Merge ZFS version 15 and almost all OpenSolaris bugfixes referenced

in Solaris 10 updates 141445-09 and 142901-14.

Detailed information:
(OpenSolaris revisions and Bug IDs, Solaris 10 patch numbers)

7844:effed23820ae
6755435	zfs_open() and zfs_close() needs to use ZFS_ENTER/ZFS_VERIFY_ZP (141445-01)

7897:e520d8258820
6748436	inconsistent zpool.cache in boot_archive could panic a zfs root filesystem upon boot-up (141445-01)

7965:b795da521357
6740164	zpool attach can create an illegal root pool (141909-02)

8084:b811cc60d650
6769612	zpool_import() will continue to write to cachefile even if altroot is set (N/A)

8121:7fd09d4ebd9c
6757430	want an option for zdb to disable space map loading and leak tracking (141445-01)

8129:e4f45a0bfbb0
6542860	ASSERT: reason != VDEV_LABEL_REMOVE||vdev_inuse(vd, crtxg, reason, 0) (141445-01)

8188:fd00c0a81e80
6761100	want zdb option to select older uberblocks (141445-01)

8190:6eeea43ced42
6774886	zfs_setattr() won't allow ndmp to restore SUNWattr_rw (141445-01)

8225:59a9961c2aeb
6737463	panic while trying to write out config file if root pool import fails (141445-01)

8227:f7d7be9b1f56
6765294	Refactor replay (141445-01)

8228:51e9ca9ee3a5
6572357	libzfs should do more to avoid mnttab lookups (141909-01)
6572376	zfs_iter_filesystems and zfs_iter_snapshots get objset stats twice (141909-01)

8241:5a60f16123ba
6328632	zpool offline is a bit too conservative (141445-01)
6739487	ASSERT: txg <= spa_final_txg due to scrub/export race (141445-01)
6767129	ASSERT: cvd->vdev_isspare, in spa_vdev_detach() (141445-01)
6747698	checksum failures after offline -t / export / import / scrub (141445-01)
6745863	ZFS writes to disk after it has been offlined (141445-01)
6722540	50% slowdown on scrub/resilver with certain vdev configurations (141445-01)
6759999	resilver logic rewrites ditto blocks on both source and destination (141445-01)
6758107	I/O should never suspend during spa_load() (141445-01)
6776548	codereview(1) runs off the page when faced with multi-line comments (N/A)
6761406	AMD errata 91 workaround doesn't work on 64-bit systems (141445-01)

8242:e46e4b2f0a03
6770866	GRUB/ZFS should require physical path or devid, but not both (141445-01)

8269:03a7e9050cfd
6674216	"zfs share" doesn't work, but "zfs set sharenfs=on" does (141445-01)
6621164	$SRC/cmd/zfs/zfs_main.c seems to have a syntax error in the translation note (141445-01)
6635482	i18n problems in libzfs_dataset.c and zfs_main.c (141445-01)
6595194	"zfs get" VALUE column is as wide as NAME (141445-01)
6722991	vdev_disk.c: error checking for ddi_pathname_to_dev_t() must test for NODEV (141445-01)
6396518	ASSERT strings shouldn't be pre-processed (141445-01)

8274:846b39508aff
6713916	scrub/resilver needlessly decompress data (141445-01)

8343:655db2375fed
6739553	libzfs_status msgid table is out of sync (141445-01)
6784104	libzfs unfairly rejects numerical values greater than 2^63 (141445-01)
6784108	zfs_realloc() should not free original memory on failure (141445-01)

8525:e0e0e525d0f8
6788830	set large value to reservation cause core dump (141445-01)
6791064	want sysevents for ZFS scrub (141445-01)
6791066	need to be able to set cachefile on faulted pools (141445-01)
6791071	zpool_do_import() should not enable datasets on faulted pools (141445-01)
6792134	getting multiple properties on a faulted pool leads to confusion (141445-01)

8547:bcc7b46e5ff7
6792884	Vista clients cannot access .zfs (141445-01)

8632:36ef517870a3
6798384	It can take a village to raise a zio (141445-01)

