From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1757054Ab3JOCJp (ORCPT ); Mon, 14 Oct 2013 22:09:45 -0400 Received: from mail-vb0-f47.google.com ([209.85.212.47]:50286 "EHLO mail-vb0-f47.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755663Ab3JOCJo (ORCPT ); Mon, 14 Oct 2013 22:09:44 -0400 Date: Mon, 14 Oct 2013 23:09:35 -0300 From: Arnaldo Carvalho de Melo To: Adrian Hunter Cc: David Ahern , Jiri Olsa , Namhyung Kim , Linux Kernel Mailing List Subject: perf top using /proc/kcore Message-ID: <20131015020935.GB30662@ghostprotocols.net> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline 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 Hi Adrian, While testing 'perf top -U' to reply a message in another thread I noticed that ksm symbols appeared as '[kernel]', and only when I tried 'perf top -U -v' to look at the DSO long name I noticed that it was... /proc/kcore. Question is: since we have access to /proc/modules, can't we parse that, as when we have access to vmlinux, and recreate the mmaps, etc, and see: [module] symbol Instead of grouping everything into a single [kernel] bucket? - Arnaldo