Messages in this thread | | | Date | Thu, 10 Jan 2013 14:34:46 -0800 | Subject | Re: [PATCH] hardlockup: detect hard lockups without NMIs using secondary cpus | From | Colin Cross <> |
| |
On Thu, Jan 10, 2013 at 12:38 PM, Tony Lindgren <tony@atomide.com> wrote: > > * Colin Cross <ccross@android.com> [130109 18:05]: > > +static void watchdog_check_hardlockup_other_cpu(void) > > +{ > > + int cpu; > > + cpumask_t cpus = watchdog_cpus; > > + > > + /* > > + * Test for hardlockups every 3 samples. The sample period is > > + * watchdog_thresh * 2 / 5, so 3 samples gets us back to slightly over > > + * watchdog_thresh (over by 20%). > > + */ > > + if (__this_cpu_read(hrtimer_interrupts) % 3 != 0) > > + return; > > + > > + /* check for a hardlockup on the next cpu */ > > + cpu = cpumask_next(smp_processor_id(), &cpus); > > Hmm don't you want to check cpu_oneline_mask here and > return if the other CPU is offline?
watchdog_cpus is effectively a local copy of cpu_online_mask, but updated after the watchdog_nmi_touch in watchdog_nmi_enable. This avoids a false positive after hotplugging in a cpu when cpu_online_mask is true but that cpu hasn't yet run it's first hrtimer.
> > + if (cpu >= nr_cpu_ids) > > + cpu = cpumask_first(&cpus); > > + if (cpu == smp_processor_id()) > > + return; > > Regards, > > Tony
| |