From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([140.186.70.92]:46305) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1QBqcf-0005aA-6B for qemu-devel@nongnu.org; Mon, 18 Apr 2011 11:44:22 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1QBqce-00087s-11 for qemu-devel@nongnu.org; Mon, 18 Apr 2011 11:44:21 -0400 Received: from mail-gy0-f173.google.com ([209.85.160.173]:51651) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1QBqcd-00087o-Sd for qemu-devel@nongnu.org; Mon, 18 Apr 2011 11:44:19 -0400 Received: by gyg4 with SMTP id 4so1853913gyg.4 for ; Mon, 18 Apr 2011 08:44:19 -0700 (PDT) Message-ID: <4DAC5C4F.2080500@codemonkey.ws> Date: Mon, 18 Apr 2011 10:44:15 -0500 From: Anthony Liguori MIME-Version: 1.0 References: <4DAC54C1.6010006@codemonkey.ws> <4DAC5A65.6040405@siemens.com> In-Reply-To: <4DAC5A65.6040405@siemens.com> Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Subject: Re: [Qemu-devel] [PULL] Migration of misc devices List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Jan Kiszka Cc: Glauber Costa , Marcelo Tosatti , "qemu-devel@nongnu.org" , "quintela@redhat.com" On 04/18/2011 10:36 AM, Jan Kiszka wrote: > On 2011-04-18 17:12, Anthony Liguori wrote: >> On 04/17/2011 03:08 PM, Juan Quintela wrote: >>> >>> >>> The following changes since commit a0102082de4026833afbd2525e8a6320d1f92885: >>> >>> usb: fix spelling errors in usb-linux.c (2011-04-16 12:24:28 +0100) >>> >>> are available in the git repository at: >>> git://repo.or.cz/qemu/quintela.git vmstate-misc-devices-v2 >> >> Migration appears to be broken for me. It's not caused by this series >> but I'm surprised you didn't encounter it. Maybe it's my kernel version? >> >> My suspicion is that the cause of this is: >> >> commit 0ec329dab938e2d97d12a91f8ed15fec27b325e0 >> Author: Jan Kiszka >> Date: Mon Feb 7 12:19:26 2011 +0100 >> >> kvm: x86: Introduce kvmclock device to save/restore its state > > Can you prove your suspicion a bit by retrying with -cpu ...,-kvmclock? > Is your guest using kvmclock? Are you migrating from old to new or > between new versions? Heh, it turns out I was migrating from KVM to TCG :-) This changeset "breaks" that but I don't think I'd count this as a supported use-case so I don't count it as a regression. So ignore the noise here. Regards, Anthony Liguori > Jan >