Re: [PATCH] KVM: nSVM: vmentry ignores EFER.LMA and possibly RFLAGS.VM

[Paolo Bonzini <pbonzini@redhat.com>]

On 09/07/20 19:12, Jim Mattson wrote:
>> +
>> +       /* The processor ignores EFER.LMA, but svm_set_efer needs it.  */
>> +       efer &= ~EFER_LMA;
>> +       if ((nested_vmcb->save.cr0 & X86_CR0_PG)
>> +           && (nested_vmcb->save.cr4 & X86_CR4_PAE)
>> +           && (efer & EFER_LME))
>> +               efer |= EFER_LMA;
> The CR4.PAE check is unnecessary, isn't it? The combination CR0.PG=1,
> EFER.LMA=1, and CR4.PAE=0 is not a legal processor state.

Yeah, I was being a bit cautious because this is the nested VMCB and it
can be filled in with invalid state, but indeed that condition was added
just yesterday by myself in nested_vmcb_checks (while reviewing Krish's
CR0/CR3/CR4 reserved bit check series).

That said, the VMCB here is guest memory and it can change under our
feet between nested_vmcb_checks and nested_prepare_vmcb_save.  Copying
the whole save area is overkill, but we probably should copy at least
EFER/CR0/CR3/CR4 in a struct at the beginning of nested_svm_vmrun; this
way there'd be no TOC/TOU issues between nested_vmcb_checks and
nested_svm_vmrun.  This would also make it easier to reuse the checks in
svm_set_nested_state.  Maybe Maxim can look at it while I'm on vacation,
as he's eager to do more nSVM stuff. :D

I'll drop this patch for now.

Thanks for the speedy review!

Paolo

> According to the SDM,
> 
> * IA32_EFER.LME cannot be modified while paging is enabled (CR0.PG =
> 1). Attempts to do so using WRMSR cause a general-protection exception
> (#GP(0)).
> * Paging cannot be enabled (by setting CR0.PG to 1) while CR4.PAE = 0
> and IA32_EFER.LME = 1. Attempts to do so using MOV to CR0 cause a
> general-protection exception (#GP(0)).
> * CR4.PAE and CR4.LA57 cannot be modified while either 4-level paging
> or 5-level paging is in use (when CR0.PG = 1 and IA32_EFER.LME = 1).
> Attempts to do so using MOV to CR4 cause a general-protection
> exception (#GP(0)).
>