From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 27418 invoked by alias); 2 May 2003 09:29:40 -0000 Mailing-List: contact gdb-patches-help@sources.redhat.com; run by ezmlm Precedence: bulk List-Subscribe: List-Archive: List-Post: List-Help: , Sender: gdb-patches-owner@sources.redhat.com Received: (qmail 27402 invoked from network); 2 May 2003 09:29:36 -0000 Received: from unknown (HELO walton.kettenis.dyndns.org) (62.163.169.212) by sources.redhat.com with SMTP; 2 May 2003 09:29:36 -0000 Received: from elgar.kettenis.dyndns.org (elgar.kettenis.dyndns.org [192.168.0.2]) by walton.kettenis.dyndns.org (8.12.6/8.12.5) with ESMTP id h429TSG3001455; Fri, 2 May 2003 11:29:28 +0200 (CEST) (envelope-from kettenis@elgar.kettenis.dyndns.org) Received: from elgar.kettenis.dyndns.org (localhost [127.0.0.1]) by elgar.kettenis.dyndns.org (8.12.6/8.12.6) with ESMTP id h429TSJl030637; Fri, 2 May 2003 11:29:28 +0200 (CEST) (envelope-from kettenis@elgar.kettenis.dyndns.org) Received: (from kettenis@localhost) by elgar.kettenis.dyndns.org (8.12.6/8.12.6/Submit) id h429TS9O030634; Fri, 2 May 2003 11:29:28 +0200 (CEST) Date: Fri, 02 May 2003 09:29:00 -0000 Message-Id: <200305020929.h429TS9O030634@elgar.kettenis.dyndns.org> From: Mark Kettenis To: ac131313@redhat.com CC: gdb-patches@sources.redhat.com In-reply-to: <3EB18FDE.9020106@redhat.com> (message from Andrew Cagney on Thu, 01 May 2003 17:21:34 -0400) Subject: Re: [i386newframe/PATCH] New i386newframe branch References: <200304191651.h3JGp1Kr004648@elgar.kettenis.dyndns.org> <3EADA9AC.8030607@redhat.com> <200305011839.h41IdPdc000324@elgar.kettenis.dyndns.org> <3EB18FDE.9020106@redhat.com> X-SW-Source: 2003-05/txt/msg00015.txt.bz2 Date: Thu, 01 May 2003 17:21:34 -0400 From: Andrew Cagney > > + return &i386_frame_unwind; > > +} > > Have you tried adding just the sigtramp unwinder? I should get just > that addition debugged regardless - it should make migrating other ISAs > easier. > > No I haven't. I don't think I can. The problem is that I simplified > the sigtramp unwinder a bit. However in order to do so, I need to be > able to trust the normal unwinder to unwind the stack pointer > correctly. Unfortunately the current code is a bit sloppy about this. Ah, yes, I know the feeling. Tighten the camshaft cover, and a push-rods slips. Put that back, and the .... Some times it's easier to just replace the engine. It is unfortunatly something of an all or nothing afair. I might still get curious though to see just what happens. I might have something else for you to toy with. I've a prototype of a DWARF2 CFI unwinder ready, which is working fine in my i386newframe branch. I still need to clean up several hacks, but I might be able to check it in soon. Mark