lkml.org 
[lkml]   [2021]   [Jul]   [28]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
Patch in this message
/
From
Subject[PATCH] KVM: x86: Exit to userspace when kvm_check_nested_events fails
Date
From: Jim Mattson <jmattson@google.com>

If kvm_check_nested_events fails due to raising an
EXIT_REASON_INTERNAL_ERROR, propagate it to userspace
immediately, even if the vCPU would otherwise be sleeping.
This happens for example when the posted interrupt descriptor
points outside guest memory.

Reported-by: Jim Mattson <jmattson@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
arch/x86/kvm/x86.c | 22 +++++++++++++++-------
1 file changed, 15 insertions(+), 7 deletions(-)

diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index 348452bb16bc..916c976e99ab 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -9752,10 +9752,14 @@ static inline int vcpu_block(struct kvm *kvm, struct kvm_vcpu *vcpu)
return 1;
}

-static inline bool kvm_vcpu_running(struct kvm_vcpu *vcpu)
+static inline int kvm_vcpu_running(struct kvm_vcpu *vcpu)
{
- if (is_guest_mode(vcpu))
- kvm_check_nested_events(vcpu);
+ int r;
+ if (is_guest_mode(vcpu)) {
+ r = kvm_check_nested_events(vcpu);
+ if (r < 0 && r != -EBUSY)
+ return r;
+ }

return (vcpu->arch.mp_state == KVM_MP_STATE_RUNNABLE &&
!vcpu->arch.apf.halted);
@@ -9770,12 +9774,16 @@ static int vcpu_run(struct kvm_vcpu *vcpu)
vcpu->arch.l1tf_flush_l1d = true;

for (;;) {
- if (kvm_vcpu_running(vcpu)) {
- r = vcpu_enter_guest(vcpu);
- } else {
- r = vcpu_block(kvm, vcpu);
+ r = kvm_vcpu_running(vcpu);
+ if (r < 0) {
+ r = 0;
+ break;
}

+ if (r)
+ r = vcpu_enter_guest(vcpu);
+ else
+ r = vcpu_block(kvm, vcpu);
if (r <= 0)
break;

--
2.27.0
\
 
 \ /
  Last update: 2021-07-28 13:54    [W:0.119 / U:0.112 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site