From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:38619) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1b2yqo-0002e2-FA for qemu-devel@nongnu.org; Wed, 18 May 2016 06:37:15 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1b2yqk-0004yI-AY for qemu-devel@nongnu.org; Wed, 18 May 2016 06:37:13 -0400 Received: from mail-wm0-x22b.google.com ([2a00:1450:400c:c09::22b]:37977) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1b2yqk-0004y2-2h for qemu-devel@nongnu.org; Wed, 18 May 2016 06:37:10 -0400 Received: by mail-wm0-x22b.google.com with SMTP id g17so72357462wme.1 for ; Wed, 18 May 2016 03:37:09 -0700 (PDT) Sender: Paolo Bonzini References: <1463496377-9729-1-git-send-email-cornelia.huck@de.ibm.com> <1463496377-9729-10-git-send-email-cornelia.huck@de.ibm.com> <573C3F45.6050203@de.ibm.com> From: Paolo Bonzini Message-ID: <1a8d7e53-1a10-158e-30f5-e687b5fb294b@redhat.com> Date: Wed, 18 May 2016 12:37:04 +0200 MIME-Version: 1.0 In-Reply-To: <573C3F45.6050203@de.ibm.com> Content-Type: text/plain; charset=windows-1252 Content-Transfer-Encoding: 7bit Subject: Re: [Qemu-devel] [PULL 09/20] s390x/ipl: Remove redundant usage of gr7 List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Christian Borntraeger , Cornelia Huck , peter.maydell@linaro.org Cc: qemu-devel@nongnu.org, jfrei@linux.vnet.ibm.com, agraf@suse.de, Alexander Yarygin On 18/05/2016 12:09, Christian Borntraeger wrote: >> > This probably breaks reboot after migration from 2.6. >> > >> > If that's fine, all the better. :) > Our BIOS is in RAM and usually overwritten by the guest. > On reboot this is reloaded from whatever BIOS is on the target, > so migration should still work. Indeed if the BIOS is not in its own RAMBlock then it's fine. Thanks! Paolo > I have verified that migration between 2.4 and 2.6 plus these > patches forth and back still allows to reboot. > > In other words the bios and the code in ipl.c have to be updated > in lockstep, then everything is fine. > > >