp,
+	       XSEM_REG_FAST_MEMORY + XSTORM_SPQ_PAGE_BASE_OFFSET(func) + 4,
 	       U64_HI(bp->spq_mapping));
 
-	REG_WR(bp, XSEM_REG_FAST_MEMORY + XSTORM_SPQ_PROD_OFFSET(port),
+	REG_WR(bp, XSEM_REG_FAST_MEMORY + XSTORM_SPQ_PROD_OFFSET(func),
 	       bp->spq_prod_idx);
 }
 
@@ -3133,6 +3576,7 @@ static void bnx2x_init_context(struct bnx2x *bp)
 	for_each_queue(bp, i) {
 		struct eth_context *context = bnx2x_sp(bp, context[i].eth);
 		struct bnx2x_fastpath *fp = &bp->fp[i];
+		u8 sb_id = FP_SB_ID(fp);
 
 		context->xstorm_st_context.tx_bd_page_base_hi =
 						U64_HI(fp->tx_desc_mapping);
@@ -3142,26 +3586,25 @@ static void bnx2x_init_context(struct bnx2x *bp)
 						U64_HI(fp->tx_prods_mapping);
 		context->xstorm_st_context.db_data_addr_lo =
 						U64_LO(fp->tx_prods_mapping);
-
-		context->ustorm_st_context.rx_bd_page_base_hi =
+		context->xstorm_st_context.statistics_data = (BP_CL_ID(bp) |
+				XSTORM_ETH_ST_CONTEXT_STATISTICS_ENABLE);
+
+		context->ustorm_st_context.common.sb_index_numbers =
+						BNX2X_RX_SB_INDEX_NUM;
+		context->ustorm_st_context.common.clientId = FP_CL_ID(fp);
+		context->ustorm_st_context.common.status_block_id = sb_id;
+		context->ustorm_st_context.common.flags =
+			USTORM_ETH_ST_CONTEXT_CONFIG_ENABLE_MC_ALIGNMENT;
+		context->ustorm_st_context.common.mc_alignment_size = 64;
+		context->ustorm_st_context.common.bd_buff_size =
+						bp->rx_buf_use_size;
+		context->ustorm_st_context.common.bd_page_base_hi =
 						U64_HI(fp->rx_desc_mapping);
-		context->ustorm_st_context.rx_bd_page_base_lo =
+		context->ustorm_st_context.common.bd_page_base_lo =
 						U64_LO(fp->rx_desc_mapping);
-		context->ustorm_st_context.status_block_id = i;
-		context->ustorm_st_context.sb_index_number =
-						HC_INDEX_U_ETH_RX_CQ_CONS;
-		context->ustorm_st_context.rcq_base_address_hi =
-						U64_HI(fp->rx_comp_mapping);
-		context->ustorm_st_context.rcq_base_address_lo =
-						U64_LO(fp->rx_comp_mapping);
-		context->ustorm_st_context.flags =
-				USTORM_ETH_ST_CONTEXT_ENABLE_MC_ALIGNMENT;
-		context->ustorm_st_context.mc_alignment_size = 64;
-		context->ustorm_st_context.num_rss = bp->num_queues;
-
 		context->cstorm_st_context.sb_index_number =
 						HC_INDEX_C_ETH_TX_CQ_CONS;
-		context->cstorm_st_context.status_block_id = i;
+		context->cstorm_st_context.status_block_id = sb_id;
 
 		context->xstorm_ag_context.cdu_reserved =
 			CDU_RSRVD_VALUE_TYPE_A(HW_CID(bp, i),
@@ -3176,14 +3619,16 @@ static void bnx2x_init_context(struct bnx2x *bp)
 
 static void bnx2x_init_ind_table(struct bnx2x *bp)
 {
-	int port = bp->port;
+	int port = BP_PORT(bp);
 	int i;
 
 	if (!is_multi(bp))
 		return;
 
+	DP(NETIF_MSG_IFUP, "Initializing indirection table\n");
 	for (i = 0; i < TSTORM_INDIRECTION_TABLE_SIZE; i++)
-		REG_WR8(bp, TSTORM_INDIRECTION_TABLE_OFFSET(port) + i,
+		REG_WR8(bp, BAR_TSTRORM_INTMEM +
+			TSTORM_INDIRECTION_TABLE_OFFSET(port) + i,
 			i % bp->num_queues);
 
 	REG_WR(bp, PRS_REG_A_PRSU_20, 0xf);
@@ -3191,77 +3636,74 @@ static void bnx2x_init_ind_table(struct bnx2x *bp)
 
 static void bnx2x_set_client_config(struct bnx2x *bp)
 {
-#ifdef BCM_VLAN
-	int mode = bp->rx_mode;
-#endif
-	int i, port = bp->port;
 	struct tstorm_eth_client_config tstorm_client = {0};
+	int port = BP_PORT(bp);
+	int i;
 
-	tstorm_client.mtu = bp->dev->mtu;
+	tstorm_client.mtu = bp->dev->mtu + ETH_OVREHEAD;
 	tstorm_client.statistics_counter_id = 0;
 	tstorm_client.config_flags =
 				TSTORM_ETH_CLIENT_CONFIG_STATSITICS_ENABLE;
 #ifdef BCM_VLAN
-	if (mode && bp->vlgrp) {
+	if (bp->rx_mode && bp->vlgrp) {
 		tstorm_client.config_flags |=
 				TSTORM_ETH_CLIENT_CONFIG_VLAN_REMOVAL_ENABLE;
 		DP(NETIF_MSG_IFUP, "vlan removal enabled\n");
 	}
 #endif
-	if (mode != BNX2X_RX_MODE_PROMISC)
-		tstorm_client.drop_flags =
-				TSTORM_ETH_CLIENT_CONFIG_DROP_MAC_ERR;
 
 	for_each_queue(bp, i) {
 		REG_WR(bp, BAR_TSTRORM_INTMEM +
-		       TSTORM_CLIENT_CONFIG_OFFSET(port, i),
+		       TSTORM_CLIENT_CONFIG_OFFSET(port, bp->fp[i].cl_id),
 		       ((u32 *)&tstorm_client)[0]);
 		REG_WR(bp, BAR_TSTRORM_INTMEM +
-		       TSTORM_CLIENT_CONFIG_OFFSET(port, i) + 4,
+		       TSTORM_CLIENT_CONFIG_OFFSET(port, bp->fp[i].cl_id) + 4,
 		       ((u32 *)&tstorm_client)[1]);
 	}
 
-/*	DP(NETIF_MSG_IFUP, "tstorm_client: 0x%08x 0x%08x\n",
-	   ((u32 *)&tstorm_client)[0], ((u32 *)&tstorm_client)[1]); */
+	DP(BNX2X_MSG_OFF, "tstorm_client: 0x%08x 0x%08x\n",
+	   ((u32 *)&tstorm_client)[0], ((u32 *)&tstorm_client)[1]);
 }
 
 static void bnx2x_set_storm_rx_mode(struct bnx2x *bp)
 {
-	int mode = bp->rx_mode;
-	int port = bp->port;
 	struct tstorm_eth_mac_filter_config tstorm_mac_filter = {0};
+	int mode = bp->rx_mode;
+	int mask = (1 << BP_L_ID(bp));
+	int func = BP_FUNC(bp);
 	int i;
 
 	DP(NETIF_MSG_RX_STATUS, "rx mode is %d\n", mode);
 
 	switch (mode) {
 	case BNX2X_RX_MODE_NONE: /* no Rx */
-		tstorm_mac_filter.ucast_drop_all = 1;
-		tstorm_mac_filter.mcast_drop_all = 1;
-		tstorm_mac_filter.bcast_drop_all = 1;
+		tstorm_mac_filter.ucast_drop_all = mask;
+		tstorm_mac_filter.mcast_drop_all = mask;
+		tstorm_mac_filter.bcast_drop_all = mask;
 		break;
 	case BNX2X_RX_MODE_NORMAL:
-		tstorm_mac_filter.bcast_accept_all = 1;
+		tstorm_mac_filter.bcast_accept_all = mask;
 		break;
 	case BNX2X_RX_MODE_ALLMULTI:
-		tstorm_mac_filter.mcast_accept_all = 1;
-		tstorm_mac_filter.bcast_accept_all = 1;
+		tstorm_mac_filter.mcast_accept_all = mask;
+		tstorm_mac_filter.bcast_accept_all = mask;
 		break;
 	case BNX2X_RX_MODE_PROMISC:
-		tstorm_mac_filter.ucast_accept_all = 1;
-		tstorm_mac_filter.mcast_accept_all = 1;
-		tstorm_mac_filter.bcast_accept_all = 1;
+		tstorm_mac_filter.ucast_accept_all = mask;
+		tstorm_mac_filter.mcast_accept_all = mask;
+		tstorm_mac_filter.bcast_accept_all = mask;
 		break;
 	default:
-		BNX2X_ERR("bad rx mode (%d)\n", mode);
+		BNX2X_ERR("BAD rx mode (%d)\n", mode);
+		break;
 	}
 
 	for (i = 0; i < sizeof(struct tstorm_eth_mac_filter_config)/4; i++) {
 		REG_WR(bp, BAR_TSTRORM_INTMEM +
-		       TSTORM_MAC_FILTER_CONFIG_OFFSET(port) + i * 4,
+		       TSTORM_MAC_FILTER_CONFIG_OFFSET(func) + i * 4,
 		       ((u32 *)&tstorm_mac_filter)[i]);
 
-/*      	DP(NETIF_MSG_IFUP, "tstorm_mac_filter[%d]: 0x%08x\n", i,
+/*		DP(NETIF_MSG_IFUP, "tstorm_mac_filter[%d]: 0x%08x\n", i,
 		   ((u32 *)&tstorm_mac_filter)[i]); */
 	}
 
@@ -3271,26 +3713,30 @@ static void bnx2x_set_storm_rx_mode(struct bnx2x *bp)
 
 static void bnx2x_init_internal(struct bnx2x *bp)
 {
-	int port = bp->port;
 	struct tstorm_eth_function_common_config tstorm_config = {0};
 	struct stats_indication_flags stats_flags = {0};
+	int port = BP_PORT(bp);
+	int func = BP_FUNC(bp);
+	int i;
 
 	if (is_multi(bp)) {
 		tstorm_config.config_flags = MULTI_FLAGS;
 		tstorm_config.rss_result_mask = MULTI_MASK;
 	}
 
+	tstorm_config.leading_client_id = BP_L_ID(bp);
+
 	REG_WR(bp, BAR_TSTRORM_INTMEM +
-	       TSTORM_FUNCTION_COMMON_CONFIG_OFFSET(port),
+	       TSTORM_FUNCTION_COMMON_CONFIG_OFFSET(func),
 	       (*(u32 *)&tstorm_config));
 
-/*      DP(NETIF_MSG_IFUP, "tstorm_config: 0x%08x\n",
+/*	DP(NETIF_MSG_IFUP, "tstorm_config: 0x%08x\n",
 	   (*(u32 *)&tstorm_config)); */
 
 	bp->rx_mode = BNX2X_RX_MODE_NONE; /* no rx until link is up */
 	bnx2x_set_storm_rx_mode(bp);
 
-	stats_flags.collect_eth = cpu_to_le32(1);
+	stats_flags.collect_eth = 1;
 
 	REG_WR(bp, BAR_XSTRORM_INTMEM + XSTORM_STATS_FLAGS_OFFSET(port),
 	       ((u32 *)&stats_flags)[0]);
@@ -3307,8 +3753,28 @@ static void bnx2x_init_internal(struct bnx2x *bp)
 	REG_WR(bp, BAR_CSTRORM_INTMEM + CSTORM_STATS_FLAGS_OFFSET(port) + 4,
 	       ((u32 *)&stats_flags)[1]);
 
-/*      DP(NETIF_MSG_IFUP, "stats_flags: 0x%08x 0x%08x\n",
+/*	DP(NETIF_MSG_IFUP, "stats_flags: 0x%08x 0x%08x\n",
 	   ((u32 *)&stats_flags)[0], ((u32 *)&stats_flags)[1]); */
+
+	if (CHIP_IS_E1H(bp)) {
+		REG_WR8(bp, BAR_XSTRORM_INTMEM + XSTORM_FUNCTION_MODE_OFFSET,
+			IS_E1HMF(bp));
+		REG_WR8(bp, BAR_TSTRORM_INTMEM + TSTORM_FUNCTION_MODE_OFFSET,
+			IS_E1HMF(bp));
+		REG_WR8(bp, BAR_CSTRORM_INTMEM + CSTORM_FUNCTION_MODE_OFFSET,
+			IS_E1HMF(bp));
+		REG_WR8(bp, BAR_USTRORM_INTMEM + USTORM_FUNCTION_MODE_OFFSET,
+			IS_E1HMF(bp));
+
+		REG_WR16(bp, BAR_XSTRORM_INTMEM +
+			 XSTORM_E1HOV_OFFSET(func), bp->e1hov);
+	}
+
+	/* Zero this manualy as its initialization is
+	   currently missing in the initTool */
+	for (i = 0; i < USTORM_AGG_DATA_SIZE >> 2; i++)
+		REG_WR(bp, BAR_USTRORM_INTMEM +
+		       USTORM_AGG_DATA_OFFSET + 4*i, 0);
 }
 
 static void bnx2x_nic_init(struct bnx2x *bp)
@@ -3318,15 +3784,20 @@ static void bnx2x_nic_init(struct bnx2x *bp)
 	for_each_queue(bp, i) {
 		struct bnx2x_fastpath *fp = &bp->fp[i];
 
+		fp->bp = bp;
 		fp->state = BNX2X_FP_STATE_CLOSED;
-		DP(NETIF_MSG_IFUP, "bnx2x_init_sb(%p,%p,%d);\n",
-		   bp, fp->status_blk, i);
 		fp->index = i;
-		bnx2x_init_sb(bp, fp->status_blk, fp->status_blk_mapping, i);
+		fp->cl_id = BP_L_ID(bp) + i;
+		fp->sb_id = fp->cl_id;
+		DP(NETIF_MSG_IFUP,
+		   "bnx2x_init_sb(%p,%p) index %d  cl_id %d  sb %d\n",
+		   bp, fp->status_blk, i, FP_CL_ID(fp), FP_SB_ID(fp));
+		bnx2x_init_sb(bp, FP_SB_ID(fp), fp->status_blk,
+			      fp->status_blk_mapping);
 	}
 
 	bnx2x_init_def_sb(bp, bp->def_status_blk,
-			  bp->def_status_blk_mapping, 0x10);
+			  bp->def_status_blk_mapping, DEF_SB_ID);
 	bnx2x_update_coalesce(bp);
 	bnx2x_init_rx_rings(bp);
 	bnx2x_init_tx_ring(bp);
@@ -3336,7 +3807,6 @@ static void bnx2x_nic_init(struct bnx2x *bp)
 	bnx2x_init_stats(bp);
 	bnx2x_init_ind_table(bp);
 	bnx2x_int_enable(bp);
-
 }
 
 /* end of nic init */
@@ -3374,7 +3844,7 @@ gunzip_nomem2:
 
 gunzip_nomem1:
 	printk(KERN_ERR PFX "%s: Cannot allocate firmware buffer for"
-	       " uncompression\n", bp->dev->name);
+	       " un-compression\n", bp->dev->name);
 	return -ENOMEM;
 }
 
@@ -3402,7 +3872,7 @@ static int bnx2x_gunzip(struct bnx2x *bp, u8 *zbuf, int len)
 
 	n = 10;
 
-#define FNAME   			0x8
+#define FNAME				0x8
 
 	if (zbuf[3] & FNAME)
 		while ((zbuf[n++] != 0) && (n < len));
@@ -3439,41 +3909,25 @@ static int bnx2x_gunzip(struct bnx2x *bp, u8 *zbuf, int len)
 /* nic load/unload */
 
 /*
- * general service functions
+ * General service functions
  */
 
 /* send a NIG loopback debug packet */
 static void bnx2x_lb_pckt(struct bnx2x *bp)
 {
-#ifdef USE_DMAE
 	u32 wb_write[3];
-#endif
 
 	/* Ethernet source and destination addresses */
-#ifdef USE_DMAE
 	wb_write[0] = 0x55555555;
 	wb_write[1] = 0x55555555;
-	wb_write[2] = 0x20;     	/* SOP */
+	wb_write[2] = 0x20;		/* SOP */
 	REG_WR_DMAE(bp, NIG_REG_DEBUG_PACKET_LB, wb_write, 3);
-#else
-	REG_WR_IND(bp, NIG_REG_DEBUG_PACKET_LB, 0x55555555);
-	REG_WR_IND(bp, NIG_REG_DEBUG_PACKET_LB + 4, 0x55555555);
-	/* SOP */
-	REG_WR_IND(bp, NIG_REG_DEBUG_PACKET_LB + 8, 0x20);
-#endif
 
 	/* NON-IP protocol */
-#ifdef USE_DMAE
 	wb_write[0] = 0x09000000;
 	wb_write[1] = 0x55555555;
-	wb_write[2] = 0x10;     	/* EOP, eop_bvalid = 0 */
+	wb_write[2] = 0x10;		/* EOP, eop_bvalid = 0 */
 	REG_WR_DMAE(bp, NIG_REG_DEBUG_PACKET_LB, wb_write, 3);
-#else
-	REG_WR_IND(bp, NIG_REG_DEBUG_PACKET_LB, 0x09000000);
-	REG_WR_IND(bp, NIG_REG_DEBUG_PACKET_LB + 4, 0x55555555);
-	/* EOP, eop_bvalid = 0 */
-	REG_WR_IND(bp, NIG_REG_DEBUG_PACKET_LB + 8, 0x10);
-#endif
 }
 
 /* some of the internal memories
@@ -3511,13 +3965,9 @@ static int bnx2x_int_mem_test(struct bnx2x *bp)
 	/* Wait until NIG register shows 1 packet of size 0x10 */
 	count = 1000 * factor;
 	while (count) {
-#ifdef BNX2X_DMAE_RD
+
 		bnx2x_read_dmae(bp, NIG_REG_STAT2_BRB_OCTET, 2);
 		val = *bnx2x_sp(bp, wb_data[0]);
-#else
-		val = REG_RD(bp, NIG_REG_STAT2_BRB_OCTET);
-		REG_RD(bp, NIG_REG_STAT2_BRB_OCTET + 4);
-#endif
 		if (val == 0x10)
 			break;
 
@@ -3533,7 +3983,6 @@ static int bnx2x_int_mem_test(struct bnx2x *bp)
 	count = 1000 * factor;
 	while (count) {
 		val = REG_RD(bp, PRS_REG_NUM_OF_PACKETS);
-
 		if (val == 1)
 			break;
 
@@ -3546,9 +3995,9 @@ static int bnx2x_int_mem_test(struct bnx2x *bp)
 	}
 
 	/* Reset and init BRB, PRS */
-	REG_WR(bp, GRCBASE_MISC + MISC_REGISTERS_RESET_REG_1_CLEAR, 0x3);
+	REG_WR(bp, GRCBASE_MISC + MISC_REGISTERS_RESET_REG_1_CLEAR, 0x03);
 	msleep(50);
-	REG_WR(bp, GRCBASE_MISC + MISC_REGISTERS_RESET_REG_1_SET, 0x3);
+	REG_WR(bp, GRCBASE_MISC + MISC_REGISTERS_RESET_REG_1_SET, 0x03);
 	msleep(50);
 	bnx2x_init_block(bp, BRB1_COMMON_START, BRB1_COMMON_END);
 	bnx2x_init_block(bp, PRS_COMMON_START, PRS_COMMON_END);
@@ -3572,13 +4021,9 @@ static int bnx2x_int_mem_test(struct bnx2x *bp)
 	   packets of size 11*0x10 = 0xb0 */
 	count = 1000 * factor;
 	while (count) {
-#ifdef BNX2X_DMAE_RD
+
 		bnx2x_read_dmae(bp, NIG_REG_STAT2_BRB_OCTET, 2);
 		val = *bnx2x_sp(bp, wb_data[0]);
-#else
-		val = REG_RD(bp, NIG_REG_STAT2_BRB_OCTET);
-		REG_RD(bp, NIG_REG_STAT2_BRB_OCTET + 4);
-#endif
 		if (val == 0xb0)
 			break;
 
@@ -3648,85 +4093,75 @@ static void enable_blocks_attention(struct bnx2x *bp)
 	REG_WR(bp, XSDM_REG_XSDM_INT_MASK_0, 0);
 	REG_WR(bp, XSDM_REG_XSDM_INT_MASK_1, 0);
 	REG_WR(bp, XCM_REG_XCM_INT_MASK, 0);
-/*      REG_WR(bp, XSEM_REG_XSEM_INT_MASK_0, 0); */
-/*      REG_WR(bp, XSEM_REG_XSEM_INT_MASK_1, 0); */
+/*	REG_WR(bp, XSEM_REG_XSEM_INT_MASK_0, 0); */
+/*	REG_WR(bp, XSEM_REG_XSEM_INT_MASK_1, 0); */
 	REG_WR(bp, USDM_REG_USDM_INT_MASK_0, 0);
 	REG_WR(bp, USDM_REG_USDM_INT_MASK_1, 0);
 	REG_WR(bp, UCM_REG_UCM_INT_MASK, 0);
-/*      REG_WR(bp, USEM_REG_USEM_INT_MASK_0, 0); */
-/*      REG_WR(bp, USEM_REG_USEM_INT_MASK_1, 0); */
+/*	REG_WR(bp, USEM_REG_USEM_INT_MASK_0, 0); */
+/*	REG_WR(bp, USEM_REG_USEM_INT_MASK_1, 0); */
 	REG_WR(bp, GRCBASE_UPB + PB_REG_PB_INT_MASK, 0);
 	REG_WR(bp, CSDM_REG_CSDM_INT_MASK_0, 0);
 	REG_WR(bp, CSDM_REG_CSDM_INT_MASK_1, 0);
 	REG_WR(bp, CCM_REG_CCM_INT_MASK, 0);
-/*      REG_WR(bp, CSEM_REG_CSEM_INT_MASK_0, 0); */
-/*      REG_WR(bp, CSEM_REG_CSEM_INT_MASK_1, 0); */
-	REG_WR(bp, PXP2_REG_PXP2_INT_MASK, 0x480000);
+/*	REG_WR(bp, CSEM_REG_CSEM_INT_MASK_0, 0); */
+/*	REG_WR(bp, CSEM_REG_CSEM_INT_MASK_1, 0); */
+	if (CHIP_REV_IS_FPGA(bp))
+		REG_WR(bp, PXP2_REG_PXP2_INT_MASK_0, 0x580000);
+	else
+		REG_WR(bp, PXP2_REG_PXP2_INT_MASK_0, 0x480000);
 	REG_WR(bp, TSDM_REG_TSDM_INT_MASK_0, 0);
 	REG_WR(bp, TSDM_REG_TSDM_INT_MASK_1, 0);
 	REG_WR(bp, TCM_REG_TCM_INT_MASK, 0);
-/*      REG_WR(bp, TSEM_REG_TSEM_INT_MASK_0, 0); */
-/*      REG_WR(bp, TSEM_REG_TSEM_INT_MASK_1, 0); */
+/*	REG_WR(bp, TSEM_REG_TSEM_INT_MASK_0, 0); */
+/*	REG_WR(bp, TSEM_REG_TSEM_INT_MASK_1, 0); */
 	REG_WR(bp, CDU_REG_CDU_INT_MASK, 0);
 	REG_WR(bp, DMAE_REG_DMAE_INT_MASK, 0);
-/*      REG_WR(bp, MISC_REG_MISC_INT_MASK, 0); */
-	REG_WR(bp, PBF_REG_PBF_INT_MASK, 0X18); 	/* bit 3,4 masked */
+/*	REG_WR(bp, MISC_REG_MISC_INT_MASK, 0); */
+	REG_WR(bp, PBF_REG_PBF_INT_MASK, 0X18);		/* bit 3,4 masked */
 }
 
-static int bnx2x_function_init(struct bnx2x *bp, int mode)
+
+static int bnx2x_init_common(struct bnx2x *bp)
 {
-	int func = bp->port;
-	int port = func ? PORT1 : PORT0;
 	u32 val, i;
-#ifdef USE_DMAE
-	u32 wb_write[2];
-#endif
-
-	DP(BNX2X_MSG_MCP, "function is %d  mode is %x\n", func, mode);
-	if ((func != 0) && (func != 1)) {
-		BNX2X_ERR("BAD function number (%d)\n", func);
-		return -ENODEV;
-	}
 
-	bnx2x_gunzip_init(bp);
+	DP(BNX2X_MSG_MCP, "starting common init  func %d\n", BP_FUNC(bp));
 
-	if (mode & 0x1) {       /* init common */
-		DP(BNX2X_MSG_MCP, "starting common init  func %d  mode %x\n",
-		   func, mode);
-		REG_WR(bp, GRCBASE_MISC + MISC_REGISTERS_RESET_REG_1_SET,
-		       0xffffffff);
-		REG_WR(bp, GRCBASE_MISC + MISC_REGISTERS_RESET_REG_2_SET,
-		       0xfffc);
-		bnx2x_init_block(bp, MISC_COMMON_START, MISC_COMMON_END);
+	REG_WR(bp, GRCBASE_MISC + MISC_REGISTERS_RESET_REG_1_SET, 0xffffffff);
+	REG_WR(bp, GRCBASE_MISC + MISC_REGISTERS_RESET_REG_2_SET, 0xfffc);
 
-		REG_WR(bp, MISC_REG_LCPLL_CTRL_REG_2, 0x100);
-		msleep(30);
-		REG_WR(bp, MISC_REG_LCPLL_CTRL_REG_2, 0x0);
+	bnx2x_init_block(bp, MISC_COMMON_START, MISC_COMMON_END);
+	if (CHIP_IS_E1H(bp))
+		REG_WR(bp, MISC_REG_E1HMF_MODE, IS_E1HMF(bp));
 
-		bnx2x_init_block(bp, PXP_COMMON_START, PXP_COMMON_END);
-		bnx2x_init_block(bp, PXP2_COMMON_START, PXP2_COMMON_END);
+	REG_WR(bp, MISC_REG_LCPLL_CTRL_REG_2, 0x100);
+	msleep(30);
+	REG_WR(bp, MISC_REG_LCPLL_CTRL_REG_2, 0x0);
 
-		bnx2x_init_pxp(bp);
+	bnx2x_init_block(bp, PXP_COMMON_START, PXP_COMMON_END);
+	if (CHIP_IS_E1(bp)) {
+		/* enable HW interrupt from PXP on USDM overflow
+		   bit 16 on INT_MASK_0 */
+		REG_WR(bp, PXP_REG_PXP_INT_MASK_0, 0);
+	}
 
-		if (CHIP_REV(bp) == CHIP_REV_Ax) {
-			/* enable HW interrupt from PXP on USDM
-			   overflow bit 16 on INT_MASK_0 */
-			REG_WR(bp, PXP_REG_PXP_INT_MASK_0, 0);
-		}
+	bnx2x_init_block(bp, PXP2_COMMON_START, PXP2_COMMON_END);
+	bnx2x_init_pxp(bp);
 
 #ifdef __BIG_ENDIAN
-		REG_WR(bp, PXP2_REG_RQ_QM_ENDIAN_M, 1);
-		REG_WR(bp, PXP2_REG_RQ_TM_ENDIAN_M, 1);
-		REG_WR(bp, PXP2_REG_RQ_SRC_ENDIAN_M, 1);
-		REG_WR(bp, PXP2_REG_RQ_CDU_ENDIAN_M, 1);
-		REG_WR(bp, PXP2_REG_RQ_DBG_ENDIAN_M, 1);
-		REG_WR(bp, PXP2_REG_RQ_HC_ENDIAN_M, 1);
-
-/*      	REG_WR(bp, PXP2_REG_RD_PBF_SWAP_MODE, 1); */
-		REG_WR(bp, PXP2_REG_RD_QM_SWAP_MODE, 1);
-		REG_WR(bp, PXP2_REG_RD_TM_SWAP_MODE, 1);
-		REG_WR(bp, PXP2_REG_RD_SRC_SWAP_MODE, 1);
-		REG_WR(bp, PXP2_REG_RD_CDURD_SWAP_MODE, 1);
+	REG_WR(bp, PXP2_REG_RQ_QM_ENDIAN_M, 1);
+	REG_WR(bp, PXP2_REG_RQ_TM_ENDIAN_M, 1);
+	REG_WR(bp, PXP2_REG_RQ_SRC_ENDIAN_M, 1);
+	REG_WR(bp, PXP2_REG_RQ_CDU_ENDIAN_M, 1);
+	REG_WR(bp, PXP2_REG_RQ_DBG_ENDIAN_M, 1);
+	REG_WR(bp, PXP2_REG_RQ_HC_ENDIAN_M, 1);
+
+/*	REG_WR(bp, PXP2_REG_RD_PBF_SWAP_MODE, 1); */
+	REG_WR(bp, PXP2_REG_RD_QM_SWAP_MODE, 1);
+	REG_WR(bp, PXP2_REG_RD_TM_SWAP_MODE, 1);
+	REG_WR(bp, PXP2_REG_RD_SRC_SWAP_MODE, 1);
+	REG_WR(bp, PXP2_REG_RD_CDURD_SWAP_MODE, 1);
 #endif
 
 #ifndef BCM_ISCSI
@@ -3734,92 +4169,105 @@ static int bnx2x_function_init(struct bnx2x *bp, int mode)
 		REG_WR(bp, PRS_REG_NIC_MODE, 1);
 #endif
 
-		REG_WR(bp, PXP2_REG_RQ_CDU_P_SIZE, 5);
+	REG_WR(bp, PXP2_REG_RQ_CDU_P_SIZE, 2);
 #ifdef BCM_ISCSI
-		REG_WR(bp, PXP2_REG_RQ_TM_P_SIZE, 5);
-		REG_WR(bp, PXP2_REG_RQ_QM_P_SIZE, 5);
-		REG_WR(bp, PXP2_REG_RQ_SRC_P_SIZE, 5);
+	REG_WR(bp, PXP2_REG_RQ_TM_P_SIZE, 5);
+	REG_WR(bp, PXP2_REG_RQ_QM_P_SIZE, 5);
+	REG_WR(bp, PXP2_REG_RQ_SRC_P_SIZE, 5);
 #endif
 
-		bnx2x_init_block(bp, DMAE_COMMON_START, DMAE_COMMON_END);
+	if (CHIP_REV_IS_FPGA(bp) && CHIP_IS_E1H(bp))
+		REG_WR(bp, PXP2_REG_PGL_TAGS_LIMIT, 0x1);
 
-		/* let the HW do it's magic ... */
-		msleep(100);
-		/* finish PXP init
-		   (can be moved up if we want to use the DMAE) */
-		val = REG_RD(bp, PXP2_REG_RQ_CFG_DONE);
-		if (val != 1) {
-			BNX2X_ERR("PXP2 CFG failed\n");
-			return -EBUSY;
-		}
+	/* let the HW do it's magic ... */
+	msleep(100);
+	/* finish PXP init */
+	val = REG_RD(bp, PXP2_REG_RQ_CFG_DONE);
+	if (val != 1) {
+		BNX2X_ERR("PXP2 CFG failed\n");
+		return -EBUSY;
+	}
+	val = REG_RD(bp, PXP2_REG_RD_INIT_DONE);
+	if (val != 1) {
+		BNX2X_ERR("PXP2 RD_INIT failed\n");
+		return -EBUSY;
+	}
 
-		val = REG_RD(bp, PXP2_REG_RD_INIT_DONE);
-		if (val != 1) {
-			BNX2X_ERR("PXP2 RD_INIT failed\n");
-			return -EBUSY;
-		}
+	REG_WR(bp, PXP2_REG_RQ_DISABLE_INPUTS, 0);
+	REG_WR(bp, PXP2_REG_RD_DISABLE_INPUTS, 0);
 
-		REG_WR(bp, PXP2_REG_RQ_DISABLE_INPUTS, 0);
-		REG_WR(bp, PXP2_REG_RD_DISABLE_INPUTS, 0);
+	bnx2x_init_block(bp, DMAE_COMMON_START, DMAE_COMMON_END);
 
-		bnx2x_init_fill(bp, TSEM_REG_PRAM, 0, 8);
+	/* clean the DMAE memory */
+	bp->dmae_ready = 1;
+	bnx2x_init_fill(bp, TSEM_REG_PRAM, 0, 8);
 
-		bnx2x_init_block(bp, TCM_COMMON_START, TCM_COMMON_END);
-		bnx2x_init_block(bp, UCM_COMMON_START, UCM_COMMON_END);
-		bnx2x_init_block(bp, CCM_COMMON_START, CCM_COMMON_END);
-		bnx2x_init_block(bp, XCM_COMMON_START, XCM_COMMON_END);
+	bnx2x_init_block(bp, TCM_COMMON_START, TCM_COMMON_END);
+	bnx2x_init_block(bp, UCM_COMMON_START, UCM_COMMON_END);
+	bnx2x_init_block(bp, CCM_COMMON_START, CCM_COMMON_END);
+	bnx2x_init_block(bp, XCM_COMMON_START, XCM_COMMON_END);
 
-#ifdef BNX2X_DMAE_RD
-		bnx2x_read_dmae(bp, XSEM_REG_PASSIVE_BUFFER, 3);
-		bnx2x_read_dmae(bp, CSEM_REG_PASSIVE_BUFFER, 3);
-		bnx2x_read_dmae(bp, TSEM_REG_PASSIVE_BUFFER, 3);
-		bnx2x_read_dmae(bp, USEM_REG_PASSIVE_BUFFER, 3);
-#else
-		REG_RD(bp, XSEM_REG_PASSIVE_BUFFER);
-		REG_RD(bp, XSEM_REG_PASSIVE_BUFFER + 4);
-		REG_RD(bp, XSEM_REG_PASSIVE_BUFFER + 8);
-		REG_RD(bp, CSEM_REG_PASSIVE_BUFFER);
-		REG_RD(bp, CSEM_REG_PASSIVE_BUFFER + 4);
-		REG_RD(bp, CSEM_REG_PASSIVE_BUFFER + 8);
-		REG_RD(bp, TSEM_REG_PASSIVE_BUFFER);
-		REG_RD(bp, TSEM_REG_PASSIVE_BUFFER + 4);
-		REG_RD(bp, TSEM_REG_PASSIVE_BUFFER + 8);
-		REG_RD(bp, USEM_REG_PASSIVE_BUFFER);
-		REG_RD(bp, USEM_REG_PASSIVE_BUFFER + 4);
-		REG_RD(bp, USEM_REG_PASSIVE_BUFFER + 8);
-#endif
-		bnx2x_init_block(bp, QM_COMMON_START, QM_COMMON_END);
-		/* soft reset pulse */
-		REG_WR(bp, QM_REG_SOFT_RESET, 1);
-		REG_WR(bp, QM_REG_SOFT_RESET, 0);
+	bnx2x_read_dmae(bp, XSEM_REG_PASSIVE_BUFFER, 3);
+	bnx2x_read_dmae(bp, CSEM_REG_PASSIVE_BUFFER, 3);
+	bnx2x_read_dmae(bp, TSEM_REG_PASSIVE_BUFFER, 3);
+	bnx2x_read_dmae(bp, USEM_REG_PASSIVE_BUFFER, 3);
+
+	bnx2x_init_block(bp, QM_COMMON_START, QM_COMMON_END);
+	/* soft reset pulse */
+	REG_WR(bp, QM_REG_SOFT_RESET, 1);
+	REG_WR(bp, QM_REG_SOFT_RESET, 0);
 
 #ifdef BCM_ISCSI
-		bnx2x_init_block(bp, TIMERS_COMMON_START, TIMERS_COMMON_END);
+	bnx2x_init_block(bp, TIMERS_COMMON_START, TIMERS_COMMON_END);
 #endif
-		bnx2x_init_block(bp, DQ_COMMON_START, DQ_COMMON_END);
-		REG_WR(bp, DORQ_REG_DPM_CID_OFST, BCM_PAGE_BITS);
-		if (CHIP_REV(bp) == CHIP_REV_Ax) {
-			/* enable hw interrupt from doorbell Q */
-			REG_WR(bp, DORQ_REG_DORQ_INT_MASK, 0);
-		}
 
-		bnx2x_init_block(bp, BRB1_COMMON_START, BRB1_COMMON_END);
+	bnx2x_init_block(bp, DQ_COMMON_START, DQ_COMMON_END);
+	REG_WR(bp, DORQ_REG_DPM_CID_OFST, BCM_PAGE_SHIFT);
+	if (!CHIP_REV_IS_SLOW(bp)) {
+		/* enable hw interrupt from doorbell Q */
+		REG_WR(bp, DORQ_REG_DORQ_INT_MASK, 0);
+	}
 
-		if (CHIP_REV_IS_SLOW(bp)) {
-			/* fix for emulation and FPGA for no pause */
-			REG_WR(bp, BRB1_REG_PAUSE_HIGH_THRESHOLD_0, 513);
-			REG_WR(bp, BRB1_REG_PAUSE_HIGH_THRESHOLD_1, 513);
-			REG_WR(bp, BRB1_REG_PAUSE_LOW_THRESHOLD_0, 0);
-			REG_WR(bp, BRB1_REG_PAUSE_LOW_THRESHOLD_1, 0);
-		}
+	bnx2x_init_block(bp, BRB1_COMMON_START, BRB1_COMMON_END);
+	if (CHIP_REV_IS_SLOW(bp)) {
+		/* fix for emulation and FPGA for no pause */
+		REG_WR(bp, BRB1_REG_PAUSE_HIGH_THRESHOLD_0, 513);
+		REG_WR(bp, BRB1_REG_PAUSE_HIGH_THRESHOLD_1, 513);
+		REG_WR(bp, BRB1_REG_PAUSE_LOW_THRESHOLD_0, 0);
+		REG_WR(bp, BRB1_REG_PAUSE_LOW_THRESHOLD_1, 0);
+	}
 
-		bnx2x_init_block(bp, PRS_COMMON_START, PRS_COMMON_END);
+	bnx2x_init_block(bp, PRS_COMMON_START, PRS_COMMON_END);
+	if (CHIP_IS_E1H(bp))
+		REG_WR(bp, PRS_REG_E1HOV_MODE, IS_E1HMF(bp));
 
-		bnx2x_init_block(bp, TSDM_COMMON_START, TSDM_COMMON_END);
-		bnx2x_init_block(bp, CSDM_COMMON_START, CSDM_COMMON_END);
-		bnx2x_init_block(bp, USDM_COMMON_START, USDM_COMMON_END);
-		bnx2x_init_block(bp, XSDM_COMMON_START, XSDM_COMMON_END);
+	bnx2x_init_block(bp, TSDM_COMMON_START, TSDM_COMMON_END);
+	bnx2x_init_block(bp, CSDM_COMMON_START, CSDM_COMMON_END);
+	bnx2x_init_block(bp, USDM_COMMON_START, USDM_COMMON_END);
+	bnx2x_init_block(bp, XSDM_COMMON_START, XSDM_COMMON_END);
 
+	if (CHIP_IS_E1H(bp)) {
+		bnx2x_init_fill(bp, TSTORM_INTMEM_ADDR, 0,
+				STORM_INTMEM_SIZE_E1H/2);
+		bnx2x_init_fill(bp,
+				TSTORM_INTMEM_ADDR + STORM_INTMEM_SIZE_E1H/2,
+				0, STORM_INTMEM_SIZE_E1H/2);
+		bnx2x_init_fill(bp, CSTORM_INTMEM_ADDR, 0,
+				STORM_INTMEM_SIZE_E1H/2);
+		bnx2x_init_fill(bp,
+				CSTORM_INTMEM_ADDR + STORM_INTMEM_SIZE_E1H/2,
+				0, STORM_INTMEM_SIZE_E1H/2);
+		bnx2x_init_fill(bp, XSTORM_INTMEM_ADDR, 0,
+				STORM_INTMEM_SIZE_E1H/2);
+		bnx2x_init_fill(bp,
+				XSTORM_INTMEM_ADDR + STORM_INTMEM_SIZE_E1H/2,
+				0, STORM_INTMEM_SIZE_E1H/2);
+		bnx2x_init_fill(bp, USTORM_INTMEM_ADDR, 0,
+				STORM_INTMEM_SIZE_E1H/2);
+		bnx2x_init_fill(bp,
+				USTORM_INTMEM_ADDR + STORM_INTMEM_SIZE_E1H/2,
+				0, STORM_INTMEM_SIZE_E1H/2);
+	} else { /* E1 */
 		bnx2x_init_fill(bp, TSTORM_INTMEM_ADDR, 0,
 				STORM_INTMEM_SIZE_E1);
 		bnx2x_init_fill(bp, CSTORM_INTMEM_ADDR, 0,
@@ -3828,157 +4276,141 @@ static int bnx2x_function_init(struct bnx2x *bp, int mode)
 				STORM_INTMEM_SIZE_E1);
 		bnx2x_init_fill(bp, USTORM_INTMEM_ADDR, 0,
 				STORM_INTMEM_SIZE_E1);
+	}
 
-		bnx2x_init_block(bp, TSEM_COMMON_START, TSEM_COMMON_END);
-		bnx2x_init_block(bp, USEM_COMMON_START, USEM_COMMON_END);
-		bnx2x_init_block(bp, CSEM_COMMON_START, CSEM_COMMON_END);
-		bnx2x_init_block(bp, XSEM_COMMON_START, XSEM_COMMON_END);
-
-		/* sync semi rtc */
-		REG_WR(bp, GRCBASE_MISC + MISC_REGISTERS_RESET_REG_1_CLEAR,
-		       0x80000000);
-		REG_WR(bp, GRCBASE_MISC + MISC_REGISTERS_RESET_REG_1_SET,
-		       0x80000000);
-
-		bnx2x_init_block(bp, UPB_COMMON_START, UPB_COMMON_END);
-		bnx2x_init_block(bp, XPB_COMMON_START, XPB_COMMON_END);
-		bnx2x_init_block(bp, PBF_COMMON_START, PBF_COMMON_END);
-
-		REG_WR(bp, SRC_REG_SOFT_RST, 1);
-		for (i = SRC_REG_KEYRSS0_0; i <= SRC_REG_KEYRSS1_9; i += 4) {
-			REG_WR(bp, i, 0xc0cac01a);
-			/* TODO: replace with something meaningful */
-		}
-		/* SRCH COMMON comes here */
-		REG_WR(bp, SRC_REG_SOFT_RST, 0);
-
-		if (sizeof(union cdu_context) != 1024) {
-			/* we currently assume that a context is 1024 bytes */
-			printk(KERN_ALERT PFX "please adjust the size of"
-			       " cdu_context(%ld)\n",
-			       (long)sizeof(union cdu_context));
-		}
-		val = (4 << 24) + (0 << 12) + 1024;
-		REG_WR(bp, CDU_REG_CDU_GLOBAL_PARAMS, val);
-		bnx2x_init_block(bp, CDU_COMMON_START, CDU_COMMON_END);
-
-		bnx2x_init_block(bp, CFC_COMMON_START, CFC_COMMON_END);
-		REG_WR(bp, CFC_REG_INIT_REG, 0x7FF);
-
-		bnx2x_init_block(bp, HC_COMMON_START, HC_COMMON_END);
-		bnx2x_init_block(bp, MISC_AEU_COMMON_START,
-				 MISC_AEU_COMMON_END);
-		/* RXPCS COMMON comes here */
-		/* EMAC0 COMMON comes here */
-		/* EMAC1 COMMON comes here */
-		/* DBU COMMON comes here */
-		/* DBG COMMON comes here */
-		bnx2x_init_block(bp, NIG_COMMON_START, NIG_COMMON_END);
+	bnx2x_init_block(bp, TSEM_COMMON_START, TSEM_COMMON_END);
+	bnx2x_init_block(bp, USEM_COMMON_START, USEM_COMMON_END);
+	bnx2x_init_block(bp, CSEM_COMMON_START, CSEM_COMMON_END);
+	bnx2x_init_block(bp, XSEM_COMMON_START, XSEM_COMMON_END);
 
-		if (CHIP_REV_IS_SLOW(bp))
-			msleep(200);
+	/* sync semi rtc */
+	REG_WR(bp, GRCBASE_MISC + MISC_REGISTERS_RESET_REG_1_CLEAR,
+	       0x80000000);
+	REG_WR(bp, GRCBASE_MISC + MISC_REGISTERS_RESET_REG_1_SET,
+	       0x80000000);
 
-		/* finish CFC init */
-		val = REG_RD(bp, CFC_REG_LL_INIT_DONE);
-		if (val != 1) {
-			BNX2X_ERR("CFC LL_INIT failed\n");
-			return -EBUSY;
-		}
+	bnx2x_init_block(bp, UPB_COMMON_START, UPB_COMMON_END);
+	bnx2x_init_block(bp, XPB_COMMON_START, XPB_COMMON_END);
+	bnx2x_init_block(bp, PBF_COMMON_START, PBF_COMMON_END);
 
-		val = REG_RD(bp, CFC_REG_AC_INIT_DONE);
-		if (val != 1) {
-			BNX2X_ERR("CFC AC_INIT failed\n");
-			return -EBUSY;
-		}
+	REG_WR(bp, SRC_REG_SOFT_RST, 1);
+	for (i = SRC_REG_KEYRSS0_0; i <= SRC_REG_KEYRSS1_9; i += 4) {
+		REG_WR(bp, i, 0xc0cac01a);
+		/* TODO: replace with something meaningful */
+	}
+	if (CHIP_IS_E1H(bp))
+		bnx2x_init_block(bp, SRCH_COMMON_START, SRCH_COMMON_END);
+	REG_WR(bp, SRC_REG_SOFT_RST, 0);
 
-		val = REG_RD(bp, CFC_REG_CAM_INIT_DONE);
-		if (val != 1) {
-			BNX2X_ERR("CFC CAM_INIT failed\n");
-			return -EBUSY;
-		}
+	if (sizeof(union cdu_context) != 1024)
+		/* we currently assume that a context is 1024 bytes */
+		printk(KERN_ALERT PFX "please adjust the size of"
+		       " cdu_context(%ld)\n", (long)sizeof(union cdu_context));
 
-		REG_WR(bp, CFC_REG_DEBUG0, 0);
+	bnx2x_init_block(bp, CDU_COMMON_START, CDU_COMMON_END);
+	val = (4 << 24) + (0 << 12) + 1024;
+	REG_WR(bp, CDU_REG_CDU_GLOBAL_PARAMS, val);
+	if (CHIP_IS_E1(bp)) {
+		/* !!! fix pxp client crdit until excel update */
+		REG_WR(bp, CDU_REG_CDU_DEBUG, 0x264);
+		REG_WR(bp, CDU_REG_CDU_DEBUG, 0);
+	}
 
-		/* read NIG statistic
-		   to see if this is our first up since powerup */
-#ifdef BNX2X_DMAE_RD
-		bnx2x_read_dmae(bp, NIG_REG_STAT2_BRB_OCTET, 2);
-		val = *bnx2x_sp(bp, wb_data[0]);
-#else
-		val = REG_RD(bp, NIG_REG_STAT2_BRB_OCTET);
-		REG_RD(bp, NIG_REG_STAT2_BRB_OCTET + 4);
-#endif
-		/* do internal memory self test */
-		if ((val == 0) && bnx2x_int_mem_test(bp)) {
-			BNX2X_ERR("internal mem selftest failed\n");
-			return -EBUSY;
-		}
+	bnx2x_init_block(bp, CFC_COMMON_START, CFC_COMMON_END);
+	REG_WR(bp, CFC_REG_INIT_REG, 0x7FF);
 
-		/* clear PXP2 attentions */
-		REG_RD(bp, PXP2_REG_PXP2_INT_STS_CLR);
+	bnx2x_init_block(bp, HC_COMMON_START, HC_COMMON_END);
+	bnx2x_init_block(bp, MISC_AEU_COMMON_START, MISC_AEU_COMMON_END);
 
-		enable_blocks_attention(bp);
-		/* enable_blocks_parity(bp); */
+	/* PXPCS COMMON comes here */
+	/* Reset PCIE errors for debug */
+	REG_WR(bp, 0x2814, 0xffffffff);
+	REG_WR(bp, 0x3820, 0xffffffff);
 
-		switch (bp->board & SHARED_HW_CFG_BOARD_TYPE_MASK) {
-		case SHARED_HW_CFG_BOARD_TYPE_BCM957710A1022G:
-			/* Fan failure is indicated by SPIO 5 */
-			bnx2x_set_spio(bp, MISC_REGISTERS_SPIO_5,
-				       MISC_REGISTERS_SPIO_INPUT_HI_Z);
+	/* EMAC0 COMMON comes here */
+	/* EMAC1 COMMON comes here */
+	/* DBU COMMON comes here */
+	/* DBG COMMON comes here */
+
+	bnx2x_init_block(bp, NIG_COMMON_START, NIG_COMMON_END);
+	if (CHIP_IS_E1H(bp)) {
+		REG_WR(bp, NIG_REG_LLH_MF_MODE, IS_E1HMF(bp));
+		REG_WR(bp, NIG_REG_LLH_E1HOV_MODE, IS_E1HMF(bp));
+	}
+
+	if (CHIP_REV_IS_SLOW(bp))
+		msleep(200);
+
+	/* finish CFC init */
+	val = reg_poll(bp, CFC_REG_LL_INIT_DONE, 1, 100, 10);
+	if (val != 1) {
+		BNX2X_ERR("CFC LL_INIT failed\n");
+		return -EBUSY;
+	}
+	val = reg_poll(bp, CFC_REG_AC_INIT_DONE, 1, 100, 10);
+	if (val != 1) {
+		BNX2X_ERR("CFC AC_INIT failed\n");
+		return -EBUSY;
+	}
+	val = reg_poll(bp, CFC_REG_CAM_INIT_DONE, 1, 100, 10);
+	if (val != 1) {
+		BNX2X_ERR("CFC CAM_INIT failed\n");
+		return -EBUSY;
+	}
+	REG_WR(bp, CFC_REG_DEBUG0, 0);
 
-			/* set to active low mode */
-			val = REG_RD(bp, MISC_REG_SPIO_INT);
-			val |= ((1 << MISC_REGISTERS_SPIO_5) <<
+	/* read NIG statistic
+	   to see if this is our first up since powerup */
+	bnx2x_read_dmae(bp, NIG_REG_STAT2_BRB_OCTET, 2);
+	val = *bnx2x_sp(bp, wb_data[0]);
+
+	/* do internal memory self test */
+	if ((CHIP_IS_E1(bp)) && (val == 0) && bnx2x_int_mem_test(bp)) {
+		BNX2X_ERR("internal mem self test failed\n");
+		return -EBUSY;
+	}
+
+	switch (bp->common.board & SHARED_HW_CFG_BOARD_TYPE_MASK) {
+	case SHARED_HW_CFG_BOARD_TYPE_BCM957710A1022G:
+		/* Fan failure is indicated by SPIO 5 */
+		bnx2x_set_spio(bp, MISC_REGISTERS_SPIO_5,
+			       MISC_REGISTERS_SPIO_INPUT_HI_Z);
+
+		/* set to active low mode */
+		val = REG_RD(bp, MISC_REG_SPIO_INT);
+		val |= ((1 << MISC_REGISTERS_SPIO_5) <<
 					MISC_REGISTERS_SPIO_INT_OLD_SET_POS);
-			REG_WR(bp, MISC_REG_SPIO_INT, val);
+		REG_WR(bp, MISC_REG_SPIO_INT, val);
 
-			/* enable interrupt to signal the IGU */
-			val = REG_RD(bp, MISC_REG_SPIO_EVENT_EN);
-			val |= (1 << MISC_REGISTERS_SPIO_5);
-			REG_WR(bp, MISC_REG_SPIO_EVENT_EN, val);
-			break;
+		/* enable interrupt to signal the IGU */
+		val = REG_RD(bp, MISC_REG_SPIO_EVENT_EN);
+		val |= (1 << MISC_REGISTERS_SPIO_5);
+		REG_WR(bp, MISC_REG_SPIO_EVENT_EN, val);
+		break;
 
-		default:
-			break;
-		}
+	default:
+		break;
+	}
 
-	} /* end of common init */
+	/* clear PXP2 attentions */
+	REG_RD(bp, PXP2_REG_PXP2_INT_STS_CLR_0);
 
-	/* per port init */
+	enable_blocks_attention(bp);
 
-	/* the phys address is shifted right 12 bits and has an added
-	   1=valid bit added to the 53rd bit
-	   then since this is a wide register(TM)
-	   we split it into two 32 bit writes
-	 */
-#define RQ_ONCHIP_AT_PORT_SIZE  384
-#define ONCHIP_ADDR1(x)   ((u32)(((u64)x >> 12) & 0xFFFFFFFF))
-#define ONCHIP_ADDR2(x)   ((u32)((1 << 20) | ((u64)x >> 44)))
-#define PXP_ONE_ILT(x)    ((x << 10) | x)
+	return 0;
+}
 
-	DP(BNX2X_MSG_MCP, "starting per-function init port is %x\n", func);
+static int bnx2x_init_port(struct bnx2x *bp)
+{
+	int port = BP_PORT(bp);
+	u32 val;
 
-	REG_WR(bp, NIG_REG_MASK_INTERRUPT_PORT0 + func*4, 0);
+	DP(BNX2X_MSG_MCP, "starting port init  port %x\n", port);
+
+	REG_WR(bp, NIG_REG_MASK_INTERRUPT_PORT0 + port*4, 0);
 
 	/* Port PXP comes here */
 	/* Port PXP2 comes here */
-
-	/* Offset is
-	 * Port0  0
-	 * Port1  384 */
-	i = func * RQ_ONCHIP_AT_PORT_SIZE;
-#ifdef USE_DMAE
-	wb_write[0] = ONCHIP_ADDR1(bnx2x_sp_mapping(bp, context));
-	wb_write[1] = ONCHIP_ADDR2(bnx2x_sp_mapping(bp, context));
-	REG_WR_DMAE(bp, PXP2_REG_RQ_ONCHIP_AT + i*8, wb_write, 2);
-#else
-	REG_WR_IND(bp, PXP2_REG_RQ_ONCHIP_AT + i*8,
-		   ONCHIP_ADDR1(bnx2x_sp_mapping(bp, context)));
-	REG_WR_IND(bp, PXP2_REG_RQ_ONCHIP_AT + i*8 + 4,
-		   ONCHIP_ADDR2(bnx2x_sp_mapping(bp, context)));
-#endif
-	REG_WR(bp, PXP2_REG_PSWRQ_CDU0_L2P + func*4, PXP_ONE_ILT(i));
-
 #ifdef BCM_ISCSI
 	/* Port0  1
 	 * Port1  385 */
@@ -4004,30 +4436,9 @@ static int bnx2x_function_init(struct bnx2x *bp, int mode)
 	REG_WR_DMAE(bp, PXP2_REG_RQ_ONCHIP_AT + i*8, wb_write, 2);
 	REG_WR(bp, PXP2_REG_PSWRQ_SRC0_L2P + func*4, PXP_ONE_ILT(i));
 #endif
-
-	/* Port TCM comes here */
-	/* Port UCM comes here */
-	/* Port CCM comes here */
-	bnx2x_init_block(bp, func ? XCM_PORT1_START : XCM_PORT0_START,
-			     func ? XCM_PORT1_END : XCM_PORT0_END);
-
-#ifdef USE_DMAE
-	wb_write[0] = 0;
-	wb_write[1] = 0;
-#endif
-	for (i = 0; i < 32; i++) {
-		REG_WR(bp, QM_REG_BASEADDR + (func*32 + i)*4, 1024 * 4 * i);
-#ifdef USE_DMAE
-		REG_WR_DMAE(bp, QM_REG_PTRTBL + (func*32 + i)*8, wb_write, 2);
-#else
-		REG_WR_IND(bp, QM_REG_PTRTBL + (func*32 + i)*8, 0);
-		REG_WR_IND(bp, QM_REG_PTRTBL + (func*32 + i)*8 + 4, 0);
-#endif
-	}
-	REG_WR(bp, QM_REG_CONNNUM_0 + func*4, 1024/16 - 1);
+	/* Port CMs come here */
 
 	/* Port QM comes here */
-
 #ifdef BCM_ISCSI
 	REG_WR(bp, TM_REG_LIN0_SCAN_TIME + func*4, 1024/64*20);
 	REG_WR(bp, TM_REG_LIN0_MAX_ACTIVE_CID + func*4, 31);
@@ -4042,31 +4453,32 @@ static int bnx2x_function_init(struct bnx2x *bp, int mode)
 	/* Port CSDM comes here */
 	/* Port USDM comes here */
 	/* Port XSDM comes here */
-	bnx2x_init_block(bp, func ? TSEM_PORT1_START : TSEM_PORT0_START,
-			     func ? TSEM_PORT1_END : TSEM_PORT0_END);
-	bnx2x_init_block(bp, func ? USEM_PORT1_START : USEM_PORT0_START,
-			     func ? USEM_PORT1_END : USEM_PORT0_END);
-	bnx2x_init_block(bp, func ? CSEM_PORT1_START : CSEM_PORT0_START,
-			     func ? CSEM_PORT1_END : CSEM_PORT0_END);
-	bnx2x_init_block(bp, func ? XSEM_PORT1_START : XSEM_PORT0_START,
-			     func ? XSEM_PORT1_END : XSEM_PORT0_END);
+	bnx2x_init_block(bp, port ? TSEM_PORT1_START : TSEM_PORT0_START,
+			     port ? TSEM_PORT1_END : TSEM_PORT0_END);
+	bnx2x_init_block(bp, port ? USEM_PORT1_START : USEM_PORT0_START,
+			     port ? USEM_PORT1_END : USEM_PORT0_END);
+	bnx2x_init_block(bp, port ? CSEM_PORT1_START : CSEM_PORT0_START,
+			     port ? CSEM_PORT1_END : CSEM_PORT0_END);
+	bnx2x_init_block(bp, port ? XSEM_PORT1_START : XSEM_PORT0_START,
+			     port ? XSEM_PORT1_END : XSEM_PORT0_END);
 	/* Port UPB comes here */
-	/* Port XSDM comes here */
-	bnx2x_init_block(bp, func ? PBF_PORT1_START : PBF_PORT0_START,
-			     func ? PBF_PORT1_END : PBF_PORT0_END);
+	/* Port XPB comes here */
+
+	bnx2x_init_block(bp, port ? PBF_PORT1_START : PBF_PORT0_START,
+			     port ? PBF_PORT1_END : PBF_PORT0_END);
 
 	/* configure PBF to work without PAUSE mtu 9000 */
-	REG_WR(bp, PBF_REG_P0_PAUSE_ENABLE + func*4, 0);
+	REG_WR(bp, PBF_REG_P0_PAUSE_ENABLE + port*4, 0);
 
 	/* update threshold */
-	REG_WR(bp, PBF_REG_P0_ARB_THRSH + func*4, (9040/16));
+	REG_WR(bp, PBF_REG_P0_ARB_THRSH + port*4, (9040/16));
 	/* update init credit */
-	REG_WR(bp, PBF_REG_P0_INIT_CRD + func*4, (9040/16) + 553 - 22);
+	REG_WR(bp, PBF_REG_P0_INIT_CRD + port*4, (9040/16) + 553 - 22);
 
 	/* probe changes */
-	REG_WR(bp, PBF_REG_INIT_P0 + func*4, 1);
+	REG_WR(bp, PBF_REG_INIT_P0 + port*4, 1);
 	msleep(5);
-	REG_WR(bp, PBF_REG_INIT_P0 + func*4, 0);
+	REG_WR(bp, PBF_REG_INIT_P0 + port*4, 0);
 
 #ifdef BCM_ISCSI
 	/* tell the searcher where the T2 table is */
@@ -4084,23 +4496,57 @@ static int bnx2x_function_init(struct bnx2x *bp, int mode)
 #endif
 	/* Port CDU comes here */
 	/* Port CFC comes here */
-	bnx2x_init_block(bp, func ? HC_PORT1_START : HC_PORT0_START,
-			     func ? HC_PORT1_END : HC_PORT0_END);
-	bnx2x_init_block(bp, func ? MISC_AEU_PORT1_START :
+
+	if (CHIP_IS_E1(bp)) {
+		REG_WR(bp, HC_REG_LEADING_EDGE_0 + port*8, 0);
+		REG_WR(bp, HC_REG_TRAILING_EDGE_0 + port*8, 0);
+	}
+	bnx2x_init_block(bp, port ? HC_PORT1_START : HC_PORT0_START,
+			     port ? HC_PORT1_END : HC_PORT0_END);
+
+	bnx2x_init_block(bp, port ? MISC_AEU_PORT1_START :
 				    MISC_AEU_PORT0_START,
-			     func ? MISC_AEU_PORT1_END : MISC_AEU_PORT0_END);
+			     port ? MISC_AEU_PORT1_END : MISC_AEU_PORT0_END);
+	/* init aeu_mask_attn_func_0/1:
+	 *  - SF mode: bits 3-7 are masked. only bits 0-2 are in use
+	 *  - MF mode: bit 3 is masked. bits 0-2 are in use as in SF
+	 *             bits 4-7 are used for "per vn group attention" */
+	REG_WR(bp, MISC_REG_AEU_MASK_ATTN_FUNC_0 + port*4,
+	       (IS_E1HMF(bp) ? 0xF7 : 0x7));
+
 	/* Port PXPCS comes here */
 	/* Port EMAC0 comes here */
 	/* Port EMAC1 comes here */
 	/* Port DBU comes here */
 	/* Port DBG comes here */
-	bnx2x_init_block(bp, func ? NIG_PORT1_START : NIG_PORT0_START,
-			     func ? NIG_PORT1_END : NIG_PORT0_END);
-	REG_WR(bp, NIG_REG_XGXS_SERDES0_MODE_SEL + func*4, 1);
+	bnx2x_init_block(bp, port ? NIG_PORT1_START : NIG_PORT0_START,
+			     port ? NIG_PORT1_END : NIG_PORT0_END);
+
+	REG_WR(bp, NIG_REG_XGXS_SERDES0_MODE_SEL + port*4, 1);
+
+	if (CHIP_IS_E1H(bp)) {
+		u32 wsum;
+		struct cmng_struct_per_port m_cmng_port;
+		int vn;
+
+		/* 0x2 disable e1hov, 0x1 enable */
+		REG_WR(bp, NIG_REG_LLH0_BRB1_DRV_MASK_MF + port*4,
+		       (IS_E1HMF(bp) ? 0x1 : 0x2));
+
+		/* Init RATE SHAPING and FAIRNESS contexts.
+		   Initialize as if there is 10G link. */
+		wsum = bnx2x_calc_vn_wsum(bp);
+		bnx2x_init_port_minmax(bp, (int)wsum, 10000, &m_cmng_port);
+		if (IS_E1HMF(bp))
+			for (vn = VN_0; vn < E1HVN_MAX; vn++)
+				bnx2x_init_vn_minmax(bp, 2*vn + port,
+					wsum, 10000, &m_cmng_port);
+	}
+
 	/* Port MCP comes here */
 	/* Port DMAE comes here */
 
-	switch (bp->board & SHARED_HW_CFG_BOARD_TYPE_MASK) {
+	switch (bp->common.board & SHARED_HW_CFG_BOARD_TYPE_MASK) {
 	case SHARED_HW_CFG_BOARD_TYPE_BCM957710A1022G:
 		/* add SPIO 5 to group 0 */
 		val = REG_RD(bp, MISC_REG_AEU_ENABLE1_FUNC_0_OUT_0);
@@ -4114,48 +4560,150 @@ static int bnx2x_function_init(struct bnx2x *bp, int mode)
 
 	bnx2x__link_reset(bp);
 
+	return 0;
+}
+
+#define ILT_PER_FUNC		(768/2)
+#define FUNC_ILT_BASE(func)	(func * ILT_PER_FUNC)
+/* the phys address is shifted right 12 bits and has an added
+   1=valid bit added to the 53rd bit
+   then since this is a wide register(TM)
+   we split it into two 32 bit writes
+ */
+#define ONCHIP_ADDR1(x)		((u32)(((u64)x >> 12) & 0xFFFFFFFF))
+#define ONCHIP_ADDR2(x)		((u32)((1 << 20) | ((u64)x >> 44)))
+#define PXP_ONE_ILT(x)		(((x) << 10) | x)
+#define PXP_ILT_RANGE(f, l)	(((l) << 10) | f)
+
+#define CNIC_ILT_LINES		0
+
+static void bnx2x_ilt_wr(struct bnx2x *bp, u32 index, dma_addr_t addr)
+{
+	int reg;
+
+	if (CHIP_IS_E1H(bp))
+		reg = PXP2_REG_RQ_ONCHIP_AT_B0 + index*8;
+	else /* E1 */
+		reg = PXP2_REG_RQ_ONCHIP_AT + index*8;
+
+	bnx2x_wb_wr(bp, reg, ONCHIP_ADDR1(addr), ONCHIP_ADDR2(addr));
+}
+
+static int bnx2x_init_func(struct bnx2x *bp)
+{
+	int port = BP_PORT(bp);
+	int func = BP_FUNC(bp);
+	int i;
+
+	DP(BNX2X_MSG_MCP, "starting func init  func %x\n", func);
+
+	i = FUNC_ILT_BASE(func);
+
+	bnx2x_ilt_wr(bp, i, bnx2x_sp_mapping(bp, context));
+	if (CHIP_IS_E1H(bp)) {
+		REG_WR(bp, PXP2_REG_RQ_CDU_FIRST_ILT, i);
+		REG_WR(bp, PXP2_REG_RQ_CDU_LAST_ILT, i + CNIC_ILT_LINES);
+	} else /* E1 */
+		REG_WR(bp, PXP2_REG_PSWRQ_CDU0_L2P + func*4,
+		       PXP_ILT_RANGE(i, i + CNIC_ILT_LINES));
+
+
+	if (CHIP_IS_E1H(bp)) {
+		for (i = 0; i < 9; i++)
+			bnx2x_init_block(bp,
+					 cm_start[func][i], cm_end[func][i]);
+
+		REG_WR(bp, NIG_REG_LLH0_FUNC_EN + port*8, 1);
+		REG_WR(bp, NIG_REG_LLH0_FUNC_VLAN_ID + port*8, bp->e1hov);
+	}
+
+	/* HC init per function */
+	if (CHIP_IS_E1H(bp)) {
+		REG_WR(bp, MISC_REG_AEU_GENERAL_ATTN_12 + func*4, 0);
+
+		REG_WR(bp, HC_REG_LEADING_EDGE_0 + port*8, 0);
+		REG_WR(bp, HC_REG_TRAILING_EDGE_0 + port*8, 0);
+	}
+	bnx2x_init_block(bp, hc_limits[func][0], hc_limits[func][1]);
+
+	if (CHIP_IS_E1H(bp))
+		REG_WR(bp, HC_REG_FUNC_NUM_P0 + port*4, func);
+
 	/* Reset PCIE errors for debug */
 	REG_WR(bp, 0x2114, 0xffffffff);
 	REG_WR(bp, 0x2120, 0xffffffff);
-	REG_WR(bp, 0x2814, 0xffffffff);
 
-	/* !!! move to init_values.h */
-	REG_WR(bp, XSDM_REG_INIT_CREDIT_PXP_CTRL, 0x1);
-	REG_WR(bp, USDM_REG_INIT_CREDIT_PXP_CTRL, 0x1);
-	REG_WR(bp, CSDM_REG_INIT_CREDIT_PXP_CTRL, 0x1);
-	REG_WR(bp, TSDM_REG_INIT_CREDIT_PXP_CTRL, 0x1);
+	return 0;
+}
+
+static int bnx2x_init_hw(struct bnx2x *bp, u32 load_code)
+{
+	int i, rc = 0;
 
-	REG_WR(bp, DBG_REG_PCI_REQ_CREDIT, 0x1);
-	REG_WR(bp, TM_REG_PCIARB_CRDCNT_VAL, 0x1);
-	REG_WR(bp, CDU_REG_CDU_DEBUG, 0x264);
-	REG_WR(bp, CDU_REG_CDU_DEBUG, 0x0);
+	DP(BNX2X_MSG_MCP, "function %d  load_code %x\n",
+	   BP_FUNC(bp), load_code);
 
-	bnx2x_gunzip_end(bp);
+	bp->dmae_ready = 0;
+	mutex_init(&bp->dmae_mutex);
+	bnx2x_gunzip_init(bp);
 
-	if (!nomcp) {
-		port = bp->port;
+	switch (load_code) {
+	case FW_MSG_CODE_DRV_LOAD_COMMON:
+		rc = bnx2x_init_common(bp);
+		if (rc)
+			goto init_hw_err;
+		/* no break */
+
+	case FW_MSG_CODE_DRV_LOAD_PORT:
+		bp->dmae_ready = 1;
+		rc = bnx2x_init_port(bp);
+		if (rc)
+			goto init_hw_err;
+		/* no break */
+
+	case FW_MSG_CODE_DRV_LOAD_FUNCTION:
+		bp->dmae_ready = 1;
+		rc = bnx2x_init_func(bp);
+		if (rc)
+			goto init_hw_err;
+		break;
+
+	default:
+		BNX2X_ERR("Unknown load_code (0x%x) from MCP\n", load_code);
+		break;
+	}
+
+	if (!BP_NOMCP(bp)) {
+		int func = BP_FUNC(bp);
 
 		bp->fw_drv_pulse_wr_seq =
-				(SHMEM_RD(bp, func_mb[port].drv_pulse_mb) &
+				(SHMEM_RD(bp, func_mb[func].drv_pulse_mb) &
 				 DRV_PULSE_SEQ_MASK);
-		bp->fw_mb = SHMEM_RD(bp, func_mb[port].fw_mb_param);
-		DP(BNX2X_MSG_MCP, "drv_pulse 0x%x  fw_mb 0x%x\n",
-		   bp->fw_drv_pulse_wr_seq, bp->fw_mb);
-	} else {
-		bp->fw_mb = 0;
-	}
+		bp->func_stx = SHMEM_RD(bp, func_mb[func].fw_mb_param);
+		DP(BNX2X_MSG_MCP, "drv_pulse 0x%x  func_stx 0x%x\n",
+		   bp->fw_drv_pulse_wr_seq, bp->func_stx);
+	} else
+		bp->func_stx = 0;
 
-	return 0;
+	/* this needs to be done before gunzip end */
+	bnx2x_zero_def_sb(bp);
+	for_each_queue(bp, i)
+		bnx2x_zero_sb(bp, BP_L_ID(bp) + i);
+
+init_hw_err:
+	bnx2x_gunzip_end(bp);
+
+	return rc;
 }
 
 /* send the MCP a request, block until there is a reply */
 static u32 bnx2x_fw_command(struct bnx2x *bp, u32 command)
 {
-	int port = bp->port;
+	int func = BP_FUNC(bp);
 	u32 seq = ++bp->fw_seq;
 	u32 rc = 0;
 
-	SHMEM_WR(bp, func_mb[port].drv_mb_header, (command | seq));
+	SHMEM_WR(bp, func_mb[func].drv_mb_header, (command | seq));
 	DP(BNX2X_MSG_MCP, "wrote command (%x) to FW MB\n", (command | seq));
 
 	/* let the FW do it's magic ... */
@@ -4164,7 +4712,7 @@ static u32 bnx2x_fw_command(struct bnx2x *bp, u32 command)
 	if (CHIP_REV_IS_SLOW(bp))
 		msleep(900);
 
-	rc = SHMEM_RD(bp, func_mb[port].fw_mb_header);
+	rc = SHMEM_RD(bp, func_mb[func].fw_mb_header);
 	DP(BNX2X_MSG_MCP, "read (%x) seq is (%x) from FW MB\n", rc, seq);
 
 	/* is this a reply to our command? */
@@ -4229,15 +4777,13 @@ static void bnx2x_free_mem(struct bnx2x *bp)
 			       NUM_RCQ_BD);
 	}
 
-	BNX2X_FREE(bp->fp);
-
 	/* end of fastpath */
 
 	BNX2X_PCI_FREE(bp->def_status_blk, bp->def_status_blk_mapping,
-		       (sizeof(struct host_def_status_block)));
+		       sizeof(struct host_def_status_block));
 
 	BNX2X_PCI_FREE(bp->slowpath, bp->slowpath_mapping,
-		       (sizeof(struct bnx2x_slowpath)));
+		       sizeof(struct bnx2x_slowpath));
 
 #ifdef BCM_ISCSI
 	BNX2X_PCI_FREE(bp->t1, bp->t1_mapping, 64*1024);
@@ -4273,8 +4819,6 @@ static int bnx2x_alloc_mem(struct bnx2x *bp)
 	int i;
 
 	/* fastpath */
-	BNX2X_ALLOC(bp->fp, sizeof(struct bnx2x_fastpath) * bp->num_queues);
-
 	for_each_queue(bp, i) {
 		bnx2x_fp(bp, i, bp) = bp;
 
@@ -4370,8 +4914,6 @@ static void bnx2x_free_tx_skbs(struct bnx2x *bp)
 		u16 sw_prod = fp->tx_pkt_prod;
 		u16 sw_cons = fp->tx_pkt_cons;
 
-		BUG_TRAP(fp->tx_buf_ring != NULL);
-
 		while (sw_cons != sw_prod) {
 			bd_cons = bnx2x_free_tx_pkt(bp, fp, TX_BD(sw_cons));
 			sw_cons++;
@@ -4386,8 +4928,6 @@ static void bnx2x_free_rx_skbs(struct bnx2x *bp)
 	for_each_queue(bp, j) {
 		struct bnx2x_fastpath *fp = &bp->fp[j];
 
-		BUG_TRAP(fp->rx_buf_ring != NULL);
-
 		for (i = 0; i < NUM_RX_BD; i++) {
 			struct sw_rx_bd *rx_buf = &fp->rx_buf_ring[i];
 			struct sk_buff *skb = rx_buf->skb;
@@ -4414,7 +4954,7 @@ static void bnx2x_free_skbs(struct bnx2x *bp)
 
 static void bnx2x_free_msix_irqs(struct bnx2x *bp)
 {
-	int i;
+	int i, offset = 1;
 
 	free_irq(bp->msix_table[0].vector, bp->dev);
 	DP(NETIF_MSG_IFDOWN, "released sp irq (%d)\n",
@@ -4422,26 +4962,22 @@ static void bnx2x_free_msix_irqs(struct bnx2x *bp)
 
 	for_each_queue(bp, i) {
 		DP(NETIF_MSG_IFDOWN, "about to release fp #%d->%d irq  "
-		   "state(%x)\n", i, bp->msix_table[i + 1].vector,
+		   "state %x\n", i, bp->msix_table[i + offset].vector,
 		   bnx2x_fp(bp, i, state));
 
 		if (bnx2x_fp(bp, i, state) != BNX2X_FP_STATE_CLOSED)
 			BNX2X_ERR("IRQ of fp #%d being freed while "
 				  "state != closed\n", i);
 
-		free_irq(bp->msix_table[i + 1].vector, &bp->fp[i]);
+		free_irq(bp->msix_table[i + offset].vector, &bp->fp[i]);
 	}
-
 }
 
 static void bnx2x_free_irq(struct bnx2x *bp)
 {
-
 	if (bp->flags & USING_MSIX_FLAG) {
-
 		bnx2x_free_msix_irqs(bp);
 		pci_disable_msix(bp->pdev);
-
 		bp->flags &= ~USING_MSIX_FLAG;
 
 	} else
@@ -4450,87 +4986,87 @@ static void bnx2x_free_irq(struct bnx2x *bp)
 
 static int bnx2x_enable_msix(struct bnx2x *bp)
 {
-
-	int i;
+	int i, rc, offset;
 
 	bp->msix_table[0].entry = 0;
-	for_each_queue(bp, i)
-		bp->msix_table[i + 1].entry = i + 1;
+	offset = 1;
+	DP(NETIF_MSG_IFUP, "msix_table[0].entry = 0 (slowpath)\n");
 
-	if (pci_enable_msix(bp->pdev, &bp->msix_table[0],
-				     bp->num_queues + 1)){
-		BNX2X_LOG("failed to enable MSI-X\n");
-		return -1;
+	for_each_queue(bp, i) {
+		int igu_vec = offset + i + BP_L_ID(bp);
 
+		bp->msix_table[i + offset].entry = igu_vec;
+		DP(NETIF_MSG_IFUP, "msix_table[%d].entry = %d "
+		   "(fastpath #%u)\n", i + offset, igu_vec, i);
 	}
 
+	rc = pci_enable_msix(bp->pdev, &bp->msix_table[0],
+			     bp->num_queues + offset);
+	if (rc) {
+		DP(NETIF_MSG_IFUP, "MSI-X is not attainable\n");
+		return -1;
+	}
 	bp->flags |= USING_MSIX_FLAG;
 
 	return 0;
-
 }
 
-
 static int bnx2x_req_msix_irqs(struct bnx2x *bp)
 {
-
-	int i, rc;
+	int i, rc, offset = 1;
 
 	rc = request_irq(bp->msix_table[0].vector, bnx2x_msix_sp_int, 0,
 			 bp->dev->name, bp->dev);
-
 	if (rc) {
 		BNX2X_ERR("request sp irq failed\n");
 		return -EBUSY;
 	}
 
 	for_each_queue(bp, i) {
-		rc = request_irq(bp->msix_table[i + 1].vector,
+		rc = request_irq(bp->msix_table[i + offset].vector,
 				 bnx2x_msix_fp_int, 0,
 				 bp->dev->name, &bp->fp[i]);
-
 		if (rc) {
-			BNX2X_ERR("request fp #%d irq failed  "
-				  "rc %d\n", i, rc);
+			BNX2X_ERR("request fp #%d irq failed  rc %d\n",
+				  i + offset, rc);
 			bnx2x_free_msix_irqs(bp);
 			return -EBUSY;
 		}
 
 		bnx2x_fp(bp, i, state) = BNX2X_FP_STATE_IRQ;
-
 	}
 
 	return 0;
-
 }
 
 static int bnx2x_req_irq(struct bnx2x *bp)
 {
+	int rc;
 
-	int rc = request_irq(bp->pdev->irq, bnx2x_interrupt,
-			     IRQF_SHARED, bp->dev->name, bp->dev);
+	rc = request_irq(bp->pdev->irq, bnx2x_interrupt, IRQF_SHARED,
+			 bp->dev->name, bp->dev);
 	if (!rc)
 		bnx2x_fp(bp, 0, state) = BNX2X_FP_STATE_IRQ;
 
 	return rc;
-
 }
 
 /*
  * Init service functions
  */
 
-static void bnx2x_set_mac_addr(struct bnx2x *bp)
+static void bnx2x_set_mac_addr_e1(struct bnx2x *bp)
 {
 	struct mac_configuration_cmd *config = bnx2x_sp(bp, mac_config);
+	int port = BP_PORT(bp);
 
 	/* CAM allocation
 	 * unicasts 0-31:port0 32-63:port1
 	 * multicast 64-127:port0 128-191:port1
 	 */
 	config->hdr.length_6b = 2;
-	config->hdr.offset = bp->port ? 31 : 0;
-	config->hdr.reserved0 = 0;
+	config->hdr.offset = port ? 31 : 0;
+	config->hdr.client_id = BP_CL_ID(bp);
 	config->hdr.reserved1 = 0;
 
 	/* primary MAC */
@@ -4540,7 +5076,7 @@ static void bnx2x_set_mac_addr(struct bnx2x *bp)
 					swab16(*(u16 *)&bp->dev->dev_addr[2]);
 	config->config_table[0].cam_entry.lsb_mac_addr =
 					swab16(*(u16 *)&bp->dev->dev_addr[4]);
-	config->config_table[0].cam_entry.flags = cpu_to_le16(bp->port);
+	config->config_table[0].cam_entry.flags = cpu_to_le16(port);
 	config->config_table[0].target_table_entry.flags = 0;
 	config->config_table[0].target_table_entry.client_id = 0;
 	config->config_table[0].target_table_entry.vlan_id = 0;
@@ -4554,7 +5090,7 @@ static void bnx2x_set_mac_addr(struct bnx2x *bp)
 	config->config_table[1].cam_entry.msb_mac_addr = 0xffff;
 	config->config_table[1].cam_entry.middle_mac_addr = 0xffff;
 	config->config_table[1].cam_entry.lsb_mac_addr = 0xffff;
-	config->config_table[1].cam_entry.flags = cpu_to_le16(bp->port);
+	config->config_table[1].cam_entry.flags = cpu_to_le16(port);
 	config->config_table[1].target_table_entry.flags =
 				TSTORM_CAM_TARGET_TABLE_ENTRY_BROADCAST;
 	config->config_table[1].target_table_entry.client_id = 0;
@@ -4565,64 +5101,105 @@ static void bnx2x_set_mac_addr(struct bnx2x *bp)
 		      U64_LO(bnx2x_sp_mapping(bp, mac_config)), 0);
 }
 
+static void bnx2x_set_mac_addr_e1h(struct bnx2x *bp)
+{
+	struct mac_configuration_cmd_e1h *config =
+		(struct mac_configuration_cmd_e1h *)bnx2x_sp(bp, mac_config);
+
+	if (bp->state != BNX2X_STATE_OPEN) {
+		DP(NETIF_MSG_IFUP, "state is %x, returning\n", bp->state);
+		return;
+	}
+
+	/* CAM allocation for E1H
+	 * unicasts: by func number
+	 * multicast: 20+FUNC*20, 20 each
+	 */
+	config->hdr.length_6b = 1;
+	config->hdr.offset = BP_FUNC(bp);
+	config->hdr.client_id = BP_CL_ID(bp);
+	config->hdr.reserved1 = 0;
+
+	/* primary MAC */
+	config->config_table[0].msb_mac_addr =
+					swab16(*(u16 *)&bp->dev->dev_addr[0]);
+	config->config_table[0].middle_mac_addr =
+					swab16(*(u16 *)&bp->dev->dev_addr[2]);
+	config->config_table[0].lsb_mac_addr =
+					swab16(*(u16 *)&bp->dev->dev_addr[4]);
+	config->config_table[0].client_id = BP_L_ID(bp);
+	config->config_table[0].vlan_id = 0;
+	config->config_table[0].e1hov_id = cpu_to_le16(bp->e1hov);
+	config->config_table[0].flags = BP_PORT(bp);
+
+	DP(NETIF_MSG_IFUP, "setting MAC (%04x:%04x:%04x)  E1HOV %d  CLID %d\n",
+	   config->config_table[0].msb_mac_addr,
+	   config->config_table[0].middle_mac_addr,
+	   config->config_table[0].lsb_mac_addr, bp->e1hov, BP_L_ID(bp));
+
+	bnx2x_sp_post(bp, RAMROD_CMD_ID_ETH_SET_MAC, 0,
+		      U64_HI(bnx2x_sp_mapping(bp, mac_config)),
+		      U64_LO(bnx2x_sp_mapping(bp, mac_config)), 0);
+}
+
 static int bnx2x_wait_ramrod(struct bnx2x *bp, int state, int idx,
 			     int *state_p, int poll)
 {
 	/* can take a while if any port is running */
-	int timeout = 500;
+	int cnt = 500;
 
 	DP(NETIF_MSG_IFUP, "%s for state to become %x on IDX [%d]\n",
 	   poll ? "polling" : "waiting", state, idx);
 
 	might_sleep();
-
-	while (timeout) {
-
+	while (cnt--) {
 		if (poll) {
 			bnx2x_rx_int(bp->fp, 10);
-			/* If index is different from 0
-			 * The reply for some commands will
+			/* if index is different from 0
+			 * the reply for some commands will
 			 * be on the none default queue
 			 */
 			if (idx)
 				bnx2x_rx_int(&bp->fp[idx], 10);
 		}
-
-		mb(); /* state is changed by bnx2x_sp_event()*/
+		mb(); /* state is changed by bnx2x_sp_event() */
 
 		if (*state_p == state)
 			return 0;
 
-		timeout--;
 		msleep(1);
-
 	}
 
 	/* timeout! */
 	BNX2X_ERR("timeout %s for state %x on IDX [%d]\n",
 		  poll ? "polling" : "waiting", state, idx);
+#ifdef BNX2X_STOP_ON_ERROR
+	bnx2x_panic();
+#endif
 
 	return -EBUSY;
 }
 
 static int bnx2x_setup_leading(struct bnx2x *bp)
 {
+	int rc;
 
 	/* reset IGU state */
-	bnx2x_ack_sb(bp, DEF_SB_ID, CSTORM_ID, 0, IGU_INT_ENABLE, 0);
+	bnx2x_ack_sb(bp, bp->fp[0].sb_id, CSTORM_ID, 0, IGU_INT_ENABLE, 0);
 
 	/* SETUP ramrod */
 	bnx2x_sp_post(bp, RAMROD_CMD_ID_ETH_PORT_SETUP, 0, 0, 0, 0);
 
-	return bnx2x_wait_ramrod(bp, BNX2X_STATE_OPEN, 0, &(bp->state), 0);
+	/* Wait for completion */
+	rc = bnx2x_wait_ramrod(bp, BNX2X_STATE_OPEN, 0, &(bp->state), 0);
 
+	return rc;
 }
 
 static int bnx2x_setup_multi(struct bnx2x *bp, int index)
 {
-
 	/* reset IGU state */
-	bnx2x_ack_sb(bp, index, CSTORM_ID, 0, IGU_INT_ENABLE, 0);
+	bnx2x_ack_sb(bp, bp->fp[index].sb_id, CSTORM_ID, 0, IGU_INT_ENABLE, 0);
 
 	/* SETUP ramrod */
 	bp->fp[index].state = BNX2X_FP_STATE_OPENING;
@@ -4631,82 +5208,107 @@ static int bnx2x_setup_multi(struct bnx2x *bp, int index)
 	/* Wait for completion */
 	return bnx2x_wait_ramrod(bp, BNX2X_FP_STATE_OPEN, index,
 				 &(bp->fp[index].state), 0);
-
 }
 
-
 static int bnx2x_poll(struct napi_struct *napi, int budget);
 static void bnx2x_set_rx_mode(struct net_device *dev);
 
-static int bnx2x_nic_load(struct bnx2x *bp, int req_irq)
+/* must be called with rtnl_lock */
+static int bnx2x_nic_load(struct bnx2x *bp, int load_mode)
 {
 	u32 load_code;
-	int i;
+	int i, rc;
+
+#ifdef BNX2X_STOP_ON_ERROR
+	if (unlikely(bp->panic))
+		return -EPERM;
+#endif
 
 	bp->state = BNX2X_STATE_OPENING_WAIT4_LOAD;
 
-	/* Send LOAD_REQUEST command to MCP.
-	   Returns the type of LOAD command: if it is the
-	   first port to be initialized common blocks should be
-	   initialized, otherwise - not.
+	/* Send LOAD_REQUEST command to MCP
+	   Returns the type of LOAD command:
+	   if it is the first port to be initialized
+	   common blocks should be initialized, otherwise - not
 	*/
-	if (!nomcp) {
+	if (!BP_NOMCP(bp)) {
 		load_code = bnx2x_fw_command(bp, DRV_MSG_CODE_LOAD_REQ);
 		if (!load_code) {
 			BNX2X_ERR("MCP response failure, unloading\n");
 			return -EBUSY;
 		}
-		if (load_code == FW_MSG_CODE_DRV_LOAD_REFUSED) {
-			BNX2X_ERR("MCP refused load request, unloading\n");
+		if (load_code == FW_MSG_CODE_DRV_LOAD_REFUSED)
 			return -EBUSY; /* other port in diagnostic mode */
-		}
+
 	} else {
-		load_code = FW_MSG_CODE_DRV_LOAD_COMMON;
+		DP(NETIF_MSG_IFUP, "NO MCP load counts before us %d, %d, %d\n",
+		   load_count[0], load_count[1], load_count[2]);
+		load_count[0]++;
+		load_count[1 + BP_PORT(bp)]++;
+		DP(NETIF_MSG_IFUP, "NO MCP new load counts       %d, %d, %d\n",
+		   load_count[0], load_count[1], load_count[2]);
+		if (load_count[0] == 1)
+			load_code = FW_MSG_CODE_DRV_LOAD_COMMON;
+		else if (load_count[1 + BP_PORT(bp)] == 1)
+			load_code = FW_MSG_CODE_DRV_LOAD_PORT;
+		else
+			load_code = FW_MSG_CODE_DRV_LOAD_FUNCTION;
 	}
 
-	/* if we can't use msix we only need one fp,
-	 * so try to enable msix with the requested number of fp's
+	if ((load_code == FW_MSG_CODE_DRV_LOAD_COMMON) ||
+	    (load_code == FW_MSG_CODE_DRV_LOAD_PORT))
+		bp->port.pmf = 1;
+	else
+		bp->port.pmf = 0;
+	DP(NETIF_MSG_LINK, "pmf %d\n", bp->port.pmf);
+
+	/* if we can't use MSI-X we only need one fp,
+	 * so try to enable MSI-X with the requested number of fp's
 	 * and fallback to inta with one fp
 	 */
-	if (req_irq) {
-		if (use_inta) {
+	if (use_inta) {
+		bp->num_queues = 1;
+
+	} else {
+		if ((use_multi > 1) && (use_multi <= BP_MAX_QUEUES(bp)))
+			/* user requested number */
+			bp->num_queues = use_multi;
+
+		else if (use_multi)
+			bp->num_queues = min_t(u32, num_online_cpus(),
+					       BP_MAX_QUEUES(bp));
+		else
 			bp->num_queues = 1;
-		} else {
-			if ((use_multi > 1) && (use_multi <= 16))
-				/* user requested number */
-				bp->num_queues = use_multi;
-			else if (use_multi == 1)
-				bp->num_queues = num_online_cpus();
-			else
-				bp->num_queues = 1;
-
-			if (bnx2x_enable_msix(bp)) {
-				/* failed to enable msix */
-				bp->num_queues = 1;
-				if (use_multi)
-					BNX2X_ERR("Multi requested but failed"
-						  " to enable MSI-X\n");
-			}
+
+		if (bnx2x_enable_msix(bp)) {
+			/* failed to enable MSI-X */
+			bp->num_queues = 1;
+			if (use_multi)
+				BNX2X_ERR("Multi requested but failed"
+					  " to enable MSI-X\n");
 		}
 	}
-
-	DP(NETIF_MSG_IFUP, "set number of queues to %d\n", bp->num_queues);
+	DP(NETIF_MSG_IFUP,
+	   "set number of queues to %d\n", bp->num_queues);
 
 	if (bnx2x_alloc_mem(bp))
 		return -ENOMEM;
 
-	if (req_irq) {
-		if (bp->flags & USING_MSIX_FLAG) {
-			if (bnx2x_req_msix_irqs(bp)) {
-				pci_disable_msix(bp->pdev);
-				goto load_error;
-			}
+	/* Disable interrupt handling until HW is initialized */
+	atomic_set(&bp->intr_sem, 1);
 
-		} else {
-			if (bnx2x_req_irq(bp)) {
-				BNX2X_ERR("IRQ request failed, aborting\n");
-				goto load_error;
-			}
+	if (bp->flags & USING_MSIX_FLAG) {
+		rc = bnx2x_req_msix_irqs(bp);
+		if (rc) {
+			pci_disable_msix(bp->pdev);
+			goto load_error;
+		}
+	} else {
+		bnx2x_ack_int(bp);
+		rc = bnx2x_req_irq(bp);
+		if (rc) {
+			BNX2X_ERR("IRQ request failed, aborting\n");
+			goto load_error;
 		}
 	}
 
@@ -4714,26 +5316,25 @@ static int bnx2x_nic_load(struct bnx2x *bp, int req_irq)
 		netif_napi_add(bp->dev, &bnx2x_fp(bp, i, napi),
 			       bnx2x_poll, 128);
 
-
 	/* Initialize HW */
-	if (bnx2x_function_init(bp,
-				(load_code == FW_MSG_CODE_DRV_LOAD_COMMON))) {
+	rc = bnx2x_init_hw(bp, load_code);
+	if (rc) {
 		BNX2X_ERR("HW init failed, aborting\n");
 		goto load_error;
 	}
 
-
+	/* Enable interrupt handling */
 	atomic_set(&bp->intr_sem, 0);
 
-
 	/* Setup NIC internals and enable interrupts */
 	bnx2x_nic_init(bp);
 
 	/* Send LOAD_DONE command to MCP */
-	if (!nomcp) {
+	if (!BP_NOMCP(bp)) {
 		load_code = bnx2x_fw_command(bp, DRV_MSG_CODE_LOAD_DONE);
 		if (!load_code) {
 			BNX2X_ERR("MCP response failure, unloading\n");
+			rc = -EBUSY;
 			goto load_int_disable;
 		}
 	}
@@ -4745,33 +5346,68 @@ static int bnx2x_nic_load(struct bnx2x *bp, int req_irq)
 	for_each_queue(bp, i)
 		napi_enable(&bnx2x_fp(bp, i, napi));
 
-	if (bnx2x_setup_leading(bp))
+	rc = bnx2x_setup_leading(bp);
+	if (rc) {
+#ifdef BNX2X_STOP_ON_ERROR
+		bp->panic = 1;
+#endif
 		goto load_stop_netif;
+	}
 
-	for_each_nondefault_queue(bp, i)
-		if (bnx2x_setup_multi(bp, i))
-			goto load_stop_netif;
+	if (CHIP_IS_E1H(bp))
+		if (bp->mf_config & FUNC_MF_CFG_FUNC_DISABLED) {
+			BNX2X_ERR("!!!  mf_cfg function disabled\n");
+			bp->state = BNX2X_STATE_DISABLED;
+		}
 
-	bnx2x_set_mac_addr(bp);
+	if (bp->state == BNX2X_STATE_OPEN)
+		for_each_nondefault_queue(bp, i) {
+			rc = bnx2x_setup_multi(bp, i);
+			if (rc)
+				goto load_stop_netif;
+		}
 
-	bnx2x_initial_phy_init(bp);
+	if (CHIP_IS_E1(bp))
+		bnx2x_set_mac_addr_e1(bp);
+	else
+		bnx2x_set_mac_addr_e1h(bp);
+
+	if (bp->port.pmf)
+		bnx2x_initial_phy_init(bp);
 
 	/* Start fast path */
-	if (req_irq) { /* IRQ is only requested from bnx2x_open */
+	switch (load_mode) {
+	case LOAD_NORMAL:
+		/* Tx queue should be only reenabled */
+		netif_wake_queue(bp->dev);
+		bnx2x_set_rx_mode(bp->dev);
+		break;
+
+	case LOAD_OPEN:
+		/* IRQ is only requested from bnx2x_open */
 		netif_start_queue(bp->dev);
+		bnx2x_set_rx_mode(bp->dev);
 		if (bp->flags & USING_MSIX_FLAG)
 			printk(KERN_INFO PFX "%s: using MSI-X\n",
 			       bp->dev->name);
+		break;
 
-	/* Otherwise Tx queue should be only reenabled */
-	} else if (netif_running(bp->dev)) {
-		netif_wake_queue(bp->dev);
+	case LOAD_DIAG:
 		bnx2x_set_rx_mode(bp->dev);
+		bp->state = BNX2X_STATE_DIAG;
+		break;
+
+	default:
+		break;
 	}
 
+	if (!bp->port.pmf)
+		bnx2x__link_status_update(bp);
+
 	/* start the timer */
 	mod_timer(&bp->timer, jiffies + bp->current_interval);
 
+
 	return 0;
 
 load_stop_netif:
@@ -4781,7 +5417,7 @@ load_stop_netif:
 load_int_disable:
 	bnx2x_int_disable_sync(bp);
 
-	bnx2x_free_skbs(bp);
+	/* Release IRQs */
 	bnx2x_free_irq(bp);
 
 load_error:
@@ -4789,95 +5425,50 @@ load_error:
 
 	/* TBD we really need to reset the chip
 	   if we want to recover from this */
-	return -EBUSY;
-}
-
-
-static void bnx2x_reset_chip(struct bnx2x *bp, u32 reset_code)
-{
-	int port = bp->port;
-#ifdef USE_DMAE
-	u32 wb_write[2];
-#endif
-	int base, i;
-
-	DP(NETIF_MSG_IFDOWN, "reset called with code %x\n", reset_code);
-
-	/* Do not rcv packets to BRB */
-	REG_WR(bp, NIG_REG_LLH0_BRB1_DRV_MASK + port*4, 0x0);
-	/* Do not direct rcv packets that are not for MCP to the BRB */
-	REG_WR(bp, (port ? NIG_REG_LLH1_BRB1_NOT_MCP :
-			   NIG_REG_LLH0_BRB1_NOT_MCP), 0x0);
-
-	/* Configure IGU and AEU */
-	REG_WR(bp, HC_REG_CONFIG_0 + port*4, 0x1000);
-	REG_WR(bp, MISC_REG_AEU_MASK_ATTN_FUNC_0 + port*4, 0);
-
-	/* TODO: Close Doorbell port? */
-
-	/* Clear ILT */
-#ifdef USE_DMAE
-	wb_write[0] = 0;
-	wb_write[1] = 0;
-#endif
-	base = port * RQ_ONCHIP_AT_PORT_SIZE;
-	for (i = base; i < base + RQ_ONCHIP_AT_PORT_SIZE; i++) {
-#ifdef USE_DMAE
-		REG_WR_DMAE(bp, PXP2_REG_RQ_ONCHIP_AT + i*8, wb_write, 2);
-#else
-		REG_WR_IND(bp, PXP2_REG_RQ_ONCHIP_AT, 0);
-		REG_WR_IND(bp, PXP2_REG_RQ_ONCHIP_AT + 4, 0);
-#endif
-	}
-
-	if (reset_code == FW_MSG_CODE_DRV_UNLOAD_COMMON) {
-		/* reset_common */
-		REG_WR(bp, GRCBASE_MISC + MISC_REGISTERS_RESET_REG_1_CLEAR,
-		       0xd3ffff7f);
-		REG_WR(bp, GRCBASE_MISC + MISC_REGISTERS_RESET_REG_2_CLEAR,
-		       0x1403);
-	}
+	return rc;
 }
 
 static int bnx2x_stop_multi(struct bnx2x *bp, int index)
 {
-
 	int rc;
 
 	/* halt the connection */
 	bp->fp[index].state = BNX2X_FP_STATE_HALTING;
 	bnx2x_sp_post(bp, RAMROD_CMD_ID_ETH_HALT, index, 0, 0, 0);
 
-
+	/* Wait for completion */
 	rc = bnx2x_wait_ramrod(bp, BNX2X_FP_STATE_HALTED, index,
-				       &(bp->fp[index].state), 1);
+			       &(bp->fp[index].state), 1);
 	if (rc) /* timeout */
 		return rc;
 
 	/* delete cfc entry */
 	bnx2x_sp_post(bp, RAMROD_CMD_ID_ETH_CFC_DEL, index, 0, 0, 1);
 
-	return bnx2x_wait_ramrod(bp, BNX2X_FP_STATE_CLOSED, index,
-				 &(bp->fp[index].state), 1);
-
+	/* Wait for completion */
+	rc = bnx2x_wait_ramrod(bp, BNX2X_FP_STATE_CLOSED, index,
+			       &(bp->fp[index].state), 1);
+	return rc;
 }
 
-
 static void bnx2x_stop_leading(struct bnx2x *bp)
 {
 	u16 dsb_sp_prod_idx;
 	/* if the other port is handling traffic,
 	   this can take a lot of time */
-	int timeout = 500;
+	int cnt = 500;
+	int rc;
 
 	might_sleep();
 
 	/* Send HALT ramrod */
 	bp->fp[0].state = BNX2X_FP_STATE_HALTING;
-	bnx2x_sp_post(bp, RAMROD_CMD_ID_ETH_HALT, 0, 0, 0, 0);
+	bnx2x_sp_post(bp, RAMROD_CMD_ID_ETH_HALT, 0, 0, BP_CL_ID(bp), 0);
 
-	if (bnx2x_wait_ramrod(bp, BNX2X_FP_STATE_HALTED, 0,
-			       &(bp->fp[0].state), 1))
+	/* Wait for completion */
+	rc = bnx2x_wait_ramrod(bp, BNX2X_FP_STATE_HALTED, 0,
+			       &(bp->fp[0].state), 1);
+	if (rc) /* timeout */
 		return;
 
 	dsb_sp_prod_idx = *bp->dsb_sp_prod;
@@ -4889,29 +5480,110 @@ static void bnx2x_stop_leading(struct bnx2x *bp)
 	   we are going to reset the chip anyway
 	   so there is not much to do if this times out
 	 */
-	while ((dsb_sp_prod_idx == *bp->dsb_sp_prod) && timeout) {
-		timeout--;
+	while (dsb_sp_prod_idx == *bp->dsb_sp_prod) {
 		msleep(1);
-	}
-	if (!timeout) {
-		DP(NETIF_MSG_IFDOWN, "timeout polling for completion "
-		   "dsb_sp_prod 0x%x != dsb_sp_prod_idx 0x%x\n",
-		   *bp->dsb_sp_prod, dsb_sp_prod_idx);
+		if (!cnt) {
+			DP(NETIF_MSG_IFDOWN, "timeout waiting for port del "
+			   "dsb_sp_prod 0x%x != dsb_sp_prod_idx 0x%x\n",
+			   *bp->dsb_sp_prod, dsb_sp_prod_idx);
+#ifdef BNX2X_STOP_ON_ERROR
+			bnx2x_panic();
+#endif
+			break;
+		}
+		cnt--;
 	}
 	bp->state = BNX2X_STATE_CLOSING_WAIT4_UNLOAD;
 	bp->fp[0].state = BNX2X_FP_STATE_CLOSED;
 }
 
+static void bnx2x_reset_func(struct bnx2x *bp)
+{
+	int port = BP_PORT(bp);
+	int func = BP_FUNC(bp);
+	int base, i;
+
+	/* Configure IGU */
+	REG_WR(bp, HC_REG_LEADING_EDGE_0 + port*8, 0);
+	REG_WR(bp, HC_REG_TRAILING_EDGE_0 + port*8, 0);
+
+	REG_WR(bp, HC_REG_CONFIG_0 + port*4, 0x1000);
+
+	/* Clear ILT */
+	base = FUNC_ILT_BASE(func);
+	for (i = base; i < base + ILT_PER_FUNC; i++)
+		bnx2x_ilt_wr(bp, i, 0);
+}
+
+static void bnx2x_reset_port(struct bnx2x *bp)
+{
+	int port = BP_PORT(bp);
+	u32 val;
+
+	REG_WR(bp, NIG_REG_MASK_INTERRUPT_PORT0 + port*4, 0);
+
+	/* Do not rcv packets to BRB */
+	REG_WR(bp, NIG_REG_LLH0_BRB1_DRV_MASK + port*4, 0x0);
+	/* Do not direct rcv packets that are not for MCP to the BRB */
+	REG_WR(bp, (port ? NIG_REG_LLH1_BRB1_NOT_MCP :
+			   NIG_REG_LLH0_BRB1_NOT_MCP), 0x0);
+
+	/* Configure AEU */
+	REG_WR(bp, MISC_REG_AEU_MASK_ATTN_FUNC_0 + port*4, 0);
+
+	msleep(100);
+	/* Check for BRB port occupancy */
+	val = REG_RD(bp, BRB1_REG_PORT_NUM_OCC_BLOCKS_0 + port*4);
+	if (val)
+		DP(NETIF_MSG_IFDOWN,
+		   "BRB1 is not empty  %d blooks are occupied\n", val);
+
+	/* TODO: Close Doorbell port? */
+}
+
+static void bnx2x_reset_common(struct bnx2x *bp)
+{
+	/* reset_common */
+	REG_WR(bp, GRCBASE_MISC + MISC_REGISTERS_RESET_REG_1_CLEAR,
+	       0xd3ffff7f);
+	REG_WR(bp, GRCBASE_MISC + MISC_REGISTERS_RESET_REG_2_CLEAR, 0x1403);
+}
+
+static void bnx2x_reset_chip(struct bnx2x *bp, u32 reset_code)
+{
+	DP(BNX2X_MSG_MCP, "function %d  reset_code %x\n",
+	   BP_FUNC(bp), reset_code);
+
+	switch (reset_code) {
+	case FW_MSG_CODE_DRV_UNLOAD_COMMON:
+		bnx2x_reset_port(bp);
+		bnx2x_reset_func(bp);
+		bnx2x_reset_common(bp);
+		break;
+
+	case FW_MSG_CODE_DRV_UNLOAD_PORT:
+		bnx2x_reset_port(bp);
+		bnx2x_reset_func(bp);
+		break;
+
+	case FW_MSG_CODE_DRV_UNLOAD_FUNCTION:
+		bnx2x_reset_func(bp);
+		break;
 
-static int bnx2x_nic_unload(struct bnx2x *bp, int free_irq)
+	default:
+		BNX2X_ERR("Unknown reset_code (0x%x) from MCP\n", reset_code);
+		break;
+	}
+}
+
+/* msut be called with rtnl_lock */
+static int bnx2x_nic_unload(struct bnx2x *bp, int unload_mode)
 {
 	u32 reset_code = 0;
-	int i, timeout;
+	int i, cnt;
 
 	bp->state = BNX2X_STATE_CLOSING_WAIT4_HALT;
 
-	del_timer_sync(&bp->timer);
-
 	bp->rx_mode = BNX2X_RX_MODE_NONE;
 	bnx2x_set_storm_rx_mode(bp);
 
@@ -4920,21 +5592,44 @@ static int bnx2x_nic_unload(struct bnx2x *bp, int free_irq)
 		bp->dev->trans_start = jiffies;	/* prevent tx timeout */
 	}
 
+	del_timer_sync(&bp->timer);
+	SHMEM_WR(bp, func_mb[BP_FUNC(bp)].drv_pulse_mb,
+		 (DRV_PULSE_ALWAYS_ALIVE | bp->fw_drv_pulse_wr_seq));
+
 	/* Wait until all fast path tasks complete */
 	for_each_queue(bp, i) {
 		struct bnx2x_fastpath *fp = &bp->fp[i];
 
-		timeout = 1000;
-		while (bnx2x_has_work(fp) && (timeout--))
+#ifdef BNX2X_STOP_ON_ERROR
+#ifdef __powerpc64__
+		DP(NETIF_MSG_RX_STATUS, "fp->tpa_queue_used = 0x%lx\n",
+#else
+		DP(NETIF_MSG_IFDOWN, "fp->tpa_queue_used = 0x%llx\n",
+#endif
+		   fp->tpa_queue_used);
+#endif
+		cnt = 1000;
+		smp_rmb();
+		while (bnx2x_has_work(fp)) {
 			msleep(1);
-		if (!timeout)
-			BNX2X_ERR("timeout waiting for queue[%d]\n", i);
+			if (!cnt) {
+				BNX2X_ERR("timeout waiting for queue[%d]\n",
+					  i);
+#ifdef BNX2X_STOP_ON_ERROR
+				bnx2x_panic();
+				return -EBUSY;
+#else
+				break;
+#endif
+			}
+			cnt--;
+			smp_rmb();
+		}
 	}
 
-	/* Wait until stat ramrod returns and all SP tasks complete */
-	timeout = 1000;
-	while ((bp->stat_pending || (bp->spq_left != MAX_SPQ_PENDING)) &&
-	       (timeout--))
+	/* Wait until all slow path tasks complete */
+	cnt = 1000;
+	while ((bp->spq_left != MAX_SPQ_PENDING) && cnt--)
 		msleep(1);
 
 	for_each_queue(bp, i)
@@ -4942,59 +5637,84 @@ static int bnx2x_nic_unload(struct bnx2x *bp, int free_irq)
 	/* Disable interrupts after Tx and Rx are disabled on stack level */
 	bnx2x_int_disable_sync(bp);
 
+	/* Release IRQs */
+	bnx2x_free_irq(bp);
+
 	if (bp->flags & NO_WOL_FLAG)
 		reset_code = DRV_MSG_CODE_UNLOAD_REQ_WOL_MCP;
 
 	else if (bp->wol) {
-		u32 emac_base = bp->port ? GRCBASE_EMAC0 : GRCBASE_EMAC1;
+		u32 emac_base = BP_PORT(bp) ? GRCBASE_EMAC1 : GRCBASE_EMAC0;
 		u8 *mac_addr = bp->dev->dev_addr;
-		u32 val = (EMAC_MODE_MPKT | EMAC_MODE_MPKT_RCVD |
-			   EMAC_MODE_ACPI_RCVD);
-
-		EMAC_WR(EMAC_REG_EMAC_MODE, val);
+		u32 val;
 
+		/* The mac address is written to entries 1-4 to
+		   preserve entry 0 which is used by the PMF */
 		val = (mac_addr[0] << 8) | mac_addr[1];
-		EMAC_WR(EMAC_REG_EMAC_MAC_MATCH, val);
+		EMAC_WR(EMAC_REG_EMAC_MAC_MATCH + (BP_E1HVN(bp) + 1)*8, val);
 
 		val = (mac_addr[2] << 24) | (mac_addr[3] << 16) |
 		      (mac_addr[4] << 8) | mac_addr[5];
-		EMAC_WR(EMAC_REG_EMAC_MAC_MATCH + 4, val);
+		EMAC_WR(EMAC_REG_EMAC_MAC_MATCH + (BP_E1HVN(bp) + 1)*8 + 4,
+			val);
 
 		reset_code = DRV_MSG_CODE_UNLOAD_REQ_WOL_EN;
 
 	} else
 		reset_code = DRV_MSG_CODE_UNLOAD_REQ_WOL_DIS;
 
-	/* Close multi and leading connections */
+	/* Close multi and leading connections
+	   Completions for ramrods are collected in a synchronous way */
 	for_each_nondefault_queue(bp, i)
 		if (bnx2x_stop_multi(bp, i))
 			goto unload_error;
 
-	bnx2x_stop_leading(bp);
+	if (CHIP_IS_E1H(bp))
+		REG_WR(bp, NIG_REG_LLH0_FUNC_EN + BP_PORT(bp)*8, 0);
+
+	bnx2x_stop_leading(bp);
+#ifdef BNX2X_STOP_ON_ERROR
+	/* If ramrod completion timed out - break here! */
+	if (bp->panic) {
+		BNX2X_ERR("Stop leading failed!\n");
+		return -EBUSY;
+	}
+#endif
+
 	if ((bp->state != BNX2X_STATE_CLOSING_WAIT4_UNLOAD) ||
 	    (bp->fp[0].state != BNX2X_FP_STATE_CLOSED)) {
-		DP(NETIF_MSG_IFDOWN, "failed to close leading properly!"
-		   "state 0x%x  fp[0].state 0x%x",
+		DP(NETIF_MSG_IFDOWN, "failed to close leading properly!  "
+		   "state 0x%x  fp[0].state 0x%x\n",
 		   bp->state, bp->fp[0].state);
 	}
 
 unload_error:
-	bnx2x__link_reset(bp);
-
-	if (!nomcp)
+	if (!BP_NOMCP(bp))
 		reset_code = bnx2x_fw_command(bp, reset_code);
-	else
-		reset_code = FW_MSG_CODE_DRV_UNLOAD_COMMON;
+	else {
+		DP(NETIF_MSG_IFDOWN, "NO MCP load counts      %d, %d, %d\n",
+		   load_count[0], load_count[1], load_count[2]);
+		load_count[0]--;
+		load_count[1 + BP_PORT(bp)]--;
+		DP(NETIF_MSG_IFDOWN, "NO MCP new load counts  %d, %d, %d\n",
+		   load_count[0], load_count[1], load_count[2]);
+		if (load_count[0] == 0)
+			reset_code = FW_MSG_CODE_DRV_UNLOAD_COMMON;
+		else if (load_count[1 + BP_PORT(bp)] == 0)
+			reset_code = FW_MSG_CODE_DRV_UNLOAD_PORT;
+		else
+			reset_code = FW_MSG_CODE_DRV_UNLOAD_FUNCTION;
+	}
 
-	/* Release IRQs */
-	if (free_irq)
-		bnx2x_free_irq(bp);
+	if ((reset_code == FW_MSG_CODE_DRV_UNLOAD_COMMON) ||
+	    (reset_code == FW_MSG_CODE_DRV_UNLOAD_PORT))
+		bnx2x__link_reset(bp);
 
 	/* Reset the chip */
 	bnx2x_reset_chip(bp, reset_code);
 
 	/* Report UNLOAD_DONE to MCP */
-	if (!nomcp)
+	if (!BP_NOMCP(bp))
 		bnx2x_fw_command(bp, DRV_MSG_CODE_UNLOAD_DONE);
 
 	/* Free SKBs and driver internals */
@@ -5008,6 +5728,29 @@ unload_error:
 	return 0;
 }
 
+static void bnx2x_reset_task(struct work_struct *work)
+{
+	struct bnx2x *bp = container_of(work, struct bnx2x, reset_task);
+
+#ifdef BNX2X_STOP_ON_ERROR
+	BNX2X_ERR("reset task called but STOP_ON_ERROR defined"
+		  " so reset not done to allow debug dump,\n"
+	 KERN_ERR " you will need to reboot when done\n");
+	return;
+#endif
+
+	rtnl_lock();
+
+	if (!netif_running(bp->dev))
+		goto reset_task_exit;
+
+	bnx2x_nic_unload(bp, UNLOAD_NORMAL);
+	bnx2x_nic_load(bp, LOAD_NORMAL);
+
+reset_task_exit:
+	rtnl_unlock();
+}
+
 /* end of nic load/unload */
 
 /* ethtool_ops */
@@ -5016,9 +5759,139 @@ unload_error:
  * Init service functions
  */
 
-static void bnx2x_link_settings_supported(struct bnx2x *bp, u32 switch_cfg)
+static void __devinit bnx2x_undi_unload(struct bnx2x *bp)
+{
+	u32 val;
+
+	/* Check if there is any driver already loaded */
+	val = REG_RD(bp, MISC_REG_UNPREPARED);
+	if (val == 0x1) {
+		/* Check if it is the UNDI driver
+		 * UNDI driver initializes CID offset for normal bell to 0x7
+		 */
+		val = REG_RD(bp, DORQ_REG_NORM_CID_OFST);
+		if (val == 0x7) {
+			u32 reset_code = DRV_MSG_CODE_UNLOAD_REQ_WOL_DIS;
+			/* save our func and fw_seq */
+			int func = BP_FUNC(bp);
+			u16 fw_seq = bp->fw_seq;
+
+			BNX2X_DEV_INFO("UNDI is active! reset device\n");
+
+			/* try unload UNDI on port 0 */
+			bp->func = 0;
+			bp->fw_seq = (SHMEM_RD(bp,
+					     func_mb[bp->func].drv_mb_header) &
+				      DRV_MSG_SEQ_NUMBER_MASK);
+
+			reset_code = bnx2x_fw_command(bp, reset_code);
+			bnx2x_fw_command(bp, DRV_MSG_CODE_UNLOAD_DONE);
+
+			/* if UNDI is loaded on the other port */
+			if (reset_code != FW_MSG_CODE_DRV_UNLOAD_COMMON) {
+
+				bp->func = 1;
+				bp->fw_seq = (SHMEM_RD(bp,
+					     func_mb[bp->func].drv_mb_header) &
+					      DRV_MSG_SEQ_NUMBER_MASK);
+
+				bnx2x_fw_command(bp,
+					     DRV_MSG_CODE_UNLOAD_REQ_WOL_DIS);
+				bnx2x_fw_command(bp,
+						 DRV_MSG_CODE_UNLOAD_DONE);
+
+				/* restore our func and fw_seq */
+				bp->func = func;
+				bp->fw_seq = fw_seq;
+			}
+
+			/* reset device */
+			REG_WR(bp,
+			       GRCBASE_MISC + MISC_REGISTERS_RESET_REG_1_CLEAR,
+			       0xd3ffff7f);
+			REG_WR(bp,
+			       GRCBASE_MISC + MISC_REGISTERS_RESET_REG_2_CLEAR,
+			       0x1403);
+		}
+	}
+}
+
+static void __devinit bnx2x_get_common_hwinfo(struct bnx2x *bp)
+{
+	u32 val, val2, val3, val4, id;
+
+	/* Get the chip revision id and number. */
+	/* chip num:16-31, rev:12-15, metal:4-11, bond_id:0-3 */
+	val = REG_RD(bp, MISC_REG_CHIP_NUM);
+	id = ((val & 0xffff) << 16);
+	val = REG_RD(bp, MISC_REG_CHIP_REV);
+	id |= ((val & 0xf) << 12);
+	val = REG_RD(bp, MISC_REG_CHIP_METAL);
+	id |= ((val & 0xff) << 4);
+	REG_RD(bp, MISC_REG_BOND_ID);
+	id |= (val & 0xf);
+	bp->common.chip_id = id;
+	bp->link_params.chip_id = bp->common.chip_id;
+	BNX2X_DEV_INFO("chip ID is 0x%x\n", id);
+
+	val = REG_RD(bp, MCP_REG_MCPR_NVM_CFG4);
+	bp->common.flash_size = (NVRAM_1MB_SIZE <<
+				 (val & MCPR_NVM_CFG4_FLASH_SIZE));
+	BNX2X_DEV_INFO("flash_size 0x%x (%d)\n",
+		       bp->common.flash_size, bp->common.flash_size);
+
+	bp->common.shmem_base = REG_RD(bp, MISC_REG_SHARED_MEM_ADDR);
+	bp->link_params.shmem_base = bp->common.shmem_base;
+	BNX2X_DEV_INFO("shmem offset is 0x%x\n", bp->common.shmem_base);
+
+	if (!bp->common.shmem_base ||
+	    (bp->common.shmem_base < 0xA0000) ||
+	    (bp->common.shmem_base >= 0xC0000)) {
+		BNX2X_DEV_INFO("MCP not active\n");
+		bp->flags |= NO_MCP_FLAG;
+		return;
+	}
+
+	val = SHMEM_RD(bp, validity_map[BP_PORT(bp)]);
+	if ((val & (SHR_MEM_VALIDITY_DEV_INFO | SHR_MEM_VALIDITY_MB))
+		!= (SHR_MEM_VALIDITY_DEV_INFO | SHR_MEM_VALIDITY_MB))
+		BNX2X_ERR("BAD MCP validity signature\n");
+
+	bp->common.hw_config = SHMEM_RD(bp, dev_info.shared_hw_config.config);
+	bp->common.board = SHMEM_RD(bp, dev_info.shared_hw_config.board);
+
+	BNX2X_DEV_INFO("hw_config 0x%08x  board 0x%08x\n",
+		       bp->common.hw_config, bp->common.board);
+
+	bp->link_params.hw_led_mode = ((bp->common.hw_config &
+					SHARED_HW_CFG_LED_MODE_MASK) >>
+				       SHARED_HW_CFG_LED_MODE_SHIFT);
+
+	val = SHMEM_RD(bp, dev_info.bc_rev) >> 8;
+	bp->common.bc_ver = val;
+	BNX2X_DEV_INFO("bc_ver %X\n", val);
+	if (val < BNX2X_BC_VER) {
+		/* for now only warn
+		 * later we might need to enforce this */
+		BNX2X_ERR("This driver needs bc_ver %X but found %X,"
+			  " please upgrade BC\n", BNX2X_BC_VER, val);
+	}
+	BNX2X_DEV_INFO("%sWoL Capable\n",
+		       (bp->flags & NO_WOL_FLAG)? "Not " : "");
+
+	val = SHMEM_RD(bp, dev_info.shared_hw_config.part_num);
+	val2 = SHMEM_RD(bp, dev_info.shared_hw_config.part_num[4]);
+	val3 = SHMEM_RD(bp, dev_info.shared_hw_config.part_num[8]);
+	val4 = SHMEM_RD(bp, dev_info.shared_hw_config.part_num[12]);
+
+	printk(KERN_INFO PFX "part number %X-%X-%X-%X\n",
+	       val, val2, val3, val4);
+}
+
+static void __devinit bnx2x_link_settings_supported(struct bnx2x *bp,
+						    u32 switch_cfg)
 {
-	int port = bp->port;
+	int port = BP_PORT(bp);
 	u32 ext_phy_type;
 
 	switch (switch_cfg) {
@@ -5032,31 +5905,33 @@ static void bnx2x_link_settings_supported(struct bnx2x *bp, u32 switch_cfg)
 			BNX2X_DEV_INFO("ext_phy_type 0x%x (Direct)\n",
 				       ext_phy_type);
 
-			bp->supported |= (SUPPORTED_10baseT_Half |
-					  SUPPORTED_10baseT_Full |
-					  SUPPORTED_100baseT_Half |
-					  SUPPORTED_100baseT_Full |
-					  SUPPORTED_1000baseT_Full |
-					  SUPPORTED_2500baseX_Full |
-					  SUPPORTED_TP | SUPPORTED_FIBRE |
-					  SUPPORTED_Autoneg |
-					  SUPPORTED_Pause |
-					  SUPPORTED_Asym_Pause);
+			bp->port.supported |= (SUPPORTED_10baseT_Half |
+					       SUPPORTED_10baseT_Full |
+					       SUPPORTED_100baseT_Half |
+					       SUPPORTED_100baseT_Full |
+					       SUPPORTED_1000baseT_Full |
+					       SUPPORTED_2500baseX_Full |
+					       SUPPORTED_TP |
+					       SUPPORTED_FIBRE |
+					       SUPPORTED_Autoneg |
+					       SUPPORTED_Pause |
+					       SUPPORTED_Asym_Pause);
 			break;
 
 		case PORT_HW_CFG_SERDES_EXT_PHY_TYPE_BCM5482:
 			BNX2X_DEV_INFO("ext_phy_type 0x%x (5482)\n",
 				       ext_phy_type);
 
-			bp->supported |= (SUPPORTED_10baseT_Half |
-					  SUPPORTED_10baseT_Full |
-					  SUPPORTED_100baseT_Half |
-					  SUPPORTED_100baseT_Full |
-					  SUPPORTED_1000baseT_Full |
-					  SUPPORTED_TP | SUPPORTED_FIBRE |
-					  SUPPORTED_Autoneg |
-					  SUPPORTED_Pause |
-					  SUPPORTED_Asym_Pause);
+			bp->port.supported |= (SUPPORTED_10baseT_Half |
+					       SUPPORTED_10baseT_Full |
+					       SUPPORTED_100baseT_Half |
+					       SUPPORTED_100baseT_Full |
+					       SUPPORTED_1000baseT_Full |
+					       SUPPORTED_TP |
+					       SUPPORTED_FIBRE |
+					       SUPPORTED_Autoneg |
+					       SUPPORTED_Pause |
+					       SUPPORTED_Asym_Pause);
 			break;
 
 		default:
@@ -5066,9 +5941,9 @@ static void bnx2x_link_settings_supported(struct bnx2x *bp, u32 switch_cfg)
 			return;
 		}
 
-		bp->phy_addr = REG_RD(bp, NIG_REG_SERDES0_CTRL_PHY_ADDR +
-				      port*0x10);
-		BNX2X_DEV_INFO("phy_addr 0x%x\n", bp->phy_addr);
+		bp->port.phy_addr = REG_RD(bp, NIG_REG_SERDES0_CTRL_PHY_ADDR +
+					   port*0x10);
+		BNX2X_DEV_INFO("phy_addr 0x%x\n", bp->port.phy_addr);
 		break;
 
 	case SWITCH_CFG_10G:
@@ -5081,75 +5956,75 @@ static void bnx2x_link_settings_supported(struct bnx2x *bp, u32 switch_cfg)
 			BNX2X_DEV_INFO("ext_phy_type 0x%x (Direct)\n",
 				       ext_phy_type);
 
-			bp->supported |= (SUPPORTED_10baseT_Half |
-					  SUPPORTED_10baseT_Full |
-					  SUPPORTED_100baseT_Half |
-					  SUPPORTED_100baseT_Full |
-					  SUPPORTED_1000baseT_Full |
-					  SUPPORTED_2500baseX_Full |
-					  SUPPORTED_10000baseT_Full |
-					  SUPPORTED_TP | SUPPORTED_FIBRE |
-					  SUPPORTED_Autoneg |
-					  SUPPORTED_Pause |
-					  SUPPORTED_Asym_Pause);
+			bp->port.supported |= (SUPPORTED_10baseT_Half |
+					       SUPPORTED_10baseT_Full |
+					       SUPPORTED_100baseT_Half |
+					       SUPPORTED_100baseT_Full |
+					       SUPPORTED_1000baseT_Full |
+					       SUPPORTED_2500baseX_Full |
+					       SUPPORTED_10000baseT_Full |
+					       SUPPORTED_TP |
+					       SUPPORTED_FIBRE |
+					       SUPPORTED_Autoneg |
+					       SUPPORTED_Pause |
+					       SUPPORTED_Asym_Pause);
 			break;
 
 		case PORT_HW_CFG_XGXS_EXT_PHY_TYPE_BCM8705:
 			BNX2X_DEV_INFO("ext_phy_type 0x%x (8705)\n",
-					ext_phy_type);
+				       ext_phy_type);
 
-			bp->supported |= (SUPPORTED_10000baseT_Full |
-					  SUPPORTED_FIBRE |
-					  SUPPORTED_Pause |
-					  SUPPORTED_Asym_Pause);
+			bp->port.supported |= (SUPPORTED_10000baseT_Full |
+					       SUPPORTED_FIBRE |
+					       SUPPORTED_Pause |
+					       SUPPORTED_Asym_Pause);
 			break;
 
 		case PORT_HW_CFG_XGXS_EXT_PHY_TYPE_BCM8706:
 			BNX2X_DEV_INFO("ext_phy_type 0x%x (8706)\n",
 				       ext_phy_type);
 
-			bp->supported |= (SUPPORTED_10000baseT_Full |
-					  SUPPORTED_1000baseT_Full |
-					  SUPPORTED_Autoneg |
-					  SUPPORTED_FIBRE |
-					  SUPPORTED_Pause |
-					  SUPPORTED_Asym_Pause);
+			bp->port.supported |= (SUPPORTED_10000baseT_Full |
+					       SUPPORTED_1000baseT_Full |
+					       SUPPORTED_FIBRE |
+					       SUPPORTED_Pause |
+					       SUPPORTED_Asym_Pause);
 			break;
 
 		case PORT_HW_CFG_XGXS_EXT_PHY_TYPE_BCM8072:
 			BNX2X_DEV_INFO("ext_phy_type 0x%x (8072)\n",
 				       ext_phy_type);
 
-			bp->supported |= (SUPPORTED_10000baseT_Full |
-					  SUPPORTED_1000baseT_Full |
-					  SUPPORTED_FIBRE |
-					  SUPPORTED_Autoneg |
-					  SUPPORTED_Pause |
-					  SUPPORTED_Asym_Pause);
+			bp->port.supported |= (SUPPORTED_10000baseT_Full |
+					       SUPPORTED_1000baseT_Full |
+					       SUPPORTED_FIBRE |
+					       SUPPORTED_Autoneg |
+					       SUPPORTED_Pause |
+					       SUPPORTED_Asym_Pause);
 			break;
 
 		case PORT_HW_CFG_XGXS_EXT_PHY_TYPE_BCM8073:
 			BNX2X_DEV_INFO("ext_phy_type 0x%x (8073)\n",
 				       ext_phy_type);
 
-			bp->supported |= (SUPPORTED_10000baseT_Full |
-					  SUPPORTED_2500baseX_Full |
-					  SUPPORTED_1000baseT_Full |
-					  SUPPORTED_FIBRE |
-					  SUPPORTED_Autoneg |
-					  SUPPORTED_Pause |
-					  SUPPORTED_Asym_Pause);
+			bp->port.supported |= (SUPPORTED_10000baseT_Full |
+					       SUPPORTED_2500baseX_Full |
+					       SUPPORTED_1000baseT_Full |
+					       SUPPORTED_FIBRE |
+					       SUPPORTED_Autoneg |
+					       SUPPORTED_Pause |
+					       SUPPORTED_Asym_Pause);
 			break;
 
 		case PORT_HW_CFG_XGXS_EXT_PHY_TYPE_SFX7101:
 			BNX2X_DEV_INFO("ext_phy_type 0x%x (SFX7101)\n",
 				       ext_phy_type);
 
-			bp->supported |= (SUPPORTED_10000baseT_Full |
-					  SUPPORTED_TP |
-					  SUPPORTED_Autoneg |
-					  SUPPORTED_Pause |
-					  SUPPORTED_Asym_Pause);
+			bp->port.supported |= (SUPPORTED_10000baseT_Full |
+					       SUPPORTED_TP |
+					       SUPPORTED_Autoneg |
+					       SUPPORTED_Pause |
+					       SUPPORTED_Asym_Pause);
 			break;
 
 		case PORT_HW_CFG_XGXS_EXT_PHY_TYPE_FAILURE:
@@ -5164,61 +6039,61 @@ static void bnx2x_link_settings_supported(struct bnx2x *bp, u32 switch_cfg)
 			return;
 		}
 
-		bp->phy_addr = REG_RD(bp, NIG_REG_XGXS0_CTRL_PHY_ADDR +
-				      port*0x18);
-		BNX2X_DEV_INFO("phy_addr 0x%x\n", bp->phy_addr);
+		bp->port.phy_addr = REG_RD(bp, NIG_REG_XGXS0_CTRL_PHY_ADDR +
+					   port*0x18);
+		BNX2X_DEV_INFO("phy_addr 0x%x\n", bp->port.phy_addr);
 
 		break;
 
 	default:
 		BNX2X_ERR("BAD switch_cfg link_config 0x%x\n",
-			  bp->link_config);
+			  bp->port.link_config);
 		return;
 	}
-	bp->link_params.phy_addr = bp->phy_addr;
+	bp->link_params.phy_addr = bp->port.phy_addr;
 
 	/* mask what we support according to speed_cap_mask */
 	if (!(bp->link_params.speed_cap_mask &
 				PORT_HW_CFG_SPEED_CAPABILITY_D0_10M_HALF))
-		bp->supported &= ~SUPPORTED_10baseT_Half;
+		bp->port.supported &= ~SUPPORTED_10baseT_Half;
 
 	if (!(bp->link_params.speed_cap_mask &
 				PORT_HW_CFG_SPEED_CAPABILITY_D0_10M_FULL))
-		bp->supported &= ~SUPPORTED_10baseT_Full;
+		bp->port.supported &= ~SUPPORTED_10baseT_Full;
 
 	if (!(bp->link_params.speed_cap_mask &
 				PORT_HW_CFG_SPEED_CAPABILITY_D0_100M_HALF))
-		bp->supported &= ~SUPPORTED_100baseT_Half;
+		bp->port.supported &= ~SUPPORTED_100baseT_Half;
 
 	if (!(bp->link_params.speed_cap_mask &
 				PORT_HW_CFG_SPEED_CAPABILITY_D0_100M_FULL))
-		bp->supported &= ~SUPPORTED_100baseT_Full;
+		bp->port.supported &= ~SUPPORTED_100baseT_Full;
 
 	if (!(bp->link_params.speed_cap_mask &
 					PORT_HW_CFG_SPEED_CAPABILITY_D0_1G))
-		bp->supported &= ~(SUPPORTED_1000baseT_Half |
-				   SUPPORTED_1000baseT_Full);
+		bp->port.supported &= ~(SUPPORTED_1000baseT_Half |
+					SUPPORTED_1000baseT_Full);
 
 	if (!(bp->link_params.speed_cap_mask &
 					PORT_HW_CFG_SPEED_CAPABILITY_D0_2_5G))
-		bp->supported &= ~SUPPORTED_2500baseX_Full;
+		bp->port.supported &= ~SUPPORTED_2500baseX_Full;
 
 	if (!(bp->link_params.speed_cap_mask &
 					PORT_HW_CFG_SPEED_CAPABILITY_D0_10G))
-		bp->supported &= ~SUPPORTED_10000baseT_Full;
+		bp->port.supported &= ~SUPPORTED_10000baseT_Full;
 
-	BNX2X_DEV_INFO("supported 0x%x\n", bp->supported);
+	BNX2X_DEV_INFO("supported 0x%x\n", bp->port.supported);
 }
 
-static void bnx2x_link_settings_requested(struct bnx2x *bp)
+static void __devinit bnx2x_link_settings_requested(struct bnx2x *bp)
 {
 	bp->link_params.req_duplex = DUPLEX_FULL;
 
-	switch (bp->link_config & PORT_FEATURE_LINK_SPEED_MASK) {
+	switch (bp->port.link_config & PORT_FEATURE_LINK_SPEED_MASK) {
 	case PORT_FEATURE_LINK_SPEED_AUTO:
-		if (bp->supported & SUPPORTED_Autoneg) {
+		if (bp->port.supported & SUPPORTED_Autoneg) {
 			bp->link_params.req_line_speed = SPEED_AUTO_NEG;
-			bp->advertising = bp->supported;
+			bp->port.advertising = bp->port.supported;
 		} else {
 			u32 ext_phy_type =
 			    XGXS_EXT_PHY_TYPE(bp->link_params.ext_phy_config);
@@ -5229,7 +6104,7 @@ static void bnx2x_link_settings_requested(struct bnx2x *bp)
 			     PORT_HW_CFG_XGXS_EXT_PHY_TYPE_BCM8706)) {
 				/* force 10G, no AN */
 				bp->link_params.req_line_speed = SPEED_10000;
-				bp->advertising =
+				bp->port.advertising =
 						(ADVERTISED_10000baseT_Full |
 						 ADVERTISED_FIBRE);
 				break;
@@ -5237,98 +6112,98 @@ static void bnx2x_link_settings_requested(struct bnx2x *bp)
 			BNX2X_ERR("NVRAM config error. "
 				  "Invalid link_config 0x%x"
 				  "  Autoneg not supported\n",
-				  bp->link_config);
+				  bp->port.link_config);
 			return;
 		}
 		break;
 
 	case PORT_FEATURE_LINK_SPEED_10M_FULL:
-		if (bp->supported & SUPPORTED_10baseT_Full) {
+		if (bp->port.supported & SUPPORTED_10baseT_Full) {
 			bp->link_params.req_line_speed = SPEED_10;
-			bp->advertising = (ADVERTISED_10baseT_Full |
-					   ADVERTISED_TP);
+			bp->port.advertising = (ADVERTISED_10baseT_Full |
+						ADVERTISED_TP);
 		} else {
 			BNX2X_ERR("NVRAM config error. "
 				  "Invalid link_config 0x%x"
 				  "  speed_cap_mask 0x%x\n",
-				  bp->link_config,
+				  bp->port.link_config,
 				  bp->link_params.speed_cap_mask);
 			return;
 		}
 		break;
 
 	case PORT_FEATURE_LINK_SPEED_10M_HALF:
-		if (bp->supported & SUPPORTED_10baseT_Half) {
+		if (bp->port.supported & SUPPORTED_10baseT_Half) {
 			bp->link_params.req_line_speed = SPEED_10;
 			bp->link_params.req_duplex = DUPLEX_HALF;
-			bp->advertising = (ADVERTISED_10baseT_Half |
-					   ADVERTISED_TP);
+			bp->port.advertising = (ADVERTISED_10baseT_Half |
+						ADVERTISED_TP);
 		} else {
 			BNX2X_ERR("NVRAM config error. "
 				  "Invalid link_config 0x%x"
 				  "  speed_cap_mask 0x%x\n",
-				  bp->link_config,
+				  bp->port.link_config,
 				  bp->link_params.speed_cap_mask);
 			return;
 		}
 		break;
 
 	case PORT_FEATURE_LINK_SPEED_100M_FULL:
-		if (bp->supported & SUPPORTED_100baseT_Full) {
+		if (bp->port.supported & SUPPORTED_100baseT_Full) {
 			bp->link_params.req_line_speed = SPEED_100;
-			bp->advertising = (ADVERTISED_100baseT_Full |
-					   ADVERTISED_TP);
+			bp->port.advertising = (ADVERTISED_100baseT_Full |
+						ADVERTISED_TP);
 		} else {
 			BNX2X_ERR("NVRAM config error. "
 				  "Invalid link_config 0x%x"
 				  "  speed_cap_mask 0x%x\n",
-				  bp->link_config,
+				  bp->port.link_config,
 				  bp->link_params.speed_cap_mask);
 			return;
 		}
 		break;
 
 	case PORT_FEATURE_LINK_SPEED_100M_HALF:
-		if (bp->supported & SUPPORTED_100baseT_Half) {
+		if (bp->port.supported & SUPPORTED_100baseT_Half) {
 			bp->link_params.req_line_speed = SPEED_100;
 			bp->link_params.req_duplex = DUPLEX_HALF;
-			bp->advertising = (ADVERTISED_100baseT_Half |
-					   ADVERTISED_TP);
+			bp->port.advertising = (ADVERTISED_100baseT_Half |
+						ADVERTISED_TP);
 		} else {
 			BNX2X_ERR("NVRAM config error. "
 				  "Invalid link_config 0x%x"
 				  "  speed_cap_mask 0x%x\n",
-				  bp->link_config,
+				  bp->port.link_config,
 				  bp->link_params.speed_cap_mask);
 			return;
 		}
 		break;
 
 	case PORT_FEATURE_LINK_SPEED_1G:
-		if (bp->supported & SUPPORTED_1000baseT_Full) {
+		if (bp->port.supported & SUPPORTED_1000baseT_Full) {
 			bp->link_params.req_line_speed = SPEED_1000;
-			bp->advertising = (ADVERTISED_1000baseT_Full |
-					   ADVERTISED_TP);
+			bp->port.advertising = (ADVERTISED_1000baseT_Full |
+						ADVERTISED_TP);
 		} else {
 			BNX2X_ERR("NVRAM config error. "
 				  "Invalid link_config 0x%x"
 				  "  speed_cap_mask 0x%x\n",
-				  bp->link_config,
+				  bp->port.link_config,
 				  bp->link_params.speed_cap_mask);
 			return;
 		}
 		break;
 
 	case PORT_FEATURE_LINK_SPEED_2_5G:
-		if (bp->supported & SUPPORTED_2500baseX_Full) {
+		if (bp->port.supported & SUPPORTED_2500baseX_Full) {
 			bp->link_params.req_line_speed = SPEED_2500;
-			bp->advertising = (ADVERTISED_2500baseX_Full |
-					   ADVERTISED_TP);
+			bp->port.advertising = (ADVERTISED_2500baseX_Full |
+						ADVERTISED_TP);
 		} else {
 			BNX2X_ERR("NVRAM config error. "
 				  "Invalid link_config 0x%x"
 				  "  speed_cap_mask 0x%x\n",
-				  bp->link_config,
+				  bp->port.link_config,
 				  bp->link_params.speed_cap_mask);
 			return;
 		}
@@ -5337,15 +6212,15 @@ static void bnx2x_link_settings_requested(struct bnx2x *bp)
 	case PORT_FEATURE_LINK_SPEED_10G_CX4:
 	case PORT_FEATURE_LINK_SPEED_10G_KX4:
 	case PORT_FEATURE_LINK_SPEED_10G_KR:
-		if (bp->supported & SUPPORTED_10000baseT_Full) {
+		if (bp->port.supported & SUPPORTED_10000baseT_Full) {
 			bp->link_params.req_line_speed = SPEED_10000;
-			bp->advertising = (ADVERTISED_10000baseT_Full |
-					   ADVERTISED_FIBRE);
+			bp->port.advertising = (ADVERTISED_10000baseT_Full |
+						ADVERTISED_FIBRE);
 		} else {
 			BNX2X_ERR("NVRAM config error. "
 				  "Invalid link_config 0x%x"
 				  "  speed_cap_mask 0x%x\n",
-				  bp->link_config,
+				  bp->port.link_config,
 				  bp->link_params.speed_cap_mask);
 			return;
 		}
@@ -5354,64 +6229,33 @@ static void bnx2x_link_settings_requested(struct bnx2x *bp)
 	default:
 		BNX2X_ERR("NVRAM config error. "
 			  "BAD link speed link_config 0x%x\n",
-			  bp->link_config);
+			  bp->port.link_config);
 		bp->link_params.req_line_speed = SPEED_AUTO_NEG;
-		bp->advertising = bp->supported;
+		bp->port.advertising = bp->port.supported;
 		break;
 	}
 
-	bp->link_params.req_flow_ctrl = (bp->link_config &
-			     PORT_FEATURE_FLOW_CONTROL_MASK);
+	bp->link_params.req_flow_ctrl = (bp->port.link_config &
+					 PORT_FEATURE_FLOW_CONTROL_MASK);
 	if ((bp->link_params.req_flow_ctrl == FLOW_CTRL_AUTO) &&
-	    (!bp->supported & SUPPORTED_Autoneg))
+	    (!bp->port.supported & SUPPORTED_Autoneg))
 		bp->link_params.req_flow_ctrl = FLOW_CTRL_NONE;
 
 	BNX2X_DEV_INFO("req_line_speed %d  req_duplex %d  req_flow_ctrl 0x%x"
 		       "  advertising 0x%x\n",
 		       bp->link_params.req_line_speed,
 		       bp->link_params.req_duplex,
-		       bp->link_params.req_flow_ctrl, bp->advertising);
+		       bp->link_params.req_flow_ctrl, bp->port.advertising);
 }
 
-static void bnx2x_get_hwinfo(struct bnx2x *bp)
+static void __devinit bnx2x_get_port_hwinfo(struct bnx2x *bp)
 {
-	u32 val, val2, val3, val4, id;
-	int port = bp->port;
-
-	bp->shmem_base = REG_RD(bp, MISC_REG_SHARED_MEM_ADDR);
-	BNX2X_DEV_INFO("shmem offset is %x\n", bp->shmem_base);
-
-	/* Get the chip revision id and number. */
-	/* chip num:16-31, rev:12-15, metal:4-11, bond_id:0-3 */
-	val = REG_RD(bp, MISC_REG_CHIP_NUM);
-	id = ((val & 0xffff) << 16);
-	val = REG_RD(bp, MISC_REG_CHIP_REV);
-	id |= ((val & 0xf) << 12);
-	val = REG_RD(bp, MISC_REG_CHIP_METAL);
-	id |= ((val & 0xff) << 4);
-	REG_RD(bp, MISC_REG_BOND_ID);
-	id |= (val & 0xf);
-	bp->chip_id = id;
-	BNX2X_DEV_INFO("chip ID is %x\n", id);
+	int port = BP_PORT(bp);
+	u32 val, val2;
 
 	bp->link_params.bp = bp;
+	bp->link_params.port = port;
 
-	if (!bp->shmem_base || (bp->shmem_base != 0xAF900)) {
-		BNX2X_DEV_INFO("MCP not active\n");
-		nomcp = 1;
-		goto set_mac;
-	}
-
-	val = SHMEM_RD(bp, validity_map[port]);
-	if ((val & (SHR_MEM_VALIDITY_DEV_INFO | SHR_MEM_VALIDITY_MB))
-		!= (SHR_MEM_VALIDITY_DEV_INFO | SHR_MEM_VALIDITY_MB))
-		BNX2X_ERR("BAD MCP validity signature\n");
-
-	bp->fw_seq = (SHMEM_RD(bp, func_mb[port].drv_mb_header) &
-		      DRV_MSG_SEQ_NUMBER_MASK);
-
-	bp->hw_config = SHMEM_RD(bp, dev_info.shared_hw_config.config);
-	bp->board = SHMEM_RD(bp, dev_info.shared_hw_config.board);
 	bp->link_params.serdes_config =
 		SHMEM_RD(bp, dev_info.port_hw_config[port].serdes_config);
 	bp->link_params.lane_config =
@@ -5423,19 +6267,18 @@ static void bnx2x_get_hwinfo(struct bnx2x *bp)
 		SHMEM_RD(bp,
 			 dev_info.port_hw_config[port].speed_capability_mask);
 
-	bp->link_config =
+	bp->port.link_config =
 		SHMEM_RD(bp, dev_info.port_feature_config[port].link_config);
 
-	BNX2X_DEV_INFO("serdes_config (%08x)  lane_config (%08x)\n"
-	     KERN_INFO "  ext_phy_config (%08x)  speed_cap_mask (%08x)"
-		       "  link_config (%08x)\n",
+	BNX2X_DEV_INFO("serdes_config 0x%08x  lane_config 0x%08x\n"
+	     KERN_INFO "  ext_phy_config 0x%08x  speed_cap_mask 0x%08x"
+		       "  link_config 0x%08x\n",
 		       bp->link_params.serdes_config,
 		       bp->link_params.lane_config,
 		       bp->link_params.ext_phy_config,
-		       bp->link_params.speed_cap_mask,
-		       bp->link_config);
+		       bp->link_params.speed_cap_mask, bp->port.link_config);
 
-	bp->link_params.switch_cfg = (bp->link_config &
+	bp->link_params.switch_cfg = (bp->port.link_config &
 				      PORT_FEATURE_CONNECTED_SWITCH_MASK);
 	bnx2x_link_settings_supported(bp, bp->link_params.switch_cfg);
 
@@ -5451,43 +6294,126 @@ static void bnx2x_get_hwinfo(struct bnx2x *bp)
 	bp->dev->dev_addr[5] = (u8)(val & 0xff);
 	memcpy(bp->link_params.mac_addr, bp->dev->dev_addr, ETH_ALEN);
 	memcpy(bp->dev->perm_addr, bp->dev->dev_addr, ETH_ALEN);
+}
+
+static int __devinit bnx2x_get_hwinfo(struct bnx2x *bp)
+{
+	int func = BP_FUNC(bp);
+	u32 val, val2;
+	int rc = 0;
 
+	bnx2x_get_common_hwinfo(bp);
 
+	bp->e1hov = 0;
+	bp->e1hmf = 0;
+	if (CHIP_IS_E1H(bp)) {
+		bp->mf_config =
+			SHMEM_RD(bp, mf_cfg.func_mf_config[func].config);
 
-	val = SHMEM_RD(bp, dev_info.shared_hw_config.part_num);
-	val2 = SHMEM_RD(bp, dev_info.shared_hw_config.part_num[4]);
-	val3 = SHMEM_RD(bp, dev_info.shared_hw_config.part_num[8]);
-	val4 = SHMEM_RD(bp, dev_info.shared_hw_config.part_num[12]);
+		val =
+		   (SHMEM_RD(bp, mf_cfg.func_mf_config[func].e1hov_tag) &
+		    FUNC_MF_CFG_E1HOV_TAG_MASK);
+		if (val != FUNC_MF_CFG_E1HOV_TAG_DEFAULT) {
 
-	printk(KERN_INFO PFX "part number %X-%X-%X-%X\n",
-	       val, val2, val3, val4);
+			bp->e1hov = val;
+			bp->e1hmf = 1;
+			BNX2X_DEV_INFO("MF mode  E1HOV for func %d is %d "
+				       "(0x%04x)\n",
+				       func, bp->e1hov, bp->e1hov);
+		} else {
+			BNX2X_DEV_INFO("Single function mode\n");
+			if (BP_E1HVN(bp)) {
+				BNX2X_ERR("!!!  No valid E1HOV for func %d,"
+					  "  aborting\n", func);
+				rc = -EPERM;
+			}
+		}
+	}
 
-	/* bc ver */
-	if (!nomcp) {
-		bp->bc_ver = val = ((SHMEM_RD(bp, dev_info.bc_rev)) >> 8);
-		BNX2X_DEV_INFO("bc_ver %X\n", val);
-		if (val < BNX2X_BC_VER) {
-			/* for now only warn
-			 * later we might need to enforce this */
-			BNX2X_ERR("This driver needs bc_ver %X but found %X,"
-				  " please upgrade BC\n", BNX2X_BC_VER, val);
+	if (!BP_NOMCP(bp)) {
+		bnx2x_get_port_hwinfo(bp);
+
+		bp->fw_seq = (SHMEM_RD(bp, func_mb[func].drv_mb_header) &
+			      DRV_MSG_SEQ_NUMBER_MASK);
+		BNX2X_DEV_INFO("fw_seq 0x%08x\n", bp->fw_seq);
+	}
+
+	if (IS_E1HMF(bp)) {
+		val2 = SHMEM_RD(bp, mf_cfg.func_mf_config[func].mac_upper);
+		val = SHMEM_RD(bp,  mf_cfg.func_mf_config[func].mac_lower);
+		if ((val2 != FUNC_MF_CFG_UPPERMAC_DEFAULT) &&
+		    (val != FUNC_MF_CFG_LOWERMAC_DEFAULT)) {
+			bp->dev->dev_addr[0] = (u8)(val2 >> 8 & 0xff);
+			bp->dev->dev_addr[1] = (u8)(val2 & 0xff);
+			bp->dev->dev_addr[2] = (u8)(val >> 24 & 0xff);
+			bp->dev->dev_addr[3] = (u8)(val >> 16 & 0xff);
+			bp->dev->dev_addr[4] = (u8)(val >> 8  & 0xff);
+			bp->dev->dev_addr[5] = (u8)(val & 0xff);
+			memcpy(bp->link_params.mac_addr, bp->dev->dev_addr,
+			       ETH_ALEN);
+			memcpy(bp->dev->perm_addr, bp->dev->dev_addr,
+			       ETH_ALEN);
 		}
-	} else {
-		bp->bc_ver = 0;
+
+		return rc;
 	}
 
-	val = REG_RD(bp, MCP_REG_MCPR_NVM_CFG4);
-	bp->flash_size = (NVRAM_1MB_SIZE << (val & MCPR_NVM_CFG4_FLASH_SIZE));
-	BNX2X_DEV_INFO("flash_size 0x%x (%d)\n",
-		       bp->flash_size, bp->flash_size);
+	if (BP_NOMCP(bp)) {
+		/* only supposed to happen on emulation/FPGA */
+		BNX2X_ERR("warning rendom MAC workaround active\n");
+		random_ether_addr(bp->dev->dev_addr);
+		memcpy(bp->dev->perm_addr, bp->dev->dev_addr, ETH_ALEN);
+	}
 
-	return;
+	return rc;
+}
+
+static int __devinit bnx2x_init_bp(struct bnx2x *bp)
+{
+	int func = BP_FUNC(bp);
+	int rc;
+
+	if (nomcp)
+		bp->flags |= NO_MCP_FLAG;
 
-set_mac: /* only supposed to happen on emulation/FPGA */
-	BNX2X_ERR("warning rendom MAC workaround active\n");
-	random_ether_addr(bp->dev->dev_addr);
-	memcpy(bp->dev->perm_addr, bp->dev->dev_addr, 6);
+	mutex_init(&bp->port.phy_mutex);
 
+	INIT_WORK(&bp->sp_task, bnx2x_sp_task);
+	INIT_WORK(&bp->reset_task, bnx2x_reset_task);
+
+	rc = bnx2x_get_hwinfo(bp);
+
+	/* need to reset chip if undi was active */
+	if (!BP_NOMCP(bp))
+		bnx2x_undi_unload(bp);
+
+	if (CHIP_REV_IS_FPGA(bp))
+		printk(KERN_ERR PFX "FPGA detected\n");
+
+	if (BP_NOMCP(bp) && (func == 0))
+		printk(KERN_ERR PFX
+		       "MCP disabled, must load devices in order!\n");
+
+	bp->tx_ring_size = MAX_TX_AVAIL;
+	bp->rx_ring_size = MAX_RX_AVAIL;
+
+	bp->rx_csum = 1;
+	bp->rx_offset = 0;
+
+	bp->tx_ticks = 50;
+	bp->rx_ticks = 25;
+
+	bp->stats_ticks = 1000000 & 0xffff00;
+
+	bp->timer_interval = (CHIP_REV_IS_SLOW(bp) ? 5*HZ : HZ);
+	bp->current_interval = (poll ? poll : bp->timer_interval);
+
+	init_timer(&bp->timer);
+	bp->timer.expires = jiffies + bp->current_interval;
+	bp->timer.data = (unsigned long) bp;
+	bp->timer.function = bnx2x_timer;
+
+	return rc;
 }
 
 /*
@@ -5500,8 +6426,8 @@ static int bnx2x_get_settings(struct net_device *dev, struct ethtool_cmd *cmd)
 {
 	struct bnx2x *bp = netdev_priv(dev);
 
-	cmd->supported = bp->supported;
-	cmd->advertising = bp->advertising;
+	cmd->supported = bp->port.supported;
+	cmd->advertising = bp->port.advertising;
 
 	if (netif_carrier_ok(dev)) {
 		cmd->speed = bp->link_vars.line_speed;
@@ -5510,6 +6436,14 @@ static int bnx2x_get_settings(struct net_device *dev, struct ethtool_cmd *cmd)
 		cmd->speed = bp->link_params.req_line_speed;
 		cmd->duplex = bp->link_params.req_duplex;
 	}
+	if (IS_E1HMF(bp)) {
+		u16 vn_max_rate;
+
+		vn_max_rate = ((bp->mf_config & FUNC_MF_CFG_MAX_BW_MASK) >>
+				FUNC_MF_CFG_MAX_BW_SHIFT) * 100;
+		if (vn_max_rate < cmd->speed)
+			cmd->speed = vn_max_rate;
+	}
 
 	if (bp->link_params.switch_cfg == SWITCH_CFG_10G) {
 		u32 ext_phy_type =
@@ -5541,7 +6475,7 @@ static int bnx2x_get_settings(struct net_device *dev, struct ethtool_cmd *cmd)
 	} else
 		cmd->port = PORT_TP;
 
-	cmd->phy_address = bp->phy_addr;
+	cmd->phy_address = bp->port.phy_addr;
 	cmd->transceiver = XCVR_INTERNAL;
 
 	if (bp->link_params.req_line_speed == SPEED_AUTO_NEG)
@@ -5568,6 +6502,9 @@ static int bnx2x_set_settings(struct net_device *dev, struct ethtool_cmd *cmd)
 	struct bnx2x *bp = netdev_priv(dev);
 	u32 advertising;
 
+	if (IS_E1HMF(bp))
+		return 0;
+
 	DP(NETIF_MSG_LINK, "ethtool_cmd: cmd %d\n"
 	   DP_LEVEL "  supported 0x%x  advertising 0x%x  speed %d\n"
 	   DP_LEVEL "  duplex %d  port %d  phy_address %d  transceiver %d\n"
@@ -5577,24 +6514,25 @@ static int bnx2x_set_settings(struct net_device *dev, struct ethtool_cmd *cmd)
 	   cmd->autoneg, cmd->maxtxpkt, cmd->maxrxpkt);
 
 	if (cmd->autoneg == AUTONEG_ENABLE) {
-		if (!(bp->supported & SUPPORTED_Autoneg)) {
-			DP(NETIF_MSG_LINK, "Aotoneg not supported\n");
+		if (!(bp->port.supported & SUPPORTED_Autoneg)) {
+			DP(NETIF_MSG_LINK, "Autoneg not supported\n");
 			return -EINVAL;
 		}
 
 		/* advertise the requested speed and duplex if supported */
-		cmd->advertising &= bp->supported;
+		cmd->advertising &= bp->port.supported;
 
 		bp->link_params.req_line_speed = SPEED_AUTO_NEG;
 		bp->link_params.req_duplex = DUPLEX_FULL;
-		bp->advertising |= (ADVERTISED_Autoneg | cmd->advertising);
+		bp->port.advertising |= (ADVERTISED_Autoneg |
+					 cmd->advertising);
 
 	} else { /* forced speed */
 		/* advertise the requested speed and duplex if supported */
 		switch (cmd->speed) {
 		case SPEED_10:
 			if (cmd->duplex == DUPLEX_FULL) {
-				if (!(bp->supported &
+				if (!(bp->port.supported &
 				      SUPPORTED_10baseT_Full)) {
 					DP(NETIF_MSG_LINK,
 					   "10M full not supported\n");
@@ -5604,7 +6542,7 @@ static int bnx2x_set_settings(struct net_device *dev, struct ethtool_cmd *cmd)
 				advertising = (ADVERTISED_10baseT_Full |
 					       ADVERTISED_TP);
 			} else {
-				if (!(bp->supported &
+				if (!(bp->port.supported &
 				      SUPPORTED_10baseT_Half)) {
 					DP(NETIF_MSG_LINK,
 					   "10M half not supported\n");
@@ -5618,7 +6556,7 @@ static int bnx2x_set_settings(struct net_device *dev, struct ethtool_cmd *cmd)
 
 		case SPEED_100:
 			if (cmd->duplex == DUPLEX_FULL) {
-				if (!(bp->supported &
+				if (!(bp->port.supported &
 						SUPPORTED_100baseT_Full)) {
 					DP(NETIF_MSG_LINK,
 					   "100M full not supported\n");
@@ -5628,7 +6566,7 @@ static int bnx2x_set_settings(struct net_device *dev, struct ethtool_cmd *cmd)
 				advertising = (ADVERTISED_100baseT_Full |
 					       ADVERTISED_TP);
 			} else {
-				if (!(bp->supported &
+				if (!(bp->port.supported &
 						SUPPORTED_100baseT_Half)) {
 					DP(NETIF_MSG_LINK,
 					   "100M half not supported\n");
@@ -5646,7 +6584,7 @@ static int bnx2x_set_settings(struct net_device *dev, struct ethtool_cmd *cmd)
 				return -EINVAL;
 			}
 
-			if (!(bp->supported & SUPPORTED_1000baseT_Full)) {
+			if (!(bp->port.supported & SUPPORTED_1000baseT_Full)) {
 				DP(NETIF_MSG_LINK, "1G full not supported\n");
 				return -EINVAL;
 			}
@@ -5662,7 +6600,7 @@ static int bnx2x_set_settings(struct net_device *dev, struct ethtool_cmd *cmd)
 				return -EINVAL;
 			}
 
-			if (!(bp->supported & SUPPORTED_2500baseX_Full)) {
+			if (!(bp->port.supported & SUPPORTED_2500baseX_Full)) {
 				DP(NETIF_MSG_LINK,
 				   "2.5G full not supported\n");
 				return -EINVAL;
@@ -5678,7 +6616,7 @@ static int bnx2x_set_settings(struct net_device *dev, struct ethtool_cmd *cmd)
 				return -EINVAL;
 			}
 
-			if (!(bp->supported & SUPPORTED_10000baseT_Full)) {
+			if (!(bp->port.supported & SUPPORTED_10000baseT_Full)) {
 				DP(NETIF_MSG_LINK, "10G full not supported\n");
 				return -EINVAL;
 			}
@@ -5694,16 +6632,18 @@ static int bnx2x_set_settings(struct net_device *dev, struct ethtool_cmd *cmd)
 
 		bp->link_params.req_line_speed = cmd->speed;
 		bp->link_params.req_duplex = cmd->duplex;
-		bp->advertising = advertising;
+		bp->port.advertising = advertising;
 	}
 
 	DP(NETIF_MSG_LINK, "req_line_speed %d\n"
 	   DP_LEVEL "  req_duplex %d  advertising 0x%x\n",
 	   bp->link_params.req_line_speed, bp->link_params.req_duplex,
-	   bp->advertising);
+	   bp->port.advertising);
 
-	bnx2x_stop_stats(bp);
-	bnx2x_link_set(bp);
+	if (netif_running(dev)) {
+		bnx2x_stop_stats(bp);
+		bnx2x_link_set(bp);
+	}
 
 	return 0;
 }
@@ -5720,21 +6660,23 @@ static void bnx2x_get_drvinfo(struct net_device *dev,
 	strcpy(info->version, DRV_MODULE_VERSION);
 
 	phy_fw_ver[0] = '\0';
-	bnx2x_phy_hw_lock(bp);
-	bnx2x_get_ext_phy_fw_version(&bp->link_params,
-				     (bp->state != BNX2X_STATE_CLOSED),
-				     phy_fw_ver, PHY_FW_VER_LEN);
-	bnx2x_phy_hw_unlock(bp);
+	if (bp->port.pmf) {
+		bnx2x_phy_hw_lock(bp);
+		bnx2x_get_ext_phy_fw_version(&bp->link_params,
+					     (bp->state != BNX2X_STATE_CLOSED),
+					     phy_fw_ver, PHY_FW_VER_LEN);
+		bnx2x_phy_hw_unlock(bp);
+	}
 
 	snprintf(info->fw_version, 32, "%d.%d.%d:%d BC:%x%s%s",
 		 BCM_5710_FW_MAJOR_VERSION, BCM_5710_FW_MINOR_VERSION,
 		 BCM_5710_FW_REVISION_VERSION,
-		 BCM_5710_FW_COMPILE_FLAGS, bp->bc_ver,
+		 BCM_5710_FW_COMPILE_FLAGS, bp->common.bc_ver,
 		 ((phy_fw_ver[0] != '\0')? " PHY:":""), phy_fw_ver);
 	strcpy(info->bus_info, pci_name(bp->pdev));
 	info->n_stats = BNX2X_NUM_STATS;
 	info->testinfo_len = BNX2X_NUM_TESTS;
-	info->eedump_len = bp->flash_size;
+	info->eedump_len = bp->common.flash_size;
 	info->regdump_len = 0;
 }
 
@@ -5767,9 +6709,9 @@ static int bnx2x_set_wol(struct net_device *dev, struct ethtool_wolinfo *wol)
 			return -EINVAL;
 
 		bp->wol = 1;
-	} else {
+	} else
 		bp->wol = 0;
-	}
+
 	return 0;
 }
 
@@ -5792,13 +6734,13 @@ static int bnx2x_nway_reset(struct net_device *dev)
 {
 	struct bnx2x *bp = netdev_priv(dev);
 
-	if (bp->state != BNX2X_STATE_OPEN) {
-		DP(NETIF_MSG_PROBE, "state is %x, returning\n", bp->state);
-		return -EAGAIN;
-	}
+	if (!bp->port.pmf)
+		return 0;
 
-	bnx2x_stop_stats(bp);
-	bnx2x_link_set(bp);
+	if (netif_running(dev)) {
+		bnx2x_stop_stats(bp);
+		bnx2x_link_set(bp);
+	}
 
 	return 0;
 }
@@ -5807,12 +6749,12 @@ static int bnx2x_get_eeprom_len(struct net_device *dev)
 {
 	struct bnx2x *bp = netdev_priv(dev);
 
-	return bp->flash_size;
+	return bp->common.flash_size;
 }
 
 static int bnx2x_acquire_nvram_lock(struct bnx2x *bp)
 {
-	int port = bp->port;
+	int port = BP_PORT(bp);
 	int count, i;
 	u32 val = 0;
 
@@ -5834,7 +6776,7 @@ static int bnx2x_acquire_nvram_lock(struct bnx2x *bp)
 	}
 
 	if (!(val & (MCPR_NVM_SW_ARB_ARB_ARB1 << port))) {
-		DP(NETIF_MSG_NVM, "cannot get access to nvram interface\n");
+		DP(BNX2X_MSG_NVM, "cannot get access to nvram interface\n");
 		return -EBUSY;
 	}
 
@@ -5843,7 +6785,7 @@ static int bnx2x_acquire_nvram_lock(struct bnx2x *bp)
 
 static int bnx2x_release_nvram_lock(struct bnx2x *bp)
 {
-	int port = bp->port;
+	int port = BP_PORT(bp);
 	int count, i;
 	u32 val = 0;
 
@@ -5865,7 +6807,7 @@ static int bnx2x_release_nvram_lock(struct bnx2x *bp)
 	}
 
 	if (val & (MCPR_NVM_SW_ARB_ARB_ARB1 << port)) {
-		DP(NETIF_MSG_NVM, "cannot free access to nvram interface\n");
+		DP(BNX2X_MSG_NVM, "cannot free access to nvram interface\n");
 		return -EBUSY;
 	}
 
@@ -5929,7 +6871,6 @@ static int bnx2x_nvram_read_dword(struct bnx2x *bp, u32 offset, u32 *ret_val,
 
 		if (val & MCPR_NVM_COMMAND_DONE) {
 			val = REG_RD(bp, MCP_REG_MCPR_NVM_READ);
-			DP(NETIF_MSG_NVM, "val 0x%08x\n", val);
 			/* we read nvram data in cpu order
 			 * but ethtool sees it as an array of bytes
 			 * converting to big-endian will do the work */
@@ -5951,16 +6892,16 @@ static int bnx2x_nvram_read(struct bnx2x *bp, u32 offset, u8 *ret_buf,
 	u32 val;
 
 	if ((offset & 0x03) || (buf_size & 0x03) || (buf_size == 0)) {
-		DP(NETIF_MSG_NVM,
+		DP(BNX2X_MSG_NVM,
 		   "Invalid parameter: offset 0x%x  buf_size 0x%x\n",
 		   offset, buf_size);
 		return -EINVAL;
 	}
 
-	if (offset + buf_size > bp->flash_size) {
-		DP(NETIF_MSG_NVM, "Invalid parameter: offset (0x%x) +"
+	if (offset + buf_size > bp->common.flash_size) {
+		DP(BNX2X_MSG_NVM, "Invalid parameter: offset (0x%x) +"
 				  " buf_size (0x%x) > flash_size (0x%x)\n",
-		   offset, buf_size, bp->flash_size);
+		   offset, buf_size, bp->common.flash_size);
 		return -EINVAL;
 	}
 
@@ -6004,7 +6945,7 @@ static int bnx2x_get_eeprom(struct net_device *dev,
 	struct bnx2x *bp = netdev_priv(dev);
 	int rc;
 
-	DP(NETIF_MSG_NVM, "ethtool_eeprom: cmd %d\n"
+	DP(BNX2X_MSG_NVM, "ethtool_eeprom: cmd %d\n"
 	   DP_LEVEL "  magic 0x%x  offset 0x%x (%d)  len 0x%x (%d)\n",
 	   eeprom->cmd, eeprom->magic, eeprom->offset, eeprom->offset,
 	   eeprom->len, eeprom->len);
@@ -6066,10 +7007,10 @@ static int bnx2x_nvram_write1(struct bnx2x *bp, u32 offset, u8 *data_buf,
 	u32 align_offset;
 	u32 val;
 
-	if (offset + buf_size > bp->flash_size) {
-		DP(NETIF_MSG_NVM, "Invalid parameter: offset (0x%x) +"
+	if (offset + buf_size > bp->common.flash_size) {
+		DP(BNX2X_MSG_NVM, "Invalid parameter: offset (0x%x) +"
 				  " buf_size (0x%x) > flash_size (0x%x)\n",
-		   offset, buf_size, bp->flash_size);
+		   offset, buf_size, bp->common.flash_size);
 		return -EINVAL;
 	}
 
@@ -6093,8 +7034,6 @@ static int bnx2x_nvram_write1(struct bnx2x *bp, u32 offset, u8 *data_buf,
 		 * convert it back to cpu order */
 		val = be32_to_cpu(val);
 
-		DP(NETIF_MSG_NVM, "val 0x%08x\n", val);
-
 		rc = bnx2x_nvram_write_dword(bp, align_offset, val,
 					     cmd_flags);
 	}
@@ -6114,21 +7053,20 @@ static int bnx2x_nvram_write(struct bnx2x *bp, u32 offset, u8 *data_buf,
 	u32 val;
 	u32 written_so_far;
 
-	if (buf_size == 1) {    /* ethtool */
+	if (buf_size == 1)	/* ethtool */
 		return bnx2x_nvram_write1(bp, offset, data_buf, buf_size);
-	}
 
 	if ((offset & 0x03) || (buf_size & 0x03) || (buf_size == 0)) {
-		DP(NETIF_MSG_NVM,
+		DP(BNX2X_MSG_NVM,
 		   "Invalid parameter: offset 0x%x  buf_size 0x%x\n",
 		   offset, buf_size);
 		return -EINVAL;
 	}
 
-	if (offset + buf_size > bp->flash_size) {
-		DP(NETIF_MSG_NVM, "Invalid parameter: offset (0x%x) +"
+	if (offset + buf_size > bp->common.flash_size) {
+		DP(BNX2X_MSG_NVM, "Invalid parameter: offset (0x%x) +"
 				  " buf_size (0x%x) > flash_size (0x%x)\n",
-		   offset, buf_size, bp->flash_size);
+		   offset, buf_size, bp->common.flash_size);
 		return -EINVAL;
 	}
 
@@ -6151,7 +7089,6 @@ static int bnx2x_nvram_write(struct bnx2x *bp, u32 offset, u8 *data_buf,
 			cmd_flags |= MCPR_NVM_COMMAND_FIRST;
 
 		memcpy(&val, data_buf, 4);
-		DP(NETIF_MSG_NVM, "val 0x%08x\n", val);
 
 		rc = bnx2x_nvram_write_dword(bp, offset, val, cmd_flags);
 
@@ -6175,7 +7112,7 @@ static int bnx2x_set_eeprom(struct net_device *dev,
 	struct bnx2x *bp = netdev_priv(dev);
 	int rc;
 
-	DP(NETIF_MSG_NVM, "ethtool_eeprom: cmd %d\n"
+	DP(BNX2X_MSG_NVM, "ethtool_eeprom: cmd %d\n"
 	   DP_LEVEL "  magic 0x%x  offset 0x%x (%d)  len 0x%x (%d)\n",
 	   eeprom->cmd, eeprom->magic, eeprom->offset, eeprom->offset,
 	   eeprom->len, eeprom->len);
@@ -6183,20 +7120,23 @@ static int bnx2x_set_eeprom(struct net_device *dev,
 	/* parameters already validated in ethtool_set_eeprom */
 
 	/* If the magic number is PHY (0x00504859) upgrade the PHY FW */
-	if (eeprom->magic == 0x00504859) {
-
-		bnx2x_phy_hw_lock(bp);
-		rc = bnx2x_flash_download(bp, bp->port,
-				     bp->link_params.ext_phy_config,
-				     (bp->state != BNX2X_STATE_CLOSED),
-				     eebuf, eeprom->len);
-		rc |= bnx2x_link_reset(&bp->link_params,
-				       &bp->link_vars);
-		rc |= bnx2x_phy_init(&bp->link_params,
-				     &bp->link_vars);
-		bnx2x_phy_hw_unlock(bp);
-
-	} else
+	if (eeprom->magic == 0x00504859)
+		if (bp->port.pmf) {
+
+			bnx2x_phy_hw_lock(bp);
+			rc = bnx2x_flash_download(bp, BP_PORT(bp),
+					     bp->link_params.ext_phy_config,
+					     (bp->state != BNX2X_STATE_CLOSED),
+					     eebuf, eeprom->len);
+				rc |= bnx2x_link_reset(&bp->link_params,
+						       &bp->link_vars);
+				rc |= bnx2x_phy_init(&bp->link_params,
+						     &bp->link_vars);
+			bnx2x_phy_hw_unlock(bp);
+
+		} else /* Only the PMF can access the PHY */
+			return -EINVAL;
+	else
 		rc = bnx2x_nvram_write(bp, eeprom->offset, eebuf, eeprom->len);
 
 	return rc;
@@ -6234,7 +7174,7 @@ static int bnx2x_set_coalesce(struct net_device *dev,
 		bp->stats_ticks = 0xffff00;
 	bp->stats_ticks &= 0xffff00;
 
-	if (netif_running(bp->dev))
+	if (netif_running(dev))
 		bnx2x_update_coalesce(bp);
 
 	return 0;
@@ -6261,6 +7201,7 @@ static int bnx2x_set_ringparam(struct net_device *dev,
 			       struct ethtool_ringparam *ering)
 {
 	struct bnx2x *bp = netdev_priv(dev);
+	int rc = 0;
 
 	if ((ering->rx_pending > MAX_RX_AVAIL) ||
 	    (ering->tx_pending > MAX_TX_AVAIL) ||
@@ -6270,12 +7211,12 @@ static int bnx2x_set_ringparam(struct net_device *dev,
 	bp->rx_ring_size = ering->rx_pending;
 	bp->tx_ring_size = ering->tx_pending;
 
-	if (netif_running(bp->dev)) {
-		bnx2x_nic_unload(bp, 0);
-		bnx2x_nic_load(bp, 0);
+	if (netif_running(dev)) {
+		bnx2x_nic_unload(bp, UNLOAD_NORMAL);
+		rc = bnx2x_nic_load(bp, LOAD_NORMAL);
 	}
 
-	return 0;
+	return rc;
 }
 
 static void bnx2x_get_pauseparam(struct net_device *dev,
@@ -6301,6 +7242,9 @@ static int bnx2x_set_pauseparam(struct net_device *dev,
 {
 	struct bnx2x *bp = netdev_priv(dev);
 
+	if (IS_E1HMF(bp))
+		return 0;
+
 	DP(NETIF_MSG_LINK, "ethtool_pauseparam: cmd %d\n"
 	   DP_LEVEL "  autoneg %d  rx_pause %d  tx_pause %d\n",
 	   epause->cmd, epause->autoneg, epause->rx_pause, epause->tx_pause);
@@ -6317,7 +7261,7 @@ static int bnx2x_set_pauseparam(struct net_device *dev,
 		bp->link_params.req_flow_ctrl = FLOW_CTRL_NONE;
 
 	if (epause->autoneg) {
-		if (!(bp->supported & SUPPORTED_Autoneg)) {
+		if (!(bp->port.supported & SUPPORTED_Autoneg)) {
 			DP(NETIF_MSG_LINK, "Autoneg not supported\n");
 			return -EINVAL;
 		}
@@ -6328,8 +7272,11 @@ static int bnx2x_set_pauseparam(struct net_device *dev,
 
 	DP(NETIF_MSG_LINK,
 	   "req_flow_ctrl 0x%x\n", bp->link_params.req_flow_ctrl);
-	bnx2x_stop_stats(bp);
-	bnx2x_link_set(bp);
+
+	if (netif_running(dev)) {
+		bnx2x_stop_stats(bp);
+		bnx2x_link_set(bp);
+	}
 
 	return 0;
 }
@@ -6531,18 +7478,25 @@ static void bnx2x_get_ethtool_stats(struct net_device *dev,
 static int bnx2x_phys_id(struct net_device *dev, u32 data)
 {
 	struct bnx2x *bp = netdev_priv(dev);
+	int port = BP_PORT(bp);
 	int i;
 
+	if (!netif_running(dev))
+		return 0;
+
+	if (!bp->port.pmf)
+		return 0;
+
 	if (data == 0)
 		data = 2;
 
 	for (i = 0; i < (data * 2); i++) {
 		if ((i % 2) == 0)
-			bnx2x_set_led(bp, bp->port, LED_MODE_OPER, SPEED_1000,
+			bnx2x_set_led(bp, port, LED_MODE_OPER, SPEED_1000,
 				      bp->link_params.hw_led_mode,
 				      bp->link_params.chip_id);
 		else
-			bnx2x_set_led(bp, bp->port, LED_MODE_OFF, 0,
+			bnx2x_set_led(bp, port, LED_MODE_OFF, 0,
 				      bp->link_params.hw_led_mode,
 				      bp->link_params.chip_id);
 
@@ -6552,7 +7506,7 @@ static int bnx2x_phys_id(struct net_device *dev, u32 data)
 	}
 
 	if (bp->link_vars.link_up)
-		bnx2x_set_led(bp, bp->port, LED_MODE_OPER,
+		bnx2x_set_led(bp, port, LED_MODE_OPER,
 			      bp->link_vars.line_speed,
 			      bp->link_params.hw_led_mode,
 			      bp->link_params.chip_id);
@@ -6609,117 +7563,40 @@ static int bnx2x_set_power_state(struct bnx2x *bp, pci_power_t state)
 
 	switch (state) {
 	case PCI_D0:
-		pci_write_config_word(bp->pdev,
-				      bp->pm_cap + PCI_PM_CTRL,
+		pci_write_config_word(bp->pdev, bp->pm_cap + PCI_PM_CTRL,
 				      ((pmcsr & ~PCI_PM_CTRL_STATE_MASK) |
 				       PCI_PM_CTRL_PME_STATUS));
 
 		if (pmcsr & PCI_PM_CTRL_STATE_MASK)
 		/* delay required during transition out of D3hot */
 			msleep(20);
-		break;
-
-	case PCI_D3hot:
-		pmcsr &= ~PCI_PM_CTRL_STATE_MASK;
-		pmcsr |= 3;
-
-		if (bp->wol)
-			pmcsr |= PCI_PM_CTRL_PME_ENABLE;
-
-		pci_write_config_word(bp->pdev, bp->pm_cap + PCI_PM_CTRL,
-				      pmcsr);
-
-		/* No more memory access after this point until
-		* device is brought back to D0.
-		*/
-		break;
-
-	default:
-		return -EINVAL;
-	}
-	return 0;
-}
-
-/*
- * net_device service functions
- */
-
-/* called with netif_tx_lock from set_multicast */
-static void bnx2x_set_rx_mode(struct net_device *dev)
-{
-	struct bnx2x *bp = netdev_priv(dev);
-	u32 rx_mode = BNX2X_RX_MODE_NORMAL;
-
-	DP(NETIF_MSG_IFUP, "called dev->flags = %x\n", dev->flags);
-
-	if (dev->flags & IFF_PROMISC)
-		rx_mode = BNX2X_RX_MODE_PROMISC;
-
-	else if ((dev->flags & IFF_ALLMULTI) ||
-		 (dev->mc_count > BNX2X_MAX_MULTICAST))
-		rx_mode = BNX2X_RX_MODE_ALLMULTI;
-
-	else { /* some multicasts */
-		int i, old, offset;
-		struct dev_mc_list *mclist;
-		struct mac_configuration_cmd *config =
-						bnx2x_sp(bp, mcast_config);
-
-		for (i = 0, mclist = dev->mc_list;
-		     mclist && (i < dev->mc_count);
-		     i++, mclist = mclist->next) {
-
-			config->config_table[i].cam_entry.msb_mac_addr =
-					swab16(*(u16 *)&mclist->dmi_addr[0]);
-			config->config_table[i].cam_entry.middle_mac_addr =
-					swab16(*(u16 *)&mclist->dmi_addr[2]);
-			config->config_table[i].cam_entry.lsb_mac_addr =
-					swab16(*(u16 *)&mclist->dmi_addr[4]);
-			config->config_table[i].cam_entry.flags =
-							cpu_to_le16(bp->port);
-			config->config_table[i].target_table_entry.flags = 0;
-			config->config_table[i].target_table_entry.
-								client_id = 0;
-			config->config_table[i].target_table_entry.
-								vlan_id = 0;
-
-			DP(NETIF_MSG_IFUP,
-			   "setting MCAST[%d] (%04x:%04x:%04x)\n",
-			   i, config->config_table[i].cam_entry.msb_mac_addr,
-			   config->config_table[i].cam_entry.middle_mac_addr,
-			   config->config_table[i].cam_entry.lsb_mac_addr);
-		}
-		old = config->hdr.length_6b;
-		if (old > i) {
-			for (; i < old; i++) {
-				if (CAM_IS_INVALID(config->config_table[i])) {
-					i--; /* already invalidated */
-					break;
-				}
-				/* invalidate */
-				CAM_INVALIDATE(config->config_table[i]);
-			}
-		}
+		break;
 
-		if (CHIP_REV_IS_SLOW(bp))
-			offset = BNX2X_MAX_EMUL_MULTI*(1 + bp->port);
-		else
-			offset = BNX2X_MAX_MULTICAST*(1 + bp->port);
+	case PCI_D3hot:
+		pmcsr &= ~PCI_PM_CTRL_STATE_MASK;
+		pmcsr |= 3;
 
-		config->hdr.length_6b = i;
-		config->hdr.offset = offset;
-		config->hdr.reserved0 = 0;
-		config->hdr.reserved1 = 0;
+		if (bp->wol)
+			pmcsr |= PCI_PM_CTRL_PME_ENABLE;
 
-		bnx2x_sp_post(bp, RAMROD_CMD_ID_ETH_SET_MAC, 0,
-			      U64_HI(bnx2x_sp_mapping(bp, mcast_config)),
-			      U64_LO(bnx2x_sp_mapping(bp, mcast_config)), 0);
-	}
+		pci_write_config_word(bp->pdev, bp->pm_cap + PCI_PM_CTRL,
+				      pmcsr);
 
-	bp->rx_mode = rx_mode;
-	bnx2x_set_storm_rx_mode(bp);
+		/* No more memory access after this point until
+		* device is brought back to D0.
+		*/
+		break;
+
+	default:
+		return -EINVAL;
+	}
+	return 0;
 }
 
+/*
+ * net_device service functions
+ */
+
 static int bnx2x_poll(struct napi_struct *napi, int budget)
 {
 	struct bnx2x_fastpath *fp = container_of(napi, struct bnx2x_fastpath,
@@ -6729,7 +7606,7 @@ static int bnx2x_poll(struct napi_struct *napi, int budget)
 
 #ifdef BNX2X_STOP_ON_ERROR
 	if (unlikely(bp->panic))
-		goto out_panic;
+		goto poll_panic;
 #endif
 
 	prefetch(fp->tx_buf_ring[TX_BD(fp->tx_pkt_cons)].skb);
@@ -6738,30 +7615,28 @@ static int bnx2x_poll(struct napi_struct *napi, int budget)
 
 	bnx2x_update_fpsb_idx(fp);
 
-	if (le16_to_cpu(*fp->tx_cons_sb) != fp->tx_pkt_cons)
+	if ((fp->tx_pkt_prod != le16_to_cpu(*fp->tx_cons_sb)) ||
+	    (fp->tx_pkt_prod != fp->tx_pkt_cons))
 		bnx2x_tx_int(fp, budget);
 
-
 	if (le16_to_cpu(*fp->rx_cons_sb) != fp->rx_comp_cons)
 		work_done = bnx2x_rx_int(fp, budget);
 
-
 	rmb(); /* bnx2x_has_work() reads the status block */
 
 	/* must not complete if we consumed full budget */
 	if ((work_done < budget) && !bnx2x_has_work(fp)) {
 
 #ifdef BNX2X_STOP_ON_ERROR
-out_panic:
+poll_panic:
 #endif
 		netif_rx_complete(bp->dev, napi);
 
-		bnx2x_ack_sb(bp, fp->index, USTORM_ID,
+		bnx2x_ack_sb(bp, FP_SB_ID(fp), USTORM_ID,
 			     le16_to_cpu(fp->fp_u_idx), IGU_INT_NOP, 1);
-		bnx2x_ack_sb(bp, fp->index, CSTORM_ID,
+		bnx2x_ack_sb(bp, FP_SB_ID(fp), CSTORM_ID,
 			     le16_to_cpu(fp->fp_c_idx), IGU_INT_ENABLE, 1);
 	}
-
 	return work_done;
 }
 
@@ -7055,18 +7930,145 @@ static int bnx2x_close(struct net_device *dev)
 	return 0;
 }
 
-/* Called with rtnl_lock */
+/* called with netif_tx_lock from set_multicast */
+static void bnx2x_set_rx_mode(struct net_device *dev)
+{
+	struct bnx2x *bp = netdev_priv(dev);
+	u32 rx_mode = BNX2X_RX_MODE_NORMAL;
+	int port = BP_PORT(bp);
+
+	if (bp->state != BNX2X_STATE_OPEN) {
+		DP(NETIF_MSG_IFUP, "state is %x, returning\n", bp->state);
+		return;
+	}
+
+	DP(NETIF_MSG_IFUP, "dev->flags = %x\n", dev->flags);
+
+	if (dev->flags & IFF_PROMISC)
+		rx_mode = BNX2X_RX_MODE_PROMISC;
+
+	else if ((dev->flags & IFF_ALLMULTI) ||
+		 ((dev->mc_count > BNX2X_MAX_MULTICAST) && CHIP_IS_E1(bp)))
+		rx_mode = BNX2X_RX_MODE_ALLMULTI;
+
+	else { /* some multicasts */
+		if (CHIP_IS_E1(bp)) {
+			int i, old, offset;
+			struct dev_mc_list *mclist;
+			struct mac_configuration_cmd *config =
+						bnx2x_sp(bp, mcast_config);
+
+			for (i = 0, mclist = dev->mc_list;
+			     mclist && (i < dev->mc_count);
+			     i++, mclist = mclist->next) {
+
+				config->config_table[i].
+					cam_entry.msb_mac_addr =
+					swab16(*(u16 *)&mclist->dmi_addr[0]);
+				config->config_table[i].
+					cam_entry.middle_mac_addr =
+					swab16(*(u16 *)&mclist->dmi_addr[2]);
+				config->config_table[i].
+					cam_entry.lsb_mac_addr =
+					swab16(*(u16 *)&mclist->dmi_addr[4]);
+				config->config_table[i].cam_entry.flags =
+							cpu_to_le16(port);
+				config->config_table[i].
+					target_table_entry.flags = 0;
+				config->config_table[i].
+					target_table_entry.client_id = 0;
+				config->config_table[i].
+					target_table_entry.vlan_id = 0;
+
+				DP(NETIF_MSG_IFUP,
+				   "setting MCAST[%d] (%04x:%04x:%04x)\n", i,
+				   config->config_table[i].
+						cam_entry.msb_mac_addr,
+				   config->config_table[i].
+						cam_entry.middle_mac_addr,
+				   config->config_table[i].
+						cam_entry.lsb_mac_addr);
+			}
+			old = config->hdr.length_6b;
+			if (old > i) {
+				for (; i < old; i++) {
+					if (CAM_IS_INVALID(config->
+							   config_table[i])) {
+						i--; /* already invalidated */
+						break;
+					}
+					/* invalidate */
+					CAM_INVALIDATE(config->
+						       config_table[i]);
+				}
+			}
+
+			if (CHIP_REV_IS_SLOW(bp))
+				offset = BNX2X_MAX_EMUL_MULTI*(1 + port);
+			else
+				offset = BNX2X_MAX_MULTICAST*(1 + port);
+
+			config->hdr.length_6b = i;
+			config->hdr.offset = offset;
+			config->hdr.client_id = BP_CL_ID(bp);
+			config->hdr.reserved1 = 0;
+
+			bnx2x_sp_post(bp, RAMROD_CMD_ID_ETH_SET_MAC, 0,
+				   U64_HI(bnx2x_sp_mapping(bp, mcast_config)),
+				   U64_LO(bnx2x_sp_mapping(bp, mcast_config)),
+				      0);
+		} else { /* E1H */
+			/* Accept one or more multicasts */
+			struct dev_mc_list *mclist;
+			u32 mc_filter[MC_HASH_SIZE];
+			u32 crc, bit, regidx;
+			int i;
+
+			memset(mc_filter, 0, 4 * MC_HASH_SIZE);
+
+			for (i = 0, mclist = dev->mc_list;
+			     mclist && (i < dev->mc_count);
+			     i++, mclist = mclist->next) {
+
+				DP(NETIF_MSG_IFUP, "Adding mcast MAC: "
+				   "%02x:%02x:%02x:%02x:%02x:%02x\n",
+				   mclist->dmi_addr[0], mclist->dmi_addr[1],
+				   mclist->dmi_addr[2], mclist->dmi_addr[3],
+				   mclist->dmi_addr[4], mclist->dmi_addr[5]);
+
+				crc = crc32c_le(0, mclist->dmi_addr, ETH_ALEN);
+				bit = (crc >> 24) & 0xff;
+				regidx = bit >> 5;
+				bit &= 0x1f;
+				mc_filter[regidx] |= (1 << bit);
+			}
+
+			for (i = 0; i < MC_HASH_SIZE; i++)
+				REG_WR(bp, MC_HASH_OFFSET(bp, i),
+				       mc_filter[i]);
+		}
+	}
+
+	bp->rx_mode = rx_mode;
+	bnx2x_set_storm_rx_mode(bp);
+}
+
+/* called with rtnl_lock */
 static int bnx2x_change_mac_addr(struct net_device *dev, void *p)
 {
 	struct sockaddr *addr = p;
 	struct bnx2x *bp = netdev_priv(dev);
 
-	if (!is_valid_ether_addr(addr->sa_data))
+	if (!is_valid_ether_addr((u8 *)(addr->sa_data)))
 		return -EINVAL;
 
 	memcpy(dev->dev_addr, addr->sa_data, dev->addr_len);
-	if (netif_running(dev))
-		bnx2x_set_mac_addr(bp);
+	if (netif_running(dev)) {
+		if (CHIP_IS_E1(bp))
+			bnx2x_set_mac_addr_e1(bp);
+		else
+			bnx2x_set_mac_addr_e1h(bp);
+	}
 
 	return 0;
 }
@@ -7080,7 +8082,7 @@ static int bnx2x_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
 
 	switch (cmd) {
 	case SIOCGMIIPHY:
-		data->phy_id = bp->phy_addr;
+		data->phy_id = bp->port.phy_addr;
 
 		/* fallthrough */
 
@@ -7090,12 +8092,12 @@ static int bnx2x_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
 		if (!netif_running(dev))
 			return -EAGAIN;
 
-		mutex_lock(&bp->phy_mutex);
-		err = bnx2x_cl45_read(bp, bp->port, 0, bp->phy_addr,
+		mutex_lock(&bp->port.phy_mutex);
+		err = bnx2x_cl45_read(bp, BP_PORT(bp), 0, bp->port.phy_addr,
 				      DEFAULT_PHY_DEV_ADDR,
 				      (data->reg_num & 0x1f), &mii_regval);
 		data->val_out = mii_regval;
-		mutex_unlock(&bp->phy_mutex);
+		mutex_unlock(&bp->port.phy_mutex);
 		return err;
 	}
 
@@ -7106,11 +8108,11 @@ static int bnx2x_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
 		if (!netif_running(dev))
 			return -EAGAIN;
 
-		mutex_lock(&bp->phy_mutex);
-		err = bnx2x_cl45_write(bp, bp->port, 0, bp->phy_addr,
+		mutex_lock(&bp->port.phy_mutex);
+		err = bnx2x_cl45_write(bp, BP_PORT(bp), 0, bp->port.phy_addr,
 				       DEFAULT_PHY_DEV_ADDR,
 				       (data->reg_num & 0x1f), data->val_in);
-		mutex_unlock(&bp->phy_mutex);
+		mutex_unlock(&bp->port.phy_mutex);
 		return err;
 
 	default:
@@ -7121,10 +8123,11 @@ static int bnx2x_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
 	return -EOPNOTSUPP;
 }
 
-/* Called with rtnl_lock */
+/* called with rtnl_lock */
 static int bnx2x_change_mtu(struct net_device *dev, int new_mtu)
 {
 	struct bnx2x *bp = netdev_priv(dev);
+	int rc = 0;
 
 	if ((new_mtu > ETH_MAX_JUMBO_PACKET_SIZE) ||
 	    ((new_mtu + ETH_HLEN) < ETH_MIN_PACKET_SIZE))
@@ -7137,10 +8140,11 @@ static int bnx2x_change_mtu(struct net_device *dev, int new_mtu)
 	dev->mtu = new_mtu;
 
 	if (netif_running(dev)) {
-		bnx2x_nic_unload(bp, 0);
-		bnx2x_nic_load(bp, 0);
+		bnx2x_nic_unload(bp, UNLOAD_NORMAL);
+		rc = bnx2x_nic_load(bp, LOAD_NORMAL);
 	}
-	return 0;
+
+	return rc;
 }
 
 static void bnx2x_tx_timeout(struct net_device *dev)
@@ -7156,7 +8160,7 @@ static void bnx2x_tx_timeout(struct net_device *dev)
 }
 
 #ifdef BCM_VLAN
-/* Called with rtnl_lock */
+/* called with rtnl_lock */
 static void bnx2x_vlan_rx_register(struct net_device *dev,
 				   struct vlan_group *vlgrp)
 {
@@ -7166,6 +8170,7 @@ static void bnx2x_vlan_rx_register(struct net_device *dev,
 	if (netif_running(dev))
 		bnx2x_set_client_config(bp);
 }
+
 #endif
 
 #if defined(HAVE_POLL_CONTROLLER) || defined(CONFIG_NET_POLL_CONTROLLER)
@@ -7179,36 +8184,8 @@ static void poll_bnx2x(struct net_device *dev)
 }
 #endif
 
-static void bnx2x_reset_task(struct work_struct *work)
-{
-	struct bnx2x *bp = container_of(work, struct bnx2x, reset_task);
-
-#ifdef BNX2X_STOP_ON_ERROR
-	BNX2X_ERR("reset task called but STOP_ON_ERROR defined"
-		  " so reset not done to allow debug dump,\n"
-	 KERN_ERR " you will need to reboot when done\n");
-	return;
-#endif
-
-	if (!netif_running(bp->dev))
-		return;
-
-	rtnl_lock();
-
-	if (bp->state != BNX2X_STATE_OPEN) {
-		DP(NETIF_MSG_TX_ERR, "state is %x, returning\n", bp->state);
-		goto reset_task_exit;
-	}
-
-	bnx2x_nic_unload(bp, 0);
-	bnx2x_nic_load(bp, 0);
-
-reset_task_exit:
-	rtnl_unlock();
-}
-
-static int __devinit bnx2x_init_board(struct pci_dev *pdev,
-				      struct net_device *dev)
+static int __devinit bnx2x_init_dev(struct pci_dev *pdev,
+				    struct net_device *dev)
 {
 	struct bnx2x *bp;
 	int rc;
@@ -7216,8 +8193,10 @@ static int __devinit bnx2x_init_board(struct pci_dev *pdev,
 	SET_NETDEV_DEV(dev, &pdev->dev);
 	bp = netdev_priv(dev);
 
+	bp->dev = dev;
+	bp->pdev = pdev;
 	bp->flags = 0;
-	bp->port = PCI_FUNC(pdev->devfn);
+	bp->func = PCI_FUNC(pdev->devfn);
 
 	rc = pci_enable_device(pdev);
 	if (rc) {
@@ -7239,14 +8218,17 @@ static int __devinit bnx2x_init_board(struct pci_dev *pdev,
 		goto err_out_disable;
 	}
 
-	rc = pci_request_regions(pdev, DRV_MODULE_NAME);
-	if (rc) {
-		printk(KERN_ERR PFX "Cannot obtain PCI resources,"
-		       " aborting\n");
-		goto err_out_disable;
-	}
+	if (atomic_read(&pdev->enable_cnt) == 1) {
+		rc = pci_request_regions(pdev, DRV_MODULE_NAME);
+		if (rc) {
+			printk(KERN_ERR PFX "Cannot obtain PCI resources,"
+			       " aborting\n");
+			goto err_out_disable;
+		}
 
-	pci_set_master(pdev);
+		pci_set_master(pdev);
+		pci_save_state(pdev);
+	}
 
 	bp->pm_cap = pci_find_capability(pdev, PCI_CAP_ID_PM);
 	if (bp->pm_cap == 0) {
@@ -7280,13 +8262,9 @@ static int __devinit bnx2x_init_board(struct pci_dev *pdev,
 		goto err_out_release;
 	}
 
-	bp->dev = dev;
-	bp->pdev = pdev;
-
-	INIT_WORK(&bp->reset_task, bnx2x_reset_task);
-	INIT_WORK(&bp->sp_task, bnx2x_sp_task);
-
-	dev->base_addr = pci_resource_start(pdev, 0);
+	dev->mem_start = pci_resource_start(pdev, 0);
+	dev->base_addr = dev->mem_start;
+	dev->mem_end = pci_resource_end(pdev, 0);
 
 	dev->irq = pdev->irq;
 
@@ -7298,8 +8276,9 @@ static int __devinit bnx2x_init_board(struct pci_dev *pdev,
 		goto err_out_release;
 	}
 
-	bp->doorbells = ioremap_nocache(pci_resource_start(pdev , 2),
-					pci_resource_len(pdev, 2));
+	bp->doorbells = ioremap_nocache(pci_resource_start(pdev, 2),
+					min_t(u64, BNX2X_DB_SIZE,
+					      pci_resource_len(pdev, 2)));
 	if (!bp->doorbells) {
 		printk(KERN_ERR PFX "Cannot map doorbell space, aborting\n");
 		rc = -ENOMEM;
@@ -7308,47 +8287,43 @@ static int __devinit bnx2x_init_board(struct pci_dev *pdev,
 
 	bnx2x_set_power_state(bp, PCI_D0);
 
-	bnx2x_get_hwinfo(bp);
-
-
-	if (nomcp) {
-		printk(KERN_ERR PFX "MCP disabled, will only"
-		       " init first device\n");
-		onefunc = 1;
-	}
-
-	if (onefunc && bp->port) {
-		printk(KERN_ERR PFX "Second device disabled, exiting\n");
-		rc = -ENODEV;
-		goto err_out_unmap;
-	}
-
-	bp->tx_ring_size = MAX_TX_AVAIL;
-	bp->rx_ring_size = MAX_RX_AVAIL;
-
-	bp->rx_csum = 1;
-
-	bp->rx_offset = 0;
-
-	bp->tx_quick_cons_trip_int = 0xff;
-	bp->tx_quick_cons_trip = 0xff;
-	bp->tx_ticks_int = 50;
-	bp->tx_ticks = 50;
+	/* clean indirect addresses */
+	pci_write_config_dword(bp->pdev, PCICFG_GRC_ADDRESS,
+			       PCICFG_VENDOR_ID_OFFSET);
+	REG_WR(bp, PXP2_REG_PGL_ADDR_88_F0 + BP_PORT(bp)*16, 0);
+	REG_WR(bp, PXP2_REG_PGL_ADDR_8C_F0 + BP_PORT(bp)*16, 0);
+	REG_WR(bp, PXP2_REG_PGL_ADDR_90_F0 + BP_PORT(bp)*16, 0);
+	REG_WR(bp, PXP2_REG_PGL_ADDR_94_F0 + BP_PORT(bp)*16, 0);
 
-	bp->rx_quick_cons_trip_int = 0xff;
-	bp->rx_quick_cons_trip = 0xff;
-	bp->rx_ticks_int = 25;
-	bp->rx_ticks = 25;
+	dev->hard_start_xmit = bnx2x_start_xmit;
+	dev->watchdog_timeo = TX_TIMEOUT;
 
-	bp->stats_ticks = 1000000 & 0xffff00;
+	dev->ethtool_ops = &bnx2x_ethtool_ops;
+	dev->open = bnx2x_open;
+	dev->stop = bnx2x_close;
+	dev->set_multicast_list = bnx2x_set_rx_mode;
+	dev->set_mac_address = bnx2x_change_mac_addr;
+	dev->do_ioctl = bnx2x_ioctl;
+	dev->change_mtu = bnx2x_change_mtu;
+	dev->tx_timeout = bnx2x_tx_timeout;
+#ifdef BCM_VLAN
+	dev->vlan_rx_register = bnx2x_vlan_rx_register;
+#endif
+#if defined(HAVE_POLL_CONTROLLER) || defined(CONFIG_NET_POLL_CONTROLLER)
+	dev->poll_controller = poll_bnx2x;
+#endif
+	dev->features |= NETIF_F_SG;
+	dev->features |= NETIF_F_HW_CSUM;
+	if (bp->flags & USING_DAC_FLAG)
+		dev->features |= NETIF_F_HIGHDMA;
+#ifdef BCM_VLAN
+	dev->features |= (NETIF_F_HW_VLAN_TX | NETIF_F_HW_VLAN_RX);
+#endif
+	dev->features |= (NETIF_F_TSO | NETIF_F_TSO_ECN);
 
 	bp->timer_interval = HZ;
 	bp->current_interval = (poll ? poll : HZ);
 
-	init_timer(&bp->timer);
-	bp->timer.expires = jiffies + bp->current_interval;
-	bp->timer.data = (unsigned long) bp;
-	bp->timer.function = bnx2x_timer;
 
 	return 0;
 
@@ -7357,14 +8332,14 @@ err_out_unmap:
 		iounmap(bp->regview);
 		bp->regview = NULL;
 	}
-
 	if (bp->doorbells) {
 		iounmap(bp->doorbells);
 		bp->doorbells = NULL;
 	}
 
 err_out_release:
-	pci_release_regions(pdev);
+	if (atomic_read(&pdev->enable_cnt) == 1)
+		pci_release_regions(pdev);
 
 err_out_disable:
 	pci_disable_device(pdev);
@@ -7398,7 +8373,6 @@ static int __devinit bnx2x_init_one(struct pci_dev *pdev,
 	struct net_device *dev = NULL;
 	struct bnx2x *bp;
 	int rc;
-	int port = PCI_FUNC(pdev->devfn);
 	DECLARE_MAC_BUF(mac);
 
 	if (version_printed++ == 0)
@@ -7406,78 +8380,62 @@ static int __devinit bnx2x_init_one(struct pci_dev *pdev,
 
 	/* dev zeroed in init_etherdev */
 	dev = alloc_etherdev(sizeof(*bp));
-	if (!dev)
+	if (!dev) {
+		printk(KERN_ERR PFX "Cannot allocate net device\n");
 		return -ENOMEM;
+	}
 
 	netif_carrier_off(dev);
 
 	bp = netdev_priv(dev);
 	bp->msglevel = debug;
 
-	if (port && onefunc) {
-		printk(KERN_ERR PFX "second function disabled. exiting\n");
-		free_netdev(dev);
-		return 0;
-	}
-
-	rc = bnx2x_init_board(pdev, dev);
+	rc = bnx2x_init_dev(pdev, dev);
 	if (rc < 0) {
 		free_netdev(dev);
 		return rc;
 	}
 
-	dev->hard_start_xmit = bnx2x_start_xmit;
-	dev->watchdog_timeo = TX_TIMEOUT;
-
-	dev->ethtool_ops = &bnx2x_ethtool_ops;
-	dev->open = bnx2x_open;
-	dev->stop = bnx2x_close;
-	dev->set_multicast_list = bnx2x_set_rx_mode;
-	dev->set_mac_address = bnx2x_change_mac_addr;
-	dev->do_ioctl = bnx2x_ioctl;
-	dev->change_mtu = bnx2x_change_mtu;
-	dev->tx_timeout = bnx2x_tx_timeout;
-#ifdef BCM_VLAN
-	dev->vlan_rx_register = bnx2x_vlan_rx_register;
-#endif
-#if defined(HAVE_POLL_CONTROLLER) || defined(CONFIG_NET_POLL_CONTROLLER)
-	dev->poll_controller = poll_bnx2x;
-#endif
-	dev->features |= NETIF_F_SG;
-	if (bp->flags & USING_DAC_FLAG)
-		dev->features |= NETIF_F_HIGHDMA;
-	dev->features |= NETIF_F_IP_CSUM;
-#ifdef BCM_VLAN
-	dev->features |= NETIF_F_HW_VLAN_TX | NETIF_F_HW_VLAN_RX;
-#endif
-	dev->features |= NETIF_F_TSO | NETIF_F_TSO_ECN;
-
 	rc = register_netdev(dev);
 	if (rc) {
 		dev_err(&pdev->dev, "Cannot register net device\n");
-		if (bp->regview)
-			iounmap(bp->regview);
-		if (bp->doorbells)
-			iounmap(bp->doorbells);
-		pci_release_regions(pdev);
-		pci_disable_device(pdev);
-		pci_set_drvdata(pdev, NULL);
-		free_netdev(dev);
-		return rc;
+		goto init_one_exit;
 	}
 
 	pci_set_drvdata(pdev, dev);
 
-	bp->name = board_info[ent->driver_data].name;
+	rc = bnx2x_init_bp(bp);
+	if (rc) {
+		unregister_netdev(dev);
+		goto init_one_exit;
+	}
+
+	bp->common.name = board_info[ent->driver_data].name;
 	printk(KERN_INFO "%s: %s (%c%d) PCI-E x%d %s found at mem %lx,"
-	       " IRQ %d, ", dev->name, bp->name,
-	       ((CHIP_ID(bp) & 0xf000) >> 12) + 'A',
-	       ((CHIP_ID(bp) & 0x0ff0) >> 4),
+	       " IRQ %d, ", dev->name, bp->common.name,
+	       (CHIP_REV(bp) >> 12) + 'A', (CHIP_METAL(bp) >> 4),
 	       bnx2x_get_pcie_width(bp),
 	       (bnx2x_get_pcie_speed(bp) == 2) ? "5GHz (Gen2)" : "2.5GHz",
 	       dev->base_addr, bp->pdev->irq);
 	printk(KERN_CONT "node addr %s\n", print_mac(mac, dev->dev_addr));
 	return 0;
+
+init_one_exit:
+	if (bp->regview)
+		iounmap(bp->regview);
+
+	if (bp->doorbells)
+		iounmap(bp->doorbells);
+
+	free_netdev(dev);
+
+	if (atomic_read(&pdev->enable_cnt) == 1)
+		pci_release_regions(pdev);
+
+	pci_disable_device(pdev);
+	pci_set_drvdata(pdev, NULL);
+
+	return rc;
 }
 
 static void __devexit bnx2x_remove_one(struct pci_dev *pdev)
@@ -7486,11 +8444,9 @@ static void __devexit bnx2x_remove_one(struct pci_dev *pdev)
 	struct bnx2x *bp;
 
 	if (!dev) {
-		/* we get here if init_one() fails */
 		printk(KERN_ERR PFX "BAD net device from bnx2x_init_one\n");
 		return;
 	}
-
 	bp = netdev_priv(dev);
 
 	unregister_netdev(dev);
@@ -7502,7 +8458,10 @@ static void __devexit bnx2x_remove_one(struct pci_dev *pdev)
 		iounmap(bp->doorbells);
 
 	free_netdev(dev);
-	pci_release_regions(pdev);
+
+	if (atomic_read(&pdev->enable_cnt) == 1)
+		pci_release_regions(pdev);
+
 	pci_disable_device(pdev);
 	pci_set_drvdata(pdev, NULL);
 }
@@ -7512,21 +8471,29 @@ static int bnx2x_suspend(struct pci_dev *pdev, pm_message_t state)
 	struct net_device *dev = pci_get_drvdata(pdev);
 	struct bnx2x *bp;
 
-	if (!dev)
-		return 0;
+	if (!dev) {
+		printk(KERN_ERR PFX "BAD net device from bnx2x_init_one\n");
+		return -ENODEV;
+	}
+	bp = netdev_priv(dev);
 
-	if (!netif_running(dev))
-		return 0;
+	rtnl_lock();
 
-	bp = netdev_priv(dev);
+	pci_save_state(pdev);
 
-	bnx2x_nic_unload(bp, 0);
+	if (!netif_running(dev)) {
+		rtnl_unlock();
+		return 0;
+	}
 
 	netif_device_detach(dev);
 
-	pci_save_state(pdev);
+	bnx2x_nic_unload(bp, UNLOAD_NORMAL);
+
 	bnx2x_set_power_state(bp, pci_choose_state(pdev, state));
 
+	rtnl_unlock();
+
 	return 0;
 }
 
@@ -7540,21 +8507,25 @@ static int bnx2x_resume(struct pci_dev *pdev)
 		printk(KERN_ERR PFX "BAD net device from bnx2x_init_one\n");
 		return -ENODEV;
 	}
-
-	if (!netif_running(dev))
-		return 0;
-
 	bp = netdev_priv(dev);
 
+	rtnl_lock();
+
 	pci_restore_state(pdev);
+
+	if (!netif_running(dev)) {
+		rtnl_unlock();
+		return 0;
+	}
+
 	bnx2x_set_power_state(bp, PCI_D0);
 	netif_device_attach(dev);
 
-	rc = bnx2x_nic_load(bp, 0);
-	if (rc)
-		return rc;
+	rc = bnx2x_nic_load(bp, LOAD_NORMAL);
 
-	return 0;
+	rtnl_unlock();
+
+	return rc;
 }
 
 static struct pci_driver bnx2x_pci_driver = {
diff --git a/drivers/net/bnx2x_reg.h b/drivers/net/bnx2x_reg.h
index 8707c0d05d9a..15c9a9946724 100644
--- a/drivers/net/bnx2x_reg.h
+++ b/drivers/net/bnx2x_reg.h
@@ -38,21 +38,19 @@
    was asserted. */
 #define BRB1_REG_NUM_OF_FULL_CYCLES_0				 0x600c8
 #define BRB1_REG_NUM_OF_FULL_CYCLES_1				 0x600cc
-#define BRB1_REG_NUM_OF_FULL_CYCLES_2				 0x600d0
-#define BRB1_REG_NUM_OF_FULL_CYCLES_3				 0x600d4
 #define BRB1_REG_NUM_OF_FULL_CYCLES_4				 0x600d8
 /* [ST 32] The number of cycles that the pause signal towards MAC #0 was
    asserted. */
 #define BRB1_REG_NUM_OF_PAUSE_CYCLES_0				 0x600b8
 #define BRB1_REG_NUM_OF_PAUSE_CYCLES_1				 0x600bc
-#define BRB1_REG_NUM_OF_PAUSE_CYCLES_2				 0x600c0
-#define BRB1_REG_NUM_OF_PAUSE_CYCLES_3				 0x600c4
 /* [RW 10] Write client 0: De-assert pause threshold. */
 #define BRB1_REG_PAUSE_HIGH_THRESHOLD_0 			 0x60078
 #define BRB1_REG_PAUSE_HIGH_THRESHOLD_1 			 0x6007c
 /* [RW 10] Write client 0: Assert pause threshold. */
 #define BRB1_REG_PAUSE_LOW_THRESHOLD_0				 0x60068
 #define BRB1_REG_PAUSE_LOW_THRESHOLD_1				 0x6006c
+/* [R 24] The number of full blocks occpied by port. */
+#define BRB1_REG_PORT_NUM_OCC_BLOCKS_0				 0x60094
 /* [RW 1] Reset the design by software. */
 #define BRB1_REG_SOFT_RESET					 0x600dc
 /* [R 5] Used to read the value of the XX protection CAM occupancy counter. */
@@ -513,7 +511,6 @@
 /* [RW 15] Interrupt table Read and write access to it is not possible in
    the middle of the work */
 #define CSEM_REG_INT_TABLE					 0x200400
-#define CSEM_REG_INT_TABLE_SIZE 				 256
 /* [ST 24] Statistics register. The number of messages that entered through
    FIC0 */
 #define CSEM_REG_MSG_NUM_FIC0					 0x200000
@@ -587,13 +584,10 @@
 #define DBG_REG_DBG_PRTY_MASK					 0xc0a8
 /* [R 1] Parity register #0 read */
 #define DBG_REG_DBG_PRTY_STS					 0xc09c
-/* [RW 2] debug only: These bits indicate the credit for PCI request type 4
-   interface; MUST be configured AFTER pci_ext_buffer_strt_addr_lsb/msb are
-   configured */
-#define DBG_REG_PCI_REQ_CREDIT					 0xc120
 /* [RW 32] Commands memory. The address to command X; row Y is to calculated
    as 14*X+Y. */
 #define DMAE_REG_CMD_MEM					 0x102400
+#define DMAE_REG_CMD_MEM_SIZE					 224
 /* [RW 1] If 0 - the CRC-16c initial value is all zeroes; if 1 - the CRC-16c
    initial value is all ones. */
 #define DMAE_REG_CRC16C_INIT					 0x10201c
@@ -1626,7 +1620,7 @@
    is reset to 0x080; giving a default blink period of approximately 8Hz. */
 #define NIG_REG_LED_CONTROL_BLINK_RATE_P0			 0x10310
 /* [RW 1] Port0: If set along with the
-   nig_registers_led_control_override_traffic_p0.led_control_override_traffic_p0
+ ~nig_registers_led_control_override_traffic_p0.led_control_override_traffic_p0
    bit and ~nig_registers_led_control_traffic_p0.led_control_traffic_p0 LED
    bit; the Traffic LED will blink with the blink rate specified in
    ~nig_registers_led_control_blink_rate_p0.led_control_blink_rate_p0 and
@@ -1733,9 +1727,21 @@
 /* [R 32] Rx statistics : In user packets discarded due to BRB backpressure
    for port0 */
 #define NIG_REG_STAT0_BRB_DISCARD				 0x105f0
+/* [WB_R 36] Tx statistics : Number of packets from emac0 or bmac0 that
+   between 1024 and 1522 bytes for port0 */
+#define NIG_REG_STAT0_EGRESS_MAC_PKT0				 0x10750
+/* [WB_R 36] Tx statistics : Number of packets from emac0 or bmac0 that
+   between 1523 bytes and above for port0 */
+#define NIG_REG_STAT0_EGRESS_MAC_PKT1				 0x10760
 /* [R 32] Rx statistics : In user packets discarded due to BRB backpressure
    for port1 */
 #define NIG_REG_STAT1_BRB_DISCARD				 0x10628
+/* [WB_R 36] Tx statistics : Number of packets from emac1 or bmac1 that
+   between 1024 and 1522 bytes for port1 */
+#define NIG_REG_STAT1_EGRESS_MAC_PKT0				 0x107a0
+/* [WB_R 36] Tx statistics : Number of packets from emac1 or bmac1 that
+   between 1523 bytes and above for port1 */
+#define NIG_REG_STAT1_EGRESS_MAC_PKT1				 0x107b0
 /* [WB_R 64] Rx statistics : User octets received for LP */
 #define NIG_REG_STAT2_BRB_OCTET 				 0x107e0
 #define NIG_REG_STATUS_INTERRUPT_PORT0				 0x10328
@@ -1849,7 +1855,6 @@
 #define PRS_REG_CFC_SEARCH_INITIAL_CREDIT			 0x4011c
 /* [RW 24] CID for port 0 if no match */
 #define PRS_REG_CID_PORT_0					 0x400fc
-#define PRS_REG_CID_PORT_1					 0x40100
 /* [RW 32] The CM header for flush message where 'load existed' bit in CFC
    load response is reset and packet type is 0. Used in packet start message
    to TCM. */
@@ -1957,6 +1962,10 @@
 #define PXP2_REG_HST_DATA_FIFO_STATUS				 0x12047c
 /* [R 7] Debug only: Number of used entries in the header FIFO */
 #define PXP2_REG_HST_HEADER_FIFO_STATUS 			 0x120478
+#define PXP2_REG_PGL_ADDR_88_F0 				 0x120534
+#define PXP2_REG_PGL_ADDR_8C_F0 				 0x120538
+#define PXP2_REG_PGL_ADDR_90_F0 				 0x12053c
+#define PXP2_REG_PGL_ADDR_94_F0 				 0x120540
 #define PXP2_REG_PGL_CONTROL0					 0x120490
 #define PXP2_REG_PGL_CONTROL1					 0x120514
 /* [RW 32] third dword data of expansion rom request. this register is
@@ -2060,12 +2069,13 @@
 #define PXP2_REG_PSWRQ_SRC0_L2P 				 0x120054
 #define PXP2_REG_PSWRQ_TM0_L2P					 0x12001c
 #define PXP2_REG_PSWRQ_TSDM0_L2P				 0x1200e0
-/* [RW 25] Interrupt mask register #0 read/write */
-#define PXP2_REG_PXP2_INT_MASK					 0x120578
-/* [R 25] Interrupt register #0 read */
-#define PXP2_REG_PXP2_INT_STS					 0x12056c
-/* [RC 25] Interrupt register #0 read clear */
-#define PXP2_REG_PXP2_INT_STS_CLR				 0x120570
+/* [RW 32] Interrupt mask register #0 read/write */
+#define PXP2_REG_PXP2_INT_MASK_0				 0x120578
+/* [R 32] Interrupt register #0 read */
+#define PXP2_REG_PXP2_INT_STS_0 				 0x12056c
+#define PXP2_REG_PXP2_INT_STS_1 				 0x120608
+/* [RC 32] Interrupt register #0 read clear */
+#define PXP2_REG_PXP2_INT_STS_CLR_0				 0x120570
 /* [RW 32] Parity mask register #0 read/write */
 #define PXP2_REG_PXP2_PRTY_MASK_0				 0x120588
 #define PXP2_REG_PXP2_PRTY_MASK_1				 0x120598
@@ -2811,22 +2821,6 @@
 #define QM_REG_QVOQIDX_97					 0x16e490
 #define QM_REG_QVOQIDX_98					 0x16e494
 #define QM_REG_QVOQIDX_99					 0x16e498
-/* [R 24] Remaining pause timeout for queues 15-0 */
-#define QM_REG_REMAINPAUSETM0					 0x168418
-/* [R 24] Remaining pause timeout for queues 31-16 */
-#define QM_REG_REMAINPAUSETM1					 0x16841c
-/* [R 24] Remaining pause timeout for queues 47-32 */
-#define QM_REG_REMAINPAUSETM2					 0x16e69c
-/* [R 24] Remaining pause timeout for queues 63-48 */
-#define QM_REG_REMAINPAUSETM3					 0x16e6a0
-/* [R 24] Remaining pause timeout for queues 79-64 */
-#define QM_REG_REMAINPAUSETM4					 0x16e6a4
-/* [R 24] Remaining pause timeout for queues 95-80 */
-#define QM_REG_REMAINPAUSETM5					 0x16e6a8
-/* [R 24] Remaining pause timeout for queues 111-96 */
-#define QM_REG_REMAINPAUSETM6					 0x16e6ac
-/* [R 24] Remaining pause timeout for queues 127-112 */
-#define QM_REG_REMAINPAUSETM7					 0x16e6b0
 /* [RW 1] Initialization bit command */
 #define QM_REG_SOFT_RESET					 0x168428
 /* [RW 8] The credit cost per every task in the QM. A value per each VOQ */
@@ -3826,7 +3820,6 @@
 /* [RW 15] Interrupt table Read and write access to it is not possible in
    the middle of the work */
 #define TSEM_REG_INT_TABLE					 0x180400
-#define TSEM_REG_INT_TABLE_SIZE 				 256
 /* [ST 24] Statistics register. The number of messages that entered through
    FIC0 */
 #define TSEM_REG_MSG_NUM_FIC0					 0x180000
@@ -4283,7 +4276,6 @@
 /* [RW 15] Interrupt table Read and write access to it is not possible in
    the middle of the work */
 #define USEM_REG_INT_TABLE					 0x300400
-#define USEM_REG_INT_TABLE_SIZE 				 256
 /* [ST 24] Statistics register. The number of messages that entered through
    FIC0 */
 #define USEM_REG_MSG_NUM_FIC0					 0x300000
@@ -4802,7 +4794,6 @@
 /* [RW 15] Interrupt table Read and write access to it is not possible in
    the middle of the work */
 #define XSEM_REG_INT_TABLE					 0x280400
-#define XSEM_REG_INT_TABLE_SIZE 				 256
 /* [ST 24] Statistics register. The number of messages that entered through
    FIC0 */
 #define XSEM_REG_MSG_NUM_FIC0					 0x280000
@@ -4930,10 +4921,7 @@
 #define EMAC_MDIO_MODE_CLOCK_CNT				 (0x3fL<<16)
 #define EMAC_MDIO_MODE_CLOCK_CNT_BITSHIFT			 16
 #define EMAC_MODE_25G_MODE					 (1L<<5)
-#define EMAC_MODE_ACPI_RCVD					 (1L<<20)
 #define EMAC_MODE_HALF_DUPLEX					 (1L<<1)
-#define EMAC_MODE_MPKT						 (1L<<18)
-#define EMAC_MODE_MPKT_RCVD					 (1L<<19)
 #define EMAC_MODE_PORT_GMII					 (2L<<2)
 #define EMAC_MODE_PORT_MII					 (1L<<2)
 #define EMAC_MODE_PORT_MII_10M					 (3L<<2)
diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h
index caa000596b25..e74b14acf8e0 100644
--- a/include/linux/pci_ids.h
+++ b/include/linux/pci_ids.h
@@ -1949,6 +1949,8 @@
 #define PCI_DEVICE_ID_NX2_5708		0x164c
 #define PCI_DEVICE_ID_TIGON3_5702FE	0x164d
 #define PCI_DEVICE_ID_NX2_57710		0x164e
+#define PCI_DEVICE_ID_NX2_57711		0x164f
+#define PCI_DEVICE_ID_NX2_57711E	0x1650
 #define PCI_DEVICE_ID_TIGON3_5705	0x1653
 #define PCI_DEVICE_ID_TIGON3_5705_2	0x1654
 #define PCI_DEVICE_ID_TIGON3_5720	0x1658
-- 
cgit v1.2.3


From f8dd0ecbb74d4b220b105d77c0633945ebb5453e Mon Sep 17 00:00:00 2001
From: Ben Dooks <ben-linux@fluff.org>
Date: Tue, 24 Jun 2008 22:16:04 +0100
Subject: DM9000: Allow the use of the NSR register to get link status.

The DM9000's internal PHY reports a copy of the link status
in the NSR register of the chip. Reading the status when
polling for link status is faster as it eliminates the need
to sleep, but does not print as much information.

Add an platform flag to force this behaviour, and a Kconfig
option to allow it to be forced to the faster method always.

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
---
 drivers/net/Kconfig    |  9 +++++++++
 drivers/net/dm9000.c   | 41 +++++++++++++++++++++++++++++++++++++++--
 include/linux/dm9000.h |  1 +
 3 files changed, 49 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/drivers/net/Kconfig b/drivers/net/Kconfig
index 287d0873c60d..4d69474b6125 100644
--- a/drivers/net/Kconfig
+++ b/drivers/net/Kconfig
@@ -938,6 +938,15 @@ config DM9000
 	  To compile this driver as a module, choose M here.  The module
 	  will be called dm9000.
 
