Speed up 'zfs list -t snapshot -o name -s name'

FreeBSD #xxx:  Dramatically optimize listing snapshots when user
requests only snapshot names and wants to sort them by name, ie.
when executes:

  # zfs list -t snapshot -o name -s name

Because only name is needed we don't have to read all snapshot
properties.

Below you can find how long does it take to list 34509 snapshots
from a single disk pool before and after this change with cold and
warm cache:

    before:

        # time zfs list -t snapshot -o name -s name > /dev/null
        cold cache: 525s
        warm cache: 218s

    after:

        # time zfs list -t snapshot -o name -s name > /dev/null
        cold cache: 1.7s
        warm cache: 1.1s

NOTE: This patch only appears in FreeBSD.  If/when Illumos picks up
the change we may want to drop this patch and adopt their version.
However, for now this addresses a real issue.

Ported-by: Brian Behlendorf <[email protected]>
Issue #450

Author: pjd

cmd/zfs/zfs_iter.c

@@ -20,6 +20,7 @@
  */
 /*
  * Copyright (c) 2005, 2010, Oracle and/or its affiliates. All rights reserved.
+ * Copyright (c) 2012 Pawel Jakub Dawidek <[email protected]>.
  */
 
 #include <libintl.h>
@@ -129,8 +130,11 @@ 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)
-			(void) zfs_iter_snapshots(zhp, zfs_callback, data);
+		if ((zfs_get_type(zhp) != ZFS_TYPE_SNAPSHOT) && include_snaps) {
+			(void) zfs_iter_snapshots(zhp,
+			    (cb->cb_flags & ZFS_ITER_SIMPLE) != 0, zfs_callback,
+			    data);
+		}
 		cb->cb_depth--;
 	}
 
@@ -184,6 +188,13 @@ zfs_free_sort_columns(zfs_sort_column_t *sc)
 	}
 }
 
+int
+zfs_sort_only_by_name(const zfs_sort_column_t *sc)
+{
+	return (sc != NULL && sc->sc_next == NULL &&
+	    sc->sc_prop == ZFS_PROP_NAME);
+}
+
 /* ARGSUSED */
 static int
 zfs_compare(const void *larg, const void *rarg, void *unused)
@@ -224,7 +235,13 @@ zfs_compare(const void *larg, const void *rarg, void *unused)
 			lcreate = zfs_prop_get_int(l, ZFS_PROP_CREATETXG);
 			rcreate = zfs_prop_get_int(r, ZFS_PROP_CREATETXG);
 
-			if (lcreate < rcreate)
+			/*
+			 * Both lcreate and rcreate being 0 means we don't have
+			 * properties and we should compare full name.
+			 */
+			if (lcreate == 0 && rcreate == 0)
+				ret = strcmp(lat + 1, rat + 1);
+			else if (lcreate < rcreate)
 				ret = -1;
 			else if (lcreate > rcreate)
 				ret = 1;
@@ -290,7 +307,14 @@ zfs_sort(const void *larg, const void *rarg, void *data)
 			if (rvalid)
 				verify(nvlist_lookup_string(rval,
 				    ZPROP_VALUE, &rstr) == 0);
+		} else if (psc->sc_prop == ZFS_PROP_NAME) {
+			lvalid = rvalid = B_TRUE;
+
+			(void) strlcpy(lbuf, zfs_get_name(l), sizeof(lbuf));
+			(void) strlcpy(rbuf, zfs_get_name(r), sizeof(rbuf));
 
+			lstr = lbuf;
+			rstr = rbuf;
 		} else if (zfs_prop_is_string(psc->sc_prop)) {
 			lvalid = (zfs_prop_get(l, psc->sc_prop, lbuf,
 			    sizeof (lbuf), NULL, NULL, 0, B_TRUE) == 0);

cmd/zfs/zfs_iter.h

@@ -43,11 +43,13 @@ typedef struct zfs_sort_column {
 #define	ZFS_ITER_PROP_LISTSNAPS    (1 << 2)
 #define	ZFS_ITER_DEPTH_LIMIT	   (1 << 3)
 #define	ZFS_ITER_RECVD_PROPS	   (1 << 4)
+#define	ZFS_ITER_SIMPLE		   (1 << 5)
 
 int zfs_for_each(int, char **, int options, zfs_type_t,
     zfs_sort_column_t *, zprop_list_t **, int, zfs_iter_f, void *);
 int zfs_add_sort_column(zfs_sort_column_t **, const char *, boolean_t);
 void zfs_free_sort_columns(zfs_sort_column_t *);
+int zfs_sort_only_by_name(const zfs_sort_column_t *);
 
 #ifdef	__cplusplus
 }

cmd/zfs/zfs_main.c

@@ -2650,7 +2650,12 @@ print_dataset(zfs_handle_t *zhp, zprop_list_t *pl, boolean_t scripted)
 			first = B_FALSE;
 		}
 
