From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 2938 invoked by alias); 23 Apr 2002 23:30:07 -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 2878 invoked from network); 23 Apr 2002 23:30:01 -0000 Received: from unknown (HELO gandalf.codesourcery.com) (66.60.148.227) by sources.redhat.com with SMTP; 23 Apr 2002 23:30:01 -0000 Received: from gandalf.codesourcery.com (localhost.localdomain [127.0.0.1]) by gandalf.codesourcery.com (8.11.6/8.11.6) with ESMTP id g3NNR7L07107; Tue, 23 Apr 2002 16:27:07 -0700 Date: Tue, 23 Apr 2002 16:53:00 -0000 From: Mark Mitchell To: Richard Henderson cc: "gcc@gcc.gnu.org" , Jason Merrill Subject: Re: mips n64 eh failures Message-ID: <43820000.1019604427@gandalf.codesourcery.com> In-Reply-To: <20020423161543.B939@redhat.com> References: <14800000.1019552485@gandalf.codesourcery.com> <20020423122254.A775@redhat.com> <29950000.1019593338@gandalf.codesourcery.com> <20020423132840.A823@redhat.com> <32140000.1019594105@gandalf.codesourcery.com> <20020423161543.B939@redhat.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit Content-Disposition: inline X-SW-Source: 2002-04/txt/msg01212.txt.bz2 --On Tuesday, April 23, 2002 04:15:43 PM -0700 Richard Henderson wrote: > On Tue, Apr 23, 2002 at 01:35:05PM -0700, Mark Mitchell wrote: >> I meant in the unwinder to try to see what it thought it was doing. > > Hum. Your advice to try printf was sound. It wasn't getting > as far as I thought. In fact, the entire problem appears to > be in the fde sorting routines. Ooooohhh. That gave me some horrible flashback from GCC 2.95/GCC 3.0 where I remember trying to debug something that sounds very much like this on IRIX. The weird thing was that it seemed semi-transient; I think it went away before I ever figured out what it was. I remember thinking that there was a miscompilation of one of the fde-sorting routines, but I don't remember much more than that. I remember thinking it could be some kind of alignment issue; that depending on where things happen to get put in memory something ends up 64-bit aligned (good!) or not (bad!). -- Mark Mitchell mark@codesourcery.com CodeSourcery, LLC http://www.codesourcery.com