From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1758023Ab1KROOe (ORCPT ); Fri, 18 Nov 2011 09:14:34 -0500 Received: from mx1.redhat.com ([209.132.183.28]:3701 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1757913Ab1KROOd (ORCPT ); Fri, 18 Nov 2011 09:14:33 -0500 Date: Fri, 18 Nov 2011 12:14:16 -0200 From: Arnaldo Carvalho de Melo To: Jiri Olsa Cc: a.p.zijlstra@chello.nl, mingo@elte.hu, paulus@samba.org, linux-kernel@vger.kernel.org Subject: Re: [RFC,PATCH] perf tool: Refactoring IO data files code Message-ID: <20111118141416.GA13052@ghostprotocols.net> References: <1321624005-6889-1-git-send-email-jolsa@redhat.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1321624005-6889-1-git-send-email-jolsa@redhat.com> X-Url: http://acmel.wordpress.com User-Agent: Mutt/1.5.19 (2009-01-05) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Em Fri, Nov 18, 2011 at 02:46:40PM +0100, Jiri Olsa escreveu: > Hi, > > based on following discussion: > http://marc.info/?l=linux-kernel&m=131731212425421&w=2 > http://marc.info/?l=linux-kernel&m=131736842010972&w=2 > > I made some changes to have event memory mapping and input/output > data code in one place. The reason was to make the change for > multiple event data files at least possible ;) > > This patchset is marked as RFC since it's quite big change > and I expect some better design might come up. > > However, patches 1 and 2 are not big deal, since they are > just small (not so related) changes. > > Patches 3-4 are the main change of the patchset. > > 1/5 perf tool: Fix session host_nachine retrieval > 2/5 perf tool: Initialize events IDs in a single function > 3/5 perf tool: Introducing perf_mmap object > 4/5 perf tool: Introducing perf_data object > 5/5 perf tool: Putting mmap support to perf_data object Reading those patches now. > I did some overall testing of all the changed command, > and so far so good. Just hit issue with 'diff' command, > but this one seems not to work even without my changes. Yeah, it needs more work to properly suport multi-event perf.data files. > Also the current perf python binsing code is broken > wit missing dependencies so I could not test my > changes there. Which ones? I fixed some and posted to my perf/urgent branch, waiting for Ingo to merge. - Arnaldo