[illumos-gate merge]

commit cab3a55e158118937e07d059c46f1bc14d1f254d
    9962 zil_commit should omit cache thrash

6 files changed, 203 insertions, 76 deletions

diff --git a/usr/src/cmd/ztest/ztest.c b/usr/src/cmd/ztest/ztest.c
index ca1c7a3cc0..46a8dd2c4e 100644
--- a/usr/src/cmd/ztest/ztest.c
+++ b/usr/src/cmd/ztest/ztest.c
@@ -1840,6 +1840,7 @@ zil_replay_func_t *ztest_replay_vector[TX_MAX_TYPE] = {
  * ZIL get_data callbacks
  */
 
+/* ARGSUSED */
 static void
 ztest_get_done(zgd_t *zgd, int error)
 {
@@ -1852,9 +1853,6 @@ ztest_get_done(zgd_t *zgd, int error)
 	ztest_range_unlock(zgd->zgd_rl);
 	ztest_object_unlock(zd, object);
 
-	if (error == 0 && zgd->zgd_bp)
-		zil_lwb_add_block(zgd->zgd_lwb, zgd->zgd_bp);
-
 	umem_free(zgd, sizeof (*zgd));
 }
 
diff --git a/usr/src/uts/common/fs/zfs/dmu.c b/usr/src/uts/common/fs/zfs/dmu.c
index ce0c42d1ee..44157a3f62 100644
--- a/usr/src/uts/common/fs/zfs/dmu.c
+++ b/usr/src/uts/common/fs/zfs/dmu.c
@@ -1706,6 +1706,15 @@ dmu_sync_done(zio_t *zio, arc_buf_t *buf, void *varg)
 	dmu_sync_arg_t *dsa = varg;
 	dbuf_dirty_record_t *dr = dsa->dsa_dr;
 	dmu_buf_impl_t *db = dr->dr_dbuf;
+	zgd_t *zgd = dsa->dsa_zgd;
+
+	/*
+	 * Record the vdev(s) backing this blkptr so they can be flushed after
+	 * the writes for the lwb have completed.
+	 */
+	if (zio->io_error == 0) {
+		zil_lwb_add_block(zgd->zgd_lwb, zgd->zgd_bp);
+	}
 
 	mutex_enter(&db->db_mtx);
 	ASSERT(dr->dt.dl.dr_override_state == DR_IN_DMU_SYNC);
@@ -1756,13 +1765,22 @@ dmu_sync_late_arrival_done(zio_t *zio)
 	blkptr_t *bp = zio->io_bp;
 	dmu_sync_arg_t *dsa = zio->io_private;
 	blkptr_t *bp_orig = &zio->io_bp_orig;
+	zgd_t *zgd = dsa->dsa_zgd;
 
-	if (zio->io_error == 0 && !BP_IS_HOLE(bp)) {
-		ASSERT(!(zio->io_flags & ZIO_FLAG_NOPWRITE));
-		ASSERT(BP_IS_HOLE(bp_orig) || !BP_EQUAL(bp, bp_orig));
-		ASSERT(zio->io_bp->blk_birth == zio->io_txg);
-		ASSERT(zio->io_txg > spa_syncing_txg(zio->io_spa));
-		zio_free(zio->io_spa, zio->io_txg, zio->io_bp);
+	if (zio->io_error == 0) {
+		/*
+		 * Record the vdev(s) backing this blkptr so they can be
+		 * flushed after the writes for the lwb have completed.
+		 */
+		zil_lwb_add_block(zgd->zgd_lwb, zgd->zgd_bp);
+
+		if (!BP_IS_HOLE(bp)) {
+			ASSERT(!(zio->io_flags & ZIO_FLAG_NOPWRITE));
+			ASSERT(BP_IS_HOLE(bp_orig) || !BP_EQUAL(bp, bp_orig));
+			ASSERT(zio->io_bp->blk_birth == zio->io_txg);
+			ASSERT(zio->io_txg > spa_syncing_txg(zio->io_spa));
+			zio_free(zio->io_spa, zio->io_txg, zio->io_bp);
+		}
 	}
 
