From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753943AbaBZVQB (ORCPT ); Wed, 26 Feb 2014 16:16:01 -0500 Received: from merlin.infradead.org ([205.233.59.134]:33037 "EHLO merlin.infradead.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751326AbaBZVP7 (ORCPT ); Wed, 26 Feb 2014 16:15:59 -0500 Date: Wed, 26 Feb 2014 22:15:40 +0100 From: Peter Zijlstra To: Andi Kleen Cc: David Ahern , Andy Lutomirski , Stephane Eranian , "Yan, Zheng" , LKML , Ingo Molnar , Arnaldo Carvalho de Melo Subject: Re: [PATCH v3 00/14] perf, x86: Haswell LBR call stack support Message-ID: <20140226211540.GE22802@laptop.programming.kicks-ass.net> References: <1392703661-15104-1-git-send-email-zheng.z.yan@intel.com> <530D53EF.9090706@amacapital.net> <20140226185513.GL22728@two.firstfloor.org> <530E3E47.8010205@gmail.com> <530E4B42.5090401@gmail.com> <20140226205322.GM22728@two.firstfloor.org> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20140226205322.GM22728@two.firstfloor.org> User-Agent: Mutt/1.5.21 (2012-12-30) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Wed, Feb 26, 2014 at 09:53:22PM +0100, Andi Kleen wrote: > > Is there some reason not to enable frame pointers? > > It makes code slower. > > Especially on Atom CPUs, where it causes pipeline stalls, but Yeah, but nobody sane cares about the in-order atom crap CPUs. > also to some degree on others, because you lose one register and > spend a little bit of time setting it up, so making small > functions more expensive. Luckily GCC is rather good at inlining a lot of those. Esp. with LTO like stuff. > Another issue is that you can't enable it on a lot of existing > libraries, sometimes not even with a recompile. For example > glibc assembler functions do not support it at all, which > is a very common case. They're mostly all leaf functions, so it doesn't matter much if anything. > They are designed to use dwarf, but in practice dwarf > is very slow (perf has to save the stack for every sample) > and in practice doesn't always work (too small stack saving, > wrong annotations, out of date or broken dwarf library etc.) > > LBR callstack mode is not perfect either, and it has > its own tradeoffs, but in many cases it seems to be a good > and more efficient replacement for dwarf, when FP is not available. But except for the lobbying Intel put into disabling FP because of that piece of shit Atom we'd all still have it enabled.