linux_dsm_epyc7002/kernel/power/process.c

242 lines
5.5 KiB
C
Raw Normal View History

/*
* drivers/power/process.c - Functions for starting/stopping processes on
* suspend transitions.
*
* Originally from swsusp.
*/
#undef DEBUG
#include <linux/interrupt.h>
#include <linux/oom.h>
#include <linux/suspend.h>
#include <linux/module.h>
#include <linux/sched/debug.h>
#include <linux/sched/task.h>
#include <linux/syscalls.h>
#include <linux/freezer.h>
#include <linux/delay.h>
#include <linux/workqueue.h>
#include <linux/kmod.h>
#include <trace/events/power.h>
/*
* Timeout for stopping processes
*/
unsigned int __read_mostly freeze_timeout_msecs = 20 * MSEC_PER_SEC;
static int try_to_freeze_tasks(bool user_only)
{
struct task_struct *g, *p;
unsigned long end_time;
unsigned int todo;
bool wq_busy = false;
ktime_t start, end, elapsed;
unsigned int elapsed_msecs;
bool wakeup = false;
int sleep_usecs = USEC_PER_MSEC;
start = ktime_get_boottime();
end_time = jiffies + msecs_to_jiffies(freeze_timeout_msecs);
if (!user_only)
freeze_workqueues_begin();
while (true) {
todo = 0;
read_lock(&tasklist_lock);
for_each_process_thread(g, p) {
if (p == current || !freeze_task(p))
continue;
if (!freezer_should_skip(p))
todo++;
}
read_unlock(&tasklist_lock);
if (!user_only) {
wq_busy = freeze_workqueues_busy();
todo += wq_busy;
}
if (!todo || time_after(jiffies, end_time))
break;
if (pm_wakeup_pending()) {
wakeup = true;
break;
}
/*
* We need to retry, but first give the freezing tasks some
* time to enter the refrigerator. Start with an initial
* 1 ms sleep followed by exponential backoff until 8 ms.
*/
usleep_range(sleep_usecs / 2, sleep_usecs);
if (sleep_usecs < 8 * USEC_PER_MSEC)
sleep_usecs *= 2;
}
end = ktime_get_boottime();
elapsed = ktime_sub(end, start);
elapsed_msecs = ktime_to_ms(elapsed);
if (todo) {
pr_cont("\n");
pr_err("Freezing of tasks %s after %d.%03d seconds "
"(%d tasks refusing to freeze, wq_busy=%d):\n",
wakeup ? "aborted" : "failed",
elapsed_msecs / 1000, elapsed_msecs % 1000,
todo - wq_busy, wq_busy);
if (wq_busy)
show_workqueue_state();
if (!wakeup) {
read_lock(&tasklist_lock);
for_each_process_thread(g, p) {
if (p != current && !freezer_should_skip(p)
&& freezing(p) && !frozen(p))
sched_show_task(p);
}
read_unlock(&tasklist_lock);
}
} else {
pr_cont("(elapsed %d.%03d seconds) ", elapsed_msecs / 1000,
elapsed_msecs % 1000);
}
return todo ? -EBUSY : 0;
}
/**
* freeze_processes - Signal user space processes to enter the refrigerator.
* The current thread will not be frozen. The same process that calls
* freeze_processes must later call thaw_processes.
*
* On success, returns 0. On failure, -errno and system is fully thawed.
