From 18cb2aef91b37dbce2bec2f39bb1dddd0e9dd838 Mon Sep 17 00:00:00 2001 From: Namhyung Kim Date: Sat, 7 Aug 2010 03:26:23 +0900 Subject: percpu: handle __percpu notations in UP accessors UP accessors didn't take care of __percpu notations leading to a lot of spurious sparse warnings on UP configurations. Fix it. Signed-off-by: Namhyung Kim Signed-off-by: Tejun Heo --- include/linux/percpu.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/percpu.h b/include/linux/percpu.h index b8b9084527b1..49466b13c5c6 100644 --- a/include/linux/percpu.h +++ b/include/linux/percpu.h @@ -149,7 +149,7 @@ extern void __init percpu_init_late(void); #else /* CONFIG_SMP */ -#define per_cpu_ptr(ptr, cpu) ({ (void)(cpu); (ptr); }) +#define per_cpu_ptr(ptr, cpu) ({ (void)(cpu); VERIFY_PERCPU_PTR((ptr)); }) /* can't distinguish from other static vars, always false */ static inline bool is_kernel_percpu_address(unsigned long addr) -- cgit v1.2.3 From 92298e668372f2f6c8a79fb272f13d65161a4876 Mon Sep 17 00:00:00 2001 From: Dave Airlie Date: Fri, 13 Aug 2010 10:22:17 +1000 Subject: PCI: provide stub pci_domain_nr function for !CONFIG_PCI configs Allows the new PCI domain aware DRM code to compile on m68k. Reported-by: Geert Uytterhoeven Signed-off-by: Dave Airlie Signed-off-by: Jesse Barnes --- include/linux/pci.h | 3 +++ 1 file changed, 3 insertions(+) (limited to 'include/linux') diff --git a/include/linux/pci.h b/include/linux/pci.h index b1d17956a153..c8d95e369ff4 100644 --- a/include/linux/pci.h +++ b/include/linux/pci.h @@ -1214,6 +1214,9 @@ static inline struct pci_dev *pci_get_bus_and_slot(unsigned int bus, unsigned int devfn) { return NULL; } +static inline int pci_domain_nr(struct pci_bus *bus) +{ return 0; } + #define dev_is_pci(d) (false) #define dev_is_pf(d) (false) #define dev_num_vf(d) (0) -- cgit v1.2.3 From 5dd531a03ad721b41911ddb32e6e0481404e7aaf Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Mon, 23 Aug 2010 13:52:19 +0200 Subject: block: add function call to switch the IO scheduler from a driver Currently drivers must do an elevator_exit() + elevator_init() to switch IO schedulers. There are a few problems with this: - Since commit 1abec4fdbb142e3ccb6ce99832fae42129134a96, elevator_init() requires a zeroed out q->elevator pointer. The two existing in-kernel users don't do that. - It will only work at initialization time, since using the above two-staged construct does not properly quisce the queue. So add elevator_change() which takes care of this, and convert the elv_iosched_store() sysfs interface to use this helper as well. Reported-by: Peter Oberparleiter Reported-by: Kevin Vigor Signed-off-by: Jens Axboe --- block/elevator.c | 44 +++++++++++++++++++++++++++++++------------- include/linux/elevator.h | 1 + 2 files changed, 32 insertions(+), 13 deletions(-) (limited to 'include/linux') diff --git a/block/elevator.c b/block/elevator.c index ec585c9554d3..205b09a5bd9e 100644 --- a/block/elevator.c +++ b/block/elevator.c @@ -1009,18 +1009,19 @@ static int elevator_switch(struct request_queue *q, struct elevator_type *new_e) { struct elevator_queue *old_elevator, *e; void *data; + int err; /* * Allocate new elevator */ e = elevator_alloc(q, new_e); if (!e) - return 0; + return -ENOMEM; data = elevator_init_queue(q, e); if (!data) { kobject_put(&e->kobj); - return 0; + return -ENOMEM; } /* @@ -1043,7 +1044,8 @@ static int elevator_switch(struct request_queue *q, struct elevator_type *new_e) __elv_unregister_queue(old_elevator); - if (elv_register_queue(q)) + err = elv_register_queue(q); + if (err) goto fail_register; /* @@ -1056,7 +1058,7 @@ static int elevator_switch(struct request_queue *q, struct elevator_type *new_e) blk_add_trace_msg(q, "elv switch: %s", e->elevator_type->elevator_name); - return 1; + return 0; fail_register: /* @@ -1071,17 +1073,19 @@ fail_register: queue_flag_clear(QUEUE_FLAG_ELVSWITCH, q); spin_unlock_irq(q->queue_lock); - return 0; + return err; } -ssize_t elv_iosched_store(struct request_queue *q, const char *name, - size_t count) +/* + * Switch this queue to the given IO scheduler. + */ +int elevator_change(struct request_queue *q, const char *name) { char elevator_name[ELV_NAME_MAX]; struct elevator_type *e; if (!q->elevator) - return count; + return -ENXIO; strlcpy(elevator_name, name, sizeof(elevator_name)); e = elevator_get(strstrip(elevator_name)); @@ -1092,13 +1096,27 @@ ssize_t elv_iosched_store(struct request_queue *q, const char *name, if (!strcmp(elevator_name, q->elevator->elevator_type->elevator_name)) { elevator_put(e); - return count; + return 0; } - if (!elevator_switch(q, e)) - printk(KERN_ERR "elevator: switch to %s failed\n", - elevator_name); - return count; + return elevator_switch(q, e); +} +EXPORT_SYMBOL(elevator_change); + +ssize_t elv_iosched_store(struct request_queue *q, const char *name, + size_t count) +{ + int ret; + + if (!q->elevator) + return count; + + ret = elevator_change(q, name); + if (!ret) + return count; + + printk(KERN_ERR "elevator: switch to %s failed\n", name); + return ret; } ssize_t elv_iosched_show(struct request_queue *q, char *name) diff --git a/include/linux/elevator.h b/include/linux/elevator.h index 2c958f4fce1e..926b50322a46 100644 --- a/include/linux/elevator.h +++ b/include/linux/elevator.h @@ -136,6 +136,7 @@ extern ssize_t elv_iosched_store(struct request_queue *, const char *, size_t); extern int elevator_init(struct request_queue *, char *); extern void elevator_exit(struct elevator_queue *); +extern int elevator_change(struct request_queue *, const char *); extern int elv_rq_merge_ok(struct request *, struct bio *); /* -- cgit v1.2.3 From e41e704bc4f49057fc68b643108366e6e6781aa3 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Tue, 24 Aug 2010 14:22:47 +0200 Subject: workqueue: improve destroy_workqueue() debuggability Now that the worklist is global, having works pending after wq destruction can easily lead to oops and destroy_workqueue() have several BUG_ON()s to catch these cases. Unfortunately, BUG_ON() doesn't tell much about how the work became pending after the final flush_workqueue(). This patch adds WQ_DYING which is set before the final flush begins. If a work is requested to be queued on a dying workqueue, WARN_ON_ONCE() is triggered and the request is ignored. This clearly indicates which caller is trying to queue a work on a dying workqueue and keeps the system working in most cases. Locking rule comment is updated such that the 'I' rule includes modifying the field from destruction path. Signed-off-by: Tejun Heo --- include/linux/workqueue.h | 2 ++ kernel/workqueue.c | 7 ++++++- 2 files changed, 8 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/workqueue.h b/include/linux/workqueue.h index 4f9d277bcd9a..c959666eafca 100644 --- a/include/linux/workqueue.h +++ b/include/linux/workqueue.h @@ -241,6 +241,8 @@ enum { WQ_HIGHPRI = 1 << 4, /* high priority */ WQ_CPU_INTENSIVE = 1 << 5, /* cpu instensive workqueue */ + WQ_DYING = 1 << 6, /* internal: workqueue is dying */ + WQ_MAX_ACTIVE = 512, /* I like 512, better ideas? */ WQ_MAX_UNBOUND_PER_CPU = 4, /* 4 * #cpus for unbound wq */ WQ_DFL_ACTIVE = WQ_MAX_ACTIVE / 2, diff --git a/kernel/workqueue.c b/kernel/workqueue.c index cc3456f96c56..362b50d092e2 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -87,7 +87,8 @@ enum { /* * Structure fields follow one of the following exclusion rules. * - * I: Set during initialization and read-only afterwards. + * I: Modifiable by initialization/destruction paths and read-only for + * everyone else. * * P: Preemption protected. Disabling preemption is enough and should * only be modified and accessed from the local cpu. @@ -944,6 +945,9 @@ static void __queue_work(unsigned int cpu, struct workqueue_struct *wq, debug_work_activate(work); + if (WARN_ON_ONCE(wq->flags & WQ_DYING)) + return; + /* determine gcwq to use */ if (!(wq->flags & WQ_UNBOUND)) { struct global_cwq *last_gcwq; @@ -2828,6 +2832,7 @@ void destroy_workqueue(struct workqueue_struct *wq) { unsigned int cpu; + wq->flags |= WQ_DYING; flush_workqueue(wq); /* -- cgit v1.2.3 From 75fb60f26befb59dbfa05cb122972642b7bdd219 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Mon, 23 Aug 2010 23:53:11 +0200 Subject: ACPI/PCI: Negotiate _OSC control bits before requesting them It is possible that the BIOS will not grant control of all _OSC features requested via acpi_pci_osc_control_set(), so it is recommended to negotiate the final set of _OSC features with the query flag set before calling _OSC to request control of these features. To implement it, rework acpi_pci_osc_control_set() so that the caller can specify the mask of _OSC control bits to negotiate and the mask of _OSC control bits that are absolutely necessary to it. Then, acpi_pci_osc_control_set() will run _OSC queries in a loop until the mask of _OSC control bits returned by the BIOS is equal to the mask passed to it. Also, before running the _OSC request acpi_pci_osc_control_set() will check if the caller's required control bits are present in the final mask. Using this mechanism we will be able to avoid situations in which the BIOS doesn't grant control of certain _OSC features, because they depend on some other _OSC features that have not been requested. Signed-off-by: Rafael J. Wysocki Signed-off-by: Jesse Barnes --- drivers/acpi/pci_root.c | 59 +++++++++++++++++++++++------------- drivers/pci/hotplug/acpi_pcihp.c | 2 +- drivers/pci/pcie/aer/aerdrv_acpi.c | 6 ++-- drivers/pci/pcie/pme/pcie_pme_acpi.c | 8 +++-- include/linux/acpi.h | 4 +-- 5 files changed, 49 insertions(+), 30 deletions(-) (limited to 'include/linux') diff --git a/drivers/acpi/pci_root.c b/drivers/acpi/pci_root.c index 77cd19697b1e..c34713112520 100644 --- a/drivers/acpi/pci_root.c +++ b/drivers/acpi/pci_root.c @@ -374,21 +374,32 @@ out: EXPORT_SYMBOL_GPL(acpi_get_pci_dev); /** - * acpi_pci_osc_control_set - commit requested control to Firmware - * @handle: acpi_handle for the target ACPI object - * @flags: driver's requested control bits + * acpi_pci_osc_control_set - Request control of PCI root _OSC features. + * @handle: ACPI handle of a PCI root bridge (or PCIe Root Complex). + * @mask: Mask of _OSC bits to request control of, place to store control mask. + * @req: Mask of _OSC bits the control of is essential to the caller. * - * Attempt to take control from Firmware on requested control bits. + * Run _OSC query for @mask and if that is successful, compare the returned + * mask of control bits with @req. If all of the @req bits are set in the + * returned mask, run _OSC request for it. + * + * The variable at the @mask address may be modified regardless of whether or + * not the function returns success. On success it will contain the mask of + * _OSC bits the BIOS has granted control of, but its contents are meaningless + * on failure. **/ -acpi_status acpi_pci_osc_control_set(acpi_handle handle, u32 flags) +acpi_status acpi_pci_osc_control_set(acpi_handle handle, u32 *mask, u32 req) { + struct acpi_pci_root *root; acpi_status status; - u32 control_req, result, capbuf[3]; + u32 ctrl, capbuf[3]; acpi_handle tmp; - struct acpi_pci_root *root; - control_req = (flags & OSC_PCI_CONTROL_MASKS); - if (!control_req) + if (!mask) + return AE_BAD_PARAMETER; + + ctrl = *mask & OSC_PCI_CONTROL_MASKS; + if ((ctrl & req) != req) return AE_TYPE; root = acpi_pci_find_root(handle); @@ -400,27 +411,33 @@ acpi_status acpi_pci_osc_control_set(acpi_handle handle, u32 flags) return status; mutex_lock(&osc_lock); + + *mask = ctrl | root->osc_control_set; /* No need to evaluate _OSC if the control was already granted. */ - if ((root->osc_control_set & control_req) == control_req) + if ((root->osc_control_set & ctrl) == ctrl) goto out; - /* Need to query controls first before requesting them */ - flags = control_req; - status = acpi_pci_query_osc(root, root->osc_support_set, &flags); - if (ACPI_FAILURE(status)) - goto out; + /* Need to check the available controls bits before requesting them. */ + while (*mask) { + status = acpi_pci_query_osc(root, root->osc_support_set, mask); + if (ACPI_FAILURE(status)) + goto out; + if (ctrl == *mask) + break; + ctrl = *mask; + } - if (flags != control_req) { + if ((ctrl & req) != req) { status = AE_SUPPORT; goto out; } capbuf[OSC_QUERY_TYPE] = 0; capbuf[OSC_SUPPORT_TYPE] = root->osc_support_set; - capbuf[OSC_CONTROL_TYPE] = root->osc_control_set | control_req; - status = acpi_pci_run_osc(handle, capbuf, &result); + capbuf[OSC_CONTROL_TYPE] = ctrl; + status = acpi_pci_run_osc(handle, capbuf, mask); if (ACPI_SUCCESS(status)) - root->osc_control_set = result; + root->osc_control_set = *mask; out: mutex_unlock(&osc_lock); return status; @@ -551,8 +568,8 @@ static int __devinit acpi_pci_root_add(struct acpi_device *device) if (flags != base_flags) acpi_pci_osc_support(root, flags); - status = acpi_pci_osc_control_set(root->device->handle, - OSC_PCI_EXPRESS_CAP_STRUCTURE_CONTROL); + flags = OSC_PCI_EXPRESS_CAP_STRUCTURE_CONTROL; + status = acpi_pci_osc_control_set(root->device->handle, &flags, flags); if (ACPI_FAILURE(status)) { printk(KERN_INFO "Unable to assume PCIe control: Disabling ASPM\n"); diff --git a/drivers/pci/hotplug/acpi_pcihp.c b/drivers/pci/hotplug/acpi_pcihp.c index 45fcc1e96df9..3d93d529a7bd 100644 --- a/drivers/pci/hotplug/acpi_pcihp.c +++ b/drivers/pci/hotplug/acpi_pcihp.c @@ -360,7 +360,7 @@ int acpi_get_hp_hw_control_from_firmware(struct pci_dev *pdev, u32 flags) acpi_get_name(handle, ACPI_FULL_PATHNAME, &string); dbg("Trying to get hotplug control for %s\n", (char *)string.pointer); - status = acpi_pci_osc_control_set(handle, flags); + status = acpi_pci_osc_control_set(handle, &flags, flags); if (ACPI_SUCCESS(status)) goto got_one; if (status == AE_SUPPORT) diff --git a/drivers/pci/pcie/aer/aerdrv_acpi.c b/drivers/pci/pcie/aer/aerdrv_acpi.c index f278d7b0d95d..3a276a0cea93 100644 --- a/drivers/pci/pcie/aer/aerdrv_acpi.c +++ b/drivers/pci/pcie/aer/aerdrv_acpi.c @@ -39,9 +39,9 @@ int aer_osc_setup(struct pcie_device *pciedev) handle = acpi_find_root_bridge_handle(pdev); if (handle) { - status = acpi_pci_osc_control_set(handle, - OSC_PCI_EXPRESS_AER_CONTROL | - OSC_PCI_EXPRESS_CAP_STRUCTURE_CONTROL); + u32 flags = OSC_PCI_EXPRESS_AER_CONTROL | + OSC_PCI_EXPRESS_CAP_STRUCTURE_CONTROL; + status = acpi_pci_osc_control_set(handle, &flags, flags); } if (ACPI_FAILURE(status)) { diff --git a/drivers/pci/pcie/pme/pcie_pme_acpi.c b/drivers/pci/pcie/pme/pcie_pme_acpi.c index 83ab2287ae3f..be20222b12d3 100644 --- a/drivers/pci/pcie/pme/pcie_pme_acpi.c +++ b/drivers/pci/pcie/pme/pcie_pme_acpi.c @@ -28,6 +28,7 @@ int pcie_pme_acpi_setup(struct pcie_device *srv) acpi_status status = AE_NOT_FOUND; struct pci_dev *port = srv->port; acpi_handle handle; + u32 flags; int error = 0; if (acpi_pci_disabled) @@ -39,9 +40,10 @@ int pcie_pme_acpi_setup(struct pcie_device *srv) if (!handle) return -EINVAL; - status = acpi_pci_osc_control_set(handle, - OSC_PCI_EXPRESS_PME_CONTROL | - OSC_PCI_EXPRESS_CAP_STRUCTURE_CONTROL); + flags = OSC_PCI_EXPRESS_PME_CONTROL | + OSC_PCI_EXPRESS_CAP_STRUCTURE_CONTROL; + + status = acpi_pci_osc_control_set(handle, &flags, flags); if (ACPI_FAILURE(status)) { dev_info(&port->dev, "Failed to receive control of PCIe PME service: %s\n", diff --git a/include/linux/acpi.h b/include/linux/acpi.h index ccf94dc5acdf..c227757feb06 100644 --- a/include/linux/acpi.h +++ b/include/linux/acpi.h @@ -304,8 +304,8 @@ acpi_status acpi_run_osc(acpi_handle handle, struct acpi_osc_context *context); OSC_PCI_EXPRESS_PME_CONTROL | \ OSC_PCI_EXPRESS_AER_CONTROL | \ OSC_PCI_EXPRESS_CAP_STRUCTURE_CONTROL) - -extern acpi_status acpi_pci_osc_control_set(acpi_handle handle, u32 flags); +extern acpi_status acpi_pci_osc_control_set(acpi_handle handle, + u32 *mask, u32 req); extern void acpi_early_init(void); #else /* !CONFIG_ACPI */ -- cgit v1.2.3 From 8a2e8e5dec7e29c56a46ba176c664ab6a3d04118 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Wed, 25 Aug 2010 10:33:56 +0200 Subject: workqueue: fix cwq->nr_active underflow cwq->nr_active is used to keep track of how many work items are active for the cpu workqueue, where 'active' is defined as either pending on global worklist or executing. This is used to implement the max_active limit and workqueue freezing. If a work item is queued after nr_active has already reached max_active, the work item doesn't increment nr_active and is put on the delayed queue and gets activated later as previous active work items retire. try_to_grab_pending() which is used in the cancellation path unconditionally decremented nr_active whether the work item being cancelled is currently active or delayed, so cancelling a delayed work item makes nr_active underflow. This breaks max_active enforcement and triggers BUG_ON() in destroy_workqueue() later on. This patch fixes this bug by adding a flag WORK_STRUCT_DELAYED, which is set while a work item in on the delayed list and making try_to_grab_pending() decrement nr_active iff the work item is currently active. The addition of the flag enlarges cwq alignment to 256 bytes which is getting a bit too large. It's scheduled to be reduced back to 128 bytes by merging WORK_STRUCT_PENDING and WORK_STRUCT_CWQ in the next devel cycle. Signed-off-by: Tejun Heo Reported-by: Johannes Berg --- include/linux/workqueue.h | 16 +++++++++------- kernel/workqueue.c | 30 ++++++++++++++++++++---------- 2 files changed, 29 insertions(+), 17 deletions(-) (limited to 'include/linux') diff --git a/include/linux/workqueue.h b/include/linux/workqueue.h index c959666eafca..f11100f96482 100644 --- a/include/linux/workqueue.h +++ b/include/linux/workqueue.h @@ -25,18 +25,20 @@ typedef void (*work_func_t)(struct work_struct *work); enum { WORK_STRUCT_PENDING_BIT = 0, /* work item is pending execution */ - WORK_STRUCT_CWQ_BIT = 1, /* data points to cwq */ - WORK_STRUCT_LINKED_BIT = 2, /* next work is linked to this one */ + WORK_STRUCT_DELAYED_BIT = 1, /* work item is delayed */ + WORK_STRUCT_CWQ_BIT = 2, /* data points to cwq */ + WORK_STRUCT_LINKED_BIT = 3, /* next work is linked to this one */ #ifdef CONFIG_DEBUG_OBJECTS_WORK - WORK_STRUCT_STATIC_BIT = 3, /* static initializer (debugobjects) */ - WORK_STRUCT_COLOR_SHIFT = 4, /* color for workqueue flushing */ + WORK_STRUCT_STATIC_BIT = 4, /* static initializer (debugobjects) */ + WORK_STRUCT_COLOR_SHIFT = 5, /* color for workqueue flushing */ #else - WORK_STRUCT_COLOR_SHIFT = 3, /* color for workqueue flushing */ + WORK_STRUCT_COLOR_SHIFT = 4, /* color for workqueue flushing */ #endif WORK_STRUCT_COLOR_BITS = 4, WORK_STRUCT_PENDING = 1 << WORK_STRUCT_PENDING_BIT, + WORK_STRUCT_DELAYED = 1 << WORK_STRUCT_DELAYED_BIT, WORK_STRUCT_CWQ = 1 << WORK_STRUCT_CWQ_BIT, WORK_STRUCT_LINKED = 1 << WORK_STRUCT_LINKED_BIT, #ifdef CONFIG_DEBUG_OBJECTS_WORK @@ -59,8 +61,8 @@ enum { /* * Reserve 7 bits off of cwq pointer w/ debugobjects turned - * off. This makes cwqs aligned to 128 bytes which isn't too - * excessive while allowing 15 workqueue flush colors. + * off. This makes cwqs aligned to 256 bytes and allows 15 + * workqueue flush colors. */ WORK_STRUCT_FLAG_BITS = WORK_STRUCT_COLOR_SHIFT + WORK_STRUCT_COLOR_BITS, diff --git a/kernel/workqueue.c b/kernel/workqueue.c index 362b50d092e2..a2dccfca03ba 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -941,6 +941,7 @@ static void __queue_work(unsigned int cpu, struct workqueue_struct *wq, struct global_cwq *gcwq; struct cpu_workqueue_struct *cwq; struct list_head *worklist; + unsigned int work_flags; unsigned long flags; debug_work_activate(work); @@ -990,14 +991,17 @@ static void __queue_work(unsigned int cpu, struct workqueue_struct *wq, BUG_ON(!list_empty(&work->entry)); cwq->nr_in_flight[cwq->work_color]++; + work_flags = work_color_to_flags(cwq->work_color); if (likely(cwq->nr_active < cwq->max_active)) { cwq->nr_active++; worklist = gcwq_determine_ins_pos(gcwq, cwq); - } else + } else { + work_flags |= WORK_STRUCT_DELAYED; worklist = &cwq->delayed_works; + } - insert_work(cwq, work, worklist, work_color_to_flags(cwq->work_color)); + insert_work(cwq, work, worklist, work_flags); spin_unlock_irqrestore(&gcwq->lock, flags); } @@ -1666,6 +1670,7 @@ static void cwq_activate_first_delayed(struct cpu_workqueue_struct *cwq) struct list_head *pos = gcwq_determine_ins_pos(cwq->gcwq, cwq); move_linked_works(work, pos, NULL); + __clear_bit(WORK_STRUCT_DELAYED_BIT, work_data_bits(work)); cwq->nr_active++; } @@ -1673,6 +1678,7 @@ static void cwq_activate_first_delayed(struct cpu_workqueue_struct *cwq) * cwq_dec_nr_in_flight - decrement cwq's nr_in_flight * @cwq: cwq of interest * @color: color of work which left the queue + * @delayed: for a delayed work * * A work either has completed or is removed from pending queue, * decrement nr_in_flight of its cwq and handle workqueue flushing. @@ -1680,19 +1686,22 @@ static void cwq_activate_first_delayed(struct cpu_workqueue_struct *cwq) * CONTEXT: * spin_lock_irq(gcwq->lock). */ -static void cwq_dec_nr_in_flight(struct cpu_workqueue_struct *cwq, int color) +static void cwq_dec_nr_in_flight(struct cpu_workqueue_struct *cwq, int color, + bool delayed) { /* ignore uncolored works */ if (color == WORK_NO_COLOR) return; cwq->nr_in_flight[color]--; - cwq->nr_active--; - if (!list_empty(&cwq->delayed_works)) { - /* one down, submit a delayed one */ - if (cwq->nr_active < cwq->max_active) - cwq_activate_first_delayed(cwq); + if (!delayed) { + cwq->nr_active--; + if (!list_empty(&cwq->delayed_works)) { + /* one down, submit a delayed one */ + if (cwq->nr_active < cwq->max_active) + cwq_activate_first_delayed(cwq); + } } /* is flush in progress and are we at the flushing tip? */ @@ -1823,7 +1832,7 @@ __acquires(&gcwq->lock) hlist_del_init(&worker->hentry); worker->current_work = NULL; worker->current_cwq = NULL; - cwq_dec_nr_in_flight(cwq, work_color); + cwq_dec_nr_in_flight(cwq, work_color, false); } /** @@ -2388,7 +2397,8 @@ static int try_to_grab_pending(struct work_struct *work) debug_work_deactivate(work); list_del_init(&work->entry); cwq_dec_nr_in_flight(get_work_cwq(work), - get_work_color(work)); + get_work_color(work), + *work_data_bits(work) & WORK_STRUCT_DELAYED); ret = 1; } } -- cgit v1.2.3 From a28dec2f26013aad89446b1f708f948617bc28a2 Mon Sep 17 00:00:00 2001 From: Anton Vorontsov Date: Sun, 8 Aug 2010 18:03:33 +0400 Subject: powerpc/85xx: Add P1021 PCI IDs and quirks This is needed for proper PCI-E support on P1021 SoCs. Signed-off-by: Anton Vorontsov Signed-off-by: Kumar Gala --- arch/powerpc/sysdev/fsl_pci.c | 2 ++ include/linux/pci_ids.h | 2 ++ 2 files changed, 4 insertions(+) (limited to 'include/linux') diff --git a/arch/powerpc/sysdev/fsl_pci.c b/arch/powerpc/sysdev/fsl_pci.c index 209384b6e039..4ae933225251 100644 --- a/arch/powerpc/sysdev/fsl_pci.c +++ b/arch/powerpc/sysdev/fsl_pci.c @@ -399,6 +399,8 @@ DECLARE_PCI_FIXUP_HEADER(0x1957, PCI_DEVICE_ID_P1013E, quirk_fsl_pcie_header); DECLARE_PCI_FIXUP_HEADER(0x1957, PCI_DEVICE_ID_P1013, quirk_fsl_pcie_header); DECLARE_PCI_FIXUP_HEADER(0x1957, PCI_DEVICE_ID_P1020E, quirk_fsl_pcie_header); DECLARE_PCI_FIXUP_HEADER(0x1957, PCI_DEVICE_ID_P1020, quirk_fsl_pcie_header); +DECLARE_PCI_FIXUP_HEADER(0x1957, PCI_DEVICE_ID_P1021E, quirk_fsl_pcie_header); +DECLARE_PCI_FIXUP_HEADER(0x1957, PCI_DEVICE_ID_P1021, quirk_fsl_pcie_header); DECLARE_PCI_FIXUP_HEADER(0x1957, PCI_DEVICE_ID_P1022E, quirk_fsl_pcie_header); DECLARE_PCI_FIXUP_HEADER(0x1957, PCI_DEVICE_ID_P1022, quirk_fsl_pcie_header); DECLARE_PCI_FIXUP_HEADER(0x1957, PCI_DEVICE_ID_P2010E, quirk_fsl_pcie_header); diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h index