*
* CDDL HEADER END
*/
+
/*
* Copyright (c) 2005, 2010, Oracle and/or its affiliates. All rights reserved.
- * Copyright (c) 2012 by Delphix. All rights reserved.
- * Copyright (c) 2012, Joyent, Inc. All rights reserved.
+ * Copyright (c) 2011, 2018 by Delphix. All rights reserved.
+ * Copyright (c) 2014, Joyent, Inc. All rights reserved.
+ * Copyright (c) 2014 RackTop Systems.
+ * Copyright (c) 2014 Spectra Logic Corporation, All rights reserved.
+ * Copyright (c) 2016 Actifio, Inc. All rights reserved.
+ * Copyright 2016, OmniTI Computer Consulting, Inc. All rights reserved.
+ * Copyright 2017 Nexenta Systems, Inc.
*/
#include <sys/dmu_objset.h>
#include <sys/zfs_context.h>
#include <sys/zfs_ioctl.h>
#include <sys/spa.h>
+#include <sys/spa_impl.h>
+#include <sys/vdev.h>
#include <sys/zfs_znode.h>
#include <sys/zfs_onexit.h>
#include <sys/zvol.h>
#include <sys/dsl_scan.h>
#include <sys/dsl_deadlist.h>
+#include <sys/dsl_destroy.h>
+#include <sys/dsl_userhold.h>
+#include <sys/dsl_bookmark.h>
+#include <sys/policy.h>
+#include <sys/dmu_recv.h>
+#include <sys/zio_compress.h>
+#include <zfs_fletcher.h>
+#include <sys/zio_checksum.h>
-static char *dsl_reaper = "the grim reaper";
-
-static dsl_checkfunc_t dsl_dataset_destroy_begin_check;
-static dsl_syncfunc_t dsl_dataset_destroy_begin_sync;
-static dsl_syncfunc_t dsl_dataset_set_reservation_sync;
+/*
+ * The SPA supports block sizes up to 16MB. However, very large blocks
+ * can have an impact on i/o latency (e.g. tying up a spinning disk for
+ * ~300ms), and also potentially on the memory allocator. Therefore,
+ * we do not allow the recordsize to be set larger than zfs_max_recordsize
+ * (default 1MB). Larger blocks can be created by changing this tunable,
+ * and pools with larger blocks can always be imported and used, regardless
+ * of this setting.
+ */
+int zfs_max_recordsize = 1 * 1024 * 1024;
#define SWITCH64(x, y) \
{ \
#define DS_REF_MAX (1ULL << 62)
-#define DSL_DEADLIST_BLOCKSIZE SPA_MAXBLOCKSIZE
+extern inline dsl_dataset_phys_t *dsl_dataset_phys(dsl_dataset_t *ds);
-#define DSL_DATASET_IS_DESTROYED(ds) ((ds)->ds_owner == dsl_reaper)
+static void dsl_dataset_set_remap_deadlist_object(dsl_dataset_t *ds,
+ uint64_t obj, dmu_tx_t *tx);
+static void dsl_dataset_unset_remap_deadlist_object(dsl_dataset_t *ds,
+ dmu_tx_t *tx);
+static void unload_zfeature(dsl_dataset_t *ds, spa_feature_t f);
+
+extern int spa_asize_inflation;
+
+static zil_header_t zero_zil;
/*
- * Figure out how much of this delta should be propogated to the dsl_dir
+ * Figure out how much of this delta should be propagated to the dsl_dir
* layer. If there's a refreservation, that space has already been
* partially accounted for in our ancestors.
*/
static int64_t
parent_delta(dsl_dataset_t *ds, int64_t delta)
{
+ dsl_dataset_phys_t *ds_phys;
uint64_t old_bytes, new_bytes;
if (ds->ds_reserved == 0)
return (delta);
- old_bytes = MAX(ds->ds_phys->ds_unique_bytes, ds->ds_reserved);
- new_bytes = MAX(ds->ds_phys->ds_unique_bytes + delta, ds->ds_reserved);
+ ds_phys = dsl_dataset_phys(ds);
+ old_bytes = MAX(ds_phys->ds_unique_bytes, ds->ds_reserved);
+ new_bytes = MAX(ds_phys->ds_unique_bytes + delta, ds->ds_reserved);
ASSERT3U(ABS((int64_t)(new_bytes - old_bytes)), <=, ABS(delta));
return (new_bytes - old_bytes);
used, compressed, uncompressed);
return;
}
- dmu_buf_will_dirty(ds->ds_dbuf, tx);
- mutex_enter(&ds->ds_dir->dd_lock);
+ ASSERT3U(bp->blk_birth, >, dsl_dataset_phys(ds)->ds_prev_snap_txg);
+ dmu_buf_will_dirty(ds->ds_dbuf, tx);
mutex_enter(&ds->ds_lock);
delta = parent_delta(ds, used);
- ds->ds_phys->ds_referenced_bytes += used;
- ds->ds_phys->ds_compressed_bytes += compressed;
- ds->ds_phys->ds_uncompressed_bytes += uncompressed;
- ds->ds_phys->ds_unique_bytes += used;
+ dsl_dataset_phys(ds)->ds_referenced_bytes += used;
+ dsl_dataset_phys(ds)->ds_compressed_bytes += compressed;
+ dsl_dataset_phys(ds)->ds_uncompressed_bytes += uncompressed;
+ dsl_dataset_phys(ds)->ds_unique_bytes += used;
+
+ if (BP_GET_LSIZE(bp) > SPA_OLD_MAXBLOCKSIZE) {
+ ds->ds_feature_activation[SPA_FEATURE_LARGE_BLOCKS] =
+ (void *)B_TRUE;
+ }
+
+ spa_feature_t f = zio_checksum_to_feature(BP_GET_CHECKSUM(bp));
+ if (f != SPA_FEATURE_NONE) {
+ ASSERT3S(spa_feature_table[f].fi_type, ==,
+ ZFEATURE_TYPE_BOOLEAN);
+ ds->ds_feature_activation[f] = (void *)B_TRUE;
+ }
+
mutex_exit(&ds->ds_lock);
dsl_dir_diduse_space(ds->ds_dir, DD_USED_HEAD, delta,
compressed, uncompressed, tx);
dsl_dir_transfer_space(ds->ds_dir, used - delta,
DD_USED_REFRSRV, DD_USED_HEAD, tx);
- mutex_exit(&ds->ds_dir->dd_lock);
+}
+
+/*
+ * Called when the specified segment has been remapped, and is thus no
+ * longer referenced in the head dataset. The vdev must be indirect.
+ *
+ * If the segment is referenced by a snapshot, put it on the remap deadlist.
+ * Otherwise, add this segment to the obsolete spacemap.
+ */
+void
+dsl_dataset_block_remapped(dsl_dataset_t *ds, uint64_t vdev, uint64_t offset,
+ uint64_t size, uint64_t birth, dmu_tx_t *tx)
+{
+ spa_t *spa = ds->ds_dir->dd_pool->dp_spa;
+
+ ASSERT(dmu_tx_is_syncing(tx));
+ ASSERT(birth <= tx->tx_txg);
+ ASSERT(!ds->ds_is_snapshot);
+
+ if (birth > dsl_dataset_phys(ds)->ds_prev_snap_txg) {
+ spa_vdev_indirect_mark_obsolete(spa, vdev, offset, size, tx);
+ } else {
+ blkptr_t fakebp;
+ dva_t *dva = &fakebp.blk_dva[0];
+
+ ASSERT(ds != NULL);
+
+ mutex_enter(&ds->ds_remap_deadlist_lock);
+ if (!dsl_dataset_remap_deadlist_exists(ds)) {
+ dsl_dataset_create_remap_deadlist(ds, tx);
+ }
+ mutex_exit(&ds->ds_remap_deadlist_lock);
+
+ BP_ZERO(&fakebp);
+ fakebp.blk_birth = birth;
+ DVA_SET_VDEV(dva, vdev);
+ DVA_SET_OFFSET(dva, offset);
+ DVA_SET_ASIZE(dva, size);
+
+ dsl_deadlist_insert(&ds->ds_remap_deadlist, &fakebp, tx);
+ }
}
int
dsl_dataset_block_kill(dsl_dataset_t *ds, const blkptr_t *bp, dmu_tx_t *tx,
boolean_t async)
{
- int used, compressed, uncompressed;
+ spa_t *spa = dmu_tx_pool(tx)->dp_spa;
+
+ int used = bp_get_dsize_sync(spa, bp);
+ int compressed = BP_GET_PSIZE(bp);
+ int uncompressed = BP_GET_UCSIZE(bp);
if (BP_IS_HOLE(bp))
return (0);
ASSERT(dmu_tx_is_syncing(tx));
ASSERT(bp->blk_birth <= tx->tx_txg);
- used = bp_get_dsize_sync(tx->tx_pool->dp_spa, bp);
- compressed = BP_GET_PSIZE(bp);
- uncompressed = BP_GET_UCSIZE(bp);
-
- ASSERT(used > 0);
if (ds == NULL) {
dsl_free(tx->tx_pool, tx->tx_txg, bp);
dsl_pool_mos_diduse_space(tx->tx_pool,
}
ASSERT3P(tx->tx_pool, ==, ds->ds_dir->dd_pool);
- ASSERT(!dsl_dataset_is_snapshot(ds));
+ ASSERT(!ds->ds_is_snapshot);
dmu_buf_will_dirty(ds->ds_dbuf, tx);
- if (bp->blk_birth > ds->ds_phys->ds_prev_snap_txg) {
+ if (bp->blk_birth > dsl_dataset_phys(ds)->ds_prev_snap_txg) {
int64_t delta;
dprintf_bp(bp, "freeing ds=%llu", ds->ds_object);
dsl_free(tx->tx_pool, tx->tx_txg, bp);
- mutex_enter(&ds->ds_dir->dd_lock);
mutex_enter(&ds->ds_lock);
- ASSERT(ds->ds_phys->ds_unique_bytes >= used ||
+ ASSERT(dsl_dataset_phys(ds)->ds_unique_bytes >= used ||
!DS_UNIQUE_IS_ACCURATE(ds));
delta = parent_delta(ds, -used);
- ds->ds_phys->ds_unique_bytes -= used;
+ dsl_dataset_phys(ds)->ds_unique_bytes -= used;
mutex_exit(&ds->ds_lock);
dsl_dir_diduse_space(ds->ds_dir, DD_USED_HEAD,
delta, -compressed, -uncompressed, tx);
dsl_dir_transfer_space(ds->ds_dir, -used - delta,
DD_USED_REFRSRV, DD_USED_HEAD, tx);
- mutex_exit(&ds->ds_dir->dd_lock);
} else {
dprintf_bp(bp, "putting on dead list: %s", "");
if (async) {
dsl_deadlist_insert(&ds->ds_deadlist, bp, tx);
}
ASSERT3U(ds->ds_prev->ds_object, ==,
- ds->ds_phys->ds_prev_snap_obj);
- ASSERT(ds->ds_prev->ds_phys->ds_num_children > 0);
+ dsl_dataset_phys(ds)->ds_prev_snap_obj);
+ ASSERT(dsl_dataset_phys(ds->ds_prev)->ds_num_children > 0);
/* if (bp->blk_birth > prev prev snap txg) prev unique += bs */
- if (ds->ds_prev->ds_phys->ds_next_snap_obj ==
+ if (dsl_dataset_phys(ds->ds_prev)->ds_next_snap_obj ==
ds->ds_object && bp->blk_birth >
- ds->ds_prev->ds_phys->ds_prev_snap_txg) {
+ dsl_dataset_phys(ds->ds_prev)->ds_prev_snap_txg) {
dmu_buf_will_dirty(ds->ds_prev->ds_dbuf, tx);
mutex_enter(&ds->ds_prev->ds_lock);
- ds->ds_prev->ds_phys->ds_unique_bytes += used;
+ dsl_dataset_phys(ds->ds_prev)->ds_unique_bytes += used;
mutex_exit(&ds->ds_prev->ds_lock);
}
if (bp->blk_birth > ds->ds_dir->dd_origin_txg) {
}
}
mutex_enter(&ds->ds_lock);
- ASSERT3U(ds->ds_phys->ds_referenced_bytes, >=, used);
- ds->ds_phys->ds_referenced_bytes -= used;
- ASSERT3U(ds->ds_phys->ds_compressed_bytes, >=, compressed);
- ds->ds_phys->ds_compressed_bytes -= compressed;
- ASSERT3U(ds->ds_phys->ds_uncompressed_bytes, >=, uncompressed);
- ds->ds_phys->ds_uncompressed_bytes -= uncompressed;
+ ASSERT3U(dsl_dataset_phys(ds)->ds_referenced_bytes, >=, used);
+ dsl_dataset_phys(ds)->ds_referenced_bytes -= used;
+ ASSERT3U(dsl_dataset_phys(ds)->ds_compressed_bytes, >=, compressed);
+ dsl_dataset_phys(ds)->ds_compressed_bytes -= compressed;
+ ASSERT3U(dsl_dataset_phys(ds)->ds_uncompressed_bytes, >=, uncompressed);
+ dsl_dataset_phys(ds)->ds_uncompressed_bytes -= uncompressed;
mutex_exit(&ds->ds_lock);
return (used);
}
-uint64_t
-dsl_dataset_prev_snap_txg(dsl_dataset_t *ds)
+struct feature_type_uint64_array_arg {
+ uint64_t length;
+ uint64_t *array;
+};
+
+static void
+unload_zfeature(dsl_dataset_t *ds, spa_feature_t f)
{
- uint64_t trysnap = 0;
+ switch (spa_feature_table[f].fi_type) {
+ case ZFEATURE_TYPE_BOOLEAN:
+ break;
+ case ZFEATURE_TYPE_UINT64_ARRAY:
+ {
+ struct feature_type_uint64_array_arg *ftuaa = ds->ds_feature[f];
+ kmem_free(ftuaa->array, ftuaa->length * sizeof (uint64_t));
+ kmem_free(ftuaa, sizeof (*ftuaa));
+ break;
+ }
+ default:
+ panic("Invalid zfeature type %d", spa_feature_table[f].fi_type);
+ }
+}
- if (ds == NULL)
- return (0);
- /*
- * The snapshot creation could fail, but that would cause an
- * incorrect FALSE return, which would only result in an
- * overestimation of the amount of space that an operation would
- * consume, which is OK.
- *
- * There's also a small window where we could miss a pending
- * snapshot, because we could set the sync task in the quiescing
- * phase. So this should only be used as a guess.
- */
- if (ds->ds_trysnap_txg >
- spa_last_synced_txg(ds->ds_dir->dd_pool->dp_spa))
- trysnap = ds->ds_trysnap_txg;
- return (MAX(ds->ds_phys->ds_prev_snap_txg, trysnap));
+static int
+load_zfeature(objset_t *mos, dsl_dataset_t *ds, spa_feature_t f)
+{
+ int err = 0;
+ switch (spa_feature_table[f].fi_type) {
+ case ZFEATURE_TYPE_BOOLEAN:
+ err = zap_contains(mos, ds->ds_object,
+ spa_feature_table[f].fi_guid);
+ if (err == 0) {
+ ds->ds_feature[f] = (void *)B_TRUE;
+ } else {
+ ASSERT3U(err, ==, ENOENT);
+ err = 0;
+ }
+ break;
+ case ZFEATURE_TYPE_UINT64_ARRAY:
+ {
+ uint64_t int_size, num_int;
+ uint64_t *data;
+ err = zap_length(mos, ds->ds_object,
+ spa_feature_table[f].fi_guid, &int_size, &num_int);
+ if (err != 0) {
+ ASSERT3U(err, ==, ENOENT);
+ err = 0;
+ break;
+ }
+ ASSERT3U(int_size, ==, sizeof (uint64_t));
+ data = kmem_alloc(int_size * num_int, KM_SLEEP);
+ VERIFY0(zap_lookup(mos, ds->ds_object,
+ spa_feature_table[f].fi_guid, int_size, num_int, data));
+ struct feature_type_uint64_array_arg *ftuaa =
+ kmem_alloc(sizeof (*ftuaa), KM_SLEEP);
+ ftuaa->length = num_int;
+ ftuaa->array = data;
+ ds->ds_feature[f] = ftuaa;
+ break;
+ }
+ default:
+ panic("Invalid zfeature type %d", spa_feature_table[f].fi_type);
+ }
+ return (err);
}
-boolean_t
-dsl_dataset_block_freeable(dsl_dataset_t *ds, const blkptr_t *bp,
- uint64_t blk_birth)
+/*
+ * We have to release the fsid syncronously or we risk that a subsequent
+ * mount of the same dataset will fail to unique_insert the fsid. This
+ * failure would manifest itself as the fsid of this dataset changing
+ * between mounts which makes NFS clients quite unhappy.