*/
int freeze_processes(void)
{
int error;
PM / Sleep: Mitigate race between the freezer and request_firmware() There is a race condition between the freezer and request_firmware() such that if request_firmware() is run on one CPU and freeze_processes() is run on another CPU and usermodehelper_disable() called by it succeeds to grab umhelper_sem for writing before usermodehelper_read_trylock() called from request_firmware() acquires it for reading, the request_firmware() will fail and trigger a WARN_ON() complaining that it was called at a wrong time. However, in fact, it wasn't called at a wrong time and freeze_processes() simply happened to be executed simultaneously. To avoid this race, at least in some cases, modify usermodehelper_read_trylock() so that it doesn't fail if the freezing of tasks has just started and hasn't been completed yet. Instead, during the freezing of tasks, it will try to freeze the task that has called it so that it can wait until user space is thawed without triggering the scary warning. For this purpose, change usermodehelper_disabled so that it can take three different values, UMH_ENABLED (0), UMH_FREEZING and UMH_DISABLED. The first one means that usermode helpers are enabled, the last one means "hard disable" (i.e. the system is not ready for usermode helpers to be used) and the second one is reserved for the freezer. Namely, when freeze_processes() is started, it sets usermodehelper_disabled to UMH_FREEZING which tells usermodehelper_read_trylock() that it shouldn't fail just yet and should call try_to_freeze() if woken up and cannot return immediately. This way all freezable tasks that happen to call request_firmware() right before freeze_processes() is started and lose the race for umhelper_sem with it will be frozen and will sleep until thaw_processes() unsets usermodehelper_disabled. [For the non-freezable callers of request_firmware() the race for umhelper_sem against freeze_processes() is unfortunately unavoidable.] Reported-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl> Acked-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Cc: stable@vger.kernel.org
2012-03-29 04:30:28 +07:00
error = __usermodehelper_disable(UMH_FREEZING);
if (error)
return error;
/* Make sure this task doesn't get frozen */
current->flags |= PF_SUSPEND_TASK;
freezer: make freezing() test freeze conditions in effect instead of TIF_FREEZE Using TIF_FREEZE for freezing worked when there was only single freezing condition (the PM one); however, now there is also the cgroup_freezer and single bit flag is getting clumsy. thaw_processes() is already testing whether cgroup freezing in in effect to avoid thawing tasks which were frozen by both PM and cgroup freezers. This is racy (nothing prevents race against cgroup freezing) and fragile. A much simpler way is to test actual freeze conditions from freezing() - ie. directly test whether PM or cgroup freezing is in effect. This patch adds variables to indicate whether and what type of freezing conditions are in effect and reimplements freezing() such that it directly tests whether any of the two freezing conditions is active and the task should freeze. On fast path, freezing() is still very cheap - it only tests system_freezing_cnt. This makes the clumsy dancing aroung TIF_FREEZE unnecessary and freeze/thaw operations more usual - updating state variables for the new state and nudging target tasks so that they notice the new state and comply. As long as the nudging happens after state update, it's race-free. * This allows use of freezing() in freeze_task(). Replace the open coded tests with freezing(). * p != current test is added to warning printing conditions in try_to_freeze_tasks() failure path. This is necessary as freezing() is now true for the task which initiated freezing too. -v2: Oleg pointed out that re-freezing FROZEN cgroup could increment system_freezing_cnt. Fixed. Signed-off-by: Tejun Heo <tj@kernel.org> Acked-by: Paul Menage <paul@paulmenage.org> (for the cgroup portions)
2011-11-22 03:32:25 +07:00
if (!pm_freezing)
atomic_inc(&system_freezing_cnt);
ACPI / PM: Ignore spurious SCI wakeups from suspend-to-idle The ACPI SCI (System Control Interrupt) is set up as a wakeup IRQ during suspend-to-idle transitions and, consequently, any events signaled through it wake up the system from that state. However, on some systems some of the events signaled via the ACPI SCI while suspended to idle should not cause the system to wake up. In fact, quite often they should just be discarded. Arguably, systems should not resume entirely on such events, but in order to decide which events really should cause the system to resume and which are spurious, it is necessary to resume up to the point when ACPI SCIs are actually handled and processed, which is after executing dpm_resume_noirq() in the system resume path. For this reasons, add a loop around freeze_enter() in which the platforms can process events signaled via multiplexed IRQ lines like the ACPI SCI and add suspend-to-idle hooks that can be used for this purpose to struct platform_freeze_ops. In the ACPI case, the ->wake hook is used for checking if the SCI has triggered while suspended and deferring the interrupt-induced system wakeup until the events signaled through it are actually processed sufficiently to decide whether or not the system should resume. In turn, the ->sync hook allows all of the relevant event queues to be flushed so as to prevent events from being missed due to race conditions. In addition to that, some ACPI code processing wakeup events needs to be modified to use the "hard" version of wakeup triggers, so that it will cause a system resume to happen on device-induced wakeup events even if the "soft" mechanism to prevent the system from suspending is not enabled. However, to preserve the existing behavior with respect to suspend-to-RAM, this only is done in the suspend-to-idle case and only if an SCI has occurred while suspended. Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2017-06-13 03:56:34 +07:00
