On Wed, Oct 24, 2012 at 02:01:18PM +0200, Peter Zijlstra wrote: > On Tue, 2012-10-23 at 18:50 +0200, Jiri Olsa wrote: > > On Tue, Oct 23, 2012 at 06:13:09PM +0200, Peter Zijlstra wrote: > > > On Sat, 2012-10-20 at 16:33 +0200, Jiri Olsa wrote: > > > > It's possible some of the counters in the group could be > > > > disabled when sampling member of the event group is reading > > > > the rest via PERF_SAMPLE_READ sample type processing. Disabled > > > > counters could then produce wrong numbers. > > > > > > > > Fixing that by reading only enabled counters for PERF_SAMPLE_READ > > > > sample type processing. > > > > > > > > > > However did you run into this? > > > > yep, with perf record -a > > > > hm, I just checked and we enable/disable event groups atomicaly.. > > I haven't checked that before because it seemed obvious :-/ > > > > So, I'm not sure now about the exact code path that triggered it > > in my test.. however you could always disable child event from > > group and hit this issue, but thats not what happened in perf. > > > > might be some other bug... I'll check > > Right, so I don't object to the patch per-se, I was just curious how you > ran into it, because ISTR what you just said, we enable all this stuff > together. > > Also, why would disabled counters give strange values? They'd simply > return the same old value time after time, right?
well, x86_pmu_read calls x86_perf_event_update, which expects the event is active.. if it's not it'll update the count from whatever left in event.hw.idx counter.. could be uninitialized or used by others.. I can easily reproduce this one, so let's see.. ;) jirka -- 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/