+ */
+static void
+dsl_dataset_evict_sync(void *dbu)
{
- if (blk_birth <= dsl_dataset_prev_snap_txg(ds))
- return (B_FALSE);
+ dsl_dataset_t *ds = dbu;
- ddt_prefetch(dsl_dataset_get_spa(ds), bp);
+ ASSERT(ds->ds_owner == NULL);
- return (B_TRUE);
+ unique_remove(ds->ds_fsid_guid);
}
-/* ARGSUSED */
static void
-dsl_dataset_evict(dmu_buf_t *db, void *dsv)
+dsl_dataset_evict_async(void *dbu)
{
- dsl_dataset_t *ds = dsv;
+ dsl_dataset_t *ds = dbu;
- ASSERT(ds->ds_owner == NULL || DSL_DATASET_IS_DESTROYED(ds));
+ ASSERT(ds->ds_owner == NULL);
- unique_remove(ds->ds_fsid_guid);
+ ds->ds_dbuf = NULL;
if (ds->ds_objset != NULL)
dmu_objset_evict(ds->ds_objset);
if (ds->ds_prev) {
- dsl_dataset_drop_ref(ds->ds_prev, ds);
+ dsl_dataset_rele(ds->ds_prev, ds);
ds->ds_prev = NULL;
}
bplist_destroy(&ds->ds_pending_deadlist);
- if (db != NULL) {
+ if (dsl_deadlist_is_open(&ds->ds_deadlist))
dsl_deadlist_close(&ds->ds_deadlist);
- } else {
- ASSERT(ds->ds_deadlist.dl_dbuf == NULL);
- ASSERT(!ds->ds_deadlist.dl_oldfmt);
- }
+ if (dsl_deadlist_is_open(&ds->ds_remap_deadlist))
+ dsl_deadlist_close(&ds->ds_remap_deadlist);
if (ds->ds_dir)
- dsl_dir_close(ds->ds_dir, ds);
+ dsl_dir_async_rele(ds->ds_dir, ds);
ASSERT(!list_link_active(&ds->ds_synced_link));
+ for (spa_feature_t f = 0; f < SPA_FEATURES; f++) {
+ if (dsl_dataset_feature_is_active(ds, f))
+ unload_zfeature(ds, f);
+ }
+
+ list_destroy(&ds->ds_prop_cbs);
mutex_destroy(&ds->ds_lock);
- mutex_destroy(&ds->ds_recvlock);
mutex_destroy(&ds->ds_opening_lock);
- rw_destroy(&ds->ds_rwlock);
- cv_destroy(&ds->ds_exclusive_cv);
+ mutex_destroy(&ds->ds_sendstream_lock);
+ mutex_destroy(&ds->ds_remap_deadlist_lock);
+ zfs_refcount_destroy(&ds->ds_longholds);
+ rrw_destroy(&ds->ds_bp_rwlock);
kmem_free(ds, sizeof (dsl_dataset_t));
}
-static int
+int
dsl_dataset_get_snapname(dsl_dataset_t *ds)
{
dsl_dataset_phys_t *headphys;
if (ds->ds_snapname[0])
return (0);
- if (ds->ds_phys->ds_next_snap_obj == 0)
+ if (dsl_dataset_phys(ds)->ds_next_snap_obj == 0)
return (0);
- err = dmu_bonus_hold(mos, ds->ds_dir->dd_phys->dd_head_dataset_obj,
+ err = dmu_bonus_hold(mos, dsl_dir_phys(ds->ds_dir)->dd_head_dataset_obj,
FTAG, &headdbuf);
- if (err)
+ if (err != 0)
return (err);
headphys = headdbuf->db_data;
err = zap_value_search(dp->dp_meta_objset,
headphys->ds_snapnames_zapobj, ds->ds_object, 0, ds->ds_snapname);
+ if (err != 0 && zfs_recover == B_TRUE) {
+ err = 0;
+ (void) snprintf(ds->ds_snapname, sizeof (ds->ds_snapname),
+ "SNAPOBJ=%llu-ERR=%d",
+ (unsigned long long)ds->ds_object, err);
+ }
dmu_buf_rele(headdbuf, FTAG);
return (err);
}
-static int
+int
dsl_dataset_snap_lookup(dsl_dataset_t *ds, const char *name, uint64_t *value)
{
objset_t *mos = ds->ds_dir->dd_pool->dp_meta_objset;
- uint64_t snapobj = ds->ds_phys->ds_snapnames_zapobj;
- matchtype_t mt;
+ uint64_t snapobj = dsl_dataset_phys(ds)->ds_snapnames_zapobj;
+ matchtype_t mt = 0;
int err;
- if (ds->ds_phys->ds_flags & DS_FLAG_CI_DATASET)
- mt = MT_FIRST;
- else
- mt = MT_EXACT;
+ if (dsl_dataset_phys(ds)->ds_flags & DS_FLAG_CI_DATASET)
+ mt = MT_NORMALIZE;
err = zap_lookup_norm(mos, snapobj, name, 8, 1,
value, mt, NULL, 0, NULL);
- if (err == ENOTSUP && mt == MT_FIRST)
+ if (err == ENOTSUP && (mt & MT_NORMALIZE))
err = zap_lookup(mos, snapobj, name, 8, 1, value);
return (err);
}
-static int
-dsl_dataset_snap_remove(dsl_dataset_t *ds, char *name, dmu_tx_t *tx)
+int
+dsl_dataset_snap_remove(dsl_dataset_t *ds, const char *name, dmu_tx_t *tx,
+ boolean_t adj_cnt)
{
objset_t *mos = ds->ds_dir->dd_pool->dp_meta_objset;
- uint64_t snapobj = ds->ds_phys->ds_snapnames_zapobj;
- matchtype_t mt;
+ uint64_t snapobj = dsl_dataset_phys(ds)->ds_snapnames_zapobj;
+ matchtype_t mt = 0;
int err;
dsl_dir_snap_cmtime_update(ds->ds_dir);
- if (ds->ds_phys->ds_flags & DS_FLAG_CI_DATASET)
- mt = MT_FIRST;
- else
- mt = MT_EXACT;
+ if (dsl_dataset_phys(ds)->ds_flags & DS_FLAG_CI_DATASET)
+ mt = MT_NORMALIZE;
err = zap_remove_norm(mos, snapobj, name, mt, tx);
- if (err == ENOTSUP && mt == MT_FIRST)
+ if (err == ENOTSUP && (mt & MT_NORMALIZE))
err = zap_remove(mos, snapobj, name, tx);
+
+ if (err == 0 && adj_cnt)
+ dsl_fs_ss_count_adjust(ds->ds_dir, -1,
+ DD_FIELD_SNAPSHOT_COUNT, tx);
+
return (err);
}
-static int
-dsl_dataset_get_ref(dsl_pool_t *dp, uint64_t dsobj, void *tag,
+boolean_t
+dsl_dataset_try_add_ref(dsl_pool_t *dp, dsl_dataset_t *ds, void *tag)
+{
+ dmu_buf_t *dbuf = ds->ds_dbuf;
+ boolean_t result = B_FALSE;
+
+ if (dbuf != NULL && dmu_buf_try_add_ref(dbuf, dp->dp_meta_objset,
+ ds->ds_object, DMU_BONUS_BLKID, tag)) {
+
+ if (ds == dmu_buf_get_user(dbuf))
+ result = B_TRUE;
+ else
+ dmu_buf_rele(dbuf, tag);
+ }
+
+ return (result);
+}
+
+int
+dsl_dataset_hold_obj(dsl_pool_t *dp, uint64_t dsobj, void *tag,
dsl_dataset_t **dsp)
{
objset_t *mos = dp->dp_meta_objset;
int err;
dmu_object_info_t doi;
- ASSERT(RW_LOCK_HELD(&dp->dp_config_rwlock) ||
- dsl_pool_sync_context(dp));
+ ASSERT(dsl_pool_config_held(dp));
err = dmu_bonus_hold(mos, dsobj, tag, &dbuf);
- if (err)
+ if (err != 0)
return (err);
/* Make sure dsobj has the correct object type. */
dmu_object_info_from_db(dbuf, &doi);
- if (doi.doi_type != DMU_OT_DSL_DATASET)
- return (EINVAL);
+ if (doi.doi_bonus_type != DMU_OT_DSL_DATASET) {
+ dmu_buf_rele(dbuf, tag);
+ return (SET_ERROR(EINVAL));
+ }
ds = dmu_buf_get_user(dbuf);
if (ds == NULL) {
dsl_dataset_t *winner = NULL;
- ds = kmem_zalloc(sizeof (dsl_dataset_t), KM_PUSHPAGE);
+ ds = kmem_zalloc(sizeof (dsl_dataset_t), KM_SLEEP);
ds->ds_dbuf = dbuf;
ds->ds_object = dsobj;
- ds->ds_phys = dbuf->db_data;
+ ds->ds_is_snapshot = dsl_dataset_phys(ds)->ds_num_children != 0;
list_link_init(&ds->ds_synced_link);
+ err = dsl_dir_hold_obj(dp, dsl_dataset_phys(ds)->ds_dir_obj,
+ NULL, ds, &ds->ds_dir);
+ if (err != 0) {
+ kmem_free(ds, sizeof (dsl_dataset_t));
+ dmu_buf_rele(dbuf, tag);
+ return (err);
+ }
+
mutex_init(&ds->ds_lock, NULL, MUTEX_DEFAULT, NULL);
- mutex_init(&ds->ds_recvlock, NULL, MUTEX_DEFAULT, NULL);
mutex_init(&ds->ds_opening_lock, NULL, MUTEX_DEFAULT, NULL);
mutex_init(&ds->ds_sendstream_lock, NULL, MUTEX_DEFAULT, NULL);
-
- rw_init(&ds->ds_rwlock, NULL, RW_DEFAULT, NULL);
- cv_init(&ds->ds_exclusive_cv, NULL, CV_DEFAULT, NULL);
+ mutex_init(&ds->ds_remap_deadlist_lock,
+ NULL, MUTEX_DEFAULT, NULL);
+ rrw_init(&ds->ds_bp_rwlock, B_FALSE);
+ zfs_refcount_create(&ds->ds_longholds);
bplist_create(&ds->ds_pending_deadlist);
- dsl_deadlist_open(&ds->ds_deadlist,
- mos, ds->ds_phys->ds_deadlist_obj);
list_create(&ds->ds_sendstreams, sizeof (dmu_sendarg_t),
offsetof(dmu_sendarg_t, dsa_link));
- if (err == 0) {
- err = dsl_dir_open_obj(dp,
- ds->ds_phys->ds_dir_obj, NULL, ds, &ds->ds_dir);
- }
- if (err) {
- mutex_destroy(&ds->ds_lock);
- mutex_destroy(&ds->ds_recvlock);
- mutex_destroy(&ds->ds_opening_lock);
- rw_destroy(&ds->ds_rwlock);
- cv_destroy(&ds->ds_exclusive_cv);
- bplist_destroy(&ds->ds_pending_deadlist);
- dsl_deadlist_close(&ds->ds_deadlist);
- kmem_free(ds, sizeof (dsl_dataset_t));
- dmu_buf_rele(dbuf, tag);
- return (err);
+ list_create(&ds->ds_prop_cbs, sizeof (dsl_prop_cb_record_t),
+ offsetof(dsl_prop_cb_record_t, cbr_ds_node));
+
+ if (doi.doi_type == DMU_OTN_ZAP_METADATA) {
+ spa_feature_t f;
+
+ for (f = 0; f < SPA_FEATURES; f++) {
+ if (!(spa_feature_table[f].fi_flags &
+ ZFEATURE_FLAG_PER_DATASET))
+ continue;
+ err = load_zfeature(mos, ds, f);
+ }
}
- if (!dsl_dataset_is_snapshot(ds)) {
+ if (!ds->ds_is_snapshot) {
ds->ds_snapname[0] = '\0';
- if (ds->ds_phys->ds_prev_snap_obj) {
- err = dsl_dataset_get_ref(dp,
- ds->ds_phys->ds_prev_snap_obj,
+ if (dsl_dataset_phys(ds)->ds_prev_snap_obj != 0) {
+ err = dsl_dataset_hold_obj(dp,
+ dsl_dataset_phys(ds)->ds_prev_snap_obj,
ds, &ds->ds_prev);
}
+ if (doi.doi_type == DMU_OTN_ZAP_METADATA) {
+ int zaperr = zap_lookup(mos, ds->ds_object,
+ DS_FIELD_BOOKMARK_NAMES,
+ sizeof (ds->ds_bookmarks), 1,
+ &ds->ds_bookmarks);
+ if (zaperr != ENOENT)
+ VERIFY0(zaperr);
+ }
} else {
if (zfs_flags & ZFS_DEBUG_SNAPNAMES)
err = dsl_dataset_get_snapname(ds);
- if (err == 0 && ds->ds_phys->ds_userrefs_obj != 0) {
+ if (err == 0 &&
+ dsl_dataset_phys(ds)->ds_userrefs_obj != 0) {
err = zap_count(
ds->ds_dir->dd_pool->dp_meta_objset,
- ds->ds_phys->ds_userrefs_obj,
+ dsl_dataset_phys(ds)->ds_userrefs_obj,
&ds->ds_userrefs);
}
}
- if (err == 0 && !dsl_dataset_is_snapshot(ds)) {
- /*
- * In sync context, we're called with either no lock
- * or with the write lock. If we're not syncing,
- * we're always called with the read lock held.
- */
- boolean_t need_lock =
- !RW_WRITE_HELD(&dp->dp_config_rwlock) &&
- dsl_pool_sync_context(dp);
-
- if (need_lock)
- rw_enter(&dp->dp_config_rwlock, RW_READER);
-
- err = dsl_prop_get_ds(ds,
- "refreservation", sizeof (uint64_t), 1,
- &ds->ds_reserved, NULL);
+ if (err == 0 && !ds->ds_is_snapshot) {
+ err = dsl_prop_get_int_ds(ds,
+ zfs_prop_to_name(ZFS_PROP_REFRESERVATION),
+ &ds->ds_reserved);
if (err == 0) {
- err = dsl_prop_get_ds(ds,
- "refquota", sizeof (uint64_t), 1,
- &ds->ds_quota, NULL);
+ err = dsl_prop_get_int_ds(ds,
+ zfs_prop_to_name(ZFS_PROP_REFQUOTA),
+ &ds->ds_quota);
}
-
- if (need_lock)
- rw_exit(&dp->dp_config_rwlock);
} else {
ds->ds_reserved = ds->ds_quota = 0;
}
- if (err == 0) {
- winner = dmu_buf_set_user_ie(dbuf, ds, &ds->ds_phys,
- dsl_dataset_evict);
+ if (err == 0 && ds->ds_dir->dd_crypto_obj != 0 &&
+ ds->ds_is_snapshot &&
+ zap_contains(mos, dsobj, DS_FIELD_IVSET_GUID) != 0) {
+ dp->dp_spa->spa_errata =
+ ZPOOL_ERRATA_ZOL_8308_ENCRYPTION;
+ }
+
+ dsl_deadlist_open(&ds->ds_deadlist,
+ mos, dsl_dataset_phys(ds)->ds_deadlist_obj);
+ uint64_t remap_deadlist_obj =
+ dsl_dataset_get_remap_deadlist_object(ds);
+ if (remap_deadlist_obj != 0) {
+ dsl_deadlist_open(&ds->ds_remap_deadlist, mos,
+ remap_deadlist_obj);
}
- if (err || winner) {
+
+ dmu_buf_init_user(&ds->ds_dbu, dsl_dataset_evict_sync,
+ dsl_dataset_evict_async, &ds->ds_dbuf);
+ if (err == 0)
+ winner = dmu_buf_set_user_ie(dbuf, &ds->ds_dbu);
+
+ if (err != 0 || winner != NULL) {
bplist_destroy(&ds->ds_pending_deadlist);
dsl_deadlist_close(&ds->ds_deadlist);
+ if (dsl_deadlist_is_open(&ds->ds_remap_deadlist))
+ dsl_deadlist_close(&ds->ds_remap_deadlist);
if (ds->ds_prev)
- dsl_dataset_drop_ref(ds->ds_prev, ds);
- dsl_dir_close(ds->ds_dir, ds);
+ dsl_dataset_rele(ds->ds_prev, ds);
+ dsl_dir_rele(ds->ds_dir, ds);
+ list_destroy(&ds->ds_prop_cbs);
+ list_destroy(&ds->ds_sendstreams);
mutex_destroy(&ds->ds_lock);
- mutex_destroy(&ds->ds_recvlock);
mutex_destroy(&ds->ds_opening_lock);
- rw_destroy(&ds->ds_rwlock);
- cv_destroy(&ds->ds_exclusive_cv);
+ mutex_destroy(&ds->ds_sendstream_lock);
+ mutex_destroy(&ds->ds_remap_deadlist_lock);
+ zfs_refcount_destroy(&ds->ds_longholds);
+ rrw_destroy(&ds->ds_bp_rwlock);
kmem_free(ds, sizeof (dsl_dataset_t));
- if (err) {
+ if (err != 0) {
dmu_buf_rele(dbuf, tag);
return (err);
}
ds = winner;
} else {
ds->ds_fsid_guid =
- unique_insert(ds->ds_phys->ds_fsid_guid);
+ unique_insert(dsl_dataset_phys(ds)->ds_fsid_guid);
+ if (ds->ds_fsid_guid !=
+ dsl_dataset_phys(ds)->ds_fsid_guid) {
+ zfs_dbgmsg("ds_fsid_guid changed from "
+ "%llx to %llx for pool %s dataset id %llu",
+ (long long)
+ dsl_dataset_phys(ds)->ds_fsid_guid,
+ (long long)ds->ds_fsid_guid,
+ spa_name(dp->dp_spa),
+ dsobj);
+ }
}
}
+
ASSERT3P(ds->ds_dbuf, ==, dbuf);
- ASSERT3P(ds->ds_phys, ==, dbuf->db_data);
- ASSERT(ds->ds_phys->ds_prev_snap_obj != 0 ||
+ ASSERT3P(dsl_dataset_phys(ds), ==, dbuf->db_data);
+ ASSERT(dsl_dataset_phys(ds)->ds_prev_snap_obj != 0 ||
spa_version(dp->dp_spa) < SPA_VERSION_ORIGIN ||
dp->dp_origin_snap == NULL || ds == dp->dp_origin_snap);
- mutex_enter(&ds->ds_lock);
- if (!dsl_pool_sync_context(dp) && DSL_DATASET_IS_DESTROYED(ds)) {
- mutex_exit(&ds->ds_lock);
- dmu_buf_rele(ds->ds_dbuf, tag);
- return (ENOENT);
- }
- mutex_exit(&ds->ds_lock);
*dsp = ds;
+
return (0);
}
-static int
-dsl_dataset_hold_ref(dsl_dataset_t *ds, void *tag)
+int
+dsl_dataset_create_key_mapping(dsl_dataset_t *ds)
{
- dsl_pool_t *dp = ds->ds_dir->dd_pool;
+ dsl_dir_t *dd = ds->ds_dir;
- /*
- * In syncing context we don't want the rwlock lock: there
- * may be an existing writer waiting for sync phase to
- * finish. We don't need to worry about such writers, since
- * sync phase is single-threaded, so the writer can't be
- * doing anything while we are active.