8636:7e4ce9158df3
6551866	deadlock between zfs_write(), zfs_freesp(), and zfs_putapage() (141909-01)
6504953	zfs_getpage() misunderstands VOP_GETPAGE() interface (141909-01)
6702206	ZFS read/writer lock contention throttles sendfile() benchmark (141445-01)
6780491	Zone on a ZFS filesystem has poor fork/exec performance (141445-01)
6747596	assertion failed: DVA_EQUAL(BP_IDENTITY(&zio->io_bp_orig), BP_IDENTITY(zio->io_bp))); (141445-01)

8692:692d4668b40d
6801507	ZFS read aggregation should not mind the gap (141445-01)

8697:e62d2612c14d
6633095	creating a filesystem with many properties set is slow (141445-01)

8768:dfecfdbb27ed
6775697	oracle crashes when overwriting after hitting quota on zfs (141909-01)

8811:f8deccf701cf
6790687	libzfs mnttab caching ignores external changes (141445-01)
6791101	memory leak from libzfs_mnttab_init (141445-01)

8845:91af0d9c0790
6800942	smb_session_create() incorrectly stores IP addresses (N/A)
6582163	Access Control List (ACL) for shares (141445-01)
6804954	smb_search - shortname field should be space padded following the NULL terminator (N/A)
6800184	Panic at smb_oplock_conflict+0x35() (N/A)

8876:59d2e67b4b65
6803822	Reboot after replacement of system disk in a ZFS mirror drops to grub> prompt (141445-01)

8924:5af812f84759
6789318	coredump when issue zdb -uuuu poolname/ (141445-01)
6790345 zdb -dddd -e poolname coredump (141445-01)
6797109 zdb: 'zdb -dddddd pool_name/fs_name inode' coredump if the file with inode was deleted (141445-01)
6797118 zdb: 'zdb -dddddd poolname inum' coredump if I miss the fs name (141445-01)
6803343 shareiscsi=on failed, iscsitgtd failed request to share (141445-01)

9030:243fd360d81f
6815893	hang mounting a dataset after booting into a new boot environment (141445-01)

9056:826e1858a846
6809691	'zpool create -f' no longer overwrites ufs infomation (141445-01)

9179:d8fbd96b79b3
6790064	zfs needs to determine uid and gid earlier in create process (141445-01)

9214:8d350e5d04aa
6604992	forced unmount + being in .zfs/snapshot/<snap1> = not happy (141909-01)
6810367	assertion failed: dvp->v_flag & VROOT, file: ../../common/fs/gfs.c, line: 426 (141909-01)

9229:e3f8b41e5db4
6807765	ztest_dsl_dataset_promote_busy needs to clean up after ENOSPC (141445-01)

9230:e4561e3eb1ef
6821169	offlining a device results in checksum errors (141445-01)
6821170	ZFS should not increment error stats for unavailable devices (141445-01)
6824006	need to increase issue and interrupt taskqs threads in zfs (141445-01)

9234:bffdc4fc05c4
6792139	recovering from a suspended pool needs some work (141445-01)
6794830	reboot command hangs on a failed zfs pool (141445-01)

9246:67c03c93c071
6824062	System panicked in zfs_mount due to NULL pointer dereference when running btts and svvs tests (141909-01)

9276:a8a7fc849933
6816124	System crash running zpool destroy on broken zpool (141445-03)

9355:09928982c591
6818183	zfs snapshot -r is slow due to set_snap_props() doing txg_wait_synced() for each new snapshot (141445-03)

9391:413d0661ef33
6710376	log device can show incorrect status when other parts of pool are degraded (141445-03)

9396:f41cf682d0d3 (part already merged)
6501037	want user/group quotas on ZFS (141445-03)
6827260	assertion failed in arc_read(): hdr == pbuf->b_hdr (141445-03)
6815592	panic: No such hold X on refcount Y from zfs_znode_move (141445-03)
6759986	zfs list shows temporary %clone when doing online zfs recv (141445-03)

9404:319573cd93f8
6774713	zfs ignores canmount=noauto when sharenfs property != off (141445-03)

