From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 10532 invoked by alias); 7 Jan 2004 18:03:45 -0000 Mailing-List: contact insight-help@sources.redhat.com; run by ezmlm Precedence: bulk List-Subscribe: List-Archive: List-Post: List-Help: , Sender: insight-owner@sources.redhat.com Received: (qmail 10525 invoked from network); 7 Jan 2004 18:03:42 -0000 Received: from unknown (HELO bowmore.compxs.com) (194.153.106.142) by sources.redhat.com with SMTP; 7 Jan 2004 18:03:42 -0000 Received: from Kindrogan ([192.168.189.6]) by bowmore.compxs.com (8.12.5/8.12.5) with ESMTP id i07ICbcW000908 for ; Wed, 7 Jan 2004 18:12:45 GMT Reply-To: From: "Jon Beniston" To: Subject: h/w breakpoints Date: Wed, 07 Jan 2004 18:03:00 -0000 Organization: CompXs Message-ID: <004301c3d548$925c7400$06bda8c0@Kindrogan> MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit X-SW-Source: 2004-q1/txt/msg00003.txt.bz2 Hi, I'm am trying to debug a remote target that only has h/w breakpoints available (the code is in ROM). Is this known to be handled well by Insight? Currently I have to set all breakpoints via the command line. Looking at the remote debug output (set debug remote 1), I can see that only Z0, X and M commands (all which write memory) are sent to the target when I set a breakpoint via Insight. Should it not try Z1 as well if all the previous fail? The fact that s/w breakpoints don't get set has the nasty side effect of causing continue not to do anything without reporting any error (removing all s/w breakpoints allows it to resume). Another downside of using the hbreak command is that the breakpoints don't appear in the GUI. Now I could hack my debug target code to write a h/w breakpoint when receiving a Z0, but feel that this should probably be handled better by Insight. I'm willing to have a crack at modifying this if deemed appropriate. Cheers, Jon