KVM: x86/mmu: fast_page_fault support for the TDP MMU

Make fast_page_fault interoperate with the TDP MMU by leveraging
walk_shadow_page_lockless_{begin,end} to acquire the RCU read lock and
introducing a new helper function kvm_tdp_mmu_fast_pf_get_last_sptep to
grab the lowest level sptep.

Suggested-by: Ben Gardon <bgardon@google.com>
Signed-off-by: David Matlack <dmatlack@google.com>
Message-Id: <20210713220957.3493520-5-dmatlack@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

3 files changed, 80 insertions, 13 deletions

diff --git a/arch/x86/kvm/mmu/mmu.c b/arch/x86/kvm/mmu/mmu.c
index d5b0c8b0e9e9..29010abb659c 100644
--- a/arch/x86/kvm/mmu/mmu.c
+++ b/arch/x86/kvm/mmu/mmu.c
@@ -3109,14 +3109,40 @@ static bool is_access_allowed(u32 fault_err_code, u64 spte)
 }
 
 /*
+ * Returns the last level spte pointer of the shadow page walk for the given
+ * gpa, and sets *spte to the spte value. This spte may be non-preset. If no
+ * walk could be performed, returns NULL and *spte does not contain valid data.
+ *
+ * Contract:
+ *  - Must be called between walk_shadow_page_lockless_{begin,end}.
+ *  - The returned sptep must not be used after walk_shadow_page_lockless_end.
+ */
+static u64 *fast_pf_get_last_sptep(struct kvm_vcpu *vcpu, gpa_t gpa, u64 *spte)
+{
+	struct kvm_shadow_walk_iterator iterator;
+	u64 old_spte;
+	u64 *sptep = NULL;
+
+	for_each_shadow_entry_lockless(vcpu, gpa, iterator, old_spte) {
+		sptep = iterator.sptep;
+		*spte = old_spte;
+
+		if (!is_shadow_present_pte(old_spte))
+			break;
+	}
+
+	return sptep;
+}
+
+/*
  * Returns one of RET_PF_INVALID, RET_PF_FIXED or RET_PF_SPURIOUS.
  */
 static int fast_page_fault(struct kvm_vcpu *vcpu, gpa_t gpa, u32 error_code)
 {
-	struct kvm_shadow_walk_iterator iterator;
 	struct kvm_mmu_page *sp;
 	int ret = RET_PF_INVALID;
 	u64 spte = 0ull;
+	u64 *sptep = NULL;
 	uint retry_count = 0;
 
 	if (!page_fault_can_be_fast(error_code))
@@ -3127,14 +3153,15 @@ static int fast_page_fault(struct kvm_vcpu *vcpu, gpa_t gpa, u32 error_code)
 	do {
 		u64 new_spte;
 
-		for_each_shadow_entry_lockless(vcpu, gpa, iterator, spte)
-			if (!is_shadow_present_pte(spte))
-				break;
+		if (is_tdp_mmu(vcpu->arch.mmu))
+			sptep = kvm_tdp_mmu_fast_pf_get_last_sptep(vcpu, gpa, &spte);
+		else
+			sptep = fast_pf_get_last_sptep(vcpu, gpa, &spte);
 
 		if (!is_shadow_present_pte(spte))
 			break;
 
-		sp = sptep_to_sp(iterator.sptep);
+		sp = sptep_to_sp(sptep);
 		if (!is_last_spte(spte, sp->role.level))
 			break;
 
@@ -3192,8 +3219,7 @@ static int fast_page_fault(struct kvm_vcpu *vcpu, gpa_t gpa, u32 error_code)
 		 * since the gfn is not stable for indirect shadow page. See
 		 * Documentation/virt/kvm/locking.rst to get more detail.
 		 */
-		if (fast_pf_fix_direct_spte(vcpu, sp, iterator.sptep, spte,
-					    new_spte)) {
+		if (fast_pf_fix_direct_spte(vcpu, sp, sptep, spte, new_spte)) {
 			ret = RET_PF_FIXED;
 			break;
 		}
@@ -3206,7 +3232,7 @@ static int fast_page_fault(struct kvm_vcpu *vcpu, gpa_t gpa, u32 error_code)
 
 	} while (true);
 