9412:4aefd8704ce0
6717022	ZFS DMU needs zero-copy support (141445-03)

9425:e7ffacaec3a8
6799895	spa_add_spares() needs to be protected by config lock (141445-03)
6826466	want to post sysevents on hot spare activation (141445-03)
6826468	spa 'allowfaulted' needs some work (141445-03)
6826469	kernel support for storing vdev FRU information (141445-03)
6826470	skip posting checksum errors from DTL regions of leaf vdevs (141445-03)
6826471	I/O errors after device remove probe can confuse FMA (141445-03)
6826472	spares should enjoy some of the benefits of cache devices (141445-03)

9443:2a96d8478e95
6833711	gang leaders shouldn't have to be logical (141445-03)

9463:d0bd231c7518
6764124	want zdb to be able to checksum metadata blocks only (141445-03)

9465:8372081b8019
6830237	zfs panic in zfs_groupmember() (141445-03)

9466:1fdfd1fed9c4
6833162	phantom log device in zpool status (141445-03)

9469:4f68f041ddcd
6824968	add ZFS userquota support to rquotad (141445-03)

9470:6d827468d7b5
6834217	godfather I/O should reexecute (141445-03)

9480:fcff33da767f
6596237	Stop looking and start ganging (141909-02)

9493:9933d599bc93
6623978	lwb->lwb_buf != NULL, file ../../../uts/common/fs/zfs/zil.c, line 787, function zil_lwb_commit (141445-06)

9512:64cafcbcc337
6801810	Commit of aligned streaming rewrites to ZIL device causes unwanted disk reads (N/A)

9515:d3b739d9d043
6586537	async zio taskqs can block out userland commands (142901-09)

9554:787363635b6a
6836768	zfs_userspace() callback has no way to indicate failure (N/A)

9574:1eb6a6ab2c57
6838062	zfs panics when an error is encountered in space_map_load() (141909-02)

9583:b0696cd037cc
6794136	Panic BAD TRAP: type=e when importing degraded zraid pool. (141909-03)

9630:e25a03f552e0
6776104	"zfs import" deadlock between spa_unload() and spa_async_thread() (141445-06)

9653:a70048a304d1
6664765	Unable to remove files when using fat-zap and quota exceeded on ZFS filesystem (141445-06)

9688:127be1845343
6841321	zfs userspace / zfs get userused@ doesn't work on mounted snapshot (N/A)
6843069	zfs get userused@S-1-... doesn't work (N/A)

9873:8ddc892eca6e
6847229	assertion failed: refcount_count(&tx->tx_space_written) + delta <= tx->tx_space_towrite in dmu_tx.c (141445-06)

9904:d260bd3fd47c
6838344	kernel heap corruption detected on zil while stress testing (141445-06)

9951:a4895b3dd543
6844900	zfs_ioc_userspace_upgrade leaks (N/A)

10040:38b25aeeaf7a
6857012	zfs panics on zpool import (141445-06)

10000:241a51d8720c
6848242	zdb -e no longer works as expected (N/A)

10100:4a6965f6bef8
6856634	snv_117 not booting: zfs_parse_bootfs: error2 (141445-07)

10160:a45b03783d44
6861983	zfs should use new name <-> SID interfaces (N/A)
6862984	userquota commands can hang (141445-06)

10299:80845694147f
6696858	zfs receive of incremental replication stream can dereference NULL pointer and crash (N/A)

10302:a9e3d1987706
6696858	zfs receive of incremental replication stream can dereference NULL pointer and crash (fix lint) (N/A)

10575:2a8816c5173b (partial merge)
6882227 spa_async_remove() shouldn't do a full clear (142901-14)

10800:469478b180d9
6880764	fsync on zfs is broken if writes are greater than 32kb on a hard crash and no log attached (142901-09)
6793430 zdb -ivvvv assertion failure: bp->blk_cksum.zc_word[2] == dmu_objset_id(zilog->zl_os) (N/A)

10801:e0bf032e8673 (partial merge)
6822816 assertion failed: zap_remove_int(ds_next_clones_obj) returns ENOENT (142901-09)

