diff options
| author | Paolo Bonzini <pbonzini@redhat.com> | 2016-06-15 13:23:11 (GMT) |
|---|---|---|
| committer | Paolo Bonzini <pbonzini@redhat.com> | 2016-07-01 09:03:38 (GMT) |
| commit | f2485b3e0c6c0aa3a9546babc2fad3739e964ebb (patch) | |
| tree | 07e778e8128e70ad5b38458686ed0cc0f3b87d98 /arch/x86/kvm/vmx.c | |
| parent | 91fa0f8e9e2937fd9360f326ad60d51908347afd (diff) | |
| download | linux-f2485b3e0c6c0aa3a9546babc2fad3739e964ebb.tar.xz | |
KVM: x86: use guest_exit_irqoff
This gains a few clock cycles per vmexit. On Intel there is no need
anymore to enable the interrupts in vmx_handle_external_intr, since
we are using the "acknowledge interrupt on exit" feature. AMD
needs to do that, and must be careful to avoid the interrupt shadow.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Diffstat (limited to 'arch/x86/kvm/vmx.c')
| -rw-r--r-- | arch/x86/kvm/vmx.c | 4 |
1 files changed, 1 insertions, 3 deletions
diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c index 1b413a52..c1d655c 100644 --- a/arch/x86/kvm/vmx.c +++ b/arch/x86/kvm/vmx.c @@ -8574,7 +8574,6 @@ static void vmx_handle_external_intr(struct kvm_vcpu *vcpu) "push %[sp]\n\t" #endif "pushf\n\t" - "orl $0x200, (%%" _ASM_SP ")\n\t" __ASM_SIZE(push) " $%c[cs]\n\t" "call *%[entry]\n\t" : @@ -8587,8 +8586,7 @@ static void vmx_handle_external_intr(struct kvm_vcpu *vcpu) [ss]"i"(__KERNEL_DS), [cs]"i"(__KERNEL_CS) ); - } else - local_irq_enable(); + } } static bool vmx_has_high_real_mode_segbase(void) |