From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 23937 invoked by alias); 14 Nov 2007 22:51:37 -0000 Received: (qmail 23928 invoked by uid 22791); 14 Nov 2007 22:51:36 -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 22:51:33 +0000 Received: (qmail 28959 invoked from network); 14 Nov 2007 22:51:31 -0000 Received: from unknown (HELO localhost) (jimb@127.0.0.2) by mail.codesourcery.com with ESMTPA; 14 Nov 2007 22:51:31 -0000 To: Rob Quill Cc: gdb@sourceware.org Subject: Re: Removing deprecated breakpoint functions References: <20071114215851.GA3539@caradoc.them.org> From: Jim Blandy Date: Wed, 14 Nov 2007 22:51:00 -0000 In-Reply-To: <20071114215851.GA3539@caradoc.them.org> (Daniel Jacobowitz's message of "Wed, 14 Nov 2007 16:58:51 -0500") 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-help@sourceware.org; run by ezmlm Precedence: bulk List-Id: List-Subscribe: List-Archive: List-Post: List-Help: , Sender: gdb-owner@sourceware.org X-SW-Source: 2007-11/txt/msg00150.txt.bz2 Daniel Jacobowitz writes: > On Wed, Nov 14, 2007 at 01:45:35PM -0800, Jim Blandy wrote: >> >> "Rob Quill" writes: >> > I am trying to figure out how to remove >> > deprecated_insert/remove_raw_breakpoint and was wondering if it is >> > acceptable to substitute a call to them with a call to the >> > break_command() function as this will handle all the breakpoint chain >> > related things, or is this function too high level and another one >> > should be used? >> >> Yes, those are too high-level; I think 'set_raw_breakpoint' is more >> what you're looking for. See the uses in breakpoint.c, things like >> 'create_fork_event_catchpoint' might serve as helpful examples. > > The breakpoints that use deprecated_insert_raw_breakpoint do not > currently go on the breakpoint chain. The most important of these > is the software single-step breakpoint. Nothing is really prepared > to find it on the chain (what priority does it have compared to other > breakpoints, for instance) - so this is going to be a bit tricky. So, bpstat_what and its relatives will need to learn about them?