f6a3b2d36cad..10d33309e9a6 100644 --- a/include/linux/pci_ids.h +++ b/include/linux/pci_ids.h @@ -2300,6 +2300,8 @@ #define PCI_DEVICE_ID_P2010 0x0079 #define PCI_DEVICE_ID_P1020E 0x0100 #define PCI_DEVICE_ID_P1020 0x0101 +#define PCI_DEVICE_ID_P1021E 0x0102 +#define PCI_DEVICE_ID_P1021 0x0103 #define PCI_DEVICE_ID_P1011E 0x0108 #define PCI_DEVICE_ID_P1011 0x0109 #define PCI_DEVICE_ID_P1022E 0x0110 -- cgit v1.2.3 From ef5dc121d5a0bb1fa477c5395277259f07d318a3 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Thu, 2 Sep 2010 15:48:16 -0700 Subject: mutex: Fix annotations to include it in kernel-locking docbook Fix kernel-doc notation in linux/mutex.h and kernel/mutex.c, then add these 2 files to the kernel-locking docbook as the Mutex API reference chapter. Add one API function to mutex-design.txt and correct a typo in that file. Signed-off-by: Randy Dunlap Cc: Rusty Russell LKML-Reference: <20100902154816.6cc2f9ad.randy.dunlap@oracle.com> Signed-off-by: Ingo Molnar --- Documentation/DocBook/kernel-locking.tmpl | 6 ++++++ Documentation/mutex-design.txt | 3 ++- include/linux/mutex.h | 8 ++++++++ kernel/mutex.c | 23 +++++++---------------- 4 files changed, 23 insertions(+), 17 deletions(-) (limited to 'include/linux') diff --git a/Documentation/DocBook/kernel-locking.tmpl b/Documentation/DocBook/kernel-locking.tmpl index 0b1a3f97f285..a0d479d1e1dd 100644 --- a/Documentation/DocBook/kernel-locking.tmpl +++ b/Documentation/DocBook/kernel-locking.tmpl @@ -1961,6 +1961,12 @@ machines due to caching. + + Mutex API reference +!Iinclude/linux/mutex.h +!Ekernel/mutex.c + + Further reading diff --git a/Documentation/mutex-design.txt b/Documentation/mutex-design.txt index c91ccc0720fa..38c10fd7f411 100644 --- a/Documentation/mutex-design.txt +++ b/Documentation/mutex-design.txt @@ -9,7 +9,7 @@ firstly, there's nothing wrong with semaphores. But if the simpler mutex semantics are sufficient for your code, then there are a couple of advantages of mutexes: - - 'struct mutex' is smaller on most architectures: .e.g on x86, + - 'struct mutex' is smaller on most architectures: E.g. on x86, 'struct semaphore' is 20 bytes, 'struct mutex' is 16 bytes. A smaller structure size means less RAM footprint, and better CPU-cache utilization. @@ -136,3 +136,4 @@ the APIs of 'struct mutex' have been streamlined: void mutex_lock_nested(struct mutex *lock, unsigned int subclass); int mutex_lock_interruptible_nested(struct mutex *lock, unsigned int subclass); + int atomic_dec_and_mutex_lock(atomic_t *cnt, struct mutex *lock); diff --git a/include/linux/mutex.h b/include/linux/mutex.h index 878cab4f5fcc..f363bc8fdc74 100644 --- a/include/linux/mutex.h +++ b/include/linux/mutex.h @@ -78,6 +78,14 @@ struct mutex_waiter { # include #else # define __DEBUG_MUTEX_INITIALIZER(lockname) +/** + * mutex_init - initialize the mutex + * @mutex: the mutex to be initialized + * + * Initialize the mutex to unlocked state. + * + * It is not allowed to initialize an already locked mutex. + */ # define mutex_init(mutex) \ do { \ static struct lock_class_key __key; \ diff --git a/kernel/mutex.c b/kernel/mutex.c index 4c0b7b3e6d2e..200407c1502f 100644 --- a/kernel/mutex.c +++ b/kernel/mutex.c @@ -36,15 +36,6 @@ # include #endif -/*** - * mutex_init - initialize the mutex - * @lock: the mutex to be initialized - * @key: the lock_class_key for the class; used by mutex lock debugging - * - * Initialize the mutex to unlocked state. - * - * It is not allowed to initialize an already locked mutex. - */ void __mutex_init(struct mutex *lock, const char *name, struct lock_class_key *key) { @@ -68,7 +59,7 @@ EXPORT_SYMBOL(__mutex_init); static __used noinline void __sched __mutex_lock_slowpath(atomic_t *lock_count); -/*** +/** * mutex_lock - acquire the mutex * @lock: the mutex to be acquired * @@ -105,7 +96,7 @@ EXPORT_SYMBOL(mutex_lock); static __used noinline void __sched __mutex_unlock_slowpath(atomic_t *lock_count); -/*** +/** * mutex_unlock - release the mutex * @lock: the mutex to be released * @@ -364,8 +355,8 @@ __mutex_lock_killable_slowpath(atomic_t *lock_count); static noinline int __sched __mutex_lock_interruptible_slowpath(atomic_t *lock_count); -/*** - * mutex_lock_interruptible - acquire the mutex, interruptable +/** + * mutex_lock_interruptible - acquire the mutex, interruptible * @lock: the mutex to be acquired * * Lock the mutex like mutex_lock(), and return 0 if the mutex has @@ -456,15 +447,15 @@ static inline int __mutex_trylock_slowpath(atomic_t *lock_count) return prev == 1; } -/*** - * mutex_trylock - try acquire the mutex, without waiting +/** + * mutex_trylock - try to acquire the mutex, without waiting * @lock: the mutex to be acquired * * Try to acquire the mutex atomically. Returns 1 if the mutex * has been acquired successfully, and 0 on contention. * * NOTE: this function follows the spin_trylock() convention, so - * it is negated to the down_trylock() return values! Be careful + * it is negated from the down_trylock() return values! Be careful * about this when converting semaphore users to mutexes. * * This function must not be used in interrupt context. The -- cgit v1.2.3 From 71cad0554956de87c3fc413b1eac9313887eb14f Mon Sep 17 00:00:00 2001 From: Philippe Langlais Date: Tue, 31 Aug 2010 14:19:09 +0200 Subject: serial: fix port type conflict between NS16550A & U6_16550A Bug seen by Dr. David Alan Gilbert with sparse Signed-off-by: Philippe Langlais Cc: stable Signed-off-by: Greg Kroah-Hartman --- include/linux/serial.h | 3 +-- include/linux/serial_core.h | 3 ++- 2 files changed, 3 insertions(+), 3 deletions(-) (limited to 'include/linux') diff --git a/include/linux/serial.h b/include/linux/serial.h index 1ebc694a6d52..ef914061511e 100644 --- a/include/linux/serial.h +++ b/include/linux/serial.h @@ -77,8 +77,7 @@ struct serial_struct { #define PORT_16654 11 #define PORT_16850 12 #define PORT_RSA 13 /* RSA-DV II/S card */ -#define PORT_U6_16550A 14 -#define PORT_MAX 14 +#define PORT_MAX 13 #define SERIAL_IO_PORT 0 #define SERIAL_IO_HUB6 1 diff --git a/include/linux/serial_core.h b/include/linux/serial_core.h index 64458a9a8938..563e23400913 100644 --- a/include/linux/serial_core.h +++ b/include/linux/serial_core.h @@ -44,7 +44,8 @@ #define PORT_RM9000 16 /* PMC-Sierra RM9xxx internal UART */ #define PORT_OCTEON 17 /* Cavium OCTEON internal UART */ #define PORT_AR7 18 /* Texas Instruments AR7 internal UART */ -#define PORT_MAX_8250 18 /* max port ID */ +#define PORT_U6_16550A 19 /* ST-Ericsson U6xxx internal UART */ +#define PORT_MAX_8250 19 /* max port ID */ /* * ARM specific type numbers. These are not currently guaranteed -- cgit v1.2.3 From 29bc17ecb856ffb2b47c7009a71971c6f9334205 Mon Sep 17 00:00:00 2001 From: Francisco Jerez Date: Sat, 4 Sep 2010 22:56:44 +0200 Subject: io-mapping: Fix the address space annotations Fixes a bunch of sparse warnings in io-mapping.h because of the inconsistent __iomem usage. Signed-off-by: Francisco Jerez LKML-Reference: <1283633804-11749-2-git-send-email-currojerez@riseup.net> Cc: Andrew Morton Signed-off-by: Ingo Molnar --- include/linux/io-mapping.h | 24 ++++++++++++------------ 1 file changed, 12 insertions(+), 12 deletions(-) (limited to 'include/linux') diff --git a/include/linux/io-mapping.h b/include/linux/io-mapping.h index 0a6b3d5c490c..7fb592793738 100644 --- a/include/linux/io-mapping.h +++ b/include/linux/io-mapping.h @@ -79,7 +79,7 @@ io_mapping_free(struct io_mapping *mapping) } /* Atomic map/unmap */ -static inline void * +static inline void __iomem * io_mapping_map_atomic_wc(struct io_mapping *mapping, unsigned long offset, int slot) @@ -94,12 +94,12 @@ io_mapping_map_atomic_wc(struct io_mapping *mapping, } static inline void -io_mapping_unmap_atomic(void *vaddr, int slot) +io_mapping_unmap_atomic(void __iomem *vaddr, int slot) { iounmap_atomic(vaddr, slot); } -static inline void * +static inline void __iomem * io_mapping_map_wc(struct io_mapping *mapping, unsigned long offset) { resource_size_t phys_addr; @@ -111,7 +111,7 @@ io_mapping_map_wc(struct io_mapping *mapping, unsigned long offset) } static inline void -io_mapping_unmap(void *vaddr) +io_mapping_unmap(void __iomem *vaddr) { iounmap(vaddr); } @@ -125,38 +125,38 @@ struct io_mapping; static inline struct io_mapping * io_mapping_create_wc(resource_size_t base, unsigned long size) { - return (struct io_mapping *) ioremap_wc(base, size); + return (struct io_mapping __force *) ioremap_wc(base, size); } static inline void io_mapping_free(struct io_mapping *mapping) { - iounmap(mapping); + iounmap((void __force __iomem *) mapping); } /* Atomic map/unmap */ -static inline void * +static inline void __iomem * io_mapping_map_atomic_wc(struct io_mapping *mapping, unsigned long offset, int slot) { - return ((char *) mapping) + offset; + return ((char __force __iomem *) mapping) + offset; } static inline void -io_mapping_unmap_atomic(void *vaddr, int slot) +io_mapping_unmap_atomic(void __iomem *vaddr, int slot) { } /* Non-atomic map/unmap */ -static inline void * +static inline void __iomem * io_mapping_map_wc(struct io_mapping *mapping, unsigned long offset) { - return ((char *) mapping) + offset; + return ((char __force __iomem *) mapping) + offset; } static inline void -io_mapping_unmap(void *vaddr) +io_mapping_unmap(void __iomem *vaddr) { } -- cgit v1.2.3 From f8f235e5bbf4e61f3e0886a44afb1dc4cfe8f337 Mon Sep 17 00:00:00 2001 From: Zhenyu Wang Date: Fri, 27 Aug 2010 11:08:57 +0800 Subject: agp/intel: Fix cache control for Sandybridge Sandybridge GTT has new cache control bits in PTE, which controls graphics page cache in LLC or LLC/MLC, so we need to extend the mask function to respect the new bits. And set cache control to always LLC only by default on Gen6. Signed-off-by: Zhenyu Wang Cc: stable@kernel.org Signed-off-by: Chris Wilson --- drivers/char/agp/intel-agp.c | 1 + drivers/char/agp/intel-gtt.c | 50 ++++++++++++++++++++++++++++++++--------- drivers/gpu/drm/i915/i915_gem.c | 1 + include/linux/intel-gtt.h | 20 +++++++++++++++++ 4 files changed, 62 insertions(+), 10 deletions(-) create mode 100644 include/linux/intel-gtt.h (limited to 'include/linux') diff --git a/drivers/char/agp/intel-agp.c b/drivers/char/agp/intel-agp.c index 710af89b176d..74461d177baf 100644 --- a/drivers/char/agp/intel-agp.c +++ b/drivers/char/agp/intel-agp.c @@ -12,6 +12,7 @@ #include #include "agp.h" #include "intel-agp.h" +#include #include "intel-gtt.c" diff --git a/drivers/char/agp/intel-gtt.c b/drivers/char/agp/intel-gtt.c index 7f35854d33a3..64b10551a3f8 100644 --- a/drivers/char/agp/intel-gtt.c +++ b/drivers/char/agp/intel-gtt.c @@ -49,6 +49,26 @@ static struct gatt_mask intel_i810_masks[] = .type = INTEL_AGP_CACHED_MEMORY} }; +#define INTEL_AGP_UNCACHED_MEMORY 0 +#define INTEL_AGP_CACHED_MEMORY_LLC 1 +#define INTEL_AGP_CACHED_MEMORY_LLC_GFDT 2 +#define INTEL_AGP_CACHED_MEMORY_LLC_MLC 3 +#define INTEL_AGP_CACHED_MEMORY_LLC_MLC_GFDT 4 + +static struct gatt_mask intel_gen6_masks[] = +{ + {.mask = I810_PTE_VALID | GEN6_PTE_UNCACHED, + .type = INTEL_AGP_UNCACHED_MEMORY }, + {.mask = I810_PTE_VALID | GEN6_PTE_LLC, + .type = INTEL_AGP_CACHED_MEMORY_LLC }, + {.mask = I810_PTE_VALID | GEN6_PTE_LLC | GEN6_PTE_GFDT, + .type = INTEL_AGP_CACHED_MEMORY_LLC_GFDT }, + {.mask = I810_PTE_VALID | GEN6_PTE_LLC_MLC, + .type = INTEL_AGP_CACHED_MEMORY_LLC_MLC }, + {.mask = I810_PTE_VALID | GEN6_PTE_LLC_MLC | GEN6_PTE_GFDT, + .type = INTEL_AGP_CACHED_MEMORY_LLC_MLC_GFDT }, +}; + static struct _intel_private { struct pci_dev *pcidev; /* device one */ u8 __iomem *registers; @@ -178,13 +198,6 @@ static void intel_agp_insert_sg_entries(struct agp_memory *mem, off_t pg_start, int mask_type) { int i, j; - u32 cache_bits = 0; - - if (agp_bridge->dev->device == PCI_DEVICE_ID_INTEL_SANDYBRIDGE_HB || - agp_bridge->dev->device == PCI_DEVICE_ID_INTEL_SANDYBRIDGE_M_HB) - { - cache_bits = GEN6_PTE_LLC_MLC; - } for (i = 0, j = pg_start; i < mem->page_count; i++, j++) { writel(agp_bridge->driver->mask_memory(agp_bridge, @@ -317,6 +330,23 @@ static int intel_i830_type_to_mask_type(struct agp_bridge_data *bridge, return 0; } +static int intel_gen6_type_to_mask_type(struct agp_bridge_data *bridge, + int type) +{ + unsigned int type_mask = type & ~AGP_USER_CACHED_MEMORY_GFDT; + unsigned int gfdt = type & AGP_USER_CACHED_MEMORY_GFDT; + + if (type_mask == AGP_USER_UNCACHED_MEMORY) + return INTEL_AGP_UNCACHED_MEMORY; + else if (type_mask == AGP_USER_CACHED_MEMORY_LLC_MLC) + return gfdt ? INTEL_AGP_CACHED_MEMORY_LLC_MLC_GFDT : + INTEL_AGP_CACHED_MEMORY_LLC_MLC; + else /* set 'normal'/'cached' to LLC by default */ + return gfdt ? INTEL_AGP_CACHED_MEMORY_LLC_GFDT : + INTEL_AGP_CACHED_MEMORY_LLC; +} + + static int intel_i810_insert_entries(struct agp_memory *mem, off_t pg_start, int type) { @@ -1163,7 +1193,7 @@ static int intel_i915_insert_entries(struct agp_memory *mem, off_t pg_start, mask_type = agp_bridge->driver->agp_type_to_mask_type(agp_bridge, type); - if (mask_type != 0 && mask_type != AGP_PHYS_MEMORY && + if (!IS_SNB && mask_type != 0 && mask_type != AGP_PHYS_MEMORY && mask_type != INTEL_AGP_CACHED_MEMORY) goto out_err; @@ -1563,7 +1593,7 @@ static const struct agp_bridge_driver intel_gen6_driver = { .fetch_size = intel_i9xx_fetch_size, .cleanup = intel_i915_cleanup, .mask_memory = intel_gen6_mask_memory, - .masks = intel_i810_masks, + .masks = intel_gen6_masks, .agp_enable = intel_i810_agp_enable, .cache_flush = global_cache_flush, .create_gatt_table = intel_i965_create_gatt_table, @@ -1576,7 +1606,7 @@ static const struct agp_bridge_driver intel_gen6_driver = { .agp_alloc_pages = agp_generic_alloc_pages, .agp_destroy_page = agp_generic_destroy_page, .agp_destroy_pages = agp_generic_destroy_pages, - .agp_type_to_mask_type = intel_i830_type_to_mask_type, + .agp_type_to_mask_type = intel_gen6_type_to_mask_type, .chipset_flush = intel_i915_chipset_flush, #ifdef USE_PCI_DMA_API .agp_map_page = intel_agp_map_page, diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c index 748c26340c35..16fca1d1799a 100644 --- a/drivers/gpu/drm/i915/i915_gem.c +++ b/drivers/gpu/drm/i915/i915_gem.c @@ -34,6 +34,7 @@ #include #include #include +#include static uint32_t i915_gem_get_gtt_alignment(struct drm_gem_object *obj); static int i915_gem_object_flush_gpu_write_domain(struct drm_gem_object *obj); diff --git a/include/linux/intel-gtt.h b/include/linux/intel-gtt.h new file mode 100644 index 000000000000..1d19ab2afa39 --- /dev/null +++ b/include/linux/intel-gtt.h @@ -0,0 +1,20 @@ +/* + * Common Intel AGPGART and GTT definitions. + */ +#ifndef _INTEL_GTT_H +#define _INTEL_GTT_H + +#include + +/* This is for Intel only GTT controls. + * + * Sandybridge: AGP_USER_CACHED_MEMORY default to LLC only + */ + +#define AGP_USER_CACHED_MEMORY_LLC_MLC (AGP_USER_TYPES + 2) +#define AGP_USER_UNCACHED_MEMORY (AGP_USER_TYPES + 4) + +/* flag for GFDT type */ +#define AGP_USER_CACHED_MEMORY_GFDT (1 << 3) + +#endif -- cgit v1.2.3 From febc88c5948f81114f64c3412011d695aecae233 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 7 Sep 2010 14:46:37 +0200 Subject: semaphore: Add DEFINE_SEMAPHORE The full cleanup of init_MUTEX[_LOCKED] and DECLARE_MUTEX has not been done. Some of the users are real semaphores and we should name them as such instead of confusing everyone with "MUTEX". Provide the infrastructure to get finally rid of init_MUTEX[_LOCKED] and DECLARE_MUTEX. Signed-off-by: Thomas Gleixner Cc: Peter Zijlstra Cc: Christoph Hellwig LKML-Reference: <20100907125054.795929962@linutronix.de> --- include/linux/semaphore.h | 3 +++ 1 file changed, 3 insertions(+) (limited to 'include/linux') diff --git a/include/linux/semaphore.h b/include/linux/semaphore.h index 7415839ac890..5310d27abd2a 100644 --- a/include/linux/semaphore.h +++ b/include/linux/semaphore.h @@ -26,6 +26,9 @@ struct semaphore { .wait_list = LIST_HEAD_INIT((name).wait_list), \ } +#define DEFINE_SEMAPHORE(name) \ + struct semaphore name = __SEMAPHORE_INITIALIZER(name, 1) + #define DECLARE_MUTEX(name) \ struct semaphore name = __SEMAPHORE_INITIALIZER(name, 1) -- cgit v1.2.3 From 39aa3cb3e8250db9188a6f1e3fb62ffa1a717678 Mon Sep 17 00:00:00 2001 From: Stefan Bader Date: Tue, 31 Aug 2010 15:52:27 +0200 Subject: mm: Move vma_stack_continue into mm.h So it can be used by all that need to check for that. Signed-off-by: Stefan Bader Signed-off-by: Linus Torvalds --- fs/proc/task_mmu.c | 3 ++- include/linux/mm.h | 6 ++++++ mm/mlock.c | 6 ------ 3 files changed, 8 insertions(+), 7 deletions(-) (limited to 'include/linux') diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c index 439fc1f1c1c4..271afc48b9a5 100644 --- a/fs/proc/task_mmu.c +++ b/fs/proc/task_mmu.c @@ -224,7 +224,8 @@ static void show_map_vma(struct seq_file *m, struct vm_area_struct *vma) /* We don't show the stack guard page in /proc/maps */ start = vma->vm_start; if (vma->vm_flags & VM_GROWSDOWN) - start += PAGE_SIZE; + if (!vma_stack_continue(vma->vm_prev, vma->vm_start)) + start += PAGE_SIZE; seq_printf(m, "%08lx-%08lx %c%c%c%c %08llx %02x:%02x %lu %n", start, diff --git a/include/linux/mm.h b/include/linux/mm.h index e6b1210772ce..74949fbef8c6 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -864,6 +864,12 @@ int set_page_dirty(struct page *page); int set_page_dirty_lock(struct page *page); int clear_page_dirty_for_io(struct page *page); +/* Is the vma a continuation of the stack vma above it? */ +static inline int vma_stack_continue(struct vm_area_struct *vma, unsigned long addr) +{ + return vma && (vma->vm_end == addr) && (vma->vm_flags & VM_GROWSDOWN); +} + extern unsigned long move_page_tables(struct vm_area_struct *vma, unsigned long old_addr, struct vm_area_struct *new_vma, unsigned long new_addr, unsigned long len); diff --git a/mm/mlock.c b/mm/mlock.c index cbae7c5b9568..b70919ce4f72 100644 --- a/mm/mlock.c +++ b/mm/mlock.c @@ -135,12 +135,6 @@ void munlock_vma_page(struct page *page) } } -/* Is the vma a continuation of the stack vma above it? */ -static inline int vma_stack_continue(struct vm_area_struct *vma, unsigned long addr) -{ - return vma && (vma->vm_end == addr) && (vma->vm_flags & VM_GROWSDOWN); -} - static inline int stack_guard_page(struct vm_area_struct *vma, unsigned long addr) { return (vma->vm_flags & VM_GROWSDOWN) && -- cgit v1.2.3 From a73f8844e1fc54c3762555c1cf1f71774142ca91 Mon Sep 17 00:00:00 2001 From: Jonathan Corbet Date: Wed, 8 Sep 2010 16:54:54 -0600 Subject: lglock: make lg_lock_global() actually lock globally lg_lock_global() currently only acquires spinlocks for online CPUs, but it's meant to lock all possible CPUs. Lglock-protected resources may be associated with removed CPUs - and, indeed, that could happen with the per-superblock open files lists. At Nick's suggestion, change for_each_online_cpu() to for_each_possible_cpu() to protect accesses to those resources. Cc: Al Viro Acked-by: Nick Piggin Signed-off-by: Jonathan Corbet Signed-off-by: Linus Torvalds --- include/linux/lglock.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/lglock.h b/include/linux/lglock.h index b288cb713b90..f549056fb20b 100644 --- a/include/linux/lglock.h +++ b/include/linux/lglock.h @@ -150,7 +150,7 @@ int i; \ preempt_disable(); \ rwlock_acquire(&name##_lock_dep_map, 0, 0, _RET_IP_); \ - for_each_online_cpu(i) { \ + for_each_possible_cpu(i) { \ arch_spinlock_t *lock; \ lock = &per_cpu(name##_lock, i); \ arch_spin_lock(lock); \ @@ -161,7 +161,7 @@ void name##_global_unlock(void) { \ int i; \ rwlock_release(&name##_lock_dep_map, 1, _RET_IP_); \ - for_each_online_cpu(i) { \ + for_each_possible_cpu(i) { \ arch_spinlock_t *lock; \ lock = &per_cpu(name##_lock, i); \ arch_spin_unlock(lock); \ -- cgit v1.2.3 From f3c65b2870f2481f3646bc410a58a12989ecc704 Mon Sep 17 00:00:00 2001 From: David Vrabel Date: Thu, 9 Sep 2010 16:37:24 -0700 Subject: mmc: avoid getting CID on SDIO-only cards The introduction of support for SD combo cards breaks the initialization of all CSR SDIO chips. The GO_IDLE (CMD0) in mmc_sd_get_cid() causes CSR chips to be reset (this is non-standard behavior). When initializing an SDIO card check for a combo card by using the memory present bit in the R4 response to IO_SEND_OP_COND (CMD5). This avoids the call to mmc_sd_get_cid() on an SDIO-only card. Signed-off-by: David Vrabel Acked-by: Michal Mirolaw Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/mmc/core/sdio.c | 5 ++--- include/linux/mmc/sdio.h | 2 ++ 2 files changed, 4 insertions(+), 3 deletions(-) (limited to 'include/linux') diff --git a/drivers/mmc/core/sdio.c b/drivers/mmc/core/sdio.c index bd2755e8d9a3..f332c52968b7 100644 --- a/drivers/mmc/core/sdio.c +++ b/drivers/mmc/core/sdio.c @@ -362,9 +362,8 @@ static int mmc_sdio_init_card(struct mmc_host *host, u32 ocr, goto err; } - err = mmc_sd_get_cid(host, host->ocr & ocr, card->raw_cid); - - if (!err) { + if (ocr & R4_MEMORY_PRESENT + && mmc_sd_get_cid(host, host->ocr & ocr, card->raw_cid) == 0) { card->type = MMC_TYPE_SD_COMBO; if (oldcard && (oldcard->type != MMC_TYPE_SD_COMBO || diff --git a/include/linux/mmc/sdio.h b/include/linux/mmc/sdio.h index 329a8faa6e37..245cdacee544 100644 --- a/include/linux/mmc/sdio.h +++ b/include/linux/mmc/sdio.h @@ -38,6 +38,8 @@ * [8:0] Byte/block count */ +#define R4_MEMORY_PRESENT (1 << 27) + /* SDIO status in R5 Type -- cgit v1.2.3 From e0bf1024b36be90da241af3c2767311e055b612c Mon Sep 17 00:00:00 2001 From: Huang Ying Date: Thu, 9 Sep 2010 16:37:26 -0700 Subject: kfifo: add parenthesis for macro parameter reference Some macro parameter references inside typeof() operator are not enclosed with parenthesis. It should be safer to add them. Signed-off-by: Huang Ying Acked-by: Stefani Seibold Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/kfifo.h | 58 +++++++++++++++++++++++++-------------------------- 1 file changed, 29 insertions(+), 29 deletions(-) (limited to 'include/linux') diff --git a/include/linux/kfifo.h b/include/linux/kfifo.h index 4aa95f203f3e..62dbee554f60 100644 --- a/include/linux/kfifo.h +++ b/include/linux/kfifo.h @@ -214,7 +214,7 @@ __kfifo_must_check_helper(unsigned int val) */ #define kfifo_reset(fifo) \ (void)({ \ - typeof(fifo + 1) __tmp = (fifo); \ + typeof((fifo) + 1) __tmp = (fifo); \ __tmp->kfifo.in = __tmp->kfifo.out = 0; \ }) @@ -228,7 +228,7 @@ __kfifo_must_check_helper(unsigned int val) */ #define kfifo_reset_out(fifo) \ (void)({ \ - typeof(fifo + 1) __tmp = (fifo); \ + typeof((fifo) + 1) __tmp = (fifo); \ __tmp->kfifo.out = __tmp->kfifo.in; \ }) @@ -238,7 +238,7 @@ __kfifo_must_check_helper(unsigned int val) */ #define kfifo_len(fifo) \ ({ \ - typeof(fifo + 1) __tmpl = (fifo); \ + typeof((fifo) + 1) __tmpl = (fifo); \ __tmpl->kfifo.in - __tmpl->kfifo.out; \ }) @@ -248,7 +248,7 @@ __kfifo_must_check_helper(unsigned int val) */ #define kfifo_is_empty(fifo) \ ({ \ - typeof(fifo + 1) __tmpq = (fifo); \ + typeof((fifo) + 1) __tmpq = (fifo); \ __tmpq->kfifo.in == __tmpq->kfifo.out; \ }) @@ -258,7 +258,7 @@ __kfifo_must_check_helper(unsigned int val) */ #define kfifo_is_full(fifo) \ ({ \ - typeof(fifo + 1) __tmpq = (fifo); \ + typeof((fifo) + 1) __tmpq = (fifo); \ kfifo_len(__tmpq) > __tmpq->kfifo.mask; \ }) @@ -269,7 +269,7 @@ __kfifo_must_check_helper(unsigned int val) #define kfifo_avail(fifo) \ __kfifo_must_check_helper( \ ({ \ - typeof(fifo + 1) __tmpq = (fifo); \ + typeof((fifo) + 1) __tmpq = (fifo); \ const size_t __recsize = sizeof(*__tmpq->rectype); \ unsigned int __avail = kfifo_size(__tmpq) - kfifo_len(__tmpq); \ (__recsize) ? ((__avail <= __recsize) ? 