Re: [PATCH 2/2] kernel/events: Add option to enable counting sideband events in wakeup_events

[Naveen N. Rao]

On 2017/07/12 01:48PM, Peter Zijlstra wrote:
> On Mon, Jun 19, 2017 at 08:01:08PM +0530, Naveen N. Rao wrote:
> > diff --git a/kernel/events/ring_buffer.c b/kernel/events/ring_buffer.c
> > index 4e7c728569a8..f43a6081141f 100644
> > --- a/kernel/events/ring_buffer.c
> > +++ b/kernel/events/ring_buffer.c
> > @@ -197,6 +197,19 @@ __perf_output_begin(struct perf_output_handle *handle,
> >      * none of the data stores below can be lifted up by the compiler.
> >      */
> >  
> > +   if (event->attr.count_sb_events && !event->attr.watermark) {
> > +           int wakeup_events = event->attr.wakeup_events;
> > +
> > +           if (wakeup_events) {
> > +                   int events = local_inc_return(&rb->events);
> > +
> > +                   if (events >= wakeup_events) {
> > +                           local_sub(wakeup_events, &rb->events);
> > +                           local_inc(&rb->wakeup);
> > +                   }
> > +           }
> > +   }
> > +
> >     if (unlikely(head - local_read(&rb->wakeup) > rb->watermark))
> >             local_add(rb->watermark, &rb->wakeup);
> >  
> 
> So this is a very performance sensitive function; not at all happy to
> add bits here ... :/

:O

Does it at all help if the above is instead guarded by:
        if (unlikely(event->attr.count_sb_events)) {
                ...
        }

That should hopefully limit the impact to only when that option is used?


Thanks,
Naveen