From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 27407 invoked by alias); 20 May 2003 20:59:27 -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 27355 invoked from network); 20 May 2003 20:59:25 -0000 Received: from unknown (HELO caip.rutgers.edu) (128.6.236.10) by sources.redhat.com with SMTP; 20 May 2003 20:59:25 -0000 Received: (from ghazi@localhost) by caip.rutgers.edu (8.9.3/8.9.3) id QAA18111; Tue, 20 May 2003 16:59:23 -0400 (EDT) Date: Tue, 20 May 2003 21:06:00 -0000 From: "Kaveh R. Ghazi" Message-Id: <200305202059.QAA18111@caip.rutgers.edu> To: dj@redhat.com Subject: Re: failure building gcc-3.3 (broken libiberty/vsprintf.c or build?) Cc: crossgcc@sources.redhat.com, gcc-patches@gcc.gnu.org, gcc@gcc.gnu.org, peter@baradas.org References: <20030520053204.ADDB298982@baradas.org> <200305201609.h4KG9jR20778@greed.delorie.com> X-SW-Source: 2003-05/txt/msg01905.txt.bz2 > > gcc-3.3/lbiberty/vsprintf.c, indeed includes (which gcc > > supplies), so if gcc-3.3 has killed support for varargs,h, why is > > libiberty/vsprintf.c trying to include it. > > That code in libiberty is ancient. If it's broken now, that means > that whoever dropped support for varargs.h didn't test it well enough. FWIW, it appears to be fixed on mainline. We can backport vsprintf.c to 3.3.1 if you think it's worth fixing. Here's the diff: --- cvs-egcc-3.3/gcc/libiberty/vsprintf.c 2002-01-22 15:03:29.000000000 -0500 +++ cvs-egcc/gcc/libiberty/vsprintf.c 2003-04-15 16:34:40.516093000 -0400 @@ -26,9 +26,13 @@ the resulting executable to be covered b This exception does not however invalidate any other reasons why the executable file might be covered by the GNU General Public License. */ +#include +#ifdef ANSI_PROTOTYPES +#include +#else #include +#endif #include -#include #undef vsprintf #if defined _IOSTRG && defined _IOWRT