KVM: nVMX: Re-evaluate L1 pending events when running L2 and L1 got posted-interrupt
authorLiran Alon <liran.alon@oracle.com>
Sun, 24 Dec 2017 16:12:55 +0000 (18:12 +0200)
committerRadim Krčmář <rkrcmar@redhat.com>
Tue, 16 Jan 2018 15:40:09 +0000 (16:40 +0100)
commitf27a85c4988d408a2918a3bcbc3d7fe4fb2dc2b3
tree6783014be25e51b83b6e66e22475b7574585f1e1
parente7387b0e27ec3a203bda4910dc4a107f6c5f912f
KVM: nVMX: Re-evaluate L1 pending events when running L2 and L1 got posted-interrupt

In case posted-interrupt was delivered to CPU while it is in host
(outside guest), then posted-interrupt delivery will be done by
calling sync_pir_to_irr() at vmentry after interrupts are disabled.

sync_pir_to_irr() will check vmx->pi_desc.control ON bit and if
set, it will sync vmx->pi_desc.pir to IRR and afterwards update RVI to
ensure virtual-interrupt-delivery will dispatch interrupt to guest.

However, it is possible that L1 will receive a posted-interrupt while
CPU runs at host and is about to enter L2. In this case, the call to
sync_pir_to_irr() will indeed update the L1's APIC IRR but
vcpu_enter_guest() will then just resume into L2 guest without
re-evaluating if it should exit from L2 to L1 as a result of this
new pending L1 event.

To address this case, if sync_pir_to_irr() has a new L1 injectable
interrupt and CPU is running L2, we force exit GUEST_MODE which will
result in another iteration of vcpu_run() run loop which will call
kvm_vcpu_running() which will call check_nested_events() which will
handle the pending L1 event properly.

Signed-off-by: Liran Alon <liran.alon@oracle.com>
Reviewed-by: Nikita Leshenko <nikita.leshchenko@oracle.com>
Reviewed-by: Krish Sadhukhan <krish.sadhukhan@oracle.com>
Reviewed-by: Liam Merwick <liam.merwick@oracle.com>
Signed-off-by: Liam Merwick <liam.merwick@oracle.com>
Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
arch/x86/kvm/vmx.c