Merge tag 'locking-urgent-2024-09-29' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip

Pull locking updates from Ingo Molnar:
 "lockdep:
    - Fix potential deadlock between lockdep and RCU (Zhiguo Niu)
    - Use str_plural() to address Coccinelle warning (Thorsten Blum)
    - Add debuggability enhancement (Luis Claudio R. Goncalves)

  static keys & calls:
    - Fix static_key_slow_dec() yet again (Peter Zijlstra)
    - Handle module init failure correctly in static_call_del_module()
      (Thomas Gleixner)
    - Replace pointless WARN_ON() in static_call_module_notify() (Thomas
      Gleixner)

  <linux/cleanup.h>:
    - Add usage and style documentation (Dan Williams)

  rwsems:
    - Move is_rwsem_reader_owned() and rwsem_owner() under
      CONFIG_DEBUG_RWSEMS (Waiman Long)

  atomic ops, x86:
    - Redeclare x86_32 arch_atomic64_{add,sub}() as void (Uros Bizjak)
    - Introduce the read64_nonatomic macro to x86_32 with cx8 (Uros
      Bizjak)"

Signed-off-by: Ingo Molnar <mingo@kernel.org>

* tag 'locking-urgent-2024-09-29' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  locking/rwsem: Move is_rwsem_reader_owned() and rwsem_owner() under CONFIG_DEBUG_RWSEMS
  jump_label: Fix static_key_slow_dec() yet again
  static_call: Replace pointless WARN_ON() in static_call_module_notify()
  static_call: Handle module init failure correctly in static_call_del_module()
  locking/lockdep: Simplify character output in seq_line()
  lockdep: fix deadlock issue between lockdep and rcu
  lockdep: Use str_plural() to fix Coccinelle warning
  cleanup: Add usage and style documentation
  lockdep: suggest the fix for "lockdep bfs error:-1" on print_bfs_bug
  locking/atomic/x86: Redeclare x86_32 arch_atomic64_{add,sub}() as void
  locking/atomic/x86: Introduce the read64_nonatomic macro to x86_32 with cx8

kernel/jump_label.c

@@ -168,7 +168,7 @@ bool static_key_slow_inc_cpuslocked(struct static_key *key)
 		jump_label_update(key);
 		/*
 		 * Ensure that when static_key_fast_inc_not_disabled() or
-		 * static_key_slow_try_dec() observe the positive value,
+		 * static_key_dec_not_one() observe the positive value,
 		 * they must also observe all the text changes.
 		 */
 		atomic_set_release(&key->enabled, 1);
@@ -250,7 +250,7 @@ void static_key_disable(struct static_key *key)
 }
 EXPORT_SYMBOL_GPL(static_key_disable);
 
-static bool static_key_slow_try_dec(struct static_key *key)
+static bool static_key_dec_not_one(struct static_key *key)
 {
 	int v;
 
@@ -274,6 +274,14 @@ static bool static_key_slow_try_dec(struct static_key *key)
 		 * enabled. This suggests an ordering problem on the user side.
 		 */
 		WARN_ON_ONCE(v < 0);
+
+		/*
+		 * Warn about underflow, and lie about success in an attempt to
+		 * not make things worse.
+		 */
+		if (WARN_ON_ONCE(v == 0))
+			return true;
+
 		if (v <= 1)
 			return false;
 	} while (!likely(atomic_try_cmpxchg(&key->enabled, &v, v - 1)));
@@ -284,15 +292,27 @@ static bool static_key_slow_try_dec(struct static_key *key)
 static void __static_key_slow_dec_cpuslocked(struct static_key *key)
 {
 	lockdep_assert_cpus_held();
+	int val;
 
-	if (static_key_slow_try_dec(key))
+	if (static_key_dec_not_one(key))
 		return;
 
 	guard(mutex)(&jump_label_mutex);
-	if (atomic_cmpxchg(&key->enabled, 1, 0) == 1)
+	val = atomic_read(&key->enabled);
+	/*
+	 * It should be impossible to observe -1 with jump_label_mutex held,
+	 * see static_key_slow_inc_cpuslocked().
+	 */
+	if (WARN_ON_ONCE(val == -1))
+		return;
+	/*
+	 * Cannot already be 0, something went sideways.
+	 */
+	if (WARN_ON_ONCE(val == 0))
+		return;
+
+	if (atomic_dec_and_test(&key->enabled))
 		jump_label_update(key);
-	else
-		WARN_ON_ONCE(!static_key_slow_try_dec(key));
 }
 
 static void __static_key_slow_dec(struct static_key *key)
@@ -329,7 +349,7 @@ void __static_key_slow_dec_deferred(struct static_key *key,
 {
 	STATIC_KEY_CHECK_USE(key);
 
