From 95b72eb0bdef6476b7e73061f0382adf46c5495a Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Fri, 20 Apr 2012 19:24:51 -0400 Subject: NFSv4: Ensure we do not reuse open owner names The NFSv4 spec is ambiguous about whether or not it is permissible to reuse open owner names, so play it safe. This patch adds a timestamp to the state_owner structure, and combines that with the IDA based uniquifier. Fixes a regression whereby the Linux server returns NFS4ERR_BAD_SEQID. Signed-off-by: Trond Myklebust --- include/linux/nfs_xdr.h | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h index bfd0d1bf6707..7ba3551a0414 100644 --- a/include/linux/nfs_xdr.h +++ b/include/linux/nfs_xdr.h @@ -312,6 +312,11 @@ struct nfs4_layoutreturn { int rpc_status; }; +struct stateowner_id { + __u64 create_time; + __u32 uniquifier; +}; + /* * Arguments to the open call. */ @@ -321,7 +326,7 @@ struct nfs_openargs { int open_flags; fmode_t fmode; __u64 clientid; - __u64 id; + struct stateowner_id id; union { struct { struct iattr * attrs; /* UNCHECKED, GUARDED */ -- cgit v1.2.3 From b2024459252a9d2d312ee562f86f332a1498f412 Mon Sep 17 00:00:00 2001 From: Dan Williams Date: Wed, 21 Mar 2012 21:09:07 -0700 Subject: [SCSI] libsas, libata: fix start of life for a sas ata_port This changes the ordering of initialization and probing events from: 1/ allocate rphy in PORTE_BYTES_DMAED, DISCE_REVALIDATE_DOMAIN 2/ allocate ata_port and schedule port probe in DISCE_PROBE ...to: 1/ allocate ata_port in PORTE_BYTES_DMAED, DISCE_REVALIDATE_DOMAIN 2/ allocate rphy in PORTE_BYTES_DMAED, DISCE_REVALIDATE_DOMAIN 3/ schedule port probe in DISCE_PROBE This ordering prevents PHYE_SIGNAL_LOSS_EVENTS from sneaking in to destrory ata devices before they have been fully initialized: BUG: unable to handle kernel paging request at 0000000000003b10 IP: [] sas_ata_end_eh+0x12/0x5e [libsas] ... [] sas_unregister_common_dev+0x78/0xc9 [libsas] [] sas_unregister_dev+0x4f/0xad [libsas] [] sas_unregister_domain_devices+0x7f/0xbf [libsas] [] sas_deform_port+0x61/0x1b8 [libsas] [] sas_phye_loss_of_signal+0x29/0x2b [libsas] ...and kills the awkward "sata domain_device briefly existing in the domain without an ata_port" state. Reported-by: Michal Kosciowski Signed-off-by: Dan Williams Acked-by: Jeff Garzik Signed-off-by: James Bottomley --- drivers/ata/libata-scsi.c | 35 ++++++++++++++++++++--------------- drivers/scsi/ipr.c | 6 +++++- drivers/scsi/libsas/sas_ata.c | 33 ++++++++++----------------------- drivers/scsi/libsas/sas_discover.c | 13 ++++++++++--- drivers/scsi/libsas/sas_expander.c | 8 +++++--- include/linux/libata.h | 3 ++- include/scsi/sas_ata.h | 4 ++-- 7 files changed, 54 insertions(+), 48 deletions(-) (limited to 'include/linux') diff --git a/drivers/ata/libata-scsi.c b/drivers/ata/libata-scsi.c index 7832b1ad2327..22226350cd0c 100644 --- a/drivers/ata/libata-scsi.c +++ b/drivers/ata/libata-scsi.c @@ -3839,18 +3839,25 @@ void ata_sas_port_stop(struct ata_port *ap) } EXPORT_SYMBOL_GPL(ata_sas_port_stop); -int ata_sas_async_port_init(struct ata_port *ap) +/** + * ata_sas_async_probe - simply schedule probing and return + * @ap: Port to probe + * + * For batch scheduling of probe for sas attached ata devices, assumes + * the port has already been through ata_sas_port_init() + */ +void ata_sas_async_probe(struct ata_port *ap) { - int rc = ap->ops->port_start(ap); - - if (!rc) { - ap->print_id = atomic_inc_return(&ata_print_id); - __ata_port_probe(ap); - } + __ata_port_probe(ap); +} +EXPORT_SYMBOL_GPL(ata_sas_async_probe); - return rc; +int ata_sas_sync_probe(struct ata_port *ap) +{ + return ata_port_probe(ap); } -EXPORT_SYMBOL_GPL(ata_sas_async_port_init); +EXPORT_SYMBOL_GPL(ata_sas_sync_probe); + /** * ata_sas_port_init - Initialize a SATA device @@ -3867,12 +3874,10 @@ int ata_sas_port_init(struct ata_port *ap) { int rc = ap->ops->port_start(ap); - if (!rc) { - ap->print_id = atomic_inc_return(&ata_print_id); - rc = ata_port_probe(ap); - } - - return rc; + if (rc) + return rc; + ap->print_id = atomic_inc_return(&ata_print_id); + return 0; } EXPORT_SYMBOL_GPL(ata_sas_port_init); diff --git a/drivers/scsi/ipr.c b/drivers/scsi/ipr.c index e002cd466e9a..467dc38246f9 100644 --- a/drivers/scsi/ipr.c +++ b/drivers/scsi/ipr.c @@ -4549,8 +4549,12 @@ static int ipr_ata_slave_alloc(struct scsi_device *sdev) ENTER; if (sdev->sdev_target) sata_port = sdev->sdev_target->hostdata; - if (sata_port) + if (sata_port) { rc = ata_sas_port_init(sata_port->ap); + if (rc == 0) + rc = ata_sas_sync_probe(sata_port->ap); + } + if (rc) ipr_slave_destroy(sdev); diff --git a/drivers/scsi/libsas/sas_ata.c b/drivers/scsi/libsas/sas_ata.c index bc0cecc6ad62..441d88ad99a7 100644 --- a/drivers/scsi/libsas/sas_ata.c +++ b/drivers/scsi/libsas/sas_ata.c @@ -546,11 +546,12 @@ static struct ata_port_info sata_port_info = { .port_ops = &sas_sata_ops }; -int sas_ata_init_host_and_port(struct domain_device *found_dev) +int sas_ata_init(struct domain_device *found_dev) { struct sas_ha_struct *ha = found_dev->port->ha; struct Scsi_Host *shost = ha->core.shost; struct ata_port *ap; + int rc; ata_host_init(&found_dev->sata_dev.ata_host, ha->dev, @@ -567,8 +568,11 @@ int sas_ata_init_host_and_port(struct domain_device *found_dev) ap->private_data = found_dev; ap->cbl = ATA_CBL_SATA; ap->scsi_host = shost; - /* publish initialized ata port */ - smp_wmb(); + rc = ata_sas_port_init(ap); + if (rc) { + ata_sas_port_destroy(ap); + return rc; + } found_dev->sata_dev.ap = ap; return 0; @@ -648,18 +652,13 @@ static void sas_get_ata_command_set(struct domain_device *dev) void sas_probe_sata(struct asd_sas_port *port) { struct domain_device *dev, *n; - int err; mutex_lock(&port->ha->disco_mutex); - list_for_each_entry_safe(dev, n, &port->disco_list, disco_list_node) { + list_for_each_entry(dev, &port->disco_list, disco_list_node) { if (!dev_is_sata(dev)) continue; - err = sas_ata_init_host_and_port(dev); - if (err) - sas_fail_probe(dev, __func__, err); - else - ata_sas_async_port_init(dev->sata_dev.ap); + ata_sas_async_probe(dev->sata_dev.ap); } mutex_unlock(&port->ha->disco_mutex); @@ -718,18 +717,6 @@ static void async_sas_ata_eh(void *data, async_cookie_t