From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 14966 invoked by alias); 1 Dec 2004 21:15:23 -0000 Mailing-List: contact gcc-help@gcc.gnu.org; run by ezmlm Precedence: bulk List-Archive: List-Post: List-Help: Sender: gcc-owner@gcc.gnu.org Received: (qmail 14900 invoked from network); 1 Dec 2004 21:15:18 -0000 Received: from unknown (HELO vortex.ices.utexas.edu) (128.83.68.102) by sourceware.org with SMTP; 1 Dec 2004 21:15:18 -0000 Received: from terra.ices.utexas.edu (terra.ices.utexas.edu [128.83.68.97]) by vortex.ices.utexas.edu (8.12.8/8.12.9) with ESMTP id iB1LFG9O026317; Wed, 1 Dec 2004 15:15:16 -0600 Received: by terra.ices.utexas.edu (Postfix, from userid 364) id C90D82FE99; Wed, 1 Dec 2004 15:15:16 -0600 (CST) From: Wolfgang Bangerth To: Richard Henderson Subject: Re: Mainline bootstrap failure in toplev.c Date: Wed, 01 Dec 2004 21:15:00 -0000 User-Agent: KMail/1.6.1 Cc: Zack Weinberg , Andrew Pinski , gcc@gcc.gnu.org, rth@gcc.gnu.org References: <200412010932.02561.bangerth@ices.utexas.edu> <200412011412.30688.bangerth@ices.utexas.edu> <20041201203820.GA13841@redhat.com> In-Reply-To: <20041201203820.GA13841@redhat.com> MIME-Version: 1.0 Content-Disposition: inline Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Message-Id: <200412011515.16530.bangerth@ices.utexas.edu> X-SW-Source: 2004-12/txt/msg00091.txt.bz2 > Can you please find out? =A0Whatever it is, *that's* the place we > probably need to fixinclude. This happens only when compiled with stage1/xgcc. And the reason is a littl= e=20 more complicated than I had thought: in /usr/include/linux/compiler.h, we=20 have #if __GNUC__ > 3 # include /* catch-all for GCC 4, 5, etc. */ #elif __GNUC__ =3D=3D 3 # include For mainline, we have __GNUC__=3D=3D4, so we include the second include fil= e in=20 the first stage, and the first include file when compiling with stage1/xgcc= .=20 Now, in /usr/include/compiler-gcc+.h we find the evil: #ifndef __KERNEL__ #define inline __inline__ __attribute__((always_inline)) #define __inline__ __inline__ __attribute__((always_inline)) #define __inline __inline__ __attribute__((always_inline)) #endif Note *ifndef*, whereas in /usr/include/compiler-gcc3.h we had the exact=20 opposite condition of whether __KERNEL__ is defined. I would claim that thi= s=20 is a genuine typo/bug in the kernel headers. W. ------------------------------------------------------------------------- Wolfgang Bangerth email: bangerth@ices.utexas.edu www: http://www.ices.utexas.edu/~bangerth/