-	if (static_key_slow_try_dec(key))
+	if (static_key_dec_not_one(key))
 		return;
 
 	schedule_delayed_work(work, timeout);

kernel/locking/lockdep.c

@@ -788,7 +788,7 @@ static void lockdep_print_held_locks(struct task_struct *p)
 		printk("no locks held by %s/%d.\n", p->comm, task_pid_nr(p));
 	else
 		printk("%d lock%s held by %s/%d:\n", depth,
-		       depth > 1 ? "s" : "", p->comm, task_pid_nr(p));
+		       str_plural(depth), p->comm, task_pid_nr(p));
 	/*
 	 * It's not reliable to print a task's held locks if it's not sleeping
 	 * and it's not the current task.
@@ -2084,6 +2084,9 @@ static noinline void print_bfs_bug(int ret)
 	/*
 	 * Breadth-first-search failed, graph got corrupted?
 	 */
+	if (ret == BFS_EQUEUEFULL)
+		pr_warn("Increase LOCKDEP_CIRCULAR_QUEUE_BITS to avoid this warning:\n");
+
 	WARN(1, "lockdep bfs error:%d\n", ret);
 }
 
@@ -6263,25 +6266,27 @@ static struct pending_free *get_pending_free(void)
 static void free_zapped_rcu(struct rcu_head *cb);
 
 /*
- * Schedule an RCU callback if no RCU callback is pending. Must be called with
- * the graph lock held.
- */
-static void call_rcu_zapped(struct pending_free *pf)
+* See if we need to queue an RCU callback, must called with
+* the lockdep lock held, returns false if either we don't have
+* any pending free or the callback is already scheduled.
+* Otherwise, a call_rcu() must follow this function call.
+*/
+static bool prepare_call_rcu_zapped(struct pending_free *pf)
 {
 	WARN_ON_ONCE(inside_selftest());
 
 	if (list_empty(&pf->zapped))
-		return;
+		return false;
 
 	if (delayed_free.scheduled)
-		return;
+		return false;
 
 	delayed_free.scheduled = true;
 
 	WARN_ON_ONCE(delayed_free.pf + delayed_free.index != pf);
 	delayed_free.index ^= 1;
 
-	call_rcu(&delayed_free.rcu_head, free_zapped_rcu);
+	return true;
 }
 
 /* The caller must hold the graph lock. May be called from RCU context. */
@@ -6307,6 +6312,7 @@ static void free_zapped_rcu(struct rcu_head *ch)
 {
 	struct pending_free *pf;
 	unsigned long flags;
+	bool need_callback;
 
 	if (WARN_ON_ONCE(ch != &delayed_free.rcu_head))
 		return;
@@ -6318,14 +6324,18 @@ static void free_zapped_rcu(struct rcu_head *ch)
 	pf = delayed_free.pf + (delayed_free.index ^ 1);
 	__free_zapped_classes(pf);
 	delayed_free.scheduled = false;
-
-	/*
-	 * If there's anything on the open list, close and start a new callback.
-	 */
-	call_rcu_zapped(delayed_free.pf + delayed_free.index);
-
+	need_callback =
+		prepare_call_rcu_zapped(delayed_free.pf + delayed_free.index);
 	lockdep_unlock();
 	raw_local_irq_restore(flags);
+
+	/*
+	* If there's pending free and its callback has not been scheduled,
+	* queue an RCU callback.
+	*/
+	if (need_callback)
+		call_rcu(&delayed_free.rcu_head, free_zapped_rcu);
+
 }
 
 /*
@@ -6365,6 +6375,7 @@ static void lockdep_free_key_range_reg(void *start, unsigned long size)
 {
 	struct pending_free *pf;
 	unsigned long flags;
+	bool need_callback;
 
 	init_data_structures_once();
 
@@ -6372,10 +6383,11 @@ static void lockdep_free_key_range_reg(void *start, unsigned long size)
 	lockdep_lock();
 	pf = get_pending_free();
 	__lockdep_free_key_range(pf, start, size);
-	call_rcu_zapped(pf);
+	need_callback = prepare_call_rcu_zapped(pf);
 	lockdep_unlock();
 	raw_local_irq_restore(flags);
-
+	if (need_callback)
+		call_rcu(&delayed_free.rcu_head, free_zapped_rcu);
 	/*
 	 * Wait for any possible iterators from look_up_lock_class() to pass
 	 * before continuing to free the memory they refer to.
@@ -6469,6 +6481,7 @@ static void lockdep_reset_lock_reg(struct lockdep_map *lock)
 	struct pending_free *pf;
 	unsigned long flags;
 	int locked;
+	bool need_callback = false;
 
 	raw_local_irq_save(flags);
 	locked = graph_lock();
@@ -6477,11 +6490,13 @@ static void lockdep_reset_lock_reg(struct lockdep_map *lock)
 
 	pf = get_pending_free();
 	__lockdep_reset_lock(pf, lock);
-	call_rcu_zapped(pf);
+	need_callback = prepare_call_rcu_zapped(pf);
 
 	graph_unlock();
 out_irq:
 	raw_local_irq_restore(flags);
+	if (need_callback)
+		call_rcu(&delayed_free.rcu_head, free_zapped_rcu);
 }
 