cookie) sas_put_device(dev); } -static bool sas_ata_dev_eh_valid(struct domain_device *dev) -{ - struct ata_port *ap; - - if (!dev_is_sata(dev)) - return false; - ap = dev->sata_dev.ap; - /* consume fully initialized ata ports */ - smp_rmb(); - return !!ap; -} - void sas_ata_strategy_handler(struct Scsi_Host *shost) { struct sas_ha_struct *sas_ha = SHOST_TO_SAS_HA(shost); @@ -753,7 +740,7 @@ void sas_ata_strategy_handler(struct Scsi_Host *shost) spin_lock(&port->dev_list_lock); list_for_each_entry(dev, &port->dev_list, dev_list_node) { - if (!sas_ata_dev_eh_valid(dev)) + if (!dev_is_sata(dev)) continue; async_schedule_domain(async_sas_ata_eh, dev, &async); } diff --git a/drivers/scsi/libsas/sas_discover.c b/drivers/scsi/libsas/sas_discover.c index 13b5891f9961..629a0865b130 100644 --- a/drivers/scsi/libsas/sas_discover.c +++ b/drivers/scsi/libsas/sas_discover.c @@ -72,6 +72,7 @@ static int sas_get_port_device(struct asd_sas_port *port) struct asd_sas_phy *phy; struct sas_rphy *rphy; struct domain_device *dev; + int rc = -ENODEV; dev = sas_alloc_device(); if (!dev) @@ -110,9 +111,16 @@ static int sas_get_port_device(struct asd_sas_port *port) sas_init_dev(dev); + dev->port = port; switch (dev->dev_type) { - case SAS_END_DEV: case SATA_DEV: + rc = sas_ata_init(dev); + if (rc) { + rphy = NULL; + break; + } + /* fall through */ + case SAS_END_DEV: rphy = sas_end_device_alloc(port->port); break; case EDGE_DEV: @@ -131,7 +139,7 @@ static int sas_get_port_device(struct asd_sas_port *port) if (!rphy) { sas_put_device(dev); - return -ENODEV; + return rc; } rphy->identify.phy_identifier = phy->phy->identify.phy_identifier; @@ -139,7 +147,6 @@ static int sas_get_port_device(struct asd_sas_port *port) sas_fill_in_rphy(dev, rphy); sas_hash_addr(dev->hashed_sas_addr, dev->sas_addr); port->port_dev = dev; - dev->port = port; dev->linkrate = port->linkrate; dev->min_linkrate = port->linkrate; dev->max_linkrate = port->linkrate; diff --git a/drivers/scsi/libsas/sas_expander.c b/drivers/scsi/libsas/sas_expander.c index c1f91b1c27c3..75247a176c6b 100644 --- a/drivers/scsi/libsas/sas_expander.c +++ b/drivers/scsi/libsas/sas_expander.c @@ -790,12 +790,14 @@ static struct domain_device *sas_ex_discover_end_dev( if (res) goto out_free; + sas_init_dev(child); + res = sas_ata_init(child); + if (res) + goto out_free; rphy = sas_end_device_alloc(phy->port); - if (unlikely(!rphy)) + if (!rphy) goto out_free; - sas_init_dev(child); - child->rphy = rphy; get_device(&rphy->dev); diff --git a/include/linux/libata.h b/include/linux/libata.h index 42378d637ffb..e926df7b54c9 100644 --- a/include/linux/libata.h +++ b/include/linux/libata.h @@ -996,7 +996,8 @@ extern int ata_sas_scsi_ioctl(struct ata_port *ap, struct scsi_device *dev, extern void ata_sas_port_destroy(struct ata_port *); extern struct ata_port *ata_sas_port_alloc(struct ata_host *, struct ata_port_info *, struct Scsi_Host *); -extern int ata_sas_async_port_init(struct ata_port *); +extern void ata_sas_async_probe(struct ata_port *ap); +extern int ata_sas_sync_probe(struct ata_port *ap); extern int ata_sas_port_init(struct ata_port *); extern int ata_sas_port_start(struct ata_port *ap); extern void ata_sas_port_stop(struct ata_port *ap); diff --git a/include/scsi/sas_ata.h b/include/scsi/sas_ata.h index cdccd2eb7b6c..77670e823ed8 100644 --- a/include/scsi/sas_ata.h +++ b/include/scsi/sas_ata.h @@ -37,7 +37,7 @@ static inline int dev_is_sata(struct domain_device *dev) } int sas_get_ata_info(struct domain_device *dev, struct ex_phy *phy); -int sas_ata_init_host_and_port(struct domain_device *found_dev); +int sas_ata_init(struct domain_device *dev); void sas_ata_task_abort(struct sas_task *task); void sas_ata_strategy_handler(struct Scsi_Host *shost); void sas_ata_eh(struct Scsi_Host *shost, struct list_head *work_q, @@ -52,7 +52,7 @@ static inline int dev_is_sata(struct domain_device *dev) { return 0; } -static inline int sas_ata_init_host_and_port(struct domain_device *found_dev) +static inline int sas_ata_init(struct domain_device *dev) { return 0; } -- cgit v1.2.3 From 5a218ceba7b64f506bf4f004b04bb457c1805a62 Mon Sep 17 00:00:00 2001 From: Carlos Chinea Date: Wed, 4 Apr 2012 14:11:45 +0300 Subject: HSI: hsi: Rework hsi_controller release Use the proper release mechanism for hsi_controller and hsi_ports structures. Free the structures through their associated device release callbacks. Signed-off-by: Carlos Chinea Acked-by: Greg Kroah-Hartman Acked-by: Linus Walleij --- drivers/hsi/hsi.c | 108 +++++++++++++++++++++++++++++------------------- include/linux/hsi/hsi.h | 6 +-- 2 files changed, 69 insertions(+), 45 deletions(-) (limited to 'include/linux') diff --git a/drivers/hsi/hsi.c b/drivers/hsi/hsi.c index 4e2d79b79334..c17d12ca8e7f 100644 --- a/drivers/hsi/hsi.c +++ b/drivers/hsi/hsi.c @@ -140,12 +140,17 @@ static int hsi_remove_port(struct device *dev, void *data __maybe_unused) return 0; } -static void hsi_controller_release(struct device *dev __maybe_unused) +static void hsi_controller_release(struct device *dev) { + struct hsi_controller *hsi = to_hsi_controller(dev); + + kfree(hsi->port); + kfree(hsi); } -static void hsi_port_release(struct device *dev __maybe_unused) +static void hsi_port_release(struct device *dev) { + kfree(to_hsi_port(dev)); } /** @@ -172,18 +177,14 @@ int hsi_register_controller(struct hsi_controller *hsi) hsi->device.type = &hsi_ctrl; hsi->device.bus = &hsi_bus_type; - hsi->device.release = hsi_controller_release; - err = device_register(&hsi->device); + err = device_add(&hsi->device); if (err < 0) return err; for (i = 0; i < hsi->num_ports; i++) { - hsi->port[i].device.parent = &hsi->device; - hsi->port[i].device.bus = &hsi_bus_type; - hsi->port[i].device.release = hsi_port_release; - hsi->port[i].device.type = &hsi_port; - INIT_LIST_HEAD(&hsi->port[i].clients); - spin_lock_init(&hsi->port[i].clock); - err = device_register(&hsi->port[i].device); + hsi->port[i]->device.parent = &hsi->device; + hsi->port[i]->device.bus = &hsi_bus_type; + hsi->port[i]->device.type = &hsi_port; + err = device_add(&hsi->port[i]->device); if (err < 0) goto out; } @@ -192,7 +193,9 @@ int hsi_register_controller(struct hsi_controller *hsi) return 0; out: - hsi_unregister_controller(hsi); + while (i-- > 0) + device_del(&hsi->port[i]->device); + device_del(&hsi->device); return err; } @@ -222,6 +225,29 @@ static inline int hsi_dummy_cl(struct hsi_client *cl __maybe_unused) return 0; } +/** + * hsi_put_controller - Free an HSI controller + * + * @hsi: Pointer to the HSI controller to freed + * + * HSI controller drivers should only use this function if they need + * to free their allocated hsi_controller structures before a successful + * call to hsi_register_controller. Other use is not allowed. + */ +void hsi_put_controller(struct hsi_controller *hsi) +{ + unsigned int i; + + if (!hsi) + return; + + for (i = 0; i < hsi->num_ports; i++) + if (hsi->port && hsi->port[i]) + put_device(&hsi->port[i]->device); + put_device(&hsi->device); +} +EXPORT_SYMBOL_GPL(hsi_put_controller); + /** * hsi_alloc_controller - Allocate an HSI controller and its ports * @n_ports: Number of ports on the HSI controller @@ -232,54 +258,52 @@ static inline int hsi_dummy_cl(struct hsi_client *cl __maybe_unused) struct hsi_controller *hsi_alloc_controller(unsigned int n_ports, gfp_t flags) { struct hsi_controller *hsi; - struct hsi_port *port; + struct hsi_port **port; unsigned int i; if (!n_ports) return NULL; - port = kzalloc(sizeof(*port)*n_ports, flags); - if (!port) - return NULL; hsi = kzalloc(sizeof(*hsi), flags); if (!hsi) - goto out; - for (i = 0; i < n_ports; i++) { - dev_set_name(&port[i].device, "port%d", i); - port[i].num = i; - port[i].async = hsi_dummy_msg; - port[i].setup = hsi_dummy_cl; - port[i].flush = hsi_dummy_cl; - port[i].start_tx = hsi_dummy_cl; - port[i].stop_tx = hsi_dummy_cl; - port[i].release = hsi_dummy_cl; - mutex_init(&port[i].lock); + return NULL; + port = kzalloc(sizeof(*port)*n_ports, flags); + if (!port) { + kfree(hsi); + return NULL; } hsi->num_ports = n_ports; hsi->port = port; + hsi->device.release = hsi_controller_release; + device_initialize(&hsi->device); + + for (i = 0; i < n_ports; i++) { + port[i] = kzalloc(sizeof(**port), flags); + if (port[i] == NULL) + goto out; + port[i]->num = i; + port[i]->async = hsi_dummy_msg; + port[i]->setup = hsi_dummy_cl; + port[i]->flush = hsi_dummy_cl; + port[i]->start_tx = hsi_dummy_cl; + port[i]->stop_tx = hsi_dummy_cl; + port[i]->release = hsi_dummy_cl; + mutex_init(&port[i]->lock); + INIT_LIST_HEAD(&hsi->port[i]->clients); + spin_lock_init(&hsi->port[i]->clock); + dev_set_name(&port[i]->device, "port%d", i); + hsi->port[i]->device.release = hsi_port_release; + device_initialize(&hsi->port[i]->device); + } return hsi; out: - kfree(port); + hsi_put_controller(hsi); return NULL; } EXPORT_SYMBOL_GPL(hsi_alloc_controller); -/** - * hsi_free_controller - Free an HSI controller - * @hsi: Pointer to HSI controller - */ -void hsi_free_controller(struct hsi_controller *hsi) -{ - if (!hsi) - return; - - kfree(hsi->port); - kfree(hsi); -} -EXPORT_SYMBOL_GPL(hsi_free_controller); - /** * hsi_free_msg - Free an HSI message * @msg: Pointer to the HSI message diff --git a/include/linux/hsi/hsi.h b/include/linux/hsi/hsi.h index 4b178067f405..7f3b7262a2b6 100644 --- a/include/linux/hsi/hsi.h +++ b/include/linux/hsi/hsi.h @@ -270,13 +270,13 @@ struct hsi_controller { struct module *owner; unsigned int id; unsigned int num_ports; - struct hsi_port *port; + struct hsi_port **port; }; #define to_hsi_controller(dev) container_of(dev, struct hsi_controller, device) struct hsi_controller *hsi_alloc_controller(unsigned int n_ports, gfp_t flags); -void hsi_free_controller(struct hsi_controller *hsi); +void hsi_put_controller(struct hsi_controller *hsi); int hsi_register_controller(struct hsi_controller *hsi); void hsi_unregister_controller(struct hsi_controller *hsi); @@ -294,7 +294,7 @@ static inline void *hsi_controller_drvdata(struct hsi_controller *hsi) static inline struct hsi_port *hsi_find_port_num(struct hsi_controller *hsi, unsigned int num) { - return (num < hsi->num_ports) ? &hsi->port[num] : NULL; + return (num < hsi->num_ports) ? hsi->port[num] : NULL; } /* -- cgit v1.2.3 From ec1c56ff813a198d656d4aa42e5de03e45751bf8 Mon Sep 17 00:00:00 2001 From: Carlos Chinea Date: Wed, 11 Apr 2012 10:55:53 +0300 Subject: HSI: hsi: Rework hsi_event interface Remove custom hack and make use of the notifier chain interfaces for delivering events from the ports to their associated clients. Clients that want to receive port events need to register their callbacks using hsi_register_port_event(). The callbacks can be called in interrupt context. Use hsi_unregestier_port_event() to undo the registration. Signed-off-by: Carlos Chinea Acked-by: Greg Kroah-Hartman Acked-by: Linus Walleij --- drivers/hsi/hsi.c | 101 ++++++++++++++++++++++++++---------------------- include/linux/hsi/hsi.h | 25 ++++++------ 2 files changed, 68 insertions(+), 58 deletions(-) (limited to 'include/linux') diff --git a/drivers/hsi/hsi.c b/drivers/hsi/hsi.c index cec1f0c04557..2d58f939d27f 100644 --- a/drivers/hsi/hsi.c +++ b/drivers/hsi/hsi.c @@ -21,12 +21,11 @@ */ #include #include -#include #include -#include #include #include #include +#include #include "hsi_core.h" static ssize_t modalias_show(struct device *dev, @@ -67,7 +66,6 @@ static void hsi_client_release(struct device *dev) static void hsi_new_client(struct hsi_port *port, struct hsi_board_info *info) { struct hsi_client *cl; - unsigned long flags; cl = kzalloc(sizeof(*cl), GFP_KERNEL); if (!cl) @@ -79,9 +77,6 @@ static void hsi_new_client(struct hsi_port *port, struct hsi_board_info *info) cl->device.release = hsi_client_release; dev_set_name(&cl->device, info->name); cl->device.platform_data = info->platform_data; - spin_lock_irqsave(&port->clock, flags); - list_add_tail(&cl->link, &port->clients); - spin_unlock_irqrestore(&port->clock, flags); if (info->archdata) cl->device.archdata = *info->archdata; if (device_register(&cl->device) < 0) { @@ -106,13 +101,6 @@ static void hsi_scan_board_info(struct hsi_controller *hsi) static int hsi_remove_client(struct device *dev, void *data __maybe_unused) { - struct hsi_client *cl = to_hsi_client(dev); - struct hsi_port *port = to_hsi_port(dev->parent); - unsigned long