1 files changed, 132 insertions, 49 deletions

diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c
index 8f72526e2f68..a10af9c87f8a 100644
--- a/arch/x86/kvm/mmu.c
+++ b/arch/x86/kvm/mmu.c
@@ -214,6 +214,7 @@ static u64 __read_mostly shadow_accessed_mask;
 static u64 __read_mostly shadow_dirty_mask;
 static u64 __read_mostly shadow_mmio_mask;
 static u64 __read_mostly shadow_mmio_value;
+static u64 __read_mostly shadow_mmio_access_mask;
 static u64 __read_mostly shadow_present_mask;
 static u64 __read_mostly shadow_me_mask;
 
@@ -299,14 +300,21 @@ static void kvm_flush_remote_tlbs_with_address(struct kvm *kvm,
 	kvm_flush_remote_tlbs_with_range(kvm, &range);
 }
 
-void kvm_mmu_set_mmio_spte_mask(u64 mmio_mask, u64 mmio_value)
+void kvm_mmu_set_mmio_spte_mask(u64 mmio_mask, u64 mmio_value, u64 access_mask)
 {
+	BUG_ON((u64)(unsigned)access_mask != access_mask);
 	BUG_ON((mmio_mask & mmio_value) != mmio_value);
 	shadow_mmio_value = mmio_value | SPTE_SPECIAL_MASK;
 	shadow_mmio_mask = mmio_mask | SPTE_SPECIAL_MASK;
+	shadow_mmio_access_mask = access_mask;
 }
 EXPORT_SYMBOL_GPL(kvm_mmu_set_mmio_spte_mask);
 
+static bool is_mmio_spte(u64 spte)
+{
+	return (spte & shadow_mmio_mask) == shadow_mmio_value;
+}
+
 static inline bool sp_ad_disabled(struct kvm_mmu_page *sp)
 {
 	return sp->role.ad_disabled;
@@ -314,19 +322,19 @@ static inline bool sp_ad_disabled(struct kvm_mmu_page *sp)
 
 static inline bool spte_ad_enabled(u64 spte)
 {
-	MMU_WARN_ON((spte & shadow_mmio_mask) == shadow_mmio_value);
+	MMU_WARN_ON(is_mmio_spte(spte));
 	return !(spte & shadow_acc_track_value);
 }
 
 static inline u64 spte_shadow_accessed_mask(u64 spte)
 {
-	MMU_WARN_ON((spte & shadow_mmio_mask) == shadow_mmio_value);
+	MMU_WARN_ON(is_mmio_spte(spte));
 	return spte_ad_enabled(spte) ? shadow_accessed_mask : 0;
 }
 
 static inline u64 spte_shadow_dirty_mask(u64 spte)
 {
-	MMU_WARN_ON((spte & shadow_mmio_mask) == shadow_mmio_value);
+	MMU_WARN_ON(is_mmio_spte(spte));
 	return spte_ad_enabled(spte) ? shadow_dirty_mask : 0;
 }
 
@@ -389,7 +397,7 @@ static void mark_mmio_spte(struct kvm_vcpu *vcpu, u64 *sptep, u64 gfn,
 	u64 mask = generation_mmio_spte_mask(gen);
 	u64 gpa = gfn << PAGE_SHIFT;
 
-	access &= ACC_WRITE_MASK | ACC_USER_MASK;
+	access &= shadow_mmio_access_mask;
 	mask |= shadow_mmio_value | access;
 	mask |= gpa | shadow_nonpresent_or_rsvd_mask;
 	mask |= (gpa & shadow_nonpresent_or_rsvd_mask)
@@ -401,11 +409,6 @@ static void mark_mmio_spte(struct kvm_vcpu *vcpu, u64 *sptep, u64 gfn,
 	mmu_spte_set(sptep, mask);
 }
 
