From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 7409 invoked by alias); 20 Oct 2008 01:46:42 -0000 Received: (qmail 7401 invoked by uid 22791); 20 Oct 2008 01:46:42 -0000 X-Spam-Check-By: sourceware.org Received: from NaN.false.org (HELO nan.false.org) (208.75.86.248) by sourceware.org (qpsmtpd/0.31) with ESMTP; Mon, 20 Oct 2008 01:46:01 +0000 Received: from nan.false.org (localhost [127.0.0.1]) by nan.false.org (Postfix) with ESMTP id 72FCD10D2D; Mon, 20 Oct 2008 01:45:58 +0000 (GMT) Received: from caradoc.them.org (209.195.188.212.nauticom.net [209.195.188.212]) by nan.false.org (Postfix) with ESMTP id 4E4AB10CF0; Mon, 20 Oct 2008 01:45:58 +0000 (GMT) Received: from drow by caradoc.them.org with local (Exim 4.69) (envelope-from ) id 1KrjqH-0006Pm-FH; Sun, 19 Oct 2008 21:45:57 -0400 Date: Mon, 20 Oct 2008 01:46:00 -0000 From: Daniel Jacobowitz To: Michael Snyder Cc: Pedro Alves , "gdb-patches@sourceware.org" , teawater Subject: Re: [reverse/record] adjust_pc_after_break in reverse execution mode? Message-ID: <20081020014557.GA24636@caradoc.them.org> Mail-Followup-To: Michael Snyder , Pedro Alves , "gdb-patches@sourceware.org" , teawater References: <200810180210.16346.pedro@codesourcery.com> <48FBB718.4040706@vmware.com> <200810200109.55661.pedro@codesourcery.com> <48FBD342.5070905@vmware.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <48FBD342.5070905@vmware.com> User-Agent: Mutt/1.5.17 (2008-05-11) 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: 2008-10/txt/msg00482.txt.bz2 On Sun, Oct 19, 2008 at 05:39:30PM -0700, Michael Snyder wrote: > See, the problem is that "adjust_pc_after_break" is assuming > memory-breakpoint semantics, but Process Record/Replay actually > implements hardware-breakpoint semantics. It watches the > instruction-address "bus" and stops when the PC matches the > address of a breakpoint. Don't x86 hardware breakpoints behave the same as x86 software breakpoints in this regard? I'd suggest the replay target make this as simple as possible for the rest of GDB: increment the PC by decr_pc_after_break. Or is that going to cause some other problem? -- Daniel Jacobowitz CodeSourcery