0 : \ @@ -284,7 +284,7 @@ __kfifo_must_check_helper( \ */ #define kfifo_skip(fifo) \ (void)({ \ - typeof(fifo + 1) __tmp = (fifo); \ + typeof((fifo) + 1) __tmp = (fifo); \ const size_t __recsize = sizeof(*__tmp->rectype); \ struct __kfifo *__kfifo = &__tmp->kfifo; \ if (__recsize) \ @@ -302,7 +302,7 @@ __kfifo_must_check_helper( \ #define kfifo_peek_len(fifo) \ __kfifo_must_check_helper( \ ({ \ - typeof(fifo + 1) __tmp = (fifo); \ + typeof((fifo) + 1) __tmp = (fifo); \ const size_t __recsize = sizeof(*__tmp->rectype); \ struct __kfifo *__kfifo = &__tmp->kfifo; \ (!__recsize) ? kfifo_len(__tmp) * sizeof(*__tmp->type) : \ @@ -325,7 +325,7 @@ __kfifo_must_check_helper( \ #define kfifo_alloc(fifo, size, gfp_mask) \ __kfifo_must_check_helper( \ ({ \ - typeof(fifo + 1) __tmp = (fifo); \ + typeof((fifo) + 1) __tmp = (fifo); \ struct __kfifo *__kfifo = &__tmp->kfifo; \ __is_kfifo_ptr(__tmp) ? \ __kfifo_alloc(__kfifo, size, sizeof(*__tmp->type), gfp_mask) : \ @@ -339,7 +339,7 @@ __kfifo_must_check_helper( \ */ #define kfifo_free(fifo) \ ({ \ - typeof(fifo + 1) __tmp = (fifo); \ + typeof((fifo) + 1) __tmp = (fifo); \ struct __kfifo *__kfifo = &__tmp->kfifo; \ if (__is_kfifo_ptr(__tmp)) \ __kfifo_free(__kfifo); \ @@ -358,7 +358,7 @@ __kfifo_must_check_helper( \ */ #define kfifo_init(fifo, buffer, size) \ ({ \ - typeof(fifo + 1) __tmp = (fifo); \ + typeof((fifo) + 1) __tmp = (fifo); \ struct __kfifo *__kfifo = &__tmp->kfifo; \ __is_kfifo_ptr(__tmp) ? \ __kfifo_init(__kfifo, buffer, size, sizeof(*__tmp->type)) : \ @@ -379,8 +379,8 @@ __kfifo_must_check_helper( \ */ #define kfifo_put(fifo, val) \ ({ \ - typeof(fifo + 1) __tmp = (fifo); \ - typeof(val + 1) __val = (val); \ + typeof((fifo) + 1) __tmp = (fifo); \ + typeof((val) + 1) __val = (val); \ unsigned int __ret; \ const size_t __recsize = sizeof(*__tmp->rectype); \ struct __kfifo *__kfifo = &__tmp->kfifo; \ @@ -421,8 +421,8 @@ __kfifo_must_check_helper( \ #define kfifo_get(fifo, val) \ __kfifo_must_check_helper( \ ({ \ - typeof(fifo + 1) __tmp = (fifo); \ - typeof(val + 1) __val = (val); \ + typeof((fifo) + 1) __tmp = (fifo); \ + typeof((val) + 1) __val = (val); \ unsigned int __ret; \ const size_t __recsize = sizeof(*__tmp->rectype); \ struct __kfifo *__kfifo = &__tmp->kfifo; \ @@ -462,8 +462,8 @@ __kfifo_must_check_helper( \ #define kfifo_peek(fifo, val) \ __kfifo_must_check_helper( \ ({ \ - typeof(fifo + 1) __tmp = (fifo); \ - typeof(val + 1) __val = (val); \ + typeof((fifo) + 1) __tmp = (fifo); \ + typeof((val) + 1) __val = (val); \ unsigned int __ret; \ const size_t __recsize = sizeof(*__tmp->rectype); \ struct __kfifo *__kfifo = &__tmp->kfifo; \ @@ -501,8 +501,8 @@ __kfifo_must_check_helper( \ */ #define kfifo_in(fifo, buf, n) \ ({ \ - typeof(fifo + 1) __tmp = (fifo); \ - typeof(buf + 1) __buf = (buf); \ + typeof((fifo) + 1) __tmp = (fifo); \ + typeof((buf) + 1) __buf = (buf); \ unsigned long __n = (n); \ const size_t __recsize = sizeof(*__tmp->rectype); \ struct __kfifo *__kfifo = &__tmp->kfifo; \ @@ -554,8 +554,8 @@ __kfifo_must_check_helper( \ #define kfifo_out(fifo, buf, n) \ __kfifo_must_check_helper( \ ({ \ - typeof(fifo + 1) __tmp = (fifo); \ - typeof(buf + 1) __buf = (buf); \ + typeof((fifo) + 1) __tmp = (fifo); \ + typeof((buf) + 1) __buf = (buf); \ unsigned long __n = (n); \ const size_t __recsize = sizeof(*__tmp->rectype); \ struct __kfifo *__kfifo = &__tmp->kfifo; \ @@ -611,7 +611,7 @@ __kfifo_must_check_helper( \ #define kfifo_from_user(fifo, from, len, copied) \ __kfifo_must_check_helper( \ ({ \ - typeof(fifo + 1) __tmp = (fifo); \ + typeof((fifo) + 1) __tmp = (fifo); \ const void __user *__from = (from); \ unsigned int __len = (len); \ unsigned int *__copied = (copied); \ @@ -639,7 +639,7 @@ __kfifo_must_check_helper( \ #define kfifo_to_user(fifo, to, len, copied) \ __kfifo_must_check_helper( \ ({ \ - typeof(fifo + 1) __tmp = (fifo); \ + typeof((fifo) + 1) __tmp = (fifo); \ void __user *__to = (to); \ unsigned int __len = (len); \ unsigned int *__copied = (copied); \ @@ -666,7 +666,7 @@ __kfifo_must_check_helper( \ */ #define kfifo_dma_in_prepare(fifo, sgl, nents, len) \ ({ \ - typeof(fifo + 1) __tmp = (fifo); \ + typeof((fifo) + 1) __tmp = (fifo); \ struct scatterlist *__sgl = (sgl); \ int __nents = (nents); \ unsigned int __len = (len); \ @@ -690,7 +690,7 @@ __kfifo_must_check_helper( \ */ #define kfifo_dma_in_finish(fifo, len) \ (void)({ \ - typeof(fifo + 1) __tmp = (fifo); \ + typeof((fifo) + 1) __tmp = (fifo); \ unsigned int __len = (len); \ const size_t __recsize = sizeof(*__tmp->rectype); \ struct __kfifo *__kfifo = &__tmp->kfifo; \ @@ -717,7 +717,7 @@ __kfifo_must_check_helper( \ */ #define kfifo_dma_out_prepare(fifo, sgl, nents, len) \ ({ \ - typeof(fifo + 1) __tmp = (fifo); \ + typeof((fifo) + 1) __tmp = (fifo); \ struct scatterlist *__sgl = (sgl); \ int __nents = (nents); \ unsigned int __len = (len); \ @@ -741,7 +741,7 @@ __kfifo_must_check_helper( \ */ #define kfifo_dma_out_finish(fifo, len) \ (void)({ \ - typeof(fifo + 1) __tmp = (fifo); \ + typeof((fifo) + 1) __tmp = (fifo); \ unsigned int __len = (len); \ const size_t __recsize = sizeof(*__tmp->rectype); \ struct __kfifo *__kfifo = &__tmp->kfifo; \ @@ -766,8 +766,8 @@ __kfifo_must_check_helper( \ #define kfifo_out_peek(fifo, buf, n) \ __kfifo_must_check_helper( \ ({ \ - typeof(fifo + 1) __tmp = (fifo); \ - typeof(buf + 1) __buf = (buf); \ + typeof((fifo) + 1) __tmp = (fifo); \ + typeof((buf) + 1) __buf = (buf); \ unsigned long __n = (n); \ const size_t __recsize = sizeof(*__tmp->rectype); \ struct __kfifo *__kfifo = &__tmp->kfifo; \ -- cgit v1.2.3 From 31583bb0cf6cc40f2a468a4d2f3b9cbefd24f891 Mon Sep 17 00:00:00 2001 From: "Michael S. Tsirkin" Date: Thu, 9 Sep 2010 16:37:37 -0700 Subject: cgroups: fix API thinko Add cgroup_attach_task_all() The existing cgroup_attach_task_current_cg() API is called by a thread to attach another thread to all of its cgroups; this is unsuitable for cases where a privileged task wants to attach itself to the cgroups of a less privileged one, since the call must be made from the context of the target task. This patch adds a more generic cgroup_attach_task_all() API that allows both the source task and to-be-moved task to be specified. cgroup_attach_task_current_cg() becomes a specialization of the more generic new function. [menage@google.com: rewrote changelog] [akpm@linux-foundation.org: address reviewer comments] Signed-off-by: Michael S. Tsirkin Tested-by: Alex Williamson Acked-by: Paul Menage Cc: Li Zefan Cc: Ben Blum Cc: Sridhar Samudrala Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/cgroup.h | 12 +++++++++++- kernel/cgroup.c | 13 +++++++------ 2 files changed, 18 insertions(+), 7 deletions(-) (limited to 'include/linux') diff --git a/include/linux/cgroup.h b/include/linux/cgroup.h index ed3e92e41c6e..0c991023ee47 100644 --- a/include/linux/cgroup.h +++ b/include/linux/cgroup.h @@ -578,7 +578,12 @@ struct task_struct *cgroup_iter_next(struct cgroup *cgrp, void cgroup_iter_end(struct cgroup *cgrp, struct cgroup_iter *it); int cgroup_scan_tasks(struct cgroup_scanner *scan); int cgroup_attach_task(struct cgroup *, struct task_struct *); -int cgroup_attach_task_current_cg(struct task_struct *); +int cgroup_attach_task_all(struct task_struct *from, struct task_struct *); + +static inline int cgroup_attach_task_current_cg(struct task_struct *tsk) +{ + return cgroup_attach_task_all(current, tsk); +} /* * CSS ID is ID for cgroup_subsys_state structs under subsys. This only works @@ -636,6 +641,11 @@ static inline int cgroupstats_build(struct cgroupstats *stats, } /* No cgroups - nothing to do */ +static inline int cgroup_attach_task_all(struct task_struct *from, + struct task_struct *t) +{ + return 0; +} static inline int cgroup_attach_task_current_cg(struct task_struct *t) { return 0; diff --git a/kernel/cgroup.c b/kernel/cgroup.c index 192f88c5b0f9..c9483d8f6140 100644 --- a/kernel/cgroup.c +++ b/kernel/cgroup.c @@ -1791,19 +1791,20 @@ out: } /** - * cgroup_attach_task_current_cg - attach task 'tsk' to current task's cgroup + * cgroup_attach_task_all - attach task 'tsk' to all cgroups of task 'from' + * @from: attach to all cgroups of a given task * @tsk: the task to be attached */ -int cgroup_attach_task_current_cg(struct task_struct *tsk) +int cgroup_attach_task_all(struct task_struct *from, struct task_struct *tsk) { struct cgroupfs_root *root; - struct cgroup *cur_cg; int retval = 0; cgroup_lock(); for_each_active_root(root) { - cur_cg = task_cgroup_from_root(current, root); - retval = cgroup_attach_task(cur_cg, tsk); + struct cgroup *from_cg = task_cgroup_from_root(from, root); + + retval = cgroup_attach_task(from_cg, tsk); if (retval) break; } @@ -1811,7 +1812,7 @@ int cgroup_attach_task_current_cg(struct task_struct *tsk) return retval; } -EXPORT_SYMBOL_GPL(cgroup_attach_task_current_cg); +EXPORT_SYMBOL_GPL(cgroup_attach_task_all); /* * Attach task with pid 'pid' to cgroup 'cgrp'. Call with cgroup_mutex -- cgit v1.2.3 From 4969c1192d15afa3389e7ae3302096ff684ba655 Mon Sep 17 00:00:00 2001 From: Andrea Arcangeli Date: Thu, 9 Sep 2010 16:37:52 -0700 Subject: mm: fix swapin race condition The pte_same check is reliable only if the swap entry remains pinned (by the page lock on swapcache). We've also to ensure the swapcache isn't removed before we take the lock as try_to_free_swap won't care about the page pin. One of the possible impacts of this patch is that a KSM-shared page can point to the anon_vma of another process, which could exit before the page is freed. This can leave a page with a pointer to a recycled anon_vma object, or worse, a pointer to something that is no longer an anon_vma. [riel@redhat.com: changelog help] Signed-off-by: Andrea Arcangeli Acked-by: Hugh Dickins Reviewed-by: Rik van Riel Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/ksm.h | 20 +++++++++----------- mm/ksm.c | 3 --- mm/memory.c | 39 ++++++++++++++++++++++++++++++++++----- 3 files changed, 43 insertions(+), 19 deletions(-) (limited to 'include/linux') diff --git a/include/linux/ksm.h b/include/linux/ksm.h index 74d691ee9121..3319a6967626 100644 --- a/include/linux/ksm.h +++ b/include/linux/ksm.h @@ -16,6 +16,9 @@ struct stable_node; struct mem_cgroup; +struct page *ksm_does_need_to_copy(struct page *page, + struct vm_area_struct *vma, unsigned long address); + #ifdef CONFIG_KSM int ksm_madvise(struct vm_area_struct *vma, unsigned long start, unsigned long end, int advice, unsigned long *vm_flags); @@ -70,19 +73,14 @@ static inline void set_page_stable_node(struct page *page, * We'd like to make this conditional on vma->vm_flags & VM_MERGEABLE, * but what if the vma was unmerged while the page was swapped out? */ -struct page *ksm_does_need_to_copy(struct page *page, - struct vm_area_struct *vma, unsigned long address); -static inline struct page *ksm_might_need_to_copy(struct page *page, +static inline int ksm_might_need_to_copy(struct page *page, struct vm_area_struct *vma, unsigned long address) { struct anon_vma *anon_vma = page_anon_vma(page); - if (!anon_vma || - (anon_vma->root == vma->anon_vma->root && - page->index == linear_page_index(vma, address))) - return page; - - return ksm_does_need_to_copy(page, vma, address); + return anon_vma && + (anon_vma->root != vma->anon_vma->root || + page->index != linear_page_index(vma, address)); } int page_referenced_ksm(struct page *page, @@ -115,10 +113,10 @@ static inline int ksm_madvise(struct vm_area_struct *vma, unsigned long start, return 0; } -static inline struct page *ksm_might_need_to_copy(struct page *page, +static inline int ksm_might_need_to_copy(struct page *page, struct vm_area_struct *vma, unsigned long address) { - return page; + return 0; } static inline int page_referenced_ksm(struct page *page, diff --git a/mm/ksm.c b/mm/ksm.c index e2ae00458320..b1873cf03ed9 100644 --- a/mm/ksm.c +++ b/mm/ksm.c @@ -1504,8 +1504,6 @@ struct page *ksm_does_need_to_copy(struct page *page, { struct