 /*
@@ -6525,6 +6540,7 @@ void lockdep_unregister_key(struct lock_class_key *key)
 	struct pending_free *pf;
 	unsigned long flags;
 	bool found = false;
+	bool need_callback = false;
 
 	might_sleep();
 
@@ -6545,11 +6561,14 @@ void lockdep_unregister_key(struct lock_class_key *key)
 	if (found) {
 		pf = get_pending_free();
 		__lockdep_free_key_range(pf, key, 1);
-		call_rcu_zapped(pf);
+		need_callback = prepare_call_rcu_zapped(pf);
 	}
 	lockdep_unlock();
 	raw_local_irq_restore(flags);
 
+	if (need_callback)
+		call_rcu(&delayed_free.rcu_head, free_zapped_rcu);
+
 	/* Wait until is_dynamic_key() has finished accessing k->hash_entry. */
 	synchronize_rcu();
 }

kernel/locking/lockdep_proc.c

@@ -424,7 +424,7 @@ static void seq_line(struct seq_file *m, char c, int offset, int length)
 	for (i = 0; i < offset; i++)
 		seq_puts(m, " ");
 	for (i = 0; i < length; i++)
-		seq_printf(m, "%c", c);
+		seq_putc(m, c);
 	seq_puts(m, "\n");
 }
 

kernel/locking/rwsem.c

@@ -181,12 +181,21 @@ static inline void rwsem_set_reader_owned(struct rw_semaphore *sem)
 	__rwsem_set_reader_owned(sem, current);
 }
 
+#ifdef CONFIG_DEBUG_RWSEMS
+/*
+ * Return just the real task structure pointer of the owner
+ */
+static inline struct task_struct *rwsem_owner(struct rw_semaphore *sem)
+{
+	return (struct task_struct *)
+		(atomic_long_read(&sem->owner) & ~RWSEM_OWNER_FLAGS_MASK);
+}
+
 /*
  * Return true if the rwsem is owned by a reader.
  */
 static inline bool is_rwsem_reader_owned(struct rw_semaphore *sem)
 {
-#ifdef CONFIG_DEBUG_RWSEMS
 	/*
 	 * Check the count to see if it is write-locked.
 	 */
@@ -194,11 +203,9 @@ static inline bool is_rwsem_reader_owned(struct rw_semaphore *sem)
 
 	if (count & RWSEM_WRITER_MASK)
 		return false;
-#endif
 	return rwsem_test_oflags(sem, RWSEM_READER_OWNED);
 }
 
-#ifdef CONFIG_DEBUG_RWSEMS
 /*
  * With CONFIG_DEBUG_RWSEMS configured, it will make sure that if there
  * is a task pointer in owner of a reader-owned rwsem, it will be the
@@ -265,15 +272,6 @@ static inline bool rwsem_write_trylock(struct rw_semaphore *sem)
 	return false;
 }
 
-/*
- * Return just the real task structure pointer of the owner
- */
-static inline struct task_struct *rwsem_owner(struct rw_semaphore *sem)
-{
-	return (struct task_struct *)
-		(atomic_long_read(&sem->owner) & ~RWSEM_OWNER_FLAGS_MASK);
-}
-
 /*
  * Return the real task structure pointer of the owner and the embedded
  * flags in the owner. pflags must be non-NULL.

kernel/static_call_inline.c

@@ -411,6 +411,17 @@ static void static_call_del_module(struct module *mod)
 
 	for (site = start; site < stop; site++) {
 		key = static_call_key(site);
+
+		/*
+		 * If the key was not updated due to a memory allocation
+		 * failure in __static_call_init() then treating key::sites
+		 * as key::mods in the code below would cause random memory
+		 * access and #GP. In that case all subsequent sites have
+		 * not been touched either, so stop iterating.
+		 */
+		if (!static_call_key_has_mods(key))
+			break;
+
 		if (key == prev_key)
 			continue;
 
@@ -442,7 +453,7 @@ static int static_call_module_notify(struct notifier_block *nb,
 	case MODULE_STATE_COMING:
 		ret = static_call_add_module(mod);
 		if (ret) {
-			WARN(1, "Failed to allocate memory for static calls");
+			pr_warn("Failed to allocate memory for static calls\n");
 			static_call_del_module(mod);
 		}
 		break;