flags; - - spin_lock_irqsave(&port->clock, flags); - list_del(&cl->link); - spin_unlock_irqrestore(&port->clock, flags); device_unregister(dev); return 0; @@ -271,8 +259,7 @@ struct hsi_controller *hsi_alloc_controller(unsigned int n_ports, gfp_t flags) port[i]->stop_tx = hsi_dummy_cl; port[i]->release = hsi_dummy_cl; mutex_init(&port[i]->lock); - INIT_LIST_HEAD(&hsi->port[i]->clients); - spin_lock_init(&hsi->port[i]->clock); + ATOMIC_INIT_NOTIFIER_HEAD(&port[i]->n_head); dev_set_name(&port[i]->device, "port%d", i); hsi->port[i]->device.release = hsi_port_release; device_initialize(&hsi->port[i]->device); @@ -420,37 +407,67 @@ void hsi_release_port(struct hsi_client *cl) } EXPORT_SYMBOL_GPL(hsi_release_port); -static int hsi_start_rx(struct hsi_client *cl, void *data __maybe_unused) +static int hsi_event_notifier_call(struct notifier_block *nb, + unsigned long event, void *data __maybe_unused) { - if (cl->hsi_start_rx) - (*cl->hsi_start_rx)(cl); + struct hsi_client *cl = container_of(nb, struct hsi_client, nb); + + (*cl->ehandler)(cl, event); return 0; } -static int hsi_stop_rx(struct hsi_client *cl, void *data __maybe_unused) +/** + * hsi_register_port_event - Register a client to receive port events + * @cl: HSI client that wants to receive port events + * @cb: Event handler callback + * + * Clients should register a callback to be able to receive + * events from the ports. Registration should happen after + * claiming the port. + * The handler can be called in interrupt context. + * + * Returns -errno on error, or 0 on success. + */ +int hsi_register_port_event(struct hsi_client *cl, + void (*handler)(struct hsi_client *, unsigned long)) { - if (cl->hsi_stop_rx) - (*cl->hsi_stop_rx)(cl); + struct hsi_port *port = hsi_get_port(cl); - return 0; + if (!handler || cl->ehandler) + return -EINVAL; + if (!hsi_port_claimed(cl)) + return -EACCES; + cl->ehandler = handler; + cl->nb.notifier_call = hsi_event_notifier_call; + + return atomic_notifier_chain_register(&port->n_head, &cl->nb); } +EXPORT_SYMBOL_GPL(hsi_register_port_event); -static int hsi_port_for_each_client(struct hsi_port *port, void *data, - int (*fn)(struct hsi_client *cl, void *data)) +/** + * hsi_unregister_port_event - Stop receiving port events for a client + * @cl: HSI client that wants to stop receiving port events + * + * Clients should call this function before releasing their associated + * port. + * + * Returns -errno on error, or 0 on success. + */ +int hsi_unregister_port_event(struct hsi_client *cl) { - struct hsi_client *cl; + struct hsi_port *port = hsi_get_port(cl); + int err; - spin_lock(&port->clock); - list_for_each_entry(cl, &port->clients, link) { - spin_unlock(&port->clock); - (*fn)(cl, data); - spin_lock(&port->clock); - } - spin_unlock(&port->clock); + WARN_ON(!hsi_port_claimed(cl)); - return 0; + err = atomic_notifier_chain_unregister(&port->n_head, &cl->nb); + if (!err) + cl->ehandler = NULL; + + return err; } +EXPORT_SYMBOL_GPL(hsi_unregister_port_event); /** * hsi_event -Notifies clients about port events @@ -464,22 +481,12 @@ static int hsi_port_for_each_client(struct hsi_port *port, void *data, * Events: * HSI_EVENT_START_RX - Incoming wake line high * HSI_EVENT_STOP_RX - Incoming wake line down + * + * Returns -errno on error, or 0 on success. */ -void hsi_event(struct hsi_port *port, unsigned int event) +int hsi_event(struct hsi_port *port, unsigned long event) { - int (*fn)(struct hsi_client *cl, void *data); - - switch (event) { - case HSI_EVENT_START_RX: - fn = hsi_start_rx; - break; - case HSI_EVENT_STOP_RX: - fn = hsi_stop_rx; - break; - default: - return; - } - hsi_port_for_each_client(port, NULL, fn); + return atomic_notifier_call_chain(&port->n_head, event, NULL); } EXPORT_SYMBOL_GPL(hsi_event); diff --git a/include/linux/hsi/hsi.h b/include/linux/hsi/hsi.h index 7f3b7262a2b6..56fae865e272 100644 --- a/include/linux/hsi/hsi.h +++ b/include/linux/hsi/hsi.h @@ -26,9 +26,9 @@ #include #include #include -#include #include #include +#include /* HSI message ttype */ #define HSI_MSG_READ 0 @@ -121,18 +121,18 @@ static inline int hsi_register_board_info(struct hsi_board_info const *info, * @device: Driver model representation of the device * @tx_cfg: HSI TX configuration * @rx_cfg: HSI RX configuration - * @hsi_start_rx: Called after incoming wake line goes high - * @hsi_stop_rx: Called after incoming wake line goes low + * @e_handler: Callback for handling port events (RX Wake High/Low) + * @pclaimed: Keeps tracks if the clients claimed its associated HSI port + * @nb: Notifier block for port events */ struct hsi_client { struct device device; struct hsi_config tx_cfg; struct hsi_config rx_cfg; - void (*hsi_start_rx)(struct hsi_client *cl); - void (*hsi_stop_rx)(struct hsi_client *cl); /* private: */ + void (*ehandler)(struct hsi_client *, unsigned long); unsigned int pclaimed:1; - struct list_head link; + struct notifier_block nb; }; #define to_hsi_client(dev) container_of(dev, struct hsi_client, device) @@ -147,6 +147,10 @@ static inline void *hsi_client_drvdata(struct hsi_client *cl) return dev_get_drvdata(&cl->device); } +int hsi_register_port_event(struct hsi_client *cl, + void (*handler)(struct hsi_client *, unsigned long)); +int hsi_unregister_port_event(struct hsi_client *cl); + /** * struct hsi_client_driver - Driver associated to an HSI client * @driver: Driver model representation of the driver @@ -214,8 +218,7 @@ void hsi_free_msg(struct hsi_msg *msg); * @start_tx: Callback to inform that a client wants to TX data * @stop_tx: Callback to inform that a client no longer wishes to TX data * @release: Callback to inform that a client no longer uses the port - * @clients: List of hsi_clients using the port. - * @clock: Lock to serialize access to the clients list. + * @n_head: Notifier chain for signaling port events to the clients. */ struct hsi_port { struct device device; @@ -231,14 +234,14 @@ struct hsi_port { int (*start_tx)(struct hsi_client *cl); int (*stop_tx)(struct hsi_client *cl); int (*release)(struct hsi_client *cl); - struct list_head clients; - spinlock_t clock; + /* private */ + struct atomic_notifier_head n_head; }; #define to_hsi_port(dev) container_of(dev, struct hsi_port, device) #define