pm_wakeup_clear(true);
pr_info("Freezing user space processes ... ");
freezer: make freezing() test freeze conditions in effect instead of TIF_FREEZE Using TIF_FREEZE for freezing worked when there was only single freezing condition (the PM one); however, now there is also the cgroup_freezer and single bit flag is getting clumsy. thaw_processes() is already testing whether cgroup freezing in in effect to avoid thawing tasks which were frozen by both PM and cgroup freezers. This is racy (nothing prevents race against cgroup freezing) and fragile. A much simpler way is to test actual freeze conditions from freezing() - ie. directly test whether PM or cgroup freezing is in effect. This patch adds variables to indicate whether and what type of freezing conditions are in effect and reimplements freezing() such that it directly tests whether any of the two freezing conditions is active and the task should freeze. On fast path, freezing() is still very cheap - it only tests system_freezing_cnt. This makes the clumsy dancing aroung TIF_FREEZE unnecessary and freeze/thaw operations more usual - updating state variables for the new state and nudging target tasks so that they notice the new state and comply. As long as the nudging happens after state update, it's race-free. * This allows use of freezing() in freeze_task(). Replace the open coded tests with freezing(). * p != current test is added to warning printing conditions in try_to_freeze_tasks() failure path. This is necessary as freezing() is now true for the task which initiated freezing too. -v2: Oleg pointed out that re-freezing FROZEN cgroup could increment system_freezing_cnt. Fixed. Signed-off-by: Tejun Heo <tj@kernel.org> Acked-by: Paul Menage <paul@paulmenage.org> (for the cgroup portions)
2011-11-22 03:32:25 +07:00
pm_freezing = true;
error = try_to_freeze_tasks(true);
if (!error) {
PM / Sleep: Mitigate race between the freezer and request_firmware() There is a race condition between the freezer and request_firmware() such that if request_firmware() is run on one CPU and freeze_processes() is run on another CPU and usermodehelper_disable() called by it succeeds to grab umhelper_sem for writing before usermodehelper_read_trylock() called from request_firmware() acquires it for reading, the request_firmware() will fail and trigger a WARN_ON() complaining that it was called at a wrong time. However, in fact, it wasn't called at a wrong time and freeze_processes() simply happened to be executed simultaneously. To avoid this race, at least in some cases, modify usermodehelper_read_trylock() so that it doesn't fail if the freezing of tasks has just started and hasn't been completed yet. Instead, during the freezing of tasks, it will try to freeze the task that has called it so that it can wait until user space is thawed without triggering the scary warning. For this purpose, change usermodehelper_disabled so that it can take three different values, UMH_ENABLED (0), UMH_FREEZING and UMH_DISABLED. The first one means that usermode helpers are enabled, the last one means "hard disable" (i.e. the system is not ready for usermode helpers to be used) and the second one is reserved for the freezer. Namely, when freeze_processes() is started, it sets usermodehelper_disabled to UMH_FREEZING which tells usermodehelper_read_trylock() that it shouldn't fail just yet and should call try_to_freeze() if woken up and cannot return immediately. This way all freezable tasks that happen to call request_firmware() right before freeze_processes() is started and lose the race for umhelper_sem with it will be frozen and will sleep until thaw_processes() unsets usermodehelper_disabled. [For the non-freezable callers of request_firmware() the race for umhelper_sem against freeze_processes() is unfortunately unavoidable.] Reported-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl> Acked-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Cc: stable@vger.kernel.org
2012-03-29 04:30:28 +07:00
__usermodehelper_set_disable_depth(UMH_DISABLED);
oom, PM: make OOM detection in the freezer path raceless Commit 5695be142e20 ("OOM, PM: OOM killed task shouldn't escape PM suspend") has left a race window when OOM killer manages to note_oom_kill after freeze_processes checks the counter. The race window is quite small and really unlikely and partial solution deemed sufficient at the time of submission. Tejun wasn't happy about this partial solution though and insisted on a full solution. That requires the full OOM and freezer's task freezing exclusion, though. This is done by this patch which introduces oom_sem RW lock and turns oom_killer_disable() into a full OOM barrier. oom_killer_disabled check is moved from the allocation path to the OOM level and we take oom_sem for reading for both the check and the whole OOM invocation. oom_killer_disable() takes oom_sem for writing so it waits for all currently running OOM killer invocations. Then it disable all the further OOMs by setting oom_killer_disabled and checks for any oom victims. Victims are counted via mark_tsk_oom_victim resp. unmark_oom_victim. The last victim wakes up all waiters enqueued by oom_killer_disable(). Therefore this function acts as the full OOM barrier. The page fault path is covered now as well although it was assumed to be safe before. As per Tejun, "We used to have freezing points deep in file system code which may be reacheable from page fault." so it would be better and more robust to not rely on freezing points here. Same applies to the memcg OOM killer. out_of_memory tells the caller whether the OOM was allowed to trigger and the callers are supposed to handle the situation. The page allocation path simply fails the allocation same as before. The page fault path will retry the fault (more on that later) and Sysrq OOM trigger will simply complain to the log. Normally there wouldn't be any unfrozen user tasks after try_to_freeze_tasks so the function will not block. But if there was an OOM killer racing with try_to_freeze_tasks and the OOM victim didn't finish yet then we have to wait for it. This should complete in a finite time, though, because - the victim cannot loop in the page fault handler (it would die on the way out from the exception) - it cannot loop in the page allocator because all the further allocation would fail and __GFP_NOFAIL allocations are not acceptable at this stage - it shouldn't be blocked on any locks held by frozen tasks (try_to_freeze expects lockless context) and kernel threads and work queues are not frozen yet Signed-off-by: Michal Hocko <mhocko@suse.cz> Suggested-by: Tejun Heo <tj@kernel.org> Cc: David Rientjes <rientjes@google.