+config DM9000_FORCE_SIMPLE_PHY_POLL
+	bool "Force simple NSR based PHY polling"
+	depends on DM9000
+	---help---
+	  This configuration forces the DM9000 to use the NSR's LinkStatus
+	  bit to determine if the link is up or down instead of the more
+	  costly MII PHY reads. Note, this will not work if the chip is
+	  operating with an external PHY.
+
 config ENC28J60
 	tristate "ENC28J60 support"
 	depends on EXPERIMENTAL && SPI && NET_ETHERNET
diff --git a/drivers/net/dm9000.c b/drivers/net/dm9000.c
index 7c38f6129b55..5ad2ec537684 100644
--- a/drivers/net/dm9000.c
+++ b/drivers/net/dm9000.c
@@ -552,15 +552,48 @@ static const struct ethtool_ops dm9000_ethtool_ops = {
  	.set_eeprom		= dm9000_set_eeprom,
 };
 
+static void dm9000_show_carrier(board_info_t *db,
+				unsigned carrier, unsigned nsr)
+{
+	struct net_device *ndev = db->ndev;
+	unsigned ncr = dm9000_read_locked(db, DM9000_NCR);
+
+	if (carrier)
+		dev_info(db->dev, "%s: link up, %dMbps, %s-duplex, no LPA\n",
+			 ndev->name, (nsr & NSR_SPEED) ? 10 : 100,
+			 (ncr & NCR_FDX) ? "full" : "half");
+	else
+		dev_info(db->dev, "%s: link down\n", ndev->name);
+}
+
 static void
 dm9000_poll_work(struct work_struct *w)
 {
 	struct delayed_work *dw = container_of(w, struct delayed_work, work);
 	board_info_t *db = container_of(dw, board_info_t, phy_poll);
+	struct net_device *ndev = db->ndev;
+
+	if (db->flags & DM9000_PLATF_SIMPLE_PHY &&
+	    !(db->flags & DM9000_PLATF_EXT_PHY)) {
+		unsigned nsr = dm9000_read_locked(db, DM9000_NSR);
+		unsigned old_carrier = netif_carrier_ok(ndev) ? 1 : 0;
+		unsigned new_carrier;
 
-	mii_check_media(&db->mii, netif_msg_link(db), 0);
+		new_carrier = (nsr & NSR_LINKST) ? 1 : 0;
+
+		if (old_carrier != new_carrier) {
+			if (netif_msg_link(db))
+				dm9000_show_carrier(db, new_carrier, nsr);
+
+			if (!new_carrier)
+				netif_carrier_off(ndev);
+			else
+				netif_carrier_on(ndev);
+		}
+	} else
+		mii_check_media(&db->mii, netif_msg_link(db), 0);
 	
-	if (netif_running(db->ndev))
+	if (netif_running(ndev))
 		dm9000_schedule_poll(db);
 }
 
@@ -1267,6 +1300,10 @@ dm9000_probe(struct platform_device *pdev)
 		db->flags = pdata->flags;
 	}
 
+#ifdef CONFIG_DM9000_FORCE_SIMPLE_PHY_POLL
+	db->flags |= DM9000_PLATF_SIMPLE_PHY;
+#endif
+
 	dm9000_reset(db);
 
 	/* try multiple times, DM9000 sometimes gets the read wrong */
diff --git a/include/linux/dm9000.h b/include/linux/dm9000.h
index a3750462f9e3..fc82446b6425 100644
--- a/include/linux/dm9000.h
+++ b/include/linux/dm9000.h
@@ -21,6 +21,7 @@
 #define DM9000_PLATF_32BITONLY	(0x0004)
 #define DM9000_PLATF_EXT_PHY	(0x0008)
 #define DM9000_PLATF_NO_EEPROM	(0x0010)
+#define DM9000_PLATF_SIMPLE_PHY (0x0020)  /* Use NSR to find LinkStatus */
 
 /* platfrom data for platfrom device structure's platfrom_data field */
 
-- 
cgit v1.2.3


From f3146aff7f283c8699e0c97df6307a705786eeba Mon Sep 17 00:00:00 2001
From: Henrique de Moraes Holschuh <hmh@hmh.eng.br>
Date: Mon, 23 Jun 2008 17:22:56 -0300
Subject: rfkill: clarify meaning of rfkill states

rfkill really should have been named rfswitch.  As it is, one can get
confused whether RFKILL_STATE_ON means the KILL switch is on (and
therefore, the radio is being *blocked* from operating), or whether it
means the RADIO rf output is on.

Clearly state that RFKILL_STATE_ON means the radio is *unblocked* from
operating (i.e. there is no rf killing going on).

Signed-off-by: Henrique de Moraes Holschuh <hmh@hmh.eng.br>
Acked-by: Ivo van Doorn <IvDoorn@gmail.com>
Cc: Dmitry Torokhov <dtor@mail.ru>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 Documentation/rfkill.txt | 7 +++++++
 include/linux/rfkill.h   | 6 +++---
 2 files changed, 10 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/Documentation/rfkill.txt b/Documentation/rfkill.txt
index a83ff23cd68c..ec75d6d34785 100644
--- a/Documentation/rfkill.txt
+++ b/Documentation/rfkill.txt
@@ -8,6 +8,13 @@ rfkill - RF switch subsystem support
 ===============================================================================
 1: Implementation details
 
+The rfkill switch subsystem exists to add a generic interface to circuitry that
+can enable or disable the RF output of a radio *transmitter* of any type.
+
+When a rfkill switch is in the RFKILL_STATE_ON, the radio transmitter is
+*enabled*.  When the rfkill switch is in the RFKILL_STATE_OFF, the radio
+transmitter is *disabled*.
+
 The rfkill switch subsystem offers support for keys often found on laptops
 to enable wireless devices like WiFi and Bluetooth.
 
diff --git a/include/linux/rfkill.h b/include/linux/rfkill.h
index e3ab21d7fc7f..ca89ae1b0219 100644
--- a/include/linux/rfkill.h
+++ b/include/linux/rfkill.h
@@ -44,8 +44,8 @@ enum rfkill_type {
 };
 
 enum rfkill_state {
-	RFKILL_STATE_OFF	= 0,
-	RFKILL_STATE_ON		= 1,
+	RFKILL_STATE_OFF	= 0,	/* Radio output blocked */
+	RFKILL_STATE_ON		= 1,	/* Radio output active */
 };
 
 /**
@@ -53,7 +53,7 @@ enum rfkill_state {
  * @name: Name of the switch.
  * @type: Radio type which the button controls, the value stored
  *	here should be a value from enum rfkill_type.
- * @state: State of the switch (on/off).
+ * @state: State of the switch, "ON" means radio can operate.
  * @user_claim_unsupported: Whether the hardware supports exclusive
  *	RF-kill control by userspace. Set this before registering.
  * @user_claim: Set when the switch is controlled exlusively by userspace.
-- 
cgit v1.2.3


From 801e49af4c1a9b988ba0d25de2b368c99c3bf2b3 Mon Sep 17 00:00:00 2001
From: Henrique de Moraes Holschuh <hmh@hmh.eng.br>
Date: Mon, 23 Jun 2008 17:23:00 -0300
Subject: rfkill: add read-write rfkill switch support

Currently, rfkill support for read/write rfkill switches is hacked through
a round-trip over the input layer and rfkill-input to let a driver sync
rfkill->state to hardware changes.

This is buggy and sub-optimal.  It causes real problems.  It is best to
think of the rfkill class as supporting only write-only switches at the
moment.

In order to implement the read/write functionality properly:

Add a get_state() hook that is called by the class every time it needs to
fetch the current state of the switch.  Add a call to this hook every time
the *current* state of the radio plays a role in a decision.

Also add a force_state() method that can be used to forcefully syncronize
the class' idea of the current state of the switch.  This allows for a
faster implementation of the read/write functionality, as a driver which
get events on switch changes can avoid the need for a get_state() hook.

If the get_state() hook is left as NULL, current behaviour is maintained,
so this change is fully backwards compatible with the current rfkill
drivers.

For hardware that issues events when the rfkill state changes, leave
get_state() NULL in the rfkill struct, set the initial state properly
before registering with the rfkill class, and use the force_state() method
in the driver to keep the rfkill interface up-to-date.

get_state() can be called by the class from atomic context. It must not
sleep.

Signed-off-by: Henrique de Moraes Holschuh <hmh@hmh.eng.br>
Acked-by: Ivo van Doorn <IvDoorn@gmail.com>
Cc: Dmitry Torokhov <dtor@mail.ru>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/linux/rfkill.h |  5 +++++
 net/rfkill/rfkill.c    | 53 +++++++++++++++++++++++++++++++++++++++++++++++---
 2 files changed, 55 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/linux/rfkill.h b/include/linux/rfkill.h
index ca89ae1b0219..844e96114861 100644
--- a/include/linux/rfkill.h
+++ b/include/linux/rfkill.h
@@ -61,6 +61,8 @@ enum rfkill_state {
  * @data: Pointer to the RF button drivers private data which will be
  *	passed along when toggling radio state.
  * @toggle_radio(): Mandatory handler to control state of the radio.
+ * @get_state(): handler to read current radio state from hardware,
+ *      may be called from atomic context, should return 0 on success.
  * @led_trigger: A LED trigger for this button's LED.
  * @dev: Device structure integrating the switch into device tree.
  * @node: Used to place switch into list of all switches known to the
@@ -80,6 +82,7 @@ struct rfkill {
 
 	void *data;
 	int (*toggle_radio)(void *data, enum rfkill_state state);
+	int (*get_state)(void *data, enum rfkill_state *state);
 
 #ifdef CONFIG_RFKILL_LEDS
 	struct led_trigger led_trigger;
@@ -95,6 +98,8 @@ void rfkill_free(struct rfkill *rfkill);
 int rfkill_register(struct rfkill *rfkill);
 void rfkill_unregister(struct rfkill *rfkill);
 
+int rfkill_force_state(struct rfkill *rfkill, enum rfkill_state state);
+
 /**
  * rfkill_get_led_name - Get the LED trigger name for the button's LED.
  * This function might return a NULL pointer if registering of the
diff --git a/net/rfkill/rfkill.c b/net/rfkill/rfkill.c
index 3edc585dcfa6..4ae4486c77ea 100644
--- a/net/rfkill/rfkill.c
+++ b/net/rfkill/rfkill.c
@@ -62,19 +62,39 @@ static void rfkill_led_trigger(struct rfkill *rfkill,
 #endif /* CONFIG_RFKILL_LEDS */
 }
 
+static void update_rfkill_state(struct rfkill *rfkill)
+{
+	enum rfkill_state newstate;
+
+	if (rfkill->get_state) {
+		mutex_lock(&rfkill->mutex);
+		if (!rfkill->get_state(rfkill->data, &newstate))
+			rfkill->state = newstate;
+		mutex_unlock(&rfkill->mutex);
+	}
+}
+
 static int rfkill_toggle_radio(struct rfkill *rfkill,
 				enum rfkill_state state)
 {
 	int retval = 0;
+	enum rfkill_state oldstate, newstate;
+
+	oldstate = rfkill->state;
+
+	if (rfkill->get_state &&
+	    !rfkill->get_state(rfkill->data, &newstate))
+		rfkill->state = newstate;
 
 	if (state != rfkill->state) {
 		retval = rfkill->toggle_radio(rfkill->data, state);
-		if (!retval) {
+		if (!retval)
 			rfkill->state = state;
-			rfkill_led_trigger(rfkill, state);
-		}
 	}
 
+	if (rfkill->state != oldstate)
+		rfkill_led_trigger(rfkill, rfkill->state);
+
 	return retval;
 }
 
@@ -105,6 +125,32 @@ void rfkill_switch_all(enum rfkill_type type, enum rfkill_state state)
 }
 EXPORT_SYMBOL(rfkill_switch_all);
 
+/**
+ * rfkill_force_state - Force the internal rfkill radio state
+ * @rfkill: pointer to the rfkill class to modify.
+ * @state: the current radio state the class should be forced to.
+ *
+ * This function updates the internal state of the radio cached
+ * by the rfkill class.  It should be used when the driver gets
+ * a notification by the firmware/hardware of the current *real*
+ * state of the radio rfkill switch.
+ *
+ * It may not be called from an atomic context.
+ */
+int rfkill_force_state(struct rfkill *rfkill, enum rfkill_state state)
+{
+	if (state != RFKILL_STATE_OFF &&
+	    state != RFKILL_STATE_ON)
+		return -EINVAL;
+
+	mutex_lock(&rfkill->mutex);
+	rfkill->state = state;
+	mutex_unlock(&rfkill->mutex);
+
+	return 0;
+}
+EXPORT_SYMBOL(rfkill_force_state);
+
 static ssize_t rfkill_name_show(struct device *dev,
 				struct device_attribute *attr,
 				char *buf)
@@ -147,6 +193,7 @@ static ssize_t rfkill_state_show(struct device *dev,
 {
 	struct rfkill *rfkill = to_rfkill(dev);
 
+	update_rfkill_state(rfkill);
 	return sprintf(buf, "%d\n", rfkill->state);
 }
 
-- 
cgit v1.2.3


From 477576a073699783abb53ae14993d5d41c66301d Mon Sep 17 00:00:00 2001
From: Henrique de Moraes Holschuh <hmh@hmh.eng.br>
Date: Mon, 23 Jun 2008 17:23:01 -0300
Subject: rfkill: add the WWAN radio type
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Unfortunately, instead of adding a generic Wireless WAN type, a technology-
specific type (WiMAX) was added.  That's useless for other WWAN devices,
such as EDGE, UMTS, X-RTT and other such radios.

Add a WWAN rfkill type for generic wireless WAN devices.  No keys are added
as most devices really want to use KEY_WLAN for WWAN control (in a cycle of
none, WLAN, WWAN, WLAN+WWAN) and need no specific keycode added.

Signed-off-by: Henrique de Moraes Holschuh <hmh@hmh.eng.br>
Acked-by: Ivo van Doorn <IvDoorn@gmail.com>
Cc: Iñaky Pérez-González <inaky.perez-gonzalez@intel.com>
Cc: David S. Miller <davem@davemloft.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/linux/rfkill.h    | 2 ++
 net/rfkill/rfkill-input.c | 4 ++++
 net/rfkill/rfkill.c       | 3 +++
 3 files changed, 9 insertions(+)

(limited to 'include')

diff --git a/include/linux/rfkill.h b/include/linux/rfkill.h
index 844e96114861..c0cab7d37828 100644
--- a/include/linux/rfkill.h
+++ b/include/linux/rfkill.h
@@ -34,12 +34,14 @@
  * RFKILL_TYPE_BLUETOOTH: switch is on a bluetooth device.
  * RFKILL_TYPE_UWB: switch is on a ultra wideband device.
  * RFKILL_TYPE_WIMAX: switch is on a WiMAX device.
+ * RFKILL_TYPE_WWAN: switch is on a wireless WAN device.
  */
 enum rfkill_type {
 	RFKILL_TYPE_WLAN ,
 	RFKILL_TYPE_BLUETOOTH,
 	RFKILL_TYPE_UWB,
 	RFKILL_TYPE_WIMAX,
+	RFKILL_TYPE_WWAN,
 	RFKILL_TYPE_MAX,
 };
 
diff --git a/net/rfkill/rfkill-input.c b/net/rfkill/rfkill-input.c
index 9d6c9255bf2c..29c13d308b31 100644
--- a/net/rfkill/rfkill-input.c
+++ b/net/rfkill/rfkill-input.c
@@ -101,6 +101,7 @@ static DEFINE_RFKILL_TASK(rfkill_wlan, RFKILL_TYPE_WLAN);
 static DEFINE_RFKILL_TASK(rfkill_bt, RFKILL_TYPE_BLUETOOTH);
 static DEFINE_RFKILL_TASK(rfkill_uwb, RFKILL_TYPE_UWB);
 static DEFINE_RFKILL_TASK(rfkill_wimax, RFKILL_TYPE_WIMAX);
+static DEFINE_RFKILL_TASK(rfkill_wwan, RFKILL_TYPE_WWAN);
 
 static void rfkill_event(struct input_handle *handle, unsigned int type,
 			unsigned int code, int data)
@@ -126,6 +127,9 @@ static void rfkill_event(struct input_handle *handle, unsigned int type,
 		switch (code) {
 		case SW_RFKILL_ALL:
 			/* EVERY radio type. data != 0 means radios ON */
+			rfkill_schedule_set(&rfkill_wwan,
+					    (data)? RFKILL_STATE_ON:
+						    RFKILL_STATE_OFF);
 			rfkill_schedule_set(&rfkill_wimax,
 					    (data)? RFKILL_STATE_ON:
 						    RFKILL_STATE_OFF);
diff --git a/net/rfkill/rfkill.c b/net/rfkill/rfkill.c
index 4ae4486c77ea..79f3bbb027ff 100644
--- a/net/rfkill/rfkill.c
+++ b/net/rfkill/rfkill.c
@@ -180,6 +180,9 @@ static ssize_t rfkill_type_show(struct device *dev,
 	case RFKILL_TYPE_WIMAX:
 		type = "wimax";
 		break;
+	case RFKILL_TYPE_WWAN:
+		type = "wwan";
+		break;
 	default:
 		BUG();
 	}
-- 
cgit v1.2.3


From 79399a8d1908f6a406e82d23c5a9937e1722ed3a Mon Sep 17 00:00:00 2001
From: Henrique de Moraes Holschuh <hmh@hmh.eng.br>
Date: Mon, 23 Jun 2008 17:23:03 -0300
Subject: rfkill: add notifier chains support

Add a notifier chain for use by the rfkill class.  This notifier chain
signals the following events (more to be added when needed):

  1. rfkill: rfkill device state has changed

A pointer to the rfkill struct will be passed as a parameter.

The notifier message types have been added to include/linux/rfkill.h
instead of to include/linux/notifier.h in order to avoid the madness of
modifying a header used globally (and that triggers an almost full tree
rebuild every time it is touched) with information that is of interest only
to code that includes the rfkill.h header.

Signed-off-by: Henrique de Moraes Holschuh <hmh@hmh.eng.br>
Acked-by: Ivo van Doorn <IvDoorn@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/linux/rfkill.h |  7 +++++
 net/rfkill/rfkill.c    | 70 +++++++++++++++++++++++++++++++++++++++++++++++---
 2 files changed, 74 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/linux/rfkill.h b/include/linux/rfkill.h
index c0cab7d37828..98667becdee4 100644
--- a/include/linux/rfkill.h
+++ b/include/linux/rfkill.h
@@ -117,4 +117,11 @@ static inline char *rfkill_get_led_name(struct rfkill *rfkill)
 #endif
 }
 
+/* rfkill notification chain */
+#define RFKILL_STATE_CHANGED		0x0001	/* state of a normal rfkill
+						   switch has changed */
+
+int register_rfkill_notifier(struct notifier_block *nb);
+int unregister_rfkill_notifier(struct notifier_block *nb);
+
 #endif /* RFKILL_H */
diff --git a/net/rfkill/rfkill.c b/net/rfkill/rfkill.c
index fb566902030a..a561e350a70a 100644
--- a/net/rfkill/rfkill.c
+++ b/net/rfkill/rfkill.c
@@ -46,6 +46,49 @@ MODULE_PARM_DESC(default_state,
 
 static enum rfkill_state rfkill_states[RFKILL_TYPE_MAX];
 
+static BLOCKING_NOTIFIER_HEAD(rfkill_notifier_list);
+
+
+/**
+ * register_rfkill_notifier - Add notifier to rfkill notifier chain
+ * @nb: pointer to the new entry to add to the chain
+ *
+ * See blocking_notifier_chain_register() for return value and further
+ * observations.
+ *
+ * Adds a notifier to the rfkill notifier chain.  The chain will be
+ * called with a pointer to the relevant rfkill structure as a parameter,
+ * refer to include/linux/rfkill.h for the possible events.
+ *
+ * Notifiers added to this chain are to always return NOTIFY_DONE.  This
+ * chain is a blocking notifier chain: notifiers can sleep.
+ *
+ * Calls to this chain may have been done through a workqueue.  One must
+ * assume unordered asynchronous behaviour, there is no way to know if
+ * actions related to the event that generated the notification have been
+ * carried out already.
+ */
+int register_rfkill_notifier(struct notifier_block *nb)
+{
+	return blocking_notifier_chain_register(&rfkill_notifier_list, nb);
+}
+EXPORT_SYMBOL_GPL(register_rfkill_notifier);
+
+/**
+ * unregister_rfkill_notifier - remove notifier from rfkill notifier chain
+ * @nb: pointer to the entry to remove from the chain
+ *
+ * See blocking_notifier_chain_unregister() for return value and further
+ * observations.
+ *
+ * Removes a notifier from the rfkill notifier chain.
+ */
+int unregister_rfkill_notifier(struct notifier_block *nb)
+{
+	return blocking_notifier_chain_unregister(&rfkill_notifier_list, nb);
+}
+EXPORT_SYMBOL_GPL(unregister_rfkill_notifier);
+
 
 static void rfkill_led_trigger(struct rfkill *rfkill,
 			       enum rfkill_state state)
@@ -62,14 +105,25 @@ static void rfkill_led_trigger(struct rfkill *rfkill,
 #endif /* CONFIG_RFKILL_LEDS */
 }
 
+static void notify_rfkill_state_change(struct rfkill *rfkill)
+{
+	blocking_notifier_call_chain(&rfkill_notifier_list,
+			RFKILL_STATE_CHANGED,
+			rfkill);
+}
+
 static void update_rfkill_state(struct rfkill *rfkill)
 {
-	enum rfkill_state newstate;
+	enum rfkill_state newstate, oldstate;
 
 	if (rfkill->get_state) {
 		mutex_lock(&rfkill->mutex);
-		if (!rfkill->get_state(rfkill->data, &newstate))
+		if (!rfkill->get_state(rfkill->data, &newstate)) {
+			oldstate = rfkill->state;
 			rfkill->state = newstate;
+			if (oldstate != newstate)
+				notify_rfkill_state_change(rfkill);
+		}
 		mutex_unlock(&rfkill->mutex);
 	}
 }
@@ -93,8 +147,10 @@ static int rfkill_toggle_radio(struct rfkill *rfkill,
 			rfkill->state = state;
 	}
 
-	if (force || rfkill->state != oldstate)
+	if (force || rfkill->state != oldstate) {
 		rfkill_led_trigger(rfkill, rfkill->state);
+		notify_rfkill_state_change(rfkill);
+	}
 
 	return retval;
 }
@@ -139,12 +195,20 @@ EXPORT_SYMBOL(rfkill_switch_all);
  */
 int rfkill_force_state(struct rfkill *rfkill, enum rfkill_state state)
 {
+	enum rfkill_state oldstate;
+
 	if (state != RFKILL_STATE_OFF &&
 	    state != RFKILL_STATE_ON)
 		return -EINVAL;
 
 	mutex_lock(&rfkill->mutex);
+
+	oldstate = rfkill->state;
 	rfkill->state = state;
+
+	if (state != oldstate)
+		notify_rfkill_state_change(rfkill);
+
 	mutex_unlock(&rfkill->mutex);
 
 	return 0;
-- 
cgit v1.2.3


From 5005657cbd0fd6f277f807c0612a6b6d4396a02c Mon Sep 17 00:00:00 2001
From: Henrique de Moraes Holschuh <hmh@hmh.eng.br>
Date: Mon, 23 Jun 2008 17:46:42 -0300
Subject: rfkill: rename the rfkill_state states and add block-locked state

The current naming of rfkill_state causes a lot of confusion: not only the
"kill" in rfkill suggests negative logic, but also the fact that rfkill cannot
turn anything on (it can just force something off or stop forcing something
off) is often forgotten.

Rename RFKILL_STATE_OFF to RFKILL_STATE_SOFT_BLOCKED (transmitter is blocked
and will not operate; state can be changed by a toggle_radio request), and
RFKILL_STATE_ON to RFKILL_STATE_UNBLOCKED (transmitter is not blocked, and may
operate).

Also, add a new third state, RFKILL_STATE_HARD_BLOCKED (transmitter is blocked
and will not operate; state cannot be changed through a toggle_radio request),
which is used by drivers to indicate a wireless transmiter was blocked by a
hardware rfkill line that accepts no overrides.

Keep the old names as #defines, but document them as deprecated.  This way,
drivers can be converted to the new names *and* verified to actually use rfkill
correctly one by one.

Signed-off-by: Henrique de Moraes Holschuh <hmh@hmh.eng.br>
Acked-by: Ivo van Doorn <IvDoorn@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 Documentation/rfkill.txt  | 56 +++++++++++++++++++++++++++++------
 include/linux/rfkill.h    | 32 ++++++++++++++++++--
 net/rfkill/rfkill-input.c | 29 +++++++++---------
 net/rfkill/rfkill.c       | 75 ++++++++++++++++++++++++++++++++++++++---------
 4 files changed, 152 insertions(+), 40 deletions(-)

(limited to 'include')

diff --git a/Documentation/rfkill.txt b/Documentation/rfkill.txt
index cf230c1ad9ef..5316cea95ce0 100644
--- a/Documentation/rfkill.txt
+++ b/Documentation/rfkill.txt
@@ -60,9 +60,20 @@ The second option provides an rfkill input handler. This handler will listen to
 all rfkill key events and will toggle the radio accordingly.  With this option
 enabled userspace could either do nothing or simply perform monitoring tasks.
 
-When a rfkill switch is in the RFKILL_STATE_ON, the wireless transmitter (radio
-TX circuit for example) is *enabled*.  When the rfkill switch is in the
-RFKILL_STATE_OFF, the wireless transmitter is to be *blocked* from operating.
+When a rfkill switch is in the RFKILL_STATE_UNBLOCKED, the wireless transmitter
+(radio TX circuit for example) is *enabled*.  When the rfkill switch is in the
+RFKILL_STATE_SOFT_BLOCKED or RFKILL_STATE_HARD_BLOCKED, the wireless
+transmitter is to be *blocked* from operating.
+
+RFKILL_STATE_SOFT_BLOCKED indicates that a call to toggle_radio() can change
+that state.  RFKILL_STATE_HARD_BLOCKED indicates that a call to toggle_radio()
+will not be able to change the state and will return with a suitable error if
+attempts are made to set the state to RFKILL_STATE_UNBLOCKED.
+
+RFKILL_STATE_HARD_BLOCKED is used by drivers to signal that the device is
+locked in the BLOCKED state by a hardwire rfkill line (typically an input pin
+that, when active, forces the transmitter to be disabled) which the driver
+CANNOT override.
 
 Full rfkill functionality requires two different subsystems to cooperate: the
 input layer and the rfkill class.  The input layer issues *commands* to the
@@ -122,10 +133,10 @@ Userspace input handlers (uevents) or kernel input handlers (rfkill-input):
 	  action).
 	* rfkill-input implements EPO by handling EV_SW SW_RFKILL_ALL 0
 	  (power off all transmitters) in a special way: it ignores any
-	  overrides and local state cache and forces all transmitters to
-	  the OFF state (including those which are already supposed to be
-	  OFF).  Note that the opposite event (power on all transmitters)
-	  is handled normally.
+	  overrides and local state cache and forces all transmitters to the
+	  RFKILL_STATE_SOFT_BLOCKED state (including those which are already
+	  supposed to be BLOCKED).  Note that the opposite event (power on all
+	  transmitters) is handled normally.
 
 Userspace uevent handler or kernel platform-specific drivers hooked to the
 rfkill notifier chain:
@@ -284,6 +295,19 @@ You should:
 	  YOU CAN ACCESS state DIRECTLY)
 	- rfkill_register()
 
+The only way to set a device to the RFKILL_STATE_HARD_BLOCKED state is through
+a suitable return of get_state() or through rfkill_force_state().
+
+When a device is in the RFKILL_STATE_HARD_BLOCKED state, the only way to switch
+it to a different state is through a suitable return of get_state() or through
+rfkill_force_state().
+
+If toggle_radio() is called to set a device to state RFKILL_STATE_SOFT_BLOCKED
+when that device is already at the RFKILL_STATE_HARD_BLOCKED state, it should
+not return an error.  Instead, it should try to double-block the transmitter,
+so that its state will change from RFKILL_STATE_HARD_BLOCKED to
+RFKILL_STATE_SOFT_BLOCKED should the hardware blocking cease.
+
 Please refer to the source for more documentation.
 
 ===============================================================================
@@ -322,13 +346,27 @@ change by writing to the "state" attribute in order for anything to happen.
 
 Take particular care to implement EV_SW SW_RFKILL_ALL properly.  When that
 switch is set to OFF, *every* rfkill device *MUST* be immediately put into the
-OFF state, no questions asked.
+RFKILL_STATE_SOFT_BLOCKED state, no questions asked.
 
 The following sysfs entries will be created:
 
 	name: Name assigned by driver to this key (interface or driver name).
 	type: Name of the key type ("wlan", "bluetooth", etc).
-	state: Current state of the key. 1: On, 0: Off.
+	state: Current state of the transmitter
+		0: RFKILL_STATE_SOFT_BLOCKED
+			transmitter is forced off, but you can override it
+			by a write to the state attribute, or through input
+			events (if rfkill-input is loaded).
+		1: RFKILL_STATE_UNBLOCKED
+			transmiter is NOT forced off, and may operate if
+			all other conditions for such operation are met
+			(such as interface is up and configured, etc).
+		2: RFKILL_STATE_HARD_BLOCKED
+			transmitter is forced off by something outside of
+			the driver's control.
+
+			You cannot set a device to this state through
+			writes to the state attribute.
 	claim: 1: Userspace handles events, 0: Kernel handles events
 
 Both the "state" and "claim" entries are also writable. For the "state" entry
diff --git a/include/linux/rfkill.h b/include/linux/rfkill.h
index 98667becdee4..c5f6e54ec6ae 100644
--- a/include/linux/rfkill.h
+++ b/include/linux/rfkill.h
@@ -46,16 +46,25 @@ enum rfkill_type {
 };
 
 enum rfkill_state {
-	RFKILL_STATE_OFF	= 0,	/* Radio output blocked */
-	RFKILL_STATE_ON		= 1,	/* Radio output active */
+	RFKILL_STATE_SOFT_BLOCKED = 0,	/* Radio output blocked */
+	RFKILL_STATE_UNBLOCKED    = 1,	/* Radio output allowed */
+	RFKILL_STATE_HARD_BLOCKED = 2,	/* Output blocked, non-overrideable */
 };
 
+/*
+ * These are DEPRECATED, drivers using them should be verified to
+ * comply with the rfkill usage guidelines in Documentation/rfkill.txt
+ * and then converted to use the new names for rfkill_state
+ */
+#define RFKILL_STATE_OFF RFKILL_STATE_SOFT_BLOCKED
+#define RFKILL_STATE_ON  RFKILL_STATE_UNBLOCKED
+
 /**
  * struct rfkill - rfkill control structure.
  * @name: Name of the switch.
  * @type: Radio type which the button controls, the value stored
  *	here should be a value from enum rfkill_type.
- * @state: State of the switch, "ON" means radio can operate.
+ * @state: State of the switch, "UNBLOCKED" means radio can operate.
  * @user_claim_unsupported: Whether the hardware supports exclusive
  *	RF-kill control by userspace. Set this before registering.
  * @user_claim: Set when the switch is controlled exlusively by userspace.
@@ -63,8 +72,12 @@ enum rfkill_state {
  * @data: Pointer to the RF button drivers private data which will be
  *	passed along when toggling radio state.
  * @toggle_radio(): Mandatory handler to control state of the radio.
+ *	only RFKILL_STATE_SOFT_BLOCKED and RFKILL_STATE_UNBLOCKED are
+ *	valid parameters.
  * @get_state(): handler to read current radio state from hardware,
  *      may be called from atomic context, should return 0 on success.
+ *      Either this handler OR judicious use of rfkill_force_state() is
+ *      MANDATORY for any driver capable of RFKILL_STATE_HARD_BLOCKED.
  * @led_trigger: A LED trigger for this button's LED.
  * @dev: Device structure integrating the switch into device tree.
  * @node: Used to place switch into list of all switches known to the
@@ -102,6 +115,19 @@ void rfkill_unregister(struct rfkill *rfkill);
 
 int rfkill_force_state(struct rfkill *rfkill, enum rfkill_state state);
 
+/**
+ * rfkill_state_complement - return complementar state
+ * @state: state to return the complement of
+ *
+ * Returns RFKILL_STATE_SOFT_BLOCKED if @state is RFKILL_STATE_UNBLOCKED,
+ * returns RFKILL_STATE_UNBLOCKED otherwise.
+ */
+static inline enum rfkill_state rfkill_state_complement(enum rfkill_state state)
+{
+	return (state == RFKILL_STATE_UNBLOCKED) ?
+		RFKILL_STATE_SOFT_BLOCKED : RFKILL_STATE_UNBLOCKED;
+}
+
 /**
  * rfkill_get_led_name - Get the LED trigger name for the button's LED.
  * This function might return a NULL pointer if registering of the
diff --git a/net/rfkill/rfkill-input.c b/net/rfkill/rfkill-input.c
index 5d4c8b2446f7..8aa822730145 100644
--- a/net/rfkill/rfkill-input.c
+++ b/net/rfkill/rfkill-input.c
@@ -84,7 +84,8 @@ static void rfkill_schedule_toggle(struct rfkill_task *task)
 	spin_lock_irqsave(&task->lock, flags);
 
 	if (time_after(jiffies, task->last + msecs_to_jiffies(200))) {
-		task->desired_state = !task->desired_state;
+		task->desired_state =
+				rfkill_state_complement(task->desired_state);
 		task->last = jiffies;
 		schedule_work(&task->work);
 	}
@@ -92,14 +93,14 @@ static void rfkill_schedule_toggle(struct rfkill_task *task)
 	spin_unlock_irqrestore(&task->lock, flags);
 }
 
-#define DEFINE_RFKILL_TASK(n, t)			\
-	struct rfkill_task n = {			\
-		.work = __WORK_INITIALIZER(n.work,	\
-				rfkill_task_handler),	\
-		.type = t,				\
-		.mutex = __MUTEX_INITIALIZER(n.mutex),	\
-		.lock = __SPIN_LOCK_UNLOCKED(n.lock),	\
-		.desired_state = RFKILL_STATE_ON,	\
+#define DEFINE_RFKILL_TASK(n, t)				\
+	struct rfkill_task n = {				\
+		.work = __WORK_INITIALIZER(n.work,		\
+				rfkill_task_handler),		\
+		.type = t,					\
+		.mutex = __MUTEX_INITIALIZER(n.mutex),		\
+		.lock = __SPIN_LOCK_UNLOCKED(n.lock),		\
+		.desired_state = RFKILL_STATE_UNBLOCKED,	\
 	}
 
 static DEFINE_RFKILL_TASK(rfkill_wlan, RFKILL_TYPE_WLAN);
@@ -135,15 +136,15 @@ static void rfkill_event(struct input_handle *handle, unsigned int type,
 			/* handle EPO (emergency power off) through shortcut */
 			if (data) {
 				rfkill_schedule_set(&rfkill_wwan,
-						    RFKILL_STATE_ON);
+						    RFKILL_STATE_UNBLOCKED);
 				rfkill_schedule_set(&rfkill_wimax,
-						    RFKILL_STATE_ON);
+						    RFKILL_STATE_UNBLOCKED);
 				rfkill_schedule_set(&rfkill_uwb,
-						    RFKILL_STATE_ON);
+						    RFKILL_STATE_UNBLOCKED);
 				rfkill_schedule_set(&rfkill_bt,
-						    RFKILL_STATE_ON);
+						    RFKILL_STATE_UNBLOCKED);
 				rfkill_schedule_set(&rfkill_wlan,
-						    RFKILL_STATE_ON);
+						    RFKILL_STATE_UNBLOCKED);
 			} else
 				rfkill_schedule_epo();
 			break;
diff --git a/net/rfkill/rfkill.c b/net/rfkill/rfkill.c
index 7d07175c407f..ce0e23148cdd 100644
--- a/net/rfkill/rfkill.c
+++ b/net/rfkill/rfkill.c
@@ -39,7 +39,7 @@ MODULE_LICENSE("GPL");
 static LIST_HEAD(rfkill_list);	/* list of registered rf switches */
 static DEFINE_MUTEX(rfkill_mutex);
 
-static unsigned int rfkill_default_state = RFKILL_STATE_ON;
+static unsigned int rfkill_default_state = RFKILL_STATE_UNBLOCKED;
 module_param_named(default_state, rfkill_default_state, uint, 0444);
 MODULE_PARM_DESC(default_state,
 		 "Default initial state for all radio types, 0 = radio off");
@@ -98,7 +98,7 @@ static void rfkill_led_trigger(struct rfkill *rfkill,
 
 	if (!led->name)
 		return;
-	if (state == RFKILL_STATE_OFF)
+	if (state != RFKILL_STATE_UNBLOCKED)
 		led_trigger_event(led, LED_OFF);
 	else
 		led_trigger_event(led, LED_FULL);
@@ -128,6 +128,28 @@ static void update_rfkill_state(struct rfkill *rfkill)
 	}
 }
 
+/**
+ * rfkill_toggle_radio - wrapper for toggle_radio hook
+ * calls toggle_radio taking into account a lot of "small"
+ * details.
+ * @rfkill: the rfkill struct to use
+ * @force: calls toggle_radio even if cache says it is not needed,
+ *	and also makes sure notifications of the state will be
+ *	sent even if it didn't change
+ * @state: the new state to call toggle_radio() with
+ *
+ * This wrappen protects and enforces the API for toggle_radio
+ * calls.  Note that @force cannot override a (possibly cached)
+ * state of RFKILL_STATE_HARD_BLOCKED.  Any device making use of
+ * RFKILL_STATE_HARD_BLOCKED implements either get_state() or
+ * rfkill_force_state(), so the cache either is bypassed or valid.
+ *
+ * Note that we do call toggle_radio for RFKILL_STATE_SOFT_BLOCKED
+ * even if the radio is in RFKILL_STATE_HARD_BLOCKED state, so as to
+ * give the driver a hint that it should double-BLOCK the transmitter.
+ *
+ * Caller must have aquired rfkill_mutex.
+ */
 static int rfkill_toggle_radio(struct rfkill *rfkill,
 				enum rfkill_state state,
 				int force)
@@ -141,9 +163,28 @@ static int rfkill_toggle_radio(struct rfkill *rfkill,
 	    !rfkill->get_state(rfkill->data, &newstate))
 		rfkill->state = newstate;
 
+	switch (state) {
+	case RFKILL_STATE_HARD_BLOCKED:
+		/* typically happens when refreshing hardware state,
+		 * such as on resume */
+		state = RFKILL_STATE_SOFT_BLOCKED;
+		break;
+	case RFKILL_STATE_UNBLOCKED:
+		/* force can't override this, only rfkill_force_state() can */
+		if (rfkill->state == RFKILL_STATE_HARD_BLOCKED)
+			return -EPERM;
+		break;
+	case RFKILL_STATE_SOFT_BLOCKED:
+		/* nothing to do, we want to give drivers the hint to double
+		 * BLOCK even a transmitter that is already in state
+		 * RFKILL_STATE_HARD_BLOCKED */
+		break;
+	}
+
 	if (force || state != rfkill->state) {
 		retval = rfkill->toggle_radio(rfkill->data, state);
-		if (!retval)
+		/* never allow a HARD->SOFT downgrade! */
+		if (!retval && rfkill->state != RFKILL_STATE_HARD_BLOCKED)
 			rfkill->state = state;
 	}
 
@@ -184,7 +225,7 @@ EXPORT_SYMBOL(rfkill_switch_all);
 /**
  * rfkill_epo - emergency power off all transmitters
  *
- * This kicks all rfkill devices to RFKILL_STATE_OFF, ignoring
+ * This kicks all rfkill devices to RFKILL_STATE_SOFT_BLOCKED, ignoring
  * everything in its path but rfkill_mutex.
  */
 void rfkill_epo(void)
@@ -193,7 +234,7 @@ void rfkill_epo(void)
 
 	mutex_lock(&rfkill_mutex);
 	list_for_each_entry(rfkill, &rfkill_list, node) {
-		rfkill_toggle_radio(rfkill, RFKILL_STATE_OFF, 1);
+		rfkill_toggle_radio(rfkill, RFKILL_STATE_SOFT_BLOCKED, 1);
 	}
 	mutex_unlock(&rfkill_mutex);
 }
@@ -215,8 +256,9 @@ int rfkill_force_state(struct rfkill *rfkill, enum rfkill_state state)
 {
 	enum rfkill_state oldstate;
 
-	if (state != RFKILL_STATE_OFF &&
-	    state != RFKILL_STATE_ON)
+	if (state != RFKILL_STATE_SOFT_BLOCKED &&
+	    state != RFKILL_STATE_UNBLOCKED &&
+	    state != RFKILL_STATE_HARD_BLOCKED)
 		return -EINVAL;
 
 	mutex_lock(&rfkill->mutex);
@@ -290,11 +332,14 @@ static ssize_t rfkill_state_store(struct device *dev,
 	if (!capable(CAP_NET_ADMIN))
 		return -EPERM;
 
+	/* RFKILL_STATE_HARD_BLOCKED is illegal here... */
+	if (state != RFKILL_STATE_UNBLOCKED &&
+	    state != RFKILL_STATE_SOFT_BLOCKED)
+		return -EINVAL;
+
 	if (mutex_lock_interruptible(&rfkill->mutex))
 		return -ERESTARTSYS;
-	error = rfkill_toggle_radio(rfkill,
-			state ? RFKILL_STATE_ON : RFKILL_STATE_OFF,
-			0);
+	error = rfkill_toggle_radio(rfkill, state, 0);
 	mutex_unlock(&rfkill->mutex);
 
 	return error ? error : count;
@@ -373,7 +418,8 @@ static int rfkill_suspend(struct device *dev, pm_message_t state)
 			update_rfkill_state(rfkill);
 
 			mutex_lock(&rfkill->mutex);
-			rfkill->toggle_radio(rfkill->data, RFKILL_STATE_OFF);
+			rfkill->toggle_radio(rfkill->data,
+						RFKILL_STATE_SOFT_BLOCKED);
 			mutex_unlock(&rfkill->mutex);
 		}
 
@@ -470,7 +516,7 @@ static void rfkill_remove_switch(struct rfkill *rfkill)
 {
 	mutex_lock(&rfkill_mutex);
 	list_del_init(&rfkill->node);
-	rfkill_toggle_radio(rfkill, RFKILL_STATE_OFF, 1);
+	rfkill_toggle_radio(rfkill, RFKILL_STATE_SOFT_BLOCKED, 1);
 	mutex_unlock(&rfkill_mutex);
 }
 
@@ -610,8 +656,9 @@ static int __init rfkill_init(void)
 	int error;
 	int i;
 
-	if (rfkill_default_state != RFKILL_STATE_OFF &&
-	    rfkill_default_state != RFKILL_STATE_ON)
+	/* RFKILL_STATE_HARD_BLOCKED is illegal here... */
+	if (rfkill_default_state != RFKILL_STATE_SOFT_BLOCKED &&
+	    rfkill_default_state != RFKILL_STATE_UNBLOCKED)
 		return -EINVAL;
 
 	for (i = 0; i < ARRAY_SIZE(rfkill_states); i++)
-- 
cgit v1.2.3


From f2df38596a81b6c24f4586b0b4befeaebf3e02db Mon Sep 17 00:00:00 2001
From: Assaf Krauss <assaf.krauss@intel.com>
Date: Sun, 15 Jun 2008 18:23:29 +0300
Subject: mac80211: 11h Infrastructure - Parsing

This patch introduces parsing of 11h and 11d related elements from incoming
management frames.

Signed-off-by: Assaf Krauss <assaf.krauss@intel.com>
Signed-off-by: Tomas Winkler <tomas.winkler@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/linux/ieee80211.h  | 54 +++++++++++++++++++++++++++++++++++++++++++---
 net/mac80211/ieee80211_i.h |  9 ++++++++
 net/mac80211/mlme.c        | 19 ++++++++++++++++
 3 files changed, 79 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/linux/ieee80211.h b/include/linux/ieee80211.h
index 2998e3b5f166..8546f09e462c 100644
--- a/include/linux/ieee80211.h
+++ b/include/linux/ieee80211.h
@@ -469,6 +469,40 @@ struct ieee80211s_hdr {
 	u8 eaddr3[6];
 } __attribute__ ((packed));
 
+/**
+ * struct ieee80211_quiet_ie
+ *
+ * This structure refers to "Quiet information element"
+ */
+struct ieee80211_quiet_ie {
+	u8 count;
+	u8 period;
+	__le16 duration;
+	__le16 offset;
+} __attribute__ ((packed));
+
+/**
+ * struct ieee80211_msrment_ie
+ *
+ * This structure refers to "Measurement Request/Report information element"
+ */
+struct ieee80211_msrment_ie {
+	u8 token;
+	u8 mode;
+	u8 type;
+	u8 request[0];
+} __attribute__ ((packed));
+
+/**
+ * struct ieee80211_channel_sw_ie
+ *
+ * This structure refers to "Channel Switch Announcement information element"
+ */
+struct ieee80211_channel_sw_ie {
+	u8 mode;
+	u8 new_ch_num;
+	u8 count;
+} __attribute__ ((packed));
 
 struct ieee80211_mgmt {
 	__le16 frame_control;
@@ -544,10 +578,15 @@ struct ieee80211_mgmt {
 					u8 action_code;
 					u8 element_id;
 					u8 length;
-					u8 switch_mode;
-					u8 new_chan;
-					u8 switch_count;
+					struct ieee80211_channel_sw_ie sw_elem;
 				} __attribute__((packed)) chan_switch;
+				struct{
+					u8 action_code;
+					u8 dialog_token;
+					u8 element_id;
+					u8 length;
+					struct ieee80211_msrment_ie msr_elem;
+				} __attribute__((packed)) measurement;
 				struct{
 					u8 action_code;
 					u8 dialog_token;
@@ -875,6 +914,15 @@ enum ieee80211_category {
 	WLAN_CATEGORY_WMM = 17,
 };
 
+/* SPECTRUM_MGMT action code */
+enum ieee80211_spectrum_mgmt_actioncode {
+	WLAN_ACTION_SPCT_MSR_REQ = 0,
+	WLAN_ACTION_SPCT_MSR_RPRT = 1,
+	WLAN_ACTION_SPCT_TPC_REQ = 2,
+	WLAN_ACTION_SPCT_TPC_RPRT = 3,
+	WLAN_ACTION_SPCT_CHL_SWITCH = 4,
+};
+
 /* BACK action code */
 enum ieee80211_back_actioncode {
 	WLAN_ACTION_ADDBA_REQ = 0,
diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h
index 80a9e7c07b47..af352c05c983 100644
--- a/net/mac80211/ieee80211_i.h
+++ b/net/mac80211/ieee80211_i.h
@@ -791,6 +791,10 @@ struct ieee802_11_elems {
 	u8 *preq;
 	u8 *prep;
 	u8 *perr;
+	u8 *ch_switch_elem;
+	u8 *country_elem;
+	u8 *pwr_constr_elem;
+	u8 *quiet_elem; 	/* first quite element */
 
 	/* length of them, respectively */
 	u8 ssid_len;
@@ -815,6 +819,11 @@ struct ieee802_11_elems {
 	u8 preq_len;
 	u8 prep_len;
 	u8 perr_len;
+	u8 ch_switch_elem_len;
+	u8 country_elem_len;
+	u8 pwr_constr_elem_len;
+	u8 quiet_elem_len;
+	u8 num_of_quiet_elem;	/* can be more the one */
 };
 
 static inline struct ieee80211_local *hw_to_local(
diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c
index e06d6450f215..32453561fe32 100644
--- a/net/mac80211/mlme.c
+++ b/net/mac80211/mlme.c
@@ -204,6 +204,25 @@ void ieee802_11_parse_elems(u8 *start, size_t len,
 			elems->perr = pos;
 			elems->perr_len = elen;
 			break;
+		case WLAN_EID_CHANNEL_SWITCH:
+			elems->ch_switch_elem = pos;
+			elems->ch_switch_elem_len = elen;
+			break;
+		case WLAN_EID_QUIET:
+			if (!elems->quiet_elem) {
+				elems->quiet_elem = pos;
+				elems->quiet_elem_len = elen;
+			}
+			elems->num_of_quiet_elem++;
+			break;
+		case WLAN_EID_COUNTRY:
+			elems->country_elem = pos;
+			elems->country_elem_len = elen;
+			break;
+		case WLAN_EID_PWR_CONSTRAINT:
+			elems->pwr_constr_elem = pos;
+			elems->pwr_constr_elem_len = elen;
+			break;
 		default:
 			break;
 		}
-- 
cgit v1.2.3


From b662348662f9661f9259c7186c1bdb65620045f1 Mon Sep 17 00:00:00 2001
From: Assaf Krauss <assaf.krauss@intel.com>
Date: Mon, 16 Jun 2008 16:09:49 +0300
Subject: mac80211: 11h - Handling measurement request

This patch handles the 11h measurement request information element.
This is minimal requested implementation - refuse measurement.

Signed-off-by: Assaf Krauss <assaf.krauss@intel.com>
Signed-off-by: Tomas Winkler <tomas.winkler@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/linux/ieee80211.h | 10 ++++++
 net/mac80211/mlme.c       | 78 +++++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 88 insertions(+)

(limited to 'include')

diff --git a/include/linux/ieee80211.h b/include/linux/ieee80211.h
index 8546f09e462c..cffd6d0094f9 100644
--- a/include/linux/ieee80211.h
+++ b/include/linux/ieee80211.h
@@ -739,11 +739,21 @@ struct ieee80211_ht_addt_info {
 #define WLAN_CAPABILITY_SHORT_PREAMBLE	(1<<5)
 #define WLAN_CAPABILITY_PBCC		(1<<6)
 #define WLAN_CAPABILITY_CHANNEL_AGILITY	(1<<7)
+
 /* 802.11h */
 #define WLAN_CAPABILITY_SPECTRUM_MGMT	(1<<8)
 #define WLAN_CAPABILITY_QOS		(1<<9)
 #define WLAN_CAPABILITY_SHORT_SLOT_TIME	(1<<10)
 #define WLAN_CAPABILITY_DSSS_OFDM	(1<<13)
+/* measurement */
+#define IEEE80211_SPCT_MSR_RPRT_MODE_LATE	(1<<0)
+#define IEEE80211_SPCT_MSR_RPRT_MODE_INCAPABLE	(1<<1)
+#define IEEE80211_SPCT_MSR_RPRT_MODE_REFUSED	(1<<2)
+
+#define IEEE80211_SPCT_MSR_RPRT_TYPE_BASIC	0
+#define IEEE80211_SPCT_MSR_RPRT_TYPE_CCA	1
+#define IEEE80211_SPCT_MSR_RPRT_TYPE_RPI	2
+
 
 /* 802.11g ERP information element */
 #define WLAN_ERP_NON_ERP_PRESENT (1<<0)
diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c
index 32453561fe32..6d9a4facf9dd 100644
--- a/net/mac80211/mlme.c
+++ b/net/mac80211/mlme.c
@@ -1720,6 +1720,71 @@ void ieee80211_sta_tear_down_BA_sessions(struct net_device *dev, u8 *addr)
 	}
 }
 
+static void ieee80211_send_refuse_measurement_request(struct net_device *dev,
+					struct ieee80211_msrment_ie *request_ie,
+					const u8 *da, const u8 *bssid,
+					u8 dialog_token)
+{
+	struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr);
+	struct sk_buff *skb;
+	struct ieee80211_mgmt *msr_report;
+
+	skb = dev_alloc_skb(sizeof(*msr_report) + local->hw.extra_tx_headroom +
+				sizeof(struct ieee80211_msrment_ie));
+
+	if (!skb) {
+		printk(KERN_ERR "%s: failed to allocate buffer for "
+				"measurement report frame\n", dev->name);
+		return;
+	}
+
+	skb_reserve(skb, local->hw.extra_tx_headroom);
+	msr_report = (struct ieee80211_mgmt *)skb_put(skb, 24);
+	memset(msr_report, 0, 24);
+	memcpy(msr_report->da, da, ETH_ALEN);
+	memcpy(msr_report->sa, dev->dev_addr, ETH_ALEN);
+	memcpy(msr_report->bssid, bssid, ETH_ALEN);
+	msr_report->frame_control = IEEE80211_FC(IEEE80211_FTYPE_MGMT,
+						IEEE80211_STYPE_ACTION);
+
+	skb_put(skb, 1 + sizeof(msr_report->u.action.u.measurement));
+	msr_report->u.action.category = WLAN_CATEGORY_SPECTRUM_MGMT;
+	msr_report->u.action.u.measurement.action_code =
+				WLAN_ACTION_SPCT_MSR_RPRT;
+	msr_report->u.action.u.measurement.dialog_token = dialog_token;
+
+	msr_report->u.action.u.measurement.element_id = WLAN_EID_MEASURE_REPORT;
+	msr_report->u.action.u.measurement.length =
+			sizeof(struct ieee80211_msrment_ie);
+
+	memset(&msr_report->u.action.u.measurement.msr_elem, 0,
+		sizeof(struct ieee80211_msrment_ie));
+	msr_report->u.action.u.measurement.msr_elem.token = request_ie->token;
+	msr_report->u.action.u.measurement.msr_elem.mode |=
+			IEEE80211_SPCT_MSR_RPRT_MODE_REFUSED;
+	msr_report->u.action.u.measurement.msr_elem.type = request_ie->type;
+
+	ieee80211_sta_tx(dev, skb, 0);
+}
+
+static void ieee80211_sta_process_measurement_req(struct net_device *dev,
+						struct ieee80211_mgmt *mgmt,
+						size_t len)
+{
+	/*
+	 * Ignoring measurement request is spec violation.
+	 * Mandatory measurements must be reported optional
+	 * measurements might be refused or reported incapable
+	 * For now just refuse
+	 * TODO: Answer basic measurement as unmeasured
+	 */
+	ieee80211_send_refuse_measurement_request(dev,
+			&mgmt->u.action.u.measurement.msr_elem,
+			mgmt->sa, mgmt->bssid,
+			mgmt->u.action.u.measurement.dialog_token);
+}
+
+
 static void ieee80211_rx_mgmt_auth(struct net_device *dev,
 				   struct ieee80211_if_sta *ifsta,
 				   struct ieee80211_mgmt *mgmt,
@@ -3044,11 +3109,24 @@ static void ieee80211_rx_mgmt_action(struct net_device *dev,
 				     struct ieee80211_rx_status *rx_status)
 {
 	struct ieee80211_sub_if_data *sdata = IEEE80211_DEV_TO_SUB_IF(dev);
+	struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr);
 
 	if (len < IEEE80211_MIN_ACTION_SIZE)
 		return;
 
 	switch (mgmt->u.action.category) {
+	case WLAN_CATEGORY_SPECTRUM_MGMT:
+		if (local->hw.conf.channel->band != IEEE80211_BAND_5GHZ)
+			break;
+		switch (mgmt->u.action.u.chan_switch.action_code) {
+		case WLAN_ACTION_SPCT_MSR_REQ:
+			if (len < (IEEE80211_MIN_ACTION_SIZE +
+				   sizeof(mgmt->u.action.u.measurement)))
+				break;
+			ieee80211_sta_process_measurement_req(dev, mgmt, len);
+			break;
+		}
+		break;
 	case WLAN_CATEGORY_BACK:
 		switch (mgmt->u.action.u.addba_req.action_code) {
 		case WLAN_ACTION_ADDBA_REQ:
-- 
cgit v1.2.3


From f225763a7d6c92c4932dbd528437997078496fcc Mon Sep 17 00:00:00 2001
From: Michael Buesch <mb@bu3sch.de>
Date: Fri, 20 Jun 2008 11:50:29 +0200
Subject: ssb, b43, b43legacy, b44: Rewrite SSB DMA API

This is a rewrite of the DMA API for SSB devices.
This is needed, because the old (non-existing) "API" made too many bad
assumptions on the API of the host-bus (PCI).
This introduces an almost complete SSB-DMA-API that maps to the lowlevel
bus-API based on the bustype.

Signed-off-by: Michael Buesch <mb@bu3sch.de>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/net/b44.c                    | 140 +++++++++++++++++-----------------
 drivers/net/wireless/b43/dma.c       |  65 ++++++++--------
 drivers/net/wireless/b43legacy/dma.c |  63 ++++++++-------
 drivers/ssb/Kconfig                  |   2 +-
 drivers/ssb/main.c                   |  75 ++++++++++++++----
 include/linux/ssb/ssb.h              | 143 ++++++++++++++++++++++++++++++++++-
 6 files changed, 336 insertions(+), 152 deletions(-)

(limited to 'include')

diff --git a/drivers/net/b44.c b/drivers/net/b44.c
index 59dce6aa0865..c3bda5ce67c4 100644
--- a/drivers/net/b44.c
+++ b/drivers/net/b44.c
@@ -148,9 +148,9 @@ static inline void b44_sync_dma_desc_for_device(struct ssb_device *sdev,
 						unsigned long offset,
 						enum dma_data_direction dir)
 {
-	dma_sync_single_range_for_device(sdev->dma_dev, dma_base,
-					 offset & dma_desc_align_mask,
-					 dma_desc_sync_size, dir);
+	ssb_dma_sync_single_range_for_device(sdev, dma_base,
+					     offset & dma_desc_align_mask,
+					     dma_desc_sync_size, dir);
 }
 
 static inline void b44_sync_dma_desc_for_cpu(struct ssb_device *sdev,
@@ -158,9 +158,9 @@ static inline void b44_sync_dma_desc_for_cpu(struct ssb_device *sdev,
 					     unsigned long offset,
 					     enum dma_data_direction dir)
 {
-	dma_sync_single_range_for_cpu(sdev->dma_dev, dma_base,
-				      offset & dma_desc_align_mask,
-				      dma_desc_sync_size, dir);
+	ssb_dma_sync_single_range_for_cpu(sdev, dma_base,
+					  offset & dma_desc_align_mask,
+					  dma_desc_sync_size, dir);
 }
 
 static inline unsigned long br32(const struct b44 *bp, unsigned long reg)
@@ -613,10 +613,10 @@ static void b44_tx(struct b44 *bp)
 
 		BUG_ON(skb == NULL);
 
-		dma_unmap_single(bp->sdev->dma_dev,
-				 rp->mapping,
-				 skb->len,
-				 DMA_TO_DEVICE);
+		ssb_dma_unmap_single(bp->sdev,
+				     rp->mapping,
+				     skb->len,
+				     DMA_TO_DEVICE);
 		rp->skb = NULL;
 		dev_kfree_skb_irq(skb);
 	}
@@ -653,29 +653,29 @@ static int b44_alloc_rx_skb(struct b44 *bp, int src_idx, u32 dest_idx_unmasked)
 	if (skb == NULL)
 		return -ENOMEM;
 
-	mapping = dma_map_single(bp->sdev->dma_dev, skb->data,
-				 RX_PKT_BUF_SZ,
-				 DMA_FROM_DEVICE);
+	mapping = ssb_dma_map_single(bp->sdev, skb->data,
+				     RX_PKT_BUF_SZ,
+				     DMA_FROM_DEVICE);
 
 	/* Hardware bug work-around, the chip is unable to do PCI DMA
 	   to/from anything above 1GB :-( */
-	if (dma_mapping_error(mapping) ||
+	if (ssb_dma_mapping_error(bp->sdev, mapping) ||
 		mapping + RX_PKT_BUF_SZ > DMA_30BIT_MASK) {
 		/* Sigh... */
-		if (!dma_mapping_error(mapping))
-			dma_unmap_single(bp->sdev->dma_dev, mapping,
-					RX_PKT_BUF_SZ, DMA_FROM_DEVICE);
+		if (!ssb_dma_mapping_error(bp->sdev, mapping))
+			ssb_dma_unmap_single(bp->sdev, mapping,
+					     RX_PKT_BUF_SZ, DMA_FROM_DEVICE);
 		dev_kfree_skb_any(skb);
 		skb = __netdev_alloc_skb(bp->dev, RX_PKT_BUF_SZ, GFP_ATOMIC|GFP_DMA);
 		if (skb == NULL)
 			return -ENOMEM;
-		mapping = dma_map_single(bp->sdev->dma_dev, skb->data,
-					 RX_PKT_BUF_SZ,
-					 DMA_FROM_DEVICE);
-		if (dma_mapping_error(mapping) ||
+		mapping = ssb_dma_map_single(bp->sdev, skb->data,
+					     RX_PKT_BUF_SZ,
+					     DMA_FROM_DEVICE);
+		if (ssb_dma_mapping_error(bp->sdev, mapping) ||
 			mapping + RX_PKT_BUF_SZ > DMA_30BIT_MASK) {
-			if (!dma_mapping_error(mapping))
-				dma_unmap_single(bp->sdev->dma_dev, mapping, RX_PKT_BUF_SZ,DMA_FROM_DEVICE);
+			if (!ssb_dma_mapping_error(bp->sdev, mapping))
+				ssb_dma_unmap_single(bp->sdev, mapping, RX_PKT_BUF_SZ,DMA_FROM_DEVICE);
 			dev_kfree_skb_any(skb);
 			return -ENOMEM;
 		}
@@ -750,9 +750,9 @@ static void b44_recycle_rx(struct b44 *bp, int src_idx, u32 dest_idx_unmasked)
 					     dest_idx * sizeof(dest_desc),
 					     DMA_BIDIRECTIONAL);
 
-	dma_sync_single_for_device(bp->sdev->dma_dev, le32_to_cpu(src_desc->addr),
-				   RX_PKT_BUF_SZ,
-				   DMA_FROM_DEVICE);
+	ssb_dma_sync_single_for_device(bp->sdev, le32_to_cpu(src_desc->addr),
+				       RX_PKT_BUF_SZ,
+				       DMA_FROM_DEVICE);
 }
 
 static int b44_rx(struct b44 *bp, int budget)
@@ -772,7 +772,7 @@ static int b44_rx(struct b44 *bp, int budget)
 		struct rx_header *rh;
 		u16 len;
 
-		dma_sync_single_for_cpu(bp->sdev->dma_dev, map,
+		ssb_dma_sync_single_for_cpu(bp->sdev, map,
 					    RX_PKT_BUF_SZ,
 					    DMA_FROM_DEVICE);
 		rh = (struct rx_header *) skb->data;
@@ -806,8 +806,8 @@ static int b44_rx(struct b44 *bp, int budget)
 			skb_size = b44_alloc_rx_skb(bp, cons, bp->rx_prod);
 			if (skb_size < 0)
 				goto drop_it;
-			dma_unmap_single(bp->sdev->dma_dev, map,
-					 skb_size, DMA_FROM_DEVICE);
+			ssb_dma_unmap_single(bp->sdev, map,
+					     skb_size, DMA_FROM_DEVICE);
 			/* Leave out rx_header */
                 	skb_put(skb, len + RX_PKT_OFFSET);
             	        skb_pull(skb, RX_PKT_OFFSET);
@@ -966,25 +966,25 @@ static int b44_start_xmit(struct sk_buff *skb, struct net_device *dev)
 		goto err_out;
 	}
 
-	mapping = dma_map_single(bp->sdev->dma_dev, skb->data, len, DMA_TO_DEVICE);
-	if (dma_mapping_error(mapping) || mapping + len > DMA_30BIT_MASK) {
+	mapping = ssb_dma_map_single(bp->sdev, skb->data, len, DMA_TO_DEVICE);
+	if (ssb_dma_mapping_error(bp->sdev, mapping) || mapping + len > DMA_30BIT_MASK) {
 		struct sk_buff *bounce_skb;
 
 		/* Chip can't handle DMA to/from >1GB, use bounce buffer */
-		if (!dma_mapping_error(mapping))
-			dma_unmap_single(bp->sdev->dma_dev, mapping, len,
-					DMA_TO_DEVICE);
+		if (!ssb_dma_mapping_error(bp->sdev, mapping))
+			ssb_dma_unmap_single(bp->sdev, mapping, len,
+					     DMA_TO_DEVICE);
 
 		bounce_skb = __dev_alloc_skb(len, GFP_ATOMIC | GFP_DMA);
 		if (!bounce_skb)
 			goto err_out;
 
-		mapping = dma_map_single(bp->sdev->dma_dev, bounce_skb->data,
-					 len, DMA_TO_DEVICE);
-		if (dma_mapping_error(mapping) || mapping + len > DMA_30BIT_MASK) {
-			if (!dma_mapping_error(mapping))
-				dma_unmap_single(bp->sdev->dma_dev, mapping,
-					 len, DMA_TO_DEVICE);
+		mapping = ssb_dma_map_single(bp->sdev, bounce_skb->data,
+					     len, DMA_TO_DEVICE);
+		if (ssb_dma_mapping_error(bp->sdev, mapping) || mapping + len > DMA_30BIT_MASK) {
+			if (!ssb_dma_mapping_error(bp->sdev, mapping))
+				ssb_dma_unmap_single(bp->sdev, mapping,
+						     len, DMA_TO_DEVICE);
 			dev_kfree_skb_any(bounce_skb);
 			goto err_out;
 		}
@@ -1082,8 +1082,8 @@ static void b44_free_rings(struct b44 *bp)
 
 		if (rp->skb == NULL)
 			continue;
-		dma_unmap_single(bp->sdev->dma_dev, rp->mapping, RX_PKT_BUF_SZ,
-					DMA_FROM_DEVICE);
+		ssb_dma_unmap_single(bp->sdev, rp->mapping, RX_PKT_BUF_SZ,
+				     DMA_FROM_DEVICE);
 		dev_kfree_skb_any(rp->skb);
 		rp->skb = NULL;
 	}
@@ -1094,8 +1094,8 @@ static void b44_free_rings(struct b44 *bp)
 
 		if (rp->skb == NULL)
 			continue;
-		dma_unmap_single(bp->sdev->dma_dev, rp->mapping, rp->skb->len,
-					DMA_TO_DEVICE);
+		ssb_dma_unmap_single(bp->sdev, rp->mapping, rp->skb->len,
+				     DMA_TO_DEVICE);
 		dev_kfree_skb_any(rp->skb);
 		rp->skb = NULL;
 	}
@@ -1117,14 +1117,14 @@ static void b44_init_rings(struct b44 *bp)
 	memset(bp->tx_ring, 0, B44_TX_RING_BYTES);
 
 	if (bp->flags & B44_FLAG_RX_RING_HACK)
-		dma_sync_single_for_device(bp->sdev->dma_dev, bp->rx_ring_dma,
-			                  DMA_TABLE_BYTES,
-			                  DMA_BIDIRECTIONAL);
+		ssb_dma_sync_single_for_device(bp->sdev, bp->rx_ring_dma,
+					       DMA_TABLE_BYTES,
+					       DMA_BIDIRECTIONAL);
 
 	if (bp->flags & B44_FLAG_TX_RING_HACK)
-		dma_sync_single_for_device(bp->sdev->dma_dev, bp->tx_ring_dma,
-			                  DMA_TABLE_BYTES,
-			                  DMA_TO_DEVICE);
+		ssb_dma_sync_single_for_device(bp->sdev, bp->tx_ring_dma,
+					       DMA_TABLE_BYTES,
+					       DMA_TO_DEVICE);
 
 	for (i = 0; i < bp->rx_pending; i++) {
 		if (b44_alloc_rx_skb(bp, -1, i) < 0)
@@ -1144,25 +1144,27 @@ static void b44_free_consistent(struct b44 *bp)
 	bp->tx_buffers = NULL;
 	if (bp->rx_ring) {
 		if (bp->flags & B44_FLAG_RX_RING_HACK) {
-			dma_unmap_single(bp->sdev->dma_dev, bp->rx_ring_dma,
-					DMA_TABLE_BYTES,
-					DMA_BIDIRECTIONAL);
+			ssb_dma_unmap_single(bp->sdev, bp->rx_ring_dma,
+					     DMA_TABLE_BYTES,
+					     DMA_BIDIRECTIONAL);
 			kfree(bp->rx_ring);
 		} else
-			dma_free_coherent(bp->sdev->dma_dev, DMA_TABLE_BYTES,
-					    bp->rx_ring, bp->rx_ring_dma);
+			ssb_dma_free_consistent(bp->sdev, DMA_TABLE_BYTES,
+						bp->rx_ring, bp->rx_ring_dma,
+						GFP_KERNEL);
 		bp->rx_ring = NULL;
 		bp->flags &= ~B44_FLAG_RX_RING_HACK;
 	}
 	if (bp->tx_ring) {
 		if (bp->flags & B44_FLAG_TX_RING_HACK) {
-			dma_unmap_single(bp->sdev->dma_dev, bp->tx_ring_dma,
-					DMA_TABLE_BYTES,
-					DMA_TO_DEVICE);
+			ssb_dma_unmap_single(bp->sdev, bp->tx_ring_dma,
+					     DMA_TABLE_BYTES,
+					     DMA_TO_DEVICE);
 			kfree(bp->tx_ring);
 		} else
-			dma_free_coherent(bp->sdev->dma_dev, DMA_TABLE_BYTES,
-					    bp->tx_ring, bp->tx_ring_dma);
+			ssb_dma_free_consistent(bp->sdev, DMA_TABLE_BYTES,
+						bp->tx_ring, bp->tx_ring_dma,
+						GFP_KERNEL);
 		bp->tx_ring = NULL;
 		bp->flags &= ~B44_FLAG_TX_RING_HACK;
 	}
@@ -1187,7 +1189,7 @@ static int b44_alloc_consistent(struct b44 *bp, gfp_t gfp)
 		goto out_err;
 
 	size = DMA_TABLE_BYTES;
-	bp->rx_ring = dma_alloc_coherent(bp->sdev->dma_dev, size, &bp->rx_ring_dma, gfp);
+	bp->rx_ring = ssb_dma_alloc_consistent(bp->sdev, size, &bp->rx_ring_dma, gfp);
 	if (!bp->rx_ring) {
 		/* Allocation may have failed due to pci_alloc_consistent
 		   insisting on use of GFP_DMA, which is more restrictive
@@ -1199,11 +1201,11 @@ static int b44_alloc_consistent(struct b44 *bp, gfp_t gfp)
 		if (!rx_ring)
 			goto out_err;
 
-		rx_ring_dma = dma_map_single(bp->sdev->dma_dev, rx_ring,
-			                    DMA_TABLE_BYTES,
-			                    DMA_BIDIRECTIONAL);
+		rx_ring_dma = ssb_dma_map_single(bp->sdev, rx_ring,
+						 DMA_TABLE_BYTES,
+						 DMA_BIDIRECTIONAL);
 
-		if (dma_mapping_error(rx_ring_dma) ||
+		if (ssb_dma_mapping_error(bp->sdev, rx_ring_dma) ||
 			rx_ring_dma + size > DMA_30BIT_MASK) {
 			kfree(rx_ring);
 			goto out_err;
@@ -1214,9 +1216,9 @@ static int b44_alloc_consistent(struct b44 *bp, gfp_t gfp)
 		bp->flags |= B44_FLAG_RX_RING_HACK;
 	}
 
-	bp->tx_ring = dma_alloc_coherent(bp->sdev->dma_dev, size, &bp->tx_ring_dma, gfp);
+	bp->tx_ring = ssb_dma_alloc_consistent(bp->sdev, size, &bp->tx_ring_dma, gfp);
 	if (!bp->tx_ring) {
-		/* Allocation may have failed due to dma_alloc_coherent
+		/* Allocation may have failed due to ssb_dma_alloc_consistent
 		   insisting on use of GFP_DMA, which is more restrictive
 		   than necessary...  */
 		struct dma_desc *tx_ring;
@@ -1226,11 +1228,11 @@ static int b44_alloc_consistent(struct b44 *bp, gfp_t gfp)
 		if (!tx_ring)
 			goto out_err;
 
-		tx_ring_dma = dma_map_single(bp->sdev->dma_dev, tx_ring,
+		tx_ring_dma = ssb_dma_map_single(bp->sdev, tx_ring,
 			                    DMA_TABLE_BYTES,
 			                    DMA_TO_DEVICE);
 
-		if (dma_mapping_error(tx_ring_dma) ||
+		if (ssb_dma_mapping_error(bp->sdev, tx_ring_dma) ||
 			tx_ring_dma + size > DMA_30BIT_MASK) {
 			kfree(tx_ring);
 			goto out_err;
diff --git a/drivers/net/wireless/b43/dma.c b/drivers/net/wireless/b43/dma.c
index 8a09a1db08db..098f886976f6 100644
--- a/drivers/net/wireless/b43/dma.c
+++ b/drivers/net/wireless/b43/dma.c
@@ -328,11 +328,11 @@ static inline
 	dma_addr_t dmaaddr;
 
 	if (tx) {
-		dmaaddr = dma_map_single(ring->dev->dev->dma_dev,
-					 buf, len, DMA_TO_DEVICE);
+		dmaaddr = ssb_dma_map_single(ring->dev->dev,
+					     buf, len, DMA_TO_DEVICE);
 	} else {
-		dmaaddr = dma_map_single(ring->dev->dev->dma_dev,
-					 buf, len, DMA_FROM_DEVICE);
+		dmaaddr = ssb_dma_map_single(ring->dev->dev,
+					     buf, len, DMA_FROM_DEVICE);
 	}
 
 	return dmaaddr;
@@ -343,11 +343,11 @@ static inline
 			  dma_addr_t addr, size_t len, int tx)
 {
 	if (tx) {
-		dma_unmap_single(ring->dev->dev->dma_dev,
-				 addr, len, DMA_TO_DEVICE);
+		ssb_dma_unmap_single(ring->dev->dev,
+				     addr, len, DMA_TO_DEVICE);
 	} else {
-		dma_unmap_single(ring->dev->dev->dma_dev,
-				 addr, len, DMA_FROM_DEVICE);
+		ssb_dma_unmap_single(ring->dev->dev,
+				     addr, len, DMA_FROM_DEVICE);
 	}
 }
 
@@ -356,8 +356,8 @@ static inline
 				 dma_addr_t addr, size_t len)
 {
 	B43_WARN_ON(ring->tx);
-	dma_sync_single_for_cpu(ring->dev->dev->dma_dev,
-				addr, len, DMA_FROM_DEVICE);
+	ssb_dma_sync_single_for_cpu(ring->dev->dev,
+				    addr, len, DMA_FROM_DEVICE);
 }
 
 static inline
@@ -365,8 +365,8 @@ static inline
 				    dma_addr_t addr, size_t len)
 {
 	B43_WARN_ON(ring->tx);
-	dma_sync_single_for_device(ring->dev->dev->dma_dev,
-				   addr, len, DMA_FROM_DEVICE);
+	ssb_dma_sync_single_for_device(ring->dev->dev,
+				       addr, len, DMA_FROM_DEVICE);
 }
 
 static inline
@@ -381,7 +381,6 @@ static inline
 
 static int alloc_ringmemory(struct b43_dmaring *ring)
 {
-	struct device *dma_dev = ring->dev->dev->dma_dev;
 	gfp_t flags = GFP_KERNEL;
 
 	/* The specs call for 4K buffers for 30- and 32-bit DMA with 4K
@@ -392,11 +391,14 @@ static int alloc_ringmemory(struct b43_dmaring *ring)
 	 * For unknown reasons - possibly a hardware error - the BCM4311 rev
 	 * 02, which uses 64-bit DMA, needs the ring buffer in very low memory,
 	 * which accounts for the GFP_DMA flag below.
+	 *
+	 * The flags here must match the flags in free_ringmemory below!
 	 */
 	if (ring->type == B43_DMA_64BIT)
 		flags |= GFP_DMA;
-	ring->descbase = dma_alloc_coherent(dma_dev, B43_DMA_RINGMEMSIZE,
-					    &(ring->dmabase), flags);
+	ring->descbase = ssb_dma_alloc_consistent(ring->dev->dev,
+						  B43_DMA_RINGMEMSIZE,
+						  &(ring->dmabase), flags);
 	if (!ring->descbase) {
 		b43err(ring->dev->wl, "DMA ringmemory allocation failed\n");
 		return -ENOMEM;
@@ -408,10 +410,13 @@ static int alloc_ringmemory(struct b43_dmaring *ring)
 
 static void free_ringmemory(struct b43_dmaring *ring)
 {
-	struct device *dma_dev = ring->dev->dev->dma_dev;
+	gfp_t flags = GFP_KERNEL;
+
+	if (ring->type == B43_DMA_64BIT)
+		flags |= GFP_DMA;
 
-	dma_free_coherent(dma_dev, B43_DMA_RINGMEMSIZE,
-			  ring->descbase, ring->dmabase);
+	ssb_dma_free_consistent(ring->dev->dev, B43_DMA_RINGMEMSIZE,
+				ring->descbase, ring->dmabase, flags);
 }
 
 /* Reset the RX DMA channel */
@@ -518,7 +523,7 @@ static bool b43_dma_mapping_error(struct b43_dmaring *ring,
 				  dma_addr_t addr,
 				  size_t buffersize, bool dma_to_device)
 {
-	if (unlikely(dma_mapping_error(addr)))
+	if (unlikely(ssb_dma_mapping_error(ring->dev->dev, addr)))
 		return 1;
 
 	switch (ring->type) {
@@ -844,10 +849,10 @@ struct b43_dmaring *b43_setup_dmaring(struct b43_wldev *dev,
 			goto err_kfree_meta;
 
 		/* test for ability to dma to txhdr_cache */
-		dma_test = dma_map_single(dev->dev->dma_dev,
-					  ring->txhdr_cache,
-					  b43_txhdr_size(dev),
-					  DMA_TO_DEVICE);
+		dma_test = ssb_dma_map_single(dev->dev,
+					      ring->txhdr_cache,
+					      b43_txhdr_size(dev),
+					      DMA_TO_DEVICE);
 
 		if (b43_dma_mapping_error(ring, dma_test,
 					  b43_txhdr_size(dev), 1)) {
@@ -859,10 +864,10 @@ struct b43_dmaring *b43_setup_dmaring(struct b43_wldev *dev,
 			if (!ring->txhdr_cache)
 				goto err_kfree_meta;
 
-			dma_test = dma_map_single(dev->dev->dma_dev,
-						  ring->txhdr_cache,
-						  b43_txhdr_size(dev),
-						  DMA_TO_DEVICE);
+			dma_test = ssb_dma_map_single(dev->dev,
+						      ring->txhdr_cache,
+						      b43_txhdr_size(dev),
+						      DMA_TO_DEVICE);
 
 			if (b43_dma_mapping_error(ring, dma_test,
 						  b43_txhdr_size(dev), 1)) {
@@ -873,9 +878,9 @@ struct b43_dmaring *b43_setup_dmaring(struct b43_wldev *dev,
 			}
 		}
 
-		dma_unmap_single(dev->dev->dma_dev,
-				 dma_test, b43_txhdr_size(dev),
-				 DMA_TO_DEVICE);
+		ssb_dma_unmap_single(dev->dev,
+				     dma_test, b43_txhdr_size(dev),
+				     DMA_TO_DEVICE);
 	}
 
 	err = alloc_ringmemory(ring);
diff --git a/drivers/net/wireless/b43legacy/dma.c b/drivers/net/wireless/b43legacy/dma.c
index 33cc256c5baf..9736b2f56a75 100644
--- a/drivers/net/wireless/b43legacy/dma.c
+++ b/drivers/net/wireless/b43legacy/dma.c
@@ -393,13 +393,13 @@ dma_addr_t map_descbuffer(struct b43legacy_dmaring *ring,
 	dma_addr_t dmaaddr;
 
 	if (tx)
-		dmaaddr = dma_map_single(ring->dev->dev->dma_dev,
-					 buf, len,
-					 DMA_TO_DEVICE);
+		dmaaddr = ssb_dma_map_single(ring->dev->dev,
+					     buf, len,
+					     DMA_TO_DEVICE);
 	else
-		dmaaddr = dma_map_single(ring->dev->dev->dma_dev,
-					 buf, len,
-					 DMA_FROM_DEVICE);
+		dmaaddr = ssb_dma_map_single(ring->dev->dev,
+					     buf, len,
+					     DMA_FROM_DEVICE);
 
 	return dmaaddr;
 }
@@ -411,13 +411,13 @@ void unmap_descbuffer(struct b43legacy_dmaring *ring,
 		      int tx)
 {
 	if (tx)
-		dma_unmap_single(ring->dev->dev->dma_dev,
-				 addr, len,
-				 DMA_TO_DEVICE);
+		ssb_dma_unmap_single(ring->dev->dev,
+				     addr, len,
+				     DMA_TO_DEVICE);
 	else
-		dma_unmap_single(ring->dev->dev->dma_dev,
-				 addr, len,
-				 DMA_FROM_DEVICE);
+		ssb_dma_unmap_single(ring->dev->dev,
+				     addr, len,
+				     DMA_FROM_DEVICE);
 }
 
 static inline
@@ -427,8 +427,8 @@ void sync_descbuffer_for_cpu(struct b43legacy_dmaring *ring,
 {
 	B43legacy_WARN_ON(ring->tx);
 
-	dma_sync_single_for_cpu(ring->dev->dev->dma_dev,
-				addr, len, DMA_FROM_DEVICE);
+	ssb_dma_sync_single_for_cpu(ring->dev->dev,
+				    addr, len, DMA_FROM_DEVICE);
 }
 
 static inline
@@ -438,8 +438,8 @@ void sync_descbuffer_for_device(struct b43legacy_dmaring *ring,
 {
 	B43legacy_WARN_ON(ring->tx);
 
-	dma_sync_single_for_device(ring->dev->dev->dma_dev,
-				   addr, len, DMA_FROM_DEVICE);
+	ssb_dma_sync_single_for_device(ring->dev->dev,
+				       addr, len, DMA_FROM_DEVICE);
 }
 
 static inline
@@ -458,10 +458,11 @@ void free_descriptor_buffer(struct b43legacy_dmaring *ring,
 
 static int alloc_ringmemory(struct b43legacy_dmaring *ring)
 {
-	struct device *dma_dev = ring->dev->dev->dma_dev;
-
-	ring->descbase = dma_alloc_coherent(dma_dev, B43legacy_DMA_RINGMEMSIZE,
-					    &(ring->dmabase), GFP_KERNEL);
+	/* GFP flags must match the flags in free_ringmemory()! */
+	ring->descbase = ssb_dma_alloc_consistent(ring->dev->dev,
+						  B43legacy_DMA_RINGMEMSIZE,
+						  &(ring->dmabase),
+						  GFP_KERNEL);
 	if (!ring->descbase) {
 		b43legacyerr(ring->dev->wl, "DMA ringmemory allocation"
 			     " failed\n");
@@ -474,10 +475,8 @@ static int alloc_ringmemory(struct b43legacy_dmaring *ring)
 
 static void free_ringmemory(struct b43legacy_dmaring *ring)
 {
-	struct device *dma_dev = ring->dev->dev->dma_dev;
-
-	dma_free_coherent(dma_dev, B43legacy_DMA_RINGMEMSIZE,
-			  ring->descbase, ring->dmabase);
+	ssb_dma_free_consistent(ring->dev->dev, B43legacy_DMA_RINGMEMSIZE,
+				ring->descbase, ring->dmabase, GFP_KERNEL);
 }
 
 /* Reset the RX DMA channel */
@@ -589,7 +588,7 @@ static bool b43legacy_dma_mapping_error(struct b43legacy_dmaring *ring,
 					 size_t buffersize,
 					 bool dma_to_device)
 {
-	if (unlikely(dma_mapping_error(addr)))
+	if (unlikely(ssb_dma_mapping_error(ring->dev->dev, addr)))
 		return 1;
 
 	switch (ring->type) {
@@ -893,9 +892,9 @@ struct b43legacy_dmaring *b43legacy_setup_dmaring(struct b43legacy_wldev *dev,
 			goto err_kfree_meta;
 
 		/* test for ability to dma to txhdr_cache */
-		dma_test = dma_map_single(dev->dev->dma_dev, ring->txhdr_cache,
-					  sizeof(struct b43legacy_txhdr_fw3),
-					  DMA_TO_DEVICE);
+		dma_test = ssb_dma_map_single(dev->dev, ring->txhdr_cache,
+					      sizeof(struct b43legacy_txhdr_fw3),
+					      DMA_TO_DEVICE);
 
 		if (b43legacy_dma_mapping_error(ring, dma_test,
 					sizeof(struct b43legacy_txhdr_fw3), 1)) {
@@ -907,7 +906,7 @@ struct b43legacy_dmaring *b43legacy_setup_dmaring(struct b43legacy_wldev *dev,
 			if (!ring->txhdr_cache)
 				goto err_kfree_meta;
 
-			dma_test = dma_map_single(dev->dev->dma_dev,
+				dma_test = ssb_dma_map_single(dev->dev,
 					ring->txhdr_cache,
 					sizeof(struct b43legacy_txhdr_fw3),
 					DMA_TO_DEVICE);
@@ -917,9 +916,9 @@ struct b43legacy_dmaring *b43legacy_setup_dmaring(struct b43legacy_wldev *dev,
 				goto err_kfree_txhdr_cache;
 		}
 
-		dma_unmap_single(dev->dev->dma_dev,
-				 dma_test, sizeof(struct b43legacy_txhdr_fw3),
-				 DMA_TO_DEVICE);
+		ssb_dma_unmap_single(dev->dev, dma_test,
+				     sizeof(struct b43legacy_txhdr_fw3),
+				     DMA_TO_DEVICE);
 	}
 
 	ring->dev = dev;
diff --git a/drivers/ssb/Kconfig b/drivers/ssb/Kconfig
index cd845b8acd17..307b1f62d949 100644
--- a/drivers/ssb/Kconfig
+++ b/drivers/ssb/Kconfig
@@ -2,7 +2,7 @@ menu "Sonics Silicon Backplane"
 
 config SSB_POSSIBLE
 	bool
-	depends on HAS_IOMEM
+	depends on HAS_IOMEM && HAS_DMA
 	default y
 
 config SSB
diff --git a/drivers/ssb/main.c b/drivers/ssb/main.c
index d184f2aea78d..d831a2beff39 100644
--- a/drivers/ssb/main.c
+++ b/drivers/ssb/main.c
@@ -462,18 +462,15 @@ static int ssb_devices_register(struct ssb_bus *bus)
 #ifdef CONFIG_SSB_PCIHOST
 			sdev->irq = bus->host_pci->irq;
 			dev->parent = &bus->host_pci->dev;
-			sdev->dma_dev = &bus->host_pci->dev;
 #endif
 			break;
 		case SSB_BUSTYPE_PCMCIA:
 #ifdef CONFIG_SSB_PCMCIAHOST
 			sdev->irq = bus->host_pcmcia->irq.AssignedIRQ;
 			dev->parent = &bus->host_pcmcia->dev;
-			sdev->dma_dev = &bus->host_pcmcia->dev;
 #endif
 			break;
 		case SSB_BUSTYPE_SSB:
-			sdev->dma_dev = dev;
 			break;
 		}
 
@@ -1156,36 +1153,82 @@ u32 ssb_dma_translation(struct ssb_device *dev)
 {
 	switch (dev->bus->bustype) {
 	case SSB_BUSTYPE_SSB:
-	case SSB_BUSTYPE_PCMCIA:
 		return 0;
 	case SSB_BUSTYPE_PCI:
 		return SSB_PCI_DMA;
+	default:
+		__ssb_dma_not_implemented(dev);
 	}
 	return 0;
 }
 EXPORT_SYMBOL(ssb_dma_translation);
 
-int ssb_dma_set_mask(struct ssb_device *ssb_dev, u64 mask)
+int ssb_dma_set_mask(struct ssb_device *dev, u64 mask)
 {
-	struct device *dma_dev = ssb_dev->dma_dev;
-	int err = 0;
+	int err;
 
-#ifdef CONFIG_SSB_PCIHOST
-	if (ssb_dev->bus->bustype == SSB_BUSTYPE_PCI) {
-		err = pci_set_dma_mask(ssb_dev->bus->host_pci, mask);
+	switch (dev->bus->bustype) {
+	case SSB_BUSTYPE_PCI:
+		err = pci_set_dma_mask(dev->bus->host_pci, mask);
 		if (err)
 			return err;
-		err = pci_set_consistent_dma_mask(ssb_dev->bus->host_pci, mask);
+		err = pci_set_consistent_dma_mask(dev->bus->host_pci, mask);
 		return err;
+	case SSB_BUSTYPE_SSB:
+		return dma_set_mask(dev->dev, mask);
+	default:
+		__ssb_dma_not_implemented(dev);
 	}
-#endif
-	dma_dev->coherent_dma_mask = mask;
-	dma_dev->dma_mask = &dma_dev->coherent_dma_mask;
-
-	return err;
+	return -ENOSYS;
 }
 EXPORT_SYMBOL(ssb_dma_set_mask);
 
+void * ssb_dma_alloc_consistent(struct ssb_device *dev, size_t size,
+				dma_addr_t *dma_handle, gfp_t gfp_flags)
+{
+	switch (dev->bus->bustype) {
+	case SSB_BUSTYPE_PCI:
+		if (gfp_flags & GFP_DMA) {
+			/* Workaround: The PCI API does not support passing
+			 * a GFP flag. */
+			return dma_alloc_coherent(&dev->bus->host_pci->dev,
+						  size, dma_handle, gfp_flags);
+		}
+		return pci_alloc_consistent(dev->bus->host_pci, size, dma_handle);
+	case SSB_BUSTYPE_SSB:
+		return dma_alloc_coherent(dev->dev, size, dma_handle, gfp_flags);
+	default:
+		__ssb_dma_not_implemented(dev);
+	}
+	return NULL;
+}
+EXPORT_SYMBOL(ssb_dma_alloc_consistent);
+
+void ssb_dma_free_consistent(struct ssb_device *dev, size_t size,
+			     void *vaddr, dma_addr_t dma_handle,
+			     gfp_t gfp_flags)
+{
+	switch (dev->bus->bustype) {
+	case SSB_BUSTYPE_PCI:
+		if (gfp_flags & GFP_DMA) {
+			/* Workaround: The PCI API does not support passing
+			 * a GFP flag. */
+			dma_free_coherent(&dev->bus->host_pci->dev,
+					  size, vaddr, dma_handle);
+			return;
+		}
+		pci_free_consistent(dev->bus->host_pci, size,
+				    vaddr, dma_handle);
+		return;
+	case SSB_BUSTYPE_SSB:
+		dma_free_coherent(dev->dev, size, vaddr, dma_handle);
+		return;
+	default:
+		__ssb_dma_not_implemented(dev);
+	}
+}
+EXPORT_SYMBOL(ssb_dma_free_consistent);
+
 int ssb_bus_may_powerdown(struct ssb_bus *bus)
 {
 	struct ssb_chipcommon *cc;
diff --git a/include/linux/ssb/ssb.h b/include/linux/ssb/ssb.h
index 50dfd0dc4093..0fe5a0ded3ea 100644
--- a/include/linux/ssb/ssb.h
+++ b/include/linux/ssb/ssb.h
@@ -137,9 +137,6 @@ struct ssb_device {
 	const struct ssb_bus_ops *ops;
 
 	struct device *dev;
-	/* Pointer to the device that has to be used for
-	 * any DMA related operation. */
-	struct device *dma_dev;
 
 	struct ssb_bus *bus;
 	struct ssb_device_id id;
@@ -399,13 +396,151 @@ static inline void ssb_block_write(struct ssb_device *dev, const void *buffer,
 #endif /* CONFIG_SSB_BLOCKIO */
 
 
+/* The SSB DMA API. Use this API for any DMA operation on the device.
+ * This API basically is a wrapper that calls the correct DMA API for
+ * the host device type the SSB device is attached to. */
+
 /* Translation (routing) bits that need to be ORed to DMA
  * addresses before they are given to a device. */
 extern u32 ssb_dma_translation(struct ssb_device *dev);
 #define SSB_DMA_TRANSLATION_MASK	0xC0000000
 #define SSB_DMA_TRANSLATION_SHIFT	30
 
-extern int ssb_dma_set_mask(struct ssb_device *ssb_dev, u64 mask);
+extern int ssb_dma_set_mask(struct ssb_device *dev, u64 mask);
+
+extern void * ssb_dma_alloc_consistent(struct ssb_device *dev, size_t size,
+				       dma_addr_t *dma_handle, gfp_t gfp_flags);
+extern void ssb_dma_free_consistent(struct ssb_device *dev, size_t size,
+				    void *vaddr, dma_addr_t dma_handle,
+				    gfp_t gfp_flags);
+
+static inline void __cold __ssb_dma_not_implemented(struct ssb_device *dev)
+{
+#ifdef CONFIG_SSB_DEBUG
+	printk(KERN_ERR "SSB: BUG! Calling DMA API for "
+	       "unsupported bustype %d\n", dev->bus->bustype);
+#endif /* DEBUG */
+}
+
+static inline int ssb_dma_mapping_error(struct ssb_device *dev, dma_addr_t addr)
+{
+	switch (dev->bus->bustype) {
+	case SSB_BUSTYPE_PCI:
+		return pci_dma_mapping_error(addr);
+	case SSB_BUSTYPE_SSB:
+		return dma_mapping_error(addr);
+	default:
+		__ssb_dma_not_implemented(dev);
+	}
+	return -ENOSYS;
+}
+
+static inline dma_addr_t ssb_dma_map_single(struct ssb_device *dev, void *p,
+					    size_t size, enum dma_data_direction dir)
+{
+	switch (dev->bus->bustype) {
+	case SSB_BUSTYPE_PCI:
+		return pci_map_single(dev->bus->host_pci, p, size, dir);
+	case SSB_BUSTYPE_SSB:
+		return dma_map_single(dev->dev, p, size, dir);
+	default:
+		__ssb_dma_not_implemented(dev);
+	}
+	return 0;
+}
+
+static inline void ssb_dma_unmap_single(struct ssb_device *dev, dma_addr_t dma_addr,
+					size_t size, enum dma_data_direction dir)
+{
+	switch (dev->bus->bustype) {
+	case SSB_BUSTYPE_PCI:
+		pci_unmap_single(dev->bus->host_pci, dma_addr, size, dir);
+		return;
+	case SSB_BUSTYPE_SSB:
+		dma_unmap_single(dev->dev, dma_addr, size, dir);
+		return;
+	default:
+		__ssb_dma_not_implemented(dev);
+	}
+}
+
+static inline void ssb_dma_sync_single_for_cpu(struct ssb_device *dev,
+					       dma_addr_t dma_addr,
+					       size_t size,
+					       enum dma_data_direction dir)
+{
+	switch (dev->bus->bustype) {
+	case SSB_BUSTYPE_PCI:
+		pci_dma_sync_single_for_cpu(dev->bus->host_pci, dma_addr,
+					    size, dir);
+		return;
+	case SSB_BUSTYPE_SSB:
+		dma_sync_single_for_cpu(dev->dev, dma_addr, size, dir);
+		return;
+	default:
+		__ssb_dma_not_implemented(dev);
+	}
+}
+
+static inline void ssb_dma_sync_single_for_device(struct ssb_device *dev,
+						  dma_addr_t dma_addr,
+						  size_t size,
+						  enum dma_data_direction dir)
+{
+	switch (dev->bus->bustype) {
+	case SSB_BUSTYPE_PCI:
+		pci_dma_sync_single_for_device(dev->bus->host_pci, dma_addr,
+					       size, dir);
+		return;
+	case SSB_BUSTYPE_SSB:
+		dma_sync_single_for_device(dev->dev, dma_addr, size, dir);
+		return;
+	default:
+		__ssb_dma_not_implemented(dev);
+	}
+}
+
+static inline void ssb_dma_sync_single_range_for_cpu(struct ssb_device *dev,
+						     dma_addr_t dma_addr,
+						     unsigned long offset,
+						     size_t size,
+						     enum dma_data_direction dir)
+{
+	switch (dev->bus->bustype) {
+	case SSB_BUSTYPE_PCI:
+		/* Just sync everything. That's all the PCI API can do. */
+		pci_dma_sync_single_for_cpu(dev->bus->host_pci, dma_addr,
+					    offset + size, dir);
+		return;
+	case SSB_BUSTYPE_SSB:
+		dma_sync_single_range_for_cpu(dev->dev, dma_addr, offset,
+					      size, dir);
+		return;
+	default:
+		__ssb_dma_not_implemented(dev);
+	}
+}
+
+static inline void ssb_dma_sync_single_range_for_device(struct ssb_device *dev,
+							dma_addr_t dma_addr,
+							unsigned long offset,
+							size_t size,
+							enum dma_data_direction dir)
+{
+	switch (dev->bus->bustype) {
+	case SSB_BUSTYPE_PCI:
+		/* Just sync everything. That's all the PCI API can do. */
+		pci_dma_sync_single_for_device(dev->bus->host_pci, dma_addr,
+					       offset + size, dir);
+		return;
+	case SSB_BUSTYPE_SSB:
+		dma_sync_single_range_for_device(dev->dev, dma_addr, offset,
+						 size, dir);
+		return;
+	default:
+		__ssb_dma_not_implemented(dev);
+	}
+}
 
 
 #ifdef CONFIG_SSB_PCIHOST
-- 
cgit v1.2.3


From ffd7891dc909b3648e87f7cf8f84a6dc12fc1cc6 Mon Sep 17 00:00:00 2001
From: "Luis R. Rodriguez" <lrodriguez@atheros.com>
Date: Sat, 21 Jun 2008 10:02:46 -0400
Subject: mac80211: Let drivers have access to TKIP key offets for TX and RX
 MIC

Some drivers may want to to use the TKIP key offsets for TX and RX
MIC so lets move this out. Lets also clear up a bit how this is used
internally in mac80211.

Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/linux/nl80211.h |  5 ++++-
 include/net/mac80211.h  |  7 ++++++-
 net/mac80211/key.h      | 37 ++++++++++++-------------------------
 net/mac80211/tkip.c     | 10 +++++-----
 net/mac80211/wpa.c      | 20 ++++++++++++++------
 5 files changed, 41 insertions(+), 38 deletions(-)

(limited to 'include')

diff --git a/include/linux/nl80211.h b/include/linux/nl80211.h
index aa8411e2a160..2be7c63bc0f2 100644
--- a/include/linux/nl80211.h
+++ b/include/linux/nl80211.h
@@ -241,7 +241,10 @@ enum nl80211_attrs {
 	NL80211_ATTR_MAX = __NL80211_ATTR_AFTER_LAST - 1
 };
 
-#define NL80211_MAX_SUPP_RATES	32
+#define NL80211_MAX_SUPP_RATES			32
+#define NL80211_TKIP_DATA_OFFSET_ENCR_KEY	0
+#define NL80211_TKIP_DATA_OFFSET_TX_MIC_KEY	16
+#define NL80211_TKIP_DATA_OFFSET_RX_MIC_KEY	24
 
 /**
  * enum nl80211_iftype - (virtual) interface types
diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index 7ab4ff6159a2..19f1e412a0f0 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -595,7 +595,12 @@ enum ieee80211_key_flags {
  * @flags: key flags, see &enum ieee80211_key_flags.
  * @keyidx: the key index (0-3)
  * @keylen: key material length
- * @key: key material
+ * @key: key material. For ALG_TKIP the key is encoded as a 256-bit (32 byte)
+ * 	data block:
+ * 	- Temporal Encryption Key (128 bits)
+ * 	- Temporal Authenticator Tx MIC Key (64 bits)
+ * 	- Temporal Authenticator Rx MIC Key (64 bits)
+ *
  */
 struct ieee80211_key_conf {
 	enum ieee80211_key_alg alg;
diff --git a/net/mac80211/key.h b/net/mac80211/key.h
index a0f774aafa45..425816e0996c 100644
--- a/net/mac80211/key.h
+++ b/net/mac80211/key.h
@@ -16,31 +16,18 @@
 #include <linux/rcupdate.h>
 #include <net/mac80211.h>
 
-/* ALG_TKIP
- * struct ieee80211_key::key is encoded as a 256-bit (32 byte) data block:
- * Temporal Encryption Key (128 bits)
- * Temporal Authenticator Tx MIC Key (64 bits)
- * Temporal Authenticator Rx MIC Key (64 bits)
- */
-
-#define WEP_IV_LEN 4
-#define WEP_ICV_LEN 4
-
-#define ALG_TKIP_KEY_LEN 32
-/* Starting offsets for each key */
-#define ALG_TKIP_TEMP_ENCR_KEY 0
-#define ALG_TKIP_TEMP_AUTH_TX_MIC_KEY 16
-#define ALG_TKIP_TEMP_AUTH_RX_MIC_KEY 24
-#define TKIP_IV_LEN 8
-#define TKIP_ICV_LEN 4
-
-#define ALG_CCMP_KEY_LEN 16
-#define CCMP_HDR_LEN 8
-#define CCMP_MIC_LEN 8
-#define CCMP_TK_LEN 16
-#define CCMP_PN_LEN 6
-
-#define NUM_RX_DATA_QUEUES 17
+#define WEP_IV_LEN		4
+#define WEP_ICV_LEN		4
+#define ALG_TKIP_KEY_LEN	32
+#define ALG_CCMP_KEY_LEN	16
+#define CCMP_HDR_LEN		8
+#define CCMP_MIC_LEN		8
+#define CCMP_TK_LEN		16
+#define CCMP_PN_LEN		6
+#define TKIP_IV_LEN		8
+#define TKIP_ICV_LEN		4
+
+#define NUM_RX_DATA_QUEUES	17
 
 struct ieee80211_local;
 struct ieee80211_sub_if_data;
diff --git a/net/mac80211/tkip.c b/net/mac80211/tkip.c
index 699807889988..995f7af3d25e 100644
--- a/net/mac80211/tkip.c
+++ b/net/mac80211/tkip.c
@@ -164,7 +164,7 @@ void ieee80211_get_tkip_key(struct ieee80211_key_conf *keyconf,
 	iv16 = data[2] | (data[0] << 8);
 	iv32 = get_unaligned_le32(&data[4]);
 
-	tk = &key->conf.key[ALG_TKIP_TEMP_ENCR_KEY];
+	tk = &key->conf.key[NL80211_TKIP_DATA_OFFSET_ENCR_KEY];
 	ctx = &key->u.tkip.tx;
 
 #ifdef CONFIG_MAC80211_TKIP_DEBUG
@@ -205,7 +205,7 @@ void ieee80211_tkip_encrypt_data(struct crypto_blkcipher *tfm,
 {
 	u8 rc4key[16];
 	struct tkip_ctx *ctx = &key->u.tkip.tx;
-	const u8 *tk = &key->conf.key[ALG_TKIP_TEMP_ENCR_KEY];
+	const u8 *tk = &key->conf.key[NL80211_TKIP_DATA_OFFSET_ENCR_KEY];
 
 	/* Calculate per-packet key */
 	if (ctx->iv16 == 0 || !ctx->initialized)
@@ -231,7 +231,7 @@ int ieee80211_tkip_decrypt_data(struct crypto_blkcipher *tfm,
 	u32 iv16;
 	u8 rc4key[16], keyid, *pos = payload;
 	int res;
-	const u8 *tk = &key->conf.key[ALG_TKIP_TEMP_ENCR_KEY];
+	const u8 *tk = &key->conf.key[NL80211_TKIP_DATA_OFFSET_ENCR_KEY];
 
 	if (payload_len < 12)
 		return -1;
@@ -286,13 +286,13 @@ int ieee80211_tkip_decrypt_data(struct crypto_blkcipher *tfm,
 #ifdef CONFIG_MAC80211_TKIP_DEBUG
 		{
 			int i;
+			u8 key_offset = NL80211_TKIP_DATA_OFFSET_ENCR_KEY;
 			DECLARE_MAC_BUF(mac);
 			printk(KERN_DEBUG "TKIP decrypt: Phase1 TA=%s"
 			       " TK=", print_mac(mac, ta));
 			for (i = 0; i < 16; i++)
 				printk("%02x ",
-				       key->conf.key[
-						ALG_TKIP_TEMP_ENCR_KEY + i]);
+				       key->conf.key[key_offset + i]);
 			printk("\n");
 			printk(KERN_DEBUG "TKIP decrypt: P1K=");
 			for (i = 0; i < 5; i++)
diff --git a/net/mac80211/wpa.c b/net/mac80211/wpa.c
index 345e10e9b313..f809761fbfb5 100644
--- a/net/mac80211/wpa.c
+++ b/net/mac80211/wpa.c
@@ -49,7 +49,7 @@ static int ieee80211_get_hdr_info(const struct sk_buff *skb, u8 **sa, u8 **da,
 ieee80211_tx_result
 ieee80211_tx_h_michael_mic_add(struct ieee80211_tx_data *tx)
 {
-	u8 *data, *sa, *da, *key, *mic, qos_tid;
+	u8 *data, *sa, *da, *key, *mic, qos_tid, key_offset;
 	size_t data_len;
 	u16 fc;
 	struct sk_buff *skb = tx->skb;
@@ -88,8 +88,12 @@ ieee80211_tx_h_michael_mic_add(struct ieee80211_tx_data *tx)
 #else
 	authenticator = 1;
 #endif
-	key = &tx->key->conf.key[authenticator ? ALG_TKIP_TEMP_AUTH_TX_MIC_KEY :
-				 ALG_TKIP_TEMP_AUTH_RX_MIC_KEY];
+	/* At this point we know we're using ALG_TKIP. To get the MIC key
+	 * we now will rely on the offset from the ieee80211_key_conf::key */
+	key_offset = authenticator ?
+		NL80211_TKIP_DATA_OFFSET_TX_MIC_KEY :
+		NL80211_TKIP_DATA_OFFSET_RX_MIC_KEY;
+	key = &tx->key->conf.key[key_offset];
 	mic = skb_put(skb, MICHAEL_MIC_LEN);
 	michael_mic(key, da, sa, qos_tid & 0x0f, data, data_len, mic);
 
@@ -100,7 +104,7 @@ ieee80211_tx_h_michael_mic_add(struct ieee80211_tx_data *tx)
 ieee80211_rx_result
 ieee80211_rx_h_michael_mic_verify(struct ieee80211_rx_data *rx)
 {
-	u8 *data, *sa, *da, *key = NULL, qos_tid;
+	u8 *data, *sa, *da, *key = NULL, qos_tid, key_offset;
 	size_t data_len;
 	u16 fc;
 	u8 mic[MICHAEL_MIC_LEN];
@@ -131,8 +135,12 @@ ieee80211_rx_h_michael_mic_verify(struct ieee80211_rx_data *rx)
 #else
 	authenticator = 1;
 #endif
-	key = &rx->key->conf.key[authenticator ? ALG_TKIP_TEMP_AUTH_RX_MIC_KEY :
-				 ALG_TKIP_TEMP_AUTH_TX_MIC_KEY];
+	/* At this point we know we're using ALG_TKIP. To get the MIC key
+	 * we now will rely on the offset from the ieee80211_key_conf::key */
+	key_offset = authenticator ?
+		NL80211_TKIP_DATA_OFFSET_RX_MIC_KEY :
+		NL80211_TKIP_DATA_OFFSET_TX_MIC_KEY;
+	key = &rx->key->conf.key[key_offset];
 	michael_mic(key, da, sa, qos_tid & 0x0f, data, data_len, mic);
 	if (memcmp(mic, data + data_len, MICHAEL_MIC_LEN) != 0 || wpa_test) {
 		if (!(rx->flags & IEEE80211_RX_RA_MATCH))
-- 
cgit v1.2.3


From 428da76523e4a9c08bdfadc25a05b520f19b9133 Mon Sep 17 00:00:00 2001
From: Ivo van Doorn <ivdoorn@gmail.com>
Date: Tue, 24 Jun 2008 19:23:36 +0200
Subject: mac80211: Add RTNL warning for workqueue

The workqueue provided by mac80211 should not be used for
scheduled tasks that acquire the RTNL lock. This could be done
when the driver uses the function ieee80211_iterate_active_interfaces()
within the scheduled work. Such behavior will end in locking
dependencies problems when an interface is being removed.

This patch will add a notification about the RTNL locking and
the mac80211 workqueue to prevent driver developers from
blindly using it.

Signed-off-by: Ivo van Doorn <IvDoorn@gmail.com>
Acked-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/net/mac80211.h | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index 19f1e412a0f0..02c79e6b309e 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -738,8 +738,11 @@ enum ieee80211_hw_flags {
  * @conf: &struct ieee80211_conf, device configuration, don't use.
  *
  * @workqueue: single threaded workqueue available for driver use,
- *	allocated by mac80211 on registration and flushed on
- *	unregistration.
+ *	allocated by mac80211 on registration and flushed when an
+ *	interface is removed.
+ *	NOTICE: All work performed on this workqueue should NEVER
+ *	acquire the RTNL lock (i.e. Don't use the function
+ *	ieee80211_iterate_active_interfaces())
  *
  * @priv: pointer to private area that was allocated for driver use
  *	along with this structure.
-- 
cgit v1.2.3


From 818727badc14ce57dc099a075b05505d50b7956e Mon Sep 17 00:00:00 2001
From: Jussi Kivilinna <jussi.kivilinna@mbnet.fi>
Date: Wed, 18 Jun 2008 15:40:12 +0300
Subject: rndis_host: pass buffer length to rndis_command

Pass buffer length to rndis_command so that rndis_command can read full
response buffer from device instead of max CONTROL_BUFFER_SIZE bytes.

Signed-off-by: Jussi Kivilinna <jussi.kivilinna@mbnet.fi>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
---
 drivers/net/usb/rndis_host.c      | 14 +++++++-------
 drivers/net/wireless/rndis_wlan.c |  4 ++--
 include/linux/usb/rndis_host.h    |  3 ++-
 3 files changed, 11 insertions(+), 10 deletions(-)

(limited to 'include')

diff --git a/drivers/net/usb/rndis_host.c b/drivers/net/usb/rndis_host.c
index ae467f182c40..61c98beb4d17 100644
--- a/drivers/net/usb/rndis_host.c
+++ b/drivers/net/usb/rndis_host.c
@@ -74,7 +74,7 @@ EXPORT_SYMBOL_GPL(rndis_status);
  * Call context is likely probe(), before interface name is known,
  * which is why we won't try to use it in the diagnostics.
  */
-int rndis_command(struct usbnet *dev, struct rndis_msg_hdr *buf)
+int rndis_command(struct usbnet *dev, struct rndis_msg_hdr *buf, int buflen)
 {
 	struct cdc_state	*info = (void *) &dev->data;
 	int			master_ifnum;
@@ -121,7 +121,7 @@ int rndis_command(struct usbnet *dev, struct rndis_msg_hdr *buf)
 			USB_CDC_GET_ENCAPSULATED_RESPONSE,
 			USB_DIR_IN | USB_TYPE_CLASS | USB_RECIP_INTERFACE,
 			0, master_ifnum,
-			buf, CONTROL_BUFFER_SIZE,
+			buf, buflen,
 			RNDIS_CONTROL_TIMEOUT_MS);
 		if (likely(retval >= 8)) {
 			msg_len = le32_to_cpu(buf->msg_len);
@@ -239,7 +239,7 @@ static int rndis_query(struct usbnet *dev, struct usb_interface *intf,
 	u.get->len = cpu_to_le32(in_len);
 	u.get->offset = ccpu2(20);
 
-	retval = rndis_command(dev, u.header);
+	retval = rndis_command(dev, u.header, CONTROL_BUFFER_SIZE);
 	if (unlikely(retval < 0)) {
 		dev_err(&intf->dev, "RNDIS_MSG_QUERY(0x%08x) failed, %d\n",
 				oid, retval);
@@ -328,7 +328,7 @@ generic_rndis_bind(struct usbnet *dev, struct usb_interface *intf, int flags)
 	u.init->max_transfer_size = cpu_to_le32(dev->rx_urb_size);
 
 	net->change_mtu = NULL;
-	retval = rndis_command(dev, u.header);
+	retval = rndis_command(dev, u.header, CONTROL_BUFFER_SIZE);
 	if (unlikely(retval < 0)) {
 		/* it might not even be an RNDIS device!! */
 		dev_err(&intf->dev, "RNDIS init failed, %d\n", retval);
@@ -409,7 +409,7 @@ generic_rndis_bind(struct usbnet *dev, struct usb_interface *intf, int flags)
 	u.set->offset = ccpu2((sizeof *u.set) - 8);
 	*(__le32 *)(u.buf + sizeof *u.set) = RNDIS_DEFAULT_FILTER;
 
-	retval = rndis_command(dev, u.header);
+	retval = rndis_command(dev, u.header, CONTROL_BUFFER_SIZE);
 	if (unlikely(retval < 0)) {
 		dev_err(&intf->dev, "rndis set packet filter, %d\n", retval);
 		goto halt_fail_and_release;
@@ -424,7 +424,7 @@ halt_fail_and_release:
 	memset(u.halt, 0, sizeof *u.halt);
 	u.halt->msg_type = RNDIS_MSG_HALT;
 	u.halt->msg_len = ccpu2(sizeof *u.halt);
-	(void) rndis_command(dev, (void *)u.halt);
+	(void) rndis_command(dev, (void *)u.halt, CONTROL_BUFFER_SIZE);
 fail_and_release:
 	usb_set_intfdata(info->data, NULL);
 	usb_driver_release_interface(driver_of(intf), info->data);
@@ -449,7 +449,7 @@ void rndis_unbind(struct usbnet *dev, struct usb_interface *intf)
 	if (halt) {
 		halt->msg_type = RNDIS_MSG_HALT;
 		halt->msg_len = ccpu2(sizeof *halt);
-		(void) rndis_command(dev, (void *)halt);
+		(void) rndis_command(dev, (void *)halt, CONTROL_BUFFER_SIZE);
 		kfree(halt);
 	}
 
diff --git a/drivers/net/wireless/rndis_wlan.c b/drivers/net/wireless/rndis_wlan.c
index a36d2c85e26e..f001f2afd05e 100644
--- a/drivers/net/wireless/rndis_wlan.c
+++ b/drivers/net/wireless/rndis_wlan.c
@@ -448,7 +448,7 @@ static int rndis_query_oid(struct usbnet *dev, __le32 oid, void *data, int *len)
 	u.get->msg_len = ccpu2(sizeof *u.get);
 	u.get->oid = oid;
 
-	ret = rndis_command(dev, u.header);
+	ret = rndis_command(dev, u.header, buflen);
 	if (ret == 0) {
 		ret = le32_to_cpu(u.get_c->len);
 		*len = (*len > ret) ? ret : *len;
@@ -498,7 +498,7 @@ static int rndis_set_oid(struct usbnet *dev, __le32 oid, void *data, int len)
 	u.set->handle = ccpu2(0);
 	memcpy(u.buf + sizeof(*u.set), data, len);
 
-	ret = rndis_command(dev, u.header);
+	ret = rndis_command(dev, u.header, buflen);
 	if (ret == 0)
 		ret = rndis_error_status(u.set_c->status);
 
diff --git a/include/linux/usb/rndis_host.h b/include/linux/usb/rndis_host.h
index 29d6458ecb8d..0a6e6d4b929a 100644
--- a/include/linux/usb/rndis_host.h
+++ b/include/linux/usb/rndis_host.h
@@ -260,7 +260,8 @@ struct rndis_keepalive_c {	/* IN (optionally OUT) */
 
 
 extern void rndis_status(struct usbnet *dev, struct urb *urb);
-extern int rndis_command(struct usbnet *dev, struct rndis_msg_hdr *buf);
+extern int
+rndis_command(struct usbnet *dev, struct rndis_msg_hdr *buf, int buflen);
 extern int
 generic_rndis_bind(struct usbnet *dev, struct usb_interface *intf, int flags);
 extern void rndis_unbind(struct usbnet *dev, struct usb_interface *intf);
-- 
cgit v1.2.3


From 06ff47bc9595848b818ac79e7d8069337c6e58b1 Mon Sep 17 00:00:00 2001
From: Tomas Winkler <tomas.winkler@intel.com>
Date: Wed, 18 Jun 2008 17:53:44 +0300
Subject: mac80211: add spectrum capabilities

This patch add spectrum capability and required information
elements to association request providing AP has requested it and
it is supported by the driver

Signed-off-by: Tomas Winkler <tomas.winkler@intel.com>
Signed-off-by: Assaf Krauss <assaf.krauss@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/net/mac80211.h |  5 +++++
 net/mac80211/mlme.c    | 24 ++++++++++++++++++++++++
 2 files changed, 29 insertions(+)

(limited to 'include')

diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index 02c79e6b309e..a7044958c75f 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -710,6 +710,10 @@ enum ieee80211_tkip_key_type {
  * @IEEE80211_HW_NOISE_DBM:
  *	Hardware can provide noise (radio interference) values in units dBm,
  *      decibel difference from one milliwatt.
+ *
+ * @IEEE80211_HW_SPECTRUM_MGMT:
+ * 	Hardware supports spectrum management defined in 802.11h
+ * 	Measurement, Channel Switch, Quieting, TPC
  */
 enum ieee80211_hw_flags {
 	IEEE80211_HW_HOST_GEN_BEACON_TEMPLATE		= 1<<0,
@@ -721,6 +725,7 @@ enum ieee80211_hw_flags {
 	IEEE80211_HW_SIGNAL_DB				= 1<<6,
 	IEEE80211_HW_SIGNAL_DBM				= 1<<7,
 	IEEE80211_HW_NOISE_DBM				= 1<<8,
+	IEEE80211_HW_SPECTRUM_MGMT			= 1<<9,
 };
 
 /**
diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c
index 7b4d4d46843b..c9f7c13d2e7e 100644
--- a/net/mac80211/mlme.c
+++ b/net/mac80211/mlme.c
@@ -747,6 +747,10 @@ static void ieee80211_send_assoc(struct net_device *dev,
 		 * b-only mode) */
 		rates_len = ieee80211_compatible_rates(bss, sband, &rates);
 
+		if ((bss->capability & WLAN_CAPABILITY_SPECTRUM_MGMT) &&
+		    (local->hw.flags & IEEE80211_HW_SPECTRUM_MGMT))
+			capab |= WLAN_CAPABILITY_SPECTRUM_MGMT;
+
 		ieee80211_rx_bss_put(dev, bss);
 	} else {
 		rates = ~0;
@@ -814,6 +818,26 @@ static void ieee80211_send_assoc(struct net_device *dev,
 		}
 	}
 
+	if (capab & WLAN_CAPABILITY_SPECTRUM_MGMT) {
+		/* 1. power capabilities */
+		pos = skb_put(skb, 4);
+		*pos++ = WLAN_EID_PWR_CAPABILITY;
+		*pos++ = 2;
+		*pos++ = 0; /* min tx power */
+		*pos++ = local->hw.conf.channel->max_power; /* max tx power */
+
+		/* 2. supported channels */
+		/* TODO: get this in reg domain format */
+		pos = skb_put(skb, 2 * sband->n_channels + 2);
+		*pos++ = WLAN_EID_SUPPORTED_CHANNELS;
+		*pos++ = 2 * sband->n_channels;
+		for (i = 0; i < sband->n_channels; i++) {
+			*pos++ = ieee80211_frequency_to_channel(
+					sband->channels[i].center_freq);
+			*pos++ = 1; /* one channel in the subband*/
+		}
+	}
+
 	if (ifsta->extra_ie) {
 		pos = skb_put(skb, ifsta->extra_ie_len);
 		memcpy(pos, ifsta->extra_ie, ifsta->extra_ie_len);
-- 
cgit v1.2.3


From 0853ad66b14feb12acde7ac13b7c3b75770a0adc Mon Sep 17 00:00:00 2001
From: Santwona Behera <santwona.behera@sun.com>
Date: Wed, 2 Jul 2008 03:47:41 -0700
Subject: netdev: Add support for rx flow hash configuration, using ethtool.

Added new interfaces to ethtool to configure receive network flow
distribution across multiple rx rings using hashing.

Signed-off-by: Santwona Behera <santwona.behera@sun.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/ethtool.h | 33 +++++++++++++++++++++++++++++++++
 net/core/ethtool.c      | 37 +++++++++++++++++++++++++++++++++++++
 2 files changed, 70 insertions(+)

(limited to 'include')

diff --git a/include/linux/ethtool.h b/include/linux/ethtool.h
index c8d216357865..8bb5e87df365 100644
--- a/include/linux/ethtool.h
+++ b/include/linux/ethtool.h
@@ -272,6 +272,12 @@ enum ethtool_flags {
 	ETH_FLAG_LRO		= (1 << 15),	/* LRO is enabled */
 };
 
+struct ethtool_rxnfc {
+	__u32		cmd;
+	__u32		flow_type;
+	__u64		data;
+};
+
 #ifdef __KERNEL__
 
 struct net_device;
@@ -396,6 +402,8 @@ struct ethtool_ops {
 	/* the following hooks are obsolete */
 	int	(*self_test_count)(struct net_device *);/* use get_sset_count */
 	int	(*get_stats_count)(struct net_device *);/* use get_sset_count */
+	int	(*get_rxhash)(struct net_device *, struct ethtool_rxnfc *);
+	int	(*set_rxhash)(struct net_device *, struct ethtool_rxnfc *);
 };
 #endif /* __KERNEL__ */
 
@@ -442,6 +450,9 @@ struct ethtool_ops {
 #define ETHTOOL_GPFLAGS		0x00000027 /* Get driver-private flags bitmap */
 #define ETHTOOL_SPFLAGS		0x00000028 /* Set driver-private flags bitmap */
 
+#define	ETHTOOL_GRXFH		0x00000029 /* Get RX flow hash configuration */
+#define	ETHTOOL_SRXFH		0x0000002a /* Set RX flow hash configuration */
+
 /* compatibility with older code */
 #define SPARC_ETH_GSET		ETHTOOL_GSET
 #define SPARC_ETH_SSET		ETHTOOL_SSET
@@ -528,4 +539,26 @@ struct ethtool_ops {
 #define WAKE_MAGIC		(1 << 5)
 #define WAKE_MAGICSECURE	(1 << 6) /* only meaningful if WAKE_MAGIC */
 
+/* L3-L4 network traffic flow types */
+#define	TCP_V4_FLOW	0x01
+#define	UDP_V4_FLOW	0x02
+#define	SCTP_V4_FLOW	0x03
+#define	AH_ESP_V4_FLOW	0x04
+#define	TCP_V6_FLOW	0x05
+#define	UDP_V6_FLOW	0x06
+#define	SCTP_V6_FLOW	0x07
+#define	AH_ESP_V6_FLOW	0x08
+
+/* L3-L4 network traffic flow hash options */
+#define	RXH_DEV_PORT	(1 << 0)
+#define	RXH_L2DA	(1 << 1)
+#define	RXH_VLAN	(1 << 2)
+#define	RXH_L3_PROTO	(1 << 3)
+#define	RXH_IP_SRC	(1 << 4)
+#define	RXH_IP_DST	(1 << 5)
+#define	RXH_L4_B_0_1	(1 << 6) /* src port in case of TCP/UDP/SCTP */
+#define	RXH_L4_B_2_3	(1 << 7) /* dst port in case of TCP/UDP/SCTP */
+#define	RXH_DISCARD	(1 << 31)
+
+
 #endif /* _LINUX_ETHTOOL_H */
diff --git a/net/core/ethtool.c b/net/core/ethtool.c
index 0133b5ebd545..14ada537f895 100644
--- a/net/core/ethtool.c
+++ b/net/core/ethtool.c
@@ -209,6 +209,36 @@ static int ethtool_get_drvinfo(struct net_device *dev, void __user *useraddr)
 	return 0;
 }
 
+static int ethtool_set_rxhash(struct net_device *dev, void __user *useraddr)
+{
+	struct ethtool_rxnfc cmd;
+
+	if (!dev->ethtool_ops->set_rxhash)
+		return -EOPNOTSUPP;
+
+	if (copy_from_user(&cmd, useraddr, sizeof(cmd)))
+		return -EFAULT;
+
+	return dev->ethtool_ops->set_rxhash(dev, &cmd);
+}
+
+static int ethtool_get_rxhash(struct net_device *dev, void __user *useraddr)
+{
+	struct ethtool_rxnfc info;
+
+	if (!dev->ethtool_ops->get_rxhash)
+		return -EOPNOTSUPP;
+
+	if (copy_from_user(&info, useraddr, sizeof(info)))
+		return -EFAULT;
+
+	dev->ethtool_ops->get_rxhash(dev, &info);
+
+	if (copy_to_user(useraddr, &info, sizeof(info)))
+		return -EFAULT;
+	return 0;
+}
+
 static int ethtool_get_regs(struct net_device *dev, char __user *useraddr)
 {
 	struct ethtool_regs regs;
@@ -826,6 +856,7 @@ int dev_ethtool(struct net *net, struct ifreq *ifr)
 	case ETHTOOL_GGSO:
 	case ETHTOOL_GFLAGS:
 	case ETHTOOL_GPFLAGS:
+	case ETHTOOL_GRXFH:
 		break;
 	default:
 		if (!capable(CAP_NET_ADMIN))
@@ -977,6 +1008,12 @@ int dev_ethtool(struct net *net, struct ifreq *ifr)
 		rc = ethtool_set_value(dev, useraddr,
 				       dev->ethtool_ops->set_priv_flags);
 		break;
+	case ETHTOOL_GRXFH:
+		rc = ethtool_get_rxhash(dev, useraddr);
+		break;
+	case ETHTOOL_SRXFH:
+		rc = ethtool_set_rxhash(dev, useraddr);
+		break;
 	default:
 		rc = -EOPNOTSUPP;
 	}
-- 
cgit v1.2.3


From 778d80be52699596bf70e0eb0761cf5e1e46088d Mon Sep 17 00:00:00 2001
From: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Date: Sat, 28 Jun 2008 14:17:11 +0900
Subject: ipv6: Add disable_ipv6 sysctl to disable IPv6 operaion on specific
 interface.

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
---
 Documentation/networking/ip-sysctl.txt |  4 ++++
 include/linux/ipv6.h                   |  2 ++
 net/ipv6/addrconf.c                    | 11 +++++++++++
 net/ipv6/ip6_input.c                   |  3 ++-
 net/ipv6/ip6_output.c                  |  7 +++++++
 5 files changed, 26 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/Documentation/networking/ip-sysctl.txt b/Documentation/networking/ip-sysctl.txt
index 71c7bea97160..dae980e8f1b9 100644
--- a/Documentation/networking/ip-sysctl.txt
+++ b/Documentation/networking/ip-sysctl.txt
@@ -1025,6 +1025,10 @@ max_addresses - INTEGER
 	autoconfigured addresses.
 	Default: 16
 
+disable_ipv6 - BOOLEAN
+	Disable IPv6 operation.
+	Default: FALSE (enable IPv6 operation)
+
 icmp/*:
 ratelimit - INTEGER
 	Limit the maximal rates for sending ICMPv6 packets.
diff --git a/include/linux/ipv6.h b/include/linux/ipv6.h
index cde056e08181..d9d7f9b69eb4 100644
--- a/include/linux/ipv6.h
+++ b/include/linux/ipv6.h
@@ -163,6 +163,7 @@ struct ipv6_devconf {
 #ifdef CONFIG_IPV6_MROUTE
 	__s32		mc_forwarding;
 #endif
+	__s32		disable_ipv6;
 	void		*sysctl;
 };
 
@@ -194,6 +195,7 @@ enum {
 	DEVCONF_OPTIMISTIC_DAD,
 	DEVCONF_ACCEPT_SOURCE_ROUTE,
 	DEVCONF_MC_FORWARDING,
+	DEVCONF_DISABLE_IPV6,
 	DEVCONF_MAX
 };
 
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index 8b6875f02039..8c5cff50bbed 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -183,6 +183,7 @@ struct ipv6_devconf ipv6_devconf __read_mostly = {
 #endif
 	.proxy_ndp		= 0,
 	.accept_source_route	= 0,	/* we do not accept RH0 by default. */
+	.disable_ipv6		= 0,
 };
 
 static struct ipv6_devconf ipv6_devconf_dflt __read_mostly = {
@@ -215,6 +216,7 @@ static struct ipv6_devconf ipv6_devconf_dflt __read_mostly = {
 #endif
 	.proxy_ndp		= 0,
 	.accept_source_route	= 0,	/* we do not accept RH0 by default. */
+	.disable_ipv6		= 0,
 };
 
 /* IPv6 Wildcard Address and Loopback Address defined by RFC2553 */
@@ -3657,6 +3659,7 @@ static inline void ipv6_store_devconf(struct ipv6_devconf *cnf,
 #ifdef CONFIG_IPV6_MROUTE
 	array[DEVCONF_MC_FORWARDING] = cnf->mc_forwarding;
 #endif
+	array[DEVCONF_DISABLE_IPV6] = cnf->disable_ipv6;
 }
 
 static inline size_t inet6_if_nlmsg_size(void)
@@ -4215,6 +4218,14 @@ static struct addrconf_sysctl_table
 			.proc_handler	=	&proc_dointvec,
 		},
 #endif
+		{
+			.ctl_name	=	CTL_UNNUMBERED,
+			.procname	=	"disable_ipv6",
+			.data		=	&ipv6_devconf.disable_ipv6,
+			.maxlen		=	sizeof(int),
+			.mode		=	0644,
+			.proc_handler	=	&proc_dointvec,
+		},
 		{
 			.ctl_name	=	0,	/* sentinel */
 		}
diff --git a/net/ipv6/ip6_input.c b/net/ipv6/ip6_input.c
index 34e5a96623ae..ea81c614dde2 100644
--- a/net/ipv6/ip6_input.c
+++ b/net/ipv6/ip6_input.c
@@ -71,7 +71,8 @@ int ipv6_rcv(struct sk_buff *skb, struct net_device *dev, struct packet_type *pt
 
 	IP6_INC_STATS_BH(idev, IPSTATS_MIB_INRECEIVES);
 
-	if ((skb = skb_share_check(skb, GFP_ATOMIC)) == NULL) {
+	if ((skb = skb_share_check(skb, GFP_ATOMIC)) == NULL ||
+	    !idev || unlikely(idev->cnf.disable_ipv6)) {
 		IP6_INC_STATS_BH(idev, IPSTATS_MIB_INDISCARDS);
 		rcu_read_unlock();
 		goto out;
diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c
index 871bdec09edb..0981c1ef3057 100644
--- a/net/ipv6/ip6_output.c
+++ b/net/ipv6/ip6_output.c
@@ -173,6 +173,13 @@ static inline int ip6_skb_dst_mtu(struct sk_buff *skb)
 
 int ip6_output(struct sk_buff *skb)
 {
+	struct inet6_dev *idev = ip6_dst_idev(skb->dst);
+	if (unlikely(idev->cnf.disable_ipv6)) {
+		IP6_INC_STATS(idev, IPSTATS_MIB_OUTDISCARDS);
+		kfree_skb(skb);
+		return 0;
+	}
+
 	if ((skb->len > ip6_skb_dst_mtu(skb) && !skb_is_gso(skb)) ||
 				dst_allfrag(skb->dst))
 		return ip6_fragment(skb, ip6_output2);
-- 
cgit v1.2.3


From 1b34be74cbf18f5d58cc85c7c4afcd9f7d74accd Mon Sep 17 00:00:00 2001
From: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Date: Sat, 28 Jun 2008 14:18:38 +0900
Subject: ipv6 addrconf: add accept_dad sysctl to control DAD operation.

- If 0, disable DAD.
- If 1, perform DAD (default).
- If >1, perform DAD and disable IPv6 operation if DAD for MAC-based
  link-local address has been failed (RFC4862 5.4.5).

We do not follow RFC4862 by default.  Refer to the netdev thread entitled
"Linux IPv6 DAD not full conform to RFC 4862 ?"
	http://www.spinics.net/lists/netdev/msg52027.html

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
---
 Documentation/networking/ip-sysctl.txt |  7 +++++++
 include/linux/ipv6.h                   |  2 ++
 net/ipv6/addrconf.c                    | 35 ++++++++++++++++++++++++++++++++++
 3 files changed, 44 insertions(+)

(limited to 'include')

diff --git a/Documentation/networking/ip-sysctl.txt b/Documentation/networking/ip-sysctl.txt
index dae980e8f1b9..72f6d52e52e6 100644
--- a/Documentation/networking/ip-sysctl.txt
+++ b/Documentation/networking/ip-sysctl.txt
@@ -1029,6 +1029,13 @@ disable_ipv6 - BOOLEAN
 	Disable IPv6 operation.
 	Default: FALSE (enable IPv6 operation)
 
+accept_dad - INTEGER
+	Whether to accept DAD (Duplicate Address Detection).
+	0: Disable DAD
+	1: Enable DAD (default)
+	2: Enable DAD, and disable IPv6 operation if MAC-based duplicate
+	   link-local address has been found.
+
 icmp/*:
 ratelimit - INTEGER
 	Limit the maximal rates for sending ICMPv6 packets.
diff --git a/include/linux/ipv6.h b/include/linux/ipv6.h
index d9d7f9b69eb4..391ad0843a46 100644
--- a/include/linux/ipv6.h
+++ b/include/linux/ipv6.h
@@ -164,6 +164,7 @@ struct ipv6_devconf {
 	__s32		mc_forwarding;
 #endif
 	__s32		disable_ipv6;
+	__s32		accept_dad;
 	void		*sysctl;
 };
 
@@ -196,6 +197,7 @@ enum {
 	DEVCONF_ACCEPT_SOURCE_ROUTE,
 	DEVCONF_MC_FORWARDING,
 	DEVCONF_DISABLE_IPV6,
+	DEVCONF_ACCEPT_DAD,
 	DEVCONF_MAX
 };
 
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index 8c5cff50bbed..2ec73e62202c 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -119,6 +119,7 @@ static void ipv6_regen_rndid(unsigned long data);
 static int desync_factor = MAX_DESYNC_FACTOR * HZ;
 #endif
 
+static int ipv6_generate_eui64(u8 *eui, struct net_device *dev);
 static int ipv6_count_addresses(struct inet6_dev *idev);
 
 /*
@@ -184,6 +185,7 @@ struct ipv6_devconf ipv6_devconf __read_mostly = {
 	.proxy_ndp		= 0,
 	.accept_source_route	= 0,	/* we do not accept RH0 by default. */
 	.disable_ipv6		= 0,
+	.accept_dad		= 1,
 };
 
 static struct ipv6_devconf ipv6_devconf_dflt __read_mostly = {
@@ -217,6 +219,7 @@ static struct ipv6_devconf ipv6_devconf_dflt __read_mostly = {
 	.proxy_ndp		= 0,
 	.accept_source_route	= 0,	/* we do not accept RH0 by default. */
 	.disable_ipv6		= 0,
+	.accept_dad		= 1,
 };
 
 /* IPv6 Wildcard Address and Loopback Address defined by RFC2553 */
@@ -380,6 +383,9 @@ static struct inet6_dev * ipv6_add_dev(struct net_device *dev)
 	 */
 	in6_dev_hold(ndev);
 
+	if (dev->flags & (IFF_NOARP | IFF_LOOPBACK))
+		ndev->cnf.accept_dad = -1;
+
 #if defined(CONFIG_IPV6_SIT) || defined(CONFIG_IPV6_SIT_MODULE)
 	if (dev->type == ARPHRD_SIT && (dev->priv_flags & IFF_ISATAP)) {
 		printk(KERN_INFO
@@ -1421,6 +1427,20 @@ static void addrconf_dad_stop(struct inet6_ifaddr *ifp)
 
 void addrconf_dad_failure(struct inet6_ifaddr *ifp)
 {
+	struct inet6_dev *idev = ifp->idev;
+	if (idev->cnf.accept_dad > 1 && !idev->cnf.disable_ipv6) {
+		struct in6_addr addr;
+
+		addr.s6_addr32[0] = htonl(0xfe800000);
+		addr.s6_addr32[1] = 0;
+
+		if (!ipv6_generate_eui64(addr.s6_addr + 8, idev->dev) &&
+		    ipv6_addr_equal(&ifp->addr, &addr)) {
+			/* DAD failed for link-local based on MAC address */
+			idev->cnf.disable_ipv6 = 1;
+		}
+	}
+
 	if (net_ratelimit())
 		printk(KERN_INFO "%s: duplicate address detected!\n", ifp->idev->dev->name);
 	addrconf_dad_stop(ifp);
@@ -2753,6 +2773,7 @@ static void addrconf_dad_start(struct inet6_ifaddr *ifp, u32 flags)
 	spin_lock_bh(&ifp->lock);
 
 	if (dev->flags&(IFF_NOARP|IFF_LOOPBACK) ||
+	    idev->cnf.accept_dad < 1 ||
 	    !(ifp->flags&IFA_F_TENTATIVE) ||
 	    ifp->flags & IFA_F_NODAD) {
 		ifp->flags &= ~(IFA_F_TENTATIVE|IFA_F_OPTIMISTIC);
@@ -2800,6 +2821,11 @@ static void addrconf_dad_timer(unsigned long data)
 		read_unlock_bh(&idev->lock);
 		goto out;
 	}
+	if (idev->cnf.accept_dad > 1 && idev->cnf.disable_ipv6) {
+		read_unlock_bh(&idev->lock);
+		addrconf_dad_failure(ifp);
+		return;
+	}
 	spin_lock_bh(&ifp->lock);
 	if (ifp->probes == 0) {
 		/*
@@ -3660,6 +3686,7 @@ static inline void ipv6_store_devconf(struct ipv6_devconf *cnf,
 	array[DEVCONF_MC_FORWARDING] = cnf->mc_forwarding;
 #endif
 	array[DEVCONF_DISABLE_IPV6] = cnf->disable_ipv6;
+	array[DEVCONF_ACCEPT_DAD] = cnf->accept_dad;
 }
 
 static inline size_t inet6_if_nlmsg_size(void)
@@ -4226,6 +4253,14 @@ static struct addrconf_sysctl_table
 			.mode		=	0644,
 			.proc_handler	=	&proc_dointvec,
 		},
+		{
+			.ctl_name	=	CTL_UNNUMBERED,
+			.procname	=	"accept_dad",
+			.data		=	&ipv6_devconf.accept_dad,
+			.maxlen		=	sizeof(int),
+			.mode		=	0644,
+			.proc_handler	=	&proc_dointvec,
+		},
 		{
 			.ctl_name	=	0,	/* sentinel */
 		}
-- 
cgit v1.2.3


From 623d1a1af77bd52a389c6eda5920e28eb2ee468b Mon Sep 17 00:00:00 2001
From: Wang Chen <wangchen@cn.fujitsu.com>
Date: Thu, 3 Jul 2008 12:13:30 +0800
Subject: ipv6: Do cleanup for ip6_mr_init.

If do not do it, we will get following issues:
1. Leaving junks after inet6_init failing halfway.
2. Leaving proc and notifier junks after ipv6 modules unloading.

Signed-off-by: Wang Chen <wangchen@cn.fujitsu.com>
Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
---
 include/linux/mroute6.h |  3 ++-
 net/ipv6/af_inet6.c     | 11 ++++++++++-
 net/ipv6/ip6mr.c        | 38 +++++++++++++++++++++++++++++++++-----
 3 files changed, 45 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/include/linux/mroute6.h b/include/linux/mroute6.h
index e7989593142b..4c4d6f57d5c5 100644
--- a/include/linux/mroute6.h
+++ b/include/linux/mroute6.h
@@ -135,7 +135,8 @@ extern int ip6_mroute_setsockopt(struct sock *, int, char __user *, int);
 extern int ip6_mroute_getsockopt(struct sock *, int, char __user *, int __user *);
 extern int ip6_mr_input(struct sk_buff *skb);
 extern int ip6mr_ioctl(struct sock *sk, int cmd, void __user *arg);
-extern void ip6_mr_init(void);
+extern int ip6_mr_init(void);
+extern void ip6_mr_cleanup(void);
 
 struct mif_device
 {
diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c
index 3ce8d2f318c6..6b39af1acb5e 100644
--- a/net/ipv6/af_inet6.c
+++ b/net/ipv6/af_inet6.c
@@ -953,7 +953,9 @@ static int __init inet6_init(void)
 	if (err)
 		goto icmp_fail;
 #ifdef CONFIG_IPV6_MROUTE
-	ip6_mr_init();
+	err = ip6_mr_init();
+	if (err)
+		goto ipmr_fail;
 #endif
 	err = ndisc_init();
 	if (err)
@@ -1057,6 +1059,10 @@ netfilter_fail:
 igmp_fail:
 	ndisc_cleanup();
 ndisc_fail:
+#ifdef CONFIG_IPV6_MROUTE
+	ip6_mr_cleanup();
+ipmr_fail:
+#endif
 	icmpv6_cleanup();
 icmp_fail:
 	unregister_pernet_subsys(&inet6_net_ops);
@@ -1111,6 +1117,9 @@ static void __exit inet6_exit(void)
 	ipv6_netfilter_fini();
 	igmp6_cleanup();
 	ndisc_cleanup();
+#ifdef CONFIG_IPV6_MROUTE
+	ip6_mr_cleanup();
+#endif
 	icmpv6_cleanup();
 	rawv6_exit();
 
diff --git a/net/ipv6/ip6mr.c b/net/ipv6/ip6mr.c
index 90e763073dc5..cfac26d674ed 100644
--- a/net/ipv6/ip6mr.c
+++ b/net/ipv6/ip6mr.c
@@ -948,23 +948,51 @@ static struct notifier_block ip6_mr_notifier = {
  *	Setup for IP multicast routing
  */
 
-void __init ip6_mr_init(void)
+int __init ip6_mr_init(void)
 {
+	int err;
+
 	mrt_cachep = kmem_cache_create("ip6_mrt_cache",
 				       sizeof(struct mfc6_cache),
 				       0, SLAB_HWCACHE_ALIGN,
 				       NULL);
 	if (!mrt_cachep)
-		panic("cannot allocate ip6_mrt_cache");
+		return -ENOMEM;
 
 	setup_timer(&ipmr_expire_timer, ipmr_expire_process, 0);
-	register_netdevice_notifier(&ip6_mr_notifier);
+	err = register_netdevice_notifier(&ip6_mr_notifier);
+	if (err)
+		goto reg_notif_fail;
+#ifdef CONFIG_PROC_FS
+	err = -ENOMEM;
+	if (!proc_net_fops_create(&init_net, "ip6_mr_vif", 0, &ip6mr_vif_fops))
+		goto proc_vif_fail;
+	if (!proc_net_fops_create(&init_net, "ip6_mr_cache",
+				     0, &ip6mr_mfc_fops))
+		goto proc_cache_fail;
+#endif
+	return 0;
+reg_notif_fail:
+	kmem_cache_destroy(mrt_cachep);
 #ifdef CONFIG_PROC_FS
-	proc_net_fops_create(&init_net, "ip6_mr_vif", 0, &ip6mr_vif_fops);
-	proc_net_fops_create(&init_net, "ip6_mr_cache", 0, &ip6mr_mfc_fops);
+proc_vif_fail:
+	unregister_netdevice_notifier(&ip6_mr_notifier);
+proc_cache_fail:
+	proc_net_remove(&init_net, "ip6_mr_vif");
 #endif
+	return err;
 }
 
+void ip6_mr_cleanup(void)
+{
+#ifdef CONFIG_PROC_FS
+	proc_net_remove(&init_net, "ip6_mr_cache");
+	proc_net_remove(&init_net, "ip6_mr_vif");
+#endif
+	unregister_netdevice_notifier(&ip6_mr_notifier);
+	del_timer(&ipmr_expire_timer);
+	kmem_cache_destroy(mrt_cachep);
+}
 
 static int ip6mr_mfc_add(struct mf6cctl *mfc, int mrtsock)
 {
-- 
cgit v1.2.3


From 03d2f897e9fb3218989baa2139a951ce7f5414bf Mon Sep 17 00:00:00 2001
From: Wang Chen <wangchen@cn.fujitsu.com>
Date: Thu, 3 Jul 2008 12:13:36 +0800
Subject: ipv4: Do cleanup for ip_mr_init

Same as ip6_mr_init(), make ip_mr_init() return errno if fails.
But do not do error handling in inet_init(), just print a msg.

Signed-off-by: Wang Chen <wangchen@cn.fujitsu.com>
Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
---
 include/linux/igmp.h   |  1 -
 include/linux/mroute.h |  3 +--
 net/ipv4/af_inet.c     |  5 +++--
 net/ipv4/ipmr.c        | 28 ++++++++++++++++++++++++----
 4 files changed, 28 insertions(+), 9 deletions(-)

(limited to 'include')

diff --git a/include/linux/igmp.h b/include/linux/igmp.h
index f5a1a0db2e8e..7bb3c095c15b 100644
--- a/include/linux/igmp.h
+++ b/include/linux/igmp.h
@@ -228,7 +228,6 @@ extern int ip_mc_msfget(struct sock *sk, struct ip_msfilter *msf,
 extern int ip_mc_gsfget(struct sock *sk, struct group_filter *gsf,
 		struct group_filter __user *optval, int __user *optlen);
 extern int ip_mc_sf_allow(struct sock *sk, __be32 local, __be32 rmt, int dif);
-extern void ip_mr_init(void);
 extern void ip_mc_init_dev(struct in_device *);
 extern void ip_mc_destroy_dev(struct in_device *);
 extern void ip_mc_up(struct in_device *);
diff --git a/include/linux/mroute.h b/include/linux/mroute.h
index de4decfa1bfc..df8efd42bf8a 100644
--- a/include/linux/mroute.h
+++ b/include/linux/mroute.h
@@ -147,8 +147,7 @@ static inline int ip_mroute_opt(int opt)
 extern int ip_mroute_setsockopt(struct sock *, int, char __user *, int);
 extern int ip_mroute_getsockopt(struct sock *, int, char __user *, int __user *);
 extern int ipmr_ioctl(struct sock *sk, int cmd, void __user *arg);
-extern void ip_mr_init(void);
-
+extern int ip_mr_init(void);
 
 struct vif_device
 {
diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c
index 42bd24b64b57..dc411335c14f 100644
--- a/net/ipv4/af_inet.c
+++ b/net/ipv4/af_inet.c
@@ -1479,14 +1479,15 @@ static int __init inet_init(void)
 	 *	Initialise the multicast router
 	 */
 #if defined(CONFIG_IP_MROUTE)
-	ip_mr_init();
+	if (ip_mr_init())
+		printk(KERN_CRIT "inet_init: Cannot init ipv4 mroute\n");
 #endif
 	/*
 	 *	Initialise per-cpu ipv4 mibs
 	 */
 
 	if (init_ipv4_mibs())
-		printk(KERN_CRIT "inet_init: Cannot init ipv4 mibs\n"); ;
+		printk(KERN_CRIT "inet_init: Cannot init ipv4 mibs\n");
 
 	ipv4_proc_init();
 
diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c
index 300ab0c2919e..438fab9c62a0 100644
--- a/net/ipv4/ipmr.c
+++ b/net/ipv4/ipmr.c
@@ -1878,16 +1878,36 @@ static struct net_protocol pim_protocol = {
  *	Setup for IP multicast routing
  */
 
-void __init ip_mr_init(void)
+int __init ip_mr_init(void)
 {
+	int err;
+
 	mrt_cachep = kmem_cache_create("ip_mrt_cache",
 				       sizeof(struct mfc_cache),
 				       0, SLAB_HWCACHE_ALIGN|SLAB_PANIC,
 				       NULL);
+	if (!mrt_cachep)
+		return -ENOMEM;
+
 	setup_timer(&ipmr_expire_timer, ipmr_expire_process, 0);
-	register_netdevice_notifier(&ip_mr_notifier);
+	err = register_netdevice_notifier(&ip_mr_notifier);
+	if (err)
+		goto reg_notif_fail;
 #ifdef CONFIG_PROC_FS
-	proc_net_fops_create(&init_net, "ip_mr_vif", 0, &ipmr_vif_fops);
-	proc_net_fops_create(&init_net, "ip_mr_cache", 0, &ipmr_mfc_fops);
+	err = -ENOMEM;
+	if (!proc_net_fops_create(&init_net, "ip_mr_vif", 0, &ipmr_vif_fops))
+		goto proc_vif_fail;
+	if (!proc_net_fops_create(&init_net, "ip_mr_cache", 0, &ipmr_mfc_fops))
+		goto proc_cache_fail;
 #endif
+	return 0;
+reg_notif_fail:
+	kmem_cache_destroy(mrt_cachep);
+#ifdef CONFIG_PROC_FS
+proc_vif_fail:
+	unregister_netdevice_notifier(&ip_mr_notifier);
+proc_cache_fail:
+	proc_net_remove(&init_net, "ip_mr_vif");
+#endif
+	return err;
 }
-- 
cgit v1.2.3


From e0835f8fa56d2d308486f8a34cf1c4480cd27f4e Mon Sep 17 00:00:00 2001
From: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Date: Thu, 3 Jul 2008 16:51:22 +0900
Subject: ipv4,ipv6 mroute: Add some helper inline functions to remove ugly
 ifdefs.

ip{,v6}_mroute_{set,get}sockopt() should not matter by optimization but
it would be better not to depend on optimization semantically.

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
---
 include/linux/mroute.h  | 27 +++++++++++++++++++++++++++
 include/linux/mroute6.h | 32 ++++++++++++++++++++++++++++++++
 net/ipv6/af_inet6.c     |  8 --------
 3 files changed, 59 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/include/linux/mroute.h b/include/linux/mroute.h
index df8efd42bf8a..07112ee9293a 100644
--- a/include/linux/mroute.h
+++ b/include/linux/mroute.h
@@ -144,10 +144,37 @@ static inline int ip_mroute_opt(int opt)
 }
 #endif
 
+#ifdef CONFIG_IP_MROUTE
 extern int ip_mroute_setsockopt(struct sock *, int, char __user *, int);
 extern int ip_mroute_getsockopt(struct sock *, int, char __user *, int __user *);
 extern int ipmr_ioctl(struct sock *sk, int cmd, void __user *arg);
 extern int ip_mr_init(void);
+#else
+static inline
+int ip_mroute_setsockopt(struct sock *sock,
+			 int optname, char __user *optval, int optlen)
+{
+	return -ENOPROTOOPT;
+}
+
+static inline
+int ip_mroute_getsockopt(struct sock *sock,
+			 int optname, char __user *optval, int __user *optlen)
+{
+	return -ENOPROTOOPT;
+}
+
+static inline
+int ipmr_ioctl(struct sock *sk, int cmd, void __user *arg)
+{
+	return -ENOIOCTLCMD;
+}
+
+static inline int ip_mr_init(void)
+{
+	return 0;
+}
+#endif
 
 struct vif_device
 {
diff --git a/include/linux/mroute6.h b/include/linux/mroute6.h
index 4c4d6f57d5c5..5cf50473a10f 100644
--- a/include/linux/mroute6.h
+++ b/include/linux/mroute6.h
@@ -131,12 +131,44 @@ static inline int ip6_mroute_opt(int opt)
 
 struct sock;
 
+#ifdef CONFIG_IPV6_MROUTE
 extern int ip6_mroute_setsockopt(struct sock *, int, char __user *, int);
 extern int ip6_mroute_getsockopt(struct sock *, int, char __user *, int __user *);
 extern int ip6_mr_input(struct sk_buff *skb);
 extern int ip6mr_ioctl(struct sock *sk, int cmd, void __user *arg);
 extern int ip6_mr_init(void);
 extern void ip6_mr_cleanup(void);
+#else
+static inline
+int ip6_mroute_setsockopt(struct sock *sock,
+			  int optname, char __user *optval, int optlen)
+{
+	return -ENOPROTOOPT;
+}
+
+static inline
+int ip6_mroute_getsockopt(struct sock *sock,
+			  int optname, char __user *optval, int __user *optlen)
+{
+	return -ENOPROTOOPT;
+}
+
+static inline
+int ip6mr_ioctl(struct sock *sk, int cmd, void __user *arg)
+{
+	return -ENOIOCTLCMD;
+}
+
+static inline int ip6_mr_init(void)
+{
+	return 0;
+}
+
+static inline void ip6_mr_cleanup(void)
+{
+	return;
+}
+#endif
 
 struct mif_device
 {
diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c
index 6b39af1acb5e..3d828bc4b1cf 100644
--- a/net/ipv6/af_inet6.c
+++ b/net/ipv6/af_inet6.c
@@ -59,9 +59,7 @@
 
 #include <asm/uaccess.h>
 #include <asm/system.h>
-#ifdef CONFIG_IPV6_MROUTE
 #include <linux/mroute6.h>
-#endif
 
 MODULE_AUTHOR("Cast of dozens");
 MODULE_DESCRIPTION("IPv6 protocol stack for Linux");
@@ -952,11 +950,9 @@ static int __init inet6_init(void)
 	err = icmpv6_init();
 	if (err)
 		goto icmp_fail;
-#ifdef CONFIG_IPV6_MROUTE
 	err = ip6_mr_init();
 	if (err)
 		goto ipmr_fail;
-#endif
 	err = ndisc_init();
 	if (err)
 		goto ndisc_fail;
@@ -1059,10 +1055,8 @@ netfilter_fail:
 igmp_fail:
 	ndisc_cleanup();
 ndisc_fail:
-#ifdef CONFIG_IPV6_MROUTE
 	ip6_mr_cleanup();
 ipmr_fail:
-#endif
 	icmpv6_cleanup();
 icmp_fail:
 	unregister_pernet_subsys(&inet6_net_ops);
@@ -1117,9 +1111,7 @@ static void __exit inet6_exit(void)
 	ipv6_netfilter_fini();
 	igmp6_cleanup();
 	ndisc_cleanup();
-#ifdef CONFIG_IPV6_MROUTE
 	ip6_mr_cleanup();
-#endif
 	icmpv6_cleanup();
 	rawv6_exit();
 
-- 
cgit v1.2.3


From 07240fd0902c872f044f523893364a1a24c9f278 Mon Sep 17 00:00:00 2001
From: Rusty Russell <rusty@rustcorp.com.au>
Date: Thu, 3 Jul 2008 03:45:32 -0700
Subject: tun: Interface to query tun/tap features.

The problem with introducing checksum offload and gso to tun is they
need to set dev->features to enable GSO and/or checksumming, which is
supposed to be done before register_netdevice(), ie. as part of
TUNSETIFF.

Unfortunately, TUNSETIFF has always just ignored flags it doesn't
understand, so there's no good way of detecting whether the kernel
supports new IFF_ flags.

This patch implements a TUNGETFEATURES ioctl which returns all the valid IFF
flags.  It could be extended later to include other features.

Here's an example program which uses it:

#include <linux/if_tun.h>
#include <sys/types.h>
#include <sys/ioctl.h>
#include <sys/stat.h>
#include <fcntl.h>
#include <err.h>
#include <stdio.h>

static struct {
	unsigned int flag;
	const char *name;
} known_flags[] = {
	{ IFF_TUN, "TUN" },
	{ IFF_TAP, "TAP" },
	{ IFF_NO_PI, "NO_PI" },
	{ IFF_ONE_QUEUE, "ONE_QUEUE" },
};

int main()
{
	unsigned int features, i;

	int netfd = open("/dev/net/tun", O_RDWR);
	if (netfd < 0)
		err(1, "Opening /dev/net/tun");

	if (ioctl(netfd, TUNGETFEATURES, &features) != 0) {
		printf("Kernel does not support TUNGETFEATURES, guessing\n");
		features = (IFF_TUN|IFF_TAP|IFF_NO_PI|IFF_ONE_QUEUE);
	}
	printf("Available features are: ");
	for (i = 0; i < sizeof(known_flags)/sizeof(known_flags[0]); i++) {
		if (features & known_flags[i].flag) {
			features &= ~known_flags[i].flag;
			printf("%s ", known_flags[i].name);
		}
	}
	if (features)
		printf("(UNKNOWN %#x)", features);
	printf("\n");
	return 0;
}

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Acked-by: Max Krasnyansky <maxk@qualcomm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/tun.c      | 8 ++++++++
 include/linux/if_tun.h | 1 +
 2 files changed, 9 insertions(+)

(limited to 'include')

diff --git a/drivers/net/tun.c b/drivers/net/tun.c
index 7ab94c825b57..3bb991fd2b51 100644
--- a/drivers/net/tun.c
+++ b/drivers/net/tun.c
@@ -640,6 +640,14 @@ static int tun_chr_ioctl(struct inode *inode, struct file *file,
 		return 0;
 	}
 
+	if (cmd == TUNGETFEATURES) {
+		/* Currently this just means: "what IFF flags are valid?".
+		 * This is needed because we never checked for invalid flags on
+		 * TUNSETIFF. */
+		return put_user(IFF_TUN | IFF_TAP | IFF_NO_PI | IFF_ONE_QUEUE,
+				(unsigned int __user*)argp);
+	}
+
 	if (!tun)
 		return -EBADFD;
 
diff --git a/include/linux/if_tun.h b/include/linux/if_tun.h
index 18f31b6187a3..94f76a112303 100644
--- a/include/linux/if_tun.h
+++ b/include/linux/if_tun.h
@@ -40,6 +40,7 @@
 #define TUNSETOWNER   _IOW('T', 204, int)
 #define TUNSETLINK    _IOW('T', 205, int)
 #define TUNSETGROUP   _IOW('T', 206, int)
+#define TUNGETFEATURES _IOR('T', 207, unsigned int)
 
 /* TUNSETIFF ifr flags */
 #define IFF_TUN		0x0001
-- 
cgit v1.2.3


From 5228ddc98fa49b3cedab4024e269d62410a0d806 Mon Sep 17 00:00:00 2001
From: Rusty Russell <rusty@rustcorp.com.au>
Date: Thu, 3 Jul 2008 03:46:16 -0700
Subject: tun: TUNSETFEATURES to set gso features.

ethtool is useful for setting (some) device fields, but it's
root-only.  Finer feature control is available through a tun-specific
ioctl.

(Includes Mark McLoughlin <markmc@redhat.com>'s fix to hold rtnl sem).

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Acked-by: Max Krasnyansky <maxk@qualcomm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/tun.c      | 49 +++++++++++++++++++++++++++++++++++++++++++++++++
 include/linux/if_tun.h |  7 +++++++
 2 files changed, 56 insertions(+)

(limited to 'include')

diff --git a/drivers/net/tun.c b/drivers/net/tun.c
index 3bb991fd2b51..a314955e6994 100644
--- a/drivers/net/tun.c
+++ b/drivers/net/tun.c
@@ -611,6 +611,46 @@ static int tun_set_iff(struct net *net, struct file *file, struct ifreq *ifr)
 	return err;
 }
 
+/* This is like a cut-down ethtool ops, except done via tun fd so no
+ * privs required. */
+static int set_offload(struct net_device *dev, unsigned long arg)
+{
+	unsigned int old_features, features;
+
+	old_features = dev->features;
+	/* Unset features, set them as we chew on the arg. */
+	features = (old_features & ~(NETIF_F_HW_CSUM|NETIF_F_SG|NETIF_F_FRAGLIST
+				    |NETIF_F_TSO_ECN|NETIF_F_TSO|NETIF_F_TSO6));
+
+	if (arg & TUN_F_CSUM) {
+		features |= NETIF_F_HW_CSUM|NETIF_F_SG|NETIF_F_FRAGLIST;
+		arg &= ~TUN_F_CSUM;
+
+		if (arg & (TUN_F_TSO4|TUN_F_TSO6)) {
+			if (arg & TUN_F_TSO_ECN) {
+				features |= NETIF_F_TSO_ECN;
+				arg &= ~TUN_F_TSO_ECN;
+			}
+			if (arg & TUN_F_TSO4)
+				features |= NETIF_F_TSO;
+			if (arg & TUN_F_TSO6)
+				features |= NETIF_F_TSO6;
+			arg &= ~(TUN_F_TSO4|TUN_F_TSO6);
+		}
+	}
+
+	/* This gives the user a way to test for new features in future by
+	 * trying to set them. */
+	if (arg)
+		return -EINVAL;
+
+	dev->features = features;
+	if (old_features != dev->features)
+		netdev_features_change(dev);
+
+	return 0;
+}
+
 static int tun_chr_ioctl(struct inode *inode, struct file *file,
 			 unsigned int cmd, unsigned long arg)
 {
@@ -715,6 +755,15 @@ static int tun_chr_ioctl(struct inode *inode, struct file *file,
 		break;
 #endif
 
+	case TUNSETOFFLOAD:
+	{
+		int ret;
+		rtnl_lock();
+		ret = set_offload(tun->dev, arg);
+		rtnl_unlock();
+		return ret;
+	}
+
 	case SIOCGIFFLAGS:
 		ifr.ifr_flags = tun->if_flags;
 		if (copy_to_user( argp, &ifr, sizeof ifr))
diff --git a/include/linux/if_tun.h b/include/linux/if_tun.h
index 94f76a112303..3f0a0995d449 100644
--- a/include/linux/if_tun.h
+++ b/include/linux/if_tun.h
@@ -41,6 +41,7 @@
 #define TUNSETLINK    _IOW('T', 205, int)
 #define TUNSETGROUP   _IOW('T', 206, int)
 #define TUNGETFEATURES _IOR('T', 207, unsigned int)
+#define TUNSETOFFLOAD _IOW('T', 208, unsigned int)
 
 /* TUNSETIFF ifr flags */
 #define IFF_TUN		0x0001
@@ -48,6 +49,12 @@
 #define IFF_NO_PI	0x1000
 #define IFF_ONE_QUEUE	0x2000
 
+/* Features for GSO (TUNSETOFFLOAD). */
+#define TUN_F_CSUM	0x01	/* You can hand me unchecksummed packets. */
+#define TUN_F_TSO4	0x02	/* I can handle TSO for IPv4 packets */
+#define TUN_F_TSO6	0x04	/* I can handle TSO for IPv6 packets */
+#define TUN_F_TSO_ECN	0x08	/* I can handle TSO with ECN bits. */
+
 struct tun_pi {
 	unsigned short flags;
 	__be16 proto;
-- 
cgit v1.2.3


From f43798c27684ab925adde7d8acc34c78c6e50df8 Mon Sep 17 00:00:00 2001
From: Rusty Russell <rusty@rustcorp.com.au>
Date: Thu, 3 Jul 2008 03:48:02 -0700
Subject: tun: Allow GSO using virtio_net_hdr

Add a IFF_VNET_HDR flag.  This uses the same ABI as virtio_net
(ie. prepending struct virtio_net_hdr to packets) to indicate GSO and
checksum information.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Acked-by: Max Krasnyansky <maxk@qualcomm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/tun.c      | 95 ++++++++++++++++++++++++++++++++++++++++++++++++--
 include/linux/if_tun.h |  2 ++
 2 files changed, 94 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/drivers/net/tun.c b/drivers/net/tun.c
index a314955e6994..aa4ee4439f04 100644
--- a/drivers/net/tun.c
+++ b/drivers/net/tun.c
@@ -63,6 +63,7 @@
 #include <linux/if_tun.h>
 #include <linux/crc32.h>
 #include <linux/nsproxy.h>
+#include <linux/virtio_net.h>
 #include <net/net_namespace.h>
 #include <net/netns/generic.h>
 
@@ -283,6 +284,7 @@ static __inline__ ssize_t tun_get_user(struct tun_struct *tun, struct iovec *iv,
 	struct tun_pi pi = { 0, __constant_htons(ETH_P_IP) };
 	struct sk_buff *skb;
 	size_t len = count, align = 0;
+	struct virtio_net_hdr gso = { 0 };
 
 	if (!(tun->flags & TUN_NO_PI)) {
 		if ((len -= sizeof(pi)) > count)
@@ -292,6 +294,17 @@ static __inline__ ssize_t tun_get_user(struct tun_struct *tun, struct iovec *iv,
 			return -EFAULT;
 	}
 
+	if (tun->flags & TUN_VNET_HDR) {
+		if ((len -= sizeof(gso)) > count)
+			return -EINVAL;
+
+		if (memcpy_fromiovec((void *)&gso, iv, sizeof(gso)))
+			return -EFAULT;
+
+		if (gso.hdr_len > len)
+			return -EINVAL;
+	}
+
 	if ((tun->flags & TUN_TYPE_MASK) == TUN_TAP_DEV) {
 		align = NET_IP_ALIGN;
 		if (unlikely(len < ETH_HLEN))
@@ -311,6 +324,16 @@ static __inline__ ssize_t tun_get_user(struct tun_struct *tun, struct iovec *iv,
 		return -EFAULT;
 	}
 
+	if (gso.flags & VIRTIO_NET_HDR_F_NEEDS_CSUM) {
+		if (!skb_partial_csum_set(skb, gso.csum_start,
+					  gso.csum_offset)) {
+			tun->dev->stats.rx_frame_errors++;
+			kfree_skb(skb);
+			return -EINVAL;
+		}
+	} else if (tun->flags & TUN_NOCHECKSUM)
+		skb->ip_summed = CHECKSUM_UNNECESSARY;
+
 	switch (tun->flags & TUN_TYPE_MASK) {
 	case TUN_TUN_DEV:
 		if (tun->flags & TUN_NO_PI) {
@@ -337,8 +360,35 @@ static __inline__ ssize_t tun_get_user(struct tun_struct *tun, struct iovec *iv,
 		break;
 	};
 
-	if (tun->flags & TUN_NOCHECKSUM)
-		skb->ip_summed = CHECKSUM_UNNECESSARY;
+	if (gso.gso_type != VIRTIO_NET_HDR_GSO_NONE) {
+		pr_debug("GSO!\n");
+		switch (gso.gso_type & ~VIRTIO_NET_HDR_GSO_ECN) {
+		case VIRTIO_NET_HDR_GSO_TCPV4:
+			skb_shinfo(skb)->gso_type = SKB_GSO_TCPV4;
+			break;
+		case VIRTIO_NET_HDR_GSO_TCPV6:
+			skb_shinfo(skb)->gso_type = SKB_GSO_TCPV6;
+			break;
+		default:
+			tun->dev->stats.rx_frame_errors++;
+			kfree_skb(skb);
+			return -EINVAL;
+		}
+
+		if (gso.gso_type & VIRTIO_NET_HDR_GSO_ECN)
+			skb_shinfo(skb)->gso_type |= SKB_GSO_TCP_ECN;
+
+		skb_shinfo(skb)->gso_size = gso.gso_size;
+		if (skb_shinfo(skb)->gso_size == 0) {
+			tun->dev->stats.rx_frame_errors++;
+			kfree_skb(skb);
+			return -EINVAL;
+		}
+
+		/* Header must be checked, and gso_segs computed. */
+		skb_shinfo(skb)->gso_type |= SKB_GSO_DODGY;
+		skb_shinfo(skb)->gso_segs = 0;
+	}
 
 	netif_rx_ni(skb);
 	tun->dev->last_rx = jiffies;
@@ -384,6 +434,39 @@ static __inline__ ssize_t tun_put_user(struct tun_struct *tun,
 		total += sizeof(pi);
 	}
 
+	if (tun->flags & TUN_VNET_HDR) {
+		struct virtio_net_hdr gso = { 0 }; /* no info leak */
+		if ((len -= sizeof(gso)) < 0)
+			return -EINVAL;
+
+		if (skb_is_gso(skb)) {
+			struct skb_shared_info *sinfo = skb_shinfo(skb);
+
+			/* This is a hint as to how much should be linear. */
+			gso.hdr_len = skb_headlen(skb);
+			gso.gso_size = sinfo->gso_size;
+			if (sinfo->gso_type & SKB_GSO_TCPV4)
+				gso.gso_type = VIRTIO_NET_HDR_GSO_TCPV4;
+			else if (sinfo->gso_type & SKB_GSO_TCPV6)
+				gso.gso_type = VIRTIO_NET_HDR_GSO_TCPV6;
+			else
+				BUG();
+			if (sinfo->gso_type & SKB_GSO_TCP_ECN)
+				gso.gso_type |= VIRTIO_NET_HDR_GSO_ECN;
+		} else
+			gso.gso_type = VIRTIO_NET_HDR_GSO_NONE;
+
+		if (skb->ip_summed == CHECKSUM_PARTIAL) {
+			gso.flags = VIRTIO_NET_HDR_F_NEEDS_CSUM;
+			gso.csum_start = skb->csum_start - skb_headroom(skb);
+			gso.csum_offset = skb->csum_offset;
+		} /* else everything is zero */
+
+		if (unlikely(memcpy_toiovec(iv, (void *)&gso, sizeof(gso))))
+			return -EFAULT;
+		total += sizeof(gso);
+	}
+
 	len = min_t(int, skb->len, len);
 
 	skb_copy_datagram_iovec(skb, 0, iv, len);
@@ -598,6 +681,11 @@ static int tun_set_iff(struct net *net, struct file *file, struct ifreq *ifr)
 	else
 		tun->flags &= ~TUN_ONE_QUEUE;
 
+	if (ifr->ifr_flags & IFF_VNET_HDR)
+		tun->flags |= TUN_VNET_HDR;
+	else
+		tun->flags &= ~TUN_VNET_HDR;
+
 	file->private_data = tun;
 	tun->attached = 1;
 	get_net(dev_net(tun->dev));
@@ -684,7 +772,8 @@ static int tun_chr_ioctl(struct inode *inode, struct file *file,
 		/* Currently this just means: "what IFF flags are valid?".
 		 * This is needed because we never checked for invalid flags on
 		 * TUNSETIFF. */
-		return put_user(IFF_TUN | IFF_TAP | IFF_NO_PI | IFF_ONE_QUEUE,
+		return put_user(IFF_TUN | IFF_TAP | IFF_NO_PI | IFF_ONE_QUEUE |
+				IFF_VNET_HDR,
 				(unsigned int __user*)argp);
 	}
 
diff --git a/include/linux/if_tun.h b/include/linux/if_tun.h
index 3f0a0995d449..563fae542da6 100644
--- a/include/linux/if_tun.h
+++ b/include/linux/if_tun.h
@@ -31,6 +31,7 @@
 #define TUN_NO_PI	0x0040
 #define TUN_ONE_QUEUE	0x0080
 #define TUN_PERSIST 	0x0100	
+#define TUN_VNET_HDR 	0x0200
 
 /* Ioctl defines */
 #define TUNSETNOCSUM  _IOW('T', 200, int) 
@@ -48,6 +49,7 @@
 #define IFF_TAP		0x0002
 #define IFF_NO_PI	0x1000
 #define IFF_ONE_QUEUE	0x2000
+#define IFF_VNET_HDR	0x4000
 
 /* Features for GSO (TUNSETOFFLOAD). */
 #define TUN_F_CSUM	0x01	/* You can hand me unchecksummed packets. */
-- 
cgit v1.2.3


From 198191c4a7ce4daba379608fb38b9bc5a4eedc61 Mon Sep 17 00:00:00 2001
From: Krzysztof Halasa <khc@pm.waw.pl>
Date: Mon, 30 Jun 2008 23:26:53 +0200
Subject: WAN: convert drivers to use built-in netdev_stats
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

There is no point in using separate net_device_stats structs when
the one in struct net_device is present. Compiles.

Signed-off-by: Krzysztof Hałasa <khc@pm.waw.pl>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
---
 drivers/char/pcmcia/synclink_cs.c | 28 +++++++--------
 drivers/char/synclink.c           | 33 +++++++++---------
 drivers/char/synclink_gt.c        | 28 +++++++--------
 drivers/char/synclinkmp.c         | 31 ++++++++---------
 drivers/net/wan/c101.c            |  6 ++--
 drivers/net/wan/dscc4.c           | 22 ++++++------
 drivers/net/wan/farsync.c         | 70 +++++++++++++++++---------------------
 drivers/net/wan/hd6457x.c         | 33 +++++++++---------
 drivers/net/wan/hdlc.c            |  2 +-
 drivers/net/wan/hdlc_cisco.c      |  4 +--
 drivers/net/wan/hdlc_fr.c         | 29 ++++++----------
 drivers/net/wan/hdlc_raw_eth.c    |  2 +-
 drivers/net/wan/hdlc_x25.c        |  6 ++--
 drivers/net/wan/pc300_drv.c       | 71 +++++++++++++++++----------------------
 drivers/net/wan/pc300_tty.c       |  6 ++--
 drivers/net/wan/wanxl.c           | 26 +++++++-------
 include/linux/hdlc.h              |  7 ----
 17 files changed, 174 insertions(+), 230 deletions(-)

(limited to 'include')

diff --git a/drivers/char/pcmcia/synclink_cs.c b/drivers/char/pcmcia/synclink_cs.c
index 1dd0e992c83d..fb2fb159faa3 100644
--- a/drivers/char/pcmcia/synclink_cs.c
+++ b/drivers/char/pcmcia/synclink_cs.c
@@ -3886,9 +3886,8 @@ static bool rx_get_frame(MGSLPC_INFO *info)
 		framesize = 0;
 #if SYNCLINK_GENERIC_HDLC
 		{
-			struct net_device_stats *stats = hdlc_stats(info->netdev);
-			stats->rx_errors++;
-			stats->rx_frame_errors++;
+			info->netdev->stats.rx_errors++;
+			info->netdev->stats.rx_frame_errors++;
 		}
 #endif
 	} else
@@ -4144,7 +4143,6 @@ static int hdlcdev_attach(struct net_device *dev, unsigned short encoding,
 static int hdlcdev_xmit(struct sk_buff *skb, struct net_device *dev)
 {
 	MGSLPC_INFO *info = dev_to_port(dev);
-	struct net_device_stats *stats = hdlc_stats(dev);
 	unsigned long flags;
 
 	if (debug_level >= DEBUG_LEVEL_INFO)
@@ -4159,8 +4157,8 @@ static int hdlcdev_xmit(struct sk_buff *skb, struct net_device *dev)
 	info->tx_put = info->tx_count = skb->len;
 
 	/* update network statistics */
-	stats->tx_packets++;
-	stats->tx_bytes += skb->len;
+	dev->stats.tx_packets++;
+	dev->stats.tx_bytes += skb->len;
 
 	/* done with socket buffer, so free it */
 	dev_kfree_skb(skb);
@@ -4376,14 +4374,13 @@ static int hdlcdev_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
 static void hdlcdev_tx_timeout(struct net_device *dev)
 {
 	MGSLPC_INFO *info = dev_to_port(dev);
-	struct net_device_stats *stats = hdlc_stats(dev);
 	unsigned long flags;
 
 	if (debug_level >= DEBUG_LEVEL_INFO)
 		printk("hdlcdev_tx_timeout(%s)\n",dev->name);
 
-	stats->tx_errors++;
-	stats->tx_aborted_errors++;
+	dev->stats.tx_errors++;
+	dev->stats.tx_aborted_errors++;
 
 	spin_lock_irqsave(&info->lock,flags);
 	tx_stop(info);
@@ -4416,27 +4413,26 @@ static void hdlcdev_rx(MGSLPC_INFO *info, char *buf, int size)
 {
 	struct sk_buff *skb = dev_alloc_skb(size);
 	struct net_device *dev = info->netdev;
-	struct net_device_stats *stats = hdlc_stats(dev);
 
 	if (debug_level >= DEBUG_LEVEL_INFO)
 		printk("hdlcdev_rx(%s)\n",dev->name);
 
 	if (skb == NULL) {
 		printk(KERN_NOTICE "%s: can't alloc skb, dropping packet\n", dev->name);
-		stats->rx_dropped++;
+		dev->stats.rx_dropped++;
 		return;
 	}
 
-	memcpy(skb_put(skb, size),buf,size);
+	memcpy(skb_put(skb, size), buf, size);
 
-	skb->protocol = hdlc_type_trans(skb, info->netdev);
+	skb->protocol = hdlc_type_trans(skb, dev);
 
-	stats->rx_packets++;
-	stats->rx_bytes += size;
+	dev->stats.rx_packets++;
+	dev->stats.rx_bytes += size;
 
 	netif_rx(skb);
 
-	info->netdev->last_rx = jiffies;
+	dev->last_rx = jiffies;
 }
 
 /**
diff --git a/drivers/char/synclink.c b/drivers/char/synclink.c
index ac5080df2565..9d247d8a87a3 100644
--- a/drivers/char/synclink.c
+++ b/drivers/char/synclink.c
@@ -6640,9 +6640,8 @@ static bool mgsl_get_rx_frame(struct mgsl_struct *info)
 		framesize = 0;
 #if SYNCLINK_GENERIC_HDLC
 		{
-			struct net_device_stats *stats = hdlc_stats(info->netdev);
-			stats->rx_errors++;
-			stats->rx_frame_errors++;
+			info->netdev->stats.rx_errors++;
+			info->netdev->stats.rx_frame_errors++;
 		}
 #endif
 	} else
@@ -7753,7 +7752,6 @@ static int hdlcdev_attach(struct net_device *dev, unsigned short encoding,
 static int hdlcdev_xmit(struct sk_buff *skb, struct net_device *dev)
 {
 	struct mgsl_struct *info = dev_to_port(dev);
-	struct net_device_stats *stats = hdlc_stats(dev);
 	unsigned long flags;
 
 	if (debug_level >= DEBUG_LEVEL_INFO)
@@ -7767,8 +7765,8 @@ static int hdlcdev_xmit(struct sk_buff *skb, struct net_device *dev)
 	mgsl_load_tx_dma_buffer(info, skb->data, skb->len);
 
 	/* update network statistics */
-	stats->tx_packets++;
-	stats->tx_bytes += skb->len;
+	dev->stats.tx_packets++;
+	dev->stats.tx_bytes += skb->len;
 
 	/* done with socket buffer, so free it */
 	dev_kfree_skb(skb);
@@ -7984,14 +7982,13 @@ static int hdlcdev_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
 static void hdlcdev_tx_timeout(struct net_device *dev)
 {
 	struct mgsl_struct *info = dev_to_port(dev);
-	struct net_device_stats *stats = hdlc_stats(dev);
 	unsigned long flags;
 
 	if (debug_level >= DEBUG_LEVEL_INFO)
 		printk("hdlcdev_tx_timeout(%s)\n",dev->name);
 
-	stats->tx_errors++;
-	stats->tx_aborted_errors++;
+	dev->stats.tx_errors++;
+	dev->stats.tx_aborted_errors++;
 
 	spin_lock_irqsave(&info->irq_spinlock,flags);
 	usc_stop_transmitter(info);
@@ -8024,27 +8021,27 @@ static void hdlcdev_rx(struct mgsl_struct *info, char *buf, int size)
 {
 	struct sk_buff *skb = dev_alloc_skb(size);
 	struct net_device *dev = info->netdev;
-	struct net_device_stats *stats = hdlc_stats(dev);
 
 	if (debug_level >= DEBUG_LEVEL_INFO)
-		printk("hdlcdev_rx(%s)\n",dev->name);
+		printk("hdlcdev_rx(%s)\n", dev->name);
 
 	if (skb == NULL) {
-		printk(KERN_NOTICE "%s: can't alloc skb, dropping packet\n", dev->name);
-		stats->rx_dropped++;
+		printk(KERN_NOTICE "%s: can't alloc skb, dropping packet\n",
+		       dev->name);
+		dev->stats.rx_dropped++;
 		return;
 	}
 
-	memcpy(skb_put(skb, size),buf,size);
+	memcpy(skb_put(skb, size), buf, size);
 
-	skb->protocol = hdlc_type_trans(skb, info->netdev);
+	skb->protocol = hdlc_type_trans(skb, dev);
 
-	stats->rx_packets++;
-	stats->rx_bytes += size;
+	dev->stats.rx_packets++;
+	dev->stats.rx_bytes += size;
 
 	netif_rx(skb);
 
-	info->netdev->last_rx = jiffies;
+	dev->last_rx = jiffies;
 }
 
 /**
diff --git a/drivers/char/synclink_gt.c b/drivers/char/synclink_gt.c
index 55c1653be00c..d88a607e34b7 100644
--- a/drivers/char/synclink_gt.c
+++ b/drivers/char/synclink_gt.c
@@ -1544,7 +1544,6 @@ static int hdlcdev_attach(struct net_device *dev, unsigned short encoding,
 static int hdlcdev_xmit(struct sk_buff *skb, struct net_device *dev)
 {
 	struct slgt_info *info = dev_to_port(dev);
-	struct net_device_stats *stats = hdlc_stats(dev);
 	unsigned long flags;
 
 	DBGINFO(("%s hdlc_xmit\n", dev->name));
@@ -1557,8 +1556,8 @@ static int hdlcdev_xmit(struct sk_buff *skb, struct net_device *dev)
 	tx_load(info, skb->data, skb->len);
 
 	/* update network statistics */
-	stats->tx_packets++;
-	stats->tx_bytes += skb->len;
+	dev->stats.tx_packets++;
+	dev->stats.tx_bytes += skb->len;
 
 	/* done with socket buffer, so free it */
 	dev_kfree_skb(skb);
@@ -1775,13 +1774,12 @@ static int hdlcdev_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
 static void hdlcdev_tx_timeout(struct net_device *dev)
 {
 	struct slgt_info *info = dev_to_port(dev);
-	struct net_device_stats *stats = hdlc_stats(dev);
 	unsigned long flags;
 
 	DBGINFO(("%s hdlcdev_tx_timeout\n", dev->name));
 
-	stats->tx_errors++;
-	stats->tx_aborted_errors++;
+	dev->stats.tx_errors++;
+	dev->stats.tx_aborted_errors++;
 
 	spin_lock_irqsave(&info->lock,flags);
 	tx_stop(info);
@@ -1814,26 +1812,25 @@ static void hdlcdev_rx(struct slgt_info *info, char *buf, int size)
 {
 	struct sk_buff *skb = dev_alloc_skb(size);
 	struct net_device *dev = info->netdev;
-	struct net_device_stats *stats = hdlc_stats(dev);
 
 	DBGINFO(("%s hdlcdev_rx\n", dev->name));
 
 	if (skb == NULL) {
 		DBGERR(("%s: can't alloc skb, drop packet\n", dev->name));
-		stats->rx_dropped++;
+		dev->stats.rx_dropped++;
 		return;
 	}
 
-	memcpy(skb_put(skb, size),buf,size);
+	memcpy(skb_put(skb, size), buf, size);
 
-	skb->protocol = hdlc_type_trans(skb, info->netdev);
+	skb->protocol = hdlc_type_trans(skb, dev);
 
-	stats->rx_packets++;
-	stats->rx_bytes += size;
+	dev->stats.rx_packets++;
+	dev->stats.rx_bytes += size;
 
 	netif_rx(skb);
 
-	info->netdev->last_rx = jiffies;
+	dev->last_rx = jiffies;
 }
 
 /**
@@ -4577,9 +4574,8 @@ check_again:
 
 #if SYNCLINK_GENERIC_HDLC
 	if (framesize == 0) {
-		struct net_device_stats *stats = hdlc_stats(info->netdev);
-		stats->rx_errors++;
-		stats->rx_frame_errors++;
+		info->netdev->stats.rx_errors++;
+		info->netdev->stats.rx_frame_errors++;
 	}
 #endif
 
diff --git a/drivers/char/synclinkmp.c b/drivers/char/synclinkmp.c
index bec54866e0bb..10241ed86100 100644
--- a/drivers/char/synclinkmp.c
+++ b/drivers/char/synclinkmp.c
@@ -1678,7 +1678,6 @@ static int hdlcdev_attach(struct net_device *dev, unsigned short encoding,
 static int hdlcdev_xmit(struct sk_buff *skb, struct net_device *dev)
 {
 	SLMP_INFO *info = dev_to_port(dev);
-	struct net_device_stats *stats = hdlc_stats(dev);
 	unsigned long flags;
 
 	if (debug_level >= DEBUG_LEVEL_INFO)
@@ -1692,8 +1691,8 @@ static int hdlcdev_xmit(struct sk_buff *skb, struct net_device *dev)
 	tx_load_dma_buffer(info, skb->data, skb->len);
 
 	/* update network statistics */
-	stats->tx_packets++;
-	stats->tx_bytes += skb->len;
+	dev->stats.tx_packets++;
+	dev->stats.tx_bytes += skb->len;
 
 	/* done with socket buffer, so free it */
 	dev_kfree_skb(skb);
@@ -1909,14 +1908,13 @@ static int hdlcdev_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
 static void hdlcdev_tx_timeout(struct net_device *dev)
 {
 	SLMP_INFO *info = dev_to_port(dev);
-	struct net_device_stats *stats = hdlc_stats(dev);
 	unsigned long flags;
 
 	if (debug_level >= DEBUG_LEVEL_INFO)
 		printk("hdlcdev_tx_timeout(%s)\n",dev->name);
 
-	stats->tx_errors++;
-	stats->tx_aborted_errors++;
+	dev->stats.tx_errors++;
+	dev->stats.tx_aborted_errors++;
 
 	spin_lock_irqsave(&info->lock,flags);
 	tx_stop(info);
@@ -1949,27 +1947,27 @@ static void hdlcdev_rx(SLMP_INFO *info, char *buf, int size)
 {
 	struct sk_buff *skb = dev_alloc_skb(size);
 	struct net_device *dev = info->netdev;
-	struct net_device_stats *stats = hdlc_stats(dev);
 
 	if (debug_level >= DEBUG_LEVEL_INFO)
 		printk("hdlcdev_rx(%s)\n",dev->name);
 
 	if (skb == NULL) {
-		printk(KERN_NOTICE "%s: can't alloc skb, dropping packet\n", dev->name);
-		stats->rx_dropped++;
+		printk(KERN_NOTICE "%s: can't alloc skb, dropping packet\n",
+		       dev->name);
+		dev->stats.rx_dropped++;
 		return;
 	}
 
-	memcpy(skb_put(skb, size),buf,size);
+	memcpy(skb_put(skb, size), buf, size);
 
-	skb->protocol = hdlc_type_trans(skb, info->netdev);
+	skb->protocol = hdlc_type_trans(skb, dev);
 
-	stats->rx_packets++;
-	stats->rx_bytes += size;
+	dev->stats.rx_packets++;
+	dev->stats.rx_bytes += size;
 
 	netif_rx(skb);
 
-	info->netdev->last_rx = jiffies;
+	dev->last_rx = jiffies;
 }
 
 /**
@@ -4983,9 +4981,8 @@ CheckAgain:
 		framesize = 0;
 #if SYNCLINK_GENERIC_HDLC
 		{
-			struct net_device_stats *stats = hdlc_stats(info->netdev);
-			stats->rx_errors++;
-			stats->rx_frame_errors++;
+			info->netdev->stats.rx_errors++;
+			info->netdev->stats.rx_frame_errors++;
 		}
 #endif
 	}
diff --git a/drivers/net/wan/c101.c b/drivers/net/wan/c101.c
index c2cc42f723d5..c8e563106a4a 100644
--- a/drivers/net/wan/c101.c
+++ b/drivers/net/wan/c101.c
@@ -133,9 +133,9 @@ static void sca_msci_intr(port_t *port)
 	sca_out(stat & (ST1_UDRN | ST1_CDCD), MSCI0_OFFSET + ST1, port);
 
 	if (stat & ST1_UDRN) {
-		struct net_device_stats *stats = hdlc_stats(port_to_dev(port));
-		stats->tx_errors++; /* TX Underrun error detected */
-		stats->tx_fifo_errors++;
+		/* TX Underrun error detected */
+		port_to_dev(port)->stats.tx_errors++;
+		port_to_dev(port)->stats.tx_fifo_errors++;
 	}
 
 	stat = sca_in(MSCI1_OFFSET + ST1, port); /* read MSCI1 ST1 status */
diff --git a/drivers/net/wan/dscc4.c b/drivers/net/wan/dscc4.c
index c6f26e28e376..50ef5b4efd6d 100644
--- a/drivers/net/wan/dscc4.c
+++ b/drivers/net/wan/dscc4.c
@@ -642,7 +642,6 @@ static inline void dscc4_rx_skb(struct dscc4_dev_priv *dpriv,
 				struct net_device *dev)
 {
 	struct RxFD *rx_fd = dpriv->rx_fd + dpriv->rx_current%RX_RING_SIZE;
-	struct net_device_stats *stats = hdlc_stats(dev);
 	struct pci_dev *pdev = dpriv->pci_priv->pdev;
 	struct sk_buff *skb;
 	int pkt_len;
@@ -656,8 +655,8 @@ static inline void dscc4_rx_skb(struct dscc4_dev_priv *dpriv,
 	pci_unmap_single(pdev, le32_to_cpu(rx_fd->data),
 			 RX_MAX(HDLC_MAX_MRU), PCI_DMA_FROMDEVICE);
 	if ((skb->data[--pkt_len] & FrameOk) == FrameOk) {
-		stats->rx_packets++;
-		stats->rx_bytes += pkt_len;
+		dev->stats.rx_packets++;
+		dev->stats.rx_bytes += pkt_len;
 		skb_put(skb, pkt_len);
 		if (netif_running(dev))
 			skb->protocol = hdlc_type_trans(skb, dev);
@@ -665,13 +664,13 @@ static inline void dscc4_rx_skb(struct dscc4_dev_priv *dpriv,
 		netif_rx(skb);
 	} else {
 		if (skb->data[pkt_len] & FrameRdo)
-			stats->rx_fifo_errors++;
+			dev->stats.rx_fifo_errors++;
 		else if (!(skb->data[pkt_len] | ~FrameCrc))
-			stats->rx_crc_errors++;
+			dev->stats.rx_crc_errors++;
 		else if (!(skb->data[pkt_len] | ~(FrameVfr | FrameRab)))
-			stats->rx_length_errors++;
+			dev->stats.rx_length_errors++;
 		else
-			stats->rx_errors++;
+			dev->stats.rx_errors++;
 		dev_kfree_skb_irq(skb);
 	}
 refill:
@@ -1569,7 +1568,6 @@ try:
 
 	if (state & SccEvt) {
 		if (state & Alls) {
-			struct net_device_stats *stats = hdlc_stats(dev);
 			struct sk_buff *skb;
 			struct TxFD *tx_fd;
 
@@ -1586,8 +1584,8 @@ try:
 				pci_unmap_single(ppriv->pdev, le32_to_cpu(tx_fd->data),
 						 skb->len, PCI_DMA_TODEVICE);
 				if (tx_fd->state & FrameEnd) {
-					stats->tx_packets++;
-					stats->tx_bytes += skb->len;
+					dev->stats.tx_packets++;
+					dev->stats.tx_bytes += skb->len;
 				}
 				dev_kfree_skb_irq(skb);
 				dpriv->tx_skbuff[cur] = NULL;
@@ -1698,7 +1696,7 @@ try:
 		}
 		if (state & Err) {
 			printk(KERN_INFO "%s: Tx ERR\n", dev->name);
-			hdlc_stats(dev)->tx_errors++;
+			dev->stats.tx_errors++;
 			state &= ~Err;
 		}
 	}
@@ -1834,7 +1832,7 @@ try:
 				if (!(rx_fd->state2 & DataComplete))
 					break;
 				if (rx_fd->state2 & FrameAborted) {
-					hdlc_stats(dev)->rx_over_errors++;
+					dev->stats.rx_over_errors++;
 					rx_fd->state1 |= Hold;
 					rx_fd->state2 = 0x00000000;
 					rx_fd->end = cpu_to_le32(0xbabeface);
diff --git a/drivers/net/wan/farsync.c b/drivers/net/wan/farsync.c
index 547368e9633d..754f00809e3e 100644
--- a/drivers/net/wan/farsync.c
+++ b/drivers/net/wan/farsync.c
@@ -845,7 +845,6 @@ fst_tx_dma_complete(struct fst_card_info *card, struct fst_port_info *port,
 		    int len, int txpos)
 {
 	struct net_device *dev = port_to_dev(port);
-	struct net_device_stats *stats = hdlc_stats(dev);
 
 	/*
 	 * Everything is now set, just tell the card to go
@@ -853,8 +852,8 @@ fst_tx_dma_complete(struct fst_card_info *card, struct fst_port_info *port,
 	dbg(DBG_TX, "fst_tx_dma_complete\n");
 	FST_WRB(card, txDescrRing[port->index][txpos].bits,
 		DMA_OWN | TX_STP | TX_ENP);
-	stats->tx_packets++;
-	stats->tx_bytes += len;
+	dev->stats.tx_packets++;
+	dev->stats.tx_bytes += len;
 	dev->trans_start = jiffies;
 }
 
@@ -876,7 +875,6 @@ fst_rx_dma_complete(struct fst_card_info *card, struct fst_port_info *port,
 		    int len, struct sk_buff *skb, int rxp)
 {
 	struct net_device *dev = port_to_dev(port);
-	struct net_device_stats *stats = hdlc_stats(dev);
 	int pi;
 	int rx_status;
 
@@ -888,8 +886,8 @@ fst_rx_dma_complete(struct fst_card_info *card, struct fst_port_info *port,
 	FST_WRB(card, rxDescrRing[pi][rxp].bits, DMA_OWN);
 
 	/* Update stats */
-	stats->rx_packets++;
-	stats->rx_bytes += len;
+	dev->stats.rx_packets++;
+	dev->stats.rx_bytes += len;
 
 	/* Push upstream */
 	dbg(DBG_RX, "Pushing the frame up the stack\n");
@@ -900,7 +898,7 @@ fst_rx_dma_complete(struct fst_card_info *card, struct fst_port_info *port,
 	rx_status = netif_rx(skb);
 	fst_process_rx_status(rx_status, port_to_dev(port)->name);
 	if (rx_status == NET_RX_DROP)
-		stats->rx_dropped++;
+		dev->stats.rx_dropped++;
 	dev->last_rx = jiffies;
 }
 
@@ -1163,29 +1161,28 @@ fst_log_rx_error(struct fst_card_info *card, struct fst_port_info *port,
 		 unsigned char dmabits, int rxp, unsigned short len)
 {
 	struct net_device *dev = port_to_dev(port);
-	struct net_device_stats *stats = hdlc_stats(dev);
 
-	/* 
+	/*
 	 * Increment the appropriate error counter
 	 */
-	stats->rx_errors++;
+	dev->stats.rx_errors++;
 	if (dmabits & RX_OFLO) {
-		stats->rx_fifo_errors++;
+		dev->stats.rx_fifo_errors++;
 		dbg(DBG_ASS, "Rx fifo error on card %d port %d buffer %d\n",
 		    card->card_no, port->index, rxp);
 	}
 	if (dmabits & RX_CRC) {
-		stats->rx_crc_errors++;
+		dev->stats.rx_crc_errors++;
 		dbg(DBG_ASS, "Rx crc error on card %d port %d\n",
 		    card->card_no, port->index);
 	}
 	if (dmabits & RX_FRAM) {
-		stats->rx_frame_errors++;
+		dev->stats.rx_frame_errors++;
 		dbg(DBG_ASS, "Rx frame error on card %d port %d\n",
 		    card->card_no, port->index);
 	}
 	if (dmabits == (RX_STP | RX_ENP)) {
-		stats->rx_length_errors++;
+		dev->stats.rx_length_errors++;
 		dbg(DBG_ASS, "Rx length error (%d) on card %d port %d\n",
 		    len, card->card_no, port->index);
 	}
@@ -1242,7 +1239,6 @@ fst_intr_rx(struct fst_card_info *card, struct fst_port_info *port)
 	unsigned short len;
 	struct sk_buff *skb;
 	struct net_device *dev = port_to_dev(port);
-	struct net_device_stats *stats = hdlc_stats(dev);
 
 	/* Check we have a buffer to process */
 	pi = port->index;
@@ -1291,7 +1287,7 @@ fst_intr_rx(struct fst_card_info *card, struct fst_port_info *port)
 	if ((skb = dev_alloc_skb(len)) == NULL) {
 		dbg(DBG_RX, "intr_rx: can't allocate buffer\n");
 
-		stats->rx_dropped++;
+		dev->stats.rx_dropped++;
 
 		/* Return descriptor to card */
 		FST_WRB(card, rxDescrRing[pi][rxp].bits, DMA_OWN);
@@ -1316,8 +1312,8 @@ fst_intr_rx(struct fst_card_info *card, struct fst_port_info *port)
 		FST_WRB(card, rxDescrRing[pi][rxp].bits, DMA_OWN);
 
 		/* Update stats */
-		stats->rx_packets++;
-		stats->rx_bytes += len;
+		dev->stats.rx_packets++;
+		dev->stats.rx_bytes += len;
 
 		/* Push upstream */
 		dbg(DBG_RX, "Pushing frame up the stack\n");
@@ -1327,9 +1323,8 @@ fst_intr_rx(struct fst_card_info *card, struct fst_port_info *port)
 			skb->protocol = hdlc_type_trans(skb, dev);
 		rx_status = netif_rx(skb);
 		fst_process_rx_status(rx_status, port_to_dev(port)->name);
-		if (rx_status == NET_RX_DROP) {
-			stats->rx_dropped++;
-		}
+		if (rx_status == NET_RX_DROP)
+			dev->stats.rx_dropped++;
 		dev->last_rx = jiffies;
 	} else {
 		card->dma_skb_rx = skb;
@@ -1361,7 +1356,6 @@ do_bottom_half_tx(struct fst_card_info *card)
 	struct sk_buff *skb;
 	unsigned long flags;
 	struct net_device *dev;
-	struct net_device_stats *stats;
 
 	/*
 	 *  Find a free buffer for the transmit
@@ -1373,12 +1367,10 @@ do_bottom_half_tx(struct fst_card_info *card)
 		if (!port->run)
 			continue;
 
-                dev = port_to_dev(port);
-                stats = hdlc_stats(dev);
-		while (!
-		       (FST_RDB(card, txDescrRing[pi][port->txpos].bits) &
-			DMA_OWN)
-                       && !(card->dmatx_in_progress)) {
+		dev = port_to_dev(port);
+		while (!(FST_RDB(card, txDescrRing[pi][port->txpos].bits) &
+			 DMA_OWN)
+		       && !(card->dmatx_in_progress)) {
 			/*
 			 * There doesn't seem to be a txdone event per-se
 			 * We seem to have to deduce it, by checking the DMA_OWN
@@ -1422,8 +1414,8 @@ do_bottom_half_tx(struct fst_card_info *card)
 						txDescrRing[pi][port->txpos].
 						bits,
 						DMA_OWN | TX_STP | TX_ENP);
-					stats->tx_packets++;
-					stats->tx_bytes += skb->len;
+					dev->stats.tx_packets++;
+					dev->stats.tx_bytes += skb->len;
 					dev->trans_start = jiffies;
 				} else {
 					/* Or do it through dma */
@@ -1628,8 +1620,8 @@ fst_intr(int dummy, void *dev_id)
 			 * always load up the entire packet for DMA.
 			 */
 			dbg(DBG_TX, "Tx underflow port %d\n", port->index);
-                        hdlc_stats(port_to_dev(port))->tx_errors++;
-                        hdlc_stats(port_to_dev(port))->tx_fifo_errors++;
+			port_to_dev(port)->stats.tx_errors++;
+			port_to_dev(port)->stats.tx_fifo_errors++;
 			dbg(DBG_ASS, "Tx underflow on card %d port %d\n",
 			    card->card_no, port->index);
 			break;
@@ -2292,12 +2284,11 @@ fst_tx_timeout(struct net_device *dev)
 {
 	struct fst_port_info *port;
 	struct fst_card_info *card;
-	struct net_device_stats *stats = hdlc_stats(dev);
 
 	port = dev_to_port(dev);
 	card = port->card;
-	stats->tx_errors++;
-	stats->tx_aborted_errors++;
+	dev->stats.tx_errors++;
+	dev->stats.tx_aborted_errors++;
 	dbg(DBG_ASS, "Tx timeout card %d port %d\n",
 	    card->card_no, port->index);
 	fst_issue_cmd(port, ABORTTX);
@@ -2312,7 +2303,6 @@ fst_start_xmit(struct sk_buff *skb, struct net_device *dev)
 {
 	struct fst_card_info *card;
 	struct fst_port_info *port;
-	struct net_device_stats *stats = hdlc_stats(dev);
 	unsigned long flags;
 	int txq_length;
 
@@ -2323,8 +2313,8 @@ fst_start_xmit(struct sk_buff *skb, struct net_device *dev)
 	/* Drop packet with error if we don't have carrier */
 	if (!netif_carrier_ok(dev)) {
 		dev_kfree_skb(skb);
-		stats->tx_errors++;
-		stats->tx_carrier_errors++;
+		dev->stats.tx_errors++;
+		dev->stats.tx_carrier_errors++;
 		dbg(DBG_ASS,
 		    "Tried to transmit but no carrier on card %d port %d\n",
 		    card->card_no, port->index);
@@ -2336,7 +2326,7 @@ fst_start_xmit(struct sk_buff *skb, struct net_device *dev)
 		dbg(DBG_ASS, "Packet too large %d vs %d\n", skb->len,
 		    LEN_TX_BUFFER);
 		dev_kfree_skb(skb);
-		stats->tx_errors++;
+		dev->stats.tx_errors++;
 		return 0;
 	}
 
@@ -2368,7 +2358,7 @@ fst_start_xmit(struct sk_buff *skb, struct net_device *dev)
 		 * This shouldn't have happened but such is life
 		 */
 		dev_kfree_skb(skb);
-		stats->tx_errors++;
+		dev->stats.tx_errors++;
 		dbg(DBG_ASS, "Tx queue overflow card %d port %d\n",
 		    card->card_no, port->index);
 		return 0;
diff --git a/drivers/net/wan/hd6457x.c b/drivers/net/wan/hd6457x.c
index 8d0a1f2f00e5..591fb45a7c68 100644
--- a/drivers/net/wan/hd6457x.c
+++ b/drivers/net/wan/hd6457x.c
@@ -271,9 +271,9 @@ static inline void sca_msci_intr(port_t *port)
 	sca_out(stat & (ST1_UDRN | ST1_CDCD), msci + ST1, card);
 
 	if (stat & ST1_UDRN) {
-		struct net_device_stats *stats = hdlc_stats(port_to_dev(port));
-		stats->tx_errors++; /* TX Underrun error detected */
-		stats->tx_fifo_errors++;
+		/* TX Underrun error detected */
+		port_to_dev(port)->stats.tx_errors++;
+		port_to_dev(port)->stats.tx_fifo_errors++;
 	}
 
 	if (stat & ST1_CDCD)
@@ -286,7 +286,6 @@ static inline void sca_msci_intr(port_t *port)
 static inline void sca_rx(card_t *card, port_t *port, pkt_desc __iomem *desc, u16 rxin)
 {
 	struct net_device *dev = port_to_dev(port);
-	struct net_device_stats *stats = hdlc_stats(dev);
 	struct sk_buff *skb;
 	u16 len;
 	u32 buff;
@@ -298,7 +297,7 @@ static inline void sca_rx(card_t *card, port_t *port, pkt_desc __iomem *desc, u1
 	len = readw(&desc->len);
 	skb = dev_alloc_skb(len);
 	if (!skb) {
-		stats->rx_dropped++;
+		dev->stats.rx_dropped++;
 		return;
 	}
 
@@ -327,8 +326,8 @@ static inline void sca_rx(card_t *card, port_t *port, pkt_desc __iomem *desc, u1
 	printk(KERN_DEBUG "%s RX(%i):", dev->name, skb->len);
 	debug_frame(skb);
 #endif
-	stats->rx_packets++;
-	stats->rx_bytes += skb->len;
+	dev->stats.rx_packets++;
+	dev->stats.rx_bytes += skb->len;
 	dev->last_rx = jiffies;
 	skb->protocol = hdlc_type_trans(skb, dev);
 	netif_rx(skb);
@@ -339,17 +338,18 @@ static inline void sca_rx(card_t *card, port_t *port, pkt_desc __iomem *desc, u1
 /* Receive DMA interrupt service */
 static inline void sca_rx_intr(port_t *port)
 {
+	struct net_device *dev = port_to_dev(port);
 	u16 dmac = get_dmac_rx(port);
 	card_t *card = port_to_card(port);
 	u8 stat = sca_in(DSR_RX(phy_node(port)), card); /* read DMA Status */
-	struct net_device_stats *stats = hdlc_stats(port_to_dev(port));
 
 	/* Reset DSR status bits */
 	sca_out((stat & (DSR_EOT | DSR_EOM | DSR_BOF | DSR_COF)) | DSR_DWE,
 		DSR_RX(phy_node(port)), card);
 
 	if (stat & DSR_BOF)
-		stats->rx_over_errors++; /* Dropped one or more frames */
+		/* Dropped one or more frames */
+		dev->stats.rx_over_errors++;
 
 	while (1) {
 		u32 desc_off = desc_offset(port, port->rxin, 0);
@@ -364,12 +364,14 @@ static inline void sca_rx_intr(port_t *port)
 		if (!(stat & ST_RX_EOM))
 			port->rxpart = 1; /* partial frame received */
 		else if ((stat & ST_ERROR_MASK) || port->rxpart) {
-			stats->rx_errors++;
-			if (stat & ST_RX_OVERRUN) stats->rx_fifo_errors++;
+			dev->stats.rx_errors++;
+			if (stat & ST_RX_OVERRUN)
+				dev->stats.rx_fifo_errors++;
 			else if ((stat & (ST_RX_SHORT | ST_RX_ABORT |
 					  ST_RX_RESBIT)) || port->rxpart)
-				stats->rx_frame_errors++;
-			else if (stat & ST_RX_CRC) stats->rx_crc_errors++;
+				dev->stats.rx_frame_errors++;
+			else if (stat & ST_RX_CRC)
+				dev->stats.rx_crc_errors++;
 			if (stat & ST_RX_EOM)
 				port->rxpart = 0; /* received last fragment */
 		} else
@@ -390,7 +392,6 @@ static inline void sca_rx_intr(port_t *port)
 static inline void sca_tx_intr(port_t *port)
 {
 	struct net_device *dev = port_to_dev(port);
-	struct net_device_stats *stats = hdlc_stats(dev);
 	u16 dmac = get_dmac_tx(port);
 	card_t* card = port_to_card(port);
 	u8 stat;
@@ -412,8 +413,8 @@ static inline void sca_tx_intr(port_t *port)
 			break;	/* Transmitter is/will_be sending this frame */
 
 		desc = desc_address(port, port->txlast, 1);
-		stats->tx_packets++;
-		stats->tx_bytes += readw(&desc->len);
+		dev->stats.tx_packets++;
+		dev->stats.tx_bytes += readw(&desc->len);
 		writeb(0, &desc->stat);	/* Free descriptor */
 		port->txlast = next_desc(port, port->txlast, 1);
 	}
diff --git a/drivers/net/wan/hdlc.c b/drivers/net/wan/hdlc.c
index 7f984895b0d5..e3a536477c7e 100644
--- a/drivers/net/wan/hdlc.c
+++ b/drivers/net/wan/hdlc.c
@@ -57,7 +57,7 @@ static int hdlc_change_mtu(struct net_device *dev, int new_mtu)
 
 static struct net_device_stats *hdlc_get_stats(struct net_device *dev)
 {
-	return hdlc_stats(dev);
+	return &dev->stats;
 }
 
 
diff --git a/drivers/net/wan/hdlc_cisco.c b/drivers/net/wan/hdlc_cisco.c
index 762d21c1c703..849819c2552d 100644
--- a/drivers/net/wan/hdlc_cisco.c
+++ b/drivers/net/wan/hdlc_cisco.c
@@ -252,8 +252,8 @@ static int cisco_rx(struct sk_buff *skb)
 	dev_kfree_skb_any(skb);
 	return NET_RX_DROP;
 
- rx_error:
-	dev_to_hdlc(dev)->stats.rx_errors++; /* Mark error */
+rx_error:
+	dev->stats.rx_errors++; /* Mark error */
 	dev_kfree_skb_any(skb);
 	return NET_RX_DROP;
 }
diff --git a/drivers/net/wan/hdlc_fr.c b/drivers/net/wan/hdlc_fr.c
index 520bb0b1a9a2..cb1b415053eb 100644
--- a/drivers/net/wan/hdlc_fr.c
+++ b/drivers/net/wan/hdlc_fr.c
@@ -136,7 +136,6 @@ typedef struct pvc_device_struct {
 }pvc_device;
 
 struct pvc_desc {
-	struct net_device_stats stats;
 	pvc_device *pvc;
 };
 
@@ -184,11 +183,6 @@ static inline struct pvc_desc* pvcdev_to_desc(struct net_device *dev)
 	return dev->priv;
 }
 
-static inline struct net_device_stats* pvc_get_stats(struct net_device *dev)
-{
-	return &pvcdev_to_desc(dev)->stats;
-}
-
 static inline pvc_device* find_pvc(hdlc_device *hdlc, u16 dlci)
 {
 	pvc_device *pvc = state(hdlc)->first_pvc;
@@ -425,7 +419,6 @@ static int pvc_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
 static int pvc_xmit(struct sk_buff *skb, struct net_device *dev)
 {
 	pvc_device *pvc = pvcdev_to_desc(dev)->pvc;
-	struct net_device_stats *stats = pvc_get_stats(dev);
 
 	if (pvc->state.active) {
 		if (dev->type == ARPHRD_ETHER) {
@@ -435,7 +428,7 @@ static int pvc_xmit(struct sk_buff *skb, struct net_device *dev)
 				if (skb_tailroom(skb) < pad)
 					if (pskb_expand_head(skb, 0, pad,
 							     GFP_ATOMIC)) {
-						stats->tx_dropped++;
+						dev->stats.tx_dropped++;
 						dev_kfree_skb(skb);
 						return 0;
 					}
@@ -445,17 +438,17 @@ static int pvc_xmit(struct sk_buff *skb, struct net_device *dev)
 			skb->protocol = __constant_htons(ETH_P_802_3);
 		}
 		if (!fr_hard_header(&skb, pvc->dlci)) {
-			stats->tx_bytes += skb->len;
-			stats->tx_packets++;
+			dev->stats.tx_bytes += skb->len;
+			dev->stats.tx_packets++;
 			if (pvc->state.fecn) /* TX Congestion counter */
-				stats->tx_compressed++;
+				dev->stats.tx_compressed++;
 			skb->dev = pvc->frad;
 			dev_queue_xmit(skb);
 			return 0;
 		}
 	}
 
-	stats->tx_dropped++;
+	dev->stats.tx_dropped++;
 	dev_kfree_skb(skb);
 	return 0;
 }
@@ -955,7 +948,7 @@ static int fr_rx(struct sk_buff *skb)
 
 
 	if ((skb = skb_share_check(skb, GFP_ATOMIC)) == NULL) {
-		dev_to_hdlc(frad)->stats.rx_dropped++;
+		frad->stats.rx_dropped++;
 		return NET_RX_DROP;
 	}
 
@@ -1003,11 +996,10 @@ static int fr_rx(struct sk_buff *skb)
 	}
 
 	if (dev) {
-		struct net_device_stats *stats = pvc_get_stats(dev);
-		stats->rx_packets++; /* PVC traffic */
-		stats->rx_bytes += skb->len;
+		dev->stats.rx_packets++; /* PVC traffic */
+		dev->stats.rx_bytes += skb->len;
 		if (pvc->state.becn)
-			stats->rx_compressed++;
+			dev->stats.rx_compressed++;
 		netif_rx(skb);
 		return NET_RX_SUCCESS;
 	} else {
@@ -1016,7 +1008,7 @@ static int fr_rx(struct sk_buff *skb)
 	}
 
  rx_error:
-	dev_to_hdlc(frad)->stats.rx_errors++; /* Mark error */
+	frad->stats.rx_errors++; /* Mark error */
 	dev_kfree_skb_any(skb);
 	return NET_RX_DROP;
 }
@@ -1122,7 +1114,6 @@ static int fr_add_pvc(struct net_device *frad, unsigned int dlci, int type)
 		dlci_to_q922(dev->broadcast, dlci);
 	}
 	dev->hard_start_xmit = pvc_xmit;
-	dev->get_stats = pvc_get_stats;
 	dev->open = pvc_open;
 	dev->stop = pvc_close;
 	dev->do_ioctl = pvc_ioctl;
diff --git a/drivers/net/wan/hdlc_raw_eth.c b/drivers/net/wan/hdlc_raw_eth.c
index d20c685f6711..26dee600506f 100644
--- a/drivers/net/wan/hdlc_raw_eth.c
+++ b/drivers/net/wan/hdlc_raw_eth.c
@@ -33,7 +33,7 @@ static int eth_tx(struct sk_buff *skb, struct net_device *dev)
 		int len = skb->len;
 		if (skb_tailroom(skb) < pad)
 			if (pskb_expand_head(skb, 0, pad, GFP_ATOMIC)) {
-				hdlc_stats(dev)->tx_dropped++;
+				dev->stats.tx_dropped++;
 				dev_kfree_skb(skb);
 				return 0;
 			}
diff --git a/drivers/net/wan/hdlc_x25.c b/drivers/net/wan/hdlc_x25.c
index c15cc11e399b..e808720030ef 100644
--- a/drivers/net/wan/hdlc_x25.c
+++ b/drivers/net/wan/hdlc_x25.c
@@ -164,17 +164,15 @@ static void x25_close(struct net_device *dev)
 
 static int x25_rx(struct sk_buff *skb)
 {
-	struct hdlc_device *hdlc = dev_to_hdlc(skb->dev);
-
 	if ((skb = skb_share_check(skb, GFP_ATOMIC)) == NULL) {
-		hdlc->stats.rx_dropped++;
+		skb->dev->stats.rx_dropped++;
 		return NET_RX_DROP;
 	}
 
 	if (lapb_data_received(skb->dev, skb) == LAPB_OK)
 		return NET_RX_SUCCESS;
 
-	hdlc->stats.rx_errors++;
+	skb->dev->stats.rx_errors++;
 	dev_kfree_skb_any(skb);
 	return NET_RX_DROP;
 }
diff --git a/drivers/net/wan/pc300_drv.c b/drivers/net/wan/pc300_drv.c
index 57914fbd41d3..334170527755 100644
--- a/drivers/net/wan/pc300_drv.c
+++ b/drivers/net/wan/pc300_drv.c
@@ -285,7 +285,6 @@ static void rx_dma_buf_init(pc300_t *, int);
 static void tx_dma_buf_check(pc300_t *, int);
 static void rx_dma_buf_check(pc300_t *, int);
 static irqreturn_t cpc_intr(int, void *);
-static struct net_device_stats *cpc_get_stats(struct net_device *);
 static int clock_rate_calc(uclong, uclong, int *);
 static uclong detect_ram(pc300_t *);
 static void plx_init(pc300_t *);
@@ -1775,13 +1774,12 @@ static void cpc_tx_timeout(struct net_device *dev)
 	pc300dev_t *d = (pc300dev_t *) dev->priv;
 	pc300ch_t *chan = (pc300ch_t *) d->chan;
 	pc300_t *card = (pc300_t *) chan->card;
-	struct net_device_stats *stats = hdlc_stats(dev);
 	int ch = chan->channel;
 	unsigned long flags;
 	ucchar ilar;
 
-	stats->tx_errors++;
-	stats->tx_aborted_errors++;
+	dev->stats.tx_errors++;
+	dev->stats.tx_aborted_errors++;
 	CPC_LOCK(card, flags);
 	if ((ilar = cpc_readb(card->hw.scabase + ILAR)) != 0) {
 		printk("%s: ILAR=0x%x\n", dev->name, ilar);
@@ -1803,7 +1801,6 @@ static int cpc_queue_xmit(struct sk_buff *skb, struct net_device *dev)
 	pc300dev_t *d = (pc300dev_t *) dev->priv;
 	pc300ch_t *chan = (pc300ch_t *) d->chan;
 	pc300_t *card = (pc300_t *) chan->card;
-	struct net_device_stats *stats = hdlc_stats(dev);
 	int ch = chan->channel;
 	unsigned long flags;
 #ifdef PC300_DEBUG_TX
@@ -1817,13 +1814,13 @@ static int cpc_queue_xmit(struct sk_buff *skb, struct net_device *dev)
 	} else if (!netif_carrier_ok(dev)) {
 		/* DCD must be OFF: drop packet */
 		dev_kfree_skb(skb);
-		stats->tx_errors++;
-		stats->tx_carrier_errors++;
+		dev->stats.tx_errors++;
+		dev->stats.tx_carrier_errors++;
 		return 0;
 	} else if (cpc_readb(card->hw.scabase + M_REG(ST3, ch)) & ST3_DCD) {
 		printk("%s: DCD is OFF. Going administrative down.\n", dev->name);
-		stats->tx_errors++;
-		stats->tx_carrier_errors++;
+		dev->stats.tx_errors++;
+		dev->stats.tx_carrier_errors++;
 		dev_kfree_skb(skb);
 		netif_carrier_off(dev);
 		CPC_LOCK(card, flags);
@@ -1843,8 +1840,8 @@ static int cpc_queue_xmit(struct sk_buff *skb, struct net_device *dev)
 //		printk("%s: write error. Dropping TX packet.\n", dev->name);
 		netif_stop_queue(dev);
 		dev_kfree_skb(skb);
-		stats->tx_errors++;
-		stats->tx_dropped++;
+		dev->stats.tx_errors++;
+		dev->stats.tx_dropped++;
 		return 0;
 	}
 #ifdef PC300_DEBUG_TX
@@ -1886,7 +1883,6 @@ static void cpc_net_rx(struct net_device *dev)
 	pc300dev_t *d = (pc300dev_t *) dev->priv;
 	pc300ch_t *chan = (pc300ch_t *) d->chan;
 	pc300_t *card = (pc300_t *) chan->card;
-	struct net_device_stats *stats = hdlc_stats(dev);
 	int ch = chan->channel;
 #ifdef PC300_DEBUG_RX
 	int i;
@@ -1922,24 +1918,24 @@ static void cpc_net_rx(struct net_device *dev)
 #endif
 			if ((skb == NULL) && (rxb > 0)) {
 				/* rxb > dev->mtu */
-				stats->rx_errors++;
-				stats->rx_length_errors++;
+				dev->stats.rx_errors++;
+				dev->stats.rx_length_errors++;
 				continue;
 			}
 
 			if (rxb < 0) {	/* Invalid frame */
 				rxb = -rxb;
 				if (rxb & DST_OVR) {
-					stats->rx_errors++;
-					stats->rx_fifo_errors++;
+					dev->stats.rx_errors++;
+					dev->stats.rx_fifo_errors++;
 				}
 				if (rxb & DST_CRC) {
-					stats->rx_errors++;
-					stats->rx_crc_errors++;
+					dev->stats.rx_errors++;
+					dev->stats.rx_crc_errors++;
 				}
 				if (rxb & (DST_RBIT | DST_SHRT | DST_ABT)) {
-					stats->rx_errors++;
-					stats->rx_frame_errors++;
+					dev->stats.rx_errors++;
+					dev->stats.rx_frame_errors++;
 				}
 			}
 			if (skb) {
@@ -1948,7 +1944,7 @@ static void cpc_net_rx(struct net_device *dev)
 			continue;
 		}
 
-		stats->rx_bytes += rxb;
+		dev->stats.rx_bytes += rxb;
 
 #ifdef PC300_DEBUG_RX
 		printk("%s R:", dev->name);
@@ -1959,7 +1955,7 @@ static void cpc_net_rx(struct net_device *dev)
 		if (d->trace_on) {
 			cpc_trace(dev, skb, 'R');
 		}
-		stats->rx_packets++;
+		dev->stats.rx_packets++;
 		skb->protocol = hdlc_type_trans(skb, dev);
 		netif_rx(skb);
 	}
@@ -1974,16 +1970,15 @@ static void sca_tx_intr(pc300dev_t *dev)
 	pc300_t *card = (pc300_t *)chan->card; 
 	int ch = chan->channel; 
 	volatile pcsca_bd_t __iomem * ptdescr; 
-	struct net_device_stats *stats = hdlc_stats(dev->dev);
 
     /* Clean up descriptors from previous transmission */
 	ptdescr = (card->hw.rambase +
 						TX_BD_ADDR(ch,chan->tx_first_bd));
-	while ((cpc_readl(card->hw.scabase + DTX_REG(CDAL,ch)) != 
-							TX_BD_ADDR(ch,chan->tx_first_bd)) && 
-			(cpc_readb(&ptdescr->status) & DST_OSB)) {
-		stats->tx_packets++;
-		stats->tx_bytes += cpc_readw(&ptdescr->len);
+	while ((cpc_readl(card->hw.scabase + DTX_REG(CDAL,ch)) !=
+		TX_BD_ADDR(ch,chan->tx_first_bd)) &&
+	       (cpc_readb(&ptdescr->status) & DST_OSB)) {
+		dev->dev->stats.tx_packets++;
+		dev->dev->stats.tx_bytes += cpc_readw(&ptdescr->len);
 		cpc_writeb(&ptdescr->status, DST_OSB);
 		cpc_writew(&ptdescr->len, 0);
 		chan->nfree_tx_bd++;
@@ -2048,8 +2043,8 @@ static void sca_intr(pc300_t * card)
 							}
 							cpc_net_rx(dev);
 							/* Discard invalid frames */
-							hdlc_stats(dev)->rx_errors++;
-							hdlc_stats(dev)->rx_over_errors++;
+							dev->stats.rx_errors++;
+							dev->stats.rx_over_errors++;
 							chan->rx_first_bd = 0;
 							chan->rx_last_bd = N_DMA_RX_BUF - 1;
 							rx_dma_start(card, ch);
@@ -2115,8 +2110,8 @@ static void sca_intr(pc300_t * card)
 										   card->hw.cpld_reg2) &
 								   ~ (CPLD_REG2_FALC_LED1 << (2 * ch)));
 						}
-						hdlc_stats(dev)->tx_errors++;
-						hdlc_stats(dev)->tx_fifo_errors++;
+						dev->stats.tx_errors++;
+						dev->stats.tx_fifo_errors++;
 						sca_tx_intr(d);
 					}
 				}
@@ -2604,7 +2599,7 @@ static int cpc_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
 		case SIOCGPC300UTILSTATS:
 			{
 				if (!arg) {	/* clear statistics */
-					memset(hdlc_stats(dev), 0, sizeof(struct net_device_stats));
+					memset(&dev->stats, 0, sizeof(dev->stats));
 					if (card->hw.type == PC300_TE) {
 						memset(&chan->falc, 0, sizeof(falc_t));
 					}
@@ -2615,8 +2610,8 @@ static int cpc_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
 					pc300stats.hw_type = card->hw.type;
 					pc300stats.line_on = card->chan[ch].d.line_on;
 					pc300stats.line_off = card->chan[ch].d.line_off;
-					memcpy(&pc300stats.gen_stats, hdlc_stats(dev),
-					       sizeof(struct net_device_stats));
+					memcpy(&pc300stats.gen_stats, &dev->stats,
+					       sizeof(dev->stats));
 					if (card->hw.type == PC300_TE)
 						memcpy(&pc300stats.te_stats,&chan->falc,sizeof(falc_t));
 				    	if (copy_to_user(arg, &pc300stats, sizeof(pc300stats_t)))
@@ -2823,11 +2818,6 @@ static int cpc_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
 	}
 }
 
-static struct net_device_stats *cpc_get_stats(struct net_device *dev)
-{
-	return hdlc_stats(dev);
-}
-
 static int clock_rate_calc(uclong rate, uclong clock, int *br_io)
 {
 	int br, tc;
@@ -3394,7 +3384,6 @@ static void cpc_init_card(pc300_t * card)
 		dev->stop = cpc_close;
 		dev->tx_timeout = cpc_tx_timeout;
 		dev->watchdog_timeo = PC300_TX_TIMEOUT;
-		dev->get_stats = cpc_get_stats;
 		dev->set_multicast_list = NULL;
 		dev->set_mac_address = NULL;
 		dev->change_mtu = cpc_change_mtu;
diff --git a/drivers/net/wan/pc300_tty.c b/drivers/net/wan/pc300_tty.c
index e03eef2f2282..eae94ab6b818 100644
--- a/drivers/net/wan/pc300_tty.c
+++ b/drivers/net/wan/pc300_tty.c
@@ -458,7 +458,7 @@ static int cpc_tty_write(struct tty_struct *tty, const unsigned char *buf, int c
 	CPC_TTY_DBG("%s: cpc_tty_write data len=%i\n",cpc_tty->name,count);
 	
 	pc300chan = (pc300ch_t *)((pc300dev_t*)cpc_tty->pc300dev)->chan; 
-	stats = hdlc_stats(((pc300dev_t*)cpc_tty->pc300dev)->dev);
+	stats = &cpc_tty->pc300dev->dev->stats;
 	card = (pc300_t *) pc300chan->card;
 	ch = pc300chan->channel; 
 
@@ -743,7 +743,7 @@ void cpc_tty_receive(pc300dev_t *pc300dev)
 	pc300_t *card = (pc300_t *)pc300chan->card; 
 	int ch = pc300chan->channel; 
 	volatile pcsca_bd_t  __iomem * ptdescr; 
-	struct net_device_stats *stats = hdlc_stats(pc300dev->dev);
+	struct net_device_stats *stats = &pc300dev->dev->stats;
 	int rx_len, rx_aux; 
 	volatile unsigned char status; 
 	unsigned short first_bd = pc300chan->rx_first_bd;
@@ -917,7 +917,7 @@ static int cpc_tty_send_to_card(pc300dev_t *dev,void* buf, int len)
 	pc300ch_t *chan = (pc300ch_t *)dev->chan; 
 	pc300_t *card = (pc300_t *)chan->card; 
 	int ch = chan->channel; 
-	struct net_device_stats *stats = hdlc_stats(dev->dev);
+	struct net_device_stats *stats = &dev->dev->stats;
 	unsigned long flags; 
 	volatile pcsca_bd_t __iomem *ptdescr; 
 	int i, nchar;
diff --git a/drivers/net/wan/wanxl.c b/drivers/net/wan/wanxl.c
index d4aab8a28b61..a8a5ca0ee6c2 100644
--- a/drivers/net/wan/wanxl.c
+++ b/drivers/net/wan/wanxl.c
@@ -161,7 +161,6 @@ static inline void wanxl_cable_intr(port_t *port)
 static inline void wanxl_tx_intr(port_t *port)
 {
 	struct net_device *dev = port->dev;
-	struct net_device_stats *stats = hdlc_stats(dev);
 	while (1) {
                 desc_t *desc = &get_status(port)->tx_descs[port->tx_in];
 		struct sk_buff *skb = port->tx_skbs[port->tx_in];
@@ -173,13 +172,13 @@ static inline void wanxl_tx_intr(port_t *port)
 			return;
 
 		case PACKET_UNDERRUN:
-			stats->tx_errors++;
-			stats->tx_fifo_errors++;
+			dev->stats.tx_errors++;
+			dev->stats.tx_fifo_errors++;
 			break;
 
 		default:
-			stats->tx_packets++;
-			stats->tx_bytes += skb->len;
+			dev->stats.tx_packets++;
+			dev->stats.tx_bytes += skb->len;
 		}
                 desc->stat = PACKET_EMPTY; /* Free descriptor */
 		pci_unmap_single(port->card->pdev, desc->address, skb->len,
@@ -205,10 +204,9 @@ static inline void wanxl_rx_intr(card_t *card)
 			port_t *port = &card->ports[desc->stat &
 						    PACKET_PORT_MASK];
 			struct net_device *dev = port->dev;
-			struct net_device_stats *stats = hdlc_stats(dev);
 
 			if (!skb)
-				stats->rx_dropped++;
+				dev->stats.rx_dropped++;
 			else {
 				pci_unmap_single(card->pdev, desc->address,
 						 BUFFER_LENGTH,
@@ -220,8 +218,8 @@ static inline void wanxl_rx_intr(card_t *card)
 				       skb->len);
 				debug_frame(skb);
 #endif
-				stats->rx_packets++;
-				stats->rx_bytes += skb->len;
+				dev->stats.rx_packets++;
+				dev->stats.rx_bytes += skb->len;
 				dev->last_rx = jiffies;
 				skb->protocol = hdlc_type_trans(skb, dev);
 				netif_rx(skb);
@@ -468,13 +466,13 @@ static int wanxl_close(struct net_device *dev)
 
 static struct net_device_stats *wanxl_get_stats(struct net_device *dev)
 {
-	struct net_device_stats *stats = hdlc_stats(dev);
 	port_t *port = dev_to_port(dev);
 
-	stats->rx_over_errors = get_status(port)->rx_overruns;
-	stats->rx_frame_errors = get_status(port)->rx_frame_errors;
-	stats->rx_errors = stats->rx_over_errors + stats->rx_frame_errors;
-        return stats;
+	dev->stats.rx_over_errors = get_status(port)->rx_overruns;
+	dev->stats.rx_frame_errors = get_status(port)->rx_frame_errors;
+	dev->stats.rx_errors = dev->stats.rx_over_errors +
+		dev->stats.rx_frame_errors;
+	return &dev->stats;
 }
 
 
diff --git a/include/linux/hdlc.h b/include/linux/hdlc.h
index 6115545a5b9c..c59769693bee 100644
--- a/include/linux/hdlc.h
+++ b/include/linux/hdlc.h
@@ -45,7 +45,6 @@ struct hdlc_proto {
 
 /* Pointed to by dev->priv */
 typedef struct hdlc_device {
-	struct net_device_stats stats;
 	/* used by HDLC layer to take control over HDLC device from hw driver*/
 	int (*attach)(struct net_device *dev,
 		      unsigned short encoding, unsigned short parity);
@@ -109,12 +108,6 @@ int attach_hdlc_protocol(struct net_device *dev, struct hdlc_proto *proto,
 /* May be used by hardware driver to gain control over HDLC device */
 void detach_hdlc_protocol(struct net_device *dev);
 
-static __inline__ struct net_device_stats *hdlc_stats(struct net_device *dev)
-{
-	return &dev_to_hdlc(dev)->stats;
-}
-
-
 static __inline__ __be16 hdlc_type_trans(struct sk_buff *skb,
 					 struct net_device *dev)
 {
-- 
cgit v1.2.3


From 76e6ebfb40a2455c18234dcb0f9df37533215461 Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Sat, 5 Jul 2008 19:00:44 -0700
Subject: netns: add namespace parameter to rt_cache_flush

Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/route.h     |  2 +-
 net/ipv4/arp.c          |  2 +-
 net/ipv4/devinet.c      |  8 +++++---
 net/ipv4/fib_frontend.c | 17 +++++++++--------
 net/ipv4/fib_hash.c     |  6 +++---
 net/ipv4/fib_rules.c    |  2 +-
 net/ipv4/fib_trie.c     |  6 +++---
 net/ipv4/route.c        |  8 ++++----
 8 files changed, 27 insertions(+), 24 deletions(-)

(limited to 'include')

diff --git a/include/net/route.h b/include/net/route.h
index fc836ff824cc..3140cc500854 100644
--- a/include/net/route.h
+++ b/include/net/route.h
@@ -111,7 +111,7 @@ struct in_device;
 extern int		ip_rt_init(void);
 extern void		ip_rt_redirect(__be32 old_gw, __be32 dst, __be32 new_gw,
 				       __be32 src, struct net_device *dev);
-extern void		rt_cache_flush(int how);
+extern void		rt_cache_flush(struct net *net, int how);
 extern int		__ip_route_output_key(struct net *, struct rtable **, const struct flowi *flp);
 extern int		ip_route_output_key(struct net *, struct rtable **, struct flowi *flp);
 extern int		ip_route_output_flow(struct net *, struct rtable **rp, struct flowi *flp, struct sock *sk, int flags);
diff --git a/net/ipv4/arp.c b/net/ipv4/arp.c
index 20c515a1be28..29df75a6bcc7 100644
--- a/net/ipv4/arp.c
+++ b/net/ipv4/arp.c
@@ -1197,7 +1197,7 @@ static int arp_netdev_event(struct notifier_block *this, unsigned long event, vo
 	switch (event) {
 	case NETDEV_CHANGEADDR:
 		neigh_changeaddr(&arp_tbl, dev);
-		rt_cache_flush(0);
+		rt_cache_flush(dev_net(dev), 0);
 		break;
 	default:
 		break;
diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c
index 9de2514946ca..2e667e2f90df 100644
--- a/net/ipv4/devinet.c
+++ b/net/ipv4/devinet.c
@@ -1348,7 +1348,7 @@ static int devinet_sysctl_forward(ctl_table *ctl, int write,
 				dev_disable_lro(idev->dev);
 			}
 			rtnl_unlock();
-			rt_cache_flush(0);
+			rt_cache_flush(net, 0);
 		}
 	}
 
@@ -1362,9 +1362,10 @@ int ipv4_doint_and_flush(ctl_table *ctl, int write,
 	int *valp = ctl->data;
 	int val = *valp;
 	int ret = proc_dointvec(ctl, write, filp, buffer, lenp, ppos);
+	struct net *net = ctl->extra2;
 
 	if (write && *valp != val)
-		rt_cache_flush(0);
+		rt_cache_flush(net, 0);
 
 	return ret;
 }
@@ -1375,9 +1376,10 @@ int ipv4_doint_and_flush_strategy(ctl_table *table, int __user *name, int nlen,
 {
 	int ret = devinet_conf_sysctl(table, name, nlen, oldval, oldlenp,
 				      newval, newlen);
+	struct net *net = table->extra2;
 
 	if (ret == 1)
-		rt_cache_flush(0);
+		rt_cache_flush(net, 0);
 
 	return ret;
 }
diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c
index 5ad01d63f83b..65c1503f8cc8 100644
--- a/net/ipv4/fib_frontend.c
+++ b/net/ipv4/fib_frontend.c
@@ -144,7 +144,7 @@ static void fib_flush(struct net *net)
 	}
 
 	if (flushed)
-		rt_cache_flush(-1);
+		rt_cache_flush(net, -1);
 }
 
 /*
@@ -897,21 +897,22 @@ static void fib_disable_ip(struct net_device *dev, int force)
 {
 	if (fib_sync_down_dev(dev, force))
 		fib_flush(dev_net(dev));
-	rt_cache_flush(0);
+	rt_cache_flush(dev_net(dev), 0);
 	arp_ifdown(dev);
 }
 
 static int fib_inetaddr_event(struct notifier_block *this, unsigned long event, void *ptr)
 {
 	struct in_ifaddr *ifa = (struct in_ifaddr*)ptr;
+	struct net_device *dev = ifa->ifa_dev->dev;
 
 	switch (event) {
 	case NETDEV_UP:
 		fib_add_ifaddr(ifa);
 #ifdef CONFIG_IP_ROUTE_MULTIPATH
-		fib_sync_up(ifa->ifa_dev->dev);
+		fib_sync_up(dev);
 #endif
-		rt_cache_flush(-1);
+		rt_cache_flush(dev_net(dev), -1);
 		break;
 	case NETDEV_DOWN:
 		fib_del_ifaddr(ifa);
@@ -919,9 +920,9 @@ static int fib_inetaddr_event(struct notifier_block *this, unsigned long event,
 			/* Last address was deleted from this interface.
 			   Disable IP.
 			 */
-			fib_disable_ip(ifa->ifa_dev->dev, 1);
+			fib_disable_ip(dev, 1);
 		} else {
-			rt_cache_flush(-1);
+			rt_cache_flush(dev_net(dev), -1);
 		}
 		break;
 	}
@@ -949,14 +950,14 @@ static int fib_netdev_event(struct notifier_block *this, unsigned long event, vo
 #ifdef CONFIG_IP_ROUTE_MULTIPATH
 		fib_sync_up(dev);
 #endif
-		rt_cache_flush(-1);
+		rt_cache_flush(dev_net(dev), -1);
 		break;
 	case NETDEV_DOWN:
 		fib_disable_ip(dev, 0);
 		break;
 	case NETDEV_CHANGEMTU:
 	case NETDEV_CHANGE:
-		rt_cache_flush(0);
+		rt_cache_flush(dev_net(dev), 0);
 		break;
 	}
 	return NOTIFY_DONE;
diff --git a/net/ipv4/fib_hash.c b/net/ipv4/fib_hash.c
index eeec4bf982b8..c8cac6c7f881 100644
--- a/net/ipv4/fib_hash.c
+++ b/net/ipv4/fib_hash.c
@@ -472,7 +472,7 @@ static int fn_hash_insert(struct fib_table *tb, struct fib_config *cfg)
 
 			fib_release_info(fi_drop);
 			if (state & FA_S_ACCESSED)
-				rt_cache_flush(-1);
+				rt_cache_flush(cfg->fc_nlinfo.nl_net, -1);
 			rtmsg_fib(RTM_NEWROUTE, key, fa, cfg->fc_dst_len, tb->tb_id,
 				  &cfg->fc_nlinfo, NLM_F_REPLACE);
 			return 0;
@@ -532,7 +532,7 @@ static int fn_hash_insert(struct fib_table *tb, struct fib_config *cfg)
 
 	if (new_f)
 		fz->fz_nent++;
-	rt_cache_flush(-1);
+	rt_cache_flush(cfg->fc_nlinfo.nl_net, -1);
 
 	rtmsg_fib(RTM_NEWROUTE, key, new_fa, cfg->fc_dst_len, tb->tb_id,
 		  &cfg->fc_nlinfo, 0);
@@ -614,7 +614,7 @@ static int fn_hash_delete(struct fib_table *tb, struct fib_config *cfg)
 		write_unlock_bh(&fib_hash_lock);
 
 		if (fa->fa_state & FA_S_ACCESSED)
-			rt_cache_flush(-1);
+			rt_cache_flush(cfg->fc_nlinfo.nl_net, -1);
 		fn_free_alias(fa, f);
 		if (kill_fn) {
 			fn_free_node(f);
diff --git a/net/ipv4/fib_rules.c b/net/ipv4/fib_rules.c
index 1fb56876be54..bc05de413087 100644
--- a/net/ipv4/fib_rules.c
+++ b/net/ipv4/fib_rules.c
@@ -260,7 +260,7 @@ static size_t fib4_rule_nlmsg_payload(struct fib_rule *rule)
 
 static void fib4_rule_flush_cache(void)
 {
-	rt_cache_flush(-1);
+	rt_cache_flush(&init_net, -1);
 }
 
 static struct fib_rules_ops fib4_rules_ops_template = {
diff --git a/net/ipv4/fib_trie.c b/net/ipv4/fib_trie.c
index 394db9c941a1..d16ae4623be6 100644
--- a/net/ipv4/fib_trie.c
+++ b/net/ipv4/fib_trie.c
@@ -1271,7 +1271,7 @@ static int fn_trie_insert(struct fib_table *tb, struct fib_config *cfg)
 
 			fib_release_info(fi_drop);
 			if (state & FA_S_ACCESSED)
-				rt_cache_flush(-1);
+				rt_cache_flush(cfg->fc_nlinfo.nl_net, -1);
 			rtmsg_fib(RTM_NEWROUTE, htonl(key), new_fa, plen,
 				tb->tb_id, &cfg->fc_nlinfo, NLM_F_REPLACE);
 
@@ -1316,7 +1316,7 @@ static int fn_trie_insert(struct fib_table *tb, struct fib_config *cfg)
 	list_add_tail_rcu(&new_fa->fa_list,
 			  (fa ? &fa->fa_list : fa_head));
 
-	rt_cache_flush(-1);
+	rt_cache_flush(cfg->fc_nlinfo.nl_net, -1);
 	rtmsg_fib(RTM_NEWROUTE, htonl(key), new_fa, plen, tb->tb_id,
 		  &cfg->fc_nlinfo, 0);
 succeeded:
@@ -1664,7 +1664,7 @@ static int fn_trie_delete(struct fib_table *tb, struct fib_config *cfg)
 		trie_leaf_remove(t, l);
 
 	if (fa->fa_state & FA_S_ACCESSED)
-		rt_cache_flush(-1);
+		rt_cache_flush(cfg->fc_nlinfo.nl_net, -1);
 
 	fib_release_info(fa->fa_info);
 	alias_free_mem_rcu(fa);
diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index fe3a02237286..cedc366505bd 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -791,7 +791,7 @@ static void rt_cache_invalidate(void)
  * delay < 0  : invalidate cache (fast : entries will be deleted later)
  * delay >= 0 : invalidate & flush cache (can be long)
  */
-void rt_cache_flush(int delay)
+void rt_cache_flush(struct net *net, int delay)
 {
 	rt_cache_invalidate();
 	if (delay >= 0)
@@ -2825,7 +2825,7 @@ done:
 
 void ip_rt_multicast_event(struct in_device *in_dev)
 {
-	rt_cache_flush(0);
+	rt_cache_flush(dev_net(in_dev->dev), 0);
 }
 
 #ifdef CONFIG_SYSCTL
@@ -2837,7 +2837,7 @@ static int ipv4_sysctl_rtcache_flush(ctl_table *ctl, int write,
 {
 	if (write) {
 		proc_dointvec(ctl, write, filp, buffer, lenp, ppos);
-		rt_cache_flush(flush_delay);
+		rt_cache_flush(&init_net, flush_delay);
 		return 0;
 	}
 
@@ -2857,7 +2857,7 @@ static int ipv4_sysctl_rtcache_flush_strategy(ctl_table *table,
 		return -EINVAL;
 	if (get_user(delay, (int __user *)newval))
 		return -EFAULT;
-	rt_cache_flush(delay);
+	rt_cache_flush(&init_net, delay);
 	return 0;
 }
 
-- 
cgit v1.2.3


From ae299fc051aa68ca6ef1807c37bb92d9b6ff817c Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Sat, 5 Jul 2008 19:01:28 -0700
Subject: net: add fib_rules_ops to flush_cache method

This is required to pass namespace context into rt_cache_flush called from
->flush_cache.

Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/fib_rules.h | 2 +-
 net/core/fib_rules.c    | 2 +-
 net/decnet/dn_rules.c   | 2 +-
 net/ipv4/fib_rules.c    | 4 ++--
 4 files changed, 5 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/include/net/fib_rules.h b/include/net/fib_rules.h
index a5c6ccc5bb19..c2bb5cae6515 100644
--- a/include/net/fib_rules.h
+++ b/include/net/fib_rules.h
@@ -62,7 +62,7 @@ struct fib_rules_ops
 
 	/* Called after modifications to the rules set, must flush
 	 * the route cache if one exists. */
-	void			(*flush_cache)(void);
+	void			(*flush_cache)(struct fib_rules_ops *ops);
 
 	int			nlgroup;
 	const struct nla_policy	*policy;
diff --git a/net/core/fib_rules.c b/net/core/fib_rules.c
index e3e9ab0f74e3..1c2943a119f3 100644
--- a/net/core/fib_rules.c
+++ b/net/core/fib_rules.c
@@ -69,7 +69,7 @@ static void rules_ops_put(struct fib_rules_ops *ops)
 static void flush_route_cache(struct fib_rules_ops *ops)
 {
 	if (ops->flush_cache)
-		ops->flush_cache();
+		ops->flush_cache(ops);
 }
 
 int fib_rules_register(struct fib_rules_ops *ops)
diff --git a/net/decnet/dn_rules.c b/net/decnet/dn_rules.c
index 5b7539b7fe0c..14fbca55e908 100644
--- a/net/decnet/dn_rules.c
+++ b/net/decnet/dn_rules.c
@@ -229,7 +229,7 @@ static u32 dn_fib_rule_default_pref(struct fib_rules_ops *ops)
 	return 0;
 }
 
-static void dn_fib_rule_flush_cache(void)
+static void dn_fib_rule_flush_cache(struct fib_rules_ops *ops)
 {
 	dn_rt_cache_flush(-1);
 }
diff --git a/net/ipv4/fib_rules.c b/net/ipv4/fib_rules.c
index bc05de413087..6080d7120821 100644
--- a/net/ipv4/fib_rules.c
+++ b/net/ipv4/fib_rules.c
@@ -258,9 +258,9 @@ static size_t fib4_rule_nlmsg_payload(struct fib_rule *rule)
 	       + nla_total_size(4); /* flow */
 }
 
-static void fib4_rule_flush_cache(void)
+static void fib4_rule_flush_cache(struct fib_rules_ops *ops)
 {
-	rt_cache_flush(&init_net, -1);
+	rt_cache_flush(ops->fro_net, -1);
 }
 
 static struct fib_rules_ops fib4_rules_ops_template = {
-- 
cgit v1.2.3


From 39a23e75087ce815abbddbd565b9a2e567ac47da Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Sat, 5 Jul 2008 19:02:33 -0700
Subject: netns: register net.ipv4.route.flush in each namespace

Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/netns/ipv4.h |  1 +
 net/ipv4/route.c         | 79 ++++++++++++++++++++++++++++++++++++++++++------
 2 files changed, 70 insertions(+), 10 deletions(-)

(limited to 'include')

diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h
index 6ef90b5fafb3..a29adf1ba0ff 100644
--- a/include/net/netns/ipv4.h
+++ b/include/net/netns/ipv4.h
@@ -18,6 +18,7 @@ struct netns_ipv4 {
 	struct ctl_table_header	*forw_hdr;
 	struct ctl_table_header	*frags_hdr;
 	struct ctl_table_header	*ipv4_hdr;
+	struct ctl_table_header *route_hdr;
 #endif
 	struct ipv4_devconf	*devconf_all;
 	struct ipv4_devconf	*devconf_dflt;
diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index 790de32cd7d4..6fe799de9b96 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -2835,6 +2835,7 @@ static int ipv4_sysctl_rtcache_flush(ctl_table *ctl, int write,
 {
 	if (write) {
 		int flush_delay;
+		struct net *net;
 		static DEFINE_MUTEX(flush_mutex);
 
 		mutex_lock(&flush_mutex);
@@ -2843,7 +2844,8 @@ static int ipv4_sysctl_rtcache_flush(ctl_table *ctl, int write,
 		ctl->data = NULL;
 		mutex_unlock(&flush_mutex);
 
-		rt_cache_flush(&init_net, flush_delay);
+		net = (struct net *)ctl->extra1;
+		rt_cache_flush(net, flush_delay);
 		return 0;
 	}
 
@@ -2859,23 +2861,17 @@ static int ipv4_sysctl_rtcache_flush_strategy(ctl_table *table,
 						size_t newlen)
 {
 	int delay;
+	struct net *net;
 	if (newlen != sizeof(int))
 		return -EINVAL;
 	if (get_user(delay, (int __user *)newval))
 		return -EFAULT;
-	rt_cache_flush(&init_net, delay);
+	net = (struct net *)table->extra1;
+	rt_cache_flush(net, delay);
 	return 0;
 }
 
 ctl_table ipv4_route_table[] = {
-	{
-		.ctl_name 	= NET_IPV4_ROUTE_FLUSH,
-		.procname	= "flush",
-		.maxlen		= sizeof(int),
-		.mode		= 0200,
-		.proc_handler	= &ipv4_sysctl_rtcache_flush,
-		.strategy	= &ipv4_sysctl_rtcache_flush_strategy,
-	},
 	{
 		.ctl_name	= NET_IPV4_ROUTE_GC_THRESH,
 		.procname	= "gc_thresh",
@@ -3014,6 +3010,66 @@ ctl_table ipv4_route_table[] = {
 	},
 	{ .ctl_name = 0 }
 };
+
+static __net_initdata struct ctl_path ipv4_route_path[] = {
+	{ .procname = "net", .ctl_name = CTL_NET, },
+	{ .procname = "ipv4", .ctl_name = NET_IPV4, },
+	{ .procname = "route", .ctl_name = NET_IPV4_ROUTE, },
+	{ },
+};
+
+
+static struct ctl_table ipv4_route_flush_table[] = {
+	{
+		.ctl_name 	= NET_IPV4_ROUTE_FLUSH,
+		.procname	= "flush",
+		.maxlen		= sizeof(int),
+		.mode		= 0200,
+		.proc_handler	= &ipv4_sysctl_rtcache_flush,
+		.strategy	= &ipv4_sysctl_rtcache_flush_strategy,
+	},
+	{ .ctl_name = 0 },
+};
+
+static __net_init int sysctl_route_net_init(struct net *net)
+{
+	struct ctl_table *tbl;
+
+	tbl = ipv4_route_flush_table;
+	if (net != &init_net) {
+		tbl = kmemdup(tbl, sizeof(ipv4_route_flush_table), GFP_KERNEL);
+		if (tbl == NULL)
+			goto err_dup;
+	}
+	tbl[0].extra1 = net;
+
+	net->ipv4.route_hdr =
+		register_net_sysctl_table(net, ipv4_route_path, tbl);
+	if (net->ipv4.route_hdr == NULL)
+		goto err_reg;
+	return 0;
+
+err_reg:
+	if (tbl != ipv4_route_flush_table)
+		kfree(tbl);
+err_dup:
+	return -ENOMEM;
+}
+
+static __net_exit void sysctl_route_net_exit(struct net *net)
+{
+	struct ctl_table *tbl;
+
+	tbl = net->ipv4.route_hdr->ctl_table_arg;
+	unregister_net_sysctl_table(net->ipv4.route_hdr);
+	BUG_ON(tbl == ipv4_route_flush_table);
+	kfree(tbl);
+}
+
+static __net_initdata struct pernet_operations sysctl_route_ops = {
+	.init = sysctl_route_net_init,
+	.exit = sysctl_route_net_exit,
+};
 #endif
 
 #ifdef CONFIG_NET_CLS_ROUTE
@@ -3090,6 +3146,9 @@ int __init ip_rt_init(void)
 #endif
 	rtnl_register(PF_INET, RTM_GETROUTE, inet_rtm_getroute, NULL);
 
+#ifdef CONFIG_SYSCTL
+	register_pernet_subsys(&sysctl_route_ops);
+#endif
 	return rc;
 }
 
-- 
cgit v1.2.3


From 9f5e97e53675caeda48e9988122a30470f4d309d Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Sat, 5 Jul 2008 19:02:59 -0700
Subject: netns: make rt_secret_rebuild timer per namespace

Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/netns/ipv4.h |  2 ++
 net/ipv4/route.c         | 40 ++++++++++++++++++++++++++++++----------
 2 files changed, 32 insertions(+), 10 deletions(-)

(limited to 'include')

diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h
index a29adf1ba0ff..356617f81beb 100644
--- a/include/net/netns/ipv4.h
+++ b/include/net/netns/ipv4.h
@@ -46,5 +46,7 @@ struct netns_ipv4 {
 	int sysctl_icmp_ratelimit;
 	int sysctl_icmp_ratemask;
 	int sysctl_icmp_errors_use_inbound_ifaddr;
+
+	struct timer_list rt_secret_timer;
 };
 #endif
diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index 6fe799de9b96..f99d9dbb7720 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -132,7 +132,6 @@ static int ip_rt_secret_interval __read_mostly	= 10 * 60 * HZ;
 
 static void rt_worker_func(struct work_struct *work);
 static DECLARE_DELAYED_WORK(expires_work, rt_worker_func);
-static struct timer_list rt_secret_timer;
 
 /*
  *	Interface to generic destination cache.
@@ -801,10 +800,11 @@ void rt_cache_flush(struct net *net, int delay)
 /*
  * We change rt_genid and let gc do the cleanup
  */
-static void rt_secret_rebuild(unsigned long dummy)
+static void rt_secret_rebuild(unsigned long __net)
 {
+	struct net *net = (struct net *)__net;
 	rt_cache_invalidate();
-	mod_timer(&rt_secret_timer, jiffies + ip_rt_secret_interval);
+	mod_timer(&net->ipv4.rt_secret_timer, jiffies + ip_rt_secret_interval);
 }
 
 /*
@@ -3072,6 +3072,31 @@ static __net_initdata struct pernet_operations sysctl_route_ops = {
 };
 #endif
 
+
+static __net_init int rt_secret_timer_init(struct net *net)
+{
+	net->ipv4.rt_secret_timer.function = rt_secret_rebuild;
+	net->ipv4.rt_secret_timer.data = (unsigned long)net;
+	init_timer_deferrable(&net->ipv4.rt_secret_timer);
+
+	net->ipv4.rt_secret_timer.expires =
+		jiffies + net_random() % ip_rt_secret_interval +
+		ip_rt_secret_interval;
+	add_timer(&net->ipv4.rt_secret_timer);
+	return 0;
+}
+
+static __net_exit void rt_secret_timer_exit(struct net *net)
+{
+	del_timer_sync(&net->ipv4.rt_secret_timer);
+}
+
+static __net_initdata struct pernet_operations rt_secret_timer_ops = {
+	.init = rt_secret_timer_init,
+	.exit = rt_secret_timer_exit,
+};
+
+
 #ifdef CONFIG_NET_CLS_ROUTE
 struct ip_rt_acct *ip_rt_acct __read_mostly;
 #endif /* CONFIG_NET_CLS_ROUTE */
@@ -3124,19 +3149,14 @@ int __init ip_rt_init(void)
 	devinet_init();
 	ip_fib_init();
 
-	rt_secret_timer.function = rt_secret_rebuild;
-	rt_secret_timer.data = 0;
-	init_timer_deferrable(&rt_secret_timer);
-
 	/* All the timers, started at system startup tend
 	   to synchronize. Perturb it a bit.
 	 */
 	schedule_delayed_work(&expires_work,
 		net_random() % ip_rt_gc_interval + ip_rt_gc_interval);
 
-	rt_secret_timer.expires = jiffies + net_random() % ip_rt_secret_interval +
-		ip_rt_secret_interval;
-	add_timer(&rt_secret_timer);
+	if (register_pernet_subsys(&rt_secret_timer_ops))
+		printk(KERN_ERR "Unable to setup rt_secret_timer\n");
 
 	if (ip_rt_proc_init())
 		printk(KERN_ERR "Unable to create route proc files\n");
-- 
cgit v1.2.3


From e84f84f276473dcc673f360e8ff3203148bdf0e2 Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Sat, 5 Jul 2008 19:04:32 -0700
Subject: netns: place rt_genid into struct net

Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/netns/ipv4.h |  1 +
 net/ipv4/route.c         | 76 +++++++++++++++++++++++++++---------------------
 2 files changed, 44 insertions(+), 33 deletions(-)

(limited to 'include')

diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h
index 356617f81beb..a6ed83853dcc 100644
--- a/include/net/netns/ipv4.h
+++ b/include/net/netns/ipv4.h
@@ -48,5 +48,6 @@ struct netns_ipv4 {
 	int sysctl_icmp_errors_use_inbound_ifaddr;
 
 	struct timer_list rt_secret_timer;
+	atomic_t rt_genid;
 };
 #endif
diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index e4e37edbad60..67c3ed772c27 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -250,7 +250,6 @@ static inline void rt_hash_lock_init(void)
 static struct rt_hash_bucket 	*rt_hash_table __read_mostly;
 static unsigned			rt_hash_mask __read_mostly;
 static unsigned int		rt_hash_log  __read_mostly;
-static atomic_t			rt_genid __read_mostly;
 
 static DEFINE_PER_CPU(struct rt_cache_stat, rt_cache_stat);
 #define RT_CACHE_STAT_INC(field) \
@@ -265,6 +264,11 @@ static inline unsigned int rt_hash(__be32 daddr, __be32 saddr, int idx,
 		& rt_hash_mask;
 }
 
+static inline int rt_genid(struct net *net)
+{
+	return atomic_read(&net->ipv4.rt_genid);
+}
+
 #ifdef CONFIG_PROC_FS
 struct rt_cache_iter_state {
 	struct seq_net_private p;
@@ -334,7 +338,7 @@ static void *rt_cache_seq_start(struct seq_file *seq, loff_t *pos)
 	struct rt_cache_iter_state *st = seq->private;
 	if (*pos)
 		return rt_cache_get_idx(seq, *pos - 1);
-	st->genid = atomic_read(&rt_genid);
+	st->genid = rt_genid(seq_file_net(seq));
 	return SEQ_START_TOKEN;
 }
 
@@ -681,6 +685,11 @@ static inline int compare_netns(struct rtable *rt1, struct rtable *rt2)
 	return dev_net(rt1->u.dst.dev) == dev_net(rt2->u.dst.dev);
 }
 
+static inline int rt_is_expired(struct rtable *rth)
+{
+	return rth->rt_genid != rt_genid(dev_net(rth->u.dst.dev));
+}
+
 /*
  * Perform a full scan of hash table and free all entries.
  * Can be called by a softirq or a process.
@@ -736,7 +745,7 @@ static void rt_check_expire(void)
 			continue;
 		spin_lock_bh(rt_hash_lock_addr(i));
 		while ((rth = *rthp) != NULL) {
-			if (rth->rt_genid != atomic_read(&rt_genid)) {
+			if (rt_is_expired(rth)) {
 				*rthp = rth->u.dst.rt_next;
 				rt_free(rth);
 				continue;
@@ -784,7 +793,7 @@ static void rt_cache_invalidate(struct net *net)
 	unsigned char shuffle;
 
 	get_random_bytes(&shuffle, sizeof(shuffle));
-	atomic_add(shuffle + 1U, &rt_genid);
+	atomic_add(shuffle + 1U, &net->ipv4.rt_genid);
 }
 
 /*
@@ -881,7 +890,7 @@ static int rt_garbage_collect(struct dst_ops *ops)
 			rthp = &rt_hash_table[k].chain;
 			spin_lock_bh(rt_hash_lock_addr(k));
 			while ((rth = *rthp) != NULL) {
-				if (rth->rt_genid == atomic_read(&rt_genid) &&
+				if (!rt_is_expired(rth) &&
 					!rt_may_expire(rth, tmo, expire)) {
 					tmo >>= 1;
 					rthp = &rth->u.dst.rt_next;
@@ -963,7 +972,7 @@ restart:
 
 	spin_lock_bh(rt_hash_lock_addr(hash));
 	while ((rth = *rthp) != NULL) {
-		if (rth->rt_genid != atomic_read(&rt_genid)) {
+		if (rt_is_expired(rth)) {
 			*rthp = rth->u.dst.rt_next;
 			rt_free(rth);
 			continue;
@@ -1139,7 +1148,7 @@ static void rt_del(unsigned hash, struct rtable *rt)
 	spin_lock_bh(rt_hash_lock_addr(hash));
 	ip_rt_put(rt);
 	while ((aux = *rthp) != NULL) {
-		if (aux == rt || (aux->rt_genid != atomic_read(&rt_genid))) {
+		if (aux == rt || rt_is_expired(aux)) {
 			*rthp = aux->u.dst.rt_next;
 			rt_free(aux);
 			continue;
@@ -1182,7 +1191,7 @@ void ip_rt_redirect(__be32 old_gw, __be32 daddr, __be32 new_gw,
 	for (i = 0; i < 2; i++) {
 		for (k = 0; k < 2; k++) {
 			unsigned hash = rt_hash(daddr, skeys[i], ikeys[k],
-						atomic_read(&rt_genid));
+						rt_genid(net));
 
 			rthp=&rt_hash_table[hash].chain;
 
@@ -1194,7 +1203,7 @@ void ip_rt_redirect(__be32 old_gw, __be32 daddr, __be32 new_gw,
 				    rth->fl.fl4_src != skeys[i] ||
 				    rth->fl.oif != ikeys[k] ||
 				    rth->fl.iif != 0 ||
-				    rth->rt_genid != atomic_read(&rt_genid) ||
+				    rt_is_expired(rth) ||
 				    !net_eq(dev_net(rth->u.dst.dev), net)) {
 					rthp = &rth->u.dst.rt_next;
 					continue;
@@ -1233,7 +1242,7 @@ void ip_rt_redirect(__be32 old_gw, __be32 daddr, __be32 new_gw,
 				rt->u.dst.neighbour	= NULL;
 				rt->u.dst.hh		= NULL;
 				rt->u.dst.xfrm		= NULL;
-				rt->rt_genid		= atomic_read(&rt_genid);
+				rt->rt_genid		= rt_genid(net);
 				rt->rt_flags		|= RTCF_REDIRECTED;
 
 				/* Gateway is different ... */
@@ -1298,7 +1307,7 @@ static struct dst_entry *ipv4_negative_advice(struct dst_entry *dst)
 			   rt->u.dst.expires) {
 			unsigned hash = rt_hash(rt->fl.fl4_dst, rt->fl.fl4_src,
 						rt->fl.oif,
-						atomic_read(&rt_genid));
+						rt_genid(dev_net(dst->dev)));
 #if RT_CACHE_DEBUG >= 1
 			printk(KERN_DEBUG "ipv4_negative_advice: redirect to "
 					  NIPQUAD_FMT "/%02x dropped\n",
@@ -1448,7 +1457,7 @@ unsigned short ip_rt_frag_needed(struct net *net, struct iphdr *iph,
 	for (k = 0; k < 2; k++) {
 		for (i = 0; i < 2; i++) {
 			unsigned hash = rt_hash(daddr, skeys[i], ikeys[k],
-						atomic_read(&rt_genid));
+						rt_genid(net));
 
 			rcu_read_lock();
 			for (rth = rcu_dereference(rt_hash_table[hash].chain); rth;
@@ -1463,7 +1472,7 @@ unsigned short ip_rt_frag_needed(struct net *net, struct iphdr *iph,
 				    rth->fl.iif != 0 ||
 				    dst_metric_locked(&rth->u.dst, RTAX_MTU) ||
 				    !net_eq(dev_net(rth->u.dst.dev), net) ||
-				    rth->rt_genid != atomic_read(&rt_genid))
+				    !rt_is_expired(rth))
 					continue;
 
 				if (new_mtu < 68 || new_mtu >= old_mtu) {
@@ -1698,7 +1707,7 @@ static int ip_route_input_mc(struct sk_buff *skb, __be32 daddr, __be32 saddr,
 	rth->fl.oif	= 0;
 	rth->rt_gateway	= daddr;
 	rth->rt_spec_dst= spec_dst;
-	rth->rt_genid	= atomic_read(&rt_genid);
+	rth->rt_genid	= rt_genid(dev_net(dev));
 	rth->rt_flags	= RTCF_MULTICAST;
 	rth->rt_type	= RTN_MULTICAST;
 	if (our) {
@@ -1713,7 +1722,7 @@ static int ip_route_input_mc(struct sk_buff *skb, __be32 daddr, __be32 saddr,
 	RT_CACHE_STAT_INC(in_slow_mc);
 
 	in_dev_put(in_dev);
-	hash = rt_hash(daddr, saddr, dev->ifindex, atomic_read(&rt_genid));
+	hash = rt_hash(daddr, saddr, dev->ifindex, rt_genid(dev_net(dev)));
 	return rt_intern_hash(hash, rth, &skb->rtable);
 
 e_nobufs:
@@ -1839,7 +1848,7 @@ static int __mkroute_input(struct sk_buff *skb,
 
 	rth->u.dst.input = ip_forward;
 	rth->u.dst.output = ip_output;
-	rth->rt_genid = atomic_read(&rt_genid);
+	rth->rt_genid = rt_genid(dev_net(rth->u.dst.dev));
 
 	rt_set_nexthop(rth, res, itag);
 
@@ -1874,7 +1883,8 @@ static int ip_mkroute_input(struct sk_buff *skb,
 		return err;
 
 	/* put it into the cache */
-	hash = rt_hash(daddr, saddr, fl->iif, atomic_read(&rt_genid));
+	hash = rt_hash(daddr, saddr, fl->iif,
+		       rt_genid(dev_net(rth->u.dst.dev)));
 	return rt_intern_hash(hash, rth, &skb->rtable);
 }
 
@@ -2000,7 +2010,7 @@ local_input:
 		goto e_nobufs;
 
 	rth->u.dst.output= ip_rt_bug;
-	rth->rt_genid = atomic_read(&rt_genid);
+	rth->rt_genid = rt_genid(net);
 
 	atomic_set(&rth->u.dst.__refcnt, 1);
 	rth->u.dst.flags= DST_HOST;
@@ -2030,7 +2040,7 @@ local_input:
 		rth->rt_flags 	&= ~RTCF_LOCAL;
 	}
 	rth->rt_type	= res.type;
-	hash = rt_hash(daddr, saddr, fl.iif, atomic_read(&rt_genid));
+	hash = rt_hash(daddr, saddr, fl.iif, rt_genid(net));
 	err = rt_intern_hash(hash, rth, &skb->rtable);
 	goto done;
 
@@ -2081,7 +2091,7 @@ int ip_route_input(struct sk_buff *skb, __be32 daddr, __be32 saddr,
 
 	net = dev_net(dev);
 	tos &= IPTOS_RT_MASK;
-	hash = rt_hash(daddr, saddr, iif, atomic_read(&rt_genid));
+	hash = rt_hash(daddr, saddr, iif, rt_genid(net));
 
 	rcu_read_lock();
 	for (rth = rcu_dereference(rt_hash_table[hash].chain); rth;
@@ -2093,7 +2103,7 @@ int ip_route_input(struct sk_buff *skb, __be32 daddr, __be32 saddr,
 		     (rth->fl.fl4_tos ^ tos)) == 0 &&
 		    rth->fl.mark == skb->mark &&
 		    net_eq(dev_net(rth->u.dst.dev), net) &&
-		    rth->rt_genid == atomic_read(&rt_genid)) {
+		    !rt_is_expired(rth)) {
 			dst_use(&rth->u.dst, jiffies);
 			RT_CACHE_STAT_INC(in_hit);
 			rcu_read_unlock();
@@ -2221,7 +2231,7 @@ static int __mkroute_output(struct rtable **result,
 	rth->rt_spec_dst= fl->fl4_src;
 
 	rth->u.dst.output=ip_output;
-	rth->rt_genid = atomic_read(&rt_genid);
+	rth->rt_genid = rt_genid(dev_net(dev_out));
 
 	RT_CACHE_STAT_INC(out_slow_tot);
 
@@ -2271,7 +2281,7 @@ static int ip_mkroute_output(struct rtable **rp,
 	unsigned hash;
 	if (err == 0) {
 		hash = rt_hash(oldflp->fl4_dst, oldflp->fl4_src, oldflp->oif,
-			       atomic_read(&rt_genid));
+			       rt_genid(dev_net(dev_out)));
 		err = rt_intern_hash(hash, rth, rp);
 	}
 
@@ -2483,8 +2493,7 @@ int __ip_route_output_key(struct net *net, struct rtable **rp,
 	unsigned hash;
 	struct rtable *rth;
 
-	hash = rt_hash(flp->fl4_dst, flp->fl4_src, flp->oif,
-		       atomic_read(&rt_genid));
+	hash = rt_hash(flp->fl4_dst, flp->fl4_src, flp->oif, rt_genid(net));
 
 	rcu_read_lock_bh();
 	for (rth = rcu_dereference(rt_hash_table[hash].chain); rth;
@@ -2497,7 +2506,7 @@ int __ip_route_output_key(struct net *net, struct rtable **rp,
 		    !((rth->fl.fl4_tos ^ flp->fl4_tos) &
 			    (IPTOS_RT_MASK | RTO_ONLINK)) &&
 		    net_eq(dev_net(rth->u.dst.dev), net) &&
-		    rth->rt_genid == atomic_read(&rt_genid)) {
+		    !rt_is_expired(rth)) {
 			dst_use(&rth->u.dst, jiffies);
 			RT_CACHE_STAT_INC(out_hit);
 			rcu_read_unlock_bh();
@@ -2528,7 +2537,7 @@ static struct dst_ops ipv4_dst_blackhole_ops = {
 };
 
 
-static int ipv4_dst_blackhole(struct rtable **rp, struct flowi *flp)
+static int ipv4_dst_blackhole(struct net *net, struct rtable **rp, struct flowi *flp)
 {
 	struct rtable *ort = *rp;
 	struct rtable *rt = (struct rtable *)
@@ -2552,7 +2561,7 @@ static int ipv4_dst_blackhole(struct rtable **rp, struct flowi *flp)
 		rt->idev = ort->idev;
 		if (rt->idev)
 			in_dev_hold(rt->idev);
-		rt->rt_genid = atomic_read(&rt_genid);
+		rt->rt_genid = rt_genid(net);
 		rt->rt_flags = ort->rt_flags;
 		rt->rt_type = ort->rt_type;
 		rt->rt_dst = ort->rt_dst;
@@ -2588,7 +2597,7 @@ int ip_route_output_flow(struct net *net, struct rtable **rp, struct flowi *flp,
 		err = __xfrm_lookup((struct dst_entry **)rp, flp, sk,
 				    flags ? XFRM_LOOKUP_WAIT : 0);
 		if (err == -EREMOTE)
-			err = ipv4_dst_blackhole(rp, flp);
+			err = ipv4_dst_blackhole(net, rp, flp);
 
 		return err;
 	}
@@ -2807,7 +2816,7 @@ int ip_rt_dump(struct sk_buff *skb,  struct netlink_callback *cb)
 		     rt = rcu_dereference(rt->u.dst.rt_next), idx++) {
 			if (!net_eq(dev_net(rt->u.dst.dev), net) || idx < s_idx)
 				continue;
-			if (rt->rt_genid != atomic_read(&rt_genid))
+			if (rt_is_expired(rt))
 				continue;
 			skb->dst = dst_clone(&rt->u.dst);
 			if (rt_fill_info(skb, NETLINK_CB(cb->skb).pid,
@@ -3081,6 +3090,10 @@ static __net_initdata struct pernet_operations sysctl_route_ops = {
 
 static __net_init int rt_secret_timer_init(struct net *net)
 {
+	atomic_set(&net->ipv4.rt_genid,
+			(int) ((num_physpages ^ (num_physpages>>8)) ^
+			(jiffies ^ (jiffies >> 7))));
+
 	net->ipv4.rt_secret_timer.function = rt_secret_rebuild;
 	net->ipv4.rt_secret_timer.data = (unsigned long)net;
 	init_timer_deferrable(&net->ipv4.rt_secret_timer);
@@ -3121,9 +3134,6 @@ int __init ip_rt_init(void)
 {
 	int rc = 0;
 
-	atomic_set(&rt_genid, (int) ((num_physpages ^ (num_physpages>>8)) ^
-			     (jiffies ^ (jiffies >> 7))));
-
 #ifdef CONFIG_NET_CLS_ROUTE
 	ip_rt_acct = __alloc_percpu(256 * sizeof(struct ip_rt_acct));
 	if (!ip_rt_acct)
-- 
cgit v1.2.3


From 629ca23c331ec75ac87b016debbb3c4d2fe62650 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Sat, 5 Jul 2008 21:18:07 -0700
Subject: MIB: add struct net to UDP_INC_STATS_USER

Nothing special - all the places already have a struct sock
at hands, so use the sock_net() net.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Acked-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/udp.h |  2 +-
 net/ipv4/udp.c    | 11 +++++++----
 2 files changed, 8 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/include/net/udp.h b/include/net/udp.h
index 7a8684855245..13319094b134 100644
--- a/include/net/udp.h
+++ b/include/net/udp.h
@@ -158,7 +158,7 @@ DECLARE_SNMP_STAT(struct udp_mib, udplite_stats_in6);
 /*
  * 	SNMP statistics for UDP and UDP-Lite
  */
-#define UDP_INC_STATS_USER(field, is_udplite)			       do {   \
+#define UDP_INC_STATS_USER(net, field, is_udplite)	      do { (void)net; \
 	if (is_udplite) SNMP_INC_STATS_USER(udplite_statistics, field);       \
 	else		SNMP_INC_STATS_USER(udp_statistics, field);  }  while(0)
 #define UDP_INC_STATS_BH(field, is_udplite) 			       do  {  \
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index 3bbf6fb6e4f5..c97da5394d70 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -526,7 +526,8 @@ out:
 	up->len = 0;
 	up->pending = 0;
 	if (!err)
-		UDP_INC_STATS_USER(UDP_MIB_OUTDATAGRAMS, is_udplite);
+		UDP_INC_STATS_USER(sock_net(sk),
+				UDP_MIB_OUTDATAGRAMS, is_udplite);
 	return err;
 }
 
@@ -725,7 +726,8 @@ out:
 	 * seems like overkill.
 	 */
 	if (err == -ENOBUFS || test_bit(SOCK_NOSPACE, &sk->sk_socket->flags)) {
-		UDP_INC_STATS_USER(UDP_MIB_SNDBUFERRORS, is_udplite);
+		UDP_INC_STATS_USER(sock_net(sk),
+				UDP_MIB_SNDBUFERRORS, is_udplite);
 	}
 	return err;
 
@@ -888,7 +890,8 @@ try_again:
 		goto out_free;
 
 	if (!peeked)
-		UDP_INC_STATS_USER(UDP_MIB_INDATAGRAMS, is_udplite);
+		UDP_INC_STATS_USER(sock_net(sk),
+				UDP_MIB_INDATAGRAMS, is_udplite);
 
 	sock_recv_timestamp(msg, sk, skb);
 
@@ -917,7 +920,7 @@ out:
 csum_copy_err:
 	lock_sock(sk);
 	if (!skb_kill_datagram(sk, skb, flags))
-		UDP_INC_STATS_USER(UDP_MIB_INERRORS, is_udplite);
+		UDP_INC_STATS_USER(sock_net(sk), UDP_MIB_INERRORS, is_udplite);
 	release_sock(sk);
 
 	if (noblock)
-- 
cgit v1.2.3


From 0283328e2360bbf3081e97f1b720dd4c4dbae111 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Sat, 5 Jul 2008 21:18:48 -0700
Subject: MIB: add struct net to UDP_INC_STATS_BH

Two special cases here - one is rxrpc - I put init_net there
explicitly, since we haven't touched this part yet. The second
place is in __udp4_lib_rcv - we already have a struct net there,
but I have to move its initialization above to make it ready
at the "drop" label.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Acked-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/udp.h    |  6 +++---
 net/ipv4/udp.c       | 18 ++++++++++--------
 net/rxrpc/ar-input.c |  5 +++--
 3 files changed, 16 insertions(+), 13 deletions(-)

(limited to 'include')

diff --git a/include/net/udp.h b/include/net/udp.h
index 13319094b134..7b18cfb2fe0b 100644
--- a/include/net/udp.h
+++ b/include/net/udp.h
@@ -161,7 +161,7 @@ DECLARE_SNMP_STAT(struct udp_mib, udplite_stats_in6);
 #define UDP_INC_STATS_USER(net, field, is_udplite)	      do { (void)net; \
 	if (is_udplite) SNMP_INC_STATS_USER(udplite_statistics, field);       \
 	else		SNMP_INC_STATS_USER(udp_statistics, field);  }  while(0)
-#define UDP_INC_STATS_BH(field, is_udplite) 			       do  {  \
+#define UDP_INC_STATS_BH(net, field, is_udplite) 	      do { (void)net; \
 	if (is_udplite) SNMP_INC_STATS_BH(udplite_statistics, field);         \
 	else		SNMP_INC_STATS_BH(udp_statistics, field);    }  while(0)
 
@@ -176,12 +176,12 @@ DECLARE_SNMP_STAT(struct udp_mib, udplite_stats_in6);
 #define UDPX_INC_STATS_BH(sk, field) \
 	do { \
 		if ((sk)->sk_family == AF_INET) \
-			UDP_INC_STATS_BH(field, 0); \
+			UDP_INC_STATS_BH(sock_net(sk), field, 0); \
 		else \
 			UDP6_INC_STATS_BH(field, 0); \
 	} while (0);
 #else
-#define UDPX_INC_STATS_BH(sk, field) UDP_INC_STATS_BH(field, 0)
+#define UDPX_INC_STATS_BH(sk, field) UDP_INC_STATS_BH(sock_net(sk), field, 0)
 #endif
 
 /* /proc */
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index c97da5394d70..7187121e922d 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -991,7 +991,8 @@ int udp_queue_rcv_skb(struct sock * sk, struct sk_buff *skb)
 
 			ret = (*up->encap_rcv)(sk, skb);
 			if (ret <= 0) {
-				UDP_INC_STATS_BH(UDP_MIB_INDATAGRAMS,
+				UDP_INC_STATS_BH(sock_net(sk),
+						 UDP_MIB_INDATAGRAMS,
 						 is_udplite);
 				return -ret;
 			}
@@ -1044,7 +1045,8 @@ int udp_queue_rcv_skb(struct sock * sk, struct sk_buff *skb)
 	if ((rc = sock_queue_rcv_skb(sk,skb)) < 0) {
 		/* Note that an ENOMEM error is charged twice */
 		if (rc == -ENOMEM) {
-			UDP_INC_STATS_BH(UDP_MIB_RCVBUFERRORS, is_udplite);
+			UDP_INC_STATS_BH(sock_net(sk),
+					UDP_MIB_RCVBUFERRORS, is_udplite);
 			atomic_inc(&sk->sk_drops);
 		}
 		goto drop;
@@ -1053,7 +1055,7 @@ int udp_queue_rcv_skb(struct sock * sk, struct sk_buff *skb)
 	return 0;
 
 drop:
-	UDP_INC_STATS_BH(UDP_MIB_INERRORS, is_udplite);
+	UDP_INC_STATS_BH(sock_net(sk), UDP_MIB_INERRORS, is_udplite);
 	kfree_skb(skb);
 	return -1;
 }
@@ -1161,7 +1163,7 @@ int __udp4_lib_rcv(struct sk_buff *skb, struct hlist_head udptable[],
 	struct rtable *rt = (struct rtable*)skb->dst;
 	__be32 saddr = ip_hdr(skb)->saddr;
 	__be32 daddr = ip_hdr(skb)->daddr;
-	struct net *net;
+	struct net *net = dev_net(skb->dev);
 
 	/*
 	 *  Validate the packet.
@@ -1183,7 +1185,6 @@ int __udp4_lib_rcv(struct sk_buff *skb, struct hlist_head udptable[],
 	if (udp4_csum_init(skb, uh, proto))
 		goto csum_error;
 
-	net = dev_net(skb->dev);
 	if (rt->rt_flags & (RTCF_BROADCAST|RTCF_MULTICAST))
 		return __udp4_lib_mcast_deliver(net, skb, uh,
 				saddr, daddr, udptable);
@@ -1217,7 +1218,7 @@ int __udp4_lib_rcv(struct sk_buff *skb, struct hlist_head udptable[],
 	if (udp_lib_checksum_complete(skb))
 		goto csum_error;
 
-	UDP_INC_STATS_BH(UDP_MIB_NOPORTS, proto == IPPROTO_UDPLITE);
+	UDP_INC_STATS_BH(net, UDP_MIB_NOPORTS, proto == IPPROTO_UDPLITE);
 	icmp_send(skb, ICMP_DEST_UNREACH, ICMP_PORT_UNREACH, 0);
 
 	/*
@@ -1251,7 +1252,7 @@ csum_error:
 		       ntohs(uh->dest),
 		       ulen);
 drop:
-	UDP_INC_STATS_BH(UDP_MIB_INERRORS, proto == IPPROTO_UDPLITE);
+	UDP_INC_STATS_BH(net, UDP_MIB_INERRORS, proto == IPPROTO_UDPLITE);
 	kfree_skb(skb);
 	return 0;
 }
@@ -1458,7 +1459,8 @@ unsigned int udp_poll(struct file *file, struct socket *sock, poll_table *wait)
 		spin_lock_bh(&rcvq->lock);
 		while ((skb = skb_peek(rcvq)) != NULL &&
 		       udp_lib_checksum_complete(skb)) {
-			UDP_INC_STATS_BH(UDP_MIB_INERRORS, is_lite);
+			UDP_INC_STATS_BH(sock_net(sk),
+					UDP_MIB_INERRORS, is_lite);
 			__skb_unlink(skb, rcvq);
 			kfree_skb(skb);
 		}
diff --git a/net/rxrpc/ar-input.c b/net/rxrpc/ar-input.c
index f8a699e92962..f98c8027e5c1 100644
--- a/net/rxrpc/ar-input.c
+++ b/net/rxrpc/ar-input.c
@@ -21,6 +21,7 @@
 #include <net/af_rxrpc.h>
 #include <net/ip.h>
 #include <net/udp.h>
+#include <net/net_namespace.h>
 #include "ar-internal.h"
 
 unsigned long rxrpc_ack_timeout = 1;
@@ -708,12 +709,12 @@ void rxrpc_data_ready(struct sock *sk, int count)
 	if (skb_checksum_complete(skb)) {
 		rxrpc_free_skb(skb);
 		rxrpc_put_local(local);
-		UDP_INC_STATS_BH(UDP_MIB_INERRORS, 0);
+		UDP_INC_STATS_BH(&init_net, UDP_MIB_INERRORS, 0);
 		_leave(" [CSUM failed]");
 		return;
 	}
 
-	UDP_INC_STATS_BH(UDP_MIB_INDATAGRAMS, 0);
+	UDP_INC_STATS_BH(&init_net, UDP_MIB_INDATAGRAMS, 0);
 
 	/* the socket buffer we have is owned by UDP, with UDP's data all over
 	 * it, but we really want our own */
-- 
cgit v1.2.3


From 235b9f7ac53489011d32efeb89e12e308fdd2c64 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Sat, 5 Jul 2008 21:19:20 -0700
Subject: MIB: add struct net to UDP6_INC_STATS_USER

As simple as the patch #1 in this set.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Acked-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/udp.h |  2 +-
 net/ipv6/udp.c    | 11 +++++++----
 2 files changed, 8 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/include/net/udp.h b/include/net/udp.h
index 7b18cfb2fe0b..bb5b9ec85638 100644
--- a/include/net/udp.h
+++ b/include/net/udp.h
@@ -168,7 +168,7 @@ DECLARE_SNMP_STAT(struct udp_mib, udplite_stats_in6);
 #define UDP6_INC_STATS_BH(field, is_udplite) 			      do  {  \
 	if (is_udplite) SNMP_INC_STATS_BH(udplite_stats_in6, field);         \
 	else		SNMP_INC_STATS_BH(udp_stats_in6, field);    } while(0)
-#define UDP6_INC_STATS_USER(field, is_udplite)			       do {    \
+#define UDP6_INC_STATS_USER(net, field, is_udplite)	    do { (void)net;    \
 	if (is_udplite) SNMP_INC_STATS_USER(udplite_stats_in6, field);         \
 	else		SNMP_INC_STATS_USER(udp_stats_in6, field);    } while(0)
 
diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c
index f91e1df0d25e..833f715e4bff 100644
--- a/net/ipv6/udp.c
+++ b/net/ipv6/udp.c
@@ -166,7 +166,8 @@ try_again:
 		goto out_free;
 
 	if (!peeked)
-		UDP6_INC_STATS_USER(UDP_MIB_INDATAGRAMS, is_udplite);
+		UDP6_INC_STATS_USER(sock_net(sk),
+				UDP_MIB_INDATAGRAMS, is_udplite);
 
 	sock_recv_timestamp(msg, sk, skb);
 
@@ -213,7 +214,7 @@ out:
 csum_copy_err:
 	lock_sock(sk);
 	if (!skb_kill_datagram(sk, skb, flags))
-		UDP6_INC_STATS_USER(UDP_MIB_INERRORS, is_udplite);
+		UDP6_INC_STATS_USER(sock_net(sk), UDP_MIB_INERRORS, is_udplite);
 	release_sock(sk);
 
 	if (flags & MSG_DONTWAIT)
@@ -591,7 +592,8 @@ out:
 	up->len = 0;
 	up->pending = 0;
 	if (!err)
-		UDP6_INC_STATS_USER(UDP_MIB_OUTDATAGRAMS, is_udplite);
+		UDP6_INC_STATS_USER(sock_net(sk),
+				UDP_MIB_OUTDATAGRAMS, is_udplite);
 	return err;
 }
 
@@ -873,7 +875,8 @@ out:
 	 * seems like overkill.
 	 */
 	if (err == -ENOBUFS || test_bit(SOCK_NOSPACE, &sk->sk_socket->flags)) {
-		UDP6_INC_STATS_USER(UDP_MIB_SNDBUFERRORS, is_udplite);
+		UDP6_INC_STATS_USER(sock_net(sk),
+				UDP_MIB_SNDBUFERRORS, is_udplite);
 	}
 	return err;
 
-- 
cgit v1.2.3


From ef28d1a20f9f18ebf1be15ef6f097a76f9a63499 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Sat, 5 Jul 2008 21:19:40 -0700
Subject: MIB: add struct net to UDP6_INC_STATS_BH

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Acked-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/udp.h |  4 ++--
 net/ipv6/udp.c    | 13 +++++++------
 2 files changed, 9 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/include/net/udp.h b/include/net/udp.h
index bb5b9ec85638..3e551592aa76 100644
--- a/include/net/udp.h
+++ b/include/net/udp.h
@@ -165,7 +165,7 @@ DECLARE_SNMP_STAT(struct udp_mib, udplite_stats_in6);
 	if (is_udplite) SNMP_INC_STATS_BH(udplite_statistics, field);         \
 	else		SNMP_INC_STATS_BH(udp_statistics, field);    }  while(0)
 
-#define UDP6_INC_STATS_BH(field, is_udplite) 			      do  {  \
+#define UDP6_INC_STATS_BH(net, field, is_udplite) 	    do { (void)net;  \
 	if (is_udplite) SNMP_INC_STATS_BH(udplite_stats_in6, field);         \
 	else		SNMP_INC_STATS_BH(udp_stats_in6, field);    } while(0)
 #define UDP6_INC_STATS_USER(net, field, is_udplite)	    do { (void)net;    \
@@ -178,7 +178,7 @@ DECLARE_SNMP_STAT(struct udp_mib, udplite_stats_in6);
 		if ((sk)->sk_family == AF_INET) \
 			UDP_INC_STATS_BH(sock_net(sk), field, 0); \
 		else \
-			UDP6_INC_STATS_BH(field, 0); \
+			UDP6_INC_STATS_BH(sock_net(sk), field, 0); \
 	} while (0);
 #else
 #define UDPX_INC_STATS_BH(sk, field) UDP_INC_STATS_BH(sock_net(sk), field, 0)
diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c
index 833f715e4bff..d1477b350f76 100644
--- a/net/ipv6/udp.c
+++ b/net/ipv6/udp.c
@@ -299,7 +299,8 @@ int udpv6_queue_rcv_skb(struct sock * sk, struct sk_buff *skb)
 	if ((rc = sock_queue_rcv_skb(sk,skb)) < 0) {
 		/* Note that an ENOMEM error is charged twice */
 		if (rc == -ENOMEM) {
-			UDP6_INC_STATS_BH(UDP_MIB_RCVBUFERRORS, is_udplite);
+			UDP6_INC_STATS_BH(sock_net(sk),
+					UDP_MIB_RCVBUFERRORS, is_udplite);
 			atomic_inc(&sk->sk_drops);
 		}
 		goto drop;
@@ -307,7 +308,7 @@ int udpv6_queue_rcv_skb(struct sock * sk, struct sk_buff *skb)
 
 	return 0;
 drop:
-	UDP6_INC_STATS_BH(UDP_MIB_INERRORS, is_udplite);
+	UDP6_INC_STATS_BH(sock_net(sk), UDP_MIB_INERRORS, is_udplite);
 	kfree_skb(skb);
 	return -1;
 }
@@ -439,7 +440,7 @@ int __udp6_lib_rcv(struct sk_buff *skb, struct hlist_head udptable[],
 	struct net_device *dev = skb->dev;
 	struct in6_addr *saddr, *daddr;
 	u32 ulen = 0;
-	struct net *net;
+	struct net *net = dev_net(skb->dev);
 
 	if (!pskb_may_pull(skb, sizeof(struct udphdr)))
 		goto short_packet;
@@ -474,7 +475,6 @@ int __udp6_lib_rcv(struct sk_buff *skb, struct hlist_head udptable[],
 	if (udp6_csum_init(skb, uh, proto))
 		goto discard;
 
-	net = dev_net(skb->dev);
 	/*
 	 *	Multicast receive code
 	 */
@@ -497,7 +497,8 @@ int __udp6_lib_rcv(struct sk_buff *skb, struct hlist_head udptable[],
 
 		if (udp_lib_checksum_complete(skb))
 			goto discard;
-		UDP6_INC_STATS_BH(UDP_MIB_NOPORTS, proto == IPPROTO_UDPLITE);
+		UDP6_INC_STATS_BH(net, UDP_MIB_NOPORTS,
+				proto == IPPROTO_UDPLITE);
 
 		icmpv6_send(skb, ICMPV6_DEST_UNREACH, ICMPV6_PORT_UNREACH, 0, dev);
 
@@ -522,7 +523,7 @@ short_packet:
 		       ulen, skb->len);
 
 discard:
-	UDP6_INC_STATS_BH(UDP_MIB_INERRORS, proto == IPPROTO_UDPLITE);
+	UDP6_INC_STATS_BH(net, UDP_MIB_INERRORS, proto == IPPROTO_UDPLITE);
 	kfree_skb(skb);
 	return 0;
 }
-- 
cgit v1.2.3


From a19800d704177caaa5874baf5819307c5b7d5e4f Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Sat, 5 Jul 2008 21:25:39 -0700
Subject: net: Add STP demux layer

Add small STP demux layer for demuxing STP PDUs based on MAC address.
This is needed to run both GARP and STP in parallel (or even load the
modules) since both use LLC_SAP_BSPAN.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/stp.h |  14 ++++++++
 net/802/Kconfig   |   3 ++
 net/802/Makefile  |   1 +
 net/802/stp.c     | 102 ++++++++++++++++++++++++++++++++++++++++++++++++++++++
 net/Kconfig       |   1 +
 5 files changed, 121 insertions(+)
 create mode 100644 include/net/stp.h
 create mode 100644 net/802/Kconfig
 create mode 100644 net/802/stp.c

(limited to 'include')

diff --git a/include/net/stp.h b/include/net/stp.h
new file mode 100644
index 000000000000..ad447f105417
--- /dev/null
+++ b/include/net/stp.h
@@ -0,0 +1,14 @@
+#ifndef _NET_STP_H
+#define _NET_STP_H
+
+struct stp_proto {
+	unsigned char	group_address[ETH_ALEN];
+	void		(*rcv)(const struct stp_proto *, struct sk_buff *,
+			       struct net_device *);
+	void		*data;
+};
+
+extern int stp_proto_register(const struct stp_proto *proto);
+extern void stp_proto_unregister(const struct stp_proto *proto);
+
+#endif /* _NET_STP_H */
diff --git a/net/802/Kconfig b/net/802/Kconfig
new file mode 100644
index 000000000000..01cb0943626b
--- /dev/null
+++ b/net/802/Kconfig
@@ -0,0 +1,3 @@
+config STP
+	tristate
+	select LLC
diff --git a/net/802/Makefile b/net/802/Makefile
index 68569ffddea1..c441d895ac25 100644
--- a/net/802/Makefile
+++ b/net/802/Makefile
@@ -10,3 +10,4 @@ obj-$(CONFIG_FDDI)	+=                 fddi.o
 obj-$(CONFIG_HIPPI)	+=                 hippi.o
 obj-$(CONFIG_IPX)	+= p8022.o psnap.o p8023.o
 obj-$(CONFIG_ATALK)	+= p8022.o psnap.o
+obj-$(CONFIG_STP)	+= stp.o
diff --git a/net/802/stp.c b/net/802/stp.c
new file mode 100644
index 000000000000..0b7a24452d11
--- /dev/null
+++ b/net/802/stp.c
@@ -0,0 +1,102 @@
+/*
+ *	STP SAP demux
+ *
+ *	Copyright (c) 2008 Patrick McHardy <kaber@trash.net>
+ *
+ *	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/mutex.h>
+#include <linux/skbuff.h>
+#include <linux/etherdevice.h>
+#include <linux/llc.h>
+#include <net/llc.h>
+#include <net/llc_pdu.h>
+#include <net/stp.h>
+
+/* 01:80:c2:00:00:20 - 01:80:c2:00:00:2F */
+#define GARP_ADDR_MIN	0x20
+#define GARP_ADDR_MAX	0x2F
+#define GARP_ADDR_RANGE	(GARP_ADDR_MAX - GARP_ADDR_MIN)
+
+static const struct stp_proto *garp_protos[GARP_ADDR_RANGE + 1] __read_mostly;
+static const struct stp_proto *stp_proto __read_mostly;
+
+static struct llc_sap *sap __read_mostly;
+static unsigned int sap_registered;
+static DEFINE_MUTEX(stp_proto_mutex);
+
+/* Called under rcu_read_lock from LLC */
+static int stp_pdu_rcv(struct sk_buff *skb, struct net_device *dev,
+		       struct packet_type *pt, struct net_device *orig_dev)
+{
+	const struct ethhdr *eh = eth_hdr(skb);
+	const struct llc_pdu_un *pdu = llc_pdu_un_hdr(skb);
+	const struct stp_proto *proto;
+
+	if (pdu->ssap != LLC_SAP_BSPAN ||
+	    pdu->dsap != LLC_SAP_BSPAN ||
+	    pdu->ctrl_1 != LLC_PDU_TYPE_U)
+		goto err;
+
+	if (eh->h_dest[5] >= GARP_ADDR_MIN && eh->h_dest[5] <= GARP_ADDR_MAX) {
+		proto = rcu_dereference(garp_protos[eh->h_dest[5] -
+						    GARP_ADDR_MIN]);
+		if (proto &&
+		    compare_ether_addr(eh->h_dest, proto->group_address))
+			goto err;
+	} else
+		proto = rcu_dereference(stp_proto);
+
+	if (!proto)
+		goto err;
+
+	proto->rcv(proto, skb, dev);
+	return 0;
+
+err:
+	kfree_skb(skb);
+	return 0;
+}
+
+int stp_proto_register(const struct stp_proto *proto)
+{
+	int err = 0;
+
+	mutex_lock(&stp_proto_mutex);
+	if (sap_registered++ == 0) {
+		sap = llc_sap_open(LLC_SAP_BSPAN, stp_pdu_rcv);
+		if (!sap) {
+			err = -ENOMEM;
+			goto out;
+		}
+	}
+	if (is_zero_ether_addr(proto->group_address))
+		rcu_assign_pointer(stp_proto, proto);
+	else
+		rcu_assign_pointer(garp_protos[proto->group_address[5] -
+					       GARP_ADDR_MIN], proto);
+out:
+	mutex_unlock(&stp_proto_mutex);
+	return err;
+}
+EXPORT_SYMBOL_GPL(stp_proto_register);
+
+void stp_proto_unregister(const struct stp_proto *proto)
+{
+	mutex_lock(&stp_proto_mutex);
+	if (is_zero_ether_addr(proto->group_address))
+		rcu_assign_pointer(stp_proto, NULL);
+	else
+		rcu_assign_pointer(garp_protos[proto->group_address[5] -
+					       GARP_ADDR_MIN], NULL);
+	synchronize_rcu();
+
+	if (--sap_registered == 0)
+		llc_sap_put(sap);
+	mutex_unlock(&stp_proto_mutex);
+}
+EXPORT_SYMBOL_GPL(stp_proto_unregister);
+
+MODULE_LICENSE("GPL");
diff --git a/net/Kconfig b/net/Kconfig
index acbf7c60e89b..b98668751749 100644
--- a/net/Kconfig
+++ b/net/Kconfig
@@ -181,6 +181,7 @@ source "net/dccp/Kconfig"
 source "net/sctp/Kconfig"
 source "net/tipc/Kconfig"
 source "net/atm/Kconfig"
+source "net/802/Kconfig"
 source "net/bridge/Kconfig"
 source "net/8021q/Kconfig"
 source "net/decnet/Kconfig"
-- 
cgit v1.2.3


From eca9ebac651f774d8b10fce7c5d173c3c3d3394f Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Sat, 5 Jul 2008 21:26:13 -0700
Subject: net: Add GARP applicant-only participant

Add an implementation of the GARP (Generic Attribute Registration Protocol)
applicant-only participant. This will be used by the following patch to
add GVRP support to the VLAN code.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h |   2 +
 include/net/garp.h        | 127 ++++++++++
 net/802/Kconfig           |   4 +
 net/802/Makefile          |   1 +
 net/802/garp.c            | 633 ++++++++++++++++++++++++++++++++++++++++++++++
 5 files changed, 767 insertions(+)
 create mode 100644 include/net/garp.h
 create mode 100644 net/802/garp.c

(limited to 'include')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 56dadb528f67..e009c6fbf5cd 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -740,6 +740,8 @@ struct net_device
 	struct net_bridge_port	*br_port;
 	/* macvlan */
 	struct macvlan_port	*macvlan_port;
+	/* GARP */
+	struct garp_port	*garp_port;
 
 	/* class/net/name entry */
 	struct device		dev;
diff --git a/include/net/garp.h b/include/net/garp.h
new file mode 100644
index 000000000000..73c772395f5b
--- /dev/null
+++ b/include/net/garp.h
@@ -0,0 +1,127 @@
+#ifndef _NET_GARP_H
+#define _NET_GARP_H
+
+#include <net/stp.h>
+
+#define GARP_PROTOCOL_ID	0x1
+#define GARP_END_MARK		0x0
+
+struct garp_pdu_hdr {
+	__be16	protocol;
+};
+
+struct garp_msg_hdr {
+	u8	attrtype;
+};
+
+enum garp_attr_event {
+	GARP_LEAVE_ALL,
+	GARP_JOIN_EMPTY,
+	GARP_JOIN_IN,
+	GARP_LEAVE_EMPTY,
+	GARP_LEAVE_IN,
+	GARP_EMPTY,
+};
+
+struct garp_attr_hdr {
+	u8	len;
+	u8	event;
+	u8	data[];
+};
+
+struct garp_skb_cb {
+	u8	cur_type;
+};
+
+static inline struct garp_skb_cb *garp_cb(struct sk_buff *skb)
+{
+	BUILD_BUG_ON(sizeof(struct garp_skb_cb) >
+		     FIELD_SIZEOF(struct sk_buff, cb));
+	return (struct garp_skb_cb *)skb->cb;
+}
+
+enum garp_applicant_state {
+	GARP_APPLICANT_INVALID,
+	GARP_APPLICANT_VA,
+	GARP_APPLICANT_AA,
+	GARP_APPLICANT_QA,
+	GARP_APPLICANT_LA,
+	GARP_APPLICANT_VP,
+	GARP_APPLICANT_AP,
+	GARP_APPLICANT_QP,
+	GARP_APPLICANT_VO,
+	GARP_APPLICANT_AO,
+	GARP_APPLICANT_QO,
+	__GARP_APPLICANT_MAX
+};
+#define GARP_APPLICANT_MAX	(__GARP_APPLICANT_MAX - 1)
+
+enum garp_event {
+	GARP_EVENT_REQ_JOIN,
+	GARP_EVENT_REQ_LEAVE,
+	GARP_EVENT_R_JOIN_IN,
+	GARP_EVENT_R_JOIN_EMPTY,
+	GARP_EVENT_R_EMPTY,
+	GARP_EVENT_R_LEAVE_IN,
+	GARP_EVENT_R_LEAVE_EMPTY,
+	GARP_EVENT_TRANSMIT_PDU,
+	__GARP_EVENT_MAX
+};
+#define GARP_EVENT_MAX		(__GARP_EVENT_MAX - 1)
+
+enum garp_action {
+	GARP_ACTION_NONE,
+	GARP_ACTION_S_JOIN_IN,
+	GARP_ACTION_S_LEAVE_EMPTY,
+};
+
+struct garp_attr {
+	struct rb_node			node;
+	enum garp_applicant_state	state;
+	u8				type;
+	u8				dlen;
+	unsigned char			data[];
+};
+
+enum garp_applications {
+	__GARP_APPLICATION_MAX
+};
+#define GARP_APPLICATION_MAX	(__GARP_APPLICATION_MAX - 1)
+
+struct garp_application {
+	enum garp_applications	type;
+	unsigned int		maxattr;
+	struct stp_proto	proto;
+};
+
+struct garp_applicant {
+	struct garp_application	*app;
+	struct net_device	*dev;
+	struct timer_list	join_timer;
+
+	spinlock_t		lock;
+	struct sk_buff_head	queue;
+	struct sk_buff		*pdu;
+	struct rb_root		gid;
+};
+
+struct garp_port {
+	struct garp_applicant	*applicants[GARP_APPLICATION_MAX + 1];
+};
+
+extern int	garp_register_application(struct garp_application *app);
+extern void	garp_unregister_application(struct garp_application *app);
+
+extern int	garp_init_applicant(struct net_device *dev,
+				    struct garp_application *app);
+extern void	garp_uninit_applicant(struct net_device *dev,
+				      struct garp_application *app);
+
+extern int	garp_request_join(const struct net_device *dev,
+				  const struct garp_application *app,
+				  const void *data, u8 len, u8 type);
+extern void	garp_request_leave(const struct net_device *dev,
+				   const struct garp_application *app,
+				   const void *data, u8 len, u8 type);
+
+#endif /* _NET_GARP_H */
diff --git a/net/802/Kconfig b/net/802/Kconfig
index 01cb0943626b..be33d27c8e69 100644
--- a/net/802/Kconfig
+++ b/net/802/Kconfig
@@ -1,3 +1,7 @@
 config STP
 	tristate
 	select LLC
+
+config GARP
+	tristate
+	select STP
diff --git a/net/802/Makefile b/net/802/Makefile
index c441d895ac25..7893d679910c 100644
--- a/net/802/Makefile
+++ b/net/802/Makefile
@@ -11,3 +11,4 @@ obj-$(CONFIG_HIPPI)	+=                 hippi.o
 obj-$(CONFIG_IPX)	+= p8022.o psnap.o p8023.o
 obj-$(CONFIG_ATALK)	+= p8022.o psnap.o
 obj-$(CONFIG_STP)	+= stp.o
+obj-$(CONFIG_GARP)	+= garp.o
diff --git a/net/802/garp.c b/net/802/garp.c
new file mode 100644
index 000000000000..3b78f7b74fd4
--- /dev/null
+++ b/net/802/garp.c
@@ -0,0 +1,633 @@
+/*
+ *	IEEE 802.1D Generic Attribute Registration Protocol (GARP)
+ *
+ *	Copyright (c) 2008 Patrick McHardy <kaber@trash.net>
+ *
+ *	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/timer.h>
+#include <linux/skbuff.h>
+#include <linux/netdevice.h>
+#include <linux/etherdevice.h>
+#include <linux/rtnetlink.h>
+#include <linux/llc.h>
+#include <net/llc.h>
+#include <net/llc_pdu.h>
+#include <net/garp.h>
+#include <asm/unaligned.h>
+
+static unsigned int garp_join_time __read_mostly = 200;
+module_param(garp_join_time, uint, 0644);
+MODULE_PARM_DESC(garp_join_time, "Join time in ms (default 200ms)");
+MODULE_LICENSE("GPL");
+
+static const struct garp_state_trans {
+	u8	state;
+	u8	action;
+} garp_applicant_state_table[GARP_APPLICANT_MAX + 1][GARP_EVENT_MAX + 1] = {
+	[GARP_APPLICANT_VA] = {
+		[GARP_EVENT_TRANSMIT_PDU]	= { .state = GARP_APPLICANT_AA,
+						    .action = GARP_ACTION_S_JOIN_IN },
+		[GARP_EVENT_R_JOIN_IN]		= { .state = GARP_APPLICANT_AA },
+		[GARP_EVENT_R_JOIN_EMPTY]	= { .state = GARP_APPLICANT_VA },
+		[GARP_EVENT_R_EMPTY]		= { .state = GARP_APPLICANT_VA },
+		[GARP_EVENT_R_LEAVE_IN]		= { .state = GARP_APPLICANT_VA },
+		[GARP_EVENT_R_LEAVE_EMPTY]	= { .state = GARP_APPLICANT_VP },
+		[GARP_EVENT_REQ_JOIN]		= { .state = GARP_APPLICANT_INVALID },
+		[GARP_EVENT_REQ_LEAVE]		= { .state = GARP_APPLICANT_LA },
+	},
+	[GARP_APPLICANT_AA] = {
+		[GARP_EVENT_TRANSMIT_PDU]	= { .state = GARP_APPLICANT_QA,
+						    .action = GARP_ACTION_S_JOIN_IN },
+		[GARP_EVENT_R_JOIN_IN]		= { .state = GARP_APPLICANT_QA },
+		[GARP_EVENT_R_JOIN_EMPTY]	= { .state = GARP_APPLICANT_VA },
+		[GARP_EVENT_R_EMPTY]		= { .state = GARP_APPLICANT_VA },
+		[GARP_EVENT_R_LEAVE_IN]		= { .state = GARP_APPLICANT_VA },
+		[GARP_EVENT_R_LEAVE_EMPTY]	= { .state = GARP_APPLICANT_VP },
+		[GARP_EVENT_REQ_JOIN]		= { .state = GARP_APPLICANT_INVALID },
+		[GARP_EVENT_REQ_LEAVE]		= { .state = GARP_APPLICANT_LA },
+	},
+	[GARP_APPLICANT_QA] = {
+		[GARP_EVENT_TRANSMIT_PDU]	= { .state = GARP_APPLICANT_INVALID },
+		[GARP_EVENT_R_JOIN_IN]		= { .state = GARP_APPLICANT_QA },
+		[GARP_EVENT_R_JOIN_EMPTY]	= { .state = GARP_APPLICANT_VA },
+		[GARP_EVENT_R_EMPTY]		= { .state = GARP_APPLICANT_VA },
+		[GARP_EVENT_R_LEAVE_IN]		= { .state = GARP_APPLICANT_VP },
+		[GARP_EVENT_R_LEAVE_EMPTY]	= { .state = GARP_APPLICANT_VP },
+		[GARP_EVENT_REQ_JOIN]		= { .state = GARP_APPLICANT_INVALID },
+		[GARP_EVENT_REQ_LEAVE]		= { .state = GARP_APPLICANT_LA },
+	},
+	[GARP_APPLICANT_LA] = {
+		[GARP_EVENT_TRANSMIT_PDU]	= { .state = GARP_APPLICANT_VO,
+						    .action = GARP_ACTION_S_LEAVE_EMPTY },
+		[GARP_EVENT_R_JOIN_IN]		= { .state = GARP_APPLICANT_LA },
+		[GARP_EVENT_R_JOIN_EMPTY]	= { .state = GARP_APPLICANT_VO },
+		[GARP_EVENT_R_EMPTY]		= { .state = GARP_APPLICANT_LA },
+		[GARP_EVENT_R_LEAVE_IN]		= { .state = GARP_APPLICANT_LA },
+		[GARP_EVENT_R_LEAVE_EMPTY]	= { .state = GARP_APPLICANT_VO },
+		[GARP_EVENT_REQ_JOIN]		= { .state = GARP_APPLICANT_VA },
+		[GARP_EVENT_REQ_LEAVE]		= { .state = GARP_APPLICANT_INVALID },
+	},
+	[GARP_APPLICANT_VP] = {
+		[GARP_EVENT_TRANSMIT_PDU]	= { .state = GARP_APPLICANT_AA,
+						    .action = GARP_ACTION_S_JOIN_IN },
+		[GARP_EVENT_R_JOIN_IN]		= { .state = GARP_APPLICANT_AP },
+		[GARP_EVENT_R_JOIN_EMPTY]	= { .state = GARP_APPLICANT_VP },
+		[GARP_EVENT_R_EMPTY]		= { .state = GARP_APPLICANT_VP },
+		[GARP_EVENT_R_LEAVE_IN]		= { .state = GARP_APPLICANT_VP },
+		[GARP_EVENT_R_LEAVE_EMPTY]	= { .state = GARP_APPLICANT_VP },
+		[GARP_EVENT_REQ_JOIN]		= { .state = GARP_APPLICANT_INVALID },
+		[GARP_EVENT_REQ_LEAVE]		= { .state = GARP_APPLICANT_VO },
+	},
+	[GARP_APPLICANT_AP] = {
+		[GARP_EVENT_TRANSMIT_PDU]	= { .state = GARP_APPLICANT_QA,
+						    .action = GARP_ACTION_S_JOIN_IN },
+		[GARP_EVENT_R_JOIN_IN]		= { .state = GARP_APPLICANT_QP },
+		[GARP_EVENT_R_JOIN_EMPTY]	= { .state = GARP_APPLICANT_VP },
+		[GARP_EVENT_R_EMPTY]		= { .state = GARP_APPLICANT_VP },
+		[GARP_EVENT_R_LEAVE_IN]		= { .state = GARP_APPLICANT_VP },
+		[GARP_EVENT_R_LEAVE_EMPTY]	= { .state = GARP_APPLICANT_VP },
+		[GARP_EVENT_REQ_JOIN]		= { .state = GARP_APPLICANT_INVALID },
+		[GARP_EVENT_REQ_LEAVE]		= { .state = GARP_APPLICANT_AO },
+	},
+	[GARP_APPLICANT_QP] = {
+		[GARP_EVENT_TRANSMIT_PDU]	= { .state = GARP_APPLICANT_INVALID },
+		[GARP_EVENT_R_JOIN_IN]		= { .state = GARP_APPLICANT_QP },
+		[GARP_EVENT_R_JOIN_EMPTY]	= { .state = GARP_APPLICANT_VP },
+		[GARP_EVENT_R_EMPTY]		= { .state = GARP_APPLICANT_VP },
+		[GARP_EVENT_R_LEAVE_IN]		= { .state = GARP_APPLICANT_VP },
+		[GARP_EVENT_R_LEAVE_EMPTY]	= { .state = GARP_APPLICANT_VP },
+		[GARP_EVENT_REQ_JOIN]		= { .state = GARP_APPLICANT_INVALID },
+		[GARP_EVENT_REQ_LEAVE]		= { .state = GARP_APPLICANT_QO },
+	},
+	[GARP_APPLICANT_VO] = {
+		[GARP_EVENT_TRANSMIT_PDU]	= { .state = GARP_APPLICANT_INVALID },
+		[GARP_EVENT_R_JOIN_IN]		= { .state = GARP_APPLICANT_AO },
+		[GARP_EVENT_R_JOIN_EMPTY]	= { .state = GARP_APPLICANT_VO },
+		[GARP_EVENT_R_EMPTY]		= { .state = GARP_APPLICANT_VO },
+		[GARP_EVENT_R_LEAVE_IN]		= { .state = GARP_APPLICANT_VO },
+		[GARP_EVENT_R_LEAVE_EMPTY]	= { .state = GARP_APPLICANT_VO },
+		[GARP_EVENT_REQ_JOIN]		= { .state = GARP_APPLICANT_VP },
+		[GARP_EVENT_REQ_LEAVE]		= { .state = GARP_APPLICANT_INVALID },
+	},
+	[GARP_APPLICANT_AO] = {
+		[GARP_EVENT_TRANSMIT_PDU]	= { .state = GARP_APPLICANT_INVALID },
+		[GARP_EVENT_R_JOIN_IN]		= { .state = GARP_APPLICANT_QO },
+		[GARP_EVENT_R_JOIN_EMPTY]	= { .state = GARP_APPLICANT_VO },
+		[GARP_EVENT_R_EMPTY]		= { .state = GARP_APPLICANT_VO },
+		[GARP_EVENT_R_LEAVE_IN]		= { .state = GARP_APPLICANT_VO },
+		[GARP_EVENT_R_LEAVE_EMPTY]	= { .state = GARP_APPLICANT_VO },
+		[GARP_EVENT_REQ_JOIN]		= { .state = GARP_APPLICANT_AP },
+		[GARP_EVENT_REQ_LEAVE]		= { .state = GARP_APPLICANT_INVALID },
+	},
+	[GARP_APPLICANT_QO] = {
+		[GARP_EVENT_TRANSMIT_PDU]	= { .state = GARP_APPLICANT_INVALID },
+		[GARP_EVENT_R_JOIN_IN]		= { .state = GARP_APPLICANT_QO },
+		[GARP_EVENT_R_JOIN_EMPTY]	= { .state = GARP_APPLICANT_VO },
+		[GARP_EVENT_R_EMPTY]		= { .state = GARP_APPLICANT_VO },
+		[GARP_EVENT_R_LEAVE_IN]		= { .state = GARP_APPLICANT_VO },
+		[GARP_EVENT_R_LEAVE_EMPTY]	= { .state = GARP_APPLICANT_VO },
+		[GARP_EVENT_REQ_JOIN]		= { .state = GARP_APPLICANT_QP },
+		[GARP_EVENT_REQ_LEAVE]		= { .state = GARP_APPLICANT_INVALID },
+	},
+};
+
+static int garp_attr_cmp(const struct garp_attr *attr,
+			 const void *data, u8 len, u8 type)
+{
+	if (attr->type != type)
+		return attr->type - type;
+	if (attr->dlen != len)
+		return attr->dlen - len;
+	return memcmp(attr->data, data, len);
+}
+
+static struct garp_attr *garp_attr_lookup(const struct garp_applicant *app,
+					  const void *data, u8 len, u8 type)
+{
+	struct rb_node *parent = app->gid.rb_node;
+	struct garp_attr *attr;
+	int d;
+
+	while (parent) {
+		attr = rb_entry(parent, struct garp_attr, node);
+		d = garp_attr_cmp(attr, data, len, type);
+		if (d < 0)
+			parent = parent->rb_left;
+		else if (d > 0)
+			parent = parent->rb_right;
+		else
+			return attr;
+	}
+	return NULL;
+}
+
+static void garp_attr_insert(struct garp_applicant *app, struct garp_attr *new)
+{
+	struct rb_node *parent = NULL, **p = &app->gid.rb_node;
+	struct garp_attr *attr;
+	int d;
+
+	while (*p) {
+		parent = *p;
+		attr = rb_entry(parent, struct garp_attr, node);
+		d = garp_attr_cmp(attr, new->data, new->dlen, new->type);
+		if (d < 0)
+			p = &parent->rb_left;
+		else if (d > 0)
+			p = &parent->rb_right;
+	}
+	rb_link_node(&new->node, parent, p);
+	rb_insert_color(&new->node, &app->gid);
+}
+
+static struct garp_attr *garp_attr_create(struct garp_applicant *app,
+					  const void *data, u8 len, u8 type)
+{
+	struct garp_attr *attr;
+
+	attr = kmalloc(sizeof(*attr) + len, GFP_ATOMIC);
+	if (!attr)
+		return attr;
+	attr->state = GARP_APPLICANT_VO;
+	attr->type  = type;
+	attr->dlen  = len;
+	memcpy(attr->data, data, len);
+	garp_attr_insert(app, attr);
+	return attr;
+}
+
+static void garp_attr_destroy(struct garp_applicant *app, struct garp_attr *attr)
+{
+	rb_erase(&attr->node, &app->gid);
+	kfree(attr);
+}
+
+static int garp_pdu_init(struct garp_applicant *app)
+{
+	struct sk_buff *skb;
+	struct garp_pdu_hdr *gp;
+
+#define LLC_RESERVE	sizeof(struct llc_pdu_un)
+	skb = alloc_skb(app->dev->mtu + LL_RESERVED_SPACE(app->dev),
+			GFP_ATOMIC);
+	if (!skb)
+		return -ENOMEM;
+
+	skb->dev = app->dev;
+	skb->protocol = htons(ETH_P_802_2);
+	skb_reserve(skb, LL_RESERVED_SPACE(app->dev) + LLC_RESERVE);
+
+	gp = (struct garp_pdu_hdr *)__skb_put(skb, sizeof(*gp));
+	put_unaligned(htons(GARP_PROTOCOL_ID), &gp->protocol);
+
+	app->pdu = skb;
+	return 0;
+}
+
+static int garp_pdu_append_end_mark(struct garp_applicant *app)
+{
+	if (skb_tailroom(app->pdu) < sizeof(u8))
+		return -1;
+	*(u8 *)__skb_put(app->pdu, sizeof(u8)) = GARP_END_MARK;
+	return 0;
+}
+
+static void garp_pdu_queue(struct garp_applicant *app)
+{
+	if (!app->pdu)
+		return;
+
+	garp_pdu_append_end_mark(app);
+	garp_pdu_append_end_mark(app);
+
+	llc_pdu_header_init(app->pdu, LLC_PDU_TYPE_U, LLC_SAP_BSPAN,
+			    LLC_SAP_BSPAN, LLC_PDU_CMD);
+	llc_pdu_init_as_ui_cmd(app->pdu);
+	llc_mac_hdr_init(app->pdu, app->dev->dev_addr,
+			 app->app->proto.group_address);
+
+	skb_queue_tail(&app->queue, app->pdu);
+	app->pdu = NULL;
+}
+
+static void garp_queue_xmit(struct garp_applicant *app)
+{
+	struct sk_buff *skb;
+
+	while ((skb = skb_dequeue(&app->queue)))
+		dev_queue_xmit(skb);
+}
+
+static int garp_pdu_append_msg(struct garp_applicant *app, u8 attrtype)
+{
+	struct garp_msg_hdr *gm;
+
+	if (skb_tailroom(app->pdu) < sizeof(*gm))
+		return -1;
+	gm = (struct garp_msg_hdr *)__skb_put(app->pdu, sizeof(*gm));
+	gm->attrtype = attrtype;
+	garp_cb(app->pdu)->cur_type = attrtype;
+	return 0;
+}
+
+static int garp_pdu_append_attr(struct garp_applicant *app,
+				const struct garp_attr *attr,
+				enum garp_attr_event event)
+{
+	struct garp_attr_hdr *ga;
+	unsigned int len;
+	int err;
+again:
+	if (!app->pdu) {
+		err = garp_pdu_init(app);
+		if (err < 0)
+			return err;
+	}
+
+	if (garp_cb(app->pdu)->cur_type != attr->type) {
+		if (garp_cb(app->pdu)->cur_type &&
+		    garp_pdu_append_end_mark(app) < 0)
+			goto queue;
+		if (garp_pdu_append_msg(app, attr->type) < 0)
+			goto queue;
+	}
+
+	len = sizeof(*ga) + attr->dlen;
+	if (skb_tailroom(app->pdu) < len)
+		goto queue;
+	ga = (struct garp_attr_hdr *)__skb_put(app->pdu, len);
+	ga->len   = len;
+	ga->event = event;
+	memcpy(ga->data, attr->data, attr->dlen);
+	return 0;
+
+queue:
+	garp_pdu_queue(app);
+	goto again;
+}
+
+static void garp_attr_event(struct garp_applicant *app,
+			    struct garp_attr *attr, enum garp_event event)
+{
+	enum garp_applicant_state state;
+
+	state = garp_applicant_state_table[attr->state][event].state;
+	if (state == GARP_APPLICANT_INVALID)
+		return;
+
+	switch (garp_applicant_state_table[attr->state][event].action) {
+	case GARP_ACTION_NONE:
+		break;
+	case GARP_ACTION_S_JOIN_IN:
+		garp_pdu_append_attr(app, attr, GARP_JOIN_IN);
+		break;
+	case GARP_ACTION_S_LEAVE_EMPTY:
+		garp_pdu_append_attr(app, attr, GARP_LEAVE_EMPTY);
+		/* As a pure applicant, sending a leave message implies that
+		 * the attribute was unregistered and can be destroyed. */
+		garp_attr_destroy(app, attr);
+		return;
+	default:
+		WARN_ON(1);
+	}
+
+	attr->state = state;
+}
+
+int garp_request_join(const struct net_device *dev,
+		      const struct garp_application *appl,
+		      const void *data, u8 len, u8 type)
+{
+	struct garp_port *port = dev->garp_port;
+	struct garp_applicant *app = port->applicants[appl->type];
+	struct garp_attr *attr;
+
+	spin_lock_bh(&app->lock);
+	attr = garp_attr_create(app, data, len, type);
+	if (!attr) {
+		spin_unlock_bh(&app->lock);
+		return -ENOMEM;
+	}
+	garp_attr_event(app, attr, GARP_EVENT_REQ_JOIN);
+	spin_unlock_bh(&app->lock);
+	return 0;
+}
+EXPORT_SYMBOL_GPL(garp_request_join);
+
+void garp_request_leave(const struct net_device *dev,
+			const struct garp_application *appl,
+			const void *data, u8 len, u8 type)
+{
+	struct garp_port *port = dev->garp_port;
+	struct garp_applicant *app = port->applicants[appl->type];
+	struct garp_attr *attr;
+
+	spin_lock_bh(&app->lock);
+	attr = garp_attr_lookup(app, data, len, type);
+	if (!attr) {
+		spin_unlock_bh(&app->lock);
+		return;
+	}
+	garp_attr_event(app, attr, GARP_EVENT_REQ_LEAVE);
+	spin_unlock_bh(&app->lock);
+}
+EXPORT_SYMBOL_GPL(garp_request_leave);
+
+static void garp_gid_event(struct garp_applicant *app, enum garp_event event)
+{
+	struct rb_node *node, *next;
+	struct garp_attr *attr;
+
+	for (node = rb_first(&app->gid);
+	     next = node ? rb_next(node) : NULL, node != NULL;
+	     node = next) {
+		attr = rb_entry(node, struct garp_attr, node);
+		garp_attr_event(app, attr, event);
+	}
+}
+
+static void garp_join_timer_arm(struct garp_applicant *app)
+{
+	unsigned long delay;
+
+	delay = (u64)msecs_to_jiffies(garp_join_time) * net_random() >> 32;
+	mod_timer(&app->join_timer, jiffies + delay);
+}
+
+static void garp_join_timer(unsigned long data)
+{
+	struct garp_applicant *app = (struct garp_applicant *)data;
+
+	spin_lock(&app->lock);
+	garp_gid_event(app, GARP_EVENT_TRANSMIT_PDU);
+	garp_pdu_queue(app);
+	spin_unlock(&app->lock);
+
+	garp_queue_xmit(app);
+	garp_join_timer_arm(app);
+}
+
+static int garp_pdu_parse_end_mark(struct sk_buff *skb)
+{
+	if (!pskb_may_pull(skb, sizeof(u8)))
+		return -1;
+	if (*skb->data == GARP_END_MARK) {
+		skb_pull(skb, sizeof(u8));
+		return -1;
+	}
+	return 0;
+}
+
+static int garp_pdu_parse_attr(struct garp_applicant *app, struct sk_buff *skb,
+			       u8 attrtype)
+{
+	const struct garp_attr_hdr *ga;
+	struct garp_attr *attr;
+	enum garp_event event;
+	unsigned int dlen;
+
+	if (!pskb_may_pull(skb, sizeof(*ga)))
+		return -1;
+	ga = (struct garp_attr_hdr *)skb->data;
+	if (ga->len < sizeof(*ga))
+		return -1;
+
+	if (!pskb_may_pull(skb, ga->len))
+		return -1;
+	skb_pull(skb, ga->len);
+	dlen = sizeof(*ga) - ga->len;
+
+	if (attrtype > app->app->maxattr)
+		return 0;
+
+	switch (ga->event) {
+	case GARP_LEAVE_ALL:
+		if (dlen != 0)
+			return -1;
+		garp_gid_event(app, GARP_EVENT_R_LEAVE_EMPTY);
+		return 0;
+	case GARP_JOIN_EMPTY:
+		event = GARP_EVENT_R_JOIN_EMPTY;
+		break;
+	case GARP_JOIN_IN:
+		event = GARP_EVENT_R_JOIN_IN;
+		break;
+	case GARP_LEAVE_EMPTY:
+		event = GARP_EVENT_R_LEAVE_EMPTY;
+		break;
+	case GARP_EMPTY:
+		event = GARP_EVENT_R_EMPTY;
+		break;
+	default:
+		return 0;
+	}
+
+	if (dlen == 0)
+		return -1;
+	attr = garp_attr_lookup(app, ga->data, dlen, attrtype);
+	if (attr == NULL)
+		return 0;
+	garp_attr_event(app, attr, event);
+	return 0;
+}
+
+static int garp_pdu_parse_msg(struct garp_applicant *app, struct sk_buff *skb)
+{
+	const struct garp_msg_hdr *gm;
+
+	if (!pskb_may_pull(skb, sizeof(*gm)))
+		return -1;
+	gm = (struct garp_msg_hdr *)skb->data;
+	if (gm->attrtype == 0)
+		return -1;
+	skb_pull(skb, sizeof(*gm));
+
+	while (skb->len > 0) {
+		if (garp_pdu_parse_attr(app, skb, gm->attrtype) < 0)
+			return -1;
+		if (garp_pdu_parse_end_mark(skb) < 0)
+			break;
+	}
+	return 0;
+}
+
+static void garp_pdu_rcv(const struct stp_proto *proto, struct sk_buff *skb,
+			 struct net_device *dev)
+{
+	struct garp_application *appl = proto->data;
+	struct garp_port *port;
+	struct garp_applicant *app;
+	const struct garp_pdu_hdr *gp;
+
+	port = rcu_dereference(dev->garp_port);
+	if (!port)
+		goto err;
+	app = rcu_dereference(port->applicants[appl->type]);
+	if (!app)
+		goto err;
+
+	if (!pskb_may_pull(skb, sizeof(*gp)))
+		goto err;
+	gp = (struct garp_pdu_hdr *)skb->data;
+	if (get_unaligned(&gp->protocol) != htons(GARP_PROTOCOL_ID))
+		goto err;
+	skb_pull(skb, sizeof(*gp));
+
+	spin_lock(&app->lock);
+	while (skb->len > 0) {
+		if (garp_pdu_parse_msg(app, skb) < 0)
+			break;
+		if (garp_pdu_parse_end_mark(skb) < 0)
+			break;
+	}
+	spin_unlock(&app->lock);
+err:
+	kfree_skb(skb);
+}
+
+static int garp_init_port(struct net_device *dev)
+{
+	struct garp_port *port;
+
+	port = kzalloc(sizeof(*port), GFP_KERNEL);
+	if (!port)
+		return -ENOMEM;
+	rcu_assign_pointer(dev->garp_port, port);
+	return 0;
+}
+
+static void garp_release_port(struct net_device *dev)
+{
+	struct garp_port *port = dev->garp_port;
+	unsigned int i;
+
+	for (i = 0; i <= GARP_APPLICATION_MAX; i++) {
+		if (port->applicants[i])
+			return;
+	}
+	rcu_assign_pointer(dev->garp_port, NULL);
+	synchronize_rcu();
+	kfree(port);
+}
+
+int garp_init_applicant(struct net_device *dev, struct garp_application *appl)
+{
+	struct garp_applicant *app;
+	int err;
+
+	ASSERT_RTNL();
+
+	if (!dev->garp_port) {
+		err = garp_init_port(dev);
+		if (err < 0)
+			goto err1;
+	}
+
+	err = -ENOMEM;
+	app = kzalloc(sizeof(*app), GFP_KERNEL);
+	if (!app)
+		goto err2;
+
+	err = dev_mc_add(dev, appl->proto.group_address, ETH_ALEN, 0);
+	if (err < 0)
+		goto err3;
+
+	app->dev = dev;
+	app->app = appl;
+	app->gid = RB_ROOT;
+	spin_lock_init(&app->lock);
+	skb_queue_head_init(&app->queue);
+	rcu_assign_pointer(dev->garp_port->applicants[appl->type], app);
+	setup_timer(&app->join_timer, garp_join_timer, (unsigned long)app);
+	garp_join_timer_arm(app);
+	return 0;
+
+err3:
+	kfree(app);
+err2:
+	garp_release_port(dev);
+err1:
+	return err;
+}
+EXPORT_SYMBOL_GPL(garp_init_applicant);
+
+void garp_uninit_applicant(struct net_device *dev, struct garp_application *appl)
+{
+	struct garp_port *port = dev->garp_port;
+	struct garp_applicant *app = port->applicants[appl->type];
+
+	ASSERT_RTNL();
+
+	rcu_assign_pointer(port->applicants[appl->type], NULL);
+	synchronize_rcu();
+
+	/* Delete timer and generate a final TRANSMIT_PDU event to flush out
+	 * all pending messages before the applicant is gone. */
+	del_timer_sync(&app->join_timer);
+	garp_gid_event(app, GARP_EVENT_TRANSMIT_PDU);
+	garp_pdu_queue(app);
+	garp_queue_xmit(app);
+
+	dev_mc_delete(dev, appl->proto.group_address, ETH_ALEN, 0);
+	kfree(app);
+	garp_release_port(dev);
+}
+EXPORT_SYMBOL_GPL(garp_uninit_applicant);
+
+int garp_register_application(struct garp_application *appl)
+{
+	appl->proto.rcv = garp_pdu_rcv;
+	appl->proto.data = appl;
+	return stp_proto_register(&appl->proto);
+}
+EXPORT_SYMBOL_GPL(garp_register_application);
+
+void garp_unregister_application(struct garp_application *appl)
+{
+	stp_proto_unregister(&appl->proto);
+}
+EXPORT_SYMBOL_GPL(garp_unregister_application);
-- 
cgit v1.2.3


From 70c03b49b80ba3634958acc31853771019c0ebd3 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Sat, 5 Jul 2008 21:26:57 -0700
Subject: vlan: Add GVRP support

Add GVRP support for dynamically registering VLANs with switches.

By default GVRP is disabled because we only support the applicant-only
participant model, which means it should not be enabled on vlans that
are members of a bridge. Since there is currently no way to cleanly
determine that, the user is responsible for enabling it.

The code is pretty small and low impact, its wrapped in a config
option though because it depends on the GARP implementation and
the STP core.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/if_vlan.h  |  1 +
 include/net/garp.h       |  1 +
 net/8021q/Kconfig        | 10 ++++++++
 net/8021q/Makefile       |  9 +++----
 net/8021q/vlan.c         | 23 ++++++++++++++---
 net/8021q/vlan.h         | 16 ++++++++++++
 net/8021q/vlan_dev.c     | 18 +++++++++++--
 net/8021q/vlan_gvrp.c    | 66 ++++++++++++++++++++++++++++++++++++++++++++++++
 net/8021q/vlan_netlink.c |  3 ++-
 9 files changed, 134 insertions(+), 13 deletions(-)
 create mode 100644 net/8021q/vlan_gvrp.c

(limited to 'include')

diff --git a/include/linux/if_vlan.h b/include/linux/if_vlan.h
index 15ace02b7b24..5190452ac7dc 100644
--- a/include/linux/if_vlan.h
+++ b/include/linux/if_vlan.h
@@ -402,6 +402,7 @@ enum vlan_ioctl_cmds {
 
 enum vlan_flags {
 	VLAN_FLAG_REORDER_HDR	= 0x1,
+	VLAN_FLAG_GVRP		= 0x2,
 };
 
 enum vlan_name_types {
diff --git a/include/net/garp.h b/include/net/garp.h
index 73c772395f5b..825f172caba9 100644
--- a/include/net/garp.h
+++ b/include/net/garp.h
@@ -84,6 +84,7 @@ struct garp_attr {
 };
 
 enum garp_applications {
+	GARP_APPLICATION_GVRP,
 	__GARP_APPLICATION_MAX
 };
 #define GARP_APPLICATION_MAX	(__GARP_APPLICATION_MAX - 1)
diff --git a/net/8021q/Kconfig b/net/8021q/Kconfig
index c4a382e450e2..fa073a54963e 100644
--- a/net/8021q/Kconfig
+++ b/net/8021q/Kconfig
@@ -17,3 +17,13 @@ config VLAN_8021Q
 	  will be called 8021q.
 
 	  If unsure, say N.
+
+config VLAN_8021Q_GVRP
+	bool "GVRP (GARP VLAN Registration Protocol) support"
+	depends on VLAN_8021Q
+	select GARP
+	help
+	  Select this to enable GVRP end-system support. GVRP is used for
+	  automatic propagation of registered VLANs to switches.
+
+	  If unsure, say N.
diff --git a/net/8021q/Makefile b/net/8021q/Makefile
index 10ca7f486c3a..3006e9ed7b08 100644
--- a/net/8021q/Makefile
+++ b/net/8021q/Makefile
@@ -4,9 +4,6 @@
 
 obj-$(CONFIG_VLAN_8021Q) += 8021q.o
 
-8021q-objs := vlan.o vlan_dev.o vlan_netlink.o
-
-ifeq ($(CONFIG_PROC_FS),y)
-8021q-objs += vlanproc.o
-endif
-
+8021q-y				:= vlan.o vlan_dev.o vlan_netlink.o
+8021q-$(CONFIG_VLAN_8021Q_GVRP)	+= vlan_gvrp.o
+8021q-$(CONFIG_PROC_FS)		+= vlanproc.o
diff --git a/net/8021q/vlan.c b/net/8021q/vlan.c
index 8cae2daeb1cc..b529110c9355 100644
--- a/net/8021q/vlan.c
+++ b/net/8021q/vlan.c
@@ -169,6 +169,8 @@ void unregister_vlan_dev(struct net_device *dev)
 
 	/* If the group is now empty, kill off the group. */
 	if (grp->nr_vlans == 0) {
+		vlan_gvrp_uninit_applicant(real_dev);
+
 		if (real_dev->features & NETIF_F_HW_VLAN_RX)
 			real_dev->vlan_rx_register(real_dev, NULL);
 
@@ -249,15 +251,18 @@ int register_vlan_dev(struct net_device *dev)
 		ngrp = grp = vlan_group_alloc(real_dev);
 		if (!grp)
 			return -ENOBUFS;
+		err = vlan_gvrp_init_applicant(real_dev);
+		if (err < 0)
+			goto out_free_group;
 	}
 
 	err = vlan_group_prealloc_vid(grp, vlan_id);
 	if (err < 0)
-		goto out_free_group;
+		goto out_uninit_applicant;
 
 	err = register_netdevice(dev);
 	if (err < 0)
-		goto out_free_group;
+		goto out_uninit_applicant;
 
 	/* Account for reference in struct vlan_dev_info */
 	dev_hold(real_dev);
@@ -278,6 +283,9 @@ int register_vlan_dev(struct net_device *dev)
 
 	return 0;
 
+out_uninit_applicant:
+	if (ngrp)
+		vlan_gvrp_uninit_applicant(real_dev);
 out_free_group:
 	if (ngrp)
 		vlan_group_free(ngrp);
@@ -713,14 +721,20 @@ static int __init vlan_proto_init(void)
 	if (err < 0)
 		goto err2;
 
-	err = vlan_netlink_init();
+	err = vlan_gvrp_init();
 	if (err < 0)
 		goto err3;
 
+	err = vlan_netlink_init();
+	if (err < 0)
+		goto err4;
+
 	dev_add_pack(&vlan_packet_type);
 	vlan_ioctl_set(vlan_ioctl_handler);
 	return 0;
 
+err4:
+	vlan_gvrp_uninit();
 err3:
 	unregister_netdevice_notifier(&vlan_notifier_block);
 err2:
@@ -745,8 +759,9 @@ static void __exit vlan_cleanup_module(void)
 		BUG_ON(!hlist_empty(&vlan_group_hash[i]));
 
 	unregister_pernet_gen_device(vlan_net_id, &vlan_net_ops);
-
 	synchronize_net();
+
+	vlan_gvrp_uninit();
 }
 
 module_init(vlan_proto_init);
diff --git a/net/8021q/vlan.h b/net/8021q/vlan.h
index 639e2544a804..097b2e04c928 100644
--- a/net/8021q/vlan.h
+++ b/net/8021q/vlan.h
@@ -37,6 +37,22 @@ void vlan_setup(struct net_device *dev);
 int register_vlan_dev(struct net_device *dev);
 void unregister_vlan_dev(struct net_device *dev);
 
+#ifdef CONFIG_VLAN_8021Q_GVRP
+extern int vlan_gvrp_request_join(const struct net_device *dev);
+extern void vlan_gvrp_request_leave(const struct net_device *dev);
+extern int vlan_gvrp_init_applicant(struct net_device *dev);
+extern void vlan_gvrp_uninit_applicant(struct net_device *dev);
+extern int vlan_gvrp_init(void);
+extern void vlan_gvrp_uninit(void);
+#else
+static inline int vlan_gvrp_request_join(const struct net_device *dev) { return 0; }
+static inline void vlan_gvrp_request_leave(const struct net_device *dev) {}
+static inline int vlan_gvrp_init_applicant(struct net_device *dev) { return 0; }
+static inline void vlan_gvrp_uninit_applicant(struct net_device *dev) {}
+static inline int vlan_gvrp_init(void) { return 0; }
+static inline void vlan_gvrp_uninit(void) {}
+#endif
+
 int vlan_netlink_init(void);
 void vlan_netlink_fini(void);
 
diff --git a/net/8021q/vlan_dev.c b/net/8021q/vlan_dev.c
index 76c665cdab66..a0617bf7cec6 100644
--- a/net/8021q/vlan_dev.c
+++ b/net/8021q/vlan_dev.c
@@ -512,10 +512,17 @@ int vlan_dev_change_flags(const struct net_device *dev, u32 flags, u32 mask)
 	struct vlan_dev_info *vlan = vlan_dev_info(dev);
 	u32 old_flags = vlan->flags;
 
-	if (mask & ~VLAN_FLAG_REORDER_HDR)
+	if (mask & ~(VLAN_FLAG_REORDER_HDR | VLAN_FLAG_GVRP))
 		return -EINVAL;
 
 	vlan->flags = (old_flags & ~mask) | (flags & mask);
+
+	if (netif_running(dev) && (vlan->flags ^ old_flags) & VLAN_FLAG_GVRP) {
+		if (vlan->flags & VLAN_FLAG_GVRP)
+			vlan_gvrp_request_join(dev);
+		else
+			vlan_gvrp_request_leave(dev);
+	}
 	return 0;
 }
 
@@ -550,12 +557,19 @@ static int vlan_dev_open(struct net_device *dev)
 	if (dev->flags & IFF_PROMISC)
 		dev_set_promiscuity(real_dev, 1);
 
+	if (vlan->flags & VLAN_FLAG_GVRP)
+		vlan_gvrp_request_join(dev);
+
 	return 0;
 }
 
 static int vlan_dev_stop(struct net_device *dev)
 {
-	struct net_device *real_dev = vlan_dev_info(dev)->real_dev;
+	struct vlan_dev_info *vlan = vlan_dev_info(dev);
+	struct net_device *real_dev = vlan->real_dev;
+
+	if (vlan->flags & VLAN_FLAG_GVRP)
+		vlan_gvrp_request_leave(dev);
 
 	dev_mc_unsync(real_dev, dev);
 	dev_unicast_unsync(real_dev, dev);
diff --git a/net/8021q/vlan_gvrp.c b/net/8021q/vlan_gvrp.c
new file mode 100644
index 000000000000..db9781608362
--- /dev/null
+++ b/net/8021q/vlan_gvrp.c
@@ -0,0 +1,66 @@
+/*
+ * 	IEEE 802.1Q GARP VLAN Registration Protocol (GVRP)
+ *
+ * 	Copyright (c) 2008 Patrick McHardy <kaber@trash.net>
+ *
+ *	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/types.h>
+#include <linux/if_vlan.h>
+#include <net/garp.h>
+#include "vlan.h"
+
+#define GARP_GVRP_ADDRESS	{ 0x01, 0x80, 0xc2, 0x00, 0x00, 0x21 }
+
+enum gvrp_attributes {
+	GVRP_ATTR_INVALID,
+	GVRP_ATTR_VID,
+	__GVRP_ATTR_MAX
+};
+#define GVRP_ATTR_MAX	(__GVRP_ATTR_MAX - 1)
+
+static struct garp_application vlan_gvrp_app __read_mostly = {
+	.proto.group_address	= GARP_GVRP_ADDRESS,
+	.maxattr		= GVRP_ATTR_MAX,
+	.type			= GARP_APPLICATION_GVRP,
+};
+
+int vlan_gvrp_request_join(const struct net_device *dev)
+{
+	const struct vlan_dev_info *vlan = vlan_dev_info(dev);
+	__be16 vid = htons(vlan->vlan_id);
+
+	return garp_request_join(vlan->real_dev, &vlan_gvrp_app,
+				 &vid, sizeof(vid), GVRP_ATTR_VID);
+}
+
+void vlan_gvrp_request_leave(const struct net_device *dev)
+{
+	const struct vlan_dev_info *vlan = vlan_dev_info(dev);
+	__be16 vid = htons(vlan->vlan_id);
+
+	garp_request_leave(vlan->real_dev, &vlan_gvrp_app,
+			   &vid, sizeof(vid), GVRP_ATTR_VID);
+}
+
+int vlan_gvrp_init_applicant(struct net_device *dev)
+{
+	return garp_init_applicant(dev, &vlan_gvrp_app);
+}
+
+void vlan_gvrp_uninit_applicant(struct net_device *dev)
+{
+	garp_uninit_applicant(dev, &vlan_gvrp_app);
+}
+
+int __init vlan_gvrp_init(void)
+{
+	return garp_register_application(&vlan_gvrp_app);
+}
+
+void vlan_gvrp_uninit(void)
+{
+	garp_unregister_application(&vlan_gvrp_app);
+}
diff --git a/net/8021q/vlan_netlink.c b/net/8021q/vlan_netlink.c
index fd7cb195d53f..e9c91dcecc9b 100644
--- a/net/8021q/vlan_netlink.c
+++ b/net/8021q/vlan_netlink.c
@@ -59,7 +59,8 @@ static int vlan_validate(struct nlattr *tb[], struct nlattr *data[])
 	}
 	if (data[IFLA_VLAN_FLAGS]) {
 		flags = nla_data(data[IFLA_VLAN_FLAGS]);
-		if ((flags->flags & flags->mask) & ~VLAN_FLAG_REORDER_HDR)
+		if ((flags->flags & flags->mask) &
+		    ~(VLAN_FLAG_REORDER_HDR | VLAN_FLAG_GVRP))
 			return -EINVAL;
 	}
 
-- 
cgit v1.2.3


From 6fe1c7a5556807e9d7154a2d2fb938d8a9e47e5f Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Sat, 5 Jul 2008 23:21:31 -0700
Subject: net-sched: add dynamically sized qdisc class hash helpers

Currently all qdiscs which allow to create classes uses a fixed sized hash
table with size 16 to hash the classes. This causes a large bottleneck
when using thousands of classes and unbound filters.

Add helpers for dynamically sized class hashes to fix this. The following
patches will convert the qdiscs to use them.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/sch_generic.h |  42 +++++++++++++++++++
 net/sched/sch_api.c       | 104 ++++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 146 insertions(+)

(limited to 'include')

diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h
index a87fc0312edc..073f2580b83b 100644
--- a/include/net/sch_generic.h
+++ b/include/net/sch_generic.h
@@ -167,6 +167,48 @@ extern void qdisc_unlock_tree(struct net_device *dev);
 extern struct Qdisc noop_qdisc;
 extern struct Qdisc_ops noop_qdisc_ops;
 
+struct Qdisc_class_common
+{
+	u32			classid;
+	struct hlist_node	hnode;
+};
+
+struct Qdisc_class_hash
+{
+	struct hlist_head	*hash;
+	unsigned int		hashsize;
+	unsigned int		hashmask;
+	unsigned int		hashelems;
+};
+
+static inline unsigned int qdisc_class_hash(u32 id, u32 mask)
+{
+	id ^= id >> 8;
+	id ^= id >> 4;
+	return id & mask;
+}
+
+static inline struct Qdisc_class_common *
+qdisc_class_find(struct Qdisc_class_hash *hash, u32 id)
+{
+	struct Qdisc_class_common *cl;
+	struct hlist_node *n;
+	unsigned int h;
+
+	h = qdisc_class_hash(id, hash->hashmask);
+	hlist_for_each_entry(cl, n, &hash->hash[h], hnode) {
+		if (cl->classid == id)
+			return cl;
+	}
+	return NULL;
+}
+
+extern int qdisc_class_hash_init(struct Qdisc_class_hash *);
+extern void qdisc_class_hash_insert(struct Qdisc_class_hash *, struct Qdisc_class_common *);
+extern void qdisc_class_hash_remove(struct Qdisc_class_hash *, struct Qdisc_class_common *);
+extern void qdisc_class_hash_grow(struct Qdisc *, struct Qdisc_class_hash *);
+extern void qdisc_class_hash_destroy(struct Qdisc_class_hash *);
+
 extern void dev_init_scheduler(struct net_device *dev);
 extern void dev_shutdown(struct net_device *dev);
 extern void dev_activate(struct net_device *dev);
diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c
index 10f01ad04380..e9ebc7af049e 100644
--- a/net/sched/sch_api.c
+++ b/net/sched/sch_api.c
@@ -316,6 +316,110 @@ void qdisc_watchdog_cancel(struct qdisc_watchdog *wd)
 }
 EXPORT_SYMBOL(qdisc_watchdog_cancel);
 
+struct hlist_head *qdisc_class_hash_alloc(unsigned int n)
+{
+	unsigned int size = n * sizeof(struct hlist_head), i;
+	struct hlist_head *h;
+
+	if (size <= PAGE_SIZE)
+		h = kmalloc(size, GFP_KERNEL);
+	else
+		h = (struct hlist_head *)
+			__get_free_pages(GFP_KERNEL, get_order(size));
+
+	if (h != NULL) {
+		for (i = 0; i < n; i++)
+			INIT_HLIST_HEAD(&h[i]);
+	}
+	return h;
+}
+
+static void qdisc_class_hash_free(struct hlist_head *h, unsigned int n)
+{
+	unsigned int size = n * sizeof(struct hlist_head);
+
+	if (size <= PAGE_SIZE)
+		kfree(h);
+	else
+		free_pages((unsigned long)h, get_order(size));
+}
+
+void qdisc_class_hash_grow(struct Qdisc *sch, struct Qdisc_class_hash *clhash)
+{
+	struct Qdisc_class_common *cl;
+	struct hlist_node *n, *next;
+	struct hlist_head *nhash, *ohash;
+	unsigned int nsize, nmask, osize;
+	unsigned int i, h;
+
+	/* Rehash when load factor exceeds 0.75 */
+	if (clhash->hashelems * 4 <= clhash->hashsize * 3)
+		return;
+	nsize = clhash->hashsize * 2;
+	nmask = nsize - 1;
+	nhash = qdisc_class_hash_alloc(nsize);
+	if (nhash == NULL)
+		return;
+
+	ohash = clhash->hash;
+	osize = clhash->hashsize;
+
+	sch_tree_lock(sch);
+	for (i = 0; i < osize; i++) {
+		hlist_for_each_entry_safe(cl, n, next, &ohash[i], hnode) {
+			h = qdisc_class_hash(cl->classid, nmask);
+			hlist_add_head(&cl->hnode, &nhash[h]);
+		}
+	}
+	clhash->hash     = nhash;
+	clhash->hashsize = nsize;
+	clhash->hashmask = nmask;
+	sch_tree_unlock(sch);
+
+	qdisc_class_hash_free(ohash, osize);
+}
+EXPORT_SYMBOL(qdisc_class_hash_grow);
+
+int qdisc_class_hash_init(struct Qdisc_class_hash *clhash)
+{
+	unsigned int size = 4;
+
+	clhash->hash = qdisc_class_hash_alloc(size);
+	if (clhash->hash == NULL)
+		return -ENOMEM;
+	clhash->hashsize  = size;
+	clhash->hashmask  = size - 1;
+	clhash->hashelems = 0;
+	return 0;
+}
+EXPORT_SYMBOL(qdisc_class_hash_init);
+
+void qdisc_class_hash_destroy(struct Qdisc_class_hash *clhash)
+{
+	qdisc_class_hash_free(clhash->hash, clhash->hashsize);
+}
+EXPORT_SYMBOL(qdisc_class_hash_destroy);
+
+void qdisc_class_hash_insert(struct Qdisc_class_hash *clhash,
+			     struct Qdisc_class_common *cl)
+{
+	unsigned int h;
+
+	INIT_HLIST_NODE(&cl->hnode);
+	h = qdisc_class_hash(cl->classid, clhash->hashmask);
+	hlist_add_head(&cl->hnode, &clhash->hash[h]);
+	clhash->hashelems++;
+}
+EXPORT_SYMBOL(qdisc_class_hash_insert);
+
+void qdisc_class_hash_remove(struct Qdisc_class_hash *clhash,
+			     struct Qdisc_class_common *cl)
+{
+	hlist_del(&cl->hnode);
+	clhash->hashelems--;
+}
+EXPORT_SYMBOL(qdisc_class_hash_remove);
+
 /* Allocate an unique handle from space managed by kernel */
 
 static u32 qdisc_alloc_handle(struct net_device *dev)
-- 
cgit v1.2.3


From fb0305ce1b03f6ff17f84f2c63daccecb45f2805 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Sat, 5 Jul 2008 23:40:21 -0700
Subject: net-sched: consolidate default fifo qdisc setup

Signed-off-by: Patrick McHardy <kaber@trash.net>
Acked-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/pkt_sched.h |  4 ++++
 net/sched/sch_fifo.c    | 42 ++++++++++++++++++++++++++++++++++++++++++
 net/sched/sch_netem.c   | 24 +-----------------------
 net/sched/sch_red.c     | 33 +++------------------------------
 net/sched/sch_tbf.c     | 33 ++++-----------------------------
 5 files changed, 54 insertions(+), 82 deletions(-)

(limited to 'include')

diff --git a/include/net/pkt_sched.h b/include/net/pkt_sched.h
index 46fb4d80c74a..8e3a0c4e9d97 100644
--- a/include/net/pkt_sched.h
+++ b/include/net/pkt_sched.h
@@ -72,6 +72,10 @@ extern void qdisc_watchdog_cancel(struct qdisc_watchdog *wd);
 extern struct Qdisc_ops pfifo_qdisc_ops;
 extern struct Qdisc_ops bfifo_qdisc_ops;
 
+extern int fifo_set_limit(struct Qdisc *q, unsigned int limit);
+extern struct Qdisc *fifo_create_dflt(struct Qdisc *sch, struct Qdisc_ops *ops,
+				      unsigned int limit);
+
 extern int register_qdisc(struct Qdisc_ops *qops);
 extern int unregister_qdisc(struct Qdisc_ops *qops);
 extern struct Qdisc *qdisc_lookup(struct net_device *dev, u32 handle);
diff --git a/net/sched/sch_fifo.c b/net/sched/sch_fifo.c
index 95ed48221652..82d7d7bbbb16 100644
--- a/net/sched/sch_fifo.c
+++ b/net/sched/sch_fifo.c
@@ -107,3 +107,45 @@ struct Qdisc_ops bfifo_qdisc_ops __read_mostly = {
 	.owner		=	THIS_MODULE,
 };
 EXPORT_SYMBOL(bfifo_qdisc_ops);
+
+/* Pass size change message down to embedded FIFO */
+int fifo_set_limit(struct Qdisc *q, unsigned int limit)
+{
+	struct nlattr *nla;
+	int ret = -ENOMEM;
+
+	/* Hack to avoid sending change message to non-FIFO */
+	if (strncmp(q->ops->id + 1, "fifo", 4) != 0)
+		return 0;
+
+	nla = kmalloc(nla_attr_size(sizeof(struct tc_fifo_qopt)), GFP_KERNEL);
+	if (nla) {
+		nla->nla_type = RTM_NEWQDISC;
+		nla->nla_len = nla_attr_size(sizeof(struct tc_fifo_qopt));
+		((struct tc_fifo_qopt *)nla_data(nla))->limit = limit;
+
+		ret = q->ops->change(q, nla);
+		kfree(nla);
+	}
+	return ret;
+}
+EXPORT_SYMBOL(fifo_set_limit);
+
+struct Qdisc *fifo_create_dflt(struct Qdisc *sch, struct Qdisc_ops *ops,
+			       unsigned int limit)
+{
+	struct Qdisc *q;
+	int err = -ENOMEM;
+
+	q = qdisc_create_dflt(sch->dev, ops, TC_H_MAKE(sch->handle, 1));
+	if (q) {
+		err = fifo_set_limit(q, limit);
+		if (err < 0) {
+			qdisc_destroy(q);
+			q = NULL;
+		}
+	}
+
+	return q ? : ERR_PTR(err);
+}
+EXPORT_SYMBOL(fifo_create_dflt);
diff --git a/net/sched/sch_netem.c b/net/sched/sch_netem.c
index c9c649b26eaa..24697667247c 100644
--- a/net/sched/sch_netem.c
+++ b/net/sched/sch_netem.c
@@ -310,28 +310,6 @@ static void netem_reset(struct Qdisc *sch)
 	qdisc_watchdog_cancel(&q->watchdog);
 }
 
-/* Pass size change message down to embedded FIFO */
-static int set_fifo_limit(struct Qdisc *q, int limit)
-{
-	struct nlattr *nla;
-	int ret = -ENOMEM;
-
-	/* Hack to avoid sending change message to non-FIFO */
-	if (strncmp(q->ops->id + 1, "fifo", 4) != 0)
-		return 0;
-
-	nla = kmalloc(nla_attr_size(sizeof(struct tc_fifo_qopt)), GFP_KERNEL);
-	if (nla) {
-		nla->nla_type = RTM_NEWQDISC;
-		nla->nla_len = nla_attr_size(sizeof(struct tc_fifo_qopt));
-		((struct tc_fifo_qopt *)nla_data(nla))->limit = limit;
-
-		ret = q->ops->change(q, nla);
-		kfree(nla);
-	}
-	return ret;
-}
-
 /*
  * Distribution data is a variable size payload containing
  * signed 16 bit values.
@@ -416,7 +394,7 @@ static int netem_change(struct Qdisc *sch, struct nlattr *opt)
 	if (ret < 0)
 		return ret;
 
-	ret = set_fifo_limit(q->qdisc, qopt->limit);
+	ret = fifo_set_limit(q->qdisc, qopt->limit);
 	if (ret) {
 		pr_debug("netem: can't set fifo limit\n");
 		return ret;
diff --git a/net/sched/sch_red.c b/net/sched/sch_red.c
index 5c569853b9c0..77098acf0adc 100644
--- a/net/sched/sch_red.c
+++ b/net/sched/sch_red.c
@@ -174,33 +174,6 @@ static void red_destroy(struct Qdisc *sch)
 	qdisc_destroy(q->qdisc);
 }
 
-static struct Qdisc *red_create_dflt(struct Qdisc *sch, u32 limit)
-{
-	struct Qdisc *q;
-	struct nlattr *nla;
-	int ret;
-
-	q = qdisc_create_dflt(sch->dev, &bfifo_qdisc_ops,
-			      TC_H_MAKE(sch->handle, 1));
-	if (q) {
-		nla = kmalloc(nla_attr_size(sizeof(struct tc_fifo_qopt)),
-			      GFP_KERNEL);
-		if (nla) {
-			nla->nla_type = RTM_NEWQDISC;
-			nla->nla_len = nla_attr_size(sizeof(struct tc_fifo_qopt));
-			((struct tc_fifo_qopt *)nla_data(nla))->limit = limit;
-
-			ret = q->ops->change(q, nla);
-			kfree(nla);
-
-			if (ret == 0)
-				return q;
-		}
-		qdisc_destroy(q);
-	}
-	return NULL;
-}
-
 static const struct nla_policy red_policy[TCA_RED_MAX + 1] = {
 	[TCA_RED_PARMS]	= { .len = sizeof(struct tc_red_qopt) },
 	[TCA_RED_STAB]	= { .len = RED_STAB_SIZE },
@@ -228,9 +201,9 @@ static int red_change(struct Qdisc *sch, struct nlattr *opt)
 	ctl = nla_data(tb[TCA_RED_PARMS]);
 
 	if (ctl->limit > 0) {
-		child = red_create_dflt(sch, ctl->limit);
-		if (child == NULL)
-			return -ENOMEM;
+		child = fifo_create_dflt(sch, &bfifo_qdisc_ops, ctl->limit);
+		if (IS_ERR(child))
+			return PTR_ERR(child);
 	}
 
 	sch_tree_lock(sch);
diff --git a/net/sched/sch_tbf.c b/net/sched/sch_tbf.c
index 0b7d78f59d8c..444c227fcb6b 100644
--- a/net/sched/sch_tbf.c
+++ b/net/sched/sch_tbf.c
@@ -242,34 +242,6 @@ static void tbf_reset(struct Qdisc* sch)
 	qdisc_watchdog_cancel(&q->watchdog);
 }
 
-static struct Qdisc *tbf_create_dflt_qdisc(struct Qdisc *sch, u32 limit)
-{
-	struct Qdisc *q;
-	struct nlattr *nla;
-	int ret;
-
-	q = qdisc_create_dflt(sch->dev, &bfifo_qdisc_ops,
-			      TC_H_MAKE(sch->handle, 1));
-	if (q) {
-		nla = kmalloc(nla_attr_size(sizeof(struct tc_fifo_qopt)),
-			      GFP_KERNEL);
-		if (nla) {
-			nla->nla_type = RTM_NEWQDISC;
-			nla->nla_len = nla_attr_size(sizeof(struct tc_fifo_qopt));
-			((struct tc_fifo_qopt *)nla_data(nla))->limit = limit;
-
-			ret = q->ops->change(q, nla);
-			kfree(nla);
-
-			if (ret == 0)
-				return q;
-		}
-		qdisc_destroy(q);
-	}
-
-	return NULL;
-}
-
 static const struct nla_policy tbf_policy[TCA_TBF_MAX + 1] = {
 	[TCA_TBF_PARMS]	= { .len = sizeof(struct tc_tbf_qopt) },
 	[TCA_TBF_RTAB]	= { .type = NLA_BINARY, .len = TC_RTAB_SIZE },
@@ -322,8 +294,11 @@ static int tbf_change(struct Qdisc* sch, struct nlattr *opt)
 		goto done;
 
 	if (qopt->limit > 0) {
-		if ((child = tbf_create_dflt_qdisc(sch, qopt->limit)) == NULL)
+		child = fifo_create_dflt(sch, &bfifo_qdisc_ops, qopt->limit);
+		if (IS_ERR(child)) {
+			err = PTR_ERR(child);
 			goto done;
+		}
 	}
 
 	sch_tree_lock(sch);
-- 
cgit v1.2.3


From b891c5a831b13f74989dcbd7b39d04537b2a05d9 Mon Sep 17 00:00:00 2001
From: Pablo Neira Ayuso <pablo@netfilter.org>
Date: Tue, 8 Jul 2008 02:35:55 -0700
Subject: netfilter: nf_conntrack: add allocation flag to nf_conntrack_alloc

ctnetlink does not need to allocate the conntrack entries with GFP_ATOMIC
as its code is executed in user context.

Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/netfilter/nf_conntrack.h | 3 ++-
 net/netfilter/nf_conntrack_core.c    | 7 ++++---
 net/netfilter/nf_conntrack_netlink.c | 2 +-
 3 files changed, 7 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/include/net/netfilter/nf_conntrack.h b/include/net/netfilter/nf_conntrack.h
index d77dec768dc2..d5d76ec7abb0 100644
--- a/include/net/netfilter/nf_conntrack.h
+++ b/include/net/netfilter/nf_conntrack.h
@@ -258,7 +258,8 @@ nf_ct_iterate_cleanup(int (*iter)(struct nf_conn *i, void *data), void *data);
 extern void nf_conntrack_free(struct nf_conn *ct);
 extern struct nf_conn *
 nf_conntrack_alloc(const struct nf_conntrack_tuple *orig,
-		   const struct nf_conntrack_tuple *repl);
+		   const struct nf_conntrack_tuple *repl,
+		   gfp_t gfp);
 
 /* It's confirmed if it is, or has been in the hash table. */
 static inline int nf_ct_is_confirmed(struct nf_conn *ct)
diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c
index f27c99246a4c..212a0888408d 100644
--- a/net/netfilter/nf_conntrack_core.c
+++ b/net/netfilter/nf_conntrack_core.c
@@ -464,7 +464,8 @@ static noinline int early_drop(unsigned int hash)
 }
 
 struct nf_conn *nf_conntrack_alloc(const struct nf_conntrack_tuple *orig,
-				   const struct nf_conntrack_tuple *repl)
+				   const struct nf_conntrack_tuple *repl,
+				   gfp_t gfp)
 {
 	struct nf_conn *ct = NULL;
 
@@ -489,7 +490,7 @@ struct nf_conn *nf_conntrack_alloc(const struct nf_conntrack_tuple *orig,
 		}
 	}
 
-	ct = kmem_cache_zalloc(nf_conntrack_cachep, GFP_ATOMIC);
+	ct = kmem_cache_zalloc(nf_conntrack_cachep, gfp);
 	if (ct == NULL) {
 		pr_debug("nf_conntrack_alloc: Can't alloc conntrack.\n");
 		atomic_dec(&nf_conntrack_count);
@@ -542,7 +543,7 @@ init_conntrack(const struct nf_conntrack_tuple *tuple,
 		return NULL;
 	}
 
-	ct = nf_conntrack_alloc(tuple, &repl_tuple);
+	ct = nf_conntrack_alloc(tuple, &repl_tuple, GFP_ATOMIC);
 	if (ct == NULL || IS_ERR(ct)) {
 		pr_debug("Can't allocate conntrack.\n");
 		return (struct nf_conntrack_tuple_hash *)ct;
diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c
index 63c4e1f299b8..dd233393f695 100644
--- a/net/netfilter/nf_conntrack_netlink.c
+++ b/net/netfilter/nf_conntrack_netlink.c
@@ -1128,7 +1128,7 @@ ctnetlink_create_conntrack(struct nlattr *cda[],
 	struct nf_conn_help *help;
 	struct nf_conntrack_helper *helper;
 
-	ct = nf_conntrack_alloc(otuple, rtuple);
+	ct = nf_conntrack_alloc(otuple, rtuple, GFP_KERNEL);
 	if (ct == NULL || IS_ERR(ct))
 		return -ENOMEM;
 
-- 
cgit v1.2.3


From cdf060a5d3c2afc7998af94b26a6c5182419e071 Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Tue, 8 Jul 2008 02:36:40 -0700
Subject: netfilter: cleanup netfilter_ipv6.h userspace header

Kernel functions are not for userspace.

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter_ipv6.h | 4 ++++
 1 file changed, 4 insertions(+)

(limited to 'include')

diff --git a/include/linux/netfilter_ipv6.h b/include/linux/netfilter_ipv6.h
index fd50988b83ec..d654873aa25a 100644
--- a/include/linux/netfilter_ipv6.h
+++ b/include/linux/netfilter_ipv6.h
@@ -70,6 +70,8 @@ enum nf_ip6_hook_priorities {
 	NF_IP6_PRI_LAST = INT_MAX,
 };
 
+#ifdef  __KERNEL__
+
 #ifdef CONFIG_NETFILTER
 extern int ip6_route_me_harder(struct sk_buff *skb);
 extern __sum16 nf_ip6_checksum(struct sk_buff *skb, unsigned int hook,
@@ -82,4 +84,6 @@ static inline int ipv6_netfilter_init(void) { return 0; }
 static inline void ipv6_netfilter_fini(void) { return; }
 #endif /* CONFIG_NETFILTER */
 
+#endif /* __KERNEL__ */
+
 #endif /*__LINUX_IP6_NETFILTER_H*/
-- 
cgit v1.2.3


From b9c796783151678d08b1ec1ef410685b2515ba51 Mon Sep 17 00:00:00 2001
From: Joonwoo Park <joonwpark81@gmail.com>
Date: Tue, 8 Jul 2008 02:37:31 -0700
Subject: textsearch: support for case insensitive searching

The function textsearch_prepare has a new flag to support case
insensitive searching.

Signed-off-by: Joonwoo Park <joonwpark81@gmail.com>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/textsearch.h | 10 +++++-----
 lib/textsearch.c           | 14 ++++++++------
 2 files changed, 13 insertions(+), 11 deletions(-)

(limited to 'include')

diff --git a/include/linux/textsearch.h b/include/linux/textsearch.h
index 6f371f24160b..6c34cf28b7aa 100644
--- a/include/linux/textsearch.h
+++ b/include/linux/textsearch.h
@@ -10,10 +10,8 @@
 
 struct ts_config;
 
-/**
- * TS_AUTOLOAD - Automatically load textsearch modules when needed
- */
-#define TS_AUTOLOAD	1
+#define TS_AUTOLOAD	1 /* Automatically load textsearch modules when needed */
+#define TS_IGNORECASE	2 /* Searches string case insensitively */
 
 /**
  * struct ts_state - search state
@@ -39,7 +37,7 @@ struct ts_state
 struct ts_ops
 {
 	const char		*name;
-	struct ts_config *	(*init)(const void *, unsigned int, gfp_t);
+	struct ts_config *	(*init)(const void *, unsigned int, gfp_t, int);
 	unsigned int		(*find)(struct ts_config *,
 					struct ts_state *);
 	void			(*destroy)(struct ts_config *);
@@ -52,12 +50,14 @@ struct ts_ops
 /**
  * struct ts_config - search configuration
  * @ops: operations of chosen algorithm
+ * @flags: flags
  * @get_next_block: callback to fetch the next block to search in
  * @finish: callback to finalize a search
  */
 struct ts_config
 {
 	struct ts_ops		*ops;
+	int 			flags;
 
 	/**
 	 * get_next_block - fetch next block of data
diff --git a/lib/textsearch.c b/lib/textsearch.c
index be8bda3862f5..b451fcc9354c 100644
--- a/lib/textsearch.c
+++ b/lib/textsearch.c
@@ -54,10 +54,13 @@
  * USAGE
  *
  *   Before a search can be performed, a configuration must be created
- *   by calling textsearch_prepare() specyfing the searching algorithm and
- *   the pattern to look for. The returned configuration may then be used
- *   for an arbitary amount of times and even in parallel as long as a
- *   separate struct ts_state variable is provided to every instance.
+ *   by calling textsearch_prepare() specifying the searching algorithm,
+ *   the pattern to look for and flags. As a flag, you can set TS_IGNORECASE
+ *   to perform case insensitive matching. But it might slow down
+ *   performance of algorithm, so you should use it at own your risk.
+ *   The returned configuration may then be used for an arbitary
+ *   amount of times and even in parallel as long as a separate struct
+ *   ts_state variable is provided to every instance.
  *
  *   The actual search is performed by either calling textsearch_find_-
  *   continuous() for linear data or by providing an own get_next_block()
@@ -89,7 +92,6 @@
  *       panic("Oh my god, dancing chickens at %d\n", pos);
  *
  *   textsearch_destroy(conf);
- *
  * ==========================================================================
  */
 
@@ -279,7 +281,7 @@ struct ts_config *textsearch_prepare(const char *algo, const void *pattern,
 	if (ops == NULL)
 		goto errout;
 
-	conf = ops->init(pattern, len, gfp_mask);
+	conf = ops->init(pattern, len, gfp_mask, flags);
 	if (IS_ERR(conf)) {
 		err = PTR_ERR(conf);
 		goto errout;
-- 
cgit v1.2.3


From dde77e604497dada6f224a685278dfb27747ae52 Mon Sep 17 00:00:00 2001
From: Joonwoo Park <joonwpark81@gmail.com>
Date: Tue, 8 Jul 2008 02:38:40 -0700
Subject: textsearch: convert kmalloc + memset to kzalloc

convert kmalloc + memset to kzalloc for alloc_ts_config

Signed-off-by: Joonwoo Park <joonwpark81@gmail.com>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/textsearch.h | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/linux/textsearch.h b/include/linux/textsearch.h
index 6c34cf28b7aa..d9a85d616385 100644
--- a/include/linux/textsearch.h
+++ b/include/linux/textsearch.h
@@ -162,11 +162,10 @@ static inline struct ts_config *alloc_ts_config(size_t payload,
 {
 	struct ts_config *conf;
 
-	conf = kmalloc(TS_PRIV_ALIGN(sizeof(*conf)) + payload, gfp_mask);
+	conf = kzalloc(TS_PRIV_ALIGN(sizeof(*conf)) + payload, gfp_mask);
 	if (conf == NULL)
 		return ERR_PTR(-ENOMEM);
 
-	memset(conf, 0, TS_PRIV_ALIGN(sizeof(*conf)) + payload);
 	return conf;
 }
 
-- 
cgit v1.2.3


From 4ad3f26162ece5aca3045fd45e15dd99acea4a0e Mon Sep 17 00:00:00 2001
From: Joonwoo Park <joonwpark81@gmail.com>
Date: Tue, 8 Jul 2008 02:38:56 -0700
Subject: netfilter: fix string extension for case insensitive pattern matching

The flag XT_STRING_FLAG_IGNORECASE indicates case insensitive string
matching. netfilter can find cmd.exe, Cmd.exe, cMd.exe and etc easily.

A new revision 1 was added, in the meantime invert of xt_string_info
was moved into flags as a flag. If revision is 1, The flag
XT_STRING_FLAG_INVERT indicates invert matching.

Signed-off-by: Joonwoo Park <joonwpark81@gmail.com>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter/xt_string.h | 15 ++++++++++++++-
 net/netfilter/xt_string.c           | 38 +++++++++++++++++++++++++++++++++++--
 2 files changed, 50 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/linux/netfilter/xt_string.h b/include/linux/netfilter/xt_string.h
index bb21dd1aee2d..8a6ba7bbef9f 100644
--- a/include/linux/netfilter/xt_string.h
+++ b/include/linux/netfilter/xt_string.h
@@ -4,6 +4,11 @@
 #define XT_STRING_MAX_PATTERN_SIZE 128
 #define XT_STRING_MAX_ALGO_NAME_SIZE 16
 
+enum {
+	XT_STRING_FLAG_INVERT		= 0x01,
+	XT_STRING_FLAG_IGNORECASE	= 0x02
+};
+
 struct xt_string_info
 {
 	u_int16_t from_offset;
@@ -11,7 +16,15 @@ struct xt_string_info
 	char	  algo[XT_STRING_MAX_ALGO_NAME_SIZE];
 	char 	  pattern[XT_STRING_MAX_PATTERN_SIZE];
 	u_int8_t  patlen;
-	u_int8_t  invert;
+	union {
+		struct {
+			u_int8_t  invert;
+		} v0;
+
+		struct {
+			u_int8_t  flags;
+		} v1;
+	} u;
 
 	/* Used internally by the kernel */
 	struct ts_config __attribute__((aligned(8))) *config;
diff --git a/net/netfilter/xt_string.c b/net/netfilter/xt_string.c
index 72f694d947f4..4903182a062b 100644
--- a/net/netfilter/xt_string.c
+++ b/net/netfilter/xt_string.c
@@ -29,12 +29,16 @@ string_mt(const struct sk_buff *skb, const struct net_device *in,
 {
 	const struct xt_string_info *conf = matchinfo;
 	struct ts_state state;
+	int invert;
 
 	memset(&state, 0, sizeof(struct ts_state));
 
+	invert = (match->revision == 0 ? conf->u.v0.invert :
+				    conf->u.v1.flags & XT_STRING_FLAG_INVERT);
+
 	return (skb_find_text((struct sk_buff *)skb, conf->from_offset,
 			     conf->to_offset, conf->config, &state)
-			     != UINT_MAX) ^ conf->invert;
+			     != UINT_MAX) ^ invert;
 }
 
 #define STRING_TEXT_PRIV(m) ((struct xt_string_info *)(m))
@@ -46,6 +50,7 @@ string_mt_check(const char *tablename, const void *ip,
 {
 	struct xt_string_info *conf = matchinfo;
 	struct ts_config *ts_conf;
+	int flags = TS_AUTOLOAD;
 
 	/* Damn, can't handle this case properly with iptables... */
 	if (conf->from_offset > conf->to_offset)
@@ -54,8 +59,15 @@ string_mt_check(const char *tablename, const void *ip,
 		return false;
 	if (conf->patlen > XT_STRING_MAX_PATTERN_SIZE)
 		return false;
+	if (match->revision == 1) {
+		if (conf->u.v1.flags &
+		    ~(XT_STRING_FLAG_IGNORECASE | XT_STRING_FLAG_INVERT))
+			return false;
+		if (conf->u.v1.flags & XT_STRING_FLAG_IGNORECASE)
+			flags |= TS_IGNORECASE;
+	}
 	ts_conf = textsearch_prepare(conf->algo, conf->pattern, conf->patlen,
-				     GFP_KERNEL, TS_AUTOLOAD);
+				     GFP_KERNEL, flags);
 	if (IS_ERR(ts_conf))
 		return false;
 
@@ -72,6 +84,17 @@ static void string_mt_destroy(const struct xt_match *match, void *matchinfo)
 static struct xt_match string_mt_reg[] __read_mostly = {
 	{
 		.name 		= "string",
+		.revision	= 0,
+		.family		= AF_INET,
+		.checkentry	= string_mt_check,
+		.match 		= string_mt,
+		.destroy 	= string_mt_destroy,
+		.matchsize	= sizeof(struct xt_string_info),
+		.me 		= THIS_MODULE
+	},
+	{
+		.name 		= "string",
+		.revision	= 1,
 		.family		= AF_INET,
 		.checkentry	= string_mt_check,
 		.match 		= string_mt,
@@ -81,6 +104,17 @@ static struct xt_match string_mt_reg[] __read_mostly = {
 	},
 	{
 		.name 		= "string",
+		.revision	= 0,
+		.family		= AF_INET6,
+		.checkentry	= string_mt_check,
+		.match 		= string_mt,
+		.destroy 	= string_mt_destroy,
+		.matchsize	= sizeof(struct xt_string_info),
+		.me 		= THIS_MODULE
+	},
+	{
+		.name 		= "string",
+		.revision	= 1,
 		.family		= AF_INET6,
 		.checkentry	= string_mt_check,
 		.match 		= string_mt,
-- 
cgit v1.2.3


From 2c693610fe923764fe41b846fb86938a2010da6e Mon Sep 17 00:00:00 2001
From: Richard Kennedy <richard@rsk.demon.co.uk>
Date: Tue, 8 Jul 2008 03:03:01 -0700
Subject: net: remove padding from struct socket on 64bit & increase
 objects/cache

remove padding from struct socket reducing its size by 8 bytes.

This allows more objects/cache in sock_inode_cache
12 objects/cache when cacheline size is 128 (generic x86_64)

Signed-off-by: Richard Kennedy <richard@rsk.demon.co.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/net.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/linux/net.h b/include/linux/net.h
index 71f7dd559285..150a48c68d52 100644
--- a/include/linux/net.h
+++ b/include/linux/net.h
@@ -106,23 +106,23 @@ enum sock_shutdown_cmd {
 /**
  *  struct socket - general BSD socket
  *  @state: socket state (%SS_CONNECTED, etc)
+ *  @type: socket type (%SOCK_STREAM, etc)
  *  @flags: socket flags (%SOCK_ASYNC_NOSPACE, etc)
  *  @ops: protocol specific socket operations
  *  @fasync_list: Asynchronous wake up list
  *  @file: File back pointer for gc
  *  @sk: internal networking protocol agnostic socket representation
  *  @wait: wait queue for several uses
- *  @type: socket type (%SOCK_STREAM, etc)
  */
 struct socket {
 	socket_state		state;
+	short			type;
 	unsigned long		flags;
 	const struct proto_ops	*ops;
 	struct fasync_struct	*fasync_list;
 	struct file		*file;
 	struct sock		*sk;
 	wait_queue_head_t	wait;
-	short			type;
 };
 
 struct vm_area_struct;
-- 
cgit v1.2.3


From acc81e1465d29e0284008770cc4b8bc90bd93bd7 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Tue, 8 Jul 2008 03:21:27 -0700
Subject: vlan: fix network_header/mac_header adjustments

Lennert Buytenhek points out that the VLAN code incorrectly adjusts
skb->network_header to point in the middle of the VLAN header and
additionally tries to adjust skb->mac_header without checking for
validity.

The network_header should not be touched at all since we're only
adding headers in front of it, mac_header adjustments are not
necessary at all.

Based on patch by Lennert Buytenhek <buytenh@wantstofly.org>.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/if_vlan.h | 2 --
 net/8021q/vlan_dev.c    | 1 -
 2 files changed, 3 deletions(-)

(limited to 'include')

diff --git a/include/linux/if_vlan.h b/include/linux/if_vlan.h
index 5190452ac7dc..8f5bf9b676aa 100644
--- a/include/linux/if_vlan.h
+++ b/include/linux/if_vlan.h
@@ -279,8 +279,6 @@ static inline struct sk_buff *__vlan_put_tag(struct sk_buff *skb, unsigned short
 	veth->h_vlan_TCI = htons(tag);
 
 	skb->protocol = htons(ETH_P_8021Q);
-	skb->mac_header -= VLAN_HLEN;
-	skb->network_header -= VLAN_HLEN;
 
 	return skb;
 }
diff --git a/net/8021q/vlan_dev.c b/net/8021q/vlan_dev.c
index a0617bf7cec6..4a8525927c27 100644
--- a/net/8021q/vlan_dev.c
+++ b/net/8021q/vlan_dev.c
@@ -308,7 +308,6 @@ static int vlan_dev_hard_header(struct sk_buff *skb, struct net_device *dev,
 			vhdr->h_vlan_encapsulated_proto = htons(len);
 
 		skb->protocol = htons(ETH_P_8021Q);
-		skb_reset_network_header(skb);
 	}
 
 	/* Before delegating work to the lower layer, enter our MAC-address */
-- 
cgit v1.2.3


From 7750f403cbe56971336d575b354365190b4e3227 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Tue, 8 Jul 2008 03:23:36 -0700
Subject: vlan: uninline __vlan_hwaccel_rx

The function is huge and included at least once in every VLAN acceleration
capable driver. Uninline it; to avoid having drivers depend on the VLAN
module, the function is always built in statically when VLAN is enabled.

With all VLAN acceleration capable drivers that build on x86_64 enabled,
this results in:

   text    data     bss     dec     hex filename
6515227  854044  343968 7713239  75b1d7 vmlinux.inlined
6505637  854044  343968 7703649  758c61 vmlinux.uninlined
----------------------------------------------------------
  -9590

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/if_vlan.h | 64 +++++++------------------------------------------
 net/8021q/Makefile      |  9 +++----
 net/8021q/vlan.h        |  8 +++++++
 net/8021q/vlan_core.c   | 48 +++++++++++++++++++++++++++++++++++++
 net/Makefile            |  4 +++-
 5 files changed, 72 insertions(+), 61 deletions(-)
 create mode 100644 net/8021q/vlan_core.c

(limited to 'include')

diff --git a/include/linux/if_vlan.h b/include/linux/if_vlan.h
index 8f5bf9b676aa..594cd35b0074 100644
--- a/include/linux/if_vlan.h
+++ b/include/linux/if_vlan.h
@@ -150,15 +150,6 @@ static inline struct vlan_dev_info *vlan_dev_info(const struct net_device *dev)
 	return netdev_priv(dev);
 }
 
-/* inline functions */
-static inline __u32 vlan_get_ingress_priority(struct net_device *dev,
-					      unsigned short vlan_tag)
-{
-	struct vlan_dev_info *vip = vlan_dev_info(dev);
-
-	return vip->ingress_priority_map[(vlan_tag >> 13) & 0x7];
-}
-
 /* VLAN tx hw acceleration helpers. */
 struct vlan_skb_tx_cookie {
 	u32	magic;
@@ -171,56 +162,17 @@ struct vlan_skb_tx_cookie {
 	(VLAN_TX_SKB_CB(__skb)->magic == VLAN_TX_COOKIE_MAGIC)
 #define vlan_tx_tag_get(__skb)	(VLAN_TX_SKB_CB(__skb)->vlan_tag)
 
-/* VLAN rx hw acceleration helper.  This acts like netif_{rx,receive_skb}(). */
-static inline int __vlan_hwaccel_rx(struct sk_buff *skb,
-				    struct vlan_group *grp,
+#if defined(CONFIG_VLAN_8021Q) || defined(CONFIG_VLAN_8021Q_MODULE)
+extern int __vlan_hwaccel_rx(struct sk_buff *skb, struct vlan_group *grp,
+			     unsigned short vlan_tag, int polling);
+#else
+static inline int __vlan_hwaccel_rx(struct sk_buff *skb, struct vlan_group *grp,
 				    unsigned short vlan_tag, int polling)
 {
-	struct net_device_stats *stats;
-
-	if (skb_bond_should_drop(skb)) {
-		dev_kfree_skb_any(skb);
-		return NET_RX_DROP;
-	}
-
-	skb->dev = vlan_group_get_device(grp, vlan_tag & VLAN_VID_MASK);
-	if (skb->dev == NULL) {
-		dev_kfree_skb_any(skb);
-
-		/* Not NET_RX_DROP, this is not being dropped
-		 * due to congestion.
-		 */
-		return 0;
-	}
-
-	skb->dev->last_rx = jiffies;
-
-	stats = &skb->dev->stats;
-	stats->rx_packets++;
-	stats->rx_bytes += skb->len;
-
-	skb->priority = vlan_get_ingress_priority(skb->dev, vlan_tag);
-	switch (skb->pkt_type) {
-	case PACKET_BROADCAST:
-		break;
-
-	case PACKET_MULTICAST:
-		stats->multicast++;
-		break;
-
-	case PACKET_OTHERHOST:
-		/* Our lower layer thinks this is not local, let's make sure.
-		 * This allows the VLAN to have a different MAC than the underlying
-		 * device, and still route correctly.
-		 */
-		if (!compare_ether_addr(eth_hdr(skb)->h_dest,
-				       	skb->dev->dev_addr))
-			skb->pkt_type = PACKET_HOST;
-		break;
-	};
-
-	return (polling ? netif_receive_skb(skb) : netif_rx(skb));
+	BUG();
+	return NET_XMIT_SUCCESS;
 }
+#endif
 
 static inline int vlan_hwaccel_rx(struct sk_buff *skb,
 				  struct vlan_group *grp,
diff --git a/net/8021q/Makefile b/net/8021q/Makefile
index 3006e9ed7b08..9f4f174ead1c 100644
--- a/net/8021q/Makefile
+++ b/net/8021q/Makefile
@@ -1,9 +1,10 @@
 #
 # Makefile for the Linux VLAN layer.
 #
+obj-$(subst m,y,$(CONFIG_VLAN_8021Q))	+= vlan_core.o
+obj-$(CONFIG_VLAN_8021Q)		+= 8021q.o
 
-obj-$(CONFIG_VLAN_8021Q) += 8021q.o
+8021q-y					:= vlan.o vlan_dev.o vlan_netlink.o
+8021q-$(CONFIG_VLAN_8021Q_GVRP)		+= vlan_gvrp.o
+8021q-$(CONFIG_PROC_FS)			+= vlanproc.o
 
-8021q-y				:= vlan.o vlan_dev.o vlan_netlink.o
-8021q-$(CONFIG_VLAN_8021Q_GVRP)	+= vlan_gvrp.o
-8021q-$(CONFIG_PROC_FS)		+= vlanproc.o
diff --git a/net/8021q/vlan.h b/net/8021q/vlan.h
index 097b2e04c928..7cc1a97c42fc 100644
--- a/net/8021q/vlan.h
+++ b/net/8021q/vlan.h
@@ -37,6 +37,14 @@ void vlan_setup(struct net_device *dev);
 int register_vlan_dev(struct net_device *dev);
 void unregister_vlan_dev(struct net_device *dev);
 
+static inline u32 vlan_get_ingress_priority(struct net_device *dev,
+					    unsigned short vlan_tag)
+{
+	struct vlan_dev_info *vip = vlan_dev_info(dev);
+
+	return vip->ingress_priority_map[(vlan_tag >> 13) & 0x7];
+}
+
 #ifdef CONFIG_VLAN_8021Q_GVRP
 extern int vlan_gvrp_request_join(const struct net_device *dev);
 extern void vlan_gvrp_request_leave(const struct net_device *dev);
diff --git a/net/8021q/vlan_core.c b/net/8021q/vlan_core.c
new file mode 100644
index 000000000000..85c94edb000f
--- /dev/null
+++ b/net/8021q/vlan_core.c
@@ -0,0 +1,48 @@
+#include <linux/skbuff.h>
+#include <linux/netdevice.h>
+#include <linux/if_vlan.h>
+#include "vlan.h"
+
+/* VLAN rx hw acceleration helper.  This acts like netif_{rx,receive_skb}(). */
+int __vlan_hwaccel_rx(struct sk_buff *skb, struct vlan_group *grp,
+		      unsigned short vlan_tag, int polling)
+{
+	struct net_device_stats *stats;
+
+	if (skb_bond_should_drop(skb)) {
+		dev_kfree_skb_any(skb);
+		return NET_RX_DROP;
+	}
+
+	skb->dev = vlan_group_get_device(grp, vlan_tag & VLAN_VID_MASK);
+	if (skb->dev == NULL) {
+		dev_kfree_skb_any(skb);
+		/* Not NET_RX_DROP, this is not being dropped
+		 * due to congestion. */
+		return NET_RX_SUCCESS;
+	}
+	skb->dev->last_rx = jiffies;
+
+	stats = &skb->dev->stats;
+	stats->rx_packets++;
+	stats->rx_bytes += skb->len;
+
+	skb->priority = vlan_get_ingress_priority(skb->dev, vlan_tag);
+	switch (skb->pkt_type) {
+	case PACKET_BROADCAST:
+		break;
+	case PACKET_MULTICAST:
+		stats->multicast++;
+		break;
+	case PACKET_OTHERHOST:
+		/* Our lower layer thinks this is not local, let's make sure.
+		 * This allows the VLAN to have a different MAC than the
+		 * underlying device, and still route correctly. */
+		if (!compare_ether_addr(eth_hdr(skb)->h_dest,
+					skb->dev->dev_addr))
+			skb->pkt_type = PACKET_HOST;
+		break;
+	};
+	return (polling ? netif_receive_skb(skb) : netif_rx(skb));
+}
+EXPORT_SYMBOL(__vlan_hwaccel_rx);
diff --git a/net/Makefile b/net/Makefile
index b7a13643b549..4f43e7f874f3 100644
--- a/net/Makefile
+++ b/net/Makefile
@@ -42,7 +42,9 @@ obj-$(CONFIG_AF_RXRPC)		+= rxrpc/
 obj-$(CONFIG_ATM)		+= atm/
 obj-$(CONFIG_DECNET)		+= decnet/
 obj-$(CONFIG_ECONET)		+= econet/
-obj-$(CONFIG_VLAN_8021Q)	+= 8021q/
+ifneq ($(CONFIG_VLAN_8021Q),)
+obj-y				+= 8021q/
+endif
 obj-$(CONFIG_IP_DCCP)		+= dccp/
 obj-$(CONFIG_IP_SCTP)		+= sctp/
 obj-y				+= wireless/
-- 
cgit v1.2.3


From 22d1ba74bbafa96d3f425cc12714d3fe8675183f Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Tue, 8 Jul 2008 03:23:57 -0700
Subject: vlan: move struct vlan_dev_info to private header

Hide struct vlan_dev_info from drivers to prevent them from growing
more creative ways to use it. Provide accessors for the two drivers
that currently use it.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/cxgb3/l2t.c         |  2 +-
 drivers/s390/net/qeth_l3_main.c |  4 +--
 include/linux/if_vlan.h         | 56 +++++++++++------------------------------
 net/8021q/vlan.c                |  4 +--
 net/8021q/vlan.h                | 50 +++++++++++++++++++++++++++++++++++-
 net/8021q/vlan_core.c           | 12 +++++++++
 net/8021q/vlan_dev.c            |  5 ----
 7 files changed, 80 insertions(+), 53 deletions(-)

(limited to 'include')

diff --git a/drivers/net/cxgb3/l2t.c b/drivers/net/cxgb3/l2t.c
index f510140885ae..825e510bd9ed 100644
--- a/drivers/net/cxgb3/l2t.c
+++ b/drivers/net/cxgb3/l2t.c
@@ -337,7 +337,7 @@ struct l2t_entry *t3_l2t_get(struct t3cdev *cdev, struct neighbour *neigh,
 		atomic_set(&e->refcnt, 1);
 		neigh_replace(e, neigh);
 		if (neigh->dev->priv_flags & IFF_802_1Q_VLAN)
-			e->vlan = vlan_dev_info(neigh->dev)->vlan_id;
+			e->vlan = vlan_dev_vlan_id(neigh->dev);
 		else
 			e->vlan = VLAN_NONE;
 		spin_unlock(&e->lock);
diff --git a/drivers/s390/net/qeth_l3_main.c b/drivers/s390/net/qeth_l3_main.c
index 999552c83bbe..85be40abdda9 100644
--- a/drivers/s390/net/qeth_l3_main.c
+++ b/drivers/s390/net/qeth_l3_main.c
@@ -2020,7 +2020,7 @@ static int qeth_l3_verify_vlan_dev(struct net_device *dev,
 		}
 	}
 
-	if (rc && !(netdev_priv(vlan_dev_info(dev)->real_dev) == (void *)card))
+	if (rc && !(netdev_priv(vlan_dev_real_dev(dev)) == (void *)card))
 		return 0;
 
 	return rc;
@@ -2056,7 +2056,7 @@ static struct qeth_card *qeth_l3_get_card_from_dev(struct net_device *dev)
 	if (rc == QETH_REAL_CARD)
 		card = netdev_priv(dev);
 	else if (rc == QETH_VLAN_CARD)
-		card = netdev_priv(vlan_dev_info(dev)->real_dev);
+		card = netdev_priv(vlan_dev_real_dev(dev));
 	if (card && card->options.layer2)
 		card = NULL;
 	QETH_DBF_TEXT_(TRACE, 4, "%d", rc);
diff --git a/include/linux/if_vlan.h b/include/linux/if_vlan.h
index 594cd35b0074..cb2e6b480882 100644
--- a/include/linux/if_vlan.h
+++ b/include/linux/if_vlan.h
@@ -109,47 +109,6 @@ static inline void vlan_group_set_device(struct vlan_group *vg,
 	array[vlan_id % VLAN_GROUP_ARRAY_PART_LEN] = dev;
 }
 
-struct vlan_priority_tci_mapping {
-	u32 priority;
-	unsigned short vlan_qos; /* This should be shifted when first set, so we only do it
-				  * at provisioning time.
-				  * ((skb->priority << 13) & 0xE000)
-				  */
-	struct vlan_priority_tci_mapping *next;
-};
-
-/* Holds information that makes sense if this device is a VLAN device. */
-struct vlan_dev_info {
-	/** This will be the mapping that correlates skb->priority to
-	 * 3 bits of VLAN QOS tags...
-	 */
-	unsigned int nr_ingress_mappings;
-	u32 ingress_priority_map[8];
-
-	unsigned int nr_egress_mappings;
-	struct vlan_priority_tci_mapping *egress_priority_map[16]; /* hash table */
-
-	unsigned short vlan_id;        /*  The VLAN Identifier for this interface. */
-	unsigned short flags;          /* (1 << 0) re_order_header   This option will cause the
-                                        *   VLAN code to move around the ethernet header on
-                                        *   ingress to make the skb look **exactly** like it
-                                        *   came in from an ethernet port.  This destroys some of
-                                        *   the VLAN information in the skb, but it fixes programs
-                                        *   like DHCP that use packet-filtering and don't understand
-                                        *   802.1Q
-                                        */
-	struct net_device *real_dev;    /* the underlying device/interface */
-	unsigned char real_dev_addr[ETH_ALEN];
-	struct proc_dir_entry *dent;    /* Holds the proc data */
-	unsigned long cnt_inc_headroom_on_tx; /* How many times did we have to grow the skb on TX. */
-	unsigned long cnt_encap_on_xmit;      /* How many times did we have to encapsulate the skb on TX. */
-};
-
-static inline struct vlan_dev_info *vlan_dev_info(const struct net_device *dev)
-{
-	return netdev_priv(dev);
-}
-
 /* VLAN tx hw acceleration helpers. */
 struct vlan_skb_tx_cookie {
 	u32	magic;
@@ -163,9 +122,24 @@ struct vlan_skb_tx_cookie {
 #define vlan_tx_tag_get(__skb)	(VLAN_TX_SKB_CB(__skb)->vlan_tag)
 
 #if defined(CONFIG_VLAN_8021Q) || defined(CONFIG_VLAN_8021Q_MODULE)
+extern struct net_device *vlan_dev_real_dev(const struct net_device *dev);
+extern u16 vlan_dev_vlan_id(const struct net_device *dev);
+
 extern int __vlan_hwaccel_rx(struct sk_buff *skb, struct vlan_group *grp,
 			     unsigned short vlan_tag, int polling);
 #else
+static inline struct net_device *vlan_dev_real_dev(const struct net_device *dev)
+{
+	BUG();
+	return NULL;
+}
+
+static inline u16 vlan_dev_vlan_id(const struct net_device *dev)
+{
+	BUG();
+	return 0;
+}
+
 static inline int __vlan_hwaccel_rx(struct sk_buff *skb, struct vlan_group *grp,
 				    unsigned short vlan_tag, int polling)
 {
diff --git a/net/8021q/vlan.c b/net/8021q/vlan.c
index 8141e2dc510b..7a2625d2f9a0 100644
--- a/net/8021q/vlan.c
+++ b/net/8021q/vlan.c
@@ -543,7 +543,6 @@ static struct notifier_block vlan_notifier_block __read_mostly = {
 static int vlan_ioctl_handler(struct net *net, void __user *arg)
 {
 	int err;
-	unsigned short vid = 0;
 	struct vlan_ioctl_args args;
 	struct net_device *dev = NULL;
 
@@ -644,8 +643,7 @@ static int vlan_ioctl_handler(struct net *net, void __user *arg)
 
 	case GET_VLAN_VID_CMD:
 		err = 0;
-		vlan_dev_get_vid(dev, &vid);
-		args.u.VID = vid;
+		args.u.VID = vlan_dev_vlan_id(dev);
 		if (copy_to_user(arg, &args,
 				 sizeof(struct vlan_ioctl_args)))
 		      err = -EFAULT;
diff --git a/net/8021q/vlan.h b/net/8021q/vlan.h
index 7cc1a97c42fc..14c421e033f2 100644
--- a/net/8021q/vlan.h
+++ b/net/8021q/vlan.h
@@ -3,6 +3,55 @@
 
 #include <linux/if_vlan.h>
 
+
+/**
+ *	struct vlan_priority_tci_mapping - vlan egress priority mappings
+ *	@priority: skb priority
+ *	@vlan_qos: vlan priority: (skb->priority << 13) & 0xE000
+ *	@next: pointer to next struct
+ */
+struct vlan_priority_tci_mapping {
+	u32					priority;
+	unsigned short				vlan_qos;
+	struct vlan_priority_tci_mapping	*next;
+};
+
+/**
+ *	struct vlan_dev_info - VLAN private device data
+ *	@nr_ingress_mappings: number of ingress priority mappings
+ *	@ingress_priority_map: ingress priority mappings
+ *	@nr_egress_mappings: number of egress priority mappings
+ *	@egress_priority_map: hash of egress priority mappings
+ *	@vlan_id: VLAN identifier
+ *	@flags: device flags
+ *	@real_dev: underlying netdevice
+ *	@real_dev_addr: address of underlying netdevice
+ *	@dent: proc dir entry
+ *	@cnt_inc_headroom_on_tx: statistic - number of skb expansions on TX
+ *	@cnt_encap_on_xmit: statistic - number of skb encapsulations on TX
+ */
+struct vlan_dev_info {
+	unsigned int				nr_ingress_mappings;
+	u32					ingress_priority_map[8];
+	unsigned int				nr_egress_mappings;
+	struct vlan_priority_tci_mapping	*egress_priority_map[16];
+
+	unsigned short				vlan_id;
+	unsigned short				flags;
+
+	struct net_device			*real_dev;
+	unsigned char				real_dev_addr[ETH_ALEN];
+
+	struct proc_dir_entry			*dent;
+	unsigned long				cnt_inc_headroom_on_tx;
+	unsigned long				cnt_encap_on_xmit;
+};
+
+static inline struct vlan_dev_info *vlan_dev_info(const struct net_device *dev)
+{
+	return netdev_priv(dev);
+}
+
 #define VLAN_GRP_HASH_SHIFT	5
 #define VLAN_GRP_HASH_SIZE	(1 << VLAN_GRP_HASH_SHIFT)
 #define VLAN_GRP_HASH_MASK	(VLAN_GRP_HASH_SIZE - 1)
@@ -30,7 +79,6 @@ int vlan_dev_set_egress_priority(const struct net_device *dev,
 				 u32 skb_prio, short vlan_prio);
 int vlan_dev_change_flags(const struct net_device *dev, u32 flag, u32 mask);
 void vlan_dev_get_realdev_name(const struct net_device *dev, char *result);
-void vlan_dev_get_vid(const struct net_device *dev, unsigned short *result);
 
 int vlan_check_real_dev(struct net_device *real_dev, unsigned short vlan_id);
 void vlan_setup(struct net_device *dev);
diff --git a/net/8021q/vlan_core.c b/net/8021q/vlan_core.c
index 85c94edb000f..f980b9154cc3 100644
--- a/net/8021q/vlan_core.c
+++ b/net/8021q/vlan_core.c
@@ -46,3 +46,15 @@ int __vlan_hwaccel_rx(struct sk_buff *skb, struct vlan_group *grp,
 	return (polling ? netif_receive_skb(skb) : netif_rx(skb));
 }
 EXPORT_SYMBOL(__vlan_hwaccel_rx);
+
+struct net_device *vlan_dev_real_dev(const struct net_device *dev)
+{
+	return vlan_dev_info(dev)->real_dev;
+}
+EXPORT_SYMBOL_GPL(vlan_dev_real_dev);
+
+u16 vlan_dev_vlan_id(const struct net_device *dev)
+{
+	return vlan_dev_info(dev)->vlan_id;
+}
+EXPORT_SYMBOL_GPL(vlan_dev_vlan_id);
diff --git a/net/8021q/vlan_dev.c b/net/8021q/vlan_dev.c
index 722697d31e4f..2aab294c5744 100644
--- a/net/8021q/vlan_dev.c
+++ b/net/8021q/vlan_dev.c
@@ -531,11 +531,6 @@ void vlan_dev_get_realdev_name(const struct net_device *dev, char *result)
 	strncpy(result, vlan_dev_info(dev)->real_dev->name, 23);
 }
 
-void vlan_dev_get_vid(const struct net_device *dev, unsigned short *result)
-{
-	*result = vlan_dev_info(dev)->vlan_id;
-}
-
 static int vlan_dev_open(struct net_device *dev)
 {
 	struct vlan_dev_info *vlan = vlan_dev_info(dev);
-- 
cgit v1.2.3


From df6b6a0cf62afeacdeb4c1a35b8fba21fda54399 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Tue, 8 Jul 2008 03:24:14 -0700
Subject: vlan: remove useless struct hlist_node declaration from if_vlan.h

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/if_vlan.h | 4 ----
 1 file changed, 4 deletions(-)

(limited to 'include')

diff --git a/include/linux/if_vlan.h b/include/linux/if_vlan.h
index cb2e6b480882..8e68b05b13df 100644
--- a/include/linux/if_vlan.h
+++ b/include/linux/if_vlan.h
@@ -14,10 +14,6 @@
 #define _LINUX_IF_VLAN_H_
 
 #ifdef __KERNEL__
-
-/* externally defined structs */
-struct hlist_node;
-
 #include <linux/netdevice.h>
 #include <linux/etherdevice.h>
 
-- 
cgit v1.2.3


From 9bb8582efb555521c7eec595ebd34e835ddc34b8 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Tue, 8 Jul 2008 03:24:44 -0700
Subject: vlan: TCI related type and naming cleanups

The VLAN code contains multiple spots that use tag, id and tci as
identifiers for arguments and variables incorrectly and they actually
contain or are expected to contain something different. Additionally
types are used inconsistently (unsigned short vs u16) and identifiers
are sometimes capitalized.

- consistently use u16 for storing TCI, ID or QoS values
- consistently use vlan_id and vlan_tci for storing the respective values
- remove capitalization
- add kdoc comment to netif_hwaccel_{rx,receive_skb}

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/if_vlan.h | 84 ++++++++++++++++++++++++++++---------------------
 net/8021q/vlan.c        | 34 ++++++++++----------
 net/8021q/vlan.h        | 19 ++++++-----
 net/8021q/vlan_core.c   |  6 ++--
 net/8021q/vlan_dev.c    | 47 ++++++++++++++-------------
 net/8021q/vlan_gvrp.c   |  8 ++---
 6 files changed, 103 insertions(+), 95 deletions(-)

(limited to 'include')

diff --git a/include/linux/if_vlan.h b/include/linux/if_vlan.h
index 8e68b05b13df..d36515dae62f 100644
--- a/include/linux/if_vlan.h
+++ b/include/linux/if_vlan.h
@@ -87,7 +87,7 @@ struct vlan_group {
 };
 
 static inline struct net_device *vlan_group_get_device(struct vlan_group *vg,
-						       unsigned int vlan_id)
+						       u16 vlan_id)
 {
 	struct net_device **array;
 	array = vg->vlan_devices_arrays[vlan_id / VLAN_GROUP_ARRAY_PART_LEN];
@@ -95,7 +95,7 @@ static inline struct net_device *vlan_group_get_device(struct vlan_group *vg,
 }
 
 static inline void vlan_group_set_device(struct vlan_group *vg,
-					 unsigned int vlan_id,
+					 u16 vlan_id,
 					 struct net_device *dev)
 {
 	struct net_device **array;
@@ -122,7 +122,7 @@ extern struct net_device *vlan_dev_real_dev(const struct net_device *dev);
 extern u16 vlan_dev_vlan_id(const struct net_device *dev);
 
 extern int __vlan_hwaccel_rx(struct sk_buff *skb, struct vlan_group *grp,
-			     unsigned short vlan_tag, int polling);
+			     u16 vlan_tci, int polling);
 #else
 static inline struct net_device *vlan_dev_real_dev(const struct net_device *dev)
 {
@@ -137,39 +137,51 @@ static inline u16 vlan_dev_vlan_id(const struct net_device *dev)
 }
 
 static inline int __vlan_hwaccel_rx(struct sk_buff *skb, struct vlan_group *grp,
-				    unsigned short vlan_tag, int polling)
+				    u16 vlan_tci, int polling)
 {
 	BUG();
 	return NET_XMIT_SUCCESS;
 }
 #endif
 
+/**
+ * vlan_hwaccel_rx - netif_rx wrapper for VLAN RX acceleration
+ * @skb: buffer
+ * @grp: vlan group
+ * @vlan_tci: VLAN TCI as received from the card
+ */
 static inline int vlan_hwaccel_rx(struct sk_buff *skb,
 				  struct vlan_group *grp,
-				  unsigned short vlan_tag)
+				  u16 vlan_tci)
 {
-	return __vlan_hwaccel_rx(skb, grp, vlan_tag, 0);
+	return __vlan_hwaccel_rx(skb, grp, vlan_tci, 0);
 }
 
+/**
+ * vlan_hwaccel_receive_skb - netif_receive_skb wrapper for VLAN RX acceleration
+ * @skb: buffer
+ * @grp: vlan group
+ * @vlan_tci: VLAN TCI as received from the card
+ */
 static inline int vlan_hwaccel_receive_skb(struct sk_buff *skb,
 					   struct vlan_group *grp,
-					   unsigned short vlan_tag)
+					   u16 vlan_tci)
 {
-	return __vlan_hwaccel_rx(skb, grp, vlan_tag, 1);
+	return __vlan_hwaccel_rx(skb, grp, vlan_tci, 1);
 }
 
 /**
  * __vlan_put_tag - regular VLAN tag inserting
  * @skb: skbuff to tag
- * @tag: VLAN tag to insert
+ * @vlan_tci: VLAN TCI to insert
  *
  * Inserts the VLAN tag into @skb as part of the payload
  * Returns a VLAN tagged skb. If a new skb is created, @skb is freed.
- * 
+ *
  * Following the skb_unshare() example, in case of error, the calling function
  * doesn't have to worry about freeing the original skb.
  */
-static inline struct sk_buff *__vlan_put_tag(struct sk_buff *skb, unsigned short tag)
+static inline struct sk_buff *__vlan_put_tag(struct sk_buff *skb, u16 vlan_tci)
 {
 	struct vlan_ethhdr *veth;
 
@@ -197,8 +209,8 @@ static inline struct sk_buff *__vlan_put_tag(struct sk_buff *skb, unsigned short
 	/* first, the ethernet type */
 	veth->h_vlan_proto = htons(ETH_P_8021Q);
 
-	/* now, the tag */
-	veth->h_vlan_TCI = htons(tag);
+	/* now, the TCI */
+	veth->h_vlan_TCI = htons(vlan_tci);
 
 	skb->protocol = htons(ETH_P_8021Q);
 
@@ -208,17 +220,18 @@ static inline struct sk_buff *__vlan_put_tag(struct sk_buff *skb, unsigned short
 /**
  * __vlan_hwaccel_put_tag - hardware accelerated VLAN inserting
  * @skb: skbuff to tag
- * @tag: VLAN tag to insert
+ * @vlan_tci: VLAN TCI to insert
  *
- * Puts the VLAN tag in @skb->cb[] and lets the device do the rest
+ * Puts the VLAN TCI in @skb->cb[] and lets the device do the rest
  */
-static inline struct sk_buff *__vlan_hwaccel_put_tag(struct sk_buff *skb, unsigned short tag)
+static inline struct sk_buff *__vlan_hwaccel_put_tag(struct sk_buff *skb,
+						     u16 vlan_tci)
 {
 	struct vlan_skb_tx_cookie *cookie;
 
 	cookie = VLAN_TX_SKB_CB(skb);
 	cookie->magic = VLAN_TX_COOKIE_MAGIC;
-	cookie->vlan_tag = tag;
+	cookie->vlan_tag = vlan_tci;
 
 	return skb;
 }
@@ -228,28 +241,28 @@ static inline struct sk_buff *__vlan_hwaccel_put_tag(struct sk_buff *skb, unsign
 /**
  * vlan_put_tag - inserts VLAN tag according to device features
  * @skb: skbuff to tag
- * @tag: VLAN tag to insert
+ * @vlan_tci: VLAN TCI to insert
  *
  * Assumes skb->dev is the target that will xmit this frame.
  * Returns a VLAN tagged skb.
  */
-static inline struct sk_buff *vlan_put_tag(struct sk_buff *skb, unsigned short tag)
+static inline struct sk_buff *vlan_put_tag(struct sk_buff *skb, u16 vlan_tci)
 {
 	if (skb->dev->features & NETIF_F_HW_VLAN_TX) {
-		return __vlan_hwaccel_put_tag(skb, tag);
+		return __vlan_hwaccel_put_tag(skb, vlan_tci);
 	} else {
-		return __vlan_put_tag(skb, tag);
+		return __vlan_put_tag(skb, vlan_tci);
 	}
 }
 
 /**
  * __vlan_get_tag - get the VLAN ID that is part of the payload
  * @skb: skbuff to query
- * @tag: buffer to store vlaue
- * 
+ * @vlan_tci: buffer to store vlaue
+ *
  * Returns error if the skb is not of VLAN type
  */
-static inline int __vlan_get_tag(const struct sk_buff *skb, unsigned short *tag)
+static inline int __vlan_get_tag(const struct sk_buff *skb, u16 *vlan_tci)
 {
 	struct vlan_ethhdr *veth = (struct vlan_ethhdr *)skb->data;
 
@@ -257,29 +270,28 @@ static inline int __vlan_get_tag(const struct sk_buff *skb, unsigned short *tag)
 		return -EINVAL;
 	}
 
-	*tag = ntohs(veth->h_vlan_TCI);
-
+	*vlan_tci = ntohs(veth->h_vlan_TCI);
 	return 0;
 }
 
 /**
  * __vlan_hwaccel_get_tag - get the VLAN ID that is in @skb->cb[]
  * @skb: skbuff to query
- * @tag: buffer to store vlaue
- * 
+ * @vlan_tci: buffer to store vlaue
+ *
  * Returns error if @skb->cb[] is not set correctly
  */
 static inline int __vlan_hwaccel_get_tag(const struct sk_buff *skb,
-					 unsigned short *tag)
+					 u16 *vlan_tci)
 {
 	struct vlan_skb_tx_cookie *cookie;
 
 	cookie = VLAN_TX_SKB_CB(skb);
 	if (cookie->magic == VLAN_TX_COOKIE_MAGIC) {
-		*tag = cookie->vlan_tag;
+		*vlan_tci = cookie->vlan_tag;
 		return 0;
 	} else {
-		*tag = 0;
+		*vlan_tci = 0;
 		return -EINVAL;
 	}
 }
@@ -289,16 +301,16 @@ static inline int __vlan_hwaccel_get_tag(const struct sk_buff *skb,
 /**
  * vlan_get_tag - get the VLAN ID from the skb
  * @skb: skbuff to query
- * @tag: buffer to store vlaue
- * 
+ * @vlan_tci: buffer to store vlaue
+ *
  * Returns error if the skb is not VLAN tagged
  */
-static inline int vlan_get_tag(const struct sk_buff *skb, unsigned short *tag)
+static inline int vlan_get_tag(const struct sk_buff *skb, u16 *vlan_tci)
 {
 	if (skb->dev->features & NETIF_F_HW_VLAN_TX) {
-		return __vlan_hwaccel_get_tag(skb, tag);
+		return __vlan_hwaccel_get_tag(skb, vlan_tci);
 	} else {
-		return __vlan_get_tag(skb, tag);
+		return __vlan_get_tag(skb, vlan_tci);
 	}
 }
 
diff --git a/net/8021q/vlan.c b/net/8021q/vlan.c
index 7a2625d2f9a0..68bdcf4a795c 100644
--- a/net/8021q/vlan.c
+++ b/net/8021q/vlan.c
@@ -83,13 +83,12 @@ static struct vlan_group *__vlan_find_group(struct net_device *real_dev)
  *
  * Must be invoked with RCU read lock (no preempt)
  */
-struct net_device *__find_vlan_dev(struct net_device *real_dev,
-				   unsigned short VID)
+struct net_device *__find_vlan_dev(struct net_device *real_dev, u16 vlan_id)
 {
 	struct vlan_group *grp = __vlan_find_group(real_dev);
 
 	if (grp)
-		return vlan_group_get_device(grp, VID);
+		return vlan_group_get_device(grp, vlan_id);
 
 	return NULL;
 }
@@ -117,14 +116,14 @@ static struct vlan_group *vlan_group_alloc(struct net_device *real_dev)
 	return grp;
 }
 
-static int vlan_group_prealloc_vid(struct vlan_group *vg, int vid)
+static int vlan_group_prealloc_vid(struct vlan_group *vg, u16 vlan_id)
 {
 	struct net_device **array;
 	unsigned int size;
 
 	ASSERT_RTNL();
 
-	array = vg->vlan_devices_arrays[vid / VLAN_GROUP_ARRAY_PART_LEN];
+	array = vg->vlan_devices_arrays[vlan_id / VLAN_GROUP_ARRAY_PART_LEN];
 	if (array != NULL)
 		return 0;
 
@@ -133,7 +132,7 @@ static int vlan_group_prealloc_vid(struct vlan_group *vg, int vid)
 	if (array == NULL)
 		return -ENOBUFS;
 
-	vg->vlan_devices_arrays[vid / VLAN_GROUP_ARRAY_PART_LEN] = array;
+	vg->vlan_devices_arrays[vlan_id / VLAN_GROUP_ARRAY_PART_LEN] = array;
 	return 0;
 }
 
@@ -147,7 +146,7 @@ void unregister_vlan_dev(struct net_device *dev)
 	struct vlan_dev_info *vlan = vlan_dev_info(dev);
 	struct net_device *real_dev = vlan->real_dev;
 	struct vlan_group *grp;
-	unsigned short vlan_id = vlan->vlan_id;
+	u16 vlan_id = vlan->vlan_id;
 
 	ASSERT_RTNL();
 
@@ -205,7 +204,7 @@ static void vlan_transfer_operstate(const struct net_device *dev,
 	}
 }
 
-int vlan_check_real_dev(struct net_device *real_dev, unsigned short vlan_id)
+int vlan_check_real_dev(struct net_device *real_dev, u16 vlan_id)
 {
 	char *name = real_dev->name;
 
@@ -242,7 +241,7 @@ int register_vlan_dev(struct net_device *dev)
 {
 	struct vlan_dev_info *vlan = vlan_dev_info(dev);
 	struct net_device *real_dev = vlan->real_dev;
-	unsigned short vlan_id = vlan->vlan_id;
+	u16 vlan_id = vlan->vlan_id;
 	struct vlan_group *grp, *ngrp = NULL;
 	int err;
 
@@ -295,8 +294,7 @@ out_free_group:
 /*  Attach a VLAN device to a mac address (ie Ethernet Card).
  *  Returns 0 if the device was created or a negative error code otherwise.
  */
-static int register_vlan_device(struct net_device *real_dev,
-				unsigned short VLAN_ID)
+static int register_vlan_device(struct net_device *real_dev, u16 vlan_id)
 {
 	struct net_device *new_dev;
 	struct net *net = dev_net(real_dev);
@@ -304,10 +302,10 @@ static int register_vlan_device(struct net_device *real_dev,
 	char name[IFNAMSIZ];
 	int err;
 
-	if (VLAN_ID >= VLAN_VID_MASK)
+	if (vlan_id >= VLAN_VID_MASK)
 		return -ERANGE;
 
-	err = vlan_check_real_dev(real_dev, VLAN_ID);
+	err = vlan_check_real_dev(real_dev, vlan_id);
 	if (err < 0)
 		return err;
 
@@ -315,26 +313,26 @@ static int register_vlan_device(struct net_device *real_dev,
 	switch (vn->name_type) {
 	case VLAN_NAME_TYPE_RAW_PLUS_VID:
 		/* name will look like:	 eth1.0005 */
-		snprintf(name, IFNAMSIZ, "%s.%.4i", real_dev->name, VLAN_ID);
+		snprintf(name, IFNAMSIZ, "%s.%.4i", real_dev->name, vlan_id);
 		break;
 	case VLAN_NAME_TYPE_PLUS_VID_NO_PAD:
 		/* Put our vlan.VID in the name.
 		 * Name will look like:	 vlan5
 		 */
-		snprintf(name, IFNAMSIZ, "vlan%i", VLAN_ID);
+		snprintf(name, IFNAMSIZ, "vlan%i", vlan_id);
 		break;
 	case VLAN_NAME_TYPE_RAW_PLUS_VID_NO_PAD:
 		/* Put our vlan.VID in the name.
 		 * Name will look like:	 eth0.5
 		 */
-		snprintf(name, IFNAMSIZ, "%s.%i", real_dev->name, VLAN_ID);
+		snprintf(name, IFNAMSIZ, "%s.%i", real_dev->name, vlan_id);
 		break;
 	case VLAN_NAME_TYPE_PLUS_VID:
 		/* Put our vlan.VID in the name.
 		 * Name will look like:	 vlan0005
 		 */
 	default:
-		snprintf(name, IFNAMSIZ, "vlan%.4i", VLAN_ID);
+		snprintf(name, IFNAMSIZ, "vlan%.4i", vlan_id);
 	}
 
 	new_dev = alloc_netdev(sizeof(struct vlan_dev_info), name,
@@ -349,7 +347,7 @@ static int register_vlan_device(struct net_device *real_dev,
 	 */
 	new_dev->mtu = real_dev->mtu;
 
-	vlan_dev_info(new_dev)->vlan_id = VLAN_ID; /* 1 through VLAN_VID_MASK */
+	vlan_dev_info(new_dev)->vlan_id = vlan_id;
 	vlan_dev_info(new_dev)->real_dev = real_dev;
 	vlan_dev_info(new_dev)->dent = NULL;
 	vlan_dev_info(new_dev)->flags = VLAN_FLAG_REORDER_HDR;
diff --git a/net/8021q/vlan.h b/net/8021q/vlan.h
index 14c421e033f2..a6603a4d917f 100644
--- a/net/8021q/vlan.h
+++ b/net/8021q/vlan.h
@@ -12,7 +12,7 @@
  */
 struct vlan_priority_tci_mapping {
 	u32					priority;
-	unsigned short				vlan_qos;
+	u16					vlan_qos;
 	struct vlan_priority_tci_mapping	*next;
 };
 
@@ -36,8 +36,8 @@ struct vlan_dev_info {
 	unsigned int				nr_egress_mappings;
 	struct vlan_priority_tci_mapping	*egress_priority_map[16];
 
-	unsigned short				vlan_id;
-	unsigned short				flags;
+	u16					vlan_id;
+	u16					flags;
 
 	struct net_device			*real_dev;
 	unsigned char				real_dev_addr[ETH_ALEN];
@@ -67,30 +67,29 @@ static inline struct vlan_dev_info *vlan_dev_info(const struct net_device *dev)
  *  Must be invoked with rcu_read_lock (ie preempt disabled)
  *  or with RTNL.
  */
-struct net_device *__find_vlan_dev(struct net_device *real_dev,
-				   unsigned short VID); /* vlan.c */
+struct net_device *__find_vlan_dev(struct net_device *real_dev, u16 vlan_id);
 
 /* found in vlan_dev.c */
 int vlan_skb_recv(struct sk_buff *skb, struct net_device *dev,
 		  struct packet_type *ptype, struct net_device *orig_dev);
 void vlan_dev_set_ingress_priority(const struct net_device *dev,
-				   u32 skb_prio, short vlan_prio);
+				   u32 skb_prio, u16 vlan_prio);
 int vlan_dev_set_egress_priority(const struct net_device *dev,
-				 u32 skb_prio, short vlan_prio);
+				 u32 skb_prio, u16 vlan_prio);
 int vlan_dev_change_flags(const struct net_device *dev, u32 flag, u32 mask);
 void vlan_dev_get_realdev_name(const struct net_device *dev, char *result);
 
-int vlan_check_real_dev(struct net_device *real_dev, unsigned short vlan_id);
+int vlan_check_real_dev(struct net_device *real_dev, u16 vlan_id);
 void vlan_setup(struct net_device *dev);
 int register_vlan_dev(struct net_device *dev);
 void unregister_vlan_dev(struct net_device *dev);
 
 static inline u32 vlan_get_ingress_priority(struct net_device *dev,
-					    unsigned short vlan_tag)
+					    u16 vlan_tci)
 {
 	struct vlan_dev_info *vip = vlan_dev_info(dev);
 
-	return vip->ingress_priority_map[(vlan_tag >> 13) & 0x7];
+	return vip->ingress_priority_map[(vlan_tci >> 13) & 0x7];
 }
 
 #ifdef CONFIG_VLAN_8021Q_GVRP
diff --git a/net/8021q/vlan_core.c b/net/8021q/vlan_core.c
index f980b9154cc3..68df12d3664b 100644
--- a/net/8021q/vlan_core.c
+++ b/net/8021q/vlan_core.c
@@ -5,7 +5,7 @@
 
 /* VLAN rx hw acceleration helper.  This acts like netif_{rx,receive_skb}(). */
 int __vlan_hwaccel_rx(struct sk_buff *skb, struct vlan_group *grp,
-		      unsigned short vlan_tag, int polling)
+		      u16 vlan_tci, int polling)
 {
 	struct net_device_stats *stats;
 
@@ -14,7 +14,7 @@ int __vlan_hwaccel_rx(struct sk_buff *skb, struct vlan_group *grp,
 		return NET_RX_DROP;
 	}
 
-	skb->dev = vlan_group_get_device(grp, vlan_tag & VLAN_VID_MASK);
+	skb->dev = vlan_group_get_device(grp, vlan_tci & VLAN_VID_MASK);
 	if (skb->dev == NULL) {
 		dev_kfree_skb_any(skb);
 		/* Not NET_RX_DROP, this is not being dropped
@@ -27,7 +27,7 @@ int __vlan_hwaccel_rx(struct sk_buff *skb, struct vlan_group *grp,
 	stats->rx_packets++;
 	stats->rx_bytes += skb->len;
 
-	skb->priority = vlan_get_ingress_priority(skb->dev, vlan_tag);
+	skb->priority = vlan_get_ingress_priority(skb->dev, vlan_tci);
 	switch (skb->pkt_type) {
 	case PACKET_BROADCAST:
 		break;
diff --git a/net/8021q/vlan_dev.c b/net/8021q/vlan_dev.c
index 2aab294c5744..2ccac6bea57e 100644
--- a/net/8021q/vlan_dev.c
+++ b/net/8021q/vlan_dev.c
@@ -150,9 +150,9 @@ int vlan_skb_recv(struct sk_buff *skb, struct net_device *dev,
 		  struct packet_type *ptype, struct net_device *orig_dev)
 {
 	struct vlan_hdr *vhdr;
-	unsigned short vid;
 	struct net_device_stats *stats;
-	unsigned short vlan_TCI;
+	u16 vlan_id;
+	u16 vlan_tci;
 
 	skb = skb_share_check(skb, GFP_ATOMIC);
 	if (skb == NULL)
@@ -162,14 +162,14 @@ int vlan_skb_recv(struct sk_buff *skb, struct net_device *dev,
 		goto err_free;
 
 	vhdr = (struct vlan_hdr *)skb->data;
-	vlan_TCI = ntohs(vhdr->h_vlan_TCI);
-	vid = (vlan_TCI & VLAN_VID_MASK);
+	vlan_tci = ntohs(vhdr->h_vlan_TCI);
+	vlan_id = vlan_tci & VLAN_VID_MASK;
 
 	rcu_read_lock();
-	skb->dev = __find_vlan_dev(dev, vid);
+	skb->dev = __find_vlan_dev(dev, vlan_id);
 	if (!skb->dev) {
 		pr_debug("%s: ERROR: No net_device for VID: %u on dev: %s\n",
-			 __func__, (unsigned int)vid, dev->name);
+			 __func__, vlan_id, dev->name);
 		goto err_unlock;
 	}
 
@@ -181,11 +181,10 @@ int vlan_skb_recv(struct sk_buff *skb, struct net_device *dev,
 
 	skb_pull_rcsum(skb, VLAN_HLEN);
 
-	skb->priority = vlan_get_ingress_priority(skb->dev,
-						  ntohs(vhdr->h_vlan_TCI));
+	skb->priority = vlan_get_ingress_priority(skb->dev, vlan_tci);
 
 	pr_debug("%s: priority: %u for TCI: %hu\n",
-		 __func__, skb->priority, ntohs(vhdr->h_vlan_TCI));
+		 __func__, skb->priority, vlan_tci);
 
 	switch (skb->pkt_type) {
 	case PACKET_BROADCAST: /* Yeah, stats collect these together.. */
@@ -228,7 +227,7 @@ err_free:
 	return NET_RX_DROP;
 }
 
-static inline unsigned short
+static inline u16
 vlan_dev_get_egress_qos_mask(struct net_device *dev, struct sk_buff *skb)
 {
 	struct vlan_priority_tci_mapping *mp;
@@ -260,7 +259,7 @@ static int vlan_dev_hard_header(struct sk_buff *skb, struct net_device *dev,
 				unsigned int len)
 {
 	struct vlan_hdr *vhdr;
-	unsigned short veth_TCI = 0;
+	u16 vlan_tci = 0;
 	int rc = 0;
 	int build_vlan_header = 0;
 	struct net_device *vdev = dev;
@@ -292,10 +291,10 @@ static int vlan_dev_hard_header(struct sk_buff *skb, struct net_device *dev,
 		 * VLAN ID	 12 bits (low bits)
 		 *
 		 */
-		veth_TCI = vlan_dev_info(dev)->vlan_id;
-		veth_TCI |= vlan_dev_get_egress_qos_mask(dev, skb);
+		vlan_tci = vlan_dev_info(dev)->vlan_id;
+		vlan_tci |= vlan_dev_get_egress_qos_mask(dev, skb);
 
-		vhdr->h_vlan_TCI = htons(veth_TCI);
+		vhdr->h_vlan_TCI = htons(vlan_tci);
 
 		/*
 		 *  Set the protocol type. For a packet of type ETH_P_802_3 we
@@ -373,7 +372,7 @@ static int vlan_dev_hard_start_xmit(struct sk_buff *skb, struct net_device *dev)
 	if (veth->h_vlan_proto != htons(ETH_P_8021Q) ||
 		vlan_dev_info(dev)->flags & VLAN_FLAG_REORDER_HDR) {
 		int orig_headroom = skb_headroom(skb);
-		unsigned short veth_TCI;
+		u16 vlan_tci;
 
 		/* This is not a VLAN frame...but we can fix that! */
 		vlan_dev_info(dev)->cnt_encap_on_xmit++;
@@ -386,10 +385,10 @@ static int vlan_dev_hard_start_xmit(struct sk_buff *skb, struct net_device *dev)
 		 * CFI		 1 bit
 		 * VLAN ID	 12 bits (low bits)
 		 */
-		veth_TCI = vlan_dev_info(dev)->vlan_id;
-		veth_TCI |= vlan_dev_get_egress_qos_mask(dev, skb);
+		vlan_tci = vlan_dev_info(dev)->vlan_id;
+		vlan_tci |= vlan_dev_get_egress_qos_mask(dev, skb);
 
-		skb = __vlan_put_tag(skb, veth_TCI);
+		skb = __vlan_put_tag(skb, vlan_tci);
 		if (!skb) {
 			stats->tx_dropped++;
 			return 0;
@@ -422,7 +421,7 @@ static int vlan_dev_hwaccel_hard_start_xmit(struct sk_buff *skb,
 					    struct net_device *dev)
 {
 	struct net_device_stats *stats = &dev->stats;
-	unsigned short veth_TCI;
+	u16 vlan_tci;
 
 	/* Construct the second two bytes. This field looks something
 	 * like:
@@ -430,9 +429,9 @@ static int vlan_dev_hwaccel_hard_start_xmit(struct sk_buff *skb,
 	 * CFI		 1 bit
 	 * VLAN ID	 12 bits (low bits)
 	 */
-	veth_TCI = vlan_dev_info(dev)->vlan_id;
-	veth_TCI |= vlan_dev_get_egress_qos_mask(dev, skb);
-	skb = __vlan_hwaccel_put_tag(skb, veth_TCI);
+	vlan_tci = vlan_dev_info(dev)->vlan_id;
+	vlan_tci |= vlan_dev_get_egress_qos_mask(dev, skb);
+	skb = __vlan_hwaccel_put_tag(skb, vlan_tci);
 
 	stats->tx_packets++;
 	stats->tx_bytes += skb->len;
@@ -457,7 +456,7 @@ static int vlan_dev_change_mtu(struct net_device *dev, int new_mtu)
 }
 
 void vlan_dev_set_ingress_priority(const struct net_device *dev,
-				   u32 skb_prio, short vlan_prio)
+				   u32 skb_prio, u16 vlan_prio)
 {
 	struct vlan_dev_info *vlan = vlan_dev_info(dev);
 
@@ -470,7 +469,7 @@ void vlan_dev_set_ingress_priority(const struct net_device *dev,
 }
 
 int vlan_dev_set_egress_priority(const struct net_device *dev,
-				 u32 skb_prio, short vlan_prio)
+				 u32 skb_prio, u16 vlan_prio)
 {
 	struct vlan_dev_info *vlan = vlan_dev_info(dev);
 	struct vlan_priority_tci_mapping *mp = NULL;
diff --git a/net/8021q/vlan_gvrp.c b/net/8021q/vlan_gvrp.c
index db9781608362..061ceceeef12 100644
--- a/net/8021q/vlan_gvrp.c
+++ b/net/8021q/vlan_gvrp.c
@@ -30,19 +30,19 @@ static struct garp_application vlan_gvrp_app __read_mostly = {
 int vlan_gvrp_request_join(const struct net_device *dev)
 {
 	const struct vlan_dev_info *vlan = vlan_dev_info(dev);
-	__be16 vid = htons(vlan->vlan_id);
+	__be16 vlan_id = htons(vlan->vlan_id);
 
 	return garp_request_join(vlan->real_dev, &vlan_gvrp_app,
-				 &vid, sizeof(vid), GVRP_ATTR_VID);
+				 &vlan_id, sizeof(vlan_id), GVRP_ATTR_VID);
 }
 
 void vlan_gvrp_request_leave(const struct net_device *dev)
 {
 	const struct vlan_dev_info *vlan = vlan_dev_info(dev);
-	__be16 vid = htons(vlan->vlan_id);
+	__be16 vlan_id = htons(vlan->vlan_id);
 
 	garp_request_leave(vlan->real_dev, &vlan_gvrp_app,
-			   &vid, sizeof(vid), GVRP_ATTR_VID);
+			   &vlan_id, sizeof(vlan_id), GVRP_ATTR_VID);
 }
 
 int vlan_gvrp_init_applicant(struct net_device *dev)
-- 
cgit v1.2.3


From 429a380571a6e6b8525b93161544eafc9b227e44 Mon Sep 17 00:00:00 2001
From: Ron Rindjunsky <ron.rindjunsky@intel.com>
Date: Tue, 1 Jul 2008 14:16:03 +0300
Subject: mac80211: add block ack request capability

This patch adds block ack request capability

Signed-off-by: Ester Kummer <ester.kummer@intel.com>
Signed-off-by: Tomas Winkler <tomas.winkler@intel.com>
Signed-off-by: Ron Rindjunsky <ron.rindjunsky@intel.com>
Acked-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/linux/ieee80211.h  |  4 ++++
 include/net/mac80211.h     |  3 +++
 net/mac80211/ieee80211_i.h |  1 +
 net/mac80211/main.c        | 21 +++++++++++++++++++--
 net/mac80211/mlme.c        | 29 +++++++++++++++++++++++++++++
 5 files changed, 56 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/linux/ieee80211.h b/include/linux/ieee80211.h
index cffd6d0094f9..aa603c3d76d1 100644
--- a/include/linux/ieee80211.h
+++ b/include/linux/ieee80211.h
@@ -658,6 +658,10 @@ struct ieee80211_bar {
 	__le16 start_seq_num;
 } __attribute__((packed));
 
+/* 802.11 BAR control masks */
+#define IEEE80211_BAR_CTRL_ACK_POLICY_NORMAL     0x0000
+#define IEEE80211_BAR_CTRL_CBMTID_COMPRESSED_BA  0x0004
+
 /**
  * struct ieee80211_ht_cap - HT capabilities
  *
diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index 3a204acad901..0a5de3ef527a 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -235,6 +235,8 @@ struct ieee80211_bss_conf {
  * @IEEE80211_TX_STAT_ACK: Frame was acknowledged
  * @IEEE80211_TX_STAT_AMPDU: The frame was aggregated, so status
  * 	is for the whole aggregation.
+ * @IEEE80211_TX_STAT_AMPDU_NO_BACK: no block ack was returned,
+ * 	so consider using block ack request (BAR).
  */
 enum mac80211_tx_control_flags {
 	IEEE80211_TX_CTL_REQ_TX_STATUS		= BIT(0),
@@ -260,6 +262,7 @@ enum mac80211_tx_control_flags {
 	IEEE80211_TX_STAT_TX_FILTERED		= BIT(20),
 	IEEE80211_TX_STAT_ACK			= BIT(21),
 	IEEE80211_TX_STAT_AMPDU			= BIT(22),
+	IEEE80211_TX_STAT_AMPDU_NO_BACK		= BIT(23),
 };
 
 
diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h
index f90da1bbec49..175cbdd36d7f 100644
--- a/net/mac80211/ieee80211_i.h
+++ b/net/mac80211/ieee80211_i.h
@@ -904,6 +904,7 @@ void ieee80211_send_addba_request(struct net_device *dev, const u8 *da,
 				  u16 agg_size, u16 timeout);
 void ieee80211_send_delba(struct net_device *dev, const u8 *da, u16 tid,
 				u16 initiator, u16 reason_code);
+void ieee80211_send_bar(struct net_device *dev, u8 *ra, u16 tid, u16 ssn);
 
 void ieee80211_sta_stop_rx_ba_session(struct net_device *dev, u8 *da,
 				u16 tid, u16 initiator, u16 reason);
diff --git a/net/mac80211/main.c b/net/mac80211/main.c
index f18cfd727872..074f71a62a6d 100644
--- a/net/mac80211/main.c
+++ b/net/mac80211/main.c
@@ -1404,14 +1404,15 @@ void ieee80211_tx_status(struct ieee80211_hw *hw, struct sk_buff *skb)
 	struct ieee80211_local *local = hw_to_local(hw);
 	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
 	u16 frag, type;
+	__le16 fc;
 	struct ieee80211_tx_status_rtap_hdr *rthdr;
 	struct ieee80211_sub_if_data *sdata;
 	struct net_device *prev_dev = NULL;
+	struct sta_info *sta;
 
 	rcu_read_lock();
 
 	if (info->status.excessive_retries) {
-		struct sta_info *sta;
 		sta = sta_info_get(local, hdr->addr1);
 		if (sta) {
 			if (test_sta_flags(sta, WLAN_STA_PS)) {
@@ -1426,8 +1427,24 @@ void ieee80211_tx_status(struct ieee80211_hw *hw, struct sk_buff *skb)
 		}
 	}
 
+	fc = hdr->frame_control;
+
+	if ((info->flags & IEEE80211_TX_STAT_AMPDU_NO_BACK) &&
+	    (ieee80211_is_data_qos(fc))) {
+		u16 tid, ssn;
+		u8 *qc;
+		sta = sta_info_get(local, hdr->addr1);
+		if (sta) {
+			qc = ieee80211_get_qos_ctl(hdr);
+			tid = qc[0] & 0xf;
+			ssn = ((le16_to_cpu(hdr->seq_ctrl) + 0x10)
+						& IEEE80211_SCTL_SEQ);
+			ieee80211_send_bar(sta->sdata->dev, hdr->addr1,
+					   tid, ssn);
+		}
+	}
+
 	if (info->flags & IEEE80211_TX_STAT_TX_FILTERED) {
-		struct sta_info *sta;
 		sta = sta_info_get(local, hdr->addr1);
 		if (sta) {
 			ieee80211_handle_filtered_frame(local, sta, skb);
diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c
index 86abdf96390c..e080482b63cd 100644
--- a/net/mac80211/mlme.c
+++ b/net/mac80211/mlme.c
@@ -1536,6 +1536,35 @@ void ieee80211_send_delba(struct net_device *dev, const u8 *da, u16 tid,
 	ieee80211_sta_tx(dev, skb, 0);
 }
 
+void ieee80211_send_bar(struct net_device *dev, u8 *ra, u16 tid, u16 ssn)
+{
+	struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr);
+	struct sk_buff *skb;
+	struct ieee80211_bar *bar;
+	u16 bar_control = 0;
+
+	skb = dev_alloc_skb(sizeof(*bar) + local->hw.extra_tx_headroom);
+	if (!skb) {
+		printk(KERN_ERR "%s: failed to allocate buffer for "
+			"bar frame\n", dev->name);
+		return;
+	}
+	skb_reserve(skb, local->hw.extra_tx_headroom);
+	bar = (struct ieee80211_bar *)skb_put(skb, sizeof(*bar));
+	memset(bar, 0, sizeof(*bar));
+	bar->frame_control = IEEE80211_FC(IEEE80211_FTYPE_CTL,
+					IEEE80211_STYPE_BACK_REQ);
+	memcpy(bar->ra, ra, ETH_ALEN);
+	memcpy(bar->ta, dev->dev_addr, ETH_ALEN);
+	bar_control |= (u16)IEEE80211_BAR_CTRL_ACK_POLICY_NORMAL;
+	bar_control |= (u16)IEEE80211_BAR_CTRL_CBMTID_COMPRESSED_BA;
+	bar_control |= (u16)(tid << 12);
+	bar->control = cpu_to_le16(bar_control);
+	bar->start_seq_num = cpu_to_le16(ssn);
+
+	ieee80211_sta_tx(dev, skb, 0);
+}
+
 void ieee80211_sta_stop_rx_ba_session(struct net_device *dev, u8 *ra, u16 tid,
 					u16 initiator, u16 reason)
 {
-- 
cgit v1.2.3


From 238f74a227fd7de8ea1bc66dcbbd36cf9920d1cb Mon Sep 17 00:00:00 2001
From: Harvey Harrison <harvey.harrison@gmail.com>
Date: Wed, 2 Jul 2008 11:05:34 -0700
Subject: mac80211: move QOS control helpers into ieee80211.h

Also remove the WLAN_IS_QOS_DATA inline after removing the last
two users.  This starts moving away from using rx->fc to using
the header frame_control directly.

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/linux/ieee80211.h |  2 ++
 net/mac80211/rx.c         | 28 ++++++++++++++--------------
 net/mac80211/wme.c        |  4 ++--
 net/mac80211/wme.h        |  8 --------
 4 files changed, 18 insertions(+), 24 deletions(-)

(limited to 'include')

diff --git a/include/linux/ieee80211.h b/include/linux/ieee80211.h
index aa603c3d76d1..a1630ba0b87c 100644
--- a/include/linux/ieee80211.h
+++ b/include/linux/ieee80211.h
@@ -99,6 +99,8 @@
 #define IEEE80211_MAX_SSID_LEN		32
 #define IEEE80211_MAX_MESH_ID_LEN	32
 #define IEEE80211_QOS_CTL_LEN		2
+#define IEEE80211_QOS_CTL_TID_MASK	0x000F
+#define IEEE80211_QOS_CTL_TAG1D_MASK	0x0007
 
 struct ieee80211_hdr {
 	__le16 frame_control;
diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c
index 6a88e8f9bff0..a3a26e557274 100644
--- a/net/mac80211/rx.c
+++ b/net/mac80211/rx.c
@@ -321,20 +321,20 @@ ieee80211_rx_monitor(struct ieee80211_local *local, struct sk_buff *origskb,
 
 static void ieee80211_parse_qos(struct ieee80211_rx_data *rx)
 {
-	u8 *data = rx->skb->data;
+	struct ieee80211_hdr *hdr = (struct ieee80211_hdr *)rx->skb->data;
 	int tid;
 
 	/* does the frame have a qos control field? */
-	if (WLAN_FC_IS_QOS_DATA(rx->fc)) {
-		u8 *qc = data + ieee80211_get_hdrlen(rx->fc) - QOS_CONTROL_LEN;
+	if (ieee80211_is_data_qos(hdr->frame_control)) {
+		u8 *qc = ieee80211_get_qos_ctl(hdr);
 		/* frame has qos control */
-		tid = qc[0] & QOS_CONTROL_TID_MASK;
-		if (qc[0] & IEEE80211_QOS_CONTROL_A_MSDU_PRESENT)
+		tid = *qc & IEEE80211_QOS_CTL_TID_MASK;
+		if (*qc & IEEE80211_QOS_CONTROL_A_MSDU_PRESENT)
 			rx->flags |= IEEE80211_RX_AMSDU;
 		else
 			rx->flags &= ~IEEE80211_RX_AMSDU;
 	} else {
-		if (unlikely((rx->fc & IEEE80211_FCTL_FTYPE) == IEEE80211_FTYPE_MGMT)) {
+		if (unlikely(ieee80211_is_mgmt(hdr->frame_control))) {
 			/* Separate TID for management frames */
 			tid = NUM_RX_DATA_QUEUES - 1;
 		} else {
@@ -1037,19 +1037,19 @@ ieee80211_rx_h_ps_poll(struct ieee80211_rx_data *rx)
 static ieee80211_rx_result debug_noinline
 ieee80211_rx_h_remove_qos_control(struct ieee80211_rx_data *rx)
 {
-	u16 fc = rx->fc;
 	u8 *data = rx->skb->data;
-	struct ieee80211_hdr *hdr = (struct ieee80211_hdr *) data;
+	struct ieee80211_hdr *hdr = (struct ieee80211_hdr *)data;
 
-	if (!WLAN_FC_IS_QOS_DATA(fc))
+	if (!ieee80211_is_data_qos(hdr->frame_control))
 		return RX_CONTINUE;
 
 	/* remove the qos control field, update frame type and meta-data */
-	memmove(data + 2, data, ieee80211_get_hdrlen(fc) - 2);
-	hdr = (struct ieee80211_hdr *) skb_pull(rx->skb, 2);
+	memmove(data + IEEE80211_QOS_CTL_LEN, data,
+		ieee80211_hdrlen(hdr->frame_control) - IEEE80211_QOS_CTL_LEN);
+	hdr = (struct ieee80211_hdr *)skb_pull(rx->skb, IEEE80211_QOS_CTL_LEN);
 	/* change frame type to non QOS */
-	rx->fc = fc &= ~IEEE80211_STYPE_QOS_DATA;
-	hdr->frame_control = cpu_to_le16(fc);
+	rx->fc &= ~IEEE80211_STYPE_QOS_DATA;
+	hdr->frame_control &= ~cpu_to_le16(IEEE80211_STYPE_QOS_DATA);
 
 	return RX_CONTINUE;
 }
@@ -2044,7 +2044,7 @@ static u8 ieee80211_rx_reorder_ampdu(struct ieee80211_local *local,
 	if (!ieee80211_is_data_qos(hdr->frame_control))
 		goto end_reorder;
 
-	tid = *ieee80211_get_qos_ctl(hdr) & QOS_CONTROL_TID_MASK;
+	tid = *ieee80211_get_qos_ctl(hdr) & IEEE80211_QOS_CTL_TID_MASK;
 
 	if (sta->ampdu_mlme.tid_state_rx[tid] != HT_AGG_STATE_OPERATIONAL)
 		goto end_reorder;
diff --git a/net/mac80211/wme.c b/net/mac80211/wme.c
index a1a53a4f2b99..5c666f7eda8f 100644
--- a/net/mac80211/wme.c
+++ b/net/mac80211/wme.c
@@ -154,7 +154,7 @@ static int wme_qdiscop_enqueue(struct sk_buff *skb, struct Qdisc* qd)
 		queue = skb_get_queue_mapping(skb);
 		rcu_read_lock();
 		sta = sta_info_get(local, hdr->addr1);
-		tid = skb->priority & QOS_CONTROL_TAG1D_MASK;
+		tid = skb->priority & IEEE80211_QOS_CTL_TAG1D_MASK;
 		if (sta) {
 			int ampdu_queue = sta->tid_to_tx_q[tid];
 			if ((ampdu_queue < QD_NUM(hw)) &&
@@ -181,7 +181,7 @@ static int wme_qdiscop_enqueue(struct sk_buff *skb, struct Qdisc* qd)
 	if (ieee80211_is_data_qos(hdr->frame_control)) {
 		u8 *p = ieee80211_get_qos_ctl(hdr);
 		u8 ack_policy = 0;
-		tid = skb->priority & QOS_CONTROL_TAG1D_MASK;
+		tid = skb->priority & IEEE80211_QOS_CTL_TAG1D_MASK;
 		if (local->wifi_wme_noack_test)
 			ack_policy |= QOS_CONTROL_ACK_POLICY_NOACK <<
 					QOS_CONTROL_ACK_POLICY_SHIFT;
diff --git a/net/mac80211/wme.h b/net/mac80211/wme.h
index bbdb53344817..1aca609eccfc 100644
--- a/net/mac80211/wme.h
+++ b/net/mac80211/wme.h
@@ -19,18 +19,10 @@
 #define QOS_CONTROL_ACK_POLICY_NORMAL 0
 #define QOS_CONTROL_ACK_POLICY_NOACK 1
 
-#define QOS_CONTROL_TID_MASK 0x0f
 #define QOS_CONTROL_ACK_POLICY_SHIFT 5
 
-#define QOS_CONTROL_TAG1D_MASK 0x07
-
 extern const int ieee802_1d_to_ac[8];
 
-static inline int WLAN_FC_IS_QOS_DATA(u16 fc)
-{
-	return (fc & 0x8C) == 0x88;
-}
-
 #ifdef CONFIG_MAC80211_QOS
 void ieee80211_install_qdisc(struct net_device *dev);
 int ieee80211_qdisc_installed(struct net_device *dev);
-- 
cgit v1.2.3


From 6ef307bc561911c8cdda98ef3896b5982b602a43 Mon Sep 17 00:00:00 2001
From: Randy Dunlap <randy.dunlap@oracle.com>
Date: Thu, 3 Jul 2008 13:52:18 -0700
Subject: mac80211: fix lots of kernel-doc

Fix more than 50 kernel-doc warnings in ieee80211/mac80211 kernel-doc notation.
Fix a few typos also.

