Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752883Ab3E1Jyb (ORCPT ); Tue, 28 May 2013 05:54:31 -0400 Received: from mail-we0-f170.google.com ([74.125.82.170]:54236 "EHLO mail-we0-f170.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750829Ab3E1Jya (ORCPT ); Tue, 28 May 2013 05:54:30 -0400 Date: Tue, 28 May 2013 11:54:25 +0200 From: Ingo Molnar To: Stephane Eranian Cc: Peter Zijlstra , Jiri Olsa , LKML , Corey Ashford , Frederic Weisbecker , Ingo Molnar , Namhyung Kim , Paul Mackerras , Arnaldo Carvalho de Melo , Andi Kleen , David Ahern Subject: Re: [PATCH 0/9] perf: Adding better precise_ip field handling Message-ID: <20130528095425.GB6764@gmail.com> References: <20130510101823.GA18427@gmail.com> <20130510102245.GA31235@dyad.programming.kicks-ass.net> <20130510103112.GA18755@gmail.com> <20130510103436.GC31235@dyad.programming.kicks-ass.net> <20130510105536.GA18805@gmail.com> <20130510112756.GH31235@dyad.programming.kicks-ass.net> <20130511075008.GC24435@gmail.com> <20130513093624.GC3708@dyad.programming.kicks-ass.net> <20130513194313.GA30998@gmail.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: User-Agent: Mutt/1.5.21 (2010-09-15) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 2484 Lines: 63 * Stephane Eranian wrote: > On Mon, May 13, 2013 at 9:43 PM, Ingo Molnar wrote: > > > > * Peter Zijlstra wrote: > > > >> On Sat, May 11, 2013 at 09:50:08AM +0200, Ingo Molnar wrote: > >> > That's really a red herring: there's absolutely no reason why the > >> > kernel could not pass back the level of precision it provided. > >> > >> All I've been saying is that doing random precision without feedback is > >> confusing. > > > > I agree with that. > > > >> We also don't really have a good feedback channel for this kind of > >> thing. The best I can come up with is tagging each and every sample with > >> the quality it represents. I think we can do with only one extra > >> PERF_RECORD_MISC bit, but it looks like we're quickly running out of > >> those things. > > > > Hm, how about passing precision back to user-space at creation time, in > > the perf_attr data structure? There's no need to pass it back in every > > sample, precision will not really change during the life-time of an event. > > > >> But I think the biggest problem is PEBS's inability do deal with REP > >> prefixes; see this email from Stephane: > >> https://lkml.org/lkml/2011/2/1/177 > >> > >> It is really unfortunate for PEBS to have such a side-effect; but it > >> makes all memset/memcpy/memmove things appear like they have no cost. > >> I'm very sure that will surprise a number of people. > > > > I'd expect PEBS to get gradually better. > > > > Note that at least for user-space, REP MOVS is getting rarer. libc uses > > SSE based memcpy/memset variants - which is not miscounted by PEBS. The > > kernel still uses REP MOVS - but it's a special case because it cannot > > cheaply use vector registers. > > > > The vast majority of code gets measured by cycles:pp more accurately than > > cycles. > > > I don't understand how you come to that conclusion. [...] By frequently looking at cycles:pp output. > [...] I can show you simple examples where this is not true at all (even > without rep mov). That would be useful if there's any practical problem with cycles:pp. In terms of profiling typical kernel and user space functions it does appear to work very well. Thanks, Ingo -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/