KVM: x86/mmu: Always walk guest PTEs with WRITE access when unprotecting
authorSean Christopherson <seanjc@google.com>
Sat, 31 Aug 2024 00:15:28 +0000 (17:15 -0700)
committerSean Christopherson <seanjc@google.com>
Tue, 10 Sep 2024 03:16:28 +0000 (20:16 -0700)
When getting a gpa from a gva to unprotect the associated gfn when an
event is awating reinjection, walk the guest PTEs for WRITE as there's no
point in unprotecting the gfn if the guest is unable to write the page,
i.e. if write-protection can't trigger emulation.

Note, the entire flow should be guarded on the access being a write, and
even better should be conditioned on actually triggering a write-protect
fault.  This will be addressed in a future commit.

Reviewed-by: Yuan Yao <yuan.yao@intel.com>
Link: https://lore.kernel.org/r/20240831001538.336683-14-seanjc@google.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
arch/x86/kvm/mmu/mmu.c

index bafec04..937fa9a 100644 (file)
@@ -2751,7 +2751,7 @@ static int kvm_mmu_unprotect_page_virt(struct kvm_vcpu *vcpu, gva_t gva)
        if (vcpu->arch.mmu->root_role.direct)
                return 0;
 
-       gpa = kvm_mmu_gva_to_gpa_read(vcpu, gva, NULL);
+       gpa = kvm_mmu_gva_to_gpa_write(vcpu, gva, NULL);
        if (gpa == INVALID_GPA)
                return 0;