From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754214Ab3JRNNc (ORCPT ); Fri, 18 Oct 2013 09:13:32 -0400 Received: from mail-ee0-f53.google.com ([74.125.83.53]:35190 "EHLO mail-ee0-f53.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752512Ab3JRNNa (ORCPT ); Fri, 18 Oct 2013 09:13:30 -0400 Date: Fri, 18 Oct 2013 15:13:26 +0200 From: Ingo Molnar To: Tetsuo Handa Cc: peterz@infradead.org, maarten.lankhorst@canonical.com, imirkin@alum.mit.edu, linux-kernel@vger.kernel.org, daniel.vetter@ffwll.ch, robdclark@gmail.com Subject: Re: [PATCH for 3.12] mutex: Avoid gcc version dependent__builtin_constant_p() usage. Message-ID: <20131018131326.GA24572@gmail.com> References: <201310171945.AGB17114.FSQVtHOJFOOFML@I-love.SAKURA.ne.jp> <20131017105243.GM2675@laptop.programming.kicks-ass.net> <525FC1DD.4030205@canonical.com> <20131017110215.GN2675@laptop.programming.kicks-ass.net> <201310182151.AGF16889.SQOFJOtVLFFMHO@I-love.SAKURA.ne.jp> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <201310182151.AGF16889.SQOFJOtVLFFMHO@I-love.SAKURA.ne.jp> 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 * Tetsuo Handa wrote: > Peter Zijlstra wrote: > > On Thu, Oct 17, 2013 at 12:54:21PM +0200, Maarten Lankhorst wrote: > > > op 17-10-13 12:52, Peter Zijlstra schreef: > > > > On Thu, Oct 17, 2013 at 07:45:29PM +0900, Tetsuo Handa wrote: > > > >> Commit 040a0a37 "mutex: Add support for wound/wait style locks" used > > > >> "!__builtin_constant_p(p == NULL)" but gcc 3.x cannot handle such expression > > > >> correctly, leading to boot failure when built with CONFIG_DEBUG_MUTEXES=y. > > > > So I completely forgot all about this, but wouldn't something like: > > > > > > > > !(__builtin_constant_p(p) && p == NULL) > > > > > > > > Not also work and generate the same code? > > > > > > > See earlier discussion. It was already answered why that was undesirable. ;) > > > > OK; I forgot if we covered that particular option.. I so hate this > > patch; but I'm afraid we'll have to take it :-/ > > > > I see. Ingo, would you send this patch via your tree? > Then, I'll post a patch for 3.11-stable which differs white spaces. If it has Maarten's and PeterZ's Acked-by then sure. Thanks, Ingo