Le Friday 01 Feb 2019 à 16:28:54 (+0100), Vincent Guittot a écrit : > On Fri, 1 Feb 2019 at 16:02, Biju Das <biju.das@xxxxxxxxxxxxxx> wrote: > > > > Hi Vincent, > > > > I have rebased my kernel to "next-20190201". Still I am seeing dead lock. > > > > Am I missing any patch? > > No you don't miss anything. > I think that it's the opposite. > > Modification in time accounting in PM runtime has been queued but it > has not moved (yet) to ktime_get_mono_fast_ns() > > Can you try to revert c669560be6c8 ("PM-runtime: Replace jiffies-based > accounting with ktime-based accounting") ? Or instead you can apply : --- drivers/base/power/runtime.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/base/power/runtime.c b/drivers/base/power/runtime.c index 4eaf166..1c40e2a 100644 --- a/drivers/base/power/runtime.c +++ b/drivers/base/power/runtime.c @@ -66,7 +66,7 @@ static int rpm_suspend(struct device *dev, int rpmflags); */ void update_pm_runtime_accounting(struct device *dev) { - u64 now = ktime_to_ns(ktime_get()); + u64 now = ktime_get_mono_fast_ns(); u64 delta; delta = now - dev->power.accounting_timestamp; @@ -1315,7 +1315,7 @@ void pm_runtime_enable(struct device *dev) /* About to enable runtime pm, set accounting_timestamp to now */ if (!dev->power.disable_depth) - dev->power.accounting_timestamp = ktime_to_ns(ktime_get()); + dev->power.accounting_timestamp = ktime_get_mono_fast_ns(); } else { dev_warn(dev, "Unbalanced %s!\n", __func__); } -- 2.7.4 > > > > > > root@ek874:/# echo e61e0000.timer > /sys/devices/system/clocksource/clocksource0/current_clocksource > > [ 193.869423] > > [ 193.870963] ============================================ > > [ 193.876292] WARNING: possible recursive locking detected > > [ 193.881625] 5.0.0-rc4-next-20190201-00007-g731346f #3 Not tainted > > [ 193.887737] -------------------------------------------- > > [ 193.893066] migration/0/11 is trying to acquire lock: > > [ 193.898136] (____ptrval____) (tk_core.seq){----}, at: update_pm_runtime_accounting+0x14/0x68 > > [ 193.906632] > > [ 193.906632] but task is already holding lock: > > [ 193.912483] (____ptrval____) (tk_core.seq){----}, at: multi_cpu_stop+0x8c/0x140 > > [ 193.919828] > > [ 193.919828] other info that might help us debug this: > > [ 193.926377] Possible unsafe locking scenario: > > [ 193.926377] > > [ 193.932314] CPU0 > > [ 193.934765] ---- > > [ 193.937216] lock(tk_core.seq); > > [ 193.940453] lock(tk_core.seq); > > [ 193.943691] > > [ 193.943691] *** DEADLOCK *** > > [ 193.943691] > > [ 193.949634] May be due to missing lock nesting notation > > [ 193.949634] > > [ 193.956446] 3 locks held by migration/0/11: > > [ 193.960642] #0: (____ptrval____) (timekeeper_lock){-.-.}, at: change_clocksource+0x2c/0x118 > > [ 193.969125] #1: (____ptrval____) (tk_core.seq){----}, at: multi_cpu_stop+0x8c/0x140 > > [ 193.976903] #2: (____ptrval____) (&(&dev->power.lock)->rlock){....