-		if (pl->pl_prop != ZPROP_INVAL) {
+		if (pl->pl_prop == ZFS_PROP_NAME) {
+			(void) strlcpy(property, zfs_get_name(zhp),
+			    sizeof(property));
+			propstr = property;
+			right_justify = zfs_prop_align_right(pl->pl_prop);
+		} else if (pl->pl_prop != ZPROP_INVAL) {
 			if (zfs_prop_get(zhp, pl->pl_prop, property,
 			    sizeof (property), NULL, NULL, 0, B_FALSE) != 0)
 				propstr = "-";
@@ -2810,6 +2815,13 @@ zfs_do_list(int argc, char **argv)
 	if (fields == NULL)
 		fields = default_fields;
 
+	/*
+	 * If we are only going to list snapshot names and sort by name,
+	 * then we can use faster version.
+	 */
+	if (strcmp(fields, "name") == 0 && zfs_sort_only_by_name(sortcol))
+		flags |= ZFS_ITER_SIMPLE;
+
 	/*
 	 * If "-o space" and no types were specified, don't display snapshots.
 	 */

include/libzfs.h

@@ -516,7 +516,7 @@ extern int zfs_iter_root(libzfs_handle_t *, zfs_iter_f, void *);
 extern int zfs_iter_children(zfs_handle_t *, zfs_iter_f, void *);
 extern int zfs_iter_dependents(zfs_handle_t *, boolean_t, zfs_iter_f, void *);
 extern int zfs_iter_filesystems(zfs_handle_t *, zfs_iter_f, void *);
-extern int zfs_iter_snapshots(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 *);
 
 typedef struct get_all_cb {

include/sys/zfs_ioctl.h

@@ -286,7 +286,8 @@ typedef struct zfs_cmd {
 	boolean_t	zc_temphold;
 	uint64_t	zc_action_handle;
 	int		zc_cleanup_fd;
-	uint8_t		zc_pad[4];		/* alignment */
+	uint8_t		zc_simple;
+	uint8_t		zc_pad[3];		/* alignment */
 	uint64_t	zc_sendobj;
 	uint64_t	zc_fromobj;
 	uint64_t	zc_createtxg;

lib/libzfs/libzfs_dataset.c

@@ -23,6 +23,7 @@
  * Copyright (c) 2005, 2010, Oracle and/or its affiliates. All rights reserved.
  * Copyright (c) 2010 Nexenta Systems, Inc. All rights reserved.
  * Copyright (c) 2011 by Delphix. All rights reserved.
+ * Copyright (c) 2012 Pawel Jakub Dawidek <[email protected]>.
  */
 
 #include <ctype.h>
@@ -480,6 +481,23 @@ make_dataset_handle_zc(libzfs_handle_t *hdl, zfs_cmd_t *zc)
 	return (zhp);
 }
 
+static zfs_handle_t *
+make_dataset_simple_handle_zc(zfs_handle_t *pzhp, zfs_cmd_t *zc)
+{
+	zfs_handle_t *zhp = calloc(sizeof (zfs_handle_t), 1);
+
+	if (zhp == NULL)
+		return (NULL);
+
+	zhp->zfs_hdl = pzhp->zfs_hdl;
+	(void) strlcpy(zhp->zfs_name, zc->zc_name, sizeof (zhp->zfs_name));
+	zhp->zfs_head_type = pzhp->zfs_type;
+	zhp->zfs_type = ZFS_TYPE_SNAPSHOT;
+	zhp->zpool_hdl = zpool_handle(zhp);
+
+	return (zhp);
+}
+
 /*
  * Opens the given snapshot, filesystem, or volume.   The 'types'
  * argument is a mask of acceptable types.  The function will print an
@@ -2518,7 +2536,8 @@ zfs_iter_filesystems(zfs_handle_t *zhp, zfs_iter_f func, void *data)
  * Iterate over all snapshots
  */
 int
-zfs_iter_snapshots(zfs_handle_t *zhp, zfs_iter_f func, void *data)
+zfs_iter_snapshots(zfs_handle_t *zhp, boolean_t simple, zfs_iter_f func,
+    void *data)
 {
 	zfs_cmd_t zc = { "\0", "\0", "\0", "\0", 0 };
 	zfs_handle_t *nzhp;
@@ -2527,15 +2546,19 @@ zfs_iter_snapshots(zfs_handle_t *zhp, zfs_iter_f func, void *data)
 	if (zhp->zfs_type == ZFS_TYPE_SNAPSHOT)
 		return (0);
 
+	zc.zc_simple = simple;
+
 	if (zcmd_alloc_dst_nvlist(zhp->zfs_hdl, &zc, 0) != 0)
 		return (-1);
 	while ((ret = zfs_do_list_ioctl(zhp, ZFS_IOC_SNAPSHOT_LIST_NEXT,
 	    &zc)) == 0) {
 
-		if ((nzhp = make_dataset_handle_zc(zhp->zfs_hdl,
-		    &zc)) == NULL) {
+		if (simple)
+			nzhp = make_dataset_simple_handle_zc(zhp, &zc);
+		else
+			nzhp = make_dataset_handle_zc(zhp->zfs_hdl, &zc);
+		if (nzhp == NULL)
 			continue;
-		}
 
 		if ((ret = func(nzhp, data)) != 0) {
 			zcmd_free_nvlists(&zc);
@@ -2557,7 +2580,7 @@ zfs_iter_children(zfs_handle_t *zhp, zfs_iter_f func, void *data)
 	if ((ret = zfs_iter_filesystems(zhp, func, data)) != 0)
 		return (ret);
 
