sched_clock: Avoid corrupting hrtimer tree during suspend
PATCH 1/1 sched_clock: Avoid corrupting hrtimer tree during suspend
During suspend we call sched_clock_poll() to update the epoch and accumulated time and reprogram the sched_clock_timer to fire before the next wrap-around time. Unfortunately, sched_clock_poll() doesn't restart the timer, instead it relies on the hrtimer layer to do that and during suspend we aren't calling that function from the hrtimer layer. Instead, we're reprogramming the expires time while the hrtimer is enqueued, which can cause the hrtimer tree to be corrupted. Fix this problem by updating the state via update_sched_clock() and properly restarting the timer via hrtimer_start().
---
I also wonder if we should be restarting the timer during resume instead of suspend given that the resume path modifies the epoch. At that point timers can't run because interrupts are disabled and we don't really care if the timer fires earlier than it's supposed to anyway because it's just there to avoid rollover events, but does it seem better to do it that way? I didn't send that version because this patch is to fix the code intention, but I'm curious if anyone else feels like it should be changed.
kernel/time/sched_clock.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-)
kernel/time/sched_clock.c +3 -1
@@ -191,7 +191,9 @@ void __init sched_clock_postinit(void)static int sched_clock_suspend (void ) { -sched_clock_poll (&sched_clock_timer); +update_sched_clock (); +/* Restart the timer because we forced an update */ +hrtimer_start (&sched_clock_timer, cd.wrap_kt,HRTIMER_MODE_REL ); cd.suspended = true;return 0 ; }
Signature · 2 lines
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum, hosted by The Linux Foundation
Quoted 15 lines · Stephen Boyd
During suspend we call
sched_clock_poll()to update the epoch and accumulated time and reprogram thesched_clock_timerto fire before the next wrap-around time. Unfortunately,sched_clock_poll()doesn't restart the timer, instead it relies on the hrtimer layer to do that and during suspend we aren't calling that function from the hrtimer layer. Instead, we're reprogramming the expires time while the hrtimer is enqueued, which can cause the hrtimer tree to be corrupted. Fix this problem by updating the state viaupdate_sched_clock()and properly restarting the timer viahrtimer_start().Fixes a08ca5d1089d "sched_clock: Use an hrtimer instead of timer"Signed-off-by Stephen Boyd <[email protected]>---
Quoted 8 lines
I also wonder if we should be restarting the timer during resume instead of suspend given that the resume path modifies the epoch. At that point timers can't run because interrupts are disabled and we don't really care if the timer fires earlier than it's supposed to anyway because it's just there to avoid rollover events, but does it seem better to do it that way? I didn't send that version because this patch is to fix the code intention, but I'm curious if anyone else feels like it should be changed.
Yea, starting the timer on suspend seems unintuitive to me.
Is this something you were hoping to get in for 3.17 or is this a urgent 3.16 item?
thanks
-john
Quoted 24 lines · John Stultz
Quoted 15 lines · Stephen Boyd
During suspend we call
sched_clock_poll()to update the epoch and accumulated time and reprogram thesched_clock_timerto fire before the next wrap-around time. Unfortunately,sched_clock_poll()doesn't restart the timer, instead it relies on the hrtimer layer to do that and during suspend we aren't calling that function from the hrtimer layer. Instead, we're reprogramming the expires time while the hrtimer is enqueued, which can cause the hrtimer tree to be corrupted. Fix this problem by updating the state viaupdate_sched_clock()and properly restarting the timer viahrtimer_start().Fixes a08ca5d1089d "sched_clock: Use an hrtimer instead of timer"Signed-off-by Stephen Boyd <[email protected]>---
Quoted 8 lines
I also wonder if we should be restarting the timer during resume instead of suspend given that the resume path modifies the epoch. At that point timers can't run because interrupts are disabled and we don't really care if the timer fires earlier than it's supposed to anyway because it's just there to avoid rollover events, but does it seem better to do it that way? I didn't send that version because this patch is to fix the code intention, but I'm curious if anyone else feels like it should be changed.
Quoted 4 lines
Yea, starting the timer on suspend seems unintuitive to me.
Is this something you were hoping to get in for 3.17 or is this a urgent 3.16 item?
Ok I'll send a follow up patch to cancel during suspend and start during resume, unless you want that to be part of this fix? It's a regression back to v3.13 so I would think it's urgent, although I haven't seen any reports on the mailing list, just reports on some of our android kernels.
