From: Jan Kiszka <jan.kiszka@xxxxxxxxxxx> Switching the VMCS obviously invalidates what may have been cached about the guest segments. Signed-off-by: Jan Kiszka <jan.kiszka@xxxxxxxxxxx> --- Probably, the missing invalidations are harmless ATM, but you never know. arch/x86/kvm/vmx.c | 4 ++++ 1 files changed, 4 insertions(+), 0 deletions(-) diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c index 94f3b66..d45f0e0 100644 --- a/arch/x86/kvm/vmx.c +++ b/arch/x86/kvm/vmx.c @@ -7236,6 +7236,8 @@ static int nested_vmx_run(struct kvm_vcpu *vcpu, bool launch) vcpu->cpu = cpu; put_cpu(); + vmx_segment_cache_clear(vmx); + vmcs12->launch_state = 1; prepare_vmcs02(vcpu, vmcs12); @@ -7504,6 +7506,8 @@ static void nested_vmx_vmexit(struct kvm_vcpu *vcpu) vcpu->cpu = cpu; put_cpu(); + vmx_segment_cache_clear(vmx); + /* if no vmcs02 cache requested, remove the one we used */ if (VMCS02_POOL_SIZE == 0) nested_free_vmcs02(vmx, vmx->nested.current_vmptr); -- 1.7.3.4 -- To unsubscribe from this list: send the line "unsubscribe kvm" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
