From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:48650) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1cnobn-0005mB-NZ for qemu-devel@nongnu.org; Tue, 14 Mar 2017 11:43:36 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1cnobi-0004vg-T6 for qemu-devel@nongnu.org; Tue, 14 Mar 2017 11:43:35 -0400 Received: from mail-wr0-x233.google.com ([2a00:1450:400c:c0c::233]:36841) by eggs.gnu.org with esmtps (TLS1.0:RSA_AES_128_CBC_SHA1:16) (Exim 4.71) (envelope-from ) id 1cnobi-0004uk-Ms for qemu-devel@nongnu.org; Tue, 14 Mar 2017 11:43:30 -0400 Received: by mail-wr0-x233.google.com with SMTP id u108so126632977wrb.3 for ; Tue, 14 Mar 2017 08:43:30 -0700 (PDT) References: <20170303131113.25898-1-pbonzini@redhat.com> <20170303131113.25898-6-pbonzini@redhat.com> <871su1qfsu.fsf@linaro.org> <834de521-e5cc-8857-afeb-8734037f95f3@redhat.com> <87y3w9oxff.fsf@linaro.org> <2d503aaa-4853-a325-bc13-983528d02e09@redhat.com> From: Alex =?utf-8?Q?Benn=C3=A9e?= In-reply-to: Date: Tue, 14 Mar 2017 15:43:45 +0000 Message-ID: <87r31zq2xa.fsf@linaro.org> MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 8bit Subject: Re: [Qemu-devel] [PATCH 5/5] icount: process QEMU_CLOCK_VIRTUAL timers in vCPU thread List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Paolo Bonzini Cc: qemu-devel@nongnu.org Paolo Bonzini writes: > On 14/03/2017 11:05, Paolo Bonzini wrote: >> >> >> On 13/03/2017 19:15, Alex Bennée wrote: >>>> Well yeah, it's all those that pass qemu_clock_use_for_deadline. >>> >>> Have you done any testing with record/replay? So far I have one >>> reproducible run and one failure. However it is not entirely clear to me >>> how I am meant to cleanly halt and stop a machine so I don't corrupt the >>> replay log. >> >> No, I haven't tried RR. > > Tried now, it doesn't work but I won't have time to fix it. > > I tried "-kernel /path/to/bzImage -icount 3,rr=record,rrfile=replay.bin > -net none". record works, replay hangs. The hang is at a "pause" > instruction. With "-d in_asm" it still hangs, but a bit later (still in > the firmware). I'm happy to keep digging. So to be clear when you do the record step how do you save your state before you replay? I was just entering the MMI and typing quit but that seems to leave the log in a broken state. -- Alex Bennée