MFC r260183: MFV r260154 + 260182:

4369 implement zfs bookmarks
4368 zfs send filesystems from readonly pools

Illumos/illumos-gate@78f171005391b928aaf1642b3206c534ed644332

12 files changed, 767 insertions, 159 deletions

diff --git a/cddl/contrib/opensolaris/cmd/zfs/zfs.8 b/cddl/contrib/opensolaris/cmd/zfs/zfs.8
index 1d08f83..009df4f 100644
--- a/cddl/contrib/opensolaris/cmd/zfs/zfs.8
+++ b/cddl/contrib/opensolaris/cmd/zfs/zfs.8
@@ -18,7 +18,7 @@
 .\" information: Portions Copyright [yyyy] [name of copyright owner]
 .\"
 .\" Copyright (c) 2010, Sun Microsystems, Inc. All Rights Reserved.
-.\" Copyright (c) 2012 by Delphix. All rights reserved.
+.\" Copyright (c) 2013 by Delphix. All rights reserved.
 .\" Copyright (c) 2011, Pawel Jakub Dawidek <pjd@FreeBSD.org>
 .\" Copyright (c) 2012, Glen Barber <gjb@FreeBSD.org>
 .\" Copyright (c) 2012, Bryan Drewery <bdrewery@FreeBSD.org>
@@ -26,10 +26,11 @@
 .\" Copyright (c) 2013 Nexenta Systems, Inc. All Rights Reserved.
 .\" Copyright (c) 2013, Joyent, Inc. All rights reserved.
 .\" Copyright (c) 2013, Steven Hartland <smh@FreeBSD.org>
+.\" Copyright (c) 2014, Xin LI <delphij@FreeBSD.org>
 .\"
 .\" $FreeBSD$
 .\"
-.Dd December 24, 2013
+.Dd January 2, 2014
 .Dt ZFS 8
 .Os
 .Sh NAME
@@ -57,11 +58,16 @@
 .Cm destroy
 .Op Fl dnpRrv
 .Sm off
-.Ar snapshot
-.Op % Ns Ar snapname
+.Ar filesystem Ns | Ns volume
+.Ns @snap
+.Op % Ns Ar snap
+.Op , Ns Ar snap Op % Ns Ar snap
 .Op , Ns ...
 .Sm on
 .Nm
+.Cm destroy
+.Ar filesystem Ns | Ns Ar volume Ns # Ns Ar bookmark
+.Nm
 .Cm snapshot Ns | Ns Cm snap
 .Op Fl r
 .Oo Fl o Ar property Ns = Ns Ar value Oc Ns ...
@@ -168,11 +174,19 @@
 .Cm unshare
 .Fl a | Ar filesystem Ns | Ns Ar mountpoint
 .Nm
+.Cm bookmark
+.Ar snapshot
+.Ar bookmark
+.Nm
 .Cm send
 .Op Fl DnPpRv
 .Op Fl i Ar snapshot | Fl I Ar snapshot
 .Ar snapshot
 .Nm
+.Cm send
+.Op Fl i Ar snapshot Ns | Ns bookmark
+.Ar filesystem Ns | Ns Ar volume Ns | Ns Ar snapshot
+.Nm
 .Cm receive Ns | Ns Cm recv
 .Op Fl vnFu
 .Ar filesystem Ns | Ns Ar volume Ns | Ns Ar snapshot
@@ -1654,6 +1668,13 @@ options, as they can destroy large portions of a pool and cause unexpected
 behavior for mounted file systems in use.
 .It Xo
 .Nm
+.Cm destroy
+.Ar filesystem Ns | Ns Ar volume Ns # Ns Ar bookmark
+.Xc
+.Pp
+The given bookmark is destroyed.
+.It Xo
+.Nm
 .Cm snapshot Ns | Ns Cm snap
 .Op Fl r
 .Oo Fl o Ar property Ns = Ns Ar value Oc Ns ...
@@ -1686,14 +1707,24 @@ Roll back the given dataset to a previous snapshot. When a dataset is rolled
 back, all data that has changed since the snapshot is discarded, and the
 dataset reverts to the state at the time of the snapshot. By default, the
 command refuses to roll back to a snapshot other than the most recent one. In
-order to do so, all intermediate snapshots must be destroyed by specifying the
+order to do so, all intermediate snapshots and bookmarks must be destroyed
+by specifying the
 .Fl r
 option.
+.Pp
+The
+.Fl rR
+options do not recursively destroy the child snapshots of a
+recursive snapshot.
+Only direct snapshots of the specified filesystem
+are destroyed by either of these options.
+To completely roll back a
+recursive snapshot, you must rollback the individual child snapshots.
 .Bl -tag -width indent
 .It Fl r
-Recursively destroy any snapshots more recent than the one specified.
+Destroy any snapshots and bookmarks more recent than the one specified.
 .It Fl R
-Recursively destroy any more recent snapshots, as well as any clones of those
+Destroy any more recent snapshots and bookmarks, as well as any clones of those
 snapshots.
 .It Fl f
 Used with the
@@ -1868,7 +1899,7 @@ syntax.
 A comma-separated list of types to display, where
 .Ar type
 is one of
-.Sy filesystem , snapshot , snap,  volume , No or Sy all .
+.Sy filesystem , snapshot , snap , volume , bookmark , No or Sy all .
 For example, specifying
 .Fl t Cm snapshot
 displays only snapshots.
@@ -1965,7 +1996,7 @@ sections.
 The special value
 .Cm all
 can be used to display all properties that apply to the given dataset's type
-(filesystem, volume, or snapshot).
+(filesystem, volume, snapshot, or bookmark).
 .Bl -tag -width indent
 .It Fl r
 Recursively display properties for any children.
@@ -2283,6 +2314,26 @@ file system shared on the system.
 .El
 .It Xo
 .Nm
+.Cm bookmark
+.Ar snapshot
+.Ar bookmark
+.Xc
+.Pp
+Creates a bookmark of the given snapshot.
+Bookmarks mark the point in time
+when the snapshot was created, and can be used as the incremental source for
+a
+.Qq Nm Cm send
+command.
+.Pp
+This feature must be enabled to be used.
+See
+.Xr zpool-features 7
+for details on ZFS feature flags and the
+.Sy bookmark
+feature.
+.It Xo
+.Nm
 .Cm send
 .Op Fl DnPpRv
 .Op Fl i Ar snapshot | Fl I Ar snapshot
@@ -2301,17 +2352,15 @@ a file or to a different system (for example, using
 By default, a full stream is generated.
 .Bl -tag -width indent
 .It Fl i Ar snapshot
-Generate an incremental stream from the
-.Fl i Ar snapshot
-to the last
-.Ar snapshot .
-The incremental source (the
-.Fl i Ar snapshot )
-can be specified as the last component of the snapshot name (for example, the
-part after the
-.Sy @ ) ,
-and it is assumed to be from the same file system as the last
-.Ar snapshot .
+Generate an incremental stream from the first
+.Ar snapshot Pq the incremental source
+to the second
+.Ar snapshot Pq the incremental target .
+The incremental source can be specified as the last component of the
+snapshot name
+.Pq the Em @ No character and following
+and
+it is assumed to be from the same file system as the incremental target.
 .Pp
 If the destination is a clone, the source may be the origin snapshot, which
 must be fully specified (for example,
@@ -2319,15 +2368,16 @@ must be fully specified (for example,
 not just
 .Cm @origin ) .
 .It Fl I Ar snapshot
-Generate a stream package that sends all intermediary snapshots from the
-.Fl I Ar snapshot
-to the last
+Generate a stream package that sends all intermediary snapshots from the first
+.Ar snapshot
+to the second
 .Ar snapshot .
 For example,
 .Ic -I @a fs@d
 is similar to
 .Ic -i @a fs@b; -i @b fs@c; -i @c fs@d .
-The incremental source snapshot may be specified as with the
+The incremental
+source may be specified as with the
 .Fl i
 option.
 .It Fl R
@@ -2380,6 +2430,35 @@ on future versions of
 .Tn ZFS .
 .It Xo
 .Nm
+.Cm send
+.Op Fl i Ar snapshot Ns | Ns Ar bookmark
+.Ar filesystem Ns | Ns Ar volume Ns | Ns Ar snapshot
+.Xc
+.Pp
+Generate a send stream, which may be of a filesystem, and may be
+incremental from a bookmark.
+If the destination is a filesystem or volume,
+the pool must be read-only, or the filesystem must not be mounted.
+When the
+stream generated from a filesystem or volume is received, the default snapshot
+name will be
+.Pq --head-- .
+.Bl -tag -width indent
+.It Fl i Ar snapshot Ns | Ns bookmark
+Generate an incremental send stream.
+The incremental source must be an earlier
+snapshot in the destination's history.
+It will commonly be an earlier
+snapshot in the destination's filesystem, in which case it can be
+specified as the last component of the name
+.Pq the Em # No or Em @ No character and following .
+.Pp
+If the incremental target is a clone, the incremental source can
+be the origin snapshot, or an earlier snapshot in the origin's filesystem,
+or the origin's origin, etc.
+.El
+.It Xo
+.Nm
 .Cm receive Ns | Ns Cm recv
 .Op Fl vnFu
 .Ar filesystem Ns | Ns Ar volume Ns | Ns Ar snapshot
diff --git a/cddl/contrib/opensolaris/cmd/zfs/zfs_iter.c b/cddl/contrib/opensolaris/cmd/zfs/zfs_iter.c
index 099d27e..626d69c 100644
--- a/cddl/contrib/opensolaris/cmd/zfs/zfs_iter.c
+++ b/cddl/contrib/opensolaris/cmd/zfs/zfs_iter.c
@@ -24,6 +24,7 @@
  * Copyright (c) 2012 Pawel Jakub Dawidek <pawel@dawidek.net>.
  * All rights reserved.
  * Copyright 2013 Nexenta Systems, Inc.  All rights reserved.
+ * Copyright (c) 2013 by Delphix. All rights reserved.
  */
 
