From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756533Ab1EJOr5 (ORCPT ); Tue, 10 May 2011 10:47:57 -0400 Received: from bombadil.infradead.org ([18.85.46.34]:40847 "EHLO bombadil.infradead.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1756510Ab1EJOrz (ORCPT ); Tue, 10 May 2011 10:47:55 -0400 Subject: Re: [PATCH v3 5/6] x86-64: Move vread_tsc into a new file with sensible options From: Peter Zijlstra To: Andy Lutomirski Cc: x86@kernel.org, linux-kernel@vger.kernel.org, Ingo Molnar , Andi Kleen , Linus Torvalds , "David S. Miller" , Eric Dumazet , Thomas Gleixner In-Reply-To: <84f1b1e345955d38309678d578ccdff9de6eca7b.1305035945.git.luto@mit.edu> References: <84f1b1e345955d38309678d578ccdff9de6eca7b.1305035945.git.luto@mit.edu> Content-Type: text/plain; charset="UTF-8" Date: Tue, 10 May 2011 16:36:02 +0200 Message-ID: <1305038162.2914.93.camel@laptop> Mime-Version: 1.0 X-Mailer: Evolution 2.30.3 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Tue, 2011-05-10 at 10:15 -0400, Andy Lutomirski wrote: > vread_tsc is short and hot, and it's userspace code so the usual > reasons to keep frame pointers around, enable -pg, and turn off > sibling calls don't apply. > I really don't like that, turning off frame pointers here means that if your profiler interrupt hits while in the vDSO the user-space backtrace is dead. Please leave the frame pointer in. Esp on x86_64 the cost of keeping the framepointer isn't much at all since it isn't nearly as register starved as i386. I'm fine with stripping -pg, that is indeed useless for the vDSO.