[PATCH 5/6] KVM: VMX: Handle KVM-induced preemption timer exits in fastpath for L2

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



Eat VMX treemption timer exits in the fastpath regardless of whether L1 or
L2 is active.  The VM-Exit is 100% KVM-induced, i.e. there is nothing
directly related to the exit that KVM needs to do on behalf of the guest,
thus there is no reason to wait until the slow path to do nothing.

Opportunistically add comments explaining why preemption timer exits for
emulating the guest's APIC timer need to go down the slow path.

Signed-off-by: Sean Christopherson <seanjc@xxxxxxxxxx>
---
 arch/x86/kvm/vmx/vmx.c | 22 ++++++++++++++++++++--
 1 file changed, 20 insertions(+), 2 deletions(-)

diff --git a/arch/x86/kvm/vmx/vmx.c b/arch/x86/kvm/vmx/vmx.c
index a602c5b52c64..14658e794fbd 100644
--- a/arch/x86/kvm/vmx/vmx.c
+++ b/arch/x86/kvm/vmx/vmx.c
@@ -6010,13 +6010,26 @@ static fastpath_t handle_fastpath_preemption_timer(struct kvm_vcpu *vcpu)
 	if (vmx->req_immediate_exit)
 		return EXIT_FASTPATH_EXIT_HANDLED;
 
+	/*
+	 * If L2 is active, go down the slow path as emulating the guest timer
+	 * expiration likely requires synthesizing a nested VM-Exit.
+	 */
+	if (is_guest_mode(vcpu))
+		return EXIT_FASTPATH_NONE;
+
 	kvm_lapic_expired_hv_timer(vcpu);
 	return EXIT_FASTPATH_REENTER_GUEST;
 }
 
 static int handle_preemption_timer(struct kvm_vcpu *vcpu)
 {
-	handle_fastpath_preemption_timer(vcpu);
+	/*
+	 * This non-fastpath handler is reached if and only if the preemption
+	 * timer was being used to emulate a guest timer while L2 is active.
+	 * All other scenarios are supposed to be handled in the fastpath.
+	 */
+	WARN_ON_ONCE(!is_guest_mode(vcpu));
+	kvm_lapic_expired_hv_timer(vcpu);
 	return 1;
 }
 
@@ -7217,7 +7230,12 @@ void noinstr vmx_spec_ctrl_restore_host(struct vcpu_vmx *vmx,
 
 static fastpath_t vmx_exit_handlers_fastpath(struct kvm_vcpu *vcpu)
 {
-	if (is_guest_mode(vcpu))
+	/*
+	 * If L2 is active, some VMX preemption timer exits can be handled in
+	 * the fastpath even, all other exits must use the slow path.
+	 */
+	if (is_guest_mode(vcpu) &&
+	    to_vmx(vcpu)->exit_reason.basic != EXIT_REASON_PREEMPTION_TIMER)
 		return EXIT_FASTPATH_NONE;
 
 	switch (to_vmx(vcpu)->exit_reason.basic) {
-- 
2.43.0.472.g3155946c3a-goog





[Index of Archives]     [KVM ARM]     [KVM ia64]     [KVM ppc]     [Virtualization Tools]     [Spice Development]     [Libvirt]     [Libvirt Users]     [Linux USB Devel]     [Linux Audio Users]     [Yosemite Questions]     [Linux Kernel]     [Linux SCSI]     [XFree86]

  Powered by Linux