From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 25792 invoked by alias); 28 Dec 2011 15:57:33 -0000 Received: (qmail 25783 invoked by uid 22791); 28 Dec 2011 15:57:31 -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; Wed, 28 Dec 2011 15:57:14 +0000 Received: from int-mx12.intmail.prod.int.phx2.redhat.com (int-mx12.intmail.prod.int.phx2.redhat.com [10.5.11.25]) by mx1.redhat.com (8.14.4/8.14.4) with ESMTP id pBSFv1Ax021518 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=OK); Wed, 28 Dec 2011 10:57:01 -0500 Received: from host2.jankratochvil.net (ovpn-116-32.ams2.redhat.com [10.36.116.32]) by int-mx12.intmail.prod.int.phx2.redhat.com (8.14.4/8.14.4) with ESMTP id pBSFuv2W014432 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES128-SHA bits=128 verify=NO); Wed, 28 Dec 2011 10:57:00 -0500 Date: Wed, 28 Dec 2011 16:00:00 -0000 From: Jan Kratochvil To: Joel Brobecker Cc: Paul Hilfinger , gdb-patches@sourceware.org Subject: Re: [RFA] Have block_innermost_frame start from selected frame Message-ID: <20111228155657.GA10556@host2.jankratochvil.net> References: <20111227195809.672D892BF6@kwai.gnat.com> <20111228130130.GA1855@host2.jankratochvil.net> <20111228153008.GN23376@adacore.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20111228153008.GN23376@adacore.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: 2011-12/txt/msg00872.txt.bz2 On Wed, 28 Dec 2011 16:30:08 +0100, Joel Brobecker wrote: > What do you mean by ambiguous? Is it the case described here, where > we have multiple frames for the given FUNCTION? Yes. Depending on which recursive function frame you select the returned value is different. > I don't think we should query(). In fact we should menu-select them but that is offtopic now. > Users often respond negatively to query, particularly in a case like this > where we can make the behavior unambiguous and easily describable. I find that GDB should ask or at least warn more in general. Like that if you debug -O2 code it should warn you on first `step' that it will not work much. If you debug program/function without -g it should warn you (not just "no debugging symbols found" lost in the screens of "Reading symbols from " messages). It would prevent false accuses of GDB due to bugs elsewhere. GIT gives nice messages suggesting what to do in any case of a problem. > It's the only way to get the value of "var" in our testcase, and you cannot > currently do it with the old behavior. You can already do so many things with GDB, just people do not do even 5% of them because it is all too magic to learn. > I think simple is good enough, in this particular case. I am not going to try to block it but it is still the current GDB style closed to hackers. Thanks, Jan