From mboxrd@z Thu Jan 1 00:00:00 1970 From: Dor Laor Subject: Re: kvm: Unknown error 524, Fail to handle apic access vmexit Date: Wed, 16 Jul 2008 16:42:10 +0300 Message-ID: <487DFAB2.8000704@qumranet.com> References: <20080715124202.GA3001@deprecation.cyrius.com> <487CBFEB.4050409@qumranet.com> <200807161126.37615.sheng.yang@intel.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Cc: kvm@vger.kernel.org, Martin Michlmayr To: "Yang, Sheng" Return-path: Received: from il.qumranet.com ([212.179.150.194]:48450 "EHLO il.qumranet.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1758648AbYGPNnr (ORCPT ); Wed, 16 Jul 2008 09:43:47 -0400 In-Reply-To: <200807161126.37615.sheng.yang@intel.com> Sender: kvm-owner@vger.kernel.org List-ID: Yang, Sheng wrote: > On Tuesday 15 July 2008 23:19:07 Dor Laor wrote: > >> Martin Michlmayr wrote: >> >>> I installed a Windows XP SP2 guest on a Debian x86_64 host The >>> installation itself went fine but kvm aborts when when XP starts >>> during "Windows XP Setup". XP mentions something with >>> intelppm.sys (see the attached screenshot) and kvm says: >>> >>> kvm_run: Unknown error 524 >>> kvm_run returned -524 >>> >> It's a FlexPriority bug, while it should be solved, you can disable >> it by using kvm-intel module parameter. >> >> > > Dor, are you sure it's a FlexPriority bug? > > Well, I'm not sure it's the FlexPriority's fault, it's just when it is disabled it does not happen and I saw the apic access. It can be miss emulation too. It happened to me on ~ kvm-69 > If you look at where is the complain, you would find there is a result > of emulate_instruction(). > > And you will find a clearly "emulation failed (mmio) rip 7cb3d000 ff > ff 8d 85" in the bug tracker Martin metioned above the "Fail to > handle apic access vmexit! Offset is 0xf0"(Spurious Interrupt Vector > Register). > > I don't think "ff ff 8d 85" is a vaild opcode for that case. > > Maybe it's a regression? The last report is long ago... > > Hi Martin, can you show more dmesg here? And if it can be reproduce > stable? > > Thanks. > >