From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932333Ab0KOLbv (ORCPT ); Mon, 15 Nov 2010 06:31:51 -0500 Received: from one.firstfloor.org ([213.235.205.2]:33458 "EHLO one.firstfloor.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755444Ab0KOLbu (ORCPT ); Mon, 15 Nov 2010 06:31:50 -0500 Date: Mon, 15 Nov 2010 12:31:48 +0100 From: Andi Kleen To: Stephane Eranian Cc: Peter Zijlstra , Andi Kleen , linux-kernel@vger.kernel.org, cjashfor@linux.vnet.ibm.com, mingo@elte.hu, fweisbec@gmail.com, Andi Kleen Subject: Re: [PATCH 1/2] perf-events: Add support for supplementary event registers v2 Message-ID: <20101115113148.GG7269@basil.fritz.box> References: <1289580941-29744-1-git-send-email-andi@firstfloor.org> <1289582630.2084.331.camel@laptop> <20101113101148.GB6418@gargoyle.fritz.box> <1289644367.2084.526.camel@laptop> <1289819201.2109.489.camel@laptop> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: User-Agent: Mutt/1.5.20 (2009-06-14) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org > Why would you want to measure both at the same time? > Because you can measure two distinct request/response in one run. > > If you want to measure both you need to create two distinct perf events. What I did currently was to assign some cache events to 0xb7 and others to bb on Westmere. That's a bit arbitrary, but allows to measure some common combinations together at least. In theory one could teach the scheduler that these events are equivalent, but that seems like too much complexity for me currently. -Andi -- ak@linux.intel.com -- Speaking for myself only.