 #include <libintl.h>
@@ -72,7 +73,7 @@ uu_avl_pool_t *avl_pool;
  * Include snaps if they were requested or if this a zfs list where types
  * were not specified and the "listsnapshots" property is set on this pool.
  */
-static int
+static boolean_t
 zfs_include_snapshots(zfs_handle_t *zhp, callback_data_t *cb)
 {
 	zpool_handle_t *zph;
@@ -92,8 +93,9 @@ static int
 zfs_callback(zfs_handle_t *zhp, void *data)
 {
 	callback_data_t *cb = data;
-	int dontclose = 0;
-	int include_snaps = zfs_include_snapshots(zhp, cb);
+	boolean_t dontclose = B_FALSE;
+	boolean_t include_snaps = zfs_include_snapshots(zhp, cb);
+	boolean_t include_bmarks = (cb->cb_types & ZFS_TYPE_BOOKMARK);
 
 	if ((zfs_get_type(zhp) & cb->cb_types) ||
 	    ((zfs_get_type(zhp) == ZFS_TYPE_SNAPSHOT) && include_snaps)) {
@@ -119,7 +121,7 @@ zfs_callback(zfs_handle_t *zhp, void *data)
 				}
 			}
 			uu_avl_insert(cb->cb_avl, node, idx);
-			dontclose = 1;
+			dontclose = B_TRUE;
 		} else {
 			free(node);
 		}
@@ -134,11 +136,14 @@ zfs_callback(zfs_handle_t *zhp, void *data)
 		cb->cb_depth++;
 		if (zfs_get_type(zhp) == ZFS_TYPE_FILESYSTEM)
 			(void) zfs_iter_filesystems(zhp, zfs_callback, data);
-		if ((zfs_get_type(zhp) != ZFS_TYPE_SNAPSHOT) && include_snaps) {
+		if (((zfs_get_type(zhp) & (ZFS_TYPE_SNAPSHOT |
+		    ZFS_TYPE_BOOKMARK)) == 0) && include_snaps)
 			(void) zfs_iter_snapshots(zhp,
 			    (cb->cb_flags & ZFS_ITER_SIMPLE) != 0, zfs_callback,
 			    data);
-		}
+		if (((zfs_get_type(zhp) & (ZFS_TYPE_SNAPSHOT |
+		    ZFS_TYPE_BOOKMARK)) == 0) && include_bmarks)
+			(void) zfs_iter_bookmarks(zhp, zfs_callback, data);
 		cb->cb_depth--;
 	}
 
diff --git a/cddl/contrib/opensolaris/cmd/zfs/zfs_main.c b/cddl/contrib/opensolaris/cmd/zfs/zfs_main.c
index db3b3a0..84b05eb 100644
--- a/cddl/contrib/opensolaris/cmd/zfs/zfs_main.c
+++ b/cddl/contrib/opensolaris/cmd/zfs/zfs_main.c
@@ -107,6 +107,7 @@ static int zfs_do_release(int argc, char **argv);
 static int zfs_do_diff(int argc, char **argv);
 static int zfs_do_jail(int argc, char **argv);
 static int zfs_do_unjail(int argc, char **argv);
+static int zfs_do_bookmark(int argc, char **argv);
 
 /*
  * Enable a reasonable set of defaults for libumem debugging on DEBUG builds.
@@ -155,6 +156,7 @@ typedef enum {
 	HELP_HOLDS,
 	HELP_RELEASE,
 	HELP_DIFF,
+	HELP_BOOKMARK,
 } zfs_help_t;
 
 typedef struct zfs_command {
@@ -181,6 +183,7 @@ static zfs_command_t command_table[] = {
 	{ "clone",	zfs_do_clone,		HELP_CLONE		},
 	{ "promote",	zfs_do_promote,		HELP_PROMOTE		},
 	{ "rename",	zfs_do_rename,		HELP_RENAME		},
+	{ "bookmark",	zfs_do_bookmark,	HELP_BOOKMARK		},
 	{ NULL },
 	{ "list",	zfs_do_list,		HELP_LIST		},
 	{ NULL },
@@ -231,11 +234,12 @@ get_usage(zfs_help_t idx)
 	case HELP_DESTROY:
 		return (gettext("\tdestroy [-fnpRrv] <filesystem|volume>\n"
 		    "\tdestroy [-dnpRrv] "
-		    "<snapshot>[%<snapname>][,...]\n"));
+		    "<filesystem|volume>@<snap>[%<snap>][,...]\n"
+		    "\tdestroy <filesystem|volume>#<bookmark>\n"));
 	case HELP_GET:
 		return (gettext("\tget [-rHp] [-d max] "
-		    "[-o \"all\" | field[,...]] [-t type[,...]] "
-		    "[-s source[,...]]\n"
+		    "[-o \"all\" | field[,...]]\n"
+		    "\t    [-t type[,...]] [-s source[,...]]\n"
 		    "\t    <\"all\" | property[,...]> "
 		    "[filesystem|volume|snapshot] ...\n"));
 	case HELP_INHERIT:
@@ -264,15 +268,16 @@ get_usage(zfs_help_t idx)
 	case HELP_RENAME:
 		return (gettext("\trename [-f] <filesystem|volume|snapshot> "
 		    "<filesystem|volume|snapshot>\n"
-		    "\trename [-f] -p <filesystem|volume> "
-		    "<filesystem|volume>\n"
+		    "\trename [-f] -p <filesystem|volume> <filesystem|volume>\n"
 		    "\trename -r <snapshot> <snapshot>\n"
 		    "\trename -u [-p] <filesystem> <filesystem>"));
 	case HELP_ROLLBACK:
 		return (gettext("\trollback [-rRf] <snapshot>\n"));
 	case HELP_SEND:
-		return (gettext("\tsend [-DnPpRv] "
-		    "[-i snapshot | -I snapshot] <snapshot>\n"));
+		return (gettext("\tsend [-DnPpRv] [-[iI] snapshot] "
+		    "<snapshot>\n"
+		    "\tsend [-i snapshot|bookmark] "
+		    "<filesystem|volume|snapshot>\n"));
 	case HELP_SET:
 		return (gettext("\tset <property=value> "
 		    "<filesystem|volume|snapshot> ...\n"));
@@ -280,7 +285,7 @@ get_usage(zfs_help_t idx)
 		return (gettext("\tshare <-a | filesystem>\n"));
 	case HELP_SNAPSHOT:
 		return (gettext("\tsnapshot|snap [-r] [-o property=value] ... "
-		    "<filesystem@snapname|volume@snapname> ...\n"));
+		    "<filesystem|volume>@<snap> ...\n"));
 	case HELP_UNMOUNT:
 		return (gettext("\tunmount|umount [-f] "
 		    "<-a | filesystem|mountpoint>\n"));
@@ -309,11 +314,13 @@ get_usage(zfs_help_t idx)
 		    "<filesystem|volume>\n"));
 	case HELP_USERSPACE:
 		return (gettext("\tuserspace [-Hinp] [-o field[,...]] "
-		    "[-s field]...\n\t    [-S field]... [-t type[,...]] "
+		    "[-s field] ...\n"
+		    "\t    [-S field] ... [-t type[,...]] "
 		    "<filesystem|snapshot>\n"));
 	case HELP_GROUPSPACE:
 		return (gettext("\tgroupspace [-Hinp] [-o field[,...]] "
-		    "[-s field]...\n\t    [-S field]... [-t type[,...]] "
+		    "[-s field] ...\n"
+		    "\t    [-S field] ... [-t type[,...]] "
 		    "<filesystem|snapshot>\n"));
 	case HELP_HOLD:
 		return (gettext("\thold [-r] <tag> <snapshot> ...\n"));
@@ -324,6 +331,8 @@ get_usage(zfs_help_t idx)
 	case HELP_DIFF:
 		return (gettext("\tdiff [-FHt] <snapshot> "
 		    "[snapshot|filesystem]\n"));
+	case HELP_BOOKMARK:
+		return (gettext("\tbookmark <snapshot> <bookmark>\n"));
 	}
 
