Re: [PATCH] perf: fix possible divide-by-zero in perf_swevent_overflow()

[Peter Zijlstra <peterz@infradead.org>]

On Fri, 2010-08-27 at 20:21 +0800, DDD wrote:

> >> Maybe the root cause was from kgdb/hw_breakpoint_layer,
> > 
> > Yeah, I think there's a bug in the hw_breakpoint stuff, does something
> > like the below fix it?

> Thanks for your patch, but I still could reproduce the problem with your 
> patch.

Frederic, any clue as to what makes hw breakpoints go funny and have
last_period == 0?

> > ---
> >  kernel/hw_breakpoint.c |   13 ++++++++++++-
> >  1 files changed, 12 insertions(+), 1 deletions(-)
> > 
> > diff --git a/kernel/hw_breakpoint.c b/kernel/hw_breakpoint.c
> > index d71a987..f57ebee 100644
> > --- a/kernel/hw_breakpoint.c
> > +++ b/kernel/hw_breakpoint.c
> > @@ -600,9 +600,20 @@ static int __init init_hw_breakpoint(void)
> >  }
> >  core_initcall(init_hw_breakpoint);
> >  
> > +static int hw_breakpoint_enable(struct perf_event *event)
> > +{
> > +     struct hw_perf_event *hwc = &event->hw;
> > +
> > +     if (hwc->sample_period) {
> > +             hwc->last_period = hwc->sample_period;
> > +             perf_swevent_set_period(event);
> > +     }
> > +
> > +     return arch_install_hw_breakpoint(event);
> > +}
> >  
> >  struct pmu perf_ops_bp = {
> > -     .enable         = arch_install_hw_breakpoint,
> > +     .enable         = hw_breakpoint_enable,
> >       .disable        = arch_uninstall_hw_breakpoint,
> >       .read           = hw_breakpoint_pmu_read,
> >  };
> > 
> >