page *new_page; - unlock_page(page); /* any racers will COW it, not modify it */ - new_page = alloc_page_vma(GFP_HIGHUSER_MOVABLE, vma, address); if (new_page) { copy_user_highpage(new_page, page, address, vma); @@ -1521,7 +1519,6 @@ struct page *ksm_does_need_to_copy(struct page *page, add_page_to_unevictable_list(new_page); } - page_cache_release(page); return new_page; } diff --git a/mm/memory.c b/mm/memory.c index 6b2ab1051851..71b161b73bb5 100644 --- a/mm/memory.c +++ b/mm/memory.c @@ -2623,7 +2623,7 @@ static int do_swap_page(struct mm_struct *mm, struct vm_area_struct *vma, unsigned int flags, pte_t orig_pte) { spinlock_t *ptl; - struct page *page; + struct page *page, *swapcache = NULL; swp_entry_t entry; pte_t pte; struct mem_cgroup *ptr = NULL; @@ -2679,10 +2679,23 @@ static int do_swap_page(struct mm_struct *mm, struct vm_area_struct *vma, lock_page(page); delayacct_clear_flag(DELAYACCT_PF_SWAPIN); - page = ksm_might_need_to_copy(page, vma, address); - if (!page) { - ret = VM_FAULT_OOM; - goto out; + /* + * Make sure try_to_free_swap didn't release the swapcache + * from under us. The page pin isn't enough to prevent that. + */ + if (unlikely(!PageSwapCache(page))) + goto out_page; + + if (ksm_might_need_to_copy(page, vma, address)) { + swapcache = page; + page = ksm_does_need_to_copy(page, vma, address); + + if (unlikely(!page)) { + ret = VM_FAULT_OOM; + page = swapcache; + swapcache = NULL; + goto out_page; + } } if (mem_cgroup_try_charge_swapin(mm, page, GFP_KERNEL, &ptr)) { @@ -2735,6 +2748,18 @@ static int do_swap_page(struct mm_struct *mm, struct vm_area_struct *vma, if (vm_swap_full() || (vma->vm_flags & VM_LOCKED) || PageMlocked(page)) try_to_free_swap(page); unlock_page(page); + if (swapcache) { + /* + * Hold the lock to avoid the swap entry to be reused + * until we take the PT lock for the pte_same() check + * (to avoid false positives from pte_same). For + * further safety release the lock after the swap_free + * so that the swap count won't change under a + * parallel locked swapcache. + */ + unlock_page(swapcache); + page_cache_release(swapcache); + } if (flags & FAULT_FLAG_WRITE) { ret |= do_wp_page(mm, vma, address, page_table, pmd, ptl, pte); @@ -2756,6 +2781,10 @@ out_page: unlock_page(page); out_release: page_cache_release(page); + if (swapcache) { + unlock_page(swapcache); + page_cache_release(swapcache); + } return ret; } -- cgit v1.2.3 From 5affb607720d734ca572b8a77c5c7d62d3042b6f Mon Sep 17 00:00:00 2001 From: Gregory Bean Date: Thu, 9 Sep 2010 16:38:02 -0700 Subject: gpio: sx150x: correct and refine reset-on-probe behavior Replace the arbitrary software-reset call from the device-probe method, because: - It is defective. To work correctly, it should be two byte writes, not a single word write. As it stands, it does nothing. - Some devices with sx150x expanders installed have their NRESET pins ganged on the same line, so resetting one causes the others to reset - not a nice thing to do arbitrarily! - The probe, usually taking place at boot, implies a recent hard-reset, so a software reset at this point is just a waste of energy anyway. Therefore, make it optional, defaulting to off, as this will match the common case of probing at powerup and also matches the current broken no-op behavior. Signed-off-by: Gregory Bean Reviewed-by: Jean Delvare Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/gpio/sx150x.c | 26 +++++++++++++++++++++----- include/linux/i2c/sx150x.h | 4 ++++ 2 files changed, 25 insertions(+), 5 deletions(-) (limited to 'include/linux') diff --git a/drivers/gpio/sx150x.c b/drivers/gpio/sx150x.c index b42f42ca70c3..823559ab0e24 100644 --- a/drivers/gpio/sx150x.c +++ b/drivers/gpio/sx150x.c @@ -459,17 +459,33 @@ static int sx150x_init_io(struct sx150x_chip *chip, u8 base, u16 cfg) return err; } -static int sx150x_init_hw(struct sx150x_chip *chip, - struct sx150x_platform_data *pdata) +static int sx150x_reset(struct sx150x_chip *chip) { - int err = 0; + int err; - err = i2c_smbus_write_word_data(chip->client, + err = i2c_smbus_write_byte_data(chip->client, chip->dev_cfg->reg_reset, - 0x3412); + 0x12); if (err < 0) return err; + err = i2c_smbus_write_byte_data(chip->client, + chip->dev_cfg->reg_reset, + 0x34); + return err; +} + +static int sx150x_init_hw(struct sx150x_chip *chip, + struct sx150x_platform_data *pdata) +{ + int err = 0; + + if (pdata->reset_during_probe) { + err = sx150x_reset(chip); + if (err < 0) + return err; + } + err = sx150x_i2c_write(chip->client, chip->dev_cfg->reg_misc, 0x01); diff --git a/include/linux/i2c/sx150x.h b/include/linux/i2c/sx150x.h index ee3049cb9ba5..52baa79d69a7 100644 --- a/include/linux/i2c/sx150x.h +++ b/include/linux/i2c/sx150x.h @@ -63,6 +63,9 @@ * IRQ lines will appear. Similarly to gpio_base, the expander * will create a block of irqs beginning at this number. * This value is ignored if irq_summary is < 0. + * @reset_during_probe: If set to true, the driver will trigger a full + * reset of the chip at the beginning of the probe + * in order to place it in a known state. */ struct sx150x_platform_data { unsigned gpio_base; @@ -73,6 +76,7 @@ struct sx150x_platform_data { u16 io_polarity; int irq_summary; unsigned irq_base; + bool reset_during_probe; }; #endif /* __LINUX_I2C_SX150X_H */ -- cgit v1.2.3 From 910321ea817a202ff70fac666e37e2c8e2f88823 Mon Sep 17 00:00:00 2001 From: Hugh Dickins Date: Thu, 9 Sep 2010 16:38:07 -0700 Subject: swap: revert special hibernation allocation Please revert 2.6.36-rc commit d2997b1042ec150616c1963b5e5e919ffd0b0ebf "hibernation: freeze swap at hibernation". It complicated matters by adding a second swap allocation path, just for hibernation; without in any way fixing the issue that it was intended to address - page reclaim after fixing the hibernation image might free swap from a page already imaged as swapcache, letting its swap be reallocated to store a different page of the image: resulting in data corruption if the imaged page were freed as clean then swapped back in. Pages freed to si->swap_map were still in danger of being reallocated by the alternative allocation path. I guess it inadvertently fixed slow SSD swap allocation for hibernation, as reported by Nigel Cunningham: by missing out the discards that occur on the usual swap allocation path; but that was unintentional, and needs a separate fix. Signed-off-by: Hugh Dickins Cc: KAMEZAWA Hiroyuki Cc: KOSAKI Motohiro Cc: "Rafael J. Wysocki" Cc: Ondrej Zary Cc: Andrea Gelmini Cc: Balbir Singh Cc: Andrea Arcangeli Cc: Nigel Cunningham Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/swap.h | 8 +---- kernel/power/hibernate.c | 1 - kernel/power/snapshot.c | 1 - kernel/power/swap.c | 6 ++-- mm/swapfile.c | 94 ++++++++++++------------------------------------ 5 files changed, 26 insertions(+), 84 deletions(-) (limited to 'include/linux') diff --git a/include/linux/swap.h b/include/linux/swap.h index 2fee51a11b73..bf4eb62506db 100644 --- a/include/linux/swap.h +++ b/include/linux/swap.h @@ -315,6 +315,7 @@ extern long nr_swap_pages; extern long total_swap_pages; extern void si_swapinfo(struct sysinfo *); extern swp_entry_t get_swap_page(void); +extern swp_entry_t get_swap_page_of_type(int); extern int valid_swaphandles(swp_entry_t, unsigned long *); extern int add_swap_count_continuation(swp_entry_t, gfp_t); extern void swap_shmem_alloc(swp_entry_t); @@ -331,13 +332,6 @@ extern int reuse_swap_page(struct page *); extern int try_to_free_swap(struct page *); struct backing_dev_info; -#ifdef CONFIG_HIBERNATION -void hibernation_freeze_swap(void); -void hibernation_thaw_swap(void); -swp_entry_t get_swap_for_hibernation(int type); -void swap_free_for_hibernation(swp_entry_t val); -#endif - /* linux/mm/thrash.c */ extern struct mm_struct *swap_token_mm; extern void grab_swap_token(struct mm_struct *); diff --git a/kernel/power/hibernate.c b/kernel/power/hibernate.c index c77963938bca..8dc31e02ae12 100644 --- a/kernel/power/hibernate.c +++ b/kernel/power/hibernate.c @@ -338,7 +338,6 @@ int hibernation_snapshot(int platform_mode) goto Close; suspend_console(); - hibernation_freeze_swap(); saved_mask = clear_gfp_allowed_mask(GFP_IOFS); error = dpm_suspend_start(PMSG_FREEZE); if (error) diff --git a/kernel/power/snapshot.c b/kernel/power/snapshot.c index 5e7edfb05e66..f6cd6faf84fd 100644 --- a/kernel/power/snapshot.c +++ b/kernel/power/snapshot.c @@ -1086,7 +1086,6 @@ void swsusp_free(void) buffer = NULL; alloc_normal = 0; alloc_highmem = 0; - hibernation_thaw_swap(); } /* Helper functions used for the shrinking of memory. */ diff --git a/kernel/power/swap.c b/kernel/power/swap.c index 5d0059eed3e4..e6a5bdf61a37 100644 --- a/kernel/power/swap.c +++ b/kernel/power/swap.c @@ -136,10 +136,10 @@ sector_t alloc_swapdev_block(int swap) { unsigned long offset; - offset = swp_offset(get_swap_for_hibernation(swap)); + offset = swp_offset(get_swap_page_of_type(swap)); if (offset) { if (swsusp_extents_insert(offset)) - swap_free_for_hibernation(swp_entry(swap, offset)); + swap_free(swp_entry(swap, offset)); else return swapdev_block(swap, offset); } @@ -163,7 +163,7 @@ void free_all_swap_pages(int swap) ext = container_of(node, struct swsusp_extent, node); rb_erase(node, &swsusp_extents); for (offset = ext->start; offset <= ext->end; offset++) - swap_free_for_hibernation(swp_entry(swap, offset)); + swap_free(swp_entry(swap, offset)); kfree(ext); } diff --git a/mm/swapfile.c b/mm/swapfile.c index 1f3f9c59a73a..f08d165871b3 100644 --- a/mm/swapfile.c +++ b/mm/swapfile.c @@ -47,8 +47,6 @@ long nr_swap_pages; long total_swap_pages; static int least_priority; -static bool swap_for_hibernation; - static const char Bad_file[] = "Bad swap file entry "; static const char Unused_file[] = "Unused swap file entry "; static const char Bad_offset[] = "Bad swap offset entry "; @@ -453,8 +451,6 @@ swp_entry_t get_swap_page(void) spin_lock(&swap_lock); if (nr_swap_pages <= 0) goto noswap; - if (swap_for_hibernation) - goto noswap; nr_swap_pages--; for (type = swap_list.next; type >= 0 && wrapped < 2; type = next) { @@ -487,6 +483,28 @@ noswap: return (swp_entry_t) {0}; } +/* The only caller of this function is now susupend routine */ +swp_entry_t get_swap_page_of_type(int type) +{ + struct swap_info_struct *si; + pgoff_t offset; + + spin_lock(&swap_lock); + si = swap_info[type]; + if (si && (si->flags & SWP_WRITEOK)) { + nr_swap_pages--; + /* This is called for allocating swap entry, not cache */ + offset = scan_swap_map(si, 1); + if (offset) { + spin_unlock(&swap_lock); + return swp_entry(type, offset); + } + nr_swap_pages++; + } + spin_unlock(&swap_lock); + return (swp_entry_t) {0}; +} + static struct swap_info_struct *swap_info_get(swp_entry_t entry) { struct swap_info_struct *p; @@ -746,74 +764,6 @@ int mem_cgroup_count_swap_user(swp_entry_t ent, struct page **pagep) #endif #ifdef CONFIG_HIBERNATION - -static pgoff_t hibernation_offset[MAX_SWAPFILES]; -/* - * Once hibernation starts to use swap, we freeze swap_map[]. Otherwise, - * saved swap_map[] image to the disk will be an incomplete because it's - * changing without synchronization with hibernation snap shot. - * At resume, we just make swap_for_hibernation=false. We can forget - * used maps easily. - */ -void hibernation_freeze_swap(void) -{ - int i; - - spin_lock(&swap_lock); - - printk(KERN_INFO "PM: Freeze Swap\n"); - swap_for_hibernation = true; - for (i = 0; i < MAX_SWAPFILES; i++) - hibernation_offset[i] = 1; - spin_unlock(&swap_lock); -} - -void hibernation_thaw_swap(void) -{ - spin_lock(&swap_lock); - if (swap_for_hibernation) { - printk(KERN_INFO "PM: Thaw Swap\n"); - swap_for_hibernation = false; - } - spin_unlock(&swap_lock); -} - -/* - * Because updateing swap_map[] can make not-saved-status-change, - * we use our own easy allocator. - * Please see kernel/power/swap.c, Used swaps are recorded into - * RB-tree. - */ -swp_entry_t get_swap_for_hibernation(int type) -{ - pgoff_t off; - swp_entry_t val = {0}; - struct swap_info_struct *si; - - spin_lock(&swap_lock); - - si = swap_info[type]; - if (!si || !