Update ZFS from version 6 to 13 and bring some FreeBSD-specific changes.

This bring huge amount of changes, I'll enumerate only user-visible changes:

- Delegated Administration

	Allows regular users to perform ZFS operations, like file system
	creation, snapshot creation, etc.

- L2ARC

	Level 2 cache for ZFS - allows to use additional disks for cache.
	Huge performance improvements mostly for random read of mostly
	static content.

- slog

	Allow to use additional disks for ZFS Intent Log to speed up
	operations like fsync(2).

- vfs.zfs.super_owner

	Allows regular users to perform privileged operations on files stored
	on ZFS file systems owned by him. Very careful with this one.

- chflags(2)

	Not all the flags are supported. This still needs work.

- ZFSBoot

	Support to boot off of ZFS pool. Not finished, AFAIK.

	Submitted by:	dfr

- Snapshot properties

- New failure modes

	Before if write requested failed, system paniced. Now one
	can select from one of three failure modes:
	- panic - panic on write error
	- wait - wait for disk to reappear
	- continue - serve read requests if possible, block write requests

- Refquota, refreservation properties

	Just quota and reservation properties, but don't count space consumed
	by children file systems, clones and snapshots.

- Sparse volumes

	ZVOLs that don't reserve space in the pool.

- External attributes

	Compatible with extattr(2).

- NFSv4-ACLs

	Not sure about the status, might not be complete yet.

	Submitted by:	trasz

- Creation-time properties

- Regression tests for zpool(8) command.

Obtained from:	OpenSolaris

1 files changed, 154 insertions, 78 deletions

diff --git a/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/spa_config.c b/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/spa_config.c
index 9e8bcf3..1ffdb10 100644
--- a/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/spa_config.c
+++ b/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/spa_config.c
@@ -20,12 +20,10 @@
  */
 
 /*
- * Copyright 2007 Sun Microsystems, Inc.  All rights reserved.
+ * Copyright 2008 Sun Microsystems, Inc.  All rights reserved.
  * Use is subject to license terms.
  */
 
-#pragma ident	"%Z%%M%	%I%	%E% SMI"
-
 #include <sys/zfs_context.h>
 #include <sys/spa.h>
 #include <sys/spa_impl.h>
@@ -43,16 +41,18 @@
 /*
  * Pool configuration repository.
  *
- * The configuration for all pools, in addition to being stored on disk, is
- * stored in /etc/zfs/zpool.cache as a packed nvlist.  The kernel maintains
- * this list as pools are created, destroyed, or modified.
+ * Pool configuration is stored as a packed nvlist on the filesystem.  By
+ * default, all pools are stored in /etc/zfs/zpool.cache and loaded on boot
+ * (when the ZFS module is loaded).  Pools can also have the 'cachefile'
+ * property set that allows them to be stored in an alternate location until
+ * the control of external software.
  *
- * We have a single nvlist which holds all the configuration information.  When
- * the module loads, we read this information from the cache and populate the
- * SPA namespace.  This namespace is maintained independently in spa.c.
- * Whenever the namespace is modified, or the configuration of a pool is
- * changed, we call spa_config_sync(), which walks through all the active pools
- * and writes the configuration to disk.
+ * For each cache file, we have a single nvlist which holds all the
+ * configuration information.  When the module loads, we read this information
+ * from /etc/zfs/zpool.cache and populate the SPA namespace.  This namespace is
+ * maintained independently in spa.c.  Whenever the namespace is modified, or
+ * the configuration of a pool is changed, we call spa_config_sync(), which
+ * walks through all the active pools and writes the configuration to disk.
  */
 
 static uint64_t spa_config_generation = 1;
@@ -61,7 +61,7 @@ static uint64_t spa_config_generation = 1;
  * This can be overridden in userland to preserve an alternate namespace for
  * userland pools when doing testing.
  */
-const char *spa_config_dir = ZPOOL_CACHE_DIR;
+const char *spa_config_path = ZPOOL_CACHE;
 
 /*
  * Called when the module is first loaded, this routine loads the configuration
@@ -75,17 +75,21 @@ spa_config_load(void)
 	nvlist_t *nvlist, *child;
 	nvpair_t *nvpair;
 	spa_t *spa;
-	char pathname[128];
+	char *pathname;
 	struct _buf *file;
 	uint64_t fsize;
 
 	/*
 	 * Open the configuration file.
 	 */
-	(void) snprintf(pathname, sizeof (pathname), "%s/%s",
-	    spa_config_dir, ZPOOL_CACHE_FILE);
+	pathname = kmem_alloc(MAXPATHLEN, KM_SLEEP);
+
+	(void) snprintf(pathname, MAXPATHLEN, "%s", spa_config_path);
 
 	file = kobj_open_file(pathname);
+
+	kmem_free(pathname, MAXPATHLEN);
+
 	if (file == (struct _buf *)-1) {
 		ZFS_LOG(1, "Cannot open %s.", pathname);
 		return;
@@ -148,47 +152,32 @@ out:
 	kobj_close_file(file);
 }
 
