KVM: X86: support paravirtualized help for TLB shootdowns

Web:        https://git.kernel.org/torvalds/c/f38a7b75267f1fb240a8178cbcb16d66dd37aac8
Commit:     f38a7b75267f1fb240a8178cbcb16d66dd37aac8
Parent:     c2ba05ccfde2f069a66c0462e5b5ef8a517dcc9c
Refname:    refs/heads/master
Author:     Wanpeng Li <[email protected]>
AuthorDate: Tue Dec 12 17:33:04 2017 -0800
Committer:  Radim Krčmář <[email protected]>
CommitDate: Tue Jan 16 16:34:13 2018 +0100

    KVM: X86: support paravirtualized help for TLB shootdowns
    
    When running on a virtual machine, IPIs are expensive when the target
    CPU is sleeping.  Thus, it is nice to be able to avoid them for TLB
    shootdowns.  KVM can just do the flush via INVVPID on the guest's behalf
    the next time the CPU is scheduled.
    
    Cc: Paolo Bonzini <[email protected]>
    Cc: Radim Krčmář <[email protected]>
    Cc: Peter Zijlstra <[email protected]>
    Signed-off-by: Wanpeng Li <[email protected]>
    [Use "&" to test the bit instead of "==". - Paolo]
    Signed-off-by: Paolo Bonzini <[email protected]>
    Signed-off-by: Radim Krčmář <[email protected]>
---
 arch/x86/kvm/cpuid.c |  3 ++-
 arch/x86/kvm/x86.c   | 19 ++++++++++++-------
 2 files changed, 14 insertions(+), 8 deletions(-)

diff --git a/arch/x86/kvm/cpuid.c b/arch/x86/kvm/cpuid.c
index 3db9e1cd4f63..a0e6c975f3a8 100644
--- a/arch/x86/kvm/cpuid.c
+++ b/arch/x86/kvm/cpuid.c
@@ -602,7 +602,8 @@ static inline int __do_cpuid_ent(struct kvm_cpuid_entry2 *entry, u32 function,
 			     (1 << KVM_FEATURE_ASYNC_PF) |
 			     (1 << KVM_FEATURE_PV_EOI) |
 			     (1 << KVM_FEATURE_CLOCKSOURCE_STABLE_BIT) |
-			     (1 << KVM_FEATURE_PV_UNHALT);
+			     (1 << KVM_FEATURE_PV_UNHALT) |
+			     (1 << KVM_FEATURE_PV_TLB_FLUSH);
 
 		if (sched_info_on())
 			entry->eax |= (1 << KVM_FEATURE_STEAL_TIME);
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index 1e6d9f2d1f0b..35d0b3de697b 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -2122,6 +2122,12 @@ static void kvmclock_reset(struct kvm_vcpu *vcpu)
 	vcpu->arch.pv_time_enabled = false;
 }
 
+static void kvm_vcpu_flush_tlb(struct kvm_vcpu *vcpu, bool invalidate_gpa)
+{
+	++vcpu->stat.tlb_flush;
+	kvm_x86_ops->tlb_flush(vcpu, invalidate_gpa);
+}
+
 static void record_steal_time(struct kvm_vcpu *vcpu)
 {
 	if (!(vcpu->arch.st.msr_val & KVM_MSR_ENABLED))
@@ -2131,7 +2137,12 @@ static void record_steal_time(struct kvm_vcpu *vcpu)
 		&vcpu->arch.st.steal, sizeof(struct kvm_steal_time))))
 		return;
 
-	vcpu->arch.st.steal.preempted = 0;
+	/*
+	 * Doing a TLB flush here, on the guest's behalf, can avoid
+	 * expensive IPIs.
+	 */
+	if (xchg(&vcpu->arch.st.steal.preempted, 0) & KVM_VCPU_FLUSH_TLB)
+		kvm_vcpu_flush_tlb(vcpu, false);
 
 	if (vcpu->arch.st.steal.version & 1)
 		vcpu->arch.st.steal.version += 1;  /* first time write, random junk */
@@ -6781,12 +6792,6 @@ static void vcpu_scan_ioapic(struct kvm_vcpu *vcpu)
 	kvm_x86_ops->load_eoi_exitmap(vcpu, eoi_exit_bitmap);
 }
 
-static void kvm_vcpu_flush_tlb(struct kvm_vcpu *vcpu, bool invalidate_gpa)
-{
-	++vcpu->stat.tlb_flush;
-	kvm_x86_ops->tlb_flush(vcpu, invalidate_gpa);
-}