From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:34886) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1UweXb-00053O-Fd for qemu-devel@nongnu.org; Tue, 09 Jul 2013 16:29:40 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1UweXa-0006Fu-E6 for qemu-devel@nongnu.org; Tue, 09 Jul 2013 16:29:39 -0400 Received: from cantor2.suse.de ([195.135.220.15]:57765 helo=mx2.suse.de) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1UweXa-0006Fn-44 for qemu-devel@nongnu.org; Tue, 09 Jul 2013 16:29:38 -0400 Message-ID: <51DC72AD.2090209@suse.de> Date: Tue, 09 Jul 2013 22:29:33 +0200 From: =?ISO-8859-1?Q?Andreas_F=E4rber?= MIME-Version: 1.0 References: <20130709202700.GA30219@redhat.com> In-Reply-To: <20130709202700.GA30219@redhat.com> Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Subject: Re: [Qemu-devel] win7 boot failure with commit f68ec83 - ide: Set BSY bit during FLUSH List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: "Michael S. Tsirkin" Cc: Kevin Wolf , Gonglei , qemu-devel@nongnu.org, Stefan Hajnoczi Hi, Am 09.07.2013 22:27, schrieb Michael S. Tsirkin: > I am using this command line: > qemu-system-x86_64 -m 512 -enable-kvm -drive > id=3Ddisk,file=3D./images/win7-x86.qcow2,if=3Dnone -device ahci,id=3Dah= ci > -device ide-drive,drive=3Ddisk,bus=3Dahci.0 -cdrom > ./images/en_windows_7_ultimate_x86_dvd_x15-65921.iso -monitor stdio > -cpu kvm64 -snapshot >=20 > and windows 7 does not boot. The same was recently reported for q35 - Kevin found the cause but I haven't noticed a proposed solution yet. Andreas > This happens starting from this commit: >=20 > commit f68ec8379e88502b4841a110c070e9b118d3151c > Author: Andreas F=E4rber > Date: Wed Jun 5 15:17:57 2013 +0200 >=20 > ide: Set BSY bit during FLUSH > =20 > The implementation of the ATA FLUSH command invokes a flush at the = block > layer, which may on raw files on POSIX entail a synchronous fdatasy= nc(). > This may in some cases take so long that the SLES 11 SP1 guest driv= er > reports I/O errors and filesystems get corrupted or remounted read-= only. > =20 > Avoid this by setting BUSY_STAT, so that the guest is made aware we= are > in the middle of an operation and no ATA commands are attempted to = be > processed concurrently. > =20 > Addresses BNC#637297. > =20 > Suggested-by: Gonglei (Arei) > Signed-off-by: Andreas F=E4rber > Signed-off-by: Kevin Wolf > Signed-off-by: Stefan Hajnoczi >=20 >=20 > guest starts booting but never completes boot - it just keeps flashing > the flag image. Boots fine with previous commit: > commit c27d565604038c1572b16dd1cd06e277e6ef02e2 >=20 > However, reverting f68ec8379e88502b4841a110c070e9b118d3151c on > current master still has the hang on boot > behaviour - possibly a separate issue, I'll have to redo the bisect > to find out. >=20 --=20 SUSE LINUX Products GmbH, Maxfeldstr. 5, 90409 N=FCrnberg, Germany GF: Jeff Hawn, Jennifer Guild, Felix Imend=F6rffer; HRB 16746 AG N=FCrnbe= rg