From mboxrd@z Thu Jan 1 00:00:00 1970 From: George Dunlap Subject: Re: Bug: Windows 2003 fails to install on xen-unstable tip Date: Wed, 17 Apr 2013 19:13:18 +0100 Message-ID: <516EE63E.4090008@eu.citrix.com> References: <516D7DA202000078000CDA60@nat28.tlf.novell.com> <516D61A8.1080102@eu.citrix.com> <516E791902000078000CDE09@nat28.tlf.novell.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii"; Format="flowed" Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <516E791902000078000CDE09@nat28.tlf.novell.com> List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xen.org Errors-To: xen-devel-bounces@lists.xen.org To: Jan Beulich Cc: "Tim (Xen.org)" , Eddie Dong , Suravee Suthikulpanit , "xen-devel@lists.xen.org" List-Id: xen-devel@lists.xenproject.org On 17/04/13 09:27, Jan Beulich wrote: >>>> On 16.04.13 at 16:35, George Dunlap wrote: >> On 16/04/13 15:34, Jan Beulich wrote: >>>>>> On 16.04.13 at 16:22, George Dunlap wrote: >>>> The other day I just happened to try to install Windows 2003 on >>>> xen-unstable tip. After booting the CD and copying the files, the >>>> whole thing stops at "Setup is starting Windows". >>>> >>>> The VM is in the "blocked" state; which means it has called HLT and is >>>> waiting for interrupts; but no interrupts are delivered, so everything >>>> just stops. >>>> >>>> Windows XP, strangely enough, installs just fine. >>>> >>>> I have duplicated this with both qemu-upstream and qemu-traditional; >>>> and also on both AMD and Intel. >>>> >>>> I was wondering: >>>> 1. if any of you had an idea what changeset might be involved >>> Any of the RTC emulation changes since 4.2 would be possible >>> candidates, with mine much more likely than Tim's (since his mainly >>> fix issues mine introduced). >> If you've got a git hash tag for your first change, I can do a quick >> test to see if that's it. > The first of these commits is 2fe82ac11fd078485388fe7c5e8bc3b6ac9185b0. Here are the results of my bisect: 620d5dad54008e40798c4a0c4322aef274c36fa3 is the first bad commit commit 620d5dad54008e40798c4a0c4322aef274c36fa3 Author: Jan Beulich Date: Tue Feb 5 09:44:00 2013 +0100 x86/HVM: assorted RTC emulation adjustments - only call check_update_timer() on REG_B writes when SET changes - only call alarm_timer_update() on REG_B writes when relevant bits change - instead properly handle AF and PF when the guest is not also setting AIE/PIE respectively (for UF this was already the case, only a comment was slightly inaccurate), including calling the respective update functions upon REG_C reads Signed-off-by: Jan Beulich Tested-by: Olaf Hering