 	dmu_tx_commit(dsa->dsa_tx);
diff --git a/usr/src/uts/common/fs/zfs/sys/zil_impl.h b/usr/src/uts/common/fs/zfs/sys/zil_impl.h
index af71cab7f1..e2eb4025a6 100644
--- a/usr/src/uts/common/fs/zfs/sys/zil_impl.h
+++ b/usr/src/uts/common/fs/zfs/sys/zil_impl.h
@@ -48,10 +48,11 @@ extern "C" {
  * via zil_lwb_write_issue(). Again, the zilog's "zl_issuer_lock" must
  * be held when making this transition.
  *
- * After the lwb's zio completes, and the vdev's are flushed, the lwb
- * will transition into the "done" state via zil_lwb_write_done(). When
- * transitioning from "issued" to "done", the zilog's "zl_lock" must be
- * held, *not* the "zl_issuer_lock".
+ * After the lwb's write zio completes, it transitions into the "write
+ * done" state via zil_lwb_write_done(); and then into the "flush done"
+ * state via zil_lwb_flush_vdevs_done(). When transitioning from
+ * "issued" to "write done", and then from "write done" to "flush done",
+ * the zilog's "zl_lock" must be held, *not* the "zl_issuer_lock".
  *
  * The zilog's "zl_issuer_lock" can become heavily contended in certain
  * workloads, so we specifically avoid acquiring that lock when
@@ -68,13 +69,14 @@ extern "C" {
  * "zl_issuer_lock" will prevent a concurrent thread from transitioning
  * that lwb to the "issued" state. Likewise, if an lwb is already in the
  * "issued" state, holding the "zl_lock" will prevent a concurrent
- * thread from transitioning that lwb to the "done" state.
+ * thread from transitioning that lwb to the "write done" state.
  */
 typedef enum {
     LWB_STATE_CLOSED,
     LWB_STATE_OPENED,
     LWB_STATE_ISSUED,
-    LWB_STATE_DONE,
+    LWB_STATE_WRITE_DONE,
+    LWB_STATE_FLUSH_DONE,
     LWB_NUM_STATES
 } lwb_state_t;
 
diff --git a/usr/src/uts/common/fs/zfs/zfs_vnops.c b/usr/src/uts/common/fs/zfs/zfs_vnops.c
index 792dbb7bac..475020a20a 100644
--- a/usr/src/uts/common/fs/zfs/zfs_vnops.c
+++ b/usr/src/uts/common/fs/zfs/zfs_vnops.c
@@ -1017,6 +1017,7 @@ zfs_write(vnode_t *vp, uio_t *uio, int ioflag, cred_t *cr, caller_context_t *ct)
 	return (0);
 }
 
+/* ARGSUSED */
 void
 zfs_get_done(zgd_t *zgd, int error)
 {
@@ -1034,9 +1035,6 @@ zfs_get_done(zgd_t *zgd, int error)
 	 */
 	VN_RELE_ASYNC(ZTOV(zp), dsl_pool_vnrele_taskq(dmu_objset_pool(os)));
 
-	if (error == 0 && zgd->zgd_bp)
-		zil_lwb_add_block(zgd->zgd_lwb, zgd->zgd_bp);
-
 	kmem_free(zgd, sizeof (zgd_t));
 }
 
