On Tue, May 25, 2010 at 6:18 PM, Peter Zijlstra <pet...@infradead.org> wrote:
> On Tue, 2010-05-25 at 18:10 +0200, Stephane Eranian wrote:
>> > Index: linux-2.6/kernel/perf_event.c
>> > ===================================================================
>> > --- linux-2.6.orig/kernel/perf_event.c
>> > +++ linux-2.6/kernel/perf_event.c
>> > @@ -668,15 +668,9 @@ group_sched_in(struct perf_event *group_
>> > }
>> > }
>> >
>> > - if (!txn)
>> > + if (!txn || !pmu->commit_txn(pmu))
>> > return 0;
>> >
>> > - ret = pmu->commit_txn(pmu);
>> > - if (!ret) {
>> > - pmu->cancel_txn(pmu);
>> > - return 0;
>> > - }
>> > -
>> > group_error:
>> > /*
>> > * Groups can be scheduled in as one unit only, so undo any
>> >
>> Looks okay.
>>
>> I believe you can also drop the txn test in group_sched_in() after
>> group_error:,
>> given you have the if !(txn) return 0.
>
> Can't we still get in the group_error: branch with either scenario?
>
You're right. We must keep it because of failure in the siblings' loop.
------------------------------------------------------------------------------
_______________________________________________
perfmon2-devel mailing list
perfmon2-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/perfmon2-devel
