From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753349Ab1HBKil (ORCPT ); Tue, 2 Aug 2011 06:38:41 -0400 Received: from merlin.infradead.org ([205.233.59.134]:39922 "EHLO merlin.infradead.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752793Ab1HBKig convert rfc822-to-8bit (ORCPT ); Tue, 2 Aug 2011 06:38:36 -0400 Subject: Re: [PATCH 6/7] perf, x86: Example code for AMD IBS From: Peter Zijlstra To: Robert Richter Cc: Ingo Molnar , Arnaldo Carvalho de Melo , LKML , Michael Cree , Lin Ming In-Reply-To: <20110801055008.GZ4590@erda.amd.com> References: <1311860812-28748-1-git-send-email-robert.richter@amd.com> <1311860812-28748-7-git-send-email-robert.richter@amd.com> <1311958729.5890.413.camel@twins> <20110801055008.GZ4590@erda.amd.com> Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: 8BIT Date: Tue, 02 Aug 2011 12:37:54 +0200 Message-ID: <1312281474.1147.40.camel@twins> Mime-Version: 1.0 X-Mailer: Evolution 2.30.3 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Mon, 2011-08-01 at 07:50 +0200, Robert Richter wrote: > On 29.07.11 12:58:49, Peter Zijlstra wrote: > > On Thu, 2011-07-28 at 15:46 +0200, Robert Richter wrote: > > > tools/perf/Documentation/examples/ibs.c | 436 ++++++++++++++++++++++++++++ > > > > That really isn't the place for this.. > > > > Also, how similar is the Alpha PMU to AMD IBS? > > Would you prefer > > tools/perf/Documentation/examples/x86/ibs.c > > instead? Possibly, but having just looked at the example again I don't really see it doing anything perf-record doesn't already do, so why does it deserve to live at all? Initially I thought it was a record+report like example, some code interpreting the 'mess' that comes out of IBS would be most appreciated and I think we can even ship that as perf-ibs-report/perf-ibs-annotate or so (and if its still remotely similar to its Alpha precursor that might make the Alpha folks happy too). > Actually I wasn't sure if this place is ok at all. But I wanted to > publish the code for reference and testing. There isn't somthing > similar already there, and the perf tool code is also not the best > place to implement such cpu specific features. > > But putting it to the documentation is common practice for library > packages (e.g. libpapi and libpfm). If the example is not for the > given architecture it won't be added to the package documentation. > This example won't be installed at all. Even for very special features > an example is important since it is often not part of the generic > implementation. Right, its just that I'm not seeing the extra value at all, aside from maybe the IBS_*_DEFAULT values that should live as an event config (poking mlin to see where he's at with those patches..).