- */
- if (dsl_pool_sync_context(dp)) {
- ASSERT(!DSL_DATASET_IS_DESTROYED(ds));
+ if (dd->dd_crypto_obj == 0)
return (0);
- }
- /*
- * Normal users will hold the ds_rwlock as a READER until they
- * are finished (i.e., call dsl_dataset_rele()). "Owners" will
- * drop their READER lock after they set the ds_owner field.
- *
- * If the dataset is being destroyed, the destroy thread will
- * obtain a WRITER lock for exclusive access after it's done its
- * open-context work and then change the ds_owner to
- * dsl_reaper once destruction is assured. So threads
- * may block here temporarily, until the "destructability" of
- * the dataset is determined.
- */
- ASSERT(!RW_WRITE_HELD(&dp->dp_config_rwlock));
- mutex_enter(&ds->ds_lock);
- while (!rw_tryenter(&ds->ds_rwlock, RW_READER)) {
- rw_exit(&dp->dp_config_rwlock);
- cv_wait(&ds->ds_exclusive_cv, &ds->ds_lock);
- if (DSL_DATASET_IS_DESTROYED(ds)) {
- mutex_exit(&ds->ds_lock);
- dsl_dataset_drop_ref(ds, tag);
- rw_enter(&dp->dp_config_rwlock, RW_READER);
- return (ENOENT);
- }
- /*
- * The dp_config_rwlock lives above the ds_lock. And
- * we need to check DSL_DATASET_IS_DESTROYED() while
- * holding the ds_lock, so we have to drop and reacquire
- * the ds_lock here.
- */
- mutex_exit(&ds->ds_lock);
- rw_enter(&dp->dp_config_rwlock, RW_READER);
- mutex_enter(&ds->ds_lock);
- }
- mutex_exit(&ds->ds_lock);
- return (0);
+ return (spa_keystore_create_mapping(dd->dd_pool->dp_spa,
+ ds, ds, &ds->ds_key_mapping));
}
int
-dsl_dataset_hold_obj(dsl_pool_t *dp, uint64_t dsobj, void *tag,
- dsl_dataset_t **dsp)
+dsl_dataset_hold_obj_flags(dsl_pool_t *dp, uint64_t dsobj,
+ ds_hold_flags_t flags, void *tag, dsl_dataset_t **dsp)
{
- int err = dsl_dataset_get_ref(dp, dsobj, tag, dsp);
+ int err;
- if (err)
+ err = dsl_dataset_hold_obj(dp, dsobj, tag, dsp);
+ if (err != 0)
return (err);
- return (dsl_dataset_hold_ref(*dsp, tag));
-}
-int
-dsl_dataset_own_obj(dsl_pool_t *dp, uint64_t dsobj, boolean_t inconsistentok,
- void *tag, dsl_dataset_t **dsp)
-{
- int err = dsl_dataset_hold_obj(dp, dsobj, tag, dsp);
- if (err)
- return (err);
- if (!dsl_dataset_tryown(*dsp, inconsistentok, tag)) {
- dsl_dataset_rele(*dsp, tag);
- *dsp = NULL;
- return (EBUSY);
+ ASSERT3P(*dsp, !=, NULL);
+
+ if (flags & DS_HOLD_FLAG_DECRYPT) {
+ err = dsl_dataset_create_key_mapping(*dsp);
+ if (err != 0)
+ dsl_dataset_rele(*dsp, tag);
}
- return (0);
+
+ return (err);
}
int
-dsl_dataset_hold(const char *name, void *tag, dsl_dataset_t **dsp)
+dsl_dataset_hold_flags(dsl_pool_t *dp, const char *name, ds_hold_flags_t flags,
+ void *tag, dsl_dataset_t **dsp)
{
dsl_dir_t *dd;
- dsl_pool_t *dp;
const char *snapname;
uint64_t obj;
int err = 0;
+ dsl_dataset_t *ds;
- err = dsl_dir_open_spa(NULL, name, FTAG, &dd, &snapname);
- if (err)
+ err = dsl_dir_hold(dp, name, FTAG, &dd, &snapname);
+ if (err != 0)
return (err);
- dp = dd->dd_pool;
- obj = dd->dd_phys->dd_head_dataset_obj;
- rw_enter(&dp->dp_config_rwlock, RW_READER);
- if (obj)
- err = dsl_dataset_get_ref(dp, obj, tag, dsp);
+ ASSERT(dsl_pool_config_held(dp));
+ obj = dsl_dir_phys(dd)->dd_head_dataset_obj;
+ if (obj != 0)
+ err = dsl_dataset_hold_obj_flags(dp, obj, flags, tag, &ds);
else
- err = ENOENT;
- if (err)
- goto out;
-
- err = dsl_dataset_hold_ref(*dsp, tag);
+ err = SET_ERROR(ENOENT);
/* we may be looking for a snapshot */
if (err == 0 && snapname != NULL) {
- dsl_dataset_t *ds = NULL;
+ dsl_dataset_t *snap_ds;
if (*snapname++ != '@') {
- dsl_dataset_rele(*dsp, tag);
- err = ENOENT;
- goto out;
+ dsl_dataset_rele_flags(ds, flags, tag);
+ dsl_dir_rele(dd, FTAG);
+ return (SET_ERROR(ENOENT));
}
dprintf("looking for snapshot '%s'\n", snapname);
- err = dsl_dataset_snap_lookup(*dsp, snapname, &obj);
- if (err == 0)
- err = dsl_dataset_get_ref(dp, obj, tag, &ds);
- dsl_dataset_rele(*dsp, tag);
-
- ASSERT3U((err == 0), ==, (ds != NULL));
-
- if (ds) {
- mutex_enter(&ds->ds_lock);
- if (ds->ds_snapname[0] == 0)
- (void) strlcpy(ds->ds_snapname, snapname,
- sizeof (ds->ds_snapname));
- mutex_exit(&ds->ds_lock);
- err = dsl_dataset_hold_ref(ds, tag);
- *dsp = err ? NULL : ds;
+ err = dsl_dataset_snap_lookup(ds, snapname, &obj);
+ if (err == 0) {
+ err = dsl_dataset_hold_obj_flags(dp, obj, flags, tag,
+ &snap_ds);
+ }
+ dsl_dataset_rele_flags(ds, flags, tag);
+
+ if (err == 0) {
+ mutex_enter(&snap_ds->ds_lock);
+ if (snap_ds->ds_snapname[0] == 0)
+ (void) strlcpy(snap_ds->ds_snapname, snapname,
+ sizeof (snap_ds->ds_snapname));
+ mutex_exit(&snap_ds->ds_lock);
+ ds = snap_ds;
}
}
-out:
- rw_exit(&dp->dp_config_rwlock);
- dsl_dir_close(dd, FTAG);
+ if (err == 0)
+ *dsp = ds;
+ dsl_dir_rele(dd, FTAG);
return (err);
}
int
-dsl_dataset_own(const char *name, boolean_t inconsistentok,
+dsl_dataset_hold(dsl_pool_t *dp, const char *name, void *tag,
+ dsl_dataset_t **dsp)
+{
+ return (dsl_dataset_hold_flags(dp, name, 0, tag, dsp));
+}
+
+int
+dsl_dataset_own_obj(dsl_pool_t *dp, uint64_t dsobj, ds_hold_flags_t flags,
+ void *tag, dsl_dataset_t **dsp)
+{
+ int err = dsl_dataset_hold_obj_flags(dp, dsobj, flags, tag, dsp);
+ if (err != 0)
+ return (err);
+ if (!dsl_dataset_tryown(*dsp, tag)) {
+ dsl_dataset_rele_flags(*dsp, flags, tag);
+ *dsp = NULL;
+ return (SET_ERROR(EBUSY));
+ }
+ return (0);
+}
+
+int
+dsl_dataset_own(dsl_pool_t *dp, const char *name, ds_hold_flags_t flags,
void *tag, dsl_dataset_t **dsp)
{
- int err = dsl_dataset_hold(name, tag, dsp);
- if (err)
+ int err = dsl_dataset_hold_flags(dp, name, flags, tag, dsp);
+ if (err != 0)
return (err);
- if (!dsl_dataset_tryown(*dsp, inconsistentok, tag)) {
- dsl_dataset_rele(*dsp, tag);
- return (EBUSY);
+ if (!dsl_dataset_tryown(*dsp, tag)) {
+ dsl_dataset_rele_flags(*dsp, flags, tag);
+ return (SET_ERROR(EBUSY));
}
return (0);
}
+/*
+ * See the comment above dsl_pool_hold() for details. In summary, a long
+ * hold is used to prevent destruction of a dataset while the pool hold
+ * is dropped, allowing other concurrent operations (e.g. spa_sync()).
+ *
+ * The dataset and pool must be held when this function is called. After it
+ * is called, the pool hold may be released while the dataset is still held
+ * and accessed.
+ */
+void
+dsl_dataset_long_hold(dsl_dataset_t *ds, void *tag)
+{
+ ASSERT(dsl_pool_config_held(ds->ds_dir->dd_pool));
+ (void) zfs_refcount_add(&ds->ds_longholds, tag);
+}
+
+void
+dsl_dataset_long_rele(dsl_dataset_t *ds, void *tag)
+{
+ (void) zfs_refcount_remove(&ds->ds_longholds, tag);
+}
+
+/* Return B_TRUE if there are any long holds on this dataset. */
+boolean_t
+dsl_dataset_long_held(dsl_dataset_t *ds)
+{
+ return (!zfs_refcount_is_zero(&ds->ds_longholds));
+}
+
void
dsl_dataset_name(dsl_dataset_t *ds, char *name)
{
(void) strcpy(name, "mos");
} else {
dsl_dir_name(ds->ds_dir, name);
- VERIFY(0 == dsl_dataset_get_snapname(ds));
+ VERIFY0(dsl_dataset_get_snapname(ds));
if (ds->ds_snapname[0]) {
- (void) strcat(name, "@");
+ VERIFY3U(strlcat(name, "@", ZFS_MAX_DATASET_NAME_LEN),
+ <, ZFS_MAX_DATASET_NAME_LEN);
/*
* We use a "recursive" mutex so that we
* can call dprintf_ds() with ds_lock held.
*/
if (!MUTEX_HELD(&ds->ds_lock)) {
mutex_enter(&ds->ds_lock);
- (void) strcat(name, ds->ds_snapname);
+ VERIFY3U(strlcat(name, ds->ds_snapname,
+ ZFS_MAX_DATASET_NAME_LEN), <,
+ ZFS_MAX_DATASET_NAME_LEN);
mutex_exit(&ds->ds_lock);
} else {
- (void) strcat(name, ds->ds_snapname);
+ VERIFY3U(strlcat(name, ds->ds_snapname,
+ ZFS_MAX_DATASET_NAME_LEN), <,
+ ZFS_MAX_DATASET_NAME_LEN);
}
}
}
}
-static int
+int
dsl_dataset_namelen(dsl_dataset_t *ds)
{
- int result;
-
- if (ds == NULL) {
- result = 3; /* "mos" */
- } else {
- result = dsl_dir_namelen(ds->ds_dir);
- VERIFY(0 == dsl_dataset_get_snapname(ds));
- if (ds->ds_snapname[0]) {
- ++result; /* adding one for the @-sign */
- if (!MUTEX_HELD(&ds->ds_lock)) {
- mutex_enter(&ds->ds_lock);
- result += strlen(ds->ds_snapname);
- mutex_exit(&ds->ds_lock);
- } else {
- result += strlen(ds->ds_snapname);
- }
- }
- }
-
- return (result);
+ VERIFY0(dsl_dataset_get_snapname(ds));
+ mutex_enter(&ds->ds_lock);
+ int len = strlen(ds->ds_snapname);
+ mutex_exit(&ds->ds_lock);
+ /* add '@' if ds is a snap */
+ if (len > 0)
+ len++;
+ len += dsl_dir_namelen(ds->ds_dir);
+ return (len);
}
void
-dsl_dataset_drop_ref(dsl_dataset_t *ds, void *tag)
+dsl_dataset_rele(dsl_dataset_t *ds, void *tag)
{
dmu_buf_rele(ds->ds_dbuf, tag);
}
void
-dsl_dataset_rele(dsl_dataset_t *ds, void *tag)
+dsl_dataset_remove_key_mapping(dsl_dataset_t *ds)
{
- if (!dsl_pool_sync_context(ds->ds_dir->dd_pool)) {
- rw_exit(&ds->ds_rwlock);
- }
- dsl_dataset_drop_ref(ds, tag);
+ dsl_dir_t *dd = ds->ds_dir;
+
+ if (dd == NULL || dd->dd_crypto_obj == 0)
+ return;
+
+ (void) spa_keystore_remove_mapping(dd->dd_pool->dp_spa,
+ ds->ds_object, ds);
}
void
-dsl_dataset_disown(dsl_dataset_t *ds, void *tag)
+dsl_dataset_rele_flags(dsl_dataset_t *ds, ds_hold_flags_t flags, void *tag)
{
- ASSERT((ds->ds_owner == tag && ds->ds_dbuf) ||
- (DSL_DATASET_IS_DESTROYED(ds) && ds->ds_dbuf == NULL));
+ if (flags & DS_HOLD_FLAG_DECRYPT)
+ dsl_dataset_remove_key_mapping(ds);
- mutex_enter(&ds->ds_lock);
- ds->ds_owner = NULL;
- if (RW_WRITE_HELD(&ds->ds_rwlock)) {
- rw_exit(&ds->ds_rwlock);
- cv_broadcast(&ds->ds_exclusive_cv);
- }
- mutex_exit(&ds->ds_lock);
- if (ds->ds_dbuf)
- dsl_dataset_drop_ref(ds, tag);
- else
- dsl_dataset_evict(NULL, ds);
+ dsl_dataset_rele(ds, tag);
+}
+
+void
+dsl_dataset_disown(dsl_dataset_t *ds, ds_hold_flags_t flags, void *tag)
+{
+ ASSERT3P(ds->ds_owner, ==, tag);
+ ASSERT(ds->ds_dbuf != NULL);
+
+ mutex_enter(&ds->ds_lock);
+ ds->ds_owner = NULL;
+ mutex_exit(&ds->ds_lock);
+ dsl_dataset_long_rele(ds, tag);
+ dsl_dataset_rele_flags(ds, flags, tag);
}
boolean_t
-dsl_dataset_tryown(dsl_dataset_t *ds, boolean_t inconsistentok, void *tag)
+dsl_dataset_tryown(dsl_dataset_t *ds, void *tag)
{
boolean_t gotit = FALSE;
+ ASSERT(dsl_pool_config_held(ds->ds_dir->dd_pool));
mutex_enter(&ds->ds_lock);
- if (ds->ds_owner == NULL &&
- (!DS_IS_INCONSISTENT(ds) || inconsistentok)) {
+ if (ds->ds_owner == NULL && !DS_IS_INCONSISTENT(ds)) {
ds->ds_owner = tag;
- if (!dsl_pool_sync_context(ds->ds_dir->dd_pool))
- rw_exit(&ds->ds_rwlock);
+ dsl_dataset_long_hold(ds, tag);
gotit = TRUE;
}
mutex_exit(&ds->ds_lock);
return (gotit);
}
+boolean_t
+dsl_dataset_has_owner(dsl_dataset_t *ds)
+{
+ boolean_t rv;
+ mutex_enter(&ds->ds_lock);
+ rv = (ds->ds_owner != NULL);
+ mutex_exit(&ds->ds_lock);
+ return (rv);
+}
+
+static boolean_t
+zfeature_active(spa_feature_t f, void *arg)
+{
+ switch (spa_feature_table[f].fi_type) {
+ case ZFEATURE_TYPE_BOOLEAN: {
+ boolean_t val = (boolean_t)arg;
+ ASSERT(val == B_FALSE || val == B_TRUE);
+ return (val);
+ }
+ case ZFEATURE_TYPE_UINT64_ARRAY:
+ /*
+ * In this case, arg is a uint64_t array. The feature is active
+ * if the array is non-null.