-static bool is_mmio_spte(u64 spte)
-{
-	return (spte & shadow_mmio_mask) == shadow_mmio_value;
-}
-
 static gfn_t get_mmio_spte_gfn(u64 spte)
 {
 	u64 gpa = spte & shadow_nonpresent_or_rsvd_lower_gfn_mask;
@@ -418,8 +421,7 @@ static gfn_t get_mmio_spte_gfn(u64 spte)
 
 static unsigned get_mmio_spte_access(u64 spte)
 {
-	u64 mask = generation_mmio_spte_mask(MMIO_SPTE_GEN_MASK) | shadow_mmio_mask;
-	return (spte & ~mask) & ~PAGE_MASK;
+	return spte & shadow_mmio_access_mask;
 }
 
 static bool set_mmio_spte(struct kvm_vcpu *vcpu, u64 *sptep, gfn_t gfn,
@@ -2095,6 +2097,12 @@ static struct kvm_mmu_page *kvm_mmu_alloc_page(struct kvm_vcpu *vcpu, int direct
 	if (!direct)
 		sp->gfns = mmu_memory_cache_alloc(&vcpu->arch.mmu_page_cache);
 	set_page_private(virt_to_page(sp->spt), (unsigned long)sp);
+
+	/*
+	 * active_mmu_pages must be a FIFO list, as kvm_zap_obsolete_pages()
+	 * depends on valid pages being added to the head of the list.  See
+	 * comments in kvm_zap_obsolete_pages().
+	 */
 	list_add(&sp->link, &vcpu->kvm->arch.active_mmu_pages);
 	kvm_mod_used_mmu_pages(vcpu->kvm, +1);
 	return sp;
@@ -2244,7 +2252,7 @@ static void kvm_mmu_commit_zap_page(struct kvm *kvm,
 #define for_each_valid_sp(_kvm, _sp, _gfn)				\
 	hlist_for_each_entry(_sp,					\
 	  &(_kvm)->arch.mmu_page_hash[kvm_page_table_hashfn(_gfn)], hash_link) \
-		if ((_sp)->role.invalid) {    \
+		if (is_obsolete_sp((_kvm), (_sp)) || (_sp)->role.invalid) {    \
 		} else
 
 #define for_each_gfn_indirect_valid_sp(_kvm, _sp, _gfn)			\
@@ -2301,6 +2309,11 @@ static void kvm_mmu_audit(struct kvm_vcpu *vcpu, int point) { }
 static void mmu_audit_disable(void) { }
 #endif
 
+static bool is_obsolete_sp(struct kvm *kvm, struct kvm_mmu_page *sp)
+{
+	return unlikely(sp->mmu_valid_gen != kvm->arch.mmu_valid_gen);
+}
+
 static bool kvm_sync_page(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp,
 			 struct list_head *invalid_list)
 {
@@ -2525,6 +2538,7 @@ static struct kvm_mmu_page *kvm_mmu_get_page(struct kvm_vcpu *vcpu,
 		if (level > PT_PAGE_TABLE_LEVEL && need_sync)
 			flush |= kvm_sync_pages(vcpu, gfn, &invalid_list);
 	}
+	sp->mmu_valid_gen = vcpu->kvm->arch.mmu_valid_gen;
 	clear_page(sp->spt);
 	trace_kvm_mmu_get_page(sp, true);
 
@@ -3290,7 +3304,8 @@ static bool handle_abnormal_pfn(struct kvm_vcpu *vcpu, gva_t gva, gfn_t gfn,
 	}
 
 	if (unlikely(is_noslot_pfn(pfn)))
-		vcpu_cache_mmio_info(vcpu, gva, gfn, access);
+		vcpu_cache_mmio_info(vcpu, gva, gfn,
+				     access & shadow_mmio_access_mask);
 
 	return false;
 }
@@ -3466,7 +3481,7 @@ static bool fast_page_fault(struct kvm_vcpu *vcpu, gva_t gva, int level,
 		/*
 		 * Currently, fast page fault only works for direct mapping
 		 * since the gfn is not stable for indirect shadow page. See
-		 * Documentation/virtual/kvm/locking.txt to get more detail.
+		 * Documentation/virt/kvm/locking.txt to get more detail.
 		 */
 		fault_handled = fast_pf_fix_direct_spte(vcpu, sp,
 							iterator.sptep, spte,
@@ -4233,6 +4248,13 @@ static bool fast_cr3_switch(struct kvm_vcpu *vcpu, gpa_t new_cr3,
 			return false;
 
 		if (cached_root_available(vcpu, new_cr3, new_role)) {
+			/*
+			 * It is possible that the cached previous root page is
+			 * obsolete because of a change in the MMU generation
+			 * number. However, changing the generation number is
+			 * accompanied by KVM_REQ_MMU_RELOAD, which will free
+			 * the root set here and allocate a new one.
+			 */
 			kvm_make_request(KVM_REQ_LOAD_CR3, vcpu);
 			if (!skip_tlb_flush) {
 				kvm_make_request(KVM_REQ_MMU_SYNC, vcpu);
@@ -5592,13 +5614,13 @@ slot_handle_leaf(struct kvm *kvm, struct kvm_memory_slot *memslot,
 				 PT_PAGE_TABLE_LEVEL, lock_flush_tlb);
 }
 
