From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <gcc-return-61436-listarch-gcc=gcc.gnu.org@gcc.gnu.org>
Received: (qmail 12052 invoked by alias); 13 Oct 2002 03:08:57 -0000
Mailing-List: contact gcc-help@gcc.gnu.org; run by ezmlm
Precedence: bulk
List-Archive: <http://gcc.gnu.org/ml/gcc/>
List-Post: <mailto:gcc@gcc.gnu.org>
List-Help: <http://gcc.gnu.org/ml/>
Sender: gcc-owner@gcc.gnu.org
Received: (qmail 12040 invoked from network); 13 Oct 2002 03:08:56 -0000
Received: from unknown (HELO caip.rutgers.edu) (128.6.236.10)
  by sources.redhat.com with SMTP; 13 Oct 2002 03:08:56 -0000
Received: (from ghazi@localhost)
	by caip.rutgers.edu (8.9.3/8.9.3) id XAA13661;
	Sat, 12 Oct 2002 23:08:12 -0400 (EDT)
Date: Sun, 13 Oct 2002 05:12:00 -0000
From: "Kaveh R. Ghazi" <ghazi@caip.rutgers.edu>
Message-Id: <200210130308.XAA13661@caip.rutgers.edu>
To: doko@cs.tu-berlin.de
Subject: Re: gcc-3_2-branch bootstrap failure when using bison-1.50
Cc: debian-gcc@lists.debian.org, gcc-bugs@gcc.gnu.org, gcc@gcc.gnu.org,
        mark@codesourcery.com
X-SW-Source: 2002-10/txt/msg00660.txt.bz2

 > From: Matthias Klose <doko at cs dot tu-berlin dot de> 
 > 
 > Bootrapping the gcc-3_2-branch using bison-1.50 is broken. Reverting
 > back to bison-1.35 works. HEAD does work well using bison-1.50
 > 
 > cd /build/gcc/gcc-3.2-3.2.1ds3/src/gcc && \
 > if bison  -o c-p$$.c c-parse.y; then \
 >   test -f c-p$$.output && mv -f c-p$$.output c-parse.output ; \
 >   mv -f c-p$$.c c-parse.c ; \
 > else \
 >   rm -f c-p$$.* ; \
 >   false ; \
 > fi
 > c-parse.y:1432.2-1436.10: type clash (`' `ttype') on default action

I checked into this one, patches were submitted for the trunk and
installed back in June for these problems.  E.g. see:
http://gcc.gnu.org/ml/gcc-patches/2002-06/msg01614.html

I guess this and other similar patches for the other .y files weren't
backported to the 3.2 branch.  This shouldn't be too bad since the
actual GCC-3.2.x release tar file will contain the necessary bison
output of a suitable bison version.  In the mean time, I suggest just
not using bison-1.50 for testing 3.2.x CVS snapshots.

I'll leave it to our release manager to decide if this issue warrants
backporting the relevant patches or not.

	Thanks for reporting this,
		--Kaveh
--
Kaveh R. Ghazi			ghazi@caip.rutgers.edu