KVM: Defer remote tlb flushes on invlpg

Commit Message

Avi Kivity March 16, 2009, 10:45 a.m. UTC

KVM flushes tlbs on remote cpus for two purposes: to protect guest pages
that it needs to collect information about, and to prevent stale tlb entries
from pointing to pages that no longer belong to the guest.

We can defer the latter flushes to the point when we actually free the pages,
which is during an mmu notifier invocation.  To this end, we add a new state
remote_tlbs_dirty which marks whether the guest tlb might be inconsistent
with the the shadow page tables.  Whenever we do a conditional flush of
remote tlbs, we check this state, and if the remote tlbs are dirty we flush
them to ensure no inconsistency.

Signed-off-by: Avi Kivity <[e-mail]>
---
 arch/x86/include/asm/kvm_host.h |    1 -
 arch/x86/kvm/mmu.c              |    4 ++--
 arch/x86/kvm/paging_tmpl.h      |    6 ++++--
 include/linux/kvm_host.h        |    1 +
 virt/kvm/kvm_main.c             |    8 ++++++--
 5 files changed, 13 insertions(+), 7 deletions(-)

Patch

diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h
index 4627627..eb1ab29 100644
--- a/arch/x86/include/asm/kvm_host.h
+++ b/arch/x86/include/asm/kvm_host.h
@@ -383,7 +383,6 @@  struct kvm_mem_alias {
 struct kvm_arch{
 	int naliases;
 	struct kvm_mem_alias aliases[KVM_ALIAS_SLOTS];
-
 	unsigned int n_free_mmu_pages;
 	unsigned int n_requested_mmu_pages;
 	unsigned int n_alloc_mmu_pages;
diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c
index 2a36f7f..a10c96b 100644
--- a/arch/x86/kvm/mmu.c
+++ b/arch/x86/kvm/mmu.c
@@ -675,7 +675,7 @@  static int rmap_write_protect(struct kvm *kvm, u64 gfn)
 		spte = rmap_next(kvm, rmapp, spte);
 	}
 
-	return write_protected;
+	return write_protected || kvm->remote_tlbs_dirty;
 }
 
 static int kvm_unmap_rmapp(struct kvm *kvm, unsigned long *rmapp)
@@ -690,7 +690,7 @@  static int kvm_unmap_rmapp(struct kvm *kvm, unsigned long *rmapp)
 		set_shadow_pte(spte, shadow_trap_nonpresent_pte);
 		need_tlb_flush = 1;
 	}
-	return need_tlb_flush;
+	return need_tlb_flush || kvm->remote_tlbs_dirty;
 }
 
 static int kvm_handle_hva(struct kvm *kvm, unsigned long hva,
diff --git a/arch/x86/kvm/paging_tmpl.h b/arch/x86/kvm/paging_tmpl.h
index 855eb71..18abdf9 100644
--- a/arch/x86/kvm/paging_tmpl.h
+++ b/arch/x86/kvm/paging_tmpl.h
@@ -475,8 +475,10 @@  static void FNAME(invlpg)(struct kvm_vcpu *vcpu, gva_t gva)
 			break;
 	}
 
-	if (need_flush)
-		kvm_flush_remote_tlbs(vcpu->kvm);
+	if (need_flush) {
+		vcpu->kvm->remote_tlbs_dirty = true;
+		kvm_x86_ops->tlb_flush(vcpu);
+	}
 	spin_unlock(&vcpu->kvm->mmu_lock);
 
 	if (pte_gpa == -1)
diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h
index 11eb702..e7813bd 100644
--- a/include/linux/kvm_host.h
+++ b/include/linux/kvm_host.h
@@ -125,6 +125,7 @@  struct kvm_kernel_irq_routing_entry {
 struct kvm {
 	struct mutex lock; /* protects the vcpus array and APIC accesses */
 	spinlock_t mmu_lock;
+	bool remote_tlbs_dirty;
 	struct rw_semaphore slots_lock;
 	struct mm_struct *mm; /* userspace tied to this vm */
 	int nmemslots;
diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
index 68b217e..8b48891 100644
--- a/virt/kvm/kvm_main.c
+++ b/virt/kvm/kvm_main.c
@@ -758,6 +758,8 @@  static bool make_all_cpus_request(struct kvm *kvm, unsigned int req)
 
 void kvm_flush_remote_tlbs(struct kvm *kvm)
 {
+	kvm->remote_tlbs_dirty = false;
+	wmb();
 	if (make_all_cpus_request(kvm, KVM_REQ_TLB_FLUSH))
 		++kvm->stat.remote_tlb_flush;
 }
@@ -840,8 +842,9 @@  static void kvm_mmu_notifier_invalidate_page(struct mmu_notifier *mn,
 	need_tlb_flush = kvm_unmap_hva(kvm, address);
 	spin_unlock(&kvm->mmu_lock);
 
+	rmb();
 	/* we've to flush the tlb before the pages can be freed */
-	if (need_tlb_flush)
+	if (need_tlb_flush || kvm->remote_tlbs_dirty)
 		kvm_flush_remote_tlbs(kvm);
 
 }
@@ -865,8 +868,9 @@  static void kvm_mmu_notifier_invalidate_range_start(struct mmu_notifier *mn,
 		need_tlb_flush |= kvm_unmap_hva(kvm, start);
 	spin_unlock(&kvm->mmu_lock);
 
+	rmb();
 	/* we've to flush the tlb before the pages can be freed */
-	if (need_tlb_flush)
+	if (need_tlb_flush || kvm->remote_tlbs_dirty)
 		kvm_flush_remote_tlbs(kvm);
 }