}, at: __pm_runtime_resume+0x40/0x98 > > [ 193.986339] > > [ 193.986339] stack backtrace: > > [ 193.990715] CPU: 0 PID: 11 Comm: migration/0 Not tainted 5.0.0-rc4-next-20190201-00007-g731346f #3 > > [ 193.999707] Hardware name: Silicon Linux RZ/G2E evaluation kit EK874 (CAT874 + CAT875) (DT) > > [ 194.008089] Call trace: > > [ 194.010553] dump_backtrace+0x0/0x178 > > [ 194.014227] show_stack+0x14/0x20 > > [ 194.017562] dump_stack+0xb0/0xec > > [ 194.020895] __lock_acquire+0xfb4/0x1c08 > > [ 194.024832] lock_acquire+0xd0/0x268 > > [ 194.028420] ktime_get+0x5c/0x108 > > [ 194.031747] update_pm_runtime_accounting+0x14/0x68 > > [ 194.036643] rpm_resume+0x4ec/0x698 > > [ 194.040144] __pm_runtime_resume+0x50/0x98 > > [ 194.044264] sh_tmu_enable.part.1+0x24/0x50 > > [ 194.048462] sh_tmu_clocksource_enable+0x48/0x70 > > [ 194.053097] change_clocksource+0x84/0x118 > > [ 194.057208] multi_cpu_stop+0x8c/0x140 > > [ 194.060970] cpu_stopper_thread+0xac/0x120 > > [ 194.065087] smpboot_thread_fn+0x1ac/0x2c8 > > [ 194.069198] kthread+0x128/0x130 > > [ 194.072439] ret_from_fork+0x10/0x18 > > > > > > Regards, > > Biju > > > > > -----Original Message----- > > > From: Rafael J. Wysocki <rafael@xxxxxxxxxx> > > > Sent: 30 January 2019 21:53 > > > To: Vincent Guittot <vincent.guittot@xxxxxxxxxx> > > > Cc: Linux PM <linux-pm@xxxxxxxxxxxxxxx>; Linux Kernel Mailing List <linux- > > > kernel@xxxxxxxxxxxxxxx>; Linux ARM <linux-arm- > > > kernel@xxxxxxxxxxxxxxxxxxx>; Linux OMAP Mailing List <linux- > > > omap@xxxxxxxxxxxxxxx>; Rafael J. Wysocki <rjw@xxxxxxxxxxxxx>; Ulf > > > Hansson <ulf.hansson@xxxxxxxxxx>; Biju Das <biju.das@xxxxxxxxxxxxxx>; > > > Geert Uytterhoeven <geert@xxxxxxxxxxxxxx>; Linux-Renesas <linux- > > > renesas-soc@xxxxxxxxxxxxxxx> > > > Subject: Re: [PATCH v3] PM-runtime: fix deadlock with ktime > > > > > > On Wed, Jan 30, 2019 at 6:26 PM Vincent Guittot > > > <vincent.guittot@xxxxxxxxxx> wrote: > > > > > > > > A deadlock has been seen when swicthing clocksources which use PM > > > runtime. > > > > The call path is: > > > > change_clocksource > > > > ... > > > > write_seqcount_begin > > > > ... > > > > timekeeping_update > > > > ... > > > > sh_cmt_clocksource_enable > > > > ... > > > > rpm_resume > > > > pm_runtime_mark_last_busy > > > > ktime_get > > > > do > > > > read_seqcount_begin > > > > while read_seqcount_retry > > > > .... > > > > write_seqcount_end > > > > > > > > Although we should be safe because we haven't yet changed the > > > > clocksource at that time, we can't because of seqcount protection. > > > > > > > > Use ktime_get_mono_fast_ns() instead which is lock safe for such case > > > > > > > > With ktime_get_mono_fast_ns, the timestamp is not guaranteed to be > > > > monotonic across an update and as a result can goes backward. > > > > According to > > > > update_fast_timekeeper() description: "In the worst case, this can > > > > result is a slightly wrong timestamp (a few nanoseconds)". For PM > > > > runtime autosuspend, this means only that the suspend decision can be > > > > slightly sub optimal. > > > > > > > > Fixes: 8234f6734c5d ("PM-runtime: Switch autosuspend over to using > > > > hrtimers") > > > > Reported-by: Biju Das <biju.das@xxxxxxxxxxxxxx> > > > > Signed-off-by: Vincent Guittot <vincent.guittot@xxxxxxxxxx> > > > > --- > > > > > > > > Hi Rafael, > > > > > > > > Sorry, I sent the version with the typo mistake that generated the > > > > compilation error reported by kbuild-test-robot > > > > > > > > This version doesn't have the typo. > > > > > > OK, I've applied this one, thanks! > > > > > > > > Renesas Electronics Europe Ltd, Dukes Meadow, Millboard Road, Bourne End, Buckinghamshire, SL8 5FH, UK. Registered in England & Wales under Registered No. 04586709.