From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754841AbaCCUd5 (ORCPT ); Mon, 3 Mar 2014 15:33:57 -0500 Received: from mx1.redhat.com ([209.132.183.28]:1976 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752487AbaCCUd4 (ORCPT ); Mon, 3 Mar 2014 15:33:56 -0500 Date: Mon, 3 Mar 2014 15:32:57 -0500 From: Don Zickus To: Andi Kleen Cc: Peter Zijlstra , Joe Mario , Davidlohr Bueso , acme@ghostprotocols.net, LKML , jolsa@redhat.com, fowles@inreach.com, eranian@google.com, Arnaldo Carvalho de Melo , David Ahern , Frederic Weisbecker , Mike Galbraith , Paul Mackerras , Richard Fowles Subject: Re: [PATCH 08/19] perf c2c: Shared data analyser Message-ID: <20140303203257.GI25953@redhat.com> References: <1393609388-40489-1-git-send-email-dzickus@redhat.com> <1393609388-40489-9-git-send-email-dzickus@redhat.com> <87lhwvawc4.fsf@tassilo.jf.intel.com> <20140228194642.GM25953@redhat.com> <1393621411.2899.39.camel@buesod1.americas.hpqcorp.net> <20140303150531.GZ25953@redhat.com> <20140303172316.GA22728@two.firstfloor.org> <5314C4C4.1020304@redhat.com> <20140303184117.GL9987@twins.programming.kicks-ass.net> <20140303185819.GB22728@two.firstfloor.org> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20140303185819.GB22728@two.firstfloor.org> 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 On Mon, Mar 03, 2014 at 07:58:19PM +0100, Andi Kleen wrote: > On Mon, Mar 03, 2014 at 07:41:17PM +0100, Peter Zijlstra wrote: > > On Mon, Mar 03, 2014 at 01:07:00PM -0500, Joe Mario wrote: > > > If you only sample on the HITMs then you don't get the store misses. > > > That means you'll not be able to detect who is simultaneously tugging > > > on the same cache lines. That gives up much of the value of "perf > > > c2c". > > > > As long as you know which lines are hurting bringing in (loads) you can > > often figure out who is doing the stores on them. > > Yes, especially since every store is a load too (unless you're talking > WC) Thoughts on how to determine which load is a potential store? I agree every store needs to load the cacheline, but I wasn't sure if there was an approach that could be applied to determine anything useful. Cheers, Don > > The method c2c uses is more exact, but keep in mind it's a sampling > heuristic in any cases, with some potential bias. load-latency tags > the loads randomly and there's no guarantee that tagging is fully > uniform. Also you only see a subset in any case. > > > > > > As we developed this, we ended up settling on Ivy Bridge to get the > > > behavior we wanted. > > > > Wouldn't SNB also work? > > Yes. > > Haswell is best however because it can report addresses on far more > events. > > -Andi > > -- > ak@linux.intel.com -- Speaking for myself only.