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Cong Wang <xiyou.wangcong@gmail.com> Cc: "Rafael J. Wysocki" <rjw@rjwysocki.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2015-02-12 06:26:24 +07:00
pr_cont("done.");
}
pr_cont("\n");
BUG_ON(in_atomic());
oom, PM: make OOM detection in the freezer path raceless Commit 5695be142e20 ("OOM, PM: OOM killed task shouldn't escape PM suspend") has left a race window when OOM killer manages to note_oom_kill after freeze_processes checks the counter. The race window is quite small and really unlikely and partial solution deemed sufficient at the time of submission. Tejun wasn't happy about this partial solution though and insisted on a full solution. That requires the full OOM and freezer's task freezing exclusion, though. This is done by this patch which introduces oom_sem RW lock and turns oom_killer_disable() into a full OOM barrier. oom_killer_disabled check is moved from the allocation path to the OOM level and we take oom_sem for reading for both the check and the whole OOM invocation. oom_killer_disable() takes oom_sem for writing so it waits for all currently running OOM killer invocations. Then it disable all the further OOMs by setting oom_killer_disabled and checks for any oom victims. Victims are counted via mark_tsk_oom_victim resp. unmark_oom_victim. The last victim wakes up all waiters enqueued by oom_killer_disable(). Therefore this function acts as the full OOM barrier. The page fault path is covered now as well although it was assumed to be safe before. As per Tejun, "We used to have freezing points deep in file system code which may be reacheable from page fault." so it would be better and more robust to not rely on freezing points here. Same applies to the memcg OOM killer. out_of_memory tells the caller whether the OOM was allowed to trigger and the callers are supposed to handle the situation. The page allocation path simply fails the allocation same as before. The page fault path will retry the fault (more on that later) and Sysrq OOM trigger will simply complain to the log. Normally there wouldn't be any unfrozen user tasks after try_to_freeze_tasks so the function will not block. But if there was an OOM killer racing with try_to_freeze_tasks and the OOM victim didn't finish yet then we have to wait for it. This should complete in a finite time, though, because - the victim cannot loop in the page fault handler (it would die on the way out from the exception) - it cannot loop in the page allocator because all the further allocation would fail and __GFP_NOFAIL allocations are not acceptable at this stage - it shouldn't be blocked on any locks held by frozen tasks (try_to_freeze expects lockless context) and kernel threads and work queues are not frozen yet Signed-off-by: Michal Hocko <mhocko@suse.cz> Suggested-by: Tejun Heo <tj@kernel.org> Cc: David Rientjes <rientjes@google.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Cong Wang <xiyou.wangcong@gmail.com> Cc: "Rafael J. Wysocki" <rjw@rjwysocki.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2015-02-12 06:26:24 +07:00
/*
* Now that the whole userspace is frozen we need to disbale
* the OOM killer to disallow any further interference with
oom, suspend: fix oom_killer_disable vs. pm suspend properly Commit 74070542099c ("oom, suspend: fix oom_reaper vs. oom_killer_disable race") has workaround an existing race between oom_killer_disable and oom_reaper by adding another round of try_to_freeze_tasks after the oom killer was disabled. This was the easiest thing to do for a late 4.7 fix. Let's fix it properly now. After "oom: keep mm of the killed task available" we no longer have to call exit_oom_victim from the oom reaper because we have stable mm available and hide the oom_reaped mm by MMF_OOM_SKIP flag. So let's remove exit_oom_victim and the race described in the above commit doesn't exist anymore if. Unfortunately this alone is not sufficient for the oom_killer_disable usecase because now we do not have any reliable way to reach exit_oom_victim (the victim might get stuck on a way to exit for an unbounded amount of time). OOM killer can cope with that by checking mm flags and move on to another victim but we cannot do the same for oom_killer_disable as we would lose the guarantee of no further interference of the victim with the rest of the system. What we can do instead is to cap the maximum time the oom_killer_disable waits for victims. The only current user of this function (pm suspend) already has a concept of timeout for back off so we can reuse the same value there. Let's drop set_freezable for the oom_reaper kthread because it is no longer needed as the reaper doesn't wake or thaw any processes. Link: http://lkml.kernel.org/r/1472119394-11342-7-git-send-email-mhocko@kernel.org Signed-off-by: Michal Hocko <mhocko@suse.com> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Oleg Nesterov <oleg@redhat.com> Cc: David Rientjes <rientjes@google.com> Cc: Vladimir Davydov <vdavydov@parallels.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-10-08 06:59:00 +07:00
* killable tasks. There is no guarantee oom victims will
* ever reach a point they go away we have to wait with a timeout.
