From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.43) id 1NFBKa-0003k6-5Z for qemu-devel@nongnu.org; Mon, 30 Nov 2009 13:50:40 -0500 Received: from exim by lists.gnu.org with spam-scanned (Exim 4.43) id 1NFBKT-0003ar-Vj for qemu-devel@nongnu.org; Mon, 30 Nov 2009 13:50:38 -0500 Received: from [199.232.76.173] (port=60631 helo=monty-python.gnu.org) by lists.gnu.org with esmtp (Exim 4.43) id 1NFBKT-0003ah-Po for qemu-devel@nongnu.org; Mon, 30 Nov 2009 13:50:33 -0500 Received: from verein.lst.de ([213.95.11.210]:56236) by monty-python.gnu.org with esmtps (TLS-1.0:DHE_RSA_3DES_EDE_CBC_SHA1:24) (Exim 4.60) (envelope-from ) id 1NFBKT-0005PV-CN for qemu-devel@nongnu.org; Mon, 30 Nov 2009 13:50:33 -0500 Date: Mon, 30 Nov 2009 19:50:28 +0100 From: Christoph Hellwig Subject: Re: [Qemu-devel] [PATCH 0/6] Fix -kernel with SeaBIOS v2 Message-ID: <20091130185028.GA1495@lst.de> References: <20091120113113.GA26408@lst.de> <450764C8-F59A-411B-9733-07136D97981B@suse.de> <20091123202138.GA9879@lst.de> <5024501E-4BFD-4C6B-8455-20C406948C20@suse.de> <20091123213600.GA12479@lst.de> <4B0CFFD5.1010003@suse.de> <20091126114213.GA31527@lst.de> <4B0E6DD9.5030608@suse.de> <20091126121855.GA336@lst.de> <4B0E7376.3020907@suse.de> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <4B0E7376.3020907@suse.de> List-Id: qemu-devel.nongnu.org List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Alexander Graf Cc: Juan Quintela , Glauber Costa , "qemu-devel@nongnu.org" , Avi Kivity , Christoph Hellwig On Thu, Nov 26, 2009 at 01:24:22PM +0100, Alexander Graf wrote: > Hm - maybe worth a try to give it a -L to the source pc-bios directory > anyways. Doesn't change a thing. > Sounds like your guest kernel is trying to access an x86_64 register? > > You can of cause try insmod'ing kvm.ko with ignore_msrs=1. You hopefully > don't need syscalls until you get into user space. The option doesn't exist yet in 2.6.31 which I'm running on this box because 2.6.32-rc has massive regressions for my workload. I booted into it again anyway to test the option - and interestingly enough qemu with -enable-kvm boots just fine even without ignore_msrs=1. Looks like the problem only happens with older kernels.