-/*
- * Synchronize all pools to disk.  This must be called with the namespace lock
- * held.
- */
-void
-spa_config_sync(void)
+static void
+spa_config_write(spa_config_dirent_t *dp, nvlist_t *nvl)
 {
-	spa_t *spa = NULL;
-	nvlist_t *config;
 	size_t buflen;
 	char *buf;
 	vnode_t *vp;
 	int oflags = FWRITE | FTRUNC | FCREAT | FOFFMAX;
-	char pathname[128];
-	char pathname2[128];
-
-	ASSERT(MUTEX_HELD(&spa_namespace_lock));
-
-	VERIFY(nvlist_alloc(&config, NV_UNIQUE_NAME, KM_SLEEP) == 0);
+	char *temp;
 
 	/*
-	 * Add all known pools to the configuration list, ignoring those with
-	 * alternate root paths.
+	 * If the nvlist is empty (NULL), then remove the old cachefile.
 	 */
-	spa = NULL;
-	while ((spa = spa_next(spa)) != NULL) {
-		mutex_enter(&spa->spa_config_cache_lock);
-		if (spa->spa_config && spa->spa_name && spa->spa_root == NULL)
-			VERIFY(nvlist_add_nvlist(config, spa->spa_name,
-			    spa->spa_config) == 0);
-		mutex_exit(&spa->spa_config_cache_lock);
+	if (nvl == NULL) {
+		(void) vn_remove(dp->scd_path, UIO_SYSSPACE, RMFILE);
+		return;
 	}
 
 	/*
 	 * Pack the configuration into a buffer.
 	 */
-	VERIFY(nvlist_size(config, &buflen, NV_ENCODE_XDR) == 0);
+	VERIFY(nvlist_size(nvl, &buflen, NV_ENCODE_XDR) == 0);
 
 	buf = kmem_alloc(buflen, KM_SLEEP);
+	temp = kmem_zalloc(MAXPATHLEN, KM_SLEEP);
 
-	VERIFY(nvlist_pack(config, &buf, &buflen, NV_ENCODE_XDR,
+	VERIFY(nvlist_pack(nvl, &buf, &buflen, NV_ENCODE_XDR,
 	    KM_SLEEP) == 0);
 
 	/*
@@ -196,29 +185,92 @@ spa_config_sync(void)
 	 * 'write to temporary file, sync, move over original' to make sure we
 	 * always have a consistent view of the data.
 	 */
-	(void) snprintf(pathname, sizeof (pathname), "%s/%s", spa_config_dir,
-	    ZPOOL_CACHE_TMP);
+	(void) snprintf(temp, MAXPATHLEN, "%s.tmp", dp->scd_path);
 
-	if (vn_open(pathname, UIO_SYSSPACE, oflags, 0644, &vp, CRCREAT, 0) != 0)
-		goto out;
+	if (vn_open(temp, UIO_SYSSPACE, oflags, 0644, &vp, CRCREAT, 0) == 0) {
+		if (vn_rdwr(UIO_WRITE, vp, buf, buflen, 0, UIO_SYSSPACE,
+		    0, RLIM64_INFINITY, kcred, NULL) == 0 &&
+		    VOP_FSYNC(vp, FSYNC, kcred, NULL) == 0) {
+			(void) vn_rename(temp, dp->scd_path, UIO_SYSSPACE);
+		}
+		(void) VOP_CLOSE(vp, oflags, 1, 0, kcred, NULL);
+		VN_RELE(vp);
+	}
+
+	(void) vn_remove(temp, UIO_SYSSPACE, RMFILE);
+
+	kmem_free(buf, buflen);
+	kmem_free(temp, MAXPATHLEN);
+}
+
+/*
+ * Synchronize pool configuration to disk.  This must be called with the
+ * namespace lock held.
+ */
+void
+spa_config_sync(spa_t *target, boolean_t removing, boolean_t postsysevent)
+{
+	spa_config_dirent_t *dp, *tdp;
+	nvlist_t *nvl;
+
+	ASSERT(MUTEX_HELD(&spa_namespace_lock));
+
+	/*
+	 * Iterate over all cachefiles for the pool, past or present.  When the
+	 * cachefile is changed, the new one is pushed onto this list, allowing
+	 * us to update previous cachefiles that no longer contain this pool.
+	 */
+	for (dp = list_head(&target->spa_config_list); dp != NULL;
+	    dp = list_next(&target->spa_config_list, dp)) {
+		spa_t *spa = NULL;
+		if (dp->scd_path == NULL)
+			continue;
 