-	return (zfs_iter_snapshots(zhp, func, data));
+	return (zfs_iter_snapshots(zhp, B_FALSE, func, data));
 }
 
 /*
@@ -3287,7 +3310,7 @@ zfs_promote(zfs_handle_t *zhp)
 		return (-1);
 	(void) zfs_prop_get(pzhp, ZFS_PROP_MOUNTPOINT, pd.cb_mountpoint,
 	    sizeof (pd.cb_mountpoint), NULL, NULL, 0, FALSE);
-	ret = zfs_iter_snapshots(pzhp, promote_snap_cb, &pd);
+	ret = zfs_iter_snapshots(pzhp, B_FALSE, promote_snap_cb, &pd);
 	if (ret != 0) {
 		zfs_close(pzhp);
 		return (-1);
@@ -3302,7 +3325,8 @@ zfs_promote(zfs_handle_t *zhp)
 	if (ret != 0) {
 		int save_errno = errno;
 
-		(void) zfs_iter_snapshots(pzhp, promote_snap_done_cb, &pd);
+		(void) zfs_iter_snapshots(pzhp, B_FALSE, promote_snap_done_cb,
+		    &pd);
 		zfs_close(pzhp);
 
 		switch (save_errno) {
@@ -3321,7 +3345,8 @@ zfs_promote(zfs_handle_t *zhp)
 			return (zfs_standard_error(hdl, save_errno, errbuf));
 		}
 	} else {
-		(void) zfs_iter_snapshots(zhp, promote_snap_done_cb, &pd);
+		(void) zfs_iter_snapshots(zhp, B_FALSE, promote_snap_done_cb,
+		    &pd);
 	}
 
 	zfs_close(pzhp);

lib/libzfs/libzfs_sendrecv.c

@@ -21,6 +21,8 @@
 
 /*
  * Copyright (c) 2005, 2010, Oracle and/or its affiliates. All rights reserved.
+ * Copyright (c) 2012 Pawel Jakub Dawidek <[email protected]>.
+ * All rights reserved
  */
 
 #include <assert.h>
@@ -717,7 +719,7 @@ send_iterate_fs(zfs_handle_t *zhp, void *arg)
 	sd->parent_fromsnap_guid = 0;
 	VERIFY(0 == nvlist_alloc(&sd->parent_snaps, NV_UNIQUE_NAME, 0));
 	VERIFY(0 == nvlist_alloc(&sd->snapprops, NV_UNIQUE_NAME, 0));
-	(void) zfs_iter_snapshots(zhp, send_iterate_snap, sd);
+	(void) zfs_iter_snapshots(zhp, B_FALSE, send_iterate_snap, sd);
 	VERIFY(0 == nvlist_add_nvlist(nvfs, "snaps", sd->parent_snaps));
 	VERIFY(0 == nvlist_add_nvlist(nvfs, "snapprops", sd->snapprops));
 	nvlist_free(sd->parent_snaps);
@@ -843,7 +845,7 @@ zfs_iter_snapshots_sorted(zfs_handle_t *zhp, zfs_iter_f callback, void *data)
 	avl_create(&avl, zfs_snapshot_compare,
 	    sizeof (zfs_node_t), offsetof(zfs_node_t, zn_avlnode));
 
-	ret = zfs_iter_snapshots(zhp, zfs_sort_snaps, &avl);
+	ret = zfs_iter_snapshots(zhp, B_FALSE, zfs_sort_snaps, &avl);
 
 	for (node = avl_first(&avl); node != NULL; node = AVL_NEXT(&avl, node))
 		ret |= callback(node->zn_handle, data);

module/zfs/zfs_ioctl.c

@@ -21,6 +21,7 @@
 /*
  * Copyright (c) 2005, 2010, Oracle and/or its affiliates. All rights reserved.
  * Portions Copyright 2011 Martin Matuska
+ * Portions Copyright 2012 Pawel Jakub Dawidek <[email protected]>
  * Copyright (c) 2012, Joyent, Inc. All rights reserved.
  */
 
@@ -1973,7 +1974,7 @@ zfs_ioc_snapshot_list_next(zfs_cmd_t *zc)
 	int error;
 
 top:
-	if (zc->zc_cookie == 0)
+	if (zc->zc_cookie == 0 && !zc->zc_simple)
 		(void) dmu_objset_find(zc->zc_name, dmu_objset_prefetch,
 		    NULL, DS_FIND_SNAPSHOTS);
 
@@ -1995,7 +1996,7 @@ zfs_ioc_snapshot_list_next(zfs_cmd_t *zc)
 	    zc->zc_name + strlen(zc->zc_name), &zc->zc_obj, &zc->zc_cookie,
 	    NULL);
 
-	if (error == 0) {
+	if (error == 0 && !zc->zc_simple) {
 		dsl_dataset_t *ds;
 		dsl_pool_t *dp = os->os_dsl_dataset->ds_dir->dd_pool;