@@ -1160,11 +1158,7 @@ zfs_get_data(void *arg, lr_write_t *lr, char *buf, struct lwb *lwb, zio_t *zio)
 				 * TX_WRITE2 relies on the data previously
 				 * written by the TX_WRITE that caused
 				 * EALREADY.  We zero out the BP because
-				 * it is the old, currently-on-disk BP,
-				 * so there's no need to zio_flush() its
-				 * vdevs (flushing would needlesly hurt
-				 * performance, and doesn't work on
-				 * indirect vdevs).
+				 * it is the old, currently-on-disk BP.
 				 */
 				zgd->zgd_bp = NULL;
 				BP_ZERO(bp);
diff --git a/usr/src/uts/common/fs/zfs/zil.c b/usr/src/uts/common/fs/zfs/zil.c
index 07d67a0816..05c988a3bc 100644
--- a/usr/src/uts/common/fs/zfs/zil.c
+++ b/usr/src/uts/common/fs/zfs/zil.c
@@ -550,7 +550,7 @@ zil_free_lwb(zilog_t *zilog, lwb_t *lwb)
 	ASSERT3P(lwb->lwb_root_zio, ==, NULL);
 	ASSERT3U(lwb->lwb_max_txg, <=, spa_syncing_txg(zilog->zl_spa));
 	ASSERT(lwb->lwb_state == LWB_STATE_CLOSED ||
-	    lwb->lwb_state == LWB_STATE_DONE);
+	    lwb->lwb_state == LWB_STATE_FLUSH_DONE);
 
 	/*
 	 * Clear the zilog's field to indicate this lwb is no longer
@@ -962,7 +962,8 @@ zil_commit_waiter_link_lwb(zil_commit_waiter_t *zcw, lwb_t *lwb)
 	ASSERT3P(zcw->zcw_lwb, ==, NULL);
 	ASSERT3P(lwb, !=, NULL);
 	ASSERT(lwb->lwb_state == LWB_STATE_OPENED ||
-	    lwb->lwb_state == LWB_STATE_ISSUED);
+	    lwb->lwb_state == LWB_STATE_ISSUED ||
+	    lwb->lwb_state == LWB_STATE_WRITE_DONE);
 
 	list_insert_tail(&lwb->lwb_waiters, zcw);
 	zcw->zcw_lwb = lwb;
@@ -1008,6 +1009,42 @@ zil_lwb_add_block(lwb_t *lwb, const blkptr_t *bp)
 	mutex_exit(&lwb->lwb_vdev_lock);
 }
 
+static void
+zil_lwb_flush_defer(lwb_t *lwb, lwb_t *nlwb)
+{
+	avl_tree_t *src = &lwb->lwb_vdev_tree;
+	avl_tree_t *dst = &nlwb->lwb_vdev_tree;
+	void *cookie = NULL;
+	zil_vdev_node_t *zv;
+
+	ASSERT3S(lwb->lwb_state, ==, LWB_STATE_WRITE_DONE);
+	ASSERT3S(nlwb->lwb_state, !=, LWB_STATE_WRITE_DONE);
+	ASSERT3S(nlwb->lwb_state, !=, LWB_STATE_FLUSH_DONE);
+
+	/*
+	 * While 'lwb' is at a point in its lifetime where lwb_vdev_tree does
+	 * not need the protection of lwb_vdev_lock (it will only be modified
+	 * while holding zilog->zl_lock) as its writes and those of its
+	 * children have all completed.  The younger 'nlwb' may be waiting on
+	 * future writes to additional vdevs.
+	 */
+	mutex_enter(&nlwb->lwb_vdev_lock);
+	/*
+	 * Tear down the 'lwb' vdev tree, ensuring that entries which do not
+	 * exist in 'nlwb' are moved to it, freeing any would-be duplicates.
+	 */
+	while ((zv = avl_destroy_nodes(src, &cookie)) != NULL) {
+		avl_index_t where;
+
+		if (avl_find(dst, zv, &where) == NULL) {
+			avl_insert(dst, zv, where);
+		} else {
+			kmem_free(zv, sizeof (*zv));
+		}
+	}
+	mutex_exit(&nlwb->lwb_vdev_lock);
+}
+
 void
 zil_lwb_add_txg(lwb_t *lwb, uint64_t txg)
 {
@@ -1015,9 +1052,13 @@ zil_lwb_add_txg(lwb_t *lwb, uint64_t txg)
 }
 