 	abort();
@@ -926,6 +935,7 @@ typedef struct destroy_cbdata {
 	char		*cb_prevsnap;
 	int64_t		cb_snapused;
 	char		*cb_snapspec;
+	char		*cb_bookmark;
 } destroy_cbdata_t;
 
 /*
@@ -1195,7 +1205,7 @@ zfs_do_destroy(int argc, char **argv)
 	int err = 0;
 	int c;
 	zfs_handle_t *zhp = NULL;
-	char *at;
+	char *at, *pound;
 	zfs_type_t type = ZFS_TYPE_DATASET;
 
 	/* check options */
@@ -1247,6 +1257,7 @@ zfs_do_destroy(int argc, char **argv)
 	}
 
 	at = strchr(argv[0], '@');
+	pound = strchr(argv[0], '#');
 	if (at != NULL) {
 
 		/* Build the list of snaps to destroy in cb_nvl. */
@@ -1308,6 +1319,46 @@ zfs_do_destroy(int argc, char **argv)
 
 		if (err != 0)
 			rv = 1;
+	} else if (pound != NULL) {
+		int err;
+		nvlist_t *nvl;
+
+		if (cb.cb_dryrun) {
+			(void) fprintf(stderr,
+			    "dryrun is not supported with bookmark\n");
+			return (-1);
+		}
+
+		if (cb.cb_defer_destroy) {
+			(void) fprintf(stderr,
+			    "defer destroy is not supported with bookmark\n");
+			return (-1);
+		}
+
+		if (cb.cb_recurse) {
+			(void) fprintf(stderr,
+			    "recursive is not supported with bookmark\n");
+			return (-1);
+		}
+
+		if (!zfs_bookmark_exists(argv[0])) {
+			(void) fprintf(stderr, gettext("bookmark '%s' "
+			    "does not exist.\n"), argv[0]);
+			return (1);
+		}
+
+		nvl = fnvlist_alloc();
+		fnvlist_add_boolean(nvl, argv[0]);
+
+		err = lzc_destroy_bookmarks(nvl, NULL);
+		if (err != 0) {
+			(void) zfs_standard_error(g_zfs, err,
+			    "cannot destroy bookmark");
+		}
+
+		nvlist_free(cb.cb_nvl);
+
+		return (err);
 	} else {
 		/* Open the given dataset */
 		if ((zhp = zfs_open(g_zfs, argv[0], type)) == NULL)
@@ -1670,7 +1721,8 @@ zfs_do_get(int argc, char **argv)
 			flags &= ~ZFS_ITER_PROP_LISTSNAPS;
 			while (*optarg != '\0') {
 				static char *type_subopts[] = { "filesystem",
-				    "volume", "snapshot", "all", NULL };
+				    "volume", "snapshot", "bookmark",
+				    "all", NULL };
 
 				switch (getsubopt(&optarg, type_subopts,
 				    &value)) {
@@ -1684,7 +1736,11 @@ zfs_do_get(int argc, char **argv)
 					types |= ZFS_TYPE_SNAPSHOT;
 					break;
 				case 3:
-					types = ZFS_TYPE_DATASET;
+					types |= ZFS_TYPE_BOOKMARK;
+					break;
+				case 4:
+					types = ZFS_TYPE_DATASET |
+					    ZFS_TYPE_BOOKMARK;
 					break;
 
 				default:
@@ -3029,7 +3085,8 @@ zfs_do_list(int argc, char **argv)
 			flags &= ~ZFS_ITER_PROP_LISTSNAPS;
 			while (*optarg != '\0') {
 				static char *type_subopts[] = { "filesystem",
-				    "volume", "snapshot", "snap", "all", NULL };
+				    "volume", "snapshot", "snap", "bookmark",
+				    "all", NULL };
 
 				switch (getsubopt(&optarg, type_subopts,
 				    &value)) {
@@ -3044,9 +3101,12 @@ zfs_do_list(int argc, char **argv)
 					types |= ZFS_TYPE_SNAPSHOT;
 					break;
 				case 4:
-					types = ZFS_TYPE_DATASET;
+					types |= ZFS_TYPE_BOOKMARK;
+					break;
+				case 5:
+					types = ZFS_TYPE_DATASET |
+					    ZFS_TYPE_BOOKMARK;
 					break;
-
 				default:
 					(void) fprintf(stderr,
 					    gettext("invalid type '%s'\n"),
@@ -3286,9 +3346,29 @@ typedef struct rollback_cbdata {
 	char		*cb_target;
 	int		cb_error;
 	boolean_t	cb_recurse;
-	boolean_t	cb_dependent;
 } rollback_cbdata_t;
 
+static int
+rollback_check_dependent(zfs_handle_t *zhp, void *data)
+{
+	rollback_cbdata_t *cbp = data;
+
+	if (cbp->cb_first && cbp->cb_recurse) {
+		(void) fprintf(stderr, gettext("cannot rollback to "
+		    "'%s': clones of previous snapshots exist\n"),
+		    cbp->cb_target);
+		(void) fprintf(stderr, gettext("use '-R' to "
+		    "force deletion of the following clones and "
+		    "dependents:\n"));
+		cbp->cb_first = 0;
+		cbp->cb_error = 1;
+	}
+
+	(void) fprintf(stderr, "%s\n", zfs_get_name(zhp));
+
+	zfs_close(zhp);
+	return (0);
+}
 /*
  * Report any snapshots more recent than the one specified.  Used when '-r' is
  * not specified.  We reuse this same callback for the snapshot dependents - if
@@ -3305,52 +3385,30 @@ rollback_check(zfs_handle_t *zhp, void *data)
 		return (0);
 	}
 
-	if (!cbp->cb_dependent) {
-		if (strcmp(zfs_get_name(zhp), cbp->cb_target) != 0 &&
-		    zfs_get_type(zhp) == ZFS_TYPE_SNAPSHOT &&
-		    zfs_prop_get_int(zhp, ZFS_PROP_CREATETXG) >
-		    cbp->cb_create) {
-
-			if (cbp->cb_first && !cbp->cb_recurse) {
-				(void) fprintf(stderr, gettext("cannot "
-				    "rollback to '%s': more recent snapshots "
-				    "exist\n"),
-				    cbp->cb_target);
-				(void) fprintf(stderr, gettext("use '-r' to "
-				    "force deletion of the following "
-				    "snapshots:\n"));
-				cbp->cb_first = 0;
-				cbp->cb_error = 1;
-			}
-
-			if (cbp->cb_recurse) {
-				cbp->cb_dependent = B_TRUE;
-				if (zfs_iter_dependents(zhp, B_TRUE,
-				    rollback_check, cbp) != 0) {
-					zfs_close(zhp);
-					return (-1);
-				}
-				cbp->cb_dependent = B_FALSE;
-			} else {
-				(void) fprintf(stderr, "%s\n",
-				    zfs_get_name(zhp));
-			}
-		}
-	} else {
-		if (cbp->cb_first && cbp->cb_recurse) {
-			(void) fprintf(stderr, gettext("cannot rollback to "
-			    "'%s': clones of previous snapshots exist\n"),
+	if (zfs_prop_get_int(zhp, ZFS_PROP_CREATETXG) > cbp->cb_create) {
+		if (cbp->cb_first && !cbp->cb_recurse) {
+			(void) fprintf(stderr, gettext("cannot "
+			    "rollback to '%s': more recent snapshots "
+			    "or bookmarks exist\n"),
 			    cbp->cb_target);
-			(void) fprintf(stderr, gettext("use '-R' to "
-			    "force deletion of the following clones and "
-			    "dependents:\n"));
+			(void) fprintf(stderr, gettext("use '-r' to "
+			    "force deletion of the following "
+			    "snapshots and bookmarks:\n"));
 			cbp->cb_first = 0;
 			cbp->cb_error = 1;
 		}
 
