Simplify threads, mutexs, cvs and rwlocks

* Simplify threads, mutexs, cvs and rwlocks

* Update the zk_thread_create() function to use the same trick
  as Illumos.  Specifically, cast the new pthread_t to a void
  pointer and return that as the kthread_t *.  This avoids the
  issues associated with managing a wrapper structure and is
  safe as long as the callers never attempt to dereference it.

* Update all function prototypes passed to pthread_create() to
  match the expected prototype.  We were getting away this with
  before since the function were explicitly cast.

* Replaced direct zk_thread_create() calls with thread_create()
  for code consistency.  All consumers of libzpool now use the
  proper wrappers.

* The mutex_held() calls were converted to MUTEX_HELD().

* Removed all mutex_owner() calls and retired the interface.
  Instead use MUTEX_HELD() which provides the same information
  and allows the implementation details to be hidden.  In this
  case the use of the pthread_equals() function.

* The kthread_t, kmutex_t, krwlock_t, and krwlock_t types had
  any non essential fields removed.  In the case of kthread_t
  and kcondvar_t they could be directly typedef'd to pthread_t
  and pthread_cond_t respectively.

* Removed all extra ASSERTS from the thread, mutex, rwlock, and
  cv wrapper functions.  In practice, pthreads already provides
  the vast majority of checks as long as we check the return
  code.  Removing this code from our wrappers help readability.

* Added TS_JOINABLE state flag to pass to request a joinable rather
  than detached thread.  This isn't a standard thread_create() state
  but it's the least invasive way to pass this information and is
  only used by ztest.

TEST_ZTEST_TIMEOUT=3600

Chunwei Chen <tuxoko@gmail.com>
Reviewed-by: Tom Caputi <tcaputi@datto.com>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #4547 
Closes #5503 
Closes #5523 
Closes #6377 
Closes #6495

module/zfs/arc.c

@@ -4214,7 +4214,7 @@ arc_kmem_reap_now(void)
  * using mutex_tryenter() from arc_reclaim_thread().
  */
 static void
-arc_reclaim_thread(void)
+arc_reclaim_thread(void *unused)
 {
 	fstrans_cookie_t	cookie = spl_fstrans_mark();
 	hrtime_t		growtime = 0;
@@ -7515,7 +7515,7 @@ l2arc_write_buffers(spa_t *spa, l2arc_dev_t *dev, uint64_t target_sz)
  * heart of the L2ARC.
  */
 static void
-l2arc_feed_thread(void)
+l2arc_feed_thread(void *unused)
 {
 	callb_cpr_t cpr;
 	l2arc_dev_t *dev;

module/zfs/dbuf.c

@@ -531,7 +531,7 @@ dbuf_evict_one(void)
  * out of the cache it is destroyed and becomes eligible for arc eviction.
  */
 static void
-dbuf_evict_thread(void)
+dbuf_evict_thread(void *unused)
 {
 	callb_cpr_t cpr;
 

module/zfs/mmp.c

@@ -123,7 +123,7 @@ uint_t zfs_multihost_import_intervals = MMP_DEFAULT_IMPORT_INTERVALS;
  */
 uint_t zfs_multihost_fail_intervals = MMP_DEFAULT_FAIL_INTERVALS;
 
-static void mmp_thread(spa_t *spa);
+static void mmp_thread(void *arg);
 
 void
 mmp_init(spa_t *spa)
@@ -364,8 +364,9 @@ mmp_write_uberblock(spa_t *spa)
 }
 
 static void
-mmp_thread(spa_t *spa)
+mmp_thread(void *arg)
 {
+	spa_t *spa = (spa_t *)arg;
 	mmp_thread_t *mmp = &spa->spa_mmp;
 	boolean_t last_spa_suspended = spa_suspended(spa);
 	boolean_t last_spa_multihost = spa_multihost(spa);

module/zfs/spa.c

@@ -1028,6 +1028,11 @@ spa_create_zio_taskqs(spa_t *spa)
 	}
 }
 
+/*
+ * Disabled until spa_thread() can be adapted for Linux.
+ */
+#undef HAVE_SPA_THREAD
+
 #if defined(_KERNEL) && defined(HAVE_SPA_THREAD)
 static void
 spa_thread(void *arg)
@@ -3415,7 +3420,7 @@ spa_open_common(const char *pool, spa_t **spapp, void *tag, nvlist_t *nvpolicy,
 	 * up calling spa_open() again.  The real fix is to figure out how to
 	 * avoid dsl_dir_open() calling this in the first place.
 	 */
-	if (mutex_owner(&spa_namespace_lock) != curthread) {
+	if (MUTEX_NOT_HELD(&spa_namespace_lock)) {
 		mutex_enter(&spa_namespace_lock);
 		locked = B_TRUE;
 	}
@@ -6068,8 +6073,9 @@ spa_async_autoexpand(spa_t *spa, vdev_t *vd)
 }
 
 static void
-spa_async_thread(spa_t *spa)
+spa_async_thread(void *arg)
 {
+	spa_t *spa = (spa_t *)arg;
 	int tasks, i;
 
 	ASSERT(spa->spa_sync_on);

module/zfs/txg.c

@@ -108,8 +108,8 @@
  * now transition to the syncing state.
  */
 
-static void txg_sync_thread(dsl_pool_t *dp);
-static void txg_quiesce_thread(dsl_pool_t *dp);
+static void txg_sync_thread(void *dp);
+static void txg_quiesce_thread(void *dp);
 
 int zfs_txg_timeout = 5;	/* max seconds worth of delta per txg */
 
@@ -477,8 +477,9 @@ txg_wait_callbacks(dsl_pool_t *dp)
 }
 
 static void
-txg_sync_thread(dsl_pool_t *dp)
+txg_sync_thread(void *arg)
 {
+	dsl_pool_t *dp = (dsl_pool_t *)arg;
 	spa_t *spa = dp->dp_spa;
 	tx_state_t *tx = &dp->dp_tx;
 	callb_cpr_t cpr;
@@ -561,8 +562,9 @@ txg_sync_thread(dsl_pool_t *dp)
 }
 
 static void
-txg_quiesce_thread(dsl_pool_t *dp)
+txg_quiesce_thread(void *arg)
 {
+	dsl_pool_t *dp = (dsl_pool_t *)arg;
 	tx_state_t *tx = &dp->dp_tx;
 	callb_cpr_t cpr;
 

module/zfs/zfs_sa.c

@@ -300,7 +300,7 @@ zfs_sa_upgrade(sa_handle_t *hdl, dmu_tx_t *tx)
 	 * Otherwise, we know we are doing the
 	 * sa_update() that caused us to enter this function.
 	 */
-	if (mutex_owner(&zp->z_lock) != curthread) {
+	if (MUTEX_NOT_HELD(&zp->z_lock)) {
 		if (mutex_tryenter(&zp->z_lock) == 0)
 			return;
 		else