From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755604AbbJUS5Y (ORCPT ); Wed, 21 Oct 2015 14:57:24 -0400 Received: from casper.infradead.org ([85.118.1.10]:37048 "EHLO casper.infradead.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753232AbbJUS5X (ORCPT ); Wed, 21 Oct 2015 14:57:23 -0400 Date: Wed, 21 Oct 2015 20:57:35 +0200 From: Peter Zijlstra To: Andi Kleen Cc: Andi Kleen , linux-kernel@vger.kernel.org, Ingo Molnar , Thomas Gleixner Subject: Re: [PATCH] x86, perf: Use INST_RETIRED.PREC_DIST for cycles:pp on Skylake Message-ID: <20151021185735.GB2508@worktop.programming.kicks-ass.net> References: <1445295496-8550-1-git-send-email-andi@firstfloor.org> <20151020113608.GC17308@twins.programming.kicks-ass.net> <20151020222844.GK3533@two.firstfloor.org> <20151021080947.GA2881@worktop.programming.kicks-ass.net> <20151021152608.GG15102@tassilo.jf.intel.com> <20151021165252.GJ3604@twins.programming.kicks-ass.net> <20151021165542.GN3533@two.firstfloor.org> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20151021165542.GN3533@two.firstfloor.org> User-Agent: Mutt/1.5.22.1 (2013-10-16) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Wed, Oct 21, 2015 at 06:55:42PM +0200, Andi Kleen wrote: > On Wed, Oct 21, 2015 at 06:52:52PM +0200, Peter Zijlstra wrote: > > On Wed, Oct 21, 2015 at 08:26:08AM -0700, Andi Kleen wrote: > > > So it turns out that UOPS_RETIRED.ALL+pebs only works by accident > > > on Skylake. It's not in the specification. > > > > Accident or not, it works and you've tested it. Might as well keep it. > > Specs can be updated. > > No, it's better not to use it. This is really rather unfortunate; what is the failure mode? > > Also, even with the SNB 'hickup' they still haven't got the memo that > > cycles is important for PEBS ? > > INST_RETIRED.PREC_DIST is a far better cycles than UOPS_RETIRED.ALL > (see my earlier example) If you only want the one counter, yes. And since you made me look at the SDM; it looks like our skl pebs table isn't anything like the table there. Also, do the events BR_INST_RETIRED have a composable umask? That is, .event=0xc4, .umask=41 is valid and would be 'conditional far branches'. The same would be true for the HSW/BDW table I suppose, the memops retired stuff looks composable but we currently do not allow that.