From mboxrd@z Thu Jan 1 00:00:00 1970 From: Avi Kivity Subject: Re: kvm-40 and above: Win XP SP2 install crashes on AMD64 & 2.6.22.5-49 Date: Wed, 26 Sep 2007 11:02:59 +0200 Message-ID: <46FA2043.2000807@qumranet.com> References: <20070925195518.GA6265@mars.oe1kib.ampr.org> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Cc: kvm-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org To: Klaus Kudielka Return-path: In-Reply-To: <20070925195518.GA6265-OaR9+9SMzBvsjPbl2Sz6imD2FQJk+8+b@public.gmane.org> List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: kvm-devel-bounces-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org Errors-To: kvm-devel-bounces-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org List-Id: kvm.vger.kernel.org Klaus Kudielka wrote: > Hello, > > Starting with kvm-40, I am not able to install a Win XP SP2 guest any more. > I consistently get a BSOD somewhere in the 2nd stage install. > Sometimes it is IRQL_NOT_LESS_OR_EQUAL, otherwise STOP: 0x0000008E > or STOP: 0x0000000A > > In most failure cases I also get a kernel message: > emulation failed (mmio) rip 426 c4 c4 00 00 > > The hardware is an AMD Athlon(tm) 64 Processor 3500+ with 3G of RAM. > The host is a FC6 installation with a pretty recent kernel (2.6.22.5-49) > I configured kvm only with --prefix=/opt/kvm > The command to run the installation was > > /opt/kvm/bin/qemu-system-x86_64 -no-acpi -hda /dev/Guests/xp_kvm -m 1024 -net nic,mac=00:16:3e:11:80:49,model=rtl8139 -cdrom /d/3/oe1kib/img/windows_xp_sp2_en.iso -boot d > > I always select the "Standard PC" HAL during install. > > On kvm-39 & 36, the install completes fine. On kvm-40,41,43,44 it fails > with the symptoms described above. I have not tested others, but the > tendency seems clear. > > [With -no-kvm, qemu-system-x86_64 does not crash the guest, but Win XP seems > to be "Installing Devices" for the rest of its life. The session is alive, > but no progress anymore. Likely a different problem] > > Any idea? Or even a fix? > Can you do a bisect to find which commit caused the breakage? -- error compiling committee.c: too many arguments to function ------------------------------------------------------------------------- This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2005. http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/