From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jan Kiszka Subject: Re: [PATCH 00/17] kvm-userspace: Fix and improve guest debugging and x86 debug registers Date: Tue, 07 Oct 2008 14:20:39 +0200 Message-ID: <48EB5417.7080903@siemens.com> References: <20081006091415.095241851@mchn012c.ww002.siemens.net> <48EB53A9.3060409@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Cc: kvm@vger.kernel.org To: Avi Kivity Return-path: Received: from gecko.sbs.de ([194.138.37.40]:18791 "EHLO gecko.sbs.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753148AbYJGMUz (ORCPT ); Tue, 7 Oct 2008 08:20:55 -0400 In-Reply-To: <48EB53A9.3060409@redhat.com> Sender: kvm-owner@vger.kernel.org List-ID: Avi Kivity wrote: > Jan Kiszka wrote: >> While still waiting on the required merge of the QEMU bits in this >> series (sigh...), I want to provide an update of my guest debugging and >> x86 debug register improvement patches. First comes the kvm-userspace >> part, kernel bits follow in a separate series. >> >> No new features since my last posting. But thanks to heavy internal use, >> I was able to identify and fix several tricky corner case (/wrt VMX). >> The changes are: >> - rebased on top of the QEMU base series >> - fixed single-stepping over STI and MOV SS/POP SS (VMX) >> - proper compat wrapping for set_debugreg >> - cleanup special_reload_dr7 >> - proper injection of soft exceptions like #BP (VMX) >> >> To summarize the contributions of this series (+ its related kernel >> bits): >> - fully functional guest debugging via gdbstub, >> including hardware breakpoints and watchpoints >> (pick up current gdb cvs to have hbreak via remote gdb) >> - (Almost) unlimited number of standard breakpoints >> - SMP guest debugging support >> - x86 debug registers support (makes guest's gdb and kgdb happy) >> >> The patches are in daily use for several moons here and have proven to >> be very helpful for tricky kernel debugging task. Specifically, >> reproducing and then tracking down certain races/deadlocks on SMP boxes >> is far more comfortable with KVM than on "real metal". >> > > I'm no expert on the qemu debugger, but apart for a few minor comments > (sent as replies to the relevant patches) this looks good to me. Thanks a lot for reviewing, specifically also the QEMU part! Will look into the details later, but they make sense on first glance. Jan -- Siemens AG, Corporate Technology, CT SE 2 Corporate Competence Center Embedded Linux