 /*
- * This function is a called after all VDEVs associated with a given lwb
+ * This function is a called after all vdevs associated with a given lwb
  * write have completed their DKIOCFLUSHWRITECACHE command; or as soon
- * as the lwb write completes, if "zil_nocacheflush" is set.
+ * as the lwb write completes, if "zil_nocacheflush" is set. Further,
+ * all "previous" lwb's will have completed before this function is
+ * called; i.e. this function is called for all previous lwbs before
+ * it's called for "this" lwb (enforced via zio the dependencies
+ * configured in zil_lwb_set_zio_dependency()).
  *
  * The intention is for this function to be called as soon as the
  * contents of an lwb are considered "stable" on disk, and will survive
@@ -1054,7 +1095,9 @@ zil_lwb_flush_vdevs_done(zio_t *zio)
 	zilog->zl_last_lwb_latency = gethrtime() - lwb->lwb_issued_timestamp;
 
 	lwb->lwb_root_zio = NULL;
-	lwb->lwb_state = LWB_STATE_DONE;
+
+	ASSERT3S(lwb->lwb_state, ==, LWB_STATE_WRITE_DONE);
+	lwb->lwb_state = LWB_STATE_FLUSH_DONE;
 
 	if (zilog->zl_last_lwb_opened == lwb) {
 		/*
@@ -1095,14 +1138,17 @@ zil_lwb_flush_vdevs_done(zio_t *zio)
 }
 
 /*
- * This is called when an lwb write completes. This means, this specific
- * lwb was written to disk, and all dependent lwb have also been
- * written to disk.
- *
- * At this point, a DKIOCFLUSHWRITECACHE command hasn't been issued to
- * the VDEVs involved in writing out this specific lwb. The lwb will be
- * "done" once zil_lwb_flush_vdevs_done() is called, which occurs in the
- * zio completion callback for the lwb's root zio.
+ * This is called when an lwb's write zio completes. The callback's
+ * purpose is to issue the DKIOCFLUSHWRITECACHE commands for the vdevs
+ * in the lwb's lwb_vdev_tree. The tree will contain the vdevs involved
+ * in writing out this specific lwb's data, and in the case that cache
+ * flushes have been deferred, vdevs involved in writing the data for
+ * previous lwbs. The writes corresponding to all the vdevs in the
+ * lwb_vdev_tree will have completed by the time this is called, due to
+ * the zio dependencies configured in zil_lwb_set_zio_dependency(),
+ * which takes deferred flushes into account. The lwb will be "done"
+ * once zil_lwb_flush_vdevs_done() is called, which occurs in the zio
+ * completion callback for the lwb's root zio.
  */
 static void
 zil_lwb_write_done(zio_t *zio)
@@ -1113,6 +1159,7 @@ zil_lwb_write_done(zio_t *zio)
 	avl_tree_t *t = &lwb->lwb_vdev_tree;
 	void *cookie = NULL;
 	zil_vdev_node_t *zv;
+	lwb_t *nlwb;
 
 	ASSERT3S(spa_config_held(spa, SCL_STATE, RW_READER), !=, 0);
 
@@ -1126,10 +1173,11 @@ zil_lwb_write_done(zio_t *zio)
 
 	abd_put(zio->io_abd);
 
-	ASSERT3S(lwb->lwb_state, ==, LWB_STATE_ISSUED);
-
 	mutex_enter(&zilog->zl_lock);
+	ASSERT3S(lwb->lwb_state, ==, LWB_STATE_ISSUED);
+	lwb->lwb_state = LWB_STATE_WRITE_DONE;
 	lwb->lwb_write_zio = NULL;
+	nlwb = list_next(&zilog->zl_lwb_list, lwb);
 	mutex_exit(&zilog->zl_lock);
 
 	if (avl_numnodes(t) == 0)
@@ -1148,6 +1196,27 @@ zil_lwb_write_done(zio_t *zio)
 		return;
 	}
 