-		(void) fprintf(stderr, "%s\n", zfs_get_name(zhp));
+		if (cbp->cb_recurse) {
+			if (zfs_iter_dependents(zhp, B_TRUE,
+			    rollback_check_dependent, cbp) != 0) {
+				zfs_close(zhp);
+				return (-1);
+			}
+		} else {
+			(void) fprintf(stderr, "%s\n",
+			    zfs_get_name(zhp));
+		}
 	}
-
 	zfs_close(zhp);
 	return (0);
 }
@@ -3420,7 +3478,9 @@ zfs_do_rollback(int argc, char **argv)
 	cb.cb_create = zfs_prop_get_int(snap, ZFS_PROP_CREATETXG);
 	cb.cb_first = B_TRUE;
 	cb.cb_error = 0;
-	if ((ret = zfs_iter_children(zhp, rollback_check, &cb)) != 0)
+	if ((ret = zfs_iter_snapshots(zhp, B_FALSE, rollback_check, &cb)) != 0)
+		goto out;
+	if ((ret = zfs_iter_bookmarks(zhp, rollback_check, &cb)) != 0)
 		goto out;
 
 	if ((ret = cb.cb_error) != 0)
@@ -3715,12 +3775,45 @@ zfs_do_send(int argc, char **argv)
 		return (1);
 	}
 
-	cp = strchr(argv[0], '@');
-	if (cp == NULL) {
-		(void) fprintf(stderr,
-		    gettext("argument must be a snapshot\n"));
-		usage(B_FALSE);
+	/*
+	 * Special case sending a filesystem, or from a bookmark.
+	 */
+	if (strchr(argv[0], '@') == NULL ||
+	    (fromname && strchr(fromname, '#') != NULL)) {
+		char frombuf[ZFS_MAXNAMELEN];
+
+		if (flags.replicate || flags.doall || flags.props ||
+		    flags.dedup || flags.dryrun || flags.verbose ||
+		    flags.progress) {
+			(void) fprintf(stderr,
+			    gettext("Error: "
+			    "Unsupported flag with filesystem or bookmark.\n"));
+			return (1);
+		}
+
+		zhp = zfs_open(g_zfs, argv[0], ZFS_TYPE_DATASET);
+		if (zhp == NULL)
+			return (1);
+
+		if (fromname != NULL &&
+		    (fromname[0] == '#' || fromname[0] == '@')) {
+			/*
+			 * Incremental source name begins with # or @.
+			 * Default to same fs as target.
+			 */
+			(void) strncpy(frombuf, argv[0], sizeof (frombuf));
+			cp = strchr(frombuf, '@');
+			if (cp != NULL)
+				*cp = '\0';
+			(void) strlcat(frombuf, fromname, sizeof (frombuf));
+			fromname = frombuf;
+		}
+		err = zfs_send_one(zhp, fromname, STDOUT_FILENO);
+		zfs_close(zhp);
+		return (err != 0);
 	}
+
+	cp = strchr(argv[0], '@');
 	*cp = '\0';
 	toname = cp + 1;
 	zhp = zfs_open(g_zfs, argv[0], ZFS_TYPE_FILESYSTEM | ZFS_TYPE_VOLUME);
@@ -3876,6 +3969,7 @@ zfs_do_receive(int argc, char **argv)
 #define	ZFS_DELEG_PERM_HOLD		"hold"
 #define	ZFS_DELEG_PERM_RELEASE		"release"
 #define	ZFS_DELEG_PERM_DIFF		"diff"
+#define	ZFS_DELEG_PERM_BOOKMARK		"bookmark"
 
 #define	ZFS_NUM_DELEG_NOTES ZFS_DELEG_NOTE_NONE
 
@@ -3895,6 +3989,7 @@ static zfs_deleg_perm_tab_t zfs_deleg_perm_tbl[] = {
 	{ ZFS_DELEG_PERM_SEND, ZFS_DELEG_NOTE_SEND },
 	{ ZFS_DELEG_PERM_SHARE, ZFS_DELEG_NOTE_SHARE },
 	{ ZFS_DELEG_PERM_SNAPSHOT, ZFS_DELEG_NOTE_SNAPSHOT },
+	{ ZFS_DELEG_PERM_BOOKMARK, ZFS_DELEG_NOTE_BOOKMARK },
 
 	{ ZFS_DELEG_PERM_GROUPQUOTA, ZFS_DELEG_NOTE_GROUPQUOTA },
 	{ ZFS_DELEG_PERM_GROUPUSED, ZFS_DELEG_NOTE_GROUPUSED },
@@ -6666,6 +6761,108 @@ zfs_do_diff(int argc, char **argv)
 	return (err != 0);
 }
 
