Wanpeng, On Tue, 2 Jul 2019, Wanpeng Li wrote: > On Tue, 2 Jul 2019 at 06:44, Thomas Gleixner <tglx@xxxxxxxxxxxxx> wrote: > > > > While that CPU0 hotplug test case is surely an esoteric issue, the APIC > > emulation is still wrong, Even if the play_dead() code would not enable > > interrupts then the pending IRR bit would turn into an ISR .. interrupt > > when the APIC is reenabled on startup. > > >From SDM 10.4.7.2 Local APIC State After It Has Been Software Disabled > * Pending interrupts in the IRR and ISR registers are held and require > masking or handling by the CPU. Correct. > In your testing, hardware cpu will not respect soft disable APIC when > IRR has already been set or APICv posted-interrupt is in flight, so we > can skip soft disable APIC checking when clearing IRR and set ISR, > continue to respect soft disable APIC when attempting to set IRR. > Could you try below fix? > diff --git a/arch/x86/kvm/lapic.c b/arch/x86/kvm/lapic.c > index 05d8934..f857a12 100644 > --- a/arch/x86/kvm/lapic.c > +++ b/arch/x86/kvm/lapic.c > @@ -2376,7 +2376,7 @@ int kvm_apic_has_interrupt(struct kvm_vcpu *vcpu) > struct kvm_lapic *apic = vcpu->arch.apic; > u32 ppr; > > - if (!apic_enabled(apic)) > + if (!kvm_apic_hw_enabled(apic)) > return -1; > > __apic_update_ppr(apic, &ppr); Yes. That fixes it and works as expected. Thanks for the quick resolution. I surely stared at that function, but was not sure how to fix it proper. Tested-by: Thomas Gleixner <tglx@xxxxxxxxxxxxx> Please add a Cc: stable... tag when you post the patch. Thanks, tglx