-static void free_mmu_pages(struct kvm_vcpu *vcpu)
+static void free_mmu_pages(struct kvm_mmu *mmu)
 {
-	free_page((unsigned long)vcpu->arch.mmu->pae_root);
-	free_page((unsigned long)vcpu->arch.mmu->lm_root);
+	free_page((unsigned long)mmu->pae_root);
+	free_page((unsigned long)mmu->lm_root);
 }
 
-static int alloc_mmu_pages(struct kvm_vcpu *vcpu)
+static int alloc_mmu_pages(struct kvm_vcpu *vcpu, struct kvm_mmu *mmu)
 {
 	struct page *page;
 	int i;
@@ -5619,9 +5641,9 @@ static int alloc_mmu_pages(struct kvm_vcpu *vcpu)
 	if (!page)
 		return -ENOMEM;
 
-	vcpu->arch.mmu->pae_root = page_address(page);
+	mmu->pae_root = page_address(page);
 	for (i = 0; i < 4; ++i)
-		vcpu->arch.mmu->pae_root[i] = INVALID_PAGE;
+		mmu->pae_root[i] = INVALID_PAGE;
 
 	return 0;
 }
@@ -5629,6 +5651,7 @@ static int alloc_mmu_pages(struct kvm_vcpu *vcpu)
 int kvm_mmu_create(struct kvm_vcpu *vcpu)
 {
 	uint i;
+	int ret;
 
 	vcpu->arch.mmu = &vcpu->arch.root_mmu;
 	vcpu->arch.walk_mmu = &vcpu->arch.root_mmu;
@@ -5646,47 +5669,106 @@ int kvm_mmu_create(struct kvm_vcpu *vcpu)
 		vcpu->arch.guest_mmu.prev_roots[i] = KVM_MMU_ROOT_INFO_INVALID;
 
 	vcpu->arch.nested_mmu.translate_gpa = translate_nested_gpa;
-	return alloc_mmu_pages(vcpu);
-}
 
-static void kvm_mmu_invalidate_zap_pages_in_memslot(struct kvm *kvm,
-			struct kvm_memory_slot *slot,
-			struct kvm_page_track_notifier_node *node)
-{
-	struct kvm_mmu_page *sp;
-	LIST_HEAD(invalid_list);
-	unsigned long i;
-	bool flush;
-	gfn_t gfn;
+	ret = alloc_mmu_pages(vcpu, &vcpu->arch.guest_mmu);
+	if (ret)
+		return ret;
 
-	spin_lock(&kvm->mmu_lock);
+	ret = alloc_mmu_pages(vcpu, &vcpu->arch.root_mmu);
+	if (ret)
+		goto fail_allocate_root;
+
+	return ret;
+ fail_allocate_root:
+	free_mmu_pages(&vcpu->arch.guest_mmu);
+	return ret;
+}
 
-	if (list_empty(&kvm->arch.active_mmu_pages))
-		goto out_unlock;
 
-	flush = slot_handle_all_level(kvm, slot, kvm_zap_rmapp, false);
+static void kvm_zap_obsolete_pages(struct kvm *kvm)
+{
+	struct kvm_mmu_page *sp, *node;
+	LIST_HEAD(invalid_list);
+	int ign;
 
-	for (i = 0; i < slot->npages; i++) {
-		gfn = slot->base_gfn + i;
+restart:
+	list_for_each_entry_safe_reverse(sp, node,
+	      &kvm->arch.active_mmu_pages, link) {
+		/*
+		 * No obsolete valid page exists before a newly created page
+		 * since active_mmu_pages is a FIFO list.
+		 */
+		if (!is_obsolete_sp(kvm, sp))
+			break;
 