+/*
+ * zfs bookmark <fs@snap> <fs#bmark>
+ *
+ * Creates a bookmark with the given name from the given snapshot.
+ */
+static int
+zfs_do_bookmark(int argc, char **argv)
+{
+	char snapname[ZFS_MAXNAMELEN];
+	zfs_handle_t *zhp;
+	nvlist_t *nvl;
+	int ret = 0;
+	int c;
+
+	/* check options */
+	while ((c = getopt(argc, argv, "")) != -1) {
+		switch (c) {
+		case '?':
+			(void) fprintf(stderr,
+			    gettext("invalid option '%c'\n"), optopt);
+			goto usage;
+		}
+	}
+
+	argc -= optind;
+	argv += optind;
+
+	/* check number of arguments */
+	if (argc < 1) {
+		(void) fprintf(stderr, gettext("missing snapshot argument\n"));
+		goto usage;
+	}
+	if (argc < 2) {
+		(void) fprintf(stderr, gettext("missing bookmark argument\n"));
+		goto usage;
+	}
+
+	if (strchr(argv[1], '#') == NULL) {
+		(void) fprintf(stderr,
+		    gettext("invalid bookmark name '%s' -- "
+		    "must contain a '#'\n"), argv[1]);
+		goto usage;
+	}
+
+	if (argv[0][0] == '@') {
+		/*
+		 * Snapshot name begins with @.
+		 * Default to same fs as bookmark.
+		 */
+		(void) strncpy(snapname, argv[1], sizeof (snapname));
+		*strchr(snapname, '#') = '\0';
+		(void) strlcat(snapname, argv[0], sizeof (snapname));
+	} else {
+		(void) strncpy(snapname, argv[0], sizeof (snapname));
+	}
+	zhp = zfs_open(g_zfs, snapname, ZFS_TYPE_SNAPSHOT);
+	if (zhp == NULL)
+		goto usage;
+	zfs_close(zhp);
+
+
+	nvl = fnvlist_alloc();
+	fnvlist_add_string(nvl, argv[1], snapname);
+	ret = lzc_bookmark(nvl, NULL);
+	fnvlist_free(nvl);
+
+	if (ret != 0) {
+		const char *err_msg;
+		char errbuf[1024];
+
+		(void) snprintf(errbuf, sizeof (errbuf),
+		    dgettext(TEXT_DOMAIN,
+		    "cannot create bookmark '%s'"), argv[1]);
+
+		switch (ret) {
+		case EXDEV:
+			err_msg = "bookmark is in a different pool";
+			break;
+		case EEXIST:
+			err_msg = "bookmark exists";
+			break;
+		case EINVAL:
+			err_msg = "invalid argument";
+			break;
+		case ENOTSUP:
+			err_msg = "bookmark feature not enabled";
+			break;
+		default:
+			err_msg = "unknown error";
+			break;
+		}
+		(void) fprintf(stderr, "%s: %s\n", errbuf,
+		    dgettext(TEXT_DOMAIN, err_msg));
+	}
+
+	return (ret);
+
+usage:
+	usage(B_FALSE);
+	return (-1);
+}
+
 int
 main(int argc, char **argv)
 {
diff --git a/cddl/contrib/opensolaris/cmd/zpool/zpool-features.7 b/cddl/contrib/opensolaris/cmd/zpool/zpool-features.7
index 43eede5..5c6c47a 100644
--- a/cddl/contrib/opensolaris/cmd/zpool/zpool-features.7
+++ b/cddl/contrib/opensolaris/cmd/zpool/zpool-features.7
@@ -23,7 +23,7 @@
 .\"
 .\" $FreeBSD$
 .\"
-.Dd December 31, 2013
+.Dd January 2, 2014
 .Dt ZPOOL-FEATURES 7
 .Os
 .Sh NAME
@@ -222,12 +222,16 @@ command. Please note that doing so will
 immediately activate the
 .Sy lz4_compress
 feature on the underlying
-pool (even before any data is written). Since this feature is not
-read-only compatible, this operation will render the pool unimportable
-on systems without support for the
+pool
+.Pq even before any data is written ,
+and the feature will not be
+deactivated.
+Since this feature is not read-only compatible, this
+operation will render the pool unimportable on systems without support
+for the
 .Sy lz4_compress
-feature. At the
-moment, this operation cannot be reversed. Booting off of
+feature.
+Booting off of
 .Sy lz4
 -compressed root pools is supported.
 .It Sy multi_vdev_crash_dump
@@ -286,6 +290,25 @@ and will be returned to the
 .Sy enabled
 state when all datasets that use
 this feature are destroyed.
+.It Sy bookmarks
+.Bl -column "READ\-ONLY COMPATIBLE" "com.delphix:bookmarks"
+.It GUID Ta com.delphix:bookmarks
+.It READ\-ONLY COMPATIBLE Ta yes
+.It DEPENDENCIES Ta extensible_dataset
+.El
+.Pp
+This feature enables use of the
+.Nm zfs
+.Cm bookmark
+subcommand.
+.Pp
+This feature is
+.Sy active
+while any bookmarks exist in the pool.
+All bookmarks in the pool can be listed by running
+.Nm zfs
+.Cm list
+.Fl t No bookmark Fl r Ar poolname .
 .It Sy enabled_txg
 .Bl -column "READ\-ONLY COMPATIBLE" "com.delphix:enabled_txg"
 .It GUID Ta com.delphix:enabled_txg
diff --git a/cddl/contrib/opensolaris/lib/libzfs/common/libzfs.h b/cddl/contrib/opensolaris/lib/libzfs/common/libzfs.h
index 0a433dc..a1d78e2 100644
--- a/cddl/contrib/opensolaris/lib/libzfs/common/libzfs.h
+++ b/cddl/contrib/opensolaris/lib/libzfs/common/libzfs.h
@@ -23,7 +23,7 @@
  * Copyright (c) 2005, 2010, Oracle and/or its affiliates. All rights reserved.
  * Copyright (c) 2011 Pawel Jakub Dawidek <pawel@dawidek.net>.
  * All rights reserved.
- * Copyright (c) 2012 by Delphix. All rights reserved.
+ * Copyright (c) 2013 by Delphix. All rights reserved.
  * Copyright (c) 2012, Joyent, Inc. All rights reserved.
  * Copyright (c) 2012 Martin Matuska <mm@FreeBSD.org>. All rights reserved.
  * Copyright (c) 2013 Steven Hartland. All rights reserved.
@@ -193,6 +193,7 @@ extern int zpool_log_history(libzfs_handle_t *, const char *);
 extern int libzfs_errno(libzfs_handle_t *);
 extern const char *libzfs_error_action(libzfs_handle_t *);
 extern const char *libzfs_error_description(libzfs_handle_t *);
+extern int zfs_standard_error(libzfs_handle_t *, int, const char *);
 extern void libzfs_mnttab_init(libzfs_handle_t *);
 extern void libzfs_mnttab_fini(libzfs_handle_t *);
 extern void libzfs_mnttab_cache(libzfs_handle_t *, boolean_t);
@@ -537,6 +538,7 @@ extern int zfs_iter_filesystems(zfs_handle_t *, zfs_iter_f, void *);
 extern int zfs_iter_snapshots(zfs_handle_t *, boolean_t, zfs_iter_f, void *);
 extern int zfs_iter_snapshots_sorted(zfs_handle_t *, zfs_iter_f, void *);
 extern int zfs_iter_snapspec(zfs_handle_t *, const char *, zfs_iter_f, void *);
+extern int zfs_iter_bookmarks(zfs_handle_t *, zfs_iter_f, void *);
 
 typedef struct get_all_cb {
 	zfs_handle_t	**cb_handles;
@@ -611,6 +613,7 @@ typedef boolean_t (snapfilter_cb_t)(zfs_handle_t *, void *);
 
 extern int zfs_send(zfs_handle_t *, const char *, const char *,
     sendflags_t *, int, snapfilter_cb_t, void *, nvlist_t **);
+extern int zfs_send_one(zfs_handle_t *, const char *, int);
 
 extern int zfs_promote(zfs_handle_t *);
 extern int zfs_hold(zfs_handle_t *, const char *, const char *,
@@ -680,6 +683,7 @@ extern zfs_handle_t *zfs_path_to_zhandle(libzfs_handle_t *, char *, zfs_type_t);
 extern boolean_t zfs_dataset_exists(libzfs_handle_t *, const char *,
     zfs_type_t);
 extern int zfs_spa_version(zfs_handle_t *, int *);
+extern boolean_t zfs_bookmark_exists(const char *path);
 
 /*
  * Mount support functions.
diff --git a/cddl/contrib/opensolaris/lib/libzfs/common/libzfs_dataset.c b/cddl/contrib/opensolaris/lib/libzfs/common/libzfs_dataset.c
index 351eb1c..4abb105 100644
--- a/cddl/contrib/opensolaris/lib/libzfs/common/libzfs_dataset.c
+++ b/cddl/contrib/opensolaris/lib/libzfs/common/libzfs_dataset.c
@@ -295,7 +295,7 @@ zpool_handle(zfs_handle_t *zhp)
 	int len;
 	zpool_handle_t *zph;
 
-	len = strcspn(zhp->zfs_name, "/@") + 1;
+	len = strcspn(zhp->zfs_name, "/@#") + 1;
 	pool_name = zfs_alloc(zhp->zfs_hdl, len);
 	(void) strlcpy(pool_name, zhp->zfs_name, len);
 
@@ -579,6 +579,70 @@ zfs_handle_dup(zfs_handle_t *zhp_orig)
 	return (zhp);
 }
 
+boolean_t
+zfs_bookmark_exists(const char *path)
+{
+	nvlist_t *bmarks;
+	nvlist_t *props;
+	char fsname[ZFS_MAXNAMELEN];
+	char *bmark_name;
+	char *pound;
+	int err;
+	boolean_t rv;
+
+
+	(void) strlcpy(fsname, path, sizeof (fsname));
+	pound = strchr(fsname, '#');
+	if (pound == NULL)
+		return (B_FALSE);
+
+	*pound = '\0';
+	bmark_name = pound + 1;
+	props = fnvlist_alloc();
+	err = lzc_get_bookmarks(fsname, props, &bmarks);
+	nvlist_free(props);
+	if (err != 0) {
+		nvlist_free(bmarks);
+		return (B_FALSE);
+	}
+
+	rv = nvlist_exists(bmarks, bmark_name);
+	nvlist_free(bmarks);
+	return (rv);
+}
+
+zfs_handle_t *
+make_bookmark_handle(zfs_handle_t *parent, const char *path,
+    nvlist_t *bmark_props)
+{
+	zfs_handle_t *zhp = calloc(sizeof (zfs_handle_t), 1);
+
+	if (zhp == NULL)
+		return (NULL);
+
+	/* Fill in the name. */
+	zhp->zfs_hdl = parent->zfs_hdl;
+	(void) strlcpy(zhp->zfs_name, path, sizeof (zhp->zfs_name));
+
+	/* Set the property lists. */
+	if (nvlist_dup(bmark_props, &zhp->zfs_props, 0) != 0) {
+		free(zhp);
+		return (NULL);
+	}
+
+	/* Set the types. */
+	zhp->zfs_head_type = parent->zfs_head_type;
+	zhp->zfs_type = ZFS_TYPE_BOOKMARK;
+
+	if ((zhp->zpool_hdl = zpool_handle(zhp)) == NULL) {
+		nvlist_free(zhp->zfs_props);
+		free(zhp);
+		return (NULL);
+	}
+
+	return (zhp);
+}
+
 /*
  * Opens the given snapshot, filesystem, or volume.   The 'types'
  * argument is a mask of acceptable types.  The function will print an
@@ -2271,6 +2335,9 @@ zfs_prop_get(zfs_handle_t *zhp, zfs_prop_t prop, char *propbuf, size_t proplen,
 		case ZFS_TYPE_SNAPSHOT:
 			str = "snapshot";
 			break;
+		case ZFS_TYPE_BOOKMARK:
+			str = "bookmark";
+			break;
 		default:
 			abort();
 		}
@@ -3133,6 +3200,19 @@ zfs_destroy(zfs_handle_t *zhp, boolean_t defer)
 {
 	zfs_cmd_t zc = { 0 };
 