oom, PM: make OOM detection in the freezer path raceless Commit 5695be142e20 ("OOM, PM: OOM killed task shouldn't escape PM suspend") has left a race window when OOM killer manages to note_oom_kill after freeze_processes checks the counter. The race window is quite small and really unlikely and partial solution deemed sufficient at the time of submission. Tejun wasn't happy about this partial solution though and insisted on a full solution. That requires the full OOM and freezer's task freezing exclusion, though. This is done by this patch which introduces oom_sem RW lock and turns oom_killer_disable() into a full OOM barrier. oom_killer_disabled check is moved from the allocation path to the OOM level and we take oom_sem for reading for both the check and the whole OOM invocation. oom_killer_disable() takes oom_sem for writing so it waits for all currently running OOM killer invocations. Then it disable all the further OOMs by setting oom_killer_disabled and checks for any oom victims. Victims are counted via mark_tsk_oom_victim resp. unmark_oom_victim. The last victim wakes up all waiters enqueued by oom_killer_disable(). Therefore this function acts as the full OOM barrier. The page fault path is covered now as well although it was assumed to be safe before. As per Tejun, "We used to have freezing points deep in file system code which may be reacheable from page fault." so it would be better and more robust to not rely on freezing points here. Same applies to the memcg OOM killer. out_of_memory tells the caller whether the OOM was allowed to trigger and the callers are supposed to handle the situation. The page allocation path simply fails the allocation same as before. The page fault path will retry the fault (more on that later) and Sysrq OOM trigger will simply complain to the log. Normally there wouldn't be any unfrozen user tasks after try_to_freeze_tasks so the function will not block. But if there was an OOM killer racing with try_to_freeze_tasks and the OOM victim didn't finish yet then we have to wait for it. This should complete in a finite time, though, because - the victim cannot loop in the page fault handler (it would die on the way out from the exception) - it cannot loop in the page allocator because all the further allocation would fail and __GFP_NOFAIL allocations are not acceptable at this stage - it shouldn't be blocked on any locks held by frozen tasks (try_to_freeze expects lockless context) and kernel threads and work queues are not frozen yet Signed-off-by: Michal Hocko <mhocko@suse.cz> Suggested-by: Tejun Heo <tj@kernel.org> Cc: David Rientjes <rientjes@google.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Cong Wang <xiyou.wangcong@gmail.com> Cc: "Rafael J. Wysocki" <rjw@rjwysocki.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2015-02-12 06:26:24 +07:00
*/
oom, suspend: fix oom_killer_disable vs. pm suspend properly Commit 74070542099c ("oom, suspend: fix oom_reaper vs. oom_killer_disable race") has workaround an existing race between oom_killer_disable and oom_reaper by adding another round of try_to_freeze_tasks after the oom killer was disabled. This was the easiest thing to do for a late 4.7 fix. Let's fix it properly now. After "oom: keep mm of the killed task available" we no longer have to call exit_oom_victim from the oom reaper because we have stable mm available and hide the oom_reaped mm by MMF_OOM_SKIP flag. So let's remove exit_oom_victim and the race described in the above commit doesn't exist anymore if. Unfortunately this alone is not sufficient for the oom_killer_disable usecase because now we do not have any reliable way to reach exit_oom_victim (the victim might get stuck on a way to exit for an unbounded amount of time). OOM killer can cope with that by checking mm flags and move on to another victim but we cannot do the same for oom_killer_disable as we would lose the guarantee of no further interference of the victim with the rest of the system. What we can do instead is to cap the maximum time the oom_killer_disable waits for victims. The only current user of this function (pm suspend) already has a concept of timeout for back off so we can reuse the same value there. Let's drop set_freezable for the oom_reaper kthread because it is no longer needed as the reaper doesn't wake or thaw any processes. Link: http://lkml.kernel.org/r/1472119394-11342-7-git-send-email-mhocko@kernel.org Signed-off-by: Michal Hocko <mhocko@suse.com> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Oleg Nesterov <oleg@redhat.com> Cc: David Rientjes <rientjes@google.com> Cc: Vladimir Davydov <vdavydov@parallels.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-10-08 06:59:00 +07:00
if (!error && !oom_killer_disable(msecs_to_jiffies(freeze_timeout_msecs)))