+	/*
+	 * If this lwb does not have any threads waiting for it to
+	 * complete, we want to defer issuing the DKIOCFLUSHWRITECACHE
+	 * command to the vdevs written to by "this" lwb, and instead
+	 * rely on the "next" lwb to handle the DKIOCFLUSHWRITECACHE
+	 * command for those vdevs. Thus, we merge the vdev tree of
+	 * "this" lwb with the vdev tree of the "next" lwb in the list,
+	 * and assume the "next" lwb will handle flushing the vdevs (or
+	 * deferring the flush(s) again).
+	 *
+	 * This is a useful performance optimization, especially for
+	 * workloads with lots of async write activity and few sync
+	 * write and/or fsync activity, as it has the potential to
+	 * coalesce multiple flush commands to a vdev into one.
+	 */
+	if (list_head(&lwb->lwb_waiters) == NULL && nlwb != NULL) {
+		zil_lwb_flush_defer(lwb, nlwb);
+		ASSERT(avl_is_empty(&lwb->lwb_vdev_tree));
+		return;
+	}
+
 	while ((zv = avl_destroy_nodes(t, &cookie)) != NULL) {
 		vdev_t *vd = vdev_lookup_top(spa, zv->zv_vdev);
 		if (vd != NULL)
@@ -1156,6 +1225,73 @@ zil_lwb_write_done(zio_t *zio)
 	}
 }
 
+static void
+zil_lwb_set_zio_dependency(zilog_t *zilog, lwb_t *lwb)
+{
+	lwb_t *last_lwb_opened = zilog->zl_last_lwb_opened;
+
+	ASSERT(MUTEX_HELD(&zilog->zl_issuer_lock));
+	ASSERT(MUTEX_HELD(&zilog->zl_lock));
+
+	/*
+	 * The zilog's "zl_last_lwb_opened" field is used to build the
+	 * lwb/zio dependency chain, which is used to preserve the
+	 * ordering of lwb completions that is required by the semantics
+	 * of the ZIL. Each new lwb zio becomes a parent of the
+	 * "previous" lwb zio, such that the new lwb's zio cannot
+	 * complete until the "previous" lwb's zio completes.
+	 *
+	 * This is required by the semantics of zil_commit(); the commit
+	 * waiters attached to the lwbs will be woken in the lwb zio's
+	 * completion callback, so this zio dependency graph ensures the
+	 * waiters are woken in the correct order (the same order the
+	 * lwbs were created).
+	 */
+	if (last_lwb_opened != NULL &&
+	    last_lwb_opened->lwb_state != LWB_STATE_FLUSH_DONE) {
+		ASSERT(last_lwb_opened->lwb_state == LWB_STATE_OPENED ||
+		    last_lwb_opened->lwb_state == LWB_STATE_ISSUED ||
+		    last_lwb_opened->lwb_state == LWB_STATE_WRITE_DONE);
+
+		ASSERT3P(last_lwb_opened->lwb_root_zio, !=, NULL);
+		zio_add_child(lwb->lwb_root_zio,
+		    last_lwb_opened->lwb_root_zio);
+
+		/*
+		 * If the previous lwb's write hasn't already completed,
+		 * we also want to order the completion of the lwb write
+		 * zios (above, we only order the completion of the lwb
+		 * root zios). This is required because of how we can
+		 * defer the DKIOCFLUSHWRITECACHE commands for each lwb.
+		 *
+		 * When the DKIOCFLUSHWRITECACHE commands are defered,
+		 * the previous lwb will rely on this lwb to flush the
+		 * vdevs written to by that previous lwb. Thus, we need
+		 * to ensure this lwb doesn't issue the flush until
+		 * after the previous lwb's write completes. We ensure
+		 * this ordering by setting the zio parent/child
+		 * relationship here.
+		 *
+		 * Without this relationship on the lwb's write zio,
+		 * it's possible for this lwb's write to complete prior
+		 * to the previous lwb's write completing; and thus, the
+		 * vdevs for the previous lwb would be flushed prior to
+		 * that lwb's data being written to those vdevs (the
+		 * vdevs are flushed in the lwb write zio's completion
+		 * handler, zil_lwb_write_done()).
+		 */
+		if (last_lwb_opened->lwb_state != LWB_STATE_WRITE_DONE) {
+			ASSERT(last_lwb_opened->lwb_state == LWB_STATE_OPENED ||
+			    last_lwb_opened->lwb_state == LWB_STATE_ISSUED);
+
+			ASSERT3P(last_lwb_opened->lwb_write_zio, !=, NULL);
+			zio_add_child(lwb->lwb_write_zio,
+			    last_lwb_opened->lwb_write_zio);
+		}
+	}
+}
+
+
 /*
  * This function's purpose is to "open" an lwb such that it is ready to
  * accept new itxs being committed to it. To do this, the lwb's zio
@@ -1200,33 +1336,8 @@ zil_lwb_write_open(zilog_t *zilog, lwb_t *lwb)
 		lwb->lwb_state = LWB_STATE_OPENED;
 