-		for_each_valid_sp(kvm, sp, gfn) {
-			if (sp->gfn != gfn)
-				continue;
+		/*
+		 * Do not repeatedly zap a root page to avoid unnecessary
+		 * KVM_REQ_MMU_RELOAD, otherwise we may not be able to
+		 * progress:
+		 *    vcpu 0                        vcpu 1
+		 *                         call vcpu_enter_guest():
+		 *                            1): handle KVM_REQ_MMU_RELOAD
+		 *                                and require mmu-lock to
+		 *                                load mmu
+		 * repeat:
+		 *    1): zap root page and
+		 *        send KVM_REQ_MMU_RELOAD
+		 *
+		 *    2): if (cond_resched_lock(mmu-lock))
+		 *
+		 *                            2): hold mmu-lock and load mmu
+		 *
+		 *                            3): see KVM_REQ_MMU_RELOAD bit
+		 *                                on vcpu->requests is set
+		 *                                then return 1 to call
+		 *                                vcpu_enter_guest() again.
+		 *            goto repeat;
+		 *
+		 * Since we are reversely walking the list and the invalid
+		 * list will be moved to the head, skip the invalid page
+		 * can help us to avoid the infinity list walking.
+		 */
+		if (sp->role.invalid)
+			continue;
 
-			kvm_mmu_prepare_zap_page(kvm, sp, &invalid_list);
-		}
 		if (need_resched() || spin_needbreak(&kvm->mmu_lock)) {
-			kvm_mmu_remote_flush_or_zap(kvm, &invalid_list, flush);
-			flush = false;
+			kvm_mmu_commit_zap_page(kvm, &invalid_list);
 			cond_resched_lock(&kvm->mmu_lock);
+			goto restart;
 		}
+
+		if (__kvm_mmu_prepare_zap_page(kvm, sp, &invalid_list, &ign))
+			goto restart;
 	}
-	kvm_mmu_remote_flush_or_zap(kvm, &invalid_list, flush);
 
-out_unlock:
+	kvm_mmu_commit_zap_page(kvm, &invalid_list);
+}
+
+/*
+ * Fast invalidate all shadow pages and use lock-break technique
+ * to zap obsolete pages.
+ *
+ * It's required when memslot is being deleted or VM is being
+ * destroyed, in these cases, we should ensure that KVM MMU does
+ * not use any resource of the being-deleted slot or all slots
+ * after calling the function.
+ */
+static void kvm_mmu_zap_all_fast(struct kvm *kvm)
+{
+	spin_lock(&kvm->mmu_lock);
+	kvm->arch.mmu_valid_gen++;
+
+	kvm_zap_obsolete_pages(kvm);
 	spin_unlock(&kvm->mmu_lock);
 }
 
+static void kvm_mmu_invalidate_zap_pages_in_memslot(struct kvm *kvm,
+			struct kvm_memory_slot *slot,
+			struct kvm_page_track_notifier_node *node)
+{
+	kvm_mmu_zap_all_fast(kvm);
+}
+
 void kvm_mmu_init_vm(struct kvm *kvm)
 {
 	struct kvm_page_track_notifier_node *node = &kvm->arch.mmu_sp_tracker;
@@ -6028,7 +6110,7 @@ static void kvm_set_mmio_spte_mask(void)
 	if (IS_ENABLED(CONFIG_X86_64) && shadow_phys_bits == 52)
 		mask &= ~1ull;
 
-	kvm_mmu_set_mmio_spte_mask(mask, mask);
+	kvm_mmu_set_mmio_spte_mask(mask, mask, ACC_WRITE_MASK | ACC_USER_MASK);
 }
 
 int kvm_mmu_module_init(void)
@@ -6102,7 +6184,8 @@ unsigned long kvm_mmu_calculate_default_mmu_pages(struct kvm *kvm)
 void kvm_mmu_destroy(struct kvm_vcpu *vcpu)
 {
 	kvm_mmu_unload(vcpu);
-	free_mmu_pages(vcpu);
+	free_mmu_pages(&vcpu->arch.root_mmu);
+	free_mmu_pages(&vcpu->arch.guest_mmu);
 	mmu_free_memory_caches(vcpu);
 }