From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753165AbZKHPot (ORCPT ); Sun, 8 Nov 2009 10:44:49 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1752407AbZKHPot (ORCPT ); Sun, 8 Nov 2009 10:44:49 -0500 Received: from borg.medozas.de ([188.40.89.202]:48632 "EHLO borg.medozas.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751717AbZKHPos (ORCPT ); Sun, 8 Nov 2009 10:44:48 -0500 Date: Sun, 8 Nov 2009 16:44:53 +0100 (CET) From: Jan Engelhardt To: =?UTF-8?Q?Andr=C3=A9_Goddard_Rosa?= cc: Linux Kernel Mailing List Subject: Re: [PATCH v4 07/12] vsprintf: factor out skip_space code in a separate function In-Reply-To: <7206ef594e67a240a842339f520284de6569b1fc.1257602781.git.andre.goddard@gmail.com> Message-ID: References: <7206ef594e67a240a842339f520284de6569b1fc.1257602781.git.andre.goddard@gmail.com> User-Agent: Alpine 2.00 (LSU 1167 2008-08-23) MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=UTF-8 Content-Transfer-Encoding: 8BIT Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Saturday 2009-11-07 16:16, André Goddard Rosa wrote: >It decreases code size: And may increase runtime. If GCC determined that skip_space itself (without the noinline you added) is going to be inlined, let it go ahead. > text data bss dec hex filename > 15719 0 8 15727 3d6f lib/vsprintf.o-before > 15543 0 8 15551 3cbf lib/vsprintf.o-after