Re: [PATCHv3] perf powerpc: Don't call perf_event_disable from atomic context

From	Jiri Olsa <jolsa@redhat.com>
Newsgroups	linux.kernel
Subject	Re: [PATCHv3] perf powerpc: Don't call perf_event_disable from atomic context
Date	2016-10-26 17:30 +0200
Message-ID	<swyjo-5Fl-7@gated-at.bofh.it> (permalink)
References	(1 earlier) <sw3IB-1ZK-11@gated-at.bofh.it> <sw5U6-3BL-33@gated-at.bofh.it> <swlPb-5zq-5@gated-at.bofh.it> <swt0l-1OS-5@gated-at.bofh.it> <swy9H-5BD-13@gated-at.bofh.it>
Organization	linux.* mail to news gateway

Show all headers | View raw

On Wed, Oct 26, 2016 at 05:12:49PM +0200, Peter Zijlstra wrote:
> On Wed, Oct 26, 2016 at 11:48:24AM +0200, Jiri Olsa wrote:
> 
> > diff --git a/kernel/events/core.c b/kernel/events/core.c
> > index c6e47e97b33f..04477983945e 100644
> > --- a/kernel/events/core.c
> > +++ b/kernel/events/core.c
> > @@ -1960,6 +1960,13 @@ void perf_event_disable(struct perf_event *event)
> >  }
> >  EXPORT_SYMBOL_GPL(perf_event_disable);
> >  
> > +void perf_event_disable_inatomic(struct perf_event *event, int kill)
> > +{
> > +	event->pending_kill    = kill;
> > +	event->pending_disable = 1;
> > +	irq_work_queue(&event->pending);
> > +}
> > +
> >  static void perf_set_shadow_time(struct perf_event *event,
> >  				 struct perf_event_context *ctx,
> >  				 u64 tstamp)
> > @@ -7074,9 +7081,7 @@ static int __perf_event_overflow(struct perf_event *event,
> >  	event->pending_kill = POLL_IN;
> >  	if (events && atomic_dec_and_test(&event->event_limit)) {
> >  		ret = 1;
> > -		event->pending_kill = POLL_HUP;
> > -		event->pending_disable = 1;
> > -		irq_work_queue(&event->pending);
> > +		perf_event_disable_inatomic(event, POLL_HUP);
> >  	}
> 
> So the pending_kill stuff is independent of the disable here. No need to
> combine the two. I've change the patch as per the below.
> 
> That is, pending_kill is part of pending_wakeup, not of pending_disable.
> Here we simply use both, its just that on disable we need a different
> kind of wakeup (HANGUP instead of IN).
> 
> See how after ->overflow_handler() we send a wakeup if there's a
> registered signal.

ok, seems good

thanks,
jirka

Thread

Re: [lkp] [perf powerpc]  18d1796d0b: [No primary change] Peter Zijlstra <peterz@infradead.org> - 2016-10-25 11:10 +0200
  Re: [LKP] [lkp] [perf powerpc]  18d1796d0b: [No primary change] "Huang\, Ying" <ying.huang@intel.com> - 2016-10-26 04:10 +0200
    [PATCHv3] perf powerpc: Don't call perf_event_disable from atomic  context Jiri Olsa <jolsa@redhat.com> - 2016-10-26 11:50 +0200
      Re: [PATCHv3] perf powerpc: Don't call perf_event_disable from  atomic context Peter Zijlstra <peterz@infradead.org> - 2016-10-26 17:20 +0200
        Re: [PATCHv3] perf powerpc: Don't call perf_event_disable from  atomic context Jiri Olsa <jolsa@redhat.com> - 2016-10-26 17:30 +0200
      [tip:perf/urgent] perf/powerpc: Don't call perf_event_disable()  from atomic context tip-bot for Jiri Olsa <tipbot@zytor.com> - 2016-10-28 12:20 +0200

csiph-web