Note: Some fields are marked as TBD and need to have their description
corrected.

Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/net/mac80211.h  | 40 +++++++++++++++---------
 net/mac80211/sta_info.c |  1 +
 net/mac80211/sta_info.h | 81 ++++++++++++++++++++++++++++++++++++-------------
 3 files changed, 86 insertions(+), 36 deletions(-)

(limited to 'include')

diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index 0a5de3ef527a..656442c6b1c3 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -42,7 +42,7 @@
  * tasklet function.
  *
  * NOTE: If the driver opts to use the _irqsafe() functions, it may not also
- *	 use the non-irqsafe functions!
+ *	 use the non-IRQ-safe functions!
  */
 
 /**
@@ -85,7 +85,7 @@ enum ieee80211_notification_types {
  * struct ieee80211_ht_bss_info - describing BSS's HT characteristics
  *
  * This structure describes most essential parameters needed
- * to describe 802.11n HT characteristics in a BSS
+ * to describe 802.11n HT characteristics in a BSS.
  *
  * @primary_channel: channel number of primery channel
  * @bss_cap: 802.11n's general BSS capabilities (e.g. channel width)
@@ -201,9 +201,9 @@ struct ieee80211_bss_conf {
 };
 
 /**
- * enum mac80211_tx_flags - flags to transmission information/status
+ * enum mac80211_tx_control_flags - flags to describe transmission information/status
  *
- * These flags are used with the @flags member of &ieee80211_tx_info
+ * These flags are used with the @flags member of &ieee80211_tx_info.
  *
  * @IEEE80211_TX_CTL_REQ_TX_STATUS: request TX status callback for this frame.
  * @IEEE80211_TX_CTL_DO_NOT_ENCRYPT: send this frame without encryption;
@@ -212,11 +212,12 @@ struct ieee80211_bss_conf {
  * @IEEE80211_TX_CTL_USE_CTS_PROTECT: use CTS protection for the frame (e.g.,
  *	for combined 802.11g / 802.11b networks)
  * @IEEE80211_TX_CTL_NO_ACK: tell the low level not to wait for an ack
- * @IEEE80211_TX_CTL_RATE_CTRL_PROBE
+ * @IEEE80211_TX_CTL_RATE_CTRL_PROBE: TBD
  * @IEEE80211_TX_CTL_CLEAR_PS_FILT: clear powersave filter for destination
  *	station
- * @IEEE80211_TX_CTL_REQUEUE:
+ * @IEEE80211_TX_CTL_REQUEUE: TBD
  * @IEEE80211_TX_CTL_FIRST_FRAGMENT: this is a first fragment of the frame
+ * @IEEE80211_TX_CTL_SHORT_PREAMBLE: TBD
  * @IEEE80211_TX_CTL_LONG_RETRY_LIMIT: this frame should be send using the
  *	through set_retry_limit configured long retry value
  * @IEEE80211_TX_CTL_EAPOL_FRAME: internal to mac80211
@@ -230,6 +231,7 @@ struct ieee80211_bss_conf {
  * @IEEE80211_TX_CTL_40_MHZ_WIDTH: send this frame using 40 Mhz channel width
  * @IEEE80211_TX_CTL_DUP_DATA: duplicate data frame on both 20 Mhz channels
  * @IEEE80211_TX_CTL_SHORT_GI: send this frame using short guard interval
+ * @IEEE80211_TX_CTL_INJECTED: TBD
  * @IEEE80211_TX_STAT_TX_FILTERED: The frame was not transmitted
  *	because the destination STA was in powersave mode.
  * @IEEE80211_TX_STAT_ACK: Frame was acknowledged
@@ -280,6 +282,12 @@ enum mac80211_tx_control_flags {
  *  (3) TX status information - driver tells mac80211 what happened
  *
  * @flags: transmit info flags, defined above
+ * @band: TBD
+ * @tx_rate_idx: TBD
+ * @antenna_sel_tx: TBD
+ * @control: union for control data
+ * @status: union for status data
+ * @driver_data: array of driver_data pointers
  * @retry_count: number of retries
  * @excessive_retries: set to 1 if the frame was retried many times
  *	but not acknowledged
@@ -562,8 +570,8 @@ enum ieee80211_key_alg {
 
 /**
  * enum ieee80211_key_len - key length
- * @WEP40: WEP 5 byte long key
- * @WEP104: WEP 13 byte long key
+ * @LEN_WEP40: WEP 5-byte long key
+ * @LEN_WEP104: WEP 13-byte long key
  */
 enum ieee80211_key_len {
 	LEN_WEP40 = 5,
@@ -640,7 +648,7 @@ enum set_key_cmd {
  * enum sta_notify_cmd - sta notify command
  *
  * Used with the sta_notify() callback in &struct ieee80211_ops, this
- * indicates addition and removal of a station to station table
+ * indicates addition and removal of a station to station table.
  *
  * @STA_NOTIFY_ADD: a station was added to the station table
  * @STA_NOTIFY_REMOVE: a station being removed from the station table
@@ -1340,7 +1348,7 @@ void ieee80211_unregister_hw(struct ieee80211_hw *hw);
  *
  * This function frees everything that was allocated, including the
  * private data for the driver. You must call ieee80211_unregister_hw()
- * before calling this function
+ * before calling this function.
  *
  * @hw: the hardware to free
  */
@@ -1411,7 +1419,7 @@ void ieee80211_tx_status(struct ieee80211_hw *hw,
 			 struct sk_buff *skb);
 
 /**
- * ieee80211_tx_status_irqsafe - irq-safe transmit status callback
+ * ieee80211_tx_status_irqsafe - IRQ-safe transmit status callback
  *
  * Like ieee80211_tx_status() but can be called in IRQ context
  * (internally defers to a tasklet.)
@@ -1589,6 +1597,8 @@ unsigned int ieee80211_hdrlen(__le16 fc);
  * @keyconf: the parameter passed with the set key
  * @skb: the skb for which the key is needed
  * @rc4key: a buffer to which the key will be written
+ * @type: TBD
+ * @key: TBD
  */
 void ieee80211_get_tkip_key(struct ieee80211_key_conf *keyconf,
 				struct sk_buff *skb,
@@ -1639,7 +1649,7 @@ void ieee80211_wake_queues(struct ieee80211_hw *hw);
 void ieee80211_scan_completed(struct ieee80211_hw *hw);
 
 /**
- * ieee80211_iterate_active_interfaces- iterate active interfaces
+ * ieee80211_iterate_active_interfaces - iterate active interfaces
  *
  * This function iterates over the interfaces associated with a given
  * hardware that are currently active and calls the callback for them.
@@ -1706,7 +1716,7 @@ void ieee80211_start_tx_ba_cb(struct ieee80211_hw *hw, u8 *ra, u16 tid);
  *
  * This function must be called by low level driver once it has
  * finished with preparations for the BA session.
- * This version of the function is irq safe.
+ * This version of the function is IRQ-safe.
  */
 void ieee80211_start_tx_ba_cb_irqsafe(struct ieee80211_hw *hw, const u8 *ra,
 				      u16 tid);
@@ -1746,7 +1756,7 @@ void ieee80211_stop_tx_ba_cb(struct ieee80211_hw *hw, u8 *ra, u8 tid);
  *
  * This function must be called by low level driver once it has
  * finished with preparations for the BA session tear down.
- * This version of the function is irq safe.
+ * This version of the function is IRQ-safe.
  */
 void ieee80211_stop_tx_ba_cb_irqsafe(struct ieee80211_hw *hw, const u8 *ra,
 				     u16 tid);
@@ -1754,7 +1764,7 @@ void ieee80211_stop_tx_ba_cb_irqsafe(struct ieee80211_hw *hw, const u8 *ra,
 /**
  * ieee80211_notify_mac - low level driver notification
  * @hw: pointer as obtained from ieee80211_alloc_hw().
- * @notification_types: enum ieee80211_notification_types
+ * @notif_type: enum ieee80211_notification_types
  *
  * This function must be called by low level driver to inform mac80211 of
  * low level driver status change or force mac80211 to re-assoc for low
diff --git a/net/mac80211/sta_info.c b/net/mac80211/sta_info.c
index d8a16b7f6a6b..47d2c1bbfcaa 100644
--- a/net/mac80211/sta_info.c
+++ b/net/mac80211/sta_info.c
@@ -135,6 +135,7 @@ struct sta_info *sta_info_get_by_idx(struct ieee80211_local *local, int idx,
 /**
  * __sta_info_free - internal STA free helper
  *
+ * @local: pointer to the global information
  * @sta: STA info to free
  *
  * This function must undo everything done by sta_info_alloc()
diff --git a/net/mac80211/sta_info.h b/net/mac80211/sta_info.h
index fd228c198e31..94311dcfe043 100644
--- a/net/mac80211/sta_info.h
+++ b/net/mac80211/sta_info.h
@@ -160,11 +160,21 @@ struct sta_ampdu_mlme {
  * @list: global linked list entry
  * @hnext: hash table linked list pointer
  * @local: pointer to the global information
+ * @sdata: TBD
+ * @key: TBD
+ * @rate_ctrl: TBD
+ * @rate_ctrl_priv: TBD
+ * @lock: used for locking all fields that require locking, see comments
+ *	in the header file.
+ * @flaglock: spinlock for flags accesses
+ * @ht_info: HT capabilities of this STA
+ * @supp_rates: Bitmap of supported rates (per band)
  * @addr: MAC address of this STA
  * @aid: STA's unique AID (1..2007, 0 = not assigned yet),
  *	only used in AP (and IBSS?) mode
+ * @listen_interval: TBD
+ * @pin_status: TBD
  * @flags: STA flags, see &enum ieee80211_sta_info_flags
- * @flaglock: spinlock for flags accesses
  * @ps_tx_buf: buffer of frames to transmit to this station
  *	when it leaves power saving state
  * @tx_filtered: buffer of frames we already tried to transmit
@@ -172,10 +182,41 @@ struct sta_ampdu_mlme {
  *	power saving state
  * @rx_packets: Number of MSDUs received from this STA
  * @rx_bytes: Number of bytes received from this STA
- * @supp_rates: Bitmap of supported rates (per band)
- * @ht_info: HT capabilities of this STA
- * @lock: used for locking all fields that require locking, see comments
- *	in the header file.
+ * @wep_weak_iv_count: TBD
+ * @last_rx: TBD
+ * @num_duplicates: number of duplicate frames received from this STA
+ * @rx_fragments: number of received MPDUs
+ * @rx_dropped: number of dropped MPDUs from this STA
+ * @last_signal: signal of last received frame from this STA
+ * @last_qual: qual of last received frame from this STA
+ * @last_noise: noise of last received frame from this STA
+ * @last_seq_ctrl: last received seq/frag number from this STA (per RX queue)
+ * @wme_rx_queue: TBD
+ * @tx_filtered_count: TBD
+ * @tx_retry_failed: TBD
+ * @tx_retry_count: TBD
+ * @tx_num_consecutive_failures: TBD
+ * @tx_num_mpdu_ok: TBD
+ * @tx_num_mpdu_fail: TBD
+ * @fail_avg: moving percentage of failed MSDUs
+ * @tx_packets: number of RX/TX MSDUs
+ * @tx_bytes: TBD
+ * @tx_fragments: number of transmitted MPDUs
+ * @txrate_idx: TBD
+ * @last_txrate_idx: TBD
+ * @wme_tx_queue: TBD
+ * @ampdu_mlme: TBD
+ * @timer_to_tid: identity mapping to ID timers
+ * @tid_to_tx_q: map tid to tx queue
+ * @llid: Local link ID
+ * @plid: Peer link ID
+ * @reason: Cancel reason on PLINK_HOLDING state
+ * @plink_retries: Retries in establishment
+ * @ignore_plink_timer: TBD
+ * @plink_state plink_state: TBD
+ * @plink_timeout: TBD
+ * @plink_timer: TBD
+ * @debugfs: debug filesystem info
  */
 struct sta_info {
 	/* General information, mostly static */
@@ -217,14 +258,12 @@ struct sta_info {
 	unsigned long rx_packets, rx_bytes;
 	unsigned long wep_weak_iv_count;
 	unsigned long last_rx;
-	unsigned long num_duplicates; /* number of duplicate frames received
-				       * from this STA */
-	unsigned long rx_fragments; /* number of received MPDUs */
-	unsigned long rx_dropped; /* number of dropped MPDUs from this STA */
-	int last_signal; /* signal of last received frame from this STA */
-	int last_qual;  /* qual of last received frame from this STA */
-	int last_noise; /* noise of last received frame from this STA */
-	/* last received seq/frag number from this STA (per RX queue) */
+	unsigned long num_duplicates;
+	unsigned long rx_fragments;
+	unsigned long rx_dropped;
+	int last_signal;
+	int last_qual;
+	int last_noise;
 	__le16 last_seq_ctrl[NUM_RX_DATA_QUEUES];
 #ifdef CONFIG_MAC80211_DEBUG_COUNTERS
 	unsigned int wme_rx_queue[NUM_RX_DATA_QUEUES];
@@ -241,9 +280,9 @@ struct sta_info {
 	unsigned int fail_avg;
 
 	/* Updated from TX path only, no locking requirements */
-	unsigned long tx_packets; /* number of RX/TX MSDUs */
+	unsigned long tx_packets;
 	unsigned long tx_bytes;
-	unsigned long tx_fragments; /* number of transmitted MPDUs */
+	unsigned long tx_fragments;
 	int txrate_idx;
 	int last_txrate_idx;
 #ifdef CONFIG_MAC80211_DEBUG_COUNTERS
@@ -254,18 +293,18 @@ struct sta_info {
 	 * Aggregation information, locked with lock.
 	 */
 	struct sta_ampdu_mlme ampdu_mlme;
-	u8 timer_to_tid[STA_TID_NUM];	/* identity mapping to ID timers */
-	u8 tid_to_tx_q[STA_TID_NUM];	/* map tid to tx queue */
+	u8 timer_to_tid[STA_TID_NUM];
+	u8 tid_to_tx_q[STA_TID_NUM];
 
 #ifdef CONFIG_MAC80211_MESH
 	/*
 	 * Mesh peer link attributes
 	 * TODO: move to a sub-structure that is referenced with pointer?
 	 */
-	__le16 llid;		/* Local link ID */
-	__le16 plid;		/* Peer link ID */
-	__le16 reason;		/* Cancel reason on PLINK_HOLDING state */
-	u8 plink_retries;	/* Retries in establishment */
+	__le16 llid;
+	__le16 plid;
+	__le16 reason;
+	u8 plink_retries;
 	bool ignore_plink_timer;
 	enum plink_state plink_state;
 	u32 plink_timeout;
-- 
cgit v1.2.3


From 11a100f844f6096787ab20e19f17d72abc957a8f Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Tue, 8 Jul 2008 15:36:57 -0700
Subject: vlan: avoid header copying and linearisation where possible

- vlan_dev_reorder_header() is only called on the receive path after
  calling skb_share_check(). This means we can use skb_cow() since
  all we need is a writable header.

- vlan_dev_hard_header() includes a work-around for some apparently
  broken out of tree MPLS code. The hard_header functions can expect
  to always have a headroom of at least there own hard_header_len
  available, so the reallocation check is unnecessary.

- __vlan_put_tag() can use skb_cow_head() to avoid the skb_unshare()
  copy when the header is writable.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/if_vlan.h | 18 +++---------------
 net/8021q/vlan_dev.c    | 34 +++++-----------------------------
 2 files changed, 8 insertions(+), 44 deletions(-)

(limited to 'include')

diff --git a/include/linux/if_vlan.h b/include/linux/if_vlan.h
index d36515dae62f..93f5d9b0e9f9 100644
--- a/include/linux/if_vlan.h
+++ b/include/linux/if_vlan.h
@@ -185,22 +185,10 @@ static inline struct sk_buff *__vlan_put_tag(struct sk_buff *skb, u16 vlan_tci)
 {
 	struct vlan_ethhdr *veth;
 
-	if (skb_headroom(skb) < VLAN_HLEN) {
-		struct sk_buff *sk_tmp = skb;
-		skb = skb_realloc_headroom(sk_tmp, VLAN_HLEN);
-		kfree_skb(sk_tmp);
-		if (!skb) {
-			printk(KERN_ERR "vlan: failed to realloc headroom\n");
-			return NULL;
-		}
-	} else {
-		skb = skb_unshare(skb, GFP_ATOMIC);
-		if (!skb) {
-			printk(KERN_ERR "vlan: failed to unshare skbuff\n");
-			return NULL;
-		}
+	if (skb_cow_head(skb, VLAN_HLEN) < 0) {
+		kfree_skb(skb);
+		return NULL;
 	}
-
 	veth = (struct vlan_ethhdr *)skb_push(skb, VLAN_HLEN);
 
 	/* Move the mac addresses to the beginning of the new header. */
diff --git a/net/8021q/vlan_dev.c b/net/8021q/vlan_dev.c
index 2ccac6bea57e..b6e52c025fd8 100644
--- a/net/8021q/vlan_dev.c
+++ b/net/8021q/vlan_dev.c
@@ -74,11 +74,8 @@ static int vlan_dev_rebuild_header(struct sk_buff *skb)
 static inline struct sk_buff *vlan_check_reorder_header(struct sk_buff *skb)
 {
 	if (vlan_dev_info(skb->dev)->flags & VLAN_FLAG_REORDER_HDR) {
-		if (skb_shared(skb) || skb_cloned(skb)) {
-			struct sk_buff *nskb = skb_copy(skb, GFP_ATOMIC);
-			kfree_skb(skb);
-			skb = nskb;
-		}
+		if (skb_cow(skb, skb_headroom(skb)) < 0)
+			skb = NULL;
 		if (skb) {
 			/* Lifted from Gleb's VLAN code... */
 			memmove(skb->data - ETH_HLEN,
@@ -262,12 +259,14 @@ static int vlan_dev_hard_header(struct sk_buff *skb, struct net_device *dev,
 	u16 vlan_tci = 0;
 	int rc = 0;
 	int build_vlan_header = 0;
-	struct net_device *vdev = dev;
 
 	pr_debug("%s: skb: %p type: %hx len: %u vlan_id: %hx, daddr: %p\n",
 		 __func__, skb, type, len, vlan_dev_info(dev)->vlan_id,
 		 daddr);
 
+	if (WARN_ON(skb_headroom(skb) < dev->hard_header_len))
+		return -ENOSPC;
+
 	/* build vlan header only if re_order_header flag is NOT set.  This
 	 * fixes some programs that get confused when they see a VLAN device
 	 * sending a frame that is VLAN encoded (the consensus is that the VLAN
@@ -316,29 +315,6 @@ static int vlan_dev_hard_header(struct sk_buff *skb, struct net_device *dev,
 
 	dev = vlan_dev_info(dev)->real_dev;
 
-	/* MPLS can send us skbuffs w/out enough space.	This check will grow
-	 * the skb if it doesn't have enough headroom. Not a beautiful solution,
-	 * so I'll tick a counter so that users can know it's happening...
-	 * If they care...
-	 */
-
-	/* NOTE: This may still break if the underlying device is not the final
-	 * device (and thus there are more headers to add...) It should work for
-	 * good-ole-ethernet though.
-	 */
-	if (skb_headroom(skb) < dev->hard_header_len) {
-		struct sk_buff *sk_tmp = skb;
-		skb = skb_realloc_headroom(sk_tmp, dev->hard_header_len);
-		kfree_skb(sk_tmp);
-		if (skb == NULL) {
-			struct net_device_stats *stats = &vdev->stats;
-			stats->tx_dropped++;
-			return -ENOMEM;
-		}
-		vlan_dev_info(vdev)->cnt_inc_headroom_on_tx++;
-		pr_debug("%s: %s: had to grow skb\n", __func__, vdev->name);
-	}
-
 	if (build_vlan_header) {
 		/* Now make the underlying real hard header */
 		rc = dev_hard_header(skb, dev, ETH_P_8021Q, daddr, saddr,
-- 
cgit v1.2.3


From bb949fbd1878973c3539d9aecff52f284482a937 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Tue, 8 Jul 2008 16:55:56 -0700
Subject: netdev: Create netdev_queue abstraction.

A netdev_queue is an entity managed by a qdisc.

Currently there is one RX and one TX queue, and a netdev_queue merely
contains a backpointer to the net_device.

The Qdisc struct is augmented with a netdev_queue pointer as well.

Eventually the 'dev' Qdisc member will go away and we will have the
resulting hierarchy:

	net_device --> netdev_queue --> Qdisc

Also, qdisc_alloc() and qdisc_create_dflt() now take a netdev_queue
pointer argument.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h |  7 +++++++
 include/net/sch_generic.h |  6 +++++-
 net/core/dev.c            |  8 ++++++++
 net/mac80211/wme.c        |  6 ++++--
 net/sched/sch_api.c       | 12 +++++++-----
 net/sched/sch_atm.c       |  6 ++++--
 net/sched/sch_cbq.c       |  9 ++++++---
 net/sched/sch_dsmark.c    |  6 ++++--
 net/sched/sch_fifo.c      |  3 ++-
 net/sched/sch_generic.c   | 14 ++++++++++----
 net/sched/sch_hfsc.c      |  9 ++++++---
 net/sched/sch_htb.c       | 11 +++++++----
 net/sched/sch_netem.c     |  3 ++-
 net/sched/sch_prio.c      |  3 ++-
 14 files changed, 74 insertions(+), 29 deletions(-)

(limited to 'include')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index e009c6fbf5cd..515fd25bf0fc 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -448,6 +448,10 @@ static inline void napi_synchronize(const struct napi_struct *n)
 # define napi_synchronize(n)	barrier()
 #endif
 
+struct netdev_queue {
+	struct net_device	*dev;
+};
+
 /*
  *	The DEVICE structure.
  *	Actually, this whole structure is a big mistake.  It mixes I/O
@@ -624,6 +628,9 @@ struct net_device
 
 	unsigned char		broadcast[MAX_ADDR_LEN];	/* hw bcast add	*/
 
+	struct netdev_queue	rx_queue;
+	struct netdev_queue	tx_queue;
+
 	/* ingress path synchronizer */
 	spinlock_t		ingress_lock;
 	struct Qdisc		*qdisc_ingress;
diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h
index 073f2580b83b..0ab53c575f87 100644
--- a/include/net/sch_generic.h
+++ b/include/net/sch_generic.h
@@ -37,6 +37,7 @@ struct Qdisc
 	u32			parent;
 	atomic_t		refcnt;
 	struct sk_buff_head	q;
+	struct netdev_queue	*dev_queue;
 	struct net_device	*dev;
 	struct list_head	list;
 
@@ -216,8 +217,11 @@ extern void dev_deactivate(struct net_device *dev);
 extern void qdisc_reset(struct Qdisc *qdisc);
 extern void qdisc_destroy(struct Qdisc *qdisc);
 extern void qdisc_tree_decrease_qlen(struct Qdisc *qdisc, unsigned int n);
-extern struct Qdisc *qdisc_alloc(struct net_device *dev, struct Qdisc_ops *ops);
+extern struct Qdisc *qdisc_alloc(struct net_device *dev,
+				 struct netdev_queue *dev_queue,
+				 struct Qdisc_ops *ops);
 extern struct Qdisc *qdisc_create_dflt(struct net_device *dev,
+				       struct netdev_queue *dev_queue,
 				       struct Qdisc_ops *ops, u32 parentid);
 extern void tcf_destroy(struct tcf_proto *tp);
 extern void tcf_destroy_chain(struct tcf_proto **fl);
diff --git a/net/core/dev.c b/net/core/dev.c
index 75933932463d..9b281c906eb0 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -4072,6 +4072,12 @@ static struct net_device_stats *internal_stats(struct net_device *dev)
 	return &dev->stats;
 }
 
+static void netdev_init_queues(struct net_device *dev)
+{
+	dev->rx_queue.dev = dev;
+	dev->tx_queue.dev = dev;
+}
+
 /**
  *	alloc_netdev_mq - allocate network device
  *	@sizeof_priv:	size of private data to allocate space for
@@ -4124,6 +4130,8 @@ struct net_device *alloc_netdev_mq(int sizeof_priv, const char *name,
 	dev->egress_subqueue_count = queue_count;
 	dev->gso_max_size = GSO_MAX_SIZE;
 
+	netdev_init_queues(dev);
+
 	dev->get_stats = internal_stats;
 	netpoll_netdev_init(dev);
 	setup(dev);
diff --git a/net/mac80211/wme.c b/net/mac80211/wme.c
index 5c666f7eda8f..770f1c09b793 100644
--- a/net/mac80211/wme.c
+++ b/net/mac80211/wme.c
@@ -359,7 +359,8 @@ static int wme_qdiscop_init(struct Qdisc *qd, struct nlattr *opt)
 	/* create child queues */
 	for (i = 0; i < QD_NUM(hw); i++) {
 		skb_queue_head_init(&q->requeued[i]);
-		q->queues[i] = qdisc_create_dflt(qd->dev, &pfifo_qdisc_ops,
+		q->queues[i] = qdisc_create_dflt(qd->dev, qd->dev_queue,
+						 &pfifo_qdisc_ops,
 						 qd->handle);
 		if (!q->queues[i]) {
 			q->queues[i] = &noop_qdisc;
@@ -575,7 +576,8 @@ void ieee80211_install_qdisc(struct net_device *dev)
 {
 	struct Qdisc *qdisc;
 
-	qdisc = qdisc_create_dflt(dev, &wme_qdisc_ops, TC_H_ROOT);
+	qdisc = qdisc_create_dflt(dev, &dev->tx_queue,
+				  &wme_qdisc_ops, TC_H_ROOT);
 	if (!qdisc) {
 		printk(KERN_ERR "%s: qdisc installation failed\n", dev->name);
 		return;
diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c
index 69e918bb4278..b86c98bd06a3 100644
--- a/net/sched/sch_api.c
+++ b/net/sched/sch_api.c
@@ -552,8 +552,8 @@ static int qdisc_graft(struct net_device *dev, struct Qdisc *parent,
  */
 
 static struct Qdisc *
-qdisc_create(struct net_device *dev, u32 parent, u32 handle,
-	   struct nlattr **tca, int *errp)
+qdisc_create(struct net_device *dev, struct netdev_queue *dev_queue,
+	     u32 parent, u32 handle, struct nlattr **tca, int *errp)
 {
 	int err;
 	struct nlattr *kind = tca[TCA_KIND];
@@ -593,7 +593,7 @@ qdisc_create(struct net_device *dev, u32 parent, u32 handle,
 	if (ops == NULL)
 		goto err_out;
 
-	sch = qdisc_alloc(dev, ops);
+	sch = qdisc_alloc(dev, dev_queue, ops);
 	if (IS_ERR(sch)) {
 		err = PTR_ERR(sch);
 		goto err_out2;
@@ -892,10 +892,12 @@ create_n_graft:
 	if (!(n->nlmsg_flags&NLM_F_CREATE))
 		return -ENOENT;
 	if (clid == TC_H_INGRESS)
-		q = qdisc_create(dev, tcm->tcm_parent, tcm->tcm_parent,
+		q = qdisc_create(dev, &dev->rx_queue,
+				 tcm->tcm_parent, tcm->tcm_parent,
 				 tca, &err);
 	else
-		q = qdisc_create(dev, tcm->tcm_parent, tcm->tcm_handle,
+		q = qdisc_create(dev, &dev->tx_queue,
+				 tcm->tcm_parent, tcm->tcm_handle,
 				 tca, &err);
 	if (q == NULL) {
 		if (err == -EAGAIN)
diff --git a/net/sched/sch_atm.c b/net/sched/sch_atm.c
index db0e23ae85f8..3dddab531d5a 100644
--- a/net/sched/sch_atm.c
+++ b/net/sched/sch_atm.c
@@ -296,7 +296,8 @@ static int atm_tc_change(struct Qdisc *sch, u32 classid, u32 parent,
 		goto err_out;
 	}
 	flow->filter_list = NULL;
-	flow->q = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops, classid);
+	flow->q = qdisc_create_dflt(sch->dev, sch->dev_queue,
+				    &pfifo_qdisc_ops, classid);
 	if (!flow->q)
 		flow->q = &noop_qdisc;
 	pr_debug("atm_tc_change: qdisc %p\n", flow->q);
@@ -555,7 +556,8 @@ static int atm_tc_init(struct Qdisc *sch, struct nlattr *opt)
 
 	pr_debug("atm_tc_init(sch %p,[qdisc %p],opt %p)\n", sch, p, opt);
 	p->flows = &p->link;
-	p->link.q = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops, sch->handle);
+	p->link.q = qdisc_create_dflt(sch->dev, sch->dev_queue,
+				      &pfifo_qdisc_ops, sch->handle);
 	if (!p->link.q)
 		p->link.q = &noop_qdisc;
 	pr_debug("atm_tc_init: link (%p) qdisc %p\n", &p->link, p->link.q);
diff --git a/net/sched/sch_cbq.c b/net/sched/sch_cbq.c
index 968b4c73c9c1..d360dcd0818b 100644
--- a/net/sched/sch_cbq.c
+++ b/net/sched/sch_cbq.c
@@ -1401,7 +1401,8 @@ static int cbq_init(struct Qdisc *sch, struct nlattr *opt)
 	q->link.sibling = &q->link;
 	q->link.common.classid = sch->handle;
 	q->link.qdisc = sch;
-	if (!(q->link.q = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops,
+	if (!(q->link.q = qdisc_create_dflt(sch->dev, sch->dev_queue,
+					    &pfifo_qdisc_ops,
 					    sch->handle)))
 		q->link.q = &noop_qdisc;
 
@@ -1645,7 +1646,8 @@ static int cbq_graft(struct Qdisc *sch, unsigned long arg, struct Qdisc *new,
 
 	if (cl) {
 		if (new == NULL) {
-			new = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops,
+			new = qdisc_create_dflt(sch->dev, sch->dev_queue,
+						&pfifo_qdisc_ops,
 						cl->common.classid);
 			if (new == NULL)
 				return -ENOBUFS;
@@ -1877,7 +1879,8 @@ cbq_change_class(struct Qdisc *sch, u32 classid, u32 parentid, struct nlattr **t
 	cl->R_tab = rtab;
 	rtab = NULL;
 	cl->refcnt = 1;
-	if (!(cl->q = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops, classid)))
+	if (!(cl->q = qdisc_create_dflt(sch->dev, sch->dev_queue,
+					&pfifo_qdisc_ops, classid)))
 		cl->q = &noop_qdisc;
 	cl->common.classid = classid;
 	cl->tparent = parent;
diff --git a/net/sched/sch_dsmark.c b/net/sched/sch_dsmark.c
index c4c1317cd47d..c955ba24e5cf 100644
--- a/net/sched/sch_dsmark.c
+++ b/net/sched/sch_dsmark.c
@@ -60,7 +60,8 @@ static int dsmark_graft(struct Qdisc *sch, unsigned long arg,
 		sch, p, new, old);
 
 	if (new == NULL) {
-		new = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops,
+		new = qdisc_create_dflt(sch->dev, sch->dev_queue,
+					&pfifo_qdisc_ops,
 					sch->handle);
 		if (new == NULL)
 			new = &noop_qdisc;
@@ -390,7 +391,8 @@ static int dsmark_init(struct Qdisc *sch, struct nlattr *opt)
 	p->default_index = default_index;
 	p->set_tc_index = nla_get_flag(tb[TCA_DSMARK_SET_TC_INDEX]);
 
-	p->q = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops, sch->handle);
+	p->q = qdisc_create_dflt(sch->dev, sch->dev_queue,
+				 &pfifo_qdisc_ops, sch->handle);
 	if (p->q == NULL)
 		p->q = &noop_qdisc;
 
diff --git a/net/sched/sch_fifo.c b/net/sched/sch_fifo.c
index 82d7d7bbbb16..779eae85faf0 100644
--- a/net/sched/sch_fifo.c
+++ b/net/sched/sch_fifo.c
@@ -137,7 +137,8 @@ struct Qdisc *fifo_create_dflt(struct Qdisc *sch, struct Qdisc_ops *ops,
 	struct Qdisc *q;
 	int err = -ENOMEM;
 
-	q = qdisc_create_dflt(sch->dev, ops, TC_H_MAKE(sch->handle, 1));
+	q = qdisc_create_dflt(sch->dev, sch->dev_queue,
+			      ops, TC_H_MAKE(sch->handle, 1));
 	if (q) {
 		err = fifo_set_limit(q, limit);
 		if (err < 0) {
diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c
index 13afa7214392..d97086480893 100644
--- a/net/sched/sch_generic.c
+++ b/net/sched/sch_generic.c
@@ -440,7 +440,9 @@ static struct Qdisc_ops pfifo_fast_ops __read_mostly = {
 	.owner		=	THIS_MODULE,
 };
 
-struct Qdisc *qdisc_alloc(struct net_device *dev, struct Qdisc_ops *ops)
+struct Qdisc *qdisc_alloc(struct net_device *dev,
+			  struct netdev_queue *dev_queue,
+			  struct Qdisc_ops *ops)
 {
 	void *p;
 	struct Qdisc *sch;
@@ -462,6 +464,7 @@ struct Qdisc *qdisc_alloc(struct net_device *dev, struct Qdisc_ops *ops)
 	sch->ops = ops;
 	sch->enqueue = ops->enqueue;
 	sch->dequeue = ops->dequeue;
+	sch->dev_queue = dev_queue;
 	sch->dev = dev;
 	dev_hold(dev);
 	atomic_set(&sch->refcnt, 1);
@@ -471,12 +474,14 @@ errout:
 	return ERR_PTR(err);
 }
 
-struct Qdisc * qdisc_create_dflt(struct net_device *dev, struct Qdisc_ops *ops,
+struct Qdisc * qdisc_create_dflt(struct net_device *dev,
+				 struct netdev_queue *dev_queue,
+				 struct Qdisc_ops *ops,
 				 unsigned int parentid)
 {
 	struct Qdisc *sch;
 
-	sch = qdisc_alloc(dev, ops);
+	sch = qdisc_alloc(dev, dev_queue, ops);
 	if (IS_ERR(sch))
 		goto errout;
 	sch->stats_lock = &dev->queue_lock;
@@ -545,7 +550,8 @@ void dev_activate(struct net_device *dev)
 	if (dev->qdisc_sleeping == &noop_qdisc) {
 		struct Qdisc *qdisc;
 		if (dev->tx_queue_len) {
-			qdisc = qdisc_create_dflt(dev, &pfifo_fast_ops,
+			qdisc = qdisc_create_dflt(dev, &dev->tx_queue,
+						  &pfifo_fast_ops,
 						  TC_H_ROOT);
 			if (qdisc == NULL) {
 				printk(KERN_INFO "%s: activation failed\n", dev->name);
diff --git a/net/sched/sch_hfsc.c b/net/sched/sch_hfsc.c
index 3a8267246a4f..5a22fec4eadd 100644
--- a/net/sched/sch_hfsc.c
+++ b/net/sched/sch_hfsc.c
@@ -1083,7 +1083,8 @@ hfsc_change_class(struct Qdisc *sch, u32 classid, u32 parentid,
 	cl->refcnt    = 1;
 	cl->sched     = q;
 	cl->cl_parent = parent;
-	cl->qdisc = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops, classid);
+	cl->qdisc = qdisc_create_dflt(sch->dev, sch->dev_queue,
+				      &pfifo_qdisc_ops, classid);
 	if (cl->qdisc == NULL)
 		cl->qdisc = &noop_qdisc;
 	INIT_LIST_HEAD(&cl->children);
@@ -1201,7 +1202,8 @@ hfsc_graft_class(struct Qdisc *sch, unsigned long arg, struct Qdisc *new,
 	if (cl->level > 0)
 		return -EINVAL;
 	if (new == NULL) {
-		new = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops,
+		new = qdisc_create_dflt(sch->dev, sch->dev_queue,
+					&pfifo_qdisc_ops,
 					cl->cl_common.classid);
 		if (new == NULL)
 			new = &noop_qdisc;
@@ -1443,7 +1445,8 @@ hfsc_init_qdisc(struct Qdisc *sch, struct nlattr *opt)
 	q->root.cl_common.classid = sch->handle;
 	q->root.refcnt  = 1;
 	q->root.sched   = q;
-	q->root.qdisc = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops,
+	q->root.qdisc = qdisc_create_dflt(sch->dev, sch->dev_queue,
+					  &pfifo_qdisc_ops,
 					  sch->handle);
 	if (q->root.qdisc == NULL)
 		q->root.qdisc = &noop_qdisc;
diff --git a/net/sched/sch_htb.c b/net/sched/sch_htb.c
index ee8b4ffe110c..956a67f66b9c 100644
--- a/net/sched/sch_htb.c
+++ b/net/sched/sch_htb.c
@@ -1129,7 +1129,8 @@ static int htb_graft(struct Qdisc *sch, unsigned long arg, struct Qdisc *new,
 
 	if (cl && !cl->level) {
 		if (new == NULL &&
-		    (new = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops,
+		    (new = qdisc_create_dflt(sch->dev, sch->dev_queue,
+					     &pfifo_qdisc_ops,
 					     cl->common.classid))
 		    == NULL)
 			return -ENOBUFS;
@@ -1256,8 +1257,9 @@ static int htb_delete(struct Qdisc *sch, unsigned long arg)
 		return -EBUSY;
 
 	if (!cl->level && htb_parent_last_child(cl)) {
-		new_q = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops,
-						cl->parent->common.classid);
+		new_q = qdisc_create_dflt(sch->dev, sch->dev_queue,
+					  &pfifo_qdisc_ops,
+					  cl->parent->common.classid);
 		last_child = 1;
 	}
 
@@ -1376,7 +1378,8 @@ static int htb_change_class(struct Qdisc *sch, u32 classid,
 		/* create leaf qdisc early because it uses kmalloc(GFP_KERNEL)
 		   so that can't be used inside of sch_tree_lock
 		   -- thanks to Karlis Peisenieks */
-		new_q = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops, classid);
+		new_q = qdisc_create_dflt(sch->dev, sch->dev_queue,
+					  &pfifo_qdisc_ops, classid);
 		sch_tree_lock(sch);
 		if (parent && !parent->level) {
 			unsigned int qlen = parent->un.leaf.q->q.qlen;
diff --git a/net/sched/sch_netem.c b/net/sched/sch_netem.c
index 24697667247c..aa7a04e32ae9 100644
--- a/net/sched/sch_netem.c
+++ b/net/sched/sch_netem.c
@@ -536,7 +536,8 @@ static int netem_init(struct Qdisc *sch, struct nlattr *opt)
 
 	qdisc_watchdog_init(&q->watchdog, sch);
 
-	q->qdisc = qdisc_create_dflt(sch->dev, &tfifo_qdisc_ops,
+	q->qdisc = qdisc_create_dflt(sch->dev, sch->dev_queue,
+				     &tfifo_qdisc_ops,
 				     TC_H_MAKE(sch->handle, 1));
 	if (!q->qdisc) {
 		pr_debug("netem: qdisc create failed\n");
diff --git a/net/sched/sch_prio.c b/net/sched/sch_prio.c
index 5532f1031ab5..ca58a039208e 100644
--- a/net/sched/sch_prio.c
+++ b/net/sched/sch_prio.c
@@ -281,7 +281,8 @@ static int prio_tune(struct Qdisc *sch, struct nlattr *opt)
 	for (i=0; i<q->bands; i++) {
 		if (q->queues[i] == &noop_qdisc) {
 			struct Qdisc *child;
-			child = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops,
+			child = qdisc_create_dflt(sch->dev, sch->dev_queue,
+						  &pfifo_qdisc_ops,
 						  TC_H_MAKE(sch->handle, i + 1));
 			if (child) {
 				sch_tree_lock(sch);
-- 
cgit v1.2.3


From 5ce2d488fe039ddd86a638496cf704df86c74eeb Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Tue, 8 Jul 2008 17:06:30 -0700
Subject: pkt_sched: Remove 'dev' member of struct Qdisc.

It can be obtained via the netdev_queue.  So create a helper routine,
qdisc_dev(), to make the transformations nicer looking.

Now, qdisc_alloc() now no longer needs a net_device pointer argument.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/sch_generic.h | 16 +++++++++-------
 net/mac80211/wme.c        | 28 ++++++++++++++--------------
 net/sched/cls_api.c       |  2 +-
 net/sched/cls_route.c     |  4 ++--
 net/sched/sch_api.c       | 10 +++++-----
 net/sched/sch_atm.c       |  4 ++--
 net/sched/sch_cbq.c       | 22 +++++++++++-----------
 net/sched/sch_dsmark.c    |  4 ++--
 net/sched/sch_fifo.c      |  6 +++---
 net/sched/sch_generic.c   | 12 +++++-------
 net/sched/sch_gred.c      |  2 +-
 net/sched/sch_hfsc.c      | 10 +++++-----
 net/sched/sch_htb.c       | 24 ++++++++++++------------
 net/sched/sch_netem.c     | 10 +++++-----
 net/sched/sch_prio.c      | 15 ++++++++-------
 net/sched/sch_sfq.c       |  4 ++--
 net/sched/sch_teql.c      | 12 ++++++------
 17 files changed, 93 insertions(+), 92 deletions(-)

(limited to 'include')

diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h
index 0ab53c575f87..66ec36d8ac97 100644
--- a/include/net/sch_generic.h
+++ b/include/net/sch_generic.h
@@ -38,7 +38,6 @@ struct Qdisc
 	atomic_t		refcnt;
 	struct sk_buff_head	q;
 	struct netdev_queue	*dev_queue;
-	struct net_device	*dev;
 	struct list_head	list;
 
 	struct gnet_stats_basic	bstats;
@@ -156,14 +155,18 @@ struct tcf_proto
 	struct tcf_proto_ops	*ops;
 };
 
+static inline struct net_device *qdisc_dev(struct Qdisc *qdisc)
+{
+	return qdisc->dev_queue->dev;
+}
 
 extern void qdisc_lock_tree(struct net_device *dev);
 extern void qdisc_unlock_tree(struct net_device *dev);
 
-#define sch_tree_lock(q)	qdisc_lock_tree((q)->dev)
-#define sch_tree_unlock(q)	qdisc_unlock_tree((q)->dev)
-#define tcf_tree_lock(tp)	qdisc_lock_tree((tp)->q->dev)
-#define tcf_tree_unlock(tp)	qdisc_unlock_tree((tp)->q->dev)
+#define sch_tree_lock(q)	qdisc_lock_tree(qdisc_dev(q))
+#define sch_tree_unlock(q)	qdisc_unlock_tree(qdisc_dev(q))
+#define tcf_tree_lock(tp)	qdisc_lock_tree(qdisc_dev((tp)->q))
+#define tcf_tree_unlock(tp)	qdisc_unlock_tree(qdisc_dev((tp)->q))
 
 extern struct Qdisc noop_qdisc;
 extern struct Qdisc_ops noop_qdisc_ops;
@@ -217,8 +220,7 @@ extern void dev_deactivate(struct net_device *dev);
 extern void qdisc_reset(struct Qdisc *qdisc);
 extern void qdisc_destroy(struct Qdisc *qdisc);
 extern void qdisc_tree_decrease_qlen(struct Qdisc *qdisc, unsigned int n);
-extern struct Qdisc *qdisc_alloc(struct net_device *dev,
-				 struct netdev_queue *dev_queue,
+extern struct Qdisc *qdisc_alloc(struct netdev_queue *dev_queue,
 				 struct Qdisc_ops *ops);
 extern struct Qdisc *qdisc_create_dflt(struct net_device *dev,
 				       struct netdev_queue *dev_queue,
diff --git a/net/mac80211/wme.c b/net/mac80211/wme.c
index 770f1c09b793..2fbc171130bf 100644
--- a/net/mac80211/wme.c
+++ b/net/mac80211/wme.c
@@ -103,7 +103,7 @@ static inline int wme_downgrade_ac(struct sk_buff *skb)
  * negative return value indicates to drop the frame */
 static int classify80211(struct sk_buff *skb, struct Qdisc *qd)
 {
-	struct ieee80211_local *local = wdev_priv(qd->dev->ieee80211_ptr);
+	struct ieee80211_local *local = wdev_priv(qdisc_dev(qd)->ieee80211_ptr);
 	struct ieee80211_hdr *hdr = (struct ieee80211_hdr *) skb->data;
 
 	if (!ieee80211_is_data(hdr->frame_control)) {
@@ -140,7 +140,7 @@ static int classify80211(struct sk_buff *skb, struct Qdisc *qd)
 
 static int wme_qdiscop_enqueue(struct sk_buff *skb, struct Qdisc* qd)
 {
-	struct ieee80211_local *local = wdev_priv(qd->dev->ieee80211_ptr);
+	struct ieee80211_local *local = wdev_priv(qdisc_dev(qd)->ieee80211_ptr);
 	struct ieee80211_hw *hw = &local->hw;
 	struct ieee80211_sched_data *q = qdisc_priv(qd);
 	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
@@ -249,7 +249,7 @@ static int wme_qdiscop_requeue(struct sk_buff *skb, struct Qdisc* qd)
 static struct sk_buff *wme_qdiscop_dequeue(struct Qdisc* qd)
 {
 	struct ieee80211_sched_data *q = qdisc_priv(qd);
-	struct net_device *dev = qd->dev;
+	struct net_device *dev = qdisc_dev(qd);
 	struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr);
 	struct ieee80211_hw *hw = &local->hw;
 	struct sk_buff *skb;
@@ -286,7 +286,7 @@ static struct sk_buff *wme_qdiscop_dequeue(struct Qdisc* qd)
 static void wme_qdiscop_reset(struct Qdisc* qd)
 {
 	struct ieee80211_sched_data *q = qdisc_priv(qd);
-	struct ieee80211_local *local = wdev_priv(qd->dev->ieee80211_ptr);
+	struct ieee80211_local *local = wdev_priv(qdisc_dev(qd)->ieee80211_ptr);
 	struct ieee80211_hw *hw = &local->hw;
 	int queue;
 
@@ -303,7 +303,7 @@ static void wme_qdiscop_reset(struct Qdisc* qd)
 static void wme_qdiscop_destroy(struct Qdisc* qd)
 {
 	struct ieee80211_sched_data *q = qdisc_priv(qd);
-	struct ieee80211_local *local = wdev_priv(qd->dev->ieee80211_ptr);
+	struct ieee80211_local *local = wdev_priv(qdisc_dev(qd)->ieee80211_ptr);
 	struct ieee80211_hw *hw = &local->hw;
 	int queue;
 
@@ -328,7 +328,7 @@ static int wme_qdiscop_tune(struct Qdisc *qd, struct nlattr *opt)
 static int wme_qdiscop_init(struct Qdisc *qd, struct nlattr *opt)
 {
 	struct ieee80211_sched_data *q = qdisc_priv(qd);
-	struct net_device *dev = qd->dev;
+	struct net_device *dev = qdisc_dev(qd);
 	struct ieee80211_local *local;
 	struct ieee80211_hw *hw;
 	int err = 0, i;
@@ -359,7 +359,7 @@ static int wme_qdiscop_init(struct Qdisc *qd, struct nlattr *opt)
 	/* create child queues */
 	for (i = 0; i < QD_NUM(hw); i++) {
 		skb_queue_head_init(&q->requeued[i]);
-		q->queues[i] = qdisc_create_dflt(qd->dev, qd->dev_queue,
+		q->queues[i] = qdisc_create_dflt(qdisc_dev(qd), qd->dev_queue,
 						 &pfifo_qdisc_ops,
 						 qd->handle);
 		if (!q->queues[i]) {
@@ -386,7 +386,7 @@ static int wme_classop_graft(struct Qdisc *qd, unsigned long arg,
 			     struct Qdisc *new, struct Qdisc **old)
 {
 	struct ieee80211_sched_data *q = qdisc_priv(qd);
-	struct ieee80211_local *local = wdev_priv(qd->dev->ieee80211_ptr);
+	struct ieee80211_local *local = wdev_priv(qdisc_dev(qd)->ieee80211_ptr);
 	struct ieee80211_hw *hw = &local->hw;
 	unsigned long queue = arg - 1;
 
@@ -410,7 +410,7 @@ static struct Qdisc *
 wme_classop_leaf(struct Qdisc *qd, unsigned long arg)
 {
 	struct ieee80211_sched_data *q = qdisc_priv(qd);
-	struct ieee80211_local *local = wdev_priv(qd->dev->ieee80211_ptr);
+	struct ieee80211_local *local = wdev_priv(qdisc_dev(qd)->ieee80211_ptr);
 	struct ieee80211_hw *hw = &local->hw;
 	unsigned long queue = arg - 1;
 
@@ -423,7 +423,7 @@ wme_classop_leaf(struct Qdisc *qd, unsigned long arg)
 
 static unsigned long wme_classop_get(struct Qdisc *qd, u32 classid)
 {
-	struct ieee80211_local *local = wdev_priv(qd->dev->ieee80211_ptr);
+	struct ieee80211_local *local = wdev_priv(qdisc_dev(qd)->ieee80211_ptr);
 	struct ieee80211_hw *hw = &local->hw;
 	unsigned long queue = TC_H_MIN(classid);
 
@@ -450,7 +450,7 @@ static int wme_classop_change(struct Qdisc *qd, u32 handle, u32 parent,
 			      struct nlattr **tca, unsigned long *arg)
 {
 	unsigned long cl = *arg;
-	struct ieee80211_local *local = wdev_priv(qd->dev->ieee80211_ptr);
+	struct ieee80211_local *local = wdev_priv(qdisc_dev(qd)->ieee80211_ptr);
 	struct ieee80211_hw *hw = &local->hw;
 
 	if (cl - 1 > QD_NUM(hw))
@@ -467,7 +467,7 @@ static int wme_classop_change(struct Qdisc *qd, u32 handle, u32 parent,
  * when we add WMM-SA support - TSPECs may be deleted here */
 static int wme_classop_delete(struct Qdisc *qd, unsigned long cl)
 {
-	struct ieee80211_local *local = wdev_priv(qd->dev->ieee80211_ptr);
+	struct ieee80211_local *local = wdev_priv(qdisc_dev(qd)->ieee80211_ptr);
 	struct ieee80211_hw *hw = &local->hw;
 
 	if (cl - 1 > QD_NUM(hw))
@@ -480,7 +480,7 @@ static int wme_classop_dump_class(struct Qdisc *qd, unsigned long cl,
 				  struct sk_buff *skb, struct tcmsg *tcm)
 {
 	struct ieee80211_sched_data *q = qdisc_priv(qd);
-	struct ieee80211_local *local = wdev_priv(qd->dev->ieee80211_ptr);
+	struct ieee80211_local *local = wdev_priv(qdisc_dev(qd)->ieee80211_ptr);
 	struct ieee80211_hw *hw = &local->hw;
 
 	if (cl - 1 > QD_NUM(hw))
@@ -494,7 +494,7 @@ static int wme_classop_dump_class(struct Qdisc *qd, unsigned long cl,
 
 static void wme_classop_walk(struct Qdisc *qd, struct qdisc_walker *arg)
 {
-	struct ieee80211_local *local = wdev_priv(qd->dev->ieee80211_ptr);
+	struct ieee80211_local *local = wdev_priv(qdisc_dev(qd)->ieee80211_ptr);
 	struct ieee80211_hw *hw = &local->hw;
 	int queue;
 
diff --git a/net/sched/cls_api.c b/net/sched/cls_api.c
index 9360fc81e8c7..e2389f161e46 100644
--- a/net/sched/cls_api.c
+++ b/net/sched/cls_api.c
@@ -334,7 +334,7 @@ static int tcf_fill_node(struct sk_buff *skb, struct tcf_proto *tp,
 	tcm->tcm_family = AF_UNSPEC;
 	tcm->tcm__pad1 = 0;
 	tcm->tcm__pad1 = 0;
-	tcm->tcm_ifindex = tp->q->dev->ifindex;
+	tcm->tcm_ifindex = qdisc_dev(tp->q)->ifindex;
 	tcm->tcm_parent = tp->classid;
 	tcm->tcm_info = TC_H_MAKE(tp->prio, tp->protocol);
 	NLA_PUT_STRING(skb, TCA_KIND, tp->ops->kind);
diff --git a/net/sched/cls_route.c b/net/sched/cls_route.c
index 784dcb870b98..5a16ca28aa3d 100644
--- a/net/sched/cls_route.c
+++ b/net/sched/cls_route.c
@@ -302,7 +302,7 @@ static int route4_delete(struct tcf_proto *tp, unsigned long arg)
 			*fp = f->next;
 			tcf_tree_unlock(tp);
 
-			route4_reset_fastmap(tp->q->dev, head, f->id);
+			route4_reset_fastmap(qdisc_dev(tp->q), head, f->id);
 			route4_delete_filter(tp, f);
 
 			/* Strip tree */
@@ -500,7 +500,7 @@ reinsert:
 	}
 	tcf_tree_unlock(tp);
 
-	route4_reset_fastmap(tp->q->dev, head, f->id);
+	route4_reset_fastmap(qdisc_dev(tp->q), head, f->id);
 	*arg = (unsigned long)f;
 	return 0;
 
diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c
index b86c98bd06a3..1f893082a4f6 100644
--- a/net/sched/sch_api.c
+++ b/net/sched/sch_api.c
@@ -281,7 +281,7 @@ static enum hrtimer_restart qdisc_watchdog(struct hrtimer *timer)
 {
 	struct qdisc_watchdog *wd = container_of(timer, struct qdisc_watchdog,
 						 timer);
-	struct net_device *dev = wd->qdisc->dev;
+	struct net_device *dev = qdisc_dev(wd->qdisc);
 
 	wd->qdisc->flags &= ~TCQ_F_THROTTLED;
 	smp_wmb();
@@ -493,7 +493,7 @@ void qdisc_tree_decrease_qlen(struct Qdisc *sch, unsigned int n)
 		if (TC_H_MAJ(parentid) == TC_H_MAJ(TC_H_INGRESS))
 			return;
 
-		sch = qdisc_lookup(sch->dev, TC_H_MAJ(parentid));
+		sch = qdisc_lookup(qdisc_dev(sch), TC_H_MAJ(parentid));
 		if (sch == NULL) {
 			WARN_ON(parentid != TC_H_ROOT);
 			return;
@@ -593,7 +593,7 @@ qdisc_create(struct net_device *dev, struct netdev_queue *dev_queue,
 	if (ops == NULL)
 		goto err_out;
 
-	sch = qdisc_alloc(dev, dev_queue, ops);
+	sch = qdisc_alloc(dev_queue, ops);
 	if (IS_ERR(sch)) {
 		err = PTR_ERR(sch);
 		goto err_out2;
@@ -940,7 +940,7 @@ static int tc_fill_qdisc(struct sk_buff *skb, struct Qdisc *q, u32 clid,
 	tcm->tcm_family = AF_UNSPEC;
 	tcm->tcm__pad1 = 0;
 	tcm->tcm__pad2 = 0;
-	tcm->tcm_ifindex = q->dev->ifindex;
+	tcm->tcm_ifindex = qdisc_dev(q)->ifindex;
 	tcm->tcm_parent = clid;
 	tcm->tcm_handle = q->handle;
 	tcm->tcm_info = atomic_read(&q->refcnt);
@@ -1186,7 +1186,7 @@ static int tc_fill_tclass(struct sk_buff *skb, struct Qdisc *q,
 	nlh = NLMSG_NEW(skb, pid, seq, event, sizeof(*tcm), flags);
 	tcm = NLMSG_DATA(nlh);
 	tcm->tcm_family = AF_UNSPEC;
-	tcm->tcm_ifindex = q->dev->ifindex;
+	tcm->tcm_ifindex = qdisc_dev(q)->ifindex;
 	tcm->tcm_parent = q->handle;
 	tcm->tcm_handle = q->handle;
 	tcm->tcm_info = 0;
diff --git a/net/sched/sch_atm.c b/net/sched/sch_atm.c
index 3dddab531d5a..0de757e3be4a 100644
--- a/net/sched/sch_atm.c
+++ b/net/sched/sch_atm.c
@@ -296,7 +296,7 @@ static int atm_tc_change(struct Qdisc *sch, u32 classid, u32 parent,
 		goto err_out;
 	}
 	flow->filter_list = NULL;
-	flow->q = qdisc_create_dflt(sch->dev, sch->dev_queue,
+	flow->q = qdisc_create_dflt(qdisc_dev(sch), sch->dev_queue,
 				    &pfifo_qdisc_ops, classid);
 	if (!flow->q)
 		flow->q = &noop_qdisc;
@@ -556,7 +556,7 @@ static int atm_tc_init(struct Qdisc *sch, struct nlattr *opt)
 
 	pr_debug("atm_tc_init(sch %p,[qdisc %p],opt %p)\n", sch, p, opt);
 	p->flows = &p->link;
-	p->link.q = qdisc_create_dflt(sch->dev, sch->dev_queue,
+	p->link.q = qdisc_create_dflt(qdisc_dev(sch), sch->dev_queue,
 				      &pfifo_qdisc_ops, sch->handle);
 	if (!p->link.q)
 		p->link.q = &noop_qdisc;
diff --git a/net/sched/sch_cbq.c b/net/sched/sch_cbq.c
index d360dcd0818b..9f2ace585fd6 100644
--- a/net/sched/sch_cbq.c
+++ b/net/sched/sch_cbq.c
@@ -650,7 +650,7 @@ static enum hrtimer_restart cbq_undelay(struct hrtimer *timer)
 	}
 
 	sch->flags &= ~TCQ_F_THROTTLED;
-	netif_schedule(sch->dev);
+	netif_schedule(qdisc_dev(sch));
 	return HRTIMER_NORESTART;
 }
 
@@ -1077,9 +1077,9 @@ static void cbq_normalize_quanta(struct cbq_sched_data *q, int prio)
 				cl->quantum = (cl->weight*cl->allot*q->nclasses[prio])/
 					q->quanta[prio];
 			}
-			if (cl->quantum <= 0 || cl->quantum>32*cl->qdisc->dev->mtu) {
+			if (cl->quantum <= 0 || cl->quantum>32*qdisc_dev(cl->qdisc)->mtu) {
 				printk(KERN_WARNING "CBQ: class %08x has bad quantum==%ld, repaired.\n", cl->common.classid, cl->quantum);
-				cl->quantum = cl->qdisc->dev->mtu/2 + 1;
+				cl->quantum = qdisc_dev(cl->qdisc)->mtu/2 + 1;
 			}
 		}
 	}
@@ -1401,7 +1401,7 @@ static int cbq_init(struct Qdisc *sch, struct nlattr *opt)
 	q->link.sibling = &q->link;
 	q->link.common.classid = sch->handle;
 	q->link.qdisc = sch;
-	if (!(q->link.q = qdisc_create_dflt(sch->dev, sch->dev_queue,
+	if (!(q->link.q = qdisc_create_dflt(qdisc_dev(sch), sch->dev_queue,
 					    &pfifo_qdisc_ops,
 					    sch->handle)))
 		q->link.q = &noop_qdisc;
@@ -1411,7 +1411,7 @@ static int cbq_init(struct Qdisc *sch, struct nlattr *opt)
 	q->link.cpriority = TC_CBQ_MAXPRIO-1;
 	q->link.ovl_strategy = TC_CBQ_OVL_CLASSIC;
 	q->link.overlimit = cbq_ovl_classic;
-	q->link.allot = psched_mtu(sch->dev);
+	q->link.allot = psched_mtu(qdisc_dev(sch));
 	q->link.quantum = q->link.allot;
 	q->link.weight = q->link.R_tab->rate.rate;
 
@@ -1646,7 +1646,7 @@ static int cbq_graft(struct Qdisc *sch, unsigned long arg, struct Qdisc *new,
 
 	if (cl) {
 		if (new == NULL) {
-			new = qdisc_create_dflt(sch->dev, sch->dev_queue,
+			new = qdisc_create_dflt(qdisc_dev(sch), sch->dev_queue,
 						&pfifo_qdisc_ops,
 						cl->common.classid);
 			if (new == NULL)
@@ -1746,10 +1746,10 @@ static void cbq_put(struct Qdisc *sch, unsigned long arg)
 #ifdef CONFIG_NET_CLS_ACT
 		struct cbq_sched_data *q = qdisc_priv(sch);
 
-		spin_lock_bh(&sch->dev->queue_lock);
+		spin_lock_bh(&qdisc_dev(sch)->queue_lock);
 		if (q->rx_class == cl)
 			q->rx_class = NULL;
-		spin_unlock_bh(&sch->dev->queue_lock);
+		spin_unlock_bh(&qdisc_dev(sch)->queue_lock);
 #endif
 
 		cbq_destroy_class(sch, cl);
@@ -1828,7 +1828,7 @@ cbq_change_class(struct Qdisc *sch, u32 classid, u32 parentid, struct nlattr **t
 
 		if (tca[TCA_RATE])
 			gen_replace_estimator(&cl->bstats, &cl->rate_est,
-					      &sch->dev->queue_lock,
+					      &qdisc_dev(sch)->queue_lock,
 					      tca[TCA_RATE]);
 		return 0;
 	}
@@ -1879,7 +1879,7 @@ cbq_change_class(struct Qdisc *sch, u32 classid, u32 parentid, struct nlattr **t
 	cl->R_tab = rtab;
 	rtab = NULL;
 	cl->refcnt = 1;
-	if (!(cl->q = qdisc_create_dflt(sch->dev, sch->dev_queue,
+	if (!(cl->q = qdisc_create_dflt(qdisc_dev(sch), sch->dev_queue,
 					&pfifo_qdisc_ops, classid)))
 		cl->q = &noop_qdisc;
 	cl->common.classid = classid;
@@ -1919,7 +1919,7 @@ cbq_change_class(struct Qdisc *sch, u32 classid, u32 parentid, struct nlattr **t
 
 	if (tca[TCA_RATE])
 		gen_new_estimator(&cl->bstats, &cl->rate_est,
-				  &sch->dev->queue_lock, tca[TCA_RATE]);
+				  &qdisc_dev(sch)->queue_lock, tca[TCA_RATE]);
 
 	*arg = (unsigned long)cl;
 	return 0;
diff --git a/net/sched/sch_dsmark.c b/net/sched/sch_dsmark.c
index c955ba24e5cf..3aafbd17393a 100644
--- a/net/sched/sch_dsmark.c
+++ b/net/sched/sch_dsmark.c
@@ -60,7 +60,7 @@ static int dsmark_graft(struct Qdisc *sch, unsigned long arg,
 		sch, p, new, old);
 
 	if (new == NULL) {
-		new = qdisc_create_dflt(sch->dev, sch->dev_queue,
+		new = qdisc_create_dflt(qdisc_dev(sch), sch->dev_queue,
 					&pfifo_qdisc_ops,
 					sch->handle);
 		if (new == NULL)
@@ -391,7 +391,7 @@ static int dsmark_init(struct Qdisc *sch, struct nlattr *opt)
 	p->default_index = default_index;
 	p->set_tc_index = nla_get_flag(tb[TCA_DSMARK_SET_TC_INDEX]);
 
-	p->q = qdisc_create_dflt(sch->dev, sch->dev_queue,
+	p->q = qdisc_create_dflt(qdisc_dev(sch), sch->dev_queue,
 				 &pfifo_qdisc_ops, sch->handle);
 	if (p->q == NULL)
 		p->q = &noop_qdisc;
diff --git a/net/sched/sch_fifo.c b/net/sched/sch_fifo.c
index 779eae85faf0..1d97fa42c902 100644
--- a/net/sched/sch_fifo.c
+++ b/net/sched/sch_fifo.c
@@ -48,10 +48,10 @@ static int fifo_init(struct Qdisc *sch, struct nlattr *opt)
 	struct fifo_sched_data *q = qdisc_priv(sch);
 
 	if (opt == NULL) {
-		u32 limit = sch->dev->tx_queue_len ? : 1;
+		u32 limit = qdisc_dev(sch)->tx_queue_len ? : 1;
 
 		if (sch->ops == &bfifo_qdisc_ops)
-			limit *= sch->dev->mtu;
+			limit *= qdisc_dev(sch)->mtu;
 
 		q->limit = limit;
 	} else {
@@ -137,7 +137,7 @@ struct Qdisc *fifo_create_dflt(struct Qdisc *sch, struct Qdisc_ops *ops,
 	struct Qdisc *q;
 	int err = -ENOMEM;
 
-	q = qdisc_create_dflt(sch->dev, sch->dev_queue,
+	q = qdisc_create_dflt(qdisc_dev(sch), sch->dev_queue,
 			      ops, TC_H_MAKE(sch->handle, 1));
 	if (q) {
 		err = fifo_set_limit(q, limit);
diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c
index d97086480893..b626a4f32b6b 100644
--- a/net/sched/sch_generic.c
+++ b/net/sched/sch_generic.c
@@ -364,7 +364,7 @@ static int pfifo_fast_enqueue(struct sk_buff *skb, struct Qdisc* qdisc)
 {
 	struct sk_buff_head *list = prio2list(skb, qdisc);
 
-	if (skb_queue_len(list) < qdisc->dev->tx_queue_len) {
+	if (skb_queue_len(list) < qdisc_dev(qdisc)->tx_queue_len) {
 		qdisc->q.qlen++;
 		return __qdisc_enqueue_tail(skb, qdisc, list);
 	}
@@ -440,8 +440,7 @@ static struct Qdisc_ops pfifo_fast_ops __read_mostly = {
 	.owner		=	THIS_MODULE,
 };
 
-struct Qdisc *qdisc_alloc(struct net_device *dev,
-			  struct netdev_queue *dev_queue,
+struct Qdisc *qdisc_alloc(struct netdev_queue *dev_queue,
 			  struct Qdisc_ops *ops)
 {
 	void *p;
@@ -465,8 +464,7 @@ struct Qdisc *qdisc_alloc(struct net_device *dev,
 	sch->enqueue = ops->enqueue;
 	sch->dequeue = ops->dequeue;
 	sch->dev_queue = dev_queue;
-	sch->dev = dev;
-	dev_hold(dev);
+	dev_hold(qdisc_dev(sch));
 	atomic_set(&sch->refcnt, 1);
 
 	return sch;
@@ -481,7 +479,7 @@ struct Qdisc * qdisc_create_dflt(struct net_device *dev,
 {
 	struct Qdisc *sch;
 
-	sch = qdisc_alloc(dev, dev_queue, ops);
+	sch = qdisc_alloc(dev_queue, ops);
 	if (IS_ERR(sch))
 		goto errout;
 	sch->stats_lock = &dev->queue_lock;
@@ -534,7 +532,7 @@ void qdisc_destroy(struct Qdisc *qdisc)
 		ops->destroy(qdisc);
 
 	module_put(ops->owner);
-	dev_put(qdisc->dev);
+	dev_put(qdisc_dev(qdisc));
 	call_rcu(&qdisc->q_rcu, __qdisc_destroy);
 }
 EXPORT_SYMBOL(qdisc_destroy);
diff --git a/net/sched/sch_gred.c b/net/sched/sch_gred.c
index c89fba56db56..39fa28511f07 100644
--- a/net/sched/sch_gred.c
+++ b/net/sched/sch_gred.c
@@ -164,7 +164,7 @@ static int gred_enqueue(struct sk_buff *skb, struct Qdisc* sch)
 			 * if no default DP has been configured. This
 			 * allows for DP flows to be left untouched.
 			 */
-			if (skb_queue_len(&sch->q) < sch->dev->tx_queue_len)
+			if (skb_queue_len(&sch->q) < qdisc_dev(sch)->tx_queue_len)
 				return qdisc_enqueue_tail(skb, sch);
 			else
 				goto drop;
diff --git a/net/sched/sch_hfsc.c b/net/sched/sch_hfsc.c
index 5a22fec4eadd..333525422f45 100644
--- a/net/sched/sch_hfsc.c
+++ b/net/sched/sch_hfsc.c
@@ -1045,7 +1045,7 @@ hfsc_change_class(struct Qdisc *sch, u32 classid, u32 parentid,
 
 		if (tca[TCA_RATE])
 			gen_replace_estimator(&cl->bstats, &cl->rate_est,
-					      &sch->dev->queue_lock,
+					      &qdisc_dev(sch)->queue_lock,
 					      tca[TCA_RATE]);
 		return 0;
 	}
@@ -1083,7 +1083,7 @@ hfsc_change_class(struct Qdisc *sch, u32 classid, u32 parentid,
 	cl->refcnt    = 1;
 	cl->sched     = q;
 	cl->cl_parent = parent;
-	cl->qdisc = qdisc_create_dflt(sch->dev, sch->dev_queue,
+	cl->qdisc = qdisc_create_dflt(qdisc_dev(sch), sch->dev_queue,
 				      &pfifo_qdisc_ops, classid);
 	if (cl->qdisc == NULL)
 		cl->qdisc = &noop_qdisc;
@@ -1104,7 +1104,7 @@ hfsc_change_class(struct Qdisc *sch, u32 classid, u32 parentid,
 
 	if (tca[TCA_RATE])
 		gen_new_estimator(&cl->bstats, &cl->rate_est,
-				  &sch->dev->queue_lock, tca[TCA_RATE]);
+				  &qdisc_dev(sch)->queue_lock, tca[TCA_RATE]);
 	*arg = (unsigned long)cl;
 	return 0;
 }
@@ -1202,7 +1202,7 @@ hfsc_graft_class(struct Qdisc *sch, unsigned long arg, struct Qdisc *new,
 	if (cl->level > 0)
 		return -EINVAL;
 	if (new == NULL) {
-		new = qdisc_create_dflt(sch->dev, sch->dev_queue,
+		new = qdisc_create_dflt(qdisc_dev(sch), sch->dev_queue,
 					&pfifo_qdisc_ops,
 					cl->cl_common.classid);
 		if (new == NULL)
@@ -1445,7 +1445,7 @@ hfsc_init_qdisc(struct Qdisc *sch, struct nlattr *opt)
 	q->root.cl_common.classid = sch->handle;
 	q->root.refcnt  = 1;
 	q->root.sched   = q;
-	q->root.qdisc = qdisc_create_dflt(sch->dev, sch->dev_queue,
+	q->root.qdisc = qdisc_create_dflt(qdisc_dev(sch), sch->dev_queue,
 					  &pfifo_qdisc_ops,
 					  sch->handle);
 	if (q->root.qdisc == NULL)
diff --git a/net/sched/sch_htb.c b/net/sched/sch_htb.c
index 956a67f66b9c..31f7d1536e6d 100644
--- a/net/sched/sch_htb.c
+++ b/net/sched/sch_htb.c
@@ -1026,7 +1026,7 @@ static int htb_init(struct Qdisc *sch, struct nlattr *opt)
 	qdisc_watchdog_init(&q->watchdog, sch);
 	skb_queue_head_init(&q->direct_queue);
 
-	q->direct_qlen = sch->dev->tx_queue_len;
+	q->direct_qlen = qdisc_dev(sch)->tx_queue_len;
 	if (q->direct_qlen < 2)	/* some devices have zero tx_queue_len */
 		q->direct_qlen = 2;
 
@@ -1043,7 +1043,7 @@ static int htb_dump(struct Qdisc *sch, struct sk_buff *skb)
 	struct nlattr *nest;
 	struct tc_htb_glob gopt;
 
-	spin_lock_bh(&sch->dev->queue_lock);
+	spin_lock_bh(&qdisc_dev(sch)->queue_lock);
 
 	gopt.direct_pkts = q->direct_pkts;
 	gopt.version = HTB_VER;
@@ -1057,11 +1057,11 @@ static int htb_dump(struct Qdisc *sch, struct sk_buff *skb)
 	NLA_PUT(skb, TCA_HTB_INIT, sizeof(gopt), &gopt);
 	nla_nest_end(skb, nest);
 
-	spin_unlock_bh(&sch->dev->queue_lock);
+	spin_unlock_bh(&qdisc_dev(sch)->queue_lock);
 	return skb->len;
 
 nla_put_failure:
-	spin_unlock_bh(&sch->dev->queue_lock);
+	spin_unlock_bh(&qdisc_dev(sch)->queue_lock);
 	nla_nest_cancel(skb, nest);
 	return -1;
 }
@@ -1073,7 +1073,7 @@ static int htb_dump_class(struct Qdisc *sch, unsigned long arg,
 	struct nlattr *nest;
 	struct tc_htb_opt opt;
 
-	spin_lock_bh(&sch->dev->queue_lock);
+	spin_lock_bh(&qdisc_dev(sch)->queue_lock);
 	tcm->tcm_parent = cl->parent ? cl->parent->common.classid : TC_H_ROOT;
 	tcm->tcm_handle = cl->common.classid;
 	if (!cl->level && cl->un.leaf.q)
@@ -1095,11 +1095,11 @@ static int htb_dump_class(struct Qdisc *sch, unsigned long arg,
 	NLA_PUT(skb, TCA_HTB_PARMS, sizeof(opt), &opt);
 
 	nla_nest_end(skb, nest);
-	spin_unlock_bh(&sch->dev->queue_lock);
+	spin_unlock_bh(&qdisc_dev(sch)->queue_lock);
 	return skb->len;
 
 nla_put_failure:
-	spin_unlock_bh(&sch->dev->queue_lock);
+	spin_unlock_bh(&qdisc_dev(sch)->queue_lock);
 	nla_nest_cancel(skb, nest);
 	return -1;
 }
@@ -1129,7 +1129,7 @@ static int htb_graft(struct Qdisc *sch, unsigned long arg, struct Qdisc *new,
 
 	if (cl && !cl->level) {
 		if (new == NULL &&
-		    (new = qdisc_create_dflt(sch->dev, sch->dev_queue,
+		    (new = qdisc_create_dflt(qdisc_dev(sch), sch->dev_queue,
 					     &pfifo_qdisc_ops,
 					     cl->common.classid))
 		    == NULL)
@@ -1257,7 +1257,7 @@ static int htb_delete(struct Qdisc *sch, unsigned long arg)
 		return -EBUSY;
 
 	if (!cl->level && htb_parent_last_child(cl)) {
-		new_q = qdisc_create_dflt(sch->dev, sch->dev_queue,
+		new_q = qdisc_create_dflt(qdisc_dev(sch), sch->dev_queue,
 					  &pfifo_qdisc_ops,
 					  cl->parent->common.classid);
 		last_child = 1;
@@ -1365,7 +1365,7 @@ static int htb_change_class(struct Qdisc *sch, u32 classid,
 			goto failure;
 
 		gen_new_estimator(&cl->bstats, &cl->rate_est,
-				  &sch->dev->queue_lock,
+				  &qdisc_dev(sch)->queue_lock,
 				  tca[TCA_RATE] ? : &est.nla);
 		cl->refcnt = 1;
 		cl->children = 0;
@@ -1378,7 +1378,7 @@ static int htb_change_class(struct Qdisc *sch, u32 classid,
 		/* create leaf qdisc early because it uses kmalloc(GFP_KERNEL)
 		   so that can't be used inside of sch_tree_lock
 		   -- thanks to Karlis Peisenieks */
-		new_q = qdisc_create_dflt(sch->dev, sch->dev_queue,
+		new_q = qdisc_create_dflt(qdisc_dev(sch), sch->dev_queue,
 					  &pfifo_qdisc_ops, classid);
 		sch_tree_lock(sch);
 		if (parent && !parent->level) {
@@ -1420,7 +1420,7 @@ static int htb_change_class(struct Qdisc *sch, u32 classid,
 	} else {
 		if (tca[TCA_RATE])
 			gen_replace_estimator(&cl->bstats, &cl->rate_est,
-					      &sch->dev->queue_lock,
+					      &qdisc_dev(sch)->queue_lock,
 					      tca[TCA_RATE]);
 		sch_tree_lock(sch);
 	}
diff --git a/net/sched/sch_netem.c b/net/sched/sch_netem.c
index aa7a04e32ae9..790582960444 100644
--- a/net/sched/sch_netem.c
+++ b/net/sched/sch_netem.c
@@ -180,7 +180,7 @@ static int netem_enqueue(struct sk_buff *skb, struct Qdisc *sch)
 	 * skb will be queued.
 	 */
 	if (count > 1 && (skb2 = skb_clone(skb, GFP_ATOMIC)) != NULL) {
-		struct Qdisc *rootq = sch->dev->qdisc;
+		struct Qdisc *rootq = qdisc_dev(sch)->qdisc;
 		u32 dupsave = q->duplicate; /* prevent duplicating a dup... */
 		q->duplicate = 0;
 
@@ -333,9 +333,9 @@ static int get_dist_table(struct Qdisc *sch, const struct nlattr *attr)
 	for (i = 0; i < n; i++)
 		d->table[i] = data[i];
 
-	spin_lock_bh(&sch->dev->queue_lock);
+	spin_lock_bh(&qdisc_dev(sch)->queue_lock);
 	d = xchg(&q->delay_dist, d);
-	spin_unlock_bh(&sch->dev->queue_lock);
+	spin_unlock_bh(&qdisc_dev(sch)->queue_lock);
 
 	kfree(d);
 	return 0;
@@ -495,7 +495,7 @@ static int tfifo_init(struct Qdisc *sch, struct nlattr *opt)
 
 		q->limit = ctl->limit;
 	} else
-		q->limit = max_t(u32, sch->dev->tx_queue_len, 1);
+		q->limit = max_t(u32, qdisc_dev(sch)->tx_queue_len, 1);
 
 	q->oldest = PSCHED_PASTPERFECT;
 	return 0;
@@ -536,7 +536,7 @@ static int netem_init(struct Qdisc *sch, struct nlattr *opt)
 
 	qdisc_watchdog_init(&q->watchdog, sch);
 
-	q->qdisc = qdisc_create_dflt(sch->dev, sch->dev_queue,
+	q->qdisc = qdisc_create_dflt(qdisc_dev(sch), sch->dev_queue,
 				     &tfifo_qdisc_ops,
 				     TC_H_MAKE(sch->handle, 1));
 	if (!q->qdisc) {
diff --git a/net/sched/sch_prio.c b/net/sched/sch_prio.c
index ca58a039208e..39157f7bc046 100644
--- a/net/sched/sch_prio.c
+++ b/net/sched/sch_prio.c
@@ -136,7 +136,8 @@ prio_dequeue(struct Qdisc* sch)
 		 * pulling an skb.  This way we avoid excessive requeues
 		 * for slower queues.
 		 */
-		if (!__netif_subqueue_stopped(sch->dev, (q->mq ? prio : 0))) {
+		if (!__netif_subqueue_stopped(qdisc_dev(sch),
+					      (q->mq ? prio : 0))) {
 			qdisc = q->queues[prio];
 			skb = qdisc->dequeue(qdisc);
 			if (skb) {
@@ -165,8 +166,8 @@ static struct sk_buff *rr_dequeue(struct Qdisc* sch)
 		 * for slower queues.  If the queue is stopped, try the
 		 * next queue.
 		 */
-		if (!__netif_subqueue_stopped(sch->dev,
-					    (q->mq ? q->curband : 0))) {
+		if (!__netif_subqueue_stopped(qdisc_dev(sch),
+					      (q->mq ? q->curband : 0))) {
 			qdisc = q->queues[q->curband];
 			skb = qdisc->dequeue(qdisc);
 			if (skb) {
@@ -249,10 +250,10 @@ static int prio_tune(struct Qdisc *sch, struct nlattr *opt)
 	if (q->mq) {
 		if (sch->parent != TC_H_ROOT)
 			return -EINVAL;
-		if (netif_is_multiqueue(sch->dev)) {
+		if (netif_is_multiqueue(qdisc_dev(sch))) {
 			if (q->bands == 0)
-				q->bands = sch->dev->egress_subqueue_count;
-			else if (q->bands != sch->dev->egress_subqueue_count)
+				q->bands = qdisc_dev(sch)->egress_subqueue_count;
+			else if (q->bands != qdisc_dev(sch)->egress_subqueue_count)
 				return -EINVAL;
 		} else
 			return -EOPNOTSUPP;
@@ -281,7 +282,7 @@ static int prio_tune(struct Qdisc *sch, struct nlattr *opt)
 	for (i=0; i<q->bands; i++) {
 		if (q->queues[i] == &noop_qdisc) {
 			struct Qdisc *child;
-			child = qdisc_create_dflt(sch->dev, sch->dev_queue,
+			child = qdisc_create_dflt(qdisc_dev(sch), sch->dev_queue,
 						  &pfifo_qdisc_ops,
 						  TC_H_MAKE(sch->handle, i + 1));
 			if (child) {
diff --git a/net/sched/sch_sfq.c b/net/sched/sch_sfq.c
index 6a97afbfb952..8458f630fac4 100644
--- a/net/sched/sch_sfq.c
+++ b/net/sched/sch_sfq.c
@@ -461,7 +461,7 @@ static int sfq_change(struct Qdisc *sch, struct nlattr *opt)
 		return -EINVAL;
 
 	sch_tree_lock(sch);
-	q->quantum = ctl->quantum ? : psched_mtu(sch->dev);
+	q->quantum = ctl->quantum ? : psched_mtu(qdisc_dev(sch));
 	q->perturb_period = ctl->perturb_period * HZ;
 	if (ctl->limit)
 		q->limit = min_t(u32, ctl->limit, SFQ_DEPTH - 1);
@@ -502,7 +502,7 @@ static int sfq_init(struct Qdisc *sch, struct nlattr *opt)
 	q->max_depth = 0;
 	q->tail = SFQ_DEPTH;
 	if (opt == NULL) {
-		q->quantum = psched_mtu(sch->dev);
+		q->quantum = psched_mtu(qdisc_dev(sch));
 		q->perturb_period = 0;
 		q->perturbation = net_random();
 	} else {
diff --git a/net/sched/sch_teql.c b/net/sched/sch_teql.c
index 0444fd0f0d22..b3fc82623fc6 100644
--- a/net/sched/sch_teql.c
+++ b/net/sched/sch_teql.c
@@ -78,7 +78,7 @@ struct teql_sched_data
 static int
 teql_enqueue(struct sk_buff *skb, struct Qdisc* sch)
 {
-	struct net_device *dev = sch->dev;
+	struct net_device *dev = qdisc_dev(sch);
 	struct teql_sched_data *q = qdisc_priv(sch);
 
 	if (q->q.qlen < dev->tx_queue_len) {
@@ -111,7 +111,7 @@ teql_dequeue(struct Qdisc* sch)
 
 	skb = __skb_dequeue(&dat->q);
 	if (skb == NULL) {
-		struct net_device *m = dat->m->dev->qdisc->dev;
+		struct net_device *m = qdisc_dev(dat->m->dev->qdisc);
 		if (m) {
 			dat->m->slaves = sch;
 			netif_wake_queue(m);
@@ -170,7 +170,7 @@ teql_destroy(struct Qdisc* sch)
 
 static int teql_qdisc_init(struct Qdisc *sch, struct nlattr *opt)
 {
-	struct net_device *dev = sch->dev;
+	struct net_device *dev = qdisc_dev(sch);
 	struct teql_master *m = (struct teql_master*)sch->ops;
 	struct teql_sched_data *q = qdisc_priv(sch);
 
@@ -282,7 +282,7 @@ restart:
 		goto drop;
 
 	do {
-		struct net_device *slave = q->dev;
+		struct net_device *slave = qdisc_dev(q);
 
 		if (slave->qdisc_sleeping != q)
 			continue;
@@ -352,7 +352,7 @@ static int teql_master_open(struct net_device *dev)
 
 	q = m->slaves;
 	do {
-		struct net_device *slave = q->dev;
+		struct net_device *slave = qdisc_dev(q);
 
 		if (slave == NULL)
 			return -EUNATCH;
@@ -403,7 +403,7 @@ static int teql_master_mtu(struct net_device *dev, int new_mtu)
 	q = m->slaves;
 	if (q) {
 		do {
-			if (new_mtu > q->dev->mtu)
+			if (new_mtu > qdisc_dev(q)->mtu)
 				return -EINVAL;
 		} while ((q=NEXT_SLAVE(q)) != m->slaves);
 	}
-- 
cgit v1.2.3


From dc2b48475a0a36f8b3bbb2da60d3a006dc5c2c84 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Tue, 8 Jul 2008 17:18:23 -0700
Subject: netdev: Move queue_lock into struct netdev_queue.

The lock is now an attribute of the device queue.

One thing to notice is that "suspicious" places
emerge which will need specific training about
multiple queue handling.  They are so marked with
explicit "netdev->rx_queue" and "netdev->tx_queue"
references.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ifb.c         |  8 ++++----
 include/linux/netdevice.h |  4 ++--
 net/core/dev.c            | 33 ++++++++++++++++++++++-----------
 net/mac80211/main.c       | 10 +++++-----
 net/mac80211/wme.c        |  2 +-
 net/sched/sch_api.c       |  2 +-
 net/sched/sch_cbq.c       |  8 ++++----
 net/sched/sch_generic.c   | 40 ++++++++++++++++++++--------------------
 net/sched/sch_hfsc.c      |  4 ++--
 net/sched/sch_htb.c       | 16 ++++++++--------
 net/sched/sch_netem.c     |  4 ++--
 net/sched/sch_teql.c      |  4 ++--
 12 files changed, 73 insertions(+), 62 deletions(-)

(limited to 'include')

diff --git a/drivers/net/ifb.c b/drivers/net/ifb.c
index af233b591534..bc3de272a829 100644
--- a/drivers/net/ifb.c
+++ b/drivers/net/ifb.c
@@ -229,12 +229,12 @@ module_param(numifbs, int, 0);
 MODULE_PARM_DESC(numifbs, "Number of ifb devices");
 
 /*
- * dev_ifb->queue_lock is usually taken after dev->ingress_lock,
+ * dev_ifb->tx_queue.lock is usually taken after dev->ingress_lock,
  * reversely to e.g. qdisc_lock_tree(). It should be safe until
- * ifb doesn't take dev->queue_lock with dev_ifb->ingress_lock.
+ * ifb doesn't take dev->tx_queue.lock with dev_ifb->ingress_lock.
  * But lockdep should know that ifb has different locks from dev.
  */
-static struct lock_class_key ifb_queue_lock_key;
+static struct lock_class_key ifb_tx_queue_lock_key;
 static struct lock_class_key ifb_ingress_lock_key;
 
 
@@ -258,7 +258,7 @@ static int __init ifb_init_one(int index)
 	if (err < 0)
 		goto err;
 
-	lockdep_set_class(&dev_ifb->queue_lock, &ifb_queue_lock_key);
+	lockdep_set_class(&dev_ifb->tx_queue.lock, &ifb_tx_queue_lock_key);
 	lockdep_set_class(&dev_ifb->ingress_lock, &ifb_ingress_lock_key);
 
 	return 0;
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 515fd25bf0fc..e835acacb479 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -449,6 +449,7 @@ static inline void napi_synchronize(const struct napi_struct *n)
 #endif
 
 struct netdev_queue {
+	spinlock_t		lock;
 	struct net_device	*dev;
 };
 
@@ -629,7 +630,7 @@ struct net_device
 	unsigned char		broadcast[MAX_ADDR_LEN];	/* hw bcast add	*/
 
 	struct netdev_queue	rx_queue;
-	struct netdev_queue	tx_queue;
+	struct netdev_queue	tx_queue ____cacheline_aligned_in_smp;
 
 	/* ingress path synchronizer */
 	spinlock_t		ingress_lock;
@@ -639,7 +640,6 @@ struct net_device
  * Cache line mostly used on queue transmit path (qdisc)
  */
 	/* device queue lock */
-	spinlock_t		queue_lock ____cacheline_aligned_in_smp;
 	struct Qdisc		*qdisc;
 	struct Qdisc		*qdisc_sleeping;
 	struct list_head	qdisc_list;
diff --git a/net/core/dev.c b/net/core/dev.c
index 9b281c906eb0..05011048b86c 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -1667,6 +1667,7 @@ out_kfree_skb:
 int dev_queue_xmit(struct sk_buff *skb)
 {
 	struct net_device *dev = skb->dev;
+	struct netdev_queue *txq;
 	struct Qdisc *q;
 	int rc = -ENOMEM;
 
@@ -1699,14 +1700,15 @@ int dev_queue_xmit(struct sk_buff *skb)
 	}
 
 gso:
-	spin_lock_prefetch(&dev->queue_lock);
+	txq = &dev->tx_queue;
+	spin_lock_prefetch(&txq->lock);
 
 	/* Disable soft irqs for various locks below. Also
 	 * stops preemption for RCU.
 	 */
 	rcu_read_lock_bh();
 
-	/* Updates of qdisc are serialized by queue_lock.
+	/* Updates of qdisc are serialized by queue->lock.
 	 * The struct Qdisc which is pointed to by qdisc is now a
 	 * rcu structure - it may be accessed without acquiring
 	 * a lock (but the structure may be stale.) The freeing of the
@@ -1714,7 +1716,7 @@ gso:
 	 * more references to it.
 	 *
 	 * If the qdisc has an enqueue function, we still need to
-	 * hold the queue_lock before calling it, since queue_lock
+	 * hold the queue->lock before calling it, since queue->lock
 	 * also serializes access to the device queue.
 	 */
 
@@ -1724,19 +1726,19 @@ gso:
 #endif
 	if (q->enqueue) {
 		/* Grab device queue */
-		spin_lock(&dev->queue_lock);
+		spin_lock(&txq->lock);
 		q = dev->qdisc;
 		if (q->enqueue) {
 			/* reset queue_mapping to zero */
 			skb_set_queue_mapping(skb, 0);
 			rc = q->enqueue(skb, q);
 			qdisc_run(dev);
-			spin_unlock(&dev->queue_lock);
+			spin_unlock(&txq->lock);
 
 			rc = rc == NET_XMIT_BYPASS ? NET_XMIT_SUCCESS : rc;
 			goto out;
 		}
-		spin_unlock(&dev->queue_lock);
+		spin_unlock(&txq->lock);
 	}
 
 	/* The device has no queue. Common case for software devices:
@@ -1919,14 +1921,17 @@ static void net_tx_action(struct softirq_action *h)
 
 		while (head) {
 			struct net_device *dev = head;
+			struct netdev_queue *txq;
 			head = head->next_sched;
 
+			txq = &dev->tx_queue;
+
 			smp_mb__before_clear_bit();
 			clear_bit(__LINK_STATE_SCHED, &dev->state);
 
-			if (spin_trylock(&dev->queue_lock)) {
+			if (spin_trylock(&txq->lock)) {
 				qdisc_run(dev);
-				spin_unlock(&dev->queue_lock);
+				spin_unlock(&txq->lock);
 			} else {
 				netif_schedule(dev);
 			}
@@ -3787,7 +3792,6 @@ int register_netdevice(struct net_device *dev)
 	BUG_ON(!dev_net(dev));
 	net = dev_net(dev);
 
-	spin_lock_init(&dev->queue_lock);
 	spin_lock_init(&dev->_xmit_lock);
 	netdev_set_lockdep_class(&dev->_xmit_lock, dev->type);
 	dev->xmit_lock_owner = -1;
@@ -4072,10 +4076,17 @@ static struct net_device_stats *internal_stats(struct net_device *dev)
 	return &dev->stats;
 }
 
+static void netdev_init_one_queue(struct net_device *dev,
+				  struct netdev_queue *queue)
+{
+	spin_lock_init(&queue->lock);
+	queue->dev = dev;
+}
+
 static void netdev_init_queues(struct net_device *dev)
 {
-	dev->rx_queue.dev = dev;
-	dev->tx_queue.dev = dev;
+	netdev_init_one_queue(dev, &dev->rx_queue);
+	netdev_init_one_queue(dev, &dev->tx_queue);
 }
 
 /**
diff --git a/net/mac80211/main.c b/net/mac80211/main.c
index cf477ad39dac..12aeaf78ae75 100644
--- a/net/mac80211/main.c
+++ b/net/mac80211/main.c
@@ -636,7 +636,7 @@ int ieee80211_start_tx_ba_session(struct ieee80211_hw *hw, u8 *ra, u16 tid)
 
 	/* ensure that TX flow won't interrupt us
 	 * until the end of the call to requeue function */
-	spin_lock_bh(&local->mdev->queue_lock);
+	spin_lock_bh(&local->mdev->tx_queue.lock);
 
 	/* create a new queue for this aggregation */
 	ret = ieee80211_ht_agg_queue_add(local, sta, tid);
@@ -675,7 +675,7 @@ int ieee80211_start_tx_ba_session(struct ieee80211_hw *hw, u8 *ra, u16 tid)
 
 	/* Will put all the packets in the new SW queue */
 	ieee80211_requeue(local, ieee802_1d_to_ac[tid]);
-	spin_unlock_bh(&local->mdev->queue_lock);
+	spin_unlock_bh(&local->mdev->tx_queue.lock);
 	spin_unlock_bh(&sta->lock);
 
 	/* send an addBA request */
@@ -701,7 +701,7 @@ int ieee80211_start_tx_ba_session(struct ieee80211_hw *hw, u8 *ra, u16 tid)
 err_unlock_queue:
 	kfree(sta->ampdu_mlme.tid_tx[tid]);
 	sta->ampdu_mlme.tid_tx[tid] = NULL;
-	spin_unlock_bh(&local->mdev->queue_lock);
+	spin_unlock_bh(&local->mdev->tx_queue.lock);
 	ret = -EBUSY;
 err_unlock_sta:
 	spin_unlock_bh(&sta->lock);
@@ -875,10 +875,10 @@ void ieee80211_stop_tx_ba_cb(struct ieee80211_hw *hw, u8 *ra, u8 tid)
 
 	/* avoid ordering issues: we are the only one that can modify
 	 * the content of the qdiscs */
-	spin_lock_bh(&local->mdev->queue_lock);
+	spin_lock_bh(&local->mdev->tx_queue.lock);
 	/* remove the queue for this aggregation */
 	ieee80211_ht_agg_queue_remove(local, sta, tid, 1);
-	spin_unlock_bh(&local->mdev->queue_lock);
+	spin_unlock_bh(&local->mdev->tx_queue.lock);
 
 	/* we just requeued the all the frames that were in the removed
 	 * queue, and since we might miss a softirq we do netif_schedule.
diff --git a/net/mac80211/wme.c b/net/mac80211/wme.c
index 2fbc171130bf..59ed9cae66b9 100644
--- a/net/mac80211/wme.c
+++ b/net/mac80211/wme.c
@@ -648,7 +648,7 @@ int ieee80211_ht_agg_queue_add(struct ieee80211_local *local,
 }
 
 /**
- * the caller needs to hold local->mdev->queue_lock
+ * the caller needs to hold local->mdev->tx_queue.lock
  */
 void ieee80211_ht_agg_queue_remove(struct ieee80211_local *local,
 				   struct sta_info *sta, u16 tid,
diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c
index 1f893082a4f6..2a1834f8c7d8 100644
--- a/net/sched/sch_api.c
+++ b/net/sched/sch_api.c
@@ -606,7 +606,7 @@ qdisc_create(struct net_device *dev, struct netdev_queue *dev_queue,
 		sch->stats_lock = &dev->ingress_lock;
 		handle = TC_H_MAKE(TC_H_INGRESS, 0);
 	} else {
-		sch->stats_lock = &dev->queue_lock;
+		sch->stats_lock = &dev_queue->lock;
 		if (handle == 0) {
 			handle = qdisc_alloc_handle(dev);
 			err = -ENOMEM;
diff --git a/net/sched/sch_cbq.c b/net/sched/sch_cbq.c
index 9f2ace585fd6..99ce3da2b0a4 100644
--- a/net/sched/sch_cbq.c
+++ b/net/sched/sch_cbq.c
@@ -1746,10 +1746,10 @@ static void cbq_put(struct Qdisc *sch, unsigned long arg)
 #ifdef CONFIG_NET_CLS_ACT
 		struct cbq_sched_data *q = qdisc_priv(sch);
 
-		spin_lock_bh(&qdisc_dev(sch)->queue_lock);
+		spin_lock_bh(&sch->dev_queue->lock);
 		if (q->rx_class == cl)
 			q->rx_class = NULL;
-		spin_unlock_bh(&qdisc_dev(sch)->queue_lock);
+		spin_unlock_bh(&sch->dev_queue->lock);
 #endif
 
 		cbq_destroy_class(sch, cl);
@@ -1828,7 +1828,7 @@ cbq_change_class(struct Qdisc *sch, u32 classid, u32 parentid, struct nlattr **t
 
 		if (tca[TCA_RATE])
 			gen_replace_estimator(&cl->bstats, &cl->rate_est,
-					      &qdisc_dev(sch)->queue_lock,
+					      &sch->dev_queue->lock,
 					      tca[TCA_RATE]);
 		return 0;
 	}
@@ -1919,7 +1919,7 @@ cbq_change_class(struct Qdisc *sch, u32 classid, u32 parentid, struct nlattr **t
 
 	if (tca[TCA_RATE])
 		gen_new_estimator(&cl->bstats, &cl->rate_est,
-				  &qdisc_dev(sch)->queue_lock, tca[TCA_RATE]);
+				  &sch->dev_queue->lock, tca[TCA_RATE]);
 
 	*arg = (unsigned long)cl;
 	return 0;
diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c
index b626a4f32b6b..ee8f9f78a095 100644
--- a/net/sched/sch_generic.c
+++ b/net/sched/sch_generic.c
@@ -29,31 +29,31 @@
 /* Main transmission queue. */
 
 /* Modifications to data participating in scheduling must be protected with
- * dev->queue_lock spinlock.
+ * queue->lock spinlock.
  *
  * The idea is the following:
  * - enqueue, dequeue are serialized via top level device
- *   spinlock dev->queue_lock.
+ *   spinlock queue->lock.
  * - ingress filtering is serialized via top level device
  *   spinlock dev->ingress_lock.
  * - updates to tree and tree walking are only done under the rtnl mutex.
  */
 
 void qdisc_lock_tree(struct net_device *dev)
-	__acquires(dev->queue_lock)
+	__acquires(dev->tx_queue.lock)
 	__acquires(dev->ingress_lock)
 {
-	spin_lock_bh(&dev->queue_lock);
+	spin_lock_bh(&dev->tx_queue.lock);
 	spin_lock(&dev->ingress_lock);
 }
 EXPORT_SYMBOL(qdisc_lock_tree);
 
 void qdisc_unlock_tree(struct net_device *dev)
 	__releases(dev->ingress_lock)
-	__releases(dev->queue_lock)
+	__releases(dev->tx_queue.lock)
 {
 	spin_unlock(&dev->ingress_lock);
-	spin_unlock_bh(&dev->queue_lock);
+	spin_unlock_bh(&dev->tx_queue.lock);
 }
 EXPORT_SYMBOL(qdisc_unlock_tree);
 
@@ -118,15 +118,15 @@ static inline int handle_dev_cpu_collision(struct sk_buff *skb,
 }
 
 /*
- * NOTE: Called under dev->queue_lock with locally disabled BH.
+ * NOTE: Called under queue->lock with locally disabled BH.
  *
  * __LINK_STATE_QDISC_RUNNING guarantees only one CPU can process this
- * device at a time. dev->queue_lock serializes queue accesses for
+ * device at a time. queue->lock serializes queue accesses for
  * this device AND dev->qdisc pointer itself.
  *
  *  netif_tx_lock serializes accesses to device driver.
  *
- *  dev->queue_lock and netif_tx_lock are mutually exclusive,
+ *  queue->lock and netif_tx_lock are mutually exclusive,
  *  if one is grabbed, another must be free.
  *
  * Note, that this procedure can be called by a watchdog timer
@@ -148,14 +148,14 @@ static inline int qdisc_restart(struct net_device *dev)
 
 
 	/* And release queue */
-	spin_unlock(&dev->queue_lock);
+	spin_unlock(&q->dev_queue->lock);
 
 	HARD_TX_LOCK(dev, smp_processor_id());
 	if (!netif_subqueue_stopped(dev, skb))
 		ret = dev_hard_start_xmit(skb, dev);
 	HARD_TX_UNLOCK(dev);
 
-	spin_lock(&dev->queue_lock);
+	spin_lock(&q->dev_queue->lock);
 	q = dev->qdisc;
 
 	switch (ret) {
@@ -482,7 +482,7 @@ struct Qdisc * qdisc_create_dflt(struct net_device *dev,
 	sch = qdisc_alloc(dev_queue, ops);
 	if (IS_ERR(sch))
 		goto errout;
-	sch->stats_lock = &dev->queue_lock;
+	sch->stats_lock = &dev_queue->lock;
 	sch->parent = parentid;
 
 	if (!ops->init || ops->init(sch, NULL) == 0)
@@ -494,7 +494,7 @@ errout:
 }
 EXPORT_SYMBOL(qdisc_create_dflt);
 
-/* Under dev->queue_lock and BH! */
+/* Under queue->lock and BH! */
 
 void qdisc_reset(struct Qdisc *qdisc)
 {
@@ -514,7 +514,7 @@ static void __qdisc_destroy(struct rcu_head *head)
 	kfree((char *) qdisc - qdisc->padded);
 }
 
-/* Under dev->queue_lock and BH! */
+/* Under queue->lock and BH! */
 
 void qdisc_destroy(struct Qdisc *qdisc)
 {
@@ -566,13 +566,13 @@ void dev_activate(struct net_device *dev)
 		/* Delay activation until next carrier-on event */
 		return;
 
-	spin_lock_bh(&dev->queue_lock);
+	spin_lock_bh(&dev->tx_queue.lock);
 	rcu_assign_pointer(dev->qdisc, dev->qdisc_sleeping);
 	if (dev->qdisc != &noqueue_qdisc) {
 		dev->trans_start = jiffies;
 		dev_watchdog_up(dev);
 	}
-	spin_unlock_bh(&dev->queue_lock);
+	spin_unlock_bh(&dev->tx_queue.lock);
 }
 
 void dev_deactivate(struct net_device *dev)
@@ -581,7 +581,7 @@ void dev_deactivate(struct net_device *dev)
 	struct sk_buff *skb;
 	int running;
 
-	spin_lock_bh(&dev->queue_lock);
+	spin_lock_bh(&dev->tx_queue.lock);
 	qdisc = dev->qdisc;
 	dev->qdisc = &noop_qdisc;
 
@@ -589,7 +589,7 @@ void dev_deactivate(struct net_device *dev)
 
 	skb = dev->gso_skb;
 	dev->gso_skb = NULL;
-	spin_unlock_bh(&dev->queue_lock);
+	spin_unlock_bh(&dev->tx_queue.lock);
 
 	kfree_skb(skb);
 
@@ -607,9 +607,9 @@ void dev_deactivate(struct net_device *dev)
 		 * Double-check inside queue lock to ensure that all effects
 		 * of the queue run are visible when we return.
 		 */
-		spin_lock_bh(&dev->queue_lock);
+		spin_lock_bh(&dev->tx_queue.lock);
 		running = test_bit(__LINK_STATE_QDISC_RUNNING, &dev->state);
-		spin_unlock_bh(&dev->queue_lock);
+		spin_unlock_bh(&dev->tx_queue.lock);
 
 		/*
 		 * The running flag should never be set at this point because
diff --git a/net/sched/sch_hfsc.c b/net/sched/sch_hfsc.c
index 333525422f45..997d520ca580 100644
--- a/net/sched/sch_hfsc.c
+++ b/net/sched/sch_hfsc.c
@@ -1045,7 +1045,7 @@ hfsc_change_class(struct Qdisc *sch, u32 classid, u32 parentid,
 
 		if (tca[TCA_RATE])
 			gen_replace_estimator(&cl->bstats, &cl->rate_est,
-					      &qdisc_dev(sch)->queue_lock,
+					      &sch->dev_queue->lock,
 					      tca[TCA_RATE]);
 		return 0;
 	}
@@ -1104,7 +1104,7 @@ hfsc_change_class(struct Qdisc *sch, u32 classid, u32 parentid,
 
 	if (tca[TCA_RATE])
 		gen_new_estimator(&cl->bstats, &cl->rate_est,
-				  &qdisc_dev(sch)->queue_lock, tca[TCA_RATE]);
+				  &sch->dev_queue->lock, tca[TCA_RATE]);
 	*arg = (unsigned long)cl;
 	return 0;
 }
diff --git a/net/sched/sch_htb.c b/net/sched/sch_htb.c
index 31f7d1536e6d..c8ca54cc26b0 100644
--- a/net/sched/sch_htb.c
+++ b/net/sched/sch_htb.c
@@ -1043,7 +1043,7 @@ static int htb_dump(struct Qdisc *sch, struct sk_buff *skb)
 	struct nlattr *nest;
 	struct tc_htb_glob gopt;
 
-	spin_lock_bh(&qdisc_dev(sch)->queue_lock);
+	spin_lock_bh(&sch->dev_queue->lock);
 
 	gopt.direct_pkts = q->direct_pkts;
 	gopt.version = HTB_VER;
@@ -1057,11 +1057,11 @@ static int htb_dump(struct Qdisc *sch, struct sk_buff *skb)
 	NLA_PUT(skb, TCA_HTB_INIT, sizeof(gopt), &gopt);
 	nla_nest_end(skb, nest);
 
-	spin_unlock_bh(&qdisc_dev(sch)->queue_lock);
+	spin_unlock_bh(&sch->dev_queue->lock);
 	return skb->len;
 
 nla_put_failure:
-	spin_unlock_bh(&qdisc_dev(sch)->queue_lock);
+	spin_unlock_bh(&sch->dev_queue->lock);
 	nla_nest_cancel(skb, nest);
 	return -1;
 }
@@ -1073,7 +1073,7 @@ static int htb_dump_class(struct Qdisc *sch, unsigned long arg,
 	struct nlattr *nest;
 	struct tc_htb_opt opt;
 
-	spin_lock_bh(&qdisc_dev(sch)->queue_lock);
+	spin_lock_bh(&sch->dev_queue->lock);
 	tcm->tcm_parent = cl->parent ? cl->parent->common.classid : TC_H_ROOT;
 	tcm->tcm_handle = cl->common.classid;
 	if (!cl->level && cl->un.leaf.q)
@@ -1095,11 +1095,11 @@ static int htb_dump_class(struct Qdisc *sch, unsigned long arg,
 	NLA_PUT(skb, TCA_HTB_PARMS, sizeof(opt), &opt);
 
 	nla_nest_end(skb, nest);
-	spin_unlock_bh(&qdisc_dev(sch)->queue_lock);
+	spin_unlock_bh(&sch->dev_queue->lock);
 	return skb->len;
 
 nla_put_failure:
-	spin_unlock_bh(&qdisc_dev(sch)->queue_lock);
+	spin_unlock_bh(&sch->dev_queue->lock);
 	nla_nest_cancel(skb, nest);
 	return -1;
 }
@@ -1365,7 +1365,7 @@ static int htb_change_class(struct Qdisc *sch, u32 classid,
 			goto failure;
 
 		gen_new_estimator(&cl->bstats, &cl->rate_est,
-				  &qdisc_dev(sch)->queue_lock,
+				  &sch->dev_queue->lock,
 				  tca[TCA_RATE] ? : &est.nla);
 		cl->refcnt = 1;
 		cl->children = 0;
@@ -1420,7 +1420,7 @@ static int htb_change_class(struct Qdisc *sch, u32 classid,
 	} else {
 		if (tca[TCA_RATE])
 			gen_replace_estimator(&cl->bstats, &cl->rate_est,
-					      &qdisc_dev(sch)->queue_lock,
+					      &sch->dev_queue->lock,
 					      tca[TCA_RATE]);
 		sch_tree_lock(sch);
 	}
diff --git a/net/sched/sch_netem.c b/net/sched/sch_netem.c
index 790582960444..71b73c528f9b 100644
--- a/net/sched/sch_netem.c
+++ b/net/sched/sch_netem.c
@@ -333,9 +333,9 @@ static int get_dist_table(struct Qdisc *sch, const struct nlattr *attr)
 	for (i = 0; i < n; i++)
 		d->table[i] = data[i];
 
-	spin_lock_bh(&qdisc_dev(sch)->queue_lock);
+	spin_lock_bh(&sch->dev_queue->lock);
 	d = xchg(&q->delay_dist, d);
-	spin_unlock_bh(&qdisc_dev(sch)->queue_lock);
+	spin_unlock_bh(&sch->dev_queue->lock);
 
 	kfree(d);
 	return 0;
diff --git a/net/sched/sch_teql.c b/net/sched/sch_teql.c
index b3fc82623fc6..4f3054e8e1ab 100644
--- a/net/sched/sch_teql.c
+++ b/net/sched/sch_teql.c
@@ -154,9 +154,9 @@ teql_destroy(struct Qdisc* sch)
 					master->slaves = NEXT_SLAVE(q);
 					if (q == master->slaves) {
 						master->slaves = NULL;
-						spin_lock_bh(&master->dev->queue_lock);
+						spin_lock_bh(&master->dev->tx_queue.lock);
 						qdisc_reset(master->dev->qdisc);
-						spin_unlock_bh(&master->dev->queue_lock);
+						spin_unlock_bh(&master->dev->tx_queue.lock);
 					}
 				}
 				skb_queue_purge(&dat->q);
-- 
cgit v1.2.3


From 555353cfa1aee293de445bfa6de43276138ddd82 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Tue, 8 Jul 2008 17:33:13 -0700
Subject: netdev: The ingress_lock member is no longer needed.

Every qdisc is assosciated with a queue, and in the case of ingress
qdiscs that will now be netdev->rx_queue so using that queue's lock is
the thing to do.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ifb.c         |  8 ++++----
 include/linux/netdevice.h |  2 --
 net/core/dev.c            | 12 +++++++-----
 net/sched/sch_api.c       |  3 +--
 net/sched/sch_generic.c   | 10 +++++-----
 5 files changed, 17 insertions(+), 18 deletions(-)

(limited to 'include')

diff --git a/drivers/net/ifb.c b/drivers/net/ifb.c
index bc3de272a829..ccbd6554f6eb 100644
--- a/drivers/net/ifb.c
+++ b/drivers/net/ifb.c
@@ -229,13 +229,13 @@ module_param(numifbs, int, 0);
 MODULE_PARM_DESC(numifbs, "Number of ifb devices");
 
 /*
- * dev_ifb->tx_queue.lock is usually taken after dev->ingress_lock,
+ * dev_ifb->tx_queue.lock is usually taken after dev->rx_queue.lock,
  * reversely to e.g. qdisc_lock_tree(). It should be safe until
- * ifb doesn't take dev->tx_queue.lock with dev_ifb->ingress_lock.
+ * ifb doesn't take dev->tx_queue.lock with dev_ifb->rx_queue.lock.
  * But lockdep should know that ifb has different locks from dev.
  */
 static struct lock_class_key ifb_tx_queue_lock_key;
-static struct lock_class_key ifb_ingress_lock_key;
+static struct lock_class_key ifb_rx_queue_lock_key;
 
 
 static int __init ifb_init_one(int index)
@@ -259,7 +259,7 @@ static int __init ifb_init_one(int index)
 		goto err;
 
 	lockdep_set_class(&dev_ifb->tx_queue.lock, &ifb_tx_queue_lock_key);
-	lockdep_set_class(&dev_ifb->ingress_lock, &ifb_ingress_lock_key);
+	lockdep_set_class(&dev_ifb->rx_queue.lock, &ifb_rx_queue_lock_key);
 
 	return 0;
 
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index e835acacb479..633a44c6fa5e 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -632,8 +632,6 @@ struct net_device
 	struct netdev_queue	rx_queue;
 	struct netdev_queue	tx_queue ____cacheline_aligned_in_smp;
 
-	/* ingress path synchronizer */
-	spinlock_t		ingress_lock;
 	struct Qdisc		*qdisc_ingress;
 
 /*
diff --git a/net/core/dev.c b/net/core/dev.c
index 05011048b86c..2322fb69fd53 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -2014,10 +2014,11 @@ static inline struct sk_buff *handle_macvlan(struct sk_buff *skb,
  */
 static int ing_filter(struct sk_buff *skb)
 {
-	struct Qdisc *q;
 	struct net_device *dev = skb->dev;
-	int result = TC_ACT_OK;
 	u32 ttl = G_TC_RTTL(skb->tc_verd);
+	struct netdev_queue *rxq;
+	int result = TC_ACT_OK;
+	struct Qdisc *q;
 
 	if (MAX_RED_LOOP < ttl++) {
 		printk(KERN_WARNING
@@ -2029,10 +2030,12 @@ static int ing_filter(struct sk_buff *skb)
 	skb->tc_verd = SET_TC_RTTL(skb->tc_verd, ttl);
 	skb->tc_verd = SET_TC_AT(skb->tc_verd, AT_INGRESS);
 
-	spin_lock(&dev->ingress_lock);
+	rxq = &dev->rx_queue;
+
+	spin_lock(&rxq->lock);
 	if ((q = dev->qdisc_ingress) != NULL)
 		result = q->enqueue(skb, q);
-	spin_unlock(&dev->ingress_lock);
+	spin_unlock(&rxq->lock);
 
 	return result;
 }
@@ -3795,7 +3798,6 @@ int register_netdevice(struct net_device *dev)
 	spin_lock_init(&dev->_xmit_lock);
 	netdev_set_lockdep_class(&dev->_xmit_lock, dev->type);
 	dev->xmit_lock_owner = -1;
-	spin_lock_init(&dev->ingress_lock);
 
 	dev->iflink = -1;
 
diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c
index 2a1834f8c7d8..570cef2a9c5f 100644
--- a/net/sched/sch_api.c
+++ b/net/sched/sch_api.c
@@ -601,12 +601,11 @@ qdisc_create(struct net_device *dev, struct netdev_queue *dev_queue,
 
 	sch->parent = parent;
 
+	sch->stats_lock = &dev_queue->lock;
 	if (handle == TC_H_INGRESS) {
 		sch->flags |= TCQ_F_INGRESS;
-		sch->stats_lock = &dev->ingress_lock;
 		handle = TC_H_MAKE(TC_H_INGRESS, 0);
 	} else {
-		sch->stats_lock = &dev_queue->lock;
 		if (handle == 0) {
 			handle = qdisc_alloc_handle(dev);
 			err = -ENOMEM;
diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c
index ee8f9f78a095..804d44b00348 100644
--- a/net/sched/sch_generic.c
+++ b/net/sched/sch_generic.c
@@ -35,24 +35,24 @@
  * - enqueue, dequeue are serialized via top level device
  *   spinlock queue->lock.
  * - ingress filtering is serialized via top level device
- *   spinlock dev->ingress_lock.
+ *   spinlock dev->rx_queue.lock.
  * - updates to tree and tree walking are only done under the rtnl mutex.
  */
 
 void qdisc_lock_tree(struct net_device *dev)
 	__acquires(dev->tx_queue.lock)
-	__acquires(dev->ingress_lock)
+	__acquires(dev->rx_queue.lock)
 {
 	spin_lock_bh(&dev->tx_queue.lock);
-	spin_lock(&dev->ingress_lock);
+	spin_lock(&dev->rx_queue.lock);
 }
 EXPORT_SYMBOL(qdisc_lock_tree);
 
 void qdisc_unlock_tree(struct net_device *dev)
-	__releases(dev->ingress_lock)
+	__releases(dev->rx_queue.lock)
 	__releases(dev->tx_queue.lock)
 {
-	spin_unlock(&dev->ingress_lock);
+	spin_unlock(&dev->rx_queue.lock);
 	spin_unlock_bh(&dev->tx_queue.lock);
 }
 EXPORT_SYMBOL(qdisc_unlock_tree);
-- 
cgit v1.2.3


From b0e1e6462df3c5944010b3328a546d8fe5d932cd Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Tue, 8 Jul 2008 17:42:10 -0700
Subject: netdev: Move rest of qdisc state into struct netdev_queue

Now qdisc, qdisc_sleeping, and qdisc_list also live there.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/isdn/i4l/isdn_net.c    |  2 +-
 include/linux/netdevice.h      | 10 ++---
 include/net/irda/irda_device.h |  2 +-
 net/core/dev.c                 |  4 +-
 net/core/link_watch.c          |  8 +++-
 net/core/rtnetlink.c           |  6 ++-
 net/ipv6/addrconf.c            |  3 +-
 net/mac80211/wme.c             | 20 ++++++----
 net/sched/cls_api.c            |  7 +++-
 net/sched/sch_api.c            | 34 ++++++++++------
 net/sched/sch_generic.c        | 90 ++++++++++++++++++++++++++----------------
 net/sched/sch_netem.c          |  2 +-
 net/sched/sch_teql.c           | 14 ++++---
 13 files changed, 125 insertions(+), 77 deletions(-)

(limited to 'include')

diff --git a/drivers/isdn/i4l/isdn_net.c b/drivers/isdn/i4l/isdn_net.c
index ef1a300068dc..457bbd119f9b 100644
--- a/drivers/isdn/i4l/isdn_net.c
+++ b/drivers/isdn/i4l/isdn_net.c
@@ -287,7 +287,7 @@ isdn_net_unbind_channel(isdn_net_local * lp)
 		   BEWARE! This chunk of code cannot be called from hardware
 		   interrupt handler. I hope it is true. --ANK
 		 */
-		qdisc_reset(lp->netdev->dev->qdisc);
+		qdisc_reset(lp->netdev->dev->tx_queue.qdisc);
 	}
 	lp->dialstate = 0;
 	dev->rx_netdev[isdn_dc2minor(lp->isdn_device, lp->isdn_channel)] = NULL;
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 633a44c6fa5e..df702a7b3db5 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -451,6 +451,9 @@ static inline void napi_synchronize(const struct napi_struct *n)
 struct netdev_queue {
 	spinlock_t		lock;
 	struct net_device	*dev;
+	struct Qdisc		*qdisc;
+	struct Qdisc		*qdisc_sleeping;
+	struct list_head	qdisc_list;
 };
 
 /*
@@ -634,13 +637,6 @@ struct net_device
 
 	struct Qdisc		*qdisc_ingress;
 
-/*
- * Cache line mostly used on queue transmit path (qdisc)
- */
-	/* device queue lock */
-	struct Qdisc		*qdisc;
-	struct Qdisc		*qdisc_sleeping;
-	struct list_head	qdisc_list;
 	unsigned long		tx_queue_len;	/* Max frames per queue allowed */
 
 	/* Partially transmitted GSO packet. */
diff --git a/include/net/irda/irda_device.h b/include/net/irda/irda_device.h
index f70e9b39ebaf..16fbf672e0b2 100644
--- a/include/net/irda/irda_device.h
+++ b/include/net/irda/irda_device.h
@@ -223,7 +223,7 @@ int  irda_device_is_receiving(struct net_device *dev);
 /* Interface for internal use */
 static inline int irda_device_txqueue_empty(const struct net_device *dev)
 {
-	return skb_queue_empty(&dev->qdisc->q);
+	return skb_queue_empty(&dev->tx_queue.qdisc->q);
 }
 int  irda_device_set_raw_mode(struct net_device* self, int status);
 struct net_device *alloc_irdadev(int sizeof_priv);
diff --git a/net/core/dev.c b/net/core/dev.c
index 2322fb69fd53..ce79c28d739d 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -1720,14 +1720,14 @@ gso:
 	 * also serializes access to the device queue.
 	 */
 
-	q = rcu_dereference(dev->qdisc);
+	q = rcu_dereference(txq->qdisc);
 #ifdef CONFIG_NET_CLS_ACT
 	skb->tc_verd = SET_TC_AT(skb->tc_verd,AT_EGRESS);
 #endif
 	if (q->enqueue) {
 		/* Grab device queue */
 		spin_lock(&txq->lock);
-		q = dev->qdisc;
+		q = txq->qdisc;
 		if (q->enqueue) {
 			/* reset queue_mapping to zero */
 			skb_set_queue_mapping(skb, 0);
diff --git a/net/core/link_watch.c b/net/core/link_watch.c
index a5e372b9ec4d..50218218445b 100644
--- a/net/core/link_watch.c
+++ b/net/core/link_watch.c
@@ -79,8 +79,10 @@ static void rfc2863_policy(struct net_device *dev)
 
 static int linkwatch_urgent_event(struct net_device *dev)
 {
+	struct netdev_queue *txq = &dev->tx_queue;
+
 	return netif_running(dev) && netif_carrier_ok(dev) &&
-	       dev->qdisc != dev->qdisc_sleeping;
+	       txq->qdisc != txq->qdisc_sleeping;
 }
 
 
@@ -181,7 +183,9 @@ static void __linkwatch_run_queue(int urgent_only)
 		rfc2863_policy(dev);
 		if (dev->flags & IFF_UP) {
 			if (netif_carrier_ok(dev)) {
-				WARN_ON(dev->qdisc_sleeping == &noop_qdisc);
+				struct netdev_queue *txq = &dev->tx_queue;
+
+				WARN_ON(txq->qdisc_sleeping == &noop_qdisc);
 				dev_activate(dev);
 			} else
 				dev_deactivate(dev);
diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c
index 6c8d7f0ea01a..8ef9f1db610e 100644
--- a/net/core/rtnetlink.c
+++ b/net/core/rtnetlink.c
@@ -605,6 +605,7 @@ static int rtnl_fill_ifinfo(struct sk_buff *skb, struct net_device *dev,
 			    int type, u32 pid, u32 seq, u32 change,
 			    unsigned int flags)
 {
+	struct netdev_queue *txq;
 	struct ifinfomsg *ifm;
 	struct nlmsghdr *nlh;
 	struct net_device_stats *stats;
@@ -635,8 +636,9 @@ static int rtnl_fill_ifinfo(struct sk_buff *skb, struct net_device *dev,
 	if (dev->master)
 		NLA_PUT_U32(skb, IFLA_MASTER, dev->master->ifindex);
 
-	if (dev->qdisc_sleeping)
-		NLA_PUT_STRING(skb, IFLA_QDISC, dev->qdisc_sleeping->ops->id);
+	txq = &dev->tx_queue;
+	if (txq->qdisc_sleeping)
+		NLA_PUT_STRING(skb, IFLA_QDISC, txq->qdisc_sleeping->ops->id);
 
 	if (1) {
 		struct rtnl_link_ifmap map = {
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index 8572cb05fc21..5c84c798331d 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -231,7 +231,8 @@ const struct in6_addr in6addr_linklocal_allrouters = IN6ADDR_LINKLOCAL_ALLROUTER
 /* Check if a valid qdisc is available */
 static inline int addrconf_qdisc_ok(struct net_device *dev)
 {
-	return (dev->qdisc != &noop_qdisc);
+	struct netdev_queue *txq = &dev->tx_queue;
+	return (txq->qdisc != &noop_qdisc);
 }
 
 /* Check if a route is valid prefix route */
diff --git a/net/mac80211/wme.c b/net/mac80211/wme.c
index 59ed9cae66b9..6ae43a3c7726 100644
--- a/net/mac80211/wme.c
+++ b/net/mac80211/wme.c
@@ -574,9 +574,10 @@ static struct Qdisc_ops wme_qdisc_ops __read_mostly =
 
 void ieee80211_install_qdisc(struct net_device *dev)
 {
+	struct netdev_queue *txq = &dev->tx_queue;
 	struct Qdisc *qdisc;
 
-	qdisc = qdisc_create_dflt(dev, &dev->tx_queue,
+	qdisc = qdisc_create_dflt(dev, txq,
 				  &wme_qdisc_ops, TC_H_ROOT);
 	if (!qdisc) {
 		printk(KERN_ERR "%s: qdisc installation failed\n", dev->name);
@@ -587,15 +588,17 @@ void ieee80211_install_qdisc(struct net_device *dev)
 	qdisc->handle = 0x80010000;
 
 	qdisc_lock_tree(dev);
-	list_add_tail(&qdisc->list, &dev->qdisc_list);
-	dev->qdisc_sleeping = qdisc;
+	list_add_tail(&qdisc->list, &txq->qdisc_list);
+	txq->qdisc_sleeping = qdisc;
 	qdisc_unlock_tree(dev);
 }
 
 
 int ieee80211_qdisc_installed(struct net_device *dev)
 {
-	return dev->qdisc_sleeping->ops == &wme_qdisc_ops;
+	struct netdev_queue *txq = &dev->tx_queue;
+
+	return txq->qdisc_sleeping->ops == &wme_qdisc_ops;
 }
 
 
@@ -614,8 +617,9 @@ int ieee80211_ht_agg_queue_add(struct ieee80211_local *local,
 			struct sta_info *sta, u16 tid)
 {
 	int i;
+	struct netdev_queue *txq = &local->mdev->tx_queue;
 	struct ieee80211_sched_data *q =
-			qdisc_priv(local->mdev->qdisc_sleeping);
+			qdisc_priv(txq->qdisc_sleeping);
 	DECLARE_MAC_BUF(mac);
 
 	/* prepare the filter and save it for the SW queue
@@ -655,8 +659,9 @@ void ieee80211_ht_agg_queue_remove(struct ieee80211_local *local,
 				   u8 requeue)
 {
 	struct ieee80211_hw *hw = &local->hw;
+	struct netdev_queue *txq = &local->mdev->tx_queue;
 	struct ieee80211_sched_data *q =
-		qdisc_priv(local->mdev->qdisc_sleeping);
+		qdisc_priv(txq->qdisc_sleeping);
 	int agg_queue = sta->tid_to_tx_q[tid];
 
 	/* return the qdisc to the pool */
@@ -671,7 +676,8 @@ void ieee80211_ht_agg_queue_remove(struct ieee80211_local *local,
 
 void ieee80211_requeue(struct ieee80211_local *local, int queue)
 {
-	struct Qdisc *root_qd = local->mdev->qdisc_sleeping;
+	struct netdev_queue *txq = &local->mdev->tx_queue;
+	struct Qdisc *root_qd = txq->qdisc_sleeping;
 	struct ieee80211_sched_data *q = qdisc_priv(root_qd);
 	struct Qdisc *qdisc = q->queues[queue];
 	struct sk_buff *skb = NULL;
diff --git a/net/sched/cls_api.c b/net/sched/cls_api.c
index e2389f161e46..b483bbea6118 100644
--- a/net/sched/cls_api.c
+++ b/net/sched/cls_api.c
@@ -166,7 +166,8 @@ replay:
 
 	/* Find qdisc */
 	if (!parent) {
-		q = dev->qdisc_sleeping;
+		struct netdev_queue *dev_queue = &dev->tx_queue;
+		q = dev_queue->qdisc_sleeping;
 		parent = q->handle;
 	} else {
 		q = qdisc_lookup(dev, TC_H_MAJ(t->tcm_parent));
@@ -390,6 +391,7 @@ static int tcf_node_dump(struct tcf_proto *tp, unsigned long n,
 static int tc_dump_tfilter(struct sk_buff *skb, struct netlink_callback *cb)
 {
 	struct net *net = sock_net(skb->sk);
+	struct netdev_queue *dev_queue;
 	int t;
 	int s_t;
 	struct net_device *dev;
@@ -408,8 +410,9 @@ static int tc_dump_tfilter(struct sk_buff *skb, struct netlink_callback *cb)
 	if ((dev = dev_get_by_index(&init_net, tcm->tcm_ifindex)) == NULL)
 		return skb->len;
 
+	dev_queue = &dev->tx_queue;
 	if (!tcm->tcm_parent)
-		q = dev->qdisc_sleeping;
+		q = dev_queue->qdisc_sleeping;
 	else
 		q = qdisc_lookup(dev, TC_H_MAJ(tcm->tcm_parent));
 	if (!q)
diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c
index 570cef2a9c5f..2313fa7c97be 100644
--- a/net/sched/sch_api.c
+++ b/net/sched/sch_api.c
@@ -185,9 +185,10 @@ EXPORT_SYMBOL(unregister_qdisc);
 
 struct Qdisc *qdisc_lookup(struct net_device *dev, u32 handle)
 {
+	struct netdev_queue *dev_queue = &dev->tx_queue;
 	struct Qdisc *q;
 
-	list_for_each_entry(q, &dev->qdisc_list, list) {
+	list_for_each_entry(q, &dev_queue->qdisc_list, list) {
 		if (q->handle == handle)
 			return q;
 	}
@@ -441,6 +442,7 @@ static u32 qdisc_alloc_handle(struct net_device *dev)
 static struct Qdisc *
 dev_graft_qdisc(struct net_device *dev, struct Qdisc *qdisc)
 {
+	struct netdev_queue *dev_queue;
 	struct Qdisc *oqdisc;
 
 	if (dev->flags & IFF_UP)
@@ -459,8 +461,8 @@ dev_graft_qdisc(struct net_device *dev, struct Qdisc *qdisc)
 		}
 
 	} else {
-
-		oqdisc = dev->qdisc_sleeping;
+		dev_queue = &dev->tx_queue;
+		oqdisc = dev_queue->qdisc_sleeping;
 
 		/* Prune old scheduler */
 		if (oqdisc && atomic_read(&oqdisc->refcnt) <= 1)
@@ -469,8 +471,8 @@ dev_graft_qdisc(struct net_device *dev, struct Qdisc *qdisc)
 		/* ... and graft new one */
 		if (qdisc == NULL)
 			qdisc = &noop_qdisc;
-		dev->qdisc_sleeping = qdisc;
-		dev->qdisc = &noop_qdisc;
+		dev_queue->qdisc_sleeping = qdisc;
+		dev_queue->qdisc = &noop_qdisc;
 	}
 
 	qdisc_unlock_tree(dev);
@@ -633,7 +635,7 @@ qdisc_create(struct net_device *dev, struct netdev_queue *dev_queue,
 			}
 		}
 		qdisc_lock_tree(dev);
-		list_add_tail(&sch->list, &dev->qdisc_list);
+		list_add_tail(&sch->list, &dev_queue->qdisc_list);
 		qdisc_unlock_tree(dev);
 
 		return sch;
@@ -740,7 +742,8 @@ static int tc_get_qdisc(struct sk_buff *skb, struct nlmsghdr *n, void *arg)
 				q = dev->qdisc_ingress;
 			}
 		} else {
-			q = dev->qdisc_sleeping;
+			struct netdev_queue *dev_queue = &dev->tx_queue;
+			q = dev_queue->qdisc_sleeping;
 		}
 		if (!q)
 			return -ENOENT;
@@ -814,7 +817,8 @@ replay:
 				q = dev->qdisc_ingress;
 			}
 		} else {
-			q = dev->qdisc_sleeping;
+			struct netdev_queue *dev_queue = &dev->tx_queue;
+			q = dev_queue->qdisc_sleeping;
 		}
 
 		/* It may be default qdisc, ignore it */
@@ -1015,12 +1019,14 @@ static int tc_dump_qdisc(struct sk_buff *skb, struct netlink_callback *cb)
 	read_lock(&dev_base_lock);
 	idx = 0;
 	for_each_netdev(&init_net, dev) {
+		struct netdev_queue *dev_queue;
 		if (idx < s_idx)
 			goto cont;
 		if (idx > s_idx)
 			s_q_idx = 0;
 		q_idx = 0;
-		list_for_each_entry(q, &dev->qdisc_list, list) {
+		dev_queue = &dev->tx_queue;
+		list_for_each_entry(q, &dev_queue->qdisc_list, list) {
 			if (q_idx < s_q_idx) {
 				q_idx++;
 				continue;
@@ -1054,6 +1060,7 @@ done:
 static int tc_ctl_tclass(struct sk_buff *skb, struct nlmsghdr *n, void *arg)
 {
 	struct net *net = sock_net(skb->sk);
+	struct netdev_queue *dev_queue;
 	struct tcmsg *tcm = NLMSG_DATA(n);
 	struct nlattr *tca[TCA_MAX + 1];
 	struct net_device *dev;
@@ -1091,6 +1098,7 @@ static int tc_ctl_tclass(struct sk_buff *skb, struct nlmsghdr *n, void *arg)
 
 	/* Step 1. Determine qdisc handle X:0 */
 
+	dev_queue = &dev->tx_queue;
 	if (pid != TC_H_ROOT) {
 		u32 qid1 = TC_H_MAJ(pid);
 
@@ -1101,7 +1109,7 @@ static int tc_ctl_tclass(struct sk_buff *skb, struct nlmsghdr *n, void *arg)
 		} else if (qid1) {
 			qid = qid1;
 		} else if (qid == 0)
-			qid = dev->qdisc_sleeping->handle;
+			qid = dev_queue->qdisc_sleeping->handle;
 
 		/* Now qid is genuine qdisc handle consistent
 		   both with parent and child.
@@ -1112,7 +1120,7 @@ static int tc_ctl_tclass(struct sk_buff *skb, struct nlmsghdr *n, void *arg)
 			pid = TC_H_MAKE(qid, pid);
 	} else {
 		if (qid == 0)
-			qid = dev->qdisc_sleeping->handle;
+			qid = dev_queue->qdisc_sleeping->handle;
 	}
 
 	/* OK. Locate qdisc */
@@ -1248,6 +1256,7 @@ static int qdisc_class_dump(struct Qdisc *q, unsigned long cl, struct qdisc_walk
 static int tc_dump_tclass(struct sk_buff *skb, struct netlink_callback *cb)
 {
 	struct net *net = sock_net(skb->sk);
+	struct netdev_queue *dev_queue;
 	int t;
 	int s_t;
 	struct net_device *dev;
@@ -1266,7 +1275,8 @@ static int tc_dump_tclass(struct sk_buff *skb, struct netlink_callback *cb)
 	s_t = cb->args[0];
 	t = 0;
 
-	list_for_each_entry(q, &dev->qdisc_list, list) {
+	dev_queue = &dev->tx_queue;
+	list_for_each_entry(q, &dev_queue->qdisc_list, list) {
 		if (t < s_t || !q->ops->cl_ops ||
 		    (tcm->tcm_parent &&
 		     TC_H_MAJ(tcm->tcm_parent) != q->handle)) {
diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c
index 804d44b00348..3223e5ba76aa 100644
--- a/net/sched/sch_generic.c
+++ b/net/sched/sch_generic.c
@@ -122,7 +122,7 @@ static inline int handle_dev_cpu_collision(struct sk_buff *skb,
  *
  * __LINK_STATE_QDISC_RUNNING guarantees only one CPU can process this
  * device at a time. queue->lock serializes queue accesses for
- * this device AND dev->qdisc pointer itself.
+ * this device AND txq->qdisc pointer itself.
  *
  *  netif_tx_lock serializes accesses to device driver.
  *
@@ -138,7 +138,8 @@ static inline int handle_dev_cpu_collision(struct sk_buff *skb,
  */
 static inline int qdisc_restart(struct net_device *dev)
 {
-	struct Qdisc *q = dev->qdisc;
+	struct netdev_queue *txq = &dev->tx_queue;
+	struct Qdisc *q = txq->qdisc;
 	struct sk_buff *skb;
 	int ret = NETDEV_TX_BUSY;
 
@@ -148,15 +149,15 @@ static inline int qdisc_restart(struct net_device *dev)
 
 
 	/* And release queue */
-	spin_unlock(&q->dev_queue->lock);
+	spin_unlock(&txq->lock);
 
 	HARD_TX_LOCK(dev, smp_processor_id());
 	if (!netif_subqueue_stopped(dev, skb))
 		ret = dev_hard_start_xmit(skb, dev);
 	HARD_TX_UNLOCK(dev);
 
-	spin_lock(&q->dev_queue->lock);
-	q = dev->qdisc;
+	spin_lock(&txq->lock);
+	q = txq->qdisc;
 
 	switch (ret) {
 	case NETDEV_TX_OK:
@@ -207,9 +208,10 @@ void __qdisc_run(struct net_device *dev)
 static void dev_watchdog(unsigned long arg)
 {
 	struct net_device *dev = (struct net_device *)arg;
+	struct netdev_queue *txq = &dev->tx_queue;
 
 	netif_tx_lock(dev);
-	if (dev->qdisc != &noop_qdisc) {
+	if (txq->qdisc != &noop_qdisc) {
 		if (netif_device_present(dev) &&
 		    netif_running(dev) &&
 		    netif_carrier_ok(dev)) {
@@ -539,53 +541,63 @@ EXPORT_SYMBOL(qdisc_destroy);
 
 void dev_activate(struct net_device *dev)
 {
+	struct netdev_queue *txq = &dev->tx_queue;
+
 	/* No queueing discipline is attached to device;
 	   create default one i.e. pfifo_fast for devices,
 	   which need queueing and noqueue_qdisc for
 	   virtual interfaces
 	 */
 
-	if (dev->qdisc_sleeping == &noop_qdisc) {
+	if (txq->qdisc_sleeping == &noop_qdisc) {
 		struct Qdisc *qdisc;
 		if (dev->tx_queue_len) {
-			qdisc = qdisc_create_dflt(dev, &dev->tx_queue,
+			qdisc = qdisc_create_dflt(dev, txq,
 						  &pfifo_fast_ops,
 						  TC_H_ROOT);
 			if (qdisc == NULL) {
 				printk(KERN_INFO "%s: activation failed\n", dev->name);
 				return;
 			}
-			list_add_tail(&qdisc->list, &dev->qdisc_list);
+			list_add_tail(&qdisc->list, &txq->qdisc_list);
 		} else {
 			qdisc =  &noqueue_qdisc;
 		}
-		dev->qdisc_sleeping = qdisc;
+		txq->qdisc_sleeping = qdisc;
 	}
 
 	if (!netif_carrier_ok(dev))
 		/* Delay activation until next carrier-on event */
 		return;
 
-	spin_lock_bh(&dev->tx_queue.lock);
-	rcu_assign_pointer(dev->qdisc, dev->qdisc_sleeping);
-	if (dev->qdisc != &noqueue_qdisc) {
+	spin_lock_bh(&txq->lock);
+	rcu_assign_pointer(txq->qdisc, txq->qdisc_sleeping);
+	if (txq->qdisc != &noqueue_qdisc) {
 		dev->trans_start = jiffies;
 		dev_watchdog_up(dev);
 	}
-	spin_unlock_bh(&dev->tx_queue.lock);
+	spin_unlock_bh(&txq->lock);
+}
+
+static void dev_deactivate_queue(struct net_device *dev,
+				 struct netdev_queue *dev_queue,
+				 struct Qdisc *qdisc_default)
+{
+	struct Qdisc *qdisc = dev_queue->qdisc;
+
+	if (qdisc) {
+		dev_queue->qdisc = qdisc_default;
+		qdisc_reset(qdisc);
+	}
 }
 
 void dev_deactivate(struct net_device *dev)
 {
-	struct Qdisc *qdisc;
 	struct sk_buff *skb;
 	int running;
 
 	spin_lock_bh(&dev->tx_queue.lock);
-	qdisc = dev->qdisc;
-	dev->qdisc = &noop_qdisc;
-
-	qdisc_reset(qdisc);
+	dev_deactivate_queue(dev, &dev->tx_queue, &noop_qdisc);
 
 	skb = dev->gso_skb;
 	dev->gso_skb = NULL;
@@ -622,32 +634,44 @@ void dev_deactivate(struct net_device *dev)
 	} while (WARN_ON_ONCE(running));
 }
 
+static void dev_init_scheduler_queue(struct net_device *dev,
+				     struct netdev_queue *dev_queue,
+				     struct Qdisc *qdisc)
+{
+	dev_queue->qdisc = qdisc;
+	dev_queue->qdisc_sleeping = qdisc;
+	INIT_LIST_HEAD(&dev_queue->qdisc_list);
+}
+
 void dev_init_scheduler(struct net_device *dev)
 {
 	qdisc_lock_tree(dev);
-	dev->qdisc = &noop_qdisc;
-	dev->qdisc_sleeping = &noop_qdisc;
-	INIT_LIST_HEAD(&dev->qdisc_list);
+	dev_init_scheduler_queue(dev, &dev->tx_queue, &noop_qdisc);
+	dev_init_scheduler_queue(dev, &dev->rx_queue, NULL);
 	qdisc_unlock_tree(dev);
 
 	setup_timer(&dev->watchdog_timer, dev_watchdog, (unsigned long)dev);
 }
 
-void dev_shutdown(struct net_device *dev)
+static void dev_shutdown_scheduler_queue(struct net_device *dev,
+					 struct netdev_queue *dev_queue,
+					 struct Qdisc *qdisc_default)
 {
-	struct Qdisc *qdisc;
+	struct Qdisc *qdisc = dev_queue->qdisc_sleeping;
+
+	if (qdisc) {
+		dev_queue->qdisc = qdisc_default;
+		dev_queue->qdisc_sleeping = qdisc_default;
 
-	qdisc_lock_tree(dev);
-	qdisc = dev->qdisc_sleeping;
-	dev->qdisc = &noop_qdisc;
-	dev->qdisc_sleeping = &noop_qdisc;
-	qdisc_destroy(qdisc);
-#if defined(CONFIG_NET_SCH_INGRESS) || defined(CONFIG_NET_SCH_INGRESS_MODULE)
-	if ((qdisc = dev->qdisc_ingress) != NULL) {
-		dev->qdisc_ingress = NULL;
 		qdisc_destroy(qdisc);
 	}
-#endif
+}
+
+void dev_shutdown(struct net_device *dev)
+{
+	qdisc_lock_tree(dev);
+	dev_shutdown_scheduler_queue(dev, &dev->tx_queue, &noop_qdisc);
+	dev_shutdown_scheduler_queue(dev, &dev->rx_queue, NULL);
 	BUG_TRAP(!timer_pending(&dev->watchdog_timer));
 	qdisc_unlock_tree(dev);
 }
diff --git a/net/sched/sch_netem.c b/net/sched/sch_netem.c
index 71b73c528f9b..4093f1eaaf60 100644
--- a/net/sched/sch_netem.c
+++ b/net/sched/sch_netem.c
@@ -180,7 +180,7 @@ static int netem_enqueue(struct sk_buff *skb, struct Qdisc *sch)
 	 * skb will be queued.
 	 */
 	if (count > 1 && (skb2 = skb_clone(skb, GFP_ATOMIC)) != NULL) {
-		struct Qdisc *rootq = qdisc_dev(sch)->qdisc;
+		struct Qdisc *rootq = qdisc_dev(sch)->tx_queue.qdisc;
 		u32 dupsave = q->duplicate; /* prevent duplicating a dup... */
 		q->duplicate = 0;
 
diff --git a/net/sched/sch_teql.c b/net/sched/sch_teql.c
index 4f3054e8e1ab..8ac05981be20 100644
--- a/net/sched/sch_teql.c
+++ b/net/sched/sch_teql.c
@@ -107,17 +107,19 @@ static struct sk_buff *
 teql_dequeue(struct Qdisc* sch)
 {
 	struct teql_sched_data *dat = qdisc_priv(sch);
+	struct netdev_queue *dat_queue;
 	struct sk_buff *skb;
 
 	skb = __skb_dequeue(&dat->q);
+	dat_queue = &dat->m->dev->tx_queue;
 	if (skb == NULL) {
-		struct net_device *m = qdisc_dev(dat->m->dev->qdisc);
+		struct net_device *m = qdisc_dev(dat_queue->qdisc);
 		if (m) {
 			dat->m->slaves = sch;
 			netif_wake_queue(m);
 		}
 	}
-	sch->q.qlen = dat->q.qlen + dat->m->dev->qdisc->q.qlen;
+	sch->q.qlen = dat->q.qlen + dat_queue->qdisc->q.qlen;
 	return skb;
 }
 
@@ -155,7 +157,7 @@ teql_destroy(struct Qdisc* sch)
 					if (q == master->slaves) {
 						master->slaves = NULL;
 						spin_lock_bh(&master->dev->tx_queue.lock);
-						qdisc_reset(master->dev->qdisc);
+						qdisc_reset(master->dev->tx_queue.qdisc);
 						spin_unlock_bh(&master->dev->tx_queue.lock);
 					}
 				}
@@ -216,7 +218,7 @@ static int teql_qdisc_init(struct Qdisc *sch, struct nlattr *opt)
 static int
 __teql_resolve(struct sk_buff *skb, struct sk_buff *skb_res, struct net_device *dev)
 {
-	struct teql_sched_data *q = qdisc_priv(dev->qdisc);
+	struct teql_sched_data *q = qdisc_priv(dev->tx_queue.qdisc);
 	struct neighbour *mn = skb->dst->neighbour;
 	struct neighbour *n = q->ncache;
 
@@ -252,7 +254,7 @@ __teql_resolve(struct sk_buff *skb, struct sk_buff *skb_res, struct net_device *
 static inline int teql_resolve(struct sk_buff *skb,
 			       struct sk_buff *skb_res, struct net_device *dev)
 {
-	if (dev->qdisc == &noop_qdisc)
+	if (dev->tx_queue.qdisc == &noop_qdisc)
 		return -ENODEV;
 
 	if (dev->header_ops == NULL ||
@@ -284,7 +286,7 @@ restart:
 	do {
 		struct net_device *slave = qdisc_dev(q);
 
-		if (slave->qdisc_sleeping != q)
+		if (slave->tx_queue.qdisc_sleeping != q)
 			continue;
 		if (netif_queue_stopped(slave) ||
 		    __netif_subqueue_stopped(slave, subq) ||
-- 
cgit v1.2.3


From 816f3258e70db38d6d92c8d871377179fd69160f Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Tue, 8 Jul 2008 22:49:00 -0700
Subject: netdev: Kill qdisc_ingress, use netdev->rx_queue.qdisc instead.

Now that our qdisc management is bi-directional, per-queue, and fully
orthogonal, there is no reason to have a special ingress qdisc pointer
in struct net_device.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h |  3 ---
 net/core/dev.c            |  4 ++--
 net/sched/sch_api.c       | 11 ++++++-----
 3 files changed, 8 insertions(+), 10 deletions(-)

(limited to 'include')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index df702a7b3db5..e7c49246fd88 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -634,9 +634,6 @@ struct net_device
 
 	struct netdev_queue	rx_queue;
 	struct netdev_queue	tx_queue ____cacheline_aligned_in_smp;
-
-	struct Qdisc		*qdisc_ingress;
-
 	unsigned long		tx_queue_len;	/* Max frames per queue allowed */
 
 	/* Partially transmitted GSO packet. */
diff --git a/net/core/dev.c b/net/core/dev.c
index ce79c28d739d..ab760a954d99 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -2033,7 +2033,7 @@ static int ing_filter(struct sk_buff *skb)
 	rxq = &dev->rx_queue;
 
 	spin_lock(&rxq->lock);
-	if ((q = dev->qdisc_ingress) != NULL)
+	if ((q = rxq->qdisc) != NULL)
 		result = q->enqueue(skb, q);
 	spin_unlock(&rxq->lock);
 
@@ -2044,7 +2044,7 @@ static inline struct sk_buff *handle_ing(struct sk_buff *skb,
 					 struct packet_type **pt_prev,
 					 int *ret, struct net_device *orig_dev)
 {
-	if (!skb->dev->qdisc_ingress)
+	if (!skb->dev->rx_queue.qdisc)
 		goto out;
 
 	if (*pt_prev) {
diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c
index 2313fa7c97be..4003c280b69f 100644
--- a/net/sched/sch_api.c
+++ b/net/sched/sch_api.c
@@ -450,14 +450,15 @@ dev_graft_qdisc(struct net_device *dev, struct Qdisc *qdisc)
 
 	qdisc_lock_tree(dev);
 	if (qdisc && qdisc->flags&TCQ_F_INGRESS) {
-		oqdisc = dev->qdisc_ingress;
+		dev_queue = &dev->rx_queue;
+		oqdisc = dev_queue->qdisc;
 		/* Prune old scheduler */
 		if (oqdisc && atomic_read(&oqdisc->refcnt) <= 1) {
 			/* delete */
 			qdisc_reset(oqdisc);
-			dev->qdisc_ingress = NULL;
+			dev_queue->qdisc = NULL;
 		} else {  /* new */
-			dev->qdisc_ingress = qdisc;
+			dev_queue->qdisc = qdisc;
 		}
 
 	} else {
@@ -739,7 +740,7 @@ static int tc_get_qdisc(struct sk_buff *skb, struct nlmsghdr *n, void *arg)
 					return -ENOENT;
 				q = qdisc_leaf(p, clid);
 			} else { /* ingress */
-				q = dev->qdisc_ingress;
+				q = dev->rx_queue.qdisc;
 			}
 		} else {
 			struct netdev_queue *dev_queue = &dev->tx_queue;
@@ -814,7 +815,7 @@ replay:
 					return -ENOENT;
 				q = qdisc_leaf(p, clid);
 			} else { /*ingress */
-				q = dev->qdisc_ingress;
+				q = dev->rx_queue.qdisc;
 			}
 		} else {
 			struct netdev_queue *dev_queue = &dev->tx_queue;
-- 
cgit v1.2.3


From 68dfb42798e1eb2d42acbf872925cc75f1487d9b Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Tue, 8 Jul 2008 22:57:31 -0700
Subject: pkt_sched: Kill stats_lock member of struct Qdisc.

It is always equal to qdisc->dev_queue->lock

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/sch_generic.h | 1 -
 net/sched/sch_api.c       | 9 ++++-----
 net/sched/sch_generic.c   | 1 -
 3 files changed, 4 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h
index 66ec36d8ac97..ea71705e9c77 100644
--- a/include/net/sch_generic.h
+++ b/include/net/sch_generic.h
@@ -43,7 +43,6 @@ struct Qdisc
 	struct gnet_stats_basic	bstats;
 	struct gnet_stats_queue	qstats;
 	struct gnet_stats_rate_est	rate_est;
-	spinlock_t		*stats_lock;
 	struct rcu_head 	q_rcu;
 	int			(*reshape_fail)(struct sk_buff *skb,
 					struct Qdisc *q);
diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c
index 4003c280b69f..e73bd68aa7ae 100644
--- a/net/sched/sch_api.c
+++ b/net/sched/sch_api.c
@@ -604,7 +604,6 @@ qdisc_create(struct net_device *dev, struct netdev_queue *dev_queue,
 
 	sch->parent = parent;
 
-	sch->stats_lock = &dev_queue->lock;
 	if (handle == TC_H_INGRESS) {
 		sch->flags |= TCQ_F_INGRESS;
 		handle = TC_H_MAKE(TC_H_INGRESS, 0);
@@ -622,7 +621,7 @@ qdisc_create(struct net_device *dev, struct netdev_queue *dev_queue,
 	if (!ops->init || (err = ops->init(sch, tca[TCA_OPTIONS])) == 0) {
 		if (tca[TCA_RATE]) {
 			err = gen_new_estimator(&sch->bstats, &sch->rate_est,
-						sch->stats_lock,
+						&sch->dev_queue->lock,
 						tca[TCA_RATE]);
 			if (err) {
 				/*
@@ -664,7 +663,7 @@ static int qdisc_change(struct Qdisc *sch, struct nlattr **tca)
 	}
 	if (tca[TCA_RATE])
 		gen_replace_estimator(&sch->bstats, &sch->rate_est,
-			sch->stats_lock, tca[TCA_RATE]);
+				      &sch->dev_queue->lock, tca[TCA_RATE]);
 	return 0;
 }
 
@@ -954,7 +953,7 @@ static int tc_fill_qdisc(struct sk_buff *skb, struct Qdisc *q, u32 clid,
 	q->qstats.qlen = q->q.qlen;
 
 	if (gnet_stats_start_copy_compat(skb, TCA_STATS2, TCA_STATS,
-			TCA_XSTATS, q->stats_lock, &d) < 0)
+					 TCA_XSTATS, &q->dev_queue->lock, &d) < 0)
 		goto nla_put_failure;
 
 	if (q->ops->dump_stats && q->ops->dump_stats(q, &d) < 0)
@@ -1203,7 +1202,7 @@ static int tc_fill_tclass(struct sk_buff *skb, struct Qdisc *q,
 		goto nla_put_failure;
 
 	if (gnet_stats_start_copy_compat(skb, TCA_STATS2, TCA_STATS,
-			TCA_XSTATS, q->stats_lock, &d) < 0)
+					 TCA_XSTATS, &q->dev_queue->lock, &d) < 0)
 		goto nla_put_failure;
 
 	if (cl_ops->dump_stats && cl_ops->dump_stats(q, cl, &d) < 0)
diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c
index 3223e5ba76aa..dda78ee314ec 100644
--- a/net/sched/sch_generic.c
+++ b/net/sched/sch_generic.c
@@ -484,7 +484,6 @@ struct Qdisc * qdisc_create_dflt(struct net_device *dev,
 	sch = qdisc_alloc(dev_queue, ops);
 	if (IS_ERR(sch))
 		goto errout;
-	sch->stats_lock = &dev_queue->lock;
 	sch->parent = parentid;
 
 	if (!ops->init || ops->init(sch, NULL) == 0)
-- 
cgit v1.2.3


From ee609cb36220d18c0cf476b066a5ab7e6f6d3a69 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Tue, 8 Jul 2008 22:58:37 -0700
Subject: netdev: Move next_sched into struct netdev_queue.

We schedule queues, not the device, for output queue processing in BH.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h |  5 ++---
 net/core/dev.c            | 15 +++++++--------
 2 files changed, 9 insertions(+), 11 deletions(-)

(limited to 'include')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index e7c49246fd88..1379c822e51d 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -454,6 +454,7 @@ struct netdev_queue {
 	struct Qdisc		*qdisc;
 	struct Qdisc		*qdisc_sleeping;
 	struct list_head	qdisc_list;
+	struct netdev_queue	*next_sched;
 };
 
 /*
@@ -545,8 +546,6 @@ struct net_device
 #define NETIF_F_V6_CSUM		(NETIF_F_GEN_CSUM | NETIF_F_IPV6_CSUM)
 #define NETIF_F_ALL_CSUM	(NETIF_F_V4_CSUM | NETIF_F_V6_CSUM)
 
-	struct net_device	*next_sched;
-
 	/* Interface index. Unique device identifier	*/
 	int			ifindex;
 	int			iflink;
@@ -940,7 +939,7 @@ static inline int unregister_gifconf(unsigned int family)
  */
 struct softnet_data
 {
-	struct net_device	*output_queue;
+	struct netdev_queue	*output_queue;
 	struct sk_buff_head	input_pkt_queue;
 	struct list_head	poll_list;
 	struct sk_buff		*completion_queue;
diff --git a/net/core/dev.c b/net/core/dev.c
index ab760a954d99..d6b8d3c3e6ec 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -1323,13 +1323,14 @@ static void dev_queue_xmit_nit(struct sk_buff *skb, struct net_device *dev)
 void __netif_schedule(struct net_device *dev)
 {
 	if (!test_and_set_bit(__LINK_STATE_SCHED, &dev->state)) {
+		struct netdev_queue *txq = &dev->tx_queue;
 		unsigned long flags;
 		struct softnet_data *sd;
 
 		local_irq_save(flags);
 		sd = &__get_cpu_var(softnet_data);
-		dev->next_sched = sd->output_queue;
-		sd->output_queue = dev;
+		txq->next_sched = sd->output_queue;
+		sd->output_queue = txq;
 		raise_softirq_irqoff(NET_TX_SOFTIRQ);
 		local_irq_restore(flags);
 	}
@@ -1912,7 +1913,7 @@ static void net_tx_action(struct softirq_action *h)
 	}
 
 	if (sd->output_queue) {
-		struct net_device *head;
+		struct netdev_queue *head;
 
 		local_irq_disable();
 		head = sd->output_queue;
@@ -1920,12 +1921,10 @@ static void net_tx_action(struct softirq_action *h)
 		local_irq_enable();
 
 		while (head) {
-			struct net_device *dev = head;
-			struct netdev_queue *txq;
+			struct netdev_queue *txq = head;
+			struct net_device *dev = txq->dev;
 			head = head->next_sched;
 
-			txq = &dev->tx_queue;
-
 			smp_mb__before_clear_bit();
 			clear_bit(__LINK_STATE_SCHED, &dev->state);
 
@@ -4346,7 +4345,7 @@ static int dev_cpu_callback(struct notifier_block *nfb,
 			    void *ocpu)
 {
 	struct sk_buff **list_skb;
-	struct net_device **list_net;
+	struct netdev_queue **list_net;
 	struct sk_buff *skb;
 	unsigned int cpu, oldcpu = (unsigned long)ocpu;
 	struct softnet_data *sd, *oldsd;
-- 
cgit v1.2.3


From 5aa709954a32a646c0cf14fdf7c8834db4ff1c03 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Tue, 8 Jul 2008 22:59:10 -0700
Subject: pkt_sched: Add qdisc_reset_all_tx().

Isolate callers that want to simply reset all the TX qdiscs from the
details of TX queues.

Use this in the ISDN code.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/isdn/i4l/isdn_net.c | 2 +-
 include/net/sch_generic.h   | 6 ++++++
 2 files changed, 7 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/drivers/isdn/i4l/isdn_net.c b/drivers/isdn/i4l/isdn_net.c
index 457bbd119f9b..bb904a0a98bd 100644
--- a/drivers/isdn/i4l/isdn_net.c
+++ b/drivers/isdn/i4l/isdn_net.c
@@ -287,7 +287,7 @@ isdn_net_unbind_channel(isdn_net_local * lp)
 		   BEWARE! This chunk of code cannot be called from hardware
 		   interrupt handler. I hope it is true. --ANK
 		 */
-		qdisc_reset(lp->netdev->dev->tx_queue.qdisc);
+		qdisc_reset_all_tx(lp->netdev->dev);
 	}
 	lp->dialstate = 0;
 	dev->rx_netdev[isdn_dc2minor(lp->isdn_device, lp->isdn_channel)] = NULL;
diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h
index ea71705e9c77..ea33176616ff 100644
--- a/include/net/sch_generic.h
+++ b/include/net/sch_generic.h
@@ -227,6 +227,12 @@ extern struct Qdisc *qdisc_create_dflt(struct net_device *dev,
 extern void tcf_destroy(struct tcf_proto *tp);
 extern void tcf_destroy_chain(struct tcf_proto **fl);
 
+/* Reset all TX qdiscs of a device.  */
+static inline void qdisc_reset_all_tx(struct net_device *dev)
+{
+	qdisc_reset(dev->tx_queue.qdisc);
+}
+
 static inline int __qdisc_enqueue_tail(struct sk_buff *skb, struct Qdisc *sch,
 				       struct sk_buff_head *list)
 {
-- 
cgit v1.2.3


From 3e745dd6950d72d19e4cb30dc44307e8d6671864 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Tue, 8 Jul 2008 23:00:25 -0700
Subject: pkt_sched: Add qdisc_all_tx_empty()

This is a helper function, currently used by IRDA.

This is being added so that we can contain and isolate as many
explicit ->tx_queue references in the tree as possible.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/irda/irda_device.h | 2 +-
 include/net/sch_generic.h      | 9 +++++++++
 2 files changed, 10 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/net/irda/irda_device.h b/include/net/irda/irda_device.h
index 16fbf672e0b2..3025ae17ddbe 100644
--- a/include/net/irda/irda_device.h
+++ b/include/net/irda/irda_device.h
@@ -223,7 +223,7 @@ int  irda_device_is_receiving(struct net_device *dev);
 /* Interface for internal use */
 static inline int irda_device_txqueue_empty(const struct net_device *dev)
 {
-	return skb_queue_empty(&dev->tx_queue.qdisc->q);
+	return qdisc_all_tx_empty(dev);
 }
 int  irda_device_set_raw_mode(struct net_device* self, int status);
 struct net_device *alloc_irdadev(int sizeof_priv);
diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h
index ea33176616ff..8cfdaebbbab3 100644
--- a/include/net/sch_generic.h
+++ b/include/net/sch_generic.h
@@ -233,6 +233,15 @@ static inline void qdisc_reset_all_tx(struct net_device *dev)
 	qdisc_reset(dev->tx_queue.qdisc);
 }
 
+/* Are all TX queues of the device empty?  */
+static inline bool qdisc_all_tx_empty(const struct net_device *dev)
+{
+	const struct netdev_queue *txq = &dev->tx_queue;
+	const struct Qdisc *q = txq->qdisc;
+
+	return (q->q.qlen == 0);
+}
+
 static inline int __qdisc_enqueue_tail(struct sk_buff *skb, struct Qdisc *sch,
 				       struct sk_buff_head *list)
 {
-- 
cgit v1.2.3


From 6fa9864b53f0680e432a2c431c2cf2055daa3a88 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Tue, 8 Jul 2008 23:01:06 -0700
Subject: net: Clean up explicit ->tx_queue references in link watch.

First, we add a qdisc_tx_changing() helper which returns true if the
qdisc attachment is in transition.

Second, we remove an assertion warning which is of limited value and
is hard to express precisely in a multiqueue environment.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/sch_generic.h |  8 ++++++++
 net/core/link_watch.c     | 15 +++++----------
 2 files changed, 13 insertions(+), 10 deletions(-)

(limited to 'include')

diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h
index 8cfdaebbbab3..bf8f7264a778 100644
--- a/include/net/sch_generic.h
+++ b/include/net/sch_generic.h
@@ -242,6 +242,14 @@ static inline bool qdisc_all_tx_empty(const struct net_device *dev)
 	return (q->q.qlen == 0);
 }
 
+/* Are any of the TX qdiscs changing?  */
+static inline bool qdisc_tx_changing(struct net_device *dev)
+{
+	struct netdev_queue *txq = &dev->tx_queue;
+
+	return (txq->qdisc != txq->qdisc_sleeping);
+}
+
 static inline int __qdisc_enqueue_tail(struct sk_buff *skb, struct Qdisc *sch,
 				       struct sk_buff_head *list)
 {
diff --git a/net/core/link_watch.c b/net/core/link_watch.c
index 50218218445b..bf8f7af699d7 100644
--- a/net/core/link_watch.c
+++ b/net/core/link_watch.c
@@ -77,12 +77,10 @@ static void rfc2863_policy(struct net_device *dev)
 }
 
 
-static int linkwatch_urgent_event(struct net_device *dev)
+static bool linkwatch_urgent_event(struct net_device *dev)
 {
-	struct netdev_queue *txq = &dev->tx_queue;
-
 	return netif_running(dev) && netif_carrier_ok(dev) &&
-	       txq->qdisc != txq->qdisc_sleeping;
+		qdisc_tx_changing(dev);
 }
 
 
@@ -182,12 +180,9 @@ static void __linkwatch_run_queue(int urgent_only)
 
 		rfc2863_policy(dev);
 		if (dev->flags & IFF_UP) {
-			if (netif_carrier_ok(dev)) {
-				struct netdev_queue *txq = &dev->tx_queue;
-
-				WARN_ON(txq->qdisc_sleeping == &noop_qdisc);
+			if (netif_carrier_ok(dev))
 				dev_activate(dev);
-			} else
+			else
 				dev_deactivate(dev);
 
 			netdev_state_change(dev);
@@ -218,7 +213,7 @@ static void linkwatch_event(struct work_struct *dummy)
 
 void linkwatch_fire_event(struct net_device *dev)
 {
-	int urgent = linkwatch_urgent_event(dev);
+	bool urgent = linkwatch_urgent_event(dev);
 
 	if (!test_and_set_bit(__LINK_STATE_LINKWATCH_PENDING, &dev->state)) {
 		dev_hold(dev);
-- 
cgit v1.2.3


From 052979499c767268b912d25031ae524c451679d0 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Tue, 8 Jul 2008 23:01:27 -0700
Subject: pkt_sched: Add qdisc_tx_is_noop() helper and use in IPV6.

This indicates if the NOOP scheduler is what is active for TX on a
given device.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/sch_generic.h | 8 ++++++++
 net/ipv6/addrconf.c       | 5 ++---
 2 files changed, 10 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h
index bf8f7264a778..5ba66b555578 100644
--- a/include/net/sch_generic.h
+++ b/include/net/sch_generic.h
@@ -250,6 +250,14 @@ static inline bool qdisc_tx_changing(struct net_device *dev)
 	return (txq->qdisc != txq->qdisc_sleeping);
 }
 
+/* Is the device using the noop qdisc?  */
+static inline bool qdisc_tx_is_noop(const struct net_device *dev)
+{
+	const struct netdev_queue *txq = &dev->tx_queue;
+
+	return (txq->qdisc == &noop_qdisc);
+}
+
 static inline int __qdisc_enqueue_tail(struct sk_buff *skb, struct Qdisc *sch,
 				       struct sk_buff_head *list)
 {
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index 5c84c798331d..30184e0dd74c 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -229,10 +229,9 @@ const struct in6_addr in6addr_linklocal_allnodes = IN6ADDR_LINKLOCAL_ALLNODES_IN
 const struct in6_addr in6addr_linklocal_allrouters = IN6ADDR_LINKLOCAL_ALLROUTERS_INIT;
 
 /* Check if a valid qdisc is available */
-static inline int addrconf_qdisc_ok(struct net_device *dev)
+static inline bool addrconf_qdisc_ok(const struct net_device *dev)
 {
-	struct netdev_queue *txq = &dev->tx_queue;
-	return (txq->qdisc != &noop_qdisc);
+	return !qdisc_tx_is_noop(dev);
 }
 
 /* Check if a route is valid prefix route */
-- 
cgit v1.2.3


From 970565bbad0c7b98db0d14131a69e5a0f4445d49 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Tue, 8 Jul 2008 23:10:33 -0700
Subject: netdev: Move gso_skb into netdev_queue.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h |  4 +---
 net/sched/sch_generic.c   | 42 +++++++++++++++++++++++-------------------
 2 files changed, 24 insertions(+), 22 deletions(-)

(limited to 'include')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 1379c822e51d..aae6c6d153f2 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -452,6 +452,7 @@ struct netdev_queue {
 	spinlock_t		lock;
 	struct net_device	*dev;
 	struct Qdisc		*qdisc;
+	struct sk_buff		*gso_skb;
 	struct Qdisc		*qdisc_sleeping;
 	struct list_head	qdisc_list;
 	struct netdev_queue	*next_sched;
@@ -635,9 +636,6 @@ struct net_device
 	struct netdev_queue	tx_queue ____cacheline_aligned_in_smp;
 	unsigned long		tx_queue_len;	/* Max frames per queue allowed */
 
-	/* Partially transmitted GSO packet. */
-	struct sk_buff		*gso_skb;
-
 /*
  * One part is mostly used on xmit path (device)
  */
diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c
index dda78ee314ec..8247a406a401 100644
--- a/net/sched/sch_generic.c
+++ b/net/sched/sch_generic.c
@@ -63,10 +63,11 @@ static inline int qdisc_qlen(struct Qdisc *q)
 }
 
 static inline int dev_requeue_skb(struct sk_buff *skb, struct net_device *dev,
+				  struct netdev_queue *dev_queue,
 				  struct Qdisc *q)
 {
 	if (unlikely(skb->next))
-		dev->gso_skb = skb;
+		dev_queue->gso_skb = skb;
 	else
 		q->ops->requeue(skb, q);
 
@@ -75,12 +76,13 @@ static inline int dev_requeue_skb(struct sk_buff *skb, struct net_device *dev,
 }
 
 static inline struct sk_buff *dev_dequeue_skb(struct net_device *dev,
+					      struct netdev_queue *dev_queue,
 					      struct Qdisc *q)
 {
 	struct sk_buff *skb;
 
-	if ((skb = dev->gso_skb))
-		dev->gso_skb = NULL;
+	if ((skb = dev_queue->gso_skb))
+		dev_queue->gso_skb = NULL;
 	else
 		skb = q->dequeue(q);
 
@@ -89,6 +91,7 @@ static inline struct sk_buff *dev_dequeue_skb(struct net_device *dev,
 
 static inline int handle_dev_cpu_collision(struct sk_buff *skb,
 					   struct net_device *dev,
+					   struct netdev_queue *dev_queue,
 					   struct Qdisc *q)
 {
 	int ret;
@@ -111,7 +114,7 @@ static inline int handle_dev_cpu_collision(struct sk_buff *skb,
 		 * some time.
 		 */
 		__get_cpu_var(netdev_rx_stat).cpu_collision++;
-		ret = dev_requeue_skb(skb, dev, q);
+		ret = dev_requeue_skb(skb, dev, dev_queue, q);
 	}
 
 	return ret;
@@ -144,7 +147,7 @@ static inline int qdisc_restart(struct net_device *dev)
 	int ret = NETDEV_TX_BUSY;
 
 	/* Dequeue packet */
-	if (unlikely((skb = dev_dequeue_skb(dev, q)) == NULL))
+	if (unlikely((skb = dev_dequeue_skb(dev, txq, q)) == NULL))
 		return 0;
 
 
@@ -167,7 +170,7 @@ static inline int qdisc_restart(struct net_device *dev)
 
 	case NETDEV_TX_LOCKED:
 		/* Driver try lock failed */
-		ret = handle_dev_cpu_collision(skb, dev, q);
+		ret = handle_dev_cpu_collision(skb, dev, txq, q);
 		break;
 
 	default:
@@ -176,7 +179,7 @@ static inline int qdisc_restart(struct net_device *dev)
 			printk(KERN_WARNING "BUG %s code %d qlen %d\n",
 			       dev->name, ret, q->q.qlen);
 
-		ret = dev_requeue_skb(skb, dev, q);
+		ret = dev_requeue_skb(skb, dev, txq, q);
 		break;
 	}
 
@@ -578,31 +581,32 @@ void dev_activate(struct net_device *dev)
 	spin_unlock_bh(&txq->lock);
 }
 
-static void dev_deactivate_queue(struct net_device *dev,
-				 struct netdev_queue *dev_queue,
+static void dev_deactivate_queue(struct netdev_queue *dev_queue,
 				 struct Qdisc *qdisc_default)
 {
-	struct Qdisc *qdisc = dev_queue->qdisc;
+	struct Qdisc *qdisc;
+	struct sk_buff *skb;
+
+	spin_lock_bh(&dev_queue->lock);
 
+	qdisc = dev_queue->qdisc;
 	if (qdisc) {
 		dev_queue->qdisc = qdisc_default;
 		qdisc_reset(qdisc);
 	}
+	skb = dev_queue->gso_skb;
+	dev_queue->gso_skb = NULL;
+
+	spin_unlock_bh(&dev_queue->lock);
+
+	kfree_skb(skb);
 }
 
 void dev_deactivate(struct net_device *dev)
 {
-	struct sk_buff *skb;
 	int running;
 
-	spin_lock_bh(&dev->tx_queue.lock);
-	dev_deactivate_queue(dev, &dev->tx_queue, &noop_qdisc);
-
-	skb = dev->gso_skb;
-	dev->gso_skb = NULL;
-	spin_unlock_bh(&dev->tx_queue.lock);
-
-	kfree_skb(skb);
+	dev_deactivate_queue(&dev->tx_queue, &noop_qdisc);
 
 	dev_watchdog_down(dev);
 
-- 
cgit v1.2.3


From 86d804e10a37cd86f16bf72386c37e843a98a74b Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Tue, 8 Jul 2008 23:11:25 -0700
Subject: netdev: Make netif_schedule() routines work with netdev_queue
 objects.

Only plain netif_schedule() remains taking a net_device, mostly as a
compatability item while we transition the rest of these interfaces.

Everything else calls netif_schedule_queue() or __netif_schedule(),
both of which take a netdev_queue pointer.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h | 17 ++++++++++++-----
 net/core/dev.c            |  9 +++++----
 net/mac80211/main.c       |  4 ++--
 net/sched/sch_api.c       |  4 ++--
 net/sched/sch_cbq.c       |  2 +-
 net/sched/sch_generic.c   | 10 +++++-----
 6 files changed, 27 insertions(+), 19 deletions(-)

(limited to 'include')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index aae6c6d153f2..28aa8e77cee9 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -952,12 +952,19 @@ DECLARE_PER_CPU(struct softnet_data,softnet_data);
 
 #define HAVE_NETIF_QUEUE
 
-extern void __netif_schedule(struct net_device *dev);
+extern void __netif_schedule(struct netdev_queue *txq);
 
-static inline void netif_schedule(struct net_device *dev)
+static inline void netif_schedule_queue(struct netdev_queue *txq)
 {
+	struct net_device *dev = txq->dev;
+
 	if (!test_bit(__LINK_STATE_XOFF, &dev->state))
-		__netif_schedule(dev);
+		__netif_schedule(txq);
+}
+
+static inline void netif_schedule(struct net_device *dev)
+{
+	netif_schedule_queue(&dev->tx_queue);
 }
 
 /**
@@ -987,7 +994,7 @@ static inline void netif_wake_queue(struct net_device *dev)
 	}
 #endif
 	if (test_and_clear_bit(__LINK_STATE_XOFF, &dev->state))
-		__netif_schedule(dev);
+		__netif_schedule(&dev->tx_queue);
 }
 
 /**
@@ -1103,7 +1110,7 @@ static inline void netif_wake_subqueue(struct net_device *dev, u16 queue_index)
 #endif
 	if (test_and_clear_bit(__LINK_STATE_XOFF,
 			       &dev->egress_subqueue[queue_index].state))
-		__netif_schedule(dev);
+		__netif_schedule(&dev->tx_queue);
 #endif
 }
 
diff --git a/net/core/dev.c b/net/core/dev.c
index d6b8d3c3e6ec..0dc888ad4217 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -1320,12 +1320,13 @@ static void dev_queue_xmit_nit(struct sk_buff *skb, struct net_device *dev)
 }
 
 
-void __netif_schedule(struct net_device *dev)
+void __netif_schedule(struct netdev_queue *txq)
 {
+	struct net_device *dev = txq->dev;
+
 	if (!test_and_set_bit(__LINK_STATE_SCHED, &dev->state)) {
-		struct netdev_queue *txq = &dev->tx_queue;
-		unsigned long flags;
 		struct softnet_data *sd;
+		unsigned long flags;
 
 		local_irq_save(flags);
 		sd = &__get_cpu_var(softnet_data);
@@ -1932,7 +1933,7 @@ static void net_tx_action(struct softirq_action *h)
 				qdisc_run(dev);
 				spin_unlock(&txq->lock);
 			} else {
-				netif_schedule(dev);
+				netif_schedule_queue(txq);
 			}
 		}
 	}
diff --git a/net/mac80211/main.c b/net/mac80211/main.c
index 2968baa66b91..1c4d3ba6b878 100644
--- a/net/mac80211/main.c
+++ b/net/mac80211/main.c
@@ -885,10 +885,10 @@ void ieee80211_stop_tx_ba_cb(struct ieee80211_hw *hw, u8 *ra, u8 tid)
 	spin_unlock_bh(&txq->lock);
 
 	/* we just requeued the all the frames that were in the removed
-	 * queue, and since we might miss a softirq we do netif_schedule.
+	 * queue, and since we might miss a softirq we do netif_schedule_queue.
 	 * ieee80211_wake_queue is not used here as this queue is not
 	 * necessarily stopped */
-	netif_schedule(local->mdev);
+	netif_schedule_queue(txq);
 	spin_lock_bh(&sta->lock);
 	*state = HT_AGG_STATE_IDLE;
 	sta->ampdu_mlme.addba_req_num[tid] = 0;
diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c
index e73bd68aa7ae..95873f8dd37c 100644
--- a/net/sched/sch_api.c
+++ b/net/sched/sch_api.c
@@ -282,11 +282,11 @@ static enum hrtimer_restart qdisc_watchdog(struct hrtimer *timer)
 {
 	struct qdisc_watchdog *wd = container_of(timer, struct qdisc_watchdog,
 						 timer);
-	struct net_device *dev = qdisc_dev(wd->qdisc);
+	struct netdev_queue *txq = wd->qdisc->dev_queue;
 
 	wd->qdisc->flags &= ~TCQ_F_THROTTLED;
 	smp_wmb();
-	netif_schedule(dev);
+	netif_schedule_queue(txq);
 
 	return HRTIMER_NORESTART;
 }
diff --git a/net/sched/sch_cbq.c b/net/sched/sch_cbq.c
index 99ce3da2b0a4..4efc836cbf38 100644
--- a/net/sched/sch_cbq.c
+++ b/net/sched/sch_cbq.c
@@ -650,7 +650,7 @@ static enum hrtimer_restart cbq_undelay(struct hrtimer *timer)
 	}
 
 	sch->flags &= ~TCQ_F_THROTTLED;
-	netif_schedule(qdisc_dev(sch));
+	netif_schedule_queue(sch->dev_queue);
 	return HRTIMER_NORESTART;
 }
 
diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c
index 8247a406a401..407dfdb142a4 100644
--- a/net/sched/sch_generic.c
+++ b/net/sched/sch_generic.c
@@ -62,7 +62,7 @@ static inline int qdisc_qlen(struct Qdisc *q)
 	return q->q.qlen;
 }
 
-static inline int dev_requeue_skb(struct sk_buff *skb, struct net_device *dev,
+static inline int dev_requeue_skb(struct sk_buff *skb,
 				  struct netdev_queue *dev_queue,
 				  struct Qdisc *q)
 {
@@ -71,7 +71,7 @@ static inline int dev_requeue_skb(struct sk_buff *skb, struct net_device *dev,
 	else
 		q->ops->requeue(skb, q);
 
-	netif_schedule(dev);
+	netif_schedule_queue(dev_queue);
 	return 0;
 }
 
@@ -114,7 +114,7 @@ static inline int handle_dev_cpu_collision(struct sk_buff *skb,
 		 * some time.
 		 */
 		__get_cpu_var(netdev_rx_stat).cpu_collision++;
-		ret = dev_requeue_skb(skb, dev, dev_queue, q);
+		ret = dev_requeue_skb(skb, dev_queue, q);
 	}
 
 	return ret;
@@ -179,7 +179,7 @@ static inline int qdisc_restart(struct net_device *dev)
 			printk(KERN_WARNING "BUG %s code %d qlen %d\n",
 			       dev->name, ret, q->q.qlen);
 
-		ret = dev_requeue_skb(skb, dev, txq, q);
+		ret = dev_requeue_skb(skb, txq, q);
 		break;
 	}
 
@@ -200,7 +200,7 @@ void __qdisc_run(struct net_device *dev)
 		 * 2. we've been doing it for too long.
 		 */
 		if (need_resched() || jiffies != start_time) {
-			netif_schedule(dev);
+			netif_schedule_queue(&dev->tx_queue);
 			break;
 		}
 	}
-- 
cgit v1.2.3


From eb6aafe3f843cb0e939546c03540a3b4911b6964 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Tue, 8 Jul 2008 23:12:38 -0700
Subject: pkt_sched: Make qdisc_run take a netdev_queue.

This allows us to use this calling convention all the way down into
qdisc_restart().

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/pkt_sched.h |  8 +++++---
 net/core/dev.c          |  4 ++--
 net/sched/sch_generic.c | 26 ++++++++++++++------------
 3 files changed, 21 insertions(+), 17 deletions(-)

(limited to 'include')

diff --git a/include/net/pkt_sched.h b/include/net/pkt_sched.h
index 8e3a0c4e9d97..2311d242bb35 100644
--- a/include/net/pkt_sched.h
+++ b/include/net/pkt_sched.h
@@ -84,13 +84,15 @@ extern struct qdisc_rate_table *qdisc_get_rtab(struct tc_ratespec *r,
 		struct nlattr *tab);
 extern void qdisc_put_rtab(struct qdisc_rate_table *tab);
 
-extern void __qdisc_run(struct net_device *dev);
+extern void __qdisc_run(struct netdev_queue *txq);
 
-static inline void qdisc_run(struct net_device *dev)
+static inline void qdisc_run(struct netdev_queue *txq)
 {
+	struct net_device *dev = txq->dev;
+
 	if (!netif_queue_stopped(dev) &&
 	    !test_and_set_bit(__LINK_STATE_QDISC_RUNNING, &dev->state))
-		__qdisc_run(dev);
+		__qdisc_run(txq);
 }
 
 extern int tc_classify_compat(struct sk_buff *skb, struct tcf_proto *tp,
diff --git a/net/core/dev.c b/net/core/dev.c
index 0dc888ad4217..0218b0b9be80 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -1734,7 +1734,7 @@ gso:
 			/* reset queue_mapping to zero */
 			skb_set_queue_mapping(skb, 0);
 			rc = q->enqueue(skb, q);
-			qdisc_run(dev);
+			qdisc_run(txq);
 			spin_unlock(&txq->lock);
 
 			rc = rc == NET_XMIT_BYPASS ? NET_XMIT_SUCCESS : rc;
@@ -1930,7 +1930,7 @@ static void net_tx_action(struct softirq_action *h)
 			clear_bit(__LINK_STATE_SCHED, &dev->state);
 
 			if (spin_trylock(&txq->lock)) {
-				qdisc_run(dev);
+				qdisc_run(txq);
 				spin_unlock(&txq->lock);
 			} else {
 				netif_schedule_queue(txq);
diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c
index 407dfdb142a4..fcc7533f0bcc 100644
--- a/net/sched/sch_generic.c
+++ b/net/sched/sch_generic.c
@@ -75,9 +75,8 @@ static inline int dev_requeue_skb(struct sk_buff *skb,
 	return 0;
 }
 
-static inline struct sk_buff *dev_dequeue_skb(struct net_device *dev,
-					      struct netdev_queue *dev_queue,
-					      struct Qdisc *q)
+static inline struct sk_buff *dequeue_skb(struct netdev_queue *dev_queue,
+					  struct Qdisc *q)
 {
 	struct sk_buff *skb;
 
@@ -90,10 +89,10 @@ static inline struct sk_buff *dev_dequeue_skb(struct net_device *dev,
 }
 
 static inline int handle_dev_cpu_collision(struct sk_buff *skb,
-					   struct net_device *dev,
 					   struct netdev_queue *dev_queue,
 					   struct Qdisc *q)
 {
+	struct net_device *dev = dev_queue->dev;
 	int ret;
 
 	if (unlikely(dev->xmit_lock_owner == smp_processor_id())) {
@@ -139,21 +138,23 @@ static inline int handle_dev_cpu_collision(struct sk_buff *skb,
  *				>0 - queue is not empty.
  *
  */
-static inline int qdisc_restart(struct net_device *dev)
+static inline int qdisc_restart(struct netdev_queue *txq)
 {
-	struct netdev_queue *txq = &dev->tx_queue;
 	struct Qdisc *q = txq->qdisc;
-	struct sk_buff *skb;
 	int ret = NETDEV_TX_BUSY;
+	struct net_device *dev;
+	struct sk_buff *skb;
 
 	/* Dequeue packet */
-	if (unlikely((skb = dev_dequeue_skb(dev, txq, q)) == NULL))
+	if (unlikely((skb = dequeue_skb(txq, q)) == NULL))
 		return 0;
 
 
 	/* And release queue */
 	spin_unlock(&txq->lock);
 
+	dev = txq->dev;
+
 	HARD_TX_LOCK(dev, smp_processor_id());
 	if (!netif_subqueue_stopped(dev, skb))
 		ret = dev_hard_start_xmit(skb, dev);
@@ -170,7 +171,7 @@ static inline int qdisc_restart(struct net_device *dev)
 
 	case NETDEV_TX_LOCKED:
 		/* Driver try lock failed */
-		ret = handle_dev_cpu_collision(skb, dev, txq, q);
+		ret = handle_dev_cpu_collision(skb, txq, q);
 		break;
 
 	default:
@@ -186,11 +187,12 @@ static inline int qdisc_restart(struct net_device *dev)
 	return ret;
 }
 
-void __qdisc_run(struct net_device *dev)
+void __qdisc_run(struct netdev_queue *txq)
 {
+	struct net_device *dev = txq->dev;
 	unsigned long start_time = jiffies;
 
-	while (qdisc_restart(dev)) {
+	while (qdisc_restart(txq)) {
 		if (netif_queue_stopped(dev))
 			break;
 
@@ -200,7 +202,7 @@ void __qdisc_run(struct net_device *dev)
 		 * 2. we've been doing it for too long.
 		 */
 		if (need_resched() || jiffies != start_time) {
-			netif_schedule_queue(&dev->tx_queue);
+			netif_schedule_queue(txq);
 			break;
 		}
 	}
-- 
cgit v1.2.3


From c773e847ea8f6812804e40f52399c6921a00eab1 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Tue, 8 Jul 2008 23:13:53 -0700
Subject: netdev: Move _xmit_lock and xmit_lock_owner into netdev_queue.

Accesses are mostly structured such that when there are multiple TX
queues the code transformations will be a little bit simpler.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/bonding/bond_main.c         | 13 ++++++-
 drivers/net/hamradio/bpqether.c         | 12 ++++++-
 drivers/net/macvlan.c                   | 14 +++++++-
 drivers/net/wireless/hostap/hostap_hw.c | 12 ++++++-
 include/linux/netdevice.h               | 62 +++++++++++++++++++++------------
 net/8021q/vlan_dev.c                    | 15 ++++++--
 net/core/dev.c                          | 28 ++++++++++-----
 net/netrom/af_netrom.c                  | 12 ++++++-
 net/rose/af_rose.c                      | 12 ++++++-
 net/sched/sch_generic.c                 |  9 +++--
 10 files changed, 145 insertions(+), 44 deletions(-)

(limited to 'include')

diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
index d57b65dc2c72..dc733d75a5e9 100644
--- a/drivers/net/bonding/bond_main.c
+++ b/drivers/net/bonding/bond_main.c
@@ -5019,6 +5019,17 @@ static int bond_check_params(struct bond_params *params)
 
 static struct lock_class_key bonding_netdev_xmit_lock_key;
 
+static void bond_set_lockdep_class_one(struct netdev_queue *txq)
+{
+	lockdep_set_class(&txq->_xmit_lock,
+			  &bonding_netdev_xmit_lock_key);
+}
+
+static void bond_set_lockdep_class(struct net_device *dev)
+{
+	bond_set_lockdep_class_one(&dev->tx_queue);
+}
+
 /* Create a new bond based on the specified name and bonding parameters.
  * If name is NULL, obtain a suitable "bond%d" name for us.
  * Caller must NOT hold rtnl_lock; we need to release it here before we
@@ -5076,7 +5087,7 @@ int bond_create(char *name, struct bond_params *params)
 		goto out_bond;
 	}
 
-	lockdep_set_class(&bond_dev->_xmit_lock, &bonding_netdev_xmit_lock_key);
+	bond_set_lockdep_class(bond_dev);
 
 	netif_carrier_off(bond_dev);
 
diff --git a/drivers/net/hamradio/bpqether.c b/drivers/net/hamradio/bpqether.c
index 5f4b4c6c9f76..fb186b8c3d4d 100644
--- a/drivers/net/hamradio/bpqether.c
+++ b/drivers/net/hamradio/bpqether.c
@@ -124,6 +124,16 @@ static LIST_HEAD(bpq_devices);
  */
 static struct lock_class_key bpq_netdev_xmit_lock_key;
 
+static void bpq_set_lockdep_class_one(struct netdev_queue *txq)
+{
+	lockdep_set_class(&txq->_xmit_lock, &bpq_netdev_xmit_lock_key);
+}
+
+static void bpq_set_lockdep_class(struct net_device *dev)
+{
+	bpq_set_lockdep_class_one(&dev->tx_queue);
+}
+
 /* ------------------------------------------------------------------------ */
 
 
@@ -523,7 +533,7 @@ static int bpq_new_device(struct net_device *edev)
 	err = register_netdevice(ndev);
 	if (err)
 		goto error;
-	lockdep_set_class(&ndev->_xmit_lock, &bpq_netdev_xmit_lock_key);
+	bpq_set_lockdep_class(ndev);
 
 	/* List protected by RTNL */
 	list_add_rcu(&bpq->bpq_list, &bpq_devices);
diff --git a/drivers/net/macvlan.c b/drivers/net/macvlan.c
index c36a03ae9bfb..c02ceaa4a216 100644
--- a/drivers/net/macvlan.c
+++ b/drivers/net/macvlan.c
@@ -277,6 +277,17 @@ static struct lock_class_key macvlan_netdev_xmit_lock_key;
 #define MACVLAN_STATE_MASK \
 	((1<<__LINK_STATE_NOCARRIER) | (1<<__LINK_STATE_DORMANT))
 
+static void macvlan_set_lockdep_class_one(struct netdev_queue *txq)
+{
+	lockdep_set_class(&txq->_xmit_lock,
+			  &macvlan_netdev_xmit_lock_key);
+}
+
+static void macvlan_set_lockdep_class(struct net_device *dev)
+{
+	macvlan_set_lockdep_class_one(&dev->tx_queue);
+}
+
 static int macvlan_init(struct net_device *dev)
 {
 	struct macvlan_dev *vlan = netdev_priv(dev);
@@ -287,7 +298,8 @@ static int macvlan_init(struct net_device *dev)
 	dev->features 		= lowerdev->features & MACVLAN_FEATURES;
 	dev->iflink		= lowerdev->ifindex;
 
-	lockdep_set_class(&dev->_xmit_lock, &macvlan_netdev_xmit_lock_key);
+	macvlan_set_lockdep_class(dev);
+
 	return 0;
 }
 
diff --git a/drivers/net/wireless/hostap/hostap_hw.c b/drivers/net/wireless/hostap/hostap_hw.c
index 09004a632ae7..c1f4bb005d92 100644
--- a/drivers/net/wireless/hostap/hostap_hw.c
+++ b/drivers/net/wireless/hostap/hostap_hw.c
@@ -3102,6 +3102,16 @@ static void prism2_clear_set_tim_queue(local_info_t *local)
  */
 static struct lock_class_key hostap_netdev_xmit_lock_key;
 
+static void prism2_set_lockdep_class_one(struct netdev_queue *txq)
+{
+	lockdep_set_class(&txq->_xmit_lock,
+			  &hostap_netdev_xmit_lock_key);
+}
+
+static void prism2_set_lockdep_class(struct net_device *dev)
+{
+	prism2_set_lockdep_class_one(&dev->tx_queue);
+}
 
 static struct net_device *
 prism2_init_local_data(struct prism2_helper_functions *funcs, int card_idx,
@@ -3268,7 +3278,7 @@ while (0)
 	if (ret >= 0)
 		ret = register_netdevice(dev);
 
-	lockdep_set_class(&dev->_xmit_lock, &hostap_netdev_xmit_lock_key);
+	prism2_set_lockdep_class(dev);
 	rtnl_unlock();
 	if (ret < 0) {
 		printk(KERN_WARNING "%s: register netdevice failed!\n",
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 28aa8e77cee9..c8d5f128858d 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -453,6 +453,8 @@ struct netdev_queue {
 	struct net_device	*dev;
 	struct Qdisc		*qdisc;
 	struct sk_buff		*gso_skb;
+	spinlock_t		_xmit_lock;
+	int			xmit_lock_owner;
 	struct Qdisc		*qdisc_sleeping;
 	struct list_head	qdisc_list;
 	struct netdev_queue	*next_sched;
@@ -639,12 +641,6 @@ struct net_device
 /*
  * One part is mostly used on xmit path (device)
  */
-	/* hard_start_xmit synchronizer */
-	spinlock_t		_xmit_lock ____cacheline_aligned_in_smp;
-	/* cpu id of processor entered to hard_start_xmit or -1,
-	   if nobody entered there.
-	 */
-	int			xmit_lock_owner;
 	void			*priv;	/* pointer to private data	*/
 	int			(*hard_start_xmit) (struct sk_buff *skb,
 						    struct net_device *dev);
@@ -1402,52 +1398,72 @@ static inline void netif_rx_complete(struct net_device *dev,
  *
  * Get network device transmit lock
  */
-static inline void __netif_tx_lock(struct net_device *dev, int cpu)
+static inline void __netif_tx_lock(struct netdev_queue *txq, int cpu)
 {
-	spin_lock(&dev->_xmit_lock);
-	dev->xmit_lock_owner = cpu;
+	spin_lock(&txq->_xmit_lock);
+	txq->xmit_lock_owner = cpu;
 }
 
 static inline void netif_tx_lock(struct net_device *dev)
 {
-	__netif_tx_lock(dev, smp_processor_id());
+	__netif_tx_lock(&dev->tx_queue, smp_processor_id());
+}
+
+static inline void __netif_tx_lock_bh(struct netdev_queue *txq)
+{
+	spin_lock_bh(&txq->_xmit_lock);
+	txq->xmit_lock_owner = smp_processor_id();
 }
 
 static inline void netif_tx_lock_bh(struct net_device *dev)
 {
-	spin_lock_bh(&dev->_xmit_lock);
-	dev->xmit_lock_owner = smp_processor_id();
+	__netif_tx_lock_bh(&dev->tx_queue);
 }
 
-static inline int netif_tx_trylock(struct net_device *dev)
+static inline int __netif_tx_trylock(struct netdev_queue *txq)
 {
-	int ok = spin_trylock(&dev->_xmit_lock);
+	int ok = spin_trylock(&txq->_xmit_lock);
 	if (likely(ok))
-		dev->xmit_lock_owner = smp_processor_id();
+		txq->xmit_lock_owner = smp_processor_id();
 	return ok;
 }
 
+static inline int netif_tx_trylock(struct net_device *dev)
+{
+	return __netif_tx_trylock(&dev->tx_queue);
+}
+
+static inline void __netif_tx_unlock(struct netdev_queue *txq)
+{
+	txq->xmit_lock_owner = -1;
+	spin_unlock(&txq->_xmit_lock);
+}
+
 static inline void netif_tx_unlock(struct net_device *dev)
 {
-	dev->xmit_lock_owner = -1;
-	spin_unlock(&dev->_xmit_lock);
+	__netif_tx_unlock(&dev->tx_queue);
+}
+
+static inline void __netif_tx_unlock_bh(struct netdev_queue *txq)
+{
+	txq->xmit_lock_owner = -1;
+	spin_unlock_bh(&txq->_xmit_lock);
 }
 
 static inline void netif_tx_unlock_bh(struct net_device *dev)
 {
-	dev->xmit_lock_owner = -1;
-	spin_unlock_bh(&dev->_xmit_lock);
+	__netif_tx_unlock_bh(&dev->tx_queue);
 }
 
-#define HARD_TX_LOCK(dev, cpu) {			\
+#define HARD_TX_LOCK(dev, txq, cpu) {			\
 	if ((dev->features & NETIF_F_LLTX) == 0) {	\
-		__netif_tx_lock(dev, cpu);			\
+		__netif_tx_lock(txq, cpu);		\
 	}						\
 }
 
-#define HARD_TX_UNLOCK(dev) {				\
+#define HARD_TX_UNLOCK(dev, txq) {			\
 	if ((dev->features & NETIF_F_LLTX) == 0) {	\
-		netif_tx_unlock(dev);			\
+		__netif_tx_unlock(txq);			\
 	}						\
 }
 
diff --git a/net/8021q/vlan_dev.c b/net/8021q/vlan_dev.c
index b6e52c025fd8..8efa399823e3 100644
--- a/net/8021q/vlan_dev.c
+++ b/net/8021q/vlan_dev.c
@@ -627,6 +627,18 @@ static void vlan_dev_set_rx_mode(struct net_device *vlan_dev)
  */
 static struct lock_class_key vlan_netdev_xmit_lock_key;
 
+static void vlan_dev_set_lockdep_one(struct netdev_queue *txq,
+				     int subclass)
+{
+	lockdep_set_class_and_subclass(&txq->_xmit_lock,
+				       &vlan_netdev_xmit_lock_key, subclass);
+}
+
+static void vlan_dev_set_lockdep_class(struct net_device *dev, int subclass)
+{
+	vlan_dev_set_lockdep_one(&dev->tx_queue, subclass);
+}
+
 static const struct header_ops vlan_header_ops = {
 	.create	 = vlan_dev_hard_header,
 	.rebuild = vlan_dev_rebuild_header,
@@ -668,8 +680,7 @@ static int vlan_dev_init(struct net_device *dev)
 	if (is_vlan_dev(real_dev))
 		subclass = 1;
 
-	lockdep_set_class_and_subclass(&dev->_xmit_lock,
-				&vlan_netdev_xmit_lock_key, subclass);
+	vlan_dev_set_lockdep_class(dev, subclass);
 	return 0;
 }
 
diff --git a/net/core/dev.c b/net/core/dev.c
index 0218b0b9be80..a29a359b15d1 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -258,7 +258,7 @@ DEFINE_PER_CPU(struct softnet_data, softnet_data);
 
 #ifdef CONFIG_DEBUG_LOCK_ALLOC
 /*
- * register_netdevice() inits dev->_xmit_lock and sets lockdep class
+ * register_netdevice() inits txq->_xmit_lock and sets lockdep class
  * according to dev->type
  */
 static const unsigned short netdev_lock_type[] =
@@ -1758,19 +1758,19 @@ gso:
 	if (dev->flags & IFF_UP) {
 		int cpu = smp_processor_id(); /* ok because BHs are off */
 
-		if (dev->xmit_lock_owner != cpu) {
+		if (txq->xmit_lock_owner != cpu) {
 
-			HARD_TX_LOCK(dev, cpu);
+			HARD_TX_LOCK(dev, txq, cpu);
 
 			if (!netif_queue_stopped(dev) &&
 			    !netif_subqueue_stopped(dev, skb)) {
 				rc = 0;
 				if (!dev_hard_start_xmit(skb, dev)) {
-					HARD_TX_UNLOCK(dev);
+					HARD_TX_UNLOCK(dev, txq);
 					goto out;
 				}
 			}
-			HARD_TX_UNLOCK(dev);
+			HARD_TX_UNLOCK(dev, txq);
 			if (net_ratelimit())
 				printk(KERN_CRIT "Virtual device %s asks to "
 				       "queue packet!\n", dev->name);
@@ -3761,6 +3761,20 @@ static void rollback_registered(struct net_device *dev)
 	dev_put(dev);
 }
 
+static void __netdev_init_queue_locks_one(struct netdev_queue *dev_queue,
+					  struct net_device *dev)
+{
+	spin_lock_init(&dev_queue->_xmit_lock);
+	netdev_set_lockdep_class(&dev_queue->_xmit_lock, dev->type);
+	dev_queue->xmit_lock_owner = -1;
+}
+
+static void netdev_init_queue_locks(struct net_device *dev)
+{
+	__netdev_init_queue_locks_one(&dev->tx_queue, dev);
+	__netdev_init_queue_locks_one(&dev->rx_queue, dev);
+}
+
 /**
  *	register_netdevice	- register a network device
  *	@dev: device to register
@@ -3795,9 +3809,7 @@ int register_netdevice(struct net_device *dev)
 	BUG_ON(!dev_net(dev));
 	net = dev_net(dev);
 
-	spin_lock_init(&dev->_xmit_lock);
-	netdev_set_lockdep_class(&dev->_xmit_lock, dev->type);
-	dev->xmit_lock_owner = -1;
+	netdev_init_queue_locks(dev);
 
 	dev->iflink = -1;
 
diff --git a/net/netrom/af_netrom.c b/net/netrom/af_netrom.c
index 74884f4a6255..819afc449e1e 100644
--- a/net/netrom/af_netrom.c
+++ b/net/netrom/af_netrom.c
@@ -74,6 +74,16 @@ static const struct proto_ops nr_proto_ops;
  */
 static struct lock_class_key nr_netdev_xmit_lock_key;
 
+static void nr_set_lockdep_one(struct netdev_queue *txq)
+{
+	lockdep_set_class(&txq->_xmit_lock, &nr_netdev_xmit_lock_key);
+}
+
+static void nr_set_lockdep_key(struct net_device *dev)
+{
+	nr_set_lockdep_one(&dev->tx_queue);
+}
+
 /*
  *	Socket removal during an interrupt is now safe.
  */
@@ -1430,7 +1440,7 @@ static int __init nr_proto_init(void)
 			free_netdev(dev);
 			goto fail;
 		}
-		lockdep_set_class(&dev->_xmit_lock, &nr_netdev_xmit_lock_key);
+		nr_set_lockdep_key(dev);
 		dev_nr[i] = dev;
 	}
 
diff --git a/net/rose/af_rose.c b/net/rose/af_rose.c
index 46461a69cd0f..7dbbc0891623 100644
--- a/net/rose/af_rose.c
+++ b/net/rose/af_rose.c
@@ -75,6 +75,16 @@ ax25_address rose_callsign;
  */
 static struct lock_class_key rose_netdev_xmit_lock_key;
 
+static void rose_set_lockdep_one(struct netdev_queue *txq)
+{
+	lockdep_set_class(&txq->_xmit_lock, &rose_netdev_xmit_lock_key);
+}
+
+static void rose_set_lockdep_key(struct net_device *dev)
+{
+	rose_set_lockdep_one(&dev->tx_queue);
+}
+
 /*
  *	Convert a ROSE address into text.
  */
@@ -1576,7 +1586,7 @@ static int __init rose_proto_init(void)
 			free_netdev(dev);
 			goto fail;
 		}
-		lockdep_set_class(&dev->_xmit_lock, &rose_netdev_xmit_lock_key);
+		rose_set_lockdep_key(dev);
 		dev_rose[i] = dev;
 	}
 
diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c
index fcc7533f0bcc..b6a36d394663 100644
--- a/net/sched/sch_generic.c
+++ b/net/sched/sch_generic.c
@@ -92,10 +92,9 @@ static inline int handle_dev_cpu_collision(struct sk_buff *skb,
 					   struct netdev_queue *dev_queue,
 					   struct Qdisc *q)
 {
-	struct net_device *dev = dev_queue->dev;
 	int ret;
 
-	if (unlikely(dev->xmit_lock_owner == smp_processor_id())) {
+	if (unlikely(dev_queue->xmit_lock_owner == smp_processor_id())) {
 		/*
 		 * Same CPU holding the lock. It may be a transient
 		 * configuration error, when hard_start_xmit() recurses. We
@@ -105,7 +104,7 @@ static inline int handle_dev_cpu_collision(struct sk_buff *skb,
 		kfree_skb(skb);
 		if (net_ratelimit())
 			printk(KERN_WARNING "Dead loop on netdevice %s, "
-			       "fix it urgently!\n", dev->name);
+			       "fix it urgently!\n", dev_queue->dev->name);
 		ret = qdisc_qlen(q);
 	} else {
 		/*
@@ -155,10 +154,10 @@ static inline int qdisc_restart(struct netdev_queue *txq)
 
 	dev = txq->dev;
 
-	HARD_TX_LOCK(dev, smp_processor_id());
+	HARD_TX_LOCK(dev, txq, smp_processor_id());
 	if (!netif_subqueue_stopped(dev, skb))
 		ret = dev_hard_start_xmit(skb, dev);
-	HARD_TX_UNLOCK(dev);
+	HARD_TX_UNLOCK(dev, txq);
 
 	spin_lock(&txq->lock);
 	q = txq->qdisc;
-- 
cgit v1.2.3


From b19fa1fa91845234961c64dbd564671aa7c0fd27 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Tue, 8 Jul 2008 23:14:24 -0700
Subject: net: Delete NETDEVICES_MULTIQUEUE kconfig option.

Multiple TX queue support is a core networking feature.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/networking/multiqueue.txt | 79 +--------------------------------
 drivers/net/Kconfig                     |  8 ----
 drivers/net/cpmac.c                     | 14 ------
 drivers/net/ixgbe/ixgbe_ethtool.c       |  6 ---
 drivers/net/ixgbe/ixgbe_main.c          | 40 -----------------
 drivers/net/s2io.c                      | 47 ++++----------------
 include/linux/netdevice.h               | 14 ------
 include/linux/skbuff.h                  | 10 -----
 net/mac80211/Kconfig                    |  3 --
 9 files changed, 9 insertions(+), 212 deletions(-)

(limited to 'include')

diff --git a/Documentation/networking/multiqueue.txt b/Documentation/networking/multiqueue.txt
index ea5a42e8f79f..e6dc1ee9e8f1 100644
--- a/Documentation/networking/multiqueue.txt
+++ b/Documentation/networking/multiqueue.txt
@@ -3,19 +3,11 @@
 		===========================================
 
 Section 1: Base driver requirements for implementing multiqueue support
-Section 2: Qdisc support for multiqueue devices
-Section 3: Brief howto using PRIO or RR for multiqueue devices
-
 
 Intro: Kernel support for multiqueue devices
 ---------------------------------------------------------
 
-Kernel support for multiqueue devices is only an API that is presented to the
-netdevice layer for base drivers to implement.  This feature is part of the
-core networking stack, and all network devices will be running on the
-multiqueue-aware stack.  If a base driver only has one queue, then these
-changes are transparent to that driver.
-
+Kernel support for multiqueue devices is always present.
 
