Messages in this thread | | | From | Alexander Shishkin <> | Subject | Re: [PATCH v8 12/14] x86: perf: intel_pt: Intel PT PMU driver | Date | Mon, 12 Jan 2015 14:19:35 +0200 |
| |
Peter Zijlstra <peterz@infradead.org> writes:
> On Fri, Nov 14, 2014 at 03:43:45PM +0200, Alexander Shishkin wrote: >> +static void pt_event_stop(struct perf_event *event, int mode) >> +{ >> + struct pt *pt = this_cpu_ptr(&pt_ctx); >> + >> + ACCESS_ONCE(pt->handle_nmi) = 0; > > Why is this needed? Will the hardware still generate interrupts if you > stop the PT thing?
It's meant to serialize nmi handler against the pmu:: callbacks, but now you mentioned it, it looks very much redundant.
Cheers, -- Alex
| |