+ */
+ return (arg != NULL);
+ default:
+ panic("Invalid zfeature type %d", spa_feature_table[f].fi_type);
+ return (B_FALSE);
+ }
+}
+
+boolean_t
+dsl_dataset_feature_is_active(dsl_dataset_t *ds, spa_feature_t f)
+{
+ return (zfeature_active(f, ds->ds_feature[f]));
+}
+
+/*
+ * The buffers passed out by this function are references to internal buffers;
+ * they should not be freed by callers of this function, and they should not be
+ * used after the dataset has been released.
+ */
+boolean_t
+dsl_dataset_get_uint64_array_feature(dsl_dataset_t *ds, spa_feature_t f,
+ uint64_t *outlength, uint64_t **outp)
+{
+ VERIFY(spa_feature_table[f].fi_type & ZFEATURE_TYPE_UINT64_ARRAY);
+ if (!dsl_dataset_feature_is_active(ds, f)) {
+ return (B_FALSE);
+ }
+ struct feature_type_uint64_array_arg *ftuaa = ds->ds_feature[f];
+ *outp = ftuaa->array;
+ *outlength = ftuaa->length;
+ return (B_TRUE);
+}
+
+void
+dsl_dataset_activate_feature(uint64_t dsobj, spa_feature_t f, void *arg,
+ dmu_tx_t *tx)
+{
+ spa_t *spa = dmu_tx_pool(tx)->dp_spa;
+ objset_t *mos = dmu_tx_pool(tx)->dp_meta_objset;
+ uint64_t zero = 0;
+
+ VERIFY(spa_feature_table[f].fi_flags & ZFEATURE_FLAG_PER_DATASET);
+
+ spa_feature_incr(spa, f, tx);
+ dmu_object_zapify(mos, dsobj, DMU_OT_DSL_DATASET, tx);
+
+ switch (spa_feature_table[f].fi_type) {
+ case ZFEATURE_TYPE_BOOLEAN:
+ ASSERT3S((boolean_t)arg, ==, B_TRUE);
+ VERIFY0(zap_add(mos, dsobj, spa_feature_table[f].fi_guid,
+ sizeof (zero), 1, &zero, tx));
+ break;
+ case ZFEATURE_TYPE_UINT64_ARRAY:
+ {
+ struct feature_type_uint64_array_arg *ftuaa = arg;
+ VERIFY0(zap_add(mos, dsobj, spa_feature_table[f].fi_guid,
+ sizeof (uint64_t), ftuaa->length, ftuaa->array, tx));
+ break;
+ }
+ default:
+ panic("Invalid zfeature type %d", spa_feature_table[f].fi_type);
+ }
+}
+
+void
+dsl_dataset_deactivate_feature_impl(dsl_dataset_t *ds, spa_feature_t f,
+ dmu_tx_t *tx)
+{
+ spa_t *spa = dmu_tx_pool(tx)->dp_spa;
+ objset_t *mos = dmu_tx_pool(tx)->dp_meta_objset;
+ uint64_t dsobj = ds->ds_object;
+
+ VERIFY(spa_feature_table[f].fi_flags & ZFEATURE_FLAG_PER_DATASET);
+
+ VERIFY0(zap_remove(mos, dsobj, spa_feature_table[f].fi_guid, tx));
+ spa_feature_decr(spa, f, tx);
+ ds->ds_feature[f] = NULL;
+}
+
void
-dsl_dataset_make_exclusive(dsl_dataset_t *ds, void *owner)
+dsl_dataset_deactivate_feature(dsl_dataset_t *ds, spa_feature_t f, dmu_tx_t *tx)
{
- ASSERT3P(owner, ==, ds->ds_owner);
- if (!RW_WRITE_HELD(&ds->ds_rwlock))
- rw_enter(&ds->ds_rwlock, RW_WRITER);
+ unload_zfeature(ds, f);
+ dsl_dataset_deactivate_feature_impl(ds, f, tx);
}
uint64_t
dsl_dataset_create_sync_dd(dsl_dir_t *dd, dsl_dataset_t *origin,
- uint64_t flags, dmu_tx_t *tx)
+ dsl_crypto_params_t *dcp, uint64_t flags, dmu_tx_t *tx)
{
dsl_pool_t *dp = dd->dd_pool;
dmu_buf_t *dbuf;
origin = dp->dp_origin_snap;
ASSERT(origin == NULL || origin->ds_dir->dd_pool == dp);
- ASSERT(origin == NULL || origin->ds_phys->ds_num_children > 0);
+ ASSERT(origin == NULL || dsl_dataset_phys(origin)->ds_num_children > 0);
ASSERT(dmu_tx_is_syncing(tx));
- ASSERT(dd->dd_phys->dd_head_dataset_obj == 0);
+ ASSERT(dsl_dir_phys(dd)->dd_head_dataset_obj == 0);
dsobj = dmu_object_alloc(mos, DMU_OT_DSL_DATASET, 0,
DMU_OT_DSL_DATASET, sizeof (dsl_dataset_phys_t), tx);
- VERIFY(0 == dmu_bonus_hold(mos, dsobj, FTAG, &dbuf));
+ VERIFY0(dmu_bonus_hold(mos, dsobj, FTAG, &dbuf));
dmu_buf_will_dirty(dbuf, tx);
dsphys = dbuf->db_data;
bzero(dsphys, sizeof (dsl_dataset_phys_t));
if (origin == NULL) {
dsphys->ds_deadlist_obj = dsl_deadlist_alloc(mos, tx);
} else {
- dsl_dataset_t *ohds;
+ dsl_dataset_t *ohds; /* head of the origin snapshot */
dsphys->ds_prev_snap_obj = origin->ds_object;
dsphys->ds_prev_snap_txg =
- origin->ds_phys->ds_creation_txg;
+ dsl_dataset_phys(origin)->ds_creation_txg;
dsphys->ds_referenced_bytes =
- origin->ds_phys->ds_referenced_bytes;
+ dsl_dataset_phys(origin)->ds_referenced_bytes;
dsphys->ds_compressed_bytes =
- origin->ds_phys->ds_compressed_bytes;
+ dsl_dataset_phys(origin)->ds_compressed_bytes;
dsphys->ds_uncompressed_bytes =
- origin->ds_phys->ds_uncompressed_bytes;
- dsphys->ds_bp = origin->ds_phys->ds_bp;
- dsphys->ds_flags |= origin->ds_phys->ds_flags;
+ dsl_dataset_phys(origin)->ds_uncompressed_bytes;
+ rrw_enter(&origin->ds_bp_rwlock, RW_READER, FTAG);
+ dsphys->ds_bp = dsl_dataset_phys(origin)->ds_bp;
+ rrw_exit(&origin->ds_bp_rwlock, FTAG);
+
+ /*
+ * Inherit flags that describe the dataset's contents
+ * (INCONSISTENT) or properties (Case Insensitive).
+ */
+ dsphys->ds_flags |= dsl_dataset_phys(origin)->ds_flags &
+ (DS_FLAG_INCONSISTENT | DS_FLAG_CI_DATASET);
+
+ for (spa_feature_t f = 0; f < SPA_FEATURES; f++) {
+ if (zfeature_active(f, origin->ds_feature[f])) {
+ dsl_dataset_activate_feature(dsobj, f,
+ origin->ds_feature[f], tx);
+ }
+ }
dmu_buf_will_dirty(origin->ds_dbuf, tx);
- origin->ds_phys->ds_num_children++;
+ dsl_dataset_phys(origin)->ds_num_children++;
- VERIFY3U(0, ==, dsl_dataset_hold_obj(dp,
- origin->ds_dir->dd_phys->dd_head_dataset_obj, FTAG, &ohds));
+ VERIFY0(dsl_dataset_hold_obj(dp,
+ dsl_dir_phys(origin->ds_dir)->dd_head_dataset_obj,
+ FTAG, &ohds));
dsphys->ds_deadlist_obj = dsl_deadlist_clone(&ohds->ds_deadlist,
dsphys->ds_prev_snap_txg, dsphys->ds_prev_snap_obj, tx);
dsl_dataset_rele(ohds, FTAG);
if (spa_version(dp->dp_spa) >= SPA_VERSION_NEXT_CLONES) {
- if (origin->ds_phys->ds_next_clones_obj == 0) {
- origin->ds_phys->ds_next_clones_obj =
+ if (dsl_dataset_phys(origin)->ds_next_clones_obj == 0) {
+ dsl_dataset_phys(origin)->ds_next_clones_obj =
zap_create(mos,
DMU_OT_NEXT_CLONES, DMU_OT_NONE, 0, tx);
}
- VERIFY(0 == zap_add_int(mos,
- origin->ds_phys->ds_next_clones_obj,
+ VERIFY0(zap_add_int(mos,
+ dsl_dataset_phys(origin)->ds_next_clones_obj,
dsobj, tx));
}
dmu_buf_will_dirty(dd->dd_dbuf, tx);
- dd->dd_phys->dd_origin_obj = origin->ds_object;
+ dsl_dir_phys(dd)->dd_origin_obj = origin->ds_object;
if (spa_version(dp->dp_spa) >= SPA_VERSION_DIR_CLONES) {
- if (origin->ds_dir->dd_phys->dd_clones == 0) {
+ if (dsl_dir_phys(origin->ds_dir)->dd_clones == 0) {
dmu_buf_will_dirty(origin->ds_dir->dd_dbuf, tx);
- origin->ds_dir->dd_phys->dd_clones =
+ dsl_dir_phys(origin->ds_dir)->dd_clones =
zap_create(mos,
DMU_OT_DSL_CLONES, DMU_OT_NONE, 0, tx);
}
- VERIFY3U(0, ==, zap_add_int(mos,
- origin->ds_dir->dd_phys->dd_clones, dsobj, tx));
+ VERIFY0(zap_add_int(mos,
+ dsl_dir_phys(origin->ds_dir)->dd_clones,
+ dsobj, tx));
}
}
+ /* handle encryption */
+ dsl_dataset_create_crypt_sync(dsobj, dd, origin, dcp, tx);
+
if (spa_version(dp->dp_spa) >= SPA_VERSION_UNIQUE_ACCURATE)
dsphys->ds_flags |= DS_FLAG_UNIQUE_ACCURATE;
dmu_buf_rele(dbuf, FTAG);
dmu_buf_will_dirty(dd->dd_dbuf, tx);
- dd->dd_phys->dd_head_dataset_obj = dsobj;
+ dsl_dir_phys(dd)->dd_head_dataset_obj = dsobj;
return (dsobj);
}
+static void
+dsl_dataset_zero_zil(dsl_dataset_t *ds, dmu_tx_t *tx)
+{
+ objset_t *os;
+
+ VERIFY0(dmu_objset_from_ds(ds, &os));
+ if (bcmp(&os->os_zil_header, &zero_zil, sizeof (zero_zil)) != 0) {
+ dsl_pool_t *dp = ds->ds_dir->dd_pool;
+ zio_t *zio;
+
+ bzero(&os->os_zil_header, sizeof (os->os_zil_header));
+ if (os->os_encrypted)
+ os->os_next_write_raw[tx->tx_txg & TXG_MASK] = B_TRUE;
+
+ zio = zio_root(dp->dp_spa, NULL, NULL, ZIO_FLAG_MUSTSUCCEED);
+ dsl_dataset_sync(ds, zio, tx);
+ VERIFY0(zio_wait(zio));
+
+ /* dsl_dataset_sync_done will drop this reference. */
+ dmu_buf_add_ref(ds->ds_dbuf, ds);
+ dsl_dataset_sync_done(ds, tx);
+ }
+}
+
uint64_t
dsl_dataset_create_sync(dsl_dir_t *pdd, const char *lastname,
- dsl_dataset_t *origin, uint64_t flags, cred_t *cr, dmu_tx_t *tx)
+ dsl_dataset_t *origin, uint64_t flags, cred_t *cr,
+ dsl_crypto_params_t *dcp, dmu_tx_t *tx)
{
dsl_pool_t *dp = pdd->dd_pool;
uint64_t dsobj, ddobj;
dsl_dir_t *dd;
+ ASSERT(dmu_tx_is_syncing(tx));
ASSERT(lastname[0] != '@');
ddobj = dsl_dir_create_sync(dp, pdd, lastname, tx);
- VERIFY(0 == dsl_dir_open_obj(dp, ddobj, lastname, FTAG, &dd));
+ VERIFY0(dsl_dir_hold_obj(dp, ddobj, lastname, FTAG, &dd));
- dsobj = dsl_dataset_create_sync_dd(dd, origin, flags, tx);
+ dsobj = dsl_dataset_create_sync_dd(dd, origin, dcp,
+ flags & ~DS_CREATE_FLAG_NODIRTY, tx);
dsl_deleg_set_create_perms(dd, tx, cr);
- dsl_dir_close(dd, FTAG);
+ /*
+ * Since we're creating a new node we know it's a leaf, so we can
+ * initialize the counts if the limit feature is active.
+ */
+ if (spa_feature_is_active(dp->dp_spa, SPA_FEATURE_FS_SS_LIMIT)) {
+ uint64_t cnt = 0;
+ objset_t *os = dd->dd_pool->dp_meta_objset;
+
+ dsl_dir_zapify(dd, tx);
+ VERIFY0(zap_add(os, dd->dd_object, DD_FIELD_FILESYSTEM_COUNT,
+ sizeof (cnt), 1, &cnt, tx));
+ VERIFY0(zap_add(os, dd->dd_object, DD_FIELD_SNAPSHOT_COUNT,
+ sizeof (cnt), 1, &cnt, tx));
+ }
+
+ dsl_dir_rele(dd, FTAG);
/*
* If we are creating a clone, make sure we zero out any stale
* data from the origin snapshots zil header.
*/
- if (origin != NULL) {
+ if (origin != NULL && !(flags & DS_CREATE_FLAG_NODIRTY)) {
dsl_dataset_t *ds;
- objset_t *os;
- VERIFY3U(0, ==, dsl_dataset_hold_obj(dp, dsobj, FTAG, &ds));
- VERIFY3U(0, ==, dmu_objset_from_ds(ds, &os));
- bzero(&os->os_zil_header, sizeof (os->os_zil_header));
- dsl_dataset_dirty(ds, tx);
+ VERIFY0(dsl_dataset_hold_obj(dp, dsobj, FTAG, &ds));
+ dsl_dataset_zero_zil(ds, tx);
dsl_dataset_rele(ds, FTAG);
}
}
/*
- * The snapshots must all be in the same pool.
+ * The unique space in the head dataset can be calculated by subtracting
+ * the space used in the most recent snapshot, that is still being used
+ * in this file system, from the space currently in use. To figure out
+ * the space in the most recent snapshot still in use, we need to take
+ * the total space used in the snapshot and subtract out the space that
+ * has been freed up since the snapshot was taken.
