Re: [RFC PATCH] KVM: x86: hyper-v: Inhibit APICv with VP Assist on SPR/EMR
From: Chao Gao
Date: Wed Feb 19 2025 - 19:56:22 EST
On Wed, Feb 19, 2025 at 03:03:39PM +0000, Jon Kohler wrote:
>
>
>> On Aug 7, 2024, at 9:10 AM, Sean Christopherson <seanjc@xxxxxxxxxx> wrote:
>>
>> !-------------------------------------------------------------------|
>> CAUTION: External Email
>>
>> |-------------------------------------------------------------------!
>>
>> On Tue, Aug 06, 2024, Paolo Bonzini wrote:
>>> On Tue, Aug 6, 2024 at 6:03 PM Sean Christopherson <seanjc@xxxxxxxxxx> wrote:
>>>>> As is noted in [1], this issue is considered to be a microcode issue
>>>>> specific to SPR/EMR.
>>>>
>>>> I don't think we can claim that without a more explicit statement from Intel.
>>>> And I would really like Intel to clarify exactly what is going on, so that (a)
>>>> it can be properly documented and (b) we can implement a precise, targeted
>>>> workaround in KVM.
>>>
>>> It is not even clear to me why this patch has any effect at all,
>>> because PV EOI and APICv don't work together anyway: PV EOI requires
>>> apic->highest_isr_cache == -1 (see apic_sync_pv_eoi_to_guest()) but
>>> the cache is only set without APICv (see apic_set_isr()). Therefore,
>>> PV EOI should be basically a no-op with APICv in use.
>>
>> Per Chao, this is a ucode bug though. Speculating wildly, I wonder if Intel added
>> acceleration and/or redirection of HV_X64_MSR_EOI when APICv is enabled, e.g. to
>> speed up existing VMs, and something went sideways.
>
>Hey Sean, Chao, Paolo, quick follow up on this one.
>
>Eiichi was working on pulling down Intel Microcode 20250211 [1], and I had
>asked to retest this one.
>
>Knock on wood, it looks like the issue is “gone” with 20250211 on SPR/EMR
>
>The EMR [2] and SPR [3] release notes allude to some Erratum regarding
>some vmexit fixups that sound interesting, but I’m not sure if they are actually
>the backing issue,
Yes. EMR137 and SPR141 are exactly the backing issue and have been fixed in
recent microcode releases.
>or if this is sheer coincidence, or if there was another fix
>but just isn’t fully documented as an errata?
>
>These two are listed as “fixed” in the release notes:
>EMR137. VM Exit Following MOV to CR8 Instruction May Lead to Unexpected IDT Vectoring-Information
>SPR141. VM Exit Following MOV to CR8 Instruction May Lead to Unexpected IDT Vectoring-Information
>
>@Chao - could you help confirm our observations one way or the other?
>
>[1] https://github.com/intel/Intel-Linux-Processor-Microcode-Data-Files/releases/tag/microcode-20250211
>[2] https://cdrdv2.intel.com/v1/dl/getContent/793902 (EMR microcode release notes)
>[3] https://cdrdv2.intel.com/v1/dl/getContent/772415 (SPR microcode release notes)