oom, PM: make OOM detection in the freezer path raceless Commit 5695be142e20 ("OOM, PM: OOM killed task shouldn't escape PM suspend") has left a race window when OOM killer manages to note_oom_kill after freeze_processes checks the counter. The race window is quite small and really unlikely and partial solution deemed sufficient at the time of submission. Tejun wasn't happy about this partial solution though and insisted on a full solution. That requires the full OOM and freezer's task freezing exclusion, though. This is done by this patch which introduces oom_sem RW lock and turns oom_killer_disable() into a full OOM barrier. oom_killer_disabled check is moved from the allocation path to the OOM level and we take oom_sem for reading for both the check and the whole OOM invocation. oom_killer_disable() takes oom_sem for writing so it waits for all currently running OOM killer invocations. Then it disable all the further OOMs by setting oom_killer_disabled and checks for any oom victims. Victims are counted via mark_tsk_oom_victim resp. unmark_oom_victim. The last victim wakes up all waiters enqueued by oom_killer_disable(). Therefore this function acts as the full OOM barrier. The page fault path is covered now as well although it was assumed to be safe before. As per Tejun, "We used to have freezing points deep in file system code which may be reacheable from page fault." so it would be better and more robust to not rely on freezing points here. Same applies to the memcg OOM killer. out_of_memory tells the caller whether the OOM was allowed to trigger and the callers are supposed to handle the situation. The page allocation path simply fails the allocation same as before. The page fault path will retry the fault (more on that later) and Sysrq OOM trigger will simply complain to the log. Normally there wouldn't be any unfrozen user tasks after try_to_freeze_tasks so the function will not block. But if there was an OOM killer racing with try_to_freeze_tasks and the OOM victim didn't finish yet then we have to wait for it. This should complete in a finite time, though, because - the victim cannot loop in the page fault handler (it would die on the way out from the exception) - it cannot loop in the page allocator because all the further allocation would fail and __GFP_NOFAIL allocations are not acceptable at this stage - it shouldn't be blocked on any locks held by frozen tasks (try_to_freeze expects lockless context) and kernel threads and work queues are not frozen yet Signed-off-by: Michal Hocko <mhocko@suse.cz> Suggested-by: Tejun Heo <tj@kernel.org> Cc: David Rientjes <rientjes@google.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Cong Wang <xiyou.wangcong@gmail.com> Cc: "Rafael J. Wysocki" <rjw@rjwysocki.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2015-02-12 06:26:24 +07:00
error = -EBUSY;
if (error)
thaw_processes();
return error;
}
/**
* freeze_kernel_threads - Make freezable kernel threads go to the refrigerator.
*
* On success, returns 0. On failure, -errno and only the kernel threads are
* thawed, so as to give a chance to the caller to do additional cleanups
* (if any) before thawing the userspace tasks. So, it is the responsibility
* of the caller to thaw the userspace tasks, when the time is right.
*/
int freeze_kernel_threads(void)
{
int error;
pr_info("Freezing remaining freezable tasks ... ");
freezer: make freezing() test freeze conditions in effect instead of TIF_FREEZE Using TIF_FREEZE for freezing worked when there was only single freezing condition (the PM one); however, now there is also the cgroup_freezer and single bit flag is getting clumsy. thaw_processes() is already testing whether cgroup freezing in in effect to avoid thawing tasks which were frozen by both PM and cgroup freezers. This is racy (nothing prevents race against cgroup freezing) and fragile. A much simpler way is to test actual freeze conditions from freezing() - ie. directly test whether PM or cgroup freezing is in effect. This patch adds variables to indicate whether and what type of freezing conditions are in effect and reimplements freezing() such that it directly tests whether any of the two freezing conditions is active and the task should freeze. On fast path, freezing() is still very cheap - it only tests system_freezing_cnt. This makes the clumsy dancing aroung TIF_FREEZE unnecessary and freeze/thaw operations more usual - updating state variables for the new state and nudging target tasks so that they notice the new state and comply. As long as the nudging happens after state update, it's race-free. * This allows use of freezing() in freeze_task(). Replace the open coded tests with freezing(). * p != current test is added to warning printing conditions in try_to_freeze_tasks() failure path. This is necessary as freezing() is now true for the task which initiated freezing too. -v2: Oleg pointed out that re-freezing FROZEN cgroup could increment system_freezing_cnt. Fixed. Signed-off-by: Tejun Heo <tj@kernel.org> Acked-by: Paul Menage <paul@paulmenage.org> (for the cgroup portions)