 Section 1: Base driver requirements for implementing multiqueue support
 -----------------------------------------------------------------------
@@ -43,73 +35,4 @@ bitmap on device initialization.  Below is an example from e1000:
 		netdev->features |= NETIF_F_MULTI_QUEUE;
 #endif
 
-
-Section 2: Qdisc support for multiqueue devices
------------------------------------------------
-
-Currently two qdiscs support multiqueue devices.  A new round-robin qdisc,
-sch_rr, and sch_prio. The qdisc is responsible for classifying the skb's to
-bands and queues, and will store the queue mapping into skb->queue_mapping.
-Use this field in the base driver to determine which queue to send the skb
-to.
-
-sch_rr has been added for hardware that doesn't want scheduling policies from
-software, so it's a straight round-robin qdisc.  It uses the same syntax and
-classification priomap that sch_prio uses, so it should be intuitive to
-configure for people who've used sch_prio.
-
-In order to utilitize the multiqueue features of the qdiscs, the network
-device layer needs to enable multiple queue support.  This can be done by
-selecting NETDEVICES_MULTIQUEUE under Drivers.
-
-The PRIO qdisc naturally plugs into a multiqueue device.  If
-NETDEVICES_MULTIQUEUE is selected, then on qdisc load, the number of
-bands requested is compared to the number of queues on the hardware.  If they
-are equal, it sets a one-to-one mapping up between the queues and bands.  If
-they're not equal, it will not load the qdisc.  This is the same behavior
-for RR.  Once the association is made, any skb that is classified will have
-skb->queue_mapping set, which will allow the driver to properly queue skb's
-to multiple queues.
-
-
-Section 3: Brief howto using PRIO and RR for multiqueue devices
----------------------------------------------------------------
-
-The userspace command 'tc,' part of the iproute2 package, is used to configure
-qdiscs.  To add the PRIO qdisc to your network device, assuming the device is
-called eth0, run the following command:
-
-# tc qdisc add dev eth0 root handle 1: prio bands 4 multiqueue
-
-This will create 4 bands, 0 being highest priority, and associate those bands
-to the queues on your NIC.  Assuming eth0 has 4 Tx queues, the band mapping
-would look like:
-
-band 0 => queue 0
-band 1 => queue 1
-band 2 => queue 2
-band 3 => queue 3
-
-Traffic will begin flowing through each queue if your TOS values are assigning
-traffic across the various bands.  For example, ssh traffic will always try to
-go out band 0 based on TOS -> Linux priority conversion (realtime traffic),
-so it will be sent out queue 0.  ICMP traffic (pings) fall into the "normal"
-traffic classification, which is band 1.  Therefore pings will be send out
-queue 1 on the NIC.
-
-Note the use of the multiqueue keyword.  This is only in versions of iproute2
-that support multiqueue networking devices; if this is omitted when loading
-a qdisc onto a multiqueue device, the qdisc will load and operate the same
-if it were loaded onto a single-queue device (i.e. - sends all traffic to
-queue 0).
-
-Another alternative to multiqueue band allocation can be done by using the
-multiqueue option and specify 0 bands.  If this is the case, the qdisc will
-allocate the number of bands to equal the number of queues that the device
-reports, and bring the qdisc online.
-
-The behavior of tc filters remains the same, where it will override TOS priority
-classification.
-
-
 Author: Peter P. Waskiewicz Jr. <peter.p.waskiewicz.jr@intel.com>
diff --git a/drivers/net/Kconfig b/drivers/net/Kconfig
index ef733abc857d..4675c1bd6fb9 100644
--- a/drivers/net/Kconfig
+++ b/drivers/net/Kconfig
@@ -26,14 +26,6 @@ menuconfig NETDEVICES
 # that for each of the symbols.
 if NETDEVICES
 
-config NETDEVICES_MULTIQUEUE
-	bool "Netdevice multiple hardware queue support"
-	---help---
-	  Say Y here if you want to allow the network stack to use multiple
-	  hardware TX queues on an ethernet device.
-
-	  Most people will say N here.
-
 config IFB
 	tristate "Intermediate Functional Block support"
 	depends on NET_CLS_ACT
diff --git a/drivers/net/cpmac.c b/drivers/net/cpmac.c
index 7f3f62e1b113..d630e2a72f42 100644
--- a/drivers/net/cpmac.c
+++ b/drivers/net/cpmac.c
@@ -569,11 +569,7 @@ static int cpmac_start_xmit(struct sk_buff *skb, struct net_device *dev)
 
 	len = max(skb->len, ETH_ZLEN);
 	queue = skb_get_queue_mapping(skb);
-#ifdef CONFIG_NETDEVICES_MULTIQUEUE
 	netif_stop_subqueue(dev, queue);
-#else
-	netif_stop_queue(dev);
-#endif
 
 	desc = &priv->desc_ring[queue];
 	if (unlikely(desc->dataflags & CPMAC_OWN)) {
@@ -626,24 +622,14 @@ static void cpmac_end_xmit(struct net_device *dev, int queue)
 
 		dev_kfree_skb_irq(desc->skb);
 		desc->skb = NULL;
-#ifdef CONFIG_NETDEVICES_MULTIQUEUE
 		if (netif_subqueue_stopped(dev, queue))
 			netif_wake_subqueue(dev, queue);
-#else
-		if (netif_queue_stopped(dev))
-			netif_wake_queue(dev);
-#endif
 	} else {
 		if (netif_msg_tx_err(priv) && net_ratelimit())
 			printk(KERN_WARNING
 			       "%s: end_xmit: spurious interrupt\n", dev->name);
-#ifdef CONFIG_NETDEVICES_MULTIQUEUE
 		if (netif_subqueue_stopped(dev, queue))
 			netif_wake_subqueue(dev, queue);
-#else
-		if (netif_queue_stopped(dev))
-			netif_wake_queue(dev);
-#endif
 	}
 }
 
diff --git a/drivers/net/ixgbe/ixgbe_ethtool.c b/drivers/net/ixgbe/ixgbe_ethtool.c
index 12990b1fe7e4..81b769093d22 100644
--- a/drivers/net/ixgbe/ixgbe_ethtool.c
+++ b/drivers/net/ixgbe/ixgbe_ethtool.c
@@ -252,21 +252,15 @@ static int ixgbe_set_tso(struct net_device *netdev, u32 data)
 		netdev->features |= NETIF_F_TSO;
 		netdev->features |= NETIF_F_TSO6;
 	} else {
-#ifdef CONFIG_NETDEVICES_MULTIQUEUE
 		struct ixgbe_adapter *adapter = netdev_priv(netdev);
 		int i;
-#endif
 		netif_stop_queue(netdev);
-#ifdef CONFIG_NETDEVICES_MULTIQUEUE
 		for (i = 0; i < adapter->num_tx_queues; i++)
 			netif_stop_subqueue(netdev, i);
-#endif
 		netdev->features &= ~NETIF_F_TSO;
 		netdev->features &= ~NETIF_F_TSO6;
-#ifdef CONFIG_NETDEVICES_MULTIQUEUE
 		for (i = 0; i < adapter->num_tx_queues; i++)
 			netif_start_subqueue(netdev, i);
-#endif
 		netif_start_queue(netdev);
 	}
 	return 0;
diff --git a/drivers/net/ixgbe/ixgbe_main.c b/drivers/net/ixgbe/ixgbe_main.c
index b37d618d8e2a..10a1c8c5cda1 100644
--- a/drivers/net/ixgbe/ixgbe_main.c
+++ b/drivers/net/ixgbe/ixgbe_main.c
@@ -266,28 +266,16 @@ static bool ixgbe_clean_tx_irq(struct ixgbe_adapter *adapter,
 		 * sees the new next_to_clean.
 		 */
 		smp_mb();
-#ifdef CONFIG_NETDEVICES_MULTIQUEUE
 		if (__netif_subqueue_stopped(netdev, tx_ring->queue_index) &&
 		    !test_bit(__IXGBE_DOWN, &adapter->state)) {
 			netif_wake_subqueue(netdev, tx_ring->queue_index);
 			adapter->restart_queue++;
 		}
-#else
-		if (netif_queue_stopped(netdev) &&
-		    !test_bit(__IXGBE_DOWN, &adapter->state)) {
-			netif_wake_queue(netdev);
-			adapter->restart_queue++;
-		}
-#endif
 	}
 
 	if (adapter->detect_tx_hung)
 		if (ixgbe_check_tx_hang(adapter, tx_ring, eop, eop_desc))
-#ifdef CONFIG_NETDEVICES_MULTIQUEUE
 			netif_stop_subqueue(netdev, tx_ring->queue_index);
-#else
-			netif_stop_queue(netdev);
-#endif
 
 	if (total_tx_packets >= tx_ring->work_limit)
 		IXGBE_WRITE_REG(&adapter->hw, IXGBE_EICS, tx_ring->eims_value);
@@ -2192,11 +2180,7 @@ static void __devinit ixgbe_set_num_queues(struct ixgbe_adapter *adapter)
 		case (IXGBE_FLAG_RSS_ENABLED):
 			rss_m = 0xF;
 			nrq = rss_i;
-#ifdef CONFIG_NETDEVICES_MULTIQUEUE
 			ntq = rss_i;
-#else
-			ntq = 1;
-#endif
 			break;
 		case 0:
 		default:
@@ -2370,10 +2354,8 @@ try_msi:
 	}
 
 out:
-#ifdef CONFIG_NETDEVICES_MULTIQUEUE
 	/* Notify the stack of the (possibly) reduced Tx Queue count. */
 	adapter->netdev->egress_subqueue_count = adapter->num_tx_queues;
-#endif
 
 	return err;
 }
@@ -2910,9 +2892,7 @@ static void ixgbe_watchdog(unsigned long data)
 	struct net_device *netdev = adapter->netdev;
 	bool link_up;
 	u32 link_speed = 0;
-#ifdef CONFIG_NETDEVICES_MULTIQUEUE
 	int i;
-#endif
 
 	adapter->hw.mac.ops.check_link(&adapter->hw, &(link_speed), &link_up);
 
@@ -2934,10 +2914,8 @@ static void ixgbe_watchdog(unsigned long data)
 
 			netif_carrier_on(netdev);
 			netif_wake_queue(netdev);
-#ifdef CONFIG_NETDEVICES_MULTIQUEUE
 			for (i = 0; i < adapter->num_tx_queues; i++)
 				netif_wake_subqueue(netdev, i);
-#endif
 		} else {
 			/* Force detection of hung controller */
 			adapter->detect_tx_hung = true;
@@ -3264,11 +3242,7 @@ static int __ixgbe_maybe_stop_tx(struct net_device *netdev,
 {
 	struct ixgbe_adapter *adapter = netdev_priv(netdev);
 
-#ifdef CONFIG_NETDEVICES_MULTIQUEUE
 	netif_stop_subqueue(netdev, tx_ring->queue_index);
-#else
-	netif_stop_queue(netdev);
-#endif
 	/* Herbert's original patch had:
 	 *  smp_mb__after_netif_stop_queue();
 	 * but since that doesn't exist yet, just open code it. */
@@ -3280,11 +3254,7 @@ static int __ixgbe_maybe_stop_tx(struct net_device *netdev,
 		return -EBUSY;
 
 	/* A reprieve! - use start_queue because it doesn't call schedule */
-#ifdef CONFIG_NETDEVICES_MULTIQUEUE
 	netif_wake_subqueue(netdev, tx_ring->queue_index);
-#else
-	netif_wake_queue(netdev);
-#endif
 	++adapter->restart_queue;
 	return 0;
 }
@@ -3312,9 +3282,7 @@ static int ixgbe_xmit_frame(struct sk_buff *skb, struct net_device *netdev)
 	unsigned int f;
 	unsigned int nr_frags = skb_shinfo(skb)->nr_frags;
 	len -= skb->data_len;
-#ifdef CONFIG_NETDEVICES_MULTIQUEUE
 	r_idx = (adapter->num_tx_queues - 1) & skb->queue_mapping;
-#endif
 	tx_ring = &adapter->tx_ring[r_idx];
 
 
@@ -3502,11 +3470,7 @@ static int __devinit ixgbe_probe(struct pci_dev *pdev,
 	pci_set_master(pdev);
 	pci_save_state(pdev);
 
-#ifdef CONFIG_NETDEVICES_MULTIQUEUE
 	netdev = alloc_etherdev_mq(sizeof(struct ixgbe_adapter), MAX_TX_QUEUES);
-#else
-	netdev = alloc_etherdev(sizeof(struct ixgbe_adapter));
-#endif
 	if (!netdev) {
 		err = -ENOMEM;
 		goto err_alloc_etherdev;
@@ -3598,9 +3562,7 @@ static int __devinit ixgbe_probe(struct pci_dev *pdev,
 	if (pci_using_dac)
 		netdev->features |= NETIF_F_HIGHDMA;
 
-#ifdef CONFIG_NETDEVICES_MULTIQUEUE
 	netdev->features |= NETIF_F_MULTI_QUEUE;
-#endif
 
 	/* make sure the EEPROM is good */
 	if (ixgbe_validate_eeprom_checksum(hw, NULL) < 0) {
@@ -3668,10 +3630,8 @@ static int __devinit ixgbe_probe(struct pci_dev *pdev,
 
 	netif_carrier_off(netdev);
 	netif_stop_queue(netdev);
-#ifdef CONFIG_NETDEVICES_MULTIQUEUE
 	for (i = 0; i < adapter->num_tx_queues; i++)
 		netif_stop_subqueue(netdev, i);
-#endif
 
 	ixgbe_napi_add_all(adapter);
 
diff --git a/drivers/net/s2io.c b/drivers/net/s2io.c
index e7a3dbec674c..51a91154125d 100644
--- a/drivers/net/s2io.c
+++ b/drivers/net/s2io.c
@@ -546,13 +546,10 @@ static struct pci_driver s2io_driver = {
 static inline void s2io_stop_all_tx_queue(struct s2io_nic *sp)
 {
 	int i;
-#ifdef CONFIG_NETDEVICES_MULTIQUEUE
 	if (sp->config.multiq) {
 		for (i = 0; i < sp->config.tx_fifo_num; i++)
 			netif_stop_subqueue(sp->dev, i);
-	} else
-#endif
-	{
+	} else {
 		for (i = 0; i < sp->config.tx_fifo_num; i++)
 			sp->mac_control.fifos[i].queue_state = FIFO_QUEUE_STOP;
 		netif_stop_queue(sp->dev);
@@ -561,12 +558,9 @@ static inline void s2io_stop_all_tx_queue(struct s2io_nic *sp)
 
 static inline void s2io_stop_tx_queue(struct s2io_nic *sp, int fifo_no)
 {
-#ifdef CONFIG_NETDEVICES_MULTIQUEUE
 	if (sp->config.multiq)
 		netif_stop_subqueue(sp->dev, fifo_no);
-	else
-#endif
-	{
+	else {
 		sp->mac_control.fifos[fifo_no].queue_state =
 			FIFO_QUEUE_STOP;
 		netif_stop_queue(sp->dev);
@@ -576,13 +570,10 @@ static inline void s2io_stop_tx_queue(struct s2io_nic *sp, int fifo_no)
 static inline void s2io_start_all_tx_queue(struct s2io_nic *sp)
 {
 	int i;
-#ifdef CONFIG_NETDEVICES_MULTIQUEUE
 	if (sp->config.multiq) {
 		for (i = 0; i < sp->config.tx_fifo_num; i++)
 			netif_start_subqueue(sp->dev, i);
-	} else
-#endif
-	{
+	} else {
 		for (i = 0; i < sp->config.tx_fifo_num; i++)
 			sp->mac_control.fifos[i].queue_state = FIFO_QUEUE_START;
 		netif_start_queue(sp->dev);
@@ -591,12 +582,9 @@ static inline void s2io_start_all_tx_queue(struct s2io_nic *sp)
 
 static inline void s2io_start_tx_queue(struct s2io_nic *sp, int fifo_no)
 {
-#ifdef CONFIG_NETDEVICES_MULTIQUEUE
 	if (sp->config.multiq)
 		netif_start_subqueue(sp->dev, fifo_no);
-	else
-#endif
-	{
+	else {
 		sp->mac_control.fifos[fifo_no].queue_state =
 			FIFO_QUEUE_START;
 		netif_start_queue(sp->dev);
@@ -606,13 +594,10 @@ static inline void s2io_start_tx_queue(struct s2io_nic *sp, int fifo_no)
 static inline void s2io_wake_all_tx_queue(struct s2io_nic *sp)
 {
 	int i;
-#ifdef CONFIG_NETDEVICES_MULTIQUEUE
 	if (sp->config.multiq) {
 		for (i = 0; i < sp->config.tx_fifo_num; i++)
 			netif_wake_subqueue(sp->dev, i);
-	} else
-#endif
-	{
+	} else {
 		for (i = 0; i < sp->config.tx_fifo_num; i++)
 			sp->mac_control.fifos[i].queue_state = FIFO_QUEUE_START;
 		netif_wake_queue(sp->dev);
@@ -623,13 +608,10 @@ static inline void s2io_wake_tx_queue(
 	struct fifo_info *fifo, int cnt, u8 multiq)
 {
 
-#ifdef CONFIG_NETDEVICES_MULTIQUEUE
 	if (multiq) {
 		if (cnt && __netif_subqueue_stopped(fifo->dev, fifo->fifo_no))
 			netif_wake_subqueue(fifo->dev, fifo->fifo_no);
-	} else
-#endif
-	if (cnt && (fifo->queue_state == FIFO_QUEUE_STOP)) {
+	} else if (cnt && (fifo->queue_state == FIFO_QUEUE_STOP)) {
 		if (netif_queue_stopped(fifo->dev)) {
 			fifo->queue_state = FIFO_QUEUE_START;
 			netif_wake_queue(fifo->dev);
@@ -4189,15 +4171,12 @@ static int s2io_xmit(struct sk_buff *skb, struct net_device *dev)
 			return NETDEV_TX_LOCKED;
 	}
 
-#ifdef CONFIG_NETDEVICES_MULTIQUEUE
 	if (sp->config.multiq) {
 		if (__netif_subqueue_stopped(dev, fifo->fifo_no)) {
 			spin_unlock_irqrestore(&fifo->tx_lock, flags);
 			return NETDEV_TX_BUSY;
 		}
-	} else
-#endif
-	if (unlikely(fifo->queue_state == FIFO_QUEUE_STOP)) {
+	} else if (unlikely(fifo->queue_state == FIFO_QUEUE_STOP)) {
 		if (netif_queue_stopped(dev)) {
 			spin_unlock_irqrestore(&fifo->tx_lock, flags);
 			return NETDEV_TX_BUSY;
@@ -7633,12 +7612,6 @@ static int s2io_verify_parm(struct pci_dev *pdev, u8 *dev_intr_type,
 		DBG_PRINT(ERR_DBG, "tx fifos\n");
 	}
 
-#ifndef CONFIG_NETDEVICES_MULTIQUEUE
-	if (multiq) {
-		DBG_PRINT(ERR_DBG, "s2io: Multiqueue support not enabled\n");
-		multiq = 0;
-	}
-#endif
 	if (multiq)
 		*dev_multiq = multiq;
 
@@ -7783,12 +7756,10 @@ s2io_init_nic(struct pci_dev *pdev, const struct pci_device_id *pre)
 		pci_disable_device(pdev);
 		return -ENODEV;
 	}
-#ifdef CONFIG_NETDEVICES_MULTIQUEUE
 	if (dev_multiq)
 		dev = alloc_etherdev_mq(sizeof(struct s2io_nic), tx_fifo_num);
 	else
-#endif
-	dev = alloc_etherdev(sizeof(struct s2io_nic));
+		dev = alloc_etherdev(sizeof(struct s2io_nic));
 	if (dev == NULL) {
 		DBG_PRINT(ERR_DBG, "Device allocation failed\n");
 		pci_disable_device(pdev);
@@ -7979,10 +7950,8 @@ s2io_init_nic(struct pci_dev *pdev, const struct pci_device_id *pre)
 		dev->features |= NETIF_F_UFO;
 		dev->features |= NETIF_F_HW_CSUM;
 	}
-#ifdef CONFIG_NETDEVICES_MULTIQUEUE
 	if (config->multiq)
 		dev->features |= NETIF_F_MULTI_QUEUE;
-#endif
 	dev->tx_timeout = &s2io_tx_watchdog;
 	dev->watchdog_timeo = WATCH_DOG_TIMEOUT;
 	INIT_WORK(&sp->rst_timer_task, s2io_restart_nic);
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index c8d5f128858d..e2d931f9b700 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -1043,9 +1043,7 @@ static inline int netif_running(const struct net_device *dev)
  */
 static inline void netif_start_subqueue(struct net_device *dev, u16 queue_index)
 {
-#ifdef CONFIG_NETDEVICES_MULTIQUEUE
 	clear_bit(__LINK_STATE_XOFF, &dev->egress_subqueue[queue_index].state);
-#endif
 }
 
 /**
@@ -1057,13 +1055,11 @@ static inline void netif_start_subqueue(struct net_device *dev, u16 queue_index)
  */
 static inline void netif_stop_subqueue(struct net_device *dev, u16 queue_index)
 {
-#ifdef CONFIG_NETDEVICES_MULTIQUEUE
 #ifdef CONFIG_NETPOLL_TRAP
 	if (netpoll_trap())
 		return;
 #endif
 	set_bit(__LINK_STATE_XOFF, &dev->egress_subqueue[queue_index].state);
-#endif
 }
 
 /**
@@ -1076,12 +1072,8 @@ static inline void netif_stop_subqueue(struct net_device *dev, u16 queue_index)
 static inline int __netif_subqueue_stopped(const struct net_device *dev,
 					 u16 queue_index)
 {
-#ifdef CONFIG_NETDEVICES_MULTIQUEUE
 	return test_bit(__LINK_STATE_XOFF,
 			&dev->egress_subqueue[queue_index].state);
-#else
-	return 0;
-#endif
 }
 
 static inline int netif_subqueue_stopped(const struct net_device *dev,
@@ -1099,7 +1091,6 @@ static inline int netif_subqueue_stopped(const struct net_device *dev,
  */
 static inline void netif_wake_subqueue(struct net_device *dev, u16 queue_index)
 {
-#ifdef CONFIG_NETDEVICES_MULTIQUEUE
 #ifdef CONFIG_NETPOLL_TRAP
 	if (netpoll_trap())
 		return;
@@ -1107,7 +1098,6 @@ static inline void netif_wake_subqueue(struct net_device *dev, u16 queue_index)
 	if (test_and_clear_bit(__LINK_STATE_XOFF,
 			       &dev->egress_subqueue[queue_index].state))
 		__netif_schedule(&dev->tx_queue);
-#endif
 }
 
 /**
@@ -1119,11 +1109,7 @@ static inline void netif_wake_subqueue(struct net_device *dev, u16 queue_index)
  */
 static inline int netif_is_multiqueue(const struct net_device *dev)
 {
-#ifdef CONFIG_NETDEVICES_MULTIQUEUE
 	return (!!(NETIF_F_MULTI_QUEUE & dev->features));
-#else
-	return 0;
-#endif
 }
 
 /* Use this variant when it is known for sure that it
diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index 2220b9e2dab0..8f10e3d08fd9 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -305,9 +305,7 @@ struct sk_buff {
 #endif
 
 	int			iif;
-#ifdef CONFIG_NETDEVICES_MULTIQUEUE
 	__u16			queue_mapping;
-#endif
 #ifdef CONFIG_NET_SCHED
 	__u16			tc_index;	/* traffic control index */
 #ifdef CONFIG_NET_CLS_ACT
@@ -1671,25 +1669,17 @@ static inline void skb_init_secmark(struct sk_buff *skb)
 
 static inline void skb_set_queue_mapping(struct sk_buff *skb, u16 queue_mapping)
 {
-#ifdef CONFIG_NETDEVICES_MULTIQUEUE
 	skb->queue_mapping = queue_mapping;
-#endif
 }
 
 static inline u16 skb_get_queue_mapping(struct sk_buff *skb)
 {
-#ifdef CONFIG_NETDEVICES_MULTIQUEUE
 	return skb->queue_mapping;
-#else
-	return 0;
-#endif
 }
 
 static inline void skb_copy_queue_mapping(struct sk_buff *to, const struct sk_buff *from)
 {
-#ifdef CONFIG_NETDEVICES_MULTIQUEUE
 	to->queue_mapping = from->queue_mapping;
-#endif
 }
 
 static inline int skb_is_gso(const struct sk_buff *skb)
diff --git a/net/mac80211/Kconfig b/net/mac80211/Kconfig
index 40f1add17753..d2038418e2bd 100644
--- a/net/mac80211/Kconfig
+++ b/net/mac80211/Kconfig
@@ -15,14 +15,11 @@ config MAC80211_QOS
 	def_bool y
 	depends on MAC80211
 	depends on NET_SCHED
-	depends on NETDEVICES_MULTIQUEUE
 
 comment "QoS/HT support disabled"
 	depends on MAC80211 && !MAC80211_QOS
 comment "QoS/HT support needs CONFIG_NET_SCHED"
 	depends on MAC80211 && !NET_SCHED
-comment "QoS/HT support needs CONFIG_NETDEVICES_MULTIQUEUE"
-	depends on MAC80211 && !NETDEVICES_MULTIQUEUE
 
 menu "Rate control algorithm selection"
 	depends on MAC80211 != n
-- 
cgit v1.2.3


From 79d16385c7f287a33ea771c4dbe60ae43f791b49 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Tue, 8 Jul 2008 23:14:46 -0700
Subject: netdev: Move atomic queue state bits into netdev_queue.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h | 55 +++++++++++++++++++++++++++++++++--------------
 include/net/pkt_sched.h   |  2 +-
 net/sched/sch_generic.c   | 20 +++++++++--------
 3 files changed, 51 insertions(+), 26 deletions(-)

(limited to 'include')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index e2d931f9b700..203c5504fe43 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -281,14 +281,12 @@ struct header_ops {
 
 enum netdev_state_t
 {
-	__LINK_STATE_XOFF=0,
 	__LINK_STATE_START,
 	__LINK_STATE_PRESENT,
 	__LINK_STATE_SCHED,
 	__LINK_STATE_NOCARRIER,
 	__LINK_STATE_LINKWATCH_PENDING,
 	__LINK_STATE_DORMANT,
-	__LINK_STATE_QDISC_RUNNING,
 };
 
 
@@ -448,10 +446,17 @@ static inline void napi_synchronize(const struct napi_struct *n)
 # define napi_synchronize(n)	barrier()
 #endif
 
+enum netdev_queue_state_t
+{
+	__QUEUE_STATE_XOFF,
+	__QUEUE_STATE_QDISC_RUNNING,
+};
+
 struct netdev_queue {
 	spinlock_t		lock;
 	struct net_device	*dev;
 	struct Qdisc		*qdisc;
+	unsigned long		state;
 	struct sk_buff		*gso_skb;
 	spinlock_t		_xmit_lock;
 	int			xmit_lock_owner;
@@ -952,9 +957,7 @@ extern void __netif_schedule(struct netdev_queue *txq);
 
 static inline void netif_schedule_queue(struct netdev_queue *txq)
 {
-	struct net_device *dev = txq->dev;
-
-	if (!test_bit(__LINK_STATE_XOFF, &dev->state))
+	if (!test_bit(__QUEUE_STATE_XOFF, &txq->state))
 		__netif_schedule(txq);
 }
 
@@ -969,9 +972,14 @@ static inline void netif_schedule(struct net_device *dev)
  *
  *	Allow upper layers to call the device hard_start_xmit routine.
  */
+static inline void netif_tx_start_queue(struct netdev_queue *dev_queue)
+{
+	clear_bit(__QUEUE_STATE_XOFF, &dev_queue->state);
+}
+
 static inline void netif_start_queue(struct net_device *dev)
 {
-	clear_bit(__LINK_STATE_XOFF, &dev->state);
+	netif_tx_start_queue(&dev->tx_queue);
 }
 
 /**
@@ -981,16 +989,21 @@ static inline void netif_start_queue(struct net_device *dev)
  *	Allow upper layers to call the device hard_start_xmit routine.
  *	Used for flow control when transmit resources are available.
  */
-static inline void netif_wake_queue(struct net_device *dev)
+static inline void netif_tx_wake_queue(struct netdev_queue *dev_queue)
 {
 #ifdef CONFIG_NETPOLL_TRAP
 	if (netpoll_trap()) {
-		clear_bit(__LINK_STATE_XOFF, &dev->state);
+		clear_bit(__QUEUE_STATE_XOFF, &dev_queue->state);
 		return;
 	}
 #endif
-	if (test_and_clear_bit(__LINK_STATE_XOFF, &dev->state))
-		__netif_schedule(&dev->tx_queue);
+	if (test_and_clear_bit(__QUEUE_STATE_XOFF, &dev_queue->state))
+		__netif_schedule(dev_queue);
+}
+
+static inline void netif_wake_queue(struct net_device *dev)
+{
+	netif_tx_wake_queue(&dev->tx_queue);
 }
 
 /**
@@ -1000,9 +1013,14 @@ static inline void netif_wake_queue(struct net_device *dev)
  *	Stop upper layers calling the device hard_start_xmit routine.
  *	Used for flow control when transmit resources are unavailable.
  */
+static inline void netif_tx_stop_queue(struct netdev_queue *dev_queue)
+{
+	set_bit(__QUEUE_STATE_XOFF, &dev_queue->state);
+}
+
 static inline void netif_stop_queue(struct net_device *dev)
 {
-	set_bit(__LINK_STATE_XOFF, &dev->state);
+	netif_tx_stop_queue(&dev->tx_queue);
 }
 
 /**
@@ -1011,9 +1029,14 @@ static inline void netif_stop_queue(struct net_device *dev)
  *
  *	Test if transmit queue on device is currently unable to send.
  */
+static inline int netif_tx_queue_stopped(const struct netdev_queue *dev_queue)
+{
+	return test_bit(__QUEUE_STATE_XOFF, &dev_queue->state);
+}
+
 static inline int netif_queue_stopped(const struct net_device *dev)
 {
-	return test_bit(__LINK_STATE_XOFF, &dev->state);
+	return netif_tx_queue_stopped(&dev->tx_queue);
 }
 
 /**
@@ -1043,7 +1066,7 @@ static inline int netif_running(const struct net_device *dev)
  */
 static inline void netif_start_subqueue(struct net_device *dev, u16 queue_index)
 {
-	clear_bit(__LINK_STATE_XOFF, &dev->egress_subqueue[queue_index].state);
+	clear_bit(__QUEUE_STATE_XOFF, &dev->egress_subqueue[queue_index].state);
 }
 
 /**
@@ -1059,7 +1082,7 @@ static inline void netif_stop_subqueue(struct net_device *dev, u16 queue_index)
 	if (netpoll_trap())
 		return;
 #endif
-	set_bit(__LINK_STATE_XOFF, &dev->egress_subqueue[queue_index].state);
+	set_bit(__QUEUE_STATE_XOFF, &dev->egress_subqueue[queue_index].state);
 }
 
 /**
@@ -1072,7 +1095,7 @@ static inline void netif_stop_subqueue(struct net_device *dev, u16 queue_index)
 static inline int __netif_subqueue_stopped(const struct net_device *dev,
 					 u16 queue_index)
 {
-	return test_bit(__LINK_STATE_XOFF,
+	return test_bit(__QUEUE_STATE_XOFF,
 			&dev->egress_subqueue[queue_index].state);
 }
 
@@ -1095,7 +1118,7 @@ static inline void netif_wake_subqueue(struct net_device *dev, u16 queue_index)
 	if (netpoll_trap())
 		return;
 #endif
-	if (test_and_clear_bit(__LINK_STATE_XOFF,
+	if (test_and_clear_bit(__QUEUE_STATE_XOFF,
 			       &dev->egress_subqueue[queue_index].state))
 		__netif_schedule(&dev->tx_queue);
 }
diff --git a/include/net/pkt_sched.h b/include/net/pkt_sched.h
index 2311d242bb35..d58c1a5eb845 100644
--- a/include/net/pkt_sched.h
+++ b/include/net/pkt_sched.h
@@ -91,7 +91,7 @@ static inline void qdisc_run(struct netdev_queue *txq)
 	struct net_device *dev = txq->dev;
 
 	if (!netif_queue_stopped(dev) &&
-	    !test_and_set_bit(__LINK_STATE_QDISC_RUNNING, &dev->state))
+	    !test_and_set_bit(__QUEUE_STATE_QDISC_RUNNING, &txq->state))
 		__qdisc_run(txq);
 }
 
diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c
index b6a36d394663..243de935b182 100644
--- a/net/sched/sch_generic.c
+++ b/net/sched/sch_generic.c
@@ -121,9 +121,9 @@ static inline int handle_dev_cpu_collision(struct sk_buff *skb,
 /*
  * NOTE: Called under queue->lock with locally disabled BH.
  *
- * __LINK_STATE_QDISC_RUNNING guarantees only one CPU can process this
- * device at a time. queue->lock serializes queue accesses for
- * this device AND txq->qdisc pointer itself.
+ * __QUEUE_STATE_QDISC_RUNNING guarantees only one CPU can process
+ * this queue at a time. queue->lock serializes queue accesses for
+ * this queue AND txq->qdisc pointer itself.
  *
  *  netif_tx_lock serializes accesses to device driver.
  *
@@ -206,7 +206,7 @@ void __qdisc_run(struct netdev_queue *txq)
 		}
 	}
 
-	clear_bit(__LINK_STATE_QDISC_RUNNING, &dev->state);
+	clear_bit(__QUEUE_STATE_QDISC_RUNNING, &txq->state);
 }
 
 static void dev_watchdog(unsigned long arg)
@@ -605,9 +605,10 @@ static void dev_deactivate_queue(struct netdev_queue *dev_queue,
 
 void dev_deactivate(struct net_device *dev)
 {
+	struct netdev_queue *dev_queue = &dev->tx_queue;
 	int running;
 
-	dev_deactivate_queue(&dev->tx_queue, &noop_qdisc);
+	dev_deactivate_queue(dev_queue, &noop_qdisc);
 
 	dev_watchdog_down(dev);
 
@@ -616,16 +617,17 @@ void dev_deactivate(struct net_device *dev)
 
 	/* Wait for outstanding qdisc_run calls. */
 	do {
-		while (test_bit(__LINK_STATE_QDISC_RUNNING, &dev->state))
+		while (test_bit(__QUEUE_STATE_QDISC_RUNNING, &dev_queue->state))
 			yield();
 
 		/*
 		 * Double-check inside queue lock to ensure that all effects
 		 * of the queue run are visible when we return.
 		 */
-		spin_lock_bh(&dev->tx_queue.lock);
-		running = test_bit(__LINK_STATE_QDISC_RUNNING, &dev->state);
-		spin_unlock_bh(&dev->tx_queue.lock);
+		spin_lock_bh(&dev_queue->lock);
+		running = test_bit(__QUEUE_STATE_QDISC_RUNNING,
+				   &dev_queue->state);
+		spin_unlock_bh(&dev_queue->lock);
 
 		/*
 		 * The running flag should never be set at this point because
-- 
cgit v1.2.3


From d8156534040996f6a93a24d3592d5d587f2587e5 Mon Sep 17 00:00:00 2001
From: Alexander Duyck <alexander.h.duyck@intel.com>
Date: Tue, 8 Jul 2008 15:13:05 -0700
Subject: net: add netif_napi_del function to allow for removal of napistructs

Adds netif_napi_del function which is used to remove the napi struct from
the netdev napi_list in cases where CONFIG_NETPOLL was enabled.
The motivation for adding this is to handle the case in which the number of
queues on a device changes due to a configuration change.  Previously the
napi structs for each queue would be left in the list until the netdev was
freed.

Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
---
 include/linux/netdevice.h | 13 +++++++++++++
 1 file changed, 13 insertions(+)

(limited to 'include')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 203c5504fe43..b54ec16dfbda 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -828,6 +828,19 @@ static inline void netif_napi_add(struct net_device *dev,
 	set_bit(NAPI_STATE_SCHED, &napi->state);
 }
 
+/**
+ *  netif_napi_del - remove a napi context
+ *  @napi: napi context
+ *
+ *  netif_napi_del() removes a napi context from the network device napi list
+ */
+static inline void netif_napi_del(struct napi_struct *napi)
+{
+#ifdef CONFIG_NETPOLL
+	list_del(&napi->dev_list);
+#endif
+}
+
 struct packet_type {
 	__be16			type;	/* This is really htons(ether_type). */
 	struct net_device	*dev;	/* NULL is wildcarded here	     */
-- 
cgit v1.2.3


From 5b5d0e704880addfd979c262e6441f126708539c Mon Sep 17 00:00:00 2001
From: Gerrit Renker <gerrit@erg.abdn.ac.uk>
Date: Sun, 13 Jul 2008 11:51:40 +0100
Subject: dccp: Upgrade NDP count from 3 to 6 bytes

RFC 4340, 7.7 specifies up to 6 bytes for the NDP Count option, whereas the code
is currently limited to up to 3 bytes. This seems to be a relict of an earlier
draft version and is brought up to date by the patch.

Signed-off-by: Gerrit Renker <gerrit@erg.abdn.ac.uk>
---
 include/linux/dccp.h                |  6 ++----
 net/dccp/ccids/ccid3.c              |  2 +-
 net/dccp/ccids/lib/packet_history.c |  6 +++---
 net/dccp/ccids/lib/packet_history.h |  6 +++---
 net/dccp/options.c                  | 14 ++++++++------
 5 files changed, 17 insertions(+), 17 deletions(-)

(limited to 'include')

diff --git a/include/linux/dccp.h b/include/linux/dccp.h
index aa0737019e37..6080449fbec9 100644
--- a/include/linux/dccp.h
+++ b/include/linux/dccp.h
@@ -364,8 +364,6 @@ static inline unsigned int dccp_hdr_len(const struct sk_buff *skb)
 /* FIXME: for now we're default to 1 but it should really be 0 */
 #define DCCPF_INITIAL_SEND_NDP_COUNT		1
 
-#define DCCP_NDP_LIMIT 0xFFFFFF
-
 /**
   * struct dccp_minisock - Minimal DCCP connection representation
   *
@@ -437,7 +435,7 @@ extern int dccp_parse_options(struct sock *sk, struct dccp_request_sock *dreq,
 			      struct sk_buff *skb);
 
 struct dccp_options_received {
-	u32	dccpor_ndp; /* only 24 bits */
+	u64	dccpor_ndp:48;
 	u32	dccpor_timestamp;
 	u32	dccpor_timestamp_echo;
 	u32	dccpor_elapsed_time;
@@ -533,7 +531,7 @@ struct dccp_sock {
 	__u16				dccps_r_ack_ratio;
 	__u16				dccps_pcslen;
 	__u16				dccps_pcrlen;
-	unsigned long			dccps_ndp_count;
+	__u64				dccps_ndp_count:48;
 	unsigned long			dccps_rate_last;
 	struct dccp_minisock		dccps_minisock;
 	struct dccp_ackvec		*dccps_hc_rx_ackvec;
diff --git a/net/dccp/ccids/ccid3.c b/net/dccp/ccids/ccid3.c
index a1929f33d703..523db262c18f 100644
--- a/net/dccp/ccids/ccid3.c
+++ b/net/dccp/ccids/ccid3.c
@@ -794,7 +794,7 @@ static void ccid3_hc_rx_packet_recv(struct sock *sk, struct sk_buff *skb)
 {
 	struct ccid3_hc_rx_sock *hcrx = ccid3_hc_rx_sk(sk);
 	enum ccid3_fback_type do_feedback = CCID3_FBACK_NONE;
-	const u32 ndp = dccp_sk(sk)->dccps_options_received.dccpor_ndp;
+	const u64 ndp = dccp_sk(sk)->dccps_options_received.dccpor_ndp;
 	const bool is_data_packet = dccp_data_packet(skb);
 
 	if (unlikely(hcrx->ccid3hcrx_state == TFRC_RSTATE_NO_DATA)) {
diff --git a/net/dccp/ccids/lib/packet_history.c b/net/dccp/ccids/lib/packet_history.c
index 8b5c41ec7ee1..712930564c38 100644
--- a/net/dccp/ccids/lib/packet_history.c
+++ b/net/dccp/ccids/lib/packet_history.c
@@ -153,7 +153,7 @@ void tfrc_rx_packet_history_exit(void)
 
 static inline void tfrc_rx_hist_entry_from_skb(struct tfrc_rx_hist_entry *entry,
 					       const struct sk_buff *skb,
-					       const u32 ndp)
+					       const u64 ndp)
 {
 	const struct dccp_hdr *dh = dccp_hdr(skb);
 
@@ -166,7 +166,7 @@ static inline void tfrc_rx_hist_entry_from_skb(struct tfrc_rx_hist_entry *entry,
 
 void tfrc_rx_hist_add_packet(struct tfrc_rx_hist *h,
 			     const struct sk_buff *skb,
-			     const u32 ndp)
+			     const u64 ndp)
 {
 	struct tfrc_rx_hist_entry *entry = tfrc_rx_hist_last_rcv(h);
 
@@ -356,7 +356,7 @@ static void __three_after_loss(struct tfrc_rx_hist *h)
  */
 int tfrc_rx_handle_loss(struct tfrc_rx_hist *h,
 			struct tfrc_loss_hist *lh,
-			struct sk_buff *skb, u32 ndp,
+			struct sk_buff *skb, const u64 ndp,
 			u32 (*calc_first_li)(struct sock *), struct sock *sk)
 {
 	int is_new_loss = 0;
diff --git a/net/dccp/ccids/lib/packet_history.h b/net/dccp/ccids/lib/packet_history.h
index c7eeda49cb20..6976156cda3c 100644
--- a/net/dccp/ccids/lib/packet_history.h
+++ b/net/dccp/ccids/lib/packet_history.h
@@ -64,7 +64,7 @@ struct tfrc_rx_hist_entry {
 	u64		 tfrchrx_seqno:48,
 			 tfrchrx_ccval:4,
 			 tfrchrx_type:4;
-	u32		 tfrchrx_ndp; /* In fact it is from 8 to 24 bits */
+	u64		 tfrchrx_ndp:48;
 	ktime_t		 tfrchrx_tstamp;
 };
 
@@ -145,14 +145,14 @@ static inline int tfrc_rx_hist_new_loss_indicated(struct tfrc_rx_hist *h,
 }
 
 extern void tfrc_rx_hist_add_packet(struct tfrc_rx_hist *h,
-				    const struct sk_buff *skb, const u32 ndp);
+				    const struct sk_buff *skb, const u64 ndp);
 
 extern int tfrc_rx_hist_duplicate(struct tfrc_rx_hist *h, struct sk_buff *skb);
 
 struct tfrc_loss_hist;
 extern int  tfrc_rx_handle_loss(struct tfrc_rx_hist *h,
 				struct tfrc_loss_hist *lh,
-				struct sk_buff *skb, u32 ndp,
+				struct sk_buff *skb, const u64 ndp,
 				u32 (*first_li)(struct sock *sk),
 				struct sock *sk);
 extern u32 tfrc_rx_hist_sample_rtt(struct tfrc_rx_hist *h,
diff --git a/net/dccp/options.c b/net/dccp/options.c
index 43bc24e761d0..dc7c158a2f4b 100644
--- a/net/dccp/options.c
+++ b/net/dccp/options.c
@@ -124,12 +124,12 @@ int dccp_parse_options(struct sock *sk, struct dccp_request_sock *dreq,
 				mandatory = 1;
 			break;
 		case DCCPO_NDP_COUNT:
-			if (len > 3)
+			if (len > 6)
 				goto out_invalid_option;
 
 			opt_recv->dccpor_ndp = dccp_decode_value_var(value, len);
-			dccp_pr_debug("%s rx opt: NDP count=%d\n", dccp_role(sk),
-				      opt_recv->dccpor_ndp);
+			dccp_pr_debug("%s opt: NDP count=%llu\n", dccp_role(sk),
+				      (unsigned long long)opt_recv->dccpor_ndp);
 			break;
 		case DCCPO_CHANGE_L:
 			/* fall through */
@@ -307,9 +307,11 @@ static void dccp_encode_value_var(const u32 value, unsigned char *to,
 		*to++ = (value & 0xFF);
 }
 
-static inline int dccp_ndp_len(const int ndp)
+static inline u8 dccp_ndp_len(const u64 ndp)
 {
-	return likely(ndp <= 0xFF) ? 1 : ndp <= 0xFFFF ? 2 : 3;
+	if (likely(ndp <= 0xFF))
+		return 1;
+	return likely(ndp <= USHORT_MAX) ? 2 : (ndp <= UINT_MAX ? 4 : 6);
 }
 
 int dccp_insert_option(struct sock *sk, struct sk_buff *skb,
@@ -336,7 +338,7 @@ EXPORT_SYMBOL_GPL(dccp_insert_option);
 static int dccp_insert_option_ndp(struct sock *sk, struct sk_buff *skb)
 {
 	struct dccp_sock *dp = dccp_sk(sk);
-	int ndp = dp->dccps_ndp_count;
+	u64 ndp = dp->dccps_ndp_count;
 
 	if (dccp_non_data_packet(skb))
 		++dp->dccps_ndp_count;
-- 
cgit v1.2.3


From 77db1980565626471a980f0d2d17299e4bd5e7a5 Mon Sep 17 00:00:00 2001
From: Marcel Holtmann <marcel@holtmann.org>
Date: Mon, 14 Jul 2008 20:13:45 +0200
Subject: [Bluetooth] Enforce security for outgoing RFCOMM connections

Recent tests with various Bluetooth headsets have shown that some of
them don't enforce authentication and encryption when connecting. All
of them leave it up to the host stack to enforce it. Non of them should
allow unencrypted connections, but that is how it is. So in case the
link mode settings require authentication and/or encryption it will now
also be enforced on outgoing RFCOMM connections. Previously this was
only done for incoming connections.

This support has a small drawback from a protocol level point of view
since the host stack can't really tell with 100% certainty if a remote
side is already authenticated or not. So if both sides are configured
to enforce authentication it will be requested twice. Most Bluetooth
chips are caching this information and thus no extra authentication
procedure has to be triggered over-the-air, but it can happen.

Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
---
 include/net/bluetooth/rfcomm.h |  1 +
 net/bluetooth/rfcomm/core.c    | 77 +++++++++++++++++++++++++-----------------
 net/bluetooth/rfcomm/sock.c    |  8 +++--
 3 files changed, 52 insertions(+), 34 deletions(-)

(limited to 'include')

diff --git a/include/net/bluetooth/rfcomm.h b/include/net/bluetooth/rfcomm.h
index 98ec7a320689..8c54ff37ad4f 100644
--- a/include/net/bluetooth/rfcomm.h
+++ b/include/net/bluetooth/rfcomm.h
@@ -181,6 +181,7 @@ struct rfcomm_dlc {
 	u8            priority;
 	u8            v24_sig;
 	u8            mscex;
+	u8            out;
 
 	u32           link_mode;
 
diff --git a/net/bluetooth/rfcomm/core.c b/net/bluetooth/rfcomm/core.c
index 0c2c93735e93..1f92f9ab4959 100644
--- a/net/bluetooth/rfcomm/core.c
+++ b/net/bluetooth/rfcomm/core.c
@@ -53,7 +53,7 @@
 #define BT_DBG(D...)
 #endif
 
-#define VERSION "1.8"
+#define VERSION "1.9"
 
 static int disable_cfc = 0;
 static int channel_mtu = -1;
@@ -230,6 +230,21 @@ static int rfcomm_l2sock_create(struct socket **sock)
 	return err;
 }
 
+static inline int rfcomm_check_link_mode(struct rfcomm_dlc *d)
+{
+	struct sock *sk = d->session->sock->sk;
+
+	if (d->link_mode & (RFCOMM_LM_ENCRYPT | RFCOMM_LM_SECURE)) {
+		if (!hci_conn_encrypt(l2cap_pi(sk)->conn->hcon))
+			return 1;
+	} else if (d->link_mode & RFCOMM_LM_AUTH) {
+		if (!hci_conn_auth(l2cap_pi(sk)->conn->hcon))
+			return 1;
+	}
+
+	return 0;
+}
+
 /* ---- RFCOMM DLCs ---- */
 static void rfcomm_dlc_timeout(unsigned long arg)
 {
@@ -371,15 +386,23 @@ static int __rfcomm_dlc_open(struct rfcomm_dlc *d, bdaddr_t *src, bdaddr_t *dst,
 	d->addr     = __addr(s->initiator, dlci);
 	d->priority = 7;
 
-	d->state    = BT_CONFIG;
+	d->state = BT_CONFIG;
 	rfcomm_dlc_link(s, d);
 
+	d->out = 1;
+
 	d->mtu = s->mtu;
 	d->cfc = (s->cfc == RFCOMM_CFC_UNKNOWN) ? 0 : s->cfc;
 
-	if (s->state == BT_CONNECTED)
-		rfcomm_send_pn(s, 1, d);
+	if (s->state == BT_CONNECTED) {
+		if (rfcomm_check_link_mode(d))
+			set_bit(RFCOMM_AUTH_PENDING, &d->flags);
+		else
+			rfcomm_send_pn(s, 1, d);
+	}
+
 	rfcomm_dlc_set_timer(d, RFCOMM_CONN_TIMEOUT);
+
 	return 0;
 }
 
@@ -1146,21 +1169,6 @@ static int rfcomm_recv_disc(struct rfcomm_session *s, u8 dlci)
 	return 0;
 }
 
-static inline int rfcomm_check_link_mode(struct rfcomm_dlc *d)
-{
-	struct sock *sk = d->session->sock->sk;
-
-	if (d->link_mode & (RFCOMM_LM_ENCRYPT | RFCOMM_LM_SECURE)) {
-		if (!hci_conn_encrypt(l2cap_pi(sk)->conn->hcon))
-			return 1;
-	} else if (d->link_mode & RFCOMM_LM_AUTH) {
-		if (!hci_conn_auth(l2cap_pi(sk)->conn->hcon))
-			return 1;
-	}
-
-	return 0;
-}
-
 static void rfcomm_dlc_accept(struct rfcomm_dlc *d)
 {
 	struct sock *sk = d->session->sock->sk;
@@ -1205,10 +1213,8 @@ static int rfcomm_recv_sabm(struct rfcomm_session *s, u8 dlci)
 			if (rfcomm_check_link_mode(d)) {
 				set_bit(RFCOMM_AUTH_PENDING, &d->flags);
 				rfcomm_dlc_set_timer(d, RFCOMM_AUTH_TIMEOUT);
-				return 0;
-			}
-
-			rfcomm_dlc_accept(d);
+			} else
+				rfcomm_dlc_accept(d);
 		}
 		return 0;
 	}
@@ -1223,10 +1229,8 @@ static int rfcomm_recv_sabm(struct rfcomm_session *s, u8 dlci)
 		if (rfcomm_check_link_mode(d)) {
 			set_bit(RFCOMM_AUTH_PENDING, &d->flags);
 			rfcomm_dlc_set_timer(d, RFCOMM_AUTH_TIMEOUT);
-			return 0;
-		}
-
-		rfcomm_dlc_accept(d);
+		} else
+			rfcomm_dlc_accept(d);
 	} else {
 		rfcomm_send_dm(s, dlci);
 	}
@@ -1636,7 +1640,11 @@ static void rfcomm_process_connect(struct rfcomm_session *s)
 		d = list_entry(p, struct rfcomm_dlc, list);
 		if (d->state == BT_CONFIG) {
 			d->mtu = s->mtu;
-			rfcomm_send_pn(s, 1, d);
+			if (rfcomm_check_link_mode(d)) {
+				set_bit(RFCOMM_AUTH_PENDING, &d->flags);
+				rfcomm_dlc_set_timer(d, RFCOMM_AUTH_TIMEOUT);
+			} else
+				rfcomm_send_pn(s, 1, d);
 		}
 	}
 }
@@ -1709,7 +1717,11 @@ static inline void rfcomm_process_dlcs(struct rfcomm_session *s)
 
 		if (test_and_clear_bit(RFCOMM_AUTH_ACCEPT, &d->flags)) {
 			rfcomm_dlc_clear_timer(d);
-			rfcomm_dlc_accept(d);
+			if (d->out) {
+				rfcomm_send_pn(s, 1, d);
+				rfcomm_dlc_set_timer(d, RFCOMM_CONN_TIMEOUT);
+			} else
+				rfcomm_dlc_accept(d);
 			if (d->link_mode & RFCOMM_LM_SECURE) {
 				struct sock *sk = s->sock->sk;
 				hci_conn_change_link_key(l2cap_pi(sk)->conn->hcon);
@@ -1717,7 +1729,10 @@ static inline void rfcomm_process_dlcs(struct rfcomm_session *s)
 			continue;
 		} else if (test_and_clear_bit(RFCOMM_AUTH_REJECT, &d->flags)) {
 			rfcomm_dlc_clear_timer(d);
-			rfcomm_send_dm(s, d->dlci);
+			if (!d->out)
+				rfcomm_send_dm(s, d->dlci);
+			else
+				d->state = BT_CLOSED;
 			__rfcomm_dlc_close(d, ECONNREFUSED);
 			continue;
 		}
@@ -1726,7 +1741,7 @@ static inline void rfcomm_process_dlcs(struct rfcomm_session *s)
 			continue;
 
 		if ((d->state == BT_CONNECTED || d->state == BT_DISCONN) &&
-				d->mscex == RFCOMM_MSCEX_OK)
+						d->mscex == RFCOMM_MSCEX_OK)
 			rfcomm_process_tx(d);
 	}
 }
diff --git a/net/bluetooth/rfcomm/sock.c b/net/bluetooth/rfcomm/sock.c
index 5083adcbfae5..cacb1ab51f99 100644
--- a/net/bluetooth/rfcomm/sock.c
+++ b/net/bluetooth/rfcomm/sock.c
@@ -309,13 +309,13 @@ static struct sock *rfcomm_sock_alloc(struct net *net, struct socket *sock, int
 	sk->sk_destruct = rfcomm_sock_destruct;
 	sk->sk_sndtimeo = RFCOMM_CONN_TIMEOUT;
 
-	sk->sk_sndbuf   = RFCOMM_MAX_CREDITS * RFCOMM_DEFAULT_MTU * 10;
-	sk->sk_rcvbuf   = RFCOMM_MAX_CREDITS * RFCOMM_DEFAULT_MTU * 10;
+	sk->sk_sndbuf = RFCOMM_MAX_CREDITS * RFCOMM_DEFAULT_MTU * 10;
+	sk->sk_rcvbuf = RFCOMM_MAX_CREDITS * RFCOMM_DEFAULT_MTU * 10;
 
 	sock_reset_flag(sk, SOCK_ZAPPED);
 
 	sk->sk_protocol = proto;
-	sk->sk_state	= BT_OPEN;
+	sk->sk_state    = BT_OPEN;
 
 	bt_sock_link(&rfcomm_sk_list, sk);
 
@@ -413,6 +413,8 @@ static int rfcomm_sock_connect(struct socket *sock, struct sockaddr *addr, int a
 	bacpy(&bt_sk(sk)->dst, &sa->rc_bdaddr);
 	rfcomm_pi(sk)->channel = sa->rc_channel;
 
+	d->link_mode = rfcomm_pi(sk)->link_mode;
+
 	err = rfcomm_dlc_open(d, &bt_sk(sk)->src, &sa->rc_bdaddr, sa->rc_channel);
 	if (!err)
 		err = bt_sock_wait_state(sk, BT_CONNECTED,
-- 
cgit v1.2.3


From 9719f8afce34d3d04e884873a8a5e3483e30974c Mon Sep 17 00:00:00 2001
From: Marcel Holtmann <marcel@holtmann.org>
Date: Mon, 14 Jul 2008 20:13:45 +0200
Subject: [Bluetooth] Disconnect when encryption gets disabled

The Bluetooth specification allows to enable or disable the encryption
of an ACL link at any time by either the peer or the remote device. If
a L2CAP or RFCOMM connection requested an encrypted link, they will now
disconnect that link if the encryption gets disabled. Higher protocols
that don't care about encryption (like SDP) are not affected.

Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
---
 include/net/bluetooth/hci_core.h | 10 +++++-----
 net/bluetooth/l2cap.c            | 13 ++++++++++++-
 net/bluetooth/rfcomm/core.c      |  8 ++++++++
 3 files changed, 25 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h
index ea13baa3851b..c8255adee8f5 100644
--- a/include/net/bluetooth/hci_core.h
+++ b/include/net/bluetooth/hci_core.h
@@ -474,7 +474,7 @@ struct hci_proto {
 	int (*recv_acldata)	(struct hci_conn *conn, struct sk_buff *skb, __u16 flags);
 	int (*recv_scodata)	(struct hci_conn *conn, struct sk_buff *skb);
 	int (*auth_cfm)		(struct hci_conn *conn, __u8 status);
-	int (*encrypt_cfm)	(struct hci_conn *conn, __u8 status);
+	int (*encrypt_cfm)	(struct hci_conn *conn, __u8 status, __u8 encrypt);
 };
 
 static inline int hci_proto_connect_ind(struct hci_dev *hdev, bdaddr_t *bdaddr, __u8 type)
@@ -532,17 +532,17 @@ static inline void hci_proto_auth_cfm(struct hci_conn *conn, __u8 status)
 		hp->auth_cfm(conn, status);
 }
 
-static inline void hci_proto_encrypt_cfm(struct hci_conn *conn, __u8 status)
+static inline void hci_proto_encrypt_cfm(struct hci_conn *conn, __u8 status, __u8 encrypt)
 {
 	register struct hci_proto *hp;
 
 	hp = hci_proto[HCI_PROTO_L2CAP];
 	if (hp && hp->encrypt_cfm)
-		hp->encrypt_cfm(conn, status);
+		hp->encrypt_cfm(conn, status, encrypt);
 
 	hp = hci_proto[HCI_PROTO_SCO];
 	if (hp && hp->encrypt_cfm)
-		hp->encrypt_cfm(conn, status);
+		hp->encrypt_cfm(conn, status, encrypt);
 }
 
 int hci_register_proto(struct hci_proto *hproto);
@@ -579,7 +579,7 @@ static inline void hci_encrypt_cfm(struct hci_conn *conn, __u8 status, __u8 encr
 {
 	struct list_head *p;
 
-	hci_proto_encrypt_cfm(conn, status);
+	hci_proto_encrypt_cfm(conn, status, encrypt);
 
 	read_lock_bh(&hci_cb_list_lock);
 	list_for_each(p, &hci_cb_list) {
diff --git a/net/bluetooth/l2cap.c b/net/bluetooth/l2cap.c
index 2e3abdfbd69d..252264062f59 100644
--- a/net/bluetooth/l2cap.c
+++ b/net/bluetooth/l2cap.c
@@ -2197,7 +2197,7 @@ static int l2cap_auth_cfm(struct hci_conn *hcon, u8 status)
 	return 0;
 }
 
-static int l2cap_encrypt_cfm(struct hci_conn *hcon, u8 status)
+static int l2cap_encrypt_cfm(struct hci_conn *hcon, u8 status, u8 encrypt)
 {
 	struct l2cap_chan_list *l;
 	struct l2cap_conn *conn = hcon->l2cap_data;
@@ -2215,8 +2215,19 @@ static int l2cap_encrypt_cfm(struct hci_conn *hcon, u8 status)
 	read_lock(&l->lock);
 
 	for (sk = l->head; sk; sk = l2cap_pi(sk)->next_c) {
+		struct l2cap_pinfo *pi = l2cap_pi(sk);
+
 		bh_lock_sock(sk);
 
+		if ((pi->link_mode & (L2CAP_LM_ENCRYPT | L2CAP_LM_SECURE)) &&
+					(sk->sk_state == BT_CONNECTED ||
+						sk->sk_state == BT_CONFIG) &&
+						!status && encrypt == 0x00) {
+			__l2cap_sock_close(sk, ECONNREFUSED);
+			bh_unlock_sock(sk);
+			continue;
+		}
+
 		if (sk->sk_state != BT_CONNECT2) {
 			bh_unlock_sock(sk);
 			continue;
diff --git a/net/bluetooth/rfcomm/core.c b/net/bluetooth/rfcomm/core.c
index 1f92f9ab4959..e7a6a03cea37 100644
--- a/net/bluetooth/rfcomm/core.c
+++ b/net/bluetooth/rfcomm/core.c
@@ -2003,6 +2003,14 @@ static void rfcomm_encrypt_cfm(struct hci_conn *conn, u8 status, u8 encrypt)
 	list_for_each_safe(p, n, &s->dlcs) {
 		d = list_entry(p, struct rfcomm_dlc, list);
 
+		if ((d->link_mode & (RFCOMM_LM_ENCRYPT | RFCOMM_LM_SECURE)) &&
+					(d->state == BT_CONNECTED ||
+						d->state == BT_CONFIG) &&
+						!status && encrypt == 0x00) {
+			__rfcomm_dlc_close(d, ECONNREFUSED);
+			continue;
+		}
+
 		if (!test_and_clear_bit(RFCOMM_AUTH_PENDING, &d->flags))
 			continue;
 
-- 
cgit v1.2.3


From a8746417e864da1ed36dd2432a399fbeb843c2a0 Mon Sep 17 00:00:00 2001
From: Marcel Holtmann <marcel@holtmann.org>
Date: Mon, 14 Jul 2008 20:13:46 +0200
Subject: [Bluetooth] Track connection packet type changes

The connection packet type can be changed after the connection has been
established and thus needs to be properly tracked to ensure that the
host stack has always correct and valid information about it.

On incoming connections the Bluetooth core switches the supported packet
types to the configured list for this controller. However the usefulness
of this feature has been questioned a lot. The general consent is that
every Bluetooth host stack should enable as many packet types as the
hardware actually supports and leave the decision to the link manager
software running on the Bluetooth chip.

When running on Bluetooth 2.0 or later hardware, don't change the packet
type for incoming connections anymore. This hardware likely supports
Enhanced Data Rate and thus leave it completely up to the link manager
to pick the best packet type.

Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
---
 include/net/bluetooth/hci.h      |  9 +++++++++
 include/net/bluetooth/hci_core.h |  1 +
 net/bluetooth/hci_conn.c         | 32 ++++++++++++++++++++++++--------
 net/bluetooth/hci_event.c        | 32 +++++++++++++++++++++++++-------
 4 files changed, 59 insertions(+), 15 deletions(-)

(limited to 'include')

diff --git a/include/net/bluetooth/hci.h b/include/net/bluetooth/hci.h
index a8a9eb6af966..f1dc174abc2a 100644
--- a/include/net/bluetooth/hci.h
+++ b/include/net/bluetooth/hci.h
@@ -137,6 +137,8 @@ enum {
 #define ESCO_EV4	0x0010
 #define ESCO_EV5	0x0020
 
+#define SCO_ESCO_MASK  (ESCO_HV1 | ESCO_HV2 | ESCO_HV3)
+
 /* ACL flags */
 #define ACL_CONT		0x01
 #define ACL_START		0x02
@@ -696,6 +698,13 @@ struct hci_ev_clock_offset {
 	__le16   clock_offset;
 } __attribute__ ((packed));
 
+#define HCI_EV_PKT_TYPE_CHANGE		0x1d
+struct hci_ev_pkt_type_change {
+	__u8     status;
+	__le16   handle;
+	__le16   pkt_type;
+} __attribute__ ((packed));
+
 #define HCI_EV_PSCAN_REP_MODE		0x20
 struct hci_ev_pscan_rep_mode {
 	bdaddr_t bdaddr;
diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h
index c8255adee8f5..6424d63e3395 100644
--- a/include/net/bluetooth/hci_core.h
+++ b/include/net/bluetooth/hci_core.h
@@ -162,6 +162,7 @@ struct hci_conn {
 	__u8		 dev_class[3];
 	__u8             features[8];
 	__u16            interval;
+	__u16            pkt_type;
 	__u16            link_policy;
 	__u32		 link_mode;
 	__u8             power_save;
diff --git a/net/bluetooth/hci_conn.c b/net/bluetooth/hci_conn.c
index f8880261da0e..69c64ce054fb 100644
--- a/net/bluetooth/hci_conn.c
+++ b/net/bluetooth/hci_conn.c
@@ -59,7 +59,8 @@ void hci_acl_connect(struct hci_conn *conn)
 	BT_DBG("%p", conn);
 
 	conn->state = BT_CONNECT;
-	conn->out   = 1;
+	conn->out = 1;
+
 	conn->link_mode = HCI_LM_MASTER;
 
 	conn->attempt++;
@@ -76,7 +77,7 @@ void hci_acl_connect(struct hci_conn *conn)
 		memcpy(conn->dev_class, ie->data.dev_class, 3);
 	}
 
-	cp.pkt_type = cpu_to_le16(hdev->pkt_type & ACL_PTYPE_MASK);
+	cp.pkt_type = cpu_to_le16(conn->pkt_type);
 	if (lmp_rswitch_capable(hdev) && !(hdev->link_mode & HCI_LM_MASTER))
 		cp.role_switch = 0x01;
 	else
@@ -122,7 +123,7 @@ void hci_add_sco(struct hci_conn *conn, __u16 handle)
 	conn->out = 1;
 
 	cp.handle   = cpu_to_le16(handle);
-	cp.pkt_type = cpu_to_le16(hdev->pkt_type & SCO_PTYPE_MASK);
+	cp.pkt_type = cpu_to_le16(conn->pkt_type);
 
 	hci_send_cmd(hdev, HCI_OP_ADD_SCO, sizeof(cp), &cp);
 }
@@ -138,7 +139,7 @@ void hci_setup_sync(struct hci_conn *conn, __u16 handle)
 	conn->out = 1;
 
 	cp.handle   = cpu_to_le16(handle);
-	cp.pkt_type = cpu_to_le16(hdev->esco_type);
+	cp.pkt_type = cpu_to_le16(conn->pkt_type);
 
 	cp.tx_bandwidth   = cpu_to_le32(0x00001f40);
 	cp.rx_bandwidth   = cpu_to_le32(0x00001f40);
@@ -199,13 +200,28 @@ struct hci_conn *hci_conn_add(struct hci_dev *hdev, int type, bdaddr_t *dst)
 		return NULL;
 
 	bacpy(&conn->dst, dst);
-	conn->hdev   = hdev;
-	conn->type   = type;
-	conn->mode   = HCI_CM_ACTIVE;
-	conn->state  = BT_OPEN;
+	conn->hdev  = hdev;
+	conn->type  = type;
+	conn->mode  = HCI_CM_ACTIVE;
+	conn->state = BT_OPEN;
 
 	conn->power_save = 1;
 
+	switch (type) {
+	case ACL_LINK:
+		conn->pkt_type = hdev->pkt_type & ACL_PTYPE_MASK;
+		break;
+	case SCO_LINK:
+		if (lmp_esco_capable(hdev))
+			conn->pkt_type = hdev->esco_type & SCO_ESCO_MASK;
+		else
+			conn->pkt_type = hdev->pkt_type & SCO_PTYPE_MASK;
+		break;
+	case ESCO_LINK:
+		conn->pkt_type = hdev->esco_type;
+		break;
+	}
+
 	skb_queue_head_init(&conn->data_q);
 
 	setup_timer(&conn->disc_timer, hci_conn_timeout, (unsigned long)conn);
diff --git a/net/bluetooth/hci_event.c b/net/bluetooth/hci_event.c
index 6bc5a0506c6c..d4d2dcc40fc7 100644
--- a/net/bluetooth/hci_event.c
+++ b/net/bluetooth/hci_event.c
@@ -699,14 +699,12 @@ static inline void hci_conn_complete_evt(struct hci_dev *hdev, struct sk_buff *s
 		}
 
 		/* Set packet type for incoming connection */
-		if (!conn->out) {
+		if (!conn->out && hdev->hci_ver < 3) {
 			struct hci_cp_change_conn_ptype cp;
 			cp.handle = ev->handle;
-			cp.pkt_type = (conn->type == ACL_LINK) ?
-				cpu_to_le16(hdev->pkt_type & ACL_PTYPE_MASK):
-				cpu_to_le16(hdev->pkt_type & SCO_PTYPE_MASK);
-
-			hci_send_cmd(hdev, HCI_OP_CHANGE_CONN_PTYPE, sizeof(cp), &cp);
+			cp.pkt_type = cpu_to_le16(conn->pkt_type);
+			hci_send_cmd(hdev, HCI_OP_CHANGE_CONN_PTYPE,
+							sizeof(cp), &cp);
 		} else {
 			/* Update disconnect timer */
 			hci_conn_hold(conn);
@@ -786,7 +784,7 @@ static inline void hci_conn_request_evt(struct hci_dev *hdev, struct sk_buff *sk
 			struct hci_cp_accept_sync_conn_req cp;
 
 			bacpy(&cp.bdaddr, &ev->bdaddr);
-			cp.pkt_type = cpu_to_le16(hdev->esco_type);
+			cp.pkt_type = cpu_to_le16(conn->pkt_type);
 
 			cp.tx_bandwidth   = cpu_to_le32(0x00001f40);
 			cp.rx_bandwidth   = cpu_to_le32(0x00001f40);
@@ -1237,6 +1235,22 @@ static inline void hci_clock_offset_evt(struct hci_dev *hdev, struct sk_buff *sk
 	hci_dev_unlock(hdev);
 }
 
+static inline void hci_pkt_type_change_evt(struct hci_dev *hdev, struct sk_buff *skb)
+{
+	struct hci_ev_pkt_type_change *ev = (void *) skb->data;
+	struct hci_conn *conn;
+
+	BT_DBG("%s status %d", hdev->name, ev->status);
+
+	hci_dev_lock(hdev);
+
+	conn = hci_conn_hash_lookup_handle(hdev, __le16_to_cpu(ev->handle));
+	if (conn && !ev->status)
+		conn->pkt_type = __le16_to_cpu(ev->pkt_type);
+
+	hci_dev_unlock(hdev);
+}
+
 static inline void hci_pscan_rep_mode_evt(struct hci_dev *hdev, struct sk_buff *skb)
 {
 	struct hci_ev_pscan_rep_mode *ev = (void *) skb->data;
@@ -1480,6 +1494,10 @@ void hci_event_packet(struct hci_dev *hdev, struct sk_buff *skb)
 		hci_clock_offset_evt(hdev, skb);
 		break;
 
+	case HCI_EV_PKT_TYPE_CHANGE:
+		hci_pkt_type_change_evt(hdev, skb);
+		break;
+
 	case HCI_EV_PSCAN_REP_MODE:
 		hci_pscan_rep_mode_evt(hdev, skb);
 		break;
-- 
cgit v1.2.3


From e4e8e37c42bdaaefcb84eeaef0dc1bc3f696f8f6 Mon Sep 17 00:00:00 2001
From: Marcel Holtmann <marcel@holtmann.org>
Date: Mon, 14 Jul 2008 20:13:47 +0200
Subject: [Bluetooth] Make use of the default link policy settings

The Bluetooth specification supports the default link policy settings
on a per host controller basis. For every new connection the link
manager would then use these settings. It is better to use this instead
of bothering the controller on every connection setup to overwrite the
default settings.

Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
---
 include/net/bluetooth/hci.h | 11 +++++++
 net/bluetooth/hci_conn.c    |  2 ++
 net/bluetooth/hci_core.c    | 35 ++++++++++++++-------
 net/bluetooth/hci_event.c   | 75 +++++++++++++++++++++++++++++++++++++--------
 4 files changed, 100 insertions(+), 23 deletions(-)

(limited to 'include')

diff --git a/include/net/bluetooth/hci.h b/include/net/bluetooth/hci.h
index f1dc174abc2a..efc8c555c182 100644
--- a/include/net/bluetooth/hci.h
+++ b/include/net/bluetooth/hci.h
@@ -404,6 +404,17 @@ struct hci_rp_write_link_policy {
 	__le16   handle;
 } __attribute__ ((packed));
 
+#define HCI_OP_READ_DEF_LINK_POLICY	0x080e
+struct hci_rp_read_def_link_policy {
+	__u8     status;
+	__le16   policy;
+} __attribute__ ((packed));
+
+#define HCI_OP_WRITE_DEF_LINK_POLICY	0x080f
+struct hci_cp_write_def_link_policy {
+	__le16   policy;
+} __attribute__ ((packed));
+
 #define HCI_OP_SNIFF_SUBRATE		0x0811
 struct hci_cp_sniff_subrate {
 	__le16   handle;
diff --git a/net/bluetooth/hci_conn.c b/net/bluetooth/hci_conn.c
index 69c64ce054fb..6175ce841e9e 100644
--- a/net/bluetooth/hci_conn.c
+++ b/net/bluetooth/hci_conn.c
@@ -65,6 +65,8 @@ void hci_acl_connect(struct hci_conn *conn)
 
 	conn->attempt++;
 
+	conn->link_policy = hdev->link_policy;
+
 	memset(&cp, 0, sizeof(cp));
 	bacpy(&cp.bdaddr, &conn->dst);
 	cp.pscan_rep_mode = 0x02;
diff --git a/net/bluetooth/hci_core.c b/net/bluetooth/hci_core.c
index aec6929f5c16..69b2c1aac08a 100644
--- a/net/bluetooth/hci_core.c
+++ b/net/bluetooth/hci_core.c
@@ -279,10 +279,20 @@ static void hci_encrypt_req(struct hci_dev *hdev, unsigned long opt)
 
 	BT_DBG("%s %x", hdev->name, encrypt);
 
-	/* Authentication */
+	/* Encryption */
 	hci_send_cmd(hdev, HCI_OP_WRITE_ENCRYPT_MODE, 1, &encrypt);
 }
 
+static void hci_linkpol_req(struct hci_dev *hdev, unsigned long opt)
+{
+	__le16 policy = cpu_to_le16(opt);
+
+	BT_DBG("%s %x", hdev->name, opt);
+
+	/* Default link policy */
+	hci_send_cmd(hdev, HCI_OP_WRITE_DEF_LINK_POLICY, 2, &policy);
+}
+
 /* Get HCI device by index.
  * Device is held on return. */
 struct hci_dev *hci_dev_get(int index)
@@ -694,32 +704,35 @@ int hci_dev_cmd(unsigned int cmd, void __user *arg)
 					msecs_to_jiffies(HCI_INIT_TIMEOUT));
 		break;
 
-	case HCISETPTYPE:
-		hdev->pkt_type = (__u16) dr.dev_opt;
-		break;
-
 	case HCISETLINKPOL:
-		hdev->link_policy = (__u16) dr.dev_opt;
+		err = hci_request(hdev, hci_linkpol_req, dr.dev_opt,
+					msecs_to_jiffies(HCI_INIT_TIMEOUT));
 		break;
 
 	case HCISETLINKMODE:
-		hdev->link_mode = ((__u16) dr.dev_opt) & (HCI_LM_MASTER | HCI_LM_ACCEPT);
+		hdev->link_mode = ((__u16) dr.dev_opt) &
+					(HCI_LM_MASTER | HCI_LM_ACCEPT);
+		break;
+
+	case HCISETPTYPE:
+		hdev->pkt_type = (__u16) dr.dev_opt;
 		break;
 
 	case HCISETACLMTU:
-		hdev->acl_mtu  = *((__u16 *)&dr.dev_opt + 1);
-		hdev->acl_pkts = *((__u16 *)&dr.dev_opt + 0);
+		hdev->acl_mtu  = *((__u16 *) &dr.dev_opt + 1);
+		hdev->acl_pkts = *((__u16 *) &dr.dev_opt + 0);
 		break;
 
 	case HCISETSCOMTU:
-		hdev->sco_mtu  = *((__u16 *)&dr.dev_opt + 1);
-		hdev->sco_pkts = *((__u16 *)&dr.dev_opt + 0);
+		hdev->sco_mtu  = *((__u16 *) &dr.dev_opt + 1);
+		hdev->sco_pkts = *((__u16 *) &dr.dev_opt + 0);
 		break;
 
 	default:
 		err = -EINVAL;
 		break;
 	}
+
 	hci_dev_put(hdev);
 	return err;
 }
diff --git a/net/bluetooth/hci_event.c b/net/bluetooth/hci_event.c
index d4d2dcc40fc7..9af181a61650 100644
--- a/net/bluetooth/hci_event.c
+++ b/net/bluetooth/hci_event.c
@@ -110,6 +110,25 @@ static void hci_cc_role_discovery(struct hci_dev *hdev, struct sk_buff *skb)
 	hci_dev_unlock(hdev);
 }
 
+static void hci_cc_read_link_policy(struct hci_dev *hdev, struct sk_buff *skb)
+{
+	struct hci_rp_read_link_policy *rp = (void *) skb->data;
+	struct hci_conn *conn;
+
+	BT_DBG("%s status 0x%x", hdev->name, rp->status);
+
+	if (rp->status)
+		return;
+
+	hci_dev_lock(hdev);
+
+	conn = hci_conn_hash_lookup_handle(hdev, __le16_to_cpu(rp->handle));
+	if (conn)
+		conn->link_policy = __le16_to_cpu(rp->policy);
+
+	hci_dev_unlock(hdev);
+}
+
 static void hci_cc_write_link_policy(struct hci_dev *hdev, struct sk_buff *skb)
 {
 	struct hci_rp_write_link_policy *rp = (void *) skb->data;
@@ -128,13 +147,41 @@ static void hci_cc_write_link_policy(struct hci_dev *hdev, struct sk_buff *skb)
 	hci_dev_lock(hdev);
 
 	conn = hci_conn_hash_lookup_handle(hdev, __le16_to_cpu(rp->handle));
-	if (conn) {
+	if (conn)
 		conn->link_policy = get_unaligned_le16(sent + 2);
-	}
 
 	hci_dev_unlock(hdev);
 }
 
+static void hci_cc_read_def_link_policy(struct hci_dev *hdev, struct sk_buff *skb)
+{
+	struct hci_rp_read_def_link_policy *rp = (void *) skb->data;
+
+	BT_DBG("%s status 0x%x", hdev->name, rp->status);
+
+	if (rp->status)
+		return;
+
+	hdev->link_policy = __le16_to_cpu(rp->policy);
+}
+
+static void hci_cc_write_def_link_policy(struct hci_dev *hdev, struct sk_buff *skb)
+{
+	__u8 status = *((__u8 *) skb->data);
+	void *sent;
+
+	BT_DBG("%s status 0x%x", hdev->name, status);
+
+	sent = hci_sent_cmd_data(hdev, HCI_OP_WRITE_DEF_LINK_POLICY);
+	if (!sent)
+		return;
+
+	if (!status)
+		hdev->link_policy = get_unaligned_le16(sent);
+
+	hci_req_complete(hdev, status);
+}
+
 static void hci_cc_reset(struct hci_dev *hdev, struct sk_buff *skb)
 {
 	__u8 status = *((__u8 *) skb->data);
@@ -347,8 +394,8 @@ static void hci_cc_read_local_version(struct hci_dev *hdev, struct sk_buff *skb)
 		return;
 
 	hdev->hci_ver = rp->hci_ver;
-	hdev->hci_rev = btohs(rp->hci_rev);
-	hdev->manufacturer = btohs(rp->manufacturer);
+	hdev->hci_rev = __le16_to_cpu(rp->hci_rev);
+	hdev->manufacturer = __le16_to_cpu(rp->manufacturer);
 
 	BT_DBG("%s manufacturer %d hci ver %d:%d", hdev->name,
 					hdev->manufacturer,
@@ -690,14 +737,6 @@ static inline void hci_conn_complete_evt(struct hci_dev *hdev, struct sk_buff *s
 			hci_send_cmd(hdev, HCI_OP_READ_REMOTE_FEATURES, sizeof(cp), &cp);
 		}
 
-		/* Set link policy */
-		if (conn->type == ACL_LINK && hdev->link_policy) {
-			struct hci_cp_write_link_policy cp;
-			cp.handle = ev->handle;
-			cp.policy = cpu_to_le16(hdev->link_policy);
-			hci_send_cmd(hdev, HCI_OP_WRITE_LINK_POLICY, sizeof(cp), &cp);
-		}
-
 		/* Set packet type for incoming connection */
 		if (!conn->out && hdev->hci_ver < 3) {
 			struct hci_cp_change_conn_ptype cp;
@@ -974,10 +1013,22 @@ static inline void hci_cmd_complete_evt(struct hci_dev *hdev, struct sk_buff *sk
 		hci_cc_role_discovery(hdev, skb);
 		break;
 
+	case HCI_OP_READ_LINK_POLICY:
+		hci_cc_read_link_policy(hdev, skb);
+		break;
+
 	case HCI_OP_WRITE_LINK_POLICY:
 		hci_cc_write_link_policy(hdev, skb);
 		break;
 
+	case HCI_OP_READ_DEF_LINK_POLICY:
+		hci_cc_read_def_link_policy(hdev, skb);
+		break;
+
+	case HCI_OP_WRITE_DEF_LINK_POLICY:
+		hci_cc_write_def_link_policy(hdev, skb);
+		break;
+
 	case HCI_OP_RESET:
 		hci_cc_reset(hdev, skb);
 		break;
-- 
cgit v1.2.3


From 0493684ed2397e111574f343534d8e4ec440dfa5 Mon Sep 17 00:00:00 2001
From: Marcel Holtmann <marcel@holtmann.org>
Date: Mon, 14 Jul 2008 20:13:48 +0200
Subject: [Bluetooth] Disable disconnect timer during Simple Pairing

During the Simple Pairing process the HCI disconnect timer must be
disabled. The way to do this is by holding a reference count of the
HCI connection. The Simple Pairing process on both sides starts with
an IO Capabilities Request and ends with Simple Pairing Complete.

Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
---
 include/net/bluetooth/hci.h | 11 +++++++++++
 net/bluetooth/hci_event.c   | 40 ++++++++++++++++++++++++++++++++++++++++
 2 files changed, 51 insertions(+)

(limited to 'include')

diff --git a/include/net/bluetooth/hci.h b/include/net/bluetooth/hci.h
index efc8c555c182..79629ff40e3e 100644
--- a/include/net/bluetooth/hci.h
+++ b/include/net/bluetooth/hci.h
@@ -794,6 +794,17 @@ struct extended_inquiry_info {
 	__u8     data[240];
 } __attribute__ ((packed));
 
+#define HCI_EV_IO_CAPA_REQUEST		0x31
+struct hci_ev_io_capa_request {
+	bdaddr_t bdaddr;
+} __attribute__ ((packed));
+
+#define HCI_EV_SIMPLE_PAIR_COMPLETE	0x36
+struct hci_ev_simple_pair_complete {
+	__u8     status;
+	bdaddr_t bdaddr;
+} __attribute__ ((packed));
+
 /* Internal events generated by Bluetooth stack */
 #define HCI_EV_STACK_INTERNAL	0xfd
 struct hci_ev_stack_internal {
diff --git a/net/bluetooth/hci_event.c b/net/bluetooth/hci_event.c
index e47676128bb5..7c9ac01cd8f9 100644
--- a/net/bluetooth/hci_event.c
+++ b/net/bluetooth/hci_event.c
@@ -1464,6 +1464,38 @@ static inline void hci_extended_inquiry_result_evt(struct hci_dev *hdev, struct
 	hci_dev_unlock(hdev);
 }
 
+static inline void hci_io_capa_request_evt(struct hci_dev *hdev, struct sk_buff *skb)
+{
+	struct hci_ev_io_capa_request *ev = (void *) skb->data;
+	struct hci_conn *conn;
+
+	BT_DBG("%s", hdev->name);
+
+	hci_dev_lock(hdev);
+
+	conn = hci_conn_hash_lookup_ba(hdev, ACL_LINK, &ev->bdaddr);
+	if (conn)
+		hci_conn_hold(conn);
+
+	hci_dev_unlock(hdev);
+}
+
+static inline void hci_simple_pair_complete_evt(struct hci_dev *hdev, struct sk_buff *skb)
+{
+	struct hci_ev_simple_pair_complete *ev = (void *) skb->data;
+	struct hci_conn *conn;
+
+	BT_DBG("%s", hdev->name);
+
+	hci_dev_lock(hdev);
+
+	conn = hci_conn_hash_lookup_ba(hdev, ACL_LINK, &ev->bdaddr);
+	if (conn)
+		hci_conn_put(conn);
+
+	hci_dev_unlock(hdev);
+}
+
 void hci_event_packet(struct hci_dev *hdev, struct sk_buff *skb)
 {
 	struct hci_event_hdr *hdr = (void *) skb->data;
@@ -1588,6 +1620,14 @@ void hci_event_packet(struct hci_dev *hdev, struct sk_buff *skb)
 		hci_extended_inquiry_result_evt(hdev, skb);
 		break;
 
+	case HCI_EV_IO_CAPA_REQUEST:
+		hci_io_capa_request_evt(hdev, skb);
+		break;
+
+	case HCI_EV_SIMPLE_PAIR_COMPLETE:
+		hci_simple_pair_complete_evt(hdev, skb);
+		break;
+
 	default:
 		BT_DBG("%s event 0x%x", hdev->name, event);
 		break;
-- 
cgit v1.2.3


From 333140b57fa7867bc92e5ee879b6ac4ef5e1d867 Mon Sep 17 00:00:00 2001
From: Marcel Holtmann <marcel@holtmann.org>
Date: Mon, 14 Jul 2008 20:13:48 +0200
Subject: [Bluetooth] Track status of Simple Pairing mode

The Simple Pairing feature is optional and needs to be enabled by the
host stack first. The Linux kernel relies on the Bluetooth daemon to
either enable or disable it, but at any time it needs to know the
current state of the Simple Pairing mode. So track any changes made
by external entities and store the current mode in the HCI device
structure.

Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
---
 include/net/bluetooth/hci.h      | 11 +++++++++++
 include/net/bluetooth/hci_core.h |  1 +
 net/bluetooth/hci_event.c        | 37 +++++++++++++++++++++++++++++++++++++
 3 files changed, 49 insertions(+)

(limited to 'include')

diff --git a/include/net/bluetooth/hci.h b/include/net/bluetooth/hci.h
index 79629ff40e3e..6d0c04a81fc7 100644
--- a/include/net/bluetooth/hci.h
+++ b/include/net/bluetooth/hci.h
@@ -514,6 +514,17 @@ struct hci_cp_host_buffer_size {
 	__le16   sco_max_pkt;
 } __attribute__ ((packed));
 
+#define HCI_OP_READ_SSP_MODE		0x0c55
+struct hci_rp_read_ssp_mode {
+	__u8     status;
+	__u8     mode;
+} __attribute__ ((packed));
+
+#define HCI_OP_WRITE_SSP_MODE		0x0c56
+struct hci_cp_write_ssp_mode {
+	__u8     mode;
+} __attribute__ ((packed));
+
 #define HCI_OP_READ_LOCAL_VERSION	0x1001
 struct hci_rp_read_local_version {
 	__u8     status;
diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h
index 6424d63e3395..b85754e29a78 100644
--- a/include/net/bluetooth/hci_core.h
+++ b/include/net/bluetooth/hci_core.h
@@ -75,6 +75,7 @@ struct hci_dev {
 	__u8		dev_class[3];
 	__u8		features[8];
 	__u8		commands[64];
+	__u8		ssp_mode;
 	__u8		hci_ver;
 	__u16		hci_rev;
 	__u16		manufacturer;
diff --git a/net/bluetooth/hci_event.c b/net/bluetooth/hci_event.c
index 7c9ac01cd8f9..6077a651aac6 100644
--- a/net/bluetooth/hci_event.c
+++ b/net/bluetooth/hci_event.c
@@ -391,6 +391,35 @@ static void hci_cc_host_buffer_size(struct hci_dev *hdev, struct sk_buff *skb)
 	hci_req_complete(hdev, status);
 }
 
+static void hci_cc_read_ssp_mode(struct hci_dev *hdev, struct sk_buff *skb)
+{
+	struct hci_rp_read_ssp_mode *rp = (void *) skb->data;
+
+	BT_DBG("%s status 0x%x", hdev->name, rp->status);
+
+	if (rp->status)
+		return;
+
+	hdev->ssp_mode = rp->mode;
+}
+
+static void hci_cc_write_ssp_mode(struct hci_dev *hdev, struct sk_buff *skb)
+{
+	__u8 status = *((__u8 *) skb->data);
+	void *sent;
+
+	BT_DBG("%s status 0x%x", hdev->name, status);
+
+	if (status)
+		return;
+
+	sent = hci_sent_cmd_data(hdev, HCI_OP_WRITE_SSP_MODE);
+	if (!sent)
+		return;
+
+	hdev->ssp_mode = *((__u8 *) sent);
+}
+
 static void hci_cc_read_local_version(struct hci_dev *hdev, struct sk_buff *skb)
 {
 	struct hci_rp_read_local_version *rp = (void *) skb->data;
@@ -1084,6 +1113,14 @@ static inline void hci_cmd_complete_evt(struct hci_dev *hdev, struct sk_buff *sk
 		hci_cc_host_buffer_size(hdev, skb);
 		break;
 
+	case HCI_OP_READ_SSP_MODE:
+		hci_cc_read_ssp_mode(hdev, skb);
+		break;
+
+	case HCI_OP_WRITE_SSP_MODE:
+		hci_cc_write_ssp_mode(hdev, skb);
+		break;
+
 	case HCI_OP_READ_LOCAL_VERSION:
 		hci_cc_read_local_version(hdev, skb);
 		break;
-- 
cgit v1.2.3


From 41a96212b3b7b3cd59e8e8d33e6dabf0e21d9778 Mon Sep 17 00:00:00 2001
From: Marcel Holtmann <marcel@holtmann.org>
Date: Mon, 14 Jul 2008 20:13:48 +0200
Subject: [Bluetooth] Track status of remote Simple Pairing mode

The Simple Pairing process can only be used if both sides have the
support enabled in the host stack. The current Bluetooth specification
has three ways to detect this support.

If an Extended Inquiry Result has been sent during inquiry then it
is safe to assume that Simple Pairing is enabled. It is not allowed
to enable Extended Inquiry without Simple Pairing. During the remote
name request phase a notification with the remote host supported
features will be sent to indicate Simple Pairing support. Also the
second page of the remote extended features can indicate support for
Simple Pairing.

For all three cases the value of remote Simple Pairing mode is stored
in the inquiry cache for later use.

Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
---
 include/net/bluetooth/hci.h      |  6 ++++++
 include/net/bluetooth/hci_core.h |  2 ++
 net/bluetooth/hci_conn.c         | 14 ++++++++-----
 net/bluetooth/hci_event.c        | 43 ++++++++++++++++++++++++++++++++++++++++
 4 files changed, 60 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/include/net/bluetooth/hci.h b/include/net/bluetooth/hci.h
index 6d0c04a81fc7..5ac0a18db63c 100644
--- a/include/net/bluetooth/hci.h
+++ b/include/net/bluetooth/hci.h
@@ -816,6 +816,12 @@ struct hci_ev_simple_pair_complete {
 	bdaddr_t bdaddr;
 } __attribute__ ((packed));
 
+#define HCI_EV_REMOTE_HOST_FEATURES	0x3d
+struct hci_ev_remote_host_features {
+	bdaddr_t bdaddr;
+	__u8     features[8];
+} __attribute__ ((packed));
+
 /* Internal events generated by Bluetooth stack */
 #define HCI_EV_STACK_INTERNAL	0xfd
 struct hci_ev_stack_internal {
diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h
index b85754e29a78..f73cc2945700 100644
--- a/include/net/bluetooth/hci_core.h
+++ b/include/net/bluetooth/hci_core.h
@@ -40,6 +40,7 @@ struct inquiry_data {
 	__u8		dev_class[3];
 	__le16		clock_offset;
 	__s8		rssi;
+	__u8		ssp_mode;
 };
 
 struct inquiry_entry {
@@ -162,6 +163,7 @@ struct hci_conn {
 	__u8		 attempt;
 	__u8		 dev_class[3];
 	__u8             features[8];
+	__u8             ssp_mode;
 	__u16            interval;
 	__u16            pkt_type;
 	__u16            link_policy;
diff --git a/net/bluetooth/hci_conn.c b/net/bluetooth/hci_conn.c
index 6175ce841e9e..41351ba692e9 100644
--- a/net/bluetooth/hci_conn.c
+++ b/net/bluetooth/hci_conn.c
@@ -71,12 +71,16 @@ void hci_acl_connect(struct hci_conn *conn)
 	bacpy(&cp.bdaddr, &conn->dst);
 	cp.pscan_rep_mode = 0x02;
 
-	if ((ie = hci_inquiry_cache_lookup(hdev, &conn->dst)) &&
-			inquiry_entry_age(ie) <= INQUIRY_ENTRY_AGE_MAX) {
-		cp.pscan_rep_mode = ie->data.pscan_rep_mode;
-		cp.pscan_mode     = ie->data.pscan_mode;
-		cp.clock_offset   = ie->data.clock_offset | cpu_to_le16(0x8000);
+	if ((ie = hci_inquiry_cache_lookup(hdev, &conn->dst))) {
+		if (inquiry_entry_age(ie) <= INQUIRY_ENTRY_AGE_MAX) {
+			cp.pscan_rep_mode = ie->data.pscan_rep_mode;
+			cp.pscan_mode     = ie->data.pscan_mode;
+			cp.clock_offset   = ie->data.clock_offset |
+							cpu_to_le16(0x8000);
+		}
+
 		memcpy(conn->dev_class, ie->data.dev_class, 3);
+		conn->ssp_mode = ie->data.ssp_mode;
 	}
 
 	cp.pkt_type = cpu_to_le16(conn->pkt_type);
diff --git a/net/bluetooth/hci_event.c b/net/bluetooth/hci_event.c
index 6077a651aac6..c8fda7dc2986 100644
--- a/net/bluetooth/hci_event.c
+++ b/net/bluetooth/hci_event.c
@@ -736,6 +736,7 @@ static inline void hci_inquiry_result_evt(struct hci_dev *hdev, struct sk_buff *
 		memcpy(data.dev_class, info->dev_class, 3);
 		data.clock_offset	= info->clock_offset;
 		data.rssi		= 0x00;
+		data.ssp_mode		= 0x00;
 		info++;
 		hci_inquiry_cache_update(hdev, &data);
 	}
@@ -1390,6 +1391,7 @@ static inline void hci_inquiry_result_with_rssi_evt(struct hci_dev *hdev, struct
 			memcpy(data.dev_class, info->dev_class, 3);
 			data.clock_offset	= info->clock_offset;
 			data.rssi		= info->rssi;
+			data.ssp_mode		= 0x00;
 			info++;
 			hci_inquiry_cache_update(hdev, &data);
 		}
@@ -1404,6 +1406,7 @@ static inline void hci_inquiry_result_with_rssi_evt(struct hci_dev *hdev, struct
 			memcpy(data.dev_class, info->dev_class, 3);
 			data.clock_offset	= info->clock_offset;
 			data.rssi		= info->rssi;
+			data.ssp_mode		= 0x00;
 			info++;
 			hci_inquiry_cache_update(hdev, &data);
 		}
@@ -1414,7 +1417,27 @@ static inline void hci_inquiry_result_with_rssi_evt(struct hci_dev *hdev, struct
 
 static inline void hci_remote_ext_features_evt(struct hci_dev *hdev, struct sk_buff *skb)
 {
+	struct hci_ev_remote_ext_features *ev = (void *) skb->data;
+	struct hci_conn *conn;
+
 	BT_DBG("%s", hdev->name);
+
+	if (ev->status || ev->page != 0x01)
+		return;
+
+	hci_dev_lock(hdev);
+
+	conn = hci_conn_hash_lookup_handle(hdev, __le16_to_cpu(ev->handle));
+	if (conn) {
+		struct inquiry_entry *ie;
+
+		if ((ie = hci_inquiry_cache_lookup(hdev, &conn->dst)))
+			ie->data.ssp_mode = (ev->features[0] & 0x01);
+
+		conn->ssp_mode = (ev->features[0] & 0x01);
+	}
+
+	hci_dev_unlock(hdev);
 }
 
 static inline void hci_sync_conn_complete_evt(struct hci_dev *hdev, struct sk_buff *skb)
@@ -1494,6 +1517,7 @@ static inline void hci_extended_inquiry_result_evt(struct hci_dev *hdev, struct
 		memcpy(data.dev_class, info->dev_class, 3);
 		data.clock_offset       = info->clock_offset;
 		data.rssi               = info->rssi;
+		data.ssp_mode		= 0x01;
 		info++;
 		hci_inquiry_cache_update(hdev, &data);
 	}
@@ -1533,6 +1557,21 @@ static inline void hci_simple_pair_complete_evt(struct hci_dev *hdev, struct sk_
 	hci_dev_unlock(hdev);
 }
 
+static inline void hci_remote_host_features_evt(struct hci_dev *hdev, struct sk_buff *skb)
+{
+	struct hci_ev_remote_host_features *ev = (void *) skb->data;
+	struct inquiry_entry *ie;
+
+	BT_DBG("%s", hdev->name);
+
+	hci_dev_lock(hdev);
+
+	if ((ie = hci_inquiry_cache_lookup(hdev, &ev->bdaddr)))
+		ie->data.ssp_mode = (ev->features[0] & 0x01);
+
+	hci_dev_unlock(hdev);
+}
+
 void hci_event_packet(struct hci_dev *hdev, struct sk_buff *skb)
 {
 	struct hci_event_hdr *hdr = (void *) skb->data;
@@ -1665,6 +1704,10 @@ void hci_event_packet(struct hci_dev *hdev, struct sk_buff *skb)
 		hci_simple_pair_complete_evt(hdev, skb);
 		break;
 
+	case HCI_EV_REMOTE_HOST_FEATURES:
+		hci_remote_host_features_evt(hdev, skb);
+		break;
+
 	default:
 		BT_DBG("%s event 0x%x", hdev->name, event);
 		break;
-- 
cgit v1.2.3


From 769be974d0c7b4fe1a52f9cdaad22259b60953f7 Mon Sep 17 00:00:00 2001
From: Marcel Holtmann <marcel@holtmann.org>
Date: Mon, 14 Jul 2008 20:13:49 +0200
Subject: [Bluetooth] Use ACL config stage to retrieve remote features

The Bluetooth technology introduces new features on a regular basis
and for some of them it is important that the hardware on both sides
support them. For features like Simple Pairing it is important that
the host stacks on both sides have switched this feature on. To make
valid decisions, a config stage during ACL link establishment has been
introduced that retrieves remote features and if needed also the remote
extended features (known as remote host features) before signalling
this link as connected.

This change introduces full reference counting of incoming and outgoing
ACL links and the Bluetooth core will disconnect both if no owner of it
is present. To better handle interoperability during the pairing phase
the disconnect timeout for incoming connections has been increased to
10 seconds. This is five times more than for outgoing connections.

Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
---
 include/net/bluetooth/hci.h      |   2 +
 include/net/bluetooth/hci_core.h |   3 +-
 net/bluetooth/hci_conn.c         |   2 +
 net/bluetooth/hci_core.c         |   7 ++-
 net/bluetooth/hci_event.c        | 127 +++++++++++++++++++++++++++++++++------
 5 files changed, 118 insertions(+), 23 deletions(-)

(limited to 'include')

diff --git a/include/net/bluetooth/hci.h b/include/net/bluetooth/hci.h
index 5ac0a18db63c..55576e848827 100644
--- a/include/net/bluetooth/hci.h
+++ b/include/net/bluetooth/hci.h
@@ -180,6 +180,8 @@ enum {
 
 #define LMP_SNIFF_SUBR	0x02
 
+#define LMP_SIMPLE_PAIR	0x08
+
 /* Connection modes */
 #define HCI_CM_ACTIVE	0x0000
 #define HCI_CM_HOLD	0x0001
diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h
index f73cc2945700..28fbd0caa534 100644
--- a/include/net/bluetooth/hci_core.h
+++ b/include/net/bluetooth/hci_core.h
@@ -348,7 +348,7 @@ static inline void hci_conn_put(struct hci_conn *conn)
 			if (conn->state == BT_CONNECTED) {
 				timeo = msecs_to_jiffies(HCI_DISCONN_TIMEOUT);
 				if (!conn->out)
-					timeo *= 2;
+					timeo *= 5;
 			} else
 				timeo = msecs_to_jiffies(10);
 		} else
@@ -463,6 +463,7 @@ void hci_conn_del_sysfs(struct hci_conn *conn);
 #define lmp_sniff_capable(dev)     ((dev)->features[0] & LMP_SNIFF)
 #define lmp_sniffsubr_capable(dev) ((dev)->features[5] & LMP_SNIFF_SUBR)
 #define lmp_esco_capable(dev)      ((dev)->features[3] & LMP_ESCO)
+#define lmp_ssp_capable(dev)       ((dev)->features[6] & LMP_SIMPLE_PAIR)
 
 /* ----- HCI protocols ----- */
 struct hci_proto {
diff --git a/net/bluetooth/hci_conn.c b/net/bluetooth/hci_conn.c
index 41351ba692e9..6f22533e7656 100644
--- a/net/bluetooth/hci_conn.c
+++ b/net/bluetooth/hci_conn.c
@@ -170,11 +170,13 @@ static void hci_conn_timeout(unsigned long arg)
 
 	switch (conn->state) {
 	case BT_CONNECT:
+	case BT_CONNECT2:
 		if (conn->type == ACL_LINK)
 			hci_acl_connect_cancel(conn);
 		else
 			hci_acl_disconn(conn, 0x13);
 		break;
+	case BT_CONFIG:
 	case BT_CONNECTED:
 		hci_acl_disconn(conn, 0x13);
 		break;
diff --git a/net/bluetooth/hci_core.c b/net/bluetooth/hci_core.c
index 69b2c1aac08a..f5b21cb93699 100644
--- a/net/bluetooth/hci_core.c
+++ b/net/bluetooth/hci_core.c
@@ -1283,9 +1283,12 @@ static inline struct hci_conn *hci_low_sent(struct hci_dev *hdev, __u8 type, int
 		struct hci_conn *c;
 		c = list_entry(p, struct hci_conn, list);
 
-		if (c->type != type || c->state != BT_CONNECTED
-				|| skb_queue_empty(&c->data_q))
+		if (c->type != type || skb_queue_empty(&c->data_q))
 			continue;
+
+		if (c->state != BT_CONNECTED && c->state != BT_CONFIG)
+			continue;
+
 		num++;
 
 		if (c->sent < min) {
diff --git a/net/bluetooth/hci_event.c b/net/bluetooth/hci_event.c
index c8fda7dc2986..e3e360c3c536 100644
--- a/net/bluetooth/hci_event.c
+++ b/net/bluetooth/hci_event.c
@@ -624,6 +624,62 @@ static void hci_cs_remote_name_req(struct hci_dev *hdev, __u8 status)
 	BT_DBG("%s status 0x%x", hdev->name, status);
 }
 
+static void hci_cs_read_remote_features(struct hci_dev *hdev, __u8 status)
+{
+	struct hci_cp_read_remote_features *cp;
+	struct hci_conn *conn;
+
+	BT_DBG("%s status 0x%x", hdev->name, status);
+
+	if (!status)
+		return;
+
+	cp = hci_sent_cmd_data(hdev, HCI_OP_READ_REMOTE_FEATURES);
+	if (!cp)
+		return;
+
+	hci_dev_lock(hdev);
+
+	conn = hci_conn_hash_lookup_handle(hdev, __le16_to_cpu(cp->handle));
+	if (conn) {
+		if (conn->state == BT_CONFIG) {
+			conn->state = BT_CONNECTED;
+			hci_proto_connect_cfm(conn, status);
+			hci_conn_put(conn);
+		}
+	}
+
+	hci_dev_unlock(hdev);
+}
+
+static void hci_cs_read_remote_ext_features(struct hci_dev *hdev, __u8 status)
+{
+	struct hci_cp_read_remote_ext_features *cp;
+	struct hci_conn *conn;
+
+	BT_DBG("%s status 0x%x", hdev->name, status);
+
+	if (!status)
+		return;
+
+	cp = hci_sent_cmd_data(hdev, HCI_OP_READ_REMOTE_EXT_FEATURES);
+	if (!cp)
+		return;
+
+	hci_dev_lock(hdev);
+
+	conn = hci_conn_hash_lookup_handle(hdev, __le16_to_cpu(cp->handle));
+	if (conn) {
+		if (conn->state == BT_CONFIG) {
+			conn->state = BT_CONNECTED;
+			hci_proto_connect_cfm(conn, status);
+			hci_conn_put(conn);
+		}
+	}
+
+	hci_dev_unlock(hdev);
+}
+
 static void hci_cs_setup_sync_conn(struct hci_dev *hdev, __u8 status)
 {
 	struct hci_cp_setup_sync_conn *cp;
@@ -759,7 +815,12 @@ static inline void hci_conn_complete_evt(struct hci_dev *hdev, struct sk_buff *s
 
 	if (!ev->status) {
 		conn->handle = __le16_to_cpu(ev->handle);
-		conn->state  = BT_CONNECTED;
+
+		if (conn->type == ACL_LINK) {
+			conn->state = BT_CONFIG;
+			hci_conn_hold(conn);
+		} else
+			conn->state = BT_CONNECTED;
 
 		if (test_bit(HCI_AUTH, &hdev->flags))
 			conn->link_mode |= HCI_LM_AUTH;
@@ -771,7 +832,8 @@ static inline void hci_conn_complete_evt(struct hci_dev *hdev, struct sk_buff *s
 		if (conn->type == ACL_LINK) {
 			struct hci_cp_read_remote_features cp;
 			cp.handle = ev->handle;
-			hci_send_cmd(hdev, HCI_OP_READ_REMOTE_FEATURES, sizeof(cp), &cp);
+			hci_send_cmd(hdev, HCI_OP_READ_REMOTE_FEATURES,
+							sizeof(cp), &cp);
 		}
 
 		/* Set packet type for incoming connection */
@@ -781,10 +843,6 @@ static inline void hci_conn_complete_evt(struct hci_dev *hdev, struct sk_buff *s
 			cp.pkt_type = cpu_to_le16(conn->pkt_type);
 			hci_send_cmd(hdev, HCI_OP_CHANGE_CONN_PTYPE,
 							sizeof(cp), &cp);
-		} else {
-			/* Update disconnect timer */
-			hci_conn_hold(conn);
-			hci_conn_put(conn);
 		}
 	} else
 		conn->state = BT_CLOSED;
@@ -804,9 +862,10 @@ static inline void hci_conn_complete_evt(struct hci_dev *hdev, struct sk_buff *s
 		}
 	}
 
-	hci_proto_connect_cfm(conn, ev->status);
-	if (ev->status)
+	if (ev->status) {
+		hci_proto_connect_cfm(conn, ev->status);
 		hci_conn_del(conn);
+	}
 
 unlock:
 	hci_dev_unlock(hdev);
@@ -1006,14 +1065,29 @@ static inline void hci_remote_features_evt(struct hci_dev *hdev, struct sk_buff
 
 	BT_DBG("%s status %d", hdev->name, ev->status);
 
-	if (ev->status)
-		return;
-
 	hci_dev_lock(hdev);
 
 	conn = hci_conn_hash_lookup_handle(hdev, __le16_to_cpu(ev->handle));
-	if (conn)
-		memcpy(conn->features, ev->features, 8);
+	if (conn) {
+		if (!ev->status)
+			memcpy(conn->features, ev->features, 8);
+
+		if (conn->state == BT_CONFIG) {
+			if (!ev->status && lmp_ssp_capable(hdev) &&
+						lmp_ssp_capable(conn)) {
+				struct hci_cp_read_remote_ext_features cp;
+				cp.handle = ev->handle;
+				cp.page = 0x01;
+				hci_send_cmd(hdev,
+					HCI_OP_READ_REMOTE_EXT_FEATURES,
+							sizeof(cp), &cp);
+			} else {
+				conn->state = BT_CONNECTED;
+				hci_proto_connect_cfm(conn, ev->status);
+				hci_conn_put(conn);
+			}
+		}
+	}
 
 	hci_dev_unlock(hdev);
 }
@@ -1180,6 +1254,14 @@ static inline void hci_cmd_status_evt(struct hci_dev *hdev, struct sk_buff *skb)
 		hci_cs_remote_name_req(hdev, ev->status);
 		break;
 
+	case HCI_OP_READ_REMOTE_FEATURES:
+		hci_cs_read_remote_features(hdev, ev->status);
+		break;
+
+	case HCI_OP_READ_REMOTE_EXT_FEATURES:
+		hci_cs_read_remote_ext_features(hdev, ev->status);
+		break;
+
 	case HCI_OP_SETUP_SYNC_CONN:
 		hci_cs_setup_sync_conn(hdev, ev->status);
 		break;
@@ -1422,19 +1504,24 @@ static inline void hci_remote_ext_features_evt(struct hci_dev *hdev, struct sk_b
 
 	BT_DBG("%s", hdev->name);
 
-	if (ev->status || ev->page != 0x01)
-		return;
-
 	hci_dev_lock(hdev);
 
 	conn = hci_conn_hash_lookup_handle(hdev, __le16_to_cpu(ev->handle));
 	if (conn) {
-		struct inquiry_entry *ie;
+		if (!ev->status && ev->page == 0x01) {
+			struct inquiry_entry *ie;
 
-		if ((ie = hci_inquiry_cache_lookup(hdev, &conn->dst)))
-			ie->data.ssp_mode = (ev->features[0] & 0x01);
+			if ((ie = hci_inquiry_cache_lookup(hdev, &conn->dst)))
+				ie->data.ssp_mode = (ev->features[0] & 0x01);
 
-		conn->ssp_mode = (ev->features[0] & 0x01);
+			conn->ssp_mode = (ev->features[0] & 0x01);
+		}
+
+		if (conn->state == BT_CONFIG) {
+			conn->state = BT_CONNECTED;
+			hci_proto_connect_cfm(conn, ev->status);
+			hci_conn_put(conn);
+		}
 	}
 
 	hci_dev_unlock(hdev);
-- 
cgit v1.2.3


From 40be492fe4fab829951681860c2bb26fa1d5fe4a Mon Sep 17 00:00:00 2001
From: Marcel Holtmann <marcel@holtmann.org>
Date: Mon, 14 Jul 2008 20:13:50 +0200
Subject: [Bluetooth] Export details about authentication requirements

With the Simple Pairing support, the authentication requirements are
an explicit setting during the bonding process. Track and enforce the
requirements and allow higher layers like L2CAP and RFCOMM to increase
them if needed.

This patch introduces a new IOCTL that allows to query the current
authentication requirements. It is also possible to detect Simple
Pairing support in the kernel this way.

Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
---
 fs/compat_ioctl.c                |  1 +
 include/net/bluetooth/hci.h      | 18 ++++++++++++++----
 include/net/bluetooth/hci_core.h |  2 ++
 net/bluetooth/hci_conn.c         | 38 ++++++++++++++++++++++++++++++++++----
 net/bluetooth/hci_sock.c         | 18 +++++-------------
 net/bluetooth/l2cap.c            | 14 ++++++++++----
 net/bluetooth/rfcomm/core.c      |  3 ++-
 7 files changed, 68 insertions(+), 26 deletions(-)

(limited to 'include')

diff --git a/fs/compat_ioctl.c b/fs/compat_ioctl.c
index 97dba0d92348..d63d430d3b26 100644
--- a/fs/compat_ioctl.c
+++ b/fs/compat_ioctl.c
@@ -2399,6 +2399,7 @@ COMPATIBLE_IOCTL(HCIGETDEVLIST)
 COMPATIBLE_IOCTL(HCIGETDEVINFO)
 COMPATIBLE_IOCTL(HCIGETCONNLIST)
 COMPATIBLE_IOCTL(HCIGETCONNINFO)
+COMPATIBLE_IOCTL(HCIGETAUTHINFO)
 COMPATIBLE_IOCTL(HCISETRAW)
 COMPATIBLE_IOCTL(HCISETSCAN)
 COMPATIBLE_IOCTL(HCISETAUTH)
diff --git a/include/net/bluetooth/hci.h b/include/net/bluetooth/hci.h
index 55576e848827..3cc294919312 100644
--- a/include/net/bluetooth/hci.h
+++ b/include/net/bluetooth/hci.h
@@ -72,8 +72,6 @@ enum {
 	HCI_INQUIRY,
 
 	HCI_RAW,
-
-	HCI_SECMGR
 };
 
 /* HCI ioctl defines */
@@ -86,6 +84,7 @@ enum {
 #define HCIGETDEVINFO	_IOR('H', 211, int)
 #define HCIGETCONNLIST	_IOR('H', 212, int)
 #define HCIGETCONNINFO	_IOR('H', 213, int)
+#define HCIGETAUTHINFO	_IOR('H', 215, int)
 
 #define HCISETRAW	_IOW('H', 220, int)
 #define HCISETSCAN	_IOW('H', 221, int)
@@ -97,8 +96,6 @@ enum {
 #define HCISETACLMTU	_IOW('H', 227, int)
 #define HCISETSCOMTU	_IOW('H', 228, int)
 
-#define HCISETSECMGR	_IOW('H', 230, int)
-
 #define HCIINQUIRY	_IOR('H', 240, int)
 
 /* HCI timeouts */
@@ -203,6 +200,14 @@ enum {
 #define HCI_LM_RELIABLE	0x0010
 #define HCI_LM_SECURE	0x0020
 
+/* Authentication types */
+#define HCI_AT_NO_BONDING		0x00
+#define HCI_AT_NO_BONDING_MITM		0x01
+#define HCI_AT_DEDICATED_BONDING	0x02
+#define HCI_AT_DEDICATED_BONDING_MITM	0x03
+#define HCI_AT_GENERAL_BONDING		0x04
+#define HCI_AT_GENERAL_BONDING_MITM	0x05
+
 /* -----  HCI Commands ---- */
 #define HCI_OP_INQUIRY			0x0401
 struct hci_cp_inquiry {
@@ -1001,6 +1006,11 @@ struct hci_conn_info_req {
 	struct   hci_conn_info conn_info[0];
 };
 
+struct hci_auth_info_req {
+	bdaddr_t bdaddr;
+	__u8     type;
+};
+
 struct hci_inquiry_req {
 	__u16 dev_id;
 	__u16 flags;
diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h
index 28fbd0caa534..cbf751094688 100644
--- a/include/net/bluetooth/hci_core.h
+++ b/include/net/bluetooth/hci_core.h
@@ -168,6 +168,7 @@ struct hci_conn {
 	__u16            pkt_type;
 	__u16            link_policy;
 	__u32		 link_mode;
+	__u8             auth_type;
 	__u8             power_save;
 	unsigned long	 pend;
 
@@ -422,6 +423,7 @@ int hci_get_dev_list(void __user *arg);
 int hci_get_dev_info(void __user *arg);
 int hci_get_conn_list(void __user *arg);
 int hci_get_conn_info(struct hci_dev *hdev, void __user *arg);
+int hci_get_auth_info(struct hci_dev *hdev, void __user *arg);
 int hci_inquiry(void __user *arg);
 
 void hci_event_packet(struct hci_dev *hdev, struct sk_buff *skb);
diff --git a/net/bluetooth/hci_conn.c b/net/bluetooth/hci_conn.c
index 6f22533e7656..0d4b8aeb8e09 100644
--- a/net/bluetooth/hci_conn.c
+++ b/net/bluetooth/hci_conn.c
@@ -379,13 +379,21 @@ int hci_conn_auth(struct hci_conn *conn)
 {
 	BT_DBG("conn %p", conn);
 
+	if (conn->ssp_mode > 0 && conn->hdev->ssp_mode > 0) {
+		if (!(conn->auth_type & 0x01)) {
+			conn->auth_type = HCI_AT_GENERAL_BONDING_MITM;
+			conn->link_mode &= ~HCI_LM_AUTH;
+		}
+	}
+
 	if (conn->link_mode & HCI_LM_AUTH)
 		return 1;
 
 	if (!test_and_set_bit(HCI_CONN_AUTH_PEND, &conn->pend)) {
 		struct hci_cp_auth_requested cp;
 		cp.handle = cpu_to_le16(conn->handle);
-		hci_send_cmd(conn->hdev, HCI_OP_AUTH_REQUESTED, sizeof(cp), &cp);
+		hci_send_cmd(conn->hdev, HCI_OP_AUTH_REQUESTED,
+							sizeof(cp), &cp);
 	}
 	return 0;
 }
@@ -397,7 +405,7 @@ int hci_conn_encrypt(struct hci_conn *conn)
 	BT_DBG("conn %p", conn);
 
 	if (conn->link_mode & HCI_LM_ENCRYPT)
-		return 1;
+		return hci_conn_auth(conn);
 
 	if (test_and_set_bit(HCI_CONN_ENCRYPT_PEND, &conn->pend))
 		return 0;
@@ -406,7 +414,8 @@ int hci_conn_encrypt(struct hci_conn *conn)
 		struct hci_cp_set_conn_encrypt cp;
 		cp.handle  = cpu_to_le16(conn->handle);
 		cp.encrypt = 1;
-		hci_send_cmd(conn->hdev, HCI_OP_SET_CONN_ENCRYPT, sizeof(cp), &cp);
+		hci_send_cmd(conn->hdev, HCI_OP_SET_CONN_ENCRYPT,
+							sizeof(cp), &cp);
 	}
 	return 0;
 }
@@ -420,7 +429,8 @@ int hci_conn_change_link_key(struct hci_conn *conn)
 	if (!test_and_set_bit(HCI_CONN_AUTH_PEND, &conn->pend)) {
 		struct hci_cp_change_conn_link_key cp;
 		cp.handle = cpu_to_le16(conn->handle);
-		hci_send_cmd(conn->hdev, HCI_OP_CHANGE_CONN_LINK_KEY, sizeof(cp), &cp);
+		hci_send_cmd(conn->hdev, HCI_OP_CHANGE_CONN_LINK_KEY,
+							sizeof(cp), &cp);
 	}
 	return 0;
 }
@@ -624,3 +634,23 @@ int hci_get_conn_info(struct hci_dev *hdev, void __user *arg)
 
 	return copy_to_user(ptr, &ci, sizeof(ci)) ? -EFAULT : 0;
 }
+
+int hci_get_auth_info(struct hci_dev *hdev, void __user *arg)
+{
+	struct hci_auth_info_req req;
+	struct hci_conn *conn;
+
+	if (copy_from_user(&req, arg, sizeof(req)))
+		return -EFAULT;
+
+	hci_dev_lock_bh(hdev);
+	conn = hci_conn_hash_lookup_ba(hdev, ACL_LINK, &req.bdaddr);
+	if (conn)
+		req.type = conn->auth_type;
+	hci_dev_unlock_bh(hdev);
+
+	if (!conn)
+		return -ENOENT;
+
+	return copy_to_user(arg, &req, sizeof(req)) ? -EFAULT : 0;
+}
diff --git a/net/bluetooth/hci_sock.c b/net/bluetooth/hci_sock.c
index 747fabd735d2..d62579b67959 100644
--- a/net/bluetooth/hci_sock.c
+++ b/net/bluetooth/hci_sock.c
@@ -193,19 +193,11 @@ static inline int hci_sock_bound_ioctl(struct sock *sk, unsigned int cmd, unsign
 
 		return 0;
 
-	case HCISETSECMGR:
-		if (!capable(CAP_NET_ADMIN))
-			return -EACCES;
-
-		if (arg)
-			set_bit(HCI_SECMGR, &hdev->flags);
-		else
-			clear_bit(HCI_SECMGR, &hdev->flags);
-
-		return 0;
-
 	case HCIGETCONNINFO:
-		return hci_get_conn_info(hdev, (void __user *)arg);
+		return hci_get_conn_info(hdev, (void __user *) arg);
+
+	case HCIGETAUTHINFO:
+		return hci_get_auth_info(hdev, (void __user *) arg);
 
 	default:
 		if (hdev->ioctl)
@@ -217,7 +209,7 @@ static inline int hci_sock_bound_ioctl(struct sock *sk, unsigned int cmd, unsign
 static int hci_sock_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg)
 {
 	struct sock *sk = sock->sk;
-	void __user *argp = (void __user *)arg;
+	void __user *argp = (void __user *) arg;
 	int err;
 
 	BT_DBG("cmd %x arg %lx", cmd, arg);
diff --git a/net/bluetooth/l2cap.c b/net/bluetooth/l2cap.c
index 252264062f59..30ad59b717d5 100644
--- a/net/bluetooth/l2cap.c
+++ b/net/bluetooth/l2cap.c
@@ -2150,7 +2150,7 @@ static int l2cap_disconn_ind(struct hci_conn *hcon, u8 reason)
 static int l2cap_auth_cfm(struct hci_conn *hcon, u8 status)
 {
 	struct l2cap_chan_list *l;
-	struct l2cap_conn *conn = conn = hcon->l2cap_data;
+	struct l2cap_conn *conn = hcon->l2cap_data;
 	struct l2cap_conn_rsp rsp;
 	struct sock *sk;
 	int result;
@@ -2165,11 +2165,17 @@ static int l2cap_auth_cfm(struct hci_conn *hcon, u8 status)
 	read_lock(&l->lock);
 
 	for (sk = l->head; sk; sk = l2cap_pi(sk)->next_c) {
+		struct l2cap_pinfo *pi = l2cap_pi(sk);
+
 		bh_lock_sock(sk);
 
-		if (sk->sk_state != BT_CONNECT2 ||
-				(l2cap_pi(sk)->link_mode & L2CAP_LM_ENCRYPT) ||
-				(l2cap_pi(sk)->link_mode & L2CAP_LM_SECURE)) {
+		if (sk->sk_state != BT_CONNECT2) {
+			bh_unlock_sock(sk);
+			continue;
+		}
+
+		if ((pi->link_mode & (L2CAP_LM_ENCRYPT | L2CAP_LM_SECURE)) &&
+					!(hcon->link_mode & HCI_LM_ENCRYPT)) {
 			bh_unlock_sock(sk);
 			continue;
 		}
diff --git a/net/bluetooth/rfcomm/core.c b/net/bluetooth/rfcomm/core.c
index e7a6a03cea37..e56bcfc35a49 100644
--- a/net/bluetooth/rfcomm/core.c
+++ b/net/bluetooth/rfcomm/core.c
@@ -1969,7 +1969,8 @@ static void rfcomm_auth_cfm(struct hci_conn *conn, u8 status)
 	list_for_each_safe(p, n, &s->dlcs) {
 		d = list_entry(p, struct rfcomm_dlc, list);
 
-		if (d->link_mode & (RFCOMM_LM_ENCRYPT | RFCOMM_LM_SECURE))
+		if ((d->link_mode & (RFCOMM_LM_ENCRYPT | RFCOMM_LM_SECURE)) &&
+				!(conn->link_mode & HCI_LM_ENCRYPT) && !status)
 			continue;
 
 		if (!test_and_clear_bit(RFCOMM_AUTH_PENDING, &d->flags))
-- 
cgit v1.2.3


From 3241ad820dbb172021e0268b5611031991431626 Mon Sep 17 00:00:00 2001
From: Marcel Holtmann <marcel@holtmann.org>
Date: Mon, 14 Jul 2008 20:13:50 +0200
Subject: [Bluetooth] Add timestamp support to L2CAP, RFCOMM and SCO

Enable the common timestamp functionality that the network subsystem
provides for L2CAP, RFCOMM and SCO sockets. It is possible to either
use SO_TIMESTAMP or the IOCTLs to retrieve the timestamp of the
current packet.

Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
---
 include/net/bluetooth/bluetooth.h |  1 +
 net/bluetooth/af_bluetooth.c      | 28 ++++++++++++++++++++++++++++
 net/bluetooth/l2cap.c             |  2 +-
 net/bluetooth/rfcomm/sock.c       | 15 +++++++++++----
 net/bluetooth/sco.c               |  2 +-
 5 files changed, 42 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/include/net/bluetooth/bluetooth.h b/include/net/bluetooth/bluetooth.h
index 750648df13f4..6f8418bf4241 100644
--- a/include/net/bluetooth/bluetooth.h
+++ b/include/net/bluetooth/bluetooth.h
@@ -121,6 +121,7 @@ void bt_sock_link(struct bt_sock_list *l, struct sock *s);
 void bt_sock_unlink(struct bt_sock_list *l, struct sock *s);
 int  bt_sock_recvmsg(struct kiocb *iocb, struct socket *sock, struct msghdr *msg, size_t len, int flags);
 uint bt_sock_poll(struct file * file, struct socket *sock, poll_table *wait);
+int  bt_sock_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg);
 int  bt_sock_wait_state(struct sock *sk, int state, unsigned long timeo);
 
 void bt_accept_enqueue(struct sock *parent, struct sock *sk);
diff --git a/net/bluetooth/af_bluetooth.c b/net/bluetooth/af_bluetooth.c
index d366423c8392..88afe25003d8 100644
--- a/net/bluetooth/af_bluetooth.c
+++ b/net/bluetooth/af_bluetooth.c
@@ -36,6 +36,7 @@
 #include <linux/init.h>
 #include <linux/poll.h>
 #include <net/sock.h>
+#include <asm/ioctls.h>
 
 #if defined(CONFIG_KMOD)
 #include <linux/kmod.h>
@@ -266,6 +267,8 @@ int bt_sock_recvmsg(struct kiocb *iocb, struct socket *sock,
 
 	skb_reset_transport_header(skb);
 	err = skb_copy_datagram_iovec(skb, 0, msg->msg_iov, copied);
+	if (err == 0)
+		sock_recv_timestamp(msg, sk, skb);
 
 	skb_free_datagram(sk, skb);
 
@@ -329,6 +332,31 @@ unsigned int bt_sock_poll(struct file * file, struct socket *sock, poll_table *w
 }
 EXPORT_SYMBOL(bt_sock_poll);
 
+int bt_sock_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg)
+{
+	struct sock *sk = sock->sk;
+	int err;
+
+	BT_DBG("sk %p cmd %x arg %lx", sk, cmd, arg);
+
+	switch (cmd) {
+	case SIOCGSTAMP:
+		err = sock_get_timestamp(sk, (struct timeval __user *) arg);
+		break;
+
+	case SIOCGSTAMPNS:
+		err = sock_get_timestampns(sk, (struct timespec __user *) arg);
+		break;
+
+	default:
+		err = -ENOIOCTLCMD;
+		break;
+	}
+
+	return err;
+}
+EXPORT_SYMBOL(bt_sock_ioctl);
+
 int bt_sock_wait_state(struct sock *sk, int state, unsigned long timeo)
 {
 	DECLARE_WAITQUEUE(wait, current);
diff --git a/net/bluetooth/l2cap.c b/net/bluetooth/l2cap.c
index 30ad59b717d5..4fcf24af7590 100644
--- a/net/bluetooth/l2cap.c
+++ b/net/bluetooth/l2cap.c
@@ -2388,9 +2388,9 @@ static const struct proto_ops l2cap_sock_ops = {
 	.sendmsg	= l2cap_sock_sendmsg,
 	.recvmsg	= bt_sock_recvmsg,
 	.poll		= bt_sock_poll,
+	.ioctl		= bt_sock_ioctl,
 	.mmap		= sock_no_mmap,
 	.socketpair	= sock_no_socketpair,
-	.ioctl		= sock_no_ioctl,
 	.shutdown	= l2cap_sock_shutdown,
 	.setsockopt	= l2cap_sock_setsockopt,
 	.getsockopt	= l2cap_sock_getsockopt
diff --git a/net/bluetooth/rfcomm/sock.c b/net/bluetooth/rfcomm/sock.c
index cacb1ab51f99..c3ed076481d8 100644
--- a/net/bluetooth/rfcomm/sock.c
+++ b/net/bluetooth/rfcomm/sock.c
@@ -690,6 +690,8 @@ static int rfcomm_sock_recvmsg(struct kiocb *iocb, struct socket *sock,
 		copied += chunk;
 		size   -= chunk;
 
+		sock_recv_timestamp(msg, sk, skb);
+
 		if (!(flags & MSG_PEEK)) {
 			atomic_sub(chunk, &sk->sk_rmem_alloc);
 
@@ -795,15 +797,20 @@ static int rfcomm_sock_ioctl(struct socket *sock, unsigned int cmd, unsigned lon
 	struct sock *sk = sock->sk;
 	int err;
 
-	lock_sock(sk);
+	BT_DBG("sk %p cmd %x arg %lx", sk, cmd, arg);
 
+	err = bt_sock_ioctl(sock, cmd, arg);
+
+	if (err == -ENOIOCTLCMD) {
 #ifdef CONFIG_BT_RFCOMM_TTY
-	err = rfcomm_dev_ioctl(sk, cmd, (void __user *)arg);
+		lock_sock(sk);
+		err = rfcomm_dev_ioctl(sk, cmd, (void __user *) arg);
+		release_sock(sk);
 #else
-	err = -EOPNOTSUPP;
+		err = -EOPNOTSUPP;
 #endif
+	}
 
-	release_sock(sk);
 	return err;
 }
 
diff --git a/net/bluetooth/sco.c b/net/bluetooth/sco.c
index b0d487e2db20..1ad226c9788c 100644
--- a/net/bluetooth/sco.c
+++ b/net/bluetooth/sco.c
@@ -921,7 +921,7 @@ static const struct proto_ops sco_sock_ops = {
 	.sendmsg	= sco_sock_sendmsg,
 	.recvmsg	= bt_sock_recvmsg,
 	.poll		= bt_sock_poll,
-	.ioctl		= sock_no_ioctl,
+	.ioctl		= bt_sock_ioctl,
 	.mmap		= sock_no_mmap,
 	.socketpair	= sock_no_socketpair,
 	.shutdown	= sock_no_shutdown,
-- 
cgit v1.2.3


From 8b6b3da765af9600b5edd8e3e84a20523e975884 Mon Sep 17 00:00:00 2001
From: Marcel Holtmann <marcel@holtmann.org>
Date: Mon, 14 Jul 2008 20:13:52 +0200
Subject: [Bluetooth] Store remote modem status for RFCOMM TTY

When switching a RFCOMM socket to a TTY, the remote modem status might
be needed later. Currently it is lost since the original configuration
is done via the socket interface. So store the modem status and reply
it when the socket has been converted to a TTY.

Signed-off-by: Denis Kenzior <denis.kenzior@trolltech.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
---
 include/net/bluetooth/rfcomm.h | 1 +
 net/bluetooth/rfcomm/core.c    | 4 ++++
 net/bluetooth/rfcomm/tty.c     | 3 +++
 3 files changed, 8 insertions(+)

(limited to 'include')

diff --git a/include/net/bluetooth/rfcomm.h b/include/net/bluetooth/rfcomm.h
index 8c54ff37ad4f..4dc8d92a4638 100644
--- a/include/net/bluetooth/rfcomm.h
+++ b/include/net/bluetooth/rfcomm.h
@@ -180,6 +180,7 @@ struct rfcomm_dlc {
 	u8            addr;
 	u8            priority;
 	u8            v24_sig;
+	u8            remote_v24_sig;
 	u8            mscex;
 	u8            out;
 
diff --git a/net/bluetooth/rfcomm/core.c b/net/bluetooth/rfcomm/core.c
index e56bcfc35a49..fcd2cafe70c8 100644
--- a/net/bluetooth/rfcomm/core.c
+++ b/net/bluetooth/rfcomm/core.c
@@ -1463,8 +1463,12 @@ static int rfcomm_recv_msc(struct rfcomm_session *s, int cr, struct sk_buff *skb
 			clear_bit(RFCOMM_TX_THROTTLED, &d->flags);
 
 		rfcomm_dlc_lock(d);
+
+		d->remote_v24_sig = msc->v24_sig;
+
 		if (d->modem_status)
 			d->modem_status(d, msc->v24_sig);
+
 		rfcomm_dlc_unlock(d);
 
 		rfcomm_send_msc(s, 0, dlci, msc->v24_sig);
diff --git a/net/bluetooth/rfcomm/tty.c b/net/bluetooth/rfcomm/tty.c
index 248802796e12..8fcca08cef8e 100644
--- a/net/bluetooth/rfcomm/tty.c
+++ b/net/bluetooth/rfcomm/tty.c
@@ -271,6 +271,9 @@ static int rfcomm_dev_add(struct rfcomm_dev_req *req, struct rfcomm_dlc *dlc)
 
 	dlc->owner = dev;
 	dev->dlc   = dlc;
+
+	rfcomm_dev_modem_status(dlc, dlc->remote_v24_sig);
+
 	rfcomm_dlc_unlock(dlc);
 
 	/* It's safe to call __module_get() here because socket already
-- 
cgit v1.2.3


From 49292d56352a6ab90d04c3448dd8b6106dfef2d6 Mon Sep 17 00:00:00 2001
From: Samuel Ortiz <samuel@sortiz.org>
Date: Fri, 4 Jul 2008 10:49:31 +0200
Subject: mac80211: power management wext hooks

This patch implements the power management routines wireless extensions
for mac80211.
For now we only support switching PS mode between on and off.

Signed-off-by: Samuel Ortiz <sameo@openedhand.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/net/mac80211.h |  2 ++
 net/mac80211/wext.c    | 43 +++++++++++++++++++++++++++++++++++++++++--
 2 files changed, 43 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index 656442c6b1c3..9672a04c4f7b 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -407,11 +407,13 @@ struct ieee80211_rx_status {
  * @IEEE80211_CONF_SHORT_SLOT_TIME: use 802.11g short slot time
  * @IEEE80211_CONF_RADIOTAP: add radiotap header at receive time (if supported)
  * @IEEE80211_CONF_SUPPORT_HT_MODE: use 802.11n HT capabilities (if supported)
+ * @IEEE80211_CONF_PS: Enable 802.11 power save mode
  */
 enum ieee80211_conf_flags {
 	IEEE80211_CONF_SHORT_SLOT_TIME	= (1<<0),
 	IEEE80211_CONF_RADIOTAP		= (1<<1),
 	IEEE80211_CONF_SUPPORT_HT_MODE	= (1<<2),
+	IEEE80211_CONF_PS		= (1<<3),
 };
 
 /**
diff --git a/net/mac80211/wext.c b/net/mac80211/wext.c
index 736c32e340f2..207971e9ad72 100644
--- a/net/mac80211/wext.c
+++ b/net/mac80211/wext.c
@@ -1009,6 +1009,45 @@ static int ieee80211_ioctl_giwencode(struct net_device *dev,
 	return 0;
 }
 
+static int ieee80211_ioctl_siwpower(struct net_device *dev,
+				    struct iw_request_info *info,
+				    struct iw_param *wrq,
+				    char *extra)
+{
+	struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr);
+	struct ieee80211_conf *conf = &local->hw.conf;
+
+	if (wrq->disabled) {
+		conf->flags &= ~IEEE80211_CONF_PS;
+		return ieee80211_hw_config(local);
+	}
+
+	switch (wrq->flags & IW_POWER_MODE) {
+	case IW_POWER_ON:       /* If not specified */
+	case IW_POWER_MODE:     /* If set all mask */
+	case IW_POWER_ALL_R:    /* If explicitely state all */
+		conf->flags |= IEEE80211_CONF_PS;
+		break;
+	default:                /* Otherwise we don't support it */
+		return -EINVAL;
+	}
+
+	return ieee80211_hw_config(local);
+}
+
+static int ieee80211_ioctl_giwpower(struct net_device *dev,
+				    struct iw_request_info *info,
+				    union iwreq_data *wrqu,
+				    char *extra)
+{
+	struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr);
+	struct ieee80211_conf *conf = &local->hw.conf;
+
+	wrqu->power.disabled = !(conf->flags & IEEE80211_CONF_PS);
+
+	return 0;
+}
+
 static int ieee80211_ioctl_siwauth(struct net_device *dev,
 				   struct iw_request_info *info,
 				   struct iw_param *data, char *extra)
@@ -1211,8 +1250,8 @@ static const iw_handler ieee80211_handler[] =
 	(iw_handler) ieee80211_ioctl_giwretry,		/* SIOCGIWRETRY */
 	(iw_handler) ieee80211_ioctl_siwencode,		/* SIOCSIWENCODE */
 	(iw_handler) ieee80211_ioctl_giwencode,		/* SIOCGIWENCODE */
-	(iw_handler) NULL,				/* SIOCSIWPOWER */
-	(iw_handler) NULL,				/* SIOCGIWPOWER */
+	(iw_handler) ieee80211_ioctl_siwpower,		/* SIOCSIWPOWER */
+	(iw_handler) ieee80211_ioctl_giwpower,		/* SIOCGIWPOWER */
 	(iw_handler) NULL,				/* -- hole -- */
 	(iw_handler) NULL,				/* -- hole -- */
 	(iw_handler) ieee80211_ioctl_siwgenie,		/* SIOCSIWGENIE */
-- 
cgit v1.2.3


From 9d139c810a2aa17365cc548d0cd2a189d8433c65 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes@sipsolutions.net>
Date: Wed, 9 Jul 2008 14:40:37 +0200
Subject: mac80211: revamp beacon configuration

This patch changes mac80211's beacon configuration handling
to never pass skbs to the driver directly but rather always
require the driver to use ieee80211_beacon_get(). Additionally,
it introduces "change flags" on the config_interface() call
to enable drivers to figure out what is changing. Finally, it
removes the beacon_update() driver callback in favour of
having IBSS beacon delivered by ieee80211_beacon_get() as well.

Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/net/wireless/ath5k/base.c           |  13 +++-
 drivers/net/wireless/b43/main.c             |  48 +++++-------
 drivers/net/wireless/b43legacy/main.c       |  45 +++++------
 drivers/net/wireless/iwlwifi/iwl3945-base.c |  19 ++++-
 drivers/net/wireless/iwlwifi/iwl4965-base.c |  18 ++++-
 drivers/net/wireless/rt2x00/rt2400pci.c     |   2 +-
 drivers/net/wireless/rt2x00/rt2500pci.c     |   2 +-
 drivers/net/wireless/rt2x00/rt2500usb.c     |   2 +-
 drivers/net/wireless/rt2x00/rt2x00.h        |   2 +
 drivers/net/wireless/rt2x00/rt2x00dev.c     |  10 +--
 drivers/net/wireless/rt2x00/rt2x00mac.c     |  22 ++++--
 drivers/net/wireless/rt2x00/rt61pci.c       |   2 +-
 drivers/net/wireless/rt2x00/rt73usb.c       |   2 +-
 drivers/net/wireless/zd1211rw/zd_mac.c      |  22 ++----
 include/net/mac80211.h                      |  49 ++++--------
 net/mac80211/cfg.c                          |   4 +-
 net/mac80211/ieee80211_i.h                  |   3 +-
 net/mac80211/main.c                         |  59 +++++++--------
 net/mac80211/mlme.c                         | 112 +++++++++++-----------------
 net/mac80211/tx.c                           |  36 +++++----
 net/mac80211/wext.c                         |   2 +-
 21 files changed, 226 insertions(+), 248 deletions(-)

(limited to 'include')

diff --git a/drivers/net/wireless/ath5k/base.c b/drivers/net/wireless/ath5k/base.c
index a43e9b25169b..217d506527a9 100644
--- a/drivers/net/wireless/ath5k/base.c
+++ b/drivers/net/wireless/ath5k/base.c
@@ -207,7 +207,6 @@ static struct ieee80211_ops ath5k_hw_ops = {
 	.get_tx_stats 	= ath5k_get_tx_stats,
 	.get_tsf 	= ath5k_get_tsf,
 	.reset_tsf 	= ath5k_reset_tsf,
-	.beacon_update 	= ath5k_beacon_update,
 };
 
 /*
@@ -2785,6 +2784,18 @@ ath5k_config_interface(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 		 * a clean way of letting us retrieve this yet. */
 		ath5k_hw_set_associd(ah, ah->ah_bssid, 0);
 	}
+
+	if (conf->changed & IEEE80211_IFCC_BEACON &&
+	    vif->type == IEEE80211_IF_TYPE_IBSS) {
+		struct sk_buff *beacon = ieee80211_beacon_get(hw, vif);
+		if (!beacon) {
+			ret = -ENOMEM;
+			goto unlock;
+		}
+		/* call old handler for now */
+		ath5k_beacon_update(hw, beacon);
+	}
+
 	mutex_unlock(&sc->lock);
 
 	return ath5k_reset(hw);
diff --git a/drivers/net/wireless/b43/main.c b/drivers/net/wireless/b43/main.c
index 9d2eb273b726..381dbd33dfc2 100644
--- a/drivers/net/wireless/b43/main.c
+++ b/drivers/net/wireless/b43/main.c
@@ -1675,14 +1675,24 @@ static void b43_beacon_update_trigger_work(struct work_struct *work)
 
 /* Asynchronously update the packet templates in template RAM.
  * Locking: Requires wl->irq_lock to be locked. */
-static void b43_update_templates(struct b43_wl *wl, struct sk_buff *beacon)
+static void b43_update_templates(struct b43_wl *wl)
 {
+	struct sk_buff *beacon;
+
 	/* This is the top half of the ansynchronous beacon update.
 	 * The bottom half is the beacon IRQ.
 	 * Beacon update must be asynchronous to avoid sending an
 	 * invalid beacon. This can happen for example, if the firmware
 	 * transmits a beacon while we are updating it. */
 
+	/* We could modify the existing beacon and set the aid bit in
+	 * the TIM field, but that would probably require resizing and
+	 * moving of data within the beacon template.
+	 * Simply request a new beacon and let mac80211 do the hard work. */
+	beacon = ieee80211_beacon_get(wl->hw, wl->vif);
+	if (unlikely(!beacon))
+		return;
+
 	if (wl->current_beacon)
 		dev_kfree_skb_any(wl->current_beacon);
 	wl->current_beacon = beacon;
@@ -3645,10 +3655,14 @@ static int b43_op_config_interface(struct ieee80211_hw *hw,
 	if (b43_status(dev) >= B43_STAT_INITIALIZED) {
 		if (b43_is_mode(wl, IEEE80211_IF_TYPE_AP) ||
 		    b43_is_mode(wl, IEEE80211_IF_TYPE_MESH_POINT)) {
-			B43_WARN_ON(conf->type != wl->if_type);
-			b43_set_ssid(dev, conf->ssid, conf->ssid_len);
-			if (conf->beacon)
-				b43_update_templates(wl, conf->beacon);
+			B43_WARN_ON(vif->type != wl->if_type);
+			if (conf->changed & IEEE80211_IFCC_SSID)
+				b43_set_ssid(dev, conf->ssid, conf->ssid_len);
+			if (conf->changed & IEEE80211_IFCC_BEACON)
+				b43_update_templates(wl);
+		} else if (b43_is_mode(wl, IEEE80211_IF_TYPE_IBSS)) {
+			if (conf->changed & IEEE80211_IFCC_BEACON)
+				b43_update_templates(wl);
 		}
 		b43_write_mac_bssid_templates(dev);
 	}
@@ -4334,33 +4348,12 @@ out_unlock:
 }
 
 static int b43_op_beacon_set_tim(struct ieee80211_hw *hw, int aid, int set)
-{
-	struct b43_wl *wl = hw_to_b43_wl(hw);
-	struct sk_buff *beacon;
-	unsigned long flags;
-
-	/* We could modify the existing beacon and set the aid bit in
-	 * the TIM field, but that would probably require resizing and
-	 * moving of data within the beacon template.
-	 * Simply request a new beacon and let mac80211 do the hard work. */
-	beacon = ieee80211_beacon_get(hw, wl->vif);
-	if (unlikely(!beacon))
-		return -ENOMEM;
-	spin_lock_irqsave(&wl->irq_lock, flags);
-	b43_update_templates(wl, beacon);
-	spin_unlock_irqrestore(&wl->irq_lock, flags);
-
-	return 0;
-}
-
-static int b43_op_ibss_beacon_update(struct ieee80211_hw *hw,
-				     struct sk_buff *beacon)
 {
 	struct b43_wl *wl = hw_to_b43_wl(hw);
 	unsigned long flags;
 
 	spin_lock_irqsave(&wl->irq_lock, flags);
-	b43_update_templates(wl, beacon);
+	b43_update_templates(wl);
 	spin_unlock_irqrestore(&wl->irq_lock, flags);
 
 	return 0;
@@ -4391,7 +4384,6 @@ static const struct ieee80211_ops b43_hw_ops = {
 	.stop			= b43_op_stop,
 	.set_retry_limit	= b43_op_set_retry_limit,
 	.set_tim		= b43_op_beacon_set_tim,
-	.beacon_update		= b43_op_ibss_beacon_update,
 	.sta_notify		= b43_op_sta_notify,
 };
 
diff --git a/drivers/net/wireless/b43legacy/main.c b/drivers/net/wireless/b43legacy/main.c
index 069157eea05c..a1b8bf3ee732 100644
--- a/drivers/net/wireless/b43legacy/main.c
+++ b/drivers/net/wireless/b43legacy/main.c
@@ -1138,14 +1138,22 @@ static void b43legacy_write_probe_resp_template(struct b43legacy_wldev *dev,
 
 /* Asynchronously update the packet templates in template RAM.
  * Locking: Requires wl->irq_lock to be locked. */
-static void b43legacy_update_templates(struct b43legacy_wl *wl,
-				       struct sk_buff *beacon)
+static void b43legacy_update_templates(struct b43legacy_wl *wl)
 {
+	struct sk_buff *beacon;
 	/* This is the top half of the ansynchronous beacon update. The bottom
 	 * half is the beacon IRQ. Beacon update must be asynchronous to avoid
 	 * sending an invalid beacon. This can happen for example, if the
 	 * firmware transmits a beacon while we are updating it. */
 
+	/* We could modify the existing beacon and set the aid bit in the TIM
+	 * field, but that would probably require resizing and moving of data
+	 * within the beacon template. Simply request a new beacon and let
+	 * mac80211 do the hard work. */
+	beacon = ieee80211_beacon_get(wl->hw, wl->vif);
+	if (unlikely(!beacon))
+		return;
+
 	if (wl->current_beacon)
 		dev_kfree_skb_any(wl->current_beacon);
 	wl->current_beacon = beacon;
@@ -2727,10 +2735,13 @@ static int b43legacy_op_config_interface(struct ieee80211_hw *hw,
 		memset(wl->bssid, 0, ETH_ALEN);
 	if (b43legacy_status(dev) >= B43legacy_STAT_INITIALIZED) {
 		if (b43legacy_is_mode(wl, IEEE80211_IF_TYPE_AP)) {
-			B43legacy_WARN_ON(conf->type != IEEE80211_IF_TYPE_AP);
+			B43legacy_WARN_ON(vif->type != IEEE80211_IF_TYPE_AP);
 			b43legacy_set_ssid(dev, conf->ssid, conf->ssid_len);
-			if (conf->beacon)
-				b43legacy_update_templates(wl, conf->beacon);
+			if (conf->changed & IEEE80211_IFCC_BEACON)
+				b43legacy_update_templates(wl);
+		} else if (b43legacy_is_mode(wl, IEEE80211_IF_TYPE_IBSS)) {
+			if (conf->changed & IEEE80211_IFCC_BEACON)
+				b43legacy_update_templates(wl);
 		}
 		b43legacy_write_mac_bssid_templates(dev);
 	}
@@ -3394,33 +3405,12 @@ out_unlock:
 
 static int b43legacy_op_beacon_set_tim(struct ieee80211_hw *hw,
 				       int aid, int set)
-{
-	struct b43legacy_wl *wl = hw_to_b43legacy_wl(hw);
-	struct sk_buff *beacon;
-	unsigned long flags;
-
-	/* We could modify the existing beacon and set the aid bit in the TIM
-	 * field, but that would probably require resizing and moving of data
-	 * within the beacon template. Simply request a new beacon and let
-	 * mac80211 do the hard work. */
-	beacon = ieee80211_beacon_get(hw, wl->vif);
-	if (unlikely(!beacon))
-		return -ENOMEM;
-	spin_lock_irqsave(&wl->irq_lock, flags);
-	b43legacy_update_templates(wl, beacon);
-	spin_unlock_irqrestore(&wl->irq_lock, flags);
-
-	return 0;
-}
-
-static int b43legacy_op_ibss_beacon_update(struct ieee80211_hw *hw,
-					   struct sk_buff *beacon)
 {
 	struct b43legacy_wl *wl = hw_to_b43legacy_wl(hw);
 	unsigned long flags;
 
 	spin_lock_irqsave(&wl->irq_lock, flags);
-	b43legacy_update_templates(wl, beacon);
+	b43legacy_update_templates(wl);
 	spin_unlock_irqrestore(&wl->irq_lock, flags);
 
 	return 0;
@@ -3440,7 +3430,6 @@ static const struct ieee80211_ops b43legacy_hw_ops = {
 	.stop			= b43legacy_op_stop,
 	.set_retry_limit	= b43legacy_op_set_retry_limit,
 	.set_tim		= b43legacy_op_beacon_set_tim,
-	.beacon_update		= b43legacy_op_ibss_beacon_update,
 };
 
 /* Hard-reset the chip. Do not call this directly.
diff --git a/drivers/net/wireless/iwlwifi/iwl3945-base.c b/drivers/net/wireless/iwlwifi/iwl3945-base.c
index 1a7d18fea89d..7d015f86ee8c 100644
--- a/drivers/net/wireless/iwlwifi/iwl3945-base.c
+++ b/drivers/net/wireless/iwlwifi/iwl3945-base.c
@@ -6907,6 +6907,9 @@ static void iwl3945_config_ap(struct iwl3945_priv *priv)
 	 * clear sta table, add BCAST sta... */
 }
 
+/* temporary */
+static int iwl3945_mac_beacon_update(struct ieee80211_hw *hw, struct sk_buff *skb);
+
 static int iwl3945_mac_config_interface(struct ieee80211_hw *hw,
 					struct ieee80211_vif *vif,
 				    struct ieee80211_if_conf *conf)
@@ -6924,10 +6927,21 @@ static int iwl3945_mac_config_interface(struct ieee80211_hw *hw,
 		return 0;
 	}
 
+	/* handle this temporarily here */
+	if (priv->iw_mode == IEEE80211_IF_TYPE_IBSS &&
+	    conf->changed & IEEE80211_IFCC_BEACON) {
+		struct sk_buff *beacon = ieee80211_beacon_get(hw, vif);
+		if (!beacon)
+			return -ENOMEM;
+		rc = iwl3945_mac_beacon_update(hw, beacon);
+		if (rc)
+			return rc;
+	}
+
 	/* XXX: this MUST use conf->mac_addr */
 
 	if ((priv->iw_mode == IEEE80211_IF_TYPE_AP) &&
-	    (!conf->beacon || !conf->ssid_len)) {
+	    (!conf->ssid_len)) {
 		IWL_DEBUG_MAC80211
 		    ("Leaving in AP mode because HostAPD is not ready.\n");
 		return 0;
@@ -6959,7 +6973,7 @@ static int iwl3945_mac_config_interface(struct ieee80211_hw *hw,
 		if (priv->ibss_beacon)
 			dev_kfree_skb(priv->ibss_beacon);
 
-		priv->ibss_beacon = conf->beacon;
+		priv->ibss_beacon = ieee80211_beacon_get(hw, vif);
 	}
 
 	if (iwl3945_is_rfkill(priv))
@@ -7940,7 +7954,6 @@ static struct ieee80211_ops iwl3945_hw_ops = {
 	.conf_tx = iwl3945_mac_conf_tx,
 	.get_tsf = iwl3945_mac_get_tsf,
 	.reset_tsf = iwl3945_mac_reset_tsf,
-	.beacon_update = iwl3945_mac_beacon_update,
 	.hw_scan = iwl3945_mac_hw_scan
 };
 
diff --git a/drivers/net/wireless/iwlwifi/iwl4965-base.c b/drivers/net/wireless/iwlwifi/iwl4965-base.c
index 7f65d9123b2a..d6fe0ded59d7 100644
--- a/drivers/net/wireless/iwlwifi/iwl4965-base.c
+++ b/drivers/net/wireless/iwlwifi/iwl4965-base.c
@@ -2912,6 +2912,9 @@ static void iwl4965_config_ap(struct iwl_priv *priv)
 	 * clear sta table, add BCAST sta... */
 }
 
+/* temporary */
+static int iwl4965_mac_beacon_update(struct ieee80211_hw *hw, struct sk_buff *skb);
+
 static int iwl4965_mac_config_interface(struct ieee80211_hw *hw,
 					struct ieee80211_vif *vif,
 				    struct ieee80211_if_conf *conf)
@@ -2929,8 +2932,18 @@ static int iwl4965_mac_config_interface(struct ieee80211_hw *hw,
 		return 0;
 	}
 
+	if (priv->iw_mode == IEEE80211_IF_TYPE_IBSS &&
+	    conf->changed & IEEE80211_IFCC_BEACON) {
+		struct sk_buff *beacon = ieee80211_beacon_get(hw, vif);
+		if (!beacon)
+			return -ENOMEM;
+		rc = iwl4965_mac_beacon_update(hw, beacon);
+		if (rc)
+			return rc;
+	}
+
 	if ((priv->iw_mode == IEEE80211_IF_TYPE_AP) &&
-	    (!conf->beacon || !conf->ssid_len)) {
+	    (!conf->ssid_len)) {
 		IWL_DEBUG_MAC80211
 		    ("Leaving in AP mode because HostAPD is not ready.\n");
 		return 0;
@@ -2962,7 +2975,7 @@ static int iwl4965_mac_config_interface(struct ieee80211_hw *hw,
 		if (priv->ibss_beacon)
 			dev_kfree_skb(priv->ibss_beacon);
 
-		priv->ibss_beacon = conf->beacon;
+		priv->ibss_beacon = ieee80211_beacon_get(hw, vif);
 	}
 
 	if (iwl_is_rfkill(priv))
@@ -4090,7 +4103,6 @@ static struct ieee80211_ops iwl4965_hw_ops = {
 	.get_tx_stats = iwl4965_mac_get_tx_stats,
 	.conf_tx = iwl4965_mac_conf_tx,
 	.reset_tsf = iwl4965_mac_reset_tsf,
-	.beacon_update = iwl4965_mac_beacon_update,
 	.bss_info_changed = iwl4965_bss_info_changed,
 	.ampdu_action = iwl4965_mac_ampdu_action,
 	.hw_scan = iwl4965_mac_hw_scan
diff --git a/drivers/net/wireless/rt2x00/rt2400pci.c b/drivers/net/wireless/rt2x00/rt2400pci.c
index ee953ca0c6a3..89b874ca6107 100644
--- a/drivers/net/wireless/rt2x00/rt2400pci.c
+++ b/drivers/net/wireless/rt2x00/rt2400pci.c
@@ -1581,7 +1581,6 @@ static const struct ieee80211_ops rt2400pci_mac80211_ops = {
 	.conf_tx		= rt2400pci_conf_tx,
 	.get_tx_stats		= rt2x00mac_get_tx_stats,
 	.get_tsf		= rt2400pci_get_tsf,
-	.beacon_update		= rt2400pci_beacon_update,
 	.tx_last_beacon		= rt2400pci_tx_last_beacon,
 };
 
@@ -1601,6 +1600,7 @@ static const struct rt2x00lib_ops rt2400pci_rt2x00_ops = {
 	.write_tx_data		= rt2x00pci_write_tx_data,
 	.kick_tx_queue		= rt2400pci_kick_tx_queue,
 	.fill_rxdone		= rt2400pci_fill_rxdone,
+	.beacon_update		= rt2400pci_beacon_update,
 	.config_filter		= rt2400pci_config_filter,
 	.config_intf		= rt2400pci_config_intf,
 	.config_erp		= rt2400pci_config_erp,
diff --git a/drivers/net/wireless/rt2x00/rt2500pci.c b/drivers/net/wireless/rt2x00/rt2500pci.c
index 0423c251c78e..a64bb18322e9 100644
--- a/drivers/net/wireless/rt2x00/rt2500pci.c
+++ b/drivers/net/wireless/rt2x00/rt2500pci.c
@@ -1875,7 +1875,6 @@ static const struct ieee80211_ops rt2500pci_mac80211_ops = {
 	.conf_tx		= rt2x00mac_conf_tx,
 	.get_tx_stats		= rt2x00mac_get_tx_stats,
 	.get_tsf		= rt2500pci_get_tsf,
-	.beacon_update		= rt2500pci_beacon_update,
 	.tx_last_beacon		= rt2500pci_tx_last_beacon,
 };
 
@@ -1895,6 +1894,7 @@ static const struct rt2x00lib_ops rt2500pci_rt2x00_ops = {
 	.write_tx_data		= rt2x00pci_write_tx_data,
 	.kick_tx_queue		= rt2500pci_kick_tx_queue,
 	.fill_rxdone		= rt2500pci_fill_rxdone,
+	.beacon_update		= rt2500pci_beacon_update,
 	.config_filter		= rt2500pci_config_filter,
 	.config_intf		= rt2500pci_config_intf,
 	.config_erp		= rt2500pci_config_erp,
diff --git a/drivers/net/wireless/rt2x00/rt2500usb.c b/drivers/net/wireless/rt2x00/rt2500usb.c
index 0dd1cb537b92..8ce1726d7508 100644
--- a/drivers/net/wireless/rt2x00/rt2500usb.c
+++ b/drivers/net/wireless/rt2x00/rt2500usb.c
@@ -1775,7 +1775,6 @@ static const struct ieee80211_ops rt2500usb_mac80211_ops = {
 	.bss_info_changed	= rt2x00mac_bss_info_changed,
 	.conf_tx		= rt2x00mac_conf_tx,
 	.get_tx_stats		= rt2x00mac_get_tx_stats,
-	.beacon_update		= rt2500usb_beacon_update,
 };
 
 static const struct rt2x00lib_ops rt2500usb_rt2x00_ops = {
@@ -1793,6 +1792,7 @@ static const struct rt2x00lib_ops rt2500usb_rt2x00_ops = {
 	.get_tx_data_len	= rt2500usb_get_tx_data_len,
 	.kick_tx_queue		= rt2500usb_kick_tx_queue,
 	.fill_rxdone		= rt2500usb_fill_rxdone,
+	.beacon_update		= rt2500usb_beacon_update,
 	.config_filter		= rt2500usb_config_filter,
 	.config_intf		= rt2500usb_config_intf,
 	.config_erp		= rt2500usb_config_erp,
diff --git a/drivers/net/wireless/rt2x00/rt2x00.h b/drivers/net/wireless/rt2x00/rt2x00.h
index c07d9ef383f0..9ad3ce43e6cd 100644
--- a/drivers/net/wireless/rt2x00/rt2x00.h
+++ b/drivers/net/wireless/rt2x00/rt2x00.h
@@ -534,6 +534,8 @@ struct rt2x00lib_ops {
 	/*
 	 * Configuration handlers.
 	 */
+	int (*beacon_update) (struct ieee80211_hw *hw, struct sk_buff *bcn);
+
 	void (*config_filter) (struct rt2x00_dev *rt2x00dev,
 			       const unsigned int filter_flags);
 	void (*config_intf) (struct rt2x00_dev *rt2x00dev,
diff --git a/drivers/net/wireless/rt2x00/rt2x00dev.c b/drivers/net/wireless/rt2x00/rt2x00dev.c
index b48c04e80a38..1b7f87799a7e 100644
--- a/drivers/net/wireless/rt2x00/rt2x00dev.c
+++ b/drivers/net/wireless/rt2x00/rt2x00dev.c
@@ -409,7 +409,6 @@ static void rt2x00lib_intf_scheduled_iter(void *data, u8 *mac,
 {
 	struct rt2x00_dev *rt2x00dev = data;
 	struct rt2x00_intf *intf = vif_to_intf(vif);
-	struct sk_buff *skb;
 	struct ieee80211_bss_conf conf;
 	int delayed_flags;
 
@@ -436,10 +435,11 @@ static void rt2x00lib_intf_scheduled_iter(void *data, u8 *mac,
 		return;
 
 	if (delayed_flags & DELAYED_UPDATE_BEACON) {
-		skb = ieee80211_beacon_get(rt2x00dev->hw, vif);
-		if (skb &&
-		    rt2x00dev->ops->hw->beacon_update(rt2x00dev->hw, skb))
-			dev_kfree_skb(skb);
+		struct ieee80211_if_conf conf;
+		conf.bssid = conf.ssid = NULL;
+		conf.ssid_len = 0;
+		conf.changed = IEEE80211_IFCC_BEACON;
+		rt2x00dev->ops->hw->config_interface(rt2x00dev->hw, vif, &conf);
 	}
 
 	if (delayed_flags & DELAYED_CONFIG_ERP)
diff --git a/drivers/net/wireless/rt2x00/rt2x00mac.c b/drivers/net/wireless/rt2x00/rt2x00mac.c
index 3a1fb6d47e5d..84b51f49175e 100644
--- a/drivers/net/wireless/rt2x00/rt2x00mac.c
+++ b/drivers/net/wireless/rt2x00/rt2x00mac.c
@@ -348,6 +348,7 @@ int rt2x00mac_config_interface(struct ieee80211_hw *hw,
 {
 	struct rt2x00_dev *rt2x00dev = hw->priv;
 	struct rt2x00_intf *intf = vif_to_intf(vif);
+	struct sk_buff *beacon;
 	int status;
 
 	/*
@@ -363,8 +364,11 @@ int rt2x00mac_config_interface(struct ieee80211_hw *hw,
 	 * If the interface does not work in master mode,
 	 * then the bssid value in the interface structure
 	 * should now be set.
+	 *
+	 * conf->bssid can be NULL if coming from the internal
+	 * beacon update routine.
 	 */
-	if (conf->type != IEEE80211_IF_TYPE_AP)
+	if (conf->bssid && vif->type != IEEE80211_IF_TYPE_AP)
 		memcpy(&intf->bssid, conf->bssid, ETH_ALEN);
 
 	spin_unlock(&intf->lock);
@@ -375,17 +379,23 @@ int rt2x00mac_config_interface(struct ieee80211_hw *hw,
 	 * values as arguments we make keep access to rt2x00_intf thread safe
 	 * even without the lock.
 	 */
-	rt2x00lib_config_intf(rt2x00dev, intf, conf->type, NULL, conf->bssid);
+	rt2x00lib_config_intf(rt2x00dev, intf, vif->type, NULL, conf->bssid);
 
 	/*
-	 * We only need to initialize the beacon when master mode is enabled.
+	 * We only need to initialize the beacon when in master/ibss mode.
 	 */
-	if (conf->type != IEEE80211_IF_TYPE_AP || !conf->beacon)
+	if ((vif->type != IEEE80211_IF_TYPE_AP &&
+	     vif->type != IEEE80211_IF_TYPE_IBSS) ||
+	    !(conf->changed & IEEE80211_IFCC_BEACON))
 		return 0;
 
-	status = rt2x00dev->ops->hw->beacon_update(rt2x00dev->hw, conf->beacon);
+	beacon = ieee80211_beacon_get(rt2x00dev->hw, vif);
+	if (!beacon)
+		return -ENOMEM;
+
+	status = rt2x00dev->ops->lib->beacon_update(rt2x00dev->hw, beacon);
 	if (status)
-		dev_kfree_skb(conf->beacon);
+		dev_kfree_skb(beacon);
 
 	return status;
 }
diff --git a/drivers/net/wireless/rt2x00/rt61pci.c b/drivers/net/wireless/rt2x00/rt61pci.c
index bbf1048f6400..852d193a11a9 100644
--- a/drivers/net/wireless/rt2x00/rt61pci.c
+++ b/drivers/net/wireless/rt2x00/rt61pci.c
@@ -2427,7 +2427,6 @@ static const struct ieee80211_ops rt61pci_mac80211_ops = {
 	.conf_tx		= rt2x00mac_conf_tx,
 	.get_tx_stats		= rt2x00mac_get_tx_stats,
 	.get_tsf		= rt61pci_get_tsf,
-	.beacon_update		= rt61pci_beacon_update,
 };
 
 static const struct rt2x00lib_ops rt61pci_rt2x00_ops = {
@@ -2449,6 +2448,7 @@ static const struct rt2x00lib_ops rt61pci_rt2x00_ops = {
 	.write_tx_data		= rt2x00pci_write_tx_data,
 	.kick_tx_queue		= rt61pci_kick_tx_queue,
 	.fill_rxdone		= rt61pci_fill_rxdone,
+	.beacon_update		= rt61pci_beacon_update,
 	.config_filter		= rt61pci_config_filter,
 	.config_intf		= rt61pci_config_intf,
 	.config_erp		= rt61pci_config_erp,
diff --git a/drivers/net/wireless/rt2x00/rt73usb.c b/drivers/net/wireless/rt2x00/rt73usb.c
index 3ef318e098e7..657200972424 100644
--- a/drivers/net/wireless/rt2x00/rt73usb.c
+++ b/drivers/net/wireless/rt2x00/rt73usb.c
@@ -2031,7 +2031,6 @@ static const struct ieee80211_ops rt73usb_mac80211_ops = {
 	.conf_tx		= rt2x00mac_conf_tx,
 	.get_tx_stats		= rt2x00mac_get_tx_stats,
 	.get_tsf		= rt73usb_get_tsf,
-	.beacon_update		= rt73usb_beacon_update,
 };
 
 static const struct rt2x00lib_ops rt73usb_rt2x00_ops = {
@@ -2052,6 +2051,7 @@ static const struct rt2x00lib_ops rt73usb_rt2x00_ops = {
 	.get_tx_data_len	= rt73usb_get_tx_data_len,
 	.kick_tx_queue		= rt73usb_kick_tx_queue,
 	.fill_rxdone		= rt73usb_fill_rxdone,
+	.beacon_update		= rt73usb_beacon_update,
 	.config_filter		= rt73usb_config_filter,
 	.config_intf		= rt73usb_config_intf,
 	.config_erp		= rt73usb_config_erp,
diff --git a/drivers/net/wireless/zd1211rw/zd_mac.c b/drivers/net/wireless/zd1211rw/zd_mac.c
index 665f76af2fec..68e2749b2ce8 100644
--- a/drivers/net/wireless/zd1211rw/zd_mac.c
+++ b/drivers/net/wireless/zd1211rw/zd_mac.c
@@ -727,15 +727,19 @@ static int zd_op_config_interface(struct ieee80211_hw *hw,
 	if (mac->type == IEEE80211_IF_TYPE_MESH_POINT ||
 	    mac->type == IEEE80211_IF_TYPE_IBSS) {
 		associated = true;
-		if (conf->beacon) {
-			r = zd_mac_config_beacon(hw, conf->beacon);
+		if (conf->changed & IEEE80211_IFCC_BEACON) {
+			struct sk_buff *beacon = ieee80211_beacon_get(hw, vif);
+
+			if (!beacon)
+				return -ENOMEM;
+			r = zd_mac_config_beacon(hw, beacon);
 			if (r < 0)
 				return r;
 			r = zd_set_beacon_interval(&mac->chip, BCN_MODE_IBSS |
 					hw->conf.beacon_int);
 			if (r < 0)
 				return r;
-			kfree_skb(conf->beacon);
+			kfree_skb(beacon);
 		}
 	} else
 		associated = is_valid_ether_addr(conf->bssid);
@@ -889,17 +893,6 @@ static void zd_op_bss_info_changed(struct ieee80211_hw *hw,
 	}
 }
 
-static int zd_op_beacon_update(struct ieee80211_hw *hw,
-			       struct sk_buff *skb)
-{
-	struct zd_mac *mac = zd_hw_mac(hw);
-	zd_mac_config_beacon(hw, skb);
-	kfree_skb(skb);
-	zd_set_beacon_interval(&mac->chip, BCN_MODE_IBSS |
-					hw->conf.beacon_int);
-	return 0;
-}
-
 static const struct ieee80211_ops zd_ops = {
 	.tx			= zd_op_tx,
 	.start			= zd_op_start,
@@ -910,7 +903,6 @@ static const struct ieee80211_ops zd_ops = {
 	.config_interface	= zd_op_config_interface,
 	.configure_filter	= zd_op_configure_filter,
 	.bss_info_changed	= zd_op_bss_info_changed,
-	.beacon_update		= zd_op_beacon_update,
 };
 
 struct ieee80211_hw *zd_mac_alloc_hw(struct usb_interface *intf)
diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index 9672a04c4f7b..47c072eab42c 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -528,34 +528,39 @@ struct ieee80211_if_init_conf {
 	void *mac_addr;
 };
 
+/**
+ * enum ieee80211_if_conf_change - interface config change flags
+ *
+ * @IEEE80211_IFCC_BSSID: The BSSID changed.
+ * @IEEE80211_IFCC_SSID: The SSID changed.
+ * @IEEE80211_IFCC_BEACON: The beacon for this interface changed
+ *	(currently AP and MESH only), use ieee80211_beacon_get().
+ */
+enum ieee80211_if_conf_change {
+	IEEE80211_IFCC_BSSID	= BIT(0),
+	IEEE80211_IFCC_SSID	= BIT(1),
+	IEEE80211_IFCC_BEACON	= BIT(2),
+};
+
 /**
  * struct ieee80211_if_conf - configuration of an interface
  *
- * @type: type of the interface. This is always the same as was specified in
- *	&struct ieee80211_if_init_conf. The type of an interface never changes
- *	during the life of the interface; this field is present only for
- *	convenience.
+ * @changed: parameters that have changed, see &enum ieee80211_if_conf_change.
  * @bssid: BSSID of the network we are associated to/creating.
  * @ssid: used (together with @ssid_len) by drivers for hardware that
  *	generate beacons independently. The pointer is valid only during the
  *	config_interface() call, so copy the value somewhere if you need
  *	it.
  * @ssid_len: length of the @ssid field.
- * @beacon: beacon template. Valid only if @host_gen_beacon_template in
- *	&struct ieee80211_hw is set. The driver is responsible of freeing
- *	the sk_buff.
- * @beacon_control: tx_control for the beacon template, this field is only
- *	valid when the @beacon field was set.
  *
  * This structure is passed to the config_interface() callback of
  * &struct ieee80211_hw.
  */
 struct ieee80211_if_conf {
-	int type;
+	u32 changed;
 	u8 *bssid;
 	u8 *ssid;
 	size_t ssid_len;
-	struct sk_buff *beacon;
 };
 
 /**
@@ -683,15 +688,6 @@ enum ieee80211_tkip_key_type {
  * any particular flags. There are some exceptions to this rule,
  * however, so you are advised to review these flags carefully.
  *
- * @IEEE80211_HW_HOST_GEN_BEACON_TEMPLATE:
- *	The device only needs to be supplied with a beacon template.
- *	If you need the host to generate each beacon then don't use
- *	this flag and call ieee80211_beacon_get() when you need the
- *	next beacon frame. Note that if you set this flag, you must
- *	implement the set_tim() callback for powersave mode to work
- *	properly.
- *	This flag is only relevant for access-point mode.
- *
  * @IEEE80211_HW_RX_INCLUDES_FCS:
  *	Indicates that received frames passed to the stack include
  *	the FCS at the end.
@@ -1151,17 +1147,6 @@ enum ieee80211_ampdu_mlme_action {
  *	function is optional if the firmware/hardware takes full care of
  *	TSF synchronization.
  *
- * @beacon_update: Setup beacon data for IBSS beacons. Unlike access point,
- *	IBSS uses a fixed beacon frame which is configured using this
- *	function.
- *	If the driver returns success (0) from this callback, it owns
- *	the skb. That means the driver is responsible to kfree_skb() it.
- *	The control structure is not dynamically allocated. That means the
- *	driver does not own the pointer and if it needs it somewhere
- *	outside of the context of this function, it must copy it
- *	somewhere else.
- *	This handler is required only for IBSS mode.
- *
  * @tx_last_beacon: Determine whether the last IBSS beacon was sent by us.
  *	This is needed only for IBSS mode and the result of this function is
  *	used to determine whether to reply to Probe Requests.
@@ -1219,8 +1204,6 @@ struct ieee80211_ops {
 			    struct ieee80211_tx_queue_stats *stats);
 	u64 (*get_tsf)(struct ieee80211_hw *hw);
 	void (*reset_tsf)(struct ieee80211_hw *hw);
-	int (*beacon_update)(struct ieee80211_hw *hw,
-			     struct sk_buff *skb);
 	int (*tx_last_beacon)(struct ieee80211_hw *hw);
 	int (*ampdu_action)(struct ieee80211_hw *hw,
 			    enum ieee80211_ampdu_mlme_action action,
diff --git a/net/mac80211/cfg.c b/net/mac80211/cfg.c
index ea0301025c15..8e7ba0e62cf5 100644
--- a/net/mac80211/cfg.c
+++ b/net/mac80211/cfg.c
@@ -469,7 +469,7 @@ static int ieee80211_config_beacon(struct ieee80211_sub_if_data *sdata,
 
 	kfree(old);
 
-	return ieee80211_if_config_beacon(sdata->dev);
+	return ieee80211_if_config(sdata, IEEE80211_IFCC_BEACON);
 }
 
 static int ieee80211_add_beacon(struct wiphy *wiphy, struct net_device *dev,
@@ -523,7 +523,7 @@ static int ieee80211_del_beacon(struct wiphy *wiphy, struct net_device *dev)
 	synchronize_rcu();
 	kfree(old);
 
-	return ieee80211_if_config_beacon(dev);
+	return ieee80211_if_config(sdata, IEEE80211_IFCC_BEACON);
 }
 
 /* Layer 2 Update frame (802.2 Type 1 LLC XID Update response) */
diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h
index 2146c0c436d2..934c3ef4f0bc 100644
--- a/net/mac80211/ieee80211_i.h
+++ b/net/mac80211/ieee80211_i.h
@@ -854,8 +854,7 @@ static inline int ieee80211_bssid_match(const u8 *raddr, const u8 *addr)
 
 /* ieee80211.c */
 int ieee80211_hw_config(struct ieee80211_local *local);
-int ieee80211_if_config(struct net_device *dev);
-int ieee80211_if_config_beacon(struct net_device *dev);
+int ieee80211_if_config(struct ieee80211_sub_if_data *sdata, u32 changed);
 void ieee80211_tx_set_protected(struct ieee80211_tx_data *tx);
 u32 ieee80211_handle_ht(struct ieee80211_local *local, int enable_ht,
 			struct ieee80211_ht_info *req_ht_cap,
diff --git a/net/mac80211/main.c b/net/mac80211/main.c
index 0759ab2ca3ff..36859e794928 100644
--- a/net/mac80211/main.c
+++ b/net/mac80211/main.c
@@ -307,7 +307,8 @@ static int ieee80211_open(struct net_device *dev)
 		if (res)
 			goto err_stop;
 
-		ieee80211_if_config(dev);
+		if (ieee80211_vif_is_mesh(&sdata->vif))
+			ieee80211_start_mesh(sdata->dev);
 		changed |= ieee80211_reset_erp_info(dev);
 		ieee80211_bss_info_change_notify(sdata, changed);
 		ieee80211_enable_keys(sdata);
@@ -985,57 +986,47 @@ void ieee80211_if_setup(struct net_device *dev)
 
 /* everything else */
 
-static int __ieee80211_if_config(struct net_device *dev,
-				 struct sk_buff *beacon)
+int ieee80211_if_config(struct ieee80211_sub_if_data *sdata, u32 changed)
 {
-	struct ieee80211_sub_if_data *sdata = IEEE80211_DEV_TO_SUB_IF(dev);
-	struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr);
+	struct ieee80211_local *local = sdata->local;
 	struct ieee80211_if_conf conf;
 
-	if (!local->ops->config_interface || !netif_running(dev))
+	if (WARN_ON(!netif_running(sdata->dev)))
+		return 0;
+
+	if (!local->ops->config_interface)
 		return 0;
 
 	memset(&conf, 0, sizeof(conf));
-	conf.type = sdata->vif.type;
+	conf.changed = changed;
+
 	if (sdata->vif.type == IEEE80211_IF_TYPE_STA ||
 	    sdata->vif.type == IEEE80211_IF_TYPE_IBSS) {
 		conf.bssid = sdata->u.sta.bssid;
 		conf.ssid = sdata->u.sta.ssid;
 		conf.ssid_len = sdata->u.sta.ssid_len;
-	} else if (ieee80211_vif_is_mesh(&sdata->vif)) {
-		conf.beacon = beacon;
-		ieee80211_start_mesh(dev);
 	} else if (sdata->vif.type == IEEE80211_IF_TYPE_AP) {
+		conf.bssid = sdata->dev->dev_addr;
 		conf.ssid = sdata->u.ap.ssid;
 		conf.ssid_len = sdata->u.ap.ssid_len;
-		conf.beacon = beacon;
+	} else if (ieee80211_vif_is_mesh(&sdata->vif)) {
+		u8 zero[ETH_ALEN] = { 0 };
+		conf.bssid = zero;
+		conf.ssid = zero;
+		conf.ssid_len = 0;
+	} else {
+		WARN_ON(1);
+		return -EINVAL;
 	}
-	return local->ops->config_interface(local_to_hw(local),
-					    &sdata->vif, &conf);
-}
 
-int ieee80211_if_config(struct net_device *dev)
-{
-	struct ieee80211_sub_if_data *sdata = IEEE80211_DEV_TO_SUB_IF(dev);
-	struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr);
-	if (sdata->vif.type == IEEE80211_IF_TYPE_MESH_POINT &&
-	    (local->hw.flags & IEEE80211_HW_HOST_GEN_BEACON_TEMPLATE))
-		return ieee80211_if_config_beacon(dev);
-	return __ieee80211_if_config(dev, NULL);
-}
+	if (WARN_ON(!conf.bssid && (changed & IEEE80211_IFCC_BSSID)))
+		return -EINVAL;
 
-int ieee80211_if_config_beacon(struct net_device *dev)
-{
-	struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr);
-	struct ieee80211_sub_if_data *sdata = IEEE80211_DEV_TO_SUB_IF(dev);
-	struct sk_buff *skb;
+	if (WARN_ON(!conf.ssid && (changed & IEEE80211_IFCC_SSID)))
+		return -EINVAL;
 
-	if (!(local->hw.flags & IEEE80211_HW_HOST_GEN_BEACON_TEMPLATE))
-		return 0;
-	skb = ieee80211_beacon_get(local_to_hw(local), &sdata->vif);
-	if (!skb)
-		return -ENOMEM;
-	return __ieee80211_if_config(dev, skb);
+	return local->ops->config_interface(local_to_hw(local),
+					    &sdata->vif, &conf);
 }
 
 int ieee80211_hw_config(struct ieee80211_local *local)
diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c
index 64d710a88b86..61d7f81bf45e 100644
--- a/net/mac80211/mlme.c
+++ b/net/mac80211/mlme.c
@@ -2406,8 +2406,6 @@ static int ieee80211_sta_join_ibss(struct net_device *dev,
 	int res, rates, i, j;
 	struct sk_buff *skb;
 	struct ieee80211_mgmt *mgmt;
-	struct ieee80211_tx_info *control;
-	struct rate_selection ratesel;
 	u8 *pos;
 	struct ieee80211_sub_if_data *sdata;
 	struct ieee80211_supported_band *sband;
@@ -2425,7 +2423,7 @@ static int ieee80211_sta_join_ibss(struct net_device *dev,
 		local->ops->reset_tsf(local_to_hw(local));
 	}
 	memcpy(ifsta->bssid, bss->bssid, ETH_ALEN);
-	res = ieee80211_if_config(dev);
+	res = ieee80211_if_config(sdata, IEEE80211_IFCC_BSSID);
 	if (res)
 		return res;
 
@@ -2439,19 +2437,16 @@ static int ieee80211_sta_join_ibss(struct net_device *dev,
 	if (res)
 		return res;
 
-	/* Set beacon template */
+	/* Build IBSS probe response */
 	skb = dev_alloc_skb(local->hw.extra_tx_headroom + 400);
-	do {
-		if (!skb)
-			break;
-
+	if (skb) {
 		skb_reserve(skb, local->hw.extra_tx_headroom);
 
 		mgmt = (struct ieee80211_mgmt *)
 			skb_put(skb, 24 + sizeof(mgmt->u.beacon));
 		memset(mgmt, 0, 24 + sizeof(mgmt->u.beacon));
 		mgmt->frame_control = IEEE80211_FC(IEEE80211_FTYPE_MGMT,
-						   IEEE80211_STYPE_BEACON);
+						   IEEE80211_STYPE_PROBE_RESP);
 		memset(mgmt->da, 0xff, ETH_ALEN);
 		memcpy(mgmt->sa, dev->dev_addr, ETH_ALEN);
 		memcpy(mgmt->bssid, ifsta->bssid, ETH_ALEN);
@@ -2495,61 +2490,22 @@ static int ieee80211_sta_join_ibss(struct net_device *dev,
 			memcpy(pos, &bss->supp_rates[8], rates);
 		}
 
-		control = IEEE80211_SKB_CB(skb);
-
-		rate_control_get_rate(dev, sband, skb, &ratesel);
-		if (ratesel.rate_idx < 0) {
-			printk(KERN_DEBUG "%s: Failed to determine TX rate "
-			       "for IBSS beacon\n", dev->name);
-			break;
-		}
-		control->control.vif = &sdata->vif;
-		control->tx_rate_idx = ratesel.rate_idx;
-		if (sdata->bss_conf.use_short_preamble &&
-		    sband->bitrates[ratesel.rate_idx].flags & IEEE80211_RATE_SHORT_PREAMBLE)
-			control->flags |= IEEE80211_TX_CTL_SHORT_PREAMBLE;
-		control->antenna_sel_tx = local->hw.conf.antenna_sel_tx;
-		control->flags |= IEEE80211_TX_CTL_NO_ACK;
-		control->flags |= IEEE80211_TX_CTL_DO_NOT_ENCRYPT;
-		control->control.retry_limit = 1;
-
-		ifsta->probe_resp = skb_copy(skb, GFP_ATOMIC);
-		if (ifsta->probe_resp) {
-			mgmt = (struct ieee80211_mgmt *)
-				ifsta->probe_resp->data;
-			mgmt->frame_control =
-				IEEE80211_FC(IEEE80211_FTYPE_MGMT,
-					     IEEE80211_STYPE_PROBE_RESP);
-		} else {
-			printk(KERN_DEBUG "%s: Could not allocate ProbeResp "
-			       "template for IBSS\n", dev->name);
-		}
-
-		if (local->ops->beacon_update &&
-		    local->ops->beacon_update(local_to_hw(local), skb) == 0) {
-			printk(KERN_DEBUG "%s: Configured IBSS beacon "
-			       "template\n", dev->name);
-			skb = NULL;
-		}
-
-		rates = 0;
-		sband = local->hw.wiphy->bands[local->hw.conf.channel->band];
-		for (i = 0; i < bss->supp_rates_len; i++) {
-			int bitrate = (bss->supp_rates[i] & 0x7f) * 5;
-			for (j = 0; j < sband->n_bitrates; j++)
-				if (sband->bitrates[j].bitrate == bitrate)
-					rates |= BIT(j);
-		}
-		ifsta->supp_rates_bits[local->hw.conf.channel->band] = rates;
+		ifsta->probe_resp = skb;
 
-		ieee80211_sta_def_wmm_params(dev, bss, 1);
-	} while (0);
+		ieee80211_if_config(sdata, IEEE80211_IFCC_BEACON);
+	}
 
-	if (skb) {
-		printk(KERN_DEBUG "%s: Failed to configure IBSS beacon "
-		       "template\n", dev->name);
-		dev_kfree_skb(skb);
+	rates = 0;
+	sband = local->hw.wiphy->bands[local->hw.conf.channel->band];
+	for (i = 0; i < bss->supp_rates_len; i++) {
+		int bitrate = (bss->supp_rates[i] & 0x7f) * 5;
+		for (j = 0; j < sband->n_bitrates; j++)
+			if (sband->bitrates[j].bitrate == bitrate)
+				rates |= BIT(j);
 	}
+	ifsta->supp_rates_bits[local->hw.conf.channel->band] = rates;
+
+	ieee80211_sta_def_wmm_params(dev, bss, 1);
 
 	ifsta->state = IEEE80211_IBSS_JOINED;
 	mod_timer(&ifsta->timer, jiffies + IEEE80211_IBSS_MERGE_INTERVAL);
@@ -3333,7 +3289,7 @@ static void ieee80211_mesh_housekeeping(struct net_device *dev,
 
 	free_plinks = mesh_plink_availables(sdata);
 	if (free_plinks != sdata->u.sta.accepting_plinks)
-		ieee80211_if_config_beacon(dev);
+		ieee80211_if_config(sdata, IEEE80211_IFCC_BEACON);
 
 	mod_timer(&ifsta->timer, jiffies +
 			IEEE80211_MESH_HOUSEKEEPING_INTERVAL);
@@ -3757,28 +3713,45 @@ int ieee80211_sta_set_ssid(struct net_device *dev, char *ssid, size_t len)
 {
 	struct ieee80211_sub_if_data *sdata = IEEE80211_DEV_TO_SUB_IF(dev);
 	struct ieee80211_if_sta *ifsta;
+	int res;
 
 	if (len > IEEE80211_MAX_SSID_LEN)
 		return -EINVAL;
 
 	ifsta = &sdata->u.sta;
 
-	if (ifsta->ssid_len != len || memcmp(ifsta->ssid, ssid, len) != 0)
+	if (ifsta->ssid_len != len || memcmp(ifsta->ssid, ssid, len) != 0) {
+		memset(ifsta->ssid, 0, sizeof(ifsta->ssid));
+		memcpy(ifsta->ssid, ssid, len);
+		ifsta->ssid_len = len;
 		ifsta->flags &= ~IEEE80211_STA_PREV_BSSID_SET;
-	memcpy(ifsta->ssid, ssid, len);
-	memset(ifsta->ssid + len, 0, IEEE80211_MAX_SSID_LEN - len);
-	ifsta->ssid_len = len;
+
+		res = 0;
+		/*
+		 * Hack! MLME code needs to be cleaned up to have different
+		 * entry points for configuration and internal selection change
+		 */
+		if (netif_running(sdata->dev))
+			res = ieee80211_if_config(sdata, IEEE80211_IFCC_SSID);
+		if (res) {
+			printk(KERN_DEBUG "%s: Failed to config new SSID to "
+			       "the low-level driver\n", dev->name);
+			return res;
+		}
+	}
 
 	if (len)
 		ifsta->flags |= IEEE80211_STA_SSID_SET;
 	else
 		ifsta->flags &= ~IEEE80211_STA_SSID_SET;
+
 	if (sdata->vif.type == IEEE80211_IF_TYPE_IBSS &&
 	    !(ifsta->flags & IEEE80211_STA_BSSID_SET)) {
 		ifsta->ibss_join_req = jiffies;
 		ifsta->state = IEEE80211_IBSS_SEARCH;
 		return ieee80211_sta_find_ibss(dev, ifsta);
 	}
+
 	return 0;
 }
 
@@ -3804,7 +3777,12 @@ int ieee80211_sta_set_bssid(struct net_device *dev, u8 *bssid)
 
 	if (memcmp(ifsta->bssid, bssid, ETH_ALEN) != 0) {
 		memcpy(ifsta->bssid, bssid, ETH_ALEN);
-		res = ieee80211_if_config(dev);
+		res = 0;
+		/*
+		 * Hack! See also ieee80211_sta_set_ssid.
+		 */
+		if (netif_running(sdata->dev))
+			res = ieee80211_if_config(sdata, IEEE80211_IFCC_BSSID);
 		if (res) {
 			printk(KERN_DEBUG "%s: Failed to config new BSSID to "
 			       "the low-level driver\n", dev->name);
diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c
index a757dcc1208d..8843416e1460 100644
--- a/net/mac80211/tx.c
+++ b/net/mac80211/tx.c
@@ -1788,17 +1788,17 @@ struct sk_buff *ieee80211_beacon_get(struct ieee80211_hw *hw,
 				     struct ieee80211_vif *vif)
 {
 	struct ieee80211_local *local = hw_to_local(hw);
-	struct sk_buff *skb;
+	struct sk_buff *skb = NULL;
 	struct ieee80211_tx_info *info;
 	struct net_device *bdev;
 	struct ieee80211_sub_if_data *sdata = NULL;
 	struct ieee80211_if_ap *ap = NULL;
+	struct ieee80211_if_sta *ifsta = NULL;
 	struct rate_selection rsel;
 	struct beacon_data *beacon;
 	struct ieee80211_supported_band *sband;
 	struct ieee80211_mgmt *mgmt;
 	int *num_beacons;
-	bool err = true;
 	enum ieee80211_band band = local->hw.conf.channel->band;
 	u8 *pos;
 
@@ -1852,9 +1852,24 @@ struct sk_buff *ieee80211_beacon_get(struct ieee80211_hw *hw,
 				       beacon->tail, beacon->tail_len);
 
 			num_beacons = &ap->num_beacons;
+		} else
+			goto out;
+	} else if (sdata->vif.type == IEEE80211_IF_TYPE_IBSS) {
+		struct ieee80211_hdr *hdr;
+		ifsta = &sdata->u.sta;
 
-			err = false;
-		}
+		if (!ifsta->probe_resp)
+			goto out;
+
+		skb = skb_copy(ifsta->probe_resp, GFP_ATOMIC);
+		if (!skb)
+			goto out;
+
+		hdr = (struct ieee80211_hdr *) skb->data;
+		hdr->frame_control = IEEE80211_FC(IEEE80211_FTYPE_MGMT,
+						  IEEE80211_STYPE_BEACON);
+
+		num_beacons = &ifsta->num_beacons;
 	} else if (ieee80211_vif_is_mesh(&sdata->vif)) {
 		/* headroom, head length, tail length and maximum TIM length */
 		skb = dev_alloc_skb(local->tx_headroom + 400);
@@ -1881,17 +1896,8 @@ struct sk_buff *ieee80211_beacon_get(struct ieee80211_hw *hw,
 		mesh_mgmt_ies_add(skb, sdata->dev);
 
 		num_beacons = &sdata->u.sta.num_beacons;
-
-		err = false;
-	}
-
-	if (err) {
-#ifdef CONFIG_MAC80211_VERBOSE_DEBUG
-		if (net_ratelimit())
-			printk(KERN_DEBUG "no beacon data avail for %s\n",
-			       bdev->name);
-#endif /* CONFIG_MAC80211_VERBOSE_DEBUG */
-		skb = NULL;
+	} else {
+		WARN_ON(1);
 		goto out;
 	}
 
diff --git a/net/mac80211/wext.c b/net/mac80211/wext.c
index c041db9556c7..34fa8ed1e784 100644
--- a/net/mac80211/wext.c
+++ b/net/mac80211/wext.c
@@ -444,7 +444,7 @@ static int ieee80211_ioctl_siwessid(struct net_device *dev,
 		memset(sdata->u.ap.ssid + len, 0,
 		       IEEE80211_MAX_SSID_LEN - len);
 		sdata->u.ap.ssid_len = len;
-		return ieee80211_if_config(dev);
+		return ieee80211_if_config(sdata, IEEE80211_IFCC_SSID);
 	}
 	return -EOPNOTSUPP;
 }
-- 
cgit v1.2.3


From 9c0c7a429a0cf02c2ac1998d5cf4c26f6be5c989 Mon Sep 17 00:00:00 2001
From: Michael Buesch <mb@bu3sch.de>
Date: Wed, 9 Jul 2008 15:48:45 +0200
Subject: ssb: Include dma-mapping.h

ssb.h implements DMA mapping functions, so it should
include dma-mapping.h. This fixes compile failures on certain architectures.

Reported-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Michael Buesch <mb@bu3sch.de>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/linux/ssb/ssb.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include')

diff --git a/include/linux/ssb/ssb.h b/include/linux/ssb/ssb.h
index 0fe5a0ded3ea..4bf8cade9dbc 100644
--- a/include/linux/ssb/ssb.h
+++ b/include/linux/ssb/ssb.h
@@ -7,6 +7,7 @@
 #include <linux/spinlock.h>
 #include <linux/pci.h>
 #include <linux/mod_devicetable.h>
+#include <linux/dma-mapping.h>
 
 #include <linux/ssb/ssb_regs.h>
 
-- 
cgit v1.2.3


From f591fa5dbbbeaebd95c9c019b3a536a327fb79de Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes@sipsolutions.net>
Date: Thu, 10 Jul 2008 11:21:26 +0200
Subject: mac80211: fix TX sequence numbers

This patch makes mac80211 assign proper sequence numbers to
QoS-data frames. It also removes the old sequence number code
because we noticed that only the driver or hardware can assign
sequence numbers to non-QoS-data and especially management
frames in a race-free manner because beacons aren't passed
through mac80211's TX path.

This patch also adds temporary code to the rt2x00 drivers to
not break them completely, that code will have to be reworked
for proper sequence numbers on beacons.

It also moves sequence number assignment down in the TX path
so no sequence numbers are assigned to frames that are dropped.

Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/net/wireless/b43/xmit.c         |  3 +-
 drivers/net/wireless/b43legacy/xmit.c   |  3 +-
 drivers/net/wireless/rt2x00/rt2x00.h    |  2 +
 drivers/net/wireless/rt2x00/rt2x00mac.c | 13 ++++++
 include/net/mac80211.h                  | 12 +++++
 net/mac80211/ieee80211_i.h              |  2 -
 net/mac80211/iface.c                    |  1 -
 net/mac80211/sta_info.h                 |  1 +
 net/mac80211/tx.c                       | 79 +++++++++++++++++++++------------
 9 files changed, 82 insertions(+), 34 deletions(-)

(limited to 'include')

diff --git a/drivers/net/wireless/b43/xmit.c b/drivers/net/wireless/b43/xmit.c
index bf6f6c1ed4cf..8d54502222a6 100644
--- a/drivers/net/wireless/b43/xmit.c
+++ b/drivers/net/wireless/b43/xmit.c
@@ -317,7 +317,8 @@ int b43_generate_txhdr(struct b43_wldev *dev,
 	/* MAC control */
 	if (!(info->flags & IEEE80211_TX_CTL_NO_ACK))
 		mac_ctl |= B43_TXH_MAC_ACK;
-	if (!ieee80211_is_pspoll(fctl))
+	/* use hardware sequence counter as the non-TID counter */
+	if (info->flags & IEEE80211_TX_CTL_ASSIGN_SEQ)
 		mac_ctl |= B43_TXH_MAC_HWSEQ;
 	if (info->flags & IEEE80211_TX_CTL_FIRST_FRAGMENT)
 		mac_ctl |= B43_TXH_MAC_STMSDU;
diff --git a/drivers/net/wireless/b43legacy/xmit.c b/drivers/net/wireless/b43legacy/xmit.c
index a3540787eb50..e969ed8d412d 100644
--- a/drivers/net/wireless/b43legacy/xmit.c
+++ b/drivers/net/wireless/b43legacy/xmit.c
@@ -295,8 +295,7 @@ static int generate_txhdr_fw3(struct b43legacy_wldev *dev,
 	/* MAC control */
 	if (!(info->flags & IEEE80211_TX_CTL_NO_ACK))
 		mac_ctl |= B43legacy_TX4_MAC_ACK;
-	if (!(((fctl & IEEE80211_FCTL_FTYPE) == IEEE80211_FTYPE_CTL) &&
-	      ((fctl & IEEE80211_FCTL_STYPE) == IEEE80211_STYPE_PSPOLL)))
+	if (info->flags & IEEE80211_TX_CTL_ASSIGN_SEQ)
 		mac_ctl |= B43legacy_TX4_MAC_HWSEQ;
 	if (info->flags & IEEE80211_TX_CTL_FIRST_FRAGMENT)
 		mac_ctl |= B43legacy_TX4_MAC_STMSDU;
diff --git a/drivers/net/wireless/rt2x00/rt2x00.h b/drivers/net/wireless/rt2x00/rt2x00.h
index f0d7e083d8f8..9fab0df18c3c 100644
--- a/drivers/net/wireless/rt2x00/rt2x00.h
+++ b/drivers/net/wireless/rt2x00/rt2x00.h
@@ -364,6 +364,8 @@ struct rt2x00_intf {
 #define DELAYED_UPDATE_BEACON		0x00000001
 #define DELAYED_CONFIG_ERP		0x00000002
 #define DELAYED_LED_ASSOC		0x00000004
+
+	u16 seqno;
 };
 
 static inline struct rt2x00_intf* vif_to_intf(struct ieee80211_vif *vif)
diff --git a/drivers/net/wireless/rt2x00/rt2x00mac.c b/drivers/net/wireless/rt2x00/rt2x00mac.c
index 16b72d9ca1c3..77af1df5d899 100644
--- a/drivers/net/wireless/rt2x00/rt2x00mac.c
+++ b/drivers/net/wireless/rt2x00/rt2x00mac.c
@@ -96,6 +96,7 @@ int rt2x00mac_tx(struct ieee80211_hw *hw, struct sk_buff *skb)
 	struct ieee80211_tx_info *tx_info = IEEE80211_SKB_CB(skb);
 	struct ieee80211_hdr *ieee80211hdr = (struct ieee80211_hdr *)skb->data;
 	enum data_queue_qid qid = skb_get_queue_mapping(skb);
+	struct rt2x00_intf *intf = vif_to_intf(tx_info->control.vif);
 	struct data_queue *queue;
 	u16 frame_control;
 
@@ -151,6 +152,18 @@ int rt2x00mac_tx(struct ieee80211_hw *hw, struct sk_buff *skb)
 		}
 	}
 
+	/*
+	 * XXX: This is as wrong as the old mac80211 code was,
+	 *	due to beacons not getting sequence numbers assigned
+	 *	properly.
+	 */
+	if (tx_info->flags & IEEE80211_TX_CTL_ASSIGN_SEQ) {
+		if (tx_info->flags & IEEE80211_TX_CTL_FIRST_FRAGMENT)
+			intf->seqno += 0x10;
+		ieee80211hdr->seq_ctrl &= cpu_to_le16(IEEE80211_SCTL_FRAG);
+		ieee80211hdr->seq_ctrl |= cpu_to_le16(intf->seqno);
+	}
+
 	if (rt2x00queue_write_tx_frame(queue, skb)) {
 		ieee80211_stop_queue(rt2x00dev->hw, qid);
 		return NETDEV_TX_BUSY;
diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index 47c072eab42c..1dbd49fc557e 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -239,6 +239,17 @@ struct ieee80211_bss_conf {
  * 	is for the whole aggregation.
  * @IEEE80211_TX_STAT_AMPDU_NO_BACK: no block ack was returned,
  * 	so consider using block ack request (BAR).
+ * @IEEE80211_TX_CTL_ASSIGN_SEQ: The driver has to assign a sequence
+ *	number to this frame, taking care of not overwriting the fragment
+ *	number and increasing the sequence number only when the
+ *	IEEE80211_TX_CTL_FIRST_FRAGMENT flags is set. mac80211 will properly
+ *	assign sequence numbers to QoS-data frames but cannot do so correctly
+ *	for non-QoS-data and management frames because beacons need them from
+ *	that counter as well and mac80211 cannot guarantee proper sequencing.
+ *	If this flag is set, the driver should instruct the hardware to
+ *	assign a sequence number to the frame or assign one itself. Cf. IEEE
+ *	802.11-2007 7.1.3.4.1 paragraph 3. This flag will always be set for
+ *	beacons always be clear for frames without a sequence number field.
  */
 enum mac80211_tx_control_flags {
 	IEEE80211_TX_CTL_REQ_TX_STATUS		= BIT(0),
@@ -265,6 +276,7 @@ enum mac80211_tx_control_flags {
 	IEEE80211_TX_STAT_ACK			= BIT(21),
 	IEEE80211_TX_STAT_AMPDU			= BIT(22),
 	IEEE80211_TX_STAT_AMPDU_NO_BACK		= BIT(23),
+	IEEE80211_TX_CTL_ASSIGN_SEQ		= BIT(24),
 };
 
 
diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h
index 934c3ef4f0bc..cbea0154ee3a 100644
--- a/net/mac80211/ieee80211_i.h
+++ b/net/mac80211/ieee80211_i.h
@@ -419,8 +419,6 @@ struct ieee80211_sub_if_data {
 	 */
 	u64 basic_rates;
 
-	u16 sequence;
-
 	/* Fragment table for host-based reassembly */
 	struct ieee80211_fragment_entry	fragments[IEEE80211_FRAGMENT_MAX];
 	unsigned int fragment_next;
diff --git a/net/mac80211/iface.c b/net/mac80211/iface.c
index 2e3adcb3ce23..610ed1d9893a 100644
--- a/net/mac80211/iface.c
+++ b/net/mac80211/iface.c
@@ -162,7 +162,6 @@ int ieee80211_if_change_type(struct ieee80211_sub_if_data *sdata,
 	/* reset some values that shouldn't be kept across type changes */
 	sdata->basic_rates = 0;
 	sdata->drop_unencrypted = 0;
-	sdata->sequence = 0;
 
 	return 0;
 }
diff --git a/net/mac80211/sta_info.h b/net/mac80211/sta_info.h
index 94311dcfe043..109db787ccb7 100644
--- a/net/mac80211/sta_info.h
+++ b/net/mac80211/sta_info.h
@@ -285,6 +285,7 @@ struct sta_info {
 	unsigned long tx_fragments;
 	int txrate_idx;
 	int last_txrate_idx;
+	u16 tid_seq[IEEE80211_QOS_CTL_TID_MASK + 1];
 #ifdef CONFIG_MAC80211_DEBUG_COUNTERS
 	unsigned int wme_tx_queue[NUM_RX_DATA_QUEUES];
 #endif
diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c
index 8843416e1460..0fbadd8b983c 100644
--- a/net/mac80211/tx.c
+++ b/net/mac80211/tx.c
@@ -38,16 +38,6 @@
 
 /* misc utils */
 
-static inline void ieee80211_include_sequence(struct ieee80211_sub_if_data *sdata,
-					      struct ieee80211_hdr *hdr)
-{
-	/* Set the sequence number for this frame. */
-	hdr->seq_ctrl = cpu_to_le16(sdata->sequence);
-
-	/* Increase the sequence number. */
-	sdata->sequence = (sdata->sequence + 0x10) & IEEE80211_SCTL_SEQ;
-}
-
 #ifdef CONFIG_MAC80211_LOWTX_FRAME_DUMP
 static void ieee80211_dump_frame(const char *ifname, const char *title,
 				 const struct sk_buff *skb)
@@ -274,17 +264,6 @@ ieee80211_tx_h_check_assoc(struct ieee80211_tx_data *tx)
 	return TX_CONTINUE;
 }
 
-static ieee80211_tx_result debug_noinline
-ieee80211_tx_h_sequence(struct ieee80211_tx_data *tx)
-{
-	struct ieee80211_hdr *hdr = (struct ieee80211_hdr *)tx->skb->data;
-
-	if (ieee80211_hdrlen(hdr->frame_control) >= 24)
-		ieee80211_include_sequence(tx->sdata, hdr);
-
-	return TX_CONTINUE;
-}
-
 /* This function is called whenever the AP is about to exceed the maximum limit
  * of buffered frames for power saving STAs. This situation should not really
  * happen often during normal operation, so dropping the oldest buffered packet
@@ -641,6 +620,49 @@ ieee80211_tx_h_misc(struct ieee80211_tx_data *tx)
 	return TX_CONTINUE;
 }
 
+static ieee80211_tx_result debug_noinline
+ieee80211_tx_h_sequence(struct ieee80211_tx_data *tx)
+{
+	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(tx->skb);
+	struct ieee80211_hdr *hdr = (struct ieee80211_hdr *)tx->skb->data;
+	u16 *seq;
+	u8 *qc;
+	int tid;
+
+	/* only for injected frames */
+	if (unlikely(ieee80211_is_ctl(hdr->frame_control)))
+		return TX_CONTINUE;
+
+	if (ieee80211_hdrlen(hdr->frame_control) < 24)
+		return TX_CONTINUE;
+
+	if (!ieee80211_is_data_qos(hdr->frame_control)) {
+		info->flags |= IEEE80211_TX_CTL_ASSIGN_SEQ;
+		return TX_CONTINUE;
+	}
+
+	/*
+	 * This should be true for injected/management frames only, for
+	 * management frames we have set the IEEE80211_TX_CTL_ASSIGN_SEQ
+	 * above since they are not QoS-data frames.
+	 */
+	if (!tx->sta)
+		return TX_CONTINUE;
+
+	/* include per-STA, per-TID sequence counter */
+
+	qc = ieee80211_get_qos_ctl(hdr);
+	tid = *qc & IEEE80211_QOS_CTL_TID_MASK;
+	seq = &tx->sta->tid_seq[tid];
+
+	hdr->seq_ctrl = cpu_to_le16(*seq);
+
+	/* Increase the sequence number. */
+	*seq = (*seq + 0x10) & IEEE80211_SCTL_SEQ;
+
+	return TX_CONTINUE;
+}
+
 static ieee80211_tx_result debug_noinline
 ieee80211_tx_h_fragment(struct ieee80211_tx_data *tx)
 {
@@ -1110,12 +1132,12 @@ static int invoke_tx_handlers(struct ieee80211_tx_data *tx)
 		goto txh_done;
 
 	CALL_TXH(ieee80211_tx_h_check_assoc)
-	CALL_TXH(ieee80211_tx_h_sequence)
 	CALL_TXH(ieee80211_tx_h_ps_buf)
 	CALL_TXH(ieee80211_tx_h_select_key)
 	CALL_TXH(ieee80211_tx_h_michael_mic_add)
 	CALL_TXH(ieee80211_tx_h_rate_ctrl)
 	CALL_TXH(ieee80211_tx_h_misc)
+	CALL_TXH(ieee80211_tx_h_sequence)
 	CALL_TXH(ieee80211_tx_h_fragment)
 	/* handlers after fragment must be aware of tx info fragmentation! */
 	CALL_TXH(ieee80211_tx_h_encrypt)
@@ -1827,9 +1849,6 @@ struct sk_buff *ieee80211_beacon_get(struct ieee80211_hw *hw,
 			memcpy(skb_put(skb, beacon->head_len), beacon->head,
 			       beacon->head_len);
 
-			ieee80211_include_sequence(sdata,
-					(struct ieee80211_hdr *)skb->data);
-
 			/*
 			 * Not very nice, but we want to allow the driver to call
 			 * ieee80211_beacon_get() as a response to the set_tim()
@@ -1919,14 +1938,18 @@ struct sk_buff *ieee80211_beacon_get(struct ieee80211_hw *hw,
 
 	info->control.vif = vif;
 	info->tx_rate_idx = rsel.rate_idx;
+
+	info->flags |= IEEE80211_TX_CTL_NO_ACK;
+	info->flags |= IEEE80211_TX_CTL_DO_NOT_ENCRYPT;
+	info->flags |= IEEE80211_TX_CTL_CLEAR_PS_FILT;
+	info->flags |= IEEE80211_TX_CTL_ASSIGN_SEQ;
 	if (sdata->bss_conf.use_short_preamble &&
 	    sband->bitrates[rsel.rate_idx].flags & IEEE80211_RATE_SHORT_PREAMBLE)
 		info->flags |= IEEE80211_TX_CTL_SHORT_PREAMBLE;
+
 	info->antenna_sel_tx = local->hw.conf.antenna_sel_tx;
-	info->flags |= IEEE80211_TX_CTL_NO_ACK;
-	info->flags |= IEEE80211_TX_CTL_DO_NOT_ENCRYPT;
 	info->control.retry_limit = 1;
-	info->flags |= IEEE80211_TX_CTL_CLEAR_PS_FILT;
+
 	(*num_beacons)++;
 out:
 	rcu_read_unlock();
-- 
cgit v1.2.3


From f434b2d111d9ff84ebdd0f11a7ae42c761453259 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes@sipsolutions.net>
Date: Thu, 10 Jul 2008 11:22:31 +0200
Subject: mac80211: fix struct ieee80211_tx_queue_params

Multiple issues:
 - there are no "default" values needed
 - cw_min/cw_max can be larger than documented
 - restructure to decrease size
 - use get_unaligned_le16

Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/net/mac80211.h | 10 +++++-----
 net/mac80211/mlme.c    |  2 +-
 2 files changed, 6 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index 1dbd49fc557e..24a69f6075c2 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -115,17 +115,17 @@ enum ieee80211_max_queues {
  * The information provided in this structure is required for QoS
  * transmit queue configuration. Cf. IEEE 802.11 7.3.2.29.
  *
- * @aifs: arbitration interface space [0..255, -1: use default]
- * @cw_min: minimum contention window [will be a value of the form
- *	2^n-1 in the range 1..1023; 0: use default]
+ * @aifs: arbitration interface space [0..255]
+ * @cw_min: minimum contention window [a value of the form
+ *	2^n-1 in the range 1..32767]
  * @cw_max: maximum contention window [like @cw_min]
  * @txop: maximum burst time in units of 32 usecs, 0 meaning disabled
  */
 struct ieee80211_tx_queue_params {
-	s16 aifs;
+	u16 txop;
 	u16 cw_min;
 	u16 cw_max;
-	u16 txop;
+	u8 aifs;
 };
 
 /**
diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c
index 61d7f81bf45e..a4bbc8d6d0e6 100644
--- a/net/mac80211/mlme.c
+++ b/net/mac80211/mlme.c
@@ -345,7 +345,7 @@ static void ieee80211_sta_wmm_params(struct net_device *dev,
 		params.aifs = pos[0] & 0x0f;
 		params.cw_max = ecw2cw((pos[1] & 0xf0) >> 4);
 		params.cw_min = ecw2cw(pos[1] & 0x0f);
-		params.txop = pos[2] | (pos[3] << 8);
+		params.txop = get_unaligned_le16(pos + 2);
 #ifdef CONFIG_MAC80211_VERBOSE_DEBUG
 		printk(KERN_DEBUG "%s: WMM queue=%d aci=%d acm=%d aifs=%d "
 		       "cWmin=%d cWmax=%d txop=%d\n",
-- 
cgit v1.2.3


From 4c8894980010536915c4f5513ee180e3614aeca9 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Mon, 14 Jul 2008 20:22:38 -0700
Subject: netfilter: Let nf_ct_kill() callers know if del_timer() returned
 true.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/netfilter/nf_conntrack.h | 20 ++++++++++----------
 net/netfilter/nf_conntrack_core.c    | 13 ++++++++-----
 2 files changed, 18 insertions(+), 15 deletions(-)

(limited to 'include')

diff --git a/include/net/netfilter/nf_conntrack.h b/include/net/netfilter/nf_conntrack.h
index d5d76ec7abb0..8f5b75734dd0 100644
--- a/include/net/netfilter/nf_conntrack.h
+++ b/include/net/netfilter/nf_conntrack.h
@@ -223,23 +223,23 @@ static inline void nf_ct_refresh(struct nf_conn *ct,
 	__nf_ct_refresh_acct(ct, 0, skb, extra_jiffies, 0);
 }
 
-extern void __nf_ct_kill_acct(struct nf_conn *ct,
-				enum ip_conntrack_info ctinfo,
-				const struct sk_buff *skb,
-				int do_acct);
+extern bool __nf_ct_kill_acct(struct nf_conn *ct,
+			      enum ip_conntrack_info ctinfo,
+			      const struct sk_buff *skb,
+			      int do_acct);
 
 /* kill conntrack and do accounting */
-static inline void nf_ct_kill_acct(struct nf_conn *ct,
-				enum ip_conntrack_info ctinfo,
-				const struct sk_buff *skb)
+static inline bool nf_ct_kill_acct(struct nf_conn *ct,
+				   enum ip_conntrack_info ctinfo,
+				   const struct sk_buff *skb)
 {
-	__nf_ct_kill_acct(ct, ctinfo, skb, 1);
+	return __nf_ct_kill_acct(ct, ctinfo, skb, 1);
 }
 
 /* kill conntrack without accounting */
-static inline void nf_ct_kill(struct nf_conn *ct)
+static inline bool nf_ct_kill(struct nf_conn *ct)
 {
-	__nf_ct_kill_acct(ct, 0, NULL, 0);
+	return __nf_ct_kill_acct(ct, 0, NULL, 0);
 }
 
 /* These are for NAT.  Icky. */
diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c
index 212a0888408d..28d03e64200b 100644
--- a/net/netfilter/nf_conntrack_core.c
+++ b/net/netfilter/nf_conntrack_core.c
@@ -848,10 +848,10 @@ acct:
 }
 EXPORT_SYMBOL_GPL(__nf_ct_refresh_acct);
 
-void __nf_ct_kill_acct(struct nf_conn *ct,
-		enum ip_conntrack_info ctinfo,
-		const struct sk_buff *skb,
-		int do_acct)
+bool __nf_ct_kill_acct(struct nf_conn *ct,
+		       enum ip_conntrack_info ctinfo,
+		       const struct sk_buff *skb,
+		       int do_acct)
 {
 #ifdef CONFIG_NF_CT_ACCT
 	if (do_acct) {
@@ -862,8 +862,11 @@ void __nf_ct_kill_acct(struct nf_conn *ct,
 		spin_unlock_bh(&nf_conntrack_lock);
 	}
 #endif
-	if (del_timer(&ct->timeout))
+	if (del_timer(&ct->timeout)) {
 		ct->timeout.function((unsigned long)ct);
+		return true;
+	}
+	return false;
 }
 EXPORT_SYMBOL_GPL(__nf_ct_kill_acct);
 
-- 
cgit v1.2.3


From 72d9794f444734af56ef12833b496326643e2964 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Mon, 14 Jul 2008 20:36:32 -0700
Subject: net-sched: cls_flow: add perturbation support

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/pkt_cls.h |  1 +
 net/sched/cls_flow.c    | 52 ++++++++++++++++++++++++++++++++++++++++---------
 2 files changed, 44 insertions(+), 9 deletions(-)

(limited to 'include')

diff --git a/include/linux/pkt_cls.h b/include/linux/pkt_cls.h
index 99efbed81fa2..7cf7824df778 100644
--- a/include/linux/pkt_cls.h
+++ b/include/linux/pkt_cls.h
@@ -374,6 +374,7 @@ enum
 	TCA_FLOW_ACT,
 	TCA_FLOW_POLICE,
 	TCA_FLOW_EMATCHES,
+	TCA_FLOW_PERTURB,
 	__TCA_FLOW_MAX
 };
 
diff --git a/net/sched/cls_flow.c b/net/sched/cls_flow.c
index 971b867e0484..8f63a1a94014 100644
--- a/net/sched/cls_flow.c
+++ b/net/sched/cls_flow.c
@@ -36,6 +36,8 @@ struct flow_filter {
 	struct list_head	list;
 	struct tcf_exts		exts;
 	struct tcf_ematch_tree	ematches;
+	struct timer_list	perturb_timer;
+	u32			perturb_period;
 	u32			handle;
 
 	u32			nkeys;
@@ -47,11 +49,9 @@ struct flow_filter {
 	u32			addend;
 	u32			divisor;
 	u32			baseclass;
+	u32			hashrnd;
 };
 
-static u32 flow_hashrnd __read_mostly;
-static int flow_hashrnd_initted __read_mostly;
-
 static const struct tcf_ext_map flow_ext_map = {
 	.action	= TCA_FLOW_ACT,
 	.police	= TCA_FLOW_POLICE,
@@ -348,7 +348,7 @@ static int flow_classify(struct sk_buff *skb, struct tcf_proto *tp,
 		}
 
 		if (f->mode == FLOW_MODE_HASH)
-			classid = jhash2(keys, f->nkeys, flow_hashrnd);
+			classid = jhash2(keys, f->nkeys, f->hashrnd);
 		else {
 			classid = keys[0];
 			classid = (classid & f->mask) ^ f->xor;
@@ -369,6 +369,15 @@ static int flow_classify(struct sk_buff *skb, struct tcf_proto *tp,
 	return -1;
 }
 
+static void flow_perturbation(unsigned long arg)
+{
+	struct flow_filter *f = (struct flow_filter *)arg;
+
+	get_random_bytes(&f->hashrnd, 4);
+	if (f->perturb_period)
+		mod_timer(&f->perturb_timer, jiffies + f->perturb_period);
+}
+
 static const struct nla_policy flow_policy[TCA_FLOW_MAX + 1] = {
 	[TCA_FLOW_KEYS]		= { .type = NLA_U32 },
 	[TCA_FLOW_MODE]		= { .type = NLA_U32 },
@@ -381,6 +390,7 @@ static const struct nla_policy flow_policy[TCA_FLOW_MAX + 1] = {
 	[TCA_FLOW_ACT]		= { .type = NLA_NESTED },
 	[TCA_FLOW_POLICE]	= { .type = NLA_NESTED },
 	[TCA_FLOW_EMATCHES]	= { .type = NLA_NESTED },
+	[TCA_FLOW_PERTURB]	= { .type = NLA_U32 },
 };
 
 static int flow_change(struct tcf_proto *tp, unsigned long base,
@@ -394,6 +404,7 @@ static int flow_change(struct tcf_proto *tp, unsigned long base,
 	struct tcf_exts e;
 	struct tcf_ematch_tree t;
 	unsigned int nkeys = 0;
+	unsigned int perturb_period = 0;
 	u32 baseclass = 0;
 	u32 keymask = 0;
 	u32 mode;
@@ -442,6 +453,14 @@ static int flow_change(struct tcf_proto *tp, unsigned long base,
 			mode = nla_get_u32(tb[TCA_FLOW_MODE]);
 		if (mode != FLOW_MODE_HASH && nkeys > 1)
 			goto err2;
+
+		if (mode == FLOW_MODE_HASH)
+			perturb_period = f->perturb_period;
+		if (tb[TCA_FLOW_PERTURB]) {
+			if (mode != FLOW_MODE_HASH)
+				goto err2;
+			perturb_period = nla_get_u32(tb[TCA_FLOW_PERTURB]) * HZ;
+		}
 	} else {
 		err = -EINVAL;
 		if (!handle)
@@ -455,6 +474,12 @@ static int flow_change(struct tcf_proto *tp, unsigned long base,
 		if (mode != FLOW_MODE_HASH && nkeys > 1)
 			goto err2;
 
+		if (tb[TCA_FLOW_PERTURB]) {
+			if (mode != FLOW_MODE_HASH)
+				goto err2;
+			perturb_period = nla_get_u32(tb[TCA_FLOW_PERTURB]) * HZ;
+		}
+
 		if (TC_H_MAJ(baseclass) == 0)
 			baseclass = TC_H_MAKE(tp->q->handle, baseclass);
 		if (TC_H_MIN(baseclass) == 0)
@@ -467,6 +492,11 @@ static int flow_change(struct tcf_proto *tp, unsigned long base,
 
 		f->handle = handle;
 		f->mask	  = ~0U;
+
+		get_random_bytes(&f->hashrnd, 4);
+		f->perturb_timer.function = flow_perturbation;
+		f->perturb_timer.data = (unsigned long)f;
+		init_timer_deferrable(&f->perturb_timer);
 	}
 
 	tcf_exts_change(tp, &f->exts, &e);
@@ -495,6 +525,11 @@ static int flow_change(struct tcf_proto *tp, unsigned long base,
 	if (baseclass)
 		f->baseclass = baseclass;
 
+	f->perturb_period = perturb_period;
+	del_timer(&f->perturb_timer);
+	if (perturb_period)
+		mod_timer(&f->perturb_timer, jiffies + perturb_period);
+
 	if (*arg == 0)
 		list_add_tail(&f->list, &head->filters);
 
@@ -512,6 +547,7 @@ err1:
 
 static void flow_destroy_filter(struct tcf_proto *tp, struct flow_filter *f)
 {
+	del_timer_sync(&f->perturb_timer);
 	tcf_exts_destroy(tp, &f->exts);
 	tcf_em_tree_destroy(tp, &f->ematches);
 	kfree(f);
@@ -532,11 +568,6 @@ static int flow_init(struct tcf_proto *tp)
 {
 	struct flow_head *head;
 
-	if (!flow_hashrnd_initted) {
-		get_random_bytes(&flow_hashrnd, 4);
-		flow_hashrnd_initted = 1;
-	}
-
 	head = kzalloc(sizeof(*head), GFP_KERNEL);
 	if (head == NULL)
 		return -ENOBUFS;
@@ -605,6 +636,9 @@ static int flow_dump(struct tcf_proto *tp, unsigned long fh,
 	if (f->baseclass)
 		NLA_PUT_U32(skb, TCA_FLOW_BASECLASS, f->baseclass);
 
+	if (f->perturb_period)
+		NLA_PUT_U32(skb, TCA_FLOW_PERTURB, f->perturb_period / HZ);
+
 	if (tcf_exts_dump(skb, &f->exts, &flow_ext_map) < 0)
 		goto nla_put_failure;
 #ifdef CONFIG_NET_EMATCH
-- 
cgit v1.2.3


From f271b2cc78f09c93ccd00a2056d3237134bf994c Mon Sep 17 00:00:00 2001
From: Max Krasnyansky <maxk@qualcomm.com>
Date: Mon, 14 Jul 2008 22:18:19 -0700
Subject: tun: Fix/rewrite packet filtering logic

Please see the following thread to get some context on this
	http://marc.info/?l=linux-netdev&m=121564433018903&w=2

Basically the issue is that current multi-cast filtering stuff in
the TUN/TAP driver is seriously broken.
Original patch went in without proper review and ACK. It was broken and
confusing to start with and subsequent patches broke it completely.
To give you an idea of what's broken here are some of the issues:

- Very confusing comments throughout the code that imply that the
character device is a network interface in its own right, and that packets
are passed between the two nics. Which is completely wrong.

- Wrong set of ioctls is used for setting up filters. They look like
shortcuts for manipulating state of the tun/tap network interface but
in reality manipulate the state of the TX filter.

- ioctls that were originally used for setting address of the the TX filter
got "fixed" and now set the address of the network interface itself. Which
made filter totaly useless.

- Filtering is done too late. Instead of filtering early on, to avoid
unnecessary wakeups, filtering is done in the read() call.

The list goes on and on :)

So the patch cleans all that up. It introduces simple and clean interface for
setting up TX filters (TUNSETTXFILTER + tun_filter spec) and does filtering
before enqueuing the packets.

TX filtering is useful in the scenarios where TAP is part of a bridge, in
which case it gets all broadcast, multicast and potentially other packets when
the bridge is learning. So for example Ethernet tunnelling app may want to
setup TX filters to avoid tunnelling multicast traffic. QEMU and other
hypervisors can push RX filtering that is currently done in the guest into the
host context therefore saving wakeups and unnecessary data transfer.

Signed-off-by: Max Krasnyansky <maxk@qualcomm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/tun.c      | 316 ++++++++++++++++++++++++-------------------------
 include/linux/if_tun.h |  24 +++-
 2 files changed, 174 insertions(+), 166 deletions(-)

(limited to 'include')

diff --git a/drivers/net/tun.c b/drivers/net/tun.c
index 2693f883ecda..901551c8ca0e 100644
--- a/drivers/net/tun.c
+++ b/drivers/net/tun.c
@@ -18,15 +18,11 @@
 /*
  *  Changes:
  *
- *  Brian Braunstein <linuxkernel@bristyle.com> 2007/03/23
- *    Fixed hw address handling.  Now net_device.dev_addr is kept consistent
- *    with tun.dev_addr when the address is set by this module.
- *
  *  Mike Kershaw <dragorn@kismetwireless.net> 2005/08/14
  *    Add TUNSETLINK ioctl to set the link encapsulation
  *
  *  Mark Smith <markzzzsmith@yahoo.com.au>
- *   Use random_ether_addr() for tap MAC address.
+ *    Use random_ether_addr() for tap MAC address.
  *
  *  Harald Roelle <harald.roelle@ifi.lmu.de>  2004/04/20
  *    Fixes in packet dropping, queue length setting and queue wakeup.
@@ -83,9 +79,16 @@ static int debug;
 #define DBG1( a... )
 #endif
 
+#define FLT_EXACT_COUNT 8
+struct tap_filter {
+	unsigned int    count;    /* Number of addrs. Zero means disabled */
+	u32             mask[2];  /* Mask of the hashed addrs */
+	unsigned char	addr[FLT_EXACT_COUNT][ETH_ALEN];
+};
+
 struct tun_struct {
 	struct list_head        list;
-	unsigned long 		flags;
+	unsigned int 		flags;
 	int			attached;
 	uid_t			owner;
 	gid_t			group;
@@ -94,19 +97,119 @@ struct tun_struct {
 	struct sk_buff_head	readq;
 
 	struct net_device	*dev;
+	struct fasync_struct	*fasync;
 
-	struct fasync_struct    *fasync;
-
-	unsigned long if_flags;
-	u8 dev_addr[ETH_ALEN];
-	u32 chr_filter[2];
-	u32 net_filter[2];
+	struct tap_filter       txflt;
 
 #ifdef TUN_DEBUG
 	int debug;
 #endif
 };
 
+/* TAP filterting */
+static void addr_hash_set(u32 *mask, const u8 *addr)
+{
+	int n = ether_crc(ETH_ALEN, addr) >> 26;
+	mask[n >> 5] |= (1 << (n & 31));
+}
+
+static unsigned int addr_hash_test(const u32 *mask, const u8 *addr)
+{
+	int n = ether_crc(ETH_ALEN, addr) >> 26;
+	return mask[n >> 5] & (1 << (n & 31));
+}
+
+static int update_filter(struct tap_filter *filter, void __user *arg)
+{
+	struct { u8 u[ETH_ALEN]; } *addr;
+	struct tun_filter uf;
+	int err, alen, n, nexact;
+
+	if (copy_from_user(&uf, arg, sizeof(uf)))
+		return -EFAULT;
+
+	if (!uf.count) {
+		/* Disabled */
+		filter->count = 0;
+		return 0;
+	}
+
+	alen = ETH_ALEN * uf.count;
+	addr = kmalloc(alen, GFP_KERNEL);
+	if (!addr)
+		return -ENOMEM;
+
+	if (copy_from_user(addr, arg + sizeof(uf), alen)) {
+		err = -EFAULT;
+		goto done;
+	}
+
+	/* The filter is updated without holding any locks. Which is
+	 * perfectly safe. We disable it first and in the worst
+	 * case we'll accept a few undesired packets. */
+	filter->count = 0;
+	wmb();
+
+	/* Use first set of addresses as an exact filter */
+	for (n = 0; n < uf.count && n < FLT_EXACT_COUNT; n++)
+		memcpy(filter->addr[n], addr[n].u, ETH_ALEN);
+
+	nexact = n;
+
+	/* The rest is hashed */
+	memset(filter->mask, 0, sizeof(filter->mask));
+	for (; n < uf.count; n++)
+		addr_hash_set(filter->mask, addr[n].u);
+
+	/* For ALLMULTI just set the mask to all ones.
+	 * This overrides the mask populated above. */
+	if ((uf.flags & TUN_FLT_ALLMULTI))
+		memset(filter->mask, ~0, sizeof(filter->mask));
+
+	/* Now enable the filter */
+	wmb();
+	filter->count = nexact;
+
+	/* Return the number of exact filters */
+	err = nexact;
+
+done:
+	kfree(addr);
+	return err;
+}
+
+/* Returns: 0 - drop, !=0 - accept */
+static int run_filter(struct tap_filter *filter, const struct sk_buff *skb)
+{
+	/* Cannot use eth_hdr(skb) here because skb_mac_hdr() is incorrect
+	 * at this point. */
+	struct ethhdr *eh = (struct ethhdr *) skb->data;
+	int i;
+
+	/* Exact match */
+	for (i = 0; i < filter->count; i++)
+		if (!compare_ether_addr(eh->h_dest, filter->addr[i]))
+			return 1;
+
+	/* Inexact match (multicast only) */
+	if (is_multicast_ether_addr(eh->h_dest))
+		return addr_hash_test(filter->mask, eh->h_dest);
+
+	return 0;
+}
+
+/*
+ * Checks whether the packet is accepted or not.
+ * Returns: 0 - drop, !=0 - accept
+ */
+static int check_filter(struct tap_filter *filter, const struct sk_buff *skb)
+{
+	if (!filter->count)
+		return 1;
+
+	return run_filter(filter, skb);
+}
+
 /* Network device part of the driver */
 
 static unsigned int tun_net_id;
@@ -141,7 +244,12 @@ static int tun_net_xmit(struct sk_buff *skb, struct net_device *dev)
 	if (!tun->attached)
 		goto drop;
 
-	/* Packet dropping */
+	/* Drop if the filter does not like it.
+	 * This is a noop if the filter is disabled.
+	 * Filter can be enabled only for the TAP devices. */
+	if (!check_filter(&tun->txflt, skb))
+		goto drop;
+
 	if (skb_queue_len(&tun->readq) >= dev->tx_queue_len) {
 		if (!(tun->flags & TUN_ONE_QUEUE)) {
 			/* Normal queueing mode. */
@@ -158,7 +266,7 @@ static int tun_net_xmit(struct sk_buff *skb, struct net_device *dev)
 		}
 	}
 
-	/* Queue packet */
+	/* Enqueue packet */
 	skb_queue_tail(&tun->readq, skb);
 	dev->trans_start = jiffies;
 
@@ -174,41 +282,14 @@ drop:
 	return 0;
 }
 
-/** Add the specified Ethernet address to this multicast filter. */
-static void
-add_multi(u32* filter, const u8* addr)
-{
-	int bit_nr = ether_crc(ETH_ALEN, addr) >> 26;
-	filter[bit_nr >> 5] |= 1 << (bit_nr & 31);
-}
-
-/** Remove the specified Ethernet addres from this multicast filter. */
-static void
-del_multi(u32* filter, const u8* addr)
+static void tun_net_mclist(struct net_device *dev)
 {
-	int bit_nr = ether_crc(ETH_ALEN, addr) >> 26;
-	filter[bit_nr >> 5] &= ~(1 << (bit_nr & 31));
-}
-
-/** Update the list of multicast groups to which the network device belongs.
- * This list is used to filter packets being sent from the character device to
- * the network device. */
-static void
-tun_net_mclist(struct net_device *dev)
-{
-	struct tun_struct *tun = netdev_priv(dev);
-	const struct dev_mc_list *mclist;
-	int i;
-	DECLARE_MAC_BUF(mac);
-	DBG(KERN_DEBUG "%s: tun_net_mclist: mc_count %d\n",
-			dev->name, dev->mc_count);
-	memset(tun->chr_filter, 0, sizeof tun->chr_filter);
-	for (i = 0, mclist = dev->mc_list; i < dev->mc_count && mclist != NULL;
-			i++, mclist = mclist->next) {
-		add_multi(tun->net_filter, mclist->dmi_addr);
-		DBG(KERN_DEBUG "%s: tun_net_mclist: %s\n",
-		    dev->name, print_mac(mac, mclist->dmi_addr));
-	}
+	/*
+	 * This callback is supposed to deal with mc filter in
+	 * _rx_ path and has nothing to do with the _tx_ path.
+	 * In rx path we always accept everything userspace gives us.
+	 */
+	return;
 }
 
 #define MIN_MTU 68
@@ -244,13 +325,11 @@ static void tun_net_init(struct net_device *dev)
 
 	case TUN_TAP_DEV:
 		/* Ethernet TAP Device */
-		dev->set_multicast_list = tun_net_mclist;
-
 		ether_setup(dev);
-		dev->change_mtu = tun_net_change_mtu;
+		dev->change_mtu         = tun_net_change_mtu;
+		dev->set_multicast_list = tun_net_mclist;
 
-		/* random address already created for us by tun_set_iff, use it */
-		memcpy(dev->dev_addr, tun->dev_addr, min(sizeof(tun->dev_addr), sizeof(dev->dev_addr)) );
+		random_ether_addr(dev->dev_addr);
 
 		dev->tx_queue_len = TUN_READQ_SIZE;  /* We prefer our own queue length */
 		break;
@@ -486,7 +565,6 @@ static ssize_t tun_chr_aio_read(struct kiocb *iocb, const struct iovec *iv,
 	DECLARE_WAITQUEUE(wait, current);
 	struct sk_buff *skb;
 	ssize_t len, ret = 0;
-	DECLARE_MAC_BUF(mac);
 
 	if (!tun)
 		return -EBADFD;
@@ -499,10 +577,6 @@ static ssize_t tun_chr_aio_read(struct kiocb *iocb, const struct iovec *iv,
 
 	add_wait_queue(&tun->read_wait, &wait);
 	while (len) {
-		const u8 ones[ ETH_ALEN] = { 0xff, 0xff, 0xff, 0xff, 0xff, 0xff };
-		u8 addr[ ETH_ALEN];
-		int bit_nr;
-
 		current->state = TASK_INTERRUPTIBLE;
 
 		/* Read frames from the queue */
@@ -522,36 +596,9 @@ static ssize_t tun_chr_aio_read(struct kiocb *iocb, const struct iovec *iv,
 		}
 		netif_wake_queue(tun->dev);
 
-		/** Decide whether to accept this packet. This code is designed to
-		 * behave identically to an Ethernet interface. Accept the packet if
-		 * - we are promiscuous.
-		 * - the packet is addressed to us.
-		 * - the packet is broadcast.
-		 * - the packet is multicast and
-		 *   - we are multicast promiscous.
-		 *   - we belong to the multicast group.
-		 */
-		skb_copy_from_linear_data(skb, addr, min_t(size_t, sizeof addr,
-								   skb->len));
-		bit_nr = ether_crc(sizeof addr, addr) >> 26;
-		if ((tun->if_flags & IFF_PROMISC) ||
-				memcmp(addr, tun->dev_addr, sizeof addr) == 0 ||
-				memcmp(addr, ones, sizeof addr) == 0 ||
-				(((addr[0] == 1 && addr[1] == 0 && addr[2] == 0x5e) ||
-				  (addr[0] == 0x33 && addr[1] == 0x33)) &&
-				 ((tun->if_flags & IFF_ALLMULTI) ||
-				  (tun->chr_filter[bit_nr >> 5] & (1 << (bit_nr & 31)))))) {
-			DBG(KERN_DEBUG "%s: tun_chr_readv: accepted: %s\n",
-					tun->dev->name, print_mac(mac, addr));
-			ret = tun_put_user(tun, skb, (struct iovec *) iv, len);
-			kfree_skb(skb);
-			break;
-		} else {
-			DBG(KERN_DEBUG "%s: tun_chr_readv: rejected: %s\n",
-					tun->dev->name, print_mac(mac, addr));
-			kfree_skb(skb);
-			continue;
-		}
+		ret = tun_put_user(tun, skb, (struct iovec *) iv, len);
+		kfree_skb(skb);
+		break;
 	}
 
 	current->state = TASK_RUNNING;
@@ -647,12 +694,7 @@ static int tun_set_iff(struct net *net, struct file *file, struct ifreq *ifr)
 		tun = netdev_priv(dev);
 		tun->dev = dev;
 		tun->flags = flags;
-		/* Be promiscuous by default to maintain previous behaviour. */
-		tun->if_flags = IFF_PROMISC;
-		/* Generate random Ethernet address. */
-		*(__be16 *)tun->dev_addr = htons(0x00FF);
-		get_random_bytes(tun->dev_addr + sizeof(u16), 4);
-		memset(tun->chr_filter, 0, sizeof tun->chr_filter);
+		tun->txflt.count = 0;
 
 		tun_net_init(dev);
 
@@ -751,6 +793,7 @@ static int tun_chr_ioctl(struct inode *inode, struct file *file,
 	struct tun_struct *tun = file->private_data;
 	void __user* argp = (void __user*)arg;
 	struct ifreq ifr;
+	int ret;
 	DECLARE_MAC_BUF(mac);
 
 	if (cmd == TUNSETIFF || _IOC_TYPE(cmd) == 0x89)
@@ -826,9 +869,6 @@ static int tun_chr_ioctl(struct inode *inode, struct file *file,
 		break;
 
 	case TUNSETLINK:
-	{
-		int ret;
-
 		/* Only allow setting the type when the interface is down */
 		rtnl_lock();
 		if (tun->dev->flags & IFF_UP) {
@@ -842,94 +882,44 @@ static int tun_chr_ioctl(struct inode *inode, struct file *file,
 		}
 		rtnl_unlock();
 		return ret;
-	}
 
 #ifdef TUN_DEBUG
 	case TUNSETDEBUG:
 		tun->debug = arg;
 		break;
 #endif
-
 	case TUNSETOFFLOAD:
-	{
-		int ret;
 		rtnl_lock();
 		ret = set_offload(tun->dev, arg);
 		rtnl_unlock();
 		return ret;
-	}
 
-	case SIOCGIFFLAGS:
-		ifr.ifr_flags = tun->if_flags;
-		if (copy_to_user( argp, &ifr, sizeof ifr))
-			return -EFAULT;
-		return 0;
-
-	case SIOCSIFFLAGS:
-		/** Set the character device's interface flags. Currently only
-		 * IFF_PROMISC and IFF_ALLMULTI are used. */
-		tun->if_flags = ifr.ifr_flags;
-		DBG(KERN_INFO "%s: interface flags 0x%lx\n",
-				tun->dev->name, tun->if_flags);
-		return 0;
+	case TUNSETTXFILTER:
+		/* Can be set only for TAPs */
+		if ((tun->flags & TUN_TYPE_MASK) != TUN_TAP_DEV)
+			return -EINVAL;
+		rtnl_lock();
+		ret = update_filter(&tun->txflt, (void *) __user arg);
+		rtnl_unlock();
+		return ret;
 
 	case SIOCGIFHWADDR:
-		/* Note: the actual net device's address may be different */
-		memcpy(ifr.ifr_hwaddr.sa_data, tun->dev_addr,
-				min(sizeof ifr.ifr_hwaddr.sa_data, sizeof tun->dev_addr));
-		if (copy_to_user( argp, &ifr, sizeof ifr))
+		/* Get hw addres */
+		memcpy(ifr.ifr_hwaddr.sa_data, tun->dev->dev_addr, ETH_ALEN);
+		ifr.ifr_hwaddr.sa_family = tun->dev->type;
+		if (copy_to_user(argp, &ifr, sizeof ifr))
 			return -EFAULT;
 		return 0;
 
 	case SIOCSIFHWADDR:
-	{
-		/* try to set the actual net device's hw address */
-		int ret;
+		/* Set hw address */
+		DBG(KERN_DEBUG "%s: set hw address: %s\n",
+			tun->dev->name, print_mac(mac, ifr.ifr_hwaddr.sa_data));
 
 		rtnl_lock();
 		ret = dev_set_mac_address(tun->dev, &ifr.ifr_hwaddr);
 		rtnl_unlock();
-
-		if (ret == 0) {
-			/** Set the character device's hardware address. This is used when
-			 * filtering packets being sent from the network device to the character
-			 * device. */
-			memcpy(tun->dev_addr, ifr.ifr_hwaddr.sa_data,
-					min(sizeof ifr.ifr_hwaddr.sa_data, sizeof tun->dev_addr));
-			DBG(KERN_DEBUG "%s: set hardware address: %x:%x:%x:%x:%x:%x\n",
-					tun->dev->name,
-					tun->dev_addr[0], tun->dev_addr[1], tun->dev_addr[2],
-					tun->dev_addr[3], tun->dev_addr[4], tun->dev_addr[5]);
-		}
-
-		return  ret;
-	}
-
-	case SIOCADDMULTI:
-		/** Add the specified group to the character device's multicast filter
-		 * list. */
-		rtnl_lock();
-		netif_tx_lock_bh(tun->dev);
-		add_multi(tun->chr_filter, ifr.ifr_hwaddr.sa_data);
-		netif_tx_unlock_bh(tun->dev);
-		rtnl_unlock();
-
-		DBG(KERN_DEBUG "%s: add multi: %s\n",
-		    tun->dev->name, print_mac(mac, ifr.ifr_hwaddr.sa_data));
-		return 0;
-
-	case SIOCDELMULTI:
-		/** Remove the specified group from the character device's multicast
-		 * filter list. */
-		rtnl_lock();
-		netif_tx_lock_bh(tun->dev);
-		del_multi(tun->chr_filter, ifr.ifr_hwaddr.sa_data);
-		netif_tx_unlock_bh(tun->dev);
-		rtnl_unlock();
-
-		DBG(KERN_DEBUG "%s: del multi: %s\n",
-		    tun->dev->name, print_mac(mac, ifr.ifr_hwaddr.sa_data));
-		return 0;
+		return ret;
 
 	default:
 		return -EINVAL;
diff --git a/include/linux/if_tun.h b/include/linux/if_tun.h
index 563fae542da6..4c6307ad9fdb 100644
--- a/include/linux/if_tun.h
+++ b/include/linux/if_tun.h
@@ -17,6 +17,7 @@
 #define __IF_TUN_H
 
 #include <linux/types.h>
+#include <linux/if_ether.h>
 
 /* Read queue size */
 #define TUN_READQ_SIZE	500
@@ -42,7 +43,8 @@
 #define TUNSETLINK    _IOW('T', 205, int)
 #define TUNSETGROUP   _IOW('T', 206, int)
 #define TUNGETFEATURES _IOR('T', 207, unsigned int)
-#define TUNSETOFFLOAD _IOW('T', 208, unsigned int)
+#define TUNSETOFFLOAD  _IOW('T', 208, unsigned int)
+#define TUNSETTXFILTER _IOW('T', 209, unsigned int)
 
 /* TUNSETIFF ifr flags */
 #define IFF_TUN		0x0001
@@ -57,10 +59,26 @@
 #define TUN_F_TSO6	0x04	/* I can handle TSO for IPv6 packets */
 #define TUN_F_TSO_ECN	0x08	/* I can handle TSO with ECN bits. */
 
+/* Protocol info prepended to the packets (when IFF_NO_PI is not set) */
+#define TUN_PKT_STRIP	0x0001
 struct tun_pi {
-	unsigned short flags;
+	__u16  flags;
 	__be16 proto;
 };
-#define TUN_PKT_STRIP	0x0001
+
+/*
+ * Filter spec (used for SETXXFILTER ioctls)
+ * This stuff is applicable only to the TAP (Ethernet) devices.
+ * If the count is zero the filter is disabled and the driver accepts
+ * all packets (promisc mode).
+ * If the filter is enabled in order to accept broadcast packets
+ * broadcast addr must be explicitly included in the addr list.
+ */
+#define TUN_FLT_ALLMULTI 0x0001 /* Accept all multicast packets */
+struct tun_filter {
+	__u16  flags; /* TUN_FLT_ flags see above */
+	__u16  count; /* Number of addresses */
+	__u8   addr[0][ETH_ALEN];
+};
 
 #endif /* __IF_TUN_H */
-- 
cgit v1.2.3


From 0ea522416b658dedfc9d565b331624a55a6260ad Mon Sep 17 00:00:00 2001
From: Allan Stephens <allan.stephens@windriver.com>
Date: Mon, 14 Jul 2008 22:42:19 -0700
Subject: tipc: Remove unneeded parameter to tipc_createport_raw()

This patch eliminates an unneeded parameter when creating a low-level
TIPC port object.  Instead of returning both the pointer to the port
structure and the port's reference ID, it now returns only the pointer
since the port structure contains the reference ID as one of its fields.