*/
-int
-dmu_snapshots_destroy_nvl(nvlist_t *snaps, boolean_t defer, char *failed)
+void
+dsl_dataset_recalc_head_uniq(dsl_dataset_t *ds)
{
- int err;
- dsl_sync_task_t *dst;
- spa_t *spa;
- nvpair_t *pair;
- dsl_sync_task_group_t *dstg;
-
- pair = nvlist_next_nvpair(snaps, NULL);
- if (pair == NULL)
- return (0);
-
- err = spa_open(nvpair_name(pair), &spa, FTAG);
- if (err)
- return (err);
- dstg = dsl_sync_task_group_create(spa_get_dsl(spa));
-
- for (pair = nvlist_next_nvpair(snaps, NULL); pair != NULL;
- pair = nvlist_next_nvpair(snaps, pair)) {
- dsl_dataset_t *ds;
-
- err = dsl_dataset_own(nvpair_name(pair), B_TRUE, dstg, &ds);
- if (err == 0) {
- struct dsl_ds_destroyarg *dsda;
-
- dsl_dataset_make_exclusive(ds, dstg);
- dsda = kmem_zalloc(sizeof (struct dsl_ds_destroyarg),
- KM_SLEEP);
- dsda->ds = ds;
- dsda->defer = defer;
- dsl_sync_task_create(dstg, dsl_dataset_destroy_check,
- dsl_dataset_destroy_sync, dsda, dstg, 0);
- } else if (err == ENOENT) {
- err = 0;
- } else {
- (void) strcpy(failed, nvpair_name(pair));
- break;
- }
- }
-
- if (err == 0)
- err = dsl_sync_task_group_wait(dstg);
-
- for (dst = list_head(&dstg->dstg_tasks); dst;
- dst = list_next(&dstg->dstg_tasks, dst)) {
- struct dsl_ds_destroyarg *dsda = dst->dst_arg1;
- dsl_dataset_t *ds = dsda->ds;
-
- /*
- * Return the file system name that triggered the error
- */
- if (dst->dst_err) {
- dsl_dataset_name(ds, failed);
- }
- ASSERT3P(dsda->rm_origin, ==, NULL);
- dsl_dataset_disown(ds, dstg);
- kmem_free(dsda, sizeof (struct dsl_ds_destroyarg));
- }
-
- dsl_sync_task_group_destroy(dstg);
- spa_close(spa, FTAG);
- return (err);
-
-}
+ uint64_t mrs_used;
+ uint64_t dlused, dlcomp, dluncomp;
-static boolean_t
-dsl_dataset_might_destroy_origin(dsl_dataset_t *ds)
-{
- boolean_t might_destroy = B_FALSE;
+ ASSERT(!ds->ds_is_snapshot);
- mutex_enter(&ds->ds_lock);
- if (ds->ds_phys->ds_num_children == 2 && ds->ds_userrefs == 0 &&
- DS_IS_DEFER_DESTROY(ds))
- might_destroy = B_TRUE;
- mutex_exit(&ds->ds_lock);
+ if (dsl_dataset_phys(ds)->ds_prev_snap_obj != 0)
+ mrs_used = dsl_dataset_phys(ds->ds_prev)->ds_referenced_bytes;
+ else
+ mrs_used = 0;
- return (might_destroy);
-}
+ dsl_deadlist_space(&ds->ds_deadlist, &dlused, &dlcomp, &dluncomp);
-/*
- * If we're removing a clone, and these three conditions are true:
- * 1) the clone's origin has no other children
- * 2) the clone's origin has no user references
- * 3) the clone's origin has been marked for deferred destruction
- * Then, prepare to remove the origin as part of this sync task group.
- */
-static int
-dsl_dataset_origin_rm_prep(struct dsl_ds_destroyarg *dsda, void *tag)
-{
- dsl_dataset_t *ds = dsda->ds;
- dsl_dataset_t *origin = ds->ds_prev;
-
- if (dsl_dataset_might_destroy_origin(origin)) {
- char *name;
- int namelen;
- int error;
-
- namelen = dsl_dataset_namelen(origin) + 1;
- name = kmem_alloc(namelen, KM_SLEEP);
- dsl_dataset_name(origin, name);
-#ifdef _KERNEL
- error = zfs_unmount_snap(name, NULL);
- if (error) {
- kmem_free(name, namelen);
- return (error);
- }
-#endif
- error = dsl_dataset_own(name, B_TRUE, tag, &origin);
- kmem_free(name, namelen);
- if (error)
- return (error);
- dsda->rm_origin = origin;
- dsl_dataset_make_exclusive(origin, tag);
- }
+ ASSERT3U(dlused, <=, mrs_used);
+ dsl_dataset_phys(ds)->ds_unique_bytes =
+ dsl_dataset_phys(ds)->ds_referenced_bytes - (mrs_used - dlused);
- return (0);
+ if (spa_version(ds->ds_dir->dd_pool->dp_spa) >=
+ SPA_VERSION_UNIQUE_ACCURATE)
+ dsl_dataset_phys(ds)->ds_flags |= DS_FLAG_UNIQUE_ACCURATE;
}
-/*
- * ds must be opened as OWNER. On return (whether successful or not),
- * ds will be closed and caller can no longer dereference it.
- */
-int
-dsl_dataset_destroy(dsl_dataset_t *ds, void *tag, boolean_t defer)
+void
+dsl_dataset_remove_from_next_clones(dsl_dataset_t *ds, uint64_t obj,
+ dmu_tx_t *tx)
{
+ objset_t *mos = ds->ds_dir->dd_pool->dp_meta_objset;
+ ASSERTV(uint64_t count);
int err;
- dsl_sync_task_group_t *dstg;
- objset_t *os;
- dsl_dir_t *dd;
- uint64_t obj;
- struct dsl_ds_destroyarg dsda = { 0 };
- dsl_dataset_t *dummy_ds;
-
- dsda.ds = ds;
-
- if (dsl_dataset_is_snapshot(ds)) {
- /* Destroying a snapshot is simpler */
- dsl_dataset_make_exclusive(ds, tag);
-
- dsda.defer = defer;
- err = dsl_sync_task_do(ds->ds_dir->dd_pool,
- dsl_dataset_destroy_check, dsl_dataset_destroy_sync,
- &dsda, tag, 0);
- ASSERT3P(dsda.rm_origin, ==, NULL);
- goto out;
- } else if (defer) {
- err = EINVAL;
- goto out;
- }
- dd = ds->ds_dir;
- dummy_ds = kmem_zalloc(sizeof (dsl_dataset_t), KM_SLEEP);
- dummy_ds->ds_dir = dd;
- dummy_ds->ds_object = ds->ds_object;
-
- if (!spa_feature_is_enabled(dsl_dataset_get_spa(ds),
- &spa_feature_table[SPA_FEATURE_ASYNC_DESTROY])) {
- /*
- * Check for errors and mark this ds as inconsistent, in
- * case we crash while freeing the objects.
- */
- err = dsl_sync_task_do(dd->dd_pool,
- dsl_dataset_destroy_begin_check,
- dsl_dataset_destroy_begin_sync, ds, NULL, 0);
- if (err)
- goto out_free;
-
- err = dmu_objset_from_ds(ds, &os);
- if (err)
- goto out_free;
-
- /*
- * Remove all objects while in the open context so that
- * there is less work to do in the syncing context.
- */
- for (obj = 0; err == 0; err = dmu_object_next(os, &obj, FALSE,
- ds->ds_phys->ds_prev_snap_txg)) {
- /*
- * Ignore errors, if there is not enough disk space
- * we will deal with it in dsl_dataset_destroy_sync().
- */
- (void) dmu_free_object(os, obj);
- }
- if (err != ESRCH)
- goto out_free;
-
- /*
- * Sync out all in-flight IO.
- */
- txg_wait_synced(dd->dd_pool, 0);
-
- /*
- * If we managed to free all the objects in open
- * context, the user space accounting should be zero.
- */
- if (ds->ds_phys->ds_bp.blk_fill == 0 &&
- dmu_objset_userused_enabled(os)) {
- ASSERTV(uint64_t count);
-
- ASSERT(zap_count(os, DMU_USERUSED_OBJECT,
- &count) != 0 || count == 0);
- ASSERT(zap_count(os, DMU_GROUPUSED_OBJECT,
- &count) != 0 || count == 0);
- }
- }
-
- rw_enter(&dd->dd_pool->dp_config_rwlock, RW_READER);
- err = dsl_dir_open_obj(dd->dd_pool, dd->dd_object, NULL, FTAG, &dd);
- rw_exit(&dd->dd_pool->dp_config_rwlock);
-
- if (err)
- goto out_free;
-
- /*
- * Blow away the dsl_dir + head dataset.
- */
- dsl_dataset_make_exclusive(ds, tag);
+ ASSERT(dsl_dataset_phys(ds)->ds_num_children >= 2);
+ err = zap_remove_int(mos, dsl_dataset_phys(ds)->ds_next_clones_obj,
+ obj, tx);
/*
- * If we're removing a clone, we might also need to remove its
- * origin.
+ * The err should not be ENOENT, but a bug in a previous version
+ * of the code could cause upgrade_clones_cb() to not set
+ * ds_next_snap_obj when it should, leading to a missing entry.
+ * If we knew that the pool was created after
+ * SPA_VERSION_NEXT_CLONES, we could assert that it isn't
+ * ENOENT. However, at least we can check that we don't have
+ * too many entries in the next_clones_obj even after failing to
+ * remove this one.
*/
- do {
- dsda.need_prep = B_FALSE;
- if (dsl_dir_is_clone(dd)) {
- err = dsl_dataset_origin_rm_prep(&dsda, tag);
- if (err) {
- dsl_dir_close(dd, FTAG);
- goto out_free;
- }
- }
-
- dstg = dsl_sync_task_group_create(ds->ds_dir->dd_pool);
- dsl_sync_task_create(dstg, dsl_dataset_destroy_check,
- dsl_dataset_destroy_sync, &dsda, tag, 0);
- dsl_sync_task_create(dstg, dsl_dir_destroy_check,
- dsl_dir_destroy_sync, dummy_ds, FTAG, 0);
- err = dsl_sync_task_group_wait(dstg);
- dsl_sync_task_group_destroy(dstg);
-
- /*
- * We could be racing against 'zfs release' or 'zfs destroy -d'
- * on the origin snap, in which case we can get EBUSY if we
- * needed to destroy the origin snap but were not ready to
- * do so.
- */
- if (dsda.need_prep) {
- ASSERT(err == EBUSY);
- ASSERT(dsl_dir_is_clone(dd));
- ASSERT(dsda.rm_origin == NULL);
- }
- } while (dsda.need_prep);
-
- if (dsda.rm_origin != NULL)
- dsl_dataset_disown(dsda.rm_origin, tag);
-
- /* if it is successful, dsl_dir_destroy_sync will close the dd */
- if (err)
- dsl_dir_close(dd, FTAG);
-
-out_free:
- kmem_free(dummy_ds, sizeof (dsl_dataset_t));
-out:
- dsl_dataset_disown(ds, tag);
- return (err);
+ if (err != ENOENT)
+ VERIFY0(err);
+ ASSERT0(zap_count(mos, dsl_dataset_phys(ds)->ds_next_clones_obj,
+ &count));
+ ASSERT3U(count, <=, dsl_dataset_phys(ds)->ds_num_children - 2);
}
+
blkptr_t *
dsl_dataset_get_blkptr(dsl_dataset_t *ds)
{
- return (&ds->ds_phys->ds_bp);
-}
-
-void
-dsl_dataset_set_blkptr(dsl_dataset_t *ds, blkptr_t *bp, dmu_tx_t *tx)
-{
- ASSERT(dmu_tx_is_syncing(tx));
- /* If it's the meta-objset, set dp_meta_rootbp */
- if (ds == NULL) {
- tx->tx_pool->dp_meta_rootbp = *bp;
- } else {
- dmu_buf_will_dirty(ds->ds_dbuf, tx);
- ds->ds_phys->ds_bp = *bp;
- }
+ return (&dsl_dataset_phys(ds)->ds_bp);
}
spa_t *
ASSERT(ds->ds_objset != NULL);
- if (ds->ds_phys->ds_next_snap_obj != 0)
+ if (dsl_dataset_phys(ds)->ds_next_snap_obj != 0)
panic("dirtying snapshot!");
+ /* Must not dirty a dataset in the same txg where it got snapshotted. */
+ ASSERT3U(tx->tx_txg, >, dsl_dataset_phys(ds)->ds_prev_snap_txg);
+
dp = ds->ds_dir->dd_pool;
+ if (txg_list_add(&dp->dp_dirty_datasets, ds, tx->tx_txg)) {
+ objset_t *os = ds->ds_objset;
- if (txg_list_add(&dp->dp_dirty_datasets, ds, tx->tx_txg) == 0) {
/* up the hold count until we can be written out */
dmu_buf_add_ref(ds->ds_dbuf, ds);
- }
-}
-
-boolean_t
-dsl_dataset_is_dirty(dsl_dataset_t *ds)
-{
- int t;
- for (t = 0; t < TXG_SIZE; t++) {
- if (txg_list_member(&ds->ds_dir->dd_pool->dp_dirty_datasets,
- ds, t))
- return (B_TRUE);
+ /* if this dataset is encrypted, grab a reference to the DCK */
+ if (ds->ds_dir->dd_crypto_obj != 0 &&
+ !os->os_raw_receive &&
+ !os->os_next_write_raw[tx->tx_txg & TXG_MASK]) {
+ ASSERT3P(ds->ds_key_mapping, !=, NULL);
+ key_mapping_add_ref(ds->ds_key_mapping, ds);
+ }
}
- return (B_FALSE);
}
-/*
- * The unique space in the head dataset can be calculated by subtracting
- * the space used in the most recent snapshot, that is still being used
- * in this file system, from the space currently in use. To figure out
- * the space in the most recent snapshot still in use, we need to take
- * the total space used in the snapshot and subtract out the space that
- * has been freed up since the snapshot was taken.
- */
-static void
-dsl_dataset_recalc_head_uniq(dsl_dataset_t *ds)
+static int
+dsl_dataset_snapshot_reserve_space(dsl_dataset_t *ds, dmu_tx_t *tx)
{
- uint64_t mrs_used;
- uint64_t dlused, dlcomp, dluncomp;
+ uint64_t asize;
- ASSERT(!dsl_dataset_is_snapshot(ds));
+ if (!dmu_tx_is_syncing(tx))
+ return (0);
- if (ds->ds_phys->ds_prev_snap_obj != 0)
- mrs_used = ds->ds_prev->ds_phys->ds_referenced_bytes;
- else
- mrs_used = 0;
+ /*
+ * If there's an fs-only reservation, any blocks that might become
+ * owned by the snapshot dataset must be accommodated by space
+ * outside of the reservation.
+ */
+ ASSERT(ds->ds_reserved == 0 || DS_UNIQUE_IS_ACCURATE(ds));
+ asize = MIN(dsl_dataset_phys(ds)->ds_unique_bytes, ds->ds_reserved);
+ if (asize > dsl_dir_space_available(ds->ds_dir, NULL, 0, TRUE))
+ return (SET_ERROR(ENOSPC));
- dsl_deadlist_space(&ds->ds_deadlist, &dlused, &dlcomp, &dluncomp);
+ /*
+ * Propagate any reserved space for this snapshot to other
+ * snapshot checks in this sync group.
+ */
+ if (asize > 0)
+ dsl_dir_willuse_space(ds->ds_dir, asize, tx);
- ASSERT3U(dlused, <=, mrs_used);
- ds->ds_phys->ds_unique_bytes =
- ds->ds_phys->ds_referenced_bytes - (mrs_used - dlused);
-
- if (spa_version(ds->ds_dir->dd_pool->dp_spa) >=
- SPA_VERSION_UNIQUE_ACCURATE)
- ds->ds_phys->ds_flags |= DS_FLAG_UNIQUE_ACCURATE;
+ return (0);
}
-struct killarg {
- dsl_dataset_t *ds;
- dmu_tx_t *tx;
-};
-
-/* ARGSUSED */
-static int
-kill_blkptr(spa_t *spa, zilog_t *zilog, const blkptr_t *bp, arc_buf_t *pbuf,
- const zbookmark_t *zb, const dnode_phys_t *dnp, void *arg)
+int
+dsl_dataset_snapshot_check_impl(dsl_dataset_t *ds, const char *snapname,
+ dmu_tx_t *tx, boolean_t recv, uint64_t cnt, cred_t *cr)
{
- struct killarg *ka = arg;
- dmu_tx_t *tx = ka->tx;
+ int error;
+ uint64_t value;
- if (bp == NULL)
- return (0);
+ ds->ds_trysnap_txg = tx->tx_txg;
- if (zb->zb_level == ZB_ZIL_LEVEL) {
- ASSERT(zilog != NULL);
- /*
- * It's a block in the intent log. It has no
- * accounting, so just free it.