hsi_get_port(cl) to_hsi_port((cl)->device.parent) -void hsi_event(struct hsi_port *port, unsigned int event); +int hsi_event(struct hsi_port *port, unsigned long event); int hsi_claim_port(struct hsi_client *cl, unsigned int share); void hsi_release_port(struct hsi_client *cl); -- cgit v1.2.3 From a881e963c7fe1f226e991ee9bbe8907acda93294 Mon Sep 17 00:00:00 2001 From: "Peter Huang (Peng)" Date: Thu, 19 Apr 2012 20:12:51 +0000 Subject: set fake_rtable's dst to NULL to avoid kernel Oops bridge: set fake_rtable's dst to NULL to avoid kernel Oops when bridge is deleted before tap/vif device's delete, kernel may encounter an oops because of NULL reference to fake_rtable's dst. Set fake_rtable's dst to NULL before sending packets out can solve this problem. v4 reformat, change br_drop_fake_rtable(skb) to {} v3 enrich commit header v2 introducing new flag DST_FAKE_RTABLE to dst_entry struct. [ Use "do { } while (0)" for nop br_drop_fake_rtable() implementation -DaveM ] Acked-by: Eric Dumazet Signed-off-by: Peter Huang Signed-off-by: David S. Miller --- include/linux/netfilter_bridge.h | 9 +++++++++ include/net/dst.h | 1 + net/bridge/br_forward.c | 1 + net/bridge/br_netfilter.c | 8 ++------ 4 files changed, 13 insertions(+), 6 deletions(-) (limited to 'include/linux') diff --git a/include/linux/netfilter_bridge.h b/include/linux/netfilter_bridge.h index 0ddd161f3b06..31d2844e6572 100644 --- a/include/linux/netfilter_bridge.h +++ b/include/linux/netfilter_bridge.h @@ -104,9 +104,18 @@ struct bridge_skb_cb { } daddr; }; +static inline void br_drop_fake_rtable(struct sk_buff *skb) +{ + struct dst_entry *dst = skb_dst(skb); + + if (dst && (dst->flags & DST_FAKE_RTABLE)) + skb_dst_drop(skb); +} + #else #define nf_bridge_maybe_copy_header(skb) (0) #define nf_bridge_pad(skb) (0) +#define br_drop_fake_rtable(skb) do { } while (0) #endif /* CONFIG_BRIDGE_NETFILTER */ #endif /* __KERNEL__ */ diff --git a/include/net/dst.h b/include/net/dst.h index ff4da42fcfc6..bed833d9796a 100644 --- a/include/net/dst.h +++ b/include/net/dst.h @@ -59,6 +59,7 @@ struct dst_entry { #define DST_NOCACHE 0x0010 #define DST_NOCOUNT 0x0020 #define DST_NOPEER 0x0040 +#define DST_FAKE_RTABLE 0x0080 short error; short obsolete; diff --git a/net/bridge/br_forward.c b/net/bridge/br_forward.c index 61f65344e711..a2098e3de500 100644 --- a/net/bridge/br_forward.c +++ b/net/bridge/br_forward.c @@ -47,6 +47,7 @@ int br_dev_queue_push_xmit(struct sk_buff *skb) kfree_skb(skb); } else { skb_push(skb, ETH_HLEN); + br_drop_fake_rtable(skb); dev_queue_xmit(skb); } diff --git a/net/bridge/br_netfilter.c b/net/bridge/br_netfilter.c index dec4f3817133..d7f49b63ab0f 100644 --- a/net/bridge/br_netfilter.c +++ b/net/bridge/br_netfilter.c @@ -156,7 +156,7 @@ void br_netfilter_rtable_init(struct net_bridge *br) rt->dst.dev = br->dev; rt->dst.path = &rt->dst; dst_init_metrics(&rt->dst, br_dst_default_metrics, true); - rt->dst.flags = DST_NOXFRM | DST_NOPEER; + rt->dst.flags = DST_NOXFRM | DST_NOPEER | DST_FAKE_RTABLE; rt->dst.ops = &fake_dst_ops; } @@ -694,11 +694,7 @@ static unsigned int br_nf_local_in(unsigned int hook, struct sk_buff *skb, const struct net_device *out, int (*okfn)(struct sk_buff *)) { - struct rtable *rt = skb_rtable(skb); - - if (rt && rt == bridge_parent_rtable(in)) - skb_dst_drop(skb); - + br_drop_fake_rtable(skb); return NF_ACCEPT; } -- cgit v1.2.3 From 151b61284776be2d6f02d48c23c3625678960b97 Mon Sep 17 00:00:00 2001 From: Alan Stern Date: Tue, 24 Apr 2012 14:07:22 -0400 Subject: USB: EHCI: fix crash during suspend on ASUS computers This patch (as1545) fixes a problem affecting several ASUS computers: The machine crashes or corrupts memory when going into suspend if the ehci-hcd driver is bound to any controllers. Users have been forced to unbind or unload ehci-hcd before putting their systems to sleep. After extensive testing, it was determined that the machines don't like going into suspend when any EHCI controllers are in the PCI D3 power state. Presumably this is a firmware bug, but there's nothing we can do about it except to avoid putting the controllers in D3 during system sleep. The patch adds a new flag to indicate whether the problem is present, and avoids changing the controller's power state if the flag is set. Runtime suspend is unaffected; this matters only for system suspend. However as a side effect, the controller will not respond to remote wakeup requests while the system is asleep. Hence USB wakeup is not functional -- but of course, this is already true in the current state of affairs. This fixes Bugzilla #42728. Signed-off-by: Alan Stern Tested-by: Steven Rostedt Tested-by: Andrey Rahmatullin Tested-by: Oleksij Rempel (fishor) Cc: stable Signed-off-by: Greg Kroah-Hartman --- drivers/usb/core/hcd-pci.c | 9 +++++++++ drivers/usb/host/ehci-pci.c | 8 ++++++++ include/linux/usb/hcd.h | 2 ++ 3 files changed, 19 insertions(+) (limited to 'include/linux') diff --git a/drivers/usb/core/hcd-pci.c b/drivers/usb/core/hcd-pci.c index 622b4a48e732..57ed9e400c06 100644 --- a/drivers/usb/core/hcd-pci.c +++ b/drivers/usb/core/hcd-pci.c @@ -493,6 +493,15 @@ static int hcd_pci_suspend_noirq(struct device *dev) pci_save_state(pci_dev); + /* + * Some systems crash if an EHCI controller is in D3 during + * a sleep transition. We have to leave such controllers in D0. + */ + if (hcd->broken_pci_sleep) { + dev_dbg(dev, "Staying in PCI D0\n"); + return retval; + } + /* If the root hub is dead rather than suspended, disallow remote * wakeup. usb_hc_died() should ensure that both hosts are marked as * dying, so we only need to check the primary roothub. diff --git a/drivers/usb/host/ehci-pci.c b/drivers/usb/host/ehci-pci.c index 01bb7241d6ef..fe8dc069164e 100644 --- a/drivers/usb/host/ehci-pci.c +++ b/drivers/usb/host/ehci-pci.c @@ -144,6 +144,14 @@ static int ehci_pci_setup(struct usb_hcd *hcd) hcd->has_tt = 1; tdi_reset(ehci); } + if (pdev->subsystem_vendor == PCI_VENDOR_ID_ASUSTEK) { + /* EHCI #1 or #2 on 6 Series/C200 Series chipset */ + if (pdev->device == 0x1c26 || pdev->device == 0x1c2d) { + ehci_info(ehci, "broken D3 during system sleep on ASUS\n"); + hcd->broken_pci_sleep = 1; + device_set_wakeup_capable(&pdev->dev, false); + } + } break; case