From: David Woodhouse <dwmw@xxxxxxxxxxxx>
Booting a guest with 'noapic' with split irqchip and APICv enabled leads
to a livelock in vcpu_run().
When the userspace PIC has an IRQ to deliver, the VMM sets
kvm_run->request_interrupt_window and vcpu_enter_guest() duly enables the
corresponding VMexit, which happens immediately.
However, if an interrupt also exists in the local APIC, that causes
kvm_cpu_has_interrupt() to be true and thus causes
kvm_vcpu_ready_for_interrupt_injection() to return false; the intent
being that the kernel will use this interrupt window to inject its own
interrupt from the LAPIC. So vcpu_run() doesn't exit all the way to
userspace, and just loops.
However, when APICv is in use there is no interrupt to be injected since
that happens automatically. So the guest vCPU is launched again, exits
again immediately, and the loop repeats ad infinitum without making any
progress.
It looks like this was introduced in commit 782d422bcaee, when
dm_request_for_irq_injection() started returning true based purely
on the fact that userspace had requested the interrupt window, without
heed to kvm_cpu_has_interrupt() also being true.
Fix it by allowing userspace to use the interrupt window with priority
over the interrupt that is already in the LAPIC, for both APICv and
legacy injection alike. Instead of '!kvm_cpu_has_interrupt()', the
condition becomes '!(lapic_in_kernel(vcpu) && kvm_cpu_has_extint(vcpu))'
Fixes: 782d422bcaee ("KVM: x86: split kvm_vcpu_ready_for_interrupt_injection out of dm_request_for_irq_injection")
Signed-off-by: David Woodhouse <dwmw@xxxxxxxxxxxx>
Cc: stable@xxxxxxxxxxxxxxx
---
arch/x86/include/asm/kvm_host.h | 1 +
arch/x86/kvm/irq.c | 2 +-
arch/x86/kvm/x86.c | 6 ++++--
3 files changed, 6 insertions(+), 3 deletions(-)
diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h
index f002cdb13a0b..e85e2f1c661c 100644
--- a/arch/x86/include/asm/kvm_host.h
+++ b/arch/x86/include/asm/kvm_host.h
@@ -1656,6 +1656,7 @@ int kvm_unmap_hva_range(struct kvm *kvm, unsigned long start, unsigned long end,
int kvm_age_hva(struct kvm *kvm, unsigned long start, unsigned long end);
int kvm_test_age_hva(struct kvm *kvm, unsigned long hva);
int kvm_set_spte_hva(struct kvm *kvm, unsigned long hva, pte_t pte);
+int kvm_cpu_has_extint(struct kvm_vcpu *v);
int kvm_cpu_has_injectable_intr(struct kvm_vcpu *v);
int kvm_cpu_has_interrupt(struct kvm_vcpu *vcpu);
int kvm_arch_interrupt_allowed(struct kvm_vcpu *vcpu);
diff --git a/arch/x86/kvm/irq.c b/arch/x86/kvm/irq.c
index 99d118ffc67d..9c4ef1682b81 100644
--- a/arch/x86/kvm/irq.c
+++ b/arch/x86/kvm/irq.c
@@ -40,7 +40,7 @@ static int pending_userspace_extint(struct kvm_vcpu *v)
* check if there is pending interrupt from
* non-APIC source without intack.
*/
-static int kvm_cpu_has_extint(struct kvm_vcpu *v)
+int kvm_cpu_has_extint(struct kvm_vcpu *v)
{
u8 accept = kvm_apic_accept_pic_intr(v);
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index a3fdc16cfd6f..b50ae8ba66e9 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -4080,12 +4080,14 @@ static int kvm_cpu_accept_dm_intr(struct kvm_vcpu *vcpu)
* if userspace requested an interrupt window, check that the
* interrupt window is open.
*
- * No need to exit to userspace if we already have an interrupt queued.
+ * If there is already an event which needs reinjection or a
+ * pending ExtINT, allow that to be processed by the kernel
+ * before letting userspace have the opportunity.
*/
static int kvm_vcpu_ready_for_interrupt_injection(struct kvm_vcpu *vcpu)
{
return kvm_arch_interrupt_allowed(vcpu) &&
- !kvm_cpu_has_interrupt(vcpu) &&
+ !(lapic_in_kernel(vcpu) && kvm_cpu_has_extint(vcpu)) &&
!kvm_event_needs_reinjection(vcpu) &&
kvm_cpu_accept_dm_intr(vcpu);
}
--
2.17.1
Attachment:
smime.p7s
Description: S/MIME cryptographic signature
