From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753739Ab3C0U7Z (ORCPT ); Wed, 27 Mar 2013 16:59:25 -0400 Received: from mail-ve0-f175.google.com ([209.85.128.175]:44509 "EHLO mail-ve0-f175.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752268Ab3C0U7Y (ORCPT ); Wed, 27 Mar 2013 16:59:24 -0400 Date: Wed, 27 Mar 2013 17:59:17 -0300 From: Arnaldo Carvalho de Melo To: Namhyung Kim Cc: Ingo Molnar , Peter Zijlstra , Paul Mackerras , Namhyung Kim , LKML Subject: Re: [PATCH 1/3] perf sort: Add 'addr' sort key Message-ID: <20130327205917.GA2027@ghostprotocols.net> References: <1364299918-31175-1-git-send-email-namhyung@kernel.org> <20130326144023.GA1621@gmail.com> <87ehf1gylp.fsf@sejong.aot.lge.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <87ehf1gylp.fsf@sejong.aot.lge.com> X-Url: http://acmel.wordpress.com 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 Em Wed, Mar 27, 2013 at 03:12:34PM +0900, Namhyung Kim escreveu: > On Tue, 26 Mar 2013 15:40:23 +0100, Ingo Molnar wrote: > > * Namhyung Kim wrote: > >> New addr sort key provides a way to sort the entries by the symbol > >> addresses. > > No objections from me - just wondering about the motivation: why would we > > want to sort by symbol address? > In fact, it came from bugzilla: > https://bugzilla.kernel.org/show_bug.cgi?id=55561 Will Cohen asked if this was possible while trying to figure out why some libxul.so samples were not being resolved to symbols, Will? Namhyung, I'm using: BugLink: https://bugzilla.kernel.org/show_bug.cgi?id=55561 Just before the: Link: entry, consider adding it when addressing some bugzilla entry, be it from bugzilla.kernel.org or from some other bugzilla where problems or feature requests were made. - Arnaldo > > > > Perhaps to see the overhead layout/distribution within a DSO, to better > > cache-pack hot functions by placing them next to each other? > > > > Putting a short usecase into the changelog (or even better, the > > documentation) would be nice. > > Well, I didn't thought it much, but yeah, aforementioned usecase looks > promising. Also it might be useful to find out which part is the > hotpath in a function if you have some really big functions (possibly > due to auto-inlining or something). > > Arnaldo, do you have other usecase/scenario in mind? See above.