From mboxrd@z Thu Jan 1 00:00:00 1970 From: Andrew Cooper Subject: Re: HVM crash system on AMD APU A8-6600K Date: Mon, 3 Feb 2014 14:09:08 +0000 Message-ID: <52EFA304.4000206@citrix.com> References: <52EF9EEF.8050301@citrix.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xen.org Errors-To: xen-devel-bounces@lists.xen.org To: Vitaliy Tomin Cc: xen-devel@lists.xensource.com, Jan Beulich , SuraveeSuthikulpanit List-Id: xen-devel@lists.xenproject.org On 03/02/14 13:58, Vitaliy Tomin wrote: > lspci output attached. > > I have never managed to crash system with debug=y, but I can provide > serial log captured with debug=y and HVM domain running up. That is a curious data point - it would imply that debug mode is doing something which non-debug mode fails to do. Could you provide the log please? Can you explain "=== whole system crashed ===" a little more. Given the lack of stack trace or any hint of a problem from Xen, is it a system hang? Does adding "watchdog" to the Xen command line cause the failure to change? Looking back at the debug=n serial log in combination with the PCI topology, (XEN) AMD-Vi: No iommu for device 0000:00:00.2 (XEN) setup 0000:00:00.2 for d0 failed (-19) Device 00:00.2 is the IOMMU itself. I would have thought applying IOMMU translation to the IOMMU is going to end in tears. Suravee; Can you comment about this? is the IOMMU expected to have an IVRS entry? ~Andrew