10810:b6b161a6ae4a
6892298 buf->b_hdr->b_state != arc_anon, file: ../../common/fs/zfs/arc.c, line: 2849 (142901-09)

10890:499786962772
6807339	spurious checksum errors when replacing a vdev (142901-13)

11249:6c30f7dfc97b
6906110 bad trap panic in zil_replay_log_record (142901-13)
6906946 zfs replay isn't handling uid/gid correctly (142901-13)

11454:6e69bacc1a5a
6898245 suspended zpool should not cause rest of the zfs/zpool commands to hang (142901-10)

11546:42ea6be8961b (partial merge)
6833999 3-way deadlock in dsl_dataset_hold_ref() and dsl_sync_task_group_sync() (142901-09)

Discussed with:	pjd
Approved by:	delphij (mentor)
Obtained from:	OpenSolaris (multiple Bug IDs)
MFC after:	2 months

1 files changed, 226 insertions, 35 deletions

diff --git a/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/metaslab.c b/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/metaslab.c
index 47f8f5f..d216154 100644
--- a/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/metaslab.c
+++ b/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/metaslab.c
@@ -19,7 +19,7 @@
  * CDDL HEADER END
  */
 /*
- * Copyright 2008 Sun Microsystems, Inc.  All rights reserved.
+ * Copyright 2009 Sun Microsystems, Inc.  All rights reserved.
  * Use is subject to license terms.
  */
 
@@ -36,18 +36,35 @@ uint64_t metaslab_aliquot = 512ULL << 10;
 uint64_t metaslab_gang_bang = SPA_MAXBLOCKSIZE + 1;	/* force gang blocks */
 
 /*
+ * Minimum size which forces the dynamic allocator to change
+ * it's allocation strategy. Once the space map cannot satisfy
+ * an allocation of this size then it switches to using more
+ * aggressive strategy (i.e search by size rather than offset).
+ */
+uint64_t metaslab_df_alloc_threshold = SPA_MAXBLOCKSIZE;
+
+/*
+ * The minimum free space, in percent, which must be available
+ * in a space map to continue allocations in a first-fit fashion.
+ * Once the space_map's free space drops below this level we dynamically
+ * switch to using best-fit allocations.
+ */
+int metaslab_df_free_pct = 30;
+
+/*
  * ==========================================================================
  * Metaslab classes
  * ==========================================================================
  */
 metaslab_class_t *
-metaslab_class_create(void)
+metaslab_class_create(space_map_ops_t *ops)
 {
 	metaslab_class_t *mc;
 
 	mc = kmem_zalloc(sizeof (metaslab_class_t), KM_SLEEP);
 
 	mc->mc_rotor = NULL;
+	mc->mc_ops = ops;
 
 	return (mc);
 }
@@ -202,30 +219,14 @@ metaslab_group_sort(metaslab_group_t *mg, metaslab_t *msp, uint64_t weight)
 }
 
 /*
- * ==========================================================================
- * The first-fit block allocator
- * ==========================================================================
+ * This is a helper function that can be used by the allocator to find
+ * a suitable block to allocate. This will search the specified AVL
+ * tree looking for a block that matches the specified criteria.
  */
-static void
-metaslab_ff_load(space_map_t *sm)
-{
-	ASSERT(sm->sm_ppd == NULL);
-	sm->sm_ppd = kmem_zalloc(64 * sizeof (uint64_t), KM_SLEEP);
-}
-
-static void
-metaslab_ff_unload(space_map_t *sm)
-{
-	kmem_free(sm->sm_ppd, 64 * sizeof (uint64_t));
-	sm->sm_ppd = NULL;
-}
-
 static uint64_t
-metaslab_ff_alloc(space_map_t *sm, uint64_t size)
+metaslab_block_picker(avl_tree_t *t, uint64_t *cursor, uint64_t size,
+    uint64_t align)
 {
-	avl_tree_t *t = &sm->sm_root;
-	uint64_t align = size & -size;
-	uint64_t *cursor = (uint64_t *)sm->sm_ppd + highbit(align) - 1;
 	space_seg_t *ss, ssearch;
 	avl_index_t where;
 