Signed-off-by: Allan Stephens <allan.stephens@windriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/tipc/tipc_port.h | 13 +++----------
 net/tipc/port.c              | 28 ++++++++++------------------
 net/tipc/socket.c            | 15 +++++++--------
 3 files changed, 20 insertions(+), 36 deletions(-)

(limited to 'include')

diff --git a/include/net/tipc/tipc_port.h b/include/net/tipc/tipc_port.h
index 9923e41a8215..c54917cbfa48 100644
--- a/include/net/tipc/tipc_port.h
+++ b/include/net/tipc/tipc_port.h
@@ -2,7 +2,7 @@
  * include/net/tipc/tipc_port.h: Include file for privileged access to TIPC ports
  * 
  * Copyright (c) 1994-2007, Ericsson AB
- * Copyright (c) 2005-2007, Wind River Systems
+ * Copyright (c) 2005-2008, Wind River Systems
  * All rights reserved.
  *
  * Redistribution and use in source and binary forms, with or without
@@ -75,17 +75,10 @@ struct tipc_port {
 };
 
 
-/**
- * tipc_createport_raw - create a native TIPC port and return it's reference
- *
- * Note: 'dispatcher' and 'wakeup' deliver a locked port.
- */
-
-u32 tipc_createport_raw(void *usr_handle,
+struct tipc_port *tipc_createport_raw(void *usr_handle,
 			u32 (*dispatcher)(struct tipc_port *, struct sk_buff *),
 			void (*wakeup)(struct tipc_port *),
-			const u32 importance,
-			struct tipc_port **tp_ptr);
+			const u32 importance);
 
 int tipc_reject_msg(struct sk_buff *buf, u32 err);
 
diff --git a/net/tipc/port.c b/net/tipc/port.c
index 2e0cff408ff9..ffba1e7f06d2 100644
--- a/net/tipc/port.c
+++ b/net/tipc/port.c
@@ -2,7 +2,7 @@
  * net/tipc/port.c: TIPC port code
  *
  * Copyright (c) 1992-2007, Ericsson AB
- * Copyright (c) 2004-2007, Wind River Systems
+ * Copyright (c) 2004-2008, Wind River Systems
  * All rights reserved.
  *
  * Redistribution and use in source and binary forms, with or without
@@ -213,16 +213,13 @@ exit:
 /**
  * tipc_createport_raw - create a generic TIPC port
  *
- * Returns port reference, or 0 if unable to create it
- *
- * Note: The newly created port is returned in the locked state.
+ * Returns pointer to (locked) TIPC port, or NULL if unable to create it
  */
 
-u32 tipc_createport_raw(void *usr_handle,
+struct tipc_port *tipc_createport_raw(void *usr_handle,
 			u32 (*dispatcher)(struct tipc_port *, struct sk_buff *),
 			void (*wakeup)(struct tipc_port *),
-			const u32 importance,
-			struct tipc_port **tp_ptr)
+			const u32 importance)
 {
 	struct port *p_ptr;
 	struct tipc_msg *msg;
@@ -231,13 +228,13 @@ u32 tipc_createport_raw(void *usr_handle,
 	p_ptr = kzalloc(sizeof(*p_ptr), GFP_ATOMIC);
 	if (!p_ptr) {
 		warn("Port creation failed, no memory\n");
-		return 0;
+		return NULL;
 	}
 	ref = tipc_ref_acquire(p_ptr, &p_ptr->publ.lock);
 	if (!ref) {
 		warn("Port creation failed, reference table exhausted\n");
 		kfree(p_ptr);
-		return 0;
+		return NULL;
 	}
 
 	p_ptr->publ.usr_handle = usr_handle;
@@ -260,8 +257,7 @@ u32 tipc_createport_raw(void *usr_handle,
 	INIT_LIST_HEAD(&p_ptr->port_list);
 	list_add_tail(&p_ptr->port_list, &ports);
 	spin_unlock_bh(&tipc_port_list_lock);
-	*tp_ptr = &p_ptr->publ;
-	return ref;
+	return &(p_ptr->publ);
 }
 
 int tipc_deleteport(u32 ref)
@@ -1044,21 +1040,18 @@ int tipc_createport(u32 user_ref,
 {
 	struct user_port *up_ptr;
 	struct port *p_ptr;
-	struct tipc_port *tp_ptr;
-	u32 ref;
 
 	up_ptr = kmalloc(sizeof(*up_ptr), GFP_ATOMIC);
 	if (!up_ptr) {
 		warn("Port creation failed, no memory\n");
 		return -ENOMEM;
 	}
-	ref = tipc_createport_raw(NULL, port_dispatcher, port_wakeup,
-				  importance, &tp_ptr);
-	if (ref == 0) {
+	p_ptr = (struct port *)tipc_createport_raw(NULL, port_dispatcher,
+						   port_wakeup, importance);
+	if (!p_ptr) {
 		kfree(up_ptr);
 		return -ENOMEM;
 	}
-	p_ptr = (struct port *)tp_ptr;
 
 	p_ptr->user_port = up_ptr;
 	up_ptr->user_ref = user_ref;
@@ -1074,7 +1067,6 @@ int tipc_createport(u32 user_ref,
 	INIT_LIST_HEAD(&up_ptr->uport_list);
 	tipc_reg_add_port(up_ptr);
 	*portref = p_ptr->publ.ref;
-	dbg(" tipc_createport: %x with ref %u\n", p_ptr, p_ptr->publ.ref);
 	tipc_port_unlock(p_ptr);
 	return TIPC_OK;
 }
diff --git a/net/tipc/socket.c b/net/tipc/socket.c
index 38f48795b40e..9c362c5759ba 100644
--- a/net/tipc/socket.c
+++ b/net/tipc/socket.c
@@ -2,7 +2,7 @@
  * net/tipc/socket.c: TIPC socket API
  *
  * Copyright (c) 2001-2007, Ericsson AB
- * Copyright (c) 2004-2007, Wind River Systems
+ * Copyright (c) 2004-2008, Wind River Systems
  * All rights reserved.
  *
  * Redistribution and use in source and binary forms, with or without
@@ -189,7 +189,6 @@ static int tipc_create(struct net *net, struct socket *sock, int protocol)
 	socket_state state;
 	struct sock *sk;
 	struct tipc_port *tp_ptr;
-	u32 portref;
 
 	/* Validate arguments */
 
@@ -225,9 +224,9 @@ static int tipc_create(struct net *net, struct socket *sock, int protocol)
 
 	/* Allocate TIPC port for socket to use */
 
-	portref = tipc_createport_raw(sk, &dispatch, &wakeupdispatch,
-				      TIPC_LOW_IMPORTANCE, &tp_ptr);
-	if (unlikely(portref == 0)) {
+	tp_ptr = tipc_createport_raw(sk, &dispatch, &wakeupdispatch,
+				     TIPC_LOW_IMPORTANCE);
+	if (unlikely(!tp_ptr)) {
 		sk_free(sk);
 		return -ENOMEM;
 	}
@@ -240,14 +239,14 @@ static int tipc_create(struct net *net, struct socket *sock, int protocol)
 	sock_init_data(sock, sk);
 	sk->sk_rcvtimeo = msecs_to_jiffies(CONN_TIMEOUT_DEFAULT);
 	sk->sk_backlog_rcv = backlog_rcv;
-	tipc_sk(sk)->p = tipc_get_port(portref);
+	tipc_sk(sk)->p = tp_ptr;
 
 	spin_unlock_bh(tp_ptr->lock);
 
 	if (sock->state == SS_READY) {
-		tipc_set_portunreturnable(portref, 1);
+		tipc_set_portunreturnable(tp_ptr->ref, 1);
 		if (sock->type == SOCK_DGRAM)
-			tipc_set_portunreliable(portref, 1);
+			tipc_set_portunreliable(tp_ptr->ref, 1);
 	}
 
 	atomic_inc(&tipc_user_count);
-- 
cgit v1.2.3


From 6aa895b047720f71ec4eb11452f7c3ce8426941f Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Mon, 14 Jul 2008 22:49:06 -0700
Subject: vlan: Don't store VLAN tag in cb

Use a real skb member to store the skb to avoid clashes with qdiscs,
which are allowed to use the cb area themselves. As currently only real
devices that consume the skb set the NETIF_F_HW_VLAN_TX flag, no explicit
invalidation is neccessary.

The new member fills a hole on 64 bit, the skb layout changes from:

        __u32                      mark;                 /*   172     4 */
        sk_buff_data_t             transport_header;     /*   176     4 */
        sk_buff_data_t             network_header;       /*   180     4 */
        sk_buff_data_t             mac_header;           /*   184     4 */
        sk_buff_data_t             tail;                 /*   188     4 */
        /* --- cacheline 3 boundary (192 bytes) --- */
        sk_buff_data_t             end;                  /*   192     4 */

        /* XXX 4 bytes hole, try to pack */

to

        __u32                      mark;                 /*   172     4 */
        __u16                      vlan_tci;             /*   176     2 */

        /* XXX 2 bytes hole, try to pack */

        sk_buff_data_t             transport_header;     /*   180     4 */
        sk_buff_data_t             network_header;       /*   184     4 */

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/if_vlan.h | 31 +++++++------------------------
 include/linux/skbuff.h  |  3 +++
 net/core/skbuff.c       |  3 +++
 3 files changed, 13 insertions(+), 24 deletions(-)

(limited to 'include')

diff --git a/include/linux/if_vlan.h b/include/linux/if_vlan.h
index 93f5d9b0e9f9..9e7b49b8062d 100644
--- a/include/linux/if_vlan.h
+++ b/include/linux/if_vlan.h
@@ -105,17 +105,8 @@ static inline void vlan_group_set_device(struct vlan_group *vg,
 	array[vlan_id % VLAN_GROUP_ARRAY_PART_LEN] = dev;
 }
 
-/* VLAN tx hw acceleration helpers. */
-struct vlan_skb_tx_cookie {
-	u32	magic;
-	u32	vlan_tag;
-};
-
-#define VLAN_TX_COOKIE_MAGIC	0x564c414e	/* "VLAN" in ascii. */
-#define VLAN_TX_SKB_CB(__skb)	((struct vlan_skb_tx_cookie *)&((__skb)->cb[0]))
-#define vlan_tx_tag_present(__skb) \
-	(VLAN_TX_SKB_CB(__skb)->magic == VLAN_TX_COOKIE_MAGIC)
-#define vlan_tx_tag_get(__skb)	(VLAN_TX_SKB_CB(__skb)->vlan_tag)
+#define vlan_tx_tag_present(__skb)	((__skb)->vlan_tci)
+#define vlan_tx_tag_get(__skb)		((__skb)->vlan_tci)
 
 #if defined(CONFIG_VLAN_8021Q) || defined(CONFIG_VLAN_8021Q_MODULE)
 extern struct net_device *vlan_dev_real_dev(const struct net_device *dev);
@@ -210,17 +201,12 @@ static inline struct sk_buff *__vlan_put_tag(struct sk_buff *skb, u16 vlan_tci)
  * @skb: skbuff to tag
  * @vlan_tci: VLAN TCI to insert
  *
- * Puts the VLAN TCI in @skb->cb[] and lets the device do the rest
+ * Puts the VLAN TCI in @skb->vlan_tci and lets the device do the rest
  */
 static inline struct sk_buff *__vlan_hwaccel_put_tag(struct sk_buff *skb,
 						     u16 vlan_tci)
 {
-	struct vlan_skb_tx_cookie *cookie;
-
-	cookie = VLAN_TX_SKB_CB(skb);
-	cookie->magic = VLAN_TX_COOKIE_MAGIC;
-	cookie->vlan_tag = vlan_tci;
-
+	skb->vlan_tci = vlan_tci;
 	return skb;
 }
 
@@ -267,16 +253,13 @@ static inline int __vlan_get_tag(const struct sk_buff *skb, u16 *vlan_tci)
  * @skb: skbuff to query
  * @vlan_tci: buffer to store vlaue
  *
- * Returns error if @skb->cb[] is not set correctly
+ * Returns error if @skb->vlan_tci is not set correctly
  */
 static inline int __vlan_hwaccel_get_tag(const struct sk_buff *skb,
 					 u16 *vlan_tci)
 {
-	struct vlan_skb_tx_cookie *cookie;
-
-	cookie = VLAN_TX_SKB_CB(skb);
-	if (cookie->magic == VLAN_TX_COOKIE_MAGIC) {
-		*vlan_tci = cookie->vlan_tag;
+	if (vlan_tx_tag_present(skb)) {
+		*vlan_tci = skb->vlan_tci;
 		return 0;
 	} else {
 		*vlan_tci = 0;
diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index 8f10e3d08fd9..7ea44f6621f2 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -246,6 +246,7 @@ typedef unsigned char *sk_buff_data_t;
  *	@dma_cookie: a cookie to one of several possible DMA operations
  *		done by skb DMA functions
  *	@secmark: security marking
+ *	@vlan_tci: vlan tag control information
  */
 
 struct sk_buff {
@@ -326,6 +327,8 @@ struct sk_buff {
 
 	__u32			mark;
 
+	__u16			vlan_tci;
+
 	sk_buff_data_t		transport_header;
 	sk_buff_data_t		network_header;
 	sk_buff_data_t		mac_header;
diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index 7c571560e9d2..50a853f7cd8e 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -459,6 +459,8 @@ static void __copy_skb_header(struct sk_buff *new, const struct sk_buff *old)
 	new->tc_verd		= old->tc_verd;
 #endif
 #endif
+	new->vlan_tci		= old->vlan_tci;
+
 	skb_copy_secmark(new, old);
 }
 
@@ -2286,6 +2288,7 @@ struct sk_buff *skb_segment(struct sk_buff *skb, int features)
 		skb_copy_queue_mapping(nskb, skb);
 		nskb->priority = skb->priority;
 		nskb->protocol = skb->protocol;
+		nskb->vlan_tci = skb->vlan_tci;
 		nskb->dst = dst_clone(skb->dst);
 		memcpy(nskb->cb, skb->cb, sizeof(skb->cb));
 		nskb->pkt_type = skb->pkt_type;
-- 
cgit v1.2.3


From bc1d0411b804ad190cdadabac48a10067f17b9e6 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Mon, 14 Jul 2008 22:49:30 -0700
Subject: vlan: deliver packets received with VLAN acceleration to network taps

When VLAN header stripping is used, packets currently bypass packet
sockets (and other network taps) completely. For locally existing
VLANs, they appear directly on the VLAN device, for unknown VLANs
they are silently dropped.

Add a new function netif_nit_deliver() to deliver incoming packets
to all network interface taps and use it in __vlan_hwaccel_rx() to
make VLAN packets visible on the underlying device.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h |  1 +
 net/8021q/vlan_core.c     |  4 ++++
 net/core/dev.c            | 27 +++++++++++++++++++++++++++
 3 files changed, 32 insertions(+)

(limited to 'include')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index b54ec16dfbda..ba5c4639ea91 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -1165,6 +1165,7 @@ extern int		netif_rx(struct sk_buff *skb);
 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		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 *);
 extern int		dev_ethtool(struct net *net, struct ifreq *);
diff --git a/net/8021q/vlan_core.c b/net/8021q/vlan_core.c
index 68df12d3664b..916061f681b6 100644
--- a/net/8021q/vlan_core.c
+++ b/net/8021q/vlan_core.c
@@ -14,6 +14,9 @@ int __vlan_hwaccel_rx(struct sk_buff *skb, struct vlan_group *grp,
 		return NET_RX_DROP;
 	}
 
+	skb->vlan_tci = vlan_tci;
+	netif_nit_deliver(skb);
+
 	skb->dev = vlan_group_get_device(grp, vlan_tci & VLAN_VID_MASK);
 	if (skb->dev == NULL) {
 		dev_kfree_skb_any(skb);
@@ -22,6 +25,7 @@ int __vlan_hwaccel_rx(struct sk_buff *skb, struct vlan_group *grp,
 		return NET_RX_SUCCESS;
 	}
 	skb->dev->last_rx = jiffies;
+	skb->vlan_tci = 0;
 
 	stats = &skb->dev->stats;
 	stats->rx_packets++;
diff --git a/net/core/dev.c b/net/core/dev.c
index a29a359b15d1..feaab4898a5b 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -2068,6 +2068,33 @@ out:
 }
 #endif
 
+/*
+ * 	netif_nit_deliver - deliver received packets to network taps
+ * 	@skb: buffer
+ *
+ * 	This function is used to deliver incoming packets to network
+ * 	taps. It should be used when the normal netif_receive_skb path
+ * 	is bypassed, for example because of VLAN acceleration.
+ */
+void netif_nit_deliver(struct sk_buff *skb)
+{
+	struct packet_type *ptype;
+
+	if (list_empty(&ptype_all))
+		return;
+
+	skb_reset_network_header(skb);
+	skb_reset_transport_header(skb);
+	skb->mac_len = skb->network_header - skb->mac_header;
+
+	rcu_read_lock();
+	list_for_each_entry_rcu(ptype, &ptype_all, list) {
+		if (!ptype->dev || ptype->dev == skb->dev)
+			deliver_skb(skb, ptype, skb->dev);
+	}
+	rcu_read_unlock();
+}
+
 /**
  *	netif_receive_skb - process receive buffer from network
  *	@skb: buffer to process
-- 
cgit v1.2.3


From bbd6ef87c544d88c30e4b762b1b61ef267a7d279 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Mon, 14 Jul 2008 22:50:15 -0700
Subject: packet: support extensible, 64 bit clean mmaped ring structure

The tpacket_hdr is not 64 bit clean due to use of an unsigned long
and can't be extended because the following struct sockaddr_ll needs
to be at a fixed offset.

Add support for a version 2 tpacket protocol that removes these
limitations.

Userspace can query the header size through a new getsockopt option
and change the protocol version through a setsockopt option. The
changes needed to switch to the new protocol version are:

1. replace struct tpacket_hdr by struct tpacket2_hdr
2. query header len and save
3. set protocol version to 2
 - set up ring as usual
4. for getting the sockaddr_ll, use (void *)hdr + TPACKET_ALIGN(hdrlen)
   instead of (void *)hdr + TPACKET_ALIGN(sizeof(struct tpacket_hdr))

Steps 2 and 4 can be omitted if the struct sockaddr_ll isn't needed.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/if_packet.h |  21 ++++++
 net/packet/af_packet.c    | 179 +++++++++++++++++++++++++++++++++++++---------
 2 files changed, 167 insertions(+), 33 deletions(-)

(limited to 'include')

diff --git a/include/linux/if_packet.h b/include/linux/if_packet.h
index ad09609227ff..d4d3c82448f5 100644
--- a/include/linux/if_packet.h
+++ b/include/linux/if_packet.h
@@ -43,6 +43,8 @@ struct sockaddr_ll
 #define PACKET_COPY_THRESH		7
 #define PACKET_AUXDATA			8
 #define PACKET_ORIGDEV			9
+#define PACKET_VERSION			10
+#define PACKET_HDRLEN			11
 
 struct tpacket_stats
 {
@@ -79,6 +81,25 @@ struct tpacket_hdr
 #define TPACKET_ALIGN(x)	(((x)+TPACKET_ALIGNMENT-1)&~(TPACKET_ALIGNMENT-1))
 #define TPACKET_HDRLEN		(TPACKET_ALIGN(sizeof(struct tpacket_hdr)) + sizeof(struct sockaddr_ll))
 
+struct tpacket2_hdr
+{
+	__u32		tp_status;
+	__u32		tp_len;
+	__u32		tp_snaplen;
+	__u16		tp_mac;
+	__u16		tp_net;
+	__u32		tp_sec;
+	__u32		tp_nsec;
+};
+
+#define TPACKET2_HDRLEN		(TPACKET_ALIGN(sizeof(struct tpacket2_hdr)) + sizeof(struct sockaddr_ll))
+
+enum tpacket_versions
+{
+	TPACKET_V1,
+	TPACKET_V2,
+};
+
 /*
    Frame structure:
 
diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c
index 9f2269166687..4f059775d48f 100644
--- a/net/packet/af_packet.c
+++ b/net/packet/af_packet.c
@@ -186,6 +186,8 @@ struct packet_sock {
 	unsigned int            pg_vec_order;
 	unsigned int		pg_vec_pages;
 	unsigned int		pg_vec_len;
+	enum tpacket_versions	tp_version;
+	unsigned int		tp_hdrlen;
 #endif
 };
 
@@ -201,14 +203,52 @@ struct packet_skb_cb {
 
 #ifdef CONFIG_PACKET_MMAP
 
-static inline struct tpacket_hdr *packet_lookup_frame(struct packet_sock *po, unsigned int position)
+static void *packet_lookup_frame(struct packet_sock *po, unsigned int position,
+				 int status)
 {
 	unsigned int pg_vec_pos, frame_offset;
+	union {
+		struct tpacket_hdr *h1;
+		struct tpacket2_hdr *h2;
+		void *raw;
+	} h;
 
 	pg_vec_pos = position / po->frames_per_block;
 	frame_offset = position % po->frames_per_block;
 
-	return (struct tpacket_hdr *)(po->pg_vec[pg_vec_pos] + (frame_offset * po->frame_size));
+	h.raw = po->pg_vec[pg_vec_pos] + (frame_offset * po->frame_size);
+	switch (po->tp_version) {
+	case TPACKET_V1:
+		if (status != h.h1->tp_status ? TP_STATUS_USER :
+						TP_STATUS_KERNEL)
+			return NULL;
+		break;
+	case TPACKET_V2:
+		if (status != h.h2->tp_status ? TP_STATUS_USER :
+						TP_STATUS_KERNEL)
+			return NULL;
+		break;
+	}
+	return h.raw;
+}
+
+static void __packet_set_status(struct packet_sock *po, void *frame, int status)
+{
+	union {
+		struct tpacket_hdr *h1;
+		struct tpacket2_hdr *h2;
+		void *raw;
+	} h;
+
+	h.raw = frame;
+	switch (po->tp_version) {
+	case TPACKET_V1:
+		h.h1->tp_status = status;
+		break;
+	case TPACKET_V2:
+		h.h2->tp_status = status;
+		break;
+	}
 }
 #endif
 
@@ -551,14 +591,19 @@ static int tpacket_rcv(struct sk_buff *skb, struct net_device *dev, struct packe
 	struct sock *sk;
 	struct packet_sock *po;
 	struct sockaddr_ll *sll;
-	struct tpacket_hdr *h;
+	union {
+		struct tpacket_hdr *h1;
+		struct tpacket2_hdr *h2;
+		void *raw;
+	} h;
 	u8 * skb_head = skb->data;
 	int skb_len = skb->len;
 	unsigned int snaplen, res;
 	unsigned long status = TP_STATUS_LOSING|TP_STATUS_USER;
-	unsigned short macoff, netoff;
+	unsigned short macoff, netoff, hdrlen;
 	struct sk_buff *copy_skb = NULL;
 	struct timeval tv;
+	struct timespec ts;
 
 	if (skb->pkt_type == PACKET_LOOPBACK)
 		goto drop;
@@ -590,10 +635,11 @@ static int tpacket_rcv(struct sk_buff *skb, struct net_device *dev, struct packe
 		snaplen = res;
 
 	if (sk->sk_type == SOCK_DGRAM) {
-		macoff = netoff = TPACKET_ALIGN(TPACKET_HDRLEN) + 16;
+		macoff = netoff = TPACKET_ALIGN(po->tp_hdrlen) + 16;
 	} else {
 		unsigned maclen = skb_network_offset(skb);
-		netoff = TPACKET_ALIGN(TPACKET_HDRLEN + (maclen < 16 ? 16 : maclen));
+		netoff = TPACKET_ALIGN(po->tp_hdrlen +
+				       (maclen < 16 ? 16 : maclen));
 		macoff = netoff - maclen;
 	}
 
@@ -616,9 +662,8 @@ static int tpacket_rcv(struct sk_buff *skb, struct net_device *dev, struct packe
 	}
 
 	spin_lock(&sk->sk_receive_queue.lock);
-	h = packet_lookup_frame(po, po->head);
-
-	if (h->tp_status)
+	h.raw = packet_lookup_frame(po, po->head, TP_STATUS_KERNEL);
+	if (!h.raw)
 		goto ring_is_full;
 	po->head = po->head != po->frame_max ? po->head+1 : 0;
 	po->stats.tp_packets++;
@@ -630,20 +675,40 @@ static int tpacket_rcv(struct sk_buff *skb, struct net_device *dev, struct packe
 		status &= ~TP_STATUS_LOSING;
 	spin_unlock(&sk->sk_receive_queue.lock);
 
-	skb_copy_bits(skb, 0, (u8*)h + macoff, snaplen);
+	skb_copy_bits(skb, 0, h.raw + macoff, snaplen);
 
-	h->tp_len = skb->len;
-	h->tp_snaplen = snaplen;
-	h->tp_mac = macoff;
-	h->tp_net = netoff;
-	if (skb->tstamp.tv64)
-		tv = ktime_to_timeval(skb->tstamp);
-	else
-		do_gettimeofday(&tv);
-	h->tp_sec = tv.tv_sec;
-	h->tp_usec = tv.tv_usec;
+	switch (po->tp_version) {
+	case TPACKET_V1:
+		h.h1->tp_len = skb->len;
+		h.h1->tp_snaplen = snaplen;
+		h.h1->tp_mac = macoff;
+		h.h1->tp_net = netoff;
+		if (skb->tstamp.tv64)
+			tv = ktime_to_timeval(skb->tstamp);
+		else
+			do_gettimeofday(&tv);
+		h.h1->tp_sec = tv.tv_sec;
+		h.h1->tp_usec = tv.tv_usec;
+		hdrlen = sizeof(*h.h1);
+		break;
+	case TPACKET_V2:
+		h.h2->tp_len = skb->len;
+		h.h2->tp_snaplen = snaplen;
+		h.h2->tp_mac = macoff;
+		h.h2->tp_net = netoff;
+		if (skb->tstamp.tv64)
+			ts = ktime_to_timespec(skb->tstamp);
+		else
+			getnstimeofday(&ts);
+		h.h2->tp_sec = ts.tv_sec;
+		h.h2->tp_nsec = ts.tv_nsec;
+		hdrlen = sizeof(*h.h2);
+		break;
+	default:
+		BUG();
+	}
 
-	sll = (struct sockaddr_ll*)((u8*)h + TPACKET_ALIGN(sizeof(*h)));
+	sll = h.raw + TPACKET_ALIGN(hdrlen);
 	sll->sll_halen = dev_parse_header(skb, sll->sll_addr);
 	sll->sll_family = AF_PACKET;
 	sll->sll_hatype = dev->type;
@@ -654,14 +719,14 @@ static int tpacket_rcv(struct sk_buff *skb, struct net_device *dev, struct packe
 	else
 		sll->sll_ifindex = dev->ifindex;
 
-	h->tp_status = status;
+	__packet_set_status(po, h.raw, status);
 	smp_mb();
 
 	{
 		struct page *p_start, *p_end;
-		u8 *h_end = (u8 *)h + macoff + snaplen - 1;
+		u8 *h_end = h.raw + macoff + snaplen - 1;
 
-		p_start = virt_to_page(h);
+		p_start = virt_to_page(h.raw);
 		p_end = virt_to_page(h_end);
 		while (p_start <= p_end) {
 			flush_dcache_page(p_start);
@@ -1362,6 +1427,25 @@ packet_setsockopt(struct socket *sock, int level, int optname, char __user *optv
 		pkt_sk(sk)->copy_thresh = val;
 		return 0;
 	}
+	case PACKET_VERSION:
+	{
+		int val;
+
+		if (optlen != sizeof(val))
+			return -EINVAL;
+		if (po->pg_vec)
+			return -EBUSY;
+		if (copy_from_user(&val, optval, sizeof(val)))
+			return -EFAULT;
+		switch (val) {
+		case TPACKET_V1:
+		case TPACKET_V2:
+			po->tp_version = val;
+			return 0;
+		default:
+			return -EINVAL;
+		}
+	}
 #endif
 	case PACKET_AUXDATA:
 	{
@@ -1437,6 +1521,31 @@ static int packet_getsockopt(struct socket *sock, int level, int optname,
 
 		data = &val;
 		break;
+#ifdef CONFIG_PACKET_MMAP
+	case PACKET_VERSION:
+		if (len > sizeof(int))
+			len = sizeof(int);
+		val = po->tp_version;
+		data = &val;
+		break;
+	case PACKET_HDRLEN:
+		if (len > sizeof(int))
+			len = sizeof(int);
+		if (copy_from_user(&val, optval, len))
+			return -EFAULT;
+		switch (val) {
+		case TPACKET_V1:
+			val = sizeof(struct tpacket_hdr);
+			break;
+		case TPACKET_V2:
+			val = sizeof(struct tpacket2_hdr);
+			break;
+		default:
+			return -EINVAL;
+		}
+		data = &val;
+		break;
+#endif
 	default:
 		return -ENOPROTOOPT;
 	}
@@ -1570,11 +1679,8 @@ static unsigned int packet_poll(struct file * file, struct socket *sock,
 	spin_lock_bh(&sk->sk_receive_queue.lock);
 	if (po->pg_vec) {
 		unsigned last = po->head ? po->head-1 : po->frame_max;
-		struct tpacket_hdr *h;
-
-		h = packet_lookup_frame(po, last);
 
-		if (h->tp_status)
+		if (packet_lookup_frame(po, last, TP_STATUS_USER))
 			mask |= POLLIN | POLLRDNORM;
 	}
 	spin_unlock_bh(&sk->sk_receive_queue.lock);
@@ -1669,11 +1775,20 @@ static int packet_set_ring(struct sock *sk, struct tpacket_req *req, int closing
 		if (unlikely(po->pg_vec))
 			return -EBUSY;
 
+		switch (po->tp_version) {
+		case TPACKET_V1:
+			po->tp_hdrlen = TPACKET_HDRLEN;
+			break;
+		case TPACKET_V2:
+			po->tp_hdrlen = TPACKET2_HDRLEN;
+			break;
+		}
+
 		if (unlikely((int)req->tp_block_size <= 0))
 			return -EINVAL;
 		if (unlikely(req->tp_block_size & (PAGE_SIZE - 1)))
 			return -EINVAL;
-		if (unlikely(req->tp_frame_size < TPACKET_HDRLEN))
+		if (unlikely(req->tp_frame_size < po->tp_hdrlen))
 			return -EINVAL;
 		if (unlikely(req->tp_frame_size & (TPACKET_ALIGNMENT - 1)))
 			return -EINVAL;
@@ -1692,13 +1807,11 @@ static int packet_set_ring(struct sock *sk, struct tpacket_req *req, int closing
 			goto out;
 
 		for (i = 0; i < req->tp_block_nr; i++) {
-			char *ptr = pg_vec[i];
-			struct tpacket_hdr *header;
+			void *ptr = pg_vec[i];
 			int k;
 
 			for (k = 0; k < po->frames_per_block; k++) {
-				header = (struct tpacket_hdr *) ptr;
-				header->tp_status = TP_STATUS_KERNEL;
+				__packet_set_status(po, ptr, TP_STATUS_KERNEL);
 				ptr += req->tp_frame_size;
 			}
 		}
-- 
cgit v1.2.3


From 393e52e33c6c26ec7db290dab803bac1bed962d4 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Mon, 14 Jul 2008 22:50:39 -0700
Subject: packet: deliver VLAN TCI to userspace

Store the VLAN tag in the auxillary data/tpacket2_hdr so userspace can
properly deal with hardware VLAN tagging/stripping.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/if_packet.h | 2 ++
 net/packet/af_packet.c    | 2 ++
 2 files changed, 4 insertions(+)

(limited to 'include')

diff --git a/include/linux/if_packet.h b/include/linux/if_packet.h
index d4d3c82448f5..a630295b255f 100644
--- a/include/linux/if_packet.h
+++ b/include/linux/if_packet.h
@@ -59,6 +59,7 @@ struct tpacket_auxdata
 	__u32		tp_snaplen;
 	__u16		tp_mac;
 	__u16		tp_net;
+	__u16		tp_vlan_tci;
 };
 
 struct tpacket_hdr
@@ -90,6 +91,7 @@ struct tpacket2_hdr
 	__u16		tp_net;
 	__u32		tp_sec;
 	__u32		tp_nsec;
+	__u16		tp_vlan_tci;
 };
 
 #define TPACKET2_HDRLEN		(TPACKET_ALIGN(sizeof(struct tpacket2_hdr)) + sizeof(struct sockaddr_ll))
diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c
index 4f059775d48f..db792e02a37f 100644
--- a/net/packet/af_packet.c
+++ b/net/packet/af_packet.c
@@ -702,6 +702,7 @@ static int tpacket_rcv(struct sk_buff *skb, struct net_device *dev, struct packe
 			getnstimeofday(&ts);
 		h.h2->tp_sec = ts.tv_sec;
 		h.h2->tp_nsec = ts.tv_nsec;
+		h.h2->tp_vlan_tci = skb->vlan_tci;
 		hdrlen = sizeof(*h.h2);
 		break;
 	default:
@@ -1172,6 +1173,7 @@ static int packet_recvmsg(struct kiocb *iocb, struct socket *sock,
 		aux.tp_snaplen = skb->len;
 		aux.tp_mac = 0;
 		aux.tp_net = skb_network_offset(skb);
+		aux.tp_vlan_tci = skb->vlan_tci;
 
 		put_cmsg(msg, SOL_PACKET, PACKET_AUXDATA, sizeof(aux), &aux);
 	}
-- 
cgit v1.2.3


From 0388b0042624714e6f8db8cc7994101a0a02d392 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Mon, 14 Jul 2008 23:00:43 -0700
Subject: icmp: add struct net argument to icmp_out_count

This routine deals with ICMP statistics, but doesn't have a
struct net at hands, so add one.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/icmp.h   | 3 ++-
 net/ipv4/icmp.c      | 2 +-
 net/ipv4/ip_output.c | 3 ++-
 net/ipv4/raw.c       | 3 ++-
 4 files changed, 7 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/include/net/icmp.h b/include/net/icmp.h
index dddb839ff4b5..38ca2f39db70 100644
--- a/include/net/icmp.h
+++ b/include/net/icmp.h
@@ -44,12 +44,13 @@ DECLARE_SNMP_STAT(struct icmpmsg_mib, icmpmsg_statistics);
 struct dst_entry;
 struct net_proto_family;
 struct sk_buff;
+struct net;
 
 extern void	icmp_send(struct sk_buff *skb_in,  int type, int code, __be32 info);
 extern int	icmp_rcv(struct sk_buff *skb);
 extern int	icmp_ioctl(struct sock *sk, int cmd, unsigned long arg);
 extern int	icmp_init(void);
-extern void	icmp_out_count(unsigned char type);
+extern void	icmp_out_count(struct net *net, unsigned char type);
 
 /* Move into dst.h ? */
 extern int 	xrlim_allow(struct dst_entry *dst, int timeout);
diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c
index aa7cf46853b7..1ffe7add4920 100644
--- a/net/ipv4/icmp.c
+++ b/net/ipv4/icmp.c
@@ -296,7 +296,7 @@ out:
 /*
  *	Maintain the counters used in the SNMP statistics for outgoing ICMP
  */
-void icmp_out_count(unsigned char type)
+void icmp_out_count(struct net *net, unsigned char type)
 {
 	ICMPMSGOUT_INC_STATS(type);
 	ICMP_INC_STATS(ICMP_MIB_OUTMSGS);
diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c
index f1278eecf56d..f003186b93b1 100644
--- a/net/ipv4/ip_output.c
+++ b/net/ipv4/ip_output.c
@@ -1211,6 +1211,7 @@ int ip_push_pending_frames(struct sock *sk)
 	struct sk_buff *skb, *tmp_skb;
 	struct sk_buff **tail_skb;
 	struct inet_sock *inet = inet_sk(sk);
+	struct net *net = sock_net(sk);
 	struct ip_options *opt = NULL;
 	struct rtable *rt = (struct rtable *)inet->cork.dst;
 	struct iphdr *iph;
@@ -1280,7 +1281,7 @@ int ip_push_pending_frames(struct sock *sk)
 	skb->dst = dst_clone(&rt->u.dst);
 
 	if (iph->protocol == IPPROTO_ICMP)
-		icmp_out_count(((struct icmphdr *)
+		icmp_out_count(net, ((struct icmphdr *)
 			skb_transport_header(skb))->type);
 
 	/* Netfilter gets whole the not fragmented skb. */
diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c
index 925fdf18cf92..7f39ea443ec8 100644
--- a/net/ipv4/raw.c
+++ b/net/ipv4/raw.c
@@ -320,6 +320,7 @@ static int raw_send_hdrinc(struct sock *sk, void *from, size_t length,
 			unsigned int flags)
 {
 	struct inet_sock *inet = inet_sk(sk);
+	struct net *net = sock_net(sk);
 	struct iphdr *iph;
 	struct sk_buff *skb;
 	unsigned int iphlen;
@@ -368,7 +369,7 @@ static int raw_send_hdrinc(struct sock *sk, void *from, size_t length,
 		iph->check = ip_fast_csum((unsigned char *)iph, iph->ihl);
 	}
 	if (iph->protocol == IPPROTO_ICMP)
-		icmp_out_count(((struct icmphdr *)
+		icmp_out_count(net, ((struct icmphdr *)
 			skb_transport_header(skb))->type);
 
 	err = NF_HOOK(PF_INET, NF_INET_LOCAL_OUT, skb, NULL, rt->u.dst.dev,
-- 
cgit v1.2.3


From 43589aa93c9ce49f0e87a280014a4dd254b37fcf Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Mon, 14 Jul 2008 23:02:09 -0700
Subject: icmp: drop unused MIB accounting wrappers

There are ICMP_XXX_STATS that are not used in the kernel, so I remove
them, not to "just patch" them later. But if there's some sense in
keeping them, kick me - I will remake this set keeping them.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/icmp.h | 5 -----
 1 file changed, 5 deletions(-)

(limited to 'include')

diff --git a/include/net/icmp.h b/include/net/icmp.h
index 38ca2f39db70..e34e981fe742 100644
--- a/include/net/icmp.h
+++ b/include/net/icmp.h
@@ -33,13 +33,8 @@ DECLARE_SNMP_STAT(struct icmp_mib, icmp_statistics);
 DECLARE_SNMP_STAT(struct icmpmsg_mib, icmpmsg_statistics);
 #define ICMP_INC_STATS(field)		SNMP_INC_STATS(icmp_statistics, field)
 #define ICMP_INC_STATS_BH(field)	SNMP_INC_STATS_BH(icmp_statistics, field)
-#define ICMP_INC_STATS_USER(field) 	SNMP_INC_STATS_USER(icmp_statistics, field)
 #define ICMPMSGOUT_INC_STATS(field)	SNMP_INC_STATS(icmpmsg_statistics, field+256)
-#define ICMPMSGOUT_INC_STATS_BH(field)	SNMP_INC_STATS_BH(icmpmsg_statistics, field+256)
-#define ICMPMSGOUT_INC_STATS_USER(field) 	SNMP_INC_STATS_USER(icmpmsg_statistics, field+256)
-#define ICMPMSGIN_INC_STATS(field)	SNMP_INC_STATS(icmpmsg_statistics, field)
 #define ICMPMSGIN_INC_STATS_BH(field)	SNMP_INC_STATS_BH(icmpmsg_statistics, field)
-#define ICMPMSGIN_INC_STATS_USER(field) SNMP_INC_STATS_USER(icmpmsg_statistics, field)
 
 struct dst_entry;
 struct net_proto_family;
-- 
cgit v1.2.3


From 75c939bb4d6da790f758a2a3dcc7432f6d8778ee Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Mon, 14 Jul 2008 23:02:35 -0700
Subject: mib: add struct net to ICMP_INC_STATS

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/icmp.h | 2 +-
 net/ipv4/icmp.c    | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/net/icmp.h b/include/net/icmp.h
index e34e981fe742..c86fef8c45e5 100644
--- a/include/net/icmp.h
+++ b/include/net/icmp.h
@@ -31,7 +31,7 @@ struct icmp_err {
 extern struct icmp_err icmp_err_convert[];
 DECLARE_SNMP_STAT(struct icmp_mib, icmp_statistics);
 DECLARE_SNMP_STAT(struct icmpmsg_mib, icmpmsg_statistics);
-#define ICMP_INC_STATS(field)		SNMP_INC_STATS(icmp_statistics, field)
+#define ICMP_INC_STATS(net, field)	SNMP_INC_STATS(icmp_statistics, field)
 #define ICMP_INC_STATS_BH(field)	SNMP_INC_STATS_BH(icmp_statistics, field)
 #define ICMPMSGOUT_INC_STATS(field)	SNMP_INC_STATS(icmpmsg_statistics, field+256)
 #define ICMPMSGIN_INC_STATS_BH(field)	SNMP_INC_STATS_BH(icmpmsg_statistics, field)
diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c
index 56d6b943345c..e94de411cfa0 100644
--- a/net/ipv4/icmp.c
+++ b/net/ipv4/icmp.c
@@ -299,7 +299,7 @@ out:
 void icmp_out_count(struct net *net, unsigned char type)
 {
 	ICMPMSGOUT_INC_STATS(type);
-	ICMP_INC_STATS(ICMP_MIB_OUTMSGS);
+	ICMP_INC_STATS(net, ICMP_MIB_OUTMSGS);
 }
 
 /*
-- 
cgit v1.2.3


From dcfc23cac103b54dbc00a6f52f47656ad5c75844 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Mon, 14 Jul 2008 23:03:00 -0700
Subject: mib: add struct net to ICMP_INC_STATS_BH

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/icmp.h  |  2 +-
 net/dccp/ipv4.c     |  4 ++--
 net/ipv4/icmp.c     | 10 +++++-----
 net/ipv4/tcp_ipv4.c |  4 ++--
 net/ipv4/udp.c      |  2 +-
 net/sctp/input.c    |  5 +++--
 6 files changed, 14 insertions(+), 13 deletions(-)

(limited to 'include')

diff --git a/include/net/icmp.h b/include/net/icmp.h
index c86fef8c45e5..e2ae6409ff80 100644
--- a/include/net/icmp.h
+++ b/include/net/icmp.h
@@ -32,7 +32,7 @@ extern struct icmp_err icmp_err_convert[];
 DECLARE_SNMP_STAT(struct icmp_mib, icmp_statistics);
 DECLARE_SNMP_STAT(struct icmpmsg_mib, icmpmsg_statistics);
 #define ICMP_INC_STATS(net, field)	SNMP_INC_STATS(icmp_statistics, field)
-#define ICMP_INC_STATS_BH(field)	SNMP_INC_STATS_BH(icmp_statistics, field)
+#define ICMP_INC_STATS_BH(net, field)	SNMP_INC_STATS_BH(icmp_statistics, field)
 #define ICMPMSGOUT_INC_STATS(field)	SNMP_INC_STATS(icmpmsg_statistics, field+256)
 #define ICMPMSGIN_INC_STATS_BH(field)	SNMP_INC_STATS_BH(icmpmsg_statistics, field)
 
diff --git a/net/dccp/ipv4.c b/net/dccp/ipv4.c
index cc4f33461f2d..2b03c47cab1b 100644
--- a/net/dccp/ipv4.c
+++ b/net/dccp/ipv4.c
@@ -208,7 +208,7 @@ static void dccp_v4_err(struct sk_buff *skb, u32 info)
 	struct net *net = dev_net(skb->dev);
 
 	if (skb->len < (iph->ihl << 2) + 8) {
-		ICMP_INC_STATS_BH(ICMP_MIB_INERRORS);
+		ICMP_INC_STATS_BH(net, ICMP_MIB_INERRORS);
 		return;
 	}
 
@@ -216,7 +216,7 @@ static void dccp_v4_err(struct sk_buff *skb, u32 info)
 			iph->daddr, dh->dccph_dport,
 			iph->saddr, dh->dccph_sport, inet_iif(skb));
 	if (sk == NULL) {
-		ICMP_INC_STATS_BH(ICMP_MIB_INERRORS);
+		ICMP_INC_STATS_BH(net, ICMP_MIB_INERRORS);
 		return;
 	}
 
diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c
index e94de411cfa0..56a7bbc7950c 100644
--- a/net/ipv4/icmp.c
+++ b/net/ipv4/icmp.c
@@ -763,7 +763,7 @@ static void icmp_unreach(struct sk_buff *skb)
 out:
 	return;
 out_err:
-	ICMP_INC_STATS_BH(ICMP_MIB_INERRORS);
+	ICMP_INC_STATS_BH(net, ICMP_MIB_INERRORS);
 	goto out;
 }
 
@@ -803,7 +803,7 @@ static void icmp_redirect(struct sk_buff *skb)
 out:
 	return;
 out_err:
-	ICMP_INC_STATS_BH(ICMP_MIB_INERRORS);
+	ICMP_INC_STATS_BH(dev_net(skb->dev), ICMP_MIB_INERRORS);
 	goto out;
 }
 
@@ -874,7 +874,7 @@ static void icmp_timestamp(struct sk_buff *skb)
 out:
 	return;
 out_err:
-	ICMP_INC_STATS_BH(ICMP_MIB_INERRORS);
+	ICMP_INC_STATS_BH(dev_net(skb->dst->dev), ICMP_MIB_INERRORS);
 	goto out;
 }
 
@@ -994,7 +994,7 @@ int icmp_rcv(struct sk_buff *skb)
 		skb_set_network_header(skb, nh);
 	}
 
-	ICMP_INC_STATS_BH(ICMP_MIB_INMSGS);
+	ICMP_INC_STATS_BH(net, ICMP_MIB_INMSGS);
 
 	switch (skb->ip_summed) {
 	case CHECKSUM_COMPLETE:
@@ -1053,7 +1053,7 @@ drop:
 	kfree_skb(skb);
 	return 0;
 error:
-	ICMP_INC_STATS_BH(ICMP_MIB_INERRORS);
+	ICMP_INC_STATS_BH(net, ICMP_MIB_INERRORS);
 	goto drop;
 }
 
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index ca41b77f3f32..0fefd4409413 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -346,14 +346,14 @@ void tcp_v4_err(struct sk_buff *skb, u32 info)
 	struct net *net = dev_net(skb->dev);
 
 	if (skb->len < (iph->ihl << 2) + 8) {
-		ICMP_INC_STATS_BH(ICMP_MIB_INERRORS);
+		ICMP_INC_STATS_BH(net, ICMP_MIB_INERRORS);
 		return;
 	}
 
 	sk = inet_lookup(net, &tcp_hashinfo, iph->daddr, th->dest,
 			iph->saddr, th->source, inet_iif(skb));
 	if (!sk) {
-		ICMP_INC_STATS_BH(ICMP_MIB_INERRORS);
+		ICMP_INC_STATS_BH(net, ICMP_MIB_INERRORS);
 		return;
 	}
 	if (sk->sk_state == TCP_TIME_WAIT) {
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index 9342cfda3d0a..fdd4faa1f120 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -359,7 +359,7 @@ void __udp4_lib_err(struct sk_buff *skb, u32 info, struct hlist_head udptable[])
 	sk = __udp4_lib_lookup(net, iph->daddr, uh->dest,
 			iph->saddr, uh->source, skb->dev->ifindex, udptable);
 	if (sk == NULL) {
-		ICMP_INC_STATS_BH(ICMP_MIB_INERRORS);
+		ICMP_INC_STATS_BH(net, ICMP_MIB_INERRORS);
 		return;	/* No socket for error */
 	}
 
diff --git a/net/sctp/input.c b/net/sctp/input.c
index d354a23972d4..ed8834e7f143 100644
--- a/net/sctp/input.c
+++ b/net/sctp/input.c
@@ -61,6 +61,7 @@
 #include <net/sctp/sctp.h>
 #include <net/sctp/sm.h>
 #include <net/sctp/checksum.h>
+#include <net/net_namespace.h>
 
 /* Forward declarations for internal helpers. */
 static int sctp_rcv_ootb(struct sk_buff *);
@@ -534,7 +535,7 @@ void sctp_v4_err(struct sk_buff *skb, __u32 info)
 	int err;
 
 	if (skb->len < ihlen + 8) {
-		ICMP_INC_STATS_BH(ICMP_MIB_INERRORS);
+		ICMP_INC_STATS_BH(&init_net, ICMP_MIB_INERRORS);
 		return;
 	}
 
@@ -548,7 +549,7 @@ void sctp_v4_err(struct sk_buff *skb, __u32 info)
 	skb->network_header = saveip;
 	skb->transport_header = savesctp;
 	if (!sk) {
-		ICMP_INC_STATS_BH(ICMP_MIB_INERRORS);
+		ICMP_INC_STATS_BH(&init_net, ICMP_MIB_INERRORS);
 		return;
 	}
 	/* Warning:  The sock lock is held.  Remember to call
-- 
cgit v1.2.3


From 903fc1964e746b8d8e2971ea20c89b7aeab8bd9a Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Mon, 14 Jul 2008 23:03:35 -0700
Subject: mib: add struct net to ICMPMSGOUT_INC_STATS

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/icmp.h | 2 +-
 net/ipv4/icmp.c    | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/net/icmp.h b/include/net/icmp.h
index e2ae6409ff80..ee9c8d4b7ba4 100644
--- a/include/net/icmp.h
+++ b/include/net/icmp.h
@@ -33,7 +33,7 @@ DECLARE_SNMP_STAT(struct icmp_mib, icmp_statistics);
 DECLARE_SNMP_STAT(struct icmpmsg_mib, icmpmsg_statistics);
 #define ICMP_INC_STATS(net, field)	SNMP_INC_STATS(icmp_statistics, field)
 #define ICMP_INC_STATS_BH(net, field)	SNMP_INC_STATS_BH(icmp_statistics, field)
-#define ICMPMSGOUT_INC_STATS(field)	SNMP_INC_STATS(icmpmsg_statistics, field+256)
+#define ICMPMSGOUT_INC_STATS(net, field)	SNMP_INC_STATS(icmpmsg_statistics, field+256)
 #define ICMPMSGIN_INC_STATS_BH(field)	SNMP_INC_STATS_BH(icmpmsg_statistics, field)
 
 struct dst_entry;
diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c
index 56a7bbc7950c..6f4c65b898b4 100644
--- a/net/ipv4/icmp.c
+++ b/net/ipv4/icmp.c
@@ -298,7 +298,7 @@ out:
  */
 void icmp_out_count(struct net *net, unsigned char type)
 {
-	ICMPMSGOUT_INC_STATS(type);
+	ICMPMSGOUT_INC_STATS(net, type);
 	ICMP_INC_STATS(net, ICMP_MIB_OUTMSGS);
 }
 
-- 
cgit v1.2.3


From f66ac03d497c162c70cd0ccc802ce1777073cdf3 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Mon, 14 Jul 2008 23:04:00 -0700
Subject: mib: add struct net to ICMPMSGIN_INC_STATS_BH

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/icmp.h | 2 +-
 net/ipv4/icmp.c    | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/net/icmp.h b/include/net/icmp.h
index ee9c8d4b7ba4..03b9972b8754 100644
--- a/include/net/icmp.h
+++ b/include/net/icmp.h
@@ -34,7 +34,7 @@ DECLARE_SNMP_STAT(struct icmpmsg_mib, icmpmsg_statistics);
 #define ICMP_INC_STATS(net, field)	SNMP_INC_STATS(icmp_statistics, field)
 #define ICMP_INC_STATS_BH(net, field)	SNMP_INC_STATS_BH(icmp_statistics, field)
 #define ICMPMSGOUT_INC_STATS(net, field)	SNMP_INC_STATS(icmpmsg_statistics, field+256)
-#define ICMPMSGIN_INC_STATS_BH(field)	SNMP_INC_STATS_BH(icmpmsg_statistics, field)
+#define ICMPMSGIN_INC_STATS_BH(net, field)	SNMP_INC_STATS_BH(icmpmsg_statistics, field)
 
 struct dst_entry;
 struct net_proto_family;
diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c
index 6f4c65b898b4..ea60ad41008c 100644
--- a/net/ipv4/icmp.c
+++ b/net/ipv4/icmp.c
@@ -1012,7 +1012,7 @@ int icmp_rcv(struct sk_buff *skb)
 
 	icmph = icmp_hdr(skb);
 
-	ICMPMSGIN_INC_STATS_BH(icmph->type);
+	ICMPMSGIN_INC_STATS_BH(net, icmph->type);
 	/*
 	 *	18 is the highest 'known' ICMP type. Anything else is a mystery
 	 *
-- 
cgit v1.2.3


From f1f28aa3510ddb84c966bac65611bb866c77a092 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Tue, 15 Jul 2008 00:08:33 -0700
Subject: netdev: Add addr_list_lock to struct net_device.

This will be used to protect the per-device unicast and multicast
address lists, as well as the callbacks into the drivers which
configure such state such as ->set_rx_mode() and ->set_multicast_list().

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h | 1 +
 net/core/dev.c            | 1 +
 2 files changed, 2 insertions(+)

(limited to 'include')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index ba5c4639ea91..fd0365219181 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -609,6 +609,7 @@ struct net_device
 	unsigned char		addr_len;	/* hardware address length	*/
 	unsigned short          dev_id;		/* for shared network cards */
 
+	spinlock_t		addr_list_lock;
 	struct dev_addr_list	*uc_list;	/* Secondary unicast mac addresses */
 	int			uc_count;	/* Number of installed ucasts	*/
 	int			uc_promisc;
diff --git a/net/core/dev.c b/net/core/dev.c
index feaab4898a5b..d933d1bfa6fa 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -3836,6 +3836,7 @@ int register_netdevice(struct net_device *dev)
 	BUG_ON(!dev_net(dev));
 	net = dev_net(dev);
 
+	spin_lock_init(&dev->addr_list_lock);
 	netdev_init_queue_locks(dev);
 
 	dev->iflink = -1;
-- 
cgit v1.2.3


From e308a5d806c852f56590ffdd3834d0df0cbed8d7 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Tue, 15 Jul 2008 00:13:44 -0700
Subject: netdev: Add netdev->addr_list_lock protection.

Add netif_addr_{lock,unlock}{,_bh}() helpers.

Use them to protect operations that operate on or read
the network device unicast and multicast address lists.

Also use them in cases where the code simply wants to
block calls into the driver's ->set_rx_mode() and
->set_multicast_list() methods.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/infiniband/ulp/ipoib/ipoib_multicast.c |  2 ++
 drivers/media/dvb/dvb-core/dvb_net.c           |  2 ++
 drivers/net/bonding/bond_main.c                |  8 ++++++++
 drivers/net/forcedeth.c                        | 16 ++++++++++++++++
 drivers/net/hamradio/6pack.c                   |  2 ++
 drivers/net/hamradio/mkiss.c                   |  2 ++
 drivers/net/ibm_newemac/core.c                 |  4 ++++
 drivers/net/sfc/efx.c                          |  2 ++
 drivers/net/wireless/libertas/main.c           |  2 ++
 include/linux/netdevice.h                      | 20 ++++++++++++++++++++
 net/core/dev.c                                 | 14 ++++++++++++++
 net/core/dev_mcast.c                           | 12 ++++++++++++
 net/mac80211/main.c                            |  4 ++++
 net/mac80211/mlme.c                            |  4 ++++
 14 files changed, 94 insertions(+)

(limited to 'include')

diff --git a/drivers/infiniband/ulp/ipoib/ipoib_multicast.c b/drivers/infiniband/ulp/ipoib/ipoib_multicast.c
index 3f663fb852c1..261ab7150431 100644
--- a/drivers/infiniband/ulp/ipoib/ipoib_multicast.c
+++ b/drivers/infiniband/ulp/ipoib/ipoib_multicast.c
@@ -775,6 +775,7 @@ void ipoib_mcast_restart_task(struct work_struct *work)
 
 	local_irq_save(flags);
 	netif_tx_lock(dev);
+	netif_addr_lock(dev);
 	spin_lock(&priv->lock);
 
 	/*
@@ -851,6 +852,7 @@ void ipoib_mcast_restart_task(struct work_struct *work)
 	}
 
 	spin_unlock(&priv->lock);
+	netif_addr_unlock(dev);
 	netif_tx_unlock(dev);
 	local_irq_restore(flags);
 
diff --git a/drivers/media/dvb/dvb-core/dvb_net.c b/drivers/media/dvb/dvb-core/dvb_net.c
index c2334aef4143..809d18c663bc 100644
--- a/drivers/media/dvb/dvb-core/dvb_net.c
+++ b/drivers/media/dvb/dvb-core/dvb_net.c
@@ -1134,6 +1134,7 @@ static void wq_set_multicast_list (struct work_struct *work)
 	dvb_net_feed_stop(dev);
 	priv->rx_mode = RX_MODE_UNI;
 	netif_tx_lock_bh(dev);
+	netif_addr_lock(dev);
 
 	if (dev->flags & IFF_PROMISC) {
 		dprintk("%s: promiscuous mode\n", dev->name);
@@ -1158,6 +1159,7 @@ static void wq_set_multicast_list (struct work_struct *work)
 		}
 	}
 
+	netif_addr_unlock(dev);
 	netif_tx_unlock_bh(dev);
 	dvb_net_feed_start(dev);
 }
diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
index 8ae7ff313218..ea71abd6f728 100644
--- a/drivers/net/bonding/bond_main.c
+++ b/drivers/net/bonding/bond_main.c
@@ -1568,10 +1568,12 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev)
 		}
 
 		netif_tx_lock_bh(bond_dev);
+		netif_addr_lock(bond_dev);
 		/* upload master's mc_list to new slave */
 		for (dmi = bond_dev->mc_list; dmi; dmi = dmi->next) {
 			dev_mc_add (slave_dev, dmi->dmi_addr, dmi->dmi_addrlen, 0);
 		}
+		netif_addr_unlock(bond_dev);
 		netif_tx_unlock_bh(bond_dev);
 	}
 
@@ -1937,7 +1939,9 @@ int bond_release(struct net_device *bond_dev, struct net_device *slave_dev)
 
 		/* flush master's mc_list from slave */
 		netif_tx_lock_bh(bond_dev);
+		netif_addr_lock(bond_dev);
 		bond_mc_list_flush(bond_dev, slave_dev);
+		netif_addr_unlock(bond_dev);
 		netif_tx_unlock_bh(bond_dev);
 	}
 
@@ -2060,7 +2064,9 @@ static int bond_release_all(struct net_device *bond_dev)
 
 			/* flush master's mc_list from slave */
 			netif_tx_lock_bh(bond_dev);
+			netif_addr_lock(bond_dev);
 			bond_mc_list_flush(bond_dev, slave_dev);
+			netif_addr_unlock(bond_dev);
 			netif_tx_unlock_bh(bond_dev);
 		}
 
@@ -4674,7 +4680,9 @@ static void bond_free_all(void)
 
 		bond_work_cancel_all(bond);
 		netif_tx_lock_bh(bond_dev);
+		netif_addr_lock(bond_dev);
 		bond_mc_list_destroy(bond);
+		netif_addr_unlock(bond_dev);
 		netif_tx_unlock_bh(bond_dev);
 		/* Release the bonded slaves */
 		bond_release_all(bond_dev);
diff --git a/drivers/net/forcedeth.c b/drivers/net/forcedeth.c
index 786d668c612e..4ed89fa9ae46 100644
--- a/drivers/net/forcedeth.c
+++ b/drivers/net/forcedeth.c
@@ -2831,6 +2831,7 @@ static int nv_change_mtu(struct net_device *dev, int new_mtu)
 		 */
 		nv_disable_irq(dev);
 		netif_tx_lock_bh(dev);
+		netif_addr_lock(dev);
 		spin_lock(&np->lock);
 		/* stop engines */
 		nv_stop_rxtx(dev);
@@ -2855,6 +2856,7 @@ static int nv_change_mtu(struct net_device *dev, int new_mtu)
 		/* restart rx engine */
 		nv_start_rxtx(dev);
 		spin_unlock(&np->lock);
+		netif_addr_unlock(dev);
 		netif_tx_unlock_bh(dev);
 		nv_enable_irq(dev);
 	}
@@ -2891,6 +2893,7 @@ static int nv_set_mac_address(struct net_device *dev, void *addr)
 
 	if (netif_running(dev)) {
 		netif_tx_lock_bh(dev);
+		netif_addr_lock(dev);
 		spin_lock_irq(&np->lock);
 
 		/* stop rx engine */
@@ -2902,6 +2905,7 @@ static int nv_set_mac_address(struct net_device *dev, void *addr)
 		/* restart rx engine */
 		nv_start_rx(dev);
 		spin_unlock_irq(&np->lock);
+		netif_addr_unlock(dev);
 		netif_tx_unlock_bh(dev);
 	} else {
 		nv_copy_mac_to_hw(dev);
@@ -3971,6 +3975,7 @@ static void nv_do_nic_poll(unsigned long data)
 		printk(KERN_INFO "forcedeth: MAC in recoverable error state\n");
 		if (netif_running(dev)) {
 			netif_tx_lock_bh(dev);
+			netif_addr_lock(dev);
 			spin_lock(&np->lock);
 			/* stop engines */
 			nv_stop_rxtx(dev);
@@ -3995,6 +4000,7 @@ static void nv_do_nic_poll(unsigned long data)
 			/* restart rx engine */
 			nv_start_rxtx(dev);
 			spin_unlock(&np->lock);
+			netif_addr_unlock(dev);
 			netif_tx_unlock_bh(dev);
 		}
 	}
@@ -4202,6 +4208,7 @@ static int nv_set_settings(struct net_device *dev, struct ethtool_cmd *ecmd)
 
 		nv_disable_irq(dev);
 		netif_tx_lock_bh(dev);
+		netif_addr_lock(dev);
 		/* with plain spinlock lockdep complains */
 		spin_lock_irqsave(&np->lock, flags);
 		/* stop engines */
@@ -4215,6 +4222,7 @@ static int nv_set_settings(struct net_device *dev, struct ethtool_cmd *ecmd)
 		 */
 		nv_stop_rxtx(dev);
 		spin_unlock_irqrestore(&np->lock, flags);
+		netif_addr_unlock(dev);
 		netif_tx_unlock_bh(dev);
 	}
 
@@ -4360,10 +4368,12 @@ static int nv_nway_reset(struct net_device *dev)
 		if (netif_running(dev)) {
 			nv_disable_irq(dev);
 			netif_tx_lock_bh(dev);
+			netif_addr_lock(dev);
 			spin_lock(&np->lock);
 			/* stop engines */
 			nv_stop_rxtx(dev);
 			spin_unlock(&np->lock);
+			netif_addr_unlock(dev);
 			netif_tx_unlock_bh(dev);
 			printk(KERN_INFO "%s: link down.\n", dev->name);
 		}
@@ -4471,6 +4481,7 @@ static int nv_set_ringparam(struct net_device *dev, struct ethtool_ringparam* ri
 	if (netif_running(dev)) {
 		nv_disable_irq(dev);
 		netif_tx_lock_bh(dev);
+		netif_addr_lock(dev);
 		spin_lock(&np->lock);
 		/* stop engines */
 		nv_stop_rxtx(dev);
@@ -4519,6 +4530,7 @@ static int nv_set_ringparam(struct net_device *dev, struct ethtool_ringparam* ri
 		/* restart engines */
 		nv_start_rxtx(dev);
 		spin_unlock(&np->lock);
+		netif_addr_unlock(dev);
 		netif_tx_unlock_bh(dev);
 		nv_enable_irq(dev);
 	}
@@ -4556,10 +4568,12 @@ static int nv_set_pauseparam(struct net_device *dev, struct ethtool_pauseparam*
 	if (netif_running(dev)) {
 		nv_disable_irq(dev);
 		netif_tx_lock_bh(dev);
+		netif_addr_lock(dev);
 		spin_lock(&np->lock);
 		/* stop engines */
 		nv_stop_rxtx(dev);
 		spin_unlock(&np->lock);
+		netif_addr_unlock(dev);
 		netif_tx_unlock_bh(dev);
 	}
 
@@ -4946,6 +4960,7 @@ static void nv_self_test(struct net_device *dev, struct ethtool_test *test, u64
 			napi_disable(&np->napi);
 #endif
 			netif_tx_lock_bh(dev);
+			netif_addr_lock(dev);
 			spin_lock_irq(&np->lock);
 			nv_disable_hw_interrupts(dev, np->irqmask);
 			if (!(np->msi_flags & NV_MSI_X_ENABLED)) {
@@ -4959,6 +4974,7 @@ static void nv_self_test(struct net_device *dev, struct ethtool_test *test, u64
 			/* drain rx queue */
 			nv_drain_rxtx(dev);
 			spin_unlock_irq(&np->lock);
+			netif_addr_unlock(dev);
 			netif_tx_unlock_bh(dev);
 		}
 
diff --git a/drivers/net/hamradio/6pack.c b/drivers/net/hamradio/6pack.c
index 06ad9f302b5a..ffc937f5d15d 100644
--- a/drivers/net/hamradio/6pack.c
+++ b/drivers/net/hamradio/6pack.c
@@ -300,7 +300,9 @@ static int sp_set_mac_address(struct net_device *dev, void *addr)
 	struct sockaddr_ax25 *sa = addr;
 
 	netif_tx_lock_bh(dev);
+	netif_addr_lock(dev);
 	memcpy(dev->dev_addr, &sa->sax25_call, AX25_ADDR_LEN);
+	netif_addr_unlock(dev);
 	netif_tx_unlock_bh(dev);
 
 	return 0;
diff --git a/drivers/net/hamradio/mkiss.c b/drivers/net/hamradio/mkiss.c
index 65166035aca0..b8740e6a5cec 100644
--- a/drivers/net/hamradio/mkiss.c
+++ b/drivers/net/hamradio/mkiss.c
@@ -356,7 +356,9 @@ static int ax_set_mac_address(struct net_device *dev, void *addr)
 	struct sockaddr_ax25 *sa = addr;
 
 	netif_tx_lock_bh(dev);
+	netif_addr_lock(dev);
 	memcpy(dev->dev_addr, &sa->sax25_call, AX25_ADDR_LEN);
+	netif_addr_unlock(dev);
 	netif_tx_unlock_bh(dev);
 
 	return 0;
diff --git a/drivers/net/ibm_newemac/core.c b/drivers/net/ibm_newemac/core.c
index babc79ad490b..9ca57d365599 100644
--- a/drivers/net/ibm_newemac/core.c
+++ b/drivers/net/ibm_newemac/core.c
@@ -295,7 +295,9 @@ static void emac_rx_disable(struct emac_instance *dev)
 static inline void emac_netif_stop(struct emac_instance *dev)
 {
 	netif_tx_lock_bh(dev->ndev);
+	netif_addr_lock(dev->ndev);
 	dev->no_mcast = 1;
+	netif_addr_unlock(dev->ndev);
 	netif_tx_unlock_bh(dev->ndev);
 	dev->ndev->trans_start = jiffies;	/* prevent tx timeout */
 	mal_poll_disable(dev->mal, &dev->commac);
@@ -305,9 +307,11 @@ static inline void emac_netif_stop(struct emac_instance *dev)
 static inline void emac_netif_start(struct emac_instance *dev)
 {
 	netif_tx_lock_bh(dev->ndev);
+	netif_addr_lock(dev->ndev);
 	dev->no_mcast = 0;
 	if (dev->mcast_pending && netif_running(dev->ndev))
 		__emac_set_multicast_list(dev);
+	netif_addr_unlock(dev->ndev);
 	netif_tx_unlock_bh(dev->ndev);
 
 	netif_wake_queue(dev->ndev);
diff --git a/drivers/net/sfc/efx.c b/drivers/net/sfc/efx.c
index 74265d8553b8..e1257e556e48 100644
--- a/drivers/net/sfc/efx.c
+++ b/drivers/net/sfc/efx.c
@@ -697,6 +697,8 @@ static void efx_stop_port(struct efx_nic *efx)
 	/* Serialise against efx_set_multicast_list() */
 	if (efx_dev_registered(efx)) {
 		netif_tx_lock_bh(efx->net_dev);
+		netif_addr_lock(efx->net_dev);
+		netif_addr_unlock(efx->net_dev);
 		netif_tx_unlock_bh(efx->net_dev);
 	}
 }
diff --git a/drivers/net/wireless/libertas/main.c b/drivers/net/wireless/libertas/main.c
index abd6d9ed8f4b..42e9b2771eab 100644
--- a/drivers/net/wireless/libertas/main.c
+++ b/drivers/net/wireless/libertas/main.c
@@ -594,6 +594,7 @@ static int lbs_add_mcast_addrs(struct cmd_ds_mac_multicast_adr *cmd,
 		return nr_addrs;
 
 	netif_tx_lock_bh(dev);
+	netif_addr_lock(dev);
 	for (mc_list = dev->mc_list; mc_list; mc_list = mc_list->next) {
 		if (mac_in_list(cmd->maclist, nr_addrs, mc_list->dmi_addr)) {
 			lbs_deb_net("mcast address %s:%s skipped\n", dev->name,
@@ -608,6 +609,7 @@ static int lbs_add_mcast_addrs(struct cmd_ds_mac_multicast_adr *cmd,
 			    print_mac(mac, mc_list->dmi_addr));
 		i++;
 	}
+	netif_addr_unlock(dev);
 	netif_tx_unlock_bh(dev);
 	if (mc_list)
 		return -EOVERFLOW;
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index fd0365219181..570cf7affa72 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -1498,6 +1498,26 @@ static inline void netif_tx_disable(struct net_device *dev)
 	netif_tx_unlock_bh(dev);
 }
 
+static inline void netif_addr_lock(struct net_device *dev)
+{
+	spin_lock(&dev->addr_list_lock);
+}
+
+static inline void netif_addr_lock_bh(struct net_device *dev)
+{
+	spin_lock_bh(&dev->addr_list_lock);
+}
+
+static inline void netif_addr_unlock(struct net_device *dev)
+{
+	spin_unlock(&dev->addr_list_lock);
+}
+
+static inline void netif_addr_unlock_bh(struct net_device *dev)
+{
+	spin_unlock_bh(&dev->addr_list_lock);
+}
+
 /* These functions live elsewhere (drivers/net/net_init.c, but related) */
 
 extern void		ether_setup(struct net_device *dev);
diff --git a/net/core/dev.c b/net/core/dev.c
index d933d1bfa6fa..ef1502d71f25 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -2982,7 +2982,9 @@ void __dev_set_rx_mode(struct net_device *dev)
 void dev_set_rx_mode(struct net_device *dev)
 {
 	netif_tx_lock_bh(dev);
+	netif_addr_lock(dev);
 	__dev_set_rx_mode(dev);
+	netif_addr_unlock(dev);
 	netif_tx_unlock_bh(dev);
 }
 
@@ -3062,9 +3064,11 @@ int dev_unicast_delete(struct net_device *dev, void *addr, int alen)
 	ASSERT_RTNL();
 
 	netif_tx_lock_bh(dev);
+	netif_addr_lock(dev);
 	err = __dev_addr_delete(&dev->uc_list, &dev->uc_count, addr, alen, 0);
 	if (!err)
 		__dev_set_rx_mode(dev);
+	netif_addr_unlock(dev);
 	netif_tx_unlock_bh(dev);
 	return err;
 }
@@ -3088,9 +3092,11 @@ int dev_unicast_add(struct net_device *dev, void *addr, int alen)
 	ASSERT_RTNL();
 
 	netif_tx_lock_bh(dev);
+	netif_addr_lock(dev);
 	err = __dev_addr_add(&dev->uc_list, &dev->uc_count, addr, alen, 0);
 	if (!err)
 		__dev_set_rx_mode(dev);
+	netif_addr_unlock(dev);
 	netif_tx_unlock_bh(dev);
 	return err;
 }
@@ -3159,10 +3165,12 @@ int dev_unicast_sync(struct net_device *to, struct net_device *from)
 	int err = 0;
 
 	netif_tx_lock_bh(to);
+	netif_addr_lock(to);
 	err = __dev_addr_sync(&to->uc_list, &to->uc_count,
 			      &from->uc_list, &from->uc_count);
 	if (!err)
 		__dev_set_rx_mode(to);
+	netif_addr_unlock(to);
 	netif_tx_unlock_bh(to);
 	return err;
 }
@@ -3180,13 +3188,17 @@ EXPORT_SYMBOL(dev_unicast_sync);
 void dev_unicast_unsync(struct net_device *to, struct net_device *from)
 {
 	netif_tx_lock_bh(from);
+	netif_addr_lock(from);
 	netif_tx_lock_bh(to);
+	netif_addr_lock(to);
 
 	__dev_addr_unsync(&to->uc_list, &to->uc_count,
 			  &from->uc_list, &from->uc_count);
 	__dev_set_rx_mode(to);
 
+	netif_addr_unlock(to);
 	netif_tx_unlock_bh(to);
+	netif_addr_unlock(from);
 	netif_tx_unlock_bh(from);
 }
 EXPORT_SYMBOL(dev_unicast_unsync);
@@ -3208,6 +3220,7 @@ static void __dev_addr_discard(struct dev_addr_list **list)
 static void dev_addr_discard(struct net_device *dev)
 {
 	netif_tx_lock_bh(dev);
+	netif_addr_lock(dev);
 
 	__dev_addr_discard(&dev->uc_list);
 	dev->uc_count = 0;
@@ -3215,6 +3228,7 @@ static void dev_addr_discard(struct net_device *dev)
 	__dev_addr_discard(&dev->mc_list);
 	dev->mc_count = 0;
 
+	netif_addr_unlock(dev);
 	netif_tx_unlock_bh(dev);
 }
 
diff --git a/net/core/dev_mcast.c b/net/core/dev_mcast.c
index f8a3455f4493..b6b2a129971a 100644
--- a/net/core/dev_mcast.c
+++ b/net/core/dev_mcast.c
@@ -73,6 +73,7 @@ int dev_mc_delete(struct net_device *dev, void *addr, int alen, int glbl)
 	int err;
 
 	netif_tx_lock_bh(dev);
+	netif_addr_lock(dev);
 	err = __dev_addr_delete(&dev->mc_list, &dev->mc_count,
 				addr, alen, glbl);
 	if (!err) {
@@ -83,6 +84,7 @@ int dev_mc_delete(struct net_device *dev, void *addr, int alen, int glbl)
 
 		__dev_set_rx_mode(dev);
 	}
+	netif_addr_unlock(dev);
 	netif_tx_unlock_bh(dev);
 	return err;
 }
@@ -96,9 +98,11 @@ int dev_mc_add(struct net_device *dev, void *addr, int alen, int glbl)
 	int err;
 
 	netif_tx_lock_bh(dev);
+	netif_addr_lock(dev);
 	err = __dev_addr_add(&dev->mc_list, &dev->mc_count, addr, alen, glbl);
 	if (!err)
 		__dev_set_rx_mode(dev);
+	netif_addr_unlock(dev);
 	netif_tx_unlock_bh(dev);
 	return err;
 }
@@ -120,10 +124,12 @@ int dev_mc_sync(struct net_device *to, struct net_device *from)
 	int err = 0;
 
 	netif_tx_lock_bh(to);
+	netif_addr_lock(to);
 	err = __dev_addr_sync(&to->mc_list, &to->mc_count,
 			      &from->mc_list, &from->mc_count);
 	if (!err)
 		__dev_set_rx_mode(to);
+	netif_addr_unlock(to);
 	netif_tx_unlock_bh(to);
 
 	return err;
@@ -144,13 +150,17 @@ EXPORT_SYMBOL(dev_mc_sync);
 void dev_mc_unsync(struct net_device *to, struct net_device *from)
 {
 	netif_tx_lock_bh(from);
+	netif_addr_lock(from);
 	netif_tx_lock_bh(to);
+	netif_addr_lock(to);
 
 	__dev_addr_unsync(&to->mc_list, &to->mc_count,
 			  &from->mc_list, &from->mc_count);
 	__dev_set_rx_mode(to);
 
+	netif_addr_unlock(to);
 	netif_tx_unlock_bh(to);
+	netif_addr_unlock(from);
 	netif_tx_unlock_bh(from);
 }
 EXPORT_SYMBOL(dev_mc_unsync);
@@ -165,6 +175,7 @@ static int dev_mc_seq_show(struct seq_file *seq, void *v)
 		return 0;
 
 	netif_tx_lock_bh(dev);
+	netif_addr_lock(dev);
 	for (m = dev->mc_list; m; m = m->next) {
 		int i;
 
@@ -176,6 +187,7 @@ static int dev_mc_seq_show(struct seq_file *seq, void *v)
 
 		seq_putc(seq, '\n');
 	}
+	netif_addr_unlock(dev);
 	netif_tx_unlock_bh(dev);
 	return 0;
 }
diff --git a/net/mac80211/main.c b/net/mac80211/main.c
index 36859e794928..095b7d928d64 100644
--- a/net/mac80211/main.c
+++ b/net/mac80211/main.c
@@ -292,7 +292,9 @@ static int ieee80211_open(struct net_device *dev)
 			local->fif_other_bss++;
 
 		netif_tx_lock_bh(local->mdev);
+		netif_addr_lock(local->mdev);
 		ieee80211_configure_filter(local);
+		netif_addr_unlock(local->mdev);
 		netif_tx_unlock_bh(local->mdev);
 		break;
 	case IEEE80211_IF_TYPE_STA:
@@ -491,7 +493,9 @@ static int ieee80211_stop(struct net_device *dev)
 			local->fif_other_bss--;
 
 		netif_tx_lock_bh(local->mdev);
+		netif_addr_lock(local->mdev);
 		ieee80211_configure_filter(local);
+		netif_addr_unlock(local->mdev);
 		netif_tx_unlock_bh(local->mdev);
 		break;
 	case IEEE80211_IF_TYPE_MESH_POINT:
diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c
index 8f51375317dd..1232ba25e1e9 100644
--- a/net/mac80211/mlme.c
+++ b/net/mac80211/mlme.c
@@ -3869,6 +3869,7 @@ void ieee80211_scan_completed(struct ieee80211_hw *hw)
 
 
 	netif_tx_lock_bh(local->mdev);
+	netif_addr_lock(local->mdev);
 	local->filter_flags &= ~FIF_BCN_PRBRESP_PROMISC;
 	local->ops->configure_filter(local_to_hw(local),
 				     FIF_BCN_PRBRESP_PROMISC,
@@ -3876,6 +3877,7 @@ void ieee80211_scan_completed(struct ieee80211_hw *hw)
 				     local->mdev->mc_count,
 				     local->mdev->mc_list);
 
+	netif_addr_unlock(local->mdev);
 	netif_tx_unlock_bh(local->mdev);
 
 	rcu_read_lock();
@@ -4063,12 +4065,14 @@ static int ieee80211_sta_start_scan(struct net_device *dev,
 	local->scan_dev = dev;
 
 	netif_tx_lock_bh(local->mdev);
+	netif_addr_lock(local->mdev);
 	local->filter_flags |= FIF_BCN_PRBRESP_PROMISC;
 	local->ops->configure_filter(local_to_hw(local),
 				     FIF_BCN_PRBRESP_PROMISC,
 				     &local->filter_flags,
 				     local->mdev->mc_count,
 				     local->mdev->mc_list);
+	netif_addr_unlock(local->mdev);
 	netif_tx_unlock_bh(local->mdev);
 
 	/* TODO: start scan as soon as all nullfunc frames are ACKed */
-- 
cgit v1.2.3


From c6f8f7e3bb4b2c1886ef3743e8f24521f7a60abc Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Wed, 16 Jul 2008 20:19:26 -0700
Subject: mib: drop unused IP_INC_STATS_USER

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip.h | 1 -
 1 file changed, 1 deletion(-)

(limited to 'include')

diff --git a/include/net/ip.h b/include/net/ip.h
index 3b40bc2234be..673ecdbe72fd 100644
--- a/include/net/ip.h
+++ b/include/net/ip.h
@@ -159,7 +159,6 @@ extern struct ipv4_config ipv4_config;
 DECLARE_SNMP_STAT(struct ipstats_mib, ip_statistics);
 #define IP_INC_STATS(field)		SNMP_INC_STATS(ip_statistics, field)
 #define IP_INC_STATS_BH(field)		SNMP_INC_STATS_BH(ip_statistics, field)
-#define IP_INC_STATS_USER(field) 	SNMP_INC_STATS_USER(ip_statistics, field)
 #define IP_ADD_STATS_BH(field, val)	SNMP_ADD_STATS_BH(ip_statistics, field, val)
 DECLARE_SNMP_STAT(struct linux_mib, net_statistics);
 #define NET_INC_STATS(field)		SNMP_INC_STATS(net_statistics, field)
-- 
cgit v1.2.3


From 5e38e270444f2629de7a706b5a9ca1b333d14517 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Wed, 16 Jul 2008 20:19:49 -0700
Subject: mib: add net to IP_INC_STATS

All the callers already have either the net itself, or the place
where to get it from.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip.h      |  2 +-
 net/ipv4/ip_forward.c |  2 +-
 net/ipv4/ip_output.c  | 30 +++++++++++++++---------------
 net/ipv4/raw.c        |  2 +-
 4 files changed, 18 insertions(+), 18 deletions(-)

(limited to 'include')

diff --git a/include/net/ip.h b/include/net/ip.h
index 673ecdbe72fd..b9aaa32e4759 100644
--- a/include/net/ip.h
+++ b/include/net/ip.h
@@ -157,7 +157,7 @@ struct ipv4_config
 
 extern struct ipv4_config ipv4_config;
 DECLARE_SNMP_STAT(struct ipstats_mib, ip_statistics);
-#define IP_INC_STATS(field)		SNMP_INC_STATS(ip_statistics, field)
+#define IP_INC_STATS(net, field)	do { (void)net; SNMP_INC_STATS(ip_statistics, field); } while (0)
 #define IP_INC_STATS_BH(field)		SNMP_INC_STATS_BH(ip_statistics, field)
 #define IP_ADD_STATS_BH(field, val)	SNMP_ADD_STATS_BH(ip_statistics, field, val)
 DECLARE_SNMP_STAT(struct linux_mib, net_statistics);
diff --git a/net/ipv4/ip_forward.c b/net/ipv4/ip_forward.c
index da14725916d3..7f78a5a7e1e7 100644
--- a/net/ipv4/ip_forward.c
+++ b/net/ipv4/ip_forward.c
@@ -88,7 +88,7 @@ int ip_forward(struct sk_buff *skb)
 
 	if (unlikely(skb->len > dst_mtu(&rt->u.dst) && !skb_is_gso(skb) &&
 		     (ip_hdr(skb)->frag_off & htons(IP_DF))) && !skb->local_df) {
-		IP_INC_STATS(IPSTATS_MIB_FRAGFAILS);
+		IP_INC_STATS(dev_net(rt->u.dst.dev), IPSTATS_MIB_FRAGFAILS);
 		icmp_send(skb, ICMP_DEST_UNREACH, ICMP_FRAG_NEEDED,
 			  htonl(dst_mtu(&rt->u.dst)));
 		goto drop;
diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c
index f003186b93b1..465544f6281a 100644
--- a/net/ipv4/ip_output.c
+++ b/net/ipv4/ip_output.c
@@ -182,9 +182,9 @@ static inline int ip_finish_output2(struct sk_buff *skb)
 	unsigned int hh_len = LL_RESERVED_SPACE(dev);
 
 	if (rt->rt_type == RTN_MULTICAST)
-		IP_INC_STATS(IPSTATS_MIB_OUTMCASTPKTS);
+		IP_INC_STATS(dev_net(dev), IPSTATS_MIB_OUTMCASTPKTS);
 	else if (rt->rt_type == RTN_BROADCAST)
-		IP_INC_STATS(IPSTATS_MIB_OUTBCASTPKTS);
+		IP_INC_STATS(dev_net(dev), IPSTATS_MIB_OUTBCASTPKTS);
 
 	/* Be paranoid, rather than too clever. */
 	if (unlikely(skb_headroom(skb) < hh_len && dev->header_ops)) {
@@ -244,7 +244,7 @@ int ip_mc_output(struct sk_buff *skb)
 	/*
 	 *	If the indicated interface is up and running, send the packet.
 	 */
-	IP_INC_STATS(IPSTATS_MIB_OUTREQUESTS);
+	IP_INC_STATS(dev_net(dev), IPSTATS_MIB_OUTREQUESTS);
 
 	skb->dev = dev;
 	skb->protocol = htons(ETH_P_IP);
@@ -298,7 +298,7 @@ int ip_output(struct sk_buff *skb)
 {
 	struct net_device *dev = skb->dst->dev;
 
-	IP_INC_STATS(IPSTATS_MIB_OUTREQUESTS);
+	IP_INC_STATS(dev_net(dev), IPSTATS_MIB_OUTREQUESTS);
 
 	skb->dev = dev;
 	skb->protocol = htons(ETH_P_IP);
@@ -389,7 +389,7 @@ packet_routed:
 	return ip_local_out(skb);
 
 no_route:
-	IP_INC_STATS(IPSTATS_MIB_OUTNOROUTES);
+	IP_INC_STATS(sock_net(sk), IPSTATS_MIB_OUTNOROUTES);
 	kfree_skb(skb);
 	return -EHOSTUNREACH;
 }
@@ -451,7 +451,7 @@ int ip_fragment(struct sk_buff *skb, int (*output)(struct sk_buff*))
 	iph = ip_hdr(skb);
 
 	if (unlikely((iph->frag_off & htons(IP_DF)) && !skb->local_df)) {
-		IP_INC_STATS(IPSTATS_MIB_FRAGFAILS);
+		IP_INC_STATS(dev_net(dev), IPSTATS_MIB_FRAGFAILS);
 		icmp_send(skb, ICMP_DEST_UNREACH, ICMP_FRAG_NEEDED,
 			  htonl(ip_skb_dst_mtu(skb)));
 		kfree_skb(skb);
@@ -542,7 +542,7 @@ int ip_fragment(struct sk_buff *skb, int (*output)(struct sk_buff*))
 			err = output(skb);
 
 			if (!err)
-				IP_INC_STATS(IPSTATS_MIB_FRAGCREATES);
+				IP_INC_STATS(dev_net(dev), IPSTATS_MIB_FRAGCREATES);
 			if (err || !frag)
 				break;
 
@@ -552,7 +552,7 @@ int ip_fragment(struct sk_buff *skb, int (*output)(struct sk_buff*))
 		}
 
 		if (err == 0) {
-			IP_INC_STATS(IPSTATS_MIB_FRAGOKS);
+			IP_INC_STATS(dev_net(dev), IPSTATS_MIB_FRAGOKS);
 			return 0;
 		}
 
@@ -561,7 +561,7 @@ int ip_fragment(struct sk_buff *skb, int (*output)(struct sk_buff*))
 			kfree_skb(frag);
 			frag = skb;
 		}
-		IP_INC_STATS(IPSTATS_MIB_FRAGFAILS);
+		IP_INC_STATS(dev_net(dev), IPSTATS_MIB_FRAGFAILS);
 		return err;
 	}
 
@@ -673,15 +673,15 @@ slow_path:
 		if (err)
 			goto fail;
 
-		IP_INC_STATS(IPSTATS_MIB_FRAGCREATES);
+		IP_INC_STATS(dev_net(dev), IPSTATS_MIB_FRAGCREATES);
 	}
 	kfree_skb(skb);
-	IP_INC_STATS(IPSTATS_MIB_FRAGOKS);
+	IP_INC_STATS(dev_net(dev), IPSTATS_MIB_FRAGOKS);
 	return err;
 
 fail:
 	kfree_skb(skb);
-	IP_INC_STATS(IPSTATS_MIB_FRAGFAILS);
+	IP_INC_STATS(dev_net(dev), IPSTATS_MIB_FRAGFAILS);
 	return err;
 }
 
@@ -1047,7 +1047,7 @@ alloc_new_skb:
 
 error:
 	inet->cork.length -= length;
-	IP_INC_STATS(IPSTATS_MIB_OUTDISCARDS);
+	IP_INC_STATS(sock_net(sk), IPSTATS_MIB_OUTDISCARDS);
 	return err;
 }
 
@@ -1189,7 +1189,7 @@ ssize_t	ip_append_page(struct sock *sk, struct page *page,
 
 error:
 	inet->cork.length -= size;
-	IP_INC_STATS(IPSTATS_MIB_OUTDISCARDS);
+	IP_INC_STATS(sock_net(sk), IPSTATS_MIB_OUTDISCARDS);
 	return err;
 }
 
@@ -1298,7 +1298,7 @@ out:
 	return err;
 
 error:
-	IP_INC_STATS(IPSTATS_MIB_OUTDISCARDS);
+	IP_INC_STATS(net, IPSTATS_MIB_OUTDISCARDS);
 	goto out;
 }
 
diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c
index 7f39ea443ec8..cd975743bcd2 100644
--- a/net/ipv4/raw.c
+++ b/net/ipv4/raw.c
@@ -385,7 +385,7 @@ error_fault:
 	err = -EFAULT;
 	kfree_skb(skb);
 error:
-	IP_INC_STATS(IPSTATS_MIB_OUTDISCARDS);
+	IP_INC_STATS(net, IPSTATS_MIB_OUTDISCARDS);
 	return err;
 }
 
-- 
cgit v1.2.3


From 7c73a6faffae0bfae70639113aecf06af666e714 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Wed, 16 Jul 2008 20:20:11 -0700
Subject: mib: add net to IP_INC_STATS_BH

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip.h                |  2 +-
 net/dccp/ipv4.c                 |  2 +-
 net/ipv4/datagram.c             |  2 +-
 net/ipv4/inet_connection_sock.c |  4 ++--
 net/ipv4/ip_forward.c           |  4 ++--
 net/ipv4/ip_fragment.c          | 17 ++++++++++-------
 net/ipv4/ip_input.c             | 30 ++++++++++++++++--------------
 net/ipv4/ipmr.c                 |  4 ++--
 net/ipv4/route.c                |  3 ++-
 net/ipv4/tcp_ipv4.c             |  2 +-
 net/ipv4/udp.c                  |  2 +-
 net/sctp/output.c               |  3 ++-
 12 files changed, 41 insertions(+), 34 deletions(-)

(limited to 'include')

diff --git a/include/net/ip.h b/include/net/ip.h
index b9aaa32e4759..9973ce08d8fd 100644
--- a/include/net/ip.h
+++ b/include/net/ip.h
@@ -158,7 +158,7 @@ struct ipv4_config
 extern struct ipv4_config ipv4_config;
 DECLARE_SNMP_STAT(struct ipstats_mib, ip_statistics);
 #define IP_INC_STATS(net, field)	do { (void)net; SNMP_INC_STATS(ip_statistics, field); } while (0)
-#define IP_INC_STATS_BH(field)		SNMP_INC_STATS_BH(ip_statistics, field)
+#define IP_INC_STATS_BH(net, field)	do { (void)net; SNMP_INC_STATS_BH(ip_statistics, field); } while (0)
 #define IP_ADD_STATS_BH(field, val)	SNMP_ADD_STATS_BH(ip_statistics, field, val)
 DECLARE_SNMP_STAT(struct linux_mib, net_statistics);
 #define NET_INC_STATS(field)		SNMP_INC_STATS(net_statistics, field)
diff --git a/net/dccp/ipv4.c b/net/dccp/ipv4.c
index 2b03c47cab1b..9f760a1e312c 100644
--- a/net/dccp/ipv4.c
+++ b/net/dccp/ipv4.c
@@ -465,7 +465,7 @@ static struct dst_entry* dccp_v4_route_skb(struct net *net, struct sock *sk,
 
 	security_skb_classify_flow(skb, &fl);
 	if (ip_route_output_flow(net, &rt, &fl, sk, 0)) {
-		IP_INC_STATS_BH(IPSTATS_MIB_OUTNOROUTES);
+		IP_INC_STATS_BH(net, IPSTATS_MIB_OUTNOROUTES);
 		return NULL;
 	}
 
diff --git a/net/ipv4/datagram.c b/net/ipv4/datagram.c
index 0c0c73f368ce..5e6c5a0f3fde 100644
--- a/net/ipv4/datagram.c
+++ b/net/ipv4/datagram.c
@@ -52,7 +52,7 @@ int ip4_datagram_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len)
 			       inet->sport, usin->sin_port, sk, 1);
 	if (err) {
 		if (err == -ENETUNREACH)
-			IP_INC_STATS_BH(IPSTATS_MIB_OUTNOROUTES);
+			IP_INC_STATS_BH(sock_net(sk), IPSTATS_MIB_OUTNOROUTES);
 		return err;
 	}
 
diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c
index 8338e1066654..bb81c958b744 100644
--- a/net/ipv4/inet_connection_sock.c
+++ b/net/ipv4/inet_connection_sock.c
@@ -342,12 +342,12 @@ struct dst_entry* inet_csk_route_req(struct sock *sk,
 
 	security_req_classify_flow(req, &fl);
 	if (ip_route_output_flow(net, &rt, &fl, sk, 0)) {
-		IP_INC_STATS_BH(IPSTATS_MIB_OUTNOROUTES);
+		IP_INC_STATS_BH(net, IPSTATS_MIB_OUTNOROUTES);
 		return NULL;
 	}
 	if (opt && opt->is_strictroute && rt->rt_dst != rt->rt_gateway) {
 		ip_rt_put(rt);
-		IP_INC_STATS_BH(IPSTATS_MIB_OUTNOROUTES);
+		IP_INC_STATS_BH(net, IPSTATS_MIB_OUTNOROUTES);
 		return NULL;
 	}
 	return &rt->u.dst;
diff --git a/net/ipv4/ip_forward.c b/net/ipv4/ip_forward.c
index 7f78a5a7e1e7..450016b89a18 100644
--- a/net/ipv4/ip_forward.c
+++ b/net/ipv4/ip_forward.c
@@ -42,7 +42,7 @@ static int ip_forward_finish(struct sk_buff *skb)
 {
 	struct ip_options * opt	= &(IPCB(skb)->opt);
 
-	IP_INC_STATS_BH(IPSTATS_MIB_OUTFORWDATAGRAMS);
+	IP_INC_STATS_BH(dev_net(skb->dst->dev), IPSTATS_MIB_OUTFORWDATAGRAMS);
 
 	if (unlikely(opt->optlen))
 		ip_forward_options(skb);
@@ -123,7 +123,7 @@ sr_failed:
 
 too_many_hops:
 	/* Tell the sender its packet died... */
-	IP_INC_STATS_BH(IPSTATS_MIB_INHDRERRORS);
+	IP_INC_STATS_BH(dev_net(skb->dst->dev), IPSTATS_MIB_INHDRERRORS);
 	icmp_send(skb, ICMP_TIME_EXCEEDED, ICMP_EXC_TTL, 0);
 drop:
 	kfree_skb(skb);
diff --git a/net/ipv4/ip_fragment.c b/net/ipv4/ip_fragment.c
index e23be5b36e1d..65364c06ada5 100644
--- a/net/ipv4/ip_fragment.c
+++ b/net/ipv4/ip_fragment.c
@@ -199,8 +199,8 @@ static void ip_expire(unsigned long arg)
 
 	ipq_kill(qp);
 
-	IP_INC_STATS_BH(IPSTATS_MIB_REASMTIMEOUT);
-	IP_INC_STATS_BH(IPSTATS_MIB_REASMFAILS);
+	IP_INC_STATS_BH(net, IPSTATS_MIB_REASMTIMEOUT);
+	IP_INC_STATS_BH(net, IPSTATS_MIB_REASMFAILS);
 
 	if ((qp->q.last_in & INET_FRAG_FIRST_IN) && qp->q.fragments != NULL) {
 		struct sk_buff *head = qp->q.fragments;
@@ -261,7 +261,10 @@ static inline int ip_frag_too_far(struct ipq *qp)
 	rc = qp->q.fragments && (end - start) > max;
 
 	if (rc) {
-		IP_INC_STATS_BH(IPSTATS_MIB_REASMFAILS);
+		struct net *net;
+
+		net = container_of(qp->q.net, struct net, ipv4.frags);
+		IP_INC_STATS_BH(net, IPSTATS_MIB_REASMFAILS);
 	}
 
 	return rc;
@@ -545,7 +548,7 @@ static int ip_frag_reasm(struct ipq *qp, struct sk_buff *prev,
 	iph = ip_hdr(head);
 	iph->frag_off = 0;
 	iph->tot_len = htons(len);
-	IP_INC_STATS_BH(IPSTATS_MIB_REASMOKS);
+	IP_INC_STATS_BH(dev_net(dev), IPSTATS_MIB_REASMOKS);
 	qp->q.fragments = NULL;
 	return 0;
 
@@ -560,7 +563,7 @@ out_oversize:
 			"Oversized IP packet from " NIPQUAD_FMT ".\n",
 			NIPQUAD(qp->saddr));
 out_fail:
-	IP_INC_STATS_BH(IPSTATS_MIB_REASMFAILS);
+	IP_INC_STATS_BH(dev_net(dev), IPSTATS_MIB_REASMFAILS);
 	return err;
 }
 
@@ -571,7 +574,7 @@ int ip_defrag(struct sk_buff *skb, u32 user)
 	struct net *net;
 
 	net = skb->dev ? dev_net(skb->dev) : dev_net(skb->dst->dev);
-	IP_INC_STATS_BH(IPSTATS_MIB_REASMREQDS);
+	IP_INC_STATS_BH(net, IPSTATS_MIB_REASMREQDS);
 
 	/* Start by cleaning up the memory. */
 	if (atomic_read(&net->ipv4.frags.mem) > net->ipv4.frags.high_thresh)
@@ -590,7 +593,7 @@ int ip_defrag(struct sk_buff *skb, u32 user)
 		return ret;
 	}
 
-	IP_INC_STATS_BH(IPSTATS_MIB_REASMFAILS);
+	IP_INC_STATS_BH(net, IPSTATS_MIB_REASMFAILS);
 	kfree_skb(skb);
 	return -ENOMEM;
 }
diff --git a/net/ipv4/ip_input.c b/net/ipv4/ip_input.c
index 7c26428ea67b..043f640df4b7 100644
--- a/net/ipv4/ip_input.c
+++ b/net/ipv4/ip_input.c
@@ -230,16 +230,16 @@ static int ip_local_deliver_finish(struct sk_buff *skb)
 				protocol = -ret;
 				goto resubmit;
 			}
-			IP_INC_STATS_BH(IPSTATS_MIB_INDELIVERS);
+			IP_INC_STATS_BH(net, IPSTATS_MIB_INDELIVERS);
 		} else {
 			if (!raw) {
 				if (xfrm4_policy_check(NULL, XFRM_POLICY_IN, skb)) {
-					IP_INC_STATS_BH(IPSTATS_MIB_INUNKNOWNPROTOS);
+					IP_INC_STATS_BH(net, IPSTATS_MIB_INUNKNOWNPROTOS);
 					icmp_send(skb, ICMP_DEST_UNREACH,
 						  ICMP_PROT_UNREACH, 0);
 				}
 			} else
-				IP_INC_STATS_BH(IPSTATS_MIB_INDELIVERS);
+				IP_INC_STATS_BH(net, IPSTATS_MIB_INDELIVERS);
 			kfree_skb(skb);
 		}
 	}
@@ -281,7 +281,7 @@ static inline int ip_rcv_options(struct sk_buff *skb)
 					      --ANK (980813)
 	*/
 	if (skb_cow(skb, skb_headroom(skb))) {
-		IP_INC_STATS_BH(IPSTATS_MIB_INDISCARDS);
+		IP_INC_STATS_BH(dev_net(dev), IPSTATS_MIB_INDISCARDS);
 		goto drop;
 	}
 
@@ -290,7 +290,7 @@ static inline int ip_rcv_options(struct sk_buff *skb)
 	opt->optlen = iph->ihl*4 - sizeof(struct iphdr);
 
 	if (ip_options_compile(dev_net(dev), opt, skb)) {
-		IP_INC_STATS_BH(IPSTATS_MIB_INHDRERRORS);
+		IP_INC_STATS_BH(dev_net(dev), IPSTATS_MIB_INHDRERRORS);
 		goto drop;
 	}
 
@@ -334,9 +334,11 @@ static int ip_rcv_finish(struct sk_buff *skb)
 					 skb->dev);
 		if (unlikely(err)) {
 			if (err == -EHOSTUNREACH)
-				IP_INC_STATS_BH(IPSTATS_MIB_INADDRERRORS);
+				IP_INC_STATS_BH(dev_net(skb->dev),
+						IPSTATS_MIB_INADDRERRORS);
 			else if (err == -ENETUNREACH)
-				IP_INC_STATS_BH(IPSTATS_MIB_INNOROUTES);
+				IP_INC_STATS_BH(dev_net(skb->dev),
+						IPSTATS_MIB_INNOROUTES);
 			goto drop;
 		}
 	}
@@ -357,9 +359,9 @@ static int ip_rcv_finish(struct sk_buff *skb)
 
 	rt = skb->rtable;
 	if (rt->rt_type == RTN_MULTICAST)
-		IP_INC_STATS_BH(IPSTATS_MIB_INMCASTPKTS);
+		IP_INC_STATS_BH(dev_net(rt->u.dst.dev), IPSTATS_MIB_INMCASTPKTS);
 	else if (rt->rt_type == RTN_BROADCAST)
-		IP_INC_STATS_BH(IPSTATS_MIB_INBCASTPKTS);
+		IP_INC_STATS_BH(dev_net(rt->u.dst.dev), IPSTATS_MIB_INBCASTPKTS);
 
 	return dst_input(skb);
 
@@ -382,10 +384,10 @@ int ip_rcv(struct sk_buff *skb, struct net_device *dev, struct packet_type *pt,
 	if (skb->pkt_type == PACKET_OTHERHOST)
 		goto drop;
 
-	IP_INC_STATS_BH(IPSTATS_MIB_INRECEIVES);
+	IP_INC_STATS_BH(dev_net(dev), IPSTATS_MIB_INRECEIVES);
 
 	if ((skb = skb_share_check(skb, GFP_ATOMIC)) == NULL) {
-		IP_INC_STATS_BH(IPSTATS_MIB_INDISCARDS);
+		IP_INC_STATS_BH(dev_net(dev), IPSTATS_MIB_INDISCARDS);
 		goto out;
 	}
 
@@ -418,7 +420,7 @@ int ip_rcv(struct sk_buff *skb, struct net_device *dev, struct packet_type *pt,
 
 	len = ntohs(iph->tot_len);
 	if (skb->len < len) {
-		IP_INC_STATS_BH(IPSTATS_MIB_INTRUNCATEDPKTS);
+		IP_INC_STATS_BH(dev_net(dev), IPSTATS_MIB_INTRUNCATEDPKTS);
 		goto drop;
 	} else if (len < (iph->ihl*4))
 		goto inhdr_error;
@@ -428,7 +430,7 @@ int ip_rcv(struct sk_buff *skb, struct net_device *dev, struct packet_type *pt,
 	 * Note this now means skb->len holds ntohs(iph->tot_len).
 	 */
 	if (pskb_trim_rcsum(skb, len)) {
-		IP_INC_STATS_BH(IPSTATS_MIB_INDISCARDS);
+		IP_INC_STATS_BH(dev_net(dev), IPSTATS_MIB_INDISCARDS);
 		goto drop;
 	}
 
@@ -439,7 +441,7 @@ int ip_rcv(struct sk_buff *skb, struct net_device *dev, struct packet_type *pt,
 		       ip_rcv_finish);
 
 inhdr_error:
-	IP_INC_STATS_BH(IPSTATS_MIB_INHDRERRORS);
+	IP_INC_STATS_BH(dev_net(dev), IPSTATS_MIB_INHDRERRORS);
 drop:
 	kfree_skb(skb);
 out:
diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c
index c9ab47b966b5..033c712c3a5d 100644
--- a/net/ipv4/ipmr.c
+++ b/net/ipv4/ipmr.c
@@ -1178,7 +1178,7 @@ static inline int ipmr_forward_finish(struct sk_buff *skb)
 {
 	struct ip_options * opt	= &(IPCB(skb)->opt);
 
-	IP_INC_STATS_BH(IPSTATS_MIB_OUTFORWDATAGRAMS);
+	IP_INC_STATS_BH(dev_net(skb->dst->dev), IPSTATS_MIB_OUTFORWDATAGRAMS);
 
 	if (unlikely(opt->optlen))
 		ip_forward_options(skb);
@@ -1241,7 +1241,7 @@ static void ipmr_queue_xmit(struct sk_buff *skb, struct mfc_cache *c, int vifi)
 		   to blackhole.
 		 */
 
-		IP_INC_STATS_BH(IPSTATS_MIB_FRAGFAILS);
+		IP_INC_STATS_BH(dev_net(dev), IPSTATS_MIB_FRAGFAILS);
 		ip_rt_put(rt);
 		goto out_free;
 	}
diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index 79c1e74263a1..e4ab0ac94f92 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -1429,7 +1429,8 @@ static int ip_error(struct sk_buff *skb)
 			break;
 		case ENETUNREACH:
 			code = ICMP_NET_UNREACH;
-			IP_INC_STATS_BH(IPSTATS_MIB_INNOROUTES);
+			IP_INC_STATS_BH(dev_net(rt->u.dst.dev),
+					IPSTATS_MIB_INNOROUTES);
 			break;
 		case EACCES:
 			code = ICMP_PKT_FILTERED;
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 0fefd4409413..7797d528701b 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -175,7 +175,7 @@ int tcp_v4_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len)
 			       inet->sport, usin->sin_port, sk, 1);
 	if (tmp < 0) {
 		if (tmp == -ENETUNREACH)
-			IP_INC_STATS_BH(IPSTATS_MIB_OUTNOROUTES);
+			IP_INC_STATS_BH(sock_net(sk), IPSTATS_MIB_OUTNOROUTES);
 		return tmp;
 	}
 
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index 048ef57edc1d..1560d11ba6ef 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -662,7 +662,7 @@ int udp_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg,
 		err = ip_route_output_flow(net, &rt, &fl, sk, 1);
 		if (err) {
 			if (err == -ENETUNREACH)
-				IP_INC_STATS_BH(IPSTATS_MIB_OUTNOROUTES);
+				IP_INC_STATS_BH(net, IPSTATS_MIB_OUTNOROUTES);
 			goto out;
 		}
 
diff --git a/net/sctp/output.c b/net/sctp/output.c
index abcd00dc05eb..9a63a3fb9011 100644
--- a/net/sctp/output.c
+++ b/net/sctp/output.c
@@ -50,6 +50,7 @@
 #include <linux/init.h>
 #include <net/inet_ecn.h>
 #include <net/icmp.h>
+#include <net/net_namespace.h>
 
 #ifndef TEST_FRAME
 #include <net/tcp.h>
@@ -595,7 +596,7 @@ out:
 	return err;
 no_route:
 	kfree_skb(nskb);
-	IP_INC_STATS_BH(IPSTATS_MIB_OUTNOROUTES);
+	IP_INC_STATS_BH(&init_net, IPSTATS_MIB_OUTNOROUTES);
 
 	/* FIXME: Returning the 'err' will effect all the associations
 	 * associated with a socket, although only one of the paths of the
-- 
cgit v1.2.3


From c5346fe396f5e22bbfb3ec037c43891c3c57d3e6 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Wed, 16 Jul 2008 20:20:33 -0700
Subject: mib: add net to IP_ADD_STATS_BH

Very simple - only ip_evictor (fragments) requires such.
This patch ends up the IP_XXX_STATS patching.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip.h       | 2 +-
 net/ipv4/ip_fragment.c | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/net/ip.h b/include/net/ip.h
index 9973ce08d8fd..93d0e093ff49 100644
--- a/include/net/ip.h
+++ b/include/net/ip.h
@@ -159,7 +159,7 @@ extern struct ipv4_config ipv4_config;
 DECLARE_SNMP_STAT(struct ipstats_mib, ip_statistics);
 #define IP_INC_STATS(net, field)	do { (void)net; SNMP_INC_STATS(ip_statistics, field); } while (0)
 #define IP_INC_STATS_BH(net, field)	do { (void)net; SNMP_INC_STATS_BH(ip_statistics, field); } while (0)
-#define IP_ADD_STATS_BH(field, val)	SNMP_ADD_STATS_BH(ip_statistics, field, val)
+#define IP_ADD_STATS_BH(net, field, val) SNMP_ADD_STATS_BH(ip_statistics, field, val)
 DECLARE_SNMP_STAT(struct linux_mib, net_statistics);
 #define NET_INC_STATS(field)		SNMP_INC_STATS(net_statistics, field)
 #define NET_INC_STATS_BH(field)		SNMP_INC_STATS_BH(net_statistics, field)
diff --git a/net/ipv4/ip_fragment.c b/net/ipv4/ip_fragment.c
index 65364c06ada5..38d38f058018 100644
--- a/net/ipv4/ip_fragment.c
+++ b/net/ipv4/ip_fragment.c
@@ -178,7 +178,7 @@ static void ip_evictor(struct net *net)
 
 	evicted = inet_frag_evictor(&net->ipv4.frags, &ip4_frags);
 	if (evicted)
-		IP_ADD_STATS_BH(IPSTATS_MIB_REASMFAILS, evicted);
+		IP_ADD_STATS_BH(net, IPSTATS_MIB_REASMFAILS, evicted);
 }
 
 /*
-- 
cgit v1.2.3


From f10f84314d5adc0ba6ca2a0877442de63e226130 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Wed, 16 Jul 2008 20:21:20 -0700
Subject: mib: drop unused TCP_XXX_STATS macros

TCP_INC_STATS_USER and TCP_ADD_STATS_BH are currently unused.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/tcp.h | 2 --
 1 file changed, 2 deletions(-)

(limited to 'include')

diff --git a/include/net/tcp.h b/include/net/tcp.h
index 6b827cc6c52f..048f8de724c2 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -268,9 +268,7 @@ extern struct proto tcp_prot;
 DECLARE_SNMP_STAT(struct tcp_mib, tcp_statistics);
 #define TCP_INC_STATS(field)		SNMP_INC_STATS(tcp_statistics, field)
 #define TCP_INC_STATS_BH(field)		SNMP_INC_STATS_BH(tcp_statistics, field)
-#define TCP_INC_STATS_USER(field) 	SNMP_INC_STATS_USER(tcp_statistics, field)
 #define TCP_DEC_STATS(field)		SNMP_DEC_STATS(tcp_statistics, field)
-#define TCP_ADD_STATS_BH(field, val)	SNMP_ADD_STATS_BH(tcp_statistics, field, val)
 #define TCP_ADD_STATS_USER(field, val)	SNMP_ADD_STATS_USER(tcp_statistics, field, val)
 
 extern void			tcp_v4_err(struct sk_buff *skb, u32);
-- 
cgit v1.2.3


From a9c19329eccdb145a08a4a2e969d7b40c54c9bcc Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Wed, 16 Jul 2008 20:21:42 -0700
Subject: tcp: add net to tcp_mib_init

This one sets TCP MIBs after zeroing them, and thus requires
the net.

The existing single caller can use init_net (temporarily).

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/tcp.h  | 2 +-
 net/ipv4/af_inet.c | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/net/tcp.h b/include/net/tcp.h
index 048f8de724c2..b9d0ba618b6b 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -1024,7 +1024,7 @@ static inline int tcp_paws_check(const struct tcp_options_received *rx_opt, int
 
 #define TCP_CHECK_TIMER(sk) do { } while (0)
 
-static inline void tcp_mib_init(void)
+static inline void tcp_mib_init(struct net *net)
 {
 	/* See RFC 2012 */
 	TCP_ADD_STATS_USER(TCP_MIB_RTOALGORITHM, 1);
diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c
index dc411335c14f..95a966dd1916 100644
--- a/net/ipv4/af_inet.c
+++ b/net/ipv4/af_inet.c
@@ -1363,7 +1363,7 @@ static int __init init_ipv4_mibs(void)
 			  sizeof(struct udp_mib)) < 0)
 		goto err_udplite_mib;
 
-	tcp_mib_init();
+	tcp_mib_init(&init_net);
 
 	return 0;
 
-- 
cgit v1.2.3


From 81cc8a75d944fa39fc333c2c329c8e8b3c62cada Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Wed, 16 Jul 2008 20:22:04 -0700
Subject: mib: add net to TCP_INC_STATS

Fortunately (almost) all the TCP code has a sock to get the net from :)

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/tcp.h     |  2 +-
 net/ipv4/tcp.c        |  4 ++--
 net/ipv4/tcp_output.c | 10 +++++-----
 3 files changed, 8 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/include/net/tcp.h b/include/net/tcp.h
index b9d0ba618b6b..15065157afc9 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -266,7 +266,7 @@ static inline int tcp_too_many_orphans(struct sock *sk, int num)
 extern struct proto tcp_prot;
 
 DECLARE_SNMP_STAT(struct tcp_mib, tcp_statistics);
-#define TCP_INC_STATS(field)		SNMP_INC_STATS(tcp_statistics, field)
+#define TCP_INC_STATS(net, field)	do { (void)net; SNMP_INC_STATS(tcp_statistics, field); } while (0)
 #define TCP_INC_STATS_BH(field)		SNMP_INC_STATS_BH(tcp_statistics, field)
 #define TCP_DEC_STATS(field)		SNMP_DEC_STATS(tcp_statistics, field)
 #define TCP_ADD_STATS_USER(field, val)	SNMP_ADD_STATS_USER(tcp_statistics, field, val)
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index 631133ebb20a..aa79198fb02e 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -1668,12 +1668,12 @@ void tcp_set_state(struct sock *sk, int state)
 	switch (state) {
 	case TCP_ESTABLISHED:
 		if (oldstate != TCP_ESTABLISHED)
-			TCP_INC_STATS(TCP_MIB_CURRESTAB);
+			TCP_INC_STATS(sock_net(sk), TCP_MIB_CURRESTAB);
 		break;
 
 	case TCP_CLOSE:
 		if (oldstate == TCP_CLOSE_WAIT || oldstate == TCP_ESTABLISHED)
-			TCP_INC_STATS(TCP_MIB_ESTABRESETS);
+			TCP_INC_STATS(sock_net(sk), TCP_MIB_ESTABRESETS);
 
 		sk->sk_prot->unhash(sk);
 		if (inet_csk(sk)->icsk_bind_hash &&
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index edef2afe905e..c3b0da826980 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -618,7 +618,7 @@ static int tcp_transmit_skb(struct sock *sk, struct sk_buff *skb, int clone_it,
 		tcp_event_data_sent(tp, skb, sk);
 
 	if (after(tcb->end_seq, tp->snd_nxt) || tcb->seq == tcb->end_seq)
-		TCP_INC_STATS(TCP_MIB_OUTSEGS);
+		TCP_INC_STATS(sock_net(sk), TCP_MIB_OUTSEGS);
 
 	err = icsk->icsk_af_ops->queue_xmit(skb, 0);
 	if (likely(err <= 0))
@@ -1910,7 +1910,7 @@ int tcp_retransmit_skb(struct sock *sk, struct sk_buff *skb)
 
 	if (err == 0) {
 		/* Update global TCP statistics. */
-		TCP_INC_STATS(TCP_MIB_RETRANSSEGS);
+		TCP_INC_STATS(sock_net(sk), TCP_MIB_RETRANSSEGS);
 
 		tp->total_retrans++;
 
@@ -2132,7 +2132,7 @@ void tcp_send_active_reset(struct sock *sk, gfp_t priority)
 	if (tcp_transmit_skb(sk, skb, 0, priority))
 		NET_INC_STATS(LINUX_MIB_TCPABORTFAILED);
 
-	TCP_INC_STATS(TCP_MIB_OUTRSTS);
+	TCP_INC_STATS(sock_net(sk), TCP_MIB_OUTRSTS);
 }
 
 /* WARNING: This routine must only be called when we have already sent
@@ -2258,7 +2258,7 @@ struct sk_buff *tcp_make_synack(struct sock *sk, struct dst_entry *dst,
 			      );
 
 	th->doff = (tcp_header_size >> 2);
-	TCP_INC_STATS(TCP_MIB_OUTSEGS);
+	TCP_INC_STATS(sock_net(sk), TCP_MIB_OUTSEGS);
 
 #ifdef CONFIG_TCP_MD5SIG
 	/* Okay, we have all we need - do the md5 hash if needed */
@@ -2367,7 +2367,7 @@ int tcp_connect(struct sock *sk)
 	 */
 	tp->snd_nxt = tp->write_seq;
 	tp->pushed_seq = tp->write_seq;
-	TCP_INC_STATS(TCP_MIB_ACTIVEOPENS);
+	TCP_INC_STATS(sock_net(sk), TCP_MIB_ACTIVEOPENS);
 
 	/* Timer for repeating the SYN until an answer. */
 	inet_csk_reset_xmit_timer(sk, ICSK_TIME_RETRANS,
-- 
cgit v1.2.3


From 63231bddf6514778792d3784f63822473d250fc0 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Wed, 16 Jul 2008 20:22:25 -0700
Subject: mib: add net to TCP_INC_STATS_BH

Same as before - the sock is always there to get the net from,
but there are also some places with the net already saved on
the stack.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/tcp.h        |  2 +-
 net/ipv4/tcp.c           |  2 +-
 net/ipv4/tcp_input.c     |  6 +++---
 net/ipv4/tcp_ipv4.c      | 14 +++++++-------
 net/ipv4/tcp_minisocks.c |  4 ++--
 net/ipv6/tcp_ipv6.c      | 14 +++++++-------
 6 files changed, 21 insertions(+), 21 deletions(-)

(limited to 'include')

diff --git a/include/net/tcp.h b/include/net/tcp.h
index 15065157afc9..8a2cd076be82 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -267,7 +267,7 @@ extern struct proto tcp_prot;
 
 DECLARE_SNMP_STAT(struct tcp_mib, tcp_statistics);
 #define TCP_INC_STATS(net, field)	do { (void)net; SNMP_INC_STATS(tcp_statistics, field); } while (0)
-#define TCP_INC_STATS_BH(field)		SNMP_INC_STATS_BH(tcp_statistics, field)
+#define TCP_INC_STATS_BH(net, field)	do { (void)net; SNMP_INC_STATS_BH(tcp_statistics, field); } while (0)
 #define TCP_DEC_STATS(field)		SNMP_DEC_STATS(tcp_statistics, field)
 #define TCP_ADD_STATS_USER(field, val)	SNMP_ADD_STATS_USER(tcp_statistics, field, val)
 
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index aa79198fb02e..f742f84026b5 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -2663,7 +2663,7 @@ EXPORT_SYMBOL(__tcp_put_md5sig_pool);
 void tcp_done(struct sock *sk)
 {
 	if(sk->sk_state == TCP_SYN_SENT || sk->sk_state == TCP_SYN_RECV)
-		TCP_INC_STATS_BH(TCP_MIB_ATTEMPTFAILS);
+		TCP_INC_STATS_BH(sock_net(sk), TCP_MIB_ATTEMPTFAILS);
 
 	tcp_set_state(sk, TCP_CLOSE);
 	tcp_clear_xmit_timers(sk);
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index d6ea970a1513..01e8004911b4 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -4795,7 +4795,7 @@ int tcp_rcv_established(struct sock *sk, struct sk_buff *skb,
 				tcp_data_snd_check(sk);
 				return 0;
 			} else { /* Header too small */
-				TCP_INC_STATS_BH(TCP_MIB_INERRS);
+				TCP_INC_STATS_BH(sock_net(sk), TCP_MIB_INERRS);
 				goto discard;
 			}
 		} else {
@@ -4934,7 +4934,7 @@ slow_path:
 	tcp_replace_ts_recent(tp, TCP_SKB_CB(skb)->seq);
 
 	if (th->syn && !before(TCP_SKB_CB(skb)->seq, tp->rcv_nxt)) {
-		TCP_INC_STATS_BH(TCP_MIB_INERRS);
+		TCP_INC_STATS_BH(sock_net(sk), TCP_MIB_INERRS);
 		NET_INC_STATS_BH(LINUX_MIB_TCPABORTONSYN);
 		tcp_reset(sk);
 		return 1;
@@ -4957,7 +4957,7 @@ step5:
 	return 0;
 
 csum_error:
-	TCP_INC_STATS_BH(TCP_MIB_INERRS);
+	TCP_INC_STATS_BH(sock_net(sk), TCP_MIB_INERRS);
 
 discard:
 	__kfree_skb(skb);
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index db3bf9be076f..e876312b950a 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -599,8 +599,8 @@ static void tcp_v4_send_reset(struct sock *sk, struct sk_buff *skb)
 	ip_send_reply(net->ipv4.tcp_sock, skb,
 		      &arg, arg.iov[0].iov_len);
 
-	TCP_INC_STATS_BH(TCP_MIB_OUTSEGS);
-	TCP_INC_STATS_BH(TCP_MIB_OUTRSTS);
+	TCP_INC_STATS_BH(net, TCP_MIB_OUTSEGS);
+	TCP_INC_STATS_BH(net, TCP_MIB_OUTRSTS);
 }
 
 /* The code following below sending ACKs in SYN-RECV and TIME-WAIT states
@@ -674,7 +674,7 @@ static void tcp_v4_send_ack(struct sk_buff *skb, u32 seq, u32 ack,
 	ip_send_reply(net->ipv4.tcp_sock, skb,
 		      &arg, arg.iov[0].iov_len);
 
-	TCP_INC_STATS_BH(TCP_MIB_OUTSEGS);
+	TCP_INC_STATS_BH(net, TCP_MIB_OUTSEGS);
 }
 
 static void tcp_v4_timewait_ack(struct sock *sk, struct sk_buff *skb)
@@ -1494,7 +1494,7 @@ discard:
 	return 0;
 
 csum_err:
-	TCP_INC_STATS_BH(TCP_MIB_INERRS);
+	TCP_INC_STATS_BH(sock_net(sk), TCP_MIB_INERRS);
 	goto discard;
 }
 
@@ -1514,7 +1514,7 @@ int tcp_v4_rcv(struct sk_buff *skb)
 		goto discard_it;
 
 	/* Count it even if it's bad */
-	TCP_INC_STATS_BH(TCP_MIB_INSEGS);
+	TCP_INC_STATS_BH(net, TCP_MIB_INSEGS);
 
 	if (!pskb_may_pull(skb, sizeof(struct tcphdr)))
 		goto discard_it;
@@ -1590,7 +1590,7 @@ no_tcp_socket:
 
 	if (skb->len < (th->doff << 2) || tcp_checksum_complete(skb)) {
 bad_packet:
-		TCP_INC_STATS_BH(TCP_MIB_INERRS);
+		TCP_INC_STATS_BH(net, TCP_MIB_INERRS);
 	} else {
 		tcp_v4_send_reset(NULL, skb);
 	}
@@ -1611,7 +1611,7 @@ do_time_wait:
 	}
 
 	if (skb->len < (th->doff << 2) || tcp_checksum_complete(skb)) {
-		TCP_INC_STATS_BH(TCP_MIB_INERRS);
+		TCP_INC_STATS_BH(net, TCP_MIB_INERRS);
 		inet_twsk_put(inet_twsk(sk));
 		goto discard_it;
 	}
diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c
index ea68a478fad6..8b02b1039968 100644
--- a/net/ipv4/tcp_minisocks.c
+++ b/net/ipv4/tcp_minisocks.c
@@ -480,7 +480,7 @@ struct sock *tcp_create_openreq_child(struct sock *sk, struct request_sock *req,
 		newtp->rx_opt.mss_clamp = req->mss;
 		TCP_ECN_openreq_child(newtp, req);
 
-		TCP_INC_STATS_BH(TCP_MIB_PASSIVEOPENS);
+		TCP_INC_STATS_BH(sock_net(sk), TCP_MIB_PASSIVEOPENS);
 	}
 	return newsk;
 }
@@ -630,7 +630,7 @@ struct sock *tcp_check_req(struct sock *sk,struct sk_buff *skb,
 		 *	   "fourth, check the SYN bit"
 		 */
 		if (flg & (TCP_FLAG_RST|TCP_FLAG_SYN)) {
-			TCP_INC_STATS_BH(TCP_MIB_ATTEMPTFAILS);
+			TCP_INC_STATS_BH(sock_net(sk), TCP_MIB_ATTEMPTFAILS);
 			goto embryonic_reset;
 		}
 
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index fc5f7166208a..3895d91ea9fa 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -1004,8 +1004,8 @@ static void tcp_v6_send_reset(struct sock *sk, struct sk_buff *skb)
 
 		if (xfrm_lookup(&buff->dst, &fl, NULL, 0) >= 0) {
 			ip6_xmit(ctl_sk, buff, &fl, NULL, 0);
-			TCP_INC_STATS_BH(TCP_MIB_OUTSEGS);
-			TCP_INC_STATS_BH(TCP_MIB_OUTRSTS);
+			TCP_INC_STATS_BH(net, TCP_MIB_OUTSEGS);
+			TCP_INC_STATS_BH(net, TCP_MIB_OUTRSTS);
 			return;
 		}
 	}
@@ -1089,7 +1089,7 @@ static void tcp_v6_send_ack(struct sk_buff *skb, u32 seq, u32 ack, u32 win, u32
 	if (!ip6_dst_lookup(ctl_sk, &buff->dst, &fl)) {
 		if (xfrm_lookup(&buff->dst, &fl, NULL, 0) >= 0) {
 			ip6_xmit(ctl_sk, buff, &fl, NULL, 0);
-			TCP_INC_STATS_BH(TCP_MIB_OUTSEGS);
+			TCP_INC_STATS_BH(net, TCP_MIB_OUTSEGS);
 			return;
 		}
 	}
@@ -1579,7 +1579,7 @@ discard:
 	kfree_skb(skb);
 	return 0;
 csum_err:
-	TCP_INC_STATS_BH(TCP_MIB_INERRS);
+	TCP_INC_STATS_BH(sock_net(sk), TCP_MIB_INERRS);
 	goto discard;
 
 
@@ -1625,7 +1625,7 @@ static int tcp_v6_rcv(struct sk_buff *skb)
 	/*
 	 *	Count it even if it's bad.
 	 */
-	TCP_INC_STATS_BH(TCP_MIB_INSEGS);
+	TCP_INC_STATS_BH(net, TCP_MIB_INSEGS);
 
 	if (!pskb_may_pull(skb, sizeof(struct tcphdr)))
 		goto discard_it;
@@ -1697,7 +1697,7 @@ no_tcp_socket:
 
 	if (skb->len < (th->doff<<2) || tcp_checksum_complete(skb)) {
 bad_packet:
-		TCP_INC_STATS_BH(TCP_MIB_INERRS);
+		TCP_INC_STATS_BH(net, TCP_MIB_INERRS);
 	} else {
 		tcp_v6_send_reset(NULL, skb);
 	}
@@ -1722,7 +1722,7 @@ do_time_wait:
 	}
 
 	if (skb->len < (th->doff<<2) || tcp_checksum_complete(skb)) {
-		TCP_INC_STATS_BH(TCP_MIB_INERRS);
+		TCP_INC_STATS_BH(net, TCP_MIB_INERRS);
 		inet_twsk_put(inet_twsk(sk));
 		goto discard_it;
 	}
-- 
cgit v1.2.3


From 74688e487a407a33d42879957b478601aca616b8 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Wed, 16 Jul 2008 20:22:46 -0700
Subject: mib: add net to TCP_DEC_STATS

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/tcp.h | 2 +-
 net/ipv4/tcp.c    | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/net/tcp.h b/include/net/tcp.h
index 8a2cd076be82..92d87691aa61 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -268,7 +268,7 @@ extern struct proto tcp_prot;
 DECLARE_SNMP_STAT(struct tcp_mib, tcp_statistics);
 #define TCP_INC_STATS(net, field)	do { (void)net; SNMP_INC_STATS(tcp_statistics, field); } while (0)
 #define TCP_INC_STATS_BH(net, field)	do { (void)net; SNMP_INC_STATS_BH(tcp_statistics, field); } while (0)
-#define TCP_DEC_STATS(field)		SNMP_DEC_STATS(tcp_statistics, field)
+#define TCP_DEC_STATS(net, field)	do { (void)net; SNMP_DEC_STATS(tcp_statistics, field); } while (0)
 #define TCP_ADD_STATS_USER(field, val)	SNMP_ADD_STATS_USER(tcp_statistics, field, val)
 
 extern void			tcp_v4_err(struct sk_buff *skb, u32);
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index f742f84026b5..525dcf534153 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -1682,7 +1682,7 @@ void tcp_set_state(struct sock *sk, int state)
 		/* fall through */
 	default:
 		if (oldstate==TCP_ESTABLISHED)
-			TCP_DEC_STATS(TCP_MIB_CURRESTAB);
+			TCP_DEC_STATS(sock_net(sk), TCP_MIB_CURRESTAB);
 	}
 
 	/* Change state AFTER socket is unhashed to avoid closed
-- 
cgit v1.2.3


From cf1100a7a4f2573f50f9a923b53373977328e3c8 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Wed, 16 Jul 2008 20:27:38 -0700
Subject: mib: add net to TCP_ADD_STATS_USER

Now we're done with the TCP_XXX_STATS macros.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/tcp.h | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/include/net/tcp.h b/include/net/tcp.h
index 92d87691aa61..4d788181654e 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -269,7 +269,7 @@ DECLARE_SNMP_STAT(struct tcp_mib, tcp_statistics);
 #define TCP_INC_STATS(net, field)	do { (void)net; SNMP_INC_STATS(tcp_statistics, field); } while (0)
 #define TCP_INC_STATS_BH(net, field)	do { (void)net; SNMP_INC_STATS_BH(tcp_statistics, field); } while (0)
 #define TCP_DEC_STATS(net, field)	do { (void)net; SNMP_DEC_STATS(tcp_statistics, field); } while (0)
-#define TCP_ADD_STATS_USER(field, val)	SNMP_ADD_STATS_USER(tcp_statistics, field, val)
+#define TCP_ADD_STATS_USER(net, field, val) do { (void)net; SNMP_ADD_STATS_USER(tcp_statistics, field, val); } while (0)
 
 extern void			tcp_v4_err(struct sk_buff *skb, u32);
 
@@ -1027,10 +1027,10 @@ static inline int tcp_paws_check(const struct tcp_options_received *rx_opt, int
 static inline void tcp_mib_init(struct net *net)
 {
 	/* See RFC 2012 */
-	TCP_ADD_STATS_USER(TCP_MIB_RTOALGORITHM, 1);
-	TCP_ADD_STATS_USER(TCP_MIB_RTOMIN, TCP_RTO_MIN*1000/HZ);
-	TCP_ADD_STATS_USER(TCP_MIB_RTOMAX, TCP_RTO_MAX*1000/HZ);
-	TCP_ADD_STATS_USER(TCP_MIB_MAXCONN, -1);
+	TCP_ADD_STATS_USER(net, TCP_MIB_RTOALGORITHM, 1);
+	TCP_ADD_STATS_USER(net, TCP_MIB_RTOMIN, TCP_RTO_MIN*1000/HZ);
+	TCP_ADD_STATS_USER(net, TCP_MIB_RTOMAX, TCP_RTO_MAX*1000/HZ);
+	TCP_ADD_STATS_USER(net, TCP_MIB_MAXCONN, -1);
 }
 
 /* from STCP */
-- 
cgit v1.2.3


From 5c52ba170f8167511bdb65b981f4582100c40675 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Wed, 16 Jul 2008 20:28:10 -0700
Subject: sock: add net to prot->enter_memory_pressure callback

The tcp_enter_memory_pressure calls NET_INC_STATS, but doesn't
have where to get the net from.

I decided to add a sk argument, not the net itself, only to factor
all the required sock_net(sk) calls inside the enter_memory_pressure
callback itself.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/sock.h     | 4 ++--
 include/net/tcp.h      | 2 +-
 net/core/sock.c        | 2 +-
 net/decnet/af_decnet.c | 2 +-
 net/ipv4/tcp.c         | 4 ++--
 net/sctp/socket.c      | 2 +-
 6 files changed, 8 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/include/net/sock.h b/include/net/sock.h
index 3f4897ab432e..06c5259aff30 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -565,7 +565,7 @@ struct proto {
 #endif
 
 	/* Memory pressure */
-	void			(*enter_memory_pressure)(void);
+	void			(*enter_memory_pressure)(struct sock *sk);
 	atomic_t		*memory_allocated;	/* Current allocated memory. */
 	atomic_t		*sockets_allocated;	/* Current number of sockets. */
 	/*
@@ -1210,7 +1210,7 @@ static inline struct page *sk_stream_alloc_page(struct sock *sk)
 
 	page = alloc_pages(sk->sk_allocation, 0);
 	if (!page) {
-		sk->sk_prot->enter_memory_pressure();
+		sk->sk_prot->enter_memory_pressure(sk);
 		sk_stream_moderate_sndbuf(sk);
 	}
 	return page;
diff --git a/include/net/tcp.h b/include/net/tcp.h
index 4d788181654e..c25cb5278b95 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -975,7 +975,7 @@ static inline void tcp_openreq_init(struct request_sock *req,
 	ireq->rmt_port = tcp_hdr(skb)->source;
 }
 
-extern void tcp_enter_memory_pressure(void);
+extern void tcp_enter_memory_pressure(struct sock *sk);
 
 static inline int keepalive_intvl_when(const struct tcp_sock *tp)
 {
diff --git a/net/core/sock.c b/net/core/sock.c
index 2c0ba52e5303..10a64d57078c 100644
--- a/net/core/sock.c
+++ b/net/core/sock.c
@@ -1442,7 +1442,7 @@ int __sk_mem_schedule(struct sock *sk, int size, int kind)
 	/* Under pressure. */
 	if (allocated > prot->sysctl_mem[1])
 		if (prot->enter_memory_pressure)
-			prot->enter_memory_pressure();
+			prot->enter_memory_pressure(sk);
 
 	/* Over hard limit. */
 	if (allocated > prot->sysctl_mem[2])
diff --git a/net/decnet/af_decnet.c b/net/decnet/af_decnet.c
index 931bdf9cb756..61b7df577ddd 100644
--- a/net/decnet/af_decnet.c
+++ b/net/decnet/af_decnet.c
@@ -451,7 +451,7 @@ static void dn_destruct(struct sock *sk)
 
 static int dn_memory_pressure;
 
-static void dn_enter_memory_pressure(void)
+static void dn_enter_memory_pressure(struct sock *sk)
 {
 	if (!dn_memory_pressure) {
 		dn_memory_pressure = 1;
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index 525dcf534153..bc8559a6f7e5 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -316,7 +316,7 @@ int tcp_memory_pressure __read_mostly;
 
 EXPORT_SYMBOL(tcp_memory_pressure);
 
-void tcp_enter_memory_pressure(void)
+void tcp_enter_memory_pressure(struct sock *sk)
 {
 	if (!tcp_memory_pressure) {
 		NET_INC_STATS(LINUX_MIB_TCPMEMORYPRESSURES);
@@ -649,7 +649,7 @@ struct sk_buff *sk_stream_alloc_skb(struct sock *sk, int size, gfp_t gfp)
 		}
 		__kfree_skb(skb);
 	} else {
-		sk->sk_prot->enter_memory_pressure();
+		sk->sk_prot->enter_memory_pressure(sk);
 		sk_stream_moderate_sndbuf(sk);
 	}
 	return NULL;
diff --git a/net/sctp/socket.c b/net/sctp/socket.c
index df5572c39f0c..6aba01b0ce4e 100644
--- a/net/sctp/socket.c
+++ b/net/sctp/socket.c
@@ -116,7 +116,7 @@ static int sctp_memory_pressure;
 static atomic_t sctp_memory_allocated;
 static atomic_t sctp_sockets_allocated;
 
-static void sctp_enter_memory_pressure(void)
+static void sctp_enter_memory_pressure(struct sock *sk)
 {
 	sctp_memory_pressure = 1;
 }
-- 
cgit v1.2.3


From 4e6734447dbc7a0a85e09616821c0782d9fb1141 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Wed, 16 Jul 2008 20:30:14 -0700
Subject: mib: add net to NET_INC_STATS

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip.h      | 2 +-
 net/ipv4/tcp.c        | 2 +-
 net/ipv4/tcp_output.c | 4 ++--
 3 files changed, 4 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/include/net/ip.h b/include/net/ip.h
index 93d0e093ff49..b42a43499856 100644
--- a/include/net/ip.h
+++ b/include/net/ip.h
@@ -161,7 +161,7 @@ DECLARE_SNMP_STAT(struct ipstats_mib, ip_statistics);
 #define IP_INC_STATS_BH(net, field)	do { (void)net; SNMP_INC_STATS_BH(ip_statistics, field); } while (0)
 #define IP_ADD_STATS_BH(net, field, val) SNMP_ADD_STATS_BH(ip_statistics, field, val)
 DECLARE_SNMP_STAT(struct linux_mib, net_statistics);
-#define NET_INC_STATS(field)		SNMP_INC_STATS(net_statistics, field)
+#define NET_INC_STATS(net, field)	do { (void)net; SNMP_INC_STATS(net_statistics, field); } while (0)
 #define NET_INC_STATS_BH(field)		SNMP_INC_STATS_BH(net_statistics, field)
 #define NET_INC_STATS_USER(field) 	SNMP_INC_STATS_USER(net_statistics, field)
 #define NET_ADD_STATS_BH(field, adnd)	SNMP_ADD_STATS_BH(net_statistics, field, adnd)
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index bc8559a6f7e5..85f08291e928 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -319,7 +319,7 @@ EXPORT_SYMBOL(tcp_memory_pressure);
 void tcp_enter_memory_pressure(struct sock *sk)
 {
 	if (!tcp_memory_pressure) {
-		NET_INC_STATS(LINUX_MIB_TCPMEMORYPRESSURES);
+		NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPMEMORYPRESSURES);
 		tcp_memory_pressure = 1;
 	}
 }
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index c3b0da826980..176f0702b8ac 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -2119,7 +2119,7 @@ void tcp_send_active_reset(struct sock *sk, gfp_t priority)
 	/* NOTE: No TCP options attached and we never retransmit this. */
 	skb = alloc_skb(MAX_TCP_HEADER, priority);
 	if (!skb) {
-		NET_INC_STATS(LINUX_MIB_TCPABORTFAILED);
+		NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPABORTFAILED);
 		return;
 	}
 
@@ -2130,7 +2130,7 @@ void tcp_send_active_reset(struct sock *sk, gfp_t priority)
 	/* Send it off. */
 	TCP_SKB_CB(skb)->when = tcp_time_stamp;
 	if (tcp_transmit_skb(sk, skb, 0, priority))
-		NET_INC_STATS(LINUX_MIB_TCPABORTFAILED);
+		NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPABORTFAILED);
 
 	TCP_INC_STATS(sock_net(sk), TCP_MIB_OUTRSTS);
 }
-- 
cgit v1.2.3


From de0744af1fe2d0a3d428f6af0f2fe1f6179b1a9c Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Wed, 16 Jul 2008 20:31:16 -0700
Subject: mib: add net to NET_INC_STATS_BH

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip.h            |  2 +-
 include/net/tcp.h           |  2 +-
 net/dccp/ipv4.c             | 10 +++----
 net/dccp/ipv6.c             |  8 +++---
 net/dccp/timer.c            |  4 +--
 net/ipv4/arp.c              |  2 +-
 net/ipv4/inet_hashtables.c  |  4 +--
 net/ipv4/syncookies.c       |  6 ++---
 net/ipv4/tcp.c              |  6 +++--
 net/ipv4/tcp_input.c        | 65 +++++++++++++++++++++++----------------------
 net/ipv4/tcp_ipv4.c         | 12 ++++-----
 net/ipv4/tcp_minisocks.c    |  6 ++---
 net/ipv4/tcp_output.c       |  4 +--
 net/ipv4/tcp_timer.c        | 12 ++++-----
 net/ipv6/inet6_hashtables.c |  4 +--
 net/ipv6/syncookies.c       |  6 ++---
 net/ipv6/tcp_ipv6.c         | 10 +++----
 net/sctp/input.c            |  2 +-
 18 files changed, 84 insertions(+), 81 deletions(-)

(limited to 'include')

diff --git a/include/net/ip.h b/include/net/ip.h
index b42a43499856..21df167a8e5b 100644
--- a/include/net/ip.h
+++ b/include/net/ip.h
@@ -162,7 +162,7 @@ DECLARE_SNMP_STAT(struct ipstats_mib, ip_statistics);
 #define IP_ADD_STATS_BH(net, field, val) SNMP_ADD_STATS_BH(ip_statistics, field, val)
 DECLARE_SNMP_STAT(struct linux_mib, net_statistics);
 #define NET_INC_STATS(net, field)	do { (void)net; SNMP_INC_STATS(net_statistics, field); } while (0)
-#define NET_INC_STATS_BH(field)		SNMP_INC_STATS_BH(net_statistics, field)
+#define NET_INC_STATS_BH(net, field)	do { (void)net; SNMP_INC_STATS_BH(net_statistics, field); } while (0)
 #define NET_INC_STATS_USER(field) 	SNMP_INC_STATS_USER(net_statistics, field)
 #define NET_ADD_STATS_BH(field, adnd)	SNMP_ADD_STATS_BH(net_statistics, field, adnd)
 #define NET_ADD_STATS_USER(field, adnd)	SNMP_ADD_STATS_USER(net_statistics, field, adnd)
diff --git a/include/net/tcp.h b/include/net/tcp.h
index c25cb5278b95..60e5be8b925b 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -894,7 +894,7 @@ static inline int tcp_prequeue(struct sock *sk, struct sk_buff *skb)
 
 			while ((skb1 = __skb_dequeue(&tp->ucopy.prequeue)) != NULL) {
 				sk->sk_backlog_rcv(sk, skb1);
-				NET_INC_STATS_BH(LINUX_MIB_TCPPREQUEUEDROPPED);
+				NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPPREQUEUEDROPPED);
 			}
 
 			tp->ucopy.memory = 0;
diff --git a/net/dccp/ipv4.c b/net/dccp/ipv4.c
index 9f760a1e312c..2622ace17c46 100644
--- a/net/dccp/ipv4.c
+++ b/net/dccp/ipv4.c
@@ -230,7 +230,7 @@ static void dccp_v4_err(struct sk_buff *skb, u32 info)
 	 * servers this needs to be solved differently.
 	 */
 	if (sock_owned_by_user(sk))
-		NET_INC_STATS_BH(LINUX_MIB_LOCKDROPPEDICMPS);
+		NET_INC_STATS_BH(net, LINUX_MIB_LOCKDROPPEDICMPS);
 
 	if (sk->sk_state == DCCP_CLOSED)
 		goto out;
@@ -239,7 +239,7 @@ static void dccp_v4_err(struct sk_buff *skb, u32 info)
 	seq = dccp_hdr_seq(dh);
 	if ((1 << sk->sk_state) & ~(DCCPF_REQUESTING | DCCPF_LISTEN) &&
 	    !between48(seq, dp->dccps_swl, dp->dccps_swh)) {
-		NET_INC_STATS_BH(LINUX_MIB_OUTOFWINDOWICMPS);
+		NET_INC_STATS_BH(net, LINUX_MIB_OUTOFWINDOWICMPS);
 		goto out;
 	}
 
@@ -286,7 +286,7 @@ static void dccp_v4_err(struct sk_buff *skb, u32 info)
 		BUG_TRAP(!req->sk);
 
 		if (seq != dccp_rsk(req)->dreq_iss) {
-			NET_INC_STATS_BH(LINUX_MIB_OUTOFWINDOWICMPS);
+			NET_INC_STATS_BH(net, LINUX_MIB_OUTOFWINDOWICMPS);
 			goto out;
 		}
 		/*
@@ -409,9 +409,9 @@ struct sock *dccp_v4_request_recv_sock(struct sock *sk, struct sk_buff *skb,
 	return newsk;
 
 exit_overflow:
-	NET_INC_STATS_BH(LINUX_MIB_LISTENOVERFLOWS);
+	NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_LISTENOVERFLOWS);
 exit:
-	NET_INC_STATS_BH(LINUX_MIB_LISTENDROPS);
+	NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_LISTENDROPS);
 	dst_release(dst);
 	return NULL;
 }
diff --git a/net/dccp/ipv6.c b/net/dccp/ipv6.c
index 83cc9bbc6208..b74e8b2cbe55 100644
--- a/net/dccp/ipv6.c
+++ b/net/dccp/ipv6.c
@@ -111,7 +111,7 @@ static void dccp_v6_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
 
 	bh_lock_sock(sk);
 	if (sock_owned_by_user(sk))
-		NET_INC_STATS_BH(LINUX_MIB_LOCKDROPPEDICMPS);
+		NET_INC_STATS_BH(net, LINUX_MIB_LOCKDROPPEDICMPS);
 
 	if (sk->sk_state == DCCP_CLOSED)
 		goto out;
@@ -189,7 +189,7 @@ static void dccp_v6_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
 		BUG_TRAP(req->sk == NULL);
 
 		if (seq != dccp_rsk(req)->dreq_iss) {
-			NET_INC_STATS_BH(LINUX_MIB_OUTOFWINDOWICMPS);
+			NET_INC_STATS_BH(net, LINUX_MIB_OUTOFWINDOWICMPS);
 			goto out;
 		}
 
@@ -630,9 +630,9 @@ static struct sock *dccp_v6_request_recv_sock(struct sock *sk,
 	return newsk;
 
 out_overflow:
-	NET_INC_STATS_BH(LINUX_MIB_LISTENOVERFLOWS);
+	NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_LISTENOVERFLOWS);
 out:
-	NET_INC_STATS_BH(LINUX_MIB_LISTENDROPS);
+	NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_LISTENDROPS);
 	if (opt != NULL && opt != np->opt)
 		sock_kfree_s(sk, opt, opt->tot_len);
 	dst_release(dst);
diff --git a/net/dccp/timer.c b/net/dccp/timer.c
index 8703a792b560..3608d5342ca2 100644
--- a/net/dccp/timer.c
+++ b/net/dccp/timer.c
@@ -224,7 +224,7 @@ static void dccp_delack_timer(unsigned long data)
 	if (sock_owned_by_user(sk)) {
 		/* Try again later. */
 		icsk->icsk_ack.blocked = 1;
-		NET_INC_STATS_BH(LINUX_MIB_DELAYEDACKLOCKED);
+		NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_DELAYEDACKLOCKED);
 		sk_reset_timer(sk, &icsk->icsk_delack_timer,
 			       jiffies + TCP_DELACK_MIN);
 		goto out;
@@ -254,7 +254,7 @@ static void dccp_delack_timer(unsigned long data)
 			icsk->icsk_ack.ato = TCP_ATO_MIN;
 		}
 		dccp_send_ack(sk);
-		NET_INC_STATS_BH(LINUX_MIB_DELAYEDACKS);
+		NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_DELAYEDACKS);
 	}
 out:
 	bh_unlock_sock(sk);
diff --git a/net/ipv4/arp.c b/net/ipv4/arp.c
index aab98b8a9945..b043eda60b04 100644
--- a/net/ipv4/arp.c
+++ b/net/ipv4/arp.c
@@ -426,7 +426,7 @@ static int arp_filter(__be32 sip, __be32 tip, struct net_device *dev)
 	if (ip_route_output_key(net, &rt, &fl) < 0)
 		return 1;
 	if (rt->u.dst.dev != dev) {
-		NET_INC_STATS_BH(LINUX_MIB_ARPFILTER);
+		NET_INC_STATS_BH(net, LINUX_MIB_ARPFILTER);
 		flag = 1;
 	}
 	ip_rt_put(rt);
diff --git a/net/ipv4/inet_hashtables.c b/net/ipv4/inet_hashtables.c
index eca5899729e3..115f53722d20 100644
--- a/net/ipv4/inet_hashtables.c
+++ b/net/ipv4/inet_hashtables.c
@@ -312,11 +312,11 @@ unique:
 
 	if (twp) {
 		*twp = tw;
-		NET_INC_STATS_BH(LINUX_MIB_TIMEWAITRECYCLED);
+		NET_INC_STATS_BH(net, LINUX_MIB_TIMEWAITRECYCLED);
 	} else if (tw) {
 		/* Silly. Should hash-dance instead... */
 		inet_twsk_deschedule(tw, death_row);
-		NET_INC_STATS_BH(LINUX_MIB_TIMEWAITRECYCLED);
+		NET_INC_STATS_BH(net, LINUX_MIB_TIMEWAITRECYCLED);
 
 		inet_twsk_put(tw);
 	}
diff --git a/net/ipv4/syncookies.c b/net/ipv4/syncookies.c
index fdde2ae07e24..51bc24d3b8a7 100644
--- a/net/ipv4/syncookies.c
+++ b/net/ipv4/syncookies.c
@@ -173,7 +173,7 @@ __u32 cookie_v4_init_sequence(struct sock *sk, struct sk_buff *skb, __u16 *mssp)
 		;
 	*mssp = msstab[mssind] + 1;
 
-	NET_INC_STATS_BH(LINUX_MIB_SYNCOOKIESSENT);
+	NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_SYNCOOKIESSENT);
 
 	return secure_tcp_syn_cookie(iph->saddr, iph->daddr,
 				     th->source, th->dest, ntohl(th->seq),
@@ -269,11 +269,11 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb,
 
 	if (time_after(jiffies, tp->last_synq_overflow + TCP_TIMEOUT_INIT) ||
 	    (mss = cookie_check(skb, cookie)) == 0) {
-		NET_INC_STATS_BH(LINUX_MIB_SYNCOOKIESFAILED);
+		NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_SYNCOOKIESFAILED);
 		goto out;
 	}
 
-	NET_INC_STATS_BH(LINUX_MIB_SYNCOOKIESRECV);
+	NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_SYNCOOKIESRECV);
 
 	/* check for timestamp cookie support */
 	memset(&tcp_opt, 0, sizeof(tcp_opt));
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index 85f08291e928..9e0e45c37806 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -1871,7 +1871,8 @@ adjudge_to_death:
 		if (tp->linger2 < 0) {
 			tcp_set_state(sk, TCP_CLOSE);
 			tcp_send_active_reset(sk, GFP_ATOMIC);
-			NET_INC_STATS_BH(LINUX_MIB_TCPABORTONLINGER);
+			NET_INC_STATS_BH(sock_net(sk),
+					LINUX_MIB_TCPABORTONLINGER);
 		} else {
 			const int tmo = tcp_fin_time(sk);
 
@@ -1893,7 +1894,8 @@ adjudge_to_death:
 				       "sockets\n");
 			tcp_set_state(sk, TCP_CLOSE);
 			tcp_send_active_reset(sk, GFP_ATOMIC);
-			NET_INC_STATS_BH(LINUX_MIB_TCPABORTONMEMORY);
+			NET_INC_STATS_BH(sock_net(sk),
+					LINUX_MIB_TCPABORTONMEMORY);
 		}
 	}
 
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index f50d8433f042..fac49a6e1611 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -961,7 +961,7 @@ static void tcp_update_reordering(struct sock *sk, const int metric,
 		else
 			mib_idx = LINUX_MIB_TCPSACKREORDER;
 
-		NET_INC_STATS_BH(mib_idx);
+		NET_INC_STATS_BH(sock_net(sk), mib_idx);
 #if FASTRETRANS_DEBUG > 1
 		printk(KERN_DEBUG "Disorder%d %d %u f%u s%u rr%d\n",
 		       tp->rx_opt.sack_ok, inet_csk(sk)->icsk_ca_state,
@@ -1157,7 +1157,7 @@ static void tcp_mark_lost_retrans(struct sock *sk)
 				tp->lost_out += tcp_skb_pcount(skb);
 				TCP_SKB_CB(skb)->sacked |= TCPCB_LOST;
 			}
-			NET_INC_STATS_BH(LINUX_MIB_TCPLOSTRETRANSMIT);
+			NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPLOSTRETRANSMIT);
 		} else {
 			if (before(ack_seq, new_low_seq))
 				new_low_seq = ack_seq;
@@ -1181,7 +1181,7 @@ static int tcp_check_dsack(struct sock *sk, struct sk_buff *ack_skb,
 	if (before(start_seq_0, TCP_SKB_CB(ack_skb)->ack_seq)) {
 		dup_sack = 1;
 		tcp_dsack_seen(tp);
-		NET_INC_STATS_BH(LINUX_MIB_TCPDSACKRECV);
+		NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPDSACKRECV);
 	} else if (num_sacks > 1) {
 		u32 end_seq_1 = get_unaligned_be32(&sp[1].end_seq);
 		u32 start_seq_1 = get_unaligned_be32(&sp[1].start_seq);
@@ -1190,7 +1190,8 @@ static int tcp_check_dsack(struct sock *sk, struct sk_buff *ack_skb,
 		    !before(start_seq_0, start_seq_1)) {
 			dup_sack = 1;
 			tcp_dsack_seen(tp);
-			NET_INC_STATS_BH(LINUX_MIB_TCPDSACKOFORECV);
+			NET_INC_STATS_BH(sock_net(sk),
+					LINUX_MIB_TCPDSACKOFORECV);
 		}
 	}
 
@@ -1476,7 +1477,7 @@ tcp_sacktag_write_queue(struct sock *sk, struct sk_buff *ack_skb,
 				mib_idx = LINUX_MIB_TCPSACKDISCARD;
 			}
 
-			NET_INC_STATS_BH(mib_idx);
+			NET_INC_STATS_BH(sock_net(sk), mib_idx);
 			if (i == 0)
 				first_sack_index = -1;
 			continue;
@@ -1969,7 +1970,7 @@ static int tcp_check_sack_reneging(struct sock *sk, int flag)
 {
 	if (flag & FLAG_SACK_RENEGING) {
 		struct inet_connection_sock *icsk = inet_csk(sk);
-		NET_INC_STATS_BH(LINUX_MIB_TCPSACKRENEGING);
+		NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPSACKRENEGING);
 
 		tcp_enter_loss(sk, 1);
 		icsk->icsk_retransmits++;
@@ -2401,7 +2402,7 @@ static int tcp_try_undo_recovery(struct sock *sk)
 		else
 			mib_idx = LINUX_MIB_TCPFULLUNDO;
 
-		NET_INC_STATS_BH(mib_idx);
+		NET_INC_STATS_BH(sock_net(sk), mib_idx);
 		tp->undo_marker = 0;
 	}
 	if (tp->snd_una == tp->high_seq && tcp_is_reno(tp)) {
@@ -2424,7 +2425,7 @@ static void tcp_try_undo_dsack(struct sock *sk)
 		DBGUNDO(sk, "D-SACK");
 		tcp_undo_cwr(sk, 1);
 		tp->undo_marker = 0;
-		NET_INC_STATS_BH(LINUX_MIB_TCPDSACKUNDO);
+		NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPDSACKUNDO);
 	}
 }
 
@@ -2447,7 +2448,7 @@ static int tcp_try_undo_partial(struct sock *sk, int acked)
 
 		DBGUNDO(sk, "Hoe");
 		tcp_undo_cwr(sk, 0);
-		NET_INC_STATS_BH(LINUX_MIB_TCPPARTIALUNDO);
+		NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPPARTIALUNDO);
 
 		/* So... Do not make Hoe's retransmit yet.
 		 * If the first packet was delayed, the rest
@@ -2476,7 +2477,7 @@ static int tcp_try_undo_loss(struct sock *sk)
 		DBGUNDO(sk, "partial loss");
 		tp->lost_out = 0;
 		tcp_undo_cwr(sk, 1);
-		NET_INC_STATS_BH(LINUX_MIB_TCPLOSSUNDO);
+		NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPLOSSUNDO);
 		inet_csk(sk)->icsk_retransmits = 0;
 		tp->undo_marker = 0;
 		if (tcp_is_sack(tp))
@@ -2595,7 +2596,7 @@ static void tcp_fastretrans_alert(struct sock *sk, int pkts_acked, int flag)
 	    icsk->icsk_ca_state != TCP_CA_Open &&
 	    tp->fackets_out > tp->reordering) {
 		tcp_mark_head_lost(sk, tp->fackets_out - tp->reordering);
-		NET_INC_STATS_BH(LINUX_MIB_TCPLOSS);
+		NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPLOSS);
 	}
 
 	/* D. Check consistency of the current state. */
@@ -2700,7 +2701,7 @@ static void tcp_fastretrans_alert(struct sock *sk, int pkts_acked, int flag)
 		else
 			mib_idx = LINUX_MIB_TCPSACKRECOVERY;
 
-		NET_INC_STATS_BH(mib_idx);
+		NET_INC_STATS_BH(sock_net(sk), mib_idx);
 
 		tp->high_seq = tp->snd_nxt;
 		tp->prior_ssthresh = 0;
@@ -3211,7 +3212,7 @@ static int tcp_process_frto(struct sock *sk, int flag)
 		}
 		tp->frto_counter = 0;
 		tp->undo_marker = 0;
-		NET_INC_STATS_BH(LINUX_MIB_TCPSPURIOUSRTOS);
+		NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPSPURIOUSRTOS);
 	}
 	return 0;
 }
@@ -3264,12 +3265,12 @@ static int tcp_ack(struct sock *sk, struct sk_buff *skb, int flag)
 
 		tcp_ca_event(sk, CA_EVENT_FAST_ACK);
 
-		NET_INC_STATS_BH(LINUX_MIB_TCPHPACKS);
+		NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPHPACKS);
 	} else {
 		if (ack_seq != TCP_SKB_CB(skb)->end_seq)
 			flag |= FLAG_DATA;
 		else
-			NET_INC_STATS_BH(LINUX_MIB_TCPPUREACKS);
+			NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPPUREACKS);
 
 		flag |= tcp_ack_update_window(sk, skb, ack, ack_seq);
 
@@ -3724,7 +3725,7 @@ static void tcp_dsack_set(struct sock *sk, u32 seq, u32 end_seq)
 		else
 			mib_idx = LINUX_MIB_TCPDSACKOFOSENT;
 
-		NET_INC_STATS_BH(mib_idx);
+		NET_INC_STATS_BH(sock_net(sk), mib_idx);
 
 		tp->rx_opt.dsack = 1;
 		tp->duplicate_sack[0].start_seq = seq;
@@ -3750,7 +3751,7 @@ static void tcp_send_dupack(struct sock *sk, struct sk_buff *skb)
 
 	if (TCP_SKB_CB(skb)->end_seq != TCP_SKB_CB(skb)->seq &&
 	    before(TCP_SKB_CB(skb)->seq, tp->rcv_nxt)) {
-		NET_INC_STATS_BH(LINUX_MIB_DELAYEDACKLOST);
+		NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_DELAYEDACKLOST);
 		tcp_enter_quickack_mode(sk);
 
 		if (tcp_is_sack(tp) && sysctl_tcp_dsack) {
@@ -4039,7 +4040,7 @@ queue_and_out:
 
 	if (!after(TCP_SKB_CB(skb)->end_seq, tp->rcv_nxt)) {
 		/* A retransmit, 2nd most common case.  Force an immediate ack. */
-		NET_INC_STATS_BH(LINUX_MIB_DELAYEDACKLOST);
+		NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_DELAYEDACKLOST);
 		tcp_dsack_set(sk, TCP_SKB_CB(skb)->seq, TCP_SKB_CB(skb)->end_seq);
 
 out_of_window:
@@ -4181,7 +4182,7 @@ tcp_collapse(struct sock *sk, struct sk_buff_head *list,
 			struct sk_buff *next = skb->next;
 			__skb_unlink(skb, list);
 			__kfree_skb(skb);
-			NET_INC_STATS_BH(LINUX_MIB_TCPRCVCOLLAPSED);
+			NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPRCVCOLLAPSED);
 			skb = next;
 			continue;
 		}
@@ -4249,7 +4250,7 @@ tcp_collapse(struct sock *sk, struct sk_buff_head *list,
 				struct sk_buff *next = skb->next;
 				__skb_unlink(skb, list);
 				__kfree_skb(skb);
-				NET_INC_STATS_BH(LINUX_MIB_TCPRCVCOLLAPSED);
+				NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPRCVCOLLAPSED);
 				skb = next;
 				if (skb == tail ||
 				    tcp_hdr(skb)->syn ||
@@ -4312,7 +4313,7 @@ static int tcp_prune_ofo_queue(struct sock *sk)
 	int res = 0;
 
 	if (!skb_queue_empty(&tp->out_of_order_queue)) {
-		NET_INC_STATS_BH(LINUX_MIB_OFOPRUNED);
+		NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_OFOPRUNED);
 		__skb_queue_purge(&tp->out_of_order_queue);
 
 		/* Reset SACK state.  A conforming SACK implementation will
@@ -4341,7 +4342,7 @@ static int tcp_prune_queue(struct sock *sk)
 
 	SOCK_DEBUG(sk, "prune_queue: c=%x\n", tp->copied_seq);
 
-	NET_INC_STATS_BH(LINUX_MIB_PRUNECALLED);
+	NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_PRUNECALLED);
 
 	if (atomic_read(&sk->sk_rmem_alloc) >= sk->sk_rcvbuf)
 		tcp_clamp_window(sk);
@@ -4370,7 +4371,7 @@ static int tcp_prune_queue(struct sock *sk)
 	 * drop receive data on the floor.  It will get retransmitted
 	 * and hopefully then we'll have sufficient space.
 	 */
-	NET_INC_STATS_BH(LINUX_MIB_RCVPRUNED);
+	NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_RCVPRUNED);
 
 	/* Massive buffer overcommit. */
 	tp->pred_flags = 0;
@@ -4837,7 +4838,7 @@ int tcp_rcv_established(struct sock *sk, struct sk_buff *skb,
 
 					__skb_pull(skb, tcp_header_len);
 					tp->rcv_nxt = TCP_SKB_CB(skb)->end_seq;
-					NET_INC_STATS_BH(LINUX_MIB_TCPHPHITSTOUSER);
+					NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPHPHITSTOUSER);
 				}
 				if (copied_early)
 					tcp_cleanup_rbuf(sk, skb->len);
@@ -4860,7 +4861,7 @@ int tcp_rcv_established(struct sock *sk, struct sk_buff *skb,
 				if ((int)skb->truesize > sk->sk_forward_alloc)
 					goto step5;
 
-				NET_INC_STATS_BH(LINUX_MIB_TCPHPHITS);
+				NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPHPHITS);
 
 				/* Bulk data transfer: receiver */
 				__skb_pull(skb, tcp_header_len);
@@ -4904,7 +4905,7 @@ slow_path:
 	if (tcp_fast_parse_options(skb, th, tp) && tp->rx_opt.saw_tstamp &&
 	    tcp_paws_discard(sk, skb)) {
 		if (!th->rst) {
-			NET_INC_STATS_BH(LINUX_MIB_PAWSESTABREJECTED);
+			NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_PAWSESTABREJECTED);
 			tcp_send_dupack(sk, skb);
 			goto discard;
 		}
@@ -4940,7 +4941,7 @@ slow_path:
 
 	if (th->syn && !before(TCP_SKB_CB(skb)->seq, tp->rcv_nxt)) {
 		TCP_INC_STATS_BH(sock_net(sk), TCP_MIB_INERRS);
-		NET_INC_STATS_BH(LINUX_MIB_TCPABORTONSYN);
+		NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPABORTONSYN);
 		tcp_reset(sk);
 		return 1;
 	}
@@ -4996,7 +4997,7 @@ static int tcp_rcv_synsent_state_process(struct sock *sk, struct sk_buff *skb,
 		if (tp->rx_opt.saw_tstamp && tp->rx_opt.rcv_tsecr &&
 		    !between(tp->rx_opt.rcv_tsecr, tp->retrans_stamp,
 			     tcp_time_stamp)) {
-			NET_INC_STATS_BH(LINUX_MIB_PAWSACTIVEREJECTED);
+			NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_PAWSACTIVEREJECTED);
 			goto reset_and_undo;
 		}
 
@@ -5280,7 +5281,7 @@ int tcp_rcv_state_process(struct sock *sk, struct sk_buff *skb,
 	if (tcp_fast_parse_options(skb, th, tp) && tp->rx_opt.saw_tstamp &&
 	    tcp_paws_discard(sk, skb)) {
 		if (!th->rst) {
-			NET_INC_STATS_BH(LINUX_MIB_PAWSESTABREJECTED);
+			NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_PAWSESTABREJECTED);
 			tcp_send_dupack(sk, skb);
 			goto discard;
 		}
@@ -5309,7 +5310,7 @@ int tcp_rcv_state_process(struct sock *sk, struct sk_buff *skb,
 	 *	Check for a SYN in window.
 	 */
 	if (th->syn && !before(TCP_SKB_CB(skb)->seq, tp->rcv_nxt)) {
-		NET_INC_STATS_BH(LINUX_MIB_TCPABORTONSYN);
+		NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPABORTONSYN);
 		tcp_reset(sk);
 		return 1;
 	}
@@ -5391,7 +5392,7 @@ int tcp_rcv_state_process(struct sock *sk, struct sk_buff *skb,
 					    (TCP_SKB_CB(skb)->end_seq != TCP_SKB_CB(skb)->seq &&
 					     after(TCP_SKB_CB(skb)->end_seq - th->fin, tp->rcv_nxt))) {
 						tcp_done(sk);
-						NET_INC_STATS_BH(LINUX_MIB_TCPABORTONDATA);
+						NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPABORTONDATA);
 						return 1;
 					}
 
@@ -5451,7 +5452,7 @@ int tcp_rcv_state_process(struct sock *sk, struct sk_buff *skb,
 		if (sk->sk_shutdown & RCV_SHUTDOWN) {
 			if (TCP_SKB_CB(skb)->end_seq != TCP_SKB_CB(skb)->seq &&
 			    after(TCP_SKB_CB(skb)->end_seq - th->fin, tp->rcv_nxt)) {
-				NET_INC_STATS_BH(LINUX_MIB_TCPABORTONDATA);
+				NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPABORTONDATA);
 				tcp_reset(sk);
 				return 1;
 			}
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index e876312b950a..29adc668ad51 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -366,7 +366,7 @@ void tcp_v4_err(struct sk_buff *skb, u32 info)
 	 * servers this needs to be solved differently.
 	 */
 	if (sock_owned_by_user(sk))
-		NET_INC_STATS_BH(LINUX_MIB_LOCKDROPPEDICMPS);
+		NET_INC_STATS_BH(net, LINUX_MIB_LOCKDROPPEDICMPS);
 
 	if (sk->sk_state == TCP_CLOSE)
 		goto out;
@@ -375,7 +375,7 @@ void tcp_v4_err(struct sk_buff *skb, u32 info)
 	seq = ntohl(th->seq);
 	if (sk->sk_state != TCP_LISTEN &&
 	    !between(seq, tp->snd_una, tp->snd_nxt)) {
-		NET_INC_STATS_BH(LINUX_MIB_OUTOFWINDOWICMPS);
+		NET_INC_STATS_BH(net, LINUX_MIB_OUTOFWINDOWICMPS);
 		goto out;
 	}
 
@@ -422,7 +422,7 @@ void tcp_v4_err(struct sk_buff *skb, u32 info)
 		BUG_TRAP(!req->sk);
 
 		if (seq != tcp_rsk(req)->snt_isn) {
-			NET_INC_STATS_BH(LINUX_MIB_OUTOFWINDOWICMPS);
+			NET_INC_STATS_BH(net, LINUX_MIB_OUTOFWINDOWICMPS);
 			goto out;
 		}
 
@@ -1251,7 +1251,7 @@ int tcp_v4_conn_request(struct sock *sk, struct sk_buff *skb)
 			if (get_seconds() < peer->tcp_ts_stamp + TCP_PAWS_MSL &&
 			    (s32)(peer->tcp_ts - req->ts_recent) >
 							TCP_PAWS_WINDOW) {
-				NET_INC_STATS_BH(LINUX_MIB_PAWSPASSIVEREJECTED);
+				NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_PAWSPASSIVEREJECTED);
 				goto drop_and_release;
 			}
 		}
@@ -1365,9 +1365,9 @@ struct sock *tcp_v4_syn_recv_sock(struct sock *sk, struct sk_buff *skb,
 	return newsk;
 
 exit_overflow:
-	NET_INC_STATS_BH(LINUX_MIB_LISTENOVERFLOWS);
+	NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_LISTENOVERFLOWS);
 exit:
-	NET_INC_STATS_BH(LINUX_MIB_LISTENDROPS);
+	NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_LISTENDROPS);
 	dst_release(dst);
 	return NULL;
 }
diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c
index 8b02b1039968..204c42162660 100644
--- a/net/ipv4/tcp_minisocks.c
+++ b/net/ipv4/tcp_minisocks.c
@@ -244,7 +244,7 @@ kill:
 	}
 
 	if (paws_reject)
-		NET_INC_STATS_BH(LINUX_MIB_PAWSESTABREJECTED);
+		NET_INC_STATS_BH(twsk_net(tw), LINUX_MIB_PAWSESTABREJECTED);
 
 	if (!th->rst) {
 		/* In this case we must reset the TIMEWAIT timer.
@@ -611,7 +611,7 @@ struct sock *tcp_check_req(struct sock *sk,struct sk_buff *skb,
 		if (!(flg & TCP_FLAG_RST))
 			req->rsk_ops->send_ack(skb, req);
 		if (paws_reject)
-			NET_INC_STATS_BH(LINUX_MIB_PAWSESTABREJECTED);
+			NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_PAWSESTABREJECTED);
 		return NULL;
 	}
 
@@ -695,7 +695,7 @@ struct sock *tcp_check_req(struct sock *sk,struct sk_buff *skb,
 		}
 
 	embryonic_reset:
-		NET_INC_STATS_BH(LINUX_MIB_EMBRYONICRSTS);
+		NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_EMBRYONICRSTS);
 		if (!(flg & TCP_FLAG_RST))
 			req->rsk_ops->send_reset(sk, skb);
 
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index 176f0702b8ac..36a19707f67f 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -1995,7 +1995,7 @@ void tcp_xmit_retransmit_queue(struct sock *sk)
 						mib_idx = LINUX_MIB_TCPFASTRETRANS;
 					else
 						mib_idx = LINUX_MIB_TCPSLOWSTARTRETRANS;
-					NET_INC_STATS_BH(mib_idx);
+					NET_INC_STATS_BH(sock_net(sk), mib_idx);
 
 					if (skb == tcp_write_queue_head(sk))
 						inet_csk_reset_xmit_timer(sk, ICSK_TIME_RETRANS,
@@ -2065,7 +2065,7 @@ void tcp_xmit_retransmit_queue(struct sock *sk)
 						  inet_csk(sk)->icsk_rto,
 						  TCP_RTO_MAX);
 
-		NET_INC_STATS_BH(LINUX_MIB_TCPFORWARDRETRANS);
+		NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPFORWARDRETRANS);
 	}
 }
 
diff --git a/net/ipv4/tcp_timer.c b/net/ipv4/tcp_timer.c
index 6a480d1fd8f6..328e0cf42b3c 100644
--- a/net/ipv4/tcp_timer.c
+++ b/net/ipv4/tcp_timer.c
@@ -48,7 +48,7 @@ static void tcp_write_err(struct sock *sk)
 	sk->sk_error_report(sk);
 
 	tcp_done(sk);
-	NET_INC_STATS_BH(LINUX_MIB_TCPABORTONTIMEOUT);
+	NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPABORTONTIMEOUT);
 }
 
 /* Do not allow orphaned sockets to eat all our resources.
@@ -89,7 +89,7 @@ static int tcp_out_of_resources(struct sock *sk, int do_reset)
 		if (do_reset)
 			tcp_send_active_reset(sk, GFP_ATOMIC);
 		tcp_done(sk);
-		NET_INC_STATS_BH(LINUX_MIB_TCPABORTONMEMORY);
+		NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPABORTONMEMORY);
 		return 1;
 	}
 	return 0;
@@ -179,7 +179,7 @@ static void tcp_delack_timer(unsigned long data)
 	if (sock_owned_by_user(sk)) {
 		/* Try again later. */
 		icsk->icsk_ack.blocked = 1;
-		NET_INC_STATS_BH(LINUX_MIB_DELAYEDACKLOCKED);
+		NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_DELAYEDACKLOCKED);
 		sk_reset_timer(sk, &icsk->icsk_delack_timer, jiffies + TCP_DELACK_MIN);
 		goto out_unlock;
 	}
@@ -198,7 +198,7 @@ static void tcp_delack_timer(unsigned long data)
 	if (!skb_queue_empty(&tp->ucopy.prequeue)) {
 		struct sk_buff *skb;
 
-		NET_INC_STATS_BH(LINUX_MIB_TCPSCHEDULERFAILED);
+		NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPSCHEDULERFAILED);
 
 		while ((skb = __skb_dequeue(&tp->ucopy.prequeue)) != NULL)
 			sk->sk_backlog_rcv(sk, skb);
@@ -218,7 +218,7 @@ static void tcp_delack_timer(unsigned long data)
 			icsk->icsk_ack.ato      = TCP_ATO_MIN;
 		}
 		tcp_send_ack(sk);
-		NET_INC_STATS_BH(LINUX_MIB_DELAYEDACKS);
+		NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_DELAYEDACKS);
 	}
 	TCP_CHECK_TIMER(sk);
 
@@ -346,7 +346,7 @@ static void tcp_retransmit_timer(struct sock *sk)
 		} else {
 			mib_idx = LINUX_MIB_TCPTIMEOUTS;
 		}
-		NET_INC_STATS_BH(mib_idx);
+		NET_INC_STATS_BH(sock_net(sk), mib_idx);
 	}
 
 	if (tcp_use_frto(sk)) {
diff --git a/net/ipv6/inet6_hashtables.c b/net/ipv6/inet6_hashtables.c
index a9cc8ab33a49..00a8a5f9380c 100644
--- a/net/ipv6/inet6_hashtables.c
+++ b/net/ipv6/inet6_hashtables.c
@@ -210,11 +210,11 @@ unique:
 
 	if (twp != NULL) {
 		*twp = tw;
-		NET_INC_STATS_BH(LINUX_MIB_TIMEWAITRECYCLED);
+		NET_INC_STATS_BH(twsk_net(tw), LINUX_MIB_TIMEWAITRECYCLED);
 	} else if (tw != NULL) {
 		/* Silly. Should hash-dance instead... */
 		inet_twsk_deschedule(tw, death_row);
-		NET_INC_STATS_BH(LINUX_MIB_TIMEWAITRECYCLED);
+		NET_INC_STATS_BH(twsk_net(tw), LINUX_MIB_TIMEWAITRECYCLED);
 
 		inet_twsk_put(tw);
 	}
diff --git a/net/ipv6/syncookies.c b/net/ipv6/syncookies.c
index 3ecc1157994e..6a68eeb7bbf8 100644
--- a/net/ipv6/syncookies.c
+++ b/net/ipv6/syncookies.c
@@ -137,7 +137,7 @@ __u32 cookie_v6_init_sequence(struct sock *sk, struct sk_buff *skb, __u16 *mssp)
 		;
 	*mssp = msstab[mssind] + 1;
 
-	NET_INC_STATS_BH(LINUX_MIB_SYNCOOKIESSENT);
+	NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_SYNCOOKIESSENT);
 
 	return secure_tcp_syn_cookie(&iph->saddr, &iph->daddr, th->source,
 				     th->dest, ntohl(th->seq),
@@ -177,11 +177,11 @@ struct sock *cookie_v6_check(struct sock *sk, struct sk_buff *skb)
 
 	if (time_after(jiffies, tp->last_synq_overflow + TCP_TIMEOUT_INIT) ||
 		(mss = cookie_check(skb, cookie)) == 0) {
-		NET_INC_STATS_BH(LINUX_MIB_SYNCOOKIESFAILED);
+		NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_SYNCOOKIESFAILED);
 		goto out;
 	}
 
-	NET_INC_STATS_BH(LINUX_MIB_SYNCOOKIESRECV);
+	NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_SYNCOOKIESRECV);
 
 	/* check for timestamp cookie support */
 	memset(&tcp_opt, 0, sizeof(tcp_opt));
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index d58b83ac06fb..ca5b93a5c02a 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -340,7 +340,7 @@ static void tcp_v6_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
 
 	bh_lock_sock(sk);
 	if (sock_owned_by_user(sk))
-		NET_INC_STATS_BH(LINUX_MIB_LOCKDROPPEDICMPS);
+		NET_INC_STATS_BH(net, LINUX_MIB_LOCKDROPPEDICMPS);
 
 	if (sk->sk_state == TCP_CLOSE)
 		goto out;
@@ -349,7 +349,7 @@ static void tcp_v6_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
 	seq = ntohl(th->seq);
 	if (sk->sk_state != TCP_LISTEN &&
 	    !between(seq, tp->snd_una, tp->snd_nxt)) {
-		NET_INC_STATS_BH(LINUX_MIB_OUTOFWINDOWICMPS);
+		NET_INC_STATS_BH(net, LINUX_MIB_OUTOFWINDOWICMPS);
 		goto out;
 	}
 
@@ -424,7 +424,7 @@ static void tcp_v6_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
 		BUG_TRAP(req->sk == NULL);
 
 		if (seq != tcp_rsk(req)->snt_isn) {
-			NET_INC_STATS_BH(LINUX_MIB_OUTOFWINDOWICMPS);
+			NET_INC_STATS_BH(net, LINUX_MIB_OUTOFWINDOWICMPS);
 			goto out;
 		}
 
@@ -1449,9 +1449,9 @@ static struct sock * tcp_v6_syn_recv_sock(struct sock *sk, struct sk_buff *skb,
 	return newsk;
 
 out_overflow:
-	NET_INC_STATS_BH(LINUX_MIB_LISTENOVERFLOWS);
+	NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_LISTENOVERFLOWS);
 out:
-	NET_INC_STATS_BH(LINUX_MIB_LISTENDROPS);
+	NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_LISTENDROPS);
 	if (opt && opt != np->opt)
 		sock_kfree_s(sk, opt, opt->tot_len);
 	dst_release(dst);
diff --git a/net/sctp/input.c b/net/sctp/input.c
index ed8834e7f143..5ed93c05c23e 100644
--- a/net/sctp/input.c
+++ b/net/sctp/input.c
@@ -486,7 +486,7 @@ struct sock *sctp_err_lookup(int family, struct sk_buff *skb,
 	 * servers this needs to be solved differently.
 	 */
 	if (sock_owned_by_user(sk))
-		NET_INC_STATS_BH(LINUX_MIB_LOCKDROPPEDICMPS);
+		NET_INC_STATS_BH(&init_net, LINUX_MIB_LOCKDROPPEDICMPS);
 
 	*app = asoc;
 	*tpp = transport;
-- 
cgit v1.2.3


From 6f67c817fcfd94f5ca0f14b114b7fa25c0210c8b Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Wed, 16 Jul 2008 20:31:39 -0700
Subject: mib: add net to NET_INC_STATS_USER

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip.h | 2 +-
 net/ipv4/tcp.c   | 6 +++---
 2 files changed, 4 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/include/net/ip.h b/include/net/ip.h
index 21df167a8e5b..79d13192ecc8 100644
--- a/include/net/ip.h
+++ b/include/net/ip.h
@@ -163,7 +163,7 @@ DECLARE_SNMP_STAT(struct ipstats_mib, ip_statistics);
 DECLARE_SNMP_STAT(struct linux_mib, net_statistics);
 #define NET_INC_STATS(net, field)	do { (void)net; SNMP_INC_STATS(net_statistics, field); } while (0)
 #define NET_INC_STATS_BH(net, field)	do { (void)net; SNMP_INC_STATS_BH(net_statistics, field); } while (0)
-#define NET_INC_STATS_USER(field) 	SNMP_INC_STATS_USER(net_statistics, field)
+#define NET_INC_STATS_USER(net, field) 	do { (void)net; SNMP_INC_STATS_USER(net_statistics, field); } while (0)
 #define NET_ADD_STATS_BH(field, adnd)	SNMP_ADD_STATS_BH(net_statistics, field, adnd)
 #define NET_ADD_STATS_USER(field, adnd)	SNMP_ADD_STATS_USER(net_statistics, field, adnd)
 
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index 9e0e45c37806..dec0b8681119 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -1153,7 +1153,7 @@ static void tcp_prequeue_process(struct sock *sk)
 	struct sk_buff *skb;
 	struct tcp_sock *tp = tcp_sk(sk);
 
-	NET_INC_STATS_USER(LINUX_MIB_TCPPREQUEUED);
+	NET_INC_STATS_USER(sock_net(sk), LINUX_MIB_TCPPREQUEUED);
 
 	/* RX process wants to run with disabled BHs, though it is not
 	 * necessary */
@@ -1793,13 +1793,13 @@ void tcp_close(struct sock *sk, long timeout)
 	 */
 	if (data_was_unread) {
 		/* Unread data was tossed, zap the connection. */
-		NET_INC_STATS_USER(LINUX_MIB_TCPABORTONCLOSE);
+		NET_INC_STATS_USER(sock_net(sk), LINUX_MIB_TCPABORTONCLOSE);
 		tcp_set_state(sk, TCP_CLOSE);
 		tcp_send_active_reset(sk, GFP_KERNEL);
 	} else if (sock_flag(sk, SOCK_LINGER) && !sk->sk_lingertime) {
 		/* Check zero linger _after_ checking for unread data. */
 		sk->sk_prot->disconnect(sk, 0);
-		NET_INC_STATS_USER(LINUX_MIB_TCPABORTONDATA);
+		NET_INC_STATS_USER(sock_net(sk), LINUX_MIB_TCPABORTONDATA);
 	} else if (tcp_close_state(sk)) {
 		/* We FIN if the application ate all the data before
 		 * zapping the connection.
-- 
cgit v1.2.3


From f2bf415cfed703de5ba94d25cdb160920c01fb00 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Wed, 16 Jul 2008 20:32:25 -0700
Subject: mib: add net to NET_ADD_STATS_BH

This one is tricky.

The thing is that this macro is only used when killing tw buckets,
but since this killer is promiscuous wrt to which net each particular
tw belongs to, I have to use it only when NET_NS is off. When the net
namespaces are on, I use the INET_INC_STATS_BH for each bucket.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip.h              |  2 +-
 net/ipv4/inet_timewait_sock.c | 15 ++++++++++++---
 2 files changed, 13 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/include/net/ip.h b/include/net/ip.h
index 79d13192ecc8..a8275b1338d7 100644
--- a/include/net/ip.h
+++ b/include/net/ip.h
@@ -164,7 +164,7 @@ DECLARE_SNMP_STAT(struct linux_mib, net_statistics);
 #define NET_INC_STATS(net, field)	do { (void)net; SNMP_INC_STATS(net_statistics, field); } while (0)
 #define NET_INC_STATS_BH(net, field)	do { (void)net; SNMP_INC_STATS_BH(net_statistics, field); } while (0)
 #define NET_INC_STATS_USER(net, field) 	do { (void)net; SNMP_INC_STATS_USER(net_statistics, field); } while (0)
-#define NET_ADD_STATS_BH(field, adnd)	SNMP_ADD_STATS_BH(net_statistics, field, adnd)
+#define NET_ADD_STATS_BH(net, field, adnd) do { (void)net; SNMP_ADD_STATS_BH(net_statistics, field, adnd); } while (0)
 #define NET_ADD_STATS_USER(field, adnd)	SNMP_ADD_STATS_USER(net_statistics, field, adnd)
 
 extern unsigned long snmp_fold_field(void *mib[], int offt);
diff --git a/net/ipv4/inet_timewait_sock.c b/net/ipv4/inet_timewait_sock.c
index 06006a5ac8b9..75c2def8f9a0 100644
--- a/net/ipv4/inet_timewait_sock.c
+++ b/net/ipv4/inet_timewait_sock.c
@@ -160,6 +160,9 @@ rescan:
 		__inet_twsk_del_dead_node(tw);
 		spin_unlock(&twdr->death_lock);
 		__inet_twsk_kill(tw, twdr->hashinfo);
+#ifdef CONFIG_NET_NS
+		NET_INC_STATS_BH(twsk_net(tw), LINUX_MIB_TIMEWAITED);
+#endif
 		inet_twsk_put(tw);
 		killed++;
 		spin_lock(&twdr->death_lock);
@@ -178,8 +181,9 @@ rescan:
 	}
 
 	twdr->tw_count -= killed;
-	NET_ADD_STATS_BH(LINUX_MIB_TIMEWAITED, killed);
-
+#ifndef CONFIG_NET_NS
+	NET_ADD_STATS_BH(&init_net, LINUX_MIB_TIMEWAITED, killed);
+#endif
 	return ret;
 }
 
@@ -372,6 +376,9 @@ void inet_twdr_twcal_tick(unsigned long data)
 						       &twdr->twcal_row[slot]) {
 				__inet_twsk_del_dead_node(tw);
 				__inet_twsk_kill(tw, twdr->hashinfo);
+#ifdef CONFIG_NET_NS
+				NET_INC_STATS_BH(twsk_net(tw), LINUX_MIB_TIMEWAITKILLED);
+#endif
 				inet_twsk_put(tw);
 				killed++;
 			}
@@ -395,7 +402,9 @@ void inet_twdr_twcal_tick(unsigned long data)
 out:
 	if ((twdr->tw_count -= killed) == 0)
 		del_timer(&twdr->tw_timer);
-	NET_ADD_STATS_BH(LINUX_MIB_TIMEWAITKILLED, killed);
+#ifndef CONFIG_NET_NS
+	NET_ADD_STATS_BH(&init_net, LINUX_MIB_TIMEWAITKILLED, killed);
+#endif
 	spin_unlock(&twdr->death_lock);
 }
 
-- 
cgit v1.2.3


From ed88098e25d77bef3b2ad8c9d8e2ebf454d9ccbf Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Wed, 16 Jul 2008 20:32:45 -0700
Subject: mib: add net to NET_ADD_STATS_USER

Done with NET_XXX_STATS macros :)

To be continued...

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip.h | 2 +-
 net/ipv4/tcp.c   | 6 +++---
 2 files changed, 4 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/include/net/ip.h b/include/net/ip.h
index a8275b1338d7..02924fb4bdbf 100644
--- a/include/net/ip.h
+++ b/include/net/ip.h
@@ -165,7 +165,7 @@ DECLARE_SNMP_STAT(struct linux_mib, net_statistics);
 #define NET_INC_STATS_BH(net, field)	do { (void)net; SNMP_INC_STATS_BH(net_statistics, field); } while (0)
 #define NET_INC_STATS_USER(net, field) 	do { (void)net; SNMP_INC_STATS_USER(net_statistics, field); } while (0)
 #define NET_ADD_STATS_BH(net, field, adnd) do { (void)net; SNMP_ADD_STATS_BH(net_statistics, field, adnd); } while (0)
-#define NET_ADD_STATS_USER(field, adnd)	SNMP_ADD_STATS_USER(net_statistics, field, adnd)
+#define NET_ADD_STATS_USER(net, field, adnd) do { (void)net; SNMP_ADD_STATS_USER(net_statistics, field, adnd); } while (0)
 
 extern unsigned long snmp_fold_field(void *mib[], int offt);
 extern int snmp_mib_init(void *ptr[2], size_t mibsize);
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index dec0b8681119..eec8cf7c0247 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -1475,7 +1475,7 @@ int tcp_recvmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg,
 			/* __ Restore normal policy in scheduler __ */
 
 			if ((chunk = len - tp->ucopy.len) != 0) {
-				NET_ADD_STATS_USER(LINUX_MIB_TCPDIRECTCOPYFROMBACKLOG, chunk);
+				NET_ADD_STATS_USER(sock_net(sk), LINUX_MIB_TCPDIRECTCOPYFROMBACKLOG, chunk);
 				len -= chunk;
 				copied += chunk;
 			}
@@ -1486,7 +1486,7 @@ do_prequeue:
 				tcp_prequeue_process(sk);
 
 				if ((chunk = len - tp->ucopy.len) != 0) {
-					NET_ADD_STATS_USER(LINUX_MIB_TCPDIRECTCOPYFROMPREQUEUE, chunk);
+					NET_ADD_STATS_USER(sock_net(sk), LINUX_MIB_TCPDIRECTCOPYFROMPREQUEUE, chunk);
 					len -= chunk;
 					copied += chunk;
 				}
@@ -1601,7 +1601,7 @@ skip_copy:
 			tcp_prequeue_process(sk);
 
 			if (copied > 0 && (chunk = len - tp->ucopy.len) != 0) {
-				NET_ADD_STATS_USER(LINUX_MIB_TCPDIRECTCOPYFROMPREQUEUE, chunk);
+				NET_ADD_STATS_USER(sock_net(sk), LINUX_MIB_TCPDIRECTCOPYFROMPREQUEUE, chunk);
 				len -= chunk;
 				copied += chunk;
 			}
-- 
cgit v1.2.3


From 9a6d276e85aa3d8f308fc5e8de6892daeb60ae5f Mon Sep 17 00:00:00 2001
From: Neil Horman <nhorman@tuxdriver.com>
Date: Wed, 16 Jul 2008 20:50:49 -0700
Subject: core: add stat to track unresolved discards in neighbor cache

in __neigh_event_send, if we have a neighbour entry which is in
NUD_INCOMPLETE state, we enqueue any outbound frames to that neighbour
to the neighbours arp_queue, which is default capped to a length of 3
skbs.  If that queue exceeds its set length, it will drop an skb on
the queue to enqueue the newly arrived skb.  This results in a drop
for which we have no statistics incremented.  This patch adds an
unresolved_discards stat to /proc/net/stat/ndisc_cache to track these
lost frames.

Signed-off-by: Neil Horman <nhorman@tuxdriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/neighbour.h | 4 +++-
 net/core/neighbour.c    | 8 +++++---
 2 files changed, 8 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/include/net/neighbour.h b/include/net/neighbour.h
index dc420fecafb9..aa4b708654a4 100644
--- a/include/net/neighbour.h
+++ b/include/net/neighbour.h
@@ -75,7 +75,7 @@ struct neigh_statistics
 	unsigned long destroys;		/* number of destroyed neighs */
 	unsigned long hash_grows;	/* number of hash resizes */
 
-	unsigned long res_failed;	/* nomber of failed resolutions */
+	unsigned long res_failed;	/* number of failed resolutions */
 
 	unsigned long lookups;		/* number of lookups */
 	unsigned long hits;		/* number of hits (among lookups) */
@@ -85,6 +85,8 @@ struct neigh_statistics
 
 	unsigned long periodic_gc_runs;	/* number of periodic GC runs */
 	unsigned long forced_gc_runs;	/* number of forced GC runs */
+
+	unsigned long unres_discards;	/* number of unresolved drops */
 };
 
 #define NEIGH_CACHE_STAT_INC(tbl, field)				\
diff --git a/net/core/neighbour.c b/net/core/neighbour.c
index 65f01f71b3f3..f62c8af85d38 100644
--- a/net/core/neighbour.c
+++ b/net/core/neighbour.c
@@ -930,6 +930,7 @@ int __neigh_event_send(struct neighbour *neigh, struct sk_buff *skb)
 				buff = neigh->arp_queue.next;
 				__skb_unlink(buff, &neigh->arp_queue);
 				kfree_skb(buff);
+				NEIGH_CACHE_STAT_INC(neigh->tbl, unres_discards);
 			}
 			__skb_queue_tail(&neigh->arp_queue, skb);
 		}
@@ -2462,12 +2463,12 @@ static int neigh_stat_seq_show(struct seq_file *seq, void *v)
 	struct neigh_statistics *st = v;
 
 	if (v == SEQ_START_TOKEN) {
-		seq_printf(seq, "entries  allocs destroys hash_grows  lookups hits  res_failed  rcv_probes_mcast rcv_probes_ucast  periodic_gc_runs forced_gc_runs\n");
+		seq_printf(seq, "entries  allocs destroys hash_grows  lookups hits  res_failed  rcv_probes_mcast rcv_probes_ucast  periodic_gc_runs forced_gc_runs unresolved_discards\n");
 		return 0;
 	}
 
 	seq_printf(seq, "%08x  %08lx %08lx %08lx  %08lx %08lx  %08lx  "
-			"%08lx %08lx  %08lx %08lx\n",
+			"%08lx %08lx  %08lx %08lx %08lx\n",
 		   atomic_read(&tbl->entries),
 
 		   st->allocs,
@@ -2483,7 +2484,8 @@ static int neigh_stat_seq_show(struct seq_file *seq, void *v)
 		   st->rcv_probes_ucast,
 
 		   st->periodic_gc_runs,
-		   st->forced_gc_runs
+		   st->forced_gc_runs,
+		   st->unres_discards
 		   );
 
 	return 0;
-- 
cgit v1.2.3


From e8a0464cc950972824e2e128028ae3db666ec1ed Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Thu, 17 Jul 2008 00:34:19 -0700
Subject: netdev: Allocate multiple queues for TX.

alloc_netdev_mq() now allocates an array of netdev_queue
structures for TX, based upon the queue_count argument.

Furthermore, all accesses to the TX queues are now vectored
through the netdev_get_tx_queue() and netdev_for_each_tx_queue()
interfaces.  This makes it easy to grep the tree for all
things that want to get to a TX queue of a net device.

