From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 22885 invoked by alias); 24 Jun 2008 03:12:02 -0000 Received: (qmail 22873 invoked by uid 22791); 24 Jun 2008 03:12:01 -0000 X-Spam-Check-By: sourceware.org Received: from bluesmobile.specifix.com (HELO bluesmobile.specifix.com) (216.129.118.140) by sourceware.org (qpsmtpd/0.31) with ESMTP; Tue, 24 Jun 2008 03:11:44 +0000 Received: from [127.0.0.1] (bluesmobile.specifix.com [216.129.118.140]) by bluesmobile.specifix.com (Postfix) with ESMTP id BF3403BD49; Mon, 23 Jun 2008 20:11:42 -0700 (PDT) Subject: Re: Turn software singlestep off? From: Michael Snyder To: Joel Brobecker Cc: gdb@sourceware.org In-Reply-To: <20080624025854.GA4484@adacore.com> References: <1214275796.3601.1193.camel@localhost.localdomain> <20080624025854.GA4484@adacore.com> Content-Type: text/plain Date: Tue, 24 Jun 2008 03:12:00 -0000 Message-Id: <1214277102.3601.1199.camel@localhost.localdomain> Mime-Version: 1.0 X-Mailer: Evolution 2.10.3 (2.10.3-7.fc7) Content-Transfer-Encoding: 7bit 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: 2008-06/txt/msg00234.txt.bz2 On Mon, 2008-06-23 at 22:58 -0400, Joel Brobecker wrote: > > I thought there used to be, but now I can't find it. > > I don't think there is such a switch either. I don't think it would > be very hard to implement, though... And it could actually be useful > on architectures such as Tru64 (I think) where we enable SW single step > because the kernel stepping doesn't work for threaded program. Such > a switch would probably give the user a chance to have better > performance when debugging a program that doesn't use threads... > It would be a pretty geeky switch, though, and I would be ready to > bet that the number of people knowing about it would remain very small. Thanks. Sounds about right. There could be other contexts where it would be useful. Imagine you're talking to a remote target, but you (gdb) don't really know what it is. It could be a simulator. Or a record/replay type of thing. Something that wouldn't have a problem with singlestep, even though the native architecture would. However, here's my real reason for asking: I'm trying to implement reverse debugging for mips. Reverse debugging and software singlestep don't mix very well. GDB would have to figure out where to put the singlestep breakpoint. Singlestepping thru a jump, you can do that going forward, but not going backward! ;-)