From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from [140.186.70.92] (port=52935 helo=eggs.gnu.org) by lists.gnu.org with esmtp (Exim 4.43) id 1PkwCT-0008Cq-Qc for qemu-devel@nongnu.org; Thu, 03 Feb 2011 05:14:07 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1PkwCS-0002HQ-Il for qemu-devel@nongnu.org; Thu, 03 Feb 2011 05:14:05 -0500 Received: from thoth.sbs.de ([192.35.17.2]:26806) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1PkwCS-0002HE-8T for qemu-devel@nongnu.org; Thu, 03 Feb 2011 05:14:04 -0500 Message-ID: <4D4A7FEA.4000708@siemens.com> Date: Thu, 03 Feb 2011 11:14:02 +0100 From: Jan Kiszka MIME-Version: 1.0 Subject: Re: [Qemu-devel] KVM: Windows 64-bit troubles with user space irqchip References: <4D4946F7.1070702@siemens.com> <20110202123532.GF14984@redhat.com> <4D4952FA.8020300@siemens.com> <4D49569F.6060207@redhat.com> <4D496A8D.90000@siemens.com> <4D496BC5.10807@redhat.com> <4D496D77.2010405@siemens.com> <4D496FA6.8070301@siemens.com> <4D49738D.7080404@redhat.com> <4D4979BD.6080900@siemens.com> <20110202154611.GR14984@redhat.com> <4D497DAB.7010901@siemens.com> <4D4A64F2.8010309@redhat.com> <4D4A7629.1010506@siemens.com> <4D4A7D08.6070900@redhat.com> In-Reply-To: <4D4A7D08.6070900@redhat.com> Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit List-Id: qemu-devel.nongnu.org List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Avi Kivity Cc: kvm , Gleb Natapov , qemu-devel On 2011-02-03 11:01, Avi Kivity wrote: > On 02/03/2011 11:32 AM, Jan Kiszka wrote: >> On 2011-02-03 09:18, Avi Kivity wrote: >>> On 02/02/2011 05:52 PM, Jan Kiszka wrote: >>>>>> >>>>> If there is no problem in the logic of this commit (and I do not see >>>>> one yet) then we somewhere miss kicking vcpu when interrupt, that should be >>>>> handled, arrives? >>>> >>>> I'm not yet confident about the logic of the kernel patch: mov to cr8 is >>>> serializing. If the guest raises the tpr and then signals this with a >>>> succeeding, non vm-exiting instruction to the other vcpus, one of those >>>> could inject an interrupt with a higher priority than the previous tpr, >>>> but a lower one than current tpr. QEMU user space would accept this >>>> interrupt - and would likely surprise the guest. Do I miss something? >>> >>> apic_get_interrupt() is only called from the vcpu thread, so it should >>> see a correct tpr. >>> >>> The only difference I can see with the patch is that we may issue a >>> spurious cpu_interrupt(). But that shouldn't do anything bad, should it? >> >> I tested this yesterday, and it doesn't confuse Windows. It actually >> receives quite a few spurious IRQs in normal operation, w/ or w/o the >> kernel's tpr optimization. > > I don't see why there should be any spurious interrupts in normal > operation. From the docs, these happen due to an INTA cycle racing with > raising the TPR, but in ioapic mode, there shouldn't be any INTA cycles. > I added an instrumentation to the line of apic_get_interrupt that returns the spurious vector, and it triggered fairly often. Just didn't examined why this happens even without the tpr optimization. Jan -- Siemens AG, Corporate Technology, CT T DE IT 1 Corporate Competence Center Embedded Linux