percpu_rw_semaphore: lockdep + config

index 592f0d6..d2146a4 100644
@@ -1,14 +1,14 @@
 #ifndef _LINUX_PERCPU_RWSEM_H
 #define _LINUX_PERCPU_RWSEM_H
 
-#include <linux/mutex.h>
+#include <linux/atomic.h>
 #include <linux/rwsem.h>
 #include <linux/percpu.h>
 #include <linux/wait.h>
 
 struct percpu_rw_semaphore {
 	unsigned int __percpu	*fast_read_ctr;
-	struct mutex		writer_mutex;
+	atomic_t		write_ctr;
 	struct rw_semaphore	rw_sem;
 	atomic_t		slow_read_ctr;
 	wait_queue_head_t	write_waitq;

index 02bd157..e5a146e 100644
@@ -1,4 +1,4 @@
-#include <linux/mutex.h>
+#include <linux/atomic.h>
 #include <linux/rwsem.h>
 #include <linux/percpu.h>
 #include <linux/wait.h>

@@ -13,8 +13,8 @@ int percpu_init_rwsem(struct percpu_rw_semaphore *brw)
 	if (unlikely(!brw->fast_read_ctr))
 		return -ENOMEM;
 
-	mutex_init(&brw->writer_mutex);
 	init_rwsem(&brw->rw_sem);
+	atomic_set(&brw->write_ctr, 0);
 	atomic_set(&brw->slow_read_ctr, 0);
 	init_waitqueue_head(&brw->write_waitq);
 	return 0;

@@ -28,7 +28,7 @@ void percpu_free_rwsem(struct percpu_rw_semaphore *brw)
 
 /*
  * This is the fast-path for down_read/up_read, it only needs to ensure
- * there is no pending writer (!mutex_is_locked() check) and inc/dec the
+ * there is no pending writer (atomic_read(write_ctr) == 0) and inc/dec the
  * fast per-cpu counter. The writer uses synchronize_sched() to serialize
  * with the preempt-disabled section below.
  *

@@ -44,7 +44,7 @@ void percpu_free_rwsem(struct percpu_rw_semaphore *brw)
  * If this helper fails the callers rely on the normal rw_semaphore and
  * atomic_dec_and_test(), so in this case we have the necessary barriers.
  *
- * But if it succeeds we do not have any barriers, mutex_is_locked() or
+ * But if it succeeds we do not have any barriers, atomic_read(write_ctr) or
  * __this_cpu_add() below can be reordered with any LOAD/STORE done by the
  * reader inside the critical section. See the comments in down_write and
  * up_write below.

@@ -54,7 +54,7 @@ static bool update_fast_ctr(struct percpu_rw_semaphore *brw, unsigned int val)
 	bool success = false;
 
 	preempt_disable();
-	if (likely(!mutex_is_locked(&brw->writer_mutex))) {
+	if (likely(!atomic_read(&brw->write_ctr))) {
 		__this_cpu_add(*brw->fast_read_ctr, val);
 		success = true;
 	}

@@ -101,9 +101,8 @@ static int clear_fast_ctr(struct percpu_rw_semaphore *brw)
 }
 
 /*
- * A writer takes ->writer_mutex to exclude other writers and to force the
- * readers to switch to the slow mode, note the mutex_is_locked() check in
- * update_fast_ctr().
+ * A writer increments ->write_ctr to force the readers to switch to the
+ * slow mode, note the atomic_read() check in update_fast_ctr().
  *
  * After that the readers can only inc/dec the slow ->slow_read_ctr counter,
  * ->fast_read_ctr is stable. Once the writer moves its sum into the slow

@@ -114,11 +113,10 @@ static int clear_fast_ctr(struct percpu_rw_semaphore *brw)
  */
 void percpu_down_write(struct percpu_rw_semaphore *brw)
 {
-	/* also blocks update_fast_ctr() which checks mutex_is_locked() */
-	mutex_lock(&brw->writer_mutex);
-
+	/* tell update_fast_ctr() there is a pending writer */
+	atomic_inc(&brw->write_ctr);
 	/*
-	 * 1. Ensures mutex_is_locked() is visible to any down_read/up_read
+	 * 1. Ensures that write_ctr != 0 is visible to any down_read/up_read
 	 *    so that update_fast_ctr() can't succeed.
 	 *
 	 * 2. Ensures we see the result of every previous this_cpu_add() in

@@ -130,25 +128,25 @@ void percpu_down_write(struct percpu_rw_semaphore *brw)
 	 */
 	synchronize_sched();
 
+	/* exclude other writers, and block the new readers completely */
+	down_write(&brw->rw_sem);
+
 	/* nobody can use fast_read_ctr, move its sum into slow_read_ctr */
 	atomic_add(clear_fast_ctr(brw), &brw->slow_read_ctr);
 