+	if (zhp->zfs_type == ZFS_TYPE_BOOKMARK) {
+		nvlist_t *nv = fnvlist_alloc();
+		fnvlist_add_boolean(nv, zhp->zfs_name);
+		int error = lzc_destroy_bookmarks(nv, NULL);
+		fnvlist_free(nv);
+		if (error != 0) {
+			return (zfs_standard_error_fmt(zhp->zfs_hdl, errno,
+			    dgettext(TEXT_DOMAIN, "cannot destroy '%s'"),
+			    zhp->zfs_name));
+		}
+		return (0);
+	}
+
 	(void) strlcpy(zc.zc_name, zhp->zfs_name, sizeof (zc.zc_name));
 
 	if (ZFS_IS_VOLUME(zhp)) {
@@ -3515,45 +3595,44 @@ typedef struct rollback_data {
 	const char	*cb_target;		/* the snapshot */
 	uint64_t	cb_create;		/* creation time reference */
 	boolean_t	cb_error;
-	boolean_t	cb_dependent;
 	boolean_t	cb_force;
 } rollback_data_t;
 
 static int
-rollback_destroy(zfs_handle_t *zhp, void *data)
+rollback_destroy_dependent(zfs_handle_t *zhp, void *data)
 {
 	rollback_data_t *cbp = data;
+	prop_changelist_t *clp;
 
-	if (!cbp->cb_dependent) {
-		if (strcmp(zhp->zfs_name, cbp->cb_target) != 0 &&
-		    zfs_get_type(zhp) == ZFS_TYPE_SNAPSHOT &&
-		    zfs_prop_get_int(zhp, ZFS_PROP_CREATETXG) >
-		    cbp->cb_create) {
+	/* We must destroy this clone; first unmount it */
+	clp = changelist_gather(zhp, ZFS_PROP_NAME, 0,
+	    cbp->cb_force ? MS_FORCE: 0);
+	if (clp == NULL || changelist_prefix(clp) != 0) {
+		cbp->cb_error = B_TRUE;
+		zfs_close(zhp);
+		return (0);
+	}
+	if (zfs_destroy(zhp, B_FALSE) != 0)
+		cbp->cb_error = B_TRUE;
+	else
+		changelist_remove(clp, zhp->zfs_name);
+	(void) changelist_postfix(clp);
+	changelist_free(clp);
 
-			cbp->cb_dependent = B_TRUE;
-			cbp->cb_error |= zfs_iter_dependents(zhp, B_FALSE,
-			    rollback_destroy, cbp);
-			cbp->cb_dependent = B_FALSE;
+	zfs_close(zhp);
+	return (0);
+}
 
-			cbp->cb_error |= zfs_destroy(zhp, B_FALSE);
-		}
-	} else {
-		/* We must destroy this clone; first unmount it */
-		prop_changelist_t *clp;
+static int
+rollback_destroy(zfs_handle_t *zhp, void *data)
+{
+	rollback_data_t *cbp = data;
 
-		clp = changelist_gather(zhp, ZFS_PROP_NAME, 0,
-		    cbp->cb_force ? MS_FORCE: 0);
-		if (clp == NULL || changelist_prefix(clp) != 0) {
-			cbp->cb_error = B_TRUE;
-			zfs_close(zhp);
-			return (0);
-		}
-		if (zfs_destroy(zhp, B_FALSE) != 0)
-			cbp->cb_error = B_TRUE;
-		else
-			changelist_remove(clp, zhp->zfs_name);
-		(void) changelist_postfix(clp);
-		changelist_free(clp);
+	if (zfs_prop_get_int(zhp, ZFS_PROP_CREATETXG) > cbp->cb_create) {
+		cbp->cb_error |= zfs_iter_dependents(zhp, B_FALSE,
+		    rollback_destroy_dependent, cbp);
+
+		cbp->cb_error |= zfs_destroy(zhp, B_FALSE);
 	}
 
 	zfs_close(zhp);
@@ -3564,8 +3643,8 @@ rollback_destroy(zfs_handle_t *zhp, void *data)
  * Given a dataset, rollback to a specific snapshot, discarding any
  * data changes since then and making it the active dataset.
  *
- * Any snapshots more recent than the target are destroyed, along with
- * their dependents.
+ * Any snapshots and bookmarks more recent than the target are
+ * destroyed, along with their dependents (i.e. clones).
  */
 int
 zfs_rollback(zfs_handle_t *zhp, zfs_handle_t *snap, boolean_t force)
@@ -3585,7 +3664,8 @@ zfs_rollback(zfs_handle_t *zhp, zfs_handle_t *snap, boolean_t force)
 	cb.cb_force = force;
 	cb.cb_target = snap->zfs_name;
 	cb.cb_create = zfs_prop_get_int(snap, ZFS_PROP_CREATETXG);
-	(void) zfs_iter_children(zhp, rollback_destroy, &cb);
+	(void) zfs_iter_snapshots(zhp, B_FALSE, rollback_destroy, &cb);
+	(void) zfs_iter_bookmarks(zhp, rollback_destroy, &cb);
 
 	if (cb.cb_error)
 		return (-1);
diff --git a/cddl/contrib/opensolaris/lib/libzfs/common/libzfs_impl.h b/cddl/contrib/opensolaris/lib/libzfs/common/libzfs_impl.h
index 11a57a9..481ae52 100644
--- a/cddl/contrib/opensolaris/lib/libzfs/common/libzfs_impl.h
+++ b/cddl/contrib/opensolaris/lib/libzfs/common/libzfs_impl.h
@@ -23,7 +23,7 @@
  * Copyright (c) 2005, 2010, Oracle and/or its affiliates. All rights reserved.
  * Copyright (c) 2011 Pawel Jakub Dawidek <pawel@dawidek.net>.
  * All rights reserved.
- * Copyright (c) 2012 by Delphix. All rights reserved.
+ * Copyright (c) 2013 by Delphix. All rights reserved.
  * Copyright (c) 2013 Martin Matuska <mm@FreeBSD.org>. All rights reserved.
  */
 
@@ -191,6 +191,8 @@ int create_parents(libzfs_handle_t *, char *, int);
 boolean_t isa_child_of(const char *dataset, const char *parent);
 
 zfs_handle_t *make_dataset_handle(libzfs_handle_t *, const char *);
+zfs_handle_t *make_bookmark_handle(zfs_handle_t *, const char *,
+    nvlist_t *props);
 
 int zpool_open_silent(libzfs_handle_t *, const char *, zpool_handle_t **);
 
diff --git a/cddl/contrib/opensolaris/lib/libzfs/common/libzfs_iter.c b/cddl/contrib/opensolaris/lib/libzfs/common/libzfs_iter.c
index 259d045..9698a72 100644
--- a/cddl/contrib/opensolaris/lib/libzfs/common/libzfs_iter.c
+++ b/cddl/contrib/opensolaris/lib/libzfs/common/libzfs_iter.c
@@ -21,7 +21,7 @@
 
