Re: [PATCH] KVM/x86: fix comment that should be user fault
From: Sean Christopherson
Date: Thu Feb 23 2023 - 10:52:30 EST
On Thu, Feb 23, 2023, Quanfa Fu wrote:
> The content of comment should be user fault not read. In order to avoid
> confusion, fix the comment.
No, the existing comment is correct. Ignoring optional extensions, EPT doesn't
differntiate between Supervisor and User, but does support !READABLE mappings.
And so KVM piggybacks PFERR_USER_MASK to track whether or not an EPT fault occurred
on a read access.
>
> Signed-off-by: Quanfa Fu <quanfafu@xxxxxxxxx>
> ---
> arch/x86/kvm/vmx/vmx.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/arch/x86/kvm/vmx/vmx.c b/arch/x86/kvm/vmx/vmx.c
> index 7eec0226d56a..3c1012039517 100644
> --- a/arch/x86/kvm/vmx/vmx.c
> +++ b/arch/x86/kvm/vmx/vmx.c
> @@ -5668,7 +5668,7 @@ static int handle_ept_violation(struct kvm_vcpu *vcpu)
> gpa = vmcs_read64(GUEST_PHYSICAL_ADDRESS);
> trace_kvm_page_fault(vcpu, gpa, exit_qualification);
>
> - /* Is it a read fault? */
> + /* Is it a user fault? */
> error_code = (exit_qualification & EPT_VIOLATION_ACC_READ)
> ? PFERR_USER_MASK : 0;
> /* Is it a write fault? */
> --
> 2.31.1
>
>