(si->flags & SWP_WRITEOK)) - goto done; - - for (off = hibernation_offset[type]; off < si->max; ++off) { - if (!si->swap_map[off]) - break; - } - if (off < si->max) { - val = swp_entry(type, off); - hibernation_offset[type] = off + 1; - } -done: - spin_unlock(&swap_lock); - return val; -} - -void swap_free_for_hibernation(swp_entry_t ent) -{ - /* Nothing to do */ -} - /* * Find the swap type that corresponds to given device (if any). * -- cgit v1.2.3 From 3399446632739fcd05fd8b272b476a69c6e6d14a Mon Sep 17 00:00:00 2001 From: Hugh Dickins Date: Thu, 9 Sep 2010 16:38:11 -0700 Subject: swap: discard while swapping only if SWAP_FLAG_DISCARD Tests with recent firmware on Intel X25-M 80GB and OCZ Vertex 60GB SSDs show a shift since I last tested in December: in part because of firmware updates, in part because of the necessary move from barriers to awaiting completion at the block layer. While discard at swapon still shows as slightly beneficial on both, discarding 1MB swap cluster when allocating is now disadvanteous: adds 25% overhead on Intel, adds 230% on OCZ (YMMV). Surrender: discard as presently implemented is more hindrance than help for swap; but might prove useful on other devices, or with improvements. So continue to do the discard at swapon, but make discard while swapping conditional on a SWAP_FLAG_DISCARD to sys_swapon() (which has been using only the lower 16 bits of int flags). We can add a --discard or -d to swapon(8), and a "discard" to swap in /etc/fstab: matching the mount option for btrfs, ext4, fat, gfs2, nilfs2. Signed-off-by: Hugh Dickins Cc: Christoph Hellwig Cc: Nigel Cunningham Cc: Tejun Heo Cc: Jens Axboe Cc: James Bottomley Cc: "Martin K. Petersen" Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/swap.h | 3 ++- mm/swapfile.c | 2 +- 2 files changed, 3 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/swap.h b/include/linux/swap.h index bf4eb62506db..7cdd63366f88 100644 --- a/include/linux/swap.h +++ b/include/linux/swap.h @@ -19,6 +19,7 @@ struct bio; #define SWAP_FLAG_PREFER 0x8000 /* set if swap priority specified */ #define SWAP_FLAG_PRIO_MASK 0x7fff #define SWAP_FLAG_PRIO_SHIFT 0 +#define SWAP_FLAG_DISCARD 0x10000 /* discard swap cluster after use */ static inline int current_is_kswapd(void) { @@ -142,7 +143,7 @@ struct swap_extent { enum { SWP_USED = (1 << 0), /* is slot in swap_info[] used? */ SWP_WRITEOK = (1 << 1), /* ok to write to this swap? */ - SWP_DISCARDABLE = (1 << 2), /* blkdev supports discard */ + SWP_DISCARDABLE = (1 << 2), /* swapon+blkdev support discard */ SWP_DISCARDING = (1 << 3), /* now discarding a free cluster */ SWP_SOLIDSTATE = (1 << 4), /* blkdev seeks are cheap */ SWP_CONTINUED = (1 << 5), /* swap_map has count continuation */ diff --git a/mm/swapfile.c b/mm/swapfile.c index 1894dead0b58..7c703ff2f36f 100644 --- a/mm/swapfile.c +++ b/mm/swapfile.c @@ -2047,7 +2047,7 @@ SYSCALL_DEFINE2(swapon, const char __user *, specialfile, int, swap_flags) p->flags |= SWP_SOLIDSTATE; p->cluster_next = 1 + (random32() % p->highest_bit); } - if (discard_swap(p) == 0) + if (discard_swap(p) == 0 && (swap_flags & SWAP_FLAG_DISCARD)) p->flags |= SWP_DISCARDABLE; } -- cgit v1.2.3 From aa45484031ddee09b06350ab8528bfe5b2c76d1c Mon Sep 17 00:00:00 2001 From: Christoph Lameter Date: Thu, 9 Sep 2010 16:38:17 -0700 Subject: mm: page allocator: calculate a better estimate of NR_FREE_PAGES when memory is low and kswapd is awake Ordinarily watermark checks are based on the vmstat NR_FREE_PAGES as it is cheaper than scanning a number of lists. To avoid synchronization overhead, counter deltas are maintained on a per-cpu basis and drained both periodically and when the delta is above a threshold. On large CPU systems, the difference between the estimated and real value of NR_FREE_PAGES can be very high. If NR_FREE_PAGES is much higher than number of real free page in buddy, the VM can allocate pages below min watermark, at worst reducing the real number of pages to zero. Even if the OOM killer kills some victim for freeing memory, it may not free memory if the exit path requires a new page resulting in livelock. This patch introduces a zone_page_state_snapshot() function (courtesy of Christoph) that takes a slightly more accurate view of an arbitrary vmstat counter. It is used to read NR_FREE_PAGES while kswapd is awake to avoid the watermark being accidentally broken. The estimate is not perfect and may result in cache line bounces but is expected to be lighter than the IPI calls necessary to continually drain the per-cpu counters while kswapd is awake. Signed-off-by: Christoph Lameter Signed-off-by: Mel Gorman Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/mmzone.h | 13 +++++++++++++ include/linux/vmstat.h | 22 ++++++++++++++++++++++ mm/mmzone.c | 21 +++++++++++++++++++++ mm/page_alloc.c | 4 ++-- mm/vmstat.c | 15 ++++++++++++++- 5 files changed, 72 insertions(+), 3 deletions(-) (limited to 'include/linux') diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h index 6e6e62648a4d..3984c4eb41fd 100644 --- a/include/linux/mmzone.h +++ b/include/linux/mmzone.h @@ -283,6 +283,13 @@ struct zone { /* zone watermarks, access with *_wmark_pages(zone) macros */ unsigned long watermark[NR_WMARK]; + /* + * When free pages are below this point, additional steps are taken + * when reading the number of free pages to avoid per-cpu counter + * drift allowing watermarks to be breached + */ + unsigned long percpu_drift_mark; + /* * We don't know if the memory that we're going to allocate will be freeable * or/and it will be released eventually, so to avoid totally wasting several @@ -441,6 +448,12 @@ static inline int zone_is_oom_locked(const struct zone *zone) return test_bit(ZONE_OOM_LOCKED, &zone->flags); } +#ifdef CONFIG_SMP +unsigned long zone_nr_free_pages(struct zone *zone); +#else +#define zone_nr_free_pages(zone) zone_page_state(zone, NR_FREE_PAGES) +#endif /* CONFIG_SMP */ + /* * The "priority" of VM scanning is how much of the queues we will scan in one * go. A value of 12 for DEF_PRIORITY implies that we will scan 1/4096th of the diff --git a/include/linux/vmstat.h b/include/linux/vmstat.h index 7f43ccdc1d38..eaaea37b3b75 100644 --- a/include/linux/vmstat.h +++ b/include/linux/vmstat.h @@ -170,6 +170,28 @@ static inline unsigned long zone_page_state(struct zone *zone, return x; } +/* + * More accurate version that also considers the currently pending + * deltas. For that we need to loop over all cpus to find the current + * deltas. There is no synchronization so the result cannot be + * exactly accurate either. + */ +static inline unsigned long zone_page_state_snapshot(struct zone *zone, + enum zone_stat_item item) +{ + long x = atomic_long_read(&zone->vm_stat[item]); + +#ifdef CONFIG_SMP + int cpu; + for_each_online_cpu(cpu) + x += per_cpu_ptr(zone->pageset, cpu)->vm_stat_diff[item]; + + if (x < 0) + x = 0; +#endif + return x; +} + extern unsigned long global_reclaimable_pages(void); extern unsigned long zone_reclaimable_pages(struct zone *zone); diff --git a/mm/mmzone.c b/mm/mmzone.c index f5b7d1760213..e35bfb82c855 100644 --- a/mm/mmzone.c +++ b/mm/mmzone.c @@ -87,3 +87,24 @@ int memmap_valid_within(unsigned long pfn, return 1; } #endif /* CONFIG_ARCH_HAS_HOLES_MEMORYMODEL */ + +#ifdef CONFIG_SMP +/* Called when a more accurate view of NR_FREE_PAGES is needed */ +unsigned long zone_nr_free_pages(struct zone *zone) +{ + unsigned long nr_free_pages = zone_page_state(zone, NR_FREE_PAGES); + + /* + * While kswapd is awake, it is considered the zone is under some + * memory pressure. Under pressure, there is a risk that + * per-cpu-counter-drift will allow the min watermark to be breached + * potentially causing a live-lock. While kswapd is awake and + * free pages are low, get a better estimate for free pages + */ + if (nr_free_pages < zone->percpu_drift_mark && + !waitqueue_active(&zone->zone_pgdat->kswapd_wait)) + return zone_page_state_snapshot(zone, NR_FREE_PAGES); + + return nr_free_pages; +} +#endif /* CONFIG_SMP */ diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 452e2ba06c7c..b2d21e06d45d 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -1462,7 +1462,7 @@ int zone_watermark_ok(struct zone *z, int order, unsigned long mark, { /* free_pages my go negative - that's OK */ long min = mark; - long free_pages = zone_page_state(z, NR_FREE_PAGES) - (1 << order) + 1; + long free_pages = zone_nr_free_pages(z) - (1 << order) + 1; int o; if (alloc_flags & ALLOC_HIGH) @@ -2424,7 +2424,7 @@ void show_free_areas(void) " all_unreclaimable? %s" "\n", zone->name, - K(zone_page_state(zone, NR_FREE_PAGES)), + K(zone_nr_free_pages(zone)), K(min_wmark_pages(zone)), K(low_wmark_pages(zone)), K(high_wmark_pages(zone)), diff --git a/mm/vmstat.c b/mm/vmstat.c index a8d6b59e609a..355a9e669aaa 100644 --- a/mm/vmstat.c +++ b/mm/vmstat.c @@ -138,11 +138,24 @@ static void refresh_zone_stat_thresholds(void) int threshold; for_each_populated_zone(zone) { + unsigned long max_drift, tolerate_drift; + threshold = calculate_threshold(zone); for_each_online_cpu(cpu) per_cpu_ptr(zone->pageset, cpu)->stat_threshold = threshold; + + /* + * Only set percpu_drift_mark if there is a danger that + * NR_FREE_PAGES reports the low watermark is ok when in fact + * the min watermark could be breached by an allocation + */ + tolerate_drift = low_wmark_pages(zone) - min_wmark_pages(zone); + max_drift = num_online_cpus() * threshold; + if (max_drift > tolerate_drift) + zone->percpu_drift_mark = high_wmark_pages(zone) + + max_drift; } } @@ -813,7 +826,7 @@ static void zoneinfo_show_print(struct seq_file *m, pg_data_t *pgdat, "\n scanned %lu" "\n spanned %lu" "\n present %lu", - zone_page_state(zone, NR_FREE_PAGES), + zone_nr_free_pages(zone), min_wmark_pages(zone), low_wmark_pages(zone), high_wmark_pages(zone), -- cgit v1.2.3 From e2f3d75fc0e4a0d03c61872bad39ffa2e74a04ff Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Tue, 7 Sep 2010 14:05:31 +0200 Subject: libata: skip EH autopsy and recovery during suspend For some mysterious reason, certain hardware reacts badly to usual EH actions while the system is going for suspend. As the devices won't be needed until the system is resumed, ask EH to skip usual autopsy and recovery and proceed directly to suspend. Signed-off-by: Tejun Heo Tested-by: Stephan Diestelhorst Cc: stable@kernel.org Signed-off-by: Jeff Garzik --- drivers/ata/libata-core.c | 14 +++++++++++++- drivers/ata/libata-eh.c | 4 ++++ include/linux/libata.h | 1 + 3 files changed, 18 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c index c035b3d041ee..932eaee50245 100644 --- a/drivers/ata/libata-core.c +++ b/drivers/ata/libata-core.c @@ -5418,6 +5418,7 @@ static int ata_host_request_pm(struct ata_host *host, pm_message_t mesg, */ int ata_host_suspend(struct ata_host *host, pm_message_t mesg) { + unsigned int ehi_flags = ATA_EHI_QUIET; int rc; /* @@ -5426,7 +5427,18 @@ int ata_host_suspend(struct ata_host *host, pm_message_t mesg) */ ata_lpm_enable(host); - rc = ata_host_request_pm(host, mesg, 0, ATA_EHI_QUIET, 1); + /* + * On some hardware, device fails to respond after spun down + * for suspend. As the device won't be used before being + * resumed, we don't need to touch the device. Ask EH to skip + * the usual stuff and proceed directly to suspend. + * + * http://thread.gmane.org/gmane.linux.ide/46764 + */ + if (mesg.event == PM_EVENT_SUSPEND) + ehi_flags |= ATA_EHI_NO_AUTOPSY | ATA_EHI_NO_RECOVERY; + + rc = ata_host_request_pm(host, mesg, 0, ehi_flags, 1); if (rc == 0) host->dev->power.power_state = mesg; return rc; diff --git a/drivers/ata/libata-eh.c b/drivers/ata/libata-eh.c index c9ae299b8342..e48302eae55f 100644 --- a/drivers/ata/libata-eh.c +++ b/drivers/ata/libata-eh.c @@ -3235,6 +3235,10 @@ static int ata_eh_skip_recovery(struct ata_link *link) if (link->flags & ATA_LFLAG_DISABLED) return 1; + /* skip if explicitly requested */ + if (ehc->i.flags & ATA_EHI_NO_RECOVERY) + return 1; + /* thaw frozen port and recover failed devices */ if ((ap->pflags & ATA_PFLAG_FROZEN) || ata_link_nr_enabled(link)) return 0; diff --git a/include/linux/libata.h b/include/linux/libata.h index f010f18a0f86..7de282d8bedf 100644 --- a/include/linux/libata.h +++ b/include/linux/libata.h @@ -335,6 +335,7 @@ enum { ATA_EHI_HOTPLUGGED = (1 << 0), /* could have been hotplugged */ ATA_EHI_NO_AUTOPSY = (1 << 2), /* no autopsy */ ATA_EHI_QUIET = (1 << 3), /* be quiet */ + ATA_EHI_NO_RECOVERY = (1 << 4), /* no recovery */ ATA_EHI_DID_SOFTRESET = (1 << 16), /* already soft-reset this port */ ATA_EHI_DID_HARDRESET = (1 << 17), /* already soft-reset this port */ -- cgit v1.2.3 From ea3c64506ea7965f86f030155e6fdef381de10e2 Mon Sep 17 00:00:00 2001 From: Gwendal Grignou Date: Tue, 31 Aug 2010 16:20:36 -0700 Subject: libata-sff: Reenable Port Multiplier after libata-sff remodeling. Keep track of the link on the which the current request is in progress. It allows support of links behind port multiplier. Not all libata-sff is PMP compliant. Code for native BMDMA controller does not take in accound PMP. Tested on Marvell 7042 and Sil7526. Signed-off-by: Gwendal Grignou Signed-off-by: Jeff Garzik --- drivers/ata/libata-sff.c | 38 ++++++++++++++++++++++++++++---------- drivers/ata/sata_mv.c | 2 +- include/linux/libata.h | 3 ++- 3 files changed, 31 insertions(+), 12 deletions(-) (limited to 'include/linux') diff --git a/drivers/ata/libata-sff.c b/drivers/ata/libata-sff.c index dee3c2c52562..e30c537cce32 100644 --- a/drivers/ata/libata-sff.c +++ b/drivers/ata/libata-sff.c @@ -1045,7 +1045,8 @@ static void ata_hsm_qc_complete(struct ata_queued_cmd *qc, int in_wq) int ata_sff_hsm_move(struct ata_port *ap, struct ata_queued_cmd *qc, u8 status, int in_wq) { - struct ata_eh_info *ehi = &ap->link.eh_info; + struct ata_link *link = qc->dev->link; + struct ata_eh_info *ehi = &link->eh_info; unsigned long flags = 0; int poll_next; @@ -1301,8 +1302,14 @@ fsm_start: } EXPORT_SYMBOL_GPL(ata_sff_hsm_move); -void ata_sff_queue_pio_task(struct ata_port *ap, unsigned long delay) +void ata_sff_queue_pio_task(struct ata_link *link, unsigned long delay) { + struct ata_port *ap = link->ap; + + WARN_ON((ap->sff_pio_task_link != NULL) && + (ap->sff_pio_task_link != link)); + ap->sff_pio_task_link = link; + /* may fail if ata_sff_flush_pio_task() in progress */ queue_delayed_work(ata_sff_wq, &ap->sff_pio_task, msecs_to_jiffies(delay)); @@ -1324,14 +1331,18 @@ static void ata_sff_pio_task(struct work_struct *work) { struct ata_port *ap = container_of(work, struct ata_port, sff_pio_task.work); + struct ata_link *link = ap->sff_pio_task_link; struct ata_queued_cmd *qc; u8 status; int poll_next; + BUG_ON(ap->sff_pio_task_link == NULL); /* qc can be NULL if timeout occurred */ - qc = ata_qc_from_tag(ap, ap->link.active_tag); - if (!qc) + qc = ata_qc_from_tag(ap, link->active_tag); + if (!qc) { + ap->sff_pio_task_link = NULL; return; + } fsm_start: WARN_ON_ONCE(ap->hsm_task_state == HSM_ST_IDLE); @@ -1348,11 +1359,16 @@ fsm_start: msleep(2); status = ata_sff_busy_wait(ap, ATA_BUSY, 10); if (status & ATA_BUSY) { - ata_sff_queue_pio_task(ap, ATA_SHORT_PAUSE); + ata_sff_queue_pio_task(link, ATA_SHORT_PAUSE); return; } } + /* + * hsm_move() may trigger another command to be processed. + * clean the link beforehand. + */ + ap->sff_pio_task_link = NULL; /* move the HSM */ poll_next = ata_sff_hsm_move(ap, qc, status, 1); @@ -1379,6 +1395,7 @@ fsm_start: unsigned int ata_sff_qc_issue(struct ata_queued_cmd *qc) { struct ata_port *ap = qc->ap; + struct ata_link *link = qc->dev->link; /* Use polling pio if the LLD doesn't handle * interrupt driven pio and atapi CDB interrupt. @@ -1399,7 +1416,7 @@ unsigned int ata_sff_qc_issue(struct ata_queued_cmd *qc) ap->hsm_task_state = HSM_ST_LAST; if (qc->tf.flags & ATA_TFLAG_POLLING) - ata_sff_queue_pio_task(ap, 0); + ata_sff_queue_pio_task(link, 0); break; @@ -1412,7 +1429,7 @@ unsigned int ata_sff_qc_issue(struct ata_queued_cmd *qc) if (qc->tf.flags & ATA_TFLAG_WRITE) { /* PIO data out protocol */ ap->hsm_task_state = HSM_ST_FIRST; - ata_sff_queue_pio_task(ap, 0); + ata_sff_queue_pio_task(link, 0); /* always send first data block using the * ata_sff_pio_task() codepath. @@ -1422,7 +1439,7 @@ unsigned int ata_sff_qc_issue(struct ata_queued_cmd *qc) ap->hsm_task_state = HSM_ST; if (qc->tf.flags & ATA_TFLAG_POLLING) - ata_sff_queue_pio_task(ap, 0); + ata_sff_queue_pio_task(link, 0); /* if polling, ata_sff_pio_task() handles the * rest. otherwise, interrupt handler takes @@ -1444,7 +1461,7 @@ unsigned int ata_sff_qc_issue(struct ata_queued_cmd *qc) /* send cdb by polling if no cdb interrupt */ if ((!(qc->dev->flags & ATA_DFLAG_CDB_INTR)) || (qc->tf.flags & ATA_TFLAG_POLLING)) - ata_sff_queue_pio_task(ap, 0); + ata_sff_queue_pio_task(link, 0); break; default: @@ -2737,6 +2754,7 @@ EXPORT_SYMBOL_GPL(ata_bmdma_dumb_qc_prep); unsigned int ata_bmdma_qc_issue(struct ata_queued_cmd *qc) { struct ata_port *ap = qc->ap; + struct ata_link *link = qc->dev->link; /* defer PIO handling to sff_qc_issue */ if (!ata_is_dma(qc->tf.protocol)) @@ -2765,7 +2783,7 @@ unsigned int ata_bmdma_qc_issue(struct ata_queued_cmd *qc) /* send cdb by polling if no cdb interrupt */ if (!(qc->dev->flags & ATA_DFLAG_CDB_INTR)) - ata_sff_queue_pio_task(ap, 0); + ata_sff_queue_pio_task(link, 0); break; default: diff --git a/drivers/ata/sata_mv.c b/drivers/ata/sata_mv.c index 81982594a014..a9fd9709c262 100644 --- a/drivers/ata/sata_mv.c +++ b/drivers/ata/sata_mv.c @@ -2284,7 +2284,7 @@ static unsigned int mv_qc_issue_fis(struct ata_queued_cmd *qc) } if (qc->tf.flags & ATA_TFLAG_POLLING) - ata_sff_queue_pio_task(ap, 0); + ata_sff_queue_pio_task(link, 0); return 0; } diff --git a/include/linux/libata.h b/include/linux/libata.h index 7de282d8bedf..45fb2967b66d 100644 --- a/include/linux/libata.h +++ b/include/linux/libata.h @@ -724,6 +724,7 @@ struct ata_port { struct ata_ioports ioaddr; /* ATA cmd/ctl/dma register blocks */ u8 ctl; /* cache of ATA control register */ u8 last_ctl; /* Cache last written value */ + struct ata_link* sff_pio_task_link; /* link currently used */ struct delayed_work sff_pio_task; #ifdef CONFIG_ATA_BMDMA struct ata_bmdma_prd *bmdma_prd; /* BMDMA SG list */ @@ -1595,7 +1596,7 @@ extern void ata_sff_irq_on(struct ata_port *ap); extern void ata_sff_irq_clear(struct ata_port *ap); extern int ata_sff_hsm_move(struct ata_port *ap, struct ata_queued_cmd *qc, u8 status, int in_wq); -extern void ata_sff_queue_pio_task(struct ata_port *ap, unsigned long delay); +extern void ata_sff_queue_pio_task(struct ata_link *link, unsigned long delay); extern unsigned int ata_sff_qc_issue(struct ata_queued_cmd *qc); extern bool ata_sff_qc_fill_rtf(struct ata_queued_cmd *qc); extern unsigned int ata_sff_port_intr(struct ata_port *ap, -- cgit v1.2.3 From 006abe887c5e637d059c44310de6c92f36aded3b Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Sun, 12 Sep 2010 19:55:25 -0400 Subject: SUNRPC: Fix a race in rpc_info_open There is a race between rpc_info_open and rpc_release_client() in that nothing stops a process from opening the file after the clnt->cl_kref goes to zero. Fix this by using atomic_inc_unless_zero()... Reported-by: J. Bruce Fields Signed-off-by: Trond Myklebust Cc: stable@kernel.org --- include/linux/sunrpc/clnt.h | 2 +- net/sunrpc/clnt.c | 26 ++++++++++++-------------- net/sunrpc/rpc_pipe.c | 14 ++++++++------ 3 files changed, 21 insertions(+), 21 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sunrpc/clnt.h b/include/linux/sunrpc/clnt.h index 569dc722a600..85f38a63f098 100644 --- a/include/linux/sunrpc/clnt.h +++ b/include/linux/sunrpc/clnt.h @@ -30,7 +30,7 @@ struct rpc_inode; * The high-level client handle */ struct rpc_clnt { - struct kref cl_kref; /* Number of references */ + atomic_t cl_count; /* Number of references */ struct list_head cl_clients; /* Global list of clients */ struct list_head cl_tasks; /* List of tasks */ spinlock_t cl_lock; /* spinlock */ diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c index 657aac630fc9..3a8f53e7ba07 100644 --- a/net/sunrpc/clnt.c +++ b/net/sunrpc/clnt.c @@ -226,7 +226,7 @@ static struct rpc_clnt * rpc_new_client(const struct rpc_create_args *args, stru goto out_no_principal; } - kref_init(&clnt->cl_kref); + atomic_set(&clnt->cl_count, 1); err = rpc_setup_pipedir(clnt, program->pipe_dir_name); if (err < 0) @@ -390,14 +390,14 @@ rpc_clone_client(struct rpc_clnt *clnt) if (new->cl_principal == NULL) goto out_no_principal; } - kref_init(&new->cl_kref); + atomic_set(&new->cl_count, 1); err = rpc_setup_pipedir(new, clnt->cl_program->pipe_dir_name); if (err != 0) goto out_no_path; if (new->cl_auth) atomic_inc(&new->cl_auth->au_count); xprt_get(clnt->cl_xprt); - kref_get(&clnt->cl_kref); + atomic_inc(&clnt->cl_count); rpc_register_client(new); rpciod_up(); return new; @@ -465,10 +465,8 @@ EXPORT_SYMBOL_GPL(rpc_shutdown_client); * Free an RPC client */ static void -rpc_free_client(struct kref *kref) +rpc_free_client(struct rpc_clnt *clnt) { - struct rpc_clnt *clnt = container_of(kref, struct rpc_clnt, cl_kref); - dprintk("RPC: destroying %s client for %s\n", clnt->cl_protname, clnt->cl_server); if (!IS_ERR(clnt->cl_path.dentry)) { @@ -495,12 +493,10 @@ out_free: * Free an RPC client */ static void -rpc_free_auth(struct kref *kref) +rpc_free_auth(struct rpc_clnt *clnt) { - struct rpc_clnt *clnt = container_of(kref, struct rpc_clnt, cl_kref); - if (clnt->cl_auth == NULL) { - rpc_free_client(kref); + rpc_free_client(clnt); return; } @@ -509,10 +505,11 @@ rpc_free_auth(struct kref *kref) * release remaining GSS contexts. This mechanism ensures * that it can do so safely. */ - kref_init(kref); + atomic_inc(&clnt->cl_count); rpcauth_release(clnt->cl_auth); clnt->cl_auth = NULL; - kref_put(kref, rpc_free_client); + if (atomic_dec_and_test(&clnt->cl_count)) + rpc_free_client(clnt); } /* @@ -525,7 +522,8 @@ rpc_release_client(struct rpc_clnt *clnt) if (list_empty(&clnt->cl_tasks)) wake_up(&destroy_wait); - kref_put(&clnt->cl_kref, rpc_free_auth); + if (atomic_dec_and_test(&clnt->cl_count)) + rpc_free_auth(clnt); } /** @@ -588,7 +586,7 @@ void rpc_task_set_client(struct rpc_task *task, struct rpc_clnt *clnt) if (clnt != NULL) { rpc_task_release_client(task); task->tk_client = clnt; - kref_get(&clnt->cl_kref); + atomic_inc(&clnt->cl_count); if (clnt->cl_softrtry) task->tk_flags |= RPC_TASK_SOFT; /* Add to the client's list of all tasks */ diff --git a/net/sunrpc/rpc_pipe.c b/net/sunrpc/rpc_pipe.c index 41a762f82630..8c8eef2b8f26 100644 --- a/net/sunrpc/rpc_pipe.c +++ b/net/sunrpc/rpc_pipe.c @@ -371,21 +371,23 @@ rpc_show_info(struct seq_file *m, void *v) static int rpc_info_open(struct inode *inode, struct file *file) { - struct rpc_clnt *clnt; + struct rpc_clnt *clnt = NULL; int ret = single_open(file, rpc_show_info, NULL); if (!ret) { struct seq_file *m = file->private_data; - mutex_lock(&inode->i_mutex); - clnt = RPC_I(inode)->private; - if (clnt) { - kref_get(&clnt->cl_kref); + + spin_lock(&file->f_path.dentry->d_lock); + if (!d_unhashed(file->f_path.dentry)) + clnt = RPC_I(inode)->private; + if (clnt != NULL && atomic_inc_not_zero(&clnt->cl_count)) { + spin_unlock(&file->f_path.dentry->d_lock); m->private = clnt; } else { + spin_unlock(&file->f_path.dentry->d_lock); single_release(inode, file); ret = -EINVAL; } - mutex_unlock(&inode->i_mutex); } return ret; } -- cgit v1.2.3