- */
- dsl_free(ka->tx->tx_pool, ka->tx->tx_txg, bp);
- } else {
- ASSERT(zilog == NULL);
- ASSERT3U(bp->blk_birth, >, ka->ds->ds_phys->ds_prev_snap_txg);
- (void) dsl_dataset_block_kill(ka->ds, bp, tx, B_FALSE);
- }
+ if (!dmu_tx_is_syncing(tx))
+ return (0);
- return (0);
-}
+ /*
+ * We don't allow multiple snapshots of the same txg. If there
+ * is already one, try again.
+ */
+ if (dsl_dataset_phys(ds)->ds_prev_snap_txg >= tx->tx_txg)
+ return (SET_ERROR(EAGAIN));
-/* ARGSUSED */
-static int
-dsl_dataset_destroy_begin_check(void *arg1, void *arg2, dmu_tx_t *tx)
-{
- dsl_dataset_t *ds = arg1;
- objset_t *mos = ds->ds_dir->dd_pool->dp_meta_objset;
- uint64_t count;
- int err;
+ /*
+ * Check for conflicting snapshot name.
+ */
+ error = dsl_dataset_snap_lookup(ds, snapname, &value);
+ if (error == 0)
+ return (SET_ERROR(EEXIST));
+ if (error != ENOENT)
+ return (error);
/*
- * Can't delete a head dataset if there are snapshots of it.
- * (Except if the only snapshots are from the branch we cloned
- * from.)
+ * We don't allow taking snapshots of inconsistent datasets, such as
+ * those into which we are currently receiving. However, if we are
+ * creating this snapshot as part of a receive, this check will be
+ * executed atomically with respect to the completion of the receive
+ * itself but prior to the clearing of DS_FLAG_INCONSISTENT; in this
+ * case we ignore this, knowing it will be fixed up for us shortly in
+ * dmu_recv_end_sync().
*/
- if (ds->ds_prev != NULL &&
- ds->ds_prev->ds_phys->ds_next_snap_obj == ds->ds_object)
- return (EBUSY);
+ if (!recv && DS_IS_INCONSISTENT(ds))
+ return (SET_ERROR(EBUSY));
/*
- * This is really a dsl_dir thing, but check it here so that
- * we'll be less likely to leave this dataset inconsistent &
- * nearly destroyed.
+ * Skip the check for temporary snapshots or if we have already checked
+ * the counts in dsl_dataset_snapshot_check. This means we really only
+ * check the count here when we're receiving a stream.
*/
- err = zap_count(mos, ds->ds_dir->dd_phys->dd_child_dir_zapobj, &count);
- if (err)
- return (err);
- if (count != 0)
- return (EEXIST);
+ if (cnt != 0 && cr != NULL) {
+ error = dsl_fs_ss_limit_check(ds->ds_dir, cnt,
+ ZFS_PROP_SNAPSHOT_LIMIT, NULL, cr);
+ if (error != 0)
+ return (error);
+ }
+
+ error = dsl_dataset_snapshot_reserve_space(ds, tx);
+ if (error != 0)
+ return (error);
return (0);
}
-/* ARGSUSED */
-static void
-dsl_dataset_destroy_begin_sync(void *arg1, void *arg2, dmu_tx_t *tx)
+int
+dsl_dataset_snapshot_check(void *arg, dmu_tx_t *tx)
{
- dsl_dataset_t *ds = arg1;
- dsl_pool_t *dp = ds->ds_dir->dd_pool;
+ dsl_dataset_snapshot_arg_t *ddsa = arg;
+ dsl_pool_t *dp = dmu_tx_pool(tx);
+ nvpair_t *pair;
+ int rv = 0;
- /* Mark it as inconsistent on-disk, in case we crash */
- dmu_buf_will_dirty(ds->ds_dbuf, tx);
- ds->ds_phys->ds_flags |= DS_FLAG_INCONSISTENT;
+ /*
+ * Pre-compute how many total new snapshots will be created for each
+ * level in the tree and below. This is needed for validating the
+ * snapshot limit when either taking a recursive snapshot or when
+ * taking multiple snapshots.
+ *
+ * The problem is that the counts are not actually adjusted when
+ * we are checking, only when we finally sync. For a single snapshot,
+ * this is easy, the count will increase by 1 at each node up the tree,
+ * but its more complicated for the recursive/multiple snapshot case.
+ *
+ * The dsl_fs_ss_limit_check function does recursively check the count
+ * at each level up the tree but since it is validating each snapshot
+ * independently we need to be sure that we are validating the complete
+ * count for the entire set of snapshots. We do this by rolling up the
+ * counts for each component of the name into an nvlist and then
+ * checking each of those cases with the aggregated count.
+ *
+ * This approach properly handles not only the recursive snapshot
+ * case (where we get all of those on the ddsa_snaps list) but also
+ * the sibling case (e.g. snapshot a/b and a/c so that we will also
+ * validate the limit on 'a' using a count of 2).
+ *
+ * We validate the snapshot names in the third loop and only report
+ * name errors once.
+ */
+ if (dmu_tx_is_syncing(tx)) {
+ char *nm;
+ nvlist_t *cnt_track = NULL;
+ cnt_track = fnvlist_alloc();
+
+ nm = kmem_alloc(MAXPATHLEN, KM_SLEEP);
+
+ /* Rollup aggregated counts into the cnt_track list */
+ for (pair = nvlist_next_nvpair(ddsa->ddsa_snaps, NULL);
+ pair != NULL;
+ pair = nvlist_next_nvpair(ddsa->ddsa_snaps, pair)) {
+ char *pdelim;
+ uint64_t val;
- spa_history_log_internal(LOG_DS_DESTROY_BEGIN, dp->dp_spa, tx,
- "dataset = %llu", ds->ds_object);
-}
+ (void) strlcpy(nm, nvpair_name(pair), MAXPATHLEN);
+ pdelim = strchr(nm, '@');
+ if (pdelim == NULL)
+ continue;
+ *pdelim = '\0';
+
+ do {
+ if (nvlist_lookup_uint64(cnt_track, nm,
+ &val) == 0) {
+ /* update existing entry */
+ fnvlist_add_uint64(cnt_track, nm,
+ val + 1);
+ } else {
+ /* add to list */
+ fnvlist_add_uint64(cnt_track, nm, 1);
+ }
-static int
-dsl_dataset_origin_check(struct dsl_ds_destroyarg *dsda, void *tag,
- dmu_tx_t *tx)
-{
- dsl_dataset_t *ds = dsda->ds;
- dsl_dataset_t *ds_prev = ds->ds_prev;
+ pdelim = strrchr(nm, '/');
+ if (pdelim != NULL)
+ *pdelim = '\0';
+ } while (pdelim != NULL);
+ }
- if (dsl_dataset_might_destroy_origin(ds_prev)) {
- struct dsl_ds_destroyarg ndsda = {0};
+ kmem_free(nm, MAXPATHLEN);
+
+ /* Check aggregated counts at each level */
+ for (pair = nvlist_next_nvpair(cnt_track, NULL);
+ pair != NULL; pair = nvlist_next_nvpair(cnt_track, pair)) {
+ int error = 0;
+ char *name;
+ uint64_t cnt = 0;
+ dsl_dataset_t *ds;
+
+ name = nvpair_name(pair);
+ cnt = fnvpair_value_uint64(pair);
+ ASSERT(cnt > 0);
+
+ error = dsl_dataset_hold(dp, name, FTAG, &ds);
+ if (error == 0) {
+ error = dsl_fs_ss_limit_check(ds->ds_dir, cnt,
+ ZFS_PROP_SNAPSHOT_LIMIT, NULL,
+ ddsa->ddsa_cr);
+ dsl_dataset_rele(ds, FTAG);
+ }
- /*
- * If we're not prepared to remove the origin, don't remove
- * the clone either.
- */
- if (dsda->rm_origin == NULL) {
- dsda->need_prep = B_TRUE;
- return (EBUSY);
+ if (error != 0) {
+ if (ddsa->ddsa_errors != NULL)
+ fnvlist_add_int32(ddsa->ddsa_errors,
+ name, error);
+ rv = error;
+ /* only report one error for this check */
+ break;
+ }
}
-
- ndsda.ds = ds_prev;
- ndsda.is_origin_rm = B_TRUE;
- return (dsl_dataset_destroy_check(&ndsda, tag, tx));
+ nvlist_free(cnt_track);
}
- /*
- * If we're not going to remove the origin after all,
- * undo the open context setup.
- */
- if (dsda->rm_origin != NULL) {
- dsl_dataset_disown(dsda->rm_origin, tag);
- dsda->rm_origin = NULL;
+ for (pair = nvlist_next_nvpair(ddsa->ddsa_snaps, NULL);
+ pair != NULL; pair = nvlist_next_nvpair(ddsa->ddsa_snaps, pair)) {
+ int error = 0;
+ dsl_dataset_t *ds;
+ char *name, *atp = NULL;
+ char dsname[ZFS_MAX_DATASET_NAME_LEN];
+
+ name = nvpair_name(pair);
+ if (strlen(name) >= ZFS_MAX_DATASET_NAME_LEN)
+ error = SET_ERROR(ENAMETOOLONG);
+ if (error == 0) {
+ atp = strchr(name, '@');
+ if (atp == NULL)
+ error = SET_ERROR(EINVAL);
+ if (error == 0)
+ (void) strlcpy(dsname, name, atp - name + 1);
+ }
+ if (error == 0)
+ error = dsl_dataset_hold(dp, dsname, FTAG, &ds);
+ if (error == 0) {
+ /* passing 0/NULL skips dsl_fs_ss_limit_check */
+ error = dsl_dataset_snapshot_check_impl(ds,
+ atp + 1, tx, B_FALSE, 0, NULL);
+ dsl_dataset_rele(ds, FTAG);
+ }
+
+ if (error != 0) {
+ if (ddsa->ddsa_errors != NULL) {
+ fnvlist_add_int32(ddsa->ddsa_errors,
+ name, error);
+ }
+ rv = error;
+ }
}
- return (0);
+ return (rv);
}
-/*
- * If you add new checks here, you may need to add
- * additional checks to the "temporary" case in
- * snapshot_check() in dmu_objset.c.
- */
-/* ARGSUSED */
-int
-dsl_dataset_destroy_check(void *arg1, void *arg2, dmu_tx_t *tx)
+void
+dsl_dataset_snapshot_sync_impl(dsl_dataset_t *ds, const char *snapname,
+ dmu_tx_t *tx)
{
- struct dsl_ds_destroyarg *dsda = arg1;
- dsl_dataset_t *ds = dsda->ds;
+ dsl_pool_t *dp = ds->ds_dir->dd_pool;
+ dmu_buf_t *dbuf;
+ dsl_dataset_phys_t *dsphys;
+ uint64_t dsobj, crtxg;
+ objset_t *mos = dp->dp_meta_objset;
+ ASSERTV(static zil_header_t zero_zil);
+ ASSERTV(objset_t *os);
- /* we have an owner hold, so noone else can destroy us */
- ASSERT(!DSL_DATASET_IS_DESTROYED(ds));
+ ASSERT(RRW_WRITE_HELD(&dp->dp_config_rwlock));
/*
- * Only allow deferred destroy on pools that support it.
- * NOTE: deferred destroy is only supported on snapshots.
+ * If we are on an old pool, the zil must not be active, in which
+ * case it will be zeroed. Usually zil_suspend() accomplishes this.
*/
- if (dsda->defer) {
- if (spa_version(ds->ds_dir->dd_pool->dp_spa) <
- SPA_VERSION_USERREFS)
- return (ENOTSUP);
- ASSERT(dsl_dataset_is_snapshot(ds));
- return (0);
- }
+ ASSERT(spa_version(dmu_tx_pool(tx)->dp_spa) >= SPA_VERSION_FAST_SNAP ||
+ dmu_objset_from_ds(ds, &os) != 0 ||
+ bcmp(&os->os_phys->os_zil_header, &zero_zil,
+ sizeof (zero_zil)) == 0);
+
+ /* Should not snapshot a dirty dataset. */
+ ASSERT(!txg_list_member(&ds->ds_dir->dd_pool->dp_dirty_datasets,
+ ds, tx->tx_txg));
+
+ dsl_fs_ss_count_adjust(ds->ds_dir, 1, DD_FIELD_SNAPSHOT_COUNT, tx);
/*
- * Can't delete a head dataset if there are snapshots of it.
- * (Except if the only snapshots are from the branch we cloned
- * from.)
+ * The origin's ds_creation_txg has to be < TXG_INITIAL
*/
- if (ds->ds_prev != NULL &&
- ds->ds_prev->ds_phys->ds_next_snap_obj == ds->ds_object)
- return (EBUSY);
+ if (strcmp(snapname, ORIGIN_DIR_NAME) == 0)
+ crtxg = 1;
+ else
+ crtxg = tx->tx_txg;
+
+ dsobj = dmu_object_alloc(mos, DMU_OT_DSL_DATASET, 0,
+ DMU_OT_DSL_DATASET, sizeof (dsl_dataset_phys_t), tx);
+ VERIFY0(dmu_bonus_hold(mos, dsobj, FTAG, &dbuf));
+ dmu_buf_will_dirty(dbuf, tx);
+ dsphys = dbuf->db_data;
+ bzero(dsphys, sizeof (dsl_dataset_phys_t));
+ dsphys->ds_dir_obj = ds->ds_dir->dd_object;
+ dsphys->ds_fsid_guid = unique_create();
+ (void) random_get_pseudo_bytes((void*)&dsphys->ds_guid,
+ sizeof (dsphys->ds_guid));
+ dsphys->ds_prev_snap_obj = dsl_dataset_phys(ds)->ds_prev_snap_obj;
+ dsphys->ds_prev_snap_txg = dsl_dataset_phys(ds)->ds_prev_snap_txg;
+ dsphys->ds_next_snap_obj = ds->ds_object;
+ dsphys->ds_num_children = 1;
+ dsphys->ds_creation_time = gethrestime_sec();
+ dsphys->ds_creation_txg = crtxg;
+ dsphys->ds_deadlist_obj = dsl_dataset_phys(ds)->ds_deadlist_obj;
+ dsphys->ds_referenced_bytes = dsl_dataset_phys(ds)->ds_referenced_bytes;
+ dsphys->ds_compressed_bytes = dsl_dataset_phys(ds)->ds_compressed_bytes;
+ dsphys->ds_uncompressed_bytes =
+ dsl_dataset_phys(ds)->ds_uncompressed_bytes;
+ dsphys->ds_flags = dsl_dataset_phys(ds)->ds_flags;
+ rrw_enter(&ds->ds_bp_rwlock, RW_READER, FTAG);
+ dsphys->ds_bp = dsl_dataset_phys(ds)->ds_bp;
+ rrw_exit(&ds->ds_bp_rwlock, FTAG);
+ dmu_buf_rele(dbuf, FTAG);
+
+ for (spa_feature_t f = 0; f < SPA_FEATURES; f++) {
+ if (zfeature_active(f, ds->ds_feature[f])) {
+ dsl_dataset_activate_feature(dsobj, f,
+ ds->ds_feature[f], tx);
+ }
+ }
+
+ ASSERT3U(ds->ds_prev != 0, ==,
+ dsl_dataset_phys(ds)->ds_prev_snap_obj != 0);
+ if (ds->ds_prev) {
+ uint64_t next_clones_obj =
+ dsl_dataset_phys(ds->ds_prev)->ds_next_clones_obj;
+ ASSERT(dsl_dataset_phys(ds->ds_prev)->ds_next_snap_obj ==
+ ds->ds_object ||
+ dsl_dataset_phys(ds->ds_prev)->ds_num_children > 1);
+ if (dsl_dataset_phys(ds->ds_prev)->ds_next_snap_obj ==
+ ds->ds_object) {
+ dmu_buf_will_dirty(ds->ds_prev->ds_dbuf, tx);
+ ASSERT3U(dsl_dataset_phys(ds)->ds_prev_snap_txg, ==,
+ dsl_dataset_phys(ds->ds_prev)->ds_creation_txg);
+ dsl_dataset_phys(ds->ds_prev)->ds_next_snap_obj = dsobj;
+ } else if (next_clones_obj != 0) {
+ dsl_dataset_remove_from_next_clones(ds->ds_prev,
+ dsphys->ds_next_snap_obj, tx);
+ VERIFY0(zap_add_int(mos,
+ next_clones_obj, dsobj, tx));
+ }
+ }
/*
- * If we made changes this txg, traverse_dsl_dataset won't find
- * them. Try again.
+ * If we have a reference-reservation on this dataset, we will
+ * need to increase the amount of refreservation being charged
+ * since our unique space is going to zero.
