Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S261921AbVCaVG5 (ORCPT ); Thu, 31 Mar 2005 16:06:57 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S261943AbVCaVG5 (ORCPT ); Thu, 31 Mar 2005 16:06:57 -0500 Received: from webmail.topspin.com ([12.162.17.3]:57723 "EHLO exch-1.topspincom.com") by vger.kernel.org with ESMTP id S261936AbVCaVEw (ORCPT ); Thu, 31 Mar 2005 16:04:52 -0500 To: Adrian Bunk Cc: =?iso-8859-1?q?J=F6rn_Engel?= , Yum Rayan , linux-kernel@vger.kernel.org, mvw@planets.elm.net Subject: Re: Stack usage tasks X-Message-Flag: Warning: May contain useful information References: <20050331150548.GC19294@wohnheim.fh-wedel.de> <20050331203010.GF3185@stusta.de> From: Roland Dreier Date: Thu, 31 Mar 2005 12:43:38 -0800 In-Reply-To: <20050331203010.GF3185@stusta.de> (Adrian Bunk's message of "Thu, 31 Mar 2005 22:30:10 +0200") Message-ID: <52ll83mtqd.fsf@topspin.com> User-Agent: Gnus/5.1006 (Gnus v5.10.6) XEmacs/21.4 (Jumbo Shrimp, linux) MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-OriginalArrivalTime: 31 Mar 2005 20:43:38.0731 (UTC) FILETIME=[512103B0:01C53632] Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 836 Lines: 19 > The task I'm suggesting was therefore: > - remove the -fno-unit-at-a-time in arch/i386/Makefile in your private > kernel sources > - use gcc 3.4 > - reduce the stack usages in call paths > 3kB This is a good idea. However, I might suggest using gcc 4.0 (you'll have to use a snapshot now, but the release should only be a few weeks away). A patch went into gcc 4.0 that makes gcc more intelligent about sharing stack for variables that cannot be alive at the same time, and therefore it may be more feasible to make unit-at-a-time work for the i386 kernels. - R. - 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/