-	trace_fast_page_fault(vcpu, gpa, error_code, iterator.sptep, spte, ret);
+	trace_fast_page_fault(vcpu, gpa, error_code, sptep, spte, ret);
 	walk_shadow_page_lockless_end(vcpu);
 
 	return ret;
@@ -3841,11 +3867,9 @@ static int direct_page_fault(struct kvm_vcpu *vcpu, gpa_t gpa, u32 error_code,
 	if (page_fault_handle_page_track(vcpu, error_code, gfn))
 		return RET_PF_EMULATE;
 
-	if (!is_tdp_mmu_fault) {
-		r = fast_page_fault(vcpu, gpa, error_code);
-		if (r != RET_PF_INVALID)
-			return r;
-	}
+	r = fast_page_fault(vcpu, gpa, error_code);
+	if (r != RET_PF_INVALID)
+		return r;
 
 	r = mmu_topup_memory_caches(vcpu, false);
 	if (r)
diff --git a/arch/x86/kvm/mmu/tdp_mmu.c b/arch/x86/kvm/mmu/tdp_mmu.c
index 228f0cc5e2cf..88aff5d0512c 100644
--- a/arch/x86/kvm/mmu/tdp_mmu.c
+++ b/arch/x86/kvm/mmu/tdp_mmu.c
@@ -527,6 +527,10 @@ static inline bool tdp_mmu_set_spte_atomic_no_dirty_log(struct kvm *kvm,
 	if (is_removed_spte(iter->old_spte))
 		return false;
 
+	/*
+	 * Note, fast_pf_fix_direct_spte() can also modify TDP MMU SPTEs and
+	 * does not hold the mmu_lock.
+	 */
 	if (cmpxchg64(rcu_dereference(iter->sptep), iter->old_spte,
 		      new_spte) != iter->old_spte)
 		return false;
@@ -1536,3 +1540,40 @@ int kvm_tdp_mmu_get_walk(struct kvm_vcpu *vcpu, u64 addr, u64 *sptes,
 
 	return leaf;
 }
+
+/*
+ * Returns the last level spte pointer of the shadow page walk for the given
+ * gpa, and sets *spte to the spte value. This spte may be non-preset. If no
+ * walk could be performed, returns NULL and *spte does not contain valid data.
+ *
+ * Contract:
+ *  - Must be called between kvm_tdp_mmu_walk_lockless_{begin,end}.
+ *  - The returned sptep must not be used after kvm_tdp_mmu_walk_lockless_end.
+ *
+ * WARNING: This function is only intended to be called during fast_page_fault.
+ */
+u64 *kvm_tdp_mmu_fast_pf_get_last_sptep(struct kvm_vcpu *vcpu, u64 addr,
+					u64 *spte)
+{
+	struct tdp_iter iter;
+	struct kvm_mmu *mmu = vcpu->arch.mmu;
+	gfn_t gfn = addr >> PAGE_SHIFT;
+	tdp_ptep_t sptep = NULL;
+
+	tdp_mmu_for_each_pte(iter, mmu, gfn, gfn + 1) {
+		*spte = iter.old_spte;
+		sptep = iter.sptep;
+	}
+
+	/*
+	 * Perform the rcu_dereference to get the raw spte pointer value since
+	 * we are passing it up to fast_page_fault, which is shared with the
+	 * legacy MMU and thus does not retain the TDP MMU-specific __rcu
+	 * annotation.
+	 *
+	 * This is safe since fast_page_fault obeys the contracts of this
+	 * function as well as all TDP MMU contracts around modifying SPTEs
+	 * outside of mmu_lock.
+	 */
+	return rcu_dereference(sptep);
+}
diff --git a/arch/x86/kvm/mmu/tdp_mmu.h b/arch/x86/kvm/mmu/tdp_mmu.h
index 93e1bf5089c4..361b47f98cc5 100644
--- a/arch/x86/kvm/mmu/tdp_mmu.h
+++ b/arch/x86/kvm/mmu/tdp_mmu.h
@@ -89,6 +89,8 @@ static inline void kvm_tdp_mmu_walk_lockless_end(void)
 
 int kvm_tdp_mmu_get_walk(struct kvm_vcpu *vcpu, u64 addr, u64 *sptes,
 			 int *root_level);
+u64 *kvm_tdp_mmu_fast_pf_get_last_sptep(struct kvm_vcpu *vcpu, u64 addr,
+					u64 *spte);
 
 #ifdef CONFIG_X86_64
 bool kvm_mmu_init_tdp_mmu(struct kvm *kvm);