From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jan Kiszka Subject: Re: [PATCH 00/21] qemu-kvm: Hook cleanups and extended use of upstream code Date: Tue, 02 Feb 2010 12:19:56 +0100 Message-ID: <4B680A5C.7090102@siemens.com> References: <15E55176-5115-41A5-8F36-923FE9FC093E@suse.de> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Cc: Avi Kivity , Marcelo Tosatti , "kvm@vger.kernel.org" , Anthony Liguori , Glauber Costa , "qemu-devel@nongnu.org" To: Alexander Graf Return-path: Received: from goliath.siemens.de ([192.35.17.28]:15399 "EHLO goliath.siemens.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750736Ab0BBLU3 (ORCPT ); Tue, 2 Feb 2010 06:20:29 -0500 In-Reply-To: <15E55176-5115-41A5-8F36-923FE9FC093E@suse.de> Sender: kvm-owner@vger.kernel.org List-ID: Alexander Graf wrote: > On 02.02.2010, at 09:18, Jan Kiszka wrote: > >> Let's start with the overall stats: >> >> 31 files changed, 274 insertions(+), 822 deletions(-) >> >> So this series drops far more than 500 lines of redundant code, moving >> qemu-kvm yet a bit closer to upstream. >> >> The other highlight is the simplification of synchronization between >> in-kernel and user space VCPU states. This area used to call a lot of >> problems in the past because it was tricky to get things right, >> specifically during the multi-threaded startup. The new approach pushes >> all the sync work around reset and vmsave/load into generic code, not >> only removing the burden from developers of, say, in-kernel APIC >> support, but also dropping most of our kvm-specific hooks, especially in >> the qemu-kvm tree. >> >> While I tested this on various VMs around, and things look good so far, >> I wouldn't be surprised if there are some regressions remaining, >> specifically in the non-x86 parts that I wasn't able to test or even >> build. Please have a careful look! > > The good news on that part is that apart from IA64, all other archs are broken in qemu-kvm anyways, but work on upstream qemu. So moving towards upstream definitely helps here. > OK, then you probably want my corresponding uq/master series in order to test. Will try to roll them out ASAP. Jan -- Siemens AG, Corporate Technology, CT T DE IT 1 Corporate Competence Center Embedded Linux From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.43) id 1NcGnf-0000Ab-0d for qemu-devel@nongnu.org; Tue, 02 Feb 2010 06:20:07 -0500 Received: from [199.232.76.173] (port=34047 helo=monty-python.gnu.org) by lists.gnu.org with esmtp (Exim 4.43) id 1NcGne-00009B-4r for qemu-devel@nongnu.org; Tue, 02 Feb 2010 06:20:06 -0500 Received: from Debian-exim by monty-python.gnu.org with spam-scanned (Exim 4.60) (envelope-from ) id 1NcGnZ-0003Y4-AO for qemu-devel@nongnu.org; Tue, 02 Feb 2010 06:20:05 -0500 Received: from mx20.gnu.org ([199.232.41.8]:30231) by monty-python.gnu.org with esmtps (TLS-1.0:RSA_AES_256_CBC_SHA1:32) (Exim 4.60) (envelope-from ) id 1NcGnY-0003XY-Ai for qemu-devel@nongnu.org; Tue, 02 Feb 2010 06:20:00 -0500 Received: from goliath.siemens.de ([192.35.17.28]) by mx20.gnu.org with esmtp (Exim 4.60) (envelope-from ) id 1NcGnW-0005d9-7W for qemu-devel@nongnu.org; Tue, 02 Feb 2010 06:19:58 -0500 Message-ID: <4B680A5C.7090102@siemens.com> Date: Tue, 02 Feb 2010 12:19:56 +0100 From: Jan Kiszka MIME-Version: 1.0 References: <15E55176-5115-41A5-8F36-923FE9FC093E@suse.de> In-Reply-To: <15E55176-5115-41A5-8F36-923FE9FC093E@suse.de> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Subject: [Qemu-devel] Re: [PATCH 00/21] qemu-kvm: Hook cleanups and extended use of upstream code List-Id: qemu-devel.nongnu.org List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Alexander Graf Cc: Anthony Liguori , "kvm@vger.kernel.org" , Glauber Costa , Marcelo Tosatti , "qemu-devel@nongnu.org" , Avi Kivity Alexander Graf wrote: > On 02.02.2010, at 09:18, Jan Kiszka wrote: > >> Let's start with the overall stats: >> >> 31 files changed, 274 insertions(+), 822 deletions(-) >> >> So this series drops far more than 500 lines of redundant code, moving >> qemu-kvm yet a bit closer to upstream. >> >> The other highlight is the simplification of synchronization between >> in-kernel and user space VCPU states. This area used to call a lot of >> problems in the past because it was tricky to get things right, >> specifically during the multi-threaded startup. The new approach pushes >> all the sync work around reset and vmsave/load into generic code, not >> only removing the burden from developers of, say, in-kernel APIC >> support, but also dropping most of our kvm-specific hooks, especially in >> the qemu-kvm tree. >> >> While I tested this on various VMs around, and things look good so far, >> I wouldn't be surprised if there are some regressions remaining, >> specifically in the non-x86 parts that I wasn't able to test or even >> build. Please have a careful look! > > The good news on that part is that apart from IA64, all other archs are broken in qemu-kvm anyways, but work on upstream qemu. So moving towards upstream definitely helps here. > OK, then you probably want my corresponding uq/master series in order to test. Will try to roll them out ASAP. Jan -- Siemens AG, Corporate Technology, CT T DE IT 1 Corporate Competence Center Embedded Linux