KVM: mmu: Add slots_arch_lock for memslot arch fields

Add a new lock to protect the arch-specific fields of memslots if they
need to be modified in a kvm->srcu read critical section. A future
commit will use this lock to lazily allocate memslot rmaps for x86.

Signed-off-by: Ben Gardon <[email protected]>
Message-Id: <[email protected]>
[Add Documentation/ hunk. - Paolo]
Signed-off-by: Paolo Bonzini <[email protected]>

Documentation/virt/kvm/locking.rst

@@ -16,6 +16,11 @@ The acquisition orders for mutexes are as follows:
 - kvm->slots_lock is taken outside kvm->irq_lock, though acquiring
   them together is quite rare.
 
+- Unlike kvm->slots_lock, kvm->slots_arch_lock is released before
+  synchronize_srcu(&kvm->srcu).  Therefore kvm->slots_arch_lock
+  can be taken inside a kvm->srcu read-side critical section,
+  while kvm->slots_lock cannot.
+
 On x86:
 
 - vcpu->mutex is taken outside kvm->arch.hyperv.hv_lock

include/linux/kvm_host.h

@@ -523,6 +523,15 @@ struct kvm {
 #endif /* KVM_HAVE_MMU_RWLOCK */
 
 	struct mutex slots_lock;
+
+	/*
+	 * Protects the arch-specific fields of struct kvm_memory_slots in
+	 * use by the VM. To be used under the slots_lock (above) or in a
+	 * kvm->srcu critical section where acquiring the slots_lock would
+	 * lead to deadlock with the synchronize_srcu in
+	 * install_new_memslots.
+	 */
+	struct mutex slots_arch_lock;
 	struct mm_struct *mm; /* userspace tied to this vm */
 	struct kvm_memslots __rcu *memslots[KVM_ADDRESS_SPACE_NUM];
 	struct kvm_vcpu *vcpus[KVM_MAX_VCPUS];

virt/kvm/kvm_main.c

@@ -909,6 +909,7 @@ static struct kvm *kvm_create_vm(unsigned long type)
 	mutex_init(&kvm->lock);
 	mutex_init(&kvm->irq_lock);
 	mutex_init(&kvm->slots_lock);
+	mutex_init(&kvm->slots_arch_lock);
 	INIT_LIST_HEAD(&kvm->devices);
 
 	BUILD_BUG_ON(KVM_MEM_SLOTS_NUM > SHRT_MAX);
@@ -1281,6 +1282,14 @@ static struct kvm_memslots *install_new_memslots(struct kvm *kvm,
 	slots->generation = gen | KVM_MEMSLOT_GEN_UPDATE_IN_PROGRESS;
 
 	rcu_assign_pointer(kvm->memslots[as_id], slots);
+
+	/*
+	 * Acquired in kvm_set_memslot. Must be released before synchronize
+	 * SRCU below in order to avoid deadlock with another thread
+	 * acquiring the slots_arch_lock in an srcu critical section.
+	 */
+	mutex_unlock(&kvm->slots_arch_lock);
+
 	synchronize_srcu_expedited(&kvm->srcu);
 
 	/*
@@ -1352,9 +1361,27 @@ static int kvm_set_memslot(struct kvm *kvm,
 	struct kvm_memslots *slots;
 	int r;
 
+	/*
+	 * Released in install_new_memslots.
+	 *
+	 * Must be held from before the current memslots are copied until
+	 * after the new memslots are installed with rcu_assign_pointer,
+	 * then released before the synchronize srcu in install_new_memslots.
+	 *
+	 * When modifying memslots outside of the slots_lock, must be held
+	 * before reading the pointer to the current memslots until after all
+	 * changes to those memslots are complete.
+	 *
+	 * These rules ensure that installing new memslots does not lose
+	 * changes made to the previous memslots.
+	 */
+	mutex_lock(&kvm->slots_arch_lock);
+
 	slots = kvm_dup_memslots(__kvm_memslots(kvm, as_id), change);
-	if (!slots)
+	if (!slots) {
+		mutex_unlock(&kvm->slots_arch_lock);
 		return -ENOMEM;
+	}
 
 	if (change == KVM_MR_DELETE || change == KVM_MR_MOVE) {
 		/*
@@ -1365,10 +1392,9 @@ static int kvm_set_memslot(struct kvm *kvm,
 		slot->flags |= KVM_MEMSLOT_INVALID;
 
 		/*
-		 * We can re-use the old memslots, the only difference from the
-		 * newly installed memslots is the invalid flag, which will get
-		 * dropped by update_memslots anyway.  We'll also revert to the
-		 * old memslots if preparing the new memory region fails.
+		 * We can re-use the memory from the old memslots.
+		 * It will be overwritten with a copy of the new memslots
+		 * after reacquiring the slots_arch_lock below.
 		 */
 		slots = install_new_memslots(kvm, as_id, slots);
 
@@ -1380,6 +1406,17 @@ static int kvm_set_memslot(struct kvm *kvm,
 		 *	- kvm_is_visible_gfn (mmu_check_root)
 		 */
 		kvm_arch_flush_shadow_memslot(kvm, slot);
+
+		/* Released in install_new_memslots. */
+		mutex_lock(&kvm->slots_arch_lock);
+
+		/*
+		 * The arch-specific fields of the memslots could have changed
+		 * between releasing the slots_arch_lock in
+		 * install_new_memslots and here, so get a fresh copy of the
+		 * slots.
+		 */
+		kvm_copy_memslots(slots, __kvm_memslots(kvm, as_id));
 	}
 
 	r = kvm_arch_prepare_memory_region(kvm, new, mem, change);
@@ -1395,8 +1432,13 @@ static int kvm_set_memslot(struct kvm *kvm,
 	return 0;
 
 out_slots:
-	if (change == KVM_MR_DELETE || change == KVM_MR_MOVE)
+	if (change == KVM_MR_DELETE || change == KVM_MR_MOVE) {
+		slot = id_to_memslot(slots, old->id);
+		slot->flags &= ~KVM_MEMSLOT_INVALID;
 		slots = install_new_memslots(kvm, as_id, slots);
+	} else {
+		mutex_unlock(&kvm->slots_arch_lock);
+	}
 	kvfree(slots);
 	return r;
 }