Commit e79f245ddec1 ("X86/KVM: Properly update 'tsc_offset' to represent
the running guest") introduced a regression in enter_vmx_non_root_mode():
when nested_vmx_load_msr() fails exit_qualification needs to point to the
entry number we failed to validate. Intel's SDM states:
"VM-entry failure due to MSR loading. The exit qualification is loaded to
indicate which entry in the VM-entry MSR-load area caused the problem (1
for the first entry, 2 for the second, etc.)."
Fixes: e79f245ddec1 ("X86/KVM: Properly update 'tsc_offset' to represent the running guest")
Reported-by: Wanpeng Li <kernellwp@xxxxxxxxx>
Signed-off-by: Vitaly Kuznetsov <vkuznets@xxxxxxxxxx>
---
arch/x86/kvm/vmx.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
index 65968649b365..be468c822892 100644
--- a/arch/x86/kvm/vmx.c
+++ b/arch/x86/kvm/vmx.c
@@ -11720,8 +11720,10 @@ static int enter_vmx_non_root_mode(struct kvm_vcpu *vcpu)
msr_entry_idx = nested_vmx_load_msr(vcpu,
vmcs12->vm_entry_msr_load_addr,
vmcs12->vm_entry_msr_load_count);
- if (msr_entry_idx)
+ if (msr_entry_idx) {
+ exit_qual = msr_entry_idx;
goto fail;
+ }
/*
* Note no nested_vmx_succeed or nested_vmx_fail here. At this point
--
2.14.4