2011-11-22 03:32:25 +07:00
pm_nosig_freezing = true;
error = try_to_freeze_tasks(false);
if (!error)
pr_cont("done.");
mm, PM/Freezer: Disable OOM killer when tasks are frozen Currently, the following scenario appears to be possible in theory: * Tasks are frozen for hibernation or suspend. * Free pages are almost exhausted. * Certain piece of code in the suspend code path attempts to allocate some memory using GFP_KERNEL and allocation order less than or equal to PAGE_ALLOC_COSTLY_ORDER. * __alloc_pages_internal() cannot find a free page so it invokes the OOM killer. * The OOM killer attempts to kill a task, but the task is frozen, so it doesn't die immediately. * __alloc_pages_internal() jumps to 'restart', unsuccessfully tries to find a free page and invokes the OOM killer. * No progress can be made. Although it is now hard to trigger during hibernation due to the memory shrinking carried out by the hibernation code, it is theoretically possible to trigger during suspend after the memory shrinking has been removed from that code path. Moreover, since memory allocations are going to be used for the hibernation memory shrinking, it will be even more likely to happen during hibernation. To prevent it from happening, introduce the oom_killer_disabled switch that will cause __alloc_pages_internal() to fail in the situations in which the OOM killer would have been called and make the freezer set this switch after tasks have been successfully frozen. [akpm@linux-foundation.org: be nicer to the namespace] Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl> Cc: Fengguang Wu <fengguang.wu@gmail.com> Cc: David Rientjes <rientjes@google.com> Acked-by: Pavel Machek <pavel@ucw.cz> Cc: Mel Gorman <mel@csn.ul.ie> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-06-17 05:32:41 +07:00
pr_cont("\n");
BUG_ON(in_atomic());
mm, PM/Freezer: Disable OOM killer when tasks are frozen Currently, the following scenario appears to be possible in theory: * Tasks are frozen for hibernation or suspend. * Free pages are almost exhausted. * Certain piece of code in the suspend code path attempts to allocate some memory using GFP_KERNEL and allocation order less than or equal to PAGE_ALLOC_COSTLY_ORDER. * __alloc_pages_internal() cannot find a free page so it invokes the OOM killer. * The OOM killer attempts to kill a task, but the task is frozen, so it doesn't die immediately. * __alloc_pages_internal() jumps to 'restart', unsuccessfully tries to find a free page and invokes the OOM killer. * No progress can be made. Although it is now hard to trigger during hibernation due to the memory shrinking carried out by the hibernation code, it is theoretically possible to trigger during suspend after the memory shrinking has been removed from that code path. Moreover, since memory allocations are going to be used for the hibernation memory shrinking, it will be even more likely to happen during hibernation. To prevent it from happening, introduce the oom_killer_disabled switch that will cause __alloc_pages_internal() to fail in the situations in which the OOM killer would have been called and make the freezer set this switch after tasks have been successfully frozen. [akpm@linux-foundation.org: be nicer to the namespace] Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl> Cc: Fengguang Wu <fengguang.wu@gmail.com> Cc: David Rientjes <rientjes@google.com> Acked-by: Pavel Machek <pavel@ucw.cz> Cc: Mel Gorman <mel@csn.ul.ie> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-06-17 05:32:41 +07:00
if (error)
thaw_kernel_threads();
return error;
}
void thaw_processes(void)
{
struct task_struct *g, *p;
struct task_struct *curr = current;
trace_suspend_resume(TPS("thaw_processes"), 0, true);
freezer: make freezing() test freeze conditions in effect instead of TIF_FREEZE Using TIF_FREEZE for freezing worked when there was only single freezing condition (the PM one); however, now there is also the cgroup_freezer and single bit flag is getting clumsy. thaw_processes() is already testing whether cgroup freezing in in effect to avoid thawing tasks which were frozen by both PM and cgroup freezers. This is racy (nothing prevents race against cgroup freezing) and fragile. A much simpler way is to test actual freeze conditions from freezing() - ie. directly test whether PM or cgroup freezing is in effect. This patch adds variables to indicate whether and what type of freezing conditions are in effect and reimplements freezing() such that it directly tests whether any of the two freezing conditions is active and the task should freeze. On fast path, freezing() is still very cheap - it only tests system_freezing_cnt. This makes the clumsy dancing aroung TIF_FREEZE unnecessary and freeze/thaw operations more usual - updating state variables for the new state and nudging target tasks so that they notice the new state and comply. As long as the nudging happens after state update, it's race-free. * This allows use of freezing() in freeze_task(). Replace the open coded tests with freezing(). * p != current test is added to warning printing conditions in try_to_freeze_tasks() failure path. This is necessary as freezing() is now true for the task which initiated freezing too. -v2: Oleg pointed out that re-freezing FROZEN cgroup could increment system_freezing_cnt. Fixed. Signed-off-by: Tejun Heo <tj@kernel.org> Acked-by: Paul Menage <paul@paulmenage.org> (for the cgroup portions)
