shinys000114/linux
1
0
Fork 0
mirror of https://github.com/hardkernel/linux.git synced 2026-06-09 12:17:12 +09:00
Code Issues Packages Projects Releases Wiki Activity

KVM: x86: Reinitialize context if host userspace toggles EFER.LME

Browse Source 
commit d617429936 upstream.

While the guest runs, EFER.LME cannot change unless CR0.PG is clear, and
therefore EFER.NX is the only bit that can affect the MMU role.  However,
set_efer accepts a host-initiated change to EFER.LME even with CR0.PG=1.
In that case, the MMU has to be reset.

Fixes: 11988499e6 ("KVM: x86: Skip EFER vs. guest CPUID checks for host-initiated writes")
Cc: stable@vger.kernel.org
Reviewed-by: Sean Christopherson <seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This commit is contained in:
Paolo Bonzini
2022-02-09 04:56:05 -05:00
committed by Greg Kroah-Hartman
parent 273ad0cf56
commit 48306afcac
2 changed files with 2 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

1
arch/x86/kvm/mmu.h
View File

@@ -49,6 +49,7 @@
X86_CR4_LA57)
#define KVM_MMU_CR0_ROLE_BITS (X86_CR0_PG | X86_CR0_WP)
#define KVM_MMU_EFER_ROLE_BITS (EFER_LME | EFER_NX)
static __always_inline u64 rsvd_bits(int s, int e)
{

3
arch/x86/kvm/x86.c
View File

@@ -1605,8 +1605,7 @@ static int set_efer(struct kvm_vcpu *vcpu, struct msr_data *msr_info)
return r;
}
/* Update reserved bits */
if ((efer ^ old_efer) & EFER_NX)
if ((efer ^ old_efer) & KVM_MMU_EFER_ROLE_BITS)
kvm_mmu_reset_context(vcpu);
return 0;