@@ -254,7 +255,37 @@ metaslab_ff_alloc(space_map_t *sm, uint64_t size)
 		return (-1ULL);
 
 	*cursor = 0;
-	return (metaslab_ff_alloc(sm, size));
+	return (metaslab_block_picker(t, cursor, size, align));
+}
+
+/*
+ * ==========================================================================
+ * The first-fit block allocator
+ * ==========================================================================
+ */
+static void
+metaslab_ff_load(space_map_t *sm)
+{
+	ASSERT(sm->sm_ppd == NULL);
+	sm->sm_ppd = kmem_zalloc(64 * sizeof (uint64_t), KM_SLEEP);
+	sm->sm_pp_root = NULL;
+}
+
+static void
+metaslab_ff_unload(space_map_t *sm)
+{
+	kmem_free(sm->sm_ppd, 64 * sizeof (uint64_t));
+	sm->sm_ppd = NULL;
+}
+
+static uint64_t
+metaslab_ff_alloc(space_map_t *sm, uint64_t size)
+{
+	avl_tree_t *t = &sm->sm_root;
+	uint64_t align = size & -size;
+	uint64_t *cursor = (uint64_t *)sm->sm_ppd + highbit(align) - 1;
+
+	return (metaslab_block_picker(t, cursor, size, align));
 }
 
 /* ARGSUSED */
@@ -276,9 +307,136 @@ static space_map_ops_t metaslab_ff_ops = {
 	metaslab_ff_unload,
 	metaslab_ff_alloc,
 	metaslab_ff_claim,
-	metaslab_ff_free
+	metaslab_ff_free,
+	NULL	/* maxsize */
+};
+
+/*
+ * Dynamic block allocator -
+ * Uses the first fit allocation scheme until space get low and then
+ * adjusts to a best fit allocation method. Uses metaslab_df_alloc_threshold
+ * and metaslab_df_free_pct to determine when to switch the allocation scheme.
+ */
+
+uint64_t
+metaslab_df_maxsize(space_map_t *sm)
+{
+	avl_tree_t *t = sm->sm_pp_root;
+	space_seg_t *ss;
+
+	if (t == NULL || (ss = avl_last(t)) == NULL)
+		return (0ULL);
+
+	return (ss->ss_end - ss->ss_start);
+}
+
+static int
+metaslab_df_seg_compare(const void *x1, const void *x2)
+{
+	const space_seg_t *s1 = x1;
+	const space_seg_t *s2 = x2;
+	uint64_t ss_size1 = s1->ss_end - s1->ss_start;
+	uint64_t ss_size2 = s2->ss_end - s2->ss_start;
+
+	if (ss_size1 < ss_size2)
+		return (-1);
+	if (ss_size1 > ss_size2)
+		return (1);
+
+	if (s1->ss_start < s2->ss_start)
+		return (-1);
+	if (s1->ss_start > s2->ss_start)
+		return (1);
+
+	return (0);
+}
+
+static void
+metaslab_df_load(space_map_t *sm)
+{
+	space_seg_t *ss;
+
+	ASSERT(sm->sm_ppd == NULL);
+	sm->sm_ppd = kmem_zalloc(64 * sizeof (uint64_t), KM_SLEEP);
+
+	sm->sm_pp_root = kmem_alloc(sizeof (avl_tree_t), KM_SLEEP);
+	avl_create(sm->sm_pp_root, metaslab_df_seg_compare,
+	    sizeof (space_seg_t), offsetof(struct space_seg, ss_pp_node));
+
+	for (ss = avl_first(&sm->sm_root); ss; ss = AVL_NEXT(&sm->sm_root, ss))
+		avl_add(sm->sm_pp_root, ss);
+}
+
+static void
+metaslab_df_unload(space_map_t *sm)
+{
+	void *cookie = NULL;
+
+	kmem_free(sm->sm_ppd, 64 * sizeof (uint64_t));
+	sm->sm_ppd = NULL;
+
+	while (avl_destroy_nodes(sm->sm_pp_root, &cookie) != NULL) {
+		/* tear down the tree */
+	}
+
+	avl_destroy(sm->sm_pp_root);
+	kmem_free(sm->sm_pp_root, sizeof (avl_tree_t));
+	sm->sm_pp_root = NULL;
+}
+
+static uint64_t
+metaslab_df_alloc(space_map_t *sm, uint64_t size)
+{
+	avl_tree_t *t = &sm->sm_root;
+	uint64_t align = size & -size;
+	uint64_t *cursor = (uint64_t *)sm->sm_ppd + highbit(align) - 1;
+	uint64_t max_size = metaslab_df_maxsize(sm);
+	int free_pct = sm->sm_space * 100 / sm->sm_size;
+
+	ASSERT(MUTEX_HELD(sm->sm_lock));
+	ASSERT3U(avl_numnodes(&sm->sm_root), ==, avl_numnodes(sm->sm_pp_root));
+
+	if (max_size < size)
+		return (-1ULL);
+
+	/*
+	 * If we're running low on space switch to using the size
+	 * sorted AVL tree (best-fit).
+	 */
+	if (max_size < metaslab_df_alloc_threshold ||
+	    free_pct < metaslab_df_free_pct) {
+		t = sm->sm_pp_root;
+		*cursor = 0;
+	}
+
+	return (metaslab_block_picker(t, cursor, size, 1ULL));
+}
+
+/* ARGSUSED */
+static void
+metaslab_df_claim(space_map_t *sm, uint64_t start, uint64_t size)
+{
+	/* No need to update cursor */
+}
+
+/* ARGSUSED */
+static void
+metaslab_df_free(space_map_t *sm, uint64_t start, uint64_t size)
+{
+	/* No need to update cursor */
+}
+
+static space_map_ops_t metaslab_df_ops = {
+	metaslab_df_load,
+	metaslab_df_unload,
+	metaslab_df_alloc,
+	metaslab_df_claim,
+	metaslab_df_free,
+	metaslab_df_maxsize
 };
 