PCI_VENDOR_ID_TDI: if (pdev->device == PCI_DEVICE_ID_TDI_EHCI) { diff --git a/include/linux/usb/hcd.h b/include/linux/usb/hcd.h index 5de415707c23..d28cc78a38e4 100644 --- a/include/linux/usb/hcd.h +++ b/include/linux/usb/hcd.h @@ -126,6 +126,8 @@ struct usb_hcd { unsigned wireless:1; /* Wireless USB HCD */ unsigned authorized_default:1; unsigned has_tt:1; /* Integrated TT in root hub */ + unsigned broken_pci_sleep:1; /* Don't put the + controller in PCI-D3 for system sleep */ unsigned int irq; /* irq allocated */ void __iomem *regs; /* device memory/io */ -- cgit v1.2.3 From 904249aa68010c8e223263c922fcbb840a3f42e4 Mon Sep 17 00:00:00 2001 From: Ying Han Date: Wed, 25 Apr 2012 16:01:48 -0700 Subject: mm: fix up the vmscan stat in vmstat The "pgsteal" stat is confusing because it counts both direct reclaim as well as background reclaim. However, we have "kswapd_steal" which also counts background reclaim value. This patch fixes it and also makes it match the existng "pgscan_" stats. Test: pgsteal_kswapd_dma32 447623 pgsteal_kswapd_normal 42272677 pgsteal_kswapd_movable 0 pgsteal_direct_dma32 2801 pgsteal_direct_normal 44353270 pgsteal_direct_movable 0 Signed-off-by: Ying Han Reviewed-by: Rik van Riel Acked-by: Christoph Lameter Cc: Johannes Weiner Cc: Michal Hocko Cc: Mel Gorman Acked-by: KAMEZAWA Hiroyuki Cc: Hillf Danton Cc: Hugh Dickins Cc: Dan Magenheimer Reviewed-by: Minchan Kim Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/vm_event_item.h | 5 +++-- mm/vmscan.c | 11 ++++++++--- mm/vmstat.c | 4 ++-- 3 files changed, 13 insertions(+), 7 deletions(-) (limited to 'include/linux') diff --git a/include/linux/vm_event_item.h b/include/linux/vm_event_item.h index 03b90cdc1921..06f8e3858251 100644 --- a/include/linux/vm_event_item.h +++ b/include/linux/vm_event_item.h @@ -26,13 +26,14 @@ enum vm_event_item { PGPGIN, PGPGOUT, PSWPIN, PSWPOUT, PGFREE, PGACTIVATE, PGDEACTIVATE, PGFAULT, PGMAJFAULT, FOR_ALL_ZONES(PGREFILL), - FOR_ALL_ZONES(PGSTEAL), + FOR_ALL_ZONES(PGSTEAL_KSWAPD), + FOR_ALL_ZONES(PGSTEAL_DIRECT), FOR_ALL_ZONES(PGSCAN_KSWAPD), FOR_ALL_ZONES(PGSCAN_DIRECT), #ifdef CONFIG_NUMA PGSCAN_ZONE_RECLAIM_FAILED, #endif - PGINODESTEAL, SLABS_SCANNED, KSWAPD_STEAL, KSWAPD_INODESTEAL, + PGINODESTEAL, SLABS_SCANNED, KSWAPD_INODESTEAL, KSWAPD_LOW_WMARK_HIT_QUICKLY, KSWAPD_HIGH_WMARK_HIT_QUICKLY, KSWAPD_SKIP_CONGESTION_WAIT, PAGEOUTRUN, ALLOCSTALL, PGROTATED, diff --git a/mm/vmscan.c b/mm/vmscan.c index 1a518684a32f..33dc256033b5 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -1568,9 +1568,14 @@ shrink_inactive_list(unsigned long nr_to_scan, struct mem_cgroup_zone *mz, reclaim_stat->recent_scanned[0] += nr_anon; reclaim_stat->recent_scanned[1] += nr_file; - if (current_is_kswapd()) - __count_vm_events(KSWAPD_STEAL, nr_reclaimed); - __count_zone_vm_events(PGSTEAL, zone, nr_reclaimed); + if (global_reclaim(sc)) { + if (current_is_kswapd()) + __count_zone_vm_events(PGSTEAL_KSWAPD, zone, + nr_reclaimed); + else + __count_zone_vm_events(PGSTEAL_DIRECT, zone, + nr_reclaimed); + } putback_inactive_pages(mz, &page_list); diff --git a/mm/vmstat.c b/mm/vmstat.c index f600557a7659..7db1b9bab492 100644 --- a/mm/vmstat.c +++ b/mm/vmstat.c @@ -738,7 +738,8 @@ const char * const vmstat_text[] = { "pgmajfault", TEXTS_FOR_ZONES("pgrefill") - TEXTS_FOR_ZONES("pgsteal") + TEXTS_FOR_ZONES("pgsteal_kswapd") + TEXTS_FOR_ZONES("pgsteal_direct") TEXTS_FOR_ZONES("pgscan_kswapd") TEXTS_FOR_ZONES("pgscan_direct") @@ -747,7 +748,6 @@ const char * const vmstat_text[] = { #endif "pginodesteal", "slabs_scanned", - "kswapd_steal", "kswapd_inodesteal", "kswapd_low_wmark_hit_quickly", "kswapd_high_wmark_hit_quickly", -- cgit v1.2.3 From b95ace54a23e2f8ebb032744cebb17c9f43bf651 Mon Sep 17 00:00:00 2001 From: Robert Jarzmik Date: Sun, 22 Apr 2012 13:37:24 +0200 Subject: ARM: pxa: fix gpio wakeup setting In 3.3, gpio wakeup setting was broken. The call enable_irq_wake() didn't set up the PXA gpio registers (PWER, ...) anymore. Fix it at least for pxa27x. The driver doesn't seem to be used in pxa25x (weird ...), and the fix doesn't extend to pxa3xx and pxa95x (which don't have a gpio_set_wake() available). Signed-off-by: Robert Jarzmik Signed-off-by: Haojian Zhuang --- arch/arm/mach-pxa/pxa27x.c | 6 +++++- drivers/gpio/gpio-pxa.c | 21 +++++++++++++++++++-- include/linux/gpio-pxa.h | 4 ++++ 3 files changed, 28 insertions(+), 3 deletions(-) (limited to 'include/linux') diff --git a/arch/arm/mach-pxa/pxa27x.c b/arch/arm/mach-pxa/pxa27x.c index 6bce78edce7a..4726c246dcdc 100644 --- a/arch/arm/mach-pxa/pxa27x.c +++ b/arch/arm/mach-pxa/pxa27x.c @@ -421,8 +421,11 @@ void __init pxa27x_set_i2c_power_info(struct i2c_pxa_platform_data *info) pxa_register_device(&pxa27x_device_i2c_power, info); } +static struct pxa_gpio_platform_data pxa27x_gpio_info __initdata = { + .gpio_set_wake = gpio_set_wake, +}; + static struct platform_device *devices[] __initdata = { - &pxa_device_gpio, &pxa27x_device_udc, &pxa_device_pmu, &pxa_device_i2s, @@ -458,6 +461,7 @@ static int __init pxa27x_init(void) register_syscore_ops(&pxa2xx_mfp_syscore_ops); register_syscore_ops(&pxa2xx_clock_syscore_ops); + pxa_register_device(&pxa_device_gpio, &pxa27x_gpio_info); ret = platform_add_devices(devices, ARRAY_SIZE(devices)); } diff --git a/drivers/gpio/gpio-pxa.c b/drivers/gpio/gpio-pxa.c index 5689ce62fd81..fc3ace3fd4cb 100644 --- a/drivers/gpio/gpio-pxa.c +++ b/drivers/gpio/gpio-pxa.c @@ -64,6 +64,7 @@ struct pxa_gpio_chip { unsigned long irq_mask; unsigned long irq_edge_rise; unsigned long irq_edge_fall; + int (*set_wake)(unsigned int gpio, unsigned int on); #ifdef CONFIG_PM unsigned long saved_gplr; @@ -269,7 +270,8 @@ static void pxa_gpio_set(struct gpio_chip *chip, unsigned offset, int value) (value ? GPSR_OFFSET : GPCR_OFFSET)); } -static int __devinit pxa_init_gpio_chip(int gpio_end) +static int __devinit pxa_init_gpio_chip(int gpio_end, + int (*set_wake)(unsigned int, unsigned int)) { int i, gpio, nbanks = gpio_to_bank(gpio_end) + 1; struct pxa_gpio_chip *chips; @@ -285,6 +287,7 @@ static int __devinit pxa_init_gpio_chip(int gpio_end) sprintf(chips[i].label, "gpio-%d", i); chips[i].regbase = gpio_reg_base + BANK_OFF(i); + chips[i].set_wake = set_wake; c->base = gpio; c->label = chips[i].label; @@ -412,6 +415,17 @@ static void pxa_mask_muxed_gpio(struct irq_data *d) writel_relaxed(gfer, c->regbase + GFER_OFFSET); } +static int pxa_gpio_set_wake(struct irq_data *d, unsigned int on) +{ + int gpio = pxa_irq_to_gpio(d->irq); + struct pxa_gpio_chip *c = gpio_to_pxachip(gpio); + + if (c->set_wake) + return c->set_wake(gpio, on); + else + return 0; +} + static void pxa_unmask_muxed_gpio(struct irq_data *d) { int gpio = pxa_irq_to_gpio(d->irq); @@ -427,6 +441,7 @@ static struct irq_chip pxa_muxed_gpio_chip = { .irq_mask = pxa_mask_muxed_gpio, .irq_unmask = pxa_unmask_muxed_gpio, .irq_set_type = pxa_gpio_irq_type, + .irq_set_wake = pxa_gpio_set_wake, }; static int pxa_gpio_nums(void) @@ -471,6 +486,7 @@ static int __devinit pxa_gpio_probe(struct platform_device *pdev) struct pxa_gpio_chip *c; struct resource *res; struct clk *clk; + struct pxa_gpio_platform_data *info; int gpio, irq, ret; int irq0 = 0, irq1 = 0, irq_mux, gpio_offset = 0; @@ -516,7 +532,8 @@ static int __devinit pxa_gpio_probe(struct platform_device *pdev) } /* Initialize GPIO chips */ - pxa_init_gpio_chip(pxa_last_gpio); + info = dev_get_platdata(&pdev->dev); + pxa_init_gpio_chip(pxa_last_gpio, info ? info->gpio_set_wake : NULL); /* clear all GPIO edge detects */ for_each_gpio_chip(gpio, c) { diff --git a/include/linux/gpio-pxa.h b/include/linux/gpio-pxa.h index 05071ee34c3f..d755b28ba635 100644 --- a/include/linux/gpio-pxa.h +++ b/include/linux/gpio-pxa.h @@ -13,4 +13,8 @@ extern int pxa_last_gpio; extern int pxa_irq_to_gpio(int irq); +struct pxa_gpio_platform_data { + int (*gpio_set_wake)(unsigned int gpio, unsigned int on); +}; + #endif /* __GPIO_PXA_H */ -- cgit v1.2.3 From dbabe0d659d3cfe42830a779909ab3cd42f7b027 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Tue, 17 Apr 2012 17:03:50 -0700 Subject: spi: fix spi.h kernel-doc warning Fix kernel-doc warning in spi.h (copy/paste): Warning(include/linux/spi/spi.h:365): No description found for parameter 'unprepare_transfer_hardware' Signed-off-by: Randy Dunlap Signed-off-by: Grant Likely --- include/linux/spi/spi.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/spi/spi.h b/include/linux/spi/spi.h index 98679b061b63..fa702aeb5038 100644 --- a/include/linux/spi/spi.h +++ b/include/linux/spi/spi.h @@ -254,7 +254,7 @@ static inline void spi_unregister_driver(struct spi_driver *sdrv) * driver is finished with this message, it must call * spi_finalize_current_message() so the subsystem can issue the next * transfer - * @prepare_transfer_hardware: there are currently no more messages on the + * @unprepare_transfer_hardware: there are currently no more messages on the * queue so the subsystem notifies the driver that it may relax the * hardware by issuing this call * -- cgit v1.2.3 From 9883035ae7edef3ec62ad215611cb8e17d6a1a5d Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Sun, 29 Apr 2012 13:12:42 -0700 Subject: pipes: add a "packetized pipe" mode for writing The actual internal pipe implementation is already really about individual packets (called "pipe buffers"), and this simply exposes that as a special packetized mode. When we are in the packetized mode (marked by O_DIRECT as suggested by Alan Cox), a write() on a pipe will not merge the new data with previous writes, so each write will get a pipe buffer of its own. The pipe buffer is then marked with the PIPE_BUF_FLAG_PACKET flag, which in turn will tell the reader side to break the read at that boundary (and throw away any partial packet contents that do not fit in the read buffer). End result: as long as you do writes less than PIPE_BUF in size (so that the pipe doesn't have to split them up), you can now treat the pipe as a packet interface, where each read() system call will read one packet at a time. You can just use a sufficiently big read buffer (PIPE_BUF is sufficient, since bigger than that doesn't guarantee atomicity anyway), and the return value of the read() will naturally give you the size of the packet. NOTE! We do not support zero-sized packets, and zero-sized reads and writes to a pipe continue to be no-ops. Also note that big packets will currently be split at write time, but that the size at which that happens is not really specified (except that it's bigger than PIPE_BUF). Currently that limit is the system page size, but we might want to explicitly support bigger packets some day. The main user for this is going to be the autofs packet interface, allowing us to stop having to care so deeply about exact packet sizes (which have had bugs with 32/64-bit compatibility modes). But user space can create packetized pipes with "pipe2(fd, O_DIRECT)", which will fail with an EINVAL on kernels that do not support this interface. Tested-by: Michael Tokarev Cc: Alan Cox Cc: David Miller Cc: Ian Kent Cc: Thomas Meyer Cc: stable@kernel.org # needed for systemd/autofs interaction fix Signed-off-by: Linus Torvalds --- fs/pipe.c | 31 +++++++++++++++++++++++++++++-- include/linux/pipe_fs_i.h | 1 + 2 files changed, 30 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/fs/pipe.c b/fs/pipe.c index 25feaa3faac0..fec5e4ad071a 100644 --- a/fs/pipe.c +++ b/fs/pipe.c @@ -346,6 +346,16 @@ static const struct pipe_buf_operations anon_pipe_buf_ops = { .get = generic_pipe_buf_get, }; +static const struct pipe_buf_operations packet_pipe_buf_ops = { + .can_merge = 0, + .map = generic_pipe_buf_map, + .unmap = generic_pipe_buf_unmap, + .confirm = generic_pipe_buf_confirm, + .release = anon_pipe_buf_release, + .steal = generic_pipe_buf_steal, + .get = generic_pipe_buf_get, +}; + static ssize_t pipe_read(struct kiocb *iocb, const struct iovec *_iov, unsigned long nr_segs, loff_t pos) @@ -407,6 +417,13 @@ redo: ret += chars; buf->offset += chars; buf->len -= chars; + + /* Was it a packet buffer? Clean up and exit */ + if (buf->flags & PIPE_BUF_FLAG_PACKET) { + total_len = chars; + buf->len = 0; + } + if (!buf->len) { buf->ops = NULL; ops->release(pipe, buf); @@ -459,6 +476,11 @@ redo: return ret; } +static inline int is_packetized(struct file *file) +{ + return (file->f_flags & O_DIRECT) != 0; +} + static ssize_t