[PATCH mm-unstable v2 09/10] kvm/x86: add kvm_arch_test_clear_young()

[Yu Zhao]

Implement kvm_arch_test_clear_young() to support the fast path in
mmu_notifier_ops->test_clear_young().

It focuses on a simple case, i.e., TDP MMU sets the accessed bit in
KVM PTEs and VMs are not nested, where it can rely on RCU and
clear_bit() to safely clear the accessed bit without taking
kvm->mmu_lock. Complex cases fall back to the existing slow path
where kvm->mmu_lock is then taken.

Signed-off-by: Yu Zhao <yuzhao@xxxxxxxxxx>
---
 arch/x86/include/asm/kvm_host.h |  7 +++++++
 arch/x86/kvm/mmu/tdp_mmu.c      | 34 +++++++++++++++++++++++++++++++++
 2 files changed, 41 insertions(+)

diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h
index 753c67072c47..d6dfdebe3d94 100644
--- a/arch/x86/include/asm/kvm_host.h
+++ b/arch/x86/include/asm/kvm_host.h
@@ -2223,4 +2223,11 @@ int memslot_rmap_alloc(struct kvm_memory_slot *slot, unsigned long npages);
  */
 #define KVM_EXIT_HYPERCALL_MBZ		GENMASK_ULL(31, 1)
 
+#define kvm_arch_has_test_clear_young kvm_arch_has_test_clear_young
+static inline bool kvm_arch_has_test_clear_young(void)
+{
+	return IS_ENABLED(CONFIG_X86_64) &&
+	       (!IS_REACHABLE(CONFIG_KVM) || (tdp_mmu_enabled && shadow_accessed_mask));
+}
+
 #endif /* _ASM_X86_KVM_HOST_H */
diff --git a/arch/x86/kvm/mmu/tdp_mmu.c b/arch/x86/kvm/mmu/tdp_mmu.c
index 08340219c35a..6875a819e007 100644
--- a/arch/x86/kvm/mmu/tdp_mmu.c
+++ b/arch/x86/kvm/mmu/tdp_mmu.c
@@ -1232,6 +1232,40 @@ bool kvm_tdp_mmu_test_age_gfn(struct kvm *kvm, struct kvm_gfn_range *range)
 	return kvm_tdp_mmu_handle_gfn(kvm, range, test_age_gfn);
 }
 
+bool kvm_arch_test_clear_young(struct kvm *kvm, struct kvm_gfn_range *range)
+{
+	struct kvm_mmu_page *root;
+	int offset = ffs(shadow_accessed_mask) - 1;
+
+	if (kvm_shadow_root_allocated(kvm))
+		return true;
+
+	rcu_read_lock();
+
+	list_for_each_entry_rcu(root, &kvm->arch.tdp_mmu_roots, link) {
+		struct tdp_iter iter;
+
+		if (kvm_mmu_page_as_id(root) != range->slot->as_id)
+			continue;
+
+		tdp_root_for_each_leaf_pte(iter, root, range->start, range->end) {
+			u64 *sptep = rcu_dereference(iter.sptep);
+
+			VM_WARN_ON_ONCE(!page_count(virt_to_page(sptep)));
+
+			if (!(iter.old_spte & shadow_accessed_mask))
+				continue;
+
+			if (kvm_should_clear_young(range, iter.gfn))
+				clear_bit(offset, (unsigned long *)sptep);
+		}
+	}
+
+	rcu_read_unlock();
+
+	return false;
+}
+
 static bool set_spte_gfn(struct kvm *kvm, struct tdp_iter *iter,
 			 struct kvm_gfn_range *range)
 {
-- 
2.41.0.rc0.172.g3f132b7071-goog