[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20160817162451.GC24652@twins.programming.kicks-ass.net>
Date: Wed, 17 Aug 2016 18:24:51 +0200
From: Peter Zijlstra <peterz@...radead.org>
To: Ingo Molnar <mingo@...nel.org>
Cc: Linus Torvalds <torvalds@...ux-foundation.org>,
Dave Chinner <david@...morbit.com>, Tejun Heo <tj@...nel.org>,
Wu Fengguang <fengguang.wu@...el.com>,
"Kirill A. Shutemov" <kirill.shutemov@...ux.intel.com>,
Christoph Hellwig <hch@....de>,
"Huang, Ying" <ying.huang@...el.com>,
LKML <linux-kernel@...r.kernel.org>,
Bob Peterson <rpeterso@...hat.com>, LKP <lkp@...org>,
Arnaldo Carvalho de Melo <acme@...radead.org>
Subject: Re: [LKP] [lkp] [xfs] 68a9f5e700: aim7.jobs-per-min -13.6% regression
On Mon, Aug 15, 2016 at 07:03:00AM +0200, Ingo Molnar wrote:
>
> * Linus Torvalds <torvalds@...ux-foundation.org> wrote:
>
> > Make sure you actually use "perf record -e cycles:pp" or something
> > that uses PEBS to get real profiles using CPU performance counters.
>
> Btw., 'perf record -e cycles:pp' is the default now for modern versions
> of perf tooling (on most x86 systems) - if you do 'perf record' it will
> just use the most precise profiling mode available on that particular
> CPU model.
>
> If unsure you can check the event that was used, via:
>
> triton:~> perf report --stdio 2>&1 | grep '# Samples'
> # Samples: 27K of event 'cycles:pp'
Problem here is that Dave is using a KVM thingy. Getting hardware
counters in a guest is somewhat tricky but doable, but PEBS does not
virtualize.
Powered by blists - more mailing lists