 /*
  * Copyright (c) 2005, 2010, Oracle and/or its affiliates. All rights reserved.
- * Copyright (c) 2012 by Delphix. All rights reserved.
+ * Copyright (c) 2013 by Delphix. All rights reserved.
  * Copyright (c) 2012 Pawel Jakub Dawidek <pawel@dawidek.net>.
  * All rights reserved.
  * Copyright 2013 Nexenta Systems, Inc.  All rights reserved.
@@ -146,7 +146,8 @@ zfs_iter_snapshots(zfs_handle_t *zhp, boolean_t simple, zfs_iter_f func,
 	zfs_handle_t *nzhp;
 	int ret;
 
-	if (zhp->zfs_type == ZFS_TYPE_SNAPSHOT)
+	if (zhp->zfs_type == ZFS_TYPE_SNAPSHOT ||
+	    zhp->zfs_type == ZFS_TYPE_BOOKMARK)
 		return (0);
 
 	zc.zc_simple = simple;
@@ -173,6 +174,59 @@ zfs_iter_snapshots(zfs_handle_t *zhp, boolean_t simple, zfs_iter_f func,
 }
 
 /*
+ * Iterate over all bookmarks
+ */
+int
+zfs_iter_bookmarks(zfs_handle_t *zhp, zfs_iter_f func, void *data)
+{
+	zfs_handle_t *nzhp;
+	nvlist_t *props = NULL;
+	nvlist_t *bmarks = NULL;
+	int err;
+
+	if ((zfs_get_type(zhp) & (ZFS_TYPE_SNAPSHOT | ZFS_TYPE_BOOKMARK)) != 0)
+		return (0);
+
+	/* Setup the requested properties nvlist. */
+	props = fnvlist_alloc();
+	fnvlist_add_boolean(props, zfs_prop_to_name(ZFS_PROP_GUID));
+	fnvlist_add_boolean(props, zfs_prop_to_name(ZFS_PROP_CREATETXG));
+	fnvlist_add_boolean(props, zfs_prop_to_name(ZFS_PROP_CREATION));
+
+	/* Allocate an nvlist to hold the bookmarks. */
+	bmarks = fnvlist_alloc();
+
+	if ((err = lzc_get_bookmarks(zhp->zfs_name, props, &bmarks)) != 0)
+		goto out;
+
+	for (nvpair_t *pair = nvlist_next_nvpair(bmarks, NULL);
+	    pair != NULL; pair = nvlist_next_nvpair(bmarks, pair)) {
+		char name[ZFS_MAXNAMELEN];
+		char *bmark_name;
+		nvlist_t *bmark_props;
+
+		bmark_name = nvpair_name(pair);
+		bmark_props = fnvpair_value_nvlist(pair);
+
+		(void) snprintf(name, sizeof (name), "%s#%s", zhp->zfs_name,
+		    bmark_name);
+
+		nzhp = make_bookmark_handle(zhp, name, bmark_props);
+		if (nzhp == NULL)
+			continue;
+
+		if ((err = func(nzhp, data)) != 0)
+			goto out;
+	}
+
+out:
+	fnvlist_free(props);
+	fnvlist_free(bmarks);
+
+	return (err);
+}
+
+/*
  * Routines for dealing with the sorted snapshot functionality
  */
 typedef struct zfs_node {
@@ -406,13 +460,13 @@ static int
 iter_dependents_cb(zfs_handle_t *zhp, void *arg)
 {
 	iter_dependents_arg_t *ida = arg;
-	int err;
+	int err = 0;
 	boolean_t first = ida->first;
 	ida->first = B_FALSE;
 
 	if (zhp->zfs_type == ZFS_TYPE_SNAPSHOT) {
 		err = zfs_iter_clones(zhp, iter_dependents_cb, ida);
-	} else {
+	} else if (zhp->zfs_type != ZFS_TYPE_BOOKMARK) {
 		iter_stack_frame_t isf;
 		iter_stack_frame_t *f;
 
diff --git a/cddl/contrib/opensolaris/lib/libzfs/common/libzfs_sendrecv.c b/cddl/contrib/opensolaris/lib/libzfs/common/libzfs_sendrecv.c
index 1fc46c2..feddb69 100644
--- a/cddl/contrib/opensolaris/lib/libzfs/common/libzfs_sendrecv.c
+++ b/cddl/contrib/opensolaris/lib/libzfs/common/libzfs_sendrecv.c
@@ -21,7 +21,7 @@
 
 /*
  * Copyright (c) 2005, 2010, Oracle and/or its affiliates. All rights reserved.
- * Copyright (c) 2012 by Delphix. All rights reserved.
+ * Copyright (c) 2013 by Delphix. All rights reserved.
  * Copyright (c) 2012, Joyent, Inc. All rights reserved.
  * Copyright (c) 2012 Pawel Jakub Dawidek <pawel@dawidek.net>.
  * All rights reserved.
@@ -1619,6 +1619,62 @@ err_out:
 	return (err);
 }
 
+int
+zfs_send_one(zfs_handle_t *zhp, const char *from, int fd)
+{
+	int err;
+	libzfs_handle_t *hdl = zhp->zfs_hdl;
+
+	char errbuf[1024];
+	(void) snprintf(errbuf, sizeof (errbuf), dgettext(TEXT_DOMAIN,
+	    "warning: cannot send '%s'"), zhp->zfs_name);
+
+	err = lzc_send(zhp->zfs_name, from, fd);
+	if (err != 0) {
+		switch (errno) {
+		case EXDEV:
+			zfs_error_aux(hdl, dgettext(TEXT_DOMAIN,
+			    "not an earlier snapshot from the same fs"));
+			return (zfs_error(hdl, EZFS_CROSSTARGET, errbuf));
+
+		case ENOENT:
+		case ESRCH:
+			if (lzc_exists(zhp->zfs_name)) {
+				zfs_error_aux(hdl, dgettext(TEXT_DOMAIN,
+				    "incremental source (%s) does not exist"),
+				    from);
+			}
+			return (zfs_error(hdl, EZFS_NOENT, errbuf));
+
+		case EBUSY:
+			zfs_error_aux(hdl, dgettext(TEXT_DOMAIN,
+			    "target is busy; if a filesystem, "
+			    "it must not be mounted"));
+			return (zfs_error(hdl, EZFS_BUSY, errbuf));
+
+		case EDQUOT:
+		case EFBIG:
+		case EIO:
+		case ENOLINK:
+		case ENOSPC:
+#ifdef illumos
+		case ENOSTR:
+#endif
+		case ENXIO:
+		case EPIPE:
+		case ERANGE:
+		case EFAULT:
+		case EROFS:
+			zfs_error_aux(hdl, strerror(errno));
+			return (zfs_error(hdl, EZFS_BADBACKUP, errbuf));
+
+		default:
+			return (zfs_standard_error(hdl, errno, errbuf));
+		}
+	}
+	return (err != 0);
+}
+
 /*
  * Routines specific to "zfs recv"
  */
diff --git a/cddl/contrib/opensolaris/lib/libzfs_core/common/libzfs_core.c b/cddl/contrib/opensolaris/lib/libzfs_core/common/libzfs_core.c
index 6f4b46c..1c87223 100644
--- a/cddl/contrib/opensolaris/lib/libzfs_core/common/libzfs_core.c
+++ b/cddl/contrib/opensolaris/lib/libzfs_core/common/libzfs_core.c
@@ -486,18 +486,30 @@ lzc_get_holds(const char *snapname, nvlist_t **holdsp)
 }
 
