Re: [PATCH 2/3] perf/x86/pebs: add workaround for broken OVFL status on HSW

From: Stephane Eranian
Date: Thu Mar 10 2016 - 11:10:13 EST

On Thu, Mar 10, 2016 at 5:53 AM, Peter Zijlstra <peterz@xxxxxxxxxxxxx> wrote:
> On Mon, Mar 07, 2016 at 01:18:40PM +0100, Peter Zijlstra wrote:
>> On Mon, Mar 07, 2016 at 11:24:13AM +0100, Peter Zijlstra wrote:
>> > I suspect Andi is having something along:
>> >
>> >
>> >
>> > applied to his tree.
>> OK, I munged a bunch of patches together, please have a hard look at the
>> end result found in:
>> git:// perf/core
> So while running tip/master I just triggered the "irq loop stuck" warn
> on my IVB, with status==1, and PMC0 didn't look weird at all.
> I've not seen that for a fair while. But it does show it still happens
> on not quite antique hardware.

Yes, as I said, tip.git is missing something, with Andi's missing 3
patches I do not get
that. But with them, I was getting the problem I described earlier, so
I just posted a
patch to handle that particular case where bit 62 is not set, yet PEBS
counters show
overflow. Test it.