Signature · 2 lines
Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum, hosted by The Linux Foundation
Quoted 25 lines · Stephen Boyd
Quoted 24 lines · John Stultz
Quoted 23 lines · Stephen Boyd
During suspend we call
sched_clock_poll()to update the epoch and accumulated time and reprogram thesched_clock_timerto fire before the next wrap-around time. Unfortunately,sched_clock_poll()doesn't restart the timer, instead it relies on the hrtimer layer to do that and during suspend we aren't calling that function from the hrtimer layer. Instead, we're reprogramming the expires time while the hrtimer is enqueued, which can cause the hrtimer tree to be corrupted. Fix this problem by updating the state viaupdate_sched_clock()and properly restarting the timer viahrtimer_start().Fixes a08ca5d1089d "sched_clock: Use an hrtimer instead of timer"Signed-off-by Stephen Boyd <[email protected]>---
I also wonder if we should be restarting the timer during resume instead of suspend given that the resume path modifies the epoch. At that point timers can't run because interrupts are disabled and we don't really care if the timer fires earlier than it's supposed to anyway because it's just there to avoid rollover events, but does it seem better to do it that way? I didn't send that version because this patch is to fix the code intention, but I'm curious if anyone else feels like it should be changed.
Quoted 8 lines
Quoted 4 lines
Yea, starting the timer on suspend seems unintuitive to me.
Is this something you were hoping to get in for 3.17 or is this a urgent 3.16 item?
Ok I'll send a follow up patch to cancel during suspend and start during resume, unless you want that to be part of this fix? It's a regression back to v3.13 so I would think it's urgent, although I haven't seen any reports on the mailing list, just reports on some of our android kernels.
If its a regression (and needs -stable backports) it needs to go in via tip/timers/urgent, and not via the regular merge window.
Whats the additional risk -stable wise for canceling the timer during suspend and starting it back up during resume?
thanks
-john
Quoted 6 lines · John Stultz
If its a regression (and needs -stable backports) it needs to go in via tip/timers/urgent, and not via the regular merge window.
Whats the additional risk -stable wise for canceling the timer during suspend and starting it back up during resume?
I'd say close to zero given that we'd only be making the timer run a little bit later and we have slack in there already. Here's that version.
----8<-----
Subject: [PATCH] sched_clock: Avoid corrupting hrtimer tree during suspend
During suspend we call sched_clock_poll() to update the epoch and accumulated time and reprogram the sched_clock_timer to fire before the next wrap-around time. Unfortunately, sched_clock_poll() doesn't restart the timer, instead it relies on the hrtimer layer to do that and during suspend we aren't calling that function from the hrtimer layer. Instead, we're reprogramming the expires time while the hrtimer is enqueued, which can cause the hrtimer tree to be corrupted. Furthermore, we restart the timer during suspend but we update the epoch during resume which seems counter-intuitive.
Let's fix this by saving the accumulated state and canceling the timer during suspend. On resume we can update the epoch and restart the timer similar to what we would do if we were starting the clock for the first time.
--- kernel/time/sched_clock.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-)
kernel/time/sched_clock.c +3 -1
@@ -191,7 +191,8 @@ void __init sched_clock_postinit(void)static int sched_clock_suspend (void ) { -sched_clock_poll (&sched_clock_timer); +update_sched_clock (); +hrtimer_cancel (&sched_clock_timer); cd.suspended = true;return 0 ; }
@@ -199,6 +200,7 @@ static int sched_clock_suspend(void)static void sched_clock_resume (void ) { cd.epoch_cyc =read_sched_clock (); +hrtimer_start (&sched_clock_timer, cd.wrap_kt,HRTIMER_MODE_REL ); cd.suspended = false; }
Signature · 2 lines
Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum, hosted by The Linux Foundation
Quoted 9 lines · Stephen Boyd
Quoted 6 lines · John Stultz
If its a regression (and needs -stable backports) it needs to go in via tip/timers/urgent, and not via the regular merge window.
Whats the additional risk -stable wise for canceling the timer during suspend and starting it back up during resume?
I'd say close to zero given that we'd only be making the timer run a little bit later and we have slack in there already. Here's that version.
Ok, thanks. I'll try to do a closer review it and get it queued. Is there anyone who might be able to validate this and provide a Tested-by: ?
thanks
-john
Quoted 5 lines · John Stultz
Quoted 4 lines · Stephen Boyd
Quoted 3 lines · John Stultz
If its a regression (and needs -stable backports) it needs to go in via tip/timers/urgent, and not via the regular merge window.
Quoted 8 lines
Quoted 5 lines
Quoted 3 lines
Whats the additional risk -stable wise for canceling the timer during suspend and starting it back up during resume?
I'd say close to zero given that we'd only be making the timer run a little bit later and we have slack in there already. Here's that version.
Ok, thanks. I'll try to do a closer review it and get it queued. Is there anyone who might be able to validate this and provide a Tested-by: ?
Maybe someone from Linaro can give a Tested-by? I basically did this:
# grep -A1 'sched_clock' /proc/timer_list && echo mem > /sys/power/state && grep -A1 'sched_clock' /proc/timer_list
and made sure that the expires time was reset.
Signature · 2 lines
Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum, hosted by The Linux Foundation