From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 22265 invoked by alias); 24 Apr 2005 22:36:47 -0000 Mailing-List: contact gdb-help@sources.redhat.com; run by ezmlm Precedence: bulk List-Subscribe: List-Archive: List-Post: List-Help: , Sender: gdb-owner@sources.redhat.com Received: (qmail 22194 invoked from network); 24 Apr 2005 22:36:40 -0000 Received: from unknown (HELO nevyn.them.org) (66.93.172.17) by sourceware.org with SMTP; 24 Apr 2005 22:36:40 -0000 Received: from drow by nevyn.them.org with local (Exim 4.50 #1 (Debian)) id 1DPpiS-0003ne-D4; Sun, 24 Apr 2005 18:36:40 -0400 Date: Sun, 24 Apr 2005 22:36:00 -0000 From: Daniel Jacobowitz To: Shaun Jackman Cc: gdb@sources.redhat.com Subject: Re: Watch point for unaligned stores Message-ID: <20050424223640.GD14252@nevyn.them.org> Mail-Followup-To: Shaun Jackman , gdb@sources.redhat.com References: <7f45d939050423132561dbe16c@mail.gmail.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <7f45d939050423132561dbe16c@mail.gmail.com> User-Agent: Mutt/1.5.8i X-SW-Source: 2005-04/txt/msg00168.txt.bz2 On Sat, Apr 23, 2005 at 01:25:24PM -0700, Shaun Jackman wrote: > I had a bug where the word at 0x8 was being clobbered by some bug in > my program. I set at a watchpoint (watch *(int*)8) but it was never > triggered even though the memory was modified. I found in the end that > the errant instruction was writing a word at address 0x9, which on the > ARM architecture modifies the word at 0x8 due to alignment. Could I > set a watchpoint that would have caught this bug, other than say > setting a watchpoint at all four addresses of the word? This sounds like a bug in your remote debugging stub. -- Daniel Jacobowitz CodeSourcery, LLC