Do not iterate through filesystems unnecessarily

Currently, when attempting to list snapshots ZFS may do a lot of
extra work checking child datasets. This is because the code does
not realize that it will not be able to reach any snapshots
contained within snapshots that are at the depth limit since the
snapshots of those datasets are counted as an additional layer
deeper. This patch corrects this issue.

In addition, this patch adds the ability to do perform the commands:

$ zfs list -t snapshot <dataset>
$ zfs get -t snapshot <prop> <dataset>

as a convenient way to list out properties of all snapshots of a
given dataset without having to use the depth limit.

Reviewed-by: Alek Pinchuk <apinchuk@datto.com>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Richard Laager <rlaager@wiktel.com>
Reviewed-by: Matt Ahrens <mahrens@delphix.com>
Reviewed-by: George Melikov <mail@gmelikov.ru>
Signed-off-by: Tom Caputi <tcaputi@datto.com>
Closes openzfs#8539

cmd/zfs/zfs_iter.c

@@ -134,16 +134,31 @@ zfs_callback(zfs_handle_t *zhp, void *data)
 	    ((cb->cb_flags & ZFS_ITER_DEPTH_LIMIT) == 0 ||
 	    cb->cb_depth < cb->cb_depth_limit)) {
 		cb->cb_depth++;
-		if (zfs_get_type(zhp) == ZFS_TYPE_FILESYSTEM)
+
+		/*
+		 * If we are not looking for filesystems, we don't need to
+		 * recurse into filesystems when we are at our depth limit.
+		 */
+		if ((cb->cb_depth < cb->cb_depth_limit ||
+		    (cb->cb_flags & ZFS_ITER_DEPTH_LIMIT) == 0 ||
+		    (cb->cb_types &
+		    (ZFS_TYPE_FILESYSTEM | ZFS_TYPE_VOLUME))) &&
+		    zfs_get_type(zhp) == ZFS_TYPE_FILESYSTEM) {
 			(void) zfs_iter_filesystems(zhp, zfs_callback, data);
+		}
+
 		if (((zfs_get_type(zhp) & (ZFS_TYPE_SNAPSHOT |
-		    ZFS_TYPE_BOOKMARK)) == 0) && include_snaps)
+		    ZFS_TYPE_BOOKMARK)) == 0) && include_snaps) {
 			(void) zfs_iter_snapshots(zhp,
-			    (cb->cb_flags & ZFS_ITER_SIMPLE) != 0, zfs_callback,
-			    data, 0, 0);
+			    (cb->cb_flags & ZFS_ITER_SIMPLE) != 0,
+			    zfs_callback, data, 0, 0);
+		}
+
 		if (((zfs_get_type(zhp) & (ZFS_TYPE_SNAPSHOT |
-		    ZFS_TYPE_BOOKMARK)) == 0) && include_bmarks)
+		    ZFS_TYPE_BOOKMARK)) == 0) && include_bmarks) {
 			(void) zfs_iter_bookmarks(zhp, zfs_callback, data);
+		}
+
 		cb->cb_depth--;
 	}
 

cmd/zfs/zfs_main.c

@@ -1922,6 +1922,16 @@ zfs_do_get(int argc, char **argv)
 
 	fields = argv[0];
 
+	/*
+	 * Handle users who want to get all snapshots of the current
+	 * dataset (ex. 'zfs get -t snapshot refer <dataset>').
+	 */
+	if (types == ZFS_TYPE_SNAPSHOT &&
+	    (flags & ZFS_ITER_RECURSE) == 0 && limit == 0) {
+		flags |= (ZFS_ITER_DEPTH_LIMIT | ZFS_ITER_RECURSE);
+		limit = 1;
+	}
+
 	if (zprop_get_list(g_zfs, fields, &cb.cb_proplist, ZFS_TYPE_DATASET)
 	    != 0)
 		usage(B_FALSE);
@@ -3416,6 +3426,16 @@ zfs_do_list(int argc, char **argv)
 	if (strcmp(fields, "space") == 0 && types_specified == B_FALSE)
 		types &= ~ZFS_TYPE_SNAPSHOT;
 
+	/*
+	 * Handle users who want to list all snapshots of the current
+	 * dataset (ex. 'zfs list -t snapshot <dataset>').
+	 */
+	if (types == ZFS_TYPE_SNAPSHOT &&
+	    (flags & ZFS_ITER_RECURSE) == 0 && limit == 0) {
+		flags |= (ZFS_ITER_DEPTH_LIMIT | ZFS_ITER_RECURSE);
+		limit = 1;
+	}
+
 	/*
 	 * If the user specifies '-o all', the zprop_get_list() doesn't
 	 * normally include the name of the dataset.  For 'zfs list', we always

tests/zfs-tests/include/libtest.shlib

@@ -2594,7 +2594,6 @@ function verify_opt_p_ops
 					"when ops is $ops."
 			fi
 			log_must datasetexists $dataset
-			log_mustnot snapexists $dataset
 			;;
 		*)
 			log_fail "$ops is not supported."

tests/zfs-tests/tests/functional/cli_root/zfs_get/zfs_get_009_pos.ksh

@@ -87,5 +87,10 @@ for dp in ${depth_array[@]}; do
 	(( old_val=dp ))
 done
 
+# Ensure 'zfs get -t snapshot <dataset>' works as though -d 1 was specified
+log_must eval "zfs get -H -t snapshot -o name creation $DEPTH_FS > $DEPTH_OUTPUT"
+log_must eval "zfs get -H -t snapshot -d 1 -o name creation $DEPTH_FS > $EXPECT_OUTPUT"
+log_must diff $DEPTH_OUTPUT $EXPECT_OUTPUT
+
 log_pass "'zfs get -d <n>' should get expected output."