+space_map_ops_t *zfs_metaslab_ops = &metaslab_df_ops;
+
 /*
  * ==========================================================================
  * Metaslabs
@@ -414,20 +572,28 @@ metaslab_weight(metaslab_t *msp)
 }
 
 static int
-metaslab_activate(metaslab_t *msp, uint64_t activation_weight)
+metaslab_activate(metaslab_t *msp, uint64_t activation_weight, uint64_t size)
 {
 	space_map_t *sm = &msp->ms_map;
+	space_map_ops_t *sm_ops = msp->ms_group->mg_class->mc_ops;
 
 	ASSERT(MUTEX_HELD(&msp->ms_lock));
 
 	if ((msp->ms_weight & METASLAB_ACTIVE_MASK) == 0) {
-		int error = space_map_load(sm, &metaslab_ff_ops,
-		    SM_FREE, &msp->ms_smo,
+		int error = space_map_load(sm, sm_ops, SM_FREE, &msp->ms_smo,
 		    msp->ms_group->mg_vd->vdev_spa->spa_meta_objset);
 		if (error) {
 			metaslab_group_sort(msp->ms_group, msp, 0);
 			return (error);
 		}
+
+		/*
+		 * If we were able to load the map then make sure
+		 * that this map is still able to satisfy our request.
+		 */
+		if (msp->ms_weight < size)
+			return (ENOSPC);
+
 		metaslab_group_sort(msp->ms_group, msp,
 		    msp->ms_weight | activation_weight);
 	}
@@ -636,11 +802,16 @@ metaslab_group_alloc(metaslab_group_t *mg, uint64_t size, uint64_t txg,
 	int i;
 
 	activation_weight = METASLAB_WEIGHT_PRIMARY;
-	for (i = 0; i < d; i++)
-		if (DVA_GET_VDEV(&dva[i]) == mg->mg_vd->vdev_id)
+	for (i = 0; i < d; i++) {
+		if (DVA_GET_VDEV(&dva[i]) == mg->mg_vd->vdev_id) {
 			activation_weight = METASLAB_WEIGHT_SECONDARY;
+			break;
+		}
+	}
 
 	for (;;) {
+		boolean_t was_active;
+
 		mutex_enter(&mg->mg_lock);
 		for (msp = avl_first(t); msp; msp = AVL_NEXT(t, msp)) {
 			if (msp->ms_weight < size) {
@@ -648,6 +819,7 @@ metaslab_group_alloc(metaslab_group_t *mg, uint64_t size, uint64_t txg,
 				return (-1ULL);
 			}
 