 /*
- * If fromsnap is NULL, a full (non-incremental) stream will be sent.
+ *
+ * "snapname" is the full name of the snapshot to send (e.g. "pool/fs@snap")
+ *
+ * If "from" is NULL, a full (non-incremental) stream will be sent.
+ * If "from" is non-NULL, it must be the full name of a snapshot or
+ * bookmark to send an incremental from (e.g. "pool/fs@earlier_snap" or
+ * "pool/fs#earlier_bmark").  If non-NULL, the specified snapshot or
+ * bookmark must represent an earlier point in the history of "snapname").
+ * It can be an earlier snapshot in the same filesystem or zvol as "snapname",
+ * or it can be the origin of "snapname"'s filesystem, or an earlier
+ * snapshot in the origin, etc.
+ *
+ * "fd" is the file descriptor to write the send stream to.
  */
 int
-lzc_send(const char *snapname, const char *fromsnap, int fd)
+lzc_send(const char *snapname, const char *from, int fd)
 {
 	nvlist_t *args;
 	int err;
 
 	args = fnvlist_alloc();
 	fnvlist_add_int32(args, "fd", fd);
-	if (fromsnap != NULL)
-		fnvlist_add_string(args, "fromsnap", fromsnap);
+	if (from != NULL)
+		fnvlist_add_string(args, "fromsnap", from);
 	err = lzc_ioctl(ZFS_IOC_SEND_NEW, snapname, args, NULL);
 	nvlist_free(args);
 	return (err);
@@ -652,3 +664,97 @@ lzc_rollback(const char *fsname, char *snapnamebuf, int snapnamelen)
 	}
 	return (err);
 }
+
+/*
+ * Creates bookmarks.
+ *
+ * The bookmarks nvlist maps from name of the bookmark (e.g. "pool/fs#bmark") to
+ * the name of the snapshot (e.g. "pool/fs@snap").  All the bookmarks and
+ * snapshots must be in the same pool.
+ *
+ * The returned results nvlist will have an entry for each bookmark that failed.
+ * The value will be the (int32) error code.
+ *
+ * The return value will be 0 if all bookmarks were created, otherwise it will
+ * be the errno of a (undetermined) bookmarks that failed.
+ */
+int
+lzc_bookmark(nvlist_t *bookmarks, nvlist_t **errlist)
+{
+	nvpair_t *elem;
+	int error;
+	char pool[MAXNAMELEN];
+
+	/* determine the pool name */
+	elem = nvlist_next_nvpair(bookmarks, NULL);
+	if (elem == NULL)
+		return (0);
+	(void) strlcpy(pool, nvpair_name(elem), sizeof (pool));
+	pool[strcspn(pool, "/#")] = '\0';
+
+	error = lzc_ioctl(ZFS_IOC_BOOKMARK, pool, bookmarks, errlist);
+
+	return (error);
+}
+
+/*
+ * Retrieve bookmarks.
+ *
+ * Retrieve the list of bookmarks for the given file system. The props
+ * parameter is an nvlist of property names (with no values) that will be
+ * returned for each bookmark.
+ *
+ * The following are valid properties on bookmarks, all of which are numbers
+ * (represented as uint64 in the nvlist)
+ *
+ * "guid" - globally unique identifier of the snapshot it refers to
+ * "createtxg" - txg when the snapshot it refers to was created
+ * "creation" - timestamp when the snapshot it refers to was created
+ *
+ * The format of the returned nvlist as follows:
+ * <short name of bookmark> -> {
+ *     <name of property> -> {
+ *         "value" -> uint64
+ *     }
+ *  }
+ */
+int
+lzc_get_bookmarks(const char *fsname, nvlist_t *props, nvlist_t **bmarks)
+{
+	return (lzc_ioctl(ZFS_IOC_GET_BOOKMARKS, fsname, props, bmarks));
+}
+
+/*
+ * Destroys bookmarks.
+ *
+ * The keys in the bmarks nvlist are the bookmarks to be destroyed.
+ * They must all be in the same pool.  Bookmarks are specified as
+ * <fs>#<bmark>.
+ *
+ * Bookmarks that do not exist will be silently ignored.
+ *
+ * The return value will be 0 if all bookmarks that existed were destroyed.
+ *
+ * Otherwise the return value will be the errno of a (undetermined) bookmark
+ * that failed, no bookmarks will be destroyed, and the errlist will have an
+ * entry for each bookmarks that failed.  The value in the errlist will be
+ * the (int32) error code.
+ */
+int
+lzc_destroy_bookmarks(nvlist_t *bmarks, nvlist_t **errlist)
+{
+	nvpair_t *elem;
+	int error;
+	char pool[MAXNAMELEN];
+
+	/* determine the pool name */
+	elem = nvlist_next_nvpair(bmarks, NULL);
+	if (elem == NULL)
+		return (0);
+	(void) strlcpy(pool, nvpair_name(elem), sizeof (pool));
+	pool[strcspn(pool, "/#")] = '\0';
+
+	error = lzc_ioctl(ZFS_IOC_DESTROY_BOOKMARKS, pool, bmarks, errlist);
+
+	return (error);
+}
diff --git a/cddl/contrib/opensolaris/lib/libzfs_core/common/libzfs_core.h b/cddl/contrib/opensolaris/lib/libzfs_core/common/libzfs_core.h
index c55caf4..380560f 100644
--- a/cddl/contrib/opensolaris/lib/libzfs_core/common/libzfs_core.h
+++ b/cddl/contrib/opensolaris/lib/libzfs_core/common/libzfs_core.h
@@ -39,27 +39,27 @@ extern "C" {
 int libzfs_core_init(void);
 void libzfs_core_fini(void);
 
-int lzc_snapshot(nvlist_t *snaps, nvlist_t *props, nvlist_t **errlist);
-int lzc_create(const char *fsname, dmu_objset_type_t type, nvlist_t *props);
-int lzc_clone(const char *fsname, const char *origin, nvlist_t *props);
-int lzc_destroy_snaps(nvlist_t *snaps, boolean_t defer, nvlist_t **errlist);
+int lzc_snapshot(nvlist_t *, nvlist_t *, nvlist_t **);
+int lzc_create(const char *, dmu_objset_type_t, nvlist_t *);
+int lzc_clone(const char *, const char *, nvlist_t *);
+int lzc_destroy_snaps(nvlist_t *, boolean_t, nvlist_t **);
+int lzc_bookmark(nvlist_t *, nvlist_t **);
+int lzc_get_bookmarks(const char *, nvlist_t *, nvlist_t **);
+int lzc_destroy_bookmarks(nvlist_t *, nvlist_t **);
 
-int lzc_snaprange_space(const char *firstsnap, const char *lastsnap,
-    uint64_t *usedp);
+int lzc_snaprange_space(const char *, const char *, uint64_t *);
 
-int lzc_hold(nvlist_t *holds, int cleanup_fd, nvlist_t **errlist);
-int lzc_release(nvlist_t *holds, nvlist_t **errlist);
-int lzc_get_holds(const char *snapname, nvlist_t **holdsp);
+int lzc_hold(nvlist_t *, int, nvlist_t **);
+int lzc_release(nvlist_t *, nvlist_t **);
+int lzc_get_holds(const char *, nvlist_t **);
 
-int lzc_send(const char *snapname, const char *fromsnap, int fd);
-int lzc_receive(const char *snapname, nvlist_t *props, const char *origin,
-    boolean_t force, int fd);
-int lzc_send_space(const char *snapname, const char *fromsnap,
-    uint64_t *result);
+int lzc_send(const char *, const char *, int);
+int lzc_receive(const char *, nvlist_t *, const char *, boolean_t, int);
+int lzc_send_space(const char *, const char *, uint64_t *);
 
-boolean_t lzc_exists(const char *dataset);
+boolean_t lzc_exists(const char *);
 
-int lzc_rollback(const char *fsname, char *snapnamebuf, int snapnamelen);
+int lzc_rollback(const char *, char *, int);
 
 #ifdef	__cplusplus
 }
diff --git a/cddl/contrib/opensolaris/lib/pyzfs/common/allow.py b/cddl/contrib/opensolaris/lib/pyzfs/common/allow.py
index fa8209f..7ad4b49 100644
--- a/cddl/contrib/opensolaris/lib/pyzfs/common/allow.py
+++ b/cddl/contrib/opensolaris/lib/pyzfs/common/allow.py
@@ -20,6 +20,7 @@
 # CDDL HEADER END
 #
 # Copyright (c) 2009, 2010, Oracle and/or its affiliates. All rights reserved.
+# Copyright (c) 2013 by Delphix. All rights reserved.
 #
 
 """This module implements the "zfs allow" and "zfs unallow" subcommands.
@@ -219,6 +220,7 @@ perms_subcmd = dict(
     hold=_("Allows adding a user hold to a snapshot"),
     release=_("Allows releasing a user hold which\n\t\t\t\tmight destroy the snapshot"),
     diff=_("Allows lookup of paths within a dataset,\n\t\t\t\tgiven an object number. Ordinary users need this\n\t\t\t\tin order to use zfs diff"),
+    bookmark="",
 )
 
 perms_other = dict(