From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755571AbcA1Tm1 (ORCPT ); Thu, 28 Jan 2016 14:42:27 -0500 Received: from mx1.redhat.com ([209.132.183.28]:56422 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751692AbcA1TmZ (ORCPT ); Thu, 28 Jan 2016 14:42:25 -0500 Date: Thu, 28 Jan 2016 20:42:22 +0100 From: Jiri Olsa To: Arnaldo Carvalho de Melo Cc: Wang Nan , jolsa@kernel.org, linux-kernel@vger.kernel.org, =?iso-8859-1?Q?J=E9r=E9mie?= Galarneau , Zefan Li , pi3orama@163.com Subject: Re: [PATCH v2] perf data: Fix releasing event_class Message-ID: <20160128194222.GA12456@krava.brq.redhat.com> References: <20160128092639.GE2322@krava.brq.redhat.com> <1453987228-31418-1-git-send-email-wangnan0@huawei.com> <20160128145923.GB20826@krava.brq.redhat.com> <20160128190930.GA3121@redhat.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20160128190930.GA3121@redhat.com> User-Agent: Mutt/1.5.24 (2015-08-30) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Thu, Jan 28, 2016 at 05:09:30PM -0200, Arnaldo Carvalho de Melo wrote: SNIP > > however even with this patch I can see perf report getting > > more memory every time it reloads the data.. either there's > > something else, or I'm looking at it wrong ;-) > > So, for the last delete, IIRC that teardown was removed because it made > the exit operation to be slower than just not calling > perf_evlist__delete (or was it perf_session__delete, don't recall right > now, probably the later, because it would delete all the threads, mmaps, > etc). > > But yeah, for switching perf.data files on the fly, we better free all > that up... ook, I'll check some more and queue it in my next post jirka