*/
- if (ds->ds_phys->ds_bp.blk_birth >= tx->tx_txg)
- return (EAGAIN);
+ if (ds->ds_reserved) {
+ int64_t delta;
+ ASSERT(DS_UNIQUE_IS_ACCURATE(ds));
+ delta = MIN(dsl_dataset_phys(ds)->ds_unique_bytes,
+ ds->ds_reserved);
+ dsl_dir_diduse_space(ds->ds_dir, DD_USED_REFRSRV,
+ delta, 0, 0, tx);
+ }
- if (dsl_dataset_is_snapshot(ds)) {
- /*
- * If this snapshot has an elevated user reference count,
- * we can't destroy it yet.
- */
- if (ds->ds_userrefs > 0 && !dsda->releasing)
- return (EBUSY);
+ dmu_buf_will_dirty(ds->ds_dbuf, tx);
+ dsl_dataset_phys(ds)->ds_deadlist_obj =
+ dsl_deadlist_clone(&ds->ds_deadlist, UINT64_MAX,
+ dsl_dataset_phys(ds)->ds_prev_snap_obj, tx);
+ dsl_deadlist_close(&ds->ds_deadlist);
+ dsl_deadlist_open(&ds->ds_deadlist, mos,
+ dsl_dataset_phys(ds)->ds_deadlist_obj);
+ dsl_deadlist_add_key(&ds->ds_deadlist,
+ dsl_dataset_phys(ds)->ds_prev_snap_txg, tx);
- mutex_enter(&ds->ds_lock);
+ if (dsl_dataset_remap_deadlist_exists(ds)) {
+ uint64_t remap_deadlist_obj =
+ dsl_dataset_get_remap_deadlist_object(ds);
/*
- * Can't delete a branch point. However, if we're destroying
- * a clone and removing its origin due to it having a user
- * hold count of 0 and having been marked for deferred destroy,
- * it's OK for the origin to have a single clone.
+ * Move the remap_deadlist to the snapshot. The head
+ * will create a new remap deadlist on demand, from
+ * dsl_dataset_block_remapped().
*/
- if (ds->ds_phys->ds_num_children >
- (dsda->is_origin_rm ? 2 : 1)) {
- mutex_exit(&ds->ds_lock);
- return (EEXIST);
- }
- mutex_exit(&ds->ds_lock);
- } else if (dsl_dir_is_clone(ds->ds_dir)) {
- return (dsl_dataset_origin_check(dsda, arg2, tx));
+ dsl_dataset_unset_remap_deadlist_object(ds, tx);
+ dsl_deadlist_close(&ds->ds_remap_deadlist);
+
+ dmu_object_zapify(mos, dsobj, DMU_OT_DSL_DATASET, tx);
+ VERIFY0(zap_add(mos, dsobj, DS_FIELD_REMAP_DEADLIST,
+ sizeof (remap_deadlist_obj), 1, &remap_deadlist_obj, tx));
}
- /* XXX we should do some i/o error checking... */
- return (0);
-}
+ /*
+ * Create a ivset guid for this snapshot if the dataset is
+ * encrypted. This may be overridden by a raw receive. A
+ * previous implementation of this code did not have this
+ * field as part of the on-disk format for ZFS encryption
+ * (see errata #4). As part of the remediation for this
+ * issue, we ask the user to enable the bookmark_v2 feature
+ * which is now a dependency of the encryption feature. We
+ * use this as a heuristic to determine when the user has
+ * elected to correct any datasets created with the old code.
+ * As a result, we only do this step if the bookmark_v2
+ * feature is enabled, which limits the number of states a
+ * given pool / dataset can be in with regards to terms of
+ * correcting the issue.
+ */
+ if (ds->ds_dir->dd_crypto_obj != 0 &&
+ spa_feature_is_enabled(dp->dp_spa, SPA_FEATURE_BOOKMARK_V2)) {
+ uint64_t ivset_guid = unique_create();
-struct refsarg {
- kmutex_t lock;
- boolean_t gone;
- kcondvar_t cv;
-};
+ dmu_object_zapify(mos, dsobj, DMU_OT_DSL_DATASET, tx);
+ VERIFY0(zap_add(mos, dsobj, DS_FIELD_IVSET_GUID,
+ sizeof (ivset_guid), 1, &ivset_guid, tx));
+ }
-/* ARGSUSED */
-static void
-dsl_dataset_refs_gone(dmu_buf_t *db, void *argv)
-{
- struct refsarg *arg = argv;
+ ASSERT3U(dsl_dataset_phys(ds)->ds_prev_snap_txg, <, tx->tx_txg);
+ dsl_dataset_phys(ds)->ds_prev_snap_obj = dsobj;
+ dsl_dataset_phys(ds)->ds_prev_snap_txg = crtxg;
+ dsl_dataset_phys(ds)->ds_unique_bytes = 0;
- mutex_enter(&arg->lock);
- arg->gone = TRUE;
- cv_signal(&arg->cv);
- mutex_exit(&arg->lock);
-}
+ if (spa_version(dp->dp_spa) >= SPA_VERSION_UNIQUE_ACCURATE)
+ dsl_dataset_phys(ds)->ds_flags |= DS_FLAG_UNIQUE_ACCURATE;
-static void
-dsl_dataset_drain_refs(dsl_dataset_t *ds, void *tag)
-{
- struct refsarg arg;
+ VERIFY0(zap_add(mos, dsl_dataset_phys(ds)->ds_snapnames_zapobj,
+ snapname, 8, 1, &dsobj, tx));
- mutex_init(&arg.lock, NULL, MUTEX_DEFAULT, NULL);
- cv_init(&arg.cv, NULL, CV_DEFAULT, NULL);
- arg.gone = FALSE;
- (void) dmu_buf_update_user(ds->ds_dbuf, ds, &arg, &ds->ds_phys,
- dsl_dataset_refs_gone);
- dmu_buf_rele(ds->ds_dbuf, tag);
- mutex_enter(&arg.lock);
- while (!arg.gone)
- cv_wait(&arg.cv, &arg.lock);
- ASSERT(arg.gone);
- mutex_exit(&arg.lock);
- ds->ds_dbuf = NULL;
- ds->ds_phys = NULL;
- mutex_destroy(&arg.lock);
- cv_destroy(&arg.cv);
+ if (ds->ds_prev)
+ dsl_dataset_rele(ds->ds_prev, ds);
+ VERIFY0(dsl_dataset_hold_obj(dp,
+ dsl_dataset_phys(ds)->ds_prev_snap_obj, ds, &ds->ds_prev));
+
+ dsl_scan_ds_snapshotted(ds, tx);
+
+ dsl_dir_snap_cmtime_update(ds->ds_dir);
+
+ spa_history_log_internal_ds(ds->ds_prev, "snapshot", tx, "");
}
-static void
-remove_from_next_clones(dsl_dataset_t *ds, uint64_t obj, dmu_tx_t *tx)
+void
+dsl_dataset_snapshot_sync(void *arg, dmu_tx_t *tx)
{
- objset_t *mos = ds->ds_dir->dd_pool->dp_meta_objset;
- int err;
- ASSERTV(uint64_t count);
+ dsl_dataset_snapshot_arg_t *ddsa = arg;
+ dsl_pool_t *dp = dmu_tx_pool(tx);
+ nvpair_t *pair;
- ASSERT(ds->ds_phys->ds_num_children >= 2);
- err = zap_remove_int(mos, ds->ds_phys->ds_next_clones_obj, obj, tx);
- /*
- * The err should not be ENOENT, but a bug in a previous version
- * of the code could cause upgrade_clones_cb() to not set
- * ds_next_snap_obj when it should, leading to a missing entry.
- * If we knew that the pool was created after
- * SPA_VERSION_NEXT_CLONES, we could assert that it isn't
- * ENOENT. However, at least we can check that we don't have
- * too many entries in the next_clones_obj even after failing to
- * remove this one.
- */
- if (err != ENOENT) {
- VERIFY3U(err, ==, 0);
+ for (pair = nvlist_next_nvpair(ddsa->ddsa_snaps, NULL);
+ pair != NULL; pair = nvlist_next_nvpair(ddsa->ddsa_snaps, pair)) {
+ dsl_dataset_t *ds;
+ char *name, *atp;
+ char dsname[ZFS_MAX_DATASET_NAME_LEN];
+
+ name = nvpair_name(pair);
+ atp = strchr(name, '@');
+ (void) strlcpy(dsname, name, atp - name + 1);
+ VERIFY0(dsl_dataset_hold(dp, dsname, FTAG, &ds));
+
+ dsl_dataset_snapshot_sync_impl(ds, atp + 1, tx);
+ if (ddsa->ddsa_props != NULL) {
+ dsl_props_set_sync_impl(ds->ds_prev,
+ ZPROP_SRC_LOCAL, ddsa->ddsa_props, tx);
+ }
+ zvol_create_minors(dp->dp_spa, nvpair_name(pair), B_TRUE);
+ dsl_dataset_rele(ds, FTAG);
}
- ASSERT3U(0, ==, zap_count(mos, ds->ds_phys->ds_next_clones_obj,
- &count));
- ASSERT3U(count, <=, ds->ds_phys->ds_num_children - 2);
}
-static void
-dsl_dataset_remove_clones_key(dsl_dataset_t *ds, uint64_t mintxg, dmu_tx_t *tx)
+/*
+ * The snapshots must all be in the same pool.
+ * All-or-nothing: if there are any failures, nothing will be modified.
+ */
+int
+dsl_dataset_snapshot(nvlist_t *snaps, nvlist_t *props, nvlist_t *errors)
{
- objset_t *mos = ds->ds_dir->dd_pool->dp_meta_objset;
- zap_cursor_t zc;
- zap_attribute_t za;
+ dsl_dataset_snapshot_arg_t ddsa;
+ nvpair_t *pair;
+ boolean_t needsuspend;
+ int error;
+ spa_t *spa;
+ char *firstname;
+ nvlist_t *suspended = NULL;
- /*
- * If it is the old version, dd_clones doesn't exist so we can't
- * find the clones, but deadlist_remove_key() is a no-op so it
- * doesn't matter.
- */
- if (ds->ds_dir->dd_phys->dd_clones == 0)
- return;
+ pair = nvlist_next_nvpair(snaps, NULL);
+ if (pair == NULL)
+ return (0);
+ firstname = nvpair_name(pair);
- for (zap_cursor_init(&zc, mos, ds->ds_dir->dd_phys->dd_clones);
- zap_cursor_retrieve(&zc, &za) == 0;
- zap_cursor_advance(&zc)) {
- dsl_dataset_t *clone;
+ error = spa_open(firstname, &spa, FTAG);
+ if (error != 0)
+ return (error);
+ needsuspend = (spa_version(spa) < SPA_VERSION_FAST_SNAP);
+ spa_close(spa, FTAG);
- VERIFY3U(0, ==, dsl_dataset_hold_obj(ds->ds_dir->dd_pool,
- za.za_first_integer, FTAG, &clone));
- if (clone->ds_dir->dd_origin_txg > mintxg) {
- dsl_deadlist_remove_key(&clone->ds_deadlist,
- mintxg, tx);
- dsl_dataset_remove_clones_key(clone, mintxg, tx);
+ if (needsuspend) {
+ suspended = fnvlist_alloc();
+ for (pair = nvlist_next_nvpair(snaps, NULL); pair != NULL;
+ pair = nvlist_next_nvpair(snaps, pair)) {
+ char fsname[ZFS_MAX_DATASET_NAME_LEN];
+ char *snapname = nvpair_name(pair);
+ char *atp;
+ void *cookie;
+
+ atp = strchr(snapname, '@');
+ if (atp == NULL) {
+ error = SET_ERROR(EINVAL);
+ break;
+ }
+ (void) strlcpy(fsname, snapname, atp - snapname + 1);
+
+ error = zil_suspend(fsname, &cookie);
+ if (error != 0)
+ break;
+ fnvlist_add_uint64(suspended, fsname,
+ (uintptr_t)cookie);
}
- dsl_dataset_rele(clone, FTAG);
}
- zap_cursor_fini(&zc);
-}
-struct process_old_arg {
- dsl_dataset_t *ds;
- dsl_dataset_t *ds_prev;
- boolean_t after_branch_point;
- zio_t *pio;
- uint64_t used, comp, uncomp;
-};
+ ddsa.ddsa_snaps = snaps;
+ ddsa.ddsa_props = props;
+ ddsa.ddsa_errors = errors;
+ ddsa.ddsa_cr = CRED();
-static int
-process_old_cb(void *arg, const blkptr_t *bp, dmu_tx_t *tx)
-{
- struct process_old_arg *poa = arg;
- dsl_pool_t *dp = poa->ds->ds_dir->dd_pool;
-
- if (bp->blk_birth <= poa->ds->ds_phys->ds_prev_snap_txg) {
- dsl_deadlist_insert(&poa->ds->ds_deadlist, bp, tx);
- if (poa->ds_prev && !poa->after_branch_point &&
- bp->blk_birth >
- poa->ds_prev->ds_phys->ds_prev_snap_txg) {
- poa->ds_prev->ds_phys->ds_unique_bytes +=
- bp_get_dsize_sync(dp->dp_spa, bp);
+ if (error == 0) {
+ error = dsl_sync_task(firstname, dsl_dataset_snapshot_check,
+ dsl_dataset_snapshot_sync, &ddsa,
+ fnvlist_num_pairs(snaps) * 3, ZFS_SPACE_CHECK_NORMAL);
+ }
+
+ if (suspended != NULL) {
+ for (pair = nvlist_next_nvpair(suspended, NULL); pair != NULL;
+ pair = nvlist_next_nvpair(suspended, pair)) {
+ zil_resume((void *)(uintptr_t)
+ fnvpair_value_uint64(pair));
}
- } else {
- poa->used += bp_get_dsize_sync(dp->dp_spa, bp);
- poa->comp += BP_GET_PSIZE(bp);
- poa->uncomp += BP_GET_UCSIZE(bp);
- dsl_free_sync(poa->pio, dp, tx->tx_txg, bp);
+ fnvlist_free(suspended);
}
- return (0);
+
+ return (error);
}
-static void
-process_old_deadlist(dsl_dataset_t *ds, dsl_dataset_t *ds_prev,
- dsl_dataset_t *ds_next, boolean_t after_branch_point, dmu_tx_t *tx)
+typedef struct dsl_dataset_snapshot_tmp_arg {
+ const char *ddsta_fsname;
+ const char *ddsta_snapname;
+ minor_t ddsta_cleanup_minor;
+ const char *ddsta_htag;
+} dsl_dataset_snapshot_tmp_arg_t;
+
+static int
+dsl_dataset_snapshot_tmp_check(void *arg, dmu_tx_t *tx)
{
- struct process_old_arg poa = { 0 };
- dsl_pool_t *dp = ds->ds_dir->dd_pool;
- objset_t *mos = dp->dp_meta_objset;
+ dsl_dataset_snapshot_tmp_arg_t *ddsta = arg;
+ dsl_pool_t *dp = dmu_tx_pool(tx);
+ dsl_dataset_t *ds;
+ int error;
- ASSERT(ds->ds_deadlist.dl_oldfmt);
- ASSERT(ds_next->ds_deadlist.dl_oldfmt);
+ error = dsl_dataset_hold(dp, ddsta->ddsta_fsname, FTAG, &ds);
+ if (error != 0)
+ return (error);
- poa.ds = ds;
- poa.ds_prev = ds_prev;
- poa.after_branch_point = after_branch_point;
- poa.pio = zio_root(dp->dp_spa, NULL, NULL, ZIO_FLAG_MUSTSUCCEED);
- VERIFY3U(0, ==, bpobj_iterate(&ds_next->ds_deadlist.dl_bpobj,
- process_old_cb, &poa, tx));
- VERIFY3U(zio_wait(poa.pio), ==, 0);
- ASSERT3U(poa.used, ==, ds->ds_phys->ds_unique_bytes);
+ /* NULL cred means no limit check for tmp snapshot */
+ error = dsl_dataset_snapshot_check_impl(ds, ddsta->ddsta_snapname,
+ tx, B_FALSE, 0, NULL);
+ if (error != 0) {
+ dsl_dataset_rele(ds, FTAG);
+ return (error);
+ }
- /* change snapused */
- dsl_dir_diduse_space(ds->ds_dir, DD_USED_SNAP,
- -poa.used, -poa.comp, -poa.uncomp, tx);
+ if (spa_version(dp->dp_spa) < SPA_VERSION_USERREFS) {
+ dsl_dataset_rele(ds, FTAG);
+ return (SET_ERROR(ENOTSUP));
+ }
+ error = dsl_dataset_user_hold_check_one(NULL, ddsta->ddsta_htag,
+ B_TRUE, tx);
+ if (error != 0) {
+ dsl_dataset_rele(ds, FTAG);
+ return (error);
+ }
- /* swap next's deadlist to our deadlist */
- dsl_deadlist_close(&ds->ds_deadlist);
- dsl_deadlist_close(&ds_next->ds_deadlist);
- SWITCH64(ds_next->ds_phys->ds_deadlist_obj,
- ds->ds_phys->ds_deadlist_obj);
- dsl_deadlist_open(&ds->ds_deadlist, mos, ds->ds_phys->ds_deadlist_obj);
- dsl_deadlist_open(&ds_next->ds_deadlist, mos,
- ds_next->ds_phys->ds_deadlist_obj);
+ dsl_dataset_rele(ds, FTAG);
+ return (0);
}
-static int
-old_synchronous_dataset_destroy(dsl_dataset_t *ds, dmu_tx_t *tx)
+static void
+dsl_dataset_snapshot_tmp_sync(void *arg, dmu_tx_t *tx)
{
- int err;
- struct killarg ka;
+ dsl_dataset_snapshot_tmp_arg_t *ddsta = arg;
+ dsl_pool_t *dp = dmu_tx_pool(tx);
+ dsl_dataset_t *ds = NULL;
- /*
- * Free everything that we point to (that's born after
- * the previous snapshot, if we are a clone)
- *
- * NB: this should be very quick, because we already
- * freed all the objects in open context.
