[PATCH v2 01/22] KVM: SVM: introduce nested_run_pending
From: Paolo Bonzini
Date: Fri Apr 24 2020 - 13:25:20 EST
We want to inject vmexits immediately from svm_check_nested_events,
so that the interrupt/NMI window requests happen in inject_pending_event
right after it returns.
This however has the same issue as in vmx_check_nested_events, so
introduce a nested_run_pending flag with the exact same purpose
of delaying vmexit injection after the vmentry.
Signed-off-by: Paolo Bonzini <pbonzini@xxxxxxxxxx>
---
arch/x86/kvm/svm/nested.c | 3 ++-
arch/x86/kvm/svm/svm.c | 1 +
arch/x86/kvm/svm/svm.h | 4 ++++
3 files changed, 7 insertions(+), 1 deletion(-)
diff --git a/arch/x86/kvm/svm/nested.c b/arch/x86/kvm/svm/nested.c
index a7c3b3030e59..51cfab68428d 100644
--- a/arch/x86/kvm/svm/nested.c
+++ b/arch/x86/kvm/svm/nested.c
@@ -413,6 +413,7 @@ int nested_svm_vmrun(struct vcpu_svm *svm)
copy_vmcb_control_area(hsave, vmcb);
+ svm->nested.nested_run_pending = 1;
enter_svm_guest_mode(svm, vmcb_gpa, nested_vmcb, &map);
if (!nested_svm_vmrun_msrpm(svm)) {
@@ -792,7 +793,8 @@ static int svm_check_nested_events(struct kvm_vcpu *vcpu)
{
struct vcpu_svm *svm = to_svm(vcpu);
bool block_nested_events =
- kvm_event_needs_reinjection(vcpu) || svm->nested.exit_required;
+ kvm_event_needs_reinjection(vcpu) || svm->nested.exit_required ||
+ svm->nested.nested_run_pending;
if (kvm_cpu_has_interrupt(vcpu) && nested_exit_on_intr(svm)) {
if (block_nested_events)
diff --git a/arch/x86/kvm/svm/svm.c b/arch/x86/kvm/svm/svm.c
index c86f7278509b..77440b5953e3 100644
--- a/arch/x86/kvm/svm/svm.c
+++ b/arch/x86/kvm/svm/svm.c
@@ -3417,6 +3417,7 @@ static enum exit_fastpath_completion svm_vcpu_run(struct kvm_vcpu *vcpu)
sync_cr8_to_lapic(vcpu);
svm->next_rip = 0;
+ svm->nested.nested_run_pending = 0;
svm->vmcb->control.tlb_ctl = TLB_CONTROL_DO_NOTHING;
diff --git a/arch/x86/kvm/svm/svm.h b/arch/x86/kvm/svm/svm.h
index 98c2890d561d..435f3328c99c 100644
--- a/arch/x86/kvm/svm/svm.h
+++ b/arch/x86/kvm/svm/svm.h
@@ -97,6 +97,10 @@ struct nested_state {
/* A VMEXIT is required but not yet emulated */
bool exit_required;
+ /* A VMRUN has started but has not yet been performed, so
+ * we cannot inject a nested vmexit yet. */
+ bool nested_run_pending;
+
/* cache for intercepts of the guest */
u32 intercept_cr;
u32 intercept_dr;
--
2.18.2