Problem spots which are not really multiqueue aware yet, and
only work with one queue, can easily be spotted by grepping
for all netdev_get_tx_queue() calls that pass in a zero index.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/bonding/bond_main.c         |   6 +-
 drivers/net/hamradio/bpqether.c         |   6 +-
 drivers/net/ifb.c                       |  12 ++-
 drivers/net/macvlan.c                   |   6 +-
 drivers/net/wireless/hostap/hostap_hw.c |   6 +-
 include/linux/netdevice.h               |  69 ++++++++-----
 include/net/sch_generic.h               |  37 +++++--
 net/8021q/vlan_dev.c                    |  10 +-
 net/core/dev.c                          |  40 +++++--
 net/core/rtnetlink.c                    |   2 +-
 net/mac80211/main.c                     |   4 +-
 net/mac80211/wme.c                      |  12 +--
 net/netrom/af_netrom.c                  |   6 +-
 net/rose/af_rose.c                      |   6 +-
 net/sched/cls_api.c                     |   4 +-
 net/sched/sch_api.c                     |  32 ++++--
 net/sched/sch_generic.c                 | 178 +++++++++++++++++++++++---------
 net/sched/sch_teql.c                    |  21 ++--
 18 files changed, 320 insertions(+), 137 deletions(-)

(limited to 'include')

diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
index fd87dbe7999a..9737c06045d6 100644
--- a/drivers/net/bonding/bond_main.c
+++ b/drivers/net/bonding/bond_main.c
@@ -5042,7 +5042,9 @@ static int bond_check_params(struct bond_params *params)
 
 static struct lock_class_key bonding_netdev_xmit_lock_key;
 
-static void bond_set_lockdep_class_one(struct netdev_queue *txq)
+static void bond_set_lockdep_class_one(struct net_device *dev,
+				       struct netdev_queue *txq,
+				       void *_unused)
 {
 	lockdep_set_class(&txq->_xmit_lock,
 			  &bonding_netdev_xmit_lock_key);
@@ -5050,7 +5052,7 @@ static void bond_set_lockdep_class_one(struct netdev_queue *txq)
 
 static void bond_set_lockdep_class(struct net_device *dev)
 {
-	bond_set_lockdep_class_one(&dev->tx_queue);
+	netdev_for_each_tx_queue(dev, bond_set_lockdep_class_one, NULL);
 }
 
 /* Create a new bond based on the specified name and bonding parameters.
diff --git a/drivers/net/hamradio/bpqether.c b/drivers/net/hamradio/bpqether.c
index fb186b8c3d4d..b6500b2aacf2 100644
--- a/drivers/net/hamradio/bpqether.c
+++ b/drivers/net/hamradio/bpqether.c
@@ -124,14 +124,16 @@ static LIST_HEAD(bpq_devices);
  */
 static struct lock_class_key bpq_netdev_xmit_lock_key;
 
-static void bpq_set_lockdep_class_one(struct netdev_queue *txq)
+static void bpq_set_lockdep_class_one(struct net_device *dev,
+				      struct netdev_queue *txq,
+				      void *_unused)
 {
 	lockdep_set_class(&txq->_xmit_lock, &bpq_netdev_xmit_lock_key);
 }
 
 static void bpq_set_lockdep_class(struct net_device *dev)
 {
-	bpq_set_lockdep_class_one(&dev->tx_queue);
+	netdev_for_each_tx_queue(dev, bpq_set_lockdep_class_one, NULL);
 }
 
 /* ------------------------------------------------------------------------ */
diff --git a/drivers/net/ifb.c b/drivers/net/ifb.c
index ccbd6554f6eb..897b05e79ed0 100644
--- a/drivers/net/ifb.c
+++ b/drivers/net/ifb.c
@@ -229,14 +229,20 @@ module_param(numifbs, int, 0);
 MODULE_PARM_DESC(numifbs, "Number of ifb devices");
 
 /*
- * dev_ifb->tx_queue.lock is usually taken after dev->rx_queue.lock,
+ * dev_ifb's TX queue lock is usually taken after dev->rx_queue.lock,
  * reversely to e.g. qdisc_lock_tree(). It should be safe until
- * ifb doesn't take dev->tx_queue.lock with dev_ifb->rx_queue.lock.
+ * ifb doesn't take dev's TX queue lock with dev_ifb->rx_queue.lock.
  * But lockdep should know that ifb has different locks from dev.
  */
 static struct lock_class_key ifb_tx_queue_lock_key;
 static struct lock_class_key ifb_rx_queue_lock_key;
 
+static void set_tx_lockdep_key(struct net_device *dev,
+			       struct netdev_queue *txq,
+			       void *_unused)
+{
+	lockdep_set_class(&txq->lock, &ifb_tx_queue_lock_key);
+}
 
 static int __init ifb_init_one(int index)
 {
@@ -258,7 +264,7 @@ static int __init ifb_init_one(int index)
 	if (err < 0)
 		goto err;
 
-	lockdep_set_class(&dev_ifb->tx_queue.lock, &ifb_tx_queue_lock_key);
+	netdev_for_each_tx_queue(dev_ifb, set_tx_lockdep_key, NULL);
 	lockdep_set_class(&dev_ifb->rx_queue.lock, &ifb_rx_queue_lock_key);
 
 	return 0;
diff --git a/drivers/net/macvlan.c b/drivers/net/macvlan.c
index 980001c2cf96..72745ce588c6 100644
--- a/drivers/net/macvlan.c
+++ b/drivers/net/macvlan.c
@@ -285,7 +285,9 @@ static struct lock_class_key macvlan_netdev_xmit_lock_key;
 #define MACVLAN_STATE_MASK \
 	((1<<__LINK_STATE_NOCARRIER) | (1<<__LINK_STATE_DORMANT))
 
-static void macvlan_set_lockdep_class_one(struct netdev_queue *txq)
+static void macvlan_set_lockdep_class_one(struct net_device *dev,
+					  struct netdev_queue *txq,
+					  void *_unused)
 {
 	lockdep_set_class(&txq->_xmit_lock,
 			  &macvlan_netdev_xmit_lock_key);
@@ -293,7 +295,7 @@ static void macvlan_set_lockdep_class_one(struct netdev_queue *txq)
 
 static void macvlan_set_lockdep_class(struct net_device *dev)
 {
-	macvlan_set_lockdep_class_one(&dev->tx_queue);
+	netdev_for_each_tx_queue(dev, macvlan_set_lockdep_class_one, NULL);
 }
 
 static int macvlan_init(struct net_device *dev)
diff --git a/drivers/net/wireless/hostap/hostap_hw.c b/drivers/net/wireless/hostap/hostap_hw.c
index c1f4bb005d92..13d5882f1f21 100644
--- a/drivers/net/wireless/hostap/hostap_hw.c
+++ b/drivers/net/wireless/hostap/hostap_hw.c
@@ -3102,7 +3102,9 @@ static void prism2_clear_set_tim_queue(local_info_t *local)
  */
 static struct lock_class_key hostap_netdev_xmit_lock_key;
 
-static void prism2_set_lockdep_class_one(struct netdev_queue *txq)
+static void prism2_set_lockdep_class_one(struct net_device *dev,
+					 struct netdev_queue *txq,
+					 void *_unused)
 {
 	lockdep_set_class(&txq->_xmit_lock,
 			  &hostap_netdev_xmit_lock_key);
@@ -3110,7 +3112,7 @@ static void prism2_set_lockdep_class_one(struct netdev_queue *txq)
 
 static void prism2_set_lockdep_class(struct net_device *dev)
 {
-	prism2_set_lockdep_class_one(&dev->tx_queue);
+	netdev_for_each_tx_queue(dev, prism2_set_lockdep_class_one, NULL);
 }
 
 static struct net_device *
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 570cf7affa72..f25d4f5a31b0 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -463,7 +463,7 @@ struct netdev_queue {
 	struct Qdisc		*qdisc_sleeping;
 	struct list_head	qdisc_list;
 	struct netdev_queue	*next_sched;
-};
+} ____cacheline_aligned_in_smp;
 
 /*
  *	The DEVICE structure.
@@ -641,7 +641,9 @@ struct net_device
 	unsigned char		broadcast[MAX_ADDR_LEN];	/* hw bcast add	*/
 
 	struct netdev_queue	rx_queue;
-	struct netdev_queue	tx_queue ____cacheline_aligned_in_smp;
+
+	struct netdev_queue	*_tx ____cacheline_aligned_in_smp;
+	unsigned int		num_tx_queues;
 	unsigned long		tx_queue_len;	/* Max frames per queue allowed */
 
 /*
@@ -764,6 +766,25 @@ struct net_device
 #define	NETDEV_ALIGN		32
 #define	NETDEV_ALIGN_CONST	(NETDEV_ALIGN - 1)
 
+static inline
+struct netdev_queue *netdev_get_tx_queue(const struct net_device *dev,
+					 unsigned int index)
+{
+	return &dev->_tx[index];
+}
+
+static inline void netdev_for_each_tx_queue(struct net_device *dev,
+					    void (*f)(struct net_device *,
+						      struct netdev_queue *,
+						      void *),
+					    void *arg)
+{
+	unsigned int i;
+
+	for (i = 0; i < dev->num_tx_queues; i++)
+		f(dev, &dev->_tx[i], arg);
+}
+
 /*
  * Net namespace inlines
  */
@@ -977,7 +998,7 @@ static inline void netif_schedule_queue(struct netdev_queue *txq)
 
 static inline void netif_schedule(struct net_device *dev)
 {
-	netif_schedule_queue(&dev->tx_queue);
+	netif_schedule_queue(netdev_get_tx_queue(dev, 0));
 }
 
 /**
@@ -993,7 +1014,7 @@ static inline void netif_tx_start_queue(struct netdev_queue *dev_queue)
 
 static inline void netif_start_queue(struct net_device *dev)
 {
-	netif_tx_start_queue(&dev->tx_queue);
+	netif_tx_start_queue(netdev_get_tx_queue(dev, 0));
 }
 
 /**
@@ -1017,7 +1038,7 @@ static inline void netif_tx_wake_queue(struct netdev_queue *dev_queue)
 
 static inline void netif_wake_queue(struct net_device *dev)
 {
-	netif_tx_wake_queue(&dev->tx_queue);
+	netif_tx_wake_queue(netdev_get_tx_queue(dev, 0));
 }
 
 /**
@@ -1034,7 +1055,7 @@ static inline void netif_tx_stop_queue(struct netdev_queue *dev_queue)
 
 static inline void netif_stop_queue(struct net_device *dev)
 {
-	netif_tx_stop_queue(&dev->tx_queue);
+	netif_tx_stop_queue(netdev_get_tx_queue(dev, 0));
 }
 
 /**
@@ -1050,7 +1071,7 @@ static inline int netif_tx_queue_stopped(const struct netdev_queue *dev_queue)
 
 static inline int netif_queue_stopped(const struct net_device *dev)
 {
-	return netif_tx_queue_stopped(&dev->tx_queue);
+	return netif_tx_queue_stopped(netdev_get_tx_queue(dev, 0));
 }
 
 /**
@@ -1134,7 +1155,7 @@ static inline void netif_wake_subqueue(struct net_device *dev, u16 queue_index)
 #endif
 	if (test_and_clear_bit(__QUEUE_STATE_XOFF,
 			       &dev->egress_subqueue[queue_index].state))
-		__netif_schedule(&dev->tx_queue);
+		__netif_schedule(netdev_get_tx_queue(dev, 0));
 }
 
 /**
@@ -1430,18 +1451,19 @@ static inline void __netif_tx_lock(struct netdev_queue *txq, int cpu)
 
 static inline void netif_tx_lock(struct net_device *dev)
 {
-	__netif_tx_lock(&dev->tx_queue, smp_processor_id());
-}
+	int cpu = smp_processor_id();
+	unsigned int i;
 
-static inline void __netif_tx_lock_bh(struct netdev_queue *txq)
-{
-	spin_lock_bh(&txq->_xmit_lock);
-	txq->xmit_lock_owner = smp_processor_id();
+	for (i = 0; i < dev->num_tx_queues; i++) {
+		struct netdev_queue *txq = netdev_get_tx_queue(dev, i);
+		__netif_tx_lock(txq, cpu);
+	}
 }
 
 static inline void netif_tx_lock_bh(struct net_device *dev)
 {
-	__netif_tx_lock_bh(&dev->tx_queue);
+	local_bh_disable();
+	netif_tx_lock(dev);
 }
 
 static inline int __netif_tx_trylock(struct netdev_queue *txq)
@@ -1454,7 +1476,7 @@ static inline int __netif_tx_trylock(struct netdev_queue *txq)
 
 static inline int netif_tx_trylock(struct net_device *dev)
 {
-	return __netif_tx_trylock(&dev->tx_queue);
+	return __netif_tx_trylock(netdev_get_tx_queue(dev, 0));
 }
 
 static inline void __netif_tx_unlock(struct netdev_queue *txq)
@@ -1465,18 +1487,19 @@ static inline void __netif_tx_unlock(struct netdev_queue *txq)
 
 static inline void netif_tx_unlock(struct net_device *dev)
 {
-	__netif_tx_unlock(&dev->tx_queue);
-}
+	unsigned int i;
+
+	for (i = 0; i < dev->num_tx_queues; i++) {
+		struct netdev_queue *txq = netdev_get_tx_queue(dev, i);
+		__netif_tx_unlock(txq);
+	}
 
-static inline void __netif_tx_unlock_bh(struct netdev_queue *txq)
-{
-	txq->xmit_lock_owner = -1;
-	spin_unlock_bh(&txq->_xmit_lock);
 }
 
 static inline void netif_tx_unlock_bh(struct net_device *dev)
 {
-	__netif_tx_unlock_bh(&dev->tx_queue);
+	netif_tx_unlock(dev);
+	local_bh_enable();
 }
 
 #define HARD_TX_LOCK(dev, txq, cpu) {			\
diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h
index 5ba66b555578..b47f556c66f8 100644
--- a/include/net/sch_generic.h
+++ b/include/net/sch_generic.h
@@ -230,32 +230,47 @@ extern void tcf_destroy_chain(struct tcf_proto **fl);
 /* Reset all TX qdiscs of a device.  */
 static inline void qdisc_reset_all_tx(struct net_device *dev)
 {
-	qdisc_reset(dev->tx_queue.qdisc);
+	unsigned int i;
+	for (i = 0; i < dev->num_tx_queues; i++)
+		qdisc_reset(netdev_get_tx_queue(dev, i)->qdisc);
 }
 
 /* Are all TX queues of the device empty?  */
 static inline bool qdisc_all_tx_empty(const struct net_device *dev)
 {
-	const struct netdev_queue *txq = &dev->tx_queue;
-	const struct Qdisc *q = txq->qdisc;
+	unsigned int i;
+	for (i = 0; i < dev->num_tx_queues; i++) {
+		struct netdev_queue *txq = netdev_get_tx_queue(dev, i);
+		const struct Qdisc *q = txq->qdisc;
 
-	return (q->q.qlen == 0);
+		if (q->q.qlen)
+			return false;
+	}
+	return true;
 }
 
 /* Are any of the TX qdiscs changing?  */
 static inline bool qdisc_tx_changing(struct net_device *dev)
 {
-	struct netdev_queue *txq = &dev->tx_queue;
-
-	return (txq->qdisc != txq->qdisc_sleeping);
+	unsigned int i;
+	for (i = 0; i < dev->num_tx_queues; i++) {
+		struct netdev_queue *txq = netdev_get_tx_queue(dev, i);
+		if (txq->qdisc != txq->qdisc_sleeping)
+			return true;
+	}
+	return false;
 }
 
-/* Is the device using the noop qdisc?  */
+/* Is the device using the noop qdisc on all queues?  */
 static inline bool qdisc_tx_is_noop(const struct net_device *dev)
 {
-	const struct netdev_queue *txq = &dev->tx_queue;
-
-	return (txq->qdisc == &noop_qdisc);
+	unsigned int i;
+	for (i = 0; i < dev->num_tx_queues; i++) {
+		struct netdev_queue *txq = netdev_get_tx_queue(dev, i);
+		if (txq->qdisc != &noop_qdisc)
+			return false;
+	}
+	return true;
 }
 
 static inline int __qdisc_enqueue_tail(struct sk_buff *skb, struct Qdisc *sch,
diff --git a/net/8021q/vlan_dev.c b/net/8021q/vlan_dev.c
index 6b985f23fd9f..f42bc2b26b85 100644
--- a/net/8021q/vlan_dev.c
+++ b/net/8021q/vlan_dev.c
@@ -570,16 +570,18 @@ static void vlan_dev_set_rx_mode(struct net_device *vlan_dev)
  */
 static struct lock_class_key vlan_netdev_xmit_lock_key;
 
-static void vlan_dev_set_lockdep_one(struct netdev_queue *txq,
-				     int subclass)
+static void vlan_dev_set_lockdep_one(struct net_device *dev,
+				     struct netdev_queue *txq,
+				     void *_subclass)
 {
 	lockdep_set_class_and_subclass(&txq->_xmit_lock,
-				       &vlan_netdev_xmit_lock_key, subclass);
+				       &vlan_netdev_xmit_lock_key,
+				       *(int *)_subclass);
 }
 
 static void vlan_dev_set_lockdep_class(struct net_device *dev, int subclass)
 {
-	vlan_dev_set_lockdep_one(&dev->tx_queue, subclass);
+	netdev_for_each_tx_queue(dev, vlan_dev_set_lockdep_one, &subclass);
 }
 
 static const struct header_ops vlan_header_ops = {
diff --git a/net/core/dev.c b/net/core/dev.c
index 9b49f74a9820..69378f250695 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -1666,6 +1666,12 @@ out_kfree_skb:
  *          --BLG
  */
 
+static struct netdev_queue *dev_pick_tx(struct net_device *dev,
+					struct sk_buff *skb)
+{
+	return netdev_get_tx_queue(dev, 0);
+}
+
 int dev_queue_xmit(struct sk_buff *skb)
 {
 	struct net_device *dev = skb->dev;
@@ -1702,7 +1708,7 @@ int dev_queue_xmit(struct sk_buff *skb)
 	}
 
 gso:
-	txq = &dev->tx_queue;
+	txq = dev_pick_tx(dev, skb);
 	spin_lock_prefetch(&txq->lock);
 
 	/* Disable soft irqs for various locks below. Also
@@ -3788,8 +3794,9 @@ static void rollback_registered(struct net_device *dev)
 	dev_put(dev);
 }
 
-static void __netdev_init_queue_locks_one(struct netdev_queue *dev_queue,
-					  struct net_device *dev)
+static void __netdev_init_queue_locks_one(struct net_device *dev,
+					  struct netdev_queue *dev_queue,
+					  void *_unused)
 {
 	spin_lock_init(&dev_queue->_xmit_lock);
 	netdev_set_lockdep_class(&dev_queue->_xmit_lock, dev->type);
@@ -3798,8 +3805,8 @@ static void __netdev_init_queue_locks_one(struct netdev_queue *dev_queue,
 
 static void netdev_init_queue_locks(struct net_device *dev)
 {
-	__netdev_init_queue_locks_one(&dev->tx_queue, dev);
-	__netdev_init_queue_locks_one(&dev->rx_queue, dev);
+	netdev_for_each_tx_queue(dev, __netdev_init_queue_locks_one, NULL);
+	__netdev_init_queue_locks_one(dev, &dev->rx_queue, NULL);
 }
 
 /**
@@ -4119,7 +4126,8 @@ static struct net_device_stats *internal_stats(struct net_device *dev)
 }
 
 static void netdev_init_one_queue(struct net_device *dev,
-				  struct netdev_queue *queue)
+				  struct netdev_queue *queue,
+				  void *_unused)
 {
 	spin_lock_init(&queue->lock);
 	queue->dev = dev;
@@ -4127,8 +4135,8 @@ static void netdev_init_one_queue(struct net_device *dev,
 
 static void netdev_init_queues(struct net_device *dev)
 {
-	netdev_init_one_queue(dev, &dev->rx_queue);
-	netdev_init_one_queue(dev, &dev->tx_queue);
+	netdev_init_one_queue(dev, &dev->rx_queue, NULL);
+	netdev_for_each_tx_queue(dev, netdev_init_one_queue, NULL);
 }
 
 /**
@@ -4145,9 +4153,10 @@ static void netdev_init_queues(struct net_device *dev)
 struct net_device *alloc_netdev_mq(int sizeof_priv, const char *name,
 		void (*setup)(struct net_device *), unsigned int queue_count)
 {
-	void *p;
+	struct netdev_queue *tx;
 	struct net_device *dev;
 	int alloc_size;
+	void *p;
 
 	BUG_ON(strlen(name) >= sizeof(dev->name));
 
@@ -4167,11 +4176,22 @@ struct net_device *alloc_netdev_mq(int sizeof_priv, const char *name,
 		return NULL;
 	}
 
+	tx = kzalloc(sizeof(struct netdev_queue) * queue_count, GFP_KERNEL);
+	if (!tx) {
+		printk(KERN_ERR "alloc_netdev: Unable to allocate "
+		       "tx qdiscs.\n");
+		kfree(p);
+		return NULL;
+	}
+
 	dev = (struct net_device *)
 		(((long)p + NETDEV_ALIGN_CONST) & ~NETDEV_ALIGN_CONST);
 	dev->padded = (char *)dev - (char *)p;
 	dev_net_set(dev, &init_net);
 
+	dev->_tx = tx;
+	dev->num_tx_queues = queue_count;
+
 	if (sizeof_priv) {
 		dev->priv = ((char *)dev +
 			     ((sizeof(struct net_device) +
@@ -4205,6 +4225,8 @@ void free_netdev(struct net_device *dev)
 {
 	release_net(dev_net(dev));
 
+	kfree(dev->_tx);
+
 	/*  Compatibility with error handling in drivers */
 	if (dev->reg_state == NETREG_UNINITIALIZED) {
 		kfree((char *)dev - dev->padded);
diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c
index 8ef9f1db610e..71edb8b36341 100644
--- a/net/core/rtnetlink.c
+++ b/net/core/rtnetlink.c
@@ -636,7 +636,7 @@ static int rtnl_fill_ifinfo(struct sk_buff *skb, struct net_device *dev,
 	if (dev->master)
 		NLA_PUT_U32(skb, IFLA_MASTER, dev->master->ifindex);
 
-	txq = &dev->tx_queue;
+	txq = netdev_get_tx_queue(dev, 0);
 	if (txq->qdisc_sleeping)
 		NLA_PUT_STRING(skb, IFLA_QDISC, txq->qdisc_sleeping->ops->id);
 
diff --git a/net/mac80211/main.c b/net/mac80211/main.c
index af0056e7e5b3..b486e634f4fe 100644
--- a/net/mac80211/main.c
+++ b/net/mac80211/main.c
@@ -621,7 +621,7 @@ int ieee80211_start_tx_ba_session(struct ieee80211_hw *hw, u8 *ra, u16 tid)
 
 	/* ensure that TX flow won't interrupt us
 	 * until the end of the call to requeue function */
-	txq = &local->mdev->tx_queue;
+	txq = netdev_get_tx_queue(local->mdev, 0);
 	spin_lock_bh(&txq->lock);
 
 	/* create a new queue for this aggregation */
@@ -862,7 +862,7 @@ void ieee80211_stop_tx_ba_cb(struct ieee80211_hw *hw, u8 *ra, u8 tid)
 
 	/* avoid ordering issues: we are the only one that can modify
 	 * the content of the qdiscs */
-	txq = &local->mdev->tx_queue;
+	txq = netdev_get_tx_queue(local->mdev, 0);
 	spin_lock_bh(&txq->lock);
 	/* remove the queue for this aggregation */
 	ieee80211_ht_agg_queue_remove(local, sta, tid, 1);
diff --git a/net/mac80211/wme.c b/net/mac80211/wme.c
index 6ae43a3c7726..f014cd38c2d0 100644
--- a/net/mac80211/wme.c
+++ b/net/mac80211/wme.c
@@ -574,7 +574,7 @@ static struct Qdisc_ops wme_qdisc_ops __read_mostly =
 
 void ieee80211_install_qdisc(struct net_device *dev)
 {
-	struct netdev_queue *txq = &dev->tx_queue;
+	struct netdev_queue *txq = netdev_get_tx_queue(dev, 0);
 	struct Qdisc *qdisc;
 
 	qdisc = qdisc_create_dflt(dev, txq,
@@ -596,7 +596,7 @@ void ieee80211_install_qdisc(struct net_device *dev)
 
 int ieee80211_qdisc_installed(struct net_device *dev)
 {
-	struct netdev_queue *txq = &dev->tx_queue;
+	struct netdev_queue *txq = netdev_get_tx_queue(dev, 0);
 
 	return txq->qdisc_sleeping->ops == &wme_qdisc_ops;
 }
@@ -617,7 +617,7 @@ int ieee80211_ht_agg_queue_add(struct ieee80211_local *local,
 			struct sta_info *sta, u16 tid)
 {
 	int i;
-	struct netdev_queue *txq = &local->mdev->tx_queue;
+	struct netdev_queue *txq = netdev_get_tx_queue(local->mdev, 0);
 	struct ieee80211_sched_data *q =
 			qdisc_priv(txq->qdisc_sleeping);
 	DECLARE_MAC_BUF(mac);
@@ -652,14 +652,14 @@ int ieee80211_ht_agg_queue_add(struct ieee80211_local *local,
 }
 
 /**
- * the caller needs to hold local->mdev->tx_queue.lock
+ * the caller needs to hold netdev_get_tx_queue(local->mdev, X)->lock
  */
 void ieee80211_ht_agg_queue_remove(struct ieee80211_local *local,
 				   struct sta_info *sta, u16 tid,
 				   u8 requeue)
 {
 	struct ieee80211_hw *hw = &local->hw;
-	struct netdev_queue *txq = &local->mdev->tx_queue;
+	struct netdev_queue *txq = netdev_get_tx_queue(local->mdev, 0);
 	struct ieee80211_sched_data *q =
 		qdisc_priv(txq->qdisc_sleeping);
 	int agg_queue = sta->tid_to_tx_q[tid];
@@ -676,7 +676,7 @@ void ieee80211_ht_agg_queue_remove(struct ieee80211_local *local,
 
 void ieee80211_requeue(struct ieee80211_local *local, int queue)
 {
-	struct netdev_queue *txq = &local->mdev->tx_queue;
+	struct netdev_queue *txq = netdev_get_tx_queue(local->mdev, 0);
 	struct Qdisc *root_qd = txq->qdisc_sleeping;
 	struct ieee80211_sched_data *q = qdisc_priv(root_qd);
 	struct Qdisc *qdisc = q->queues[queue];
diff --git a/net/netrom/af_netrom.c b/net/netrom/af_netrom.c
index 819afc449e1e..d41be0d66eb0 100644
--- a/net/netrom/af_netrom.c
+++ b/net/netrom/af_netrom.c
@@ -74,14 +74,16 @@ static const struct proto_ops nr_proto_ops;
  */
 static struct lock_class_key nr_netdev_xmit_lock_key;
 
-static void nr_set_lockdep_one(struct netdev_queue *txq)
+static void nr_set_lockdep_one(struct net_device *dev,
+			       struct netdev_queue *txq,
+			       void *_unused)
 {
 	lockdep_set_class(&txq->_xmit_lock, &nr_netdev_xmit_lock_key);
 }
 
 static void nr_set_lockdep_key(struct net_device *dev)
 {
-	nr_set_lockdep_one(&dev->tx_queue);
+	netdev_for_each_tx_queue(dev, nr_set_lockdep_one, NULL);
 }
 
 /*
diff --git a/net/rose/af_rose.c b/net/rose/af_rose.c
index 7dbbc0891623..f3a691f34909 100644
--- a/net/rose/af_rose.c
+++ b/net/rose/af_rose.c
@@ -75,14 +75,16 @@ ax25_address rose_callsign;
  */
 static struct lock_class_key rose_netdev_xmit_lock_key;
 
-static void rose_set_lockdep_one(struct netdev_queue *txq)
+static void rose_set_lockdep_one(struct net_device *dev,
+				 struct netdev_queue *txq,
+				 void *_unused)
 {
 	lockdep_set_class(&txq->_xmit_lock, &rose_netdev_xmit_lock_key);
 }
 
 static void rose_set_lockdep_key(struct net_device *dev)
 {
-	rose_set_lockdep_one(&dev->tx_queue);
+	netdev_for_each_tx_queue(dev, rose_set_lockdep_one, NULL);
 }
 
 /*
diff --git a/net/sched/cls_api.c b/net/sched/cls_api.c
index b483bbea6118..d0b0a9b14394 100644
--- a/net/sched/cls_api.c
+++ b/net/sched/cls_api.c
@@ -166,7 +166,7 @@ replay:
 
 	/* Find qdisc */
 	if (!parent) {
-		struct netdev_queue *dev_queue = &dev->tx_queue;
+		struct netdev_queue *dev_queue = netdev_get_tx_queue(dev, 0);
 		q = dev_queue->qdisc_sleeping;
 		parent = q->handle;
 	} else {
@@ -410,7 +410,7 @@ static int tc_dump_tfilter(struct sk_buff *skb, struct netlink_callback *cb)
 	if ((dev = dev_get_by_index(&init_net, tcm->tcm_ifindex)) == NULL)
 		return skb->len;
 
-	dev_queue = &dev->tx_queue;
+	dev_queue = netdev_get_tx_queue(dev, 0);
 	if (!tcm->tcm_parent)
 		q = dev_queue->qdisc_sleeping;
 	else
diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c
index 95873f8dd37c..830ccc544a15 100644
--- a/net/sched/sch_api.c
+++ b/net/sched/sch_api.c
@@ -183,9 +183,8 @@ EXPORT_SYMBOL(unregister_qdisc);
    (root qdisc, all its children, children of children etc.)
  */
 
-struct Qdisc *qdisc_lookup(struct net_device *dev, u32 handle)
+static struct Qdisc *__qdisc_lookup(struct netdev_queue *dev_queue, u32 handle)
 {
-	struct netdev_queue *dev_queue = &dev->tx_queue;
 	struct Qdisc *q;
 
 	list_for_each_entry(q, &dev_queue->qdisc_list, list) {
@@ -195,6 +194,19 @@ struct Qdisc *qdisc_lookup(struct net_device *dev, u32 handle)
 	return NULL;
 }
 
+struct Qdisc *qdisc_lookup(struct net_device *dev, u32 handle)
+{
+	unsigned int i;
+
+	for (i = 0; i < dev->num_tx_queues; i++) {
+		struct netdev_queue *txq = netdev_get_tx_queue(dev, i);
+		struct Qdisc *q = __qdisc_lookup(txq, handle);
+		if (q)
+			return q;
+	}
+	return NULL;
+}
+
 static struct Qdisc *qdisc_leaf(struct Qdisc *p, u32 classid)
 {
 	unsigned long cl;
@@ -462,7 +474,7 @@ dev_graft_qdisc(struct net_device *dev, struct Qdisc *qdisc)
 		}
 
 	} else {
-		dev_queue = &dev->tx_queue;
+		dev_queue = netdev_get_tx_queue(dev, 0);
 		oqdisc = dev_queue->qdisc_sleeping;
 
 		/* Prune old scheduler */
@@ -742,7 +754,8 @@ static int tc_get_qdisc(struct sk_buff *skb, struct nlmsghdr *n, void *arg)
 				q = dev->rx_queue.qdisc;
 			}
 		} else {
-			struct netdev_queue *dev_queue = &dev->tx_queue;
+			struct netdev_queue *dev_queue;
+			dev_queue = netdev_get_tx_queue(dev, 0);
 			q = dev_queue->qdisc_sleeping;
 		}
 		if (!q)
@@ -817,7 +830,8 @@ replay:
 				q = dev->rx_queue.qdisc;
 			}
 		} else {
-			struct netdev_queue *dev_queue = &dev->tx_queue;
+			struct netdev_queue *dev_queue;
+			dev_queue = netdev_get_tx_queue(dev, 0);
 			q = dev_queue->qdisc_sleeping;
 		}
 
@@ -899,7 +913,7 @@ create_n_graft:
 				 tcm->tcm_parent, tcm->tcm_parent,
 				 tca, &err);
 	else
-		q = qdisc_create(dev, &dev->tx_queue,
+		q = qdisc_create(dev, netdev_get_tx_queue(dev, 0),
 				 tcm->tcm_parent, tcm->tcm_handle,
 				 tca, &err);
 	if (q == NULL) {
@@ -1025,7 +1039,7 @@ static int tc_dump_qdisc(struct sk_buff *skb, struct netlink_callback *cb)
 		if (idx > s_idx)
 			s_q_idx = 0;
 		q_idx = 0;
-		dev_queue = &dev->tx_queue;
+		dev_queue = netdev_get_tx_queue(dev, 0);
 		list_for_each_entry(q, &dev_queue->qdisc_list, list) {
 			if (q_idx < s_q_idx) {
 				q_idx++;
@@ -1098,7 +1112,7 @@ static int tc_ctl_tclass(struct sk_buff *skb, struct nlmsghdr *n, void *arg)
 
 	/* Step 1. Determine qdisc handle X:0 */
 
-	dev_queue = &dev->tx_queue;
+	dev_queue = netdev_get_tx_queue(dev, 0);
 	if (pid != TC_H_ROOT) {
 		u32 qid1 = TC_H_MAJ(pid);
 
@@ -1275,7 +1289,7 @@ static int tc_dump_tclass(struct sk_buff *skb, struct netlink_callback *cb)
 	s_t = cb->args[0];
 	t = 0;
 
-	dev_queue = &dev->tx_queue;
+	dev_queue = netdev_get_tx_queue(dev, 0);
 	list_for_each_entry(q, &dev_queue->qdisc_list, list) {
 		if (t < s_t || !q->ops->cl_ops ||
 		    (tcm->tcm_parent &&
diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c
index 243de935b182..4e2b865cbba0 100644
--- a/net/sched/sch_generic.c
+++ b/net/sched/sch_generic.c
@@ -40,20 +40,30 @@
  */
 
 void qdisc_lock_tree(struct net_device *dev)
-	__acquires(dev->tx_queue.lock)
 	__acquires(dev->rx_queue.lock)
 {
-	spin_lock_bh(&dev->tx_queue.lock);
+	unsigned int i;
+
+	local_bh_disable();
+	for (i = 0; i < dev->num_tx_queues; i++) {
+		struct netdev_queue *txq = netdev_get_tx_queue(dev, i);
+		spin_lock(&txq->lock);
+	}
 	spin_lock(&dev->rx_queue.lock);
 }
 EXPORT_SYMBOL(qdisc_lock_tree);
 
 void qdisc_unlock_tree(struct net_device *dev)
 	__releases(dev->rx_queue.lock)
-	__releases(dev->tx_queue.lock)
 {
+	unsigned int i;
+
 	spin_unlock(&dev->rx_queue.lock);
-	spin_unlock_bh(&dev->tx_queue.lock);
+	for (i = 0; i < dev->num_tx_queues; i++) {
+		struct netdev_queue *txq = netdev_get_tx_queue(dev, i);
+		spin_unlock(&txq->lock);
+	}
+	local_bh_enable();
 }
 EXPORT_SYMBOL(qdisc_unlock_tree);
 
@@ -212,22 +222,37 @@ void __qdisc_run(struct netdev_queue *txq)
 static void dev_watchdog(unsigned long arg)
 {
 	struct net_device *dev = (struct net_device *)arg;
-	struct netdev_queue *txq = &dev->tx_queue;
 
 	netif_tx_lock(dev);
-	if (txq->qdisc != &noop_qdisc) {
+	if (!qdisc_tx_is_noop(dev)) {
 		if (netif_device_present(dev) &&
 		    netif_running(dev) &&
 		    netif_carrier_ok(dev)) {
-			if (netif_queue_stopped(dev) &&
-			    time_after(jiffies, dev->trans_start + dev->watchdog_timeo)) {
+			int some_queue_stopped = 0;
+			unsigned int i;
+
+			for (i = 0; i < dev->num_tx_queues; i++) {
+				struct netdev_queue *txq;
+
+				txq = netdev_get_tx_queue(dev, i);
+				if (netif_tx_queue_stopped(txq)) {
+					some_queue_stopped = 1;
+					break;
+				}
+			}
 
-				printk(KERN_INFO "NETDEV WATCHDOG: %s: transmit timed out\n",
+			if (some_queue_stopped &&
+			    time_after(jiffies, (dev->trans_start +
+						 dev->watchdog_timeo))) {
+				printk(KERN_INFO "NETDEV WATCHDOG: %s: "
+				       "transmit timed out\n",
 				       dev->name);
 				dev->tx_timeout(dev);
 				WARN_ON_ONCE(1);
 			}
-			if (!mod_timer(&dev->watchdog_timer, round_jiffies(jiffies + dev->watchdog_timeo)))
+			if (!mod_timer(&dev->watchdog_timer,
+				       round_jiffies(jiffies +
+						     dev->watchdog_timeo)))
 				dev_hold(dev);
 		}
 	}
@@ -542,9 +567,55 @@ void qdisc_destroy(struct Qdisc *qdisc)
 }
 EXPORT_SYMBOL(qdisc_destroy);
 
+static bool dev_all_qdisc_sleeping_noop(struct net_device *dev)
+{
+	unsigned int i;
+
+	for (i = 0; i < dev->num_tx_queues; i++) {
+		struct netdev_queue *txq = netdev_get_tx_queue(dev, i);
+
+		if (txq->qdisc_sleeping != &noop_qdisc)
+			return false;
+	}
+	return true;
+}
+
+static void attach_one_default_qdisc(struct net_device *dev,
+				     struct netdev_queue *dev_queue,
+				     void *_unused)
+{
+	struct Qdisc *qdisc;
+
+	if (dev->tx_queue_len) {
+		qdisc = qdisc_create_dflt(dev, dev_queue,
+					  &pfifo_fast_ops, TC_H_ROOT);
+		if (!qdisc) {
+			printk(KERN_INFO "%s: activation failed\n", dev->name);
+			return;
+		}
+		list_add_tail(&qdisc->list, &dev_queue->qdisc_list);
+	} else {
+		qdisc =  &noqueue_qdisc;
+	}
+	dev_queue->qdisc_sleeping = qdisc;
+}
+
+static void transition_one_qdisc(struct net_device *dev,
+				 struct netdev_queue *dev_queue,
+				 void *_need_watchdog)
+{
+	int *need_watchdog_p = _need_watchdog;
+
+	spin_lock_bh(&dev_queue->lock);
+	rcu_assign_pointer(dev_queue->qdisc, dev_queue->qdisc_sleeping);
+	if (dev_queue->qdisc != &noqueue_qdisc)
+		*need_watchdog_p = 1;
+	spin_unlock_bh(&dev_queue->lock);
+}
+
 void dev_activate(struct net_device *dev)
 {
-	struct netdev_queue *txq = &dev->tx_queue;
+	int need_watchdog;
 
 	/* No queueing discipline is attached to device;
 	   create default one i.e. pfifo_fast for devices,
@@ -552,39 +623,27 @@ void dev_activate(struct net_device *dev)
 	   virtual interfaces
 	 */
 
-	if (txq->qdisc_sleeping == &noop_qdisc) {
-		struct Qdisc *qdisc;
-		if (dev->tx_queue_len) {
-			qdisc = qdisc_create_dflt(dev, txq,
-						  &pfifo_fast_ops,
-						  TC_H_ROOT);
-			if (qdisc == NULL) {
-				printk(KERN_INFO "%s: activation failed\n", dev->name);
-				return;
-			}
-			list_add_tail(&qdisc->list, &txq->qdisc_list);
-		} else {
-			qdisc =  &noqueue_qdisc;
-		}
-		txq->qdisc_sleeping = qdisc;
-	}
+	if (dev_all_qdisc_sleeping_noop(dev))
+		netdev_for_each_tx_queue(dev, attach_one_default_qdisc, NULL);
 
 	if (!netif_carrier_ok(dev))
 		/* Delay activation until next carrier-on event */
 		return;
 
-	spin_lock_bh(&txq->lock);
-	rcu_assign_pointer(txq->qdisc, txq->qdisc_sleeping);
-	if (txq->qdisc != &noqueue_qdisc) {
+	need_watchdog = 0;
+	netdev_for_each_tx_queue(dev, transition_one_qdisc, &need_watchdog);
+
+	if (need_watchdog) {
 		dev->trans_start = jiffies;
 		dev_watchdog_up(dev);
 	}
-	spin_unlock_bh(&txq->lock);
 }
 
-static void dev_deactivate_queue(struct netdev_queue *dev_queue,
-				 struct Qdisc *qdisc_default)
+static void dev_deactivate_queue(struct net_device *dev,
+				 struct netdev_queue *dev_queue,
+				 void *_qdisc_default)
 {
+	struct Qdisc *qdisc_default = _qdisc_default;
 	struct Qdisc *qdisc;
 	struct sk_buff *skb;
 
@@ -603,12 +662,35 @@ static void dev_deactivate_queue(struct netdev_queue *dev_queue,
 	kfree_skb(skb);
 }
 
+static bool some_qdisc_is_running(struct net_device *dev, int lock)
+{
+	unsigned int i;
+
+	for (i = 0; i < dev->num_tx_queues; i++) {
+		struct netdev_queue *dev_queue;
+		int val;
+
+		dev_queue = netdev_get_tx_queue(dev, i);
+
+		if (lock)
+			spin_lock_bh(&dev_queue->lock);
+
+		val = test_bit(__QUEUE_STATE_QDISC_RUNNING, &dev_queue->state);
+
+		if (lock)
+			spin_unlock_bh(&dev_queue->lock);
+
+		if (val)
+			return true;
+	}
+	return false;
+}
+
 void dev_deactivate(struct net_device *dev)
 {
-	struct netdev_queue *dev_queue = &dev->tx_queue;
-	int running;
+	bool running;
 
-	dev_deactivate_queue(dev_queue, &noop_qdisc);
+	netdev_for_each_tx_queue(dev, dev_deactivate_queue, &noop_qdisc);
 
 	dev_watchdog_down(dev);
 
@@ -617,17 +699,14 @@ void dev_deactivate(struct net_device *dev)
 
 	/* Wait for outstanding qdisc_run calls. */
 	do {
-		while (test_bit(__QUEUE_STATE_QDISC_RUNNING, &dev_queue->state))
+		while (some_qdisc_is_running(dev, 0))
 			yield();
 
 		/*
 		 * Double-check inside queue lock to ensure that all effects
 		 * of the queue run are visible when we return.
 		 */
-		spin_lock_bh(&dev_queue->lock);
-		running = test_bit(__QUEUE_STATE_QDISC_RUNNING,
-				   &dev_queue->state);
-		spin_unlock_bh(&dev_queue->lock);
+		running = some_qdisc_is_running(dev, 1);
 
 		/*
 		 * The running flag should never be set at this point because
@@ -642,8 +721,10 @@ void dev_deactivate(struct net_device *dev)
 
 static void dev_init_scheduler_queue(struct net_device *dev,
 				     struct netdev_queue *dev_queue,
-				     struct Qdisc *qdisc)
+				     void *_qdisc)
 {
+	struct Qdisc *qdisc = _qdisc;
+
 	dev_queue->qdisc = qdisc;
 	dev_queue->qdisc_sleeping = qdisc;
 	INIT_LIST_HEAD(&dev_queue->qdisc_list);
@@ -652,18 +733,19 @@ static void dev_init_scheduler_queue(struct net_device *dev,
 void dev_init_scheduler(struct net_device *dev)
 {
 	qdisc_lock_tree(dev);
-	dev_init_scheduler_queue(dev, &dev->tx_queue, &noop_qdisc);
+	netdev_for_each_tx_queue(dev, dev_init_scheduler_queue, &noop_qdisc);
 	dev_init_scheduler_queue(dev, &dev->rx_queue, NULL);
 	qdisc_unlock_tree(dev);
 
 	setup_timer(&dev->watchdog_timer, dev_watchdog, (unsigned long)dev);
 }
 
-static void dev_shutdown_scheduler_queue(struct net_device *dev,
-					 struct netdev_queue *dev_queue,
-					 struct Qdisc *qdisc_default)
+static void shutdown_scheduler_queue(struct net_device *dev,
+				     struct netdev_queue *dev_queue,
+				     void *_qdisc_default)
 {
 	struct Qdisc *qdisc = dev_queue->qdisc_sleeping;
+	struct Qdisc *qdisc_default = _qdisc_default;
 
 	if (qdisc) {
 		dev_queue->qdisc = qdisc_default;
@@ -676,8 +758,8 @@ static void dev_shutdown_scheduler_queue(struct net_device *dev,
 void dev_shutdown(struct net_device *dev)
 {
 	qdisc_lock_tree(dev);
-	dev_shutdown_scheduler_queue(dev, &dev->tx_queue, &noop_qdisc);
-	dev_shutdown_scheduler_queue(dev, &dev->rx_queue, NULL);
+	netdev_for_each_tx_queue(dev, shutdown_scheduler_queue, &noop_qdisc);
+	shutdown_scheduler_queue(dev, &dev->rx_queue, NULL);
 	BUG_TRAP(!timer_pending(&dev->watchdog_timer));
 	qdisc_unlock_tree(dev);
 }
diff --git a/net/sched/sch_teql.c b/net/sched/sch_teql.c
index 8ac05981be20..44a2c3451f4d 100644
--- a/net/sched/sch_teql.c
+++ b/net/sched/sch_teql.c
@@ -111,7 +111,7 @@ teql_dequeue(struct Qdisc* sch)
 	struct sk_buff *skb;
 
 	skb = __skb_dequeue(&dat->q);
-	dat_queue = &dat->m->dev->tx_queue;
+	dat_queue = netdev_get_tx_queue(dat->m->dev, 0);
 	if (skb == NULL) {
 		struct net_device *m = qdisc_dev(dat_queue->qdisc);
 		if (m) {
@@ -155,10 +155,13 @@ teql_destroy(struct Qdisc* sch)
 				if (q == master->slaves) {
 					master->slaves = NEXT_SLAVE(q);
 					if (q == master->slaves) {
+						struct netdev_queue *txq;
+
+						txq = netdev_get_tx_queue(master->dev, 0);
 						master->slaves = NULL;
-						spin_lock_bh(&master->dev->tx_queue.lock);
-						qdisc_reset(master->dev->tx_queue.qdisc);
-						spin_unlock_bh(&master->dev->tx_queue.lock);
+						spin_lock_bh(&txq->lock);
+						qdisc_reset(txq->qdisc);
+						spin_unlock_bh(&txq->lock);
 					}
 				}
 				skb_queue_purge(&dat->q);
@@ -218,7 +221,8 @@ static int teql_qdisc_init(struct Qdisc *sch, struct nlattr *opt)
 static int
 __teql_resolve(struct sk_buff *skb, struct sk_buff *skb_res, struct net_device *dev)
 {
-	struct teql_sched_data *q = qdisc_priv(dev->tx_queue.qdisc);
+	struct netdev_queue *dev_queue = netdev_get_tx_queue(dev, 0);
+	struct teql_sched_data *q = qdisc_priv(dev_queue->qdisc);
 	struct neighbour *mn = skb->dst->neighbour;
 	struct neighbour *n = q->ncache;
 
@@ -254,7 +258,8 @@ __teql_resolve(struct sk_buff *skb, struct sk_buff *skb_res, struct net_device *
 static inline int teql_resolve(struct sk_buff *skb,
 			       struct sk_buff *skb_res, struct net_device *dev)
 {
-	if (dev->tx_queue.qdisc == &noop_qdisc)
+	struct netdev_queue *txq = netdev_get_tx_queue(dev, 0);
+	if (txq->qdisc == &noop_qdisc)
 		return -ENODEV;
 
 	if (dev->header_ops == NULL ||
@@ -285,8 +290,10 @@ restart:
 
 	do {
 		struct net_device *slave = qdisc_dev(q);
+		struct netdev_queue *slave_txq;
 
-		if (slave->tx_queue.qdisc_sleeping != q)
+		slave_txq = netdev_get_tx_queue(slave, 0);
+		if (slave_txq->qdisc_sleeping != q)
 			continue;
 		if (netif_queue_stopped(slave) ||
 		    __netif_subqueue_stopped(slave, subq) ||
-- 
cgit v1.2.3


From 09e83b5d7d1878065e2453239b49b684cd0fe4e5 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Thu, 17 Jul 2008 01:52:12 -0700
Subject: netdev: Kill NETIF_F_MULTI_QUEUE.

There is no need for a feature bit for something that
can be tested by simply checking the TX queue count.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/networking/multiqueue.txt | 11 -----------
 drivers/net/cpmac.c                     |  1 -
 drivers/net/igb/igb_main.c              |  2 --
 drivers/net/ixgbe/ixgbe_main.c          |  2 --
 drivers/net/s2io.c                      |  2 --
 include/linux/netdevice.h               |  4 +---
 net/mac80211/main.c                     |  3 ---
 7 files changed, 1 insertion(+), 24 deletions(-)

(limited to 'include')

diff --git a/Documentation/networking/multiqueue.txt b/Documentation/networking/multiqueue.txt
index e6dc1ee9e8f1..d391ea631141 100644
--- a/Documentation/networking/multiqueue.txt
+++ b/Documentation/networking/multiqueue.txt
@@ -24,15 +24,4 @@ netif_{start|stop|wake}_subqueue() functions to manage each queue while the
 device is still operational.  netdev->queue_lock is still used when the device
 comes online or when it's completely shut down (unregister_netdev(), etc.).
 
-Finally, the base driver should indicate that it is a multiqueue device.  The
-feature flag NETIF_F_MULTI_QUEUE should be added to the netdev->features
-bitmap on device initialization.  Below is an example from e1000:
-
-#ifdef CONFIG_E1000_MQ
-	if ( (adapter->hw.mac.type == e1000_82571) ||
-	     (adapter->hw.mac.type == e1000_82572) ||
-	     (adapter->hw.mac.type == e1000_80003es2lan))
-		netdev->features |= NETIF_F_MULTI_QUEUE;
-#endif
-
 Author: Peter P. Waskiewicz Jr. <peter.p.waskiewicz.jr@intel.com>
diff --git a/drivers/net/cpmac.c b/drivers/net/cpmac.c
index d630e2a72f42..7c7b54e4828e 100644
--- a/drivers/net/cpmac.c
+++ b/drivers/net/cpmac.c
@@ -1165,7 +1165,6 @@ static int __devinit cpmac_probe(struct platform_device *pdev)
 	dev->set_multicast_list = cpmac_set_multicast_list;
 	dev->tx_timeout         = cpmac_tx_timeout;
 	dev->ethtool_ops        = &cpmac_ethtool_ops;
-	dev->features |= NETIF_F_MULTI_QUEUE;
 
 	netif_napi_add(dev, &priv->napi, cpmac_poll, 64);
 
diff --git a/drivers/net/igb/igb_main.c b/drivers/net/igb/igb_main.c
index 64a150a16f39..471c194cd54e 100644
--- a/drivers/net/igb/igb_main.c
+++ b/drivers/net/igb/igb_main.c
@@ -1155,8 +1155,6 @@ static int __devinit igb_probe(struct pci_dev *pdev,
 	if (pci_using_dac)
 		netdev->features |= NETIF_F_HIGHDMA;
 
-	netdev->features |= NETIF_F_MULTI_QUEUE;
-
 	netdev->features |= NETIF_F_LLTX;
 	adapter->en_mng_pt = igb_enable_mng_pass_thru(&adapter->hw);
 
diff --git a/drivers/net/ixgbe/ixgbe_main.c b/drivers/net/ixgbe/ixgbe_main.c
index 7d8bf94d3783..e6df9233f5ef 100644
--- a/drivers/net/ixgbe/ixgbe_main.c
+++ b/drivers/net/ixgbe/ixgbe_main.c
@@ -3566,8 +3566,6 @@ static int __devinit ixgbe_probe(struct pci_dev *pdev,
 	if (pci_using_dac)
 		netdev->features |= NETIF_F_HIGHDMA;
 
-	netdev->features |= NETIF_F_MULTI_QUEUE;
-
 	/* make sure the EEPROM is good */
 	if (ixgbe_validate_eeprom_checksum(hw, NULL) < 0) {
 		dev_err(&pdev->dev, "The EEPROM Checksum Is Not Valid\n");
diff --git a/drivers/net/s2io.c b/drivers/net/s2io.c
index 517425dcb77c..5f0fcb04afff 100644
--- a/drivers/net/s2io.c
+++ b/drivers/net/s2io.c
@@ -7966,8 +7966,6 @@ s2io_init_nic(struct pci_dev *pdev, const struct pci_device_id *pre)
 		dev->features |= NETIF_F_UFO;
 		dev->features |= NETIF_F_HW_CSUM;
 	}
-	if (config->multiq)
-		dev->features |= NETIF_F_MULTI_QUEUE;
 	dev->tx_timeout = &s2io_tx_watchdog;
 	dev->watchdog_timeo = WATCH_DOG_TIMEOUT;
 	INIT_WORK(&sp->rst_timer_task, s2io_restart_nic);
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index f25d4f5a31b0..c02227b9dd7b 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -533,7 +533,6 @@ struct net_device
 #define NETIF_F_LLTX		4096	/* LockLess TX - deprecated. Please */
 					/* do not use LLTX in new drivers */
 #define NETIF_F_NETNS_LOCAL	8192	/* Does not change network namespaces */
-#define NETIF_F_MULTI_QUEUE	16384	/* Has multiple TX/RX queues */
 #define NETIF_F_LRO		32768	/* large receive offload */
 
 	/* Segmentation offload features */
@@ -1163,11 +1162,10 @@ static inline void netif_wake_subqueue(struct net_device *dev, u16 queue_index)
  *	@dev: network device
  *
  * Check if device has multiple transmit queues
- * Always falls if NETDEVICE_MULTIQUEUE is not configured
  */
 static inline int netif_is_multiqueue(const struct net_device *dev)
 {
-	return (!!(NETIF_F_MULTI_QUEUE & dev->features));
+	return (dev->num_tx_queues > 1);
 }
 
 /* Use this variant when it is known for sure that it
diff --git a/net/mac80211/main.c b/net/mac80211/main.c
index b486e634f4fe..c74607eda1ee 100644
--- a/net/mac80211/main.c
+++ b/net/mac80211/main.c
@@ -1678,9 +1678,6 @@ int ieee80211_register_hw(struct ieee80211_hw *hw)
 	if (!mdev)
 		goto fail_mdev_alloc;
 
-	if (ieee80211_num_queues(hw) > 1)
-		mdev->features |= NETIF_F_MULTI_QUEUE;
-
 	mwdev = netdev_priv(mdev);
 	mdev->ieee80211_ptr = mwdev;
 	mwdev->wiphy = local->hw.wiphy;
-- 
cgit v1.2.3


From 1d8ae3fdeb001b8f534a6782c261aba6ec1779f5 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Tue, 15 Jul 2008 02:52:19 -0700
Subject: pkt_sched: Remove RR scheduler.

This actually fixes a bug added by the RR scheduler changes.  The
->bands and ->prio2band parameters were being set outside of the
sch_tree_lock() and thus could result in strange behavior and
inconsistencies.

It might be possible, in the new design (where there will be one qdisc
per device TX queue) to allow similar functionality via a TX hash
algorithm for RR but I really see no reason to export this aspect of
how these multiqueue cards actually implement the scheduling of the
the individual DMA TX rings and the single physical MAC/PHY port.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/pkt_sched.h |   9 ---
 net/sched/sch_prio.c      | 136 ++++++----------------------------------------
 2 files changed, 16 insertions(+), 129 deletions(-)

(limited to 'include')

diff --git a/include/linux/pkt_sched.h b/include/linux/pkt_sched.h
index dbb7ac37960d..87f4e0fa8f27 100644
--- a/include/linux/pkt_sched.h
+++ b/include/linux/pkt_sched.h
@@ -103,15 +103,6 @@ struct tc_prio_qopt
 	__u8	priomap[TC_PRIO_MAX+1];	/* Map: logical priority -> PRIO band */
 };
 
-enum
-{
-	TCA_PRIO_UNSPEC,
-	TCA_PRIO_MQ,
-	__TCA_PRIO_MAX
-};
-
-#define TCA_PRIO_MAX    (__TCA_PRIO_MAX - 1)
-
 /* TBF section */
 
 struct tc_tbf_qopt
diff --git a/net/sched/sch_prio.c b/net/sched/sch_prio.c
index 39157f7bc046..536ca474dc69 100644
--- a/net/sched/sch_prio.c
+++ b/net/sched/sch_prio.c
@@ -24,11 +24,9 @@
 struct prio_sched_data
 {
 	int bands;
-	int curband; /* for round-robin */
 	struct tcf_proto *filter_list;
 	u8  prio2band[TC_PRIO_MAX+1];
 	struct Qdisc *queues[TCQ_PRIO_BANDS];
-	int mq;
 };
 
 
@@ -55,17 +53,14 @@ prio_classify(struct sk_buff *skb, struct Qdisc *sch, int *qerr)
 		if (!q->filter_list || err < 0) {
 			if (TC_H_MAJ(band))
 				band = 0;
-			band = q->prio2band[band&TC_PRIO_MAX];
-			goto out;
+			return q->queues[q->prio2band[band&TC_PRIO_MAX]];
 		}
 		band = res.classid;
 	}
 	band = TC_H_MIN(band) - 1;
 	if (band >= q->bands)
-		band = q->prio2band[0];
-out:
-	if (q->mq)
-		skb_set_queue_mapping(skb, band);
+		return q->queues[q->prio2band[0]];
+
 	return q->queues[band];
 }
 
@@ -123,68 +118,23 @@ prio_requeue(struct sk_buff *skb, struct Qdisc* sch)
 }
 
 
-static struct sk_buff *
-prio_dequeue(struct Qdisc* sch)
+static struct sk_buff *prio_dequeue(struct Qdisc* sch)
 {
-	struct sk_buff *skb;
 	struct prio_sched_data *q = qdisc_priv(sch);
 	int prio;
-	struct Qdisc *qdisc;
 
 	for (prio = 0; prio < q->bands; prio++) {
-		/* Check if the target subqueue is available before
-		 * pulling an skb.  This way we avoid excessive requeues
-		 * for slower queues.
-		 */
-		if (!__netif_subqueue_stopped(qdisc_dev(sch),
-					      (q->mq ? prio : 0))) {
-			qdisc = q->queues[prio];
-			skb = qdisc->dequeue(qdisc);
-			if (skb) {
-				sch->q.qlen--;
-				return skb;
-			}
+		struct Qdisc *qdisc = q->queues[prio];
+		struct sk_buff *skb = qdisc->dequeue(qdisc);
+		if (skb) {
+			sch->q.qlen--;
+			return skb;
 		}
 	}
 	return NULL;
 
 }
 
-static struct sk_buff *rr_dequeue(struct Qdisc* sch)
-{
-	struct sk_buff *skb;
-	struct prio_sched_data *q = qdisc_priv(sch);
-	struct Qdisc *qdisc;
-	int bandcount;
-
-	/* Only take one pass through the queues.  If nothing is available,
-	 * return nothing.
-	 */
-	for (bandcount = 0; bandcount < q->bands; bandcount++) {
-		/* Check if the target subqueue is available before
-		 * pulling an skb.  This way we avoid excessive requeues
-		 * for slower queues.  If the queue is stopped, try the
-		 * next queue.
-		 */
-		if (!__netif_subqueue_stopped(qdisc_dev(sch),
-					      (q->mq ? q->curband : 0))) {
-			qdisc = q->queues[q->curband];
-			skb = qdisc->dequeue(qdisc);
-			if (skb) {
-				sch->q.qlen--;
-				q->curband++;
-				if (q->curband >= q->bands)
-					q->curband = 0;
-				return skb;
-			}
-		}
-		q->curband++;
-		if (q->curband >= q->bands)
-			q->curband = 0;
-	}
-	return NULL;
-}
-
 static unsigned int prio_drop(struct Qdisc* sch)
 {
 	struct prio_sched_data *q = qdisc_priv(sch);
@@ -229,45 +179,22 @@ static int prio_tune(struct Qdisc *sch, struct nlattr *opt)
 {
 	struct prio_sched_data *q = qdisc_priv(sch);
 	struct tc_prio_qopt *qopt;
-	struct nlattr *tb[TCA_PRIO_MAX + 1];
-	int err;
 	int i;
 
-	err = nla_parse_nested_compat(tb, TCA_PRIO_MAX, opt, NULL, qopt,
-				      sizeof(*qopt));
-	if (err < 0)
-		return err;
-
-	q->bands = qopt->bands;
-	/* If we're multiqueue, make sure the number of incoming bands
-	 * matches the number of queues on the device we're associating with.
-	 * If the number of bands requested is zero, then set q->bands to
-	 * dev->egress_subqueue_count.  Also, the root qdisc must be the
-	 * only one that is enabled for multiqueue, since it's the only one
-	 * that interacts with the underlying device.
-	 */
-	q->mq = nla_get_flag(tb[TCA_PRIO_MQ]);
-	if (q->mq) {
-		if (sch->parent != TC_H_ROOT)
-			return -EINVAL;
-		if (netif_is_multiqueue(qdisc_dev(sch))) {
-			if (q->bands == 0)
-				q->bands = qdisc_dev(sch)->egress_subqueue_count;
-			else if (q->bands != qdisc_dev(sch)->egress_subqueue_count)
-				return -EINVAL;
-		} else
-			return -EOPNOTSUPP;
-	}
+	if (nla_len(opt) < sizeof(*qopt))
+		return -EINVAL;
+	qopt = nla_data(opt);
 
-	if (q->bands > TCQ_PRIO_BANDS || q->bands < 2)
+	if (qopt->bands > TCQ_PRIO_BANDS || qopt->bands < 2)
 		return -EINVAL;
 
 	for (i=0; i<=TC_PRIO_MAX; i++) {
-		if (qopt->priomap[i] >= q->bands)
+		if (qopt->priomap[i] >= qopt->bands)
 			return -EINVAL;
 	}
 
 	sch_tree_lock(sch);
+	q->bands = qopt->bands;
 	memcpy(q->prio2band, qopt->priomap, TC_PRIO_MAX+1);
 
 	for (i=q->bands; i<TCQ_PRIO_BANDS; i++) {
@@ -333,10 +260,6 @@ static int prio_dump(struct Qdisc *sch, struct sk_buff *skb)
 	nest = nla_nest_compat_start(skb, TCA_OPTIONS, sizeof(opt), &opt);
 	if (nest == NULL)
 		goto nla_put_failure;
-	if (q->mq) {
-		if (nla_put_flag(skb, TCA_PRIO_MQ) < 0)
-			goto nla_put_failure;
-	}
 	nla_nest_compat_end(skb, nest);
 
 	return skb->len;
@@ -509,44 +432,17 @@ static struct Qdisc_ops prio_qdisc_ops __read_mostly = {
 	.owner		=	THIS_MODULE,
 };
 
-static struct Qdisc_ops rr_qdisc_ops __read_mostly = {
-	.next		=	NULL,
-	.cl_ops		=	&prio_class_ops,
-	.id		=	"rr",
-	.priv_size	=	sizeof(struct prio_sched_data),
-	.enqueue	=	prio_enqueue,
-	.dequeue	=	rr_dequeue,
-	.requeue	=	prio_requeue,
-	.drop		=	prio_drop,
-	.init		=	prio_init,
-	.reset		=	prio_reset,
-	.destroy	=	prio_destroy,
-	.change		=	prio_tune,
-	.dump		=	prio_dump,
-	.owner		=	THIS_MODULE,
-};
-
 static int __init prio_module_init(void)
 {
-	int err;
-
-	err = register_qdisc(&prio_qdisc_ops);
-	if (err < 0)
-		return err;
-	err = register_qdisc(&rr_qdisc_ops);
-	if (err < 0)
-		unregister_qdisc(&prio_qdisc_ops);
-	return err;
+	return register_qdisc(&prio_qdisc_ops);
 }
 
 static void __exit prio_module_exit(void)
 {
 	unregister_qdisc(&prio_qdisc_ops);
-	unregister_qdisc(&rr_qdisc_ops);
 }
 
 module_init(prio_module_init)
 module_exit(prio_module_exit)
 
 MODULE_LICENSE("GPL");
-MODULE_ALIAS("sch_rr");
-- 
cgit v1.2.3


From fd2ea0a79faad824258af5dcec1927aa24d81c16 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Thu, 17 Jul 2008 01:56:23 -0700
Subject: net: Use queue aware tests throughout.

This effectively "flips the switch" by making the core networking
and multiqueue-aware drivers use the new TX multiqueue structures.

Non-multiqueue drivers need no changes.  The interfaces they use such
as netif_stop_queue() degenerate into an operation on TX queue zero.
So everything "just works" for them.

Code that really wants to do "X" to all TX queues now invokes a
routine that does so, such as netif_tx_wake_all_queues(),
netif_tx_stop_all_queues(), etc.

pktgen and netpoll required a little bit more surgery than the others.

In particular the pktgen changes, whilst functional, could be largely
improved.  The initial check in pktgen_xmit() will sometimes check the
wrong queue, which is mostly harmless.  The thing to do is probably to
invoke fill_packet() earlier.

The bulk of the netpoll changes is to make the code operate solely on
the TX queue indicated by by the SKB queue mapping.

Setting of the SKB queue mapping is entirely confined inside of
net/core/dev.c:dev_pick_tx().  If we end up needing any kind of
special semantics (drops, for example) it will be implemented here.

Finally, we now have a "real_num_tx_queues" which is where the driver
indicates how many TX queues are actually active.

With IGB changes from Jeff Kirsher.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/cpmac.c               | 20 ++++------
 drivers/net/igb/igb_main.c        | 19 +++------
 drivers/net/ixgbe/ixgbe_ethtool.c | 10 +----
 drivers/net/ixgbe/ixgbe_main.c    | 15 +++----
 drivers/net/s2io.c                | 48 +++++++++--------------
 include/linux/netdevice.h         | 82 ++++++++++++++++++++++++++++++++++-----
 include/net/pkt_sched.h           |  4 +-
 net/core/dev.c                    | 28 ++++++-------
 net/core/netpoll.c                | 24 +++++++-----
 net/core/pktgen.c                 | 69 ++++++++++++++++++++------------
 net/sched/sch_generic.c           |  5 +--
 net/sched/sch_teql.c              |  6 +--
 12 files changed, 187 insertions(+), 143 deletions(-)

(limited to 'include')

diff --git a/drivers/net/cpmac.c b/drivers/net/cpmac.c
index 7c7b54e4828e..fbd4280c102c 100644
--- a/drivers/net/cpmac.c
+++ b/drivers/net/cpmac.c
@@ -544,7 +544,7 @@ fatal_error:
 
 	spin_unlock(&priv->rx_lock);
 	netif_rx_complete(priv->dev, napi);
-	netif_stop_queue(priv->dev);
+	netif_tx_stop_all_queues(priv->dev);
 	napi_disable(&priv->napi);
 
 	atomic_inc(&priv->reset_pending);
@@ -750,9 +750,7 @@ static void cpmac_hw_error(struct work_struct *work)
 	barrier();
 	atomic_dec(&priv->reset_pending);
 
-	for (i = 0; i < CPMAC_QUEUES; i++)
-		netif_wake_subqueue(priv->dev, i);
-	netif_wake_queue(priv->dev);
+	netif_tx_wake_all_queues(priv->dev);
 	cpmac_write(priv->regs, CPMAC_MAC_INT_ENABLE, 3);
 }
 
@@ -781,7 +779,7 @@ static void cpmac_check_status(struct net_device *dev)
 				     dev->name, tx_code, tx_channel, macstatus);
 		}
 
-		netif_stop_queue(dev);
+		netif_tx_stop_all_queues(dev);
 		cpmac_hw_stop(dev);
 		if (schedule_work(&priv->reset_work))
 			atomic_inc(&priv->reset_pending);
@@ -842,9 +840,7 @@ static void cpmac_tx_timeout(struct net_device *dev)
 	barrier();
 	atomic_dec(&priv->reset_pending);
 
-	netif_wake_queue(priv->dev);
-	for (i = 0; i < CPMAC_QUEUES; i++)
-		netif_wake_subqueue(dev, i);
+	netif_tx_wake_all_queues(priv->dev);
 }
 
 static int cpmac_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
@@ -935,7 +931,7 @@ static void cpmac_adjust_link(struct net_device *dev)
 
 	spin_lock(&priv->lock);
 	if (priv->phy->link) {
-		netif_start_queue(dev);
+		netif_tx_start_all_queues(dev);
 		if (priv->phy->duplex != priv->oldduplex) {
 			new_state = 1;
 			priv->oldduplex = priv->phy->duplex;
@@ -949,10 +945,10 @@ static void cpmac_adjust_link(struct net_device *dev)
 		if (!priv->oldlink) {
 			new_state = 1;
 			priv->oldlink = 1;
-			netif_schedule(dev);
+			netif_tx_schedule_all(dev);
 		}
 	} else if (priv->oldlink) {
-		netif_stop_queue(dev);
+		netif_tx_stop_all_queues(dev);
 		new_state = 1;
 		priv->oldlink = 0;
 		priv->oldspeed = 0;
@@ -1072,7 +1068,7 @@ static int cpmac_stop(struct net_device *dev)
 	struct cpmac_priv *priv = netdev_priv(dev);
 	struct resource *mem;
 
-	netif_stop_queue(dev);
+	netif_tx_stop_all_queues(dev);
 
 	cancel_work_sync(&priv->reset_work);
 	napi_disable(&priv->napi);
diff --git a/drivers/net/igb/igb_main.c b/drivers/net/igb/igb_main.c
index 471c194cd54e..81bba6983dde 100644
--- a/drivers/net/igb/igb_main.c
+++ b/drivers/net/igb/igb_main.c
@@ -533,7 +533,7 @@ msi_only:
 		adapter->flags |= IGB_FLAG_HAS_MSI;
 
 	/* Notify the stack of the (possibly) reduced Tx Queue count. */
-	adapter->netdev->egress_subqueue_count = adapter->num_tx_queues;
+	adapter->netdev->real_num_tx_queues = adapter->num_tx_queues;
 	return;
 }
 
@@ -821,9 +821,7 @@ void igb_down(struct igb_adapter *adapter)
 	wr32(E1000_RCTL, rctl & ~E1000_RCTL_EN);
 	/* flush and sleep below */
 
-	netif_stop_queue(netdev);
-	for (i = 0; i < adapter->num_tx_queues; i++)
-		netif_stop_subqueue(netdev, i);
+	netif_tx_stop_all_queues(netdev);
 
 	/* disable transmits in the hardware */
 	tctl = rd32(E1000_TCTL);
@@ -1266,9 +1264,7 @@ static int __devinit igb_probe(struct pci_dev *pdev,
 
 	/* tell the stack to leave us alone until igb_open() is called */
 	netif_carrier_off(netdev);
-	netif_stop_queue(netdev);
-	for (i = 0; i < adapter->num_tx_queues; i++)
-		netif_stop_subqueue(netdev, i);
+	netif_tx_stop_all_queues(netdev);
 
 	strcpy(netdev->name, "eth%d");
 	err = register_netdev(netdev);
@@ -2315,7 +2311,6 @@ static void igb_watchdog_task(struct work_struct *work)
 	struct e1000_mac_info *mac = &adapter->hw.mac;
 	u32 link;
 	s32 ret_val;
-	int i;
 
 	if ((netif_carrier_ok(netdev)) &&
 	    (rd32(E1000_STATUS) & E1000_STATUS_LU))
@@ -2371,9 +2366,7 @@ static void igb_watchdog_task(struct work_struct *work)
 			}
 
 			netif_carrier_on(netdev);
-			netif_wake_queue(netdev);
-			for (i = 0; i < adapter->num_tx_queues; i++)
-				netif_wake_subqueue(netdev, i);
+			netif_tx_wake_all_queues(netdev);
 
 			if (!test_bit(__IGB_DOWN, &adapter->state))
 				mod_timer(&adapter->phy_info_timer,
@@ -2385,9 +2378,7 @@ static void igb_watchdog_task(struct work_struct *work)
 			adapter->link_duplex = 0;
 			dev_info(&adapter->pdev->dev, "NIC Link is Down\n");
 			netif_carrier_off(netdev);
-			netif_stop_queue(netdev);
-			for (i = 0; i < adapter->num_tx_queues; i++)
-				netif_stop_subqueue(netdev, i);
+			netif_tx_stop_all_queues(netdev);
 			if (!test_bit(__IGB_DOWN, &adapter->state))
 				mod_timer(&adapter->phy_info_timer,
 					  round_jiffies(jiffies + 2 * HZ));
diff --git a/drivers/net/ixgbe/ixgbe_ethtool.c b/drivers/net/ixgbe/ixgbe_ethtool.c
index 81b769093d22..3efe5dda10af 100644
--- a/drivers/net/ixgbe/ixgbe_ethtool.c
+++ b/drivers/net/ixgbe/ixgbe_ethtool.c
@@ -252,16 +252,10 @@ static int ixgbe_set_tso(struct net_device *netdev, u32 data)
 		netdev->features |= NETIF_F_TSO;
 		netdev->features |= NETIF_F_TSO6;
 	} else {
-		struct ixgbe_adapter *adapter = netdev_priv(netdev);
-		int i;
-		netif_stop_queue(netdev);
-		for (i = 0; i < adapter->num_tx_queues; i++)
-			netif_stop_subqueue(netdev, i);
+		netif_tx_stop_all_queues(netdev);
 		netdev->features &= ~NETIF_F_TSO;
 		netdev->features &= ~NETIF_F_TSO6;
-		for (i = 0; i < adapter->num_tx_queues; i++)
-			netif_start_subqueue(netdev, i);
-		netif_start_queue(netdev);
+		netif_tx_start_all_queues(netdev);
 	}
 	return 0;
 }
diff --git a/drivers/net/ixgbe/ixgbe_main.c b/drivers/net/ixgbe/ixgbe_main.c
index e6df9233f5ef..6af8fb5c4b5f 100644
--- a/drivers/net/ixgbe/ixgbe_main.c
+++ b/drivers/net/ixgbe/ixgbe_main.c
@@ -2013,7 +2013,7 @@ void ixgbe_down(struct ixgbe_adapter *adapter)
 	del_timer_sync(&adapter->watchdog_timer);
 
 	netif_carrier_off(netdev);
-	netif_stop_queue(netdev);
+	netif_tx_stop_all_queues(netdev);
 
 	if (!pci_channel_offline(adapter->pdev))
 		ixgbe_reset(adapter);
@@ -2359,7 +2359,7 @@ try_msi:
 
 out:
 	/* Notify the stack of the (possibly) reduced Tx Queue count. */
-	adapter->netdev->egress_subqueue_count = adapter->num_tx_queues;
+	adapter->netdev->real_num_tx_queues = adapter->num_tx_queues;
 
 	return err;
 }
@@ -2896,7 +2896,6 @@ static void ixgbe_watchdog(unsigned long data)
 	struct net_device *netdev = adapter->netdev;
 	bool link_up;
 	u32 link_speed = 0;
-	int i;
 
 	adapter->hw.mac.ops.check_link(&adapter->hw, &(link_speed), &link_up);
 
@@ -2917,9 +2916,7 @@ static void ixgbe_watchdog(unsigned long data)
 				 (FLOW_TX ? "TX" : "None"))));
 
 			netif_carrier_on(netdev);
-			netif_wake_queue(netdev);
-			for (i = 0; i < adapter->num_tx_queues; i++)
-				netif_wake_subqueue(netdev, i);
+			netif_tx_wake_all_queues(netdev);
 		} else {
 			/* Force detection of hung controller */
 			adapter->detect_tx_hung = true;
@@ -2928,7 +2925,7 @@ static void ixgbe_watchdog(unsigned long data)
 		if (netif_carrier_ok(netdev)) {
 			DPRINTK(LINK, INFO, "NIC Link is Down\n");
 			netif_carrier_off(netdev);
-			netif_stop_queue(netdev);
+			netif_tx_stop_all_queues(netdev);
 		}
 	}
 
@@ -3631,9 +3628,7 @@ static int __devinit ixgbe_probe(struct pci_dev *pdev,
 	ixgbe_start_hw(hw);
 
 	netif_carrier_off(netdev);
-	netif_stop_queue(netdev);
-	for (i = 0; i < adapter->num_tx_queues; i++)
-		netif_stop_subqueue(netdev, i);
+	netif_tx_stop_all_queues(netdev);
 
 	ixgbe_napi_add_all(adapter);
 
diff --git a/drivers/net/s2io.c b/drivers/net/s2io.c
index 5f0fcb04afff..9dae40ccf048 100644
--- a/drivers/net/s2io.c
+++ b/drivers/net/s2io.c
@@ -545,63 +545,53 @@ static struct pci_driver s2io_driver = {
 /* netqueue manipulation helper functions */
 static inline void s2io_stop_all_tx_queue(struct s2io_nic *sp)
 {
-	int i;
-	if (sp->config.multiq) {
-		for (i = 0; i < sp->config.tx_fifo_num; i++)
-			netif_stop_subqueue(sp->dev, i);
-	} else {
+	if (!sp->config.multiq) {
+		int i;
+
 		for (i = 0; i < sp->config.tx_fifo_num; i++)
 			sp->mac_control.fifos[i].queue_state = FIFO_QUEUE_STOP;
-		netif_stop_queue(sp->dev);
 	}
+	netif_tx_stop_all_queues(sp->dev);
 }
 
 static inline void s2io_stop_tx_queue(struct s2io_nic *sp, int fifo_no)
 {
-	if (sp->config.multiq)
-		netif_stop_subqueue(sp->dev, fifo_no);
-	else {
+	if (!sp->config.multiq)
 		sp->mac_control.fifos[fifo_no].queue_state =
 			FIFO_QUEUE_STOP;
-		netif_stop_queue(sp->dev);
-	}
+
+	netif_tx_stop_all_queues(sp->dev);
 }
 
 static inline void s2io_start_all_tx_queue(struct s2io_nic *sp)
 {
-	int i;
-	if (sp->config.multiq) {
-		for (i = 0; i < sp->config.tx_fifo_num; i++)
-			netif_start_subqueue(sp->dev, i);
-	} else {
+	if (!sp->config.multiq) {
+		int i;
+
 		for (i = 0; i < sp->config.tx_fifo_num; i++)
 			sp->mac_control.fifos[i].queue_state = FIFO_QUEUE_START;
-		netif_start_queue(sp->dev);
 	}
+	netif_tx_start_all_queues(sp->dev);
 }
 
 static inline void s2io_start_tx_queue(struct s2io_nic *sp, int fifo_no)
 {
-	if (sp->config.multiq)
-		netif_start_subqueue(sp->dev, fifo_no);
-	else {
+	if (!sp->config.multiq)
 		sp->mac_control.fifos[fifo_no].queue_state =
 			FIFO_QUEUE_START;
-		netif_start_queue(sp->dev);
-	}
+
+	netif_tx_start_all_queues(sp->dev);
 }
 
 static inline void s2io_wake_all_tx_queue(struct s2io_nic *sp)
 {
-	int i;
-	if (sp->config.multiq) {
-		for (i = 0; i < sp->config.tx_fifo_num; i++)
-			netif_wake_subqueue(sp->dev, i);
-	} else {
+	if (!sp->config.multiq) {
+		int i;
+
 		for (i = 0; i < sp->config.tx_fifo_num; i++)
 			sp->mac_control.fifos[i].queue_state = FIFO_QUEUE_START;
-		netif_wake_queue(sp->dev);
 	}
+	netif_tx_wake_all_queues(sp->dev);
 }
 
 static inline void s2io_wake_tx_queue(
@@ -8691,5 +8681,5 @@ static void s2io_io_resume(struct pci_dev *pdev)
 	}
 
 	netif_device_attach(netdev);
-	netif_wake_queue(netdev);
+	netif_tx_wake_all_queues(netdev);
 }
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index c02227b9dd7b..b5c1e7df64fc 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -642,7 +642,13 @@ struct net_device
 	struct netdev_queue	rx_queue;
 
 	struct netdev_queue	*_tx ____cacheline_aligned_in_smp;
+
+	/* Number of TX queues allocated at alloc_netdev_mq() time  */
 	unsigned int		num_tx_queues;
+
+	/* Number of TX queues currently active in device  */
+	unsigned int		real_num_tx_queues;
+
 	unsigned long		tx_queue_len;	/* Max frames per queue allowed */
 
 /*
@@ -1000,6 +1006,14 @@ static inline void netif_schedule(struct net_device *dev)
 	netif_schedule_queue(netdev_get_tx_queue(dev, 0));
 }
 
+static inline void netif_tx_schedule_all(struct net_device *dev)
+{
+	unsigned int i;
+
+	for (i = 0; i < dev->num_tx_queues; i++)
+		netif_schedule_queue(netdev_get_tx_queue(dev, i));
+}
+
 /**
  *	netif_start_queue - allow transmit
  *	@dev: network device
@@ -1016,6 +1030,16 @@ static inline void netif_start_queue(struct net_device *dev)
 	netif_tx_start_queue(netdev_get_tx_queue(dev, 0));
 }
 
+static inline void netif_tx_start_all_queues(struct net_device *dev)
+{
+	unsigned int i;
+
+	for (i = 0; i < dev->num_tx_queues; i++) {
+		struct netdev_queue *txq = netdev_get_tx_queue(dev, i);
+		netif_tx_start_queue(txq);
+	}
+}
+
 /**
  *	netif_wake_queue - restart transmit
  *	@dev: network device
@@ -1040,6 +1064,16 @@ static inline void netif_wake_queue(struct net_device *dev)
 	netif_tx_wake_queue(netdev_get_tx_queue(dev, 0));
 }
 
+static inline void netif_tx_wake_all_queues(struct net_device *dev)
+{
+	unsigned int i;
+
+	for (i = 0; i < dev->num_tx_queues; i++) {
+		struct netdev_queue *txq = netdev_get_tx_queue(dev, i);
+		netif_tx_wake_queue(txq);
+	}
+}
+
 /**
  *	netif_stop_queue - stop transmitted packets
  *	@dev: network device
@@ -1057,6 +1091,16 @@ static inline void netif_stop_queue(struct net_device *dev)
 	netif_tx_stop_queue(netdev_get_tx_queue(dev, 0));
 }
 
+static inline void netif_tx_stop_all_queues(struct net_device *dev)
+{
+	unsigned int i;
+
+	for (i = 0; i < dev->num_tx_queues; i++) {
+		struct netdev_queue *txq = netdev_get_tx_queue(dev, i);
+		netif_tx_stop_queue(txq);
+	}
+}
+
 /**
  *	netif_queue_stopped - test if transmit queue is flowblocked
  *	@dev: network device
@@ -1100,7 +1144,8 @@ static inline int netif_running(const struct net_device *dev)
  */
 static inline void netif_start_subqueue(struct net_device *dev, u16 queue_index)
 {
-	clear_bit(__QUEUE_STATE_XOFF, &dev->egress_subqueue[queue_index].state);
+	struct netdev_queue *txq = netdev_get_tx_queue(dev, queue_index);
+	clear_bit(__QUEUE_STATE_XOFF, &txq->state);
 }
 
 /**
@@ -1112,11 +1157,12 @@ static inline void netif_start_subqueue(struct net_device *dev, u16 queue_index)
  */
 static inline void netif_stop_subqueue(struct net_device *dev, u16 queue_index)
 {
+	struct netdev_queue *txq = netdev_get_tx_queue(dev, queue_index);
 #ifdef CONFIG_NETPOLL_TRAP
 	if (netpoll_trap())
 		return;
 #endif
-	set_bit(__QUEUE_STATE_XOFF, &dev->egress_subqueue[queue_index].state);
+	set_bit(__QUEUE_STATE_XOFF, &txq->state);
 }
 
 /**
@@ -1129,8 +1175,8 @@ static inline void netif_stop_subqueue(struct net_device *dev, u16 queue_index)
 static inline int __netif_subqueue_stopped(const struct net_device *dev,
 					 u16 queue_index)
 {
-	return test_bit(__QUEUE_STATE_XOFF,
-			&dev->egress_subqueue[queue_index].state);
+	struct netdev_queue *txq = netdev_get_tx_queue(dev, queue_index);
+	return test_bit(__QUEUE_STATE_XOFF, &txq->state);
 }
 
 static inline int netif_subqueue_stopped(const struct net_device *dev,
@@ -1148,13 +1194,13 @@ static inline int netif_subqueue_stopped(const struct net_device *dev,
  */
 static inline void netif_wake_subqueue(struct net_device *dev, u16 queue_index)
 {
+	struct netdev_queue *txq = netdev_get_tx_queue(dev, queue_index);
 #ifdef CONFIG_NETPOLL_TRAP
 	if (netpoll_trap())
 		return;
 #endif
-	if (test_and_clear_bit(__QUEUE_STATE_XOFF,
-			       &dev->egress_subqueue[queue_index].state))
-		__netif_schedule(netdev_get_tx_queue(dev, 0));
+	if (test_and_clear_bit(__QUEUE_STATE_XOFF, &txq->state))
+		__netif_schedule(txq);
 }
 
 /**
@@ -1198,7 +1244,8 @@ extern int		dev_set_mtu(struct net_device *, int);
 extern int		dev_set_mac_address(struct net_device *,
 					    struct sockaddr *);
 extern int		dev_hard_start_xmit(struct sk_buff *skb,
-					    struct net_device *dev);
+					    struct net_device *dev,
+					    struct netdev_queue *txq);
 
 extern int		netdev_budget;
 
@@ -1447,6 +1494,12 @@ static inline void __netif_tx_lock(struct netdev_queue *txq, int cpu)
 	txq->xmit_lock_owner = cpu;
 }
 
+static inline void __netif_tx_lock_bh(struct netdev_queue *txq)
+{
+	spin_lock_bh(&txq->_xmit_lock);
+	txq->xmit_lock_owner = smp_processor_id();
+}
+
 static inline void netif_tx_lock(struct net_device *dev)
 {
 	int cpu = smp_processor_id();
@@ -1483,6 +1536,12 @@ static inline void __netif_tx_unlock(struct netdev_queue *txq)
 	spin_unlock(&txq->_xmit_lock);
 }
 
+static inline void __netif_tx_unlock_bh(struct netdev_queue *txq)
+{
+	txq->xmit_lock_owner = -1;
+	spin_unlock_bh(&txq->_xmit_lock);
+}
+
 static inline void netif_tx_unlock(struct net_device *dev)
 {
 	unsigned int i;
@@ -1514,8 +1573,13 @@ static inline void netif_tx_unlock_bh(struct net_device *dev)
 
 static inline void netif_tx_disable(struct net_device *dev)
 {
+	unsigned int i;
+
 	netif_tx_lock_bh(dev);
-	netif_stop_queue(dev);
+	for (i = 0; i < dev->num_tx_queues; i++) {
+		struct netdev_queue *txq = netdev_get_tx_queue(dev, i);
+		netif_tx_stop_queue(txq);
+	}
 	netif_tx_unlock_bh(dev);
 }
 
diff --git a/include/net/pkt_sched.h b/include/net/pkt_sched.h
index d58c1a5eb845..cb9527815606 100644
--- a/include/net/pkt_sched.h
+++ b/include/net/pkt_sched.h
@@ -88,9 +88,7 @@ extern void __qdisc_run(struct netdev_queue *txq);
 
 static inline void qdisc_run(struct netdev_queue *txq)
 {
-	struct net_device *dev = txq->dev;
-
-	if (!netif_queue_stopped(dev) &&
+	if (!netif_tx_queue_stopped(txq) &&
 	    !test_and_set_bit(__QUEUE_STATE_QDISC_RUNNING, &txq->state))
 		__qdisc_run(txq);
 }
diff --git a/net/core/dev.c b/net/core/dev.c
index 69378f250695..f027a1ac4fbb 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -1598,7 +1598,8 @@ static int dev_gso_segment(struct sk_buff *skb)
 	return 0;
 }
 
-int dev_hard_start_xmit(struct sk_buff *skb, struct net_device *dev)
+int dev_hard_start_xmit(struct sk_buff *skb, struct net_device *dev,
+			struct netdev_queue *txq)
 {
 	if (likely(!skb->next)) {
 		if (!list_empty(&ptype_all))
@@ -1627,9 +1628,7 @@ gso:
 			skb->next = nskb;
 			return rc;
 		}
-		if (unlikely((netif_queue_stopped(dev) ||
-			     netif_subqueue_stopped(dev, skb)) &&
-			     skb->next))
+		if (unlikely(netif_tx_queue_stopped(txq) && skb->next))
 			return NETDEV_TX_BUSY;
 	} while (skb->next);
 
@@ -1669,7 +1668,10 @@ out_kfree_skb:
 static struct netdev_queue *dev_pick_tx(struct net_device *dev,
 					struct sk_buff *skb)
 {
-	return netdev_get_tx_queue(dev, 0);
+	u16 queue_index = 0;
+
+	skb_set_queue_mapping(skb, queue_index);
+	return netdev_get_tx_queue(dev, queue_index);
 }
 
 int dev_queue_xmit(struct sk_buff *skb)
@@ -1737,8 +1739,6 @@ gso:
 		spin_lock(&txq->lock);
 		q = txq->qdisc;
 		if (q->enqueue) {
-			/* reset queue_mapping to zero */
-			skb_set_queue_mapping(skb, 0);
 			rc = q->enqueue(skb, q);
 			qdisc_run(txq);
 			spin_unlock(&txq->lock);
@@ -1768,10 +1768,9 @@ gso:
 
 			HARD_TX_LOCK(dev, txq, cpu);
 
-			if (!netif_queue_stopped(dev) &&
-			    !netif_subqueue_stopped(dev, skb)) {
+			if (!netif_tx_queue_stopped(txq)) {
 				rc = 0;
-				if (!dev_hard_start_xmit(skb, dev)) {
+				if (!dev_hard_start_xmit(skb, dev, txq)) {
 					HARD_TX_UNLOCK(dev, txq);
 					goto out;
 				}
@@ -4160,8 +4159,7 @@ struct net_device *alloc_netdev_mq(int sizeof_priv, const char *name,
 
 	BUG_ON(strlen(name) >= sizeof(dev->name));
 
-	alloc_size = sizeof(struct net_device) +
-		     sizeof(struct net_device_subqueue) * (queue_count - 1);
+	alloc_size = sizeof(struct net_device);
 	if (sizeof_priv) {
 		/* ensure 32-byte alignment of private area */
 		alloc_size = (alloc_size + NETDEV_ALIGN_CONST) & ~NETDEV_ALIGN_CONST;
@@ -4191,16 +4189,14 @@ struct net_device *alloc_netdev_mq(int sizeof_priv, const char *name,
 
 	dev->_tx = tx;
 	dev->num_tx_queues = queue_count;
+	dev->real_num_tx_queues = queue_count;
 
 	if (sizeof_priv) {
 		dev->priv = ((char *)dev +
-			     ((sizeof(struct net_device) +
-			       (sizeof(struct net_device_subqueue) *
-				(queue_count - 1)) + NETDEV_ALIGN_CONST)
+			     ((sizeof(struct net_device) + NETDEV_ALIGN_CONST)
 			      & ~NETDEV_ALIGN_CONST));
 	}
 
-	dev->egress_subqueue_count = queue_count;
 	dev->gso_max_size = GSO_MAX_SIZE;
 
 	netdev_init_queues(dev);
diff --git a/net/core/netpoll.c b/net/core/netpoll.c
index 8fb134da0346..c12720895ecf 100644
--- a/net/core/netpoll.c
+++ b/net/core/netpoll.c
@@ -58,25 +58,27 @@ static void queue_process(struct work_struct *work)
 
 	while ((skb = skb_dequeue(&npinfo->txq))) {
 		struct net_device *dev = skb->dev;
+		struct netdev_queue *txq;
 
 		if (!netif_device_present(dev) || !netif_running(dev)) {
 			__kfree_skb(skb);
 			continue;
 		}
 
+		txq = netdev_get_tx_queue(dev, skb_get_queue_mapping(skb));
+
 		local_irq_save(flags);
-		netif_tx_lock(dev);
-		if ((netif_queue_stopped(dev) ||
-		     netif_subqueue_stopped(dev, skb)) ||
-		     dev->hard_start_xmit(skb, dev) != NETDEV_TX_OK) {
+		__netif_tx_lock(txq, smp_processor_id());
+		if (netif_tx_queue_stopped(txq) ||
+		    dev->hard_start_xmit(skb, dev) != NETDEV_TX_OK) {
 			skb_queue_head(&npinfo->txq, skb);
-			netif_tx_unlock(dev);
+			__netif_tx_unlock(txq);
 			local_irq_restore(flags);
 
 			schedule_delayed_work(&npinfo->tx_work, HZ/10);
 			return;
 		}
-		netif_tx_unlock(dev);
+		__netif_tx_unlock(txq);
 		local_irq_restore(flags);
 	}
 }
@@ -278,17 +280,19 @@ static void netpoll_send_skb(struct netpoll *np, struct sk_buff *skb)
 
 	/* don't get messages out of order, and no recursion */
 	if (skb_queue_len(&npinfo->txq) == 0 && !netpoll_owner_active(dev)) {
+		struct netdev_queue *txq;
 		unsigned long flags;
 
+		txq = netdev_get_tx_queue(dev, skb_get_queue_mapping(skb));
+
 		local_irq_save(flags);
 		/* try until next clock tick */
 		for (tries = jiffies_to_usecs(1)/USEC_PER_POLL;
 		     tries > 0; --tries) {
-			if (netif_tx_trylock(dev)) {
-				if (!netif_queue_stopped(dev) &&
-				    !netif_subqueue_stopped(dev, skb))
+			if (__netif_tx_trylock(txq)) {
+				if (!netif_tx_queue_stopped(txq))
 					status = dev->hard_start_xmit(skb, dev);
-				netif_tx_unlock(dev);
+				__netif_tx_unlock(txq);
 
 				if (status == NETDEV_TX_OK)
 					break;
diff --git a/net/core/pktgen.c b/net/core/pktgen.c
index fdf537707e51..906802db4ed4 100644
--- a/net/core/pktgen.c
+++ b/net/core/pktgen.c
@@ -2123,6 +2123,24 @@ static void get_ipsec_sa(struct pktgen_dev *pkt_dev, int flow)
 	}
 }
 #endif
+static void set_cur_queue_map(struct pktgen_dev *pkt_dev)
+{
+	if (pkt_dev->queue_map_min < pkt_dev->queue_map_max) {
+		__u16 t;
+		if (pkt_dev->flags & F_QUEUE_MAP_RND) {
+			t = random32() %
+				(pkt_dev->queue_map_max -
+				 pkt_dev->queue_map_min + 1)
+				+ pkt_dev->queue_map_min;
+		} else {
+			t = pkt_dev->cur_queue_map + 1;
+			if (t > pkt_dev->queue_map_max)
+				t = pkt_dev->queue_map_min;
+		}
+		pkt_dev->cur_queue_map = t;
+	}
+}
+
 /* Increment/randomize headers according to flags and current values
  * for IP src/dest, UDP src/dst port, MAC-Addr src/dst
  */
@@ -2325,19 +2343,7 @@ static void mod_cur_headers(struct pktgen_dev *pkt_dev)
 		pkt_dev->cur_pkt_size = t;
 	}
 
-	if (pkt_dev->queue_map_min < pkt_dev->queue_map_max) {
-		__u16 t;
-		if (pkt_dev->flags & F_QUEUE_MAP_RND) {
-			t = random32() %
-				(pkt_dev->queue_map_max - pkt_dev->queue_map_min + 1)
-				+ pkt_dev->queue_map_min;
-		} else {
-			t = pkt_dev->cur_queue_map + 1;
-			if (t > pkt_dev->queue_map_max)
-				t = pkt_dev->queue_map_min;
-		}
-		pkt_dev->cur_queue_map = t;
-	}
+	set_cur_queue_map(pkt_dev);
 
 	pkt_dev->flows[flow].count++;
 }
@@ -2458,7 +2464,7 @@ static struct sk_buff *fill_packet_ipv4(struct net_device *odev,
 	__be16 *vlan_encapsulated_proto = NULL;  /* packet type ID field (or len) for VLAN tag */
 	__be16 *svlan_tci = NULL;                /* Encapsulates priority and SVLAN ID */
 	__be16 *svlan_encapsulated_proto = NULL; /* packet type ID field (or len) for SVLAN tag */
-
+	u16 queue_map;
 
 	if (pkt_dev->nr_labels)
 		protocol = htons(ETH_P_MPLS_UC);
@@ -2469,6 +2475,7 @@ static struct sk_buff *fill_packet_ipv4(struct net_device *odev,
 	/* Update any of the values, used when we're incrementing various
 	 * fields.
 	 */
+	queue_map = pkt_dev->cur_queue_map;
 	mod_cur_headers(pkt_dev);
 
 	datalen = (odev->hard_header_len + 16) & ~0xf;
@@ -2507,7 +2514,7 @@ static struct sk_buff *fill_packet_ipv4(struct net_device *odev,
 	skb->network_header = skb->tail;
 	skb->transport_header = skb->network_header + sizeof(struct iphdr);
 	skb_put(skb, sizeof(struct iphdr) + sizeof(struct udphdr));
-	skb_set_queue_mapping(skb, pkt_dev->cur_queue_map);
+	skb_set_queue_mapping(skb, queue_map);
 	iph = ip_hdr(skb);
 	udph = udp_hdr(skb);
 
@@ -2797,6 +2804,7 @@ static struct sk_buff *fill_packet_ipv6(struct net_device *odev,
 	__be16 *vlan_encapsulated_proto = NULL;  /* packet type ID field (or len) for VLAN tag */
 	__be16 *svlan_tci = NULL;                /* Encapsulates priority and SVLAN ID */
 	__be16 *svlan_encapsulated_proto = NULL; /* packet type ID field (or len) for SVLAN tag */
+	u16 queue_map;
 
 	if (pkt_dev->nr_labels)
 		protocol = htons(ETH_P_MPLS_UC);
@@ -2807,6 +2815,7 @@ static struct sk_buff *fill_packet_ipv6(struct net_device *odev,
 	/* Update any of the values, used when we're incrementing various
 	 * fields.
 	 */
+	queue_map = pkt_dev->cur_queue_map;
 	mod_cur_headers(pkt_dev);
 
 	skb = alloc_skb(pkt_dev->cur_pkt_size + 64 + 16 +
@@ -2844,7 +2853,7 @@ static struct sk_buff *fill_packet_ipv6(struct net_device *odev,
 	skb->network_header = skb->tail;
 	skb->transport_header = skb->network_header + sizeof(struct ipv6hdr);
 	skb_put(skb, sizeof(struct ipv6hdr) + sizeof(struct udphdr));
-	skb_set_queue_mapping(skb, pkt_dev->cur_queue_map);
+	skb_set_queue_mapping(skb, queue_map);
 	iph = ipv6_hdr(skb);
 	udph = udp_hdr(skb);
 
@@ -3263,7 +3272,9 @@ static void pktgen_rem_thread(struct pktgen_thread *t)
 static __inline__ void pktgen_xmit(struct pktgen_dev *pkt_dev)
 {
 	struct net_device *odev = NULL;
+	struct netdev_queue *txq;
 	__u64 idle_start = 0;
+	u16 queue_map;
 	int ret;
 
 	odev = pkt_dev->odev;
@@ -3285,9 +3296,15 @@ static __inline__ void pktgen_xmit(struct pktgen_dev *pkt_dev)
 		}
 	}
 
-	if ((netif_queue_stopped(odev) ||
-	     (pkt_dev->skb &&
-	      netif_subqueue_stopped(odev, pkt_dev->skb))) ||
+	if (!pkt_dev->skb) {
+		set_cur_queue_map(pkt_dev);
+		queue_map = pkt_dev->cur_queue_map;
+	} else {
+		queue_map = skb_get_queue_mapping(pkt_dev->skb);
+	}
+
+	txq = netdev_get_tx_queue(odev, queue_map);
+	if (netif_tx_queue_stopped(txq) ||
 	    need_resched()) {
 		idle_start = getCurUs();
 
@@ -3303,8 +3320,7 @@ static __inline__ void pktgen_xmit(struct pktgen_dev *pkt_dev)
 
 		pkt_dev->idle_acc += getCurUs() - idle_start;
 
-		if (netif_queue_stopped(odev) ||
-		    netif_subqueue_stopped(odev, pkt_dev->skb)) {
+		if (netif_tx_queue_stopped(txq)) {
 			pkt_dev->next_tx_us = getCurUs();	/* TODO */
 			pkt_dev->next_tx_ns = 0;
 			goto out;	/* Try the next interface */
@@ -3331,9 +3347,12 @@ static __inline__ void pktgen_xmit(struct pktgen_dev *pkt_dev)
 		}
 	}
 
-	netif_tx_lock_bh(odev);
-	if (!netif_queue_stopped(odev) &&
-	    !netif_subqueue_stopped(odev, pkt_dev->skb)) {
+	/* fill_packet() might have changed the queue */
+	queue_map = skb_get_queue_mapping(pkt_dev->skb);
+	txq = netdev_get_tx_queue(odev, queue_map);
+
+	__netif_tx_lock_bh(txq);
+	if (!netif_tx_queue_stopped(txq)) {
 
 		atomic_inc(&(pkt_dev->skb->users));
 	      retry_now:
@@ -3377,7 +3396,7 @@ static __inline__ void pktgen_xmit(struct pktgen_dev *pkt_dev)
 		pkt_dev->next_tx_ns = 0;
 	}
 
-	netif_tx_unlock_bh(odev);
+	__netif_tx_unlock_bh(txq);
 
 	/* If pkt_dev->count is zero, then run forever */
 	if ((pkt_dev->count != 0) && (pkt_dev->sofar >= pkt_dev->count)) {
diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c
index 4e2b865cbba0..2f575b9017d1 100644
--- a/net/sched/sch_generic.c
+++ b/net/sched/sch_generic.c
@@ -166,7 +166,7 @@ static inline int qdisc_restart(struct netdev_queue *txq)
 
 	HARD_TX_LOCK(dev, txq, smp_processor_id());
 	if (!netif_subqueue_stopped(dev, skb))
-		ret = dev_hard_start_xmit(skb, dev);
+		ret = dev_hard_start_xmit(skb, dev, txq);
 	HARD_TX_UNLOCK(dev, txq);
 
 	spin_lock(&txq->lock);
@@ -198,11 +198,10 @@ static inline int qdisc_restart(struct netdev_queue *txq)
 
 void __qdisc_run(struct netdev_queue *txq)
 {
-	struct net_device *dev = txq->dev;
 	unsigned long start_time = jiffies;
 
 	while (qdisc_restart(txq)) {
-		if (netif_queue_stopped(dev))
+		if (netif_tx_queue_stopped(txq))
 			break;
 
 		/*
diff --git a/net/sched/sch_teql.c b/net/sched/sch_teql.c
index 44a2c3451f4d..ade3372221c7 100644
--- a/net/sched/sch_teql.c
+++ b/net/sched/sch_teql.c
@@ -295,8 +295,7 @@ restart:
 		slave_txq = netdev_get_tx_queue(slave, 0);
 		if (slave_txq->qdisc_sleeping != q)
 			continue;
-		if (netif_queue_stopped(slave) ||
-		    __netif_subqueue_stopped(slave, subq) ||
+		if (__netif_subqueue_stopped(slave, subq) ||
 		    !netif_running(slave)) {
 			busy = 1;
 			continue;
@@ -305,8 +304,7 @@ restart:
 		switch (teql_resolve(skb, skb_res, slave)) {
 		case 0:
 			if (netif_tx_trylock(slave)) {
-				if (!netif_queue_stopped(slave) &&
-				    !__netif_subqueue_stopped(slave, subq) &&
+				if (!__netif_subqueue_stopped(slave, subq) &&
 				    slave->hard_start_xmit(skb, slave) == 0) {
 					netif_tx_unlock(slave);
 					master->slaves = NEXT_SLAVE(q);
-- 
cgit v1.2.3


From 6b0fb1261a4655613bed5dac0e935e733969e999 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Tue, 15 Jul 2008 02:58:10 -0700
Subject: netdev: Kill struct net_device_subqueue and netdev->egress_subqueue*

No longer used.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h | 12 ------------
 1 file changed, 12 deletions(-)

(limited to 'include')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index b5c1e7df64fc..a1c2c2204498 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -118,14 +118,6 @@ struct wireless_dev;
 
 #endif  /*  __KERNEL__  */
 
-struct net_device_subqueue
-{
-	/* Give a control state for each queue.  This struct may contain
-	 * per-queue locks in the future.
-	 */
-	unsigned long   state;
-};
-
 /*
  *	Network device statistics. Akin to the 2.0 ether stats but
  *	with byte counters.
@@ -761,10 +753,6 @@ struct net_device
 	/* for setting kernel sock attribute on TCP connection setup */
 #define GSO_MAX_SIZE		65536
 	unsigned int		gso_max_size;
-
-	/* The TX queue control structures */
-	unsigned int			egress_subqueue_count;
-	struct net_device_subqueue	egress_subqueue[1];
 };
 #define to_net_dev(d) container_of(d, struct net_device, dev)
 
-- 
cgit v1.2.3


From e3c50d5d25ac09efd9acbe2b2a3e365466de84ed Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Tue, 15 Jul 2008 02:58:39 -0700
Subject: netdev: netdev_priv() can now be sane again.

The private area of a netdev is now at a fixed offset once more.

Unfortunately, some assumptions that netdev_priv() == netdev->priv
crept back into the tree.  In particular this happened in the
loopback driver.  Make it use netdev->ml_priv.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/loopback.c    | 8 ++++----
 include/linux/netdevice.h | 4 +++-
 2 files changed, 7 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/drivers/net/loopback.c b/drivers/net/loopback.c
index 41b774baac4d..49f6bc036a92 100644
--- a/drivers/net/loopback.c
+++ b/drivers/net/loopback.c
@@ -153,7 +153,7 @@ static int loopback_xmit(struct sk_buff *skb, struct net_device *dev)
 	dev->last_rx = jiffies;
 
 	/* it's OK to use per_cpu_ptr() because BHs are off */
-	pcpu_lstats = netdev_priv(dev);
+	pcpu_lstats = dev->ml_priv;
 	lb_stats = per_cpu_ptr(pcpu_lstats, smp_processor_id());
 	lb_stats->bytes += skb->len;
 	lb_stats->packets++;
@@ -171,7 +171,7 @@ static struct net_device_stats *get_stats(struct net_device *dev)
 	unsigned long packets = 0;
 	int i;
 
-	pcpu_lstats = netdev_priv(dev);
+	pcpu_lstats = dev->ml_priv;
 	for_each_possible_cpu(i) {
 		const struct pcpu_lstats *lb_stats;
 
@@ -207,13 +207,13 @@ static int loopback_dev_init(struct net_device *dev)
 	if (!lstats)
 		return -ENOMEM;
 
-	dev->priv = lstats;
+	dev->ml_priv = lstats;
 	return 0;
 }
 
 static void loopback_dev_free(struct net_device *dev)
 {
-	struct pcpu_lstats *lstats = netdev_priv(dev);
+	struct pcpu_lstats *lstats = dev->ml_priv;
 
 	free_percpu(lstats);
 	free_netdev(dev);
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index a1c2c2204498..fdac1159253e 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -808,7 +808,9 @@ void dev_net_set(struct net_device *dev, struct net *net)
  */
 static inline void *netdev_priv(const struct net_device *dev)
 {
-	return dev->priv;
+	return (char *)dev + ((sizeof(struct net_device)
+			       + NETDEV_ALIGN_CONST)
+			      & ~NETDEV_ALIGN_CONST);
 }
 
 /* Set the sysfs physical device reference for the network logical device
-- 
cgit v1.2.3


From eae792b722fef08dcf3aee88266ee7def9710757 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Tue, 15 Jul 2008 03:03:33 -0700
Subject: netdev: Add netdev->select_queue() method.

Devices or device layers can set this to control the queue selection
performed by dev_pick_tx().

This function runs under RCU protection, which allows overriding
functions to have some way of synchronizing with things like dynamic
->real_num_tx_queues adjustments.

This makes the spinlock prefetch in dev_queue_xmit() a little bit
less effective, but that's the price right now for correctness.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h | 3 +++
 net/core/dev.c            | 9 ++++++---
 2 files changed, 9 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index fdac1159253e..9464e6452967 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -724,6 +724,9 @@ struct net_device
 	void                    (*poll_controller)(struct net_device *dev);
 #endif
 
+	u16			(*select_queue)(struct net_device *dev,
+						struct sk_buff *skb);
+
 #ifdef CONFIG_NET_NS
 	/* Network namespace this network device is inside */
 	struct net		*nd_net;
diff --git a/net/core/dev.c b/net/core/dev.c
index f027a1ac4fbb..7ca9564d2f44 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -1670,6 +1670,9 @@ static struct netdev_queue *dev_pick_tx(struct net_device *dev,
 {
 	u16 queue_index = 0;
 
+	if (dev->select_queue)
+		queue_index = dev->select_queue(dev, skb);
+
 	skb_set_queue_mapping(skb, queue_index);
 	return netdev_get_tx_queue(dev, queue_index);
 }
@@ -1710,14 +1713,14 @@ int dev_queue_xmit(struct sk_buff *skb)
 	}
 
 gso:
-	txq = dev_pick_tx(dev, skb);
-	spin_lock_prefetch(&txq->lock);
-
 	/* Disable soft irqs for various locks below. Also
 	 * stops preemption for RCU.
 	 */
 	rcu_read_lock_bh();
 
+	txq = dev_pick_tx(dev, skb);
+	spin_lock_prefetch(&txq->lock);
+
 	/* Updates of qdisc are serialized by queue->lock.
 	 * The struct Qdisc which is pointed to by qdisc is now a
 	 * rcu structure - it may be accessed without acquiring
-- 
cgit v1.2.3


From 51cb6db0f5654f08a4a6bfa3888dc36a51c2df3e Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Tue, 15 Jul 2008 03:34:57 -0700
Subject: mac80211: Reimplement WME using ->select_queue().

The only behavior change is that we do not drop packets under any
circumstances.  If that is absolutely needed, we could easily add it
back.

With cleanups and help from Johannes Berg.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/mac80211.h     |   8 -
 net/mac80211/Kconfig       |  11 -
 net/mac80211/Makefile      |   2 +-
 net/mac80211/ieee80211_i.h |  15 +-
 net/mac80211/main.c        |  61 ++---
 net/mac80211/util.c        |  14 +-
 net/mac80211/wme.c         | 602 ++++++++-------------------------------------
 net/mac80211/wme.h         |  35 +--
 8 files changed, 124 insertions(+), 624 deletions(-)

(limited to 'include')

diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index 24a69f6075c2..4dd3d93e1960 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -847,20 +847,12 @@ static inline void SET_IEEE80211_PERM_ADDR(struct ieee80211_hw *hw, u8 *addr)
 
 static inline int ieee80211_num_regular_queues(struct ieee80211_hw *hw)
 {
-#ifdef CONFIG_MAC80211_QOS
 	return hw->queues;
-#else
-	return 1;
-#endif
 }
 
 static inline int ieee80211_num_queues(struct ieee80211_hw *hw)
 {
-#ifdef CONFIG_MAC80211_QOS
 	return hw->queues + hw->ampdu_queues;
-#else
-	return 1;
-#endif
 }
 
 static inline struct ieee80211_rate *
diff --git a/net/mac80211/Kconfig b/net/mac80211/Kconfig
index 10579def490b..80d693392b0f 100644
--- a/net/mac80211/Kconfig
+++ b/net/mac80211/Kconfig
@@ -11,17 +11,6 @@ config MAC80211
 	  This option enables the hardware independent IEEE 802.11
 	  networking stack.
 
-config MAC80211_QOS
-	def_bool y
-	depends on MAC80211
-	depends on NET_SCHED
-	depends on BROKEN
-
-comment "QoS/HT support disabled"
-	depends on MAC80211 && !MAC80211_QOS
-comment "QoS/HT support needs CONFIG_NET_SCHED"
-	depends on MAC80211 && !NET_SCHED
-
 menu "Rate control algorithm selection"
 	depends on MAC80211 != n
 
diff --git a/net/mac80211/Makefile b/net/mac80211/Makefile
index fa47438e338f..a169b0201d61 100644
--- a/net/mac80211/Makefile
+++ b/net/mac80211/Makefile
@@ -18,10 +18,10 @@ mac80211-y := \
 	tx.o \
 	key.o \
 	util.o \
+	wme.o \
 	event.o
 
 mac80211-$(CONFIG_MAC80211_LEDS) += led.o
-mac80211-$(CONFIG_MAC80211_QOS) += wme.o
 mac80211-$(CONFIG_MAC80211_DEBUGFS) += \
 	debugfs.o \
 	debugfs_sta.o \
diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h
index cbea0154ee3a..a4f9a832722a 100644
--- a/net/mac80211/ieee80211_i.h
+++ b/net/mac80211/ieee80211_i.h
@@ -25,6 +25,7 @@
 #include <linux/etherdevice.h>
 #include <net/wireless.h>
 #include <net/iw_handler.h>
+#include <net/mac80211.h>
 #include "key.h"
 #include "sta_info.h"
 
@@ -537,6 +538,9 @@ enum {
 	IEEE80211_ADDBA_MSG	= 4,
 };
 
+/* maximum number of hardware queues we support. */
+#define QD_MAX_QUEUES (IEEE80211_MAX_AMPDU_QUEUES + IEEE80211_MAX_QUEUES)
+
 struct ieee80211_local {
 	/* embed the driver visible part.
 	 * don't cast (use the static inlines below), but we keep
@@ -545,6 +549,8 @@ struct ieee80211_local {
 
 	const struct ieee80211_ops *ops;
 
+	unsigned long queue_pool[BITS_TO_LONGS(QD_MAX_QUEUES)];
+
 	struct net_device *mdev; /* wmaster# - "master" 802.11 device */
 	int open_count;
 	int monitors, cooked_mntrs;
@@ -740,15 +746,6 @@ struct ieee80211_local {
 #endif
 };
 
-static inline int ieee80211_is_multiqueue(struct ieee80211_local *local)
-{
-#ifdef CONFIG_MAC80211_QOS
-	return netif_is_multiqueue(local->mdev);
-#else
-	return 0;
-#endif
-}
-
 static inline struct ieee80211_sub_if_data *
 IEEE80211_DEV_TO_SUB_IF(struct net_device *dev)
 {
diff --git a/net/mac80211/main.c b/net/mac80211/main.c
index c74607eda1ee..f1a83d450ea0 100644
--- a/net/mac80211/main.c
+++ b/net/mac80211/main.c
@@ -114,7 +114,7 @@ static int ieee80211_master_open(struct net_device *dev)
 	if (res)
 		return res;
 
-	netif_start_queue(local->mdev);
+	netif_tx_start_all_queues(local->mdev);
 
 	return 0;
 }
@@ -375,7 +375,7 @@ static int ieee80211_open(struct net_device *dev)
 		queue_work(local->hw.workqueue, &ifsta->work);
 	}
 
-	netif_start_queue(dev);
+	netif_tx_start_all_queues(dev);
 
 	return 0;
  err_del_interface:
@@ -400,7 +400,7 @@ static int ieee80211_stop(struct net_device *dev)
 	/*
 	 * Stop TX on this interface first.
 	 */
-	netif_stop_queue(dev);
+	netif_tx_stop_all_queues(dev);
 
 	/*
 	 * Now delete all active aggregation sessions.
@@ -554,7 +554,6 @@ static int ieee80211_stop(struct net_device *dev)
 int ieee80211_start_tx_ba_session(struct ieee80211_hw *hw, u8 *ra, u16 tid)
 {
 	struct ieee80211_local *local = hw_to_local(hw);
-	struct netdev_queue *txq;
 	struct sta_info *sta;
 	struct ieee80211_sub_if_data *sdata;
 	u16 start_seq_num = 0;
@@ -619,11 +618,6 @@ int ieee80211_start_tx_ba_session(struct ieee80211_hw *hw, u8 *ra, u16 tid)
 			(unsigned long)&sta->timer_to_tid[tid];
 	init_timer(&sta->ampdu_mlme.tid_tx[tid]->addba_resp_timer);
 
-	/* ensure that TX flow won't interrupt us
-	 * until the end of the call to requeue function */
-	txq = netdev_get_tx_queue(local->mdev, 0);
-	spin_lock_bh(&txq->lock);
-
 	/* create a new queue for this aggregation */
 	ret = ieee80211_ht_agg_queue_add(local, sta, tid);
 
@@ -650,7 +644,7 @@ int ieee80211_start_tx_ba_session(struct ieee80211_hw *hw, u8 *ra, u16 tid)
 		/* No need to requeue the packets in the agg queue, since we
 		 * held the tx lock: no packet could be enqueued to the newly
 		 * allocated queue */
-		 ieee80211_ht_agg_queue_remove(local, sta, tid, 0);
+		ieee80211_ht_agg_queue_remove(local, sta, tid, 0);
 #ifdef CONFIG_MAC80211_HT_DEBUG
 		printk(KERN_DEBUG "BA request denied - HW unavailable for"
 					" tid %d\n", tid);
@@ -661,7 +655,6 @@ int ieee80211_start_tx_ba_session(struct ieee80211_hw *hw, u8 *ra, u16 tid)
 
 	/* Will put all the packets in the new SW queue */
 	ieee80211_requeue(local, ieee802_1d_to_ac[tid]);
-	spin_unlock_bh(&txq->lock);
 	spin_unlock_bh(&sta->lock);
 
 	/* send an addBA request */
@@ -687,7 +680,6 @@ int ieee80211_start_tx_ba_session(struct ieee80211_hw *hw, u8 *ra, u16 tid)
 err_unlock_queue:
 	kfree(sta->ampdu_mlme.tid_tx[tid]);
 	sta->ampdu_mlme.tid_tx[tid] = NULL;
-	spin_unlock_bh(&txq->lock);
 	ret = -EBUSY;
 err_unlock_sta:
 	spin_unlock_bh(&sta->lock);
@@ -812,7 +804,6 @@ EXPORT_SYMBOL(ieee80211_start_tx_ba_cb);
 void ieee80211_stop_tx_ba_cb(struct ieee80211_hw *hw, u8 *ra, u8 tid)
 {
 	struct ieee80211_local *local = hw_to_local(hw);
-	struct netdev_queue *txq;
 	struct sta_info *sta;
 	u8 *state;
 	int agg_queue;
@@ -844,8 +835,9 @@ void ieee80211_stop_tx_ba_cb(struct ieee80211_hw *hw, u8 *ra, u8 tid)
 	state = &sta->ampdu_mlme.tid_state_tx[tid];
 
 	/* NOTE: no need to use sta->lock in this state check, as
-	 * ieee80211_stop_tx_ba_session will let only
-	 * one stop call to pass through per sta/tid */
+	 * ieee80211_stop_tx_ba_session will let only one stop call to
+	 * pass through per sta/tid
+	 */
 	if ((*state & HT_AGG_STATE_REQ_STOP_BA_MSK) == 0) {
 #ifdef CONFIG_MAC80211_HT_DEBUG
 		printk(KERN_DEBUG "unexpected callback to A-MPDU stop\n");
@@ -860,19 +852,14 @@ void ieee80211_stop_tx_ba_cb(struct ieee80211_hw *hw, u8 *ra, u8 tid)
 
 	agg_queue = sta->tid_to_tx_q[tid];
 
-	/* avoid ordering issues: we are the only one that can modify
-	 * the content of the qdiscs */
-	txq = netdev_get_tx_queue(local->mdev, 0);
-	spin_lock_bh(&txq->lock);
-	/* remove the queue for this aggregation */
 	ieee80211_ht_agg_queue_remove(local, sta, tid, 1);
-	spin_unlock_bh(&txq->lock);
 
-	/* we just requeued the all the frames that were in the removed
-	 * queue, and since we might miss a softirq we do netif_schedule_queue.
-	 * ieee80211_wake_queue is not used here as this queue is not
-	 * necessarily stopped */
-	netif_schedule_queue(txq);
+	/* We just requeued the all the frames that were in the
+	 * removed queue, and since we might miss a softirq we do
+	 * netif_schedule_queue.  ieee80211_wake_queue is not used
+	 * here as this queue is not necessarily stopped
+	 */
+	netif_schedule_queue(netdev_get_tx_queue(local->mdev, agg_queue));
 	spin_lock_bh(&sta->lock);
 	*state = HT_AGG_STATE_IDLE;
 	sta->ampdu_mlme.addba_req_num[tid] = 0;
@@ -1660,17 +1647,12 @@ int ieee80211_register_hw(struct ieee80211_hw *hw)
 	 * We use the number of queues for feature tests (QoS, HT) internally
 	 * so restrict them appropriately.
 	 */
-#ifdef CONFIG_MAC80211_QOS
 	if (hw->queues > IEEE80211_MAX_QUEUES)
 		hw->queues = IEEE80211_MAX_QUEUES;
 	if (hw->ampdu_queues > IEEE80211_MAX_AMPDU_QUEUES)
 		hw->ampdu_queues = IEEE80211_MAX_AMPDU_QUEUES;
 	if (hw->queues < 4)
 		hw->ampdu_queues = 0;
-#else
-	hw->queues = 1;
-	hw->ampdu_queues = 0;
-#endif
 
 	mdev = alloc_netdev_mq(sizeof(struct wireless_dev),
 			       "wmaster%d", ether_setup,
@@ -1754,7 +1736,7 @@ int ieee80211_register_hw(struct ieee80211_hw *hw)
 		goto fail_wep;
 	}
 
-	ieee80211_install_qdisc(local->mdev);
+	local->mdev->select_queue = ieee80211_select_queue;
 
 	/* add one default STA interface */
 	result = ieee80211_if_add(local, "wlan%d", NULL,
@@ -1852,23 +1834,11 @@ static int __init ieee80211_init(void)
 
 	ret = rc80211_pid_init();
 	if (ret)
-		goto out;
-
-	ret = ieee80211_wme_register();
-	if (ret) {
-		printk(KERN_DEBUG "ieee80211_init: failed to "
-		       "initialize WME (err=%d)\n", ret);
-		goto out_cleanup_pid;
-	}
+		return ret;
 
 	ieee80211_debugfs_netdev_init();
 
 	return 0;
-
- out_cleanup_pid:
-	rc80211_pid_exit();
- out:
-	return ret;
 }
 
 static void __exit ieee80211_exit(void)
@@ -1884,7 +1854,6 @@ static void __exit ieee80211_exit(void)
 	if (mesh_allocated)
 		ieee80211s_stop();
 
-	ieee80211_wme_unregister();
 	ieee80211_debugfs_netdev_exit();
 }
 
diff --git a/net/mac80211/util.c b/net/mac80211/util.c
index 89ce4e07bd84..19f85e1b3695 100644
--- a/net/mac80211/util.c
+++ b/net/mac80211/util.c
@@ -363,12 +363,7 @@ void ieee80211_wake_queue(struct ieee80211_hw *hw, int queue)
 	if (test_bit(queue, local->queues_pending)) {
 		tasklet_schedule(&local->tx_pending_tasklet);
 	} else {
-		if (ieee80211_is_multiqueue(local)) {
-			netif_wake_subqueue(local->mdev, queue);
-		} else {
-			WARN_ON(queue != 0);
-			netif_wake_queue(local->mdev);
-		}
+		netif_wake_subqueue(local->mdev, queue);
 	}
 }
 EXPORT_SYMBOL(ieee80211_wake_queue);
@@ -377,12 +372,7 @@ void ieee80211_stop_queue(struct ieee80211_hw *hw, int queue)
 {
 	struct ieee80211_local *local = hw_to_local(hw);
 
-	if (ieee80211_is_multiqueue(local)) {
-		netif_stop_subqueue(local->mdev, queue);
-	} else {
-		WARN_ON(queue != 0);
-		netif_stop_queue(local->mdev);
-	}
+	netif_stop_subqueue(local->mdev, queue);
 }
 EXPORT_SYMBOL(ieee80211_stop_queue);
 
diff --git a/net/mac80211/wme.c b/net/mac80211/wme.c
index f014cd38c2d0..b21cfec4b6ce 100644
--- a/net/mac80211/wme.c
+++ b/net/mac80211/wme.c
@@ -18,67 +18,42 @@
 #include "ieee80211_i.h"
 #include "wme.h"
 
-/* maximum number of hardware queues we support. */
-#define QD_MAX_QUEUES (IEEE80211_MAX_AMPDU_QUEUES + IEEE80211_MAX_QUEUES)
-/* current number of hardware queues we support. */
-#define QD_NUM(hw) ((hw)->queues + (hw)->ampdu_queues)
-
-/*
- * Default mapping in classifier to work with default
+/* Default mapping in classifier to work with default
  * queue setup.
  */
 const int ieee802_1d_to_ac[8] = { 2, 3, 3, 2, 1, 1, 0, 0 };
 
-struct ieee80211_sched_data
-{
-	unsigned long qdisc_pool[BITS_TO_LONGS(QD_MAX_QUEUES)];
-	struct tcf_proto *filter_list;
-	struct Qdisc *queues[QD_MAX_QUEUES];
-	struct sk_buff_head requeued[QD_MAX_QUEUES];
-};
-
 static const char llc_ip_hdr[8] = {0xAA, 0xAA, 0x3, 0, 0, 0, 0x08, 0};
 
-/* given a data frame determine the 802.1p/1d tag to use */
-static inline unsigned classify_1d(struct sk_buff *skb, struct Qdisc *qd)
+/* Given a data frame determine the 802.1p/1d tag to use.  */
+static unsigned int classify_1d(struct sk_buff *skb)
 {
-	struct iphdr *ip;
-	int dscp;
-	int offset;
-
-	struct ieee80211_sched_data *q = qdisc_priv(qd);
-	struct tcf_result res = { -1, 0 };
-
-	/* if there is a user set filter list, call out to that */
-	if (q->filter_list) {
-		tc_classify(skb, q->filter_list, &res);
-		if (res.class != -1)
-			return res.class;
-	}
+	unsigned int dscp;
 
 	/* skb->priority values from 256->263 are magic values to
-	 * directly indicate a specific 802.1d priority.
-	 * This is used to allow 802.1d priority to be passed directly in
-	 * from VLAN tags, etc. */
+	 * directly indicate a specific 802.1d priority.  This is used
+	 * to allow 802.1d priority to be passed directly in from VLAN
+	 * tags, etc.
+	 */
 	if (skb->priority >= 256 && skb->priority <= 263)
 		return skb->priority - 256;
 
-	/* check there is a valid IP header present */
-	offset = ieee80211_get_hdrlen_from_skb(skb);
-	if (skb->len < offset + sizeof(llc_ip_hdr) + sizeof(*ip) ||
-	    memcmp(skb->data + offset, llc_ip_hdr, sizeof(llc_ip_hdr)))
-		return 0;
+	switch (skb->protocol) {
+	case __constant_htons(ETH_P_IP):
+		dscp = ip_hdr(skb)->tos & 0xfc;
+		break;
 
-	ip = (struct iphdr *) (skb->data + offset + sizeof(llc_ip_hdr));
+	default:
+		return 0;
+	}
 
-	dscp = ip->tos & 0xfc;
 	if (dscp & 0x1c)
 		return 0;
 	return dscp >> 5;
 }
 
 
-static inline int wme_downgrade_ac(struct sk_buff *skb)
+static int wme_downgrade_ac(struct sk_buff *skb)
 {
 	switch (skb->priority) {
 	case 6:
@@ -99,11 +74,10 @@ static inline int wme_downgrade_ac(struct sk_buff *skb)
 }
 
 
-/* positive return value indicates which queue to use
- * negative return value indicates to drop the frame */
-static int classify80211(struct sk_buff *skb, struct Qdisc *qd)
+/* Indicate which queue to use.  */
+static u16 classify80211(struct sk_buff *skb, struct net_device *dev)
 {
-	struct ieee80211_local *local = wdev_priv(qdisc_dev(qd)->ieee80211_ptr);
+	struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr);
 	struct ieee80211_hdr *hdr = (struct ieee80211_hdr *) skb->data;
 
 	if (!ieee80211_is_data(hdr->frame_control)) {
@@ -123,13 +97,15 @@ static int classify80211(struct sk_buff *skb, struct Qdisc *qd)
 
 	/* use the data classifier to determine what 802.1d tag the
 	 * data frame has */
-	skb->priority = classify_1d(skb, qd);
+	skb->priority = classify_1d(skb);
 
 	/* in case we are a client verify acm is not set for this ac */
 	while (unlikely(local->wmm_acm & BIT(skb->priority))) {
 		if (wme_downgrade_ac(skb)) {
-			/* No AC with lower priority has acm=0, drop packet. */
-			return -1;
+			/* The old code would drop the packet in this
+			 * case.
+			 */
+			return 0;
 		}
 	}
 
@@ -137,28 +113,29 @@ static int classify80211(struct sk_buff *skb, struct Qdisc *qd)
 	return ieee802_1d_to_ac[skb->priority];
 }
 
-
-static int wme_qdiscop_enqueue(struct sk_buff *skb, struct Qdisc* qd)
+u16 ieee80211_select_queue(struct net_device *dev, struct sk_buff *skb)
 {
-	struct ieee80211_local *local = wdev_priv(qdisc_dev(qd)->ieee80211_ptr);
-	struct ieee80211_hw *hw = &local->hw;
-	struct ieee80211_sched_data *q = qdisc_priv(qd);
-	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
 	struct ieee80211_hdr *hdr = (struct ieee80211_hdr *) skb->data;
-	struct Qdisc *qdisc;
+	struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr);
+	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
 	struct sta_info *sta;
-	int err, queue;
+	u16 queue;
 	u8 tid;
 
+	queue = classify80211(skb, dev);
+	if (unlikely(queue >= local->hw.queues))
+		queue = local->hw.queues - 1;
+
 	if (info->flags & IEEE80211_TX_CTL_REQUEUE) {
-		queue = skb_get_queue_mapping(skb);
 		rcu_read_lock();
 		sta = sta_info_get(local, hdr->addr1);
 		tid = skb->priority & IEEE80211_QOS_CTL_TAG1D_MASK;
 		if (sta) {
+			struct ieee80211_hw *hw = &local->hw;
 			int ampdu_queue = sta->tid_to_tx_q[tid];
-			if ((ampdu_queue < QD_NUM(hw)) &&
-			    test_bit(ampdu_queue, q->qdisc_pool)) {
+
+			if ((ampdu_queue < ieee80211_num_queues(hw)) &&
+			    test_bit(ampdu_queue, local->queue_pool)) {
 				queue = ampdu_queue;
 				info->flags |= IEEE80211_TX_CTL_AMPDU;
 			} else {
@@ -166,17 +143,12 @@ static int wme_qdiscop_enqueue(struct sk_buff *skb, struct Qdisc* qd)
 			}
 		}
 		rcu_read_unlock();
-		skb_queue_tail(&q->requeued[queue], skb);
-		qd->q.qlen++;
-		return 0;
-	}
-
-	queue = classify80211(skb, qd);
 
-	if (unlikely(queue >= local->hw.queues))
-		queue = local->hw.queues - 1;
+		return queue;
+	}
 
-	/* now we know the 1d priority, fill in the QoS header if there is one
+	/* Now we know the 1d priority, fill in the QoS header if
+	 * there is one.
 	 */
 	if (ieee80211_is_data_qos(hdr->frame_control)) {
 		u8 *p = ieee80211_get_qos_ctl(hdr);
@@ -194,8 +166,10 @@ static int wme_qdiscop_enqueue(struct sk_buff *skb, struct Qdisc* qd)
 		sta = sta_info_get(local, hdr->addr1);
 		if (sta) {
 			int ampdu_queue = sta->tid_to_tx_q[tid];
-			if ((ampdu_queue < QD_NUM(hw)) &&
-			    test_bit(ampdu_queue, q->qdisc_pool)) {
+			struct ieee80211_hw *hw = &local->hw;
+
+			if ((ampdu_queue < ieee80211_num_queues(hw)) &&
+			    test_bit(ampdu_queue, local->queue_pool)) {
 				queue = ampdu_queue;
 				info->flags |= IEEE80211_TX_CTL_AMPDU;
 			} else {
@@ -206,421 +180,13 @@ static int wme_qdiscop_enqueue(struct sk_buff *skb, struct Qdisc* qd)
 		rcu_read_unlock();
 	}
 
-	if (unlikely(queue < 0)) {
-			kfree_skb(skb);
-			err = NET_XMIT_DROP;
-	} else {
-		skb_set_queue_mapping(skb, queue);
-		qdisc = q->queues[queue];
-		err = qdisc->enqueue(skb, qdisc);
-		if (err == NET_XMIT_SUCCESS) {
-			qd->q.qlen++;
-			qd->bstats.bytes += skb->len;
-			qd->bstats.packets++;
-			return NET_XMIT_SUCCESS;
-		}
-	}
-	qd->qstats.drops++;
-	return err;
-}
-
-
-/* TODO: clean up the cases where master_hard_start_xmit
- * returns non 0 - it shouldn't ever do that. Once done we
- * can remove this function */
-static int wme_qdiscop_requeue(struct sk_buff *skb, struct Qdisc* qd)
-{
-	struct ieee80211_sched_data *q = qdisc_priv(qd);
-	struct Qdisc *qdisc;
-	int err;
-
-	/* we recorded which queue to use earlier! */
-	qdisc = q->queues[skb_get_queue_mapping(skb)];
-
-	if ((err = qdisc->ops->requeue(skb, qdisc)) == 0) {
-		qd->q.qlen++;
-		return 0;
-	}
-	qd->qstats.drops++;
-	return err;
-}
-
-
-static struct sk_buff *wme_qdiscop_dequeue(struct Qdisc* qd)
-{
-	struct ieee80211_sched_data *q = qdisc_priv(qd);
-	struct net_device *dev = qdisc_dev(qd);
-	struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr);
-	struct ieee80211_hw *hw = &local->hw;
-	struct sk_buff *skb;
-	struct Qdisc *qdisc;
-	int queue;
-
-	/* check all the h/w queues in numeric/priority order */
-	for (queue = 0; queue < QD_NUM(hw); queue++) {
-		/* see if there is room in this hardware queue */
-		if (__netif_subqueue_stopped(local->mdev, queue) ||
-		    !test_bit(queue, q->qdisc_pool))
-			continue;
-
-		/* there is space - try and get a frame */
-		skb = skb_dequeue(&q->requeued[queue]);
-		if (skb) {
-			qd->q.qlen--;
-			return skb;
-		}
-
-		qdisc = q->queues[queue];
-		skb = qdisc->dequeue(qdisc);
-		if (skb) {
-			qd->q.qlen--;
-			return skb;
-		}
-	}
-	/* returning a NULL here when all the h/w queues are full means we
-	 * never need to call netif_stop_queue in the driver */
-	return NULL;
-}
-
-
-static void wme_qdiscop_reset(struct Qdisc* qd)
-{
-	struct ieee80211_sched_data *q = qdisc_priv(qd);
-	struct ieee80211_local *local = wdev_priv(qdisc_dev(qd)->ieee80211_ptr);
-	struct ieee80211_hw *hw = &local->hw;
-	int queue;
-
-	/* QUESTION: should we have some hardware flush functionality here? */
-
-	for (queue = 0; queue < QD_NUM(hw); queue++) {
-		skb_queue_purge(&q->requeued[queue]);
-		qdisc_reset(q->queues[queue]);
-	}
-	qd->q.qlen = 0;
-}
-
-
-static void wme_qdiscop_destroy(struct Qdisc* qd)
-{
-	struct ieee80211_sched_data *q = qdisc_priv(qd);
-	struct ieee80211_local *local = wdev_priv(qdisc_dev(qd)->ieee80211_ptr);
-	struct ieee80211_hw *hw = &local->hw;
-	int queue;
-
-	tcf_destroy_chain(&q->filter_list);
-
-	for (queue = 0; queue < QD_NUM(hw); queue++) {
-		skb_queue_purge(&q->requeued[queue]);
-		qdisc_destroy(q->queues[queue]);
-		q->queues[queue] = &noop_qdisc;
-	}
-}
-
-
-/* called whenever parameters are updated on existing qdisc */
-static int wme_qdiscop_tune(struct Qdisc *qd, struct nlattr *opt)
-{
-	return 0;
-}
-
-
-/* called during initial creation of qdisc on device */
-static int wme_qdiscop_init(struct Qdisc *qd, struct nlattr *opt)
-{
-	struct ieee80211_sched_data *q = qdisc_priv(qd);
-	struct net_device *dev = qdisc_dev(qd);
-	struct ieee80211_local *local;
-	struct ieee80211_hw *hw;
-	int err = 0, i;
-
-	/* check that device is a mac80211 device */
-	if (!dev->ieee80211_ptr ||
-	    dev->ieee80211_ptr->wiphy->privid != mac80211_wiphy_privid)
-		return -EINVAL;
-
-	local = wdev_priv(dev->ieee80211_ptr);
-	hw = &local->hw;
-
-	/* only allow on master dev */
-	if (dev != local->mdev)
-		return -EINVAL;
-
-	/* ensure that we are root qdisc */
-	if (qd->parent != TC_H_ROOT)
-		return -EINVAL;
-
-	if (qd->flags & TCQ_F_INGRESS)
-		return -EINVAL;
-
-	/* if options were passed in, set them */
-	if (opt)
-		err = wme_qdiscop_tune(qd, opt);
-
-	/* create child queues */
-	for (i = 0; i < QD_NUM(hw); i++) {
-		skb_queue_head_init(&q->requeued[i]);
-		q->queues[i] = qdisc_create_dflt(qdisc_dev(qd), qd->dev_queue,
-						 &pfifo_qdisc_ops,
-						 qd->handle);
-		if (!q->queues[i]) {
-			q->queues[i] = &noop_qdisc;
-			printk(KERN_ERR "%s child qdisc %i creation failed\n",
-			       dev->name, i);
-		}
-	}
-
-	/* non-aggregation queues: reserve/mark as used */
-	for (i = 0; i < local->hw.queues; i++)
-		set_bit(i, q->qdisc_pool);
-
-	return err;
-}
-
-static int wme_qdiscop_dump(struct Qdisc *qd, struct sk_buff *skb)
-{
-	return -1;
-}
-
-
-static int wme_classop_graft(struct Qdisc *qd, unsigned long arg,
-			     struct Qdisc *new, struct Qdisc **old)
-{
-	struct ieee80211_sched_data *q = qdisc_priv(qd);
-	struct ieee80211_local *local = wdev_priv(qdisc_dev(qd)->ieee80211_ptr);
-	struct ieee80211_hw *hw = &local->hw;
-	unsigned long queue = arg - 1;
-
-	if (queue >= QD_NUM(hw))
-		return -EINVAL;
-
-	if (!new)
-		new = &noop_qdisc;
-
-	sch_tree_lock(qd);
-	*old = q->queues[queue];
-	q->queues[queue] = new;
-	qdisc_reset(*old);
-	sch_tree_unlock(qd);
-
-	return 0;
-}
-
-
-static struct Qdisc *
-wme_classop_leaf(struct Qdisc *qd, unsigned long arg)
-{
-	struct ieee80211_sched_data *q = qdisc_priv(qd);
-	struct ieee80211_local *local = wdev_priv(qdisc_dev(qd)->ieee80211_ptr);
-	struct ieee80211_hw *hw = &local->hw;
-	unsigned long queue = arg - 1;
-
-	if (queue >= QD_NUM(hw))
-		return NULL;
-
-	return q->queues[queue];
-}
-
-
-static unsigned long wme_classop_get(struct Qdisc *qd, u32 classid)
-{
-	struct ieee80211_local *local = wdev_priv(qdisc_dev(qd)->ieee80211_ptr);
-	struct ieee80211_hw *hw = &local->hw;
-	unsigned long queue = TC_H_MIN(classid);
-
-	if (queue - 1 >= QD_NUM(hw))
-		return 0;
-
 	return queue;
 }
 
-
-static unsigned long wme_classop_bind(struct Qdisc *qd, unsigned long parent,
-				      u32 classid)
-{
-	return wme_classop_get(qd, classid);
-}
-
-
-static void wme_classop_put(struct Qdisc *q, unsigned long cl)
-{
-}
-
-
-static int wme_classop_change(struct Qdisc *qd, u32 handle, u32 parent,
-			      struct nlattr **tca, unsigned long *arg)
-{
-	unsigned long cl = *arg;
-	struct ieee80211_local *local = wdev_priv(qdisc_dev(qd)->ieee80211_ptr);
-	struct ieee80211_hw *hw = &local->hw;
-
-	if (cl - 1 > QD_NUM(hw))
-		return -ENOENT;
-
-	/* TODO: put code to program hardware queue parameters here,
-	 * to allow programming from tc command line */
-
-	return 0;
-}
-
-
-/* we don't support deleting hardware queues
- * when we add WMM-SA support - TSPECs may be deleted here */
-static int wme_classop_delete(struct Qdisc *qd, unsigned long cl)
-{
-	struct ieee80211_local *local = wdev_priv(qdisc_dev(qd)->ieee80211_ptr);
-	struct ieee80211_hw *hw = &local->hw;
-
-	if (cl - 1 > QD_NUM(hw))
-		return -ENOENT;
-	return 0;
-}
-
-
-static int wme_classop_dump_class(struct Qdisc *qd, unsigned long cl,
-				  struct sk_buff *skb, struct tcmsg *tcm)
-{
-	struct ieee80211_sched_data *q = qdisc_priv(qd);
-	struct ieee80211_local *local = wdev_priv(qdisc_dev(qd)->ieee80211_ptr);
-	struct ieee80211_hw *hw = &local->hw;
-
-	if (cl - 1 > QD_NUM(hw))
-		return -ENOENT;
-	tcm->tcm_handle = TC_H_MIN(cl);
-	tcm->tcm_parent = qd->handle;
-	tcm->tcm_info = q->queues[cl-1]->handle; /* do we need this? */
-	return 0;
-}
-
-
-static void wme_classop_walk(struct Qdisc *qd, struct qdisc_walker *arg)
-{
-	struct ieee80211_local *local = wdev_priv(qdisc_dev(qd)->ieee80211_ptr);
-	struct ieee80211_hw *hw = &local->hw;
-	int queue;
-
-	if (arg->stop)
-		return;
-
-	for (queue = 0; queue < QD_NUM(hw); queue++) {
-		if (arg->count < arg->skip) {
-			arg->count++;
-			continue;
-		}
-		/* we should return classids for our internal queues here
-		 * as well as the external ones */
-		if (arg->fn(qd, queue+1, arg) < 0) {
-			arg->stop = 1;
-			break;
-		}
-		arg->count++;
-	}
-}
-
-
-static struct tcf_proto ** wme_classop_find_tcf(struct Qdisc *qd,
-						unsigned long cl)
-{
-	struct ieee80211_sched_data *q = qdisc_priv(qd);
-
-	if (cl)
-		return NULL;
-
-	return &q->filter_list;
-}
-
-
-/* this qdisc is classful (i.e. has classes, some of which may have leaf qdiscs attached)
- * - these are the operations on the classes */
-static const struct Qdisc_class_ops class_ops =
-{
-	.graft = wme_classop_graft,
-	.leaf = wme_classop_leaf,
-
-	.get = wme_classop_get,
-	.put = wme_classop_put,
-	.change = wme_classop_change,
-	.delete = wme_classop_delete,
-	.walk = wme_classop_walk,
-
-	.tcf_chain = wme_classop_find_tcf,
-	.bind_tcf = wme_classop_bind,
-	.unbind_tcf = wme_classop_put,
-
-	.dump = wme_classop_dump_class,
-};
-
-
-/* queueing discipline operations */
-static struct Qdisc_ops wme_qdisc_ops __read_mostly =
-{
-	.next = NULL,
-	.cl_ops = &class_ops,
-	.id = "ieee80211",
-	.priv_size = sizeof(struct ieee80211_sched_data),
-
-	.enqueue = wme_qdiscop_enqueue,
-	.dequeue = wme_qdiscop_dequeue,
-	.requeue = wme_qdiscop_requeue,
-	.drop = NULL, /* drop not needed since we are always the root qdisc */
-
-	.init = wme_qdiscop_init,
-	.reset = wme_qdiscop_reset,
-	.destroy = wme_qdiscop_destroy,
-	.change = wme_qdiscop_tune,
-
-	.dump = wme_qdiscop_dump,
-};
-
-
-void ieee80211_install_qdisc(struct net_device *dev)
-{
-	struct netdev_queue *txq = netdev_get_tx_queue(dev, 0);
-	struct Qdisc *qdisc;
-
-	qdisc = qdisc_create_dflt(dev, txq,
-				  &wme_qdisc_ops, TC_H_ROOT);
-	if (!qdisc) {
-		printk(KERN_ERR "%s: qdisc installation failed\n", dev->name);
-		return;
-	}
-
-	/* same handle as would be allocated by qdisc_alloc_handle() */
-	qdisc->handle = 0x80010000;
-
-	qdisc_lock_tree(dev);
-	list_add_tail(&qdisc->list, &txq->qdisc_list);
-	txq->qdisc_sleeping = qdisc;
-	qdisc_unlock_tree(dev);
-}
-
-
-int ieee80211_qdisc_installed(struct net_device *dev)
-{
-	struct netdev_queue *txq = netdev_get_tx_queue(dev, 0);
-
-	return txq->qdisc_sleeping->ops == &wme_qdisc_ops;
-}
-
-
-int ieee80211_wme_register(void)
-{
-	return register_qdisc(&wme_qdisc_ops);
-}
-
-
-void ieee80211_wme_unregister(void)
-{
-	unregister_qdisc(&wme_qdisc_ops);
-}
-
 int ieee80211_ht_agg_queue_add(struct ieee80211_local *local,
-			struct sta_info *sta, u16 tid)
+			       struct sta_info *sta, u16 tid)
 {
 	int i;
-	struct netdev_queue *txq = netdev_get_tx_queue(local->mdev, 0);
-	struct ieee80211_sched_data *q =
-			qdisc_priv(txq->qdisc_sleeping);
-	DECLARE_MAC_BUF(mac);
 
 	/* prepare the filter and save it for the SW queue
 	 * matching the received HW queue */
@@ -629,8 +195,8 @@ int ieee80211_ht_agg_queue_add(struct ieee80211_local *local,
 		return -EPERM;
 
 	/* try to get a Qdisc from the pool */
-	for (i = local->hw.queues; i < QD_NUM(&local->hw); i++)
-		if (!test_and_set_bit(i, q->qdisc_pool)) {
+	for (i = local->hw.queues; i < ieee80211_num_queues(&local->hw); i++)
+		if (!test_and_set_bit(i, local->queue_pool)) {
 			ieee80211_stop_queue(local_to_hw(local), i);
 			sta->tid_to_tx_q[tid] = i;
 
@@ -639,11 +205,13 @@ int ieee80211_ht_agg_queue_add(struct ieee80211_local *local,
 			 * on the previous queue
 			 * since HT is strict in order */
 #ifdef CONFIG_MAC80211_HT_DEBUG
-			if (net_ratelimit())
+			if (net_ratelimit()) {
+				DECLARE_MAC_BUF(mac);
 				printk(KERN_DEBUG "allocated aggregation queue"
 					" %d tid %d addr %s pool=0x%lX\n",
 					i, tid, print_mac(mac, sta->addr),
-					q->qdisc_pool[0]);
+					local->queue_pool[0]);
+			}
 #endif /* CONFIG_MAC80211_HT_DEBUG */
 			return 0;
 		}
@@ -658,40 +226,68 @@ void ieee80211_ht_agg_queue_remove(struct ieee80211_local *local,
 				   struct sta_info *sta, u16 tid,
 				   u8 requeue)
 {
-	struct ieee80211_hw *hw = &local->hw;
-	struct netdev_queue *txq = netdev_get_tx_queue(local->mdev, 0);
-	struct ieee80211_sched_data *q =
-		qdisc_priv(txq->qdisc_sleeping);
 	int agg_queue = sta->tid_to_tx_q[tid];
+	struct ieee80211_hw *hw = &local->hw;
 
 	/* return the qdisc to the pool */
-	clear_bit(agg_queue, q->qdisc_pool);
-	sta->tid_to_tx_q[tid] = QD_NUM(hw);
+	clear_bit(agg_queue, local->queue_pool);
+	sta->tid_to_tx_q[tid] = ieee80211_num_queues(hw);
 
-	if (requeue)
+	if (requeue) {
 		ieee80211_requeue(local, agg_queue);
-	else
-		q->queues[agg_queue]->ops->reset(q->queues[agg_queue]);
+	} else {
+		struct netdev_queue *txq;
+
+		txq = netdev_get_tx_queue(local->mdev, agg_queue);
+
+		spin_lock_bh(&txq->lock);
+		qdisc_reset(txq->qdisc);
+		spin_unlock_bh(&txq->lock);
+	}
 }
 
 void ieee80211_requeue(struct ieee80211_local *local, int queue)
 {
-	struct netdev_queue *txq = netdev_get_tx_queue(local->mdev, 0);
-	struct Qdisc *root_qd = txq->qdisc_sleeping;
-	struct ieee80211_sched_data *q = qdisc_priv(root_qd);
-	struct Qdisc *qdisc = q->queues[queue];
-	struct sk_buff *skb = NULL;
+	struct netdev_queue *txq = netdev_get_tx_queue(local->mdev, queue);
+	struct sk_buff_head list;
+	struct Qdisc *qdisc;
 	u32 len;
 
+	rcu_read_lock_bh();
+
+	qdisc = rcu_dereference(txq->qdisc);
 	if (!qdisc || !qdisc->dequeue)
-		return;
+		goto out_unlock;
+
+	skb_queue_head_init(&list);
 
+	spin_lock(&txq->lock);
 	for (len = qdisc->q.qlen; len > 0; len--) {
-		skb = qdisc->dequeue(qdisc);
-		root_qd->q.qlen--;
-		/* packet will be classified again and */
-		/* skb->packet_data->queue will be overridden if needed */
+		struct sk_buff *skb = qdisc->dequeue(qdisc);
+
 		if (skb)
-			wme_qdiscop_enqueue(skb, root_qd);
+			__skb_queue_tail(&list, skb);
+	}
+	spin_unlock(&txq->lock);
+
+	for (len = list.qlen; len > 0; len--) {
+		struct sk_buff *skb = __skb_dequeue(&list);
+		u16 new_queue;
+
+		BUG_ON(!skb);
+		new_queue = ieee80211_select_queue(local->mdev, skb);
+		skb_set_queue_mapping(skb, new_queue);
+
+		txq = netdev_get_tx_queue(local->mdev, new_queue);
+
+		spin_lock(&txq->lock);
+
+		qdisc = rcu_dereference(txq->qdisc);
+		qdisc->enqueue(skb, qdisc);
+
+		spin_unlock(&txq->lock);
 	}
+
+out_unlock:
+	rcu_read_unlock_bh();
 }
diff --git a/net/mac80211/wme.h b/net/mac80211/wme.h
index 1aca609eccfc..04de28c071a6 100644
--- a/net/mac80211/wme.h
+++ b/net/mac80211/wme.h
@@ -23,45 +23,12 @@
 
 extern const int ieee802_1d_to_ac[8];
 
-#ifdef CONFIG_MAC80211_QOS
-void ieee80211_install_qdisc(struct net_device *dev);
-int ieee80211_qdisc_installed(struct net_device *dev);
+u16 ieee80211_select_queue(struct net_device *dev, struct sk_buff *skb);
 int ieee80211_ht_agg_queue_add(struct ieee80211_local *local,
 			       struct sta_info *sta, u16 tid);
 void ieee80211_ht_agg_queue_remove(struct ieee80211_local *local,
 				   struct sta_info *sta, u16 tid,
 				   u8 requeue);
 void ieee80211_requeue(struct ieee80211_local *local, int queue);
-int ieee80211_wme_register(void);
-void ieee80211_wme_unregister(void);
-#else
-static inline void ieee80211_install_qdisc(struct net_device *dev)
-{
-}
-static inline int ieee80211_qdisc_installed(struct net_device *dev)
-{
-	return 0;
-}
-static inline int ieee80211_ht_agg_queue_add(struct ieee80211_local *local,
-					     struct sta_info *sta, u16 tid)
-{
-	return -EAGAIN;
-}
-static inline void ieee80211_ht_agg_queue_remove(struct ieee80211_local *local,
-						 struct sta_info *sta, u16 tid,
-						 u8 requeue)
-{
-}
-static inline void ieee80211_requeue(struct ieee80211_local *local, int queue)
-{
-}
-static inline int ieee80211_wme_register(void)
-{
-	return 0;
-}
-static inline void ieee80211_wme_unregister(void)
-{
-}
-#endif /* CONFIG_NET_SCHED */
 
 #endif /* _WME_H */
-- 
cgit v1.2.3


From 92831bc395ac8390bf759775c50cb6f90c6eb03d Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Tue, 15 Jul 2008 03:48:01 -0700
Subject: netdev: Kill plain netif_schedule()

No more users.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h | 5 -----
 1 file changed, 5 deletions(-)

(limited to 'include')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 9464e6452967..787fbfc5aebb 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -994,11 +994,6 @@ static inline void netif_schedule_queue(struct netdev_queue *txq)
 		__netif_schedule(txq);
 }
 
-static inline void netif_schedule(struct net_device *dev)
-{
-	netif_schedule_queue(netdev_get_tx_queue(dev, 0));
-}
-
 static inline void netif_tx_schedule_all(struct net_device *dev)
 {
 	unsigned int i;
-- 
cgit v1.2.3


From d3b753db7c4f1f37a98b51974d484fda5d86dab5 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Tue, 15 Jul 2008 20:14:35 -0700
Subject: pkt_sched: Move gso_skb into Qdisc.

We liberate any dangling gso_skb during qdisc destruction.

It really only matters for the root qdisc.  But when qdiscs
can be shared by multiple netdev_queue objects, we can't
have the gso_skb in the netdev_queue any more.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h |  1 -
 include/net/sch_generic.h |  1 +
 net/sched/sch_generic.c   | 19 +++++++++----------
 3 files changed, 10 insertions(+), 11 deletions(-)

(limited to 'include')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 787fbfc5aebb..0883fcf2d16a 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -449,7 +449,6 @@ struct netdev_queue {
 	struct net_device	*dev;
 	struct Qdisc		*qdisc;
 	unsigned long		state;
-	struct sk_buff		*gso_skb;
 	spinlock_t		_xmit_lock;
 	int			xmit_lock_owner;
 	struct Qdisc		*qdisc_sleeping;
diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h
index b47f556c66f8..b96c3d9e10a8 100644
--- a/include/net/sch_generic.h
+++ b/include/net/sch_generic.h
@@ -36,6 +36,7 @@ struct Qdisc
 	u32			handle;
 	u32			parent;
 	atomic_t		refcnt;
+	struct sk_buff		*gso_skb;
 	struct sk_buff_head	q;
 	struct netdev_queue	*dev_queue;
 	struct list_head	list;
diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c
index 2f575b9017d1..2bd75befa066 100644
--- a/net/sched/sch_generic.c
+++ b/net/sched/sch_generic.c
@@ -77,7 +77,7 @@ static inline int dev_requeue_skb(struct sk_buff *skb,
 				  struct Qdisc *q)
 {
 	if (unlikely(skb->next))
-		dev_queue->gso_skb = skb;
+		q->gso_skb = skb;
 	else
 		q->ops->requeue(skb, q);
 
@@ -85,13 +85,12 @@ static inline int dev_requeue_skb(struct sk_buff *skb,
 	return 0;
 }
 
-static inline struct sk_buff *dequeue_skb(struct netdev_queue *dev_queue,
-					  struct Qdisc *q)
+static inline struct sk_buff *dequeue_skb(struct Qdisc *q)
 {
 	struct sk_buff *skb;
 
-	if ((skb = dev_queue->gso_skb))
-		dev_queue->gso_skb = NULL;
+	if ((skb = q->gso_skb))
+		q->gso_skb = NULL;
 	else
 		skb = q->dequeue(q);
 
@@ -155,10 +154,9 @@ static inline int qdisc_restart(struct netdev_queue *txq)
 	struct sk_buff *skb;
 
 	/* Dequeue packet */
-	if (unlikely((skb = dequeue_skb(txq, q)) == NULL))
+	if (unlikely((skb = dequeue_skb(q)) == NULL))
 		return 0;
 
-
 	/* And release queue */
 	spin_unlock(&txq->lock);
 
@@ -643,8 +641,8 @@ static void dev_deactivate_queue(struct net_device *dev,
 				 void *_qdisc_default)
 {
 	struct Qdisc *qdisc_default = _qdisc_default;
+	struct sk_buff *skb = NULL;
 	struct Qdisc *qdisc;
-	struct sk_buff *skb;
 
 	spin_lock_bh(&dev_queue->lock);
 
@@ -652,9 +650,10 @@ static void dev_deactivate_queue(struct net_device *dev,
 	if (qdisc) {
 		dev_queue->qdisc = qdisc_default;
 		qdisc_reset(qdisc);
+
+		skb = qdisc->gso_skb;
+		qdisc->gso_skb = NULL;
 	}
-	skb = dev_queue->gso_skb;
-	dev_queue->gso_skb = NULL;
 
 	spin_unlock_bh(&dev_queue->lock);
 
-- 
cgit v1.2.3


From e2627c8c2241bce45e368e150654d076b58a4595 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Wed, 16 Jul 2008 00:56:32 -0700
Subject: pkt_sched: Make QDISC_RUNNING a qdisc state.

Currently it is associated with a netdev_queue, but when we have
qdisc sharing that no longer makes any sense.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h |  1 -
 include/net/pkt_sched.h   |  4 +++-
 include/net/sch_generic.h |  6 ++++++
 net/sched/sch_generic.c   | 18 ++++++++++--------
 4 files changed, 19 insertions(+), 10 deletions(-)

(limited to 'include')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 0883fcf2d16a..9240a95793be 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -441,7 +441,6 @@ static inline void napi_synchronize(const struct napi_struct *n)
 enum netdev_queue_state_t
 {
 	__QUEUE_STATE_XOFF,
-	__QUEUE_STATE_QDISC_RUNNING,
 };
 
 struct netdev_queue {
diff --git a/include/net/pkt_sched.h b/include/net/pkt_sched.h
index cb9527815606..06a442d85186 100644
--- a/include/net/pkt_sched.h
+++ b/include/net/pkt_sched.h
@@ -88,8 +88,10 @@ extern void __qdisc_run(struct netdev_queue *txq);
 
 static inline void qdisc_run(struct netdev_queue *txq)
 {
+	struct Qdisc *q = txq->qdisc;
+
 	if (!netif_tx_queue_stopped(txq) &&
-	    !test_and_set_bit(__QUEUE_STATE_QDISC_RUNNING, &txq->state))
+	    !test_and_set_bit(__QDISC_STATE_RUNNING, &q->state))
 		__qdisc_run(txq);
 }
 
diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h
index b96c3d9e10a8..bc2a09da21b1 100644
--- a/include/net/sch_generic.h
+++ b/include/net/sch_generic.h
@@ -23,6 +23,11 @@ struct qdisc_rate_table
 	int		refcnt;
 };
 
+enum qdisc_state_t
+{
+	__QDISC_STATE_RUNNING,
+};
+
 struct Qdisc
 {
 	int 			(*enqueue)(struct sk_buff *skb, struct Qdisc *dev);
@@ -36,6 +41,7 @@ struct Qdisc
 	u32			handle;
 	u32			parent;
 	atomic_t		refcnt;
+	unsigned long		state;
 	struct sk_buff		*gso_skb;
 	struct sk_buff_head	q;
 	struct netdev_queue	*dev_queue;
diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c
index 2bd75befa066..ac208c2b2d10 100644
--- a/net/sched/sch_generic.c
+++ b/net/sched/sch_generic.c
@@ -130,8 +130,8 @@ static inline int handle_dev_cpu_collision(struct sk_buff *skb,
 /*
  * NOTE: Called under queue->lock with locally disabled BH.
  *
- * __QUEUE_STATE_QDISC_RUNNING guarantees only one CPU can process
- * this queue at a time. queue->lock serializes queue accesses for
+ * __QDISC_STATE_RUNNING guarantees only one CPU can process
+ * this qdisc at a time. queue->lock serializes queue accesses for
  * this queue AND txq->qdisc pointer itself.
  *
  *  netif_tx_lock serializes accesses to device driver.
@@ -146,9 +146,9 @@ static inline int handle_dev_cpu_collision(struct sk_buff *skb,
  *				>0 - queue is not empty.
  *
  */
-static inline int qdisc_restart(struct netdev_queue *txq)
+static inline int qdisc_restart(struct netdev_queue *txq,
+				struct Qdisc *q)
 {
-	struct Qdisc *q = txq->qdisc;
 	int ret = NETDEV_TX_BUSY;
 	struct net_device *dev;
 	struct sk_buff *skb;
@@ -168,7 +168,6 @@ static inline int qdisc_restart(struct netdev_queue *txq)
 	HARD_TX_UNLOCK(dev, txq);
 
 	spin_lock(&txq->lock);
-	q = txq->qdisc;
 
 	switch (ret) {
 	case NETDEV_TX_OK:
@@ -197,8 +196,9 @@ static inline int qdisc_restart(struct netdev_queue *txq)
 void __qdisc_run(struct netdev_queue *txq)
 {
 	unsigned long start_time = jiffies;
+	struct Qdisc *q = txq->qdisc;
 
-	while (qdisc_restart(txq)) {
+	while (qdisc_restart(txq, q)) {
 		if (netif_tx_queue_stopped(txq))
 			break;
 
@@ -213,7 +213,7 @@ void __qdisc_run(struct netdev_queue *txq)
 		}
 	}
 
-	clear_bit(__QUEUE_STATE_QDISC_RUNNING, &txq->state);
+	clear_bit(__QDISC_STATE_RUNNING, &q->state);
 }
 
 static void dev_watchdog(unsigned long arg)
@@ -666,14 +666,16 @@ static bool some_qdisc_is_running(struct net_device *dev, int lock)
 
 	for (i = 0; i < dev->num_tx_queues; i++) {
 		struct netdev_queue *dev_queue;
+		struct Qdisc *q;
 		int val;
 
 		dev_queue = netdev_get_tx_queue(dev, i);
+		q = dev_queue->qdisc;
 
 		if (lock)
 			spin_lock_bh(&dev_queue->lock);
 
-		val = test_bit(__QUEUE_STATE_QDISC_RUNNING, &dev_queue->state);
+		val = test_bit(__QDISC_STATE_RUNNING, &q->state);
 
 		if (lock)
 			spin_unlock_bh(&dev_queue->lock);
-- 
cgit v1.2.3


From 7698b4fcabcd790efc4f226bada1e7b5870653af Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Wed, 16 Jul 2008 01:42:40 -0700
Subject: pkt_sched: Add and use qdisc_root() and qdisc_root_lock().

When code wants to lock the qdisc tree state, the logic
operation it's doing is locking the top-level qdisc that
sits of the root of the netdev_queue.

Add qdisc_root_lock() to represent this and convert the
easiest cases.

In order for this to work out in all cases, we have to
hook up the noop_qdisc to a dummy netdev_queue.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/sch_generic.h | 12 ++++++++++++
 net/sched/sch_api.c       |  8 ++++----
 net/sched/sch_cbq.c       |  9 +++++----
 net/sched/sch_generic.c   | 21 ++++++++++++++++-----
 net/sched/sch_hfsc.c      |  4 ++--
 net/sched/sch_htb.c       | 18 ++++++++++--------
 net/sched/sch_netem.c     |  9 ++++++---
 7 files changed, 55 insertions(+), 26 deletions(-)

(limited to 'include')

diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h
index bc2a09da21b1..92417825d387 100644
--- a/include/net/sch_generic.h
+++ b/include/net/sch_generic.h
@@ -161,6 +161,18 @@ struct tcf_proto
 	struct tcf_proto_ops	*ops;
 };
 
+static inline struct Qdisc *qdisc_root(struct Qdisc *qdisc)
+{
+	return qdisc->dev_queue->qdisc;
+}
+
+static inline spinlock_t *qdisc_root_lock(struct Qdisc *qdisc)
+{
+	struct Qdisc *root = qdisc_root(qdisc);
+
+	return &root->dev_queue->lock;
+}
+
 static inline struct net_device *qdisc_dev(struct Qdisc *qdisc)
 {
 	return qdisc->dev_queue->dev;
diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c
index 830ccc544a15..19c244a00839 100644
--- a/net/sched/sch_api.c
+++ b/net/sched/sch_api.c
@@ -633,7 +633,7 @@ qdisc_create(struct net_device *dev, struct netdev_queue *dev_queue,
 	if (!ops->init || (err = ops->init(sch, tca[TCA_OPTIONS])) == 0) {
 		if (tca[TCA_RATE]) {
 			err = gen_new_estimator(&sch->bstats, &sch->rate_est,
-						&sch->dev_queue->lock,
+						qdisc_root_lock(sch),
 						tca[TCA_RATE]);
 			if (err) {
 				/*
@@ -675,7 +675,7 @@ static int qdisc_change(struct Qdisc *sch, struct nlattr **tca)
 	}
 	if (tca[TCA_RATE])
 		gen_replace_estimator(&sch->bstats, &sch->rate_est,
-				      &sch->dev_queue->lock, tca[TCA_RATE]);
+				      qdisc_root_lock(sch), tca[TCA_RATE]);
 	return 0;
 }
 
@@ -967,7 +967,7 @@ static int tc_fill_qdisc(struct sk_buff *skb, struct Qdisc *q, u32 clid,
 	q->qstats.qlen = q->q.qlen;
 
 	if (gnet_stats_start_copy_compat(skb, TCA_STATS2, TCA_STATS,
-					 TCA_XSTATS, &q->dev_queue->lock, &d) < 0)
+					 TCA_XSTATS, qdisc_root_lock(q), &d) < 0)
 		goto nla_put_failure;
 
 	if (q->ops->dump_stats && q->ops->dump_stats(q, &d) < 0)
@@ -1216,7 +1216,7 @@ static int tc_fill_tclass(struct sk_buff *skb, struct Qdisc *q,
 		goto nla_put_failure;
 
 	if (gnet_stats_start_copy_compat(skb, TCA_STATS2, TCA_STATS,
-					 TCA_XSTATS, &q->dev_queue->lock, &d) < 0)
+					 TCA_XSTATS, qdisc_root_lock(q), &d) < 0)
 		goto nla_put_failure;
 
 	if (cl_ops->dump_stats && cl_ops->dump_stats(q, cl, &d) < 0)
diff --git a/net/sched/sch_cbq.c b/net/sched/sch_cbq.c
index 4efc836cbf38..37ae653db683 100644
--- a/net/sched/sch_cbq.c
+++ b/net/sched/sch_cbq.c
@@ -1744,12 +1744,13 @@ static void cbq_put(struct Qdisc *sch, unsigned long arg)
 
 	if (--cl->refcnt == 0) {
 #ifdef CONFIG_NET_CLS_ACT
+		spinlock_t *root_lock = qdisc_root_lock(sch);
 		struct cbq_sched_data *q = qdisc_priv(sch);
 
-		spin_lock_bh(&sch->dev_queue->lock);
+		spin_lock_bh(root_lock);
 		if (q->rx_class == cl)
 			q->rx_class = NULL;
-		spin_unlock_bh(&sch->dev_queue->lock);
+		spin_unlock_bh(root_lock);
 #endif
 
 		cbq_destroy_class(sch, cl);
@@ -1828,7 +1829,7 @@ cbq_change_class(struct Qdisc *sch, u32 classid, u32 parentid, struct nlattr **t
 
 		if (tca[TCA_RATE])
 			gen_replace_estimator(&cl->bstats, &cl->rate_est,
-					      &sch->dev_queue->lock,
+					      qdisc_root_lock(sch),
 					      tca[TCA_RATE]);
 		return 0;
 	}
@@ -1919,7 +1920,7 @@ cbq_change_class(struct Qdisc *sch, u32 classid, u32 parentid, struct nlattr **t
 
 	if (tca[TCA_RATE])
 		gen_new_estimator(&cl->bstats, &cl->rate_est,
-				  &sch->dev_queue->lock, tca[TCA_RATE]);
+				  qdisc_root_lock(sch), tca[TCA_RATE]);
 
 	*arg = (unsigned long)cl;
 	return 0;
diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c
index ac208c2b2d10..739a8711ab30 100644
--- a/net/sched/sch_generic.c
+++ b/net/sched/sch_generic.c
@@ -151,14 +151,17 @@ static inline int qdisc_restart(struct netdev_queue *txq,
 {
 	int ret = NETDEV_TX_BUSY;
 	struct net_device *dev;
+	spinlock_t *root_lock;
 	struct sk_buff *skb;
 
 	/* Dequeue packet */
 	if (unlikely((skb = dequeue_skb(q)) == NULL))
 		return 0;
 
-	/* And release queue */
-	spin_unlock(&txq->lock);
+	root_lock = qdisc_root_lock(q);
+
+	/* And release qdisc */
+	spin_unlock(root_lock);
 
 	dev = txq->dev;
 
@@ -167,7 +170,7 @@ static inline int qdisc_restart(struct netdev_queue *txq,
 		ret = dev_hard_start_xmit(skb, dev, txq);
 	HARD_TX_UNLOCK(dev, txq);
 
-	spin_lock(&txq->lock);
+	spin_lock(root_lock);
 
 	switch (ret) {
 	case NETDEV_TX_OK:
@@ -345,12 +348,18 @@ struct Qdisc_ops noop_qdisc_ops __read_mostly = {
 	.owner		=	THIS_MODULE,
 };
 
+static struct netdev_queue noop_netdev_queue = {
+	.lock		=	__SPIN_LOCK_UNLOCKED(noop_netdev_queue.lock),
+	.qdisc		=	&noop_qdisc,
+};
+
 struct Qdisc noop_qdisc = {
 	.enqueue	=	noop_enqueue,
 	.dequeue	=	noop_dequeue,
 	.flags		=	TCQ_F_BUILTIN,
 	.ops		=	&noop_qdisc_ops,
 	.list		=	LIST_HEAD_INIT(noop_qdisc.list),
+	.dev_queue	=	&noop_netdev_queue,
 };
 EXPORT_SYMBOL(noop_qdisc);
 
@@ -666,19 +675,21 @@ static bool some_qdisc_is_running(struct net_device *dev, int lock)
 
 	for (i = 0; i < dev->num_tx_queues; i++) {
 		struct netdev_queue *dev_queue;
+		spinlock_t *root_lock;
 		struct Qdisc *q;
 		int val;
 
 		dev_queue = netdev_get_tx_queue(dev, i);
 		q = dev_queue->qdisc;
+		root_lock = qdisc_root_lock(q);
 
 		if (lock)
-			spin_lock_bh(&dev_queue->lock);
+			spin_lock_bh(root_lock);
 
 		val = test_bit(__QDISC_STATE_RUNNING, &q->state);
 
 		if (lock)
-			spin_unlock_bh(&dev_queue->lock);
+			spin_unlock_bh(root_lock);
 
 		if (val)
 			return true;
diff --git a/net/sched/sch_hfsc.c b/net/sched/sch_hfsc.c
index 997d520ca580..5090708ba384 100644
--- a/net/sched/sch_hfsc.c
+++ b/net/sched/sch_hfsc.c
@@ -1045,7 +1045,7 @@ hfsc_change_class(struct Qdisc *sch, u32 classid, u32 parentid,
 
 		if (tca[TCA_RATE])
 			gen_replace_estimator(&cl->bstats, &cl->rate_est,
-					      &sch->dev_queue->lock,
+					      qdisc_root_lock(sch),
 					      tca[TCA_RATE]);
 		return 0;
 	}
@@ -1104,7 +1104,7 @@ hfsc_change_class(struct Qdisc *sch, u32 classid, u32 parentid,
 
 	if (tca[TCA_RATE])
 		gen_new_estimator(&cl->bstats, &cl->rate_est,
-				  &sch->dev_queue->lock, tca[TCA_RATE]);
+				  qdisc_root_lock(sch), tca[TCA_RATE]);
 	*arg = (unsigned long)cl;
 	return 0;
 }
diff --git a/net/sched/sch_htb.c b/net/sched/sch_htb.c
index c8ca54cc26b0..ee48457eaa4a 100644
--- a/net/sched/sch_htb.c
+++ b/net/sched/sch_htb.c
@@ -1039,11 +1039,12 @@ static int htb_init(struct Qdisc *sch, struct nlattr *opt)
 
 static int htb_dump(struct Qdisc *sch, struct sk_buff *skb)
 {
+	spinlock_t *root_lock = qdisc_root_lock(sch);
 	struct htb_sched *q = qdisc_priv(sch);
 	struct nlattr *nest;
 	struct tc_htb_glob gopt;
 
-	spin_lock_bh(&sch->dev_queue->lock);
+	spin_lock_bh(root_lock);
 
 	gopt.direct_pkts = q->direct_pkts;
 	gopt.version = HTB_VER;
@@ -1057,11 +1058,11 @@ static int htb_dump(struct Qdisc *sch, struct sk_buff *skb)
 	NLA_PUT(skb, TCA_HTB_INIT, sizeof(gopt), &gopt);
 	nla_nest_end(skb, nest);
 
-	spin_unlock_bh(&sch->dev_queue->lock);
+	spin_unlock_bh(root_lock);
 	return skb->len;
 
 nla_put_failure:
-	spin_unlock_bh(&sch->dev_queue->lock);
+	spin_unlock_bh(root_lock);
 	nla_nest_cancel(skb, nest);
 	return -1;
 }
@@ -1070,10 +1071,11 @@ static int htb_dump_class(struct Qdisc *sch, unsigned long arg,
 			  struct sk_buff *skb, struct tcmsg *tcm)
 {
 	struct htb_class *cl = (struct htb_class *)arg;
+	spinlock_t *root_lock = qdisc_root_lock(sch);
 	struct nlattr *nest;
 	struct tc_htb_opt opt;
 
-	spin_lock_bh(&sch->dev_queue->lock);
+	spin_lock_bh(root_lock);
 	tcm->tcm_parent = cl->parent ? cl->parent->common.classid : TC_H_ROOT;
 	tcm->tcm_handle = cl->common.classid;
 	if (!cl->level && cl->un.leaf.q)
@@ -1095,11 +1097,11 @@ static int htb_dump_class(struct Qdisc *sch, unsigned long arg,
 	NLA_PUT(skb, TCA_HTB_PARMS, sizeof(opt), &opt);
 
 	nla_nest_end(skb, nest);
-	spin_unlock_bh(&sch->dev_queue->lock);
+	spin_unlock_bh(root_lock);
 	return skb->len;
 
 nla_put_failure:
-	spin_unlock_bh(&sch->dev_queue->lock);
+	spin_unlock_bh(root_lock);
 	nla_nest_cancel(skb, nest);
 	return -1;
 }
@@ -1365,7 +1367,7 @@ static int htb_change_class(struct Qdisc *sch, u32 classid,
 			goto failure;
 
 		gen_new_estimator(&cl->bstats, &cl->rate_est,
-				  &sch->dev_queue->lock,
+				  qdisc_root_lock(sch),
 				  tca[TCA_RATE] ? : &est.nla);
 		cl->refcnt = 1;
 		cl->children = 0;
@@ -1420,7 +1422,7 @@ static int htb_change_class(struct Qdisc *sch, u32 classid,
 	} else {
 		if (tca[TCA_RATE])
 			gen_replace_estimator(&cl->bstats, &cl->rate_est,
-					      &sch->dev_queue->lock,
+					      qdisc_root_lock(sch),
 					      tca[TCA_RATE]);
 		sch_tree_lock(sch);
 	}
diff --git a/net/sched/sch_netem.c b/net/sched/sch_netem.c
index bc585f2089ff..c5ea40c9eb21 100644
--- a/net/sched/sch_netem.c
+++ b/net/sched/sch_netem.c
@@ -180,7 +180,7 @@ static int netem_enqueue(struct sk_buff *skb, struct Qdisc *sch)
 	 * skb will be queued.
 	 */
 	if (count > 1 && (skb2 = skb_clone(skb, GFP_ATOMIC)) != NULL) {
-		struct Qdisc *rootq = sch->dev_queue->qdisc;
+		struct Qdisc *rootq = qdisc_root(sch);
 		u32 dupsave = q->duplicate; /* prevent duplicating a dup... */
 		q->duplicate = 0;
 
@@ -319,6 +319,7 @@ static int get_dist_table(struct Qdisc *sch, const struct nlattr *attr)
 	struct netem_sched_data *q = qdisc_priv(sch);
 	unsigned long n = nla_len(attr)/sizeof(__s16);
 	const __s16 *data = nla_data(attr);
+	spinlock_t *root_lock;
 	struct disttable *d;
 	int i;
 
@@ -333,9 +334,11 @@ static int get_dist_table(struct Qdisc *sch, const struct nlattr *attr)
 	for (i = 0; i < n; i++)
 		d->table[i] = data[i];
 
-	spin_lock_bh(&sch->dev_queue->lock);
+	root_lock = qdisc_root_lock(sch);
+
+	spin_lock_bh(root_lock);
 	d = xchg(&q->delay_dist, d);
-	spin_unlock_bh(&sch->dev_queue->lock);
+	spin_unlock_bh(root_lock);
 
 	kfree(d);
 	return 0;
-- 
cgit v1.2.3


From 37437bb2e1ae8af470dfcd5b4ff454110894ccaf Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Wed, 16 Jul 2008 02:15:04 -0700
Subject: pkt_sched: Schedule qdiscs instead of netdev_queue.

When we have shared qdiscs, packets come out of the qdiscs
for multiple transmit queues.

Therefore it doesn't make any sense to schedule the transmit
queue when logically we cannot know ahead of time the TX
queue of the SKB that the qdisc->dequeue() will give us.

Just for sanity I added a BUG check to make sure we never
get into a state where the noop_qdisc is scheduled.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h | 12 ++++-----
 include/net/pkt_sched.h   | 11 +++-----
 include/net/sch_generic.h |  2 ++
 net/core/dev.c            | 68 +++++++++++++++++++----------------------------
 net/sched/sch_api.c       |  3 +--
 net/sched/sch_cbq.c       |  2 +-
 net/sched/sch_generic.c   | 30 ++++++++++-----------
 7 files changed, 55 insertions(+), 73 deletions(-)

(limited to 'include')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 9240a95793be..1e839fa01434 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -275,7 +275,6 @@ enum netdev_state_t
 {
 	__LINK_STATE_START,
 	__LINK_STATE_PRESENT,
-	__LINK_STATE_SCHED,
 	__LINK_STATE_NOCARRIER,
 	__LINK_STATE_LINKWATCH_PENDING,
 	__LINK_STATE_DORMANT,
@@ -452,7 +451,6 @@ struct netdev_queue {
 	int			xmit_lock_owner;
 	struct Qdisc		*qdisc_sleeping;
 	struct list_head	qdisc_list;
-	struct netdev_queue	*next_sched;
 } ____cacheline_aligned_in_smp;
 
 /*
@@ -969,7 +967,7 @@ static inline int unregister_gifconf(unsigned int family)
  */
 struct softnet_data
 {
-	struct netdev_queue	*output_queue;
+	struct Qdisc		*output_queue;
 	struct sk_buff_head	input_pkt_queue;
 	struct list_head	poll_list;
 	struct sk_buff		*completion_queue;
@@ -984,12 +982,12 @@ DECLARE_PER_CPU(struct softnet_data,softnet_data);
 
 #define HAVE_NETIF_QUEUE
 
-extern void __netif_schedule(struct netdev_queue *txq);
+extern void __netif_schedule(struct Qdisc *q);
 
 static inline void netif_schedule_queue(struct netdev_queue *txq)
 {
 	if (!test_bit(__QUEUE_STATE_XOFF, &txq->state))
-		__netif_schedule(txq);
+		__netif_schedule(txq->qdisc);
 }
 
 static inline void netif_tx_schedule_all(struct net_device *dev)
@@ -1042,7 +1040,7 @@ static inline void netif_tx_wake_queue(struct netdev_queue *dev_queue)
 	}
 #endif
 	if (test_and_clear_bit(__QUEUE_STATE_XOFF, &dev_queue->state))
-		__netif_schedule(dev_queue);
+		__netif_schedule(dev_queue->qdisc);
 }
 
 static inline void netif_wake_queue(struct net_device *dev)
@@ -1186,7 +1184,7 @@ static inline void netif_wake_subqueue(struct net_device *dev, u16 queue_index)
 		return;
 #endif
 	if (test_and_clear_bit(__QUEUE_STATE_XOFF, &txq->state))
-		__netif_schedule(txq);
+		__netif_schedule(txq->qdisc);
 }
 
 /**
diff --git a/include/net/pkt_sched.h b/include/net/pkt_sched.h
index 06a442d85186..e4e30052e4e2 100644
--- a/include/net/pkt_sched.h
+++ b/include/net/pkt_sched.h
@@ -84,15 +84,12 @@ extern struct qdisc_rate_table *qdisc_get_rtab(struct tc_ratespec *r,
 		struct nlattr *tab);
 extern void qdisc_put_rtab(struct qdisc_rate_table *tab);
 
-extern void __qdisc_run(struct netdev_queue *txq);
+extern void __qdisc_run(struct Qdisc *q);
 
-static inline void qdisc_run(struct netdev_queue *txq)
+static inline void qdisc_run(struct Qdisc *q)
 {
-	struct Qdisc *q = txq->qdisc;
-
-	if (!netif_tx_queue_stopped(txq) &&
-	    !test_and_set_bit(__QDISC_STATE_RUNNING, &q->state))
-		__qdisc_run(txq);
+	if (!test_and_set_bit(__QDISC_STATE_RUNNING, &q->state))
+		__qdisc_run(q);
 }
 
 extern int tc_classify_compat(struct sk_buff *skb, struct tcf_proto *tp,
diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h
index 92417825d387..3cc4b5cd8c6a 100644
--- a/include/net/sch_generic.h
+++ b/include/net/sch_generic.h
@@ -26,6 +26,7 @@ struct qdisc_rate_table
 enum qdisc_state_t
 {
 	__QDISC_STATE_RUNNING,
+	__QDISC_STATE_SCHED,
 };
 
 struct Qdisc
@@ -45,6 +46,7 @@ struct Qdisc
 	struct sk_buff		*gso_skb;
 	struct sk_buff_head	q;
 	struct netdev_queue	*dev_queue;
+	struct Qdisc		*next_sched;
 	struct list_head	list;
 
 	struct gnet_stats_basic	bstats;
diff --git a/net/core/dev.c b/net/core/dev.c
index 467bfb325123..0b909b74f698 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -1323,18 +1323,18 @@ static void dev_queue_xmit_nit(struct sk_buff *skb, struct net_device *dev)
 }
 
 
-void __netif_schedule(struct netdev_queue *txq)
+void __netif_schedule(struct Qdisc *q)
 {
-	struct net_device *dev = txq->dev;
+	BUG_ON(q == &noop_qdisc);
 
-	if (!test_and_set_bit(__LINK_STATE_SCHED, &dev->state)) {
+	if (!test_and_set_bit(__QDISC_STATE_SCHED, &q->state)) {
 		struct softnet_data *sd;
 		unsigned long flags;
 
 		local_irq_save(flags);
 		sd = &__get_cpu_var(softnet_data);
-		txq->next_sched = sd->output_queue;
-		sd->output_queue = txq;
+		q->next_sched = sd->output_queue;
+		sd->output_queue = q;
 		raise_softirq_irqoff(NET_TX_SOFTIRQ);
 		local_irq_restore(flags);
 	}
@@ -1771,37 +1771,23 @@ gso:
 	rcu_read_lock_bh();
 
 	txq = dev_pick_tx(dev, skb);
-	spin_lock_prefetch(&txq->lock);
-
-	/* Updates of qdisc are serialized by queue->lock.
-	 * The struct Qdisc which is pointed to by qdisc is now a
-	 * rcu structure - it may be accessed without acquiring
-	 * a lock (but the structure may be stale.) The freeing of the
-	 * qdisc will be deferred until it's known that there are no
-	 * more references to it.
-	 *
-	 * If the qdisc has an enqueue function, we still need to
-	 * hold the queue->lock before calling it, since queue->lock
-	 * also serializes access to the device queue.
-	 */
-
 	q = rcu_dereference(txq->qdisc);
+
 #ifdef CONFIG_NET_CLS_ACT
 	skb->tc_verd = SET_TC_AT(skb->tc_verd,AT_EGRESS);
 #endif
 	if (q->enqueue) {
-		/* Grab device queue */
-		spin_lock(&txq->lock);
-		q = txq->qdisc;
-		if (q->enqueue) {
-			rc = q->enqueue(skb, q);
-			qdisc_run(txq);
-			spin_unlock(&txq->lock);
-
-			rc = rc == NET_XMIT_BYPASS ? NET_XMIT_SUCCESS : rc;
-			goto out;
-		}
-		spin_unlock(&txq->lock);
+		spinlock_t *root_lock = qdisc_root_lock(q);
+
+		spin_lock(root_lock);
+
+		rc = q->enqueue(skb, q);
+		qdisc_run(q);
+
+		spin_unlock(root_lock);
+
+		rc = rc == NET_XMIT_BYPASS ? NET_XMIT_SUCCESS : rc;
+		goto out;
 	}
 
 	/* The device has no queue. Common case for software devices:
@@ -1974,7 +1960,7 @@ static void net_tx_action(struct softirq_action *h)
 	}
 
 	if (sd->output_queue) {
-		struct netdev_queue *head;
+		struct Qdisc *head;
 
 		local_irq_disable();
 		head = sd->output_queue;
@@ -1982,18 +1968,20 @@ static void net_tx_action(struct softirq_action *h)
 		local_irq_enable();
 
 		while (head) {
-			struct netdev_queue *txq = head;
-			struct net_device *dev = txq->dev;
+			struct Qdisc *q = head;
+			spinlock_t *root_lock;
+
 			head = head->next_sched;
 
 			smp_mb__before_clear_bit();
-			clear_bit(__LINK_STATE_SCHED, &dev->state);
+			clear_bit(__QDISC_STATE_SCHED, &q->state);
 
-			if (spin_trylock(&txq->lock)) {
-				qdisc_run(txq);
-				spin_unlock(&txq->lock);
+			root_lock = qdisc_root_lock(q);
+			if (spin_trylock(root_lock)) {
+				qdisc_run(q);
+				spin_unlock(root_lock);
 			} else {
-				netif_schedule_queue(txq);
+				__netif_schedule(q);
 			}
 		}
 	}
@@ -4459,7 +4447,7 @@ static int dev_cpu_callback(struct notifier_block *nfb,
 			    void *ocpu)
 {
 	struct sk_buff **list_skb;
-	struct netdev_queue **list_net;
+	struct Qdisc **list_net;
 	struct sk_buff *skb;
 	unsigned int cpu, oldcpu = (unsigned long)ocpu;
 	struct softnet_data *sd, *oldsd;
diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c
index 19c244a00839..8e8c5becc348 100644
--- a/net/sched/sch_api.c
+++ b/net/sched/sch_api.c
@@ -294,11 +294,10 @@ static enum hrtimer_restart qdisc_watchdog(struct hrtimer *timer)
 {
 	struct qdisc_watchdog *wd = container_of(timer, struct qdisc_watchdog,
 						 timer);
-	struct netdev_queue *txq = wd->qdisc->dev_queue;
 
 	wd->qdisc->flags &= ~TCQ_F_THROTTLED;
 	smp_wmb();
-	netif_schedule_queue(txq);
+	__netif_schedule(wd->qdisc);
 
 	return HRTIMER_NORESTART;
 }
diff --git a/net/sched/sch_cbq.c b/net/sched/sch_cbq.c
index 37ae653db683..a3953bbe2d79 100644
--- a/net/sched/sch_cbq.c
+++ b/net/sched/sch_cbq.c
@@ -650,7 +650,7 @@ static enum hrtimer_restart cbq_undelay(struct hrtimer *timer)
 	}
 
 	sch->flags &= ~TCQ_F_THROTTLED;
-	netif_schedule_queue(sch->dev_queue);
+	__netif_schedule(sch);
 	return HRTIMER_NORESTART;
 }
 
diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c
index 739a8711ab30..dd5c4e70abe4 100644
--- a/net/sched/sch_generic.c
+++ b/net/sched/sch_generic.c
@@ -72,16 +72,14 @@ static inline int qdisc_qlen(struct Qdisc *q)
 	return q->q.qlen;
 }
 
-static inline int dev_requeue_skb(struct sk_buff *skb,
-				  struct netdev_queue *dev_queue,
-				  struct Qdisc *q)
+static inline int dev_requeue_skb(struct sk_buff *skb, struct Qdisc *q)
 {
 	if (unlikely(skb->next))
 		q->gso_skb = skb;
 	else
 		q->ops->requeue(skb, q);
 
-	netif_schedule_queue(dev_queue);
+	__netif_schedule(q);
 	return 0;
 }
 
@@ -121,7 +119,7 @@ static inline int handle_dev_cpu_collision(struct sk_buff *skb,
 		 * some time.
 		 */
 		__get_cpu_var(netdev_rx_stat).cpu_collision++;
-		ret = dev_requeue_skb(skb, dev_queue, q);
+		ret = dev_requeue_skb(skb, q);
 	}
 
 	return ret;
@@ -146,9 +144,9 @@ static inline int handle_dev_cpu_collision(struct sk_buff *skb,
  *				>0 - queue is not empty.
  *
  */
-static inline int qdisc_restart(struct netdev_queue *txq,
-				struct Qdisc *q)
+static inline int qdisc_restart(struct Qdisc *q)
 {
+	struct netdev_queue *txq;
 	int ret = NETDEV_TX_BUSY;
 	struct net_device *dev;
 	spinlock_t *root_lock;
@@ -163,7 +161,8 @@ static inline int qdisc_restart(struct netdev_queue *txq,
 	/* And release qdisc */
 	spin_unlock(root_lock);
 
-	dev = txq->dev;
+	dev = qdisc_dev(q);
+	txq = netdev_get_tx_queue(dev, skb_get_queue_mapping(skb));
 
 	HARD_TX_LOCK(dev, txq, smp_processor_id());
 	if (!netif_subqueue_stopped(dev, skb))
@@ -189,29 +188,28 @@ static inline int qdisc_restart(struct netdev_queue *txq,
 			printk(KERN_WARNING "BUG %s code %d qlen %d\n",
 			       dev->name, ret, q->q.qlen);
 
-		ret = dev_requeue_skb(skb, txq, q);
+		ret = dev_requeue_skb(skb, q);
 		break;
 	}
 
+	if (ret && netif_tx_queue_stopped(txq))
+		ret = 0;
+
 	return ret;
 }
 
-void __qdisc_run(struct netdev_queue *txq)
+void __qdisc_run(struct Qdisc *q)
 {
 	unsigned long start_time = jiffies;
-	struct Qdisc *q = txq->qdisc;
-
-	while (qdisc_restart(txq, q)) {
-		if (netif_tx_queue_stopped(txq))
-			break;
 
+	while (qdisc_restart(q)) {
 		/*
 		 * Postpone processing if
 		 * 1. another process needs the CPU;
 		 * 2. we've been doing it for too long.
 		 */
 		if (need_resched() || jiffies != start_time) {
-			netif_schedule_queue(txq);
+			__netif_schedule(q);
 			break;
 		}
 	}
-- 
cgit v1.2.3


From ead81cc5fc6d996db6afb20f211241612610a07a Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Thu, 17 Jul 2008 00:50:32 -0700
Subject: netdevice: Move qdisc_list back into net_device proper.

And give it it's own lock.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h |  3 ++-
 net/core/dev.c            |  2 ++
 net/sched/sch_api.c       | 31 +++++++------------------------
 net/sched/sch_generic.c   |  9 +++++++--
 4 files changed, 18 insertions(+), 27 deletions(-)

(limited to 'include')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 1e839fa01434..3170bcef734b 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -450,7 +450,6 @@ struct netdev_queue {
 	spinlock_t		_xmit_lock;
 	int			xmit_lock_owner;
 	struct Qdisc		*qdisc_sleeping;
-	struct list_head	qdisc_list;
 } ____cacheline_aligned_in_smp;
 
 /*
@@ -638,6 +637,8 @@ struct net_device
 	unsigned int		real_num_tx_queues;
 
 	unsigned long		tx_queue_len;	/* Max frames per queue allowed */
+	spinlock_t		qdisc_list_lock;
+	struct list_head	qdisc_list;
 
 /*
  * One part is mostly used on xmit path (device)
diff --git a/net/core/dev.c b/net/core/dev.c
index 0b909b74f698..6741e344ac59 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -3886,6 +3886,8 @@ int register_netdevice(struct net_device *dev)
 	net = dev_net(dev);
 
 	spin_lock_init(&dev->addr_list_lock);
+	spin_lock_init(&dev->qdisc_list_lock);
+	INIT_LIST_HEAD(&dev->qdisc_list);
 	netdev_init_queue_locks(dev);
 
 	dev->iflink = -1;
diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c
index 8e8c5becc348..6958fe7c9a77 100644
--- a/net/sched/sch_api.c
+++ b/net/sched/sch_api.c
@@ -183,30 +183,17 @@ EXPORT_SYMBOL(unregister_qdisc);
    (root qdisc, all its children, children of children etc.)
  */
 
-static struct Qdisc *__qdisc_lookup(struct netdev_queue *dev_queue, u32 handle)
+struct Qdisc *qdisc_lookup(struct net_device *dev, u32 handle)
 {
 	struct Qdisc *q;
 
-	list_for_each_entry(q, &dev_queue->qdisc_list, list) {
+	list_for_each_entry(q, &dev->qdisc_list, list) {
 		if (q->handle == handle)
 			return q;
 	}
 	return NULL;
 }
 
-struct Qdisc *qdisc_lookup(struct net_device *dev, u32 handle)
-{
-	unsigned int i;
-
-	for (i = 0; i < dev->num_tx_queues; i++) {
-		struct netdev_queue *txq = netdev_get_tx_queue(dev, i);
-		struct Qdisc *q = __qdisc_lookup(txq, handle);
-		if (q)
-			return q;
-	}
-	return NULL;
-}
-
 static struct Qdisc *qdisc_leaf(struct Qdisc *p, u32 classid)
 {
 	unsigned long cl;
@@ -645,9 +632,9 @@ qdisc_create(struct net_device *dev, struct netdev_queue *dev_queue,
 				goto err_out3;
 			}
 		}
-		qdisc_lock_tree(dev);
-		list_add_tail(&sch->list, &dev_queue->qdisc_list);
-		qdisc_unlock_tree(dev);
+		spin_lock_bh(&dev->qdisc_list_lock);
+		list_add_tail(&sch->list, &dev->qdisc_list);
+		spin_unlock_bh(&dev->qdisc_list_lock);
 
 		return sch;
 	}
@@ -1032,14 +1019,12 @@ static int tc_dump_qdisc(struct sk_buff *skb, struct netlink_callback *cb)
 	read_lock(&dev_base_lock);
 	idx = 0;
 	for_each_netdev(&init_net, dev) {
-		struct netdev_queue *dev_queue;
 		if (idx < s_idx)
 			goto cont;
 		if (idx > s_idx)
 			s_q_idx = 0;
 		q_idx = 0;
-		dev_queue = netdev_get_tx_queue(dev, 0);
-		list_for_each_entry(q, &dev_queue->qdisc_list, list) {
+		list_for_each_entry(q, &dev->qdisc_list, list) {
 			if (q_idx < s_q_idx) {
 				q_idx++;
 				continue;
@@ -1269,7 +1254,6 @@ static int qdisc_class_dump(struct Qdisc *q, unsigned long cl, struct qdisc_walk
 static int tc_dump_tclass(struct sk_buff *skb, struct netlink_callback *cb)
 {
 	struct net *net = sock_net(skb->sk);
-	struct netdev_queue *dev_queue;
 	int t;
 	int s_t;
 	struct net_device *dev;
@@ -1288,8 +1272,7 @@ static int tc_dump_tclass(struct sk_buff *skb, struct netlink_callback *cb)
 	s_t = cb->args[0];
 	t = 0;
 
-	dev_queue = netdev_get_tx_queue(dev, 0);
-	list_for_each_entry(q, &dev_queue->qdisc_list, list) {
+	list_for_each_entry(q, &dev->qdisc_list, list) {
 		if (t < s_t || !q->ops->cl_ops ||
 		    (tcm->tcm_parent &&
 		     TC_H_MAJ(tcm->tcm_parent) != q->handle)) {
diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c
index efa418a1b34e..8cdf0b4a6a5a 100644
--- a/net/sched/sch_generic.c
+++ b/net/sched/sch_generic.c
@@ -563,11 +563,15 @@ static void __qdisc_destroy(struct rcu_head *head)
 
 void qdisc_destroy(struct Qdisc *qdisc)
 {
+	struct net_device *dev = qdisc_dev(qdisc);
+
 	if (qdisc->flags & TCQ_F_BUILTIN ||
 	    !atomic_dec_and_test(&qdisc->refcnt))
 		return;
 
+	spin_lock_bh(&dev->qdisc_list_lock);
 	list_del(&qdisc->list);
+	spin_unlock_bh(&dev->qdisc_list_lock);
 
 	call_rcu(&qdisc->q_rcu, __qdisc_destroy);
 }
@@ -599,7 +603,9 @@ static void attach_one_default_qdisc(struct net_device *dev,
 			printk(KERN_INFO "%s: activation failed\n", dev->name);
 			return;
 		}
-		list_add_tail(&qdisc->list, &dev_queue->qdisc_list);
+		spin_lock_bh(&dev->qdisc_list_lock);
+		list_add_tail(&qdisc->list, &dev->qdisc_list);
+		spin_unlock_bh(&dev->qdisc_list_lock);
 	} else {
 		qdisc =  &noqueue_qdisc;
 	}
@@ -738,7 +744,6 @@ static void dev_init_scheduler_queue(struct net_device *dev,
 
 	dev_queue->qdisc = qdisc;
 	dev_queue->qdisc_sleeping = qdisc;
-	INIT_LIST_HEAD(&dev_queue->qdisc_list);
 }
 
 void dev_init_scheduler(struct net_device *dev)
-- 
cgit v1.2.3


From 78a5b30b7324b2d66bcf7d2e3935877d3c26497c Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Wed, 16 Jul 2008 03:12:24 -0700
Subject: pkt_sched: Rework {sch,tbf}_tree_lock().

Make sch_tree_lock() lock the qdisc's root.  All of the
users hold the RTNL semaphore and the root qdisc is not
changing.

Implement tbf_tree_{lock,unlock}() simply in terms of
sch_tree_{lock,unlock}().

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/sch_generic.h | 16 ++++++++++++----
 1 file changed, 12 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h
index 3cc4b5cd8c6a..60e1e9f9e453 100644
--- a/include/net/sch_generic.h
+++ b/include/net/sch_generic.h
@@ -183,10 +183,18 @@ static inline struct net_device *qdisc_dev(struct Qdisc *qdisc)
 extern void qdisc_lock_tree(struct net_device *dev);
 extern void qdisc_unlock_tree(struct net_device *dev);
 
-#define sch_tree_lock(q)	qdisc_lock_tree(qdisc_dev(q))
-#define sch_tree_unlock(q)	qdisc_unlock_tree(qdisc_dev(q))
-#define tcf_tree_lock(tp)	qdisc_lock_tree(qdisc_dev((tp)->q))
-#define tcf_tree_unlock(tp)	qdisc_unlock_tree(qdisc_dev((tp)->q))
+static inline void sch_tree_lock(struct Qdisc *q)
+{
+	spin_lock_bh(qdisc_root_lock(q));
+}
+
+static inline void sch_tree_unlock(struct Qdisc *q)
+{
+	spin_unlock_bh(qdisc_root_lock(q));
+}
+
+#define tcf_tree_lock(tp)	sch_tree_lock((tp)->q)
+#define tcf_tree_unlock(tp)	sch_tree_unlock((tp)->q)
 
 extern struct Qdisc noop_qdisc;
 extern struct Qdisc_ops noop_qdisc_ops;
-- 
cgit v1.2.3


From c7e4f3bbb4ba4e48ab3b529d5016e454cee1ccd6 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Wed, 16 Jul 2008 03:22:39 -0700
Subject: pkt_sched: Kill qdisc_lock_tree and qdisc_unlock_tree.

No longer used.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/sch_generic.h |  3 ---
 net/sched/sch_generic.c   | 36 +++---------------------------------
 2 files changed, 3 insertions(+), 36 deletions(-)

(limited to 'include')

diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h
index 60e1e9f9e453..1eef8d0c9990 100644
--- a/include/net/sch_generic.h
+++ b/include/net/sch_generic.h
@@ -180,9 +180,6 @@ static inline struct net_device *qdisc_dev(struct Qdisc *qdisc)
 	return qdisc->dev_queue->dev;
 }
 
-extern void qdisc_lock_tree(struct net_device *dev);
-extern void qdisc_unlock_tree(struct net_device *dev);
-
 static inline void sch_tree_lock(struct Qdisc *q)
 {
 	spin_lock_bh(qdisc_root_lock(q));
diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c
index 8cdf0b4a6a5a..3d53e92ad9c8 100644
--- a/net/sched/sch_generic.c
+++ b/net/sched/sch_generic.c
@@ -29,44 +29,14 @@
 /* Main transmission queue. */
 
 /* Modifications to data participating in scheduling must be protected with
- * queue->lock spinlock.
+ * qdisc_root_lock(qdisc) spinlock.
  *
  * The idea is the following:
- * - enqueue, dequeue are serialized via top level device
- *   spinlock queue->lock.
- * - ingress filtering is serialized via top level device
- *   spinlock dev->rx_queue.lock.
+ * - enqueue, dequeue are serialized via qdisc root lock
+ * - ingress filtering is also serialized via qdisc root lock
  * - updates to tree and tree walking are only done under the rtnl mutex.
  */
 
-void qdisc_lock_tree(struct net_device *dev)
-	__acquires(dev->rx_queue.lock)
-{
-	unsigned int i;
-
-	local_bh_disable();
-	for (i = 0; i < dev->num_tx_queues; i++) {
-		struct netdev_queue *txq = netdev_get_tx_queue(dev, i);
-		spin_lock(&txq->lock);
-	}
-	spin_lock(&dev->rx_queue.lock);
-}
-EXPORT_SYMBOL(qdisc_lock_tree);
-
-void qdisc_unlock_tree(struct net_device *dev)
-	__releases(dev->rx_queue.lock)
-{
-	unsigned int i;
-
-	spin_unlock(&dev->rx_queue.lock);
-	for (i = 0; i < dev->num_tx_queues; i++) {
-		struct netdev_queue *txq = netdev_get_tx_queue(dev, i);
-		spin_unlock(&txq->lock);
-	}
-	local_bh_enable();
-}
-EXPORT_SYMBOL(qdisc_unlock_tree);
-
 static inline int qdisc_qlen(struct Qdisc *q)
 {
 	return q->q.qlen;
-- 
cgit v1.2.3


From 83874000929ed63aef30b44083a9f713135ff040 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Thu, 17 Jul 2008 00:53:03 -0700
Subject: pkt_sched: Kill netdev_queue lock.

We can simply use the qdisc->q.lock for all of the
qdisc tree synchronization.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ifb.c         | 20 --------------------
 include/linux/netdevice.h |  1 -
 include/net/sch_generic.h |  7 ++++++-
 net/core/dev.c            |  9 +++++----
 net/mac80211/wme.c        | 19 ++++++++++++-------
 net/sched/sch_generic.c   | 32 +++++++++++++++-----------------
 net/sched/sch_teql.c      |  7 +++++--
 7 files changed, 43 insertions(+), 52 deletions(-)

(limited to 'include')

diff --git a/drivers/net/ifb.c b/drivers/net/ifb.c
index 897b05e79ed0..0960e69b2da4 100644
--- a/drivers/net/ifb.c
+++ b/drivers/net/ifb.c
@@ -35,7 +35,6 @@
 #include <linux/moduleparam.h>
 #include <net/pkt_sched.h>
 #include <net/net_namespace.h>
-#include <linux/lockdep.h>
 
 #define TX_TIMEOUT  (2*HZ)
 
@@ -228,22 +227,6 @@ static struct rtnl_link_ops ifb_link_ops __read_mostly = {
 module_param(numifbs, int, 0);
 MODULE_PARM_DESC(numifbs, "Number of ifb devices");
 
-/*
- * dev_ifb's TX queue lock is usually taken after dev->rx_queue.lock,
- * reversely to e.g. qdisc_lock_tree(). It should be safe until
- * ifb doesn't take dev's TX queue lock with dev_ifb->rx_queue.lock.
- * But lockdep should know that ifb has different locks from dev.
- */
-static struct lock_class_key ifb_tx_queue_lock_key;
-static struct lock_class_key ifb_rx_queue_lock_key;
-
-static void set_tx_lockdep_key(struct net_device *dev,
-			       struct netdev_queue *txq,
-			       void *_unused)
-{
-	lockdep_set_class(&txq->lock, &ifb_tx_queue_lock_key);
-}
-
 static int __init ifb_init_one(int index)
 {
 	struct net_device *dev_ifb;
@@ -264,9 +247,6 @@ static int __init ifb_init_one(int index)
 	if (err < 0)
 		goto err;
 
-	netdev_for_each_tx_queue(dev_ifb, set_tx_lockdep_key, NULL);
-	lockdep_set_class(&dev_ifb->rx_queue.lock, &ifb_rx_queue_lock_key);
-
 	return 0;
 
 err:
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 3170bcef734b..9c5a68850114 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -443,7 +443,6 @@ enum netdev_queue_state_t
 };
 
 struct netdev_queue {
-	spinlock_t		lock;
 	struct net_device	*dev;
 	struct Qdisc		*qdisc;
 	unsigned long		state;
diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h
index 1eef8d0c9990..2902a42564f0 100644
--- a/include/net/sch_generic.h
+++ b/include/net/sch_generic.h
@@ -163,6 +163,11 @@ struct tcf_proto
 	struct tcf_proto_ops	*ops;
 };
 
+static inline spinlock_t *qdisc_lock(struct Qdisc *qdisc)
+{
+	return &qdisc->q.lock;
+}
+
 static inline struct Qdisc *qdisc_root(struct Qdisc *qdisc)
 {
 	return qdisc->dev_queue->qdisc;
@@ -172,7 +177,7 @@ static inline spinlock_t *qdisc_root_lock(struct Qdisc *qdisc)
 {
 	struct Qdisc *root = qdisc_root(qdisc);
 
-	return &root->dev_queue->lock;
+	return qdisc_lock(root);
 }
 
 static inline struct net_device *qdisc_dev(struct Qdisc *qdisc)
diff --git a/net/core/dev.c b/net/core/dev.c
index 6741e344ac59..32a13772c1cb 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -2080,10 +2080,12 @@ static int ing_filter(struct sk_buff *skb)
 
 	rxq = &dev->rx_queue;
 
-	spin_lock(&rxq->lock);
-	if ((q = rxq->qdisc) != NULL)
+	q = rxq->qdisc;
+	if (q) {
+		spin_lock(qdisc_lock(q));
 		result = q->enqueue(skb, q);
-	spin_unlock(&rxq->lock);
+		spin_unlock(qdisc_lock(q));
+	}
 
 	return result;
 }
@@ -4173,7 +4175,6 @@ static void netdev_init_one_queue(struct net_device *dev,
 				  struct netdev_queue *queue,
 				  void *_unused)
 {
-	spin_lock_init(&queue->lock);
 	queue->dev = dev;
 }
 
diff --git a/net/mac80211/wme.c b/net/mac80211/wme.c
index b21cfec4b6ce..6e8099e77043 100644
--- a/net/mac80211/wme.c
+++ b/net/mac80211/wme.c
@@ -237,12 +237,14 @@ void ieee80211_ht_agg_queue_remove(struct ieee80211_local *local,
 		ieee80211_requeue(local, agg_queue);
 	} else {
 		struct netdev_queue *txq;
+		spinlock_t *root_lock;
 
 		txq = netdev_get_tx_queue(local->mdev, agg_queue);
+		root_lock = qdisc_root_lock(txq->qdisc);
 
-		spin_lock_bh(&txq->lock);
+		spin_lock_bh(root_lock);
 		qdisc_reset(txq->qdisc);
-		spin_unlock_bh(&txq->lock);
+		spin_unlock_bh(root_lock);
 	}
 }
 
@@ -250,6 +252,7 @@ void ieee80211_requeue(struct ieee80211_local *local, int queue)
 {
 	struct netdev_queue *txq = netdev_get_tx_queue(local->mdev, queue);
 	struct sk_buff_head list;
+	spinlock_t *root_lock;
 	struct Qdisc *qdisc;
 	u32 len;
 
@@ -261,14 +264,15 @@ void ieee80211_requeue(struct ieee80211_local *local, int queue)
 
 	skb_queue_head_init(&list);
 
-	spin_lock(&txq->lock);
+	root_lock = qdisc_root_lock(qdisc);
+	spin_lock(root_lock);
 	for (len = qdisc->q.qlen; len > 0; len--) {
 		struct sk_buff *skb = qdisc->dequeue(qdisc);
 
 		if (skb)
 			__skb_queue_tail(&list, skb);
 	}
-	spin_unlock(&txq->lock);
+	spin_unlock(root_lock);
 
 	for (len = list.qlen; len > 0; len--) {
 		struct sk_buff *skb = __skb_dequeue(&list);
@@ -280,12 +284,13 @@ void ieee80211_requeue(struct ieee80211_local *local, int queue)
 
 		txq = netdev_get_tx_queue(local->mdev, new_queue);
 
-		spin_lock(&txq->lock);
 
 		qdisc = rcu_dereference(txq->qdisc);
-		qdisc->enqueue(skb, qdisc);
+		root_lock = qdisc_root_lock(qdisc);
 
-		spin_unlock(&txq->lock);
+		spin_lock(root_lock);
+		qdisc->enqueue(skb, qdisc);
+		spin_unlock(root_lock);
 	}
 
 out_unlock:
diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c
index 3d53e92ad9c8..8fc580b3e173 100644
--- a/net/sched/sch_generic.c
+++ b/net/sched/sch_generic.c
@@ -96,15 +96,15 @@ static inline int handle_dev_cpu_collision(struct sk_buff *skb,
 }
 
 /*
- * NOTE: Called under queue->lock with locally disabled BH.
+ * NOTE: Called under qdisc_lock(q) with locally disabled BH.
  *
  * __QDISC_STATE_RUNNING guarantees only one CPU can process
- * this qdisc at a time. queue->lock serializes queue accesses for
- * this queue AND txq->qdisc pointer itself.
+ * this qdisc at a time. qdisc_lock(q) serializes queue accesses for
+ * this queue.
  *
  *  netif_tx_lock serializes accesses to device driver.
  *
- *  queue->lock and netif_tx_lock are mutually exclusive,
+ *  qdisc_lock(q) and netif_tx_lock are mutually exclusive,
  *  if one is grabbed, another must be free.
  *
  * Note, that this procedure can be called by a watchdog timer
@@ -317,7 +317,6 @@ struct Qdisc_ops noop_qdisc_ops __read_mostly = {
 };
 
 static struct netdev_queue noop_netdev_queue = {
-	.lock		=	__SPIN_LOCK_UNLOCKED(noop_netdev_queue.lock),
 	.qdisc		=	&noop_qdisc,
 };
 
@@ -327,6 +326,7 @@ struct Qdisc noop_qdisc = {
 	.flags		=	TCQ_F_BUILTIN,
 	.ops		=	&noop_qdisc_ops,
 	.list		=	LIST_HEAD_INIT(noop_qdisc.list),
+	.q.lock		=	__SPIN_LOCK_UNLOCKED(noop_qdisc.q.lock),
 	.dev_queue	=	&noop_netdev_queue,
 };
 EXPORT_SYMBOL(noop_qdisc);
@@ -498,7 +498,7 @@ errout:
 }
 EXPORT_SYMBOL(qdisc_create_dflt);
 
-/* Under queue->lock and BH! */
+/* Under qdisc_root_lock(qdisc) and BH! */
 
 void qdisc_reset(struct Qdisc *qdisc)
 {
@@ -526,10 +526,12 @@ static void __qdisc_destroy(struct rcu_head *head)
 	module_put(ops->owner);
 	dev_put(qdisc_dev(qdisc));
 
+	kfree_skb(qdisc->gso_skb);
+
 	kfree((char *) qdisc - qdisc->padded);
 }
 
-/* Under queue->lock and BH! */
+/* Under qdisc_root_lock(qdisc) and BH! */
 
 void qdisc_destroy(struct Qdisc *qdisc)
 {
@@ -586,13 +588,12 @@ static void transition_one_qdisc(struct net_device *dev,
 				 struct netdev_queue *dev_queue,
 				 void *_need_watchdog)
 {
+	struct Qdisc *new_qdisc = dev_queue->qdisc_sleeping;
 	int *need_watchdog_p = _need_watchdog;
 
-	spin_lock_bh(&dev_queue->lock);
-	rcu_assign_pointer(dev_queue->qdisc, dev_queue->qdisc_sleeping);
-	if (dev_queue->qdisc != &noqueue_qdisc)
+	rcu_assign_pointer(dev_queue->qdisc, new_qdisc);
+	if (new_qdisc != &noqueue_qdisc)
 		*need_watchdog_p = 1;
-	spin_unlock_bh(&dev_queue->lock);
 }
 
 void dev_activate(struct net_device *dev)
@@ -629,19 +630,16 @@ static void dev_deactivate_queue(struct net_device *dev,
 	struct sk_buff *skb = NULL;
 	struct Qdisc *qdisc;
 
-	spin_lock_bh(&dev_queue->lock);
-
 	qdisc = dev_queue->qdisc;
 	if (qdisc) {
+		spin_lock_bh(qdisc_lock(qdisc));
+
 		dev_queue->qdisc = qdisc_default;
 		qdisc_reset(qdisc);
 
-		skb = qdisc->gso_skb;
-		qdisc->gso_skb = NULL;
+		spin_unlock_bh(qdisc_lock(qdisc));
 	}
 
-	spin_unlock_bh(&dev_queue->lock);
-
 	kfree_skb(skb);
 }
 
diff --git a/net/sched/sch_teql.c b/net/sched/sch_teql.c
index ade3372221c7..8b0ff345f9da 100644
--- a/net/sched/sch_teql.c
+++ b/net/sched/sch_teql.c
@@ -156,12 +156,15 @@ teql_destroy(struct Qdisc* sch)
 					master->slaves = NEXT_SLAVE(q);
 					if (q == master->slaves) {
 						struct netdev_queue *txq;
+						spinlock_t *root_lock;
 
 						txq = netdev_get_tx_queue(master->dev, 0);
 						master->slaves = NULL;
-						spin_lock_bh(&txq->lock);
+
+						root_lock = qdisc_root_lock(txq->qdisc);
+						spin_lock_bh(root_lock);
 						qdisc_reset(txq->qdisc);
-						spin_unlock_bh(&txq->lock);
+						spin_unlock_bh(root_lock);
 					}
 				}
 				skb_queue_purge(&dat->q);
-- 
cgit v1.2.3


From 93245dd6d356b864f6676396a9f3edecbd378ed0 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Thu, 17 Jul 2008 04:03:43 -0700
Subject: pkt_sched: Don't used locked skb_queue_purge() in
 __qdisc_reset_queue()

We have to have exclusive access to the given qdisc anyways, so
doing even more locking is superfluous.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/sch_generic.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h
index 2902a42564f0..0a158ff4de12 100644
--- a/include/net/sch_generic.h
+++ b/include/net/sch_generic.h
@@ -374,7 +374,7 @@ static inline void __qdisc_reset_queue(struct Qdisc *sch,
 	 * We do not know the backlog in bytes of this list, it
 	 * is up to the caller to correct it
 	 */
-	skb_queue_purge(list);
+	__skb_queue_purge(list);
 }
 
 static inline void qdisc_reset_queue(struct Qdisc *sch)
-- 
cgit v1.2.3


From 852566f53ccdc9d1c149ffa28daa9778e13fa3da Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Fri, 18 Jul 2008 04:01:24 -0700
Subject: mib: add netns/mib.h file

The only structure declared within is the netns_mib, which will
carry all our mibs within. I didn't put the mibs in the existing
netns_xxx structures to make it possible to mark this one as
properly aligned and get in a separate "read-mostly" cache-line.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/net_namespace.h | 2 ++
 include/net/netns/mib.h     | 9 +++++++++
 2 files changed, 11 insertions(+)
 create mode 100644 include/net/netns/mib.h

(limited to 'include')

diff --git a/include/net/net_namespace.h b/include/net/net_namespace.h
index f90443045c70..3855620b78a9 100644
--- a/include/net/net_namespace.h
+++ b/include/net/net_namespace.h
@@ -9,6 +9,7 @@
 #include <linux/list.h>
 
 #include <net/netns/core.h>
+#include <net/netns/mib.h>
 #include <net/netns/unix.h>
 #include <net/netns/packet.h>
 #include <net/netns/ipv4.h>
@@ -52,6 +53,7 @@ struct net {
 	struct sock 		*rtnl;			/* rtnetlink socket */
 
 	struct netns_core	core;
+	struct netns_mib	mib;
 	struct netns_packet	packet;
 	struct netns_unix	unx;
 	struct netns_ipv4	ipv4;
diff --git a/include/net/netns/mib.h b/include/net/netns/mib.h
new file mode 100644
index 000000000000..9f4b31ed18c3
--- /dev/null
+++ b/include/net/netns/mib.h
@@ -0,0 +1,9 @@
+#ifndef __NETNS_MIB_H__
+#define __NETNS_MIB_H__
+
+#include <net/snmp.h>
+
+struct netns_mib {
+};
+
+#endif
-- 
cgit v1.2.3


From 57ef42d59d1c1d79be59fc3c6380ae14234e38c3 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Fri, 18 Jul 2008 04:02:08 -0700
Subject: mib: put tcp statistics on struct net

Proc temporary uses stats from init_net.