+			was_active = msp->ms_weight & METASLAB_ACTIVE_MASK;
 			if (activation_weight == METASLAB_WEIGHT_PRIMARY)
 				break;
 
@@ -673,7 +845,9 @@ metaslab_group_alloc(metaslab_group_t *mg, uint64_t size, uint64_t txg,
 		 * another thread may have changed the weight while we
 		 * were blocked on the metaslab lock.
 		 */
-		if (msp->ms_weight < size) {
+		if (msp->ms_weight < size || (was_active &&
+		    !(msp->ms_weight & METASLAB_ACTIVE_MASK) &&
+		    activation_weight == METASLAB_WEIGHT_PRIMARY)) {
 			mutex_exit(&msp->ms_lock);
 			continue;
 		}
@@ -686,7 +860,7 @@ metaslab_group_alloc(metaslab_group_t *mg, uint64_t size, uint64_t txg,
 			continue;
 		}
 
-		if (metaslab_activate(msp, activation_weight) != 0) {
+		if (metaslab_activate(msp, activation_weight, size) != 0) {
 			mutex_exit(&msp->ms_lock);
 			continue;
 		}
@@ -720,6 +894,8 @@ metaslab_alloc_dva(spa_t *spa, metaslab_class_t *mc, uint64_t psize,
 	vdev_t *vd;
 	int dshift = 3;
 	int all_zero;
+	int zio_lock = B_FALSE;
+	boolean_t allocatable;
 	uint64_t offset = -1ULL;
 	uint64_t asize;
 	uint64_t distance;
@@ -778,11 +954,20 @@ top:
 	all_zero = B_TRUE;
 	do {
 		vd = mg->mg_vd;
+
 		/*
 		 * Don't allocate from faulted devices.
 		 */
-		if (!vdev_allocatable(vd))
+		if (zio_lock) {
+			spa_config_enter(spa, SCL_ZIO, FTAG, RW_READER);
+			allocatable = vdev_allocatable(vd);
+			spa_config_exit(spa, SCL_ZIO, FTAG);
+		} else {
+			allocatable = vdev_allocatable(vd);
+		}
+		if (!allocatable)
 			goto next;
+
 		/*
 		 * Avoid writing single-copy data to a failing vdev
 		 */
@@ -858,6 +1043,12 @@ next:
 		goto top;
 	}
 
+	if (!allocatable && !zio_lock) {
+		dshift = 3;
+		zio_lock = B_TRUE;
+		goto top;
+	}
+
 	bzero(&dva[d], sizeof (dva_t));
 
 	return (ENOSPC);
@@ -938,7 +1129,7 @@ metaslab_claim_dva(spa_t *spa, const dva_t *dva, uint64_t txg)
 
 	mutex_enter(&msp->ms_lock);
 
-	error = metaslab_activate(msp, METASLAB_WEIGHT_SECONDARY);
+	error = metaslab_activate(msp, METASLAB_WEIGHT_SECONDARY, 0);
 	if (error || txg == 0) {	/* txg == 0 indicates dry run */
 		mutex_exit(&msp->ms_lock);
 		return (error);
@@ -946,7 +1137,7 @@ metaslab_claim_dva(spa_t *spa, const dva_t *dva, uint64_t txg)
 
 	space_map_claim(&msp->ms_map, offset, size);
 
-	if (spa_mode & FWRITE) {	/* don't dirty if we're zdb(1M) */
+	if (spa_writeable(spa)) {	/* don't dirty if we're zdb(1M) */
 		if (msp->ms_allocmap[txg & TXG_MASK].sm_space == 0)
 			vdev_dirty(vd, VDD_METASLAB, msp, txg);
 		space_map_add(&msp->ms_allocmap[txg & TXG_MASK], offset, size);