From mboxrd@z Thu Jan 1 00:00:00 1970 From: Martin Wolf Subject: Re: vga passthrough // questions about pci passthrough Date: Tue, 02 Oct 2012 10:22:15 +0200 Message-ID: <506AA437.2010806@adiumentum.com> References: <50064D87.2090809@adiumentum.com> <50068134.4060306@siemens.com> <5006A72D.3080408@adiumentum.com> <5006ADF3.9030304@adiumentum.com> <1342621405.2229.194.camel@bling.home> <50081FBC.7060208@adiumentum.com> <1342714590.3142.4.camel@ul30vt> <5064364E.7050508@adiumentum.com> <1348768586.2320.204.camel@ul30vt.home> <50649E5C.6020808@adiumentum.com> <1348773491.2320.217.camel@ul30vt.home> <50655BE6.9050305@siemens.com> <1348847417.2320.333.camel@ul30vt.home> <5065D066.4030006@siemens.com> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: kvm@vger.kernel.org, Alex Williamson To: Jan Kiszka Return-path: Received: from mail.andre-duewel.de ([178.63.55.43]:60140 "EHLO mail.andre-duewel.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754197Ab2JBIW3 (ORCPT ); Tue, 2 Oct 2012 04:22:29 -0400 In-Reply-To: <5065D066.4030006@siemens.com> Sender: kvm-owner@vger.kernel.org List-ID: Hello, would a Memory Dump from inside the VM help solving the BSOD, or is that pointless? Thank you for your patience and help ;) Am 28.09.2012 18:29, schrieb Jan Kiszka: > On 2012-09-28 17:50, Alex Williamson wrote: >> On Fri, 2012-09-28 at 10:12 +0200, Jan Kiszka wrote: >>> On 2012-09-27 21:18, Alex Williamson wrote: >>>> On Thu, 2012-09-27 at 20:43 +0200, Martin Wolf wrote: >>>>> thank you for the information. >>>>> >>>>> i will try what you mentioned... >>>>> do you have some additional information about rebooting a VM with= a >>>>> passed through videocard? >>>>> (amd / ati 7870) >>>> I don't. Is the bsod on reboot only or does it also happen on shu= tdown? >>>> There's a slim chance it could be traced by enabling debug in the >>>> pci-assign driver and analyzing what the guest driver is trying to= do. >>>> I'm hoping that q35 chipset support might resolve some issues with= vga >>>> assignment as it exposes a topology that looks a bit more like one= that >>>> a driver would expect on physical hardware. Thanks, >>> From our attempts to get more working than what NVIDIA Quadro card= s >>> support officially, my own experiments with q35 in this context and= our >>> discussions with NVIDIA, I'm pretty skeptical that this chipset wil= l >>> make a difference here. Most problems are due to those non-standard= side >>> channels to configure the hardware, memory mappings etc. And gettin= g >>> this working requires either cooperation of the vendor or *a lot* o= f >>> reverse engineering. >> I heard from an nvidia guy that the driver behaves differently depen= ding >> on whether it finds an upstream express port, so we're probably caus= ing >> ourselves more problems if it's trying to run in AGP mode. > May be a point for the low- to mid-range cards. It does not apply to = the > "virtualization-ready" Quadro series according to our information bac= k then. > >> There was >> also a lot of FUD in Xen (maybe justified) around how the BIOS >> determines the memory ranges and whether it bypasses the PCI BARs an= d >> gets them directly. That means some cards may require identity mapp= ing >> to work. It seems like the very high-end cards are possibly fixing >> this, but they're far more expensive than I can justify. Thanks, > Yes, that is what makes them virtualization ready. But they also come > with limitations. So far, you can't pass-through a primary card or us= e > it for early boot messages of the guest as the BIOS is not ready for > that - without identity mapping or even more. > > Jan > --=20 Adiumentum GmbH Gf. Martin Wolf Banderbacherstra=C3=9Fe 76 90513 Zirndorf 0911 / 9601470 mwolf@adiumentum.com