Messages in this thread | | | Date | Thu, 21 May 2020 02:44:44 +0200 | From | Frederic Weisbecker <> | Subject | Re: [RFC PATCH] tick/sched: update full_nohz status after SCHED dep is cleared |
| |
On Wed, May 20, 2020 at 08:47:10PM +0200, Juri Lelli wrote: > On 20/05/20 19:02, Frederic Weisbecker wrote: > > On Wed, May 20, 2020 at 06:49:25PM +0200, Juri Lelli wrote: > > > On 20/05/20 18:24, Frederic Weisbecker wrote: > > > > > > Hummm, so I enabled 'timer:*', anything else you think I should be > > > looking at? > > > > Are you sure you also enabled timer_expire_entry? > > Because: > > I run with > > trace-cmd record -e sched_switch -e sched_wakeup -e sched_migrate_task \ > -e 'timer:*' -e 'irq_vectors:*' -e 'irq:softirq*' \ > -e 'irq:irq_handler*' taskset --cpu-list 4-35 ./sysjitter/sysjitter \ > --runtime 10 200 > > And > > $ grep ' timer_expire_entry:' trace.txt | wc -l > 1173 > > seems to confirm that the event was indeed enabled, e.g. > > ksoftirqd/11-100 [011] 159.270023: timer_expire_entry: timer=0xffff9807df8a5a60 function=delayed_work_timer_fn now=4294826418 baseclk=4294825984
That's interesting! I'll see if I can reproduce that with threaded irqs. If not we'll have to investigate on your machine. We really don't want to trigger timer softirqs when it's not necessary.
| |