2011-11-22 03:32:25 +07:00
if (pm_freezing)
atomic_dec(&system_freezing_cnt);
pm_freezing = false;
pm_nosig_freezing = false;
oom_killer_enable();
pr_info("Restarting tasks ... ");
PM / sleep: Fix request_firmware() error at resume The commit [247bc037: PM / Sleep: Mitigate race between the freezer and request_firmware()] introduced the finer state control, but it also leads to a new bug; for example, a bug report regarding the firmware loading of intel BT device at suspend/resume: https://bugzilla.novell.com/show_bug.cgi?id=873790 The root cause seems to be a small window between the process resume and the clear of usermodehelper lock. The request_firmware() function checks the UMH lock and gives up when it's in UMH_DISABLE state. This is for avoiding the invalid f/w loading during suspend/resume phase. The problem is, however, that usermodehelper_enable() is called at the end of thaw_processes(). Thus, a thawed process in between can kick off the f/w loader code path (in this case, via btusb_setup_intel()) even before the call of usermodehelper_enable(). Then usermodehelper_read_trylock() returns an error and request_firmware() spews WARN_ON() in the end. This oneliner patch fixes the issue just by setting to UMH_FREEZING state again before restarting tasks, so that the call of request_firmware() will be blocked until the end of this function instead of returning an error. Fixes: 247bc0374254 (PM / Sleep: Mitigate race between the freezer and request_firmware()) Link: https://bugzilla.novell.com/show_bug.cgi?id=873790 Cc: 3.4+ <stable@vger.kernel.org> # 3.4+ Signed-off-by: Takashi Iwai <tiwai@suse.de> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-07-15 13:51:27 +07:00
__usermodehelper_set_disable_depth(UMH_FREEZING);
thaw_workqueues();
read_lock(&tasklist_lock);
for_each_process_thread(g, p) {
/* No other threads should have PF_SUSPEND_TASK set */
WARN_ON((p != curr) && (p->flags & PF_SUSPEND_TASK));
__thaw_task(p);
}
read_unlock(&tasklist_lock);
mm, PM/Freezer: Disable OOM killer when tasks are frozen Currently, the following scenario appears to be possible in theory: * Tasks are frozen for hibernation or suspend. * Free pages are almost exhausted. * Certain piece of code in the suspend code path attempts to allocate some memory using GFP_KERNEL and allocation order less than or equal to PAGE_ALLOC_COSTLY_ORDER. * __alloc_pages_internal() cannot find a free page so it invokes the OOM killer. * The OOM killer attempts to kill a task, but the task is frozen, so it doesn't die immediately. * __alloc_pages_internal() jumps to 'restart', unsuccessfully tries to find a free page and invokes the OOM killer. * No progress can be made. Although it is now hard to trigger during hibernation due to the memory shrinking carried out by the hibernation code, it is theoretically possible to trigger during suspend after the memory shrinking has been removed from that code path. Moreover, since memory allocations are going to be used for the hibernation memory shrinking, it will be even more likely to happen during hibernation. To prevent it from happening, introduce the oom_killer_disabled switch that will cause __alloc_pages_internal() to fail in the situations in which the OOM killer would have been called and make the freezer set this switch after tasks have been successfully frozen. [akpm@linux-foundation.org: be nicer to the namespace] Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl> Cc: Fengguang Wu <fengguang.wu@gmail.com> Cc: David Rientjes <rientjes@google.com> Acked-by: Pavel Machek <pavel@ucw.cz> Cc: Mel Gorman <mel@csn.ul.ie> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-06-17 05:32:41 +07:00
WARN_ON(!(curr->flags & PF_SUSPEND_TASK));
curr->flags &= ~PF_SUSPEND_TASK;
usermodehelper_enable();
schedule();
pr_cont("done.\n");
trace_suspend_resume(TPS("thaw_processes"), 0, false);
}
void thaw_kernel_threads(void)
{
struct task_struct *g, *p;
pm_nosig_freezing = false;
pr_info("Restarting kernel threads ... ");
thaw_workqueues();
read_lock(&tasklist_lock);
for_each_process_thread(g, p) {
if (p->flags & (PF_KTHREAD | PF_WQ_WORKER))
__thaw_task(p);
}
read_unlock(&tasklist_lock);
schedule();
pr_cont("done.\n");
}