pipe_write(struct kiocb *iocb, const struct iovec *_iov, unsigned long nr_segs, loff_t ppos) @@ -593,6 +615,11 @@ redo2: buf->ops = &anon_pipe_buf_ops; buf->offset = 0; buf->len = chars; + buf->flags = 0; + if (is_packetized(filp)) { + buf->ops = &packet_pipe_buf_ops; + buf->flags = PIPE_BUF_FLAG_PACKET; + } pipe->nrbufs = ++bufs; pipe->tmp_page = NULL; @@ -1013,7 +1040,7 @@ struct file *create_write_pipe(int flags) goto err_dentry; f->f_mapping = inode->i_mapping; - f->f_flags = O_WRONLY | (flags & O_NONBLOCK); + f->f_flags = O_WRONLY | (flags & (O_NONBLOCK | O_DIRECT)); f->f_version = 0; return f; @@ -1057,7 +1084,7 @@ int do_pipe_flags(int *fd, int flags) int error; int fdw, fdr; - if (flags & ~(O_CLOEXEC | O_NONBLOCK)) + if (flags & ~(O_CLOEXEC | O_NONBLOCK | O_DIRECT)) return -EINVAL; fw = create_write_pipe(flags); diff --git a/include/linux/pipe_fs_i.h b/include/linux/pipe_fs_i.h index 6d626ff0cfd0..e1ac1ce16fb0 100644 --- a/include/linux/pipe_fs_i.h +++ b/include/linux/pipe_fs_i.h @@ -6,6 +6,7 @@ #define PIPE_BUF_FLAG_LRU 0x01 /* page is on the LRU */ #define PIPE_BUF_FLAG_ATOMIC 0x02 /* was atomically mapped */ #define PIPE_BUF_FLAG_GIFT 0x04 /* page is a gift */ +#define PIPE_BUF_FLAG_PACKET 0x08 /* read() as a packet */ /** * struct pipe_buffer - a linux kernel pipe buffer -- cgit v1.2.3 From 41b3254c93acc56adc3c4477fef7c9512d47659e Mon Sep 17 00:00:00 2001 From: Matthew Garrett Date: Mon, 30 Apr 2012 16:11:29 -0400 Subject: efi: Add new variable attributes More recent versions of the UEFI spec have added new attributes for variables. Add them. Signed-off-by: Matthew Garrett Cc: stable@vger.kernel.org Signed-off-by: Linus Torvalds --- include/linux/efi.h | 13 ++++++++++++- 1 file changed, 12 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/efi.h b/include/linux/efi.h index 88ec80670d5f..ec45ccd8708a 100644 --- a/include/linux/efi.h +++ b/include/linux/efi.h @@ -554,7 +554,18 @@ extern int __init efi_setup_pcdp_console(char *); #define EFI_VARIABLE_NON_VOLATILE 0x0000000000000001 #define EFI_VARIABLE_BOOTSERVICE_ACCESS 0x0000000000000002 #define EFI_VARIABLE_RUNTIME_ACCESS 0x0000000000000004 - +#define EFI_VARIABLE_HARDWARE_ERROR_RECORD 0x0000000000000008 +#define EFI_VARIABLE_AUTHENTICATED_WRITE_ACCESS 0x0000000000000010 +#define EFI_VARIABLE_TIME_BASED_AUTHENTICATED_WRITE_ACCESS 0x0000000000000020 +#define EFI_VARIABLE_APPEND_WRITE 0x0000000000000040 + +#define EFI_VARIABLE_MASK (EFI_VARIABLE_NON_VOLATILE | \ + EFI_VARIABLE_BOOTSERVICE_ACCESS | \ + EFI_VARIABLE_RUNTIME_ACCESS | \ + EFI_VARIABLE_HARDWARE_ERROR_RECORD | \ + EFI_VARIABLE_AUTHENTICATED_WRITE_ACCESS | \ + EFI_VARIABLE_TIME_BASED_AUTHENTICATED_WRITE_ACCESS | \ + EFI_VARIABLE_APPEND_WRITE) /* * The type of search to perform when calling boottime->locate_handle */ -- cgit v1.2.3 From d961949660fa1c1b7eb0c3a3c157989c90f14e8e Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 30 Apr 2012 21:29:16 +0000 Subject: net: fix two typos in skbuff.h fix kernel doc typos in function names Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/linux/skbuff.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 775292a66fa4..111f26b6e28b 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -1020,7 +1020,7 @@ static inline void skb_queue_splice(const struct sk_buff_head *list, } /** - * skb_queue_splice - join two skb lists and reinitialise the emptied list + * skb_queue_splice_init - join two skb lists and reinitialise the emptied list * @list: the new list to add * @head: the place to add it in the first list * @@ -1051,7 +1051,7 @@ static inline void skb_queue_splice_tail(const struct sk_buff_head *list, } /** - * skb_queue_splice_tail - join two skb lists and reinitialise the emptied list + * skb_queue_splice_tail_init - join two skb lists and reinitialise the emptied list * @list: the new list to add * @head: the place to add it in the first list * -- cgit v1.2.3 From 1c430a727fa512500a422ffe4712166c550ea06a Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Mon, 7 May 2012 15:39:06 +0200 Subject: net: compare_ether_addr[_64bits]() has no ordering Neither compare_ether_addr() nor compare_ether_addr_64bits() (as it can fall back to the former) have comparison semantics like memcmp() where the sign of the return value indicates sort order. We had a bug in the wireless code due to a blind memcmp replacement because of this. A cursory look suggests that the wireless bug was the only one due to this semantic difference. Signed-off-by: Johannes Berg Signed-off-by: David S. Miller --- include/linux/etherdevice.h | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-) (limited to 'include/linux') diff --git a/include/linux/etherdevice.h b/include/linux/etherdevice.h index 8a1835855faa..fe5136d81454 100644 --- a/include/linux/etherdevice.h +++ b/include/linux/etherdevice.h @@ -159,7 +159,8 @@ static inline void eth_hw_addr_random(struct net_device *dev) * @addr1: Pointer to a six-byte array containing the Ethernet address * @addr2: Pointer other six-byte array containing the Ethernet address * - * Compare two ethernet addresses, returns 0 if equal + * Compare two ethernet addresses, returns 0 if equal, non-zero otherwise. + * Unlike memcmp(), it doesn't return a value suitable for sorting. */ static inline unsigned compare_ether_addr(const u8 *addr1, const u8 *addr2) { @@ -184,10 +185,10 @@ static inline unsigned long zap_last_2bytes(unsigned long value) * @addr1: Pointer to an array of 8 bytes * @addr2: Pointer to an other array of 8 bytes * - * Compare two ethernet addresses, returns 0 if equal. - * Same result than "memcmp(addr1, addr2, ETH_ALEN)" but without conditional - * branches, and possibly long word memory accesses on CPU allowing cheap - * unaligned memory reads. + * Compare two ethernet addresses, returns 0 if equal, non-zero otherwise. + * Unlike memcmp(), it doesn't return a value suitable for sorting. + * The function doesn't need any conditional branches and possibly uses + * word memory accesses on CPU allowing cheap unaligned memory reads. * arrays = { byte1, byte2, byte3, byte4, byte6, byte7, pad1, pad2} * * Please note that alignment of addr1 & addr2 is only guaranted to be 16 bits. -- cgit v1.2.3