-	if (vn_rdwr(UIO_WRITE, vp, buf, buflen, 0, UIO_SYSSPACE,
-	    0, RLIM64_INFINITY, kcred, NULL) == 0 &&
-	    VOP_FSYNC(vp, FSYNC, kcred) == 0) {
-		(void) snprintf(pathname2, sizeof (pathname2), "%s/%s",
-		    spa_config_dir, ZPOOL_CACHE_FILE);
-		(void) vn_rename(pathname, pathname2, UIO_SYSSPACE);
+		/*
+		 * Iterate over all pools, adding any matching pools to 'nvl'.
+		 */
+		nvl = NULL;
+		while ((spa = spa_next(spa)) != NULL) {
+			if (spa == target && removing)
+				continue;
+
+			mutex_enter(&spa->spa_props_lock);
+			tdp = list_head(&spa->spa_config_list);
+			if (spa->spa_config == NULL ||
+			    tdp->scd_path == NULL ||
+			    strcmp(tdp->scd_path, dp->scd_path) != 0) {
+				mutex_exit(&spa->spa_props_lock);
+				continue;
+			}
+
+			if (nvl == NULL)
+				VERIFY(nvlist_alloc(&nvl, NV_UNIQUE_NAME,
+				    KM_SLEEP) == 0);
+
+			VERIFY(nvlist_add_nvlist(nvl, spa->spa_name,
+			    spa->spa_config) == 0);
+			mutex_exit(&spa->spa_props_lock);
+		}
+
+		spa_config_write(dp, nvl);
+		nvlist_free(nvl);
 	}
 
-	(void) VOP_CLOSE(vp, oflags, 1, 0, kcred);
-	VN_RELE(vp);
+	/*
+	 * Remove any config entries older than the current one.
+	 */
+	dp = list_head(&target->spa_config_list);
+	while ((tdp = list_next(&target->spa_config_list, dp)) != NULL) {
+		list_remove(&target->spa_config_list, tdp);
+		if (tdp->scd_path != NULL)
+			spa_strfree(tdp->scd_path);
+		kmem_free(tdp, sizeof (spa_config_dirent_t));
+	}
 
-out:
-	(void) vn_remove(pathname, UIO_SYSSPACE, RMFILE);
 	spa_config_generation++;
 
-	kmem_free(buf, buflen);
-	nvlist_free(config);
+	if (postsysevent)
+		spa_event_notify(target, NULL, ESC_ZFS_CONFIG_SYNC);
 }
 
 /*
@@ -231,27 +283,25 @@ nvlist_t *
 spa_all_configs(uint64_t *generation)
 {
 	nvlist_t *pools;
-	spa_t *spa;
+	spa_t *spa = NULL;
 
 	if (*generation == spa_config_generation)
 		return (NULL);
 
 	VERIFY(nvlist_alloc(&pools, NV_UNIQUE_NAME, KM_SLEEP) == 0);
 
-	spa = NULL;
 	mutex_enter(&spa_namespace_lock);
 	while ((spa = spa_next(spa)) != NULL) {
-		if (INGLOBALZONE(curproc) ||
+		if (INGLOBALZONE(curthread) ||
 		    zone_dataset_visible(spa_name(spa), NULL)) {
-			mutex_enter(&spa->spa_config_cache_lock);
+			mutex_enter(&spa->spa_props_lock);
 			VERIFY(nvlist_add_nvlist(pools, spa_name(spa),
 			    spa->spa_config) == 0);
-			mutex_exit(&spa->spa_config_cache_lock);
+			mutex_exit(&spa->spa_props_lock);
 		}
 	}
-	mutex_exit(&spa_namespace_lock);
-
 	*generation = spa_config_generation;
+	mutex_exit(&spa_namespace_lock);
 
 	return (pools);
 }
@@ -259,11 +309,11 @@ spa_all_configs(uint64_t *generation)
 void
 spa_config_set(spa_t *spa, nvlist_t *config)
 {
-	mutex_enter(&spa->spa_config_cache_lock);
+	mutex_enter(&spa->spa_props_lock);
 	if (spa->spa_config != NULL)
 		nvlist_free(spa->spa_config);
 	spa->spa_config = config;
-	mutex_exit(&spa->spa_config_cache_lock);
+	mutex_exit(&spa->spa_props_lock);
 }
 
 /*
@@ -277,11 +327,16 @@ spa_config_generate(spa_t *spa, vdev_t *vd, uint64_t txg, int getstats)
 	nvlist_t *config, *nvroot;
 	vdev_t *rvd = spa->spa_root_vdev;
 	unsigned long hostid = 0;
+	boolean_t locked = B_FALSE;
 
