From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from [140.186.70.92] (port=41008 helo=eggs.gnu.org) by lists.gnu.org with esmtp (Exim 4.43) id 1PchgG-0006Hv-J1 for qemu-devel@nongnu.org; Tue, 11 Jan 2011 12:07:05 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1PchfR-0005Cp-VX for qemu-devel@nongnu.org; Tue, 11 Jan 2011 12:06:10 -0500 Received: from mx1.redhat.com ([209.132.183.28]:26549) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1PchfR-0005CN-M8 for qemu-devel@nongnu.org; Tue, 11 Jan 2011 12:05:57 -0500 Message-ID: <4D2C8DF3.9030205@redhat.com> Date: Tue, 11 Jan 2011 19:05:55 +0200 From: Avi Kivity MIME-Version: 1.0 Subject: Re: [Qemu-devel] Re: [PATCH 26/35] kvm: Eliminate KVMState arguments References: <4D2616D6.4080309@linux.vnet.ibm.com> <4D26D6CF.5070405@web.de> <4D27A16F.9030809@linux.vnet.ibm.com> <4D282489.90506@web.de> <4D2B6506.6070907@linux.vnet.ibm.com> <4D2B6845.7050809@web.de> <4D2B6ADD.4090505@codemonkey.ws> <4D2C1C5D.2050504@redhat.com> <4D2C6290.1060607@linux.vnet.ibm.com> <1EA102F5-B6C2-43BC-9493-0271B287FC18@suse.de> <4D2C649F.6080508@linux.vnet.ibm.com> <4D2C67C2.5080000@redhat.com> <4D2C6AFA.4040104@linux.vnet.ibm.com> <4D2C6FAB.3050209@redhat.com> <4D2C7353.2000008@linux.vnet.ibm.com> <4D2C793C.2070003@redhat.com> <4D2C7D8C.8070503@linux.vnet.ibm.com> <4D2C7F45.9060005@redhat.com> <4D2C84B6.3090104@linux.vnet.ibm.com> In-Reply-To: <4D2C84B6.3090104@linux.vnet.ibm.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit List-Id: qemu-devel.nongnu.org List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Anthony Liguori Cc: Marcelo Tosatti , Jan Kiszka , Alexander Graf , kvm@vger.kernel.org, qemu-devel@nongnu.org On 01/11/2011 06:26 PM, Anthony Liguori wrote: > >> Visible, yes, but not in live migration, or in 'info i8254', or >> similar. We can live migrate between qcow2 and qed (using block >> migration), we should be able to do the same for the two i8254 >> implementations. >> >> I'm not happy about separate implementations, but that's a minor >> details. We can change it 2n+1 times without anybody noticing. Not >> so about ABI stuff. >> >>> Imagine getting a sosreport that includes a dump of the device >>> tree. You really want to see something in there that tells you it's >>> an in-kernel PIT and not the userspace one. >> >> Sure. Not the device tree though. The command line would give all >> the information? > > Then it's a one off option. We really want as much info as possible > stored in the device tree. > >> >> Or 'info i8254' can say something about the implementation. I don't >> want to have the user say 'info i8254-kvm'. > > info doesn't take a qdev device so yes, it can show whatever we want > it to show. > It may be a qdev read-only attribute (and thus not migrated?) if we have to have it in qdev for some reason.