From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 28324 invoked by alias); 14 Nov 2007 19:21:06 -0000 Received: (qmail 28308 invoked by uid 22791); 14 Nov 2007 19:21:05 -0000 X-Spam-Check-By: sourceware.org Received: from mail.codesourcery.com (HELO mail.codesourcery.com) (65.74.133.4) by sourceware.org (qpsmtpd/0.31) with ESMTP; Wed, 14 Nov 2007 19:21:01 +0000 Received: (qmail 18079 invoked from network); 14 Nov 2007 19:20:58 -0000 Received: from unknown (HELO localhost) (jimb@127.0.0.2) by mail.codesourcery.com with ESMTPA; 14 Nov 2007 19:20:58 -0000 To: Vladimir Prus Cc: gdb-patches@sources.redhat.com Subject: Re: [8/9] multiple locations References: <18233.20228.172834.464875@kahikatea.snap.net.nz> <200711131105.10906.vladimir@codesourcery.com> <20071113123812.GA22747@caradoc.them.org> From: Jim Blandy Date: Wed, 14 Nov 2007 19:21:00 -0000 In-Reply-To: (Vladimir Prus's message of "Tue, 13 Nov 2007 22:40:57 +0300") Message-ID: User-Agent: Gnus/5.11 (Gnus v5.11) Emacs/22.0.50 (gnu/linux) MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii 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: 2007-11/txt/msg00277.txt.bz2 Vladimir Prus writes: > Daniel Jacobowitz wrote: > >> On Tue, Nov 13, 2007 at 11:05:10AM +0300, Vladimir Prus wrote: >>> > >>> > Does the (p) add anything useful to the user? The manual says: >>> > >>> > An optional `(p)' suffix marks pending breakpoints >>> > >>> > Is that not clear from the word PENDING? >>> >>> No, because if you set breakpoint in a shared library, and that library >>> is unloaded, you have some number in 'address' field, but the breakpoint >>> won't actually fire, and the "(p)" indicates that fact. >> >> Can we just change it back to by discarding the address? > > This can actually work. I'll try to implement this idea. Yeah, that's what I was going to suggest, too --- the number is basically a dangling pointer once the library's been unloaded. I agree with Vladimir that the priority of the CLI should be providing a good command-line experience --- not stability for code trying to parse the output. MI has been in GDB in some form or another since 2001.