| 1234567891011121314151617181920212223242526272829303132333435363738 |
- From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
- From: Boris Ostrovsky <[email protected]>
- Date: Fri, 31 Jan 2020 08:06:40 -0300
- Subject: [PATCH] x86/kvm: Be careful not to clear KVM_VCPU_FLUSH_TLB bit
- CVE-2019-3016
- CVE-2020-3016
- kvm_steal_time_set_preempted() may accidentally clear KVM_VCPU_FLUSH_TLB
- bit if it is called more than once while VCPU is preempted.
- This is part of CVE-2019-3016.
- (This bug was also independently discovered by Jim Mattson
- <[email protected]>)
- Signed-off-by: Boris Ostrovsky <[email protected]>
- Reviewed-by: Joao Martins <[email protected]>
- Signed-off-by: Thadeu Lima de Souza Cascardo <[email protected]>
- Signed-off-by: Thomas Lamprecht <[email protected]>
- ---
- arch/x86/kvm/x86.c | 3 +++
- 1 file changed, 3 insertions(+)
- diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
- index 778b3a899769..92d8e4ebba16 100644
- --- a/arch/x86/kvm/x86.c
- +++ b/arch/x86/kvm/x86.c
- @@ -3393,6 +3393,9 @@ static void kvm_steal_time_set_preempted(struct kvm_vcpu *vcpu)
- if (!(vcpu->arch.st.msr_val & KVM_MSR_ENABLED))
- return;
-
- + if (vcpu->arch.st.steal.preempted)
- + return;
- +
- vcpu->arch.st.steal.preempted = KVM_VCPU_PREEMPTED;
-
- kvm_write_guest_offset_cached(vcpu->kvm, &vcpu->arch.st.stime,
|