OpenZFS 7143 - dbuf_read() creates unnecessary zio_root() for bonus buf

dbuf_read() creates a zio_root() to track and wait for all the
zio's that may happen as part of this call. However, if the blkptr_t
for this buffer is NULL or a hole, we will not create any more zio's,
so this zio_root() is unnecessary. This is always the case when calling
dbuf_read() on a bonus buffer, because it has no blkptr (it's part of
the containing dnode). For workloads that read a lot of bonus buffers
(e.g. file creation and removal), creating and destroying these
unnecessary zio's can decrease performance by around 3%.

The fix is to only create/destroy the zio_root() in dbuf_read() if
the blkptr is not NULL and not a hole.

Changes sponsored by Intel Corp.

Authored by: Matthew Ahrens <mahrens@delphix.com>
Ported-by: Brian Behlendorf <behlendorf1@llnl.gov>
Issue openzfs/openzfs#137
Issue openzfs#4803

module/zfs/dbuf.c

@@ -21,7 +21,7 @@
 /*
  * Copyright (c) 2005, 2010, Oracle and/or its affiliates. All rights reserved.
  * Copyright 2011 Nexenta Systems, Inc.  All rights reserved.
- * Copyright (c) 2012, 2016 by Delphix. All rights reserved.
+ * Copyright (c) 2012, 2015 by Delphix. All rights reserved.
  * Copyright (c) 2013 by Saso Kiselkov. All rights reserved.
  * Copyright (c) 2014 Spectra Logic Corporation, All rights reserved.
  */
@@ -1207,7 +1207,8 @@ dbuf_read(dmu_buf_impl_t *db, zio_t *zio, uint32_t flags)
 	} else if (db->db_state == DB_UNCACHED) {
 		spa_t *spa = dn->dn_objset->os_spa;
 
-		if (zio == NULL)
+		if (zio == NULL &&
+		    db->db_blkptr != NULL && !BP_IS_HOLE(db->db_blkptr))
 			zio = zio_root(spa, NULL, NULL, ZIO_FLAG_CANFAIL);
 
 		err = dbuf_read_impl(db, zio, flags);
@@ -1221,7 +1222,7 @@ dbuf_read(dmu_buf_impl_t *db, zio_t *zio, uint32_t flags)
 			rw_exit(&dn->dn_struct_rwlock);
 		DB_DNODE_EXIT(db);
 
-		if (!err && !havepzio)
+		if (!err && !havepzio && zio != NULL)
 			err = zio_wait(zio);
 	} else {
 		/*