 		mutex_enter(&zilog->zl_lock);
-
-		/*
-		 * The zilog's "zl_last_lwb_opened" field is used to
-		 * build the lwb/zio dependency chain, which is used to
-		 * preserve the ordering of lwb completions that is
-		 * required by the semantics of the ZIL. Each new lwb
-		 * zio becomes a parent of the "previous" lwb zio, such
-		 * that the new lwb's zio cannot complete until the
-		 * "previous" lwb's zio completes.
-		 *
-		 * This is required by the semantics of zil_commit();
-		 * the commit waiters attached to the lwbs will be woken
-		 * in the lwb zio's completion callback, so this zio
-		 * dependency graph ensures the waiters are woken in the
-		 * correct order (the same order the lwbs were created).
-		 */
-		lwb_t *last_lwb_opened = zilog->zl_last_lwb_opened;
-		if (last_lwb_opened != NULL &&
-		    last_lwb_opened->lwb_state != LWB_STATE_DONE) {
-			ASSERT(last_lwb_opened->lwb_state == LWB_STATE_OPENED ||
-			    last_lwb_opened->lwb_state == LWB_STATE_ISSUED);
-			ASSERT3P(last_lwb_opened->lwb_root_zio, !=, NULL);
-			zio_add_child(lwb->lwb_root_zio,
-			    last_lwb_opened->lwb_root_zio);
-		}
+		zil_lwb_set_zio_dependency(zilog, lwb);
 		zilog->zl_last_lwb_opened = lwb;
-
 		mutex_exit(&zilog->zl_lock);
 	}
 
@@ -1927,7 +2038,8 @@ zil_prune_commit_list(zilog_t *zilog)
 		mutex_enter(&zilog->zl_lock);
 
 		lwb_t *last_lwb = zilog->zl_last_lwb_opened;
-		if (last_lwb == NULL || last_lwb->lwb_state == LWB_STATE_DONE) {
+		if (last_lwb == NULL ||
+		    last_lwb->lwb_state == LWB_STATE_FLUSH_DONE) {
 			/*
 			 * All of the itxs this waiter was waiting on
 			 * must have already completed (or there were
@@ -2008,7 +2120,8 @@ zil_process_commit_list(zilog_t *zilog)
 		lwb = zil_create(zilog);
 	} else {
 		ASSERT3S(lwb->lwb_state, !=, LWB_STATE_ISSUED);
-		ASSERT3S(lwb->lwb_state, !=, LWB_STATE_DONE);
+		ASSERT3S(lwb->lwb_state, !=, LWB_STATE_WRITE_DONE);
+		ASSERT3S(lwb->lwb_state, !=, LWB_STATE_FLUSH_DONE);
 	}
 
 	while (itx = list_head(&zilog->zl_itx_commit_list)) {
@@ -2110,7 +2223,8 @@ zil_process_commit_list(zilog_t *zilog)
 		ASSERT(list_is_empty(&nolwb_waiters));
 		ASSERT3P(lwb, !=, NULL);
 		ASSERT3S(lwb->lwb_state, !=, LWB_STATE_ISSUED);
-		ASSERT3S(lwb->lwb_state, !=, LWB_STATE_DONE);
+		ASSERT3S(lwb->lwb_state, !=, LWB_STATE_WRITE_DONE);
+		ASSERT3S(lwb->lwb_state, !=, LWB_STATE_FLUSH_DONE);
 
