From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 24683 invoked by alias); 14 May 2003 08:06:01 -0000 Mailing-List: contact gcc-prs-help@gcc.gnu.org; run by ezmlm Precedence: bulk List-Archive: List-Post: List-Help: Sender: gcc-prs-owner@gcc.gnu.org Received: (qmail 24664 invoked by uid 71); 14 May 2003 08:06:01 -0000 Date: Wed, 14 May 2003 08:06:00 -0000 Message-ID: <20030514080601.24663.qmail@sources.redhat.com> To: nobody@gcc.gnu.org Cc: gcc-prs@gcc.gnu.org, From: Ludovic Brenta Subject: Re: bootstrap/10740: ../../gcc/gcc/gengtype.c:430: undefined reference to "lexer_line" Reply-To: Ludovic Brenta X-SW-Source: 2003-05/txt/msg01548.txt.bz2 List-Id: The following reply was made to PR bootstrap/10740; it has been noted by GNATS. From: Ludovic Brenta To: Dara Hazeghi Cc: gcc-gnats@gcc.gnu.org Subject: Re: bootstrap/10740: ../../gcc/gcc/gengtype.c:430: undefined reference to "lexer_line" Date: 14 May 2003 09:59:59 +0200 Yes, it is a problem of lingering files. I traced it to the file gcc/gcc/gengtype-lex.c which flex creates in the source, not object, directory. For one reason or another, this file ends up being empty, but newer than its source, which is gengtype-lex.l. Then, it gets compiled into an object file which lacks "lexer_line" and other symbols. At first I thought it was a binutils issue; then I thought it was a flex issue; not I'm not sure anymore. I think it is a combination of two problems: (1) the build scripts should create gengtype-lex.c in the object directory, which I erase completely before each bootstrap; and (2) flex should fail instead of creating an empty C file. I've had success this night bootstrapping GCC 3.3 after downgrading flex to 2.5.4a (stable) and erasing the C file. I'll try again with the testing version of flex (2.5.31) to see if it's a flex problem. BTW, the build scripts also create other files in the source directory; for example, gengtype-yacc.c (from gengtype-yacc.y), and some .info files in the ada subdirectory (from .texinfo files). This does not appear to create problems for me right now, but I see this as a latent bug. -- Ludovic Brenta.