MFV r272804:

Refactor the code and stop restore_object from creating two transactions.

Illumos issue:
    3693 restore_object uses at least two transactions to restore an object

MFC after:	2 weeks

sys/cddl/contrib/opensolaris/uts/common/fs/zfs/dmu.c

@@ -1918,6 +1918,7 @@ dmu_object_info_from_dnode(dnode_t *dn, dmu_object_info_t *doi)
 	doi->doi_indirection = dn->dn_nlevels;
 	doi->doi_checksum = dn->dn_checksum;
 	doi->doi_compress = dn->dn_compress;
+	doi->doi_nblkptr = dn->dn_nblkptr;
 	doi->doi_physical_blocks_512 = (DN_USED_BYTES(dnp) + 256) >> 9;
 	doi->doi_max_offset = (dn->dn_maxblkid + 1) * dn->dn_datablksz;
 	doi->doi_fill_count = 0;

sys/cddl/contrib/opensolaris/uts/common/fs/zfs/dmu_object.c

@@ -20,7 +20,8 @@
  */
 /*
  * Copyright (c) 2005, 2010, Oracle and/or its affiliates. All rights reserved.
- * Copyright (c) 2013 by Delphix. All rights reserved.
+ * Copyright (c) 2013, 2014 by Delphix. All rights reserved.
+ * Copyright 2014 HybridCluster. All rights reserved.
  */
 
 #include <sys/dmu.h>
@@ -107,11 +108,9 @@ dmu_object_claim(objset_t *os, uint64_t object, dmu_object_type_t ot,
 
 int
 dmu_object_reclaim(objset_t *os, uint64_t object, dmu_object_type_t ot,
-    int blocksize, dmu_object_type_t bonustype, int bonuslen)
+    int blocksize, dmu_object_type_t bonustype, int bonuslen, dmu_tx_t *tx)
 {
 	dnode_t *dn;
-	dmu_tx_t *tx;
-	int nblkptr;
 	int err;
 
 	if (object == DMU_META_DNODE_OBJECT)
@@ -122,44 +121,9 @@ dmu_object_reclaim(objset_t *os, uint64_t object, dmu_object_type_t ot,
 	if (err)
 		return (err);
 
-	if (dn->dn_type == ot && dn->dn_datablksz == blocksize &&
-	    dn->dn_bonustype == bonustype && dn->dn_bonuslen == bonuslen) {
-		/* nothing is changing, this is a noop */
-		dnode_rele(dn, FTAG);
-		return (0);
-	}
-
-	if (bonustype == DMU_OT_SA) {
-		nblkptr = 1;
-	} else {
-		nblkptr = 1 + ((DN_MAX_BONUSLEN - bonuslen) >> SPA_BLKPTRSHIFT);
-	}
-
-	/*
-	 * If we are losing blkptrs or changing the block size this must
-	 * be a new file instance.   We must clear out the previous file
-	 * contents before we can change this type of metadata in the dnode.
-	 */
-	if (dn->dn_nblkptr > nblkptr || dn->dn_datablksz != blocksize) {
-		err = dmu_free_long_range(os, object, 0, DMU_OBJECT_END);
-		if (err)
-			goto out;
-	}
-
-	tx = dmu_tx_create(os);
-	dmu_tx_hold_bonus(tx, object);
-	err = dmu_tx_assign(tx, TXG_WAIT);
-	if (err) {
-		dmu_tx_abort(tx);
-		goto out;
-	}
-
 	dnode_reallocate(dn, ot, blocksize, bonustype, bonuslen, tx);
 
-	dmu_tx_commit(tx);
-out:
 	dnode_rele(dn, FTAG);
-
 	return (err);
 }
 

sys/cddl/contrib/opensolaris/uts/common/fs/zfs/dmu_send.c

@@ -24,6 +24,7 @@
  * Copyright (c) 2011, 2014 by Delphix. All rights reserved.
  * Copyright (c) 2014, Joyent, Inc. All rights reserved.
  * Copyright (c) 2012, Martin Matuska <mm@FreeBSD.org>. All rights reserved.
+ * Copyright 2014 HybridCluster. All rights reserved.
  */
 
 #include <sys/dmu.h>
@@ -1392,12 +1393,25 @@ backup_byteswap(dmu_replay_record_t *drr)
 #undef DO32
 }
 
+static inline uint8_t
+deduce_nblkptr(dmu_object_type_t bonus_type, uint64_t bonus_size)
+{
+	if (bonus_type == DMU_OT_SA) {
+		return (1);
+	} else {
+		return (1 +
+		    ((DN_MAX_BONUSLEN - bonus_size) >> SPA_BLKPTRSHIFT));
+	}
+}
+
 static int
 restore_object(struct restorearg *ra, objset_t *os, struct drr_object *drro)
 {
-	int err;
+	dmu_object_info_t doi;
 	dmu_tx_t *tx;
 	void *data = NULL;
+	uint64_t object;
+	int err;
 
 	if (drro->drr_type == DMU_OT_NONE ||
 	    !DMU_OT_IS_VALID(drro->drr_type) ||
@@ -1411,10 +1425,11 @@ restore_object(struct restorearg *ra, objset_t *os, struct drr_object *drro)
 		return (SET_ERROR(EINVAL));
 	}
 
