Messages in this thread | | | Date | Wed, 28 May 2014 10:45:45 -0700 | From | Andi Kleen <> | Subject | Re: [RFC PATCH 6/7] perf, x86: large PEBS interrupt threshold |
| |
> So the window is fairly large for another to slip in, esp if you have a > fast (say inst-ret) and a slow (say br-misp) event.
yes, but unless you have really small periods it's still very unlikely that they synchronize, so it will only happen very rarely. And the rare cases won't affect the statistics very much.
Even with multi-pebs the min period is 1000, which is several orders of magnitude larger than the typical window.
-Andi
| |