BTW, TCP_XXX_STATS are beautiful (w/o do { } while (0) facing) again :)

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/netns/mib.h |  1 +
 include/net/tcp.h       |  9 ++++-----
 net/ipv4/af_inet.c      | 16 +++++++++-------
 net/ipv4/proc.c         |  4 ++--
 net/ipv4/tcp.c          |  3 ---
 5 files changed, 16 insertions(+), 17 deletions(-)

(limited to 'include')

diff --git a/include/net/netns/mib.h b/include/net/netns/mib.h
index 9f4b31ed18c3..8f96079bf0e3 100644
--- a/include/net/netns/mib.h
+++ b/include/net/netns/mib.h
@@ -4,6 +4,7 @@
 #include <net/snmp.h>
 
 struct netns_mib {
+	DEFINE_SNMP_STAT(struct tcp_mib, tcp_statistics);
 };
 
 #endif
diff --git a/include/net/tcp.h b/include/net/tcp.h
index 60e5be8b925b..92d7b551dc55 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -265,11 +265,10 @@ static inline int tcp_too_many_orphans(struct sock *sk, int num)
 
 extern struct proto tcp_prot;
 
-DECLARE_SNMP_STAT(struct tcp_mib, tcp_statistics);
-#define TCP_INC_STATS(net, field)	do { (void)net; SNMP_INC_STATS(tcp_statistics, field); } while (0)
-#define TCP_INC_STATS_BH(net, field)	do { (void)net; SNMP_INC_STATS_BH(tcp_statistics, field); } while (0)
-#define TCP_DEC_STATS(net, field)	do { (void)net; SNMP_DEC_STATS(tcp_statistics, field); } while (0)
-#define TCP_ADD_STATS_USER(net, field, val) do { (void)net; SNMP_ADD_STATS_USER(tcp_statistics, field, val); } while (0)
+#define TCP_INC_STATS(net, field)	SNMP_INC_STATS((net)->mib.tcp_statistics, field)
+#define TCP_INC_STATS_BH(net, field)	SNMP_INC_STATS_BH((net)->mib.tcp_statistics, field)
+#define TCP_DEC_STATS(net, field)	SNMP_DEC_STATS((net)->mib.tcp_statistics, field)
+#define TCP_ADD_STATS_USER(net, field, val) SNMP_ADD_STATS_USER((net)->mib.tcp_statistics, field, val)
 
 extern void			tcp_v4_err(struct sk_buff *skb, u32);
 
diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c
index b4b77aa07953..c1a3e986f8b4 100644
--- a/net/ipv4/af_inet.c
+++ b/net/ipv4/af_inet.c
@@ -1342,11 +1342,20 @@ static struct net_protocol icmp_protocol = {
 
 static __net_init int ipv4_mib_init_net(struct net *net)
 {
+	if (snmp_mib_init((void **)net->mib.tcp_statistics,
+			  sizeof(struct tcp_mib)) < 0)
+		goto err_tcp_mib;
+
+	tcp_mib_init(net);
 	return 0;
+
+err_tcp_mib:
+	return -ENOMEM;
 }
 
 static __net_exit void ipv4_mib_exit_net(struct net *net)
 {
+	snmp_mib_free((void **)net->mib.tcp_statistics);
 }
 
 static __net_initdata struct pernet_operations ipv4_mib_ops = {
@@ -1368,9 +1377,6 @@ static int __init init_ipv4_mibs(void)
 	if (snmp_mib_init((void **)icmpmsg_statistics,
 			  sizeof(struct icmpmsg_mib)) < 0)
 		goto err_icmpmsg_mib;
-	if (snmp_mib_init((void **)tcp_statistics,
-			  sizeof(struct tcp_mib)) < 0)
-		goto err_tcp_mib;
 	if (snmp_mib_init((void **)udp_statistics,
 			  sizeof(struct udp_mib)) < 0)
 		goto err_udp_mib;
@@ -1378,8 +1384,6 @@ static int __init init_ipv4_mibs(void)
 			  sizeof(struct udp_mib)) < 0)
 		goto err_udplite_mib;
 
-	tcp_mib_init(&init_net);
-
 	if (register_pernet_subsys(&ipv4_mib_ops))
 		goto err_net;
 
@@ -1390,8 +1394,6 @@ err_net:
 err_udplite_mib:
 	snmp_mib_free((void **)udp_statistics);
 err_udp_mib:
-	snmp_mib_free((void **)tcp_statistics);
-err_tcp_mib:
 	snmp_mib_free((void **)icmpmsg_statistics);
 err_icmpmsg_mib:
 	snmp_mib_free((void **)icmp_statistics);
diff --git a/net/ipv4/proc.c b/net/ipv4/proc.c
index eb5cee279c5f..19e1d8e257da 100644
--- a/net/ipv4/proc.c
+++ b/net/ipv4/proc.c
@@ -359,11 +359,11 @@ static int snmp_seq_show(struct seq_file *seq, void *v)
 		/* MaxConn field is signed, RFC 2012 */
 		if (snmp4_tcp_list[i].entry == TCP_MIB_MAXCONN)
 			seq_printf(seq, " %ld",
-				   snmp_fold_field((void **)tcp_statistics,
+				   snmp_fold_field((void **)init_net.mib.tcp_statistics,
 						   snmp4_tcp_list[i].entry));
 		else
 			seq_printf(seq, " %lu",
-				   snmp_fold_field((void **)tcp_statistics,
+				   snmp_fold_field((void **)init_net.mib.tcp_statistics,
 						   snmp4_tcp_list[i].entry));
 	}
 
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index eec8cf7c0247..827e6132af5f 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -277,8 +277,6 @@
 
 int sysctl_tcp_fin_timeout __read_mostly = TCP_FIN_TIMEOUT;
 
-DEFINE_SNMP_STAT(struct tcp_mib, tcp_statistics) __read_mostly;
-
 atomic_t tcp_orphan_count = ATOMIC_INIT(0);
 
 EXPORT_SYMBOL_GPL(tcp_orphan_count);
@@ -2802,4 +2800,3 @@ EXPORT_SYMBOL(tcp_splice_read);
 EXPORT_SYMBOL(tcp_sendpage);
 EXPORT_SYMBOL(tcp_setsockopt);
 EXPORT_SYMBOL(tcp_shutdown);
-EXPORT_SYMBOL(tcp_statistics);
-- 
cgit v1.2.3


From a20f5799ca7ceb24d63c74b6fdad4b0c0ee91f4f Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Fri, 18 Jul 2008 04:02:42 -0700
Subject: mib: put ip statistics on struct net

Similar to tcp one.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip.h        |  7 +++----
 include/net/netns/mib.h |  1 +
 net/ipv4/af_inet.c      | 11 ++++++-----
 net/ipv4/ip_input.c     |  8 --------
 net/ipv4/proc.c         |  4 ++--
 5 files changed, 12 insertions(+), 19 deletions(-)

(limited to 'include')

diff --git a/include/net/ip.h b/include/net/ip.h
index 02924fb4bdbf..ff2535ada02e 100644
--- a/include/net/ip.h
+++ b/include/net/ip.h
@@ -156,10 +156,9 @@ struct ipv4_config
 };
 
 extern struct ipv4_config ipv4_config;
-DECLARE_SNMP_STAT(struct ipstats_mib, ip_statistics);
-#define IP_INC_STATS(net, field)	do { (void)net; SNMP_INC_STATS(ip_statistics, field); } while (0)
-#define IP_INC_STATS_BH(net, field)	do { (void)net; SNMP_INC_STATS_BH(ip_statistics, field); } while (0)
-#define IP_ADD_STATS_BH(net, field, val) SNMP_ADD_STATS_BH(ip_statistics, field, val)
+#define IP_INC_STATS(net, field)	SNMP_INC_STATS((net)->mib.ip_statistics, field)
+#define IP_INC_STATS_BH(net, field)	SNMP_INC_STATS_BH((net)->mib.ip_statistics, field)
+#define IP_ADD_STATS_BH(net, field, val) SNMP_ADD_STATS_BH((net)->mib.ip_statistics, field, val)
 DECLARE_SNMP_STAT(struct linux_mib, net_statistics);
 #define NET_INC_STATS(net, field)	do { (void)net; SNMP_INC_STATS(net_statistics, field); } while (0)
 #define NET_INC_STATS_BH(net, field)	do { (void)net; SNMP_INC_STATS_BH(net_statistics, field); } while (0)
diff --git a/include/net/netns/mib.h b/include/net/netns/mib.h
index 8f96079bf0e3..1094ebbf0b42 100644
--- a/include/net/netns/mib.h
+++ b/include/net/netns/mib.h
@@ -5,6 +5,7 @@
 
 struct netns_mib {
 	DEFINE_SNMP_STAT(struct tcp_mib, tcp_statistics);
+	DEFINE_SNMP_STAT(struct ipstats_mib, ip_statistics);
 };
 
 #endif
diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c
index c1a3e986f8b4..3090a9307c46 100644
--- a/net/ipv4/af_inet.c
+++ b/net/ipv4/af_inet.c
@@ -1345,16 +1345,22 @@ static __net_init int ipv4_mib_init_net(struct net *net)
 	if (snmp_mib_init((void **)net->mib.tcp_statistics,
 			  sizeof(struct tcp_mib)) < 0)
 		goto err_tcp_mib;
+	if (snmp_mib_init((void **)net->mib.ip_statistics,
+			  sizeof(struct ipstats_mib)) < 0)
+		goto err_ip_mib;
 
 	tcp_mib_init(net);
 	return 0;
 
+err_ip_mib:
+	snmp_mib_free((void **)net->mib.tcp_statistics);
 err_tcp_mib:
 	return -ENOMEM;
 }
 
 static __net_exit void ipv4_mib_exit_net(struct net *net)
 {
+	snmp_mib_free((void **)net->mib.ip_statistics);
 	snmp_mib_free((void **)net->mib.tcp_statistics);
 }
 
@@ -1368,9 +1374,6 @@ static int __init init_ipv4_mibs(void)
 	if (snmp_mib_init((void **)net_statistics,
 			  sizeof(struct linux_mib)) < 0)
 		goto err_net_mib;
-	if (snmp_mib_init((void **)ip_statistics,
-			  sizeof(struct ipstats_mib)) < 0)
-		goto err_ip_mib;
 	if (snmp_mib_init((void **)icmp_statistics,
 			  sizeof(struct icmp_mib)) < 0)
 		goto err_icmp_mib;
@@ -1398,8 +1401,6 @@ err_udp_mib:
 err_icmpmsg_mib:
 	snmp_mib_free((void **)icmp_statistics);
 err_icmp_mib:
-	snmp_mib_free((void **)ip_statistics);
-err_ip_mib:
 	snmp_mib_free((void **)net_statistics);
 err_net_mib:
 	return -ENOMEM;
diff --git a/net/ipv4/ip_input.c b/net/ipv4/ip_input.c
index 043f640df4b7..e0bed56c51f1 100644
--- a/net/ipv4/ip_input.c
+++ b/net/ipv4/ip_input.c
@@ -144,12 +144,6 @@
 #include <linux/mroute.h>
 #include <linux/netlink.h>
 
-/*
- *	SNMP management statistics
- */
-
-DEFINE_SNMP_STAT(struct ipstats_mib, ip_statistics) __read_mostly;
-
 /*
  *	Process Router Attention IP option
  */
@@ -447,5 +441,3 @@ drop:
 out:
 	return NET_RX_DROP;
 }
-
-EXPORT_SYMBOL(ip_statistics);
diff --git a/net/ipv4/proc.c b/net/ipv4/proc.c
index 19e1d8e257da..2698bb2ce98f 100644
--- a/net/ipv4/proc.c
+++ b/net/ipv4/proc.c
@@ -344,7 +344,7 @@ static int snmp_seq_show(struct seq_file *seq, void *v)
 
 	for (i = 0; snmp4_ipstats_list[i].name != NULL; i++)
 		seq_printf(seq, " %lu",
-			   snmp_fold_field((void **)ip_statistics,
+			   snmp_fold_field((void **)init_net.mib.ip_statistics,
 					   snmp4_ipstats_list[i].entry));
 
 	icmp_put(seq);	/* RFC 2011 compatibility */
@@ -431,7 +431,7 @@ static int netstat_seq_show(struct seq_file *seq, void *v)
 	seq_puts(seq, "\nIpExt:");
 	for (i = 0; snmp4_ipextstats_list[i].name != NULL; i++)
 		seq_printf(seq, " %lu",
-			   snmp_fold_field((void **)ip_statistics,
+			   snmp_fold_field((void **)init_net.mib.ip_statistics,
 					   snmp4_ipextstats_list[i].entry));
 
 	seq_putc(seq, '\n');
-- 
cgit v1.2.3


From 61a7e26028b94805fd686a6dc9dbd9941f8f19b0 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Fri, 18 Jul 2008 04:03:08 -0700
Subject: mib: put net statistics on struct net

Similar to ip and tcp ones :)

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip.h        | 11 +++++------
 include/net/netns/mib.h |  1 +
 net/ipv4/af_inet.c      | 14 ++++++--------
 net/ipv4/proc.c         |  2 +-
 4 files changed, 13 insertions(+), 15 deletions(-)

(limited to 'include')

diff --git a/include/net/ip.h b/include/net/ip.h
index ff2535ada02e..b5862b975207 100644
--- a/include/net/ip.h
+++ b/include/net/ip.h
@@ -159,12 +159,11 @@ extern struct ipv4_config ipv4_config;
 #define IP_INC_STATS(net, field)	SNMP_INC_STATS((net)->mib.ip_statistics, field)
 #define IP_INC_STATS_BH(net, field)	SNMP_INC_STATS_BH((net)->mib.ip_statistics, field)
 #define IP_ADD_STATS_BH(net, field, val) SNMP_ADD_STATS_BH((net)->mib.ip_statistics, field, val)
-DECLARE_SNMP_STAT(struct linux_mib, net_statistics);
-#define NET_INC_STATS(net, field)	do { (void)net; SNMP_INC_STATS(net_statistics, field); } while (0)
-#define NET_INC_STATS_BH(net, field)	do { (void)net; SNMP_INC_STATS_BH(net_statistics, field); } while (0)
-#define NET_INC_STATS_USER(net, field) 	do { (void)net; SNMP_INC_STATS_USER(net_statistics, field); } while (0)
-#define NET_ADD_STATS_BH(net, field, adnd) do { (void)net; SNMP_ADD_STATS_BH(net_statistics, field, adnd); } while (0)
-#define NET_ADD_STATS_USER(net, field, adnd) do { (void)net; SNMP_ADD_STATS_USER(net_statistics, field, adnd); } while (0)
+#define NET_INC_STATS(net, field)	SNMP_INC_STATS((net)->mib.net_statistics, field)
+#define NET_INC_STATS_BH(net, field)	SNMP_INC_STATS_BH((net)->mib.net_statistics, field)
+#define NET_INC_STATS_USER(net, field) 	SNMP_INC_STATS_USER((net)->mib.net_statistics, field)
+#define NET_ADD_STATS_BH(net, field, adnd) SNMP_ADD_STATS_BH((net)->mib.net_statistics, field, adnd)
+#define NET_ADD_STATS_USER(net, field, adnd) SNMP_ADD_STATS_USER((net)->mib.net_statistics, field, adnd)
 
 extern unsigned long snmp_fold_field(void *mib[], int offt);
 extern int snmp_mib_init(void *ptr[2], size_t mibsize);
diff --git a/include/net/netns/mib.h b/include/net/netns/mib.h
index 1094ebbf0b42..b5b1115adfa8 100644
--- a/include/net/netns/mib.h
+++ b/include/net/netns/mib.h
@@ -6,6 +6,7 @@
 struct netns_mib {
 	DEFINE_SNMP_STAT(struct tcp_mib, tcp_statistics);
 	DEFINE_SNMP_STAT(struct ipstats_mib, ip_statistics);
+	DEFINE_SNMP_STAT(struct linux_mib, net_statistics);
 };
 
 #endif
diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c
index 3090a9307c46..776ed3199b52 100644
--- a/net/ipv4/af_inet.c
+++ b/net/ipv4/af_inet.c
@@ -115,8 +115,6 @@
 #include <linux/mroute.h>
 #endif
 
-DEFINE_SNMP_STAT(struct linux_mib, net_statistics) __read_mostly;
-
 extern void ip_mc_drop_socket(struct sock *sk);
 
 /* The inetsw table contains everything that inet_create needs to
@@ -1348,10 +1346,15 @@ static __net_init int ipv4_mib_init_net(struct net *net)
 	if (snmp_mib_init((void **)net->mib.ip_statistics,
 			  sizeof(struct ipstats_mib)) < 0)
 		goto err_ip_mib;
+	if (snmp_mib_init((void **)net->mib.net_statistics,
+			  sizeof(struct linux_mib)) < 0)
+		goto err_net_mib;
 
 	tcp_mib_init(net);
 	return 0;
 
+err_net_mib:
+	snmp_mib_free((void **)net->mib.ip_statistics);
 err_ip_mib:
 	snmp_mib_free((void **)net->mib.tcp_statistics);
 err_tcp_mib:
@@ -1360,6 +1363,7 @@ err_tcp_mib:
 
 static __net_exit void ipv4_mib_exit_net(struct net *net)
 {
+	snmp_mib_free((void **)net->mib.net_statistics);
 	snmp_mib_free((void **)net->mib.ip_statistics);
 	snmp_mib_free((void **)net->mib.tcp_statistics);
 }
@@ -1371,9 +1375,6 @@ static __net_initdata struct pernet_operations ipv4_mib_ops = {
 
 static int __init init_ipv4_mibs(void)
 {
-	if (snmp_mib_init((void **)net_statistics,
-			  sizeof(struct linux_mib)) < 0)
-		goto err_net_mib;
 	if (snmp_mib_init((void **)icmp_statistics,
 			  sizeof(struct icmp_mib)) < 0)
 		goto err_icmp_mib;
@@ -1401,8 +1402,6 @@ err_udp_mib:
 err_icmpmsg_mib:
 	snmp_mib_free((void **)icmp_statistics);
 err_icmp_mib:
-	snmp_mib_free((void **)net_statistics);
-err_net_mib:
 	return -ENOMEM;
 }
 
@@ -1582,5 +1581,4 @@ EXPORT_SYMBOL(inet_sock_destruct);
 EXPORT_SYMBOL(inet_stream_connect);
 EXPORT_SYMBOL(inet_stream_ops);
 EXPORT_SYMBOL(inet_unregister_protosw);
-EXPORT_SYMBOL(net_statistics);
 EXPORT_SYMBOL(sysctl_ip_nonlocal_bind);
diff --git a/net/ipv4/proc.c b/net/ipv4/proc.c
index 2698bb2ce98f..ef38b1dccfda 100644
--- a/net/ipv4/proc.c
+++ b/net/ipv4/proc.c
@@ -421,7 +421,7 @@ static int netstat_seq_show(struct seq_file *seq, void *v)
 	seq_puts(seq, "\nTcpExt:");
 	for (i = 0; snmp4_net_list[i].name != NULL; i++)
 		seq_printf(seq, " %lu",
-			   snmp_fold_field((void **)net_statistics,
+			   snmp_fold_field((void **)init_net.mib.net_statistics,
 					   snmp4_net_list[i].entry));
 
 	seq_puts(seq, "\nIpExt:");
-- 
cgit v1.2.3


From 2f275f91a438abd8eec5321798d66a4ffe6869fa Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Fri, 18 Jul 2008 04:03:27 -0700
Subject: mib: put udp statistics on struct net

Similar to... ouch, I repeat myself.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/netns/mib.h |  1 +
 include/net/udp.h       |  9 ++++-----
 net/ipv4/af_inet.c      | 11 ++++++-----
 net/ipv4/proc.c         |  2 +-
 net/ipv4/udp.c          |  3 ---
 5 files changed, 12 insertions(+), 14 deletions(-)

(limited to 'include')

diff --git a/include/net/netns/mib.h b/include/net/netns/mib.h
index b5b1115adfa8..738c87ce183d 100644
--- a/include/net/netns/mib.h
+++ b/include/net/netns/mib.h
@@ -7,6 +7,7 @@ struct netns_mib {
 	DEFINE_SNMP_STAT(struct tcp_mib, tcp_statistics);
 	DEFINE_SNMP_STAT(struct ipstats_mib, ip_statistics);
 	DEFINE_SNMP_STAT(struct linux_mib, net_statistics);
+	DEFINE_SNMP_STAT(struct udp_mib, udp_statistics);
 };
 
 #endif
diff --git a/include/net/udp.h b/include/net/udp.h
index 3e551592aa76..ba5544152708 100644
--- a/include/net/udp.h
+++ b/include/net/udp.h
@@ -148,7 +148,6 @@ extern int 	udp_lib_setsockopt(struct sock *sk, int level, int optname,
 				   char __user *optval, int optlen,
 				   int (*push_pending_frames)(struct sock *));
 
-DECLARE_SNMP_STAT(struct udp_mib, udp_statistics);
 DECLARE_SNMP_STAT(struct udp_mib, udp_stats_in6);
 
 /* UDP-Lite does not have a standardized MIB yet, so we inherit from UDP */
@@ -158,12 +157,12 @@ DECLARE_SNMP_STAT(struct udp_mib, udplite_stats_in6);
 /*
  * 	SNMP statistics for UDP and UDP-Lite
  */
-#define UDP_INC_STATS_USER(net, field, is_udplite)	      do { (void)net; \
+#define UDP_INC_STATS_USER(net, field, is_udplite)	      do { \
 	if (is_udplite) SNMP_INC_STATS_USER(udplite_statistics, field);       \
-	else		SNMP_INC_STATS_USER(udp_statistics, field);  }  while(0)
-#define UDP_INC_STATS_BH(net, field, is_udplite) 	      do { (void)net; \
+	else		SNMP_INC_STATS_USER((net)->mib.udp_statistics, field);  }  while(0)
+#define UDP_INC_STATS_BH(net, field, is_udplite) 	      do { \
 	if (is_udplite) SNMP_INC_STATS_BH(udplite_statistics, field);         \
-	else		SNMP_INC_STATS_BH(udp_statistics, field);    }  while(0)
+	else		SNMP_INC_STATS_BH((net)->mib.udp_statistics, field);    }  while(0)
 
 #define UDP6_INC_STATS_BH(net, field, is_udplite) 	    do { (void)net;  \
 	if (is_udplite) SNMP_INC_STATS_BH(udplite_stats_in6, field);         \
diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c
index 776ed3199b52..1f418164ebf2 100644
--- a/net/ipv4/af_inet.c
+++ b/net/ipv4/af_inet.c
@@ -1349,10 +1349,15 @@ static __net_init int ipv4_mib_init_net(struct net *net)
 	if (snmp_mib_init((void **)net->mib.net_statistics,
 			  sizeof(struct linux_mib)) < 0)
 		goto err_net_mib;
+	if (snmp_mib_init((void **)net->mib.udp_statistics,
+			  sizeof(struct udp_mib)) < 0)
+		goto err_udp_mib;
 
 	tcp_mib_init(net);
 	return 0;
 
+err_udp_mib:
+	snmp_mib_free((void **)net->mib.net_statistics);
 err_net_mib:
 	snmp_mib_free((void **)net->mib.ip_statistics);
 err_ip_mib:
@@ -1363,6 +1368,7 @@ err_tcp_mib:
 
 static __net_exit void ipv4_mib_exit_net(struct net *net)
 {
+	snmp_mib_free((void **)net->mib.udp_statistics);
 	snmp_mib_free((void **)net->mib.net_statistics);
 	snmp_mib_free((void **)net->mib.ip_statistics);
 	snmp_mib_free((void **)net->mib.tcp_statistics);
@@ -1381,9 +1387,6 @@ static int __init init_ipv4_mibs(void)
 	if (snmp_mib_init((void **)icmpmsg_statistics,
 			  sizeof(struct icmpmsg_mib)) < 0)
 		goto err_icmpmsg_mib;
-	if (snmp_mib_init((void **)udp_statistics,
-			  sizeof(struct udp_mib)) < 0)
-		goto err_udp_mib;
 	if (snmp_mib_init((void **)udplite_statistics,
 			  sizeof(struct udp_mib)) < 0)
 		goto err_udplite_mib;
@@ -1396,8 +1399,6 @@ static int __init init_ipv4_mibs(void)
 err_net:
 	snmp_mib_free((void **)udplite_statistics);
 err_udplite_mib:
-	snmp_mib_free((void **)udp_statistics);
-err_udp_mib:
 	snmp_mib_free((void **)icmpmsg_statistics);
 err_icmpmsg_mib:
 	snmp_mib_free((void **)icmp_statistics);
diff --git a/net/ipv4/proc.c b/net/ipv4/proc.c
index ef38b1dccfda..869085c8e43f 100644
--- a/net/ipv4/proc.c
+++ b/net/ipv4/proc.c
@@ -374,7 +374,7 @@ static int snmp_seq_show(struct seq_file *seq, void *v)
 	seq_puts(seq, "\nUdp:");
 	for (i = 0; snmp4_udp_list[i].name != NULL; i++)
 		seq_printf(seq, " %lu",
-			   snmp_fold_field((void **)udp_statistics,
+			   snmp_fold_field((void **)init_net.mib.udp_statistics,
 					   snmp4_udp_list[i].entry));
 
 	/* the UDP and UDP-Lite MIBs are the same */
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index 1560d11ba6ef..a751770947a3 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -108,9 +108,6 @@
  *	Snmp MIB for the UDP layer
  */
 
-DEFINE_SNMP_STAT(struct udp_mib, udp_statistics) __read_mostly;
-EXPORT_SYMBOL(udp_statistics);
-
 DEFINE_SNMP_STAT(struct udp_mib, udp_stats_in6) __read_mostly;
 EXPORT_SYMBOL(udp_stats_in6);
 
-- 
cgit v1.2.3


From 386019d3514b3ed9de8d0b05b67e638a7048375b Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Fri, 18 Jul 2008 04:03:45 -0700
Subject: mib: put udplite statistics on struct net

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/netns/mib.h |  1 +
 include/net/udp.h       |  5 ++---
 net/ipv4/af_inet.c      | 11 ++++++-----
 net/ipv4/proc.c         |  2 +-
 net/ipv4/udplite.c      |  1 -
 5 files changed, 10 insertions(+), 10 deletions(-)

(limited to 'include')

diff --git a/include/net/netns/mib.h b/include/net/netns/mib.h
index 738c87ce183d..23e699fdb5cc 100644
--- a/include/net/netns/mib.h
+++ b/include/net/netns/mib.h
@@ -8,6 +8,7 @@ struct netns_mib {
 	DEFINE_SNMP_STAT(struct ipstats_mib, ip_statistics);
 	DEFINE_SNMP_STAT(struct linux_mib, net_statistics);
 	DEFINE_SNMP_STAT(struct udp_mib, udp_statistics);
+	DEFINE_SNMP_STAT(struct udp_mib, udplite_statistics);
 };
 
 #endif
diff --git a/include/net/udp.h b/include/net/udp.h
index ba5544152708..addcdc67234c 100644
--- a/include/net/udp.h
+++ b/include/net/udp.h
@@ -151,17 +151,16 @@ extern int 	udp_lib_setsockopt(struct sock *sk, int level, int optname,
 DECLARE_SNMP_STAT(struct udp_mib, udp_stats_in6);
 
 /* UDP-Lite does not have a standardized MIB yet, so we inherit from UDP */
-DECLARE_SNMP_STAT(struct udp_mib, udplite_statistics);
 DECLARE_SNMP_STAT(struct udp_mib, udplite_stats_in6);
 
 /*
  * 	SNMP statistics for UDP and UDP-Lite
  */
 #define UDP_INC_STATS_USER(net, field, is_udplite)	      do { \
-	if (is_udplite) SNMP_INC_STATS_USER(udplite_statistics, field);       \
+	if (is_udplite) SNMP_INC_STATS_USER((net)->mib.udplite_statistics, field);       \
 	else		SNMP_INC_STATS_USER((net)->mib.udp_statistics, field);  }  while(0)
 #define UDP_INC_STATS_BH(net, field, is_udplite) 	      do { \
-	if (is_udplite) SNMP_INC_STATS_BH(udplite_statistics, field);         \
+	if (is_udplite) SNMP_INC_STATS_BH((net)->mib.udplite_statistics, field);         \
 	else		SNMP_INC_STATS_BH((net)->mib.udp_statistics, field);    }  while(0)
 
 #define UDP6_INC_STATS_BH(net, field, is_udplite) 	    do { (void)net;  \
diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c
index 1f418164ebf2..bf1f200c6575 100644
--- a/net/ipv4/af_inet.c
+++ b/net/ipv4/af_inet.c
@@ -1352,10 +1352,15 @@ static __net_init int ipv4_mib_init_net(struct net *net)
 	if (snmp_mib_init((void **)net->mib.udp_statistics,
 			  sizeof(struct udp_mib)) < 0)
 		goto err_udp_mib;
+	if (snmp_mib_init((void **)net->mib.udplite_statistics,
+			  sizeof(struct udp_mib)) < 0)
+		goto err_udplite_mib;
 
 	tcp_mib_init(net);
 	return 0;
 
+err_udplite_mib:
+	snmp_mib_free((void **)net->mib.udp_statistics);
 err_udp_mib:
 	snmp_mib_free((void **)net->mib.net_statistics);
 err_net_mib:
@@ -1368,6 +1373,7 @@ err_tcp_mib:
 
 static __net_exit void ipv4_mib_exit_net(struct net *net)
 {
+	snmp_mib_free((void **)net->mib.udplite_statistics);
 	snmp_mib_free((void **)net->mib.udp_statistics);
 	snmp_mib_free((void **)net->mib.net_statistics);
 	snmp_mib_free((void **)net->mib.ip_statistics);
@@ -1387,9 +1393,6 @@ static int __init init_ipv4_mibs(void)
 	if (snmp_mib_init((void **)icmpmsg_statistics,
 			  sizeof(struct icmpmsg_mib)) < 0)
 		goto err_icmpmsg_mib;
-	if (snmp_mib_init((void **)udplite_statistics,
-			  sizeof(struct udp_mib)) < 0)
-		goto err_udplite_mib;
 
 	if (register_pernet_subsys(&ipv4_mib_ops))
 		goto err_net;
@@ -1397,8 +1400,6 @@ static int __init init_ipv4_mibs(void)
 	return 0;
 
 err_net:
-	snmp_mib_free((void **)udplite_statistics);
-err_udplite_mib:
 	snmp_mib_free((void **)icmpmsg_statistics);
 err_icmpmsg_mib:
 	snmp_mib_free((void **)icmp_statistics);
diff --git a/net/ipv4/proc.c b/net/ipv4/proc.c
index 869085c8e43f..765418334b3c 100644
--- a/net/ipv4/proc.c
+++ b/net/ipv4/proc.c
@@ -385,7 +385,7 @@ static int snmp_seq_show(struct seq_file *seq, void *v)
 	seq_puts(seq, "\nUdpLite:");
 	for (i = 0; snmp4_udp_list[i].name != NULL; i++)
 		seq_printf(seq, " %lu",
-			   snmp_fold_field((void **)udplite_statistics,
+			   snmp_fold_field((void **)init_net.mib.udplite_statistics,
 					   snmp4_udp_list[i].entry));
 
 	seq_putc(seq, '\n');
diff --git a/net/ipv4/udplite.c b/net/ipv4/udplite.c
index 4ad16b6d5138..3c807964da96 100644
--- a/net/ipv4/udplite.c
+++ b/net/ipv4/udplite.c
@@ -11,7 +11,6 @@
  *		2 of the License, or (at your option) any later version.
  */
 #include "udp_impl.h"
-DEFINE_SNMP_STAT(struct udp_mib, udplite_statistics)	__read_mostly;
 
 struct hlist_head 	udplite_hash[UDP_HTABLE_SIZE];
 
-- 
cgit v1.2.3


From b60538a0d737609213e4b758881913498d3ff0b4 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Fri, 18 Jul 2008 04:04:02 -0700
Subject: mib: put icmp statistics on struct net

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/icmp.h      |  5 ++---
 include/net/netns/mib.h |  1 +
 net/ipv4/af_inet.c      | 11 ++++++-----
 net/ipv4/icmp.c         |  2 --
 net/ipv4/proc.c         |  8 ++++----
 5 files changed, 13 insertions(+), 14 deletions(-)

(limited to 'include')

diff --git a/include/net/icmp.h b/include/net/icmp.h
index 03b9972b8754..e14f2c03f42e 100644
--- a/include/net/icmp.h
+++ b/include/net/icmp.h
@@ -29,10 +29,9 @@ struct icmp_err {
 };
 
 extern struct icmp_err icmp_err_convert[];
-DECLARE_SNMP_STAT(struct icmp_mib, icmp_statistics);
 DECLARE_SNMP_STAT(struct icmpmsg_mib, icmpmsg_statistics);
-#define ICMP_INC_STATS(net, field)	SNMP_INC_STATS(icmp_statistics, field)
-#define ICMP_INC_STATS_BH(net, field)	SNMP_INC_STATS_BH(icmp_statistics, field)
+#define ICMP_INC_STATS(net, field)	SNMP_INC_STATS((net)->mib.icmp_statistics, field)
+#define ICMP_INC_STATS_BH(net, field)	SNMP_INC_STATS_BH((net)->mib.icmp_statistics, field)
 #define ICMPMSGOUT_INC_STATS(net, field)	SNMP_INC_STATS(icmpmsg_statistics, field+256)
 #define ICMPMSGIN_INC_STATS_BH(net, field)	SNMP_INC_STATS_BH(icmpmsg_statistics, field)
 
diff --git a/include/net/netns/mib.h b/include/net/netns/mib.h
index 23e699fdb5cc..03f5abdf58e5 100644
--- a/include/net/netns/mib.h
+++ b/include/net/netns/mib.h
@@ -9,6 +9,7 @@ struct netns_mib {
 	DEFINE_SNMP_STAT(struct linux_mib, net_statistics);
 	DEFINE_SNMP_STAT(struct udp_mib, udp_statistics);
 	DEFINE_SNMP_STAT(struct udp_mib, udplite_statistics);
+	DEFINE_SNMP_STAT(struct icmp_mib, icmp_statistics);
 };
 
 #endif
diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c
index bf1f200c6575..5c72bb3adadd 100644
--- a/net/ipv4/af_inet.c
+++ b/net/ipv4/af_inet.c
@@ -1355,10 +1355,15 @@ static __net_init int ipv4_mib_init_net(struct net *net)
 	if (snmp_mib_init((void **)net->mib.udplite_statistics,
 			  sizeof(struct udp_mib)) < 0)
 		goto err_udplite_mib;
+	if (snmp_mib_init((void **)net->mib.icmp_statistics,
+			  sizeof(struct icmp_mib)) < 0)
+		goto err_icmp_mib;
 
 	tcp_mib_init(net);
 	return 0;
 
+err_icmp_mib:
+	snmp_mib_free((void **)net->mib.udplite_statistics);
 err_udplite_mib:
 	snmp_mib_free((void **)net->mib.udp_statistics);
 err_udp_mib:
@@ -1373,6 +1378,7 @@ err_tcp_mib:
 
 static __net_exit void ipv4_mib_exit_net(struct net *net)
 {
+	snmp_mib_free((void **)net->mib.icmp_statistics);
 	snmp_mib_free((void **)net->mib.udplite_statistics);
 	snmp_mib_free((void **)net->mib.udp_statistics);
 	snmp_mib_free((void **)net->mib.net_statistics);
@@ -1387,9 +1393,6 @@ static __net_initdata struct pernet_operations ipv4_mib_ops = {
 
 static int __init init_ipv4_mibs(void)
 {
-	if (snmp_mib_init((void **)icmp_statistics,
-			  sizeof(struct icmp_mib)) < 0)
-		goto err_icmp_mib;
 	if (snmp_mib_init((void **)icmpmsg_statistics,
 			  sizeof(struct icmpmsg_mib)) < 0)
 		goto err_icmpmsg_mib;
@@ -1402,8 +1405,6 @@ static int __init init_ipv4_mibs(void)
 err_net:
 	snmp_mib_free((void **)icmpmsg_statistics);
 err_icmpmsg_mib:
-	snmp_mib_free((void **)icmp_statistics);
-err_icmp_mib:
 	return -ENOMEM;
 }
 
diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c
index ea60ad41008c..33f958902d9b 100644
--- a/net/ipv4/icmp.c
+++ b/net/ipv4/icmp.c
@@ -114,7 +114,6 @@ struct icmp_bxm {
 /*
  *	Statistics
  */
-DEFINE_SNMP_STAT(struct icmp_mib, icmp_statistics) __read_mostly;
 DEFINE_SNMP_STAT(struct icmpmsg_mib, icmpmsg_statistics) __read_mostly;
 
 /* An array of errno for error messages from dest unreach. */
@@ -1213,5 +1212,4 @@ int __init icmp_init(void)
 
 EXPORT_SYMBOL(icmp_err_convert);
 EXPORT_SYMBOL(icmp_send);
-EXPORT_SYMBOL(icmp_statistics);
 EXPORT_SYMBOL(xrlim_allow);
diff --git a/net/ipv4/proc.c b/net/ipv4/proc.c
index 765418334b3c..6b43cce05587 100644
--- a/net/ipv4/proc.c
+++ b/net/ipv4/proc.c
@@ -311,15 +311,15 @@ static void icmp_put(struct seq_file *seq)
 	for (i=0; icmpmibmap[i].name != NULL; i++)
 		seq_printf(seq, " Out%s", icmpmibmap[i].name);
 	seq_printf(seq, "\nIcmp: %lu %lu",
-		snmp_fold_field((void **) icmp_statistics, ICMP_MIB_INMSGS),
-		snmp_fold_field((void **) icmp_statistics, ICMP_MIB_INERRORS));
+		snmp_fold_field((void **) init_net.mib.icmp_statistics, ICMP_MIB_INMSGS),
+		snmp_fold_field((void **) init_net.mib.icmp_statistics, ICMP_MIB_INERRORS));
 	for (i=0; icmpmibmap[i].name != NULL; i++)
 		seq_printf(seq, " %lu",
 			snmp_fold_field((void **) icmpmsg_statistics,
 				icmpmibmap[i].index));
 	seq_printf(seq, " %lu %lu",
-		snmp_fold_field((void **) icmp_statistics, ICMP_MIB_OUTMSGS),
-		snmp_fold_field((void **) icmp_statistics, ICMP_MIB_OUTERRORS));
+		snmp_fold_field((void **) init_net.mib.icmp_statistics, ICMP_MIB_OUTMSGS),
+		snmp_fold_field((void **) init_net.mib.icmp_statistics, ICMP_MIB_OUTERRORS));
 	for (i=0; icmpmibmap[i].name != NULL; i++)
 		seq_printf(seq, " %lu",
 			snmp_fold_field((void **) icmpmsg_statistics,
-- 
cgit v1.2.3


From 923c6586b0dc0a00df07a1608185437145a0c68b Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Fri, 18 Jul 2008 04:04:22 -0700
Subject: mib: put icmpmsg statistics on struct net

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/icmp.h      |  5 ++---
 include/net/netns/mib.h |  1 +
 net/ipv4/af_inet.c      | 12 ++++++------
 net/ipv4/icmp.c         |  5 -----
 net/ipv4/proc.c         | 10 +++++-----
 5 files changed, 14 insertions(+), 19 deletions(-)

(limited to 'include')

diff --git a/include/net/icmp.h b/include/net/icmp.h
index e14f2c03f42e..dfa72d4e8907 100644
--- a/include/net/icmp.h
+++ b/include/net/icmp.h
@@ -29,11 +29,10 @@ struct icmp_err {
 };
 
 extern struct icmp_err icmp_err_convert[];
-DECLARE_SNMP_STAT(struct icmpmsg_mib, icmpmsg_statistics);
 #define ICMP_INC_STATS(net, field)	SNMP_INC_STATS((net)->mib.icmp_statistics, field)
 #define ICMP_INC_STATS_BH(net, field)	SNMP_INC_STATS_BH((net)->mib.icmp_statistics, field)
-#define ICMPMSGOUT_INC_STATS(net, field)	SNMP_INC_STATS(icmpmsg_statistics, field+256)
-#define ICMPMSGIN_INC_STATS_BH(net, field)	SNMP_INC_STATS_BH(icmpmsg_statistics, field)
+#define ICMPMSGOUT_INC_STATS(net, field)	SNMP_INC_STATS((net)->mib.icmpmsg_statistics, field+256)
+#define ICMPMSGIN_INC_STATS_BH(net, field)	SNMP_INC_STATS_BH((net)->mib.icmpmsg_statistics, field)
 
 struct dst_entry;
 struct net_proto_family;
diff --git a/include/net/netns/mib.h b/include/net/netns/mib.h
index 03f5abdf58e5..449147604642 100644
--- a/include/net/netns/mib.h
+++ b/include/net/netns/mib.h
@@ -10,6 +10,7 @@ struct netns_mib {
 	DEFINE_SNMP_STAT(struct udp_mib, udp_statistics);
 	DEFINE_SNMP_STAT(struct udp_mib, udplite_statistics);
 	DEFINE_SNMP_STAT(struct icmp_mib, icmp_statistics);
+	DEFINE_SNMP_STAT(struct icmpmsg_mib, icmpmsg_statistics);
 };
 
 #endif
diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c
index 5c72bb3adadd..36ff6dc769c6 100644
--- a/net/ipv4/af_inet.c
+++ b/net/ipv4/af_inet.c
@@ -1358,10 +1358,15 @@ static __net_init int ipv4_mib_init_net(struct net *net)
 	if (snmp_mib_init((void **)net->mib.icmp_statistics,
 			  sizeof(struct icmp_mib)) < 0)
 		goto err_icmp_mib;
+	if (snmp_mib_init((void **)net->mib.icmpmsg_statistics,
+			  sizeof(struct icmpmsg_mib)) < 0)
+		goto err_icmpmsg_mib;
 
 	tcp_mib_init(net);
 	return 0;
 
+err_icmpmsg_mib:
+	snmp_mib_free((void **)net->mib.icmp_statistics);
 err_icmp_mib:
 	snmp_mib_free((void **)net->mib.udplite_statistics);
 err_udplite_mib:
@@ -1378,6 +1383,7 @@ err_tcp_mib:
 
 static __net_exit void ipv4_mib_exit_net(struct net *net)
 {
+	snmp_mib_free((void **)net->mib.icmpmsg_statistics);
 	snmp_mib_free((void **)net->mib.icmp_statistics);
 	snmp_mib_free((void **)net->mib.udplite_statistics);
 	snmp_mib_free((void **)net->mib.udp_statistics);
@@ -1393,18 +1399,12 @@ static __net_initdata struct pernet_operations ipv4_mib_ops = {
 
 static int __init init_ipv4_mibs(void)
 {
-	if (snmp_mib_init((void **)icmpmsg_statistics,
-			  sizeof(struct icmpmsg_mib)) < 0)
-		goto err_icmpmsg_mib;
-
 	if (register_pernet_subsys(&ipv4_mib_ops))
 		goto err_net;
 
 	return 0;
 
 err_net:
-	snmp_mib_free((void **)icmpmsg_statistics);
-err_icmpmsg_mib:
 	return -ENOMEM;
 }
 
diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c
index 33f958902d9b..860558633b2c 100644
--- a/net/ipv4/icmp.c
+++ b/net/ipv4/icmp.c
@@ -111,11 +111,6 @@ struct icmp_bxm {
 	unsigned char  optbuf[40];
 };
 
-/*
- *	Statistics
- */
-DEFINE_SNMP_STAT(struct icmpmsg_mib, icmpmsg_statistics) __read_mostly;
-
 /* An array of errno for error messages from dest unreach. */
 /* RFC 1122: 3.2.2.1 States that NET_UNREACH, HOST_UNREACH and SR_FAILED MUST be considered 'transient errs'. */
 
diff --git a/net/ipv4/proc.c b/net/ipv4/proc.c
index 6b43cce05587..e11144b0134a 100644
--- a/net/ipv4/proc.c
+++ b/net/ipv4/proc.c
@@ -270,7 +270,7 @@ static void icmpmsg_put(struct seq_file *seq)
 	count = 0;
 	for (i = 0; i < ICMPMSG_MIB_MAX; i++) {
 
-		if (snmp_fold_field((void **) icmpmsg_statistics, i))
+		if (snmp_fold_field((void **) init_net.mib.icmpmsg_statistics, i))
 			out[count++] = i;
 		if (count < PERLINE)
 			continue;
@@ -282,7 +282,7 @@ static void icmpmsg_put(struct seq_file *seq)
 		seq_printf(seq, "\nIcmpMsg: ");
 		for (j = 0; j < PERLINE; ++j)
 			seq_printf(seq, " %lu",
-				snmp_fold_field((void **) icmpmsg_statistics,
+				snmp_fold_field((void **) init_net.mib.icmpmsg_statistics,
 				out[j]));
 		seq_putc(seq, '\n');
 	}
@@ -294,7 +294,7 @@ static void icmpmsg_put(struct seq_file *seq)
 		seq_printf(seq, "\nIcmpMsg:");
 		for (j = 0; j < count; ++j)
 			seq_printf(seq, " %lu", snmp_fold_field((void **)
-				icmpmsg_statistics, out[j]));
+				init_net.mib.icmpmsg_statistics, out[j]));
 	}
 
 #undef PERLINE
@@ -315,14 +315,14 @@ static void icmp_put(struct seq_file *seq)
 		snmp_fold_field((void **) init_net.mib.icmp_statistics, ICMP_MIB_INERRORS));
 	for (i=0; icmpmibmap[i].name != NULL; i++)
 		seq_printf(seq, " %lu",
-			snmp_fold_field((void **) icmpmsg_statistics,
+			snmp_fold_field((void **) init_net.mib.icmpmsg_statistics,
 				icmpmibmap[i].index));
 	seq_printf(seq, " %lu %lu",
 		snmp_fold_field((void **) init_net.mib.icmp_statistics, ICMP_MIB_OUTMSGS),
 		snmp_fold_field((void **) init_net.mib.icmp_statistics, ICMP_MIB_OUTERRORS));
 	for (i=0; icmpmibmap[i].name != NULL; i++)
 		seq_printf(seq, " %lu",
-			snmp_fold_field((void **) icmpmsg_statistics,
+			snmp_fold_field((void **) init_net.mib.icmpmsg_statistics,
 				icmpmibmap[i].index | 0x100));
 }
 
-- 
cgit v1.2.3


From de05c557b24c7dffc6d392e3db120cf11c9f6ae7 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Fri, 18 Jul 2008 04:07:21 -0700
Subject: proc: consolidate per-net single_open callers

There are already 7 of them - time to kill some duplicate code.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 fs/proc/proc_net.c           | 24 +++++++++++++++++++
 include/linux/seq_file_net.h |  2 ++
 net/ipv4/fib_trie.c          | 13 +---------
 net/ipv4/proc.c              | 57 +++-----------------------------------------
 net/ipv6/proc.c              | 19 +--------------
 net/ipv6/route.c             | 26 ++------------------
 6 files changed, 33 insertions(+), 108 deletions(-)

(limited to 'include')

diff --git a/fs/proc/proc_net.c b/fs/proc/proc_net.c
index 83f357b30d71..ab232ad2e6f9 100644
--- a/fs/proc/proc_net.c
+++ b/fs/proc/proc_net.c
@@ -51,6 +51,30 @@ int seq_open_net(struct inode *ino, struct file *f,
 }
 EXPORT_SYMBOL_GPL(seq_open_net);
 
+int single_open_net(struct inode *inode, struct file *file,
+		int (*show)(struct seq_file *, void *))
+{
+	int err;
+	struct net *net;
+
+	err = -ENXIO;
+	net = get_proc_net(inode);
+	if (net == NULL)
+		goto err_net;
+
+	err = single_open(file, show, net);
+	if (err < 0)
+		goto err_open;
+
+	return 0;
+
+err_open:
+	put_net(net);
+err_net:
+	return err;
+}
+EXPORT_SYMBOL_GPL(single_open_net);
+
 int seq_release_net(struct inode *ino, struct file *f)
 {
 	struct seq_file *seq;
diff --git a/include/linux/seq_file_net.h b/include/linux/seq_file_net.h
index 4ac52542a563..87dcc0ecf6eb 100644
--- a/include/linux/seq_file_net.h
+++ b/include/linux/seq_file_net.h
@@ -14,6 +14,8 @@ struct seq_net_private {
 
 int seq_open_net(struct inode *, struct file *,
 		 const struct seq_operations *, int);
+int single_open_net(struct inode *, struct file *file,
+		int (*show)(struct seq_file *, void *));
 int seq_release_net(struct inode *, struct file *);
 static inline struct net *seq_file_net(struct seq_file *seq)
 {
diff --git a/net/ipv4/fib_trie.c b/net/ipv4/fib_trie.c
index f155a66d6ebf..6009df238ed9 100644
--- a/net/ipv4/fib_trie.c
+++ b/net/ipv4/fib_trie.c
@@ -2251,18 +2251,7 @@ static int fib_triestat_seq_show(struct seq_file *seq, void *v)
 
 static int fib_triestat_seq_open(struct inode *inode, struct file *file)
 {
-	int err;
-	struct net *net;
-
-	net = get_proc_net(inode);
-	if (net == NULL)
-		return -ENXIO;
-	err = single_open(file, fib_triestat_seq_show, net);
-	if (err < 0) {
-		put_net(net);
-		return err;
-	}
-	return 0;
+	return single_open_net(inode, file, fib_triestat_seq_show);
 }
 
 static int fib_triestat_seq_release(struct inode *ino, struct file *f)
diff --git a/net/ipv4/proc.c b/net/ipv4/proc.c
index 554390431a4e..daf5d3c80cef 100644
--- a/net/ipv4/proc.c
+++ b/net/ipv4/proc.c
@@ -71,24 +71,7 @@ static int sockstat_seq_show(struct seq_file *seq, void *v)
 
 static int sockstat_seq_open(struct inode *inode, struct file *file)
 {
-	int err;
-	struct net *net;
-
-	err = -ENXIO;
-	net = get_proc_net(inode);
-	if (net == NULL)
-		goto err_net;
-
-	err = single_open(file, sockstat_seq_show, net);
-	if (err < 0)
-		goto err_open;
-
-	return 0;
-
-err_open:
-	put_net(net);
-err_net:
-	return err;
+	return single_open_net(inode, file, sockstat_seq_show);
 }
 
 static int sockstat_seq_release(struct inode *inode, struct file *file)
@@ -397,24 +380,7 @@ static int snmp_seq_show(struct seq_file *seq, void *v)
 
 static int snmp_seq_open(struct inode *inode, struct file *file)
 {
-	int err;
-	struct net *net;
-
-	err = -ENXIO;
-	net = get_proc_net(inode);
-	if (net == NULL)
-		goto err_net;
-
-	err = single_open(file, snmp_seq_show, net);
-	if (err < 0)
-		goto err_open;
-
-	return 0;
-
-err_open:
-	put_net(net);
-err_net:
-	return err;
+	return single_open_net(inode, file, snmp_seq_show);
 }
 
 static int snmp_seq_release(struct inode *inode, struct file *file)
@@ -469,24 +435,7 @@ static int netstat_seq_show(struct seq_file *seq, void *v)
 
 static int netstat_seq_open(struct inode *inode, struct file *file)
 {
-	int err;
-	struct net *net;
-
-	err = -ENXIO;
-	net = get_proc_net(inode);
-	if (net == NULL)
-		goto err_net;
-
-	err = single_open(file, netstat_seq_show, net);
-	if (err < 0)
-		goto err_open;
-
-	return 0;
-
-err_open:
-	put_net(net);
-err_net:
-	return err;
+	return single_open_net(inode, file, netstat_seq_show);
 }
 
 static int netstat_seq_release(struct inode *inode, struct file *file)
diff --git a/net/ipv6/proc.c b/net/ipv6/proc.c
index cbc7e514d3ec..29c5a79444c2 100644
--- a/net/ipv6/proc.c
+++ b/net/ipv6/proc.c
@@ -183,24 +183,7 @@ static int snmp6_seq_show(struct seq_file *seq, void *v)
 
 static int sockstat6_seq_open(struct inode *inode, struct file *file)
 {
-	int err;
-	struct net *net;
-
-	err = -ENXIO;
-	net = get_proc_net(inode);
-	if (net == NULL)
-		goto err_net;
-
-	err = single_open(file, sockstat6_seq_show, net);
-	if (err < 0)
-		goto err_open;
-
-	return 0;
-
-err_open:
-	put_net(net);
-err_net:
-	return err;
+	return single_open_net(inode, file, sockstat6_seq_show);
 }
 
 static int sockstat6_seq_release(struct inode *inode, struct file *file)
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index 5d6c166dfbb6..fb7ff8f0c6db 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -2416,18 +2416,7 @@ static int ipv6_route_show(struct seq_file *m, void *v)
 
 static int ipv6_route_open(struct inode *inode, struct file *file)
 {
-	int err;
-	struct net *net = get_proc_net(inode);
-	if (!net)
-		return -ENXIO;
-
-	err = single_open(file, ipv6_route_show, net);
-	if (err < 0) {
-		put_net(net);
-		return err;
-	}
-
-	return 0;
+	return single_open_net(inode, file, ipv6_route_show);
 }
 
 static int ipv6_route_release(struct inode *inode, struct file *file)
@@ -2463,18 +2452,7 @@ static int rt6_stats_seq_show(struct seq_file *seq, void *v)
 
 static int rt6_stats_seq_open(struct inode *inode, struct file *file)
 {
-	int err;
-	struct net *net = get_proc_net(inode);
-	if (!net)
-		return -ENXIO;
-
-	err = single_open(file, rt6_stats_seq_show, net);
-	if (err < 0) {
-		put_net(net);
-		return err;
-	}
-
-	return 0;
+	return single_open_net(inode, file, rt6_stats_seq_show);
 }
 
 static int rt6_stats_seq_release(struct inode *inode, struct file *file)
-- 
cgit v1.2.3


From b6fcbdb4f283f7ba67cec3cda6be23da8e959031 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Fri, 18 Jul 2008 04:07:44 -0700
Subject: proc: consolidate per-net single-release callers

They are symmetrical to single_open ones :)

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 fs/proc/proc_net.c           |  8 ++++++++
 include/linux/seq_file_net.h |  1 +
 net/ipv4/fib_trie.c          |  9 +--------
 net/ipv4/proc.c              | 30 +++---------------------------
 net/ipv6/proc.c              | 10 +---------
 net/ipv6/route.c             | 20 ++------------------
 6 files changed, 16 insertions(+), 62 deletions(-)

(limited to 'include')

diff --git a/fs/proc/proc_net.c b/fs/proc/proc_net.c
index ab232ad2e6f9..b224a28e0c15 100644
--- a/fs/proc/proc_net.c
+++ b/fs/proc/proc_net.c
@@ -87,6 +87,14 @@ int seq_release_net(struct inode *ino, struct file *f)
 }
 EXPORT_SYMBOL_GPL(seq_release_net);
 
+int single_release_net(struct inode *ino, struct file *f)
+{
+	struct seq_file *seq = f->private_data;
+	put_net(seq->private);
+	return single_release(ino, f);
+}
+EXPORT_SYMBOL_GPL(single_release_net);
+
 static struct net *get_proc_task_net(struct inode *dir)
 {
 	struct task_struct *task;
diff --git a/include/linux/seq_file_net.h b/include/linux/seq_file_net.h
index 87dcc0ecf6eb..32c89bbe24a2 100644
--- a/include/linux/seq_file_net.h
+++ b/include/linux/seq_file_net.h
@@ -17,6 +17,7 @@ int seq_open_net(struct inode *, struct file *,
 int single_open_net(struct inode *, struct file *file,
 		int (*show)(struct seq_file *, void *));
 int seq_release_net(struct inode *, struct file *);
+int single_release_net(struct inode *, struct file *);
 static inline struct net *seq_file_net(struct seq_file *seq)
 {
 #ifdef CONFIG_NET_NS
diff --git a/net/ipv4/fib_trie.c b/net/ipv4/fib_trie.c
index 6009df238ed9..5cb72786a8af 100644
--- a/net/ipv4/fib_trie.c
+++ b/net/ipv4/fib_trie.c
@@ -2254,19 +2254,12 @@ static int fib_triestat_seq_open(struct inode *inode, struct file *file)
 	return single_open_net(inode, file, fib_triestat_seq_show);
 }
 
-static int fib_triestat_seq_release(struct inode *ino, struct file *f)
-{
-	struct seq_file *seq = f->private_data;
-	put_net(seq->private);
-	return single_release(ino, f);
-}
-
 static const struct file_operations fib_triestat_fops = {
 	.owner	= THIS_MODULE,
 	.open	= fib_triestat_seq_open,
 	.read	= seq_read,
 	.llseek	= seq_lseek,
-	.release = fib_triestat_seq_release,
+	.release = single_release_net,
 };
 
 static struct node *fib_trie_get_idx(struct seq_file *seq, loff_t pos)
diff --git a/net/ipv4/proc.c b/net/ipv4/proc.c
index daf5d3c80cef..834356ea99df 100644
--- a/net/ipv4/proc.c
+++ b/net/ipv4/proc.c
@@ -74,20 +74,12 @@ static int sockstat_seq_open(struct inode *inode, struct file *file)
 	return single_open_net(inode, file, sockstat_seq_show);
 }
 
-static int sockstat_seq_release(struct inode *inode, struct file *file)
-{
-	struct net *net = ((struct seq_file *)file->private_data)->private;
-
-	put_net(net);
-	return single_release(inode, file);
-}
-
 static const struct file_operations sockstat_seq_fops = {
 	.owner	 = THIS_MODULE,
 	.open	 = sockstat_seq_open,
 	.read	 = seq_read,
 	.llseek	 = seq_lseek,
-	.release = sockstat_seq_release,
+	.release = single_release_net,
 };
 
 /* snmp items */
@@ -383,20 +375,12 @@ static int snmp_seq_open(struct inode *inode, struct file *file)
 	return single_open_net(inode, file, snmp_seq_show);
 }
 
-static int snmp_seq_release(struct inode *inode, struct file *file)
-{
-	struct net *net = ((struct seq_file *)file->private_data)->private;
-
-	put_net(net);
-	return single_release(inode, file);
-}
-
 static const struct file_operations snmp_seq_fops = {
 	.owner	 = THIS_MODULE,
 	.open	 = snmp_seq_open,
 	.read	 = seq_read,
 	.llseek	 = seq_lseek,
-	.release = snmp_seq_release,
+	.release = single_release_net,
 };
 
 
@@ -438,20 +422,12 @@ static int netstat_seq_open(struct inode *inode, struct file *file)
 	return single_open_net(inode, file, netstat_seq_show);
 }
 
-static int netstat_seq_release(struct inode *inode, struct file *file)
-{
-	struct net *net = ((struct seq_file *)file->private_data)->private;
-
-	put_net(net);
-	return single_release(inode, file);
-}
-
 static const struct file_operations netstat_seq_fops = {
 	.owner	 = THIS_MODULE,
 	.open	 = netstat_seq_open,
 	.read	 = seq_read,
 	.llseek	 = seq_lseek,
-	.release = netstat_seq_release,
+	.release = single_release_net,
 };
 
 static __net_init int ip_proc_init_net(struct net *net)
diff --git a/net/ipv6/proc.c b/net/ipv6/proc.c
index 29c5a79444c2..70940b3654a1 100644
--- a/net/ipv6/proc.c
+++ b/net/ipv6/proc.c
@@ -186,20 +186,12 @@ static int sockstat6_seq_open(struct inode *inode, struct file *file)
 	return single_open_net(inode, file, sockstat6_seq_show);
 }
 
-static int sockstat6_seq_release(struct inode *inode, struct file *file)
-{
-	struct net *net = ((struct seq_file *)file->private_data)->private;
-
-	put_net(net);
-	return single_release(inode, file);
-}
-
 static const struct file_operations sockstat6_seq_fops = {
 	.owner	 = THIS_MODULE,
 	.open	 = sockstat6_seq_open,
 	.read	 = seq_read,
 	.llseek	 = seq_lseek,
-	.release = sockstat6_seq_release,
+	.release = single_release_net,
 };
 
 static int snmp6_seq_open(struct inode *inode, struct file *file)
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index fb7ff8f0c6db..cb8a51271b67 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -2419,20 +2419,12 @@ static int ipv6_route_open(struct inode *inode, struct file *file)
 	return single_open_net(inode, file, ipv6_route_show);
 }
 
-static int ipv6_route_release(struct inode *inode, struct file *file)
-{
-	struct seq_file *seq = file->private_data;
-	struct net *net = seq->private;
-	put_net(net);
-	return single_release(inode, file);
-}
-
 static const struct file_operations ipv6_route_proc_fops = {
 	.owner		= THIS_MODULE,
 	.open		= ipv6_route_open,
 	.read		= seq_read,
 	.llseek		= seq_lseek,
-	.release	= ipv6_route_release,
+	.release	= single_release_net,
 };
 
 static int rt6_stats_seq_show(struct seq_file *seq, void *v)
@@ -2455,20 +2447,12 @@ static int rt6_stats_seq_open(struct inode *inode, struct file *file)
 	return single_open_net(inode, file, rt6_stats_seq_show);
 }
 
-static int rt6_stats_seq_release(struct inode *inode, struct file *file)
-{
-	struct seq_file *seq = file->private_data;
-	struct net *net = (struct net *)seq->private;
-	put_net(net);
-	return single_release(inode, file);
-}
-
 static const struct file_operations rt6_stats_seq_fops = {
 	.owner	 = THIS_MODULE,
 	.open	 = rt6_stats_seq_open,
 	.read	 = seq_read,
 	.llseek	 = seq_lseek,
-	.release = rt6_stats_seq_release,
+	.release = single_release_net,
 };
 #endif	/* CONFIG_PROC_FS */
 
-- 
cgit v1.2.3


From 8913336a7e8d56e984109a3137d6c0e3362596a4 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Fri, 18 Jul 2008 18:05:19 -0700
Subject: packet: add PACKET_RESERVE sockopt

Add new sockopt to reserve some headroom in the mmaped ring frames in
front of the packet payload. This can be used f.i. when the VLAN header
needs to be (re)constructed to avoid moving the entire payload.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/if_packet.h |  1 +
 net/packet/af_packet.c    | 29 ++++++++++++++++++++++++++---
 2 files changed, 27 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/linux/if_packet.h b/include/linux/if_packet.h
index a630295b255f..18db0668065a 100644
--- a/include/linux/if_packet.h
+++ b/include/linux/if_packet.h
@@ -45,6 +45,7 @@ struct sockaddr_ll
 #define PACKET_ORIGDEV			9
 #define PACKET_VERSION			10
 #define PACKET_HDRLEN			11
+#define PACKET_RESERVE			12
 
 struct tpacket_stats
 {
diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c
index db792e02a37f..de73bcb5235a 100644
--- a/net/packet/af_packet.c
+++ b/net/packet/af_packet.c
@@ -188,6 +188,7 @@ struct packet_sock {
 	unsigned int		pg_vec_len;
 	enum tpacket_versions	tp_version;
 	unsigned int		tp_hdrlen;
+	unsigned int		tp_reserve;
 #endif
 };
 
@@ -635,11 +636,13 @@ static int tpacket_rcv(struct sk_buff *skb, struct net_device *dev, struct packe
 		snaplen = res;
 
 	if (sk->sk_type == SOCK_DGRAM) {
-		macoff = netoff = TPACKET_ALIGN(po->tp_hdrlen) + 16;
+		macoff = netoff = TPACKET_ALIGN(po->tp_hdrlen) + 16 +
+				  po->tp_reserve;
 	} else {
 		unsigned maclen = skb_network_offset(skb);
 		netoff = TPACKET_ALIGN(po->tp_hdrlen +
-				       (maclen < 16 ? 16 : maclen));
+				       (maclen < 16 ? 16 : maclen)) +
+			po->tp_reserve;
 		macoff = netoff - maclen;
 	}
 
@@ -1448,6 +1451,19 @@ packet_setsockopt(struct socket *sock, int level, int optname, char __user *optv
 			return -EINVAL;
 		}
 	}
+	case PACKET_RESERVE:
+	{
+		unsigned int val;
+
+		if (optlen != sizeof(val))
+			return -EINVAL;
+		if (po->pg_vec)
+			return -EBUSY;
+		if (copy_from_user(&val, optval, sizeof(val)))
+			return -EFAULT;
+		po->tp_reserve = val;
+		return 0;
+	}
 #endif
 	case PACKET_AUXDATA:
 	{
@@ -1547,6 +1563,12 @@ static int packet_getsockopt(struct socket *sock, int level, int optname,
 		}
 		data = &val;
 		break;
+	case PACKET_RESERVE:
+		if (len > sizeof(unsigned int))
+			len = sizeof(unsigned int);
+		val = po->tp_reserve;
+		data = &val;
+		break;
 #endif
 	default:
 		return -ENOPROTOOPT;
@@ -1790,7 +1812,8 @@ static int packet_set_ring(struct sock *sk, struct tpacket_req *req, int closing
 			return -EINVAL;
 		if (unlikely(req->tp_block_size & (PAGE_SIZE - 1)))
 			return -EINVAL;
-		if (unlikely(req->tp_frame_size < po->tp_hdrlen))
+		if (unlikely(req->tp_frame_size < po->tp_hdrlen +
+						  po->tp_reserve))
 			return -EINVAL;
 		if (unlikely(req->tp_frame_size & (TPACKET_ALIGNMENT - 1)))
 			return -EINVAL;
-- 
cgit v1.2.3


From 72b25a913ed9b1ab49c7022adaf3f271a65ea219 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Fri, 18 Jul 2008 20:54:17 -0700
Subject: pkt_sched: Get rid of u32_list.

The u32_list is just an indirect way of maintaining a reference
to a U32 node on a per-qdisc basis.

Just add an explicit node pointer for u32 to struct Qdisc an do
away with this global list.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/sch_generic.h |  2 ++
 net/sched/cls_u32.c       | 18 +++---------------
 2 files changed, 5 insertions(+), 15 deletions(-)

(limited to 'include')

diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h
index 0a158ff4de12..8a44386b35cf 100644
--- a/include/net/sch_generic.h
+++ b/include/net/sch_generic.h
@@ -56,6 +56,8 @@ struct Qdisc
 	int			(*reshape_fail)(struct sk_buff *skb,
 					struct Qdisc *q);
 
+	void			*u32_node;
+
 	/* This field is deprecated, but it is still used by CBQ
 	 * and it will live until better solution will be invented.
 	 */
diff --git a/net/sched/cls_u32.c b/net/sched/cls_u32.c
index 4d755444c449..527db2559dd2 100644
--- a/net/sched/cls_u32.c
+++ b/net/sched/cls_u32.c
@@ -75,7 +75,6 @@ struct tc_u_hnode
 
 struct tc_u_common
 {
-	struct tc_u_common	*next;
 	struct tc_u_hnode	*hlist;
 	struct Qdisc		*q;
 	int			refcnt;
@@ -87,8 +86,6 @@ static const struct tcf_ext_map u32_ext_map = {
 	.police = TCA_U32_POLICE
 };
 
-static struct tc_u_common *u32_list;
-
 static __inline__ unsigned u32_hash_fold(__be32 key, struct tc_u32_sel *sel, u8 fshift)
 {
 	unsigned h = ntohl(key & sel->hmask)>>fshift;
@@ -287,9 +284,7 @@ static int u32_init(struct tcf_proto *tp)
 	struct tc_u_hnode *root_ht;
 	struct tc_u_common *tp_c;
 
-	for (tp_c = u32_list; tp_c; tp_c = tp_c->next)
-		if (tp_c->q == tp->q)
-			break;
+	tp_c = tp->q->u32_node;
 
 	root_ht = kzalloc(sizeof(*root_ht), GFP_KERNEL);
 	if (root_ht == NULL)
@@ -307,8 +302,7 @@ static int u32_init(struct tcf_proto *tp)
 			return -ENOBUFS;
 		}
 		tp_c->q = tp->q;
-		tp_c->next = u32_list;
-		u32_list = tp_c;
+		tp->q->u32_node = tp_c;
 	}
 
 	tp_c->refcnt++;
@@ -402,14 +396,8 @@ static void u32_destroy(struct tcf_proto *tp)
 
 	if (--tp_c->refcnt == 0) {
 		struct tc_u_hnode *ht;
-		struct tc_u_common **tp_cp;
 
-		for (tp_cp = &u32_list; *tp_cp; tp_cp = &(*tp_cp)->next) {
-			if (*tp_cp == tp_c) {
-				*tp_cp = tp_c->next;
-				break;
-			}
-		}
+		tp->q->u32_node = NULL;
 
 		for (ht = tp_c->hlist; ht; ht = ht->next) {
 			ht->refcnt--;
-- 
cgit v1.2.3


From 3072367300aa8c779e3a14ee8e89de079e90f3ad Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Fri, 18 Jul 2008 22:50:15 -0700
Subject: pkt_sched: Manage qdisc list inside of root qdisc.

Idea is from Patrick McHardy.

Instead of managing the list of qdiscs on the device level, manage it
in the root qdisc of a netdev_queue.  This solves all kinds of
visibility issues during qdisc destruction.

The way to iterate over all qdiscs of a netdev_queue is to visit
the netdev_queue->qdisc, and then traverse it's list.

The only special case is to ignore builting qdiscs at the root when
dumping or doing a qdisc_lookup().  That was not needed previously
because builtin qdiscs were not added to the device's qdisc_list.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h |   2 -
 net/core/dev.c            |   2 -
 net/sched/sch_api.c       | 175 ++++++++++++++++++++++++++++++++++------------
 net/sched/sch_generic.c   |  10 +--
 4 files changed, 133 insertions(+), 56 deletions(-)

(limited to 'include')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 9c5a68850114..812bcd8b4363 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -636,8 +636,6 @@ struct net_device
 	unsigned int		real_num_tx_queues;
 
 	unsigned long		tx_queue_len;	/* Max frames per queue allowed */
-	spinlock_t		qdisc_list_lock;
-	struct list_head	qdisc_list;
 
 /*
  * One part is mostly used on xmit path (device)
diff --git a/net/core/dev.c b/net/core/dev.c
index e54acde839da..065b9817e209 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -3888,8 +3888,6 @@ int register_netdevice(struct net_device *dev)
 	net = dev_net(dev);
 
 	spin_lock_init(&dev->addr_list_lock);
-	spin_lock_init(&dev->qdisc_list_lock);
-	INIT_LIST_HEAD(&dev->qdisc_list);
 	netdev_init_queue_locks(dev);
 
 	dev->iflink = -1;
diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c
index b3ef8307204e..fb43731c9860 100644
--- a/net/sched/sch_api.c
+++ b/net/sched/sch_api.c
@@ -185,11 +185,20 @@ EXPORT_SYMBOL(unregister_qdisc);
 
 struct Qdisc *qdisc_lookup(struct net_device *dev, u32 handle)
 {
-	struct Qdisc *q;
+	unsigned int i;
+
+	for (i = 0; i < dev->num_tx_queues; i++) {
+		struct netdev_queue *txq = netdev_get_tx_queue(dev, i);
+		struct Qdisc *q, *txq_root = txq->qdisc;
 
-	list_for_each_entry(q, &dev->qdisc_list, list) {
-		if (q->handle == handle)
-			return q;
+		if (!(txq_root->flags & TCQ_F_BUILTIN) &&
+		    txq_root->handle == handle)
+			return txq_root;
+
+		list_for_each_entry(q, &txq_root->list, list) {
+			if (q->handle == handle)
+				return q;
+		}
 	}
 	return NULL;
 }
@@ -676,9 +685,8 @@ qdisc_create(struct net_device *dev, struct netdev_queue *dev_queue,
 				goto err_out3;
 			}
 		}
-		spin_lock_bh(&dev->qdisc_list_lock);
-		list_add_tail(&sch->list, &dev->qdisc_list);
-		spin_unlock_bh(&dev->qdisc_list_lock);
+		if (parent)
+			list_add_tail(&sch->list, &dev_queue->qdisc->list);
 
 		return sch;
 	}
@@ -1037,13 +1045,57 @@ err_out:
 	return -EINVAL;
 }
 
+static bool tc_qdisc_dump_ignore(struct Qdisc *q)
+{
+	return (q->flags & TCQ_F_BUILTIN) ? true : false;
+}
+
+static int tc_dump_qdisc_root(struct Qdisc *root, struct sk_buff *skb,
+			      struct netlink_callback *cb,
+			      int *q_idx_p, int s_q_idx)
+{
+	int ret = 0, q_idx = *q_idx_p;
+	struct Qdisc *q;
+
+	if (!root)
+		return 0;
+
+	q = root;
+	if (q_idx < s_q_idx) {
+		q_idx++;
+	} else {
+		if (!tc_qdisc_dump_ignore(q) &&
+		    tc_fill_qdisc(skb, q, q->parent, NETLINK_CB(cb->skb).pid,
+				  cb->nlh->nlmsg_seq, NLM_F_MULTI, RTM_NEWQDISC) <= 0)
+			goto done;
+		q_idx++;
+	}
+	list_for_each_entry(q, &root->list, list) {
+		if (q_idx < s_q_idx) {
+			q_idx++;
+			continue;
+		}
+		if (!tc_qdisc_dump_ignore(q) && 
+		    tc_fill_qdisc(skb, q, q->parent, NETLINK_CB(cb->skb).pid,
+				  cb->nlh->nlmsg_seq, NLM_F_MULTI, RTM_NEWQDISC) <= 0)
+			goto done;
+		q_idx++;
+	}
+
+out:
+	*q_idx_p = q_idx;
+	return ret;
+done:
+	ret = -1;
+	goto out;
+}
+
 static int tc_dump_qdisc(struct sk_buff *skb, struct netlink_callback *cb)
 {
 	struct net *net = sock_net(skb->sk);
 	int idx, q_idx;
 	int s_idx, s_q_idx;
 	struct net_device *dev;
-	struct Qdisc *q;
 
 	if (net != &init_net)
 		return 0;
@@ -1053,21 +1105,22 @@ static int tc_dump_qdisc(struct sk_buff *skb, struct netlink_callback *cb)
 	read_lock(&dev_base_lock);
 	idx = 0;
 	for_each_netdev(&init_net, dev) {
+		struct netdev_queue *dev_queue;
+
 		if (idx < s_idx)
 			goto cont;
 		if (idx > s_idx)
 			s_q_idx = 0;
 		q_idx = 0;
-		list_for_each_entry(q, &dev->qdisc_list, list) {
-			if (q_idx < s_q_idx) {
-				q_idx++;
-				continue;
-			}
-			if (tc_fill_qdisc(skb, q, q->parent, NETLINK_CB(cb->skb).pid,
-					  cb->nlh->nlmsg_seq, NLM_F_MULTI, RTM_NEWQDISC) <= 0)
-				goto done;
-			q_idx++;
-		}
+
+		dev_queue = netdev_get_tx_queue(dev, 0);
+		if (tc_dump_qdisc_root(dev_queue->qdisc, skb, cb, &q_idx, s_q_idx) < 0)
+			goto done;
+
+		dev_queue = &dev->rx_queue;
+		if (tc_dump_qdisc_root(dev_queue->qdisc, skb, cb, &q_idx, s_q_idx) < 0)
+			goto done;
+
 cont:
 		idx++;
 	}
@@ -1285,15 +1338,62 @@ static int qdisc_class_dump(struct Qdisc *q, unsigned long cl, struct qdisc_walk
 			      a->cb->nlh->nlmsg_seq, NLM_F_MULTI, RTM_NEWTCLASS);
 }
 
+static int tc_dump_tclass_qdisc(struct Qdisc *q, struct sk_buff *skb,
+				struct tcmsg *tcm, struct netlink_callback *cb,
+				int *t_p, int s_t)
+{
+	struct qdisc_dump_args arg;
+
+	if (tc_qdisc_dump_ignore(q) ||
+	    *t_p < s_t || !q->ops->cl_ops ||
+	    (tcm->tcm_parent &&
+	     TC_H_MAJ(tcm->tcm_parent) != q->handle)) {
+		(*t_p)++;
+		return 0;
+	}
+	if (*t_p > s_t)
+		memset(&cb->args[1], 0, sizeof(cb->args)-sizeof(cb->args[0]));
+	arg.w.fn = qdisc_class_dump;
+	arg.skb = skb;
+	arg.cb = cb;
+	arg.w.stop  = 0;
+	arg.w.skip = cb->args[1];
+	arg.w.count = 0;
+	q->ops->cl_ops->walk(q, &arg.w);
+	cb->args[1] = arg.w.count;
+	if (arg.w.stop)
+		return -1;
+	(*t_p)++;
+	return 0;
+}
+
+static int tc_dump_tclass_root(struct Qdisc *root, struct sk_buff *skb,
+			       struct tcmsg *tcm, struct netlink_callback *cb,
+			       int *t_p, int s_t)
+{
+	struct Qdisc *q;
+
+	if (!root)
+		return 0;
+
+	if (tc_dump_tclass_qdisc(root, skb, tcm, cb, t_p, s_t) < 0)
+		return -1;
+
+	list_for_each_entry(q, &root->list, list) {
+		if (tc_dump_tclass_qdisc(q, skb, tcm, cb, t_p, s_t) < 0)
+			return -1;
+	}
+
+	return 0;
+}
+
 static int tc_dump_tclass(struct sk_buff *skb, struct netlink_callback *cb)
 {
+	struct tcmsg *tcm = (struct tcmsg*)NLMSG_DATA(cb->nlh);
 	struct net *net = sock_net(skb->sk);
-	int t;
-	int s_t;
+	struct netdev_queue *dev_queue;
 	struct net_device *dev;
-	struct Qdisc *q;
-	struct tcmsg *tcm = (struct tcmsg*)NLMSG_DATA(cb->nlh);
-	struct qdisc_dump_args arg;
+	int t, s_t;
 
 	if (net != &init_net)
 		return 0;
@@ -1306,28 +1406,15 @@ static int tc_dump_tclass(struct sk_buff *skb, struct netlink_callback *cb)
 	s_t = cb->args[0];
 	t = 0;
 
-	list_for_each_entry(q, &dev->qdisc_list, list) {
-		if (t < s_t || !q->ops->cl_ops ||
-		    (tcm->tcm_parent &&
-		     TC_H_MAJ(tcm->tcm_parent) != q->handle)) {
-			t++;
-			continue;
-		}
-		if (t > s_t)
-			memset(&cb->args[1], 0, sizeof(cb->args)-sizeof(cb->args[0]));
-		arg.w.fn = qdisc_class_dump;
-		arg.skb = skb;
-		arg.cb = cb;
-		arg.w.stop  = 0;
-		arg.w.skip = cb->args[1];
-		arg.w.count = 0;
-		q->ops->cl_ops->walk(q, &arg.w);
-		cb->args[1] = arg.w.count;
-		if (arg.w.stop)
-			break;
-		t++;
-	}
+	dev_queue = netdev_get_tx_queue(dev, 0);
+	if (tc_dump_tclass_root(dev_queue->qdisc, skb, tcm, cb, &t, s_t) < 0)
+		goto done;
+
+	dev_queue = &dev->rx_queue;
+	if (tc_dump_tclass_root(dev_queue->qdisc, skb, tcm, cb, &t, s_t) < 0)
+		goto done;
 
+done:
 	cb->args[0] = t;
 
 	dev_put(dev);
diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c
index e244c462e6bd..14cc443d0490 100644
--- a/net/sched/sch_generic.c
+++ b/net/sched/sch_generic.c
@@ -480,15 +480,12 @@ static void __qdisc_destroy(struct rcu_head *head)
 
 void qdisc_destroy(struct Qdisc *qdisc)
 {
-	struct net_device *dev = qdisc_dev(qdisc);
-
 	if (qdisc->flags & TCQ_F_BUILTIN ||
 	    !atomic_dec_and_test(&qdisc->refcnt))
 		return;
 
-	spin_lock_bh(&dev->qdisc_list_lock);
-	list_del(&qdisc->list);
-	spin_unlock_bh(&dev->qdisc_list_lock);
+	if (qdisc->parent)
+		list_del(&qdisc->list);
 
 	call_rcu(&qdisc->q_rcu, __qdisc_destroy);
 }
@@ -520,9 +517,6 @@ static void attach_one_default_qdisc(struct net_device *dev,
 			printk(KERN_INFO "%s: activation failed\n", dev->name);
 			return;
 		}
-		spin_lock_bh(&dev->qdisc_list_lock);
-		list_add_tail(&qdisc->list, &dev->qdisc_list);
-		spin_unlock_bh(&dev->qdisc_list_lock);
 	} else {
 		qdisc =  &noqueue_qdisc;
 	}
-- 
cgit v1.2.3


From c1e20f7c8b9ccbafc9ea78f2b406738728ce6b81 Mon Sep 17 00:00:00 2001
From: Stephen Hemminger <shemminger@vyatta.com>
Date: Fri, 18 Jul 2008 23:02:15 -0700
Subject: tcp: RTT metrics scaling

Some of the metrics (RTT, RTTVAR and RTAX_RTO_MIN) are stored in
kernel units (jiffies) and this leaks out through the netlink API to
user space where the units for jiffies are unknown.

This patches changes the kernel to convert to/from milliseconds. This
changes the ABI, but milliseconds seemed like the most natural unit
for these parameters.  Values available via syscall in
/proc/net/rt_cache and netlink will be in milliseconds.

Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/dst.h    | 12 ++++++++++++
 net/ipv4/tcp_input.c | 31 ++++++++++++++++++-------------
 2 files changed, 30 insertions(+), 13 deletions(-)

(limited to 'include')

diff --git a/include/net/dst.h b/include/net/dst.h
index 002500e631f5..c5c318a628f8 100644
--- a/include/net/dst.h
+++ b/include/net/dst.h
@@ -128,6 +128,18 @@ static inline u32 dst_mtu(const struct dst_entry *dst)
 	return mtu;
 }
 
+/* RTT metrics are stored in milliseconds for user ABI, but used as jiffies */
+static inline unsigned long dst_metric_rtt(const struct dst_entry *dst, int metric)
+{
+	return msecs_to_jiffies(dst_metric(dst, metric));
+}
+
+static inline void set_dst_metric_rtt(struct dst_entry *dst, int metric,
+				      unsigned long rtt)
+{
+	dst->metrics[metric-1] = jiffies_to_msecs(rtt);
+}
+
 static inline u32
 dst_allfrag(const struct dst_entry *dst)
 {
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index fac49a6e1611..88810bc01370 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -602,7 +602,7 @@ static u32 tcp_rto_min(struct sock *sk)
 	u32 rto_min = TCP_RTO_MIN;
 
 	if (dst && dst_metric_locked(dst, RTAX_RTO_MIN))
-		rto_min = dst_metric(dst, RTAX_RTO_MIN);
+		rto_min = dst_metric_rtt(dst, RTAX_RTO_MIN);
 	return rto_min;
 }
 
@@ -729,6 +729,7 @@ void tcp_update_metrics(struct sock *sk)
 	if (dst && (dst->flags & DST_HOST)) {
 		const struct inet_connection_sock *icsk = inet_csk(sk);
 		int m;
+		unsigned long rtt;
 
 		if (icsk->icsk_backoff || !tp->srtt) {
 			/* This session failed to estimate rtt. Why?
@@ -740,7 +741,8 @@ void tcp_update_metrics(struct sock *sk)
 			return;
 		}
 
-		m = dst_metric(dst, RTAX_RTT) - tp->srtt;
+		rtt = dst_metric_rtt(dst, RTAX_RTT);
+		m = rtt - tp->srtt;
 
 		/* If newly calculated rtt larger than stored one,
 		 * store new one. Otherwise, use EWMA. Remember,
@@ -748,12 +750,13 @@ void tcp_update_metrics(struct sock *sk)
 		 */
 		if (!(dst_metric_locked(dst, RTAX_RTT))) {
 			if (m <= 0)
-				dst->metrics[RTAX_RTT - 1] = tp->srtt;
+				set_dst_metric_rtt(dst, RTAX_RTT, tp->srtt);
 			else
-				dst->metrics[RTAX_RTT - 1] -= (m >> 3);
+				set_dst_metric_rtt(dst, RTAX_RTT, rtt - (m >> 3));
 		}
 
 		if (!(dst_metric_locked(dst, RTAX_RTTVAR))) {
+			unsigned long var;
 			if (m < 0)
 				m = -m;
 
@@ -762,11 +765,13 @@ void tcp_update_metrics(struct sock *sk)
 			if (m < tp->mdev)
 				m = tp->mdev;
 
-			if (m >= dst_metric(dst, RTAX_RTTVAR))
-				dst->metrics[RTAX_RTTVAR - 1] = m;
+			var = dst_metric_rtt(dst, RTAX_RTTVAR);
+			if (m >= var)
+				var = m;
 			else
-				dst->metrics[RTAX_RTTVAR-1] -=
-					(dst_metric(dst, RTAX_RTTVAR) - m)>>2;
+				var -= (var - m) >> 2;
+
+			set_dst_metric_rtt(dst, RTAX_RTTVAR, var);
 		}
 
 		if (tp->snd_ssthresh >= 0xFFFF) {
@@ -897,7 +902,7 @@ static void tcp_init_metrics(struct sock *sk)
 	if (dst_metric(dst, RTAX_RTT) == 0)
 		goto reset;
 
-	if (!tp->srtt && dst_metric(dst, RTAX_RTT) < (TCP_TIMEOUT_INIT << 3))
+	if (!tp->srtt && dst_metric_rtt(dst, RTAX_RTT) < (TCP_TIMEOUT_INIT << 3))
 		goto reset;
 
 	/* Initial rtt is determined from SYN,SYN-ACK.
@@ -914,12 +919,12 @@ static void tcp_init_metrics(struct sock *sk)
 	 * to low value, and then abruptly stops to do it and starts to delay
 	 * ACKs, wait for troubles.
 	 */
-	if (dst_metric(dst, RTAX_RTT) > tp->srtt) {
-		tp->srtt = dst_metric(dst, RTAX_RTT);
+	if (dst_metric_rtt(dst, RTAX_RTT) > tp->srtt) {
+		tp->srtt = dst_metric_rtt(dst, RTAX_RTT);
 		tp->rtt_seq = tp->snd_nxt;
 	}
-	if (dst_metric(dst, RTAX_RTTVAR) > tp->mdev) {
-		tp->mdev = dst_metric(dst, RTAX_RTTVAR);
+	if (dst_metric_rtt(dst, RTAX_RTTVAR) > tp->mdev) {
+		tp->mdev = dst_metric_rtt(dst, RTAX_RTTVAR);
 		tp->mdev_max = tp->rttvar = max(tp->mdev, tcp_rto_min(sk));
 	}
 	tcp_set_rto(sk);
-- 
cgit v1.2.3


From 7dab83de50c7b2b7ceac695a0b56fa6c0f95b0bc Mon Sep 17 00:00:00 2001
From: Vlad Yasevich <vladislav.yasevich@hp.com>
Date: Fri, 18 Jul 2008 23:05:40 -0700
Subject: sctp: Support ipv6only AF_INET6 sockets.

Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/sctp/structs.h |  2 ++
 net/sctp/bind_addr.c       | 37 +++++++++++++++++++++++++++++++++++++
 net/sctp/ipv6.c            | 20 ++++++++++++++++----
 net/sctp/protocol.c        |  7 +++++++
 net/sctp/sm_make_chunk.c   |  7 ++++++-
 net/sctp/socket.c          | 30 +++++++++++++++++++++++++-----
 6 files changed, 93 insertions(+), 10 deletions(-)

(limited to 'include')

diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h
index 82116e84ee34..70eb64a7e1a1 100644
--- a/include/net/sctp/structs.h
+++ b/include/net/sctp/structs.h
@@ -1211,6 +1211,8 @@ int sctp_add_bind_addr(struct sctp_bind_addr *, union sctp_addr *,
 int sctp_del_bind_addr(struct sctp_bind_addr *, union sctp_addr *);
 int sctp_bind_addr_match(struct sctp_bind_addr *, const union sctp_addr *,
 			 struct sctp_sock *);
+int sctp_bind_addr_conflict(struct sctp_bind_addr *, const union sctp_addr *,
+			 struct sctp_sock *, struct sctp_sock *);
 int sctp_bind_addr_state(const struct sctp_bind_addr *bp,
 			 const union sctp_addr *addr);
 union sctp_addr *sctp_find_unmatch_addr(struct sctp_bind_addr	*bp,
diff --git a/net/sctp/bind_addr.c b/net/sctp/bind_addr.c
index 80e6df06967a..f62bc2468935 100644
--- a/net/sctp/bind_addr.c
+++ b/net/sctp/bind_addr.c
@@ -348,6 +348,43 @@ int sctp_bind_addr_match(struct sctp_bind_addr *bp,
 	return match;
 }
 
+/* Does the address 'addr' conflict with any addresses in
+ * the bp.
+ */
+int sctp_bind_addr_conflict(struct sctp_bind_addr *bp,
+			    const union sctp_addr *addr,
+			    struct sctp_sock *bp_sp,
+			    struct sctp_sock *addr_sp)
+{
+	struct sctp_sockaddr_entry *laddr;
+	int conflict = 0;
+	struct sctp_sock *sp;
+
+	/* Pick the IPv6 socket as the basis of comparison
+	 * since it's usually a superset of the IPv4.
+	 * If there is no IPv6 socket, then default to bind_addr.
+	 */
+	if (sctp_opt2sk(bp_sp)->sk_family == AF_INET6)
+		sp = bp_sp;
+	else if (sctp_opt2sk(addr_sp)->sk_family == AF_INET6)
+		sp = addr_sp;
+	else
+		sp = bp_sp;
+
+	rcu_read_lock();
+	list_for_each_entry_rcu(laddr, &bp->address_list, list) {
+		if (!laddr->valid)
+			continue;
+
+		conflict = sp->pf->cmp_addr(&laddr->a, addr, sp);
+		if (conflict)
+			break;
+	}
+	rcu_read_unlock();
+
+	return conflict;
+}
+
 /* Get the state of the entry in the bind_addr_list */
 int sctp_bind_addr_state(const struct sctp_bind_addr *bp,
 			 const union sctp_addr *addr)
diff --git a/net/sctp/ipv6.c b/net/sctp/ipv6.c
index a2f4d4d51593..a238d6834b33 100644
--- a/net/sctp/ipv6.c
+++ b/net/sctp/ipv6.c
@@ -818,7 +818,7 @@ static int sctp_inet6_af_supported(sa_family_t family, struct sctp_sock *sp)
 		return 1;
 	/* v4-mapped-v6 addresses */
 	case AF_INET:
-		if (!__ipv6_only_sock(sctp_opt2sk(sp)) && sp->v4mapped)
+		if (!__ipv6_only_sock(sctp_opt2sk(sp)))
 			return 1;
 	default:
 		return 0;
@@ -840,6 +840,11 @@ static int sctp_inet6_cmp_addr(const union sctp_addr *addr1,
 
 	if (!af1 || !af2)
 		return 0;
+
+	/* If the socket is IPv6 only, v4 addrs will not match */
+	if (__ipv6_only_sock(sctp_opt2sk(opt)) && af1 != af2)
+		return 0;
+
 	/* Today, wildcard AF_INET/AF_INET6. */
 	if (sctp_is_any(addr1) || sctp_is_any(addr2))
 		return 1;
@@ -876,7 +881,11 @@ static int sctp_inet6_bind_verify(struct sctp_sock *opt, union sctp_addr *addr)
 				return 0;
 			}
 			dev_put(dev);
+		} else if (type == IPV6_ADDR_MAPPED) {
+			if (!opt->v4mapped)
+				return 0;
 		}
+
 		af = opt->pf->af;
 	}
 	return af->available(addr, opt);
@@ -919,9 +928,12 @@ static int sctp_inet6_send_verify(struct sctp_sock *opt, union sctp_addr *addr)
 static int sctp_inet6_supported_addrs(const struct sctp_sock *opt,
 				      __be16 *types)
 {
-	types[0] = SCTP_PARAM_IPV4_ADDRESS;
-	types[1] = SCTP_PARAM_IPV6_ADDRESS;
-	return 2;
+	types[0] = SCTP_PARAM_IPV6_ADDRESS;
+	if (!opt || !ipv6_only_sock(sctp_opt2sk(opt))) {
+		types[1] = SCTP_PARAM_IPV4_ADDRESS;
+		return 2;
+	}
+	return 1;
 }
 
 static const struct proto_ops inet6_seqpacket_ops = {
diff --git a/net/sctp/protocol.c b/net/sctp/protocol.c
index dd811a8456a8..cdd881142f2c 100644
--- a/net/sctp/protocol.c
+++ b/net/sctp/protocol.c
@@ -381,6 +381,10 @@ static int sctp_v4_addr_valid(union sctp_addr *addr,
 			      struct sctp_sock *sp,
 			      const struct sk_buff *skb)
 {
+	/* IPv4 addresses not allowed */
+	if (sp && ipv6_only_sock(sctp_opt2sk(sp)))
+		return 0;
+
 	/* Is this a non-unicast address or a unusable SCTP address? */
 	if (IS_IPV4_UNUSABLE_ADDRESS(addr->v4.sin_addr.s_addr))
 		return 0;
@@ -404,6 +408,9 @@ static int sctp_v4_available(union sctp_addr *addr, struct sctp_sock *sp)
 	   !sysctl_ip_nonlocal_bind)
 		return 0;
 
+	if (ipv6_only_sock(sctp_opt2sk(sp)))
+		return 0;
+
 	return 1;
 }
 
diff --git a/net/sctp/sm_make_chunk.c b/net/sctp/sm_make_chunk.c
index bbc7107c86cf..e8ca4e54981f 100644
--- a/net/sctp/sm_make_chunk.c
+++ b/net/sctp/sm_make_chunk.c
@@ -2364,8 +2364,13 @@ static int sctp_process_param(struct sctp_association *asoc,
 	case SCTP_PARAM_IPV6_ADDRESS:
 		if (PF_INET6 != asoc->base.sk->sk_family)
 			break;
-		/* Fall through. */
+		goto do_addr_param;
+
 	case SCTP_PARAM_IPV4_ADDRESS:
+		/* v4 addresses are not allowed on v6-only socket */
+		if (ipv6_only_sock(asoc->base.sk))
+			break;
+do_addr_param:
 		af = sctp_get_af_specific(param_type2af(param.p->type));
 		af->from_addr_param(&addr, param.addr, htons(asoc->peer.port), 0);
 		scope = sctp_scope(peer_addr);
diff --git a/net/sctp/socket.c b/net/sctp/socket.c
index 6aba01b0ce4e..a0e879bb202d 100644
--- a/net/sctp/socket.c
+++ b/net/sctp/socket.c
@@ -308,9 +308,16 @@ static struct sctp_af *sctp_sockaddr_af(struct sctp_sock *opt,
 	if (len < sizeof (struct sockaddr))
 		return NULL;
 
-	/* Does this PF support this AF? */
-	if (!opt->pf->af_supported(addr->sa.sa_family, opt))
-		return NULL;
+	/* V4 mapped address are really of AF_INET family */
+	if (addr->sa.sa_family == AF_INET6 &&
+	    ipv6_addr_v4mapped(&addr->v6.sin6_addr)) {
+		if (!opt->pf->af_supported(AF_INET, opt))
+			return NULL;
+	} else {
+		/* Does this PF support this AF? */
+		if (!opt->pf->af_supported(addr->sa.sa_family, opt))
+			return NULL;
+	}
 
 	/* If we get this far, af is valid. */
 	af = sctp_get_af_specific(addr->sa.sa_family);
@@ -4395,6 +4402,11 @@ static int sctp_getsockopt_local_addrs_num_old(struct sock *sk, int len,
 				    (AF_INET6 == addr->a.sa.sa_family))
 					continue;
 
+				if ((PF_INET6 == sk->sk_family) &&
+				    inet_v6_ipv6only(sk) &&
+				    (AF_INET == addr->a.sa.sa_family))
+					continue;
+
 				cnt++;
 			}
 			rcu_read_unlock();
@@ -4435,6 +4447,10 @@ static int sctp_copy_laddrs_old(struct sock *sk, __u16 port,
 		if ((PF_INET == sk->sk_family) &&
 		    (AF_INET6 == addr->a.sa.sa_family))
 			continue;
+		if ((PF_INET6 == sk->sk_family) &&
+		    inet_v6_ipv6only(sk) &&
+		    (AF_INET == addr->a.sa.sa_family))
+			continue;
 		memcpy(&temp, &addr->a, sizeof(temp));
 		if (!temp.v4.sin_port)
 			temp.v4.sin_port = htons(port);
@@ -4470,6 +4486,10 @@ static int sctp_copy_laddrs(struct sock *sk, __u16 port, void *to,
 		if ((PF_INET == sk->sk_family) &&
 		    (AF_INET6 == addr->a.sa.sa_family))
 			continue;
+		if ((PF_INET6 == sk->sk_family) &&
+		    inet_v6_ipv6only(sk) &&
+		    (AF_INET == addr->a.sa.sa_family))
+			continue;
 		memcpy(&temp, &addr->a, sizeof(temp));
 		if (!temp.v4.sin_port)
 			temp.v4.sin_port = htons(port);
@@ -5568,8 +5588,8 @@ pp_found:
 			    sk2->sk_state != SCTP_SS_LISTENING)
 				continue;
 
-			if (sctp_bind_addr_match(&ep2->base.bind_addr, addr,
-						 sctp_sk(sk))) {
+			if (sctp_bind_addr_conflict(&ep2->base.bind_addr, addr,
+						 sctp_sk(sk2), sctp_sk(sk))) {
 				ret = (long)sk2;
 				goto fail_unlock;
 			}
-- 
cgit v1.2.3


From 336d3262df71fcd2661180bb35d5ea41b4cbca58 Mon Sep 17 00:00:00 2001
From: Harvey Harrison <harvey.harrison@gmail.com>
Date: Fri, 18 Jul 2008 23:07:09 -0700
Subject: sctp: remove unnecessary byteshifting, calculate directly in
 big-endian

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/sctp/checksum.h            | 23 ++++++++++++++---------
 net/ipv4/netfilter/nf_nat_proto_sctp.c |  4 ++--
 net/sctp/input.c                       |  4 ++--
 net/sctp/output.c                      |  4 ++--
 4 files changed, 20 insertions(+), 15 deletions(-)

(limited to 'include')

diff --git a/include/net/sctp/checksum.h b/include/net/sctp/checksum.h
index ba75c67cb992..b799fb21519a 100644
--- a/include/net/sctp/checksum.h
+++ b/include/net/sctp/checksum.h
@@ -46,9 +46,14 @@
 #include <net/sctp/sctp.h>
 #include <linux/crc32c.h>
 
-static inline __u32 sctp_start_cksum(__u8 *buffer, __u16 length)
+static inline __be32 sctp_crc32c(__be32 crc, u8 *buffer, u16 length)
 {
-	__u32 crc = ~(__u32) 0;
+	return (__force __be32)crc32c((__force u32)crc, buffer, length);
+}
+
+static inline __be32 sctp_start_cksum(__u8 *buffer, __u16 length)
+{
+	__be32 crc = ~cpu_to_be32(0);
 	__u8  zero[sizeof(__u32)] = {0};
 
 	/* Optimize this routine to be SCTP specific, knowing how
@@ -56,23 +61,23 @@ static inline __u32 sctp_start_cksum(__u8 *buffer, __u16 length)
 	 */
 
 	/* Calculate CRC up to the checksum. */
-	crc = crc32c(crc, buffer, sizeof(struct sctphdr) - sizeof(__u32));
+	crc = sctp_crc32c(crc, buffer, sizeof(struct sctphdr) - sizeof(__u32));
 
 	/* Skip checksum field of the header. */
-	crc = crc32c(crc, zero, sizeof(__u32));
+	crc = sctp_crc32c(crc, zero, sizeof(__u32));
 
 	/* Calculate the rest of the CRC. */
-	crc = crc32c(crc, &buffer[sizeof(struct sctphdr)],
+	crc = sctp_crc32c(crc, &buffer[sizeof(struct sctphdr)],
 			    length - sizeof(struct sctphdr));
 	return crc;
 }
 
-static inline __u32 sctp_update_cksum(__u8 *buffer, __u16 length, __u32 crc32)
+static inline __be32 sctp_update_cksum(__u8 *buffer, __u16 length, __be32 crc32)
 {
-	return crc32c(crc32, buffer, length);
+	return sctp_crc32c(crc32, buffer, length);
 }
 
-static inline __u32 sctp_end_cksum(__u32 crc32)
+static inline __be32 sctp_end_cksum(__be32 crc32)
 {
-	return ntohl(~crc32);
+	return ~crc32;
 }
diff --git a/net/ipv4/netfilter/nf_nat_proto_sctp.c b/net/ipv4/netfilter/nf_nat_proto_sctp.c
index 82e4c0e286b8..65e470bc6123 100644
--- a/net/ipv4/netfilter/nf_nat_proto_sctp.c
+++ b/net/ipv4/netfilter/nf_nat_proto_sctp.c
@@ -36,7 +36,7 @@ sctp_manip_pkt(struct sk_buff *skb,
 	sctp_sctphdr_t *hdr;
 	unsigned int hdroff = iphdroff + iph->ihl*4;
 	__be32 oldip, newip;
-	u32 crc32;
+	__be32 crc32;
 
 	if (!skb_make_writable(skb, hdroff + sizeof(*hdr)))
 		return false;
@@ -61,7 +61,7 @@ sctp_manip_pkt(struct sk_buff *skb,
 		crc32 = sctp_update_cksum((u8 *)skb->data, skb_headlen(skb),
 					  crc32);
 	crc32 = sctp_end_cksum(crc32);
-	hdr->checksum = htonl(crc32);
+	hdr->checksum = crc32;
 
 	return true;
 }
diff --git a/net/sctp/input.c b/net/sctp/input.c
index 5ed93c05c23e..a49fa80b57b9 100644
--- a/net/sctp/input.c
+++ b/net/sctp/input.c
@@ -83,8 +83,8 @@ static inline int sctp_rcv_checksum(struct sk_buff *skb)
 {
 	struct sk_buff *list = skb_shinfo(skb)->frag_list;
 	struct sctphdr *sh = sctp_hdr(skb);
-	__u32 cmp = ntohl(sh->checksum);
-	__u32 val = sctp_start_cksum((__u8 *)sh, skb_headlen(skb));
+	__be32 cmp = sh->checksum;
+	__be32 val = sctp_start_cksum((__u8 *)sh, skb_headlen(skb));
 
 	for (; list; list = list->next)
 		val = sctp_update_cksum((__u8 *)list->data, skb_headlen(list),
diff --git a/net/sctp/output.c b/net/sctp/output.c
index 9a63a3fb9011..45684646b1db 100644
--- a/net/sctp/output.c
+++ b/net/sctp/output.c
@@ -365,7 +365,7 @@ int sctp_packet_transmit(struct sctp_packet *packet)
 	struct sctp_transport *tp = packet->transport;
 	struct sctp_association *asoc = tp->asoc;
 	struct sctphdr *sh;
-	__u32 crc32 = 0;
+	__be32 crc32 = __constant_cpu_to_be32(0);
 	struct sk_buff *nskb;
 	struct sctp_chunk *chunk, *tmp;
 	struct sock *sk;
@@ -538,7 +538,7 @@ int sctp_packet_transmit(struct sctp_packet *packet)
 	/* 3) Put the resultant value into the checksum field in the
 	 *    common header, and leave the rest of the bits unchanged.
 	 */
-	sh->checksum = htonl(crc32);
+	sh->checksum = crc32;
 
 	/* IP layer ECN support
 	 * From RFC 2481
-- 
cgit v1.2.3


From 49a72dfb8814c2d65bd9f8c9c6daf6395a1ec58d Mon Sep 17 00:00:00 2001
From: Adam Langley <agl@imperialviolet.org>
Date: Sat, 19 Jul 2008 00:01:42 -0700
Subject: tcp: Fix MD5 signatures for non-linear skbs

Currently, the MD5 code assumes that the SKBs are linear and, in the case
that they aren't, happily goes off and hashes off the end of the SKB and
into random memory.

Reported by Stephen Hemminger in [1]. Advice thanks to Stephen and Evgeniy
Polyakov. Also includes a couple of missed route_caps from Stephen's patch
in [2].

[1] http://marc.info/?l=linux-netdev&m=121445989106145&w=2
[2] http://marc.info/?l=linux-netdev&m=121459157816964&w=2

Signed-off-by: Adam Langley <agl@imperialviolet.org>
Acked-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/tcp.h     |  29 +++++------
 net/ipv4/tcp.c        | 127 ++++++++++++++++++++-------------------------
 net/ipv4/tcp_ipv4.c   | 140 ++++++++++++++++++++++++++++++--------------------
 net/ipv4/tcp_output.c |  14 ++---
 net/ipv6/tcp_ipv6.c   | 127 +++++++++++++++++++++++++++++----------------
 5 files changed, 242 insertions(+), 195 deletions(-)

(limited to 'include')

diff --git a/include/net/tcp.h b/include/net/tcp.h
index 92d7b551dc55..31f5bbfc59bc 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -1112,20 +1112,12 @@ struct tcp_md5sig_pool {
 #define TCP_MD5SIG_MAXKEYS	(~(u32)0)	/* really?! */
 
 /* - functions */
-extern int			tcp_calc_md5_hash(char *md5_hash,
-						  struct tcp_md5sig_key *key,
-						  int bplen,
-						  struct tcphdr *th,
-						  unsigned int tcplen,
-						  struct tcp_md5sig_pool *hp);
-
-extern int			tcp_v4_calc_md5_hash(char *md5_hash,
-						     struct tcp_md5sig_key *key,
-						     struct sock *sk,
-						     struct dst_entry *dst,
-						     struct request_sock *req,
-						     struct tcphdr *th,
-						     unsigned int tcplen);
+extern int			tcp_v4_md5_hash_skb(char *md5_hash,
+						    struct tcp_md5sig_key *key,
+						    struct sock *sk,
+						    struct request_sock *req,
+						    struct sk_buff *skb);
+
 extern struct tcp_md5sig_key	*tcp_v4_md5_lookup(struct sock *sk,
 						   struct sock *addr_sk);
 
@@ -1152,6 +1144,11 @@ extern void			tcp_free_md5sig_pool(void);
 
 extern struct tcp_md5sig_pool	*__tcp_get_md5sig_pool(int cpu);
 extern void			__tcp_put_md5sig_pool(void);
+extern int tcp_md5_hash_header(struct tcp_md5sig_pool *, struct tcphdr *);
+extern int tcp_md5_hash_skb_data(struct tcp_md5sig_pool *, struct sk_buff *,
+				 unsigned header_len);
+extern int tcp_md5_hash_key(struct tcp_md5sig_pool *hp,
+			    struct tcp_md5sig_key *key);
 
 static inline
 struct tcp_md5sig_pool		*tcp_get_md5sig_pool(void)
@@ -1381,10 +1378,8 @@ struct tcp_sock_af_ops {
 	int			(*calc_md5_hash) (char *location,
 						  struct tcp_md5sig_key *md5,
 						  struct sock *sk,
-						  struct dst_entry *dst,
 						  struct request_sock *req,
-						  struct tcphdr *th,
-						  unsigned int len);
+						  struct sk_buff *skb);
 	int			(*md5_add) (struct sock *sk,
 					    struct sock *addr_sk,
 					    u8 *newkey,
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index 827e6132af5f..0b491bf03db4 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -2465,76 +2465,6 @@ static unsigned long tcp_md5sig_users;
 static struct tcp_md5sig_pool **tcp_md5sig_pool;
 static DEFINE_SPINLOCK(tcp_md5sig_pool_lock);
 
-int tcp_calc_md5_hash(char *md5_hash, struct tcp_md5sig_key *key,
-		      int bplen,
-		      struct tcphdr *th, unsigned int tcplen,
-		      struct tcp_md5sig_pool *hp)
-{
-	struct scatterlist sg[4];
-	__u16 data_len;
-	int block = 0;
-	__sum16 cksum;
-	struct hash_desc *desc = &hp->md5_desc;
-	int err;
-	unsigned int nbytes = 0;
-
-	sg_init_table(sg, 4);
-
-	/* 1. The TCP pseudo-header */
-	sg_set_buf(&sg[block++], &hp->md5_blk, bplen);
-	nbytes += bplen;
-
-	/* 2. The TCP header, excluding options, and assuming a
-	 * checksum of zero
-	 */
-	cksum = th->check;
-	th->check = 0;
-	sg_set_buf(&sg[block++], th, sizeof(*th));
-	nbytes += sizeof(*th);
-
-	/* 3. The TCP segment data (if any) */
-	data_len = tcplen - (th->doff << 2);
-	if (data_len > 0) {
-		u8 *data = (u8 *)th + (th->doff << 2);
-		sg_set_buf(&sg[block++], data, data_len);
-		nbytes += data_len;
-	}
-
-	/* 4. an independently-specified key or password, known to both
-	 * TCPs and presumably connection-specific
-	 */
-	sg_set_buf(&sg[block++], key->key, key->keylen);
-	nbytes += key->keylen;
-
-	sg_mark_end(&sg[block - 1]);
-
-	/* Now store the hash into the packet */
-	err = crypto_hash_init(desc);
-	if (err) {
-		if (net_ratelimit())
-			printk(KERN_WARNING "%s(): hash_init failed\n", __func__);
-		return -1;
-	}
-	err = crypto_hash_update(desc, sg, nbytes);
-	if (err) {
-		if (net_ratelimit())
-			printk(KERN_WARNING "%s(): hash_update failed\n", __func__);
-		return -1;
-	}
-	err = crypto_hash_final(desc, md5_hash);
-	if (err) {
-		if (net_ratelimit())
-			printk(KERN_WARNING "%s(): hash_final failed\n", __func__);
-		return -1;
-	}
-
-	/* Reset header */
-	th->check = cksum;
-
-	return 0;
-}
-EXPORT_SYMBOL(tcp_calc_md5_hash);
-
 static void __tcp_free_md5sig_pool(struct tcp_md5sig_pool **pool)
 {
 	int cpu;
@@ -2658,6 +2588,63 @@ void __tcp_put_md5sig_pool(void)
 }
 
 EXPORT_SYMBOL(__tcp_put_md5sig_pool);
+
+int tcp_md5_hash_header(struct tcp_md5sig_pool *hp,
+			struct tcphdr *th)
+{
+	struct scatterlist sg;
+	int err;
+
+	__sum16 old_checksum = th->check;
+	th->check = 0;
+	/* options aren't included in the hash */
+	sg_init_one(&sg, th, sizeof(struct tcphdr));
+	err = crypto_hash_update(&hp->md5_desc, &sg, sizeof(struct tcphdr));
+	th->check = old_checksum;
+	return err;
+}
+
+EXPORT_SYMBOL(tcp_md5_hash_header);
+
+int tcp_md5_hash_skb_data(struct tcp_md5sig_pool *hp,
+			  struct sk_buff *skb, unsigned header_len)
+{
+	struct scatterlist sg;
+	const struct tcphdr *tp = tcp_hdr(skb);
+	struct hash_desc *desc = &hp->md5_desc;
+	unsigned i;
+	const unsigned head_data_len = skb_headlen(skb) > header_len ?
+				       skb_headlen(skb) - header_len : 0;
+	const struct skb_shared_info *shi = skb_shinfo(skb);
+
+	sg_init_table(&sg, 1);
+
+	sg_set_buf(&sg, ((u8 *) tp) + header_len, head_data_len);
+	if (crypto_hash_update(desc, &sg, head_data_len))
+		return 1;
+
+	for (i = 0; i < shi->nr_frags; ++i) {
+		const struct skb_frag_struct *f = &shi->frags[i];
+		sg_set_page(&sg, f->page, f->size, f->page_offset);
+		if (crypto_hash_update(desc, &sg, f->size))
+			return 1;
+	}
+
+	return 0;
+}
+
+EXPORT_SYMBOL(tcp_md5_hash_skb_data);
+
+int tcp_md5_hash_key(struct tcp_md5sig_pool *hp, struct tcp_md5sig_key *key)
+{
+	struct scatterlist sg;
+
+	sg_init_one(&sg, key->key, key->keylen);
+	return crypto_hash_update(&hp->md5_desc, &sg, key->keylen);
+}
+
+EXPORT_SYMBOL(tcp_md5_hash_key);
+
 #endif
 
 void tcp_done(struct sock *sk)
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 29adc668ad51..5400d75ff17a 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -87,9 +87,8 @@ int sysctl_tcp_low_latency __read_mostly;
 #ifdef CONFIG_TCP_MD5SIG
 static struct tcp_md5sig_key *tcp_v4_md5_do_lookup(struct sock *sk,
 						   __be32 addr);
-static int tcp_v4_do_calc_md5_hash(char *md5_hash, struct tcp_md5sig_key *key,
-				   __be32 saddr, __be32 daddr,
-				   struct tcphdr *th, unsigned int tcplen);
+static int tcp_v4_md5_hash_hdr(char *md5_hash, struct tcp_md5sig_key *key,
+			       __be32 daddr, __be32 saddr, struct tcphdr *th);
 #else
 static inline
 struct tcp_md5sig_key *tcp_v4_md5_do_lookup(struct sock *sk, __be32 addr)
@@ -583,11 +582,9 @@ static void tcp_v4_send_reset(struct sock *sk, struct sk_buff *skb)
 		arg.iov[0].iov_len += TCPOLEN_MD5SIG_ALIGNED;
 		rep.th.doff = arg.iov[0].iov_len / 4;
 
-		tcp_v4_do_calc_md5_hash((__u8 *)&rep.opt[1],
-					key,
-					ip_hdr(skb)->daddr,
-					ip_hdr(skb)->saddr,
-					&rep.th, arg.iov[0].iov_len);
+		tcp_v4_md5_hash_hdr((__u8 *) &rep.opt[1],
+				     key, ip_hdr(skb)->daddr,
+				     ip_hdr(skb)->saddr, &rep.th);
 	}
 #endif
 	arg.csum = csum_tcpudp_nofold(ip_hdr(skb)->daddr,
@@ -657,11 +654,9 @@ static void tcp_v4_send_ack(struct sk_buff *skb, u32 seq, u32 ack,
 		arg.iov[0].iov_len += TCPOLEN_MD5SIG_ALIGNED;
 		rep.th.doff = arg.iov[0].iov_len/4;
 
-		tcp_v4_do_calc_md5_hash((__u8 *)&rep.opt[offset],
-					key,
-					ip_hdr(skb)->daddr,
-					ip_hdr(skb)->saddr,
-					&rep.th, arg.iov[0].iov_len);
+		tcp_v4_md5_hash_hdr((__u8 *) &rep.opt[offset],
+				    key, ip_hdr(skb)->daddr,
+				    ip_hdr(skb)->saddr, &rep.th);
 	}
 #endif
 	arg.csum = csum_tcpudp_nofold(ip_hdr(skb)->daddr,
@@ -989,28 +984,16 @@ static int tcp_v4_parse_md5_keys(struct sock *sk, char __user *optval,
 				 newkey, cmd.tcpm_keylen);
 }
 
-static int tcp_v4_do_calc_md5_hash(char *md5_hash, struct tcp_md5sig_key *key,
-				   __be32 saddr, __be32 daddr,
-				   struct tcphdr *th,
-				   unsigned int tcplen)
+static int tcp_v4_md5_hash_pseudoheader(struct tcp_md5sig_pool *hp,
+					__be32 daddr, __be32 saddr, int nbytes)
 {
-	struct tcp_md5sig_pool *hp;
 	struct tcp4_pseudohdr *bp;
-	int err;
-
-	/*
-	 * Okay, so RFC2385 is turned on for this connection,
-	 * so we need to generate the MD5 hash for the packet now.
-	 */
-
-	hp = tcp_get_md5sig_pool();
-	if (!hp)
-		goto clear_hash_noput;
+	struct scatterlist sg;
 
 	bp = &hp->md5_blk.ip4;
 
 	/*
-	 * The TCP pseudo-header (in the order: source IP address,
+	 * 1. the TCP pseudo-header (in the order: source IP address,
 	 * destination IP address, zero-padded protocol number, and
 	 * segment length)
 	 */
@@ -1018,48 +1001,95 @@ static int tcp_v4_do_calc_md5_hash(char *md5_hash, struct tcp_md5sig_key *key,
 	bp->daddr = daddr;
 	bp->pad = 0;
 	bp->protocol = IPPROTO_TCP;
-	bp->len = htons(tcplen);
+	bp->len = cpu_to_be16(nbytes);
 
-	err = tcp_calc_md5_hash(md5_hash, key, sizeof(*bp),
-				th, tcplen, hp);
-	if (err)
+	sg_init_one(&sg, bp, sizeof(*bp));
+	return crypto_hash_update(&hp->md5_desc, &sg, sizeof(*bp));
+}
+
+static int tcp_v4_md5_hash_hdr(char *md5_hash, struct tcp_md5sig_key *key,
+			       __be32 daddr, __be32 saddr, struct tcphdr *th)
+{
+	struct tcp_md5sig_pool *hp;
+	struct hash_desc *desc;
+
+	hp = tcp_get_md5sig_pool();
+	if (!hp)
+		goto clear_hash_noput;
+	desc = &hp->md5_desc;
+
+	if (crypto_hash_init(desc))
+		goto clear_hash;
+	if (tcp_v4_md5_hash_pseudoheader(hp, daddr, saddr, th->doff << 2))
+		goto clear_hash;
+	if (tcp_md5_hash_header(hp, th))
+		goto clear_hash;
+	if (tcp_md5_hash_key(hp, key))
+		goto clear_hash;
+	if (crypto_hash_final(desc, md5_hash))
 		goto clear_hash;
 
-	/* Free up the crypto pool */
 	tcp_put_md5sig_pool();
-out:
 	return 0;
+
 clear_hash:
 	tcp_put_md5sig_pool();
 clear_hash_noput:
 	memset(md5_hash, 0, 16);
-	goto out;
+	return 1;
 }
 
-int tcp_v4_calc_md5_hash(char *md5_hash, struct tcp_md5sig_key *key,
-			 struct sock *sk,
-			 struct dst_entry *dst,
-			 struct request_sock *req,
-			 struct tcphdr *th,
-			 unsigned int tcplen)
+int tcp_v4_md5_hash_skb(char *md5_hash, struct tcp_md5sig_key *key,
+			struct sock *sk, struct request_sock *req,
+			struct sk_buff *skb)
 {
+	struct tcp_md5sig_pool *hp;
+	struct hash_desc *desc;
+	struct tcphdr *th = tcp_hdr(skb);
 	__be32 saddr, daddr;
 
 	if (sk) {
 		saddr = inet_sk(sk)->saddr;
 		daddr = inet_sk(sk)->daddr;
+	} else if (req) {
+		saddr = inet_rsk(req)->loc_addr;
+		daddr = inet_rsk(req)->rmt_addr;
 	} else {
-		struct rtable *rt = (struct rtable *)dst;
-		BUG_ON(!rt);
-		saddr = rt->rt_src;
-		daddr = rt->rt_dst;
+		const struct iphdr *iph = ip_hdr(skb);
+		saddr = iph->saddr;
+		daddr = iph->daddr;
 	}
-	return tcp_v4_do_calc_md5_hash(md5_hash, key,
-				       saddr, daddr,
-				       th, tcplen);
+
+	hp = tcp_get_md5sig_pool();
+	if (!hp)
+		goto clear_hash_noput;
+	desc = &hp->md5_desc;
+
+	if (crypto_hash_init(desc))
+		goto clear_hash;
+
+	if (tcp_v4_md5_hash_pseudoheader(hp, daddr, saddr, skb->len))
+		goto clear_hash;
+	if (tcp_md5_hash_header(hp, th))
+		goto clear_hash;
+	if (tcp_md5_hash_skb_data(hp, skb, th->doff << 2))
+		goto clear_hash;
+	if (tcp_md5_hash_key(hp, key))
+		goto clear_hash;
+	if (crypto_hash_final(desc, md5_hash))
+		goto clear_hash;
+
+	tcp_put_md5sig_pool();
+	return 0;
+
+clear_hash:
+	tcp_put_md5sig_pool();
+clear_hash_noput:
+	memset(md5_hash, 0, 16);
+	return 1;
 }
 
-EXPORT_SYMBOL(tcp_v4_calc_md5_hash);
+EXPORT_SYMBOL(tcp_v4_md5_hash_skb);
 
 static int tcp_v4_inbound_md5_hash(struct sock *sk, struct sk_buff *skb)
 {
@@ -1104,10 +1134,9 @@ static int tcp_v4_inbound_md5_hash(struct sock *sk, struct sk_buff *skb)
 	/* Okay, so this is hash_expected and hash_location -
 	 * so we need to calculate the checksum.
 	 */
-	genhash = tcp_v4_do_calc_md5_hash(newhash,
-					  hash_expected,
-					  iph->saddr, iph->daddr,
-					  th, skb->len);
+	genhash = tcp_v4_md5_hash_skb(newhash,
+				      hash_expected,
+				      NULL, NULL, skb);
 
 	if (genhash || memcmp(hash_location, newhash, 16) != 0) {
 		if (net_ratelimit()) {
@@ -1356,6 +1385,7 @@ struct sock *tcp_v4_syn_recv_sock(struct sock *sk, struct sk_buff *skb,
 		if (newkey != NULL)
 			tcp_v4_md5_do_add(newsk, inet_sk(sk)->daddr,
 					  newkey, key->keylen);
+		newsk->sk_route_caps &= ~NETIF_F_GSO_MASK;
 	}
 #endif
 
@@ -1719,7 +1749,7 @@ struct inet_connection_sock_af_ops ipv4_specific = {
 #ifdef CONFIG_TCP_MD5SIG
 static struct tcp_sock_af_ops tcp_sock_ipv4_specific = {
 	.md5_lookup		= tcp_v4_md5_lookup,
-	.calc_md5_hash		= tcp_v4_calc_md5_hash,
+	.calc_md5_hash		= tcp_v4_md5_hash_skb,
 	.md5_add		= tcp_v4_md5_add_func,
 	.md5_parse		= tcp_v4_parse_md5_keys,
 };
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index 36a19707f67f..958ff486165f 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -540,8 +540,10 @@ static int tcp_transmit_skb(struct sock *sk, struct sk_buff *skb, int clone_it,
 	 * room for it.
 	 */
 	md5 = tp->af_specific->md5_lookup(sk, sk);
-	if (md5)
+	if (md5) {
 		tcp_header_size += TCPOLEN_MD5SIG_ALIGNED;
+		sk->sk_route_caps &= ~NETIF_F_GSO_MASK;
+	}
 #endif
 
 	skb_push(skb, tcp_header_size);
@@ -602,10 +604,7 @@ static int tcp_transmit_skb(struct sock *sk, struct sk_buff *skb, int clone_it,
 	/* Calculate the MD5 hash, as we have all we need now */
 	if (md5) {
 		tp->af_specific->calc_md5_hash(md5_hash_location,
-					       md5,
-					       sk, NULL, NULL,
-					       tcp_hdr(skb),
-					       skb->len);
+					       md5, sk, NULL, skb);
 	}
 #endif
 
@@ -2264,10 +2263,7 @@ struct sk_buff *tcp_make_synack(struct sock *sk, struct dst_entry *dst,
 	/* Okay, we have all we need - do the md5 hash if needed */
 	if (md5) {
 		tp->af_specific->calc_md5_hash(md5_hash_location,
-					       md5,
-					       NULL, dst, req,
-					       tcp_hdr(skb),
-					       skb->len);
+					       md5, NULL, req, skb);
 	}
 #endif
 
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index ca5b93a5c02a..ae45f9835014 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -736,64 +736,105 @@ static int tcp_v6_parse_md5_keys (struct sock *sk, char __user *optval,
 	return tcp_v6_md5_do_add(sk, &sin6->sin6_addr, newkey, cmd.tcpm_keylen);
 }
 
-static int tcp_v6_do_calc_md5_hash(char *md5_hash, struct tcp_md5sig_key *key,
-				   struct in6_addr *saddr,
-				   struct in6_addr *daddr,
-				   struct tcphdr *th, unsigned int tcplen)
+static int tcp_v6_md5_hash_pseudoheader(struct tcp_md5sig_pool *hp,
+					struct in6_addr *daddr,
+					struct in6_addr *saddr, int nbytes)
 {
-	struct tcp_md5sig_pool *hp;
 	struct tcp6_pseudohdr *bp;
-	int err;
-
-	hp = tcp_get_md5sig_pool();
-	if (!hp) {
-		printk(KERN_WARNING "%s(): hash pool not found...\n", __func__);
-		goto clear_hash_noput;
-	}
+	struct scatterlist sg;
 
 	bp = &hp->md5_blk.ip6;
-
 	/* 1. TCP pseudo-header (RFC2460) */
 	ipv6_addr_copy(&bp->saddr, saddr);
 	ipv6_addr_copy(&bp->daddr, daddr);
-	bp->len = htonl(tcplen);
-	bp->protocol = htonl(IPPROTO_TCP);
+	bp->protocol = cpu_to_be32(IPPROTO_TCP);
+	bp->len = cpu_to_be16(nbytes);
 
-	err = tcp_calc_md5_hash(md5_hash, key, sizeof(*bp),
-				th, tcplen, hp);
+	sg_init_one(&sg, bp, sizeof(*bp));
+	return crypto_hash_update(&hp->md5_desc, &sg, sizeof(*bp));
+}
 
-	if (err)
+static int tcp_v6_md5_hash_hdr(char *md5_hash, struct tcp_md5sig_key *key,
+			       struct in6_addr *daddr, struct in6_addr *saddr,
+			       struct tcphdr *th)
+{
+	struct tcp_md5sig_pool *hp;
+	struct hash_desc *desc;
+
+	hp = tcp_get_md5sig_pool();
+	if (!hp)
+		goto clear_hash_noput;
+	desc = &hp->md5_desc;
+
+	if (crypto_hash_init(desc))
+		goto clear_hash;
+	if (tcp_v6_md5_hash_pseudoheader(hp, daddr, saddr, th->doff << 2))
+		goto clear_hash;
+	if (tcp_md5_hash_header(hp, th))
+		goto clear_hash;
+	if (tcp_md5_hash_key(hp, key))
+		goto clear_hash;
+	if (crypto_hash_final(desc, md5_hash))
 		goto clear_hash;
 
-	/* Free up the crypto pool */
 	tcp_put_md5sig_pool();
-out:
 	return 0;
+
 clear_hash:
 	tcp_put_md5sig_pool();
 clear_hash_noput:
 	memset(md5_hash, 0, 16);
-	goto out;
+	return 1;
 }
 
-static int tcp_v6_calc_md5_hash(char *md5_hash, struct tcp_md5sig_key *key,
-				struct sock *sk,
-				struct dst_entry *dst,
-				struct request_sock *req,
-				struct tcphdr *th, unsigned int tcplen)
+static int tcp_v6_md5_hash_skb(char *md5_hash, struct tcp_md5sig_key *key,
+			       struct sock *sk, struct request_sock *req,
+			       struct sk_buff *skb)
 {
 	struct in6_addr *saddr, *daddr;
+	struct tcp_md5sig_pool *hp;
+	struct hash_desc *desc;
+	struct tcphdr *th = tcp_hdr(skb);
 
 	if (sk) {
 		saddr = &inet6_sk(sk)->saddr;
 		daddr = &inet6_sk(sk)->daddr;
-	} else {
+	} else if (req) {
 		saddr = &inet6_rsk(req)->loc_addr;
 		daddr = &inet6_rsk(req)->rmt_addr;
+	} else {
+		struct ipv6hdr *ip6h = ipv6_hdr(skb);
+		saddr = &ip6h->saddr;
+		daddr = &ip6h->daddr;
 	}
-	return tcp_v6_do_calc_md5_hash(md5_hash, key,
-				       saddr, daddr,
-				       th, tcplen);
+
+	hp = tcp_get_md5sig_pool();
+	if (!hp)
+		goto clear_hash_noput;
+	desc = &hp->md5_desc;
+
+	if (crypto_hash_init(desc))
+		goto clear_hash;
+
+	if (tcp_v6_md5_hash_pseudoheader(hp, daddr, saddr, skb->len))
+		goto clear_hash;
+	if (tcp_md5_hash_header(hp, th))
+		goto clear_hash;
+	if (tcp_md5_hash_skb_data(hp, skb, th->doff << 2))
+		goto clear_hash;
+	if (tcp_md5_hash_key(hp, key))
+		goto clear_hash;
+	if (crypto_hash_final(desc, md5_hash))
+		goto clear_hash;
+
+	tcp_put_md5sig_pool();
+	return 0;
+
+clear_hash:
+	tcp_put_md5sig_pool();
+clear_hash_noput:
+	memset(md5_hash, 0, 16);
+	return 1;
 }
 
 static int tcp_v6_inbound_md5_hash (struct sock *sk, struct sk_buff *skb)
@@ -834,10 +875,10 @@ static int tcp_v6_inbound_md5_hash (struct sock *sk, struct sk_buff *skb)
 	}
 
 	/* check the signature */
-	genhash = tcp_v6_do_calc_md5_hash(newhash,
-					  hash_expected,
-					  &ip6h->saddr, &ip6h->daddr,
-					  th, skb->len);
+	genhash = tcp_v6_md5_hash_skb(newhash,
+				      hash_expected,
+				      NULL, NULL, skb);
+
 	if (genhash || memcmp(hash_location, newhash, 16) != 0) {
 		if (net_ratelimit()) {
 			printk(KERN_INFO "MD5 Hash %s for "
@@ -974,10 +1015,9 @@ static void tcp_v6_send_reset(struct sock *sk, struct sk_buff *skb)
 			       (TCPOPT_NOP << 16) |
 			       (TCPOPT_MD5SIG << 8) |
 			       TCPOLEN_MD5SIG);
-		tcp_v6_do_calc_md5_hash((__u8 *)&opt[1], key,
-					&ipv6_hdr(skb)->daddr,
-					&ipv6_hdr(skb)->saddr,
-					t1, tot_len);
+		tcp_v6_md5_hash_hdr((__u8 *)&opt[1], key,
+				    &ipv6_hdr(skb)->daddr,
+				    &ipv6_hdr(skb)->saddr, t1);
 	}
 #endif
 
@@ -1064,10 +1104,9 @@ static void tcp_v6_send_ack(struct sk_buff *skb, u32 seq, u32 ack, u32 win, u32
 	if (key) {
 		*topt++ = htonl((TCPOPT_NOP << 24) | (TCPOPT_NOP << 16) |
 				(TCPOPT_MD5SIG << 8) | TCPOLEN_MD5SIG);
-		tcp_v6_do_calc_md5_hash((__u8 *)topt, key,
-					&ipv6_hdr(skb)->daddr,
-					&ipv6_hdr(skb)->saddr,
-					t1, tot_len);
+		tcp_v6_md5_hash_hdr((__u8 *)topt, key,
+				    &ipv6_hdr(skb)->daddr,
+				    &ipv6_hdr(skb)->saddr, t1);
 	}
 #endif
 
@@ -1783,7 +1822,7 @@ static struct inet_connection_sock_af_ops ipv6_specific = {
 #ifdef CONFIG_TCP_MD5SIG
 static struct tcp_sock_af_ops tcp_sock_ipv6_specific = {
 	.md5_lookup	=	tcp_v6_md5_lookup,
-	.calc_md5_hash	=	tcp_v6_calc_md5_hash,
+	.calc_md5_hash	=	tcp_v6_md5_hash_skb,
 	.md5_add	=	tcp_v6_md5_add_func,
 	.md5_parse	=	tcp_v6_parse_md5_keys,
 };
@@ -1815,7 +1854,7 @@ static struct inet_connection_sock_af_ops ipv6_mapped = {
 #ifdef CONFIG_TCP_MD5SIG
 static struct tcp_sock_af_ops tcp_sock_ipv6_mapped_specific = {
 	.md5_lookup	=	tcp_v4_md5_lookup,
-	.calc_md5_hash	=	tcp_v4_calc_md5_hash,
+	.calc_md5_hash	=	tcp_v4_md5_hash_skb,
 	.md5_add	=	tcp_v6_md5_add_func,
 	.md5_parse	=	tcp_v6_parse_md5_keys,
 };
-- 
cgit v1.2.3


From 33ad798c924b4a1afad3593f2796d465040aadd5 Mon Sep 17 00:00:00 2001
From: Adam Langley <agl@imperialviolet.org>
Date: Sat, 19 Jul 2008 00:04:31 -0700
Subject: tcp: options clean up

This should fix the following bugs:
  * Connections with MD5 signatures produce invalid packets whenever SACK
    options are included
  * MD5 signatures are counted twice in the MSS calculations

Behaviour changes:
  * A SYN with MD5 + SACK + TS elicits a SYNACK with MD5 + SACK

    This is because we can't fit any SACK blocks in a packet with MD5 + TS
    options. There was discussion about disabling SACK rather than TS in
    order to fit in better with old, buggy kernels, but that was deemed to
    be unnecessary.

  * SYNs with MD5 don't include a TS option

    See above.

Additionally, it removes a bunch of duplicated logic for calculating options,
which should help avoid these sort of issues in the future.

Signed-off-by: Adam Langley <agl@imperialviolet.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/tcp.h     |   2 +
 net/ipv4/tcp_output.c | 432 +++++++++++++++++++++++++++-----------------------
 2 files changed, 238 insertions(+), 196 deletions(-)

(limited to 'include')

diff --git a/include/net/tcp.h b/include/net/tcp.h
index 31f5bbfc59bc..8983386356a5 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -50,6 +50,7 @@ extern atomic_t tcp_orphan_count;
 extern void tcp_time_wait(struct sock *sk, int state, int timeo);
 
 #define MAX_TCP_HEADER	(128 + MAX_HEADER)
+#define MAX_TCP_OPTION_SPACE 40
 
 /* 
  * Never offer a window over 32767 without using window scaling. Some
@@ -184,6 +185,7 @@ extern void tcp_time_wait(struct sock *sk, int state, int timeo);
 #define TCPOLEN_SACK_BASE_ALIGNED	4
 #define TCPOLEN_SACK_PERBLOCK		8
 #define TCPOLEN_MD5SIG_ALIGNED		20
+#define TCPOLEN_MSS_ALIGNED		4
 
 /* Flags in tp->nonagle */
 #define TCP_NAGLE_OFF		1	/* Nagle's algo is disabled */
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index 958ff486165f..1fa683c0ba9b 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -345,28 +345,82 @@ static void tcp_init_nondata_skb(struct sk_buff *skb, u32 seq, u8 flags)
 	TCP_SKB_CB(skb)->end_seq = seq;
 }
 
-static void tcp_build_and_update_options(__be32 *ptr, struct tcp_sock *tp,
-					 __u32 tstamp, __u8 **md5_hash)
-{
-	if (tp->rx_opt.tstamp_ok) {
+#define OPTION_SACK_ADVERTISE	(1 << 0)
+#define OPTION_TS		(1 << 1)
+#define OPTION_MD5		(1 << 2)
+
+struct tcp_out_options {
+	u8 options;		/* bit field of OPTION_* */
+	u8 ws;			/* window scale, 0 to disable */
+	u8 num_sack_blocks;	/* number of SACK blocks to include */
+	u16 mss;		/* 0 to disable */
+	__u32 tsval, tsecr;	/* need to include OPTION_TS */
+};
+
+static void tcp_options_write(__be32 *ptr, struct tcp_sock *tp,
+			      const struct tcp_out_options *opts,
+			      __u8 **md5_hash) {
+	if (unlikely(OPTION_MD5 & opts->options)) {
 		*ptr++ = htonl((TCPOPT_NOP << 24) |
 			       (TCPOPT_NOP << 16) |
-			       (TCPOPT_TIMESTAMP << 8) |
-			       TCPOLEN_TIMESTAMP);
-		*ptr++ = htonl(tstamp);
-		*ptr++ = htonl(tp->rx_opt.ts_recent);
+			       (TCPOPT_MD5SIG << 8) |
+			       TCPOLEN_MD5SIG);
+		*md5_hash = (__u8 *)ptr;
+		ptr += 4;
+	} else {
+		*md5_hash = NULL;
 	}
-	if (tp->rx_opt.eff_sacks) {
-		struct tcp_sack_block *sp = tp->rx_opt.dsack ? tp->duplicate_sack : tp->selective_acks;
+
+	if (likely(OPTION_TS & opts->options)) {
+		if (unlikely(OPTION_SACK_ADVERTISE & opts->options)) {
+			*ptr++ = htonl((TCPOPT_SACK_PERM << 24) |
+				       (TCPOLEN_SACK_PERM << 16) |
+				       (TCPOPT_TIMESTAMP << 8) |
+				       TCPOLEN_TIMESTAMP);
+		} else {
+			*ptr++ = htonl((TCPOPT_NOP << 24) |
+				       (TCPOPT_NOP << 16) |
+				       (TCPOPT_TIMESTAMP << 8) |
+				       TCPOLEN_TIMESTAMP);
+		}
+		*ptr++ = htonl(opts->tsval);
+		*ptr++ = htonl(opts->tsecr);
+	}
+
+	if (unlikely(opts->mss)) {
+		*ptr++ = htonl((TCPOPT_MSS << 24) |
+			       (TCPOLEN_MSS << 16) |
+			       opts->mss);
+	}
+
+	if (unlikely(OPTION_SACK_ADVERTISE & opts->options &&
+		     !(OPTION_TS & opts->options))) {
+		*ptr++ = htonl((TCPOPT_NOP << 24) |
+			       (TCPOPT_NOP << 16) |
+			       (TCPOPT_SACK_PERM << 8) |
+			       TCPOLEN_SACK_PERM);
+	}
+
+	if (unlikely(opts->ws)) {
+		*ptr++ = htonl((TCPOPT_NOP << 24) |
+			       (TCPOPT_WINDOW << 16) |
+			       (TCPOLEN_WINDOW << 8) |
+			       opts->ws);
+	}
+
+	if (unlikely(opts->num_sack_blocks)) {
+		struct tcp_sack_block *sp = tp->rx_opt.dsack ?
+			tp->duplicate_sack : tp->selective_acks;
 		int this_sack;
 
 		*ptr++ = htonl((TCPOPT_NOP  << 24) |
 			       (TCPOPT_NOP  << 16) |
 			       (TCPOPT_SACK <<  8) |
-			       (TCPOLEN_SACK_BASE + (tp->rx_opt.eff_sacks *
+			       (TCPOLEN_SACK_BASE + (opts->num_sack_blocks *
 						     TCPOLEN_SACK_PERBLOCK)));
 
-		for (this_sack = 0; this_sack < tp->rx_opt.eff_sacks; this_sack++) {
+		for (this_sack = 0; this_sack < opts->num_sack_blocks;
+		     ++this_sack) {
 			*ptr++ = htonl(sp[this_sack].start_seq);
 			*ptr++ = htonl(sp[this_sack].end_seq);
 		}
@@ -376,81 +430,137 @@ static void tcp_build_and_update_options(__be32 *ptr, struct tcp_sock *tp,
 			tp->rx_opt.eff_sacks--;
 		}
 	}
+}
+
+static unsigned tcp_syn_options(struct sock *sk, struct sk_buff *skb,
+				struct tcp_out_options *opts,
+				struct tcp_md5sig_key **md5) {
+	struct tcp_sock *tp = tcp_sk(sk);
+	unsigned size = 0;
+
 #ifdef CONFIG_TCP_MD5SIG
-	if (md5_hash) {
-		*ptr++ = htonl((TCPOPT_NOP << 24) |
-			       (TCPOPT_NOP << 16) |
-			       (TCPOPT_MD5SIG << 8) |
-			       TCPOLEN_MD5SIG);
-		*md5_hash = (__u8 *)ptr;
+	*md5 = tp->af_specific->md5_lookup(sk, sk);
+	if (*md5) {
+		opts->options |= OPTION_MD5;
+		size += TCPOLEN_MD5SIG_ALIGNED;
 	}
+#else
+	*md5 = NULL;
 #endif
+
+	/* We always get an MSS option.  The option bytes which will be seen in
+	 * normal data packets should timestamps be used, must be in the MSS
+	 * advertised.  But we subtract them from tp->mss_cache so that
+	 * calculations in tcp_sendmsg are simpler etc.  So account for this
+	 * fact here if necessary.  If we don't do this correctly, as a
+	 * receiver we won't recognize data packets as being full sized when we
+	 * should, and thus we won't abide by the delayed ACK rules correctly.
+	 * SACKs don't matter, we never delay an ACK when we have any of those
+	 * going out.  */
+	opts->mss = tcp_advertise_mss(sk);
+	size += TCPOLEN_MSS_ALIGNED;
+
+	if (likely(sysctl_tcp_timestamps && *md5 == NULL)) {
+		opts->options |= OPTION_TS;
+		opts->tsval = TCP_SKB_CB(skb)->when;
+		opts->tsecr = tp->rx_opt.ts_recent;
+		size += TCPOLEN_TSTAMP_ALIGNED;
+	}
+	if (likely(sysctl_tcp_window_scaling)) {
+		opts->ws = tp->rx_opt.rcv_wscale;
+		size += TCPOLEN_WSCALE_ALIGNED;
+	}
+	if (likely(sysctl_tcp_sack)) {
+		opts->options |= OPTION_SACK_ADVERTISE;
+		if (unlikely(!OPTION_TS & opts->options))
+			size += TCPOLEN_SACKPERM_ALIGNED;
+	}
+
+	return size;
 }
 
-/* Construct a tcp options header for a SYN or SYN_ACK packet.
- * If this is every changed make sure to change the definition of
- * MAX_SYN_SIZE to match the new maximum number of options that you
- * can generate.
- *
- * Note - that with the RFC2385 TCP option, we make room for the
- * 16 byte MD5 hash. This will be filled in later, so the pointer for the
- * location to be filled is passed back up.
- */
-static void tcp_syn_build_options(__be32 *ptr, int mss, int ts, int sack,
-				  int offer_wscale, int wscale, __u32 tstamp,
-				  __u32 ts_recent, __u8 **md5_hash)
-{
-	/* We always get an MSS option.
-	 * The option bytes which will be seen in normal data
-	 * packets should timestamps be used, must be in the MSS
-	 * advertised.  But we subtract them from tp->mss_cache so
-	 * that calculations in tcp_sendmsg are simpler etc.
-	 * So account for this fact here if necessary.  If we
-	 * don't do this correctly, as a receiver we won't
-	 * recognize data packets as being full sized when we
-	 * should, and thus we won't abide by the delayed ACK
-	 * rules correctly.
-	 * SACKs don't matter, we never delay an ACK when we
-	 * have any of those going out.
-	 */
-	*ptr++ = htonl((TCPOPT_MSS << 24) | (TCPOLEN_MSS << 16) | mss);
-	if (ts) {
-		if (sack)
-			*ptr++ = htonl((TCPOPT_SACK_PERM << 24) |
-				       (TCPOLEN_SACK_PERM << 16) |
-				       (TCPOPT_TIMESTAMP << 8) |
-				       TCPOLEN_TIMESTAMP);
-		else
-			*ptr++ = htonl((TCPOPT_NOP << 24) |
-				       (TCPOPT_NOP << 16) |
-				       (TCPOPT_TIMESTAMP << 8) |
-				       TCPOLEN_TIMESTAMP);
-		*ptr++ = htonl(tstamp);		/* TSVAL */
-		*ptr++ = htonl(ts_recent);	/* TSECR */
-	} else if (sack)
-		*ptr++ = htonl((TCPOPT_NOP << 24) |
-			       (TCPOPT_NOP << 16) |
-			       (TCPOPT_SACK_PERM << 8) |
-			       TCPOLEN_SACK_PERM);
-	if (offer_wscale)
-		*ptr++ = htonl((TCPOPT_NOP << 24) |
-			       (TCPOPT_WINDOW << 16) |
-			       (TCPOLEN_WINDOW << 8) |
-			       (wscale));
+static unsigned tcp_synack_options(struct sock *sk,
+				   struct request_sock *req,
+				   unsigned mss, struct sk_buff *skb,
+				   struct tcp_out_options *opts,
+				   struct tcp_md5sig_key **md5) {
+	unsigned size = 0;
+	struct inet_request_sock *ireq = inet_rsk(req);
+	char doing_ts;
+
 #ifdef CONFIG_TCP_MD5SIG
-	/*
-	 * If MD5 is enabled, then we set the option, and include the size
-	 * (always 18). The actual MD5 hash is added just before the
-	 * packet is sent.
-	 */
-	if (md5_hash) {
-		*ptr++ = htonl((TCPOPT_NOP << 24) |
-			       (TCPOPT_NOP << 16) |
-			       (TCPOPT_MD5SIG << 8) |
-			       TCPOLEN_MD5SIG);
-		*md5_hash = (__u8 *)ptr;
+	*md5 = tcp_rsk(req)->af_specific->md5_lookup(sk, req);
+	if (*md5) {
+		opts->options |= OPTION_MD5;
+		size += TCPOLEN_MD5SIG_ALIGNED;
 	}
+#else
+	*md5 = NULL;
 #endif
+
+	/* we can't fit any SACK blocks in a packet with MD5 + TS
+	   options. There was discussion about disabling SACK rather than TS in
+	   order to fit in better with old, buggy kernels, but that was deemed
+	   to be unnecessary. */
+	doing_ts = ireq->tstamp_ok && !(*md5 && ireq->sack_ok);
+
+	opts->mss = mss;
+	size += TCPOLEN_MSS_ALIGNED;
+
+	if (likely(ireq->wscale_ok)) {
+		opts->ws = ireq->rcv_wscale;
+		size += TCPOLEN_WSCALE_ALIGNED;
+	}
+	if (likely(doing_ts)) {
+		opts->options |= OPTION_TS;
+		opts->tsval = TCP_SKB_CB(skb)->when;
+		opts->tsecr = req->ts_recent;
+		size += TCPOLEN_TSTAMP_ALIGNED;
+	}
+	if (likely(ireq->sack_ok)) {
+		opts->options |= OPTION_SACK_ADVERTISE;
+		if (unlikely(!doing_ts))
+			size += TCPOLEN_SACKPERM_ALIGNED;
+	}
+
+	return size;
+}
+
+static unsigned tcp_established_options(struct sock *sk, struct sk_buff *skb,
+					struct tcp_out_options *opts,
+					struct tcp_md5sig_key **md5) {
+	struct tcp_skb_cb *tcb = skb ? TCP_SKB_CB(skb) : NULL;
+	struct tcp_sock *tp = tcp_sk(sk);
+	unsigned size = 0;
+
+#ifdef CONFIG_TCP_MD5SIG
+	*md5 = tp->af_specific->md5_lookup(sk, sk);
+	if (unlikely(*md5)) {
+		opts->options |= OPTION_MD5;
+		size += TCPOLEN_MD5SIG_ALIGNED;
+	}
+#else
+	*md5 = NULL;
+#endif
+
+	if (likely(tp->rx_opt.tstamp_ok)) {
+		opts->options |= OPTION_TS;
+		opts->tsval = tcb ? tcb->when : 0;
+		opts->tsecr = tp->rx_opt.ts_recent;
+		size += TCPOLEN_TSTAMP_ALIGNED;
+	}
+
+	if (unlikely(tp->rx_opt.eff_sacks)) {
+		const unsigned remaining = MAX_TCP_OPTION_SPACE - size;
+		opts->num_sack_blocks =
+			min_t(unsigned, tp->rx_opt.eff_sacks,
+			      (remaining - TCPOLEN_SACK_BASE_ALIGNED) /
+			      TCPOLEN_SACK_PERBLOCK);
+		size += TCPOLEN_SACK_BASE_ALIGNED +
+			opts->num_sack_blocks * TCPOLEN_SACK_PERBLOCK;
+	}
+
+	return size;
 }
 
 /* This routine actually transmits TCP packets queued in by
@@ -471,13 +581,11 @@ static int tcp_transmit_skb(struct sock *sk, struct sk_buff *skb, int clone_it,
 	struct inet_sock *inet;
 	struct tcp_sock *tp;
 	struct tcp_skb_cb *tcb;
-	int tcp_header_size;
-#ifdef CONFIG_TCP_MD5SIG
+	struct tcp_out_options opts;
+	unsigned tcp_options_size, tcp_header_size;
 	struct tcp_md5sig_key *md5;
 	__u8 *md5_hash_location;
-#endif
 	struct tcphdr *th;
-	int sysctl_flags;
 	int err;
 
 	BUG_ON(!skb || !tcp_skb_pcount(skb));
@@ -500,52 +608,18 @@ static int tcp_transmit_skb(struct sock *sk, struct sk_buff *skb, int clone_it,
 	inet = inet_sk(sk);
 	tp = tcp_sk(sk);
 	tcb = TCP_SKB_CB(skb);
-	tcp_header_size = tp->tcp_header_len;
-
-#define SYSCTL_FLAG_TSTAMPS	0x1
-#define SYSCTL_FLAG_WSCALE	0x2
-#define SYSCTL_FLAG_SACK	0x4
+	memset(&opts, 0, sizeof(opts));
 
-	sysctl_flags = 0;
-	if (unlikely(tcb->flags & TCPCB_FLAG_SYN)) {
-		tcp_header_size = sizeof(struct tcphdr) + TCPOLEN_MSS;
-		if (sysctl_tcp_timestamps) {
-			tcp_header_size += TCPOLEN_TSTAMP_ALIGNED;
-			sysctl_flags |= SYSCTL_FLAG_TSTAMPS;
-		}
-		if (sysctl_tcp_window_scaling) {
-			tcp_header_size += TCPOLEN_WSCALE_ALIGNED;
-			sysctl_flags |= SYSCTL_FLAG_WSCALE;
-		}
-		if (sysctl_tcp_sack) {
-			sysctl_flags |= SYSCTL_FLAG_SACK;
-			if (!(sysctl_flags & SYSCTL_FLAG_TSTAMPS))
-				tcp_header_size += TCPOLEN_SACKPERM_ALIGNED;
-		}
-	} else if (unlikely(tp->rx_opt.eff_sacks)) {
-		/* A SACK is 2 pad bytes, a 2 byte header, plus
-		 * 2 32-bit sequence numbers for each SACK block.
-		 */
-		tcp_header_size += (TCPOLEN_SACK_BASE_ALIGNED +
-				    (tp->rx_opt.eff_sacks *
-				     TCPOLEN_SACK_PERBLOCK));
-	}
+	if (unlikely(tcb->flags & TCPCB_FLAG_SYN))
+		tcp_options_size = tcp_syn_options(sk, skb, &opts, &md5);
+	else
+		tcp_options_size = tcp_established_options(sk, skb, &opts,
+							   &md5);
+	tcp_header_size = tcp_options_size + sizeof(struct tcphdr);
 
 	if (tcp_packets_in_flight(tp) == 0)
 		tcp_ca_event(sk, CA_EVENT_TX_START);
 
-#ifdef CONFIG_TCP_MD5SIG
-	/*
-	 * Are we doing MD5 on this segment? If so - make
-	 * room for it.
-	 */
-	md5 = tp->af_specific->md5_lookup(sk, sk);
-	if (md5) {
-		tcp_header_size += TCPOLEN_MD5SIG_ALIGNED;
-		sk->sk_route_caps &= ~NETIF_F_GSO_MASK;
-	}
-#endif
-
 	skb_push(skb, tcp_header_size);
 	skb_reset_transport_header(skb);
 	skb_set_owner_w(skb, sk);
@@ -576,33 +650,14 @@ static int tcp_transmit_skb(struct sock *sk, struct sk_buff *skb, int clone_it,
 		th->urg			= 1;
 	}
 
-	if (unlikely(tcb->flags & TCPCB_FLAG_SYN)) {
-		tcp_syn_build_options((__be32 *)(th + 1),
-				      tcp_advertise_mss(sk),
-				      (sysctl_flags & SYSCTL_FLAG_TSTAMPS),
-				      (sysctl_flags & SYSCTL_FLAG_SACK),
-				      (sysctl_flags & SYSCTL_FLAG_WSCALE),
-				      tp->rx_opt.rcv_wscale,
-				      tcb->when,
-				      tp->rx_opt.ts_recent,
-
-#ifdef CONFIG_TCP_MD5SIG
-				      md5 ? &md5_hash_location :
-#endif
-				      NULL);
-	} else {
-		tcp_build_and_update_options((__be32 *)(th + 1),
-					     tp, tcb->when,
-#ifdef CONFIG_TCP_MD5SIG
-					     md5 ? &md5_hash_location :
-#endif
-					     NULL);
+	tcp_options_write((__be32 *)(th + 1), tp, &opts, &md5_hash_location);
+	if (likely((tcb->flags & TCPCB_FLAG_SYN) == 0))
 		TCP_ECN_send(sk, skb, tcp_header_size);
-	}
 
 #ifdef CONFIG_TCP_MD5SIG
 	/* Calculate the MD5 hash, as we have all we need now */
 	if (md5) {
+		sk->sk_route_caps &= ~NETIF_F_GSO_MASK;
 		tp->af_specific->calc_md5_hash(md5_hash_location,
 					       md5, sk, NULL, skb);
 	}
@@ -626,10 +681,6 @@ static int tcp_transmit_skb(struct sock *sk, struct sk_buff *skb, int clone_it,
 	tcp_enter_cwr(sk, 1);
 
 	return net_xmit_eval(err);
-
-#undef SYSCTL_FLAG_TSTAMPS
-#undef SYSCTL_FLAG_WSCALE
-#undef SYSCTL_FLAG_SACK
 }
 
 /* This routine just queue's the buffer
@@ -970,6 +1021,9 @@ unsigned int tcp_current_mss(struct sock *sk, int large_allowed)
 	u32 mss_now;
 	u16 xmit_size_goal;
 	int doing_tso = 0;
+	unsigned header_len;
+	struct tcp_out_options opts;
+	struct tcp_md5sig_key *md5;
 
 	mss_now = tp->mss_cache;
 
@@ -982,14 +1036,16 @@ unsigned int tcp_current_mss(struct sock *sk, int large_allowed)
 			mss_now = tcp_sync_mss(sk, mtu);
 	}
 
-	if (tp->rx_opt.eff_sacks)
-		mss_now -= (TCPOLEN_SACK_BASE_ALIGNED +
-			    (tp->rx_opt.eff_sacks * TCPOLEN_SACK_PERBLOCK));
-
-#ifdef CONFIG_TCP_MD5SIG
-	if (tp->af_specific->md5_lookup(sk, sk))
-		mss_now -= TCPOLEN_MD5SIG_ALIGNED;
-#endif
+	header_len = tcp_established_options(sk, NULL, &opts, &md5) +
+		     sizeof(struct tcphdr);
+	/* The mss_cache is sized based on tp->tcp_header_len, which assumes
+	 * some common options. If this is an odd packet (because we have SACK
+	 * blocks etc) then our calculated header_len will be different, and
+	 * we have to adjust mss_now correspondingly */
+	if (header_len != tp->tcp_header_len) {
+		int delta = (int) header_len - tp->tcp_header_len;
+		mss_now -= delta;
+	}
 
 	xmit_size_goal = mss_now;
 
@@ -2179,11 +2235,10 @@ struct sk_buff *tcp_make_synack(struct sock *sk, struct dst_entry *dst,
 	struct tcp_sock *tp = tcp_sk(sk);
 	struct tcphdr *th;
 	int tcp_header_size;
+	struct tcp_out_options opts;
 	struct sk_buff *skb;
-#ifdef CONFIG_TCP_MD5SIG
 	struct tcp_md5sig_key *md5;
 	__u8 *md5_hash_location;
-#endif
 
 	skb = sock_wmalloc(sk, MAX_TCP_HEADER + 15, 1, GFP_ATOMIC);
 	if (skb == NULL)
@@ -2194,18 +2249,27 @@ struct sk_buff *tcp_make_synack(struct sock *sk, struct dst_entry *dst,
 
 	skb->dst = dst_clone(dst);
 
-	tcp_header_size = (sizeof(struct tcphdr) + TCPOLEN_MSS +
-			   (ireq->tstamp_ok ? TCPOLEN_TSTAMP_ALIGNED : 0) +
-			   (ireq->wscale_ok ? TCPOLEN_WSCALE_ALIGNED : 0) +
-			   /* SACK_PERM is in the place of NOP NOP of TS */
-			   ((ireq->sack_ok && !ireq->tstamp_ok) ? TCPOLEN_SACKPERM_ALIGNED : 0));
+	if (req->rcv_wnd == 0) { /* ignored for retransmitted syns */
+		__u8 rcv_wscale;
+		/* Set this up on the first call only */
+		req->window_clamp = tp->window_clamp ? : dst_metric(dst, RTAX_WINDOW);
+		/* tcp_full_space because it is guaranteed to be the first packet */
+		tcp_select_initial_window(tcp_full_space(sk),
+			dst_metric(dst, RTAX_ADVMSS) - (ireq->tstamp_ok ? TCPOLEN_TSTAMP_ALIGNED : 0),
+			&req->rcv_wnd,
+			&req->window_clamp,
+			ireq->wscale_ok,
+			&rcv_wscale);
+		ireq->rcv_wscale = rcv_wscale;
+	}
+
+	memset(&opts, 0, sizeof(opts));
+	TCP_SKB_CB(skb)->when = tcp_time_stamp;
+	tcp_header_size = tcp_synack_options(sk, req,
+					     dst_metric(dst, RTAX_ADVMSS),
+					     skb, &opts, &md5) +
+			  sizeof(struct tcphdr);
 
-#ifdef CONFIG_TCP_MD5SIG
-	/* Are we doing MD5 on this segment? If so - make room for it */
-	md5 = tcp_rsk(req)->af_specific->md5_lookup(sk, req);
-	if (md5)
-		tcp_header_size += TCPOLEN_MD5SIG_ALIGNED;
-#endif
 	skb_push(skb, tcp_header_size);
 	skb_reset_transport_header(skb);
 
@@ -2223,19 +2287,6 @@ struct sk_buff *tcp_make_synack(struct sock *sk, struct dst_entry *dst,
 			     TCPCB_FLAG_SYN | TCPCB_FLAG_ACK);
 	th->seq = htonl(TCP_SKB_CB(skb)->seq);
 	th->ack_seq = htonl(tcp_rsk(req)->rcv_isn + 1);
-	if (req->rcv_wnd == 0) { /* ignored for retransmitted syns */
-		__u8 rcv_wscale;
-		/* Set this up on the first call only */
-		req->window_clamp = tp->window_clamp ? : dst_metric(dst, RTAX_WINDOW);
-		/* tcp_full_space because it is guaranteed to be the first packet */
-		tcp_select_initial_window(tcp_full_space(sk),
-			dst_metric(dst, RTAX_ADVMSS) - (ireq->tstamp_ok ? TCPOLEN_TSTAMP_ALIGNED : 0),
-			&req->rcv_wnd,
-			&req->window_clamp,
-			ireq->wscale_ok,
-			&rcv_wscale);
-		ireq->rcv_wscale = rcv_wscale;
-	}
 
 	/* RFC1323: The window in SYN & SYN/ACK segments is never scaled. */
 	th->window = htons(min(req->rcv_wnd, 65535U));
@@ -2244,18 +2295,7 @@ struct sk_buff *tcp_make_synack(struct sock *sk, struct dst_entry *dst,
 		TCP_SKB_CB(skb)->when = cookie_init_timestamp(req);
 	else
 #endif
-	TCP_SKB_CB(skb)->when = tcp_time_stamp;
-	tcp_syn_build_options((__be32 *)(th + 1), dst_metric(dst, RTAX_ADVMSS), ireq->tstamp_ok,
-			      ireq->sack_ok, ireq->wscale_ok, ireq->rcv_wscale,
-			      TCP_SKB_CB(skb)->when,
-			      req->ts_recent,
-			      (
-#ifdef CONFIG_TCP_MD5SIG
-			       md5 ? &md5_hash_location :
-#endif
-			       NULL)
-			      );
-
+	tcp_options_write((__be32 *)(th + 1), tp, &opts, &md5_hash_location);
 	th->doff = (tcp_header_size >> 2);
 	TCP_INC_STATS(sock_net(sk), TCP_MIB_OUTSEGS);
 
-- 
cgit v1.2.3


From 4389dded7767d24290463f2a8302ba3253ebdd56 Mon Sep 17 00:00:00 2001
From: Adam Langley <agl@imperialviolet.org>
Date: Sat, 19 Jul 2008 00:07:02 -0700
Subject: tcp: Remove redundant checks when setting eff_sacks

Remove redundant checks when setting eff_sacks and make the number of SACKs a
compile time constant. Now that the options code knows how many SACK blocks can
fit in the header, we don't need to have the SACK code guessing at it.

Signed-off-by: Adam Langley <agl@imperialviolet.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/tcp.h  |  6 ++++++
 net/ipv4/tcp_input.c | 25 ++++++++++---------------
 2 files changed, 16 insertions(+), 15 deletions(-)

(limited to 'include')

diff --git a/include/linux/tcp.h b/include/linux/tcp.h
index 07e79bdb9cdf..2e2557388e36 100644
--- a/include/linux/tcp.h
+++ b/include/linux/tcp.h
@@ -224,6 +224,12 @@ struct tcp_options_received {
 	u16	mss_clamp;	/* Maximal mss, negotiated at connection setup */
 };
 
+/* This is the max number of SACKS that we'll generate and process. It's safe
+ * to increse this, although since:
+ *   size = TCPOLEN_SACK_BASE_ALIGNED (4) + n * TCPOLEN_SACK_PERBLOCK (8)
+ * only four options will fit in a standard TCP header */
+#define TCP_NUM_SACKS 4
+
 struct tcp_request_sock {
 	struct inet_request_sock 	req;
 #ifdef CONFIG_TCP_MD5SIG
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index 88810bc01370..1f5e6049883e 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -1423,10 +1423,10 @@ tcp_sacktag_write_queue(struct sock *sk, struct sk_buff *ack_skb,
 	unsigned char *ptr = (skb_transport_header(ack_skb) +
 			      TCP_SKB_CB(ack_skb)->sacked);
 	struct tcp_sack_block_wire *sp_wire = (struct tcp_sack_block_wire *)(ptr+2);
-	struct tcp_sack_block sp[4];
+	struct tcp_sack_block sp[TCP_NUM_SACKS];
 	struct tcp_sack_block *cache;
 	struct sk_buff *skb;
-	int num_sacks = (ptr[1] - TCPOLEN_SACK_BASE) >> 3;
+	int num_sacks = min(TCP_NUM_SACKS, (ptr[1] - TCPOLEN_SACK_BASE) >> 3);
 	int used_sacks;
 	int reord = tp->packets_out;
 	int flag = 0;
@@ -3735,8 +3735,7 @@ static void tcp_dsack_set(struct sock *sk, u32 seq, u32 end_seq)
 		tp->rx_opt.dsack = 1;
 		tp->duplicate_sack[0].start_seq = seq;
 		tp->duplicate_sack[0].end_seq = end_seq;
-		tp->rx_opt.eff_sacks = min(tp->rx_opt.num_sacks + 1,
-					   4 - tp->rx_opt.tstamp_ok);
+		tp->rx_opt.eff_sacks = tp->rx_opt.num_sacks + 1;
 	}
 }
 
@@ -3791,9 +3790,8 @@ static void tcp_sack_maybe_coalesce(struct tcp_sock *tp)
 			 * Decrease num_sacks.
 			 */
 			tp->rx_opt.num_sacks--;
-			tp->rx_opt.eff_sacks = min(tp->rx_opt.num_sacks +
-						   tp->rx_opt.dsack,
-						   4 - tp->rx_opt.tstamp_ok);
+			tp->rx_opt.eff_sacks = tp->rx_opt.num_sacks +
+					       tp->rx_opt.dsack;
 			for (i = this_sack; i < tp->rx_opt.num_sacks; i++)
 				sp[i] = sp[i + 1];
 			continue;
@@ -3843,7 +3841,7 @@ static void tcp_sack_new_ofo_skb(struct sock *sk, u32 seq, u32 end_seq)
 	 *
 	 * If the sack array is full, forget about the last one.
 	 */
-	if (this_sack >= 4) {
+	if (this_sack >= TCP_NUM_SACKS) {
 		this_sack--;
 		tp->rx_opt.num_sacks--;
 		sp--;
@@ -3856,8 +3854,7 @@ new_sack:
 	sp->start_seq = seq;
 	sp->end_seq = end_seq;
 	tp->rx_opt.num_sacks++;
-	tp->rx_opt.eff_sacks = min(tp->rx_opt.num_sacks + tp->rx_opt.dsack,
-				   4 - tp->rx_opt.tstamp_ok);
+	tp->rx_opt.eff_sacks = tp->rx_opt.num_sacks + tp->rx_opt.dsack;
 }
 
 /* RCV.NXT advances, some SACKs should be eaten. */
@@ -3894,9 +3891,8 @@ static void tcp_sack_remove(struct tcp_sock *tp)
 	}
 	if (num_sacks != tp->rx_opt.num_sacks) {
 		tp->rx_opt.num_sacks = num_sacks;
-		tp->rx_opt.eff_sacks = min(tp->rx_opt.num_sacks +
-					   tp->rx_opt.dsack,
-					   4 - tp->rx_opt.tstamp_ok);
+		tp->rx_opt.eff_sacks = tp->rx_opt.num_sacks +
+				       tp->rx_opt.dsack;
 	}
 }
 
@@ -3975,8 +3971,7 @@ static void tcp_data_queue(struct sock *sk, struct sk_buff *skb)
 
 	if (tp->rx_opt.dsack) {
 		tp->rx_opt.dsack = 0;
-		tp->rx_opt.eff_sacks = min_t(unsigned int, tp->rx_opt.num_sacks,
-					     4 - tp->rx_opt.tstamp_ok);
+		tp->rx_opt.eff_sacks = tp->rx_opt.num_sacks;
 	}
 
 	/*  Queue data for delivery to the user.
-- 
cgit v1.2.3


From 725a8ff04a5dc473cd9d8eb7817ca96fc36c7789 Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Sat, 19 Jul 2008 00:28:58 -0700
Subject: ipv6: remove unused parameter from ip6_ra_control

Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ipv6.h       | 4 +---
 net/ipv6/ipv6_sockglue.c | 7 ++-----
 net/ipv6/raw.c           | 2 +-
 3 files changed, 4 insertions(+), 9 deletions(-)

(limited to 'include')

diff --git a/include/net/ipv6.h b/include/net/ipv6.h
index c2222ee74d66..6b7982d3dda0 100644
--- a/include/net/ipv6.h
+++ b/include/net/ipv6.h
@@ -227,9 +227,7 @@ static inline void fl6_sock_release(struct ip6_flowlabel *fl)
 		atomic_dec(&fl->users);
 }
 
-extern int 			ip6_ra_control(struct sock *sk, int sel,
-					       void (*destructor)(struct sock *));
-
+extern int 			ip6_ra_control(struct sock *sk, int sel);
 
 extern int			ipv6_parse_hopopts(struct sk_buff *skb);
 
diff --git a/net/ipv6/ipv6_sockglue.c b/net/ipv6/ipv6_sockglue.c
index 030c0c956f9d..8c6ea07f4d56 100644
--- a/net/ipv6/ipv6_sockglue.c
+++ b/net/ipv6/ipv6_sockglue.c
@@ -59,7 +59,7 @@ DEFINE_SNMP_STAT(struct ipstats_mib, ipv6_statistics) __read_mostly;
 struct ip6_ra_chain *ip6_ra_chain;
 DEFINE_RWLOCK(ip6_ra_lock);
 
-int ip6_ra_control(struct sock *sk, int sel, void (*destructor)(struct sock *))
+int ip6_ra_control(struct sock *sk, int sel)
 {
 	struct ip6_ra_chain *ra, *new_ra, **rap;
 
@@ -81,8 +81,6 @@ int ip6_ra_control(struct sock *sk, int sel, void (*destructor)(struct sock *))
 			*rap = ra->next;
 			write_unlock_bh(&ip6_ra_lock);
 
-			if (ra->destructor)
-				ra->destructor(sk);
 			sock_put(sk);
 			kfree(ra);
 			return 0;
@@ -94,7 +92,6 @@ int ip6_ra_control(struct sock *sk, int sel, void (*destructor)(struct sock *))
 	}
 	new_ra->sk = sk;
 	new_ra->sel = sel;
-	new_ra->destructor = destructor;
 	new_ra->next = ra;
 	*rap = new_ra;
 	sock_hold(sk);
@@ -632,7 +629,7 @@ done:
 	case IPV6_ROUTER_ALERT:
 		if (optlen < sizeof(int))
 			goto e_inval;
-		retv = ip6_ra_control(sk, val, NULL);
+		retv = ip6_ra_control(sk, val);
 		break;
 	case IPV6_MTU_DISCOVER:
 		if (optlen < sizeof(int))
diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c
index 34cfb3f41c2c..01d47674f7e5 100644
--- a/net/ipv6/raw.c
+++ b/net/ipv6/raw.c
@@ -1157,7 +1157,7 @@ static int rawv6_ioctl(struct sock *sk, int cmd, unsigned long arg)
 static void rawv6_close(struct sock *sk, long timeout)
 {
 	if (inet_sk(sk)->num == IPPROTO_RAW)
-		ip6_ra_control(sk, -1, NULL);
+		ip6_ra_control(sk, -1);
 	ip6mr_sk_done(sk);
 	sk_common_release(sk);
 }
-- 
cgit v1.2.3


From 7abbcd6a4c8d6179121f2915a761b1133bf1cd99 Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Sat, 19 Jul 2008 00:29:42 -0700
Subject: ipv6: remove unused macros from net/ipv6.h

Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ipv6.h | 6 ------
 1 file changed, 6 deletions(-)

(limited to 'include')

diff --git a/include/net/ipv6.h b/include/net/ipv6.h
index 6b7982d3dda0..2d5c18514a2d 100644
--- a/include/net/ipv6.h
+++ b/include/net/ipv6.h
@@ -131,7 +131,6 @@ DECLARE_SNMP_STAT(struct ipstats_mib, ipv6_statistics);
 
 #define IP6_INC_STATS(idev,field)	_DEVINC(ipv6, , idev, field)
 #define IP6_INC_STATS_BH(idev,field)	_DEVINC(ipv6, _BH, idev, field)
-#define IP6_INC_STATS_USER(idev,field)	_DEVINC(ipv6, _USER, idev, field)
 #define IP6_ADD_STATS_BH(idev,field,val) _DEVADD(ipv6, _BH, idev, field, val)
 
 DECLARE_SNMP_STAT(struct icmpv6_mib, icmpv6_statistics);
@@ -139,20 +138,15 @@ DECLARE_SNMP_STAT(struct icmpv6msg_mib, icmpv6msg_statistics);
 
 #define ICMP6_INC_STATS(idev, field)	_DEVINC(icmpv6, , idev, field)
 #define ICMP6_INC_STATS_BH(idev, field)	_DEVINC(icmpv6, _BH, idev, field)
-#define ICMP6_INC_STATS_USER(idev, field) _DEVINC(icmpv6, _USER, idev, field)
 
 #define ICMP6MSGOUT_INC_STATS(idev, field) \
 	_DEVINC(icmpv6msg, , idev, field +256)
 #define ICMP6MSGOUT_INC_STATS_BH(idev, field) \
 	_DEVINC(icmpv6msg, _BH, idev, field +256)
-#define ICMP6MSGOUT_INC_STATS_USER(idev, field) \
-	_DEVINC(icmpv6msg, _USER, idev, field +256)
 #define ICMP6MSGIN_INC_STATS(idev, field) \
 	 _DEVINC(icmpv6msg, , idev, field)
 #define ICMP6MSGIN_INC_STATS_BH(idev, field) \
 	_DEVINC(icmpv6msg, _BH, idev, field)
-#define ICMP6MSGIN_INC_STATS_USER(idev, field) \
-	_DEVINC(icmpv6msg, _USER, idev, field)
 
 struct ip6_ra_chain
 {
-- 
cgit v1.2.3


From 230b183921ecbaa5fedc0d35ad6ba7bb64b6e06a Mon Sep 17 00:00:00 2001
From: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Date: Sat, 19 Jul 2008 22:35:47 -0700
Subject: net: Use standard structures for generic socket address structures.

Use sockaddr_storage{} for generic socket address storage
and ensures proper alignment.
Use sockaddr{} for pointers to omit several casts.

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/socket.h |  6 ++--
 include/net/compat.h   |  2 +-
 net/compat.c           |  2 +-
 net/core/iovec.c       |  2 +-
 net/socket.c           | 82 ++++++++++++++++++++++++++++----------------------
 5 files changed, 52 insertions(+), 42 deletions(-)

(limited to 'include')

diff --git a/include/linux/socket.h b/include/linux/socket.h
index bd2b30a74e76..950af631e7fb 100644
--- a/include/linux/socket.h
+++ b/include/linux/socket.h
@@ -306,10 +306,10 @@ extern int csum_partial_copy_fromiovecend(unsigned char *kdata,
 					  int offset, 
 					  unsigned int len, __wsum *csump);
 
-extern int verify_iovec(struct msghdr *m, struct iovec *iov, char *address, int mode);
+extern int verify_iovec(struct msghdr *m, struct iovec *iov, struct sockaddr *address, int mode);
 extern int memcpy_toiovec(struct iovec *v, unsigned char *kdata, int len);
-extern int move_addr_to_user(void *kaddr, int klen, void __user *uaddr, int __user *ulen);
-extern int move_addr_to_kernel(void __user *uaddr, int ulen, void *kaddr);
+extern int move_addr_to_user(struct sockaddr *kaddr, int klen, void __user *uaddr, int __user *ulen);
+extern int move_addr_to_kernel(void __user *uaddr, int ulen, struct sockaddr *kaddr);
 extern int put_cmsg(struct msghdr*, int level, int type, int len, void *data);
 
 #endif
diff --git a/include/net/compat.h b/include/net/compat.h
index 164cb682e220..5bbf8bf9efea 100644
--- a/include/net/compat.h
+++ b/include/net/compat.h
@@ -32,7 +32,7 @@ extern int compat_sock_get_timestampns(struct sock *, struct timespec __user *);
 #endif /* defined(CONFIG_COMPAT) */
 
 extern int get_compat_msghdr(struct msghdr *, struct compat_msghdr __user *);
-extern int verify_compat_iovec(struct msghdr *, struct iovec *, char *, int);
+extern int verify_compat_iovec(struct msghdr *, struct iovec *, struct sockaddr *, int);
 extern asmlinkage long compat_sys_sendmsg(int,struct compat_msghdr __user *,unsigned);
 extern asmlinkage long compat_sys_recvmsg(int,struct compat_msghdr __user *,unsigned);
 extern asmlinkage long compat_sys_getsockopt(int, int, int, char __user *, int __user *);
diff --git a/net/compat.c b/net/compat.c
index c823f6f290cb..6e1b03b51933 100644
--- a/net/compat.c
+++ b/net/compat.c
@@ -75,7 +75,7 @@ int get_compat_msghdr(struct msghdr *kmsg, struct compat_msghdr __user *umsg)
 
 /* I've named the args so it is easy to tell whose space the pointers are in. */
 int verify_compat_iovec(struct msghdr *kern_msg, struct iovec *kern_iov,
-		   char *kern_address, int mode)
+		   struct sockaddr *kern_address, int mode)
 {
 	int tot_len;
 
diff --git a/net/core/iovec.c b/net/core/iovec.c
index 755c37fdaee7..4c9c0121c9da 100644
--- a/net/core/iovec.c
+++ b/net/core/iovec.c
@@ -36,7 +36,7 @@
  *	in any case.
  */
 
-int verify_iovec(struct msghdr *m, struct iovec *iov, char *address, int mode)
+int verify_iovec(struct msghdr *m, struct iovec *iov, struct sockaddr *address, int mode)
 {
 	int size, err, ct;
 
diff --git a/net/socket.c b/net/socket.c
index 81fe82513046..1ba57d888981 100644
--- a/net/socket.c
+++ b/net/socket.c
@@ -180,9 +180,9 @@ static DEFINE_PER_CPU(int, sockets_in_use) = 0;
  *	invalid addresses -EFAULT is returned. On a success 0 is returned.
  */
 
-int move_addr_to_kernel(void __user *uaddr, int ulen, void *kaddr)
+int move_addr_to_kernel(void __user *uaddr, int ulen, struct sockaddr *kaddr)
 {
-	if (ulen < 0 || ulen > MAX_SOCK_ADDR)
+	if (ulen < 0 || ulen > sizeof(struct sockaddr_storage))
 		return -EINVAL;
 	if (ulen == 0)
 		return 0;
@@ -208,7 +208,7 @@ int move_addr_to_kernel(void __user *uaddr, int ulen, void *kaddr)
  *	specified. Zero is returned for a success.
  */
 
-int move_addr_to_user(void *kaddr, int klen, void __user *uaddr,
+int move_addr_to_user(struct sockaddr *kaddr, int klen, void __user *uaddr,
 		      int __user *ulen)
 {
 	int err;
@@ -219,7 +219,7 @@ int move_addr_to_user(void *kaddr, int klen, void __user *uaddr,
 		return err;
 	if (len > klen)
 		len = klen;
-	if (len < 0 || len > MAX_SOCK_ADDR)
+	if (len < 0 || len > sizeof(struct sockaddr_storage))
 		return -EINVAL;
 	if (len) {
 		if (audit_sockaddr(klen, kaddr))
@@ -1342,20 +1342,20 @@ out_fd:
 asmlinkage long sys_bind(int fd, struct sockaddr __user *umyaddr, int addrlen)
 {
 	struct socket *sock;
-	char address[MAX_SOCK_ADDR];
+	struct sockaddr_storage address;
 	int err, fput_needed;
 
 	sock = sockfd_lookup_light(fd, &err, &fput_needed);
 	if (sock) {
-		err = move_addr_to_kernel(umyaddr, addrlen, address);
+		err = move_addr_to_kernel(umyaddr, addrlen, (struct sockaddr *)&address);
 		if (err >= 0) {
 			err = security_socket_bind(sock,
-						   (struct sockaddr *)address,
+						   (struct sockaddr *)&address,
 						   addrlen);
 			if (!err)
 				err = sock->ops->bind(sock,
 						      (struct sockaddr *)
-						      address, addrlen);
+						      &address, addrlen);
 		}
 		fput_light(sock->file, fput_needed);
 	}
@@ -1407,7 +1407,7 @@ asmlinkage long sys_accept(int fd, struct sockaddr __user *upeer_sockaddr,
 	struct socket *sock, *newsock;
 	struct file *newfile;
 	int err, len, newfd, fput_needed;
-	char address[MAX_SOCK_ADDR];
+	struct sockaddr_storage address;
 
 	sock = sockfd_lookup_light(fd, &err, &fput_needed);
 	if (!sock)
@@ -1446,13 +1446,13 @@ asmlinkage long sys_accept(int fd, struct sockaddr __user *upeer_sockaddr,
 		goto out_fd;
 
 	if (upeer_sockaddr) {
-		if (newsock->ops->getname(newsock, (struct sockaddr *)address,
+		if (newsock->ops->getname(newsock, (struct sockaddr *)&address,
 					  &len, 2) < 0) {
 			err = -ECONNABORTED;
 			goto out_fd;
 		}
-		err = move_addr_to_user(address, len, upeer_sockaddr,
-					upeer_addrlen);
+		err = move_addr_to_user((struct sockaddr *)&address,
+					len, upeer_sockaddr, upeer_addrlen);
 		if (err < 0)
 			goto out_fd;
 	}
@@ -1495,22 +1495,22 @@ asmlinkage long sys_connect(int fd, struct sockaddr __user *uservaddr,
 			    int addrlen)
 {
 	struct socket *sock;
-	char address[MAX_SOCK_ADDR];
+	struct sockaddr_storage address;
 	int err, fput_needed;
 
 	sock = sockfd_lookup_light(fd, &err, &fput_needed);
 	if (!sock)
 		goto out;
-	err = move_addr_to_kernel(uservaddr, addrlen, address);
+	err = move_addr_to_kernel(uservaddr, addrlen, (struct sockaddr *)&address);
 	if (err < 0)
 		goto out_put;
 
 	err =
-	    security_socket_connect(sock, (struct sockaddr *)address, addrlen);
+	    security_socket_connect(sock, (struct sockaddr *)&address, addrlen);
 	if (err)
 		goto out_put;
 
-	err = sock->ops->connect(sock, (struct sockaddr *)address, addrlen,
+	err = sock->ops->connect(sock, (struct sockaddr *)&address, addrlen,
 				 sock->file->f_flags);
 out_put:
 	fput_light(sock->file, fput_needed);
@@ -1527,7 +1527,7 @@ asmlinkage long sys_getsockname(int fd, struct sockaddr __user *usockaddr,
 				int __user *usockaddr_len)
 {
 	struct socket *sock;
-	char address[MAX_SOCK_ADDR];
+	struct sockaddr_storage address;
 	int len, err, fput_needed;
 
 	sock = sockfd_lookup_light(fd, &err, &fput_needed);
@@ -1538,10 +1538,10 @@ asmlinkage long sys_getsockname(int fd, struct sockaddr __user *usockaddr,
 	if (err)
 		goto out_put;
 
-	err = sock->ops->getname(sock, (struct sockaddr *)address, &len, 0);
+	err = sock->ops->getname(sock, (struct sockaddr *)&address, &len, 0);
 	if (err)
 		goto out_put;
-	err = move_addr_to_user(address, len, usockaddr, usockaddr_len);
+	err = move_addr_to_user((struct sockaddr *)&address, len, usockaddr, usockaddr_len);
 
 out_put:
 	fput_light(sock->file, fput_needed);
@@ -1558,7 +1558,7 @@ asmlinkage long sys_getpeername(int fd, struct sockaddr __user *usockaddr,
 				int __user *usockaddr_len)
 {
 	struct socket *sock;
-	char address[MAX_SOCK_ADDR];
+	struct sockaddr_storage address;
 	int len, err, fput_needed;
 
 	sock = sockfd_lookup_light(fd, &err, &fput_needed);
@@ -1570,10 +1570,10 @@ asmlinkage long sys_getpeername(int fd, struct sockaddr __user *usockaddr,
 		}
 
 		err =
-		    sock->ops->getname(sock, (struct sockaddr *)address, &len,
+		    sock->ops->getname(sock, (struct sockaddr *)&address, &len,
 				       1);
 		if (!err)
-			err = move_addr_to_user(address, len, usockaddr,
+			err = move_addr_to_user((struct sockaddr *)&address, len, usockaddr,
 						usockaddr_len);
 		fput_light(sock->file, fput_needed);
 	}
@@ -1591,7 +1591,7 @@ asmlinkage long sys_sendto(int fd, void __user *buff, size_t len,
 			   int addr_len)
 {
 	struct socket *sock;
-	char address[MAX_SOCK_ADDR];
+	struct sockaddr_storage address;
 	int err;
 	struct msghdr msg;
 	struct iovec iov;
@@ -1610,10 +1610,10 @@ asmlinkage long sys_sendto(int fd, void __user *buff, size_t len,
 	msg.msg_controllen = 0;
 	msg.msg_namelen = 0;
 	if (addr) {
-		err = move_addr_to_kernel(addr, addr_len, address);
+		err = move_addr_to_kernel(addr, addr_len, (struct sockaddr *)&address);
 		if (err < 0)
 			goto out_put;
-		msg.msg_name = address;
+		msg.msg_name = (struct sockaddr *)&address;
 		msg.msg_namelen = addr_len;
 	}
 	if (sock->file->f_flags & O_NONBLOCK)
@@ -1649,7 +1649,7 @@ asmlinkage long sys_recvfrom(int fd, void __user *ubuf, size_t size,
 	struct socket *sock;
 	struct iovec iov;
 	struct msghdr msg;
-	char address[MAX_SOCK_ADDR];
+	struct sockaddr_storage address;
 	int err, err2;
 	int fput_needed;
 
@@ -1663,14 +1663,15 @@ asmlinkage long sys_recvfrom(int fd, void __user *ubuf, size_t size,
 	msg.msg_iov = &iov;
 	iov.iov_len = size;
 	iov.iov_base = ubuf;
-	msg.msg_name = address;
-	msg.msg_namelen = MAX_SOCK_ADDR;
+	msg.msg_name = (struct sockaddr *)&address;
+	msg.msg_namelen = sizeof(address);
 	if (sock->file->f_flags & O_NONBLOCK)
 		flags |= MSG_DONTWAIT;
 	err = sock_recvmsg(sock, &msg, size, flags);
 
 	if (err >= 0 && addr != NULL) {
-		err2 = move_addr_to_user(address, msg.msg_namelen, addr, addr_len);
+		err2 = move_addr_to_user((struct sockaddr *)&address,
+					 msg.msg_namelen, addr, addr_len);
 		if (err2 < 0)
 			err = err2;
 	}
@@ -1790,7 +1791,7 @@ asmlinkage long sys_sendmsg(int fd, struct msghdr __user *msg, unsigned flags)
 	struct compat_msghdr __user *msg_compat =
 	    (struct compat_msghdr __user *)msg;
 	struct socket *sock;
-	char address[MAX_SOCK_ADDR];
+	struct sockaddr_storage address;
 	struct iovec iovstack[UIO_FASTIOV], *iov = iovstack;
 	unsigned char ctl[sizeof(struct cmsghdr) + 20]
 	    __attribute__ ((aligned(sizeof(__kernel_size_t))));
@@ -1828,9 +1829,13 @@ asmlinkage long sys_sendmsg(int fd, struct msghdr __user *msg, unsigned flags)
 
 	/* This will also move the address data into kernel space */
 	if (MSG_CMSG_COMPAT & flags) {
-		err = verify_compat_iovec(&msg_sys, iov, address, VERIFY_READ);
+		err = verify_compat_iovec(&msg_sys, iov,
+					  (struct sockaddr *)&address,
+					  VERIFY_READ);
 	} else
-		err = verify_iovec(&msg_sys, iov, address, VERIFY_READ);
+		err = verify_iovec(&msg_sys, iov,
+				   (struct sockaddr *)&address,
+				   VERIFY_READ);
 	if (err < 0)
 		goto out_freeiov;
 	total_len = err;
@@ -1901,7 +1906,7 @@ asmlinkage long sys_recvmsg(int fd, struct msghdr __user *msg,
 	int fput_needed;
 
 	/* kernel mode address */
-	char addr[MAX_SOCK_ADDR];
+	struct sockaddr_storage addr;
 
 	/* user mode address pointers */
 	struct sockaddr __user *uaddr;
@@ -1939,9 +1944,13 @@ asmlinkage long sys_recvmsg(int fd, struct msghdr __user *msg,
 	uaddr = (__force void __user *)msg_sys.msg_name;
 	uaddr_len = COMPAT_NAMELEN(msg);
 	if (MSG_CMSG_COMPAT & flags) {
-		err = verify_compat_iovec(&msg_sys, iov, addr, VERIFY_WRITE);
+		err = verify_compat_iovec(&msg_sys, iov,
+					  (struct sockaddr *)&addr,
+					  VERIFY_WRITE);
 	} else
-		err = verify_iovec(&msg_sys, iov, addr, VERIFY_WRITE);
+		err = verify_iovec(&msg_sys, iov,
+				   (struct sockaddr *)&addr,
+				   VERIFY_WRITE);
 	if (err < 0)
 		goto out_freeiov;
 	total_len = err;
@@ -1957,7 +1966,8 @@ asmlinkage long sys_recvmsg(int fd, struct msghdr __user *msg,
 	len = err;
 
 	if (uaddr != NULL) {
-		err = move_addr_to_user(addr, msg_sys.msg_namelen, uaddr,
+		err = move_addr_to_user((struct sockaddr *)&addr,
+					msg_sys.msg_namelen, uaddr,
 					uaddr_len);
 		if (err < 0)
 			goto out_freeiov;
-- 
cgit v1.2.3


From 5f86173bdf15981ca49d0434f638b68f70a35644 Mon Sep 17 00:00:00 2001
From: Jussi Kivilinna <jussi.kivilinna@mbnet.fi>
Date: Sun, 20 Jul 2008 00:08:04 -0700
Subject: net_sched: Add qdisc_enqueue wrapper

Signed-off-by: Jussi Kivilinna <jussi.kivilinna@mbnet.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/sch_generic.h | 10 ++++++++++
 net/core/dev.c            |  4 ++--
 net/mac80211/wme.c        |  2 +-
 net/sched/sch_atm.c       |  2 +-
 net/sched/sch_cbq.c       |  5 +++--
 net/sched/sch_dsmark.c    |  2 +-
 net/sched/sch_hfsc.c      |  2 +-
 net/sched/sch_htb.c       |  3 +--
 net/sched/sch_netem.c     | 20 ++++++++++++--------
 net/sched/sch_prio.c      |  3 ++-
 net/sched/sch_red.c       |  2 +-
 net/sched/sch_tbf.c       |  3 ++-
 12 files changed, 37 insertions(+), 21 deletions(-)

(limited to 'include')

diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h
index 8a44386b35cf..f396dff335a3 100644
--- a/include/net/sch_generic.h
+++ b/include/net/sch_generic.h
@@ -306,6 +306,16 @@ static inline bool qdisc_tx_is_noop(const struct net_device *dev)
 	return true;
 }
 
+static inline int qdisc_enqueue(struct sk_buff *skb, struct Qdisc *sch)
+{
+	return sch->enqueue(skb, sch);
+}
+
+static inline int qdisc_enqueue_root(struct sk_buff *skb, struct Qdisc *sch)
+{
+	return qdisc_enqueue(skb, sch);
+}
+
 static inline int __qdisc_enqueue_tail(struct sk_buff *skb, struct Qdisc *sch,
 				       struct sk_buff_head *list)
 {
diff --git a/net/core/dev.c b/net/core/dev.c
index 065b9817e209..2eed17bcb2dd 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -1781,7 +1781,7 @@ gso:
 
 		spin_lock(root_lock);
 
-		rc = q->enqueue(skb, q);
+		rc = qdisc_enqueue_root(skb, q);
 		qdisc_run(q);
 
 		spin_unlock(root_lock);
@@ -2083,7 +2083,7 @@ static int ing_filter(struct sk_buff *skb)
 	q = rxq->qdisc;
 	if (q) {
 		spin_lock(qdisc_lock(q));
-		result = q->enqueue(skb, q);
+		result = qdisc_enqueue_root(skb, q);
 		spin_unlock(qdisc_lock(q));
 	}
 
diff --git a/net/mac80211/wme.c b/net/mac80211/wme.c
index 6e8099e77043..07edda0b8a5c 100644
--- a/net/mac80211/wme.c
+++ b/net/mac80211/wme.c
@@ -289,7 +289,7 @@ void ieee80211_requeue(struct ieee80211_local *local, int queue)
 		root_lock = qdisc_root_lock(qdisc);
 
 		spin_lock(root_lock);
-		qdisc->enqueue(skb, qdisc);
+		qdisc_enqueue_root(skb, qdisc);
 		spin_unlock(root_lock);
 	}
 
diff --git a/net/sched/sch_atm.c b/net/sched/sch_atm.c
index 0de757e3be4a..68ed35e2a763 100644
--- a/net/sched/sch_atm.c
+++ b/net/sched/sch_atm.c
@@ -429,7 +429,7 @@ static int atm_tc_enqueue(struct sk_buff *skb, struct Qdisc *sch)
 #endif
 	}
 
-	ret = flow->q->enqueue(skb, flow->q);
+	ret = qdisc_enqueue(skb, flow->q);
 	if (ret != 0) {
 drop: __maybe_unused
 		sch->qstats.drops++;
diff --git a/net/sched/sch_cbq.c b/net/sched/sch_cbq.c
index a3953bbe2d79..1afe3eece627 100644
--- a/net/sched/sch_cbq.c
+++ b/net/sched/sch_cbq.c
@@ -387,7 +387,8 @@ cbq_enqueue(struct sk_buff *skb, struct Qdisc *sch)
 #ifdef CONFIG_NET_CLS_ACT
 	cl->q->__parent = sch;
 #endif
-	if ((ret = cl->q->enqueue(skb, cl->q)) == NET_XMIT_SUCCESS) {
+	ret = qdisc_enqueue(skb, cl->q);
+	if (ret == NET_XMIT_SUCCESS) {
 		sch->q.qlen++;
 		sch->bstats.packets++;
 		sch->bstats.bytes+=len;
@@ -671,7 +672,7 @@ static int cbq_reshape_fail(struct sk_buff *skb, struct Qdisc *child)
 		q->rx_class = cl;
 		cl->q->__parent = sch;
 
-		if (cl->q->enqueue(skb, cl->q) == 0) {
+		if (qdisc_enqueue(skb, cl->q) == 0) {
 			sch->q.qlen++;
 			sch->bstats.packets++;
 			sch->bstats.bytes+=len;
diff --git a/net/sched/sch_dsmark.c b/net/sched/sch_dsmark.c
index 3aafbd17393a..44d347e831cf 100644
--- a/net/sched/sch_dsmark.c
+++ b/net/sched/sch_dsmark.c
@@ -252,7 +252,7 @@ static int dsmark_enqueue(struct sk_buff *skb, struct Qdisc *sch)
 		}
 	}
 
-	err = p->q->enqueue(skb, p->q);
+	err = qdisc_enqueue(skb, p->q);
 	if (err != NET_XMIT_SUCCESS) {
 		sch->qstats.drops++;
 		return err;
diff --git a/net/sched/sch_hfsc.c b/net/sched/sch_hfsc.c
index 5090708ba384..fd61ed6ee1e7 100644
--- a/net/sched/sch_hfsc.c
+++ b/net/sched/sch_hfsc.c
@@ -1586,7 +1586,7 @@ hfsc_enqueue(struct sk_buff *skb, struct Qdisc *sch)
 	}
 
 	len = skb->len;
-	err = cl->qdisc->enqueue(skb, cl->qdisc);
+	err = qdisc_enqueue(skb, cl->qdisc);
 	if (unlikely(err != NET_XMIT_SUCCESS)) {
 		cl->qstats.drops++;
 		sch->qstats.drops++;
diff --git a/net/sched/sch_htb.c b/net/sched/sch_htb.c
index ee48457eaa4a..72b5a946178f 100644
--- a/net/sched/sch_htb.c
+++ b/net/sched/sch_htb.c
@@ -572,8 +572,7 @@ static int htb_enqueue(struct sk_buff *skb, struct Qdisc *sch)
 		kfree_skb(skb);
 		return ret;
 #endif
-	} else if (cl->un.leaf.q->enqueue(skb, cl->un.leaf.q) !=
-		   NET_XMIT_SUCCESS) {
+	} else if (qdisc_enqueue(skb, cl->un.leaf.q) != NET_XMIT_SUCCESS) {
 		sch->qstats.drops++;
 		cl->qstats.drops++;
 		return NET_XMIT_DROP;
diff --git a/net/sched/sch_netem.c b/net/sched/sch_netem.c
index c5ea40c9eb21..13c4821e42b8 100644
--- a/net/sched/sch_netem.c
+++ b/net/sched/sch_netem.c
@@ -82,6 +82,12 @@ struct netem_skb_cb {
 	psched_time_t	time_to_send;
 };
 
+static inline struct netem_skb_cb *netem_skb_cb(struct sk_buff *skb)
+{
+	BUILD_BUG_ON(sizeof(skb->cb) < sizeof(struct netem_skb_cb));
+	return (struct netem_skb_cb *)skb->cb;
+}
+
 /* init_crandom - initialize correlated random number generator
  * Use entropy source for initial seed.
  */
@@ -184,7 +190,7 @@ static int netem_enqueue(struct sk_buff *skb, struct Qdisc *sch)
 		u32 dupsave = q->duplicate; /* prevent duplicating a dup... */
 		q->duplicate = 0;
 
-		rootq->enqueue(skb2, rootq);
+		qdisc_enqueue_root(skb2, rootq);
 		q->duplicate = dupsave;
 	}
 
@@ -205,7 +211,7 @@ static int netem_enqueue(struct sk_buff *skb, struct Qdisc *sch)
 		skb->data[net_random() % skb_headlen(skb)] ^= 1<<(net_random() % 8);
 	}
 
-	cb = (struct netem_skb_cb *)skb->cb;
+	cb = netem_skb_cb(skb);
 	if (q->gap == 0 		/* not doing reordering */
 	    || q->counter < q->gap 	/* inside last reordering gap */
 	    || q->reorder < get_crandom(&q->reorder_cor)) {
@@ -218,7 +224,7 @@ static int netem_enqueue(struct sk_buff *skb, struct Qdisc *sch)
 		now = psched_get_time();
 		cb->time_to_send = now + delay;
 		++q->counter;
-		ret = q->qdisc->enqueue(skb, q->qdisc);
+		ret = qdisc_enqueue(skb, q->qdisc);
 	} else {
 		/*
 		 * Do re-ordering by putting one out of N packets at the front
@@ -277,8 +283,7 @@ static struct sk_buff *netem_dequeue(struct Qdisc *sch)
 
 	skb = q->qdisc->dequeue(q->qdisc);
 	if (skb) {
-		const struct netem_skb_cb *cb
-			= (const struct netem_skb_cb *)skb->cb;
+		const struct netem_skb_cb *cb = netem_skb_cb(skb);
 		psched_time_t now = psched_get_time();
 
 		/* if more time remaining? */
@@ -457,7 +462,7 @@ static int tfifo_enqueue(struct sk_buff *nskb, struct Qdisc *sch)
 {
 	struct fifo_sched_data *q = qdisc_priv(sch);
 	struct sk_buff_head *list = &sch->q;
-	psched_time_t tnext = ((struct netem_skb_cb *)nskb->cb)->time_to_send;
+	psched_time_t tnext = netem_skb_cb(nskb)->time_to_send;
 	struct sk_buff *skb;
 
 	if (likely(skb_queue_len(list) < q->limit)) {
@@ -468,8 +473,7 @@ static int tfifo_enqueue(struct sk_buff *nskb, struct Qdisc *sch)
 		}
 
 		skb_queue_reverse_walk(list, skb) {
-			const struct netem_skb_cb *cb
-				= (const struct netem_skb_cb *)skb->cb;
+			const struct netem_skb_cb *cb = netem_skb_cb(skb);
 
 			if (tnext >= cb->time_to_send)
 				break;
diff --git a/net/sched/sch_prio.c b/net/sched/sch_prio.c
index 536ca474dc69..d29c2f87fc0b 100644
--- a/net/sched/sch_prio.c
+++ b/net/sched/sch_prio.c
@@ -81,7 +81,8 @@ prio_enqueue(struct sk_buff *skb, struct Qdisc *sch)
 	}
 #endif
 
-	if ((ret = qdisc->enqueue(skb, qdisc)) == NET_XMIT_SUCCESS) {
+	ret = qdisc_enqueue(skb, qdisc);
+	if (ret == NET_XMIT_SUCCESS) {
 		sch->bstats.bytes += skb->len;
 		sch->bstats.packets++;
 		sch->q.qlen++;
diff --git a/net/sched/sch_red.c b/net/sched/sch_red.c
index 77098acf0adc..b48a391bc129 100644
--- a/net/sched/sch_red.c
+++ b/net/sched/sch_red.c
@@ -92,7 +92,7 @@ static int red_enqueue(struct sk_buff *skb, struct Qdisc* sch)
 			break;
 	}
 
-	ret = child->enqueue(skb, child);
+	ret = qdisc_enqueue(skb, child);
 	if (likely(ret == NET_XMIT_SUCCESS)) {
 		sch->bstats.bytes += skb->len;
 		sch->bstats.packets++;
diff --git a/net/sched/sch_tbf.c b/net/sched/sch_tbf.c
index 444c227fcb6b..7d705b86dae5 100644
--- a/net/sched/sch_tbf.c
+++ b/net/sched/sch_tbf.c
@@ -133,7 +133,8 @@ static int tbf_enqueue(struct sk_buff *skb, struct Qdisc* sch)
 		return NET_XMIT_DROP;
 	}
 
-	if ((ret = q->qdisc->enqueue(skb, q->qdisc)) != 0) {
+	ret = qdisc_enqueue(skb, q->qdisc);
+	if (ret != 0) {
 		sch->qstats.drops++;
 		return ret;
 	}
-- 
cgit v1.2.3


From 0abf77e55a2459aa9905be4b226e4729d5b4f0cb Mon Sep 17 00:00:00 2001
From: Jussi Kivilinna <jussi.kivilinna@mbnet.fi>
Date: Sun, 20 Jul 2008 00:08:27 -0700
Subject: net_sched: Add accessor function for packet length for qdiscs

Signed-off-by: Jussi Kivilinna <jussi.kivilinna@mbnet.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/sch_generic.h | 17 +++++++++++------
 net/sched/act_gact.c      |  2 +-
 net/sched/act_ipt.c       |  2 +-
 net/sched/act_mirred.c    |  4 ++--
 net/sched/act_nat.c       |  2 +-
 net/sched/act_pedit.c     |  2 +-
 net/sched/act_police.c    |  8 ++++----
 net/sched/act_simple.c    |  2 +-
 net/sched/sch_atm.c       |  4 ++--
 net/sched/sch_cbq.c       | 14 ++++++--------
 net/sched/sch_dsmark.c    |  2 +-
 net/sched/sch_fifo.c      |  2 +-
 net/sched/sch_gred.c      | 12 ++++++------
 net/sched/sch_hfsc.c      | 14 ++++++--------
 net/sched/sch_htb.c       |  9 +++++----
 net/sched/sch_ingress.c   |  2 +-
 net/sched/sch_netem.c     |  6 +++---
 net/sched/sch_prio.c      |  2 +-
 net/sched/sch_red.c       |  2 +-
 net/sched/sch_sfq.c       | 16 ++++++++--------
 net/sched/sch_tbf.c       |  6 +++---
 net/sched/sch_teql.c      |  6 +++---
 22 files changed, 69 insertions(+), 67 deletions(-)

(limited to 'include')

diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h
index f396dff335a3..8229520e088a 100644
--- a/include/net/sch_generic.h
+++ b/include/net/sch_generic.h
@@ -306,6 +306,11 @@ static inline bool qdisc_tx_is_noop(const struct net_device *dev)
 	return true;
 }
 
+static inline unsigned int qdisc_pkt_len(struct sk_buff *skb)
+{
+	return skb->len;
+}
+
 static inline int qdisc_enqueue(struct sk_buff *skb, struct Qdisc *sch)
 {
 	return sch->enqueue(skb, sch);
@@ -320,8 +325,8 @@ static inline int __qdisc_enqueue_tail(struct sk_buff *skb, struct Qdisc *sch,
 				       struct sk_buff_head *list)
 {
 	__skb_queue_tail(list, skb);
-	sch->qstats.backlog += skb->len;
-	sch->bstats.bytes += skb->len;
+	sch->qstats.backlog += qdisc_pkt_len(skb);
+	sch->bstats.bytes += qdisc_pkt_len(skb);
 	sch->bstats.packets++;
 
 	return NET_XMIT_SUCCESS;
@@ -338,7 +343,7 @@ static inline struct sk_buff *__qdisc_dequeue_head(struct Qdisc *sch,
 	struct sk_buff *skb = __skb_dequeue(list);
 
 	if (likely(skb != NULL))
-		sch->qstats.backlog -= skb->len;
+		sch->qstats.backlog -= qdisc_pkt_len(skb);
 
 	return skb;
 }
@@ -354,7 +359,7 @@ static inline struct sk_buff *__qdisc_dequeue_tail(struct Qdisc *sch,
 	struct sk_buff *skb = __skb_dequeue_tail(list);
 
 	if (likely(skb != NULL))
-		sch->qstats.backlog -= skb->len;
+		sch->qstats.backlog -= qdisc_pkt_len(skb);
 
 	return skb;
 }
@@ -368,7 +373,7 @@ static inline int __qdisc_requeue(struct sk_buff *skb, struct Qdisc *sch,
 				  struct sk_buff_head *list)
 {
 	__skb_queue_head(list, skb);
-	sch->qstats.backlog += skb->len;
+	sch->qstats.backlog += qdisc_pkt_len(skb);
 	sch->qstats.requeues++;
 
 	return NET_XMIT_SUCCESS;
@@ -401,7 +406,7 @@ static inline unsigned int __qdisc_queue_drop(struct Qdisc *sch,
 	struct sk_buff *skb = __qdisc_dequeue_tail(sch, list);
 
 	if (likely(skb != NULL)) {
-		unsigned int len = skb->len;
+		unsigned int len = qdisc_pkt_len(skb);
 		kfree_skb(skb);
 		return len;
 	}
diff --git a/net/sched/act_gact.c b/net/sched/act_gact.c
index 422872c4f14b..ac04289da5d7 100644
--- a/net/sched/act_gact.c
+++ b/net/sched/act_gact.c
@@ -139,7 +139,7 @@ static int tcf_gact(struct sk_buff *skb, struct tc_action *a, struct tcf_result
 #else
 	action = gact->tcf_action;
 #endif
-	gact->tcf_bstats.bytes += skb->len;
+	gact->tcf_bstats.bytes += qdisc_pkt_len(skb);
 	gact->tcf_bstats.packets++;
 	if (action == TC_ACT_SHOT)
 		gact->tcf_qstats.drops++;
diff --git a/net/sched/act_ipt.c b/net/sched/act_ipt.c
index da696fd3e341..d1263b3c96c3 100644
--- a/net/sched/act_ipt.c
+++ b/net/sched/act_ipt.c
@@ -205,7 +205,7 @@ static int tcf_ipt(struct sk_buff *skb, struct tc_action *a,
 	spin_lock(&ipt->tcf_lock);
 
 	ipt->tcf_tm.lastuse = jiffies;
-	ipt->tcf_bstats.bytes += skb->len;
+	ipt->tcf_bstats.bytes += qdisc_pkt_len(skb);
 	ipt->tcf_bstats.packets++;
 
 	/* yes, we have to worry about both in and out dev
diff --git a/net/sched/act_mirred.c b/net/sched/act_mirred.c
index 1aff005d95cd..70341c020b6d 100644
--- a/net/sched/act_mirred.c
+++ b/net/sched/act_mirred.c
@@ -164,7 +164,7 @@ bad_mirred:
 		if (skb2 != NULL)
 			kfree_skb(skb2);
 		m->tcf_qstats.overlimits++;
-		m->tcf_bstats.bytes += skb->len;
+		m->tcf_bstats.bytes += qdisc_pkt_len(skb);
 		m->tcf_bstats.packets++;
 		spin_unlock(&m->tcf_lock);
 		/* should we be asking for packet to be dropped?
@@ -184,7 +184,7 @@ bad_mirred:
 		goto bad_mirred;
 	}
 
-	m->tcf_bstats.bytes += skb2->len;
+	m->tcf_bstats.bytes += qdisc_pkt_len(skb2);
 	m->tcf_bstats.packets++;
 	if (!(at & AT_EGRESS))
 		if (m->tcfm_ok_push)
diff --git a/net/sched/act_nat.c b/net/sched/act_nat.c
index 0a3c8339767a..7b39ed485bca 100644
--- a/net/sched/act_nat.c
+++ b/net/sched/act_nat.c
@@ -124,7 +124,7 @@ static int tcf_nat(struct sk_buff *skb, struct tc_action *a,
 	egress = p->flags & TCA_NAT_FLAG_EGRESS;
 	action = p->tcf_action;
 
-	p->tcf_bstats.bytes += skb->len;
+	p->tcf_bstats.bytes += qdisc_pkt_len(skb);
 	p->tcf_bstats.packets++;
 
 	spin_unlock(&p->tcf_lock);
diff --git a/net/sched/act_pedit.c b/net/sched/act_pedit.c
index 3cc4cb9e500e..d5f4e3404864 100644
--- a/net/sched/act_pedit.c
+++ b/net/sched/act_pedit.c
@@ -182,7 +182,7 @@ static int tcf_pedit(struct sk_buff *skb, struct tc_action *a,
 bad:
 	p->tcf_qstats.overlimits++;
 done:
-	p->tcf_bstats.bytes += skb->len;
+	p->tcf_bstats.bytes += qdisc_pkt_len(skb);
 	p->tcf_bstats.packets++;
 	spin_unlock(&p->tcf_lock);
 	return p->tcf_action;
diff --git a/net/sched/act_police.c b/net/sched/act_police.c
index 0898120bbcc0..32c3f9d9fb7a 100644
--- a/net/sched/act_police.c
+++ b/net/sched/act_police.c
@@ -272,7 +272,7 @@ static int tcf_act_police(struct sk_buff *skb, struct tc_action *a,
 
 	spin_lock(&police->tcf_lock);
 
-	police->tcf_bstats.bytes += skb->len;
+	police->tcf_bstats.bytes += qdisc_pkt_len(skb);
 	police->tcf_bstats.packets++;
 
 	if (police->tcfp_ewma_rate &&
@@ -282,7 +282,7 @@ static int tcf_act_police(struct sk_buff *skb, struct tc_action *a,
 		return police->tcf_action;
 	}
 
-	if (skb->len <= police->tcfp_mtu) {
+	if (qdisc_pkt_len(skb) <= police->tcfp_mtu) {
 		if (police->tcfp_R_tab == NULL) {
 			spin_unlock(&police->tcf_lock);
 			return police->tcfp_result;
@@ -295,12 +295,12 @@ static int tcf_act_police(struct sk_buff *skb, struct tc_action *a,
 			ptoks = toks + police->tcfp_ptoks;
 			if (ptoks > (long)L2T_P(police, police->tcfp_mtu))
 				ptoks = (long)L2T_P(police, police->tcfp_mtu);
-			ptoks -= L2T_P(police, skb->len);
+			ptoks -= L2T_P(police, qdisc_pkt_len(skb));
 		}
 		toks += police->tcfp_toks;
 		if (toks > (long)police->tcfp_burst)
 			toks = police->tcfp_burst;
-		toks -= L2T(police, skb->len);
+		toks -= L2T(police, qdisc_pkt_len(skb));
 		if ((toks|ptoks) >= 0) {
 			police->tcfp_t_c = now;
 			police->tcfp_toks = toks;
diff --git a/net/sched/act_simple.c b/net/sched/act_simple.c
index 1d421d059caf..e7851ce92cfe 100644
--- a/net/sched/act_simple.c
+++ b/net/sched/act_simple.c
@@ -41,7 +41,7 @@ static int tcf_simp(struct sk_buff *skb, struct tc_action *a, struct tcf_result
 
 	spin_lock(&d->tcf_lock);
 	d->tcf_tm.lastuse = jiffies;
-	d->tcf_bstats.bytes += skb->len;
+	d->tcf_bstats.bytes += qdisc_pkt_len(skb);
 	d->tcf_bstats.packets++;
 
 	/* print policy string followed by _ then packet count
diff --git a/net/sched/sch_atm.c b/net/sched/sch_atm.c
index 68ed35e2a763..04faa835be17 100644
--- a/net/sched/sch_atm.c
+++ b/net/sched/sch_atm.c
@@ -437,9 +437,9 @@ drop: __maybe_unused
 			flow->qstats.drops++;
 		return ret;
 	}
-	sch->bstats.bytes += skb->len;
+	sch->bstats.bytes += qdisc_pkt_len(skb);
 	sch->bstats.packets++;
-	flow->bstats.bytes += skb->len;
+	flow->bstats.bytes += qdisc_pkt_len(skb);
 	flow->bstats.packets++;
 	/*
 	 * Okay, this may seem weird. We pretend we've dropped the packet if
diff --git a/net/sched/sch_cbq.c b/net/sched/sch_cbq.c
index 1afe3eece627..f1d2f8ec8b4c 100644
--- a/net/sched/sch_cbq.c
+++ b/net/sched/sch_cbq.c
@@ -370,7 +370,6 @@ static int
 cbq_enqueue(struct sk_buff *skb, struct Qdisc *sch)
 {
 	struct cbq_sched_data *q = qdisc_priv(sch);
-	int len = skb->len;
 	int uninitialized_var(ret);
 	struct cbq_class *cl = cbq_classify(skb, sch, &ret);
 
@@ -391,7 +390,7 @@ cbq_enqueue(struct sk_buff *skb, struct Qdisc *sch)
 	if (ret == NET_XMIT_SUCCESS) {
 		sch->q.qlen++;
 		sch->bstats.packets++;
-		sch->bstats.bytes+=len;
+		sch->bstats.bytes += qdisc_pkt_len(skb);
 		cbq_mark_toplevel(q, cl);
 		if (!cl->next_alive)
 			cbq_activate_class(cl);
@@ -658,7 +657,6 @@ static enum hrtimer_restart cbq_undelay(struct hrtimer *timer)
 #ifdef CONFIG_NET_CLS_ACT
 static int cbq_reshape_fail(struct sk_buff *skb, struct Qdisc *child)
 {
-	int len = skb->len;
 	struct Qdisc *sch = child->__parent;
 	struct cbq_sched_data *q = qdisc_priv(sch);
 	struct cbq_class *cl = q->rx_class;
@@ -675,7 +673,7 @@ static int cbq_reshape_fail(struct sk_buff *skb, struct Qdisc *child)
 		if (qdisc_enqueue(skb, cl->q) == 0) {
 			sch->q.qlen++;
 			sch->bstats.packets++;
-			sch->bstats.bytes+=len;
+			sch->bstats.bytes += qdisc_pkt_len(skb);
 			if (!cl->next_alive)
 				cbq_activate_class(cl);
 			return 0;
@@ -881,7 +879,7 @@ cbq_dequeue_prio(struct Qdisc *sch, int prio)
 			if (skb == NULL)
 				goto skip_class;
 
-			cl->deficit -= skb->len;
+			cl->deficit -= qdisc_pkt_len(skb);
 			q->tx_class = cl;
 			q->tx_borrowed = borrow;
 			if (borrow != cl) {
@@ -889,11 +887,11 @@ cbq_dequeue_prio(struct Qdisc *sch, int prio)
 				borrow->xstats.borrows++;
 				cl->xstats.borrows++;
 #else
-				borrow->xstats.borrows += skb->len;
-				cl->xstats.borrows += skb->len;
+				borrow->xstats.borrows += qdisc_pkt_len(skb);
+				cl->xstats.borrows += qdisc_pkt_len(skb);
 #endif
 			}
-			q->tx_len = skb->len;
+			q->tx_len = qdisc_pkt_len(skb);
 
 			if (cl->deficit <= 0) {
 				q->active[prio] = cl;
diff --git a/net/sched/sch_dsmark.c b/net/sched/sch_dsmark.c
index 44d347e831cf..a935676987e2 100644
--- a/net/sched/sch_dsmark.c
+++ b/net/sched/sch_dsmark.c
@@ -258,7 +258,7 @@ static int dsmark_enqueue(struct sk_buff *skb, struct Qdisc *sch)
 		return err;
 	}
 
-	sch->bstats.bytes += skb->len;
+	sch->bstats.bytes += qdisc_pkt_len(skb);
 	sch->bstats.packets++;
 	sch->q.qlen++;
 
diff --git a/net/sched/sch_fifo.c b/net/sched/sch_fifo.c
index 1d97fa42c902..23d258bfe8ac 100644
--- a/net/sched/sch_fifo.c
+++ b/net/sched/sch_fifo.c
@@ -27,7 +27,7 @@ static int bfifo_enqueue(struct sk_buff *skb, struct Qdisc* sch)
 {
 	struct fifo_sched_data *q = qdisc_priv(sch);
 
-	if (likely(sch->qstats.backlog + skb->len <= q->limit))
+	if (likely(sch->qstats.backlog + qdisc_pkt_len(skb) <= q->limit))
 		return qdisc_enqueue_tail(skb, sch);
 
 	return qdisc_reshape_fail(skb, sch);
diff --git a/net/sched/sch_gred.c b/net/sched/sch_gred.c
index 39fa28511f07..c1ad6b8de105 100644
--- a/net/sched/sch_gred.c
+++ b/net/sched/sch_gred.c
@@ -188,7 +188,7 @@ static int gred_enqueue(struct sk_buff *skb, struct Qdisc* sch)
 	}
 
 	q->packetsin++;
-	q->bytesin += skb->len;
+	q->bytesin += qdisc_pkt_len(skb);
 
 	if (gred_wred_mode(t))
 		gred_load_wred_set(t, q);
@@ -226,8 +226,8 @@ static int gred_enqueue(struct sk_buff *skb, struct Qdisc* sch)
 			break;
 	}
 
-	if (q->backlog + skb->len <= q->limit) {
-		q->backlog += skb->len;
+	if (q->backlog + qdisc_pkt_len(skb) <= q->limit) {
+		q->backlog += qdisc_pkt_len(skb);
 		return qdisc_enqueue_tail(skb, sch);
 	}
 
@@ -254,7 +254,7 @@ static int gred_requeue(struct sk_buff *skb, struct Qdisc* sch)
 	} else {
 		if (red_is_idling(&q->parms))
 			red_end_of_idle_period(&q->parms);
-		q->backlog += skb->len;
+		q->backlog += qdisc_pkt_len(skb);
 	}
 
 	return qdisc_requeue(skb, sch);
@@ -277,7 +277,7 @@ static struct sk_buff *gred_dequeue(struct Qdisc* sch)
 				       "VQ 0x%x after dequeue, screwing up "
 				       "backlog.\n", tc_index_to_dp(skb));
 		} else {
-			q->backlog -= skb->len;
+			q->backlog -= qdisc_pkt_len(skb);
 
 			if (!q->backlog && !gred_wred_mode(t))
 				red_start_of_idle_period(&q->parms);
@@ -299,7 +299,7 @@ static unsigned int gred_drop(struct Qdisc* sch)
 
 	skb = qdisc_dequeue_tail(sch);
 	if (skb) {
-		unsigned int len = skb->len;
+		unsigned int len = qdisc_pkt_len(skb);
 		struct gred_sched_data *q;
 		u16 dp = tc_index_to_dp(skb);
 
diff --git a/net/sched/sch_hfsc.c b/net/sched/sch_hfsc.c
index fd61ed6ee1e7..0ae7d19dcba8 100644
--- a/net/sched/sch_hfsc.c
+++ b/net/sched/sch_hfsc.c
@@ -895,7 +895,7 @@ qdisc_peek_len(struct Qdisc *sch)
 			printk("qdisc_peek_len: non work-conserving qdisc ?\n");
 		return 0;
 	}
-	len = skb->len;
+	len = qdisc_pkt_len(skb);
 	if (unlikely(sch->ops->requeue(skb, sch) != NET_XMIT_SUCCESS)) {
 		if (net_ratelimit())
 			printk("qdisc_peek_len: failed to requeue\n");
@@ -1574,7 +1574,6 @@ static int
 hfsc_enqueue(struct sk_buff *skb, struct Qdisc *sch)
 {
 	struct hfsc_class *cl;
-	unsigned int len;
 	int err;
 
 	cl = hfsc_classify(skb, sch, &err);
@@ -1585,7 +1584,6 @@ hfsc_enqueue(struct sk_buff *skb, struct Qdisc *sch)
 		return err;
 	}
 
-	len = skb->len;
 	err = qdisc_enqueue(skb, cl->qdisc);
 	if (unlikely(err != NET_XMIT_SUCCESS)) {
 		cl->qstats.drops++;
@@ -1594,12 +1592,12 @@ hfsc_enqueue(struct sk_buff *skb, struct Qdisc *sch)
 	}
 
 	if (cl->qdisc->q.qlen == 1)
-		set_active(cl, len);
+		set_active(cl, qdisc_pkt_len(skb));
 
 	cl->bstats.packets++;
-	cl->bstats.bytes += len;
+	cl->bstats.bytes += qdisc_pkt_len(skb);
 	sch->bstats.packets++;
-	sch->bstats.bytes += len;
+	sch->bstats.bytes += qdisc_pkt_len(skb);
 	sch->q.qlen++;
 
 	return NET_XMIT_SUCCESS;
@@ -1649,9 +1647,9 @@ hfsc_dequeue(struct Qdisc *sch)
 		return NULL;
 	}
 
-	update_vf(cl, skb->len, cur_time);
+	update_vf(cl, qdisc_pkt_len(skb), cur_time);
 	if (realtime)
-		cl->cl_cumul += skb->len;
+		cl->cl_cumul += qdisc_pkt_len(skb);
 
 	if (cl->qdisc->q.qlen != 0) {
 		if (cl->cl_flags & HFSC_RSC) {
diff --git a/net/sched/sch_htb.c b/net/sched/sch_htb.c
index 72b5a946178f..30c999c61b01 100644
--- a/net/sched/sch_htb.c
+++ b/net/sched/sch_htb.c
@@ -579,13 +579,13 @@ static int htb_enqueue(struct sk_buff *skb, struct Qdisc *sch)
 	} else {
 		cl->bstats.packets +=
 			skb_is_gso(skb)?skb_shinfo(skb)->gso_segs:1;
-		cl->bstats.bytes += skb->len;
+		cl->bstats.bytes += qdisc_pkt_len(skb);
 		htb_activate(q, cl);
 	}
 
 	sch->q.qlen++;
 	sch->bstats.packets += skb_is_gso(skb)?skb_shinfo(skb)->gso_segs:1;
-	sch->bstats.bytes += skb->len;
+	sch->bstats.bytes += qdisc_pkt_len(skb);
 	return NET_XMIT_SUCCESS;
 }
 
@@ -642,7 +642,7 @@ static int htb_requeue(struct sk_buff *skb, struct Qdisc *sch)
 static void htb_charge_class(struct htb_sched *q, struct htb_class *cl,
 			     int level, struct sk_buff *skb)
 {
-	int bytes = skb->len;
+	int bytes = qdisc_pkt_len(skb);
 	long toks, diff;
 	enum htb_cmode old_mode;
 
@@ -855,7 +855,8 @@ next:
 	} while (cl != start);
 
 	if (likely(skb != NULL)) {
-		if ((cl->un.leaf.deficit[level] -= skb->len) < 0) {
+		cl->un.leaf.deficit[level] -= qdisc_pkt_len(skb);
+		if (cl->un.leaf.deficit[level] < 0) {
 			cl->un.leaf.deficit[level] += cl->un.leaf.quantum;
 			htb_next_rb_node((level ? cl->parent->un.inner.ptr : q->
 					  ptr[0]) + prio);
diff --git a/net/sched/sch_ingress.c b/net/sched/sch_ingress.c
index 956c80ad5965..4a2b77374358 100644
--- a/net/sched/sch_ingress.c
+++ b/net/sched/sch_ingress.c
@@ -77,7 +77,7 @@ static int ingress_enqueue(struct sk_buff *skb, struct Qdisc *sch)
 	result = tc_classify(skb, p->filter_list, &res);
 
 	sch->bstats.packets++;
-	sch->bstats.bytes += skb->len;
+	sch->bstats.bytes += qdisc_pkt_len(skb);
 	switch (result) {
 	case TC_ACT_SHOT:
 		result = TC_ACT_SHOT;
diff --git a/net/sched/sch_netem.c b/net/sched/sch_netem.c
index 13c4821e42b8..ae49be00022f 100644
--- a/net/sched/sch_netem.c
+++ b/net/sched/sch_netem.c
@@ -237,7 +237,7 @@ static int netem_enqueue(struct sk_buff *skb, struct Qdisc *sch)
 
 	if (likely(ret == NET_XMIT_SUCCESS)) {
 		sch->q.qlen++;
-		sch->bstats.bytes += skb->len;
+		sch->bstats.bytes += qdisc_pkt_len(skb);
 		sch->bstats.packets++;
 	} else
 		sch->qstats.drops++;
@@ -481,8 +481,8 @@ static int tfifo_enqueue(struct sk_buff *nskb, struct Qdisc *sch)
 
 		__skb_queue_after(list, skb, nskb);
 
-		sch->qstats.backlog += nskb->len;
-		sch->bstats.bytes += nskb->len;
+		sch->qstats.backlog += qdisc_pkt_len(nskb);
+		sch->bstats.bytes += qdisc_pkt_len(nskb);
 		sch->bstats.packets++;
 
 		return NET_XMIT_SUCCESS;
diff --git a/net/sched/sch_prio.c b/net/sched/sch_prio.c
index d29c2f87fc0b..f849243eb095 100644
--- a/net/sched/sch_prio.c
+++ b/net/sched/sch_prio.c
@@ -83,7 +83,7 @@ prio_enqueue(struct sk_buff *skb, struct Qdisc *sch)
 
 	ret = qdisc_enqueue(skb, qdisc);
 	if (ret == NET_XMIT_SUCCESS) {
-		sch->bstats.bytes += skb->len;
+		sch->bstats.bytes += qdisc_pkt_len(skb);
 		sch->bstats.packets++;
 		sch->q.qlen++;
 		return NET_XMIT_SUCCESS;
diff --git a/net/sched/sch_red.c b/net/sched/sch_red.c
index b48a391bc129..3f2d1d7f3bbd 100644
--- a/net/sched/sch_red.c
+++ b/net/sched/sch_red.c
@@ -94,7 +94,7 @@ static int red_enqueue(struct sk_buff *skb, struct Qdisc* sch)
 
 	ret = qdisc_enqueue(skb, child);
 	if (likely(ret == NET_XMIT_SUCCESS)) {
-		sch->bstats.bytes += skb->len;
+		sch->bstats.bytes += qdisc_pkt_len(skb);
 		sch->bstats.packets++;
 		sch->q.qlen++;
 	} else {
diff --git a/net/sched/sch_sfq.c b/net/sched/sch_sfq.c
index 8458f630fac4..8589da666568 100644
--- a/net/sched/sch_sfq.c
+++ b/net/sched/sch_sfq.c
@@ -245,7 +245,7 @@ static unsigned int sfq_drop(struct Qdisc *sch)
 	if (d > 1) {
 		sfq_index x = q->dep[d + SFQ_DEPTH].next;
 		skb = q->qs[x].prev;
-		len = skb->len;
+		len = qdisc_pkt_len(skb);
 		__skb_unlink(skb, &q->qs[x]);
 		kfree_skb(skb);
 		sfq_dec(q, x);
@@ -261,7 +261,7 @@ static unsigned int sfq_drop(struct Qdisc *sch)
 		q->next[q->tail] = q->next[d];
 		q->allot[q->next[d]] += q->quantum;
 		skb = q->qs[d].prev;
-		len = skb->len;
+		len = qdisc_pkt_len(skb);
 		__skb_unlink(skb, &q->qs[d]);
 		kfree_skb(skb);
 		sfq_dec(q, d);
@@ -305,7 +305,7 @@ sfq_enqueue(struct sk_buff *skb, struct Qdisc *sch)
 	if (q->qs[x].qlen >= q->limit)
 		return qdisc_drop(skb, sch);
 
-	sch->qstats.backlog += skb->len;
+	sch->qstats.backlog += qdisc_pkt_len(skb);
 	__skb_queue_tail(&q->qs[x], skb);
 	sfq_inc(q, x);
 	if (q->qs[x].qlen == 1) {		/* The flow is new */
@@ -320,7 +320,7 @@ sfq_enqueue(struct sk_buff *skb, struct Qdisc *sch)
 		}
 	}
 	if (++sch->q.qlen <= q->limit) {
-		sch->bstats.bytes += skb->len;
+		sch->bstats.bytes += qdisc_pkt_len(skb);
 		sch->bstats.packets++;
 		return 0;
 	}
@@ -352,7 +352,7 @@ sfq_requeue(struct sk_buff *skb, struct Qdisc *sch)
 		q->hash[x] = hash;
 	}
 
-	sch->qstats.backlog += skb->len;
+	sch->qstats.backlog += qdisc_pkt_len(skb);
 	__skb_queue_head(&q->qs[x], skb);
 	/* If selected queue has length q->limit+1, this means that
 	 * all another queues are empty and we do simple tail drop.
@@ -363,7 +363,7 @@ sfq_requeue(struct sk_buff *skb, struct Qdisc *sch)
 		skb = q->qs[x].prev;
 		__skb_unlink(skb, &q->qs[x]);
 		sch->qstats.drops++;
-		sch->qstats.backlog -= skb->len;
+		sch->qstats.backlog -= qdisc_pkt_len(skb);
 		kfree_skb(skb);
 		return NET_XMIT_CN;
 	}
@@ -411,7 +411,7 @@ sfq_dequeue(struct Qdisc *sch)
 	skb = __skb_dequeue(&q->qs[a]);
 	sfq_dec(q, a);
 	sch->q.qlen--;
-	sch->qstats.backlog -= skb->len;
+	sch->qstats.backlog -= qdisc_pkt_len(skb);
 
 	/* Is the slot empty? */
 	if (q->qs[a].qlen == 0) {
@@ -423,7 +423,7 @@ sfq_dequeue(struct Qdisc *sch)
 		}
 		q->next[q->tail] = a;
 		q->allot[a] += q->quantum;
-	} else if ((q->allot[a] -= skb->len) <= 0) {
+	} else if ((q->allot[a] -= qdisc_pkt_len(skb)) <= 0) {
 		q->tail = a;
 		a = q->next[a];
 		q->allot[a] += q->quantum;
diff --git a/net/sched/sch_tbf.c b/net/sched/sch_tbf.c
index 7d705b86dae5..b296672f7632 100644
--- a/net/sched/sch_tbf.c
+++ b/net/sched/sch_tbf.c
@@ -123,7 +123,7 @@ static int tbf_enqueue(struct sk_buff *skb, struct Qdisc* sch)
 	struct tbf_sched_data *q = qdisc_priv(sch);
 	int ret;
 
-	if (skb->len > q->max_size) {
+	if (qdisc_pkt_len(skb) > q->max_size) {
 		sch->qstats.drops++;
 #ifdef CONFIG_NET_CLS_ACT
 		if (sch->reshape_fail == NULL || sch->reshape_fail(skb, sch))
@@ -140,7 +140,7 @@ static int tbf_enqueue(struct sk_buff *skb, struct Qdisc* sch)
 	}
 
 	sch->q.qlen++;
-	sch->bstats.bytes += skb->len;
+	sch->bstats.bytes += qdisc_pkt_len(skb);
 	sch->bstats.packets++;
 	return 0;
 }
@@ -181,7 +181,7 @@ static struct sk_buff *tbf_dequeue(struct Qdisc* sch)
 		psched_time_t now;
 		long toks;
 		long ptoks = 0;
-		unsigned int len = skb->len;
+		unsigned int len = qdisc_pkt_len(skb);
 
 		now = psched_get_time();
 		toks = psched_tdiff_bounded(now, q->t_c, q->buffer);
diff --git a/net/sched/sch_teql.c b/net/sched/sch_teql.c
index 8b0ff345f9da..537223642b6e 100644
--- a/net/sched/sch_teql.c
+++ b/net/sched/sch_teql.c
@@ -83,7 +83,7 @@ teql_enqueue(struct sk_buff *skb, struct Qdisc* sch)
 
 	if (q->q.qlen < dev->tx_queue_len) {
 		__skb_queue_tail(&q->q, skb);
-		sch->bstats.bytes += skb->len;
+		sch->bstats.bytes += qdisc_pkt_len(skb);
 		sch->bstats.packets++;
 		return 0;
 	}
@@ -278,7 +278,6 @@ static int teql_master_xmit(struct sk_buff *skb, struct net_device *dev)
 	struct Qdisc *start, *q;
 	int busy;
 	int nores;
-	int len = skb->len;
 	int subq = skb_get_queue_mapping(skb);
 	struct sk_buff *skb_res = NULL;
 
@@ -313,7 +312,8 @@ restart:
 					master->slaves = NEXT_SLAVE(q);
 					netif_wake_queue(dev);
 					master->stats.tx_packets++;
-					master->stats.tx_bytes += len;
+					master->stats.tx_bytes +=
+						qdisc_pkt_len(skb);
 					return 0;
 				}
 				netif_tx_unlock(slave);
-- 
cgit v1.2.3


From 175f9c1bba9b825d22b142d183c9e175488b260c Mon Sep 17 00:00:00 2001
From: Jussi Kivilinna <jussi.kivilinna@mbnet.fi>
Date: Sun, 20 Jul 2008 00:08:47 -0700
Subject: net_sched: Add size table for qdiscs

Add size table functions for qdiscs and calculate packet size in
qdisc_enqueue().

Based on patch by Patrick McHardy
 http://marc.info/?l=linux-netdev&m=115201979221729&w=2

Signed-off-by: Jussi Kivilinna <jussi.kivilinna@mbnet.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/pkt_sched.h |  20 ++++++
 include/linux/rtnetlink.h |   1 +
 include/net/pkt_sched.h   |   1 +
 include/net/sch_generic.h |  25 +++++++-
 net/sched/sch_api.c       | 151 +++++++++++++++++++++++++++++++++++++++++++++-
 net/sched/sch_generic.c   |   1 +
 net/sched/sch_netem.c     |   5 +-
 7 files changed, 199 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/include/linux/pkt_sched.h b/include/linux/pkt_sched.h
index 87f4e0fa8f27..e5de421ac7b4 100644
--- a/include/linux/pkt_sched.h
+++ b/include/linux/pkt_sched.h
@@ -85,6 +85,26 @@ struct tc_ratespec
 
 #define TC_RTAB_SIZE	1024
 
+struct tc_sizespec {
+	unsigned char	cell_log;
+	unsigned char	size_log;
+	short		cell_align;
+	int		overhead;
+	unsigned int	linklayer;
+	unsigned int	mpu;
+	unsigned int	mtu;
+	unsigned int	tsize;
+};
+
+enum {
+	TCA_STAB_UNSPEC,
+	TCA_STAB_BASE,
+	TCA_STAB_DATA,
+	__TCA_STAB_MAX
+};
+
+#define TCA_STAB_MAX (__TCA_STAB_MAX - 1)
+
 /* FIFO section */
 
 struct tc_fifo_qopt
diff --git a/include/linux/rtnetlink.h b/include/linux/rtnetlink.h
index b358c704d102..f4d386c191f5 100644
--- a/include/linux/rtnetlink.h
+++ b/include/linux/rtnetlink.h
@@ -482,6 +482,7 @@ enum
 	TCA_RATE,
 	TCA_FCNT,
 	TCA_STATS2,
+	TCA_STAB,
 	__TCA_MAX
 };
 
diff --git a/include/net/pkt_sched.h b/include/net/pkt_sched.h
index e4e30052e4e2..6affcfaa123e 100644
--- a/include/net/pkt_sched.h
+++ b/include/net/pkt_sched.h
@@ -83,6 +83,7 @@ extern struct Qdisc *qdisc_lookup_class(struct net_device *dev, u32 handle);
 extern struct qdisc_rate_table *qdisc_get_rtab(struct tc_ratespec *r,
 		struct nlattr *tab);
 extern void qdisc_put_rtab(struct qdisc_rate_table *tab);
+extern void qdisc_put_stab(struct qdisc_size_table *tab);
 
 extern void __qdisc_run(struct Qdisc *q);
 
diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h
index 8229520e088a..db9ad655eb8a 100644
--- a/include/net/sch_generic.h
+++ b/include/net/sch_generic.h
@@ -29,6 +29,13 @@ enum qdisc_state_t
 	__QDISC_STATE_SCHED,
 };
 
+struct qdisc_size_table {
+	struct list_head	list;
+	struct tc_sizespec	szopts;
+	int			refcnt;
+	u16			data[];
+};
+
 struct Qdisc
 {
 	int 			(*enqueue)(struct sk_buff *skb, struct Qdisc *dev);
@@ -39,6 +46,7 @@ struct Qdisc
 #define TCQ_F_INGRESS	4
 	int			padded;
 	struct Qdisc_ops	*ops;
+	struct qdisc_size_table	*stab;
 	u32			handle;
 	u32			parent;
 	atomic_t		refcnt;
@@ -165,6 +173,16 @@ struct tcf_proto
 	struct tcf_proto_ops	*ops;
 };
 
+struct qdisc_skb_cb {
+	unsigned int		pkt_len;
+	char			data[];
+};
+
+static inline struct qdisc_skb_cb *qdisc_skb_cb(struct sk_buff *skb)
+{
+	return (struct qdisc_skb_cb *)skb->cb;
+}
+
 static inline spinlock_t *qdisc_lock(struct Qdisc *qdisc)
 {
 	return &qdisc->q.lock;
@@ -257,6 +275,8 @@ extern struct Qdisc *qdisc_alloc(struct netdev_queue *dev_queue,
 extern struct Qdisc *qdisc_create_dflt(struct net_device *dev,
 				       struct netdev_queue *dev_queue,
 				       struct Qdisc_ops *ops, u32 parentid);
+extern void qdisc_calculate_pkt_len(struct sk_buff *skb,
+				   struct qdisc_size_table *stab);
 extern void tcf_destroy(struct tcf_proto *tp);
 extern void tcf_destroy_chain(struct tcf_proto **fl);
 
@@ -308,16 +328,19 @@ static inline bool qdisc_tx_is_noop(const struct net_device *dev)
 
 static inline unsigned int qdisc_pkt_len(struct sk_buff *skb)
 {
-	return skb->len;
+	return qdisc_skb_cb(skb)->pkt_len;
 }
 
 static inline int qdisc_enqueue(struct sk_buff *skb, struct Qdisc *sch)
 {
+	if (sch->stab)
+		qdisc_calculate_pkt_len(skb, sch->stab);
 	return sch->enqueue(skb, sch);
 }
 
 static inline int qdisc_enqueue_root(struct sk_buff *skb, struct Qdisc *sch)
 {
+	qdisc_skb_cb(skb)->pkt_len = skb->len;
 	return qdisc_enqueue(skb, sch);
 }
 
diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c
index fb43731c9860..5219d5f9d754 100644
--- a/net/sched/sch_api.c
+++ b/net/sched/sch_api.c
@@ -286,6 +286,129 @@ void qdisc_put_rtab(struct qdisc_rate_table *tab)
 }
 EXPORT_SYMBOL(qdisc_put_rtab);
 
+static LIST_HEAD(qdisc_stab_list);
+static DEFINE_SPINLOCK(qdisc_stab_lock);
+
+static const struct nla_policy stab_policy[TCA_STAB_MAX + 1] = {
+	[TCA_STAB_BASE]	= { .len = sizeof(struct tc_sizespec) },
+	[TCA_STAB_DATA] = { .type = NLA_BINARY },
+};
+
+static struct qdisc_size_table *qdisc_get_stab(struct nlattr *opt)
+{
+	struct nlattr *tb[TCA_STAB_MAX + 1];
+	struct qdisc_size_table *stab;
+	struct tc_sizespec *s;
+	unsigned int tsize = 0;
+	u16 *tab = NULL;
+	int err;
+
+	err = nla_parse_nested(tb, TCA_STAB_MAX, opt, stab_policy);
+	if (err < 0)
+		return ERR_PTR(err);
+	if (!tb[TCA_STAB_BASE])
+		return ERR_PTR(-EINVAL);
+
+	s = nla_data(tb[TCA_STAB_BASE]);
+
+	if (s->tsize > 0) {
+		if (!tb[TCA_STAB_DATA])
+			return ERR_PTR(-EINVAL);
+		tab = nla_data(tb[TCA_STAB_DATA]);
+		tsize = nla_len(tb[TCA_STAB_DATA]) / sizeof(u16);
+	}
+
+	if (!s || tsize != s->tsize || (!tab && tsize > 0))
+		return ERR_PTR(-EINVAL);
+
+	spin_lock(&qdisc_stab_lock);
+
+	list_for_each_entry(stab, &qdisc_stab_list, list) {
+		if (memcmp(&stab->szopts, s, sizeof(*s)))
+			continue;
+		if (tsize > 0 && memcmp(stab->data, tab, tsize * sizeof(u16)))
+			continue;
+		stab->refcnt++;
+		spin_unlock(&qdisc_stab_lock);
+		return stab;
+	}
+
+	spin_unlock(&qdisc_stab_lock);
+
+	stab = kmalloc(sizeof(*stab) + tsize * sizeof(u16), GFP_KERNEL);
+	if (!stab)
+		return ERR_PTR(-ENOMEM);
+
+	stab->refcnt = 1;
+	stab->szopts = *s;
+	if (tsize > 0)
+		memcpy(stab->data, tab, tsize * sizeof(u16));
+
+	spin_lock(&qdisc_stab_lock);
+	list_add_tail(&stab->list, &qdisc_stab_list);
+	spin_unlock(&qdisc_stab_lock);
+
+	return stab;
+}
+
+void qdisc_put_stab(struct qdisc_size_table *tab)
+{
+	if (!tab)
+		return;
+
+	spin_lock(&qdisc_stab_lock);
+
+	if (--tab->refcnt == 0) {
+		list_del(&tab->list);
+		kfree(tab);
+	}
+
+	spin_unlock(&qdisc_stab_lock);
+}
+EXPORT_SYMBOL(qdisc_put_stab);
+
+static int qdisc_dump_stab(struct sk_buff *skb, struct qdisc_size_table *stab)
+{
+	struct nlattr *nest;
+
+	nest = nla_nest_start(skb, TCA_STAB);
+	NLA_PUT(skb, TCA_STAB_BASE, sizeof(stab->szopts), &stab->szopts);
+	nla_nest_end(skb, nest);
+
+	return skb->len;
+
+nla_put_failure:
+	return -1;
+}
+
+void qdisc_calculate_pkt_len(struct sk_buff *skb, struct qdisc_size_table *stab)
+{
+	int pkt_len, slot;
+
+	pkt_len = skb->len + stab->szopts.overhead;
+	if (unlikely(!stab->szopts.tsize))
+		goto out;
+
+	slot = pkt_len + stab->szopts.cell_align;
+	if (unlikely(slot < 0))
+		slot = 0;
+
+	slot >>= stab->szopts.cell_log;
+	if (likely(slot < stab->szopts.tsize))
+		pkt_len = stab->data[slot];
+	else
+		pkt_len = stab->data[stab->szopts.tsize - 1] *
+				(slot / stab->szopts.tsize) +
+				stab->data[slot % stab->szopts.tsize];
+
+	pkt_len <<= stab->szopts.size_log;
+out:
+	if (unlikely(pkt_len < 1))
+		pkt_len = 1;
+	qdisc_skb_cb(skb)->pkt_len = pkt_len;
+}
+EXPORT_SYMBOL(qdisc_calculate_pkt_len);
+
 static enum hrtimer_restart qdisc_watchdog(struct hrtimer *timer)
 {
 	struct qdisc_watchdog *wd = container_of(timer, struct qdisc_watchdog,
@@ -613,6 +736,7 @@ qdisc_create(struct net_device *dev, struct netdev_queue *dev_queue,
 	struct nlattr *kind = tca[TCA_KIND];
 	struct Qdisc *sch;
 	struct Qdisc_ops *ops;
+	struct qdisc_size_table *stab;
 
 	ops = qdisc_lookup_ops(kind);
 #ifdef CONFIG_KMOD
@@ -670,6 +794,14 @@ qdisc_create(struct net_device *dev, struct netdev_queue *dev_queue,
 	sch->handle = handle;
 
 	if (!ops->init || (err = ops->init(sch, tca[TCA_OPTIONS])) == 0) {
+		if (tca[TCA_STAB]) {
+			stab = qdisc_get_stab(tca[TCA_STAB]);
+			if (IS_ERR(stab)) {
+				err = PTR_ERR(stab);
+				goto err_out3;
+			}
+			sch->stab = stab;
+		}
 		if (tca[TCA_RATE]) {
 			err = gen_new_estimator(&sch->bstats, &sch->rate_est,
 						qdisc_root_lock(sch),
@@ -691,6 +823,7 @@ qdisc_create(struct net_device *dev, struct netdev_queue *dev_queue,
 		return sch;
 	}
 err_out3:
+	qdisc_put_stab(sch->stab);
 	dev_put(dev);
 	kfree((char *) sch - sch->padded);
 err_out2:
@@ -702,15 +835,26 @@ err_out:
 
 static int qdisc_change(struct Qdisc *sch, struct nlattr **tca)
 {
-	if (tca[TCA_OPTIONS]) {
-		int err;
+	struct qdisc_size_table *stab = NULL;
+	int err = 0;
 
+	if (tca[TCA_OPTIONS]) {
 		if (sch->ops->change == NULL)
 			return -EINVAL;
 		err = sch->ops->change(sch, tca[TCA_OPTIONS]);
 		if (err)
 			return err;
 	}
+
+	if (tca[TCA_STAB]) {
+		stab = qdisc_get_stab(tca[TCA_STAB]);
+		if (IS_ERR(stab))
+			return PTR_ERR(stab);
+	}
+
+	qdisc_put_stab(sch->stab);
+	sch->stab = stab;
+
 	if (tca[TCA_RATE])
 		gen_replace_estimator(&sch->bstats, &sch->rate_est,
 				      qdisc_root_lock(sch), tca[TCA_RATE]);
@@ -994,6 +1138,9 @@ static int tc_fill_qdisc(struct sk_buff *skb, struct Qdisc *q, u32 clid,
 		goto nla_put_failure;
 	q->qstats.qlen = q->q.qlen;
 
+	if (q->stab && qdisc_dump_stab(skb, q->stab) < 0)
+		goto nla_put_failure;
+
 	if (gnet_stats_start_copy_compat(skb, TCA_STATS2, TCA_STATS,
 					 TCA_XSTATS, qdisc_root_lock(q), &d) < 0)
 		goto nla_put_failure;
diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c
index 522a41a9f904..27a51f04db49 100644
--- a/net/sched/sch_generic.c
+++ b/net/sched/sch_generic.c
@@ -469,6 +469,7 @@ static void __qdisc_destroy(struct rcu_head *head)
 	struct Qdisc *qdisc = container_of(head, struct Qdisc, q_rcu);
 	const struct Qdisc_ops  *ops = qdisc->ops;
 
+	qdisc_put_stab(qdisc->stab);
 	gen_kill_estimator(&qdisc->bstats, &qdisc->rate_est);
 	if (ops->reset)
 		ops->reset(qdisc);
diff --git a/net/sched/sch_netem.c b/net/sched/sch_netem.c
index ae49be00022f..a59085700678 100644
--- a/net/sched/sch_netem.c
+++ b/net/sched/sch_netem.c
@@ -84,8 +84,9 @@ struct netem_skb_cb {
 
 static inline struct netem_skb_cb *netem_skb_cb(struct sk_buff *skb)
 {
-	BUILD_BUG_ON(sizeof(skb->cb) < sizeof(struct netem_skb_cb));
-	return (struct netem_skb_cb *)skb->cb;
+	BUILD_BUG_ON(sizeof(skb->cb) <
+		sizeof(struct qdisc_skb_cb) + sizeof(struct netem_skb_cb));
+	return (struct netem_skb_cb *)qdisc_skb_cb(skb)->data;
 }
 
 /* init_crandom - initialize correlated random number generator
-- 
cgit v1.2.3