diff options
| author | Wanpeng Li <wanpengli@tencent.com> | 2019-09-05 14:26:27 +0800 |
|---|---|---|
| committer | Paolo Bonzini <pbonzini@redhat.com> | 2019-09-11 18:05:29 +0200 |
| commit | 2b0911d1311e8f3a3cda472e08df21bcc49a5e61 (patch) | |
| tree | 9a70d3e16544d98c19c268db7ff85e17ebce1f4d /arch/x86/kvm/lapic.c | |
| parent | 1cfff4d9a5d01fa61e5768a6afffc81ae1c8ecb9 (diff) | |
KVM: LAPIC: Micro optimize IPI latency
This patch optimizes the virtual IPI emulation sequence:
write ICR2 write ICR2
write ICR read ICR2
read ICR ==> send virtual IPI
read ICR2 write ICR
send virtual IPI
It can reduce kvm-unit-tests/vmexit.flat IPI testing latency(from sender
send IPI to sender receive the ACK) from 3319 cycles to 3203 cycles on
SKylake server.
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Radim Krčmář <rkrcmar@redhat.com>
Signed-off-by: Wanpeng Li <wanpengli@tencent.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Diffstat (limited to 'arch/x86/kvm/lapic.c')
| -rw-r--r-- | arch/x86/kvm/lapic.c | 9 |
1 files changed, 4 insertions, 5 deletions
diff --git a/arch/x86/kvm/lapic.c b/arch/x86/kvm/lapic.c index e904ff06a83d..559e1c4c0832 100644 --- a/arch/x86/kvm/lapic.c +++ b/arch/x86/kvm/lapic.c @@ -1198,10 +1198,8 @@ void kvm_apic_set_eoi_accelerated(struct kvm_vcpu *vcpu, int vector) } EXPORT_SYMBOL_GPL(kvm_apic_set_eoi_accelerated); -static void apic_send_ipi(struct kvm_lapic *apic) +static void apic_send_ipi(struct kvm_lapic *apic, u32 icr_low, u32 icr_high) { - u32 icr_low = kvm_lapic_get_reg(apic, APIC_ICR); - u32 icr_high = kvm_lapic_get_reg(apic, APIC_ICR2); struct kvm_lapic_irq irq; irq.vector = icr_low & APIC_VECTOR_MASK; @@ -1914,8 +1912,9 @@ int kvm_lapic_reg_write(struct kvm_lapic *apic, u32 reg, u32 val) } case APIC_ICR: /* No delay here, so we always clear the pending bit */ - kvm_lapic_set_reg(apic, APIC_ICR, val & ~(1 << 12)); - apic_send_ipi(apic); + val &= ~(1 << 12); + apic_send_ipi(apic, val, kvm_lapic_get_reg(apic, APIC_ICR2)); + kvm_lapic_set_reg(apic, APIC_ICR, val); break; case APIC_ICR2: |