From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755525Ab3HFOCl (ORCPT ); Tue, 6 Aug 2013 10:02:41 -0400 Received: from mx1.redhat.com ([209.132.183.28]:24074 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755351Ab3HFOCa (ORCPT ); Tue, 6 Aug 2013 10:02:30 -0400 Date: Tue, 6 Aug 2013 16:02:21 +0200 From: Jiri Olsa To: Michael Ellerman Cc: linux-kernel@vger.kernel.org, acme@ghostprotocols.net, a.p.zijlstra@chello.nl Subject: Re: [PATCH v2 1/2] perf tools: Add support for pinned modifier Message-ID: <20130806140221.GD1068@krava.brq.redhat.com> References: <1375795686-4226-1-git-send-email-michael@ellerman.id.au> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1375795686-4226-1-git-send-email-michael@ellerman.id.au> 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 Tue, Aug 06, 2013 at 11:28:05PM +1000, Michael Ellerman wrote: > This commit adds support for a new modifier "D", which requests that the > event, or group of events, be pinned to the PMU. > > The "p" modifier is already taken for precise, and "P" may be used in > future to mean "fully precise". > > So we use "D", which stands for pinneD - and looks like a padlock, or if > you're using the ":D" syntax perf smiles at you. > > This is an oft-requested feature from our HW folks, who want to be able > to run a large number of events, but also want 100% accurate results for > instructions per cycle. > > Comparison of results with and without pinning: > > $ perf stat -e '{cycles,instructions}:D' -e cycles,instructions,... > > 79,590,480,683 cycles # 0.000 GHz > 166,123,716,524 instructions # 2.09 insns per cycle > # 0.11 stalled cycles per insn > > 79,352,134,463 cycles # 0.000 GHz [11.11%] > 165,178,301,818 instructions # 2.08 insns per cycle > # 0.11 stalled cycles per insn [11.13%] > > As you can see although perf does a very good job of scaling the values > in the non-pinned case, there is some small discrepancy. > > The patch is fairly straight forward, the one detail is that we need to > make sure we only request pinning for the group leader when we have a > group. > > Signed-off-by: Michael Ellerman Tested-by/Acked-by: Jiri Olsa thanks, jirka