Re: [PATCH 4/6] kvm: nVMX: support EPT accessed/dirty bits

From	Paolo Bonzini <pbonzini@redhat.com>
Newsgroups	linux.kernel
Subject	Re: [PATCH 4/6] kvm: nVMX: support EPT accessed/dirty bits
Date	2017-04-14 07:20 +0200
Message-ID	<tw1Oh-5tU-5@gated-at.bofh.it> (permalink)
References	<tqF21-4or-3@gated-at.bofh.it> <tqF22-4or-25@gated-at.bofh.it> <tvdya-5fl-3@gated-at.bofh.it> <tvdRw-5mf-9@gated-at.bofh.it> <tvzyF-2Mb-9@gated-at.bofh.it>
Organization	linux.* mail to news gateway

Show all headers | View raw


On 13/04/2017 07:02, Bandan Das wrote:
>> For EPT it is, you're right it's fishy.  The "nested_access" should be
>> computed in translate_nested_gpa, which is where kvm->arch.nested_mmu
>> (non-EPT) requests to access kvm->arch.mmu (EPT).
>
> Thanks for the clarification. Is it the case when L1 runs L2 without
> EPT ? I can't figure out the case where translate_nested_gpa will actually
> be called.

It happens when L2 instruction are emulated by L0, for example when L1
is passing through I/O ports to L2 and L2 runs an "insb" instruction.  I
think this case is not covered by vmx.flat.

Paolo

> FNAME(walk_addr_nested) calls walk_addr_generic
> with &vcpu->arch.nested_mmu and init_kvm_nested_mmu() sets gva_to_gpa()
> with the appropriate "_nested" functions. But the gva_to_gpa() pointers
> don't seem to get invoked at all for the nested case.
> 
> BTW, just noticed that setting PFERR_USER_MASK is redundant since
> translate_nested_gpa does it too.

Back to linux.kernel | Previous | Next — Previous in thread | Find similar | Unroll thread

Thread

Re: [PATCH 4/6] kvm: nVMX: support EPT accessed/dirty bits Bandan Das <bsd@redhat.com> - 2017-04-12 01:40 +0200
  Re: [PATCH 4/6] kvm: nVMX: support EPT accessed/dirty bits Paolo Bonzini <pbonzini@redhat.com> - 2017-04-12 02:00 +0200
    Re: [PATCH 4/6] kvm: nVMX: support EPT accessed/dirty bits Bandan Das <bsd@redhat.com> - 2017-04-13 01:10 +0200
      Re: [PATCH 4/6] kvm: nVMX: support EPT accessed/dirty bits Paolo Bonzini <pbonzini@redhat.com> - 2017-04-14 07:20 +0200

csiph-web