 		/*
 		 * At this point, the ZIL block pointed at by the "lwb"
@@ -2231,7 +2345,8 @@ zil_commit_waiter_timeout(zilog_t *zilog, zil_commit_waiter_t *zcw)
 	 * acquiring it when it's not necessary to do so.
 	 */
 	if (lwb->lwb_state == LWB_STATE_ISSUED ||
-	    lwb->lwb_state == LWB_STATE_DONE)
+	    lwb->lwb_state == LWB_STATE_WRITE_DONE ||
+	    lwb->lwb_state == LWB_STATE_FLUSH_DONE)
 		return;
 
 	/*
@@ -2279,7 +2394,8 @@ zil_commit_waiter_timeout(zilog_t *zilog, zil_commit_waiter_t *zcw)
 	 * more details on the lwb states, and locking requirements.
 	 */
 	if (lwb->lwb_state == LWB_STATE_ISSUED ||
-	    lwb->lwb_state == LWB_STATE_DONE)
+	    lwb->lwb_state == LWB_STATE_WRITE_DONE ||
+	    lwb->lwb_state == LWB_STATE_FLUSH_DONE)
 		goto out;
 
 	ASSERT3S(lwb->lwb_state, ==, LWB_STATE_OPENED);
@@ -2452,7 +2568,8 @@ zil_commit_waiter(zilog_t *zilog, zil_commit_waiter_t *zcw)
 
 			IMPLY(lwb != NULL,
 			    lwb->lwb_state == LWB_STATE_ISSUED ||
-			    lwb->lwb_state == LWB_STATE_DONE);
+			    lwb->lwb_state == LWB_STATE_WRITE_DONE ||
+			    lwb->lwb_state == LWB_STATE_FLUSH_DONE);
 			cv_wait(&zcw->zcw_cv, &zcw->zcw_lock);
 		}
 	}
@@ -3094,13 +3211,13 @@ zil_suspend(const char *osname, void **cookiep)
 	 * to disk before proceeding. If we used zil_commit instead, it
 	 * would just call txg_wait_synced(), because zl_suspend is set.
 	 * txg_wait_synced() doesn't wait for these lwb's to be
-	 * LWB_STATE_DONE before returning.
+	 * LWB_STATE_FLUSH_DONE before returning.
 	 */
 	zil_commit_impl(zilog, 0);
 
 	/*
-	 * Now that we've ensured all lwb's are LWB_STATE_DONE, we use
-	 * txg_wait_synced() to ensure the data from the zilog has
+	 * Now that we've ensured all lwb's are LWB_STATE_FLUSH_DONE, we
+	 * use txg_wait_synced() to ensure the data from the zilog has
 	 * migrated to the main pool before calling zil_destroy().
 	 */
 	txg_wait_synced(zilog->zl_dmu_pool, 0);
diff --git a/usr/src/uts/common/fs/zfs/zvol.c b/usr/src/uts/common/fs/zfs/zvol.c
index b58cf54a17..1e787f31b2 100644
--- a/usr/src/uts/common/fs/zfs/zvol.c
+++ b/usr/src/uts/common/fs/zfs/zvol.c
@@ -980,6 +980,7 @@ zvol_close(dev_t dev, int flag, int otyp, cred_t *cr)
 	return (error);
 }
 
+/* ARGSUSED */
 static void
 zvol_get_done(zgd_t *zgd, int error)
 {
@@ -988,9 +989,6 @@ zvol_get_done(zgd_t *zgd, int error)
 
 	zfs_range_unlock(zgd->zgd_rl);
 
-	if (error == 0 && zgd->zgd_bp)
-		zil_lwb_add_block(zgd->zgd_lwb, zgd->zgd_bp);
-
 	kmem_free(zgd, sizeof (zgd_t));
 }