-	err = dmu_object_info(os, drro->drr_object, NULL);
+	err = dmu_object_info(os, drro->drr_object, &doi);
 
 	if (err != 0 && err != ENOENT)
 		return (SET_ERROR(EINVAL));
+	object = err == 0 ? drro->drr_object : DMU_NEW_OBJECT;
 
 	if (drro->drr_bonuslen) {
 		data = restore_read(ra, P2ROUNDUP(drro->drr_bonuslen, 8), NULL);
@@ -1422,37 +1437,53 @@ restore_object(struct restorearg *ra, objset_t *os, struct drr_object *drro)
 			return (ra->err);
 	}
 
-	if (err == ENOENT) {
-		/* currently free, want to be allocated */
-		tx = dmu_tx_create(os);
-		dmu_tx_hold_bonus(tx, DMU_NEW_OBJECT);
-		err = dmu_tx_assign(tx, TXG_WAIT);
-		if (err != 0) {
-			dmu_tx_abort(tx);
-			return (err);
+	/*
+	 * If we are losing blkptrs or changing the block size this must
+	 * be a new file instance.  We must clear out the previous file
+	 * contents before we can change this type of metadata in the dnode.
+	 */
+	if (err == 0) {
+		int nblkptr;
+
+		nblkptr = deduce_nblkptr(drro->drr_bonustype,
+		    drro->drr_bonuslen);
+
+		if (drro->drr_blksz != doi.doi_data_block_size ||
+		    nblkptr < doi.doi_nblkptr) {
+			err = dmu_free_long_range(os, drro->drr_object,
+			    0, DMU_OBJECT_END);
+			if (err != 0)
+				return (SET_ERROR(EINVAL));
 		}
-		err = dmu_object_claim(os, drro->drr_object,
-		    drro->drr_type, drro->drr_blksz,
-		    drro->drr_bonustype, drro->drr_bonuslen, tx);
-		dmu_tx_commit(tx);
-	} else {
-		/* currently allocated, want to be allocated */
-		err = dmu_object_reclaim(os, drro->drr_object,
-		    drro->drr_type, drro->drr_blksz,
-		    drro->drr_bonustype, drro->drr_bonuslen);
-	}
-	if (err != 0) {
-		return (SET_ERROR(EINVAL));
 	}
 
 	tx = dmu_tx_create(os);
-	dmu_tx_hold_bonus(tx, drro->drr_object);
+	dmu_tx_hold_bonus(tx, object);
 	err = dmu_tx_assign(tx, TXG_WAIT);
 	if (err != 0) {
 		dmu_tx_abort(tx);
 		return (err);
 	}
 
+	if (object == DMU_NEW_OBJECT) {
+		/* currently free, want to be allocated */
+		err = dmu_object_claim(os, drro->drr_object,
+		    drro->drr_type, drro->drr_blksz,
+		    drro->drr_bonustype, drro->drr_bonuslen, tx);
+	} else if (drro->drr_type != doi.doi_type ||
+	    drro->drr_blksz != doi.doi_data_block_size ||
+	    drro->drr_bonustype != doi.doi_bonus_type ||
+	    drro->drr_bonuslen != doi.doi_bonus_size) {
+		/* currently allocated, but with different properties */
+		err = dmu_object_reclaim(os, drro->drr_object,
+		    drro->drr_type, drro->drr_blksz,
+		    drro->drr_bonustype, drro->drr_bonuslen, tx);
+	}
+	if (err != 0) {
+		dmu_tx_commit(tx);
+		return (SET_ERROR(EINVAL));
+	}
+
 	dmu_object_set_checksum(os, drro->drr_object, drro->drr_checksumtype,
 	    tx);
 	dmu_object_set_compress(os, drro->drr_object, drro->drr_compress, tx);

sys/cddl/contrib/opensolaris/uts/common/fs/zfs/sys/dmu.h

@@ -25,6 +25,7 @@
  * Copyright 2011 Nexenta Systems, Inc. All rights reserved.
  * Copyright (c) 2012, Joyent, Inc. All rights reserved.
  * Copyright 2013 DEY Storage Systems, Inc.
+ * Copyright 2014 HybridCluster. All rights reserved.
  */
 
 /* Portions Copyright 2010 Robert Milkowski */
@@ -341,7 +342,7 @@ uint64_t dmu_object_alloc(objset_t *os, dmu_object_type_t ot,
 int dmu_object_claim(objset_t *os, uint64_t object, dmu_object_type_t ot,
     int blocksize, dmu_object_type_t bonus_type, int bonus_len, dmu_tx_t *tx);
 int dmu_object_reclaim(objset_t *os, uint64_t object, dmu_object_type_t ot,
-    int blocksize, dmu_object_type_t bonustype, int bonuslen);
+    int blocksize, dmu_object_type_t bonustype, int bonuslen, dmu_tx_t *txp);
 
 /*
  * Free an object from this objset.
@@ -662,7 +663,8 @@ typedef struct dmu_object_info {
 	uint8_t doi_indirection;		/* 2 = dnode->indirect->data */
 	uint8_t doi_checksum;
 	uint8_t doi_compress;
-	uint8_t doi_pad[5];
+	uint8_t doi_nblkptr;
+	uint8_t doi_pad[4];
 	uint64_t doi_physical_blocks_512;	/* data + metadata, 512b blks */
 	uint64_t doi_max_offset;
 	uint64_t doi_fill_count;		/* number of non-empty blocks */