-	ASSERT(spa_config_held(spa, RW_READER));
-
-	if (vd == NULL)
+	if (vd == NULL) {
 		vd = rvd;
+		locked = B_TRUE;
+		spa_config_enter(spa, SCL_CONFIG | SCL_STATE, FTAG, RW_READER);
+	}
+
+	ASSERT(spa_config_held(spa, SCL_CONFIG | SCL_STATE, RW_READER) ==
+	    (SCL_CONFIG | SCL_STATE));
 
 	/*
 	 * If txg is -1, report the current value of spa->spa_config_txg.
@@ -302,8 +357,10 @@ spa_config_generate(spa_t *spa, vdev_t *vd, uint64_t txg, int getstats)
 	VERIFY(nvlist_add_uint64(config, ZPOOL_CONFIG_POOL_GUID,
 	    spa_guid(spa)) == 0);
 	(void) ddi_strtoul(hw_serial, NULL, 10, &hostid);
-	VERIFY(nvlist_add_uint64(config, ZPOOL_CONFIG_HOSTID,
-	    hostid) == 0);
+	if (hostid != 0) {
+		VERIFY(nvlist_add_uint64(config, ZPOOL_CONFIG_HOSTID,
+		    hostid) == 0);
+	}
 	VERIFY(nvlist_add_string(config, ZPOOL_CONFIG_HOSTNAME,
 	    utsname.nodename) == 0);
 
@@ -315,30 +372,48 @@ spa_config_generate(spa_t *spa, vdev_t *vd, uint64_t txg, int getstats)
 		if (vd->vdev_isspare)
 			VERIFY(nvlist_add_uint64(config, ZPOOL_CONFIG_IS_SPARE,
 			    1ULL) == 0);
+		if (vd->vdev_islog)
+			VERIFY(nvlist_add_uint64(config, ZPOOL_CONFIG_IS_LOG,
+			    1ULL) == 0);
 		vd = vd->vdev_top;		/* label contains top config */
 	}
 
-	nvroot = vdev_config_generate(spa, vd, getstats, B_FALSE);
+	nvroot = vdev_config_generate(spa, vd, getstats, B_FALSE, B_FALSE);
 	VERIFY(nvlist_add_nvlist(config, ZPOOL_CONFIG_VDEV_TREE, nvroot) == 0);
 	nvlist_free(nvroot);
 
+	if (locked)
+		spa_config_exit(spa, SCL_CONFIG | SCL_STATE, FTAG);
+
 	return (config);
 }
 
 /*
- * Update all disk labels, generate a fresh config based on the current
- * in-core state, and sync the global config cache.
+ * For a pool that's not currently a booting rootpool, update all disk labels,
+ * generate a fresh config based on the current in-core state, and sync the
+ * global config cache.
  */
 void
 spa_config_update(spa_t *spa, int what)
 {
+	spa_config_update_common(spa, what, FALSE);
+}
+
+/*
+ * Update all disk labels, generate a fresh config based on the current
+ * in-core state, and sync the global config cache (do not sync the config
+ * cache if this is a booting rootpool).
+ */
+void
+spa_config_update_common(spa_t *spa, int what, boolean_t isroot)
+{
 	vdev_t *rvd = spa->spa_root_vdev;
 	uint64_t txg;
 	int c;
 
 	ASSERT(MUTEX_HELD(&spa_namespace_lock));
 
-	spa_config_enter(spa, RW_WRITER, FTAG);
+	spa_config_enter(spa, SCL_ALL, FTAG, RW_WRITER);
 	txg = spa_last_synced_txg(spa) + 1;
 	if (what == SPA_CONFIG_UPDATE_POOL) {
 		vdev_config_dirty(rvd);
@@ -358,7 +433,7 @@ spa_config_update(spa_t *spa, int what)
 			}
 		}
 	}
-	spa_config_exit(spa, FTAG);
+	spa_config_exit(spa, SCL_ALL, FTAG);
 
 	/*
 	 * Wait for the mosconfig to be regenerated and synced.
@@ -368,8 +443,9 @@ spa_config_update(spa_t *spa, int what)
 	/*
 	 * Update the global config cache to reflect the new mosconfig.
 	 */
-	spa_config_sync();
+	if (!isroot)
+		spa_config_sync(spa, B_FALSE, what != SPA_CONFIG_UPDATE_POOL);
 
 	if (what == SPA_CONFIG_UPDATE_POOL)
-		spa_config_update(spa, SPA_CONFIG_UPDATE_VDEVS);
+		spa_config_update_common(spa, SPA_CONFIG_UPDATE_VDEVS, isroot);
 }