-	/* block the new readers completely */
-	down_write(&brw->rw_sem);
-
 	/* wait for all readers to complete their percpu_up_read() */
 	wait_event(brw->write_waitq, !atomic_read(&brw->slow_read_ctr));
 }
 
 void percpu_up_write(struct percpu_rw_semaphore *brw)
 {
-	/* allow the new readers, but only the slow-path */
+	/* release the lock, but the readers can't use the fast-path */
 	up_write(&brw->rw_sem);
-
 	/*
 	 * Insert the barrier before the next fast-path in down_read,
 	 * see W_R case in the comment above update_fast_ctr().
 	 */
 	synchronize_sched();
-	mutex_unlock(&brw->writer_mutex);
+	/* the last writer unblocks update_fast_ctr() */
+	atomic_dec(&brw->write_ctr);
 }

index d2146a4..3e88c9a 100644
@@ -5,6 +5,7 @@
 #include <linux/rwsem.h>
 #include <linux/percpu.h>
 #include <linux/wait.h>
+#include <linux/lockdep.h>
 
 struct percpu_rw_semaphore {
 	unsigned int __percpu	*fast_read_ctr;

@@ -20,7 +21,14 @@ extern void percpu_up_read(struct percpu_rw_semaphore *);
 extern void percpu_down_write(struct percpu_rw_semaphore *);
 extern void percpu_up_write(struct percpu_rw_semaphore *);
 
-extern int percpu_init_rwsem(struct percpu_rw_semaphore *);
+extern int __percpu_init_rwsem(struct percpu_rw_semaphore *,
+				const char *, struct lock_class_key *);
 extern void percpu_free_rwsem(struct percpu_rw_semaphore *);
 
+#define percpu_init_rwsem(brw)	\
+({								\
+	static struct lock_class_key rwsem_key;			\
+	__percpu_init_rwsem(brw, #brw, &rwsem_key);		\
+})
+
 #endif

index e5a146e..ba2708f 100644
@@ -2,18 +2,21 @@
 #include <linux/rwsem.h>
 #include <linux/percpu.h>
 #include <linux/wait.h>
+#include <linux/lockdep.h>
 #include <linux/percpu-rwsem.h>
 #include <linux/rcupdate.h>
 #include <linux/sched.h>
 #include <linux/errno.h>
 
-int percpu_init_rwsem(struct percpu_rw_semaphore *brw)
+int __percpu_init_rwsem(struct percpu_rw_semaphore *brw,
+			const char *name, struct lock_class_key *rwsem_key)
 {
 	brw->fast_read_ctr = alloc_percpu(int);
 	if (unlikely(!brw->fast_read_ctr))
 		return -ENOMEM;
 
-	init_rwsem(&brw->rw_sem);
+	/* ->rw_sem represents the whole percpu_rw_semaphore for lockdep */
+	__init_rwsem(&brw->rw_sem, name, rwsem_key);
 	atomic_set(&brw->write_ctr, 0);
 	atomic_set(&brw->slow_read_ctr, 0);
 	init_waitqueue_head(&brw->write_waitq);

@@ -66,19 +69,29 @@ static bool update_fast_ctr(struct percpu_rw_semaphore *brw, unsigned int val)
 /*
  * Like the normal down_read() this is not recursive, the writer can
  * come after the first percpu_down_read() and create the deadlock.
+ *
+ * Note: returns with lock_is_held(brw->rw_sem) == T for lockdep,
+ * percpu_up_read() does rwsem_release(). This pairs with the usage
+ * of ->rw_sem in percpu_down/up_write().
  */
 void percpu_down_read(struct percpu_rw_semaphore *brw)
 {
-	if (likely(update_fast_ctr(brw, +1)))
+	might_sleep();
+	if (likely(update_fast_ctr(brw, +1))) {
+		rwsem_acquire_read(&brw->rw_sem.dep_map, 0, 0, _RET_IP_);
 		return;
+	}
 
 	down_read(&brw->rw_sem);
 	atomic_inc(&brw->slow_read_ctr);
-	up_read(&brw->rw_sem);
+	/* avoid up_read()->rwsem_release() */
+	__up_read(&brw->rw_sem);
 }
 
 void percpu_up_read(struct percpu_rw_semaphore *brw)
 {
+	rwsem_release(&brw->rw_sem.dep_map, 1, _RET_IP_);
+
 	if (likely(update_fast_ctr(brw, -1)))
 		return;
 

@@ -1935,7 +1939,8 @@ check_prevs_add(struct task_struct *curr, struct held_lock *next)
 		 * Only non-recursive-read entries get new dependencies
 		 * added:
 		 */
-		if (hlock->read != 2) {
+		if (hlock->read != 2 &&
+		    hlock->instance->key != &__lockdep_no_validate__) {
 			if (!check_prev_add(curr, hlock, next,
 						distance, trylock_loop))
 				return 0;