From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 17553 invoked by alias); 2 Jan 2012 14:53:13 -0000 Received: (qmail 17544 invoked by uid 22791); 2 Jan 2012 14:53:12 -0000 X-SWARE-Spam-Status: No, hits=-7.3 required=5.0 tests=AWL,BAYES_00,RCVD_IN_DNSWL_HI,RP_MATCHES_RCVD,SPF_HELO_PASS X-Spam-Check-By: sourceware.org Received: from mx1.redhat.com (HELO mx1.redhat.com) (209.132.183.28) by sourceware.org (qpsmtpd/0.43rc1) with ESMTP; Mon, 02 Jan 2012 14:53:00 +0000 Received: from int-mx09.intmail.prod.int.phx2.redhat.com (int-mx09.intmail.prod.int.phx2.redhat.com [10.5.11.22]) by mx1.redhat.com (8.14.4/8.14.4) with ESMTP id q02Eqxrg028699 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=OK); Mon, 2 Jan 2012 09:52:59 -0500 Received: from host2.jankratochvil.net (ovpn-116-32.ams2.redhat.com [10.36.116.32]) by int-mx09.intmail.prod.int.phx2.redhat.com (8.14.4/8.14.4) with ESMTP id q02Eqtjv016035 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES128-SHA bits=128 verify=NO); Mon, 2 Jan 2012 09:52:58 -0500 Date: Mon, 02 Jan 2012 14:53:00 -0000 From: Jan Kratochvil To: Pedro Alves Cc: gdb-patches@sourceware.org Subject: Re: [patch] Fix gdb.cp/gdb2495.exp regression with gcc-4.7 Message-ID: <20120102145255.GA20385@host2.jankratochvil.net> References: <20111222202047.GA16110@host2.jankratochvil.net> <4F01BAA6.8000004@gmail.com> <20120102141937.GA14138@host2.jankratochvil.net> <4F01C2B2.7010808@gmail.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <4F01C2B2.7010808@gmail.com> User-Agent: Mutt/1.5.21 (2010-09-15) X-IsSubscribed: yes Mailing-List: contact gdb-patches-help@sourceware.org; run by ezmlm Precedence: bulk List-Id: List-Subscribe: List-Archive: List-Post: List-Help: , Sender: gdb-patches-owner@sourceware.org X-SW-Source: 2012-01/txt/msg00039.txt.bz2 On Mon, 02 Jan 2012 15:44:02 +0100, Pedro Alves wrote: > I see. That '_Unwind_IsSignalFrame(context)' is there to cancel out > the '- 1' for signal frames. Ideally, the same treatment would be > applied for gdb's dummy frame, perhaps by having gdb itself change > `context->flags' making libunwind treat it as a signal frame. I > have no idea whether that's feasible. Oh well. It needs to be universal, at least for both libgcc and libunwind, via the _Unwind_* API. And this signal frame indication (such as 'S' in .eh_frame CIE augmentation) would need to be present in the frame for the called code. The caller frame is already too late, that one gets already wrongly unwound. I do not find it feasible myself. Thanks, Jan