Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755499Ab1DANWL (ORCPT ); Fri, 1 Apr 2011 09:22:11 -0400 Received: from mga09.intel.com ([134.134.136.24]:12105 "EHLO mga09.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750874Ab1DANWJ (ORCPT ); Fri, 1 Apr 2011 09:22:09 -0400 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="4.63,282,1299484800"; d="scan'208";a="727926654" Subject: Re: [RFC PATCH] perf report: add sort by file lines From: Lin Ming To: Peter Zijlstra Cc: Masami Hiramatsu , Arnaldo Carvalho de Melo , Frederic Weisbecker , LKML , "2nddept-manager@sdl.hitachi.co.jp" <2nddept-manager@sdl.hitachi.co.jp> In-Reply-To: <1301655903.4859.494.camel@twins> References: <1301391136.14111.98.camel@minggr.sh.intel.com> <1301392457.4859.74.camel@twins> <1301417155.3620.8.camel@localhost> <1301418183.2250.416.camel@laptop> <1301418404.2250.417.camel@laptop> <1301418533.2250.419.camel@laptop> <20110329174556.GC24129@ghostprotocols.net> <4D92818A.8020405@hitachi.com> <1301561155.14111.288.camel@minggr.sh.intel.com> <1301580117.4859.286.camel@twins> <1301582056.2271.15.camel@localhost> <1301588922.2250.498.camel@laptop> <4D95AC94.1080303@hitachi.com> <1301655903.4859.494.camel@twins> Content-Type: text/plain; charset="UTF-8" Date: Fri, 01 Apr 2011 21:22:10 +0800 Message-Id: <1301664130.2399.10.camel@localhost> Mime-Version: 1.0 X-Mailer: Evolution 2.28.0 (2.28.0-2.fc12) Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 973 Lines: 25 On Fri, 2011-04-01 at 19:05 +0800, Peter Zijlstra wrote: > On Fri, 2011-04-01 at 19:44 +0900, Masami Hiramatsu wrote: > > E.g. an indirect jump makes it hard to find where it jumps to. > > Yes indirect jumps are 'interesting', is there anything in the debug > info that will help us out with the possible target sites? > > Also what generates indirect jumps, switch() stmts? Indirect function > call that get optimized might also be, but hopefully DWARF would tell us > about that and allow us to know the state right after the jump. I'm not sure. But I'm thinking another way to trace the register assignment by LBR records. LBR will introduce overhead, but it can tell us all the branches, including the indirect jump. -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/