On Wed, 12 Aug 2015, tip-bot for Peter Zijlstra wrote:
> Commit-ID: c7999c6f3fed9e383d3131474588f282ae6d56b9
> Gitweb: http://git.kernel.org/tip/c7999c6f3fed9e383d3131474588f282ae6d56b9
> Author: Peter Zijlstra <[email protected]>
> AuthorDate: Tue, 4 Aug 2015 19:22:49 +0200
> Committer: Ingo Molnar <[email protected]>
> CommitDate: Wed, 12 Aug 2015 11:37:22 +0200
>
> perf: Fix PERF_EVENT_IOC_PERIOD migration race
>
> I ran the perf fuzzer, which triggered some WARN()s which are due to
> trying to stop/restart an event on the wrong CPU.
Do you have a list of which WARNs this should fix? Just so I can
update my bug tracker.
Thanks,
Vince
On Wed, Aug 12, 2015 at 09:49:40AM -0400, Vince Weaver wrote:
> On Wed, 12 Aug 2015, tip-bot for Peter Zijlstra wrote:
>
> > Commit-ID: c7999c6f3fed9e383d3131474588f282ae6d56b9
> > Gitweb: http://git.kernel.org/tip/c7999c6f3fed9e383d3131474588f282ae6d56b9
> > Author: Peter Zijlstra <[email protected]>
> > AuthorDate: Tue, 4 Aug 2015 19:22:49 +0200
> > Committer: Ingo Molnar <[email protected]>
> > CommitDate: Wed, 12 Aug 2015 11:37:22 +0200
> >
> > perf: Fix PERF_EVENT_IOC_PERIOD migration race
> >
> > I ran the perf fuzzer, which triggered some WARN()s which are due to
> > trying to stop/restart an event on the wrong CPU.
>
> Do you have a list of which WARNs this should fix? Just so I can
> update my bug tracker.
I think this triggered:
arch/x86/kernel/cpu/perf_event.c:x86_pmu_del()
if (WARN_ON_ONCE(i == cpuc->n_events)) /* called ->del() without ->add() ? */
Through perf_ioctl()->perf_event_period().
I didn't safe a copy of the exact backtrace.