On Thu, 20 Feb 2014, Vince Weaver wrote: > On Thu, 20 Feb 2014, Vince Weaver wrote: > > > Might be relevant: check the last_cpu values. Right before the above > > it looks like the thread gets moved from CPU 1 to CPU 0 > > (possibly as a result of the long chain started with the > > close() of the tracepoint event), > > so the problem NMI watchdog event being enabled is a different one than > > the one that was disabled just before. > > so is this a false warning? If you get scheduled to a new CPU > and there's an already running CPU-wide event, is that OK? > > Or should x86_pmu_disable() be setting PERF_HES_STOPPED on all events? > It looks like other architectures are (such as armpmu_stop() ).
an actually I take that back, other architectures don't. It's just confusing how perf_pmu_disable() calls pmu->pmu_disable() which on arm is: armpmu_disable() which calls: armpmu->stop() which is *not* the same as: armpmu_stop() but is actually armpmu->pmu->stop() Urgh. Vince -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/