- */
- ka.ds = ds;
- ka.tx = tx;
- err = traverse_dataset(ds,
- ds->ds_phys->ds_prev_snap_txg, TRAVERSE_POST,
- kill_blkptr, &ka);
- ASSERT3U(err, ==, 0);
- ASSERT(!DS_UNIQUE_IS_ACCURATE(ds) || ds->ds_phys->ds_unique_bytes == 0);
+ VERIFY0(dsl_dataset_hold(dp, ddsta->ddsta_fsname, FTAG, &ds));
- return (err);
+ dsl_dataset_snapshot_sync_impl(ds, ddsta->ddsta_snapname, tx);
+ dsl_dataset_user_hold_sync_one(ds->ds_prev, ddsta->ddsta_htag,
+ ddsta->ddsta_cleanup_minor, gethrestime_sec(), tx);
+ dsl_destroy_snapshot_sync_impl(ds->ds_prev, B_TRUE, tx);
+
+ dsl_dataset_rele(ds, FTAG);
}
-void
-dsl_dataset_destroy_sync(void *arg1, void *tag, dmu_tx_t *tx)
+int
+dsl_dataset_snapshot_tmp(const char *fsname, const char *snapname,
+ minor_t cleanup_minor, const char *htag)
{
- struct dsl_ds_destroyarg *dsda = arg1;
- dsl_dataset_t *ds = dsda->ds;
- int err = 0;
- int after_branch_point = FALSE;
- dsl_pool_t *dp = ds->ds_dir->dd_pool;
- objset_t *mos = dp->dp_meta_objset;
- dsl_dataset_t *ds_prev = NULL;
- boolean_t wont_destroy;
- uint64_t obj;
-
- wont_destroy = (dsda->defer &&
- (ds->ds_userrefs > 0 || ds->ds_phys->ds_num_children > 1));
-
- ASSERT(ds->ds_owner || wont_destroy);
- ASSERT(dsda->defer || ds->ds_phys->ds_num_children <= 1);
- ASSERT(ds->ds_prev == NULL ||
- ds->ds_prev->ds_phys->ds_next_snap_obj != ds->ds_object);
- ASSERT3U(ds->ds_phys->ds_bp.blk_birth, <=, tx->tx_txg);
-
- if (wont_destroy) {
- ASSERT(spa_version(dp->dp_spa) >= SPA_VERSION_USERREFS);
- dmu_buf_will_dirty(ds->ds_dbuf, tx);
- ds->ds_phys->ds_flags |= DS_FLAG_DEFER_DESTROY;
- return;
- }
-
- /* signal any waiters that this dataset is going away */
- mutex_enter(&ds->ds_lock);
- ds->ds_owner = dsl_reaper;
- cv_broadcast(&ds->ds_exclusive_cv);
- mutex_exit(&ds->ds_lock);
-
- /* Remove our reservation */
- if (ds->ds_reserved != 0) {
- dsl_prop_setarg_t psa;
- uint64_t value = 0;
-
- dsl_prop_setarg_init_uint64(&psa, "refreservation",
- (ZPROP_SRC_NONE | ZPROP_SRC_LOCAL | ZPROP_SRC_RECEIVED),
- &value);
- psa.psa_effective_value = 0; /* predict default value */
-
- dsl_dataset_set_reservation_sync(ds, &psa, tx);
- ASSERT3U(ds->ds_reserved, ==, 0);
- }
-
- ASSERT(RW_WRITE_HELD(&dp->dp_config_rwlock));
-
- dsl_scan_ds_destroyed(ds, tx);
+ dsl_dataset_snapshot_tmp_arg_t ddsta;
+ int error;
+ spa_t *spa;
+ boolean_t needsuspend;
+ void *cookie;
- obj = ds->ds_object;
+ ddsta.ddsta_fsname = fsname;
+ ddsta.ddsta_snapname = snapname;
+ ddsta.ddsta_cleanup_minor = cleanup_minor;
+ ddsta.ddsta_htag = htag;
- if (ds->ds_phys->ds_prev_snap_obj != 0) {
- if (ds->ds_prev) {
- ds_prev = ds->ds_prev;
- } else {
- VERIFY(0 == dsl_dataset_hold_obj(dp,
- ds->ds_phys->ds_prev_snap_obj, FTAG, &ds_prev));
- }
- after_branch_point =
- (ds_prev->ds_phys->ds_next_snap_obj != obj);
-
- dmu_buf_will_dirty(ds_prev->ds_dbuf, tx);
- if (after_branch_point &&
- ds_prev->ds_phys->ds_next_clones_obj != 0) {
- remove_from_next_clones(ds_prev, obj, tx);
- if (ds->ds_phys->ds_next_snap_obj != 0) {
- VERIFY(0 == zap_add_int(mos,
- ds_prev->ds_phys->ds_next_clones_obj,
- ds->ds_phys->ds_next_snap_obj, tx));
- }
- }
- if (after_branch_point &&
- ds->ds_phys->ds_next_snap_obj == 0) {
- /* This clone is toast. */
- ASSERT(ds_prev->ds_phys->ds_num_children > 1);
- ds_prev->ds_phys->ds_num_children--;
+ error = spa_open(fsname, &spa, FTAG);
+ if (error != 0)
+ return (error);
+ needsuspend = (spa_version(spa) < SPA_VERSION_FAST_SNAP);
+ spa_close(spa, FTAG);
- /*
- * If the clone's origin has no other clones, no
- * user holds, and has been marked for deferred
- * deletion, then we should have done the necessary
- * destroy setup for it.
- */
- if (ds_prev->ds_phys->ds_num_children == 1 &&
- ds_prev->ds_userrefs == 0 &&
- DS_IS_DEFER_DESTROY(ds_prev)) {
- ASSERT3P(dsda->rm_origin, !=, NULL);
- } else {
- ASSERT3P(dsda->rm_origin, ==, NULL);
- }
- } else if (!after_branch_point) {
- ds_prev->ds_phys->ds_next_snap_obj =
- ds->ds_phys->ds_next_snap_obj;
- }
+ if (needsuspend) {
+ error = zil_suspend(fsname, &cookie);
+ if (error != 0)
+ return (error);
}
- if (dsl_dataset_is_snapshot(ds)) {
- dsl_dataset_t *ds_next;
- uint64_t old_unique;
- uint64_t used = 0, comp = 0, uncomp = 0;
-
- VERIFY(0 == dsl_dataset_hold_obj(dp,
- ds->ds_phys->ds_next_snap_obj, FTAG, &ds_next));
- ASSERT3U(ds_next->ds_phys->ds_prev_snap_obj, ==, obj);
-
- old_unique = ds_next->ds_phys->ds_unique_bytes;
-
- dmu_buf_will_dirty(ds_next->ds_dbuf, tx);
- ds_next->ds_phys->ds_prev_snap_obj =
- ds->ds_phys->ds_prev_snap_obj;
- ds_next->ds_phys->ds_prev_snap_txg =
- ds->ds_phys->ds_prev_snap_txg;
- ASSERT3U(ds->ds_phys->ds_prev_snap_txg, ==,
- ds_prev ? ds_prev->ds_phys->ds_creation_txg : 0);
-
-
- if (ds_next->ds_deadlist.dl_oldfmt) {
- process_old_deadlist(ds, ds_prev, ds_next,
- after_branch_point, tx);
- } else {
- /* Adjust prev's unique space. */
- if (ds_prev && !after_branch_point) {
- dsl_deadlist_space_range(&ds_next->ds_deadlist,
- ds_prev->ds_phys->ds_prev_snap_txg,
- ds->ds_phys->ds_prev_snap_txg,
- &used, &comp, &uncomp);
- ds_prev->ds_phys->ds_unique_bytes += used;
- }
-
- /* Adjust snapused. */
- dsl_deadlist_space_range(&ds_next->ds_deadlist,
- ds->ds_phys->ds_prev_snap_txg, UINT64_MAX,
- &used, &comp, &uncomp);
- dsl_dir_diduse_space(ds->ds_dir, DD_USED_SNAP,
- -used, -comp, -uncomp, tx);
-
- /* Move blocks to be freed to pool's free list. */
- dsl_deadlist_move_bpobj(&ds_next->ds_deadlist,
- &dp->dp_free_bpobj, ds->ds_phys->ds_prev_snap_txg,
- tx);
- dsl_dir_diduse_space(tx->tx_pool->dp_free_dir,
- DD_USED_HEAD, used, comp, uncomp, tx);
-
- /* Merge our deadlist into next's and free it. */
- dsl_deadlist_merge(&ds_next->ds_deadlist,
- ds->ds_phys->ds_deadlist_obj, tx);
- }
- dsl_deadlist_close(&ds->ds_deadlist);
- dsl_deadlist_free(mos, ds->ds_phys->ds_deadlist_obj, tx);
-
- /* Collapse range in clone heads */
- dsl_dataset_remove_clones_key(ds,
- ds->ds_phys->ds_creation_txg, tx);
-
- if (dsl_dataset_is_snapshot(ds_next)) {
- dsl_dataset_t *ds_nextnext;
- dsl_dataset_t *hds;
-
- /*
- * Update next's unique to include blocks which
- * were previously shared by only this snapshot
- * and it. Those blocks will be born after the
- * prev snap and before this snap, and will have
- * died after the next snap and before the one
- * after that (ie. be on the snap after next's
- * deadlist).
- */
- VERIFY(0 == dsl_dataset_hold_obj(dp,
- ds_next->ds_phys->ds_next_snap_obj,
- FTAG, &ds_nextnext));
- dsl_deadlist_space_range(&ds_nextnext->ds_deadlist,
- ds->ds_phys->ds_prev_snap_txg,
- ds->ds_phys->ds_creation_txg,
- &used, &comp, &uncomp);
- ds_next->ds_phys->ds_unique_bytes += used;
- dsl_dataset_rele(ds_nextnext, FTAG);
- ASSERT3P(ds_next->ds_prev, ==, NULL);
-
- /* Collapse range in this head. */
- VERIFY3U(0, ==, dsl_dataset_hold_obj(dp,
- ds->ds_dir->dd_phys->dd_head_dataset_obj,
- FTAG, &hds));
- dsl_deadlist_remove_key(&hds->ds_deadlist,
- ds->ds_phys->ds_creation_txg, tx);
- dsl_dataset_rele(hds, FTAG);
-
- } else {
- ASSERT3P(ds_next->ds_prev, ==, ds);
- dsl_dataset_drop_ref(ds_next->ds_prev, ds_next);
- ds_next->ds_prev = NULL;
- if (ds_prev) {
- VERIFY(0 == dsl_dataset_get_ref(dp,
- ds->ds_phys->ds_prev_snap_obj,
- ds_next, &ds_next->ds_prev));
- }
-
- dsl_dataset_recalc_head_uniq(ds_next);
-
- /*
- * Reduce the amount of our unconsmed refreservation
- * being charged to our parent by the amount of
- * new unique data we have gained.
- */
- if (old_unique < ds_next->ds_reserved) {
- int64_t mrsdelta;
- uint64_t new_unique =
- ds_next->ds_phys->ds_unique_bytes;
-
- ASSERT(old_unique <= new_unique);
- mrsdelta = MIN(new_unique - old_unique,
- ds_next->ds_reserved - old_unique);
- dsl_dir_diduse_space(ds->ds_dir,
- DD_USED_REFRSRV, -mrsdelta, 0, 0, tx);
- }
- }
- dsl_dataset_rele(ds_next, FTAG);
- } else {
- zfeature_info_t *async_destroy =
- &spa_feature_table[SPA_FEATURE_ASYNC_DESTROY];
- objset_t *os;
-
- /*
- * There's no next snapshot, so this is a head dataset.
- * Destroy the deadlist. Unless it's a clone, the
- * deadlist should be empty. (If it's a clone, it's
- * safe to ignore the deadlist contents.)
- */
- dsl_deadlist_close(&ds->ds_deadlist);
- dsl_deadlist_free(mos, ds->ds_phys->ds_deadlist_obj, tx);
- ds->ds_phys->ds_deadlist_obj = 0;
-
- VERIFY3U(0, ==, dmu_objset_from_ds(ds, &os));
+ error = dsl_sync_task(fsname, dsl_dataset_snapshot_tmp_check,
+ dsl_dataset_snapshot_tmp_sync, &ddsta, 3, ZFS_SPACE_CHECK_RESERVED);
- if (!spa_feature_is_enabled(dp->dp_spa, async_destroy)) {
- err = old_synchronous_dataset_destroy(ds, tx);
- } else {
- /*
- * Move the bptree into the pool's list of trees to
- * clean up and update space accounting information.
- */
- uint64_t used, comp, uncomp;
-
- zil_destroy_sync(dmu_objset_zil(os), tx);
-
- if (!spa_feature_is_active(dp->dp_spa, async_destroy)) {
- spa_feature_incr(dp->dp_spa, async_destroy, tx);
- dp->dp_bptree_obj = bptree_alloc(mos, tx);
- VERIFY(zap_add(mos,
- DMU_POOL_DIRECTORY_OBJECT,
- DMU_POOL_BPTREE_OBJ, sizeof (uint64_t), 1,
- &dp->dp_bptree_obj, tx) == 0);
- }
-
- used = ds->ds_dir->dd_phys->dd_used_bytes;
- comp = ds->ds_dir->dd_phys->dd_compressed_bytes;
- uncomp = ds->ds_dir->dd_phys->dd_uncompressed_bytes;
-
- ASSERT(!DS_UNIQUE_IS_ACCURATE(ds) ||
- ds->ds_phys->ds_unique_bytes == used);
-
- bptree_add(mos, dp->dp_bptree_obj,
- &ds->ds_phys->ds_bp, ds->ds_phys->ds_prev_snap_txg,
- used, comp, uncomp, tx);
- dsl_dir_diduse_space(ds->ds_dir, DD_USED_HEAD,
- -used, -comp, -uncomp, tx);
- dsl_dir_diduse_space(dp->dp_free_dir, DD_USED_HEAD,
- used, comp, uncomp, tx);
- }
+ if (needsuspend)
+ zil_resume(cookie);
+ return (error);
+}
- if (ds->ds_prev != NULL) {
- if (spa_version(dp->dp_spa) >= SPA_VERSION_DIR_CLONES) {
- VERIFY3U(0, ==, zap_remove_int(mos,
- ds->ds_prev->ds_dir->dd_phys->dd_clones,
- ds->ds_object, tx));
- }
- dsl_dataset_rele(ds->ds_prev, ds);
- ds->ds_prev = ds_prev = NULL;
- }
- }
+void
+dsl_dataset_sync(dsl_dataset_t *ds, zio_t *zio, dmu_tx_t *tx)
+{
+ ASSERT(dmu_tx_is_syncing(tx));
+ ASSERT(ds->ds_objset != NULL);
+ ASSERT(dsl_dataset_phys(ds)->ds_next_snap_obj == 0);
/*
- * This must be done after the dsl_traverse(), because it will
- * re-open the objset.
+ * in case we had to change ds_fsid_guid when we opened it,
+ * sync it out now.
*/
- if (ds->ds_objset) {
- dmu_objset_evict(ds->ds_objset);
- ds->ds_objset = NULL;
+ dmu_buf_will_dirty(ds->ds_dbuf, tx);
+ dsl_dataset_phys(ds)->ds_fsid_guid = ds->ds_fsid_guid;
+
+ if (ds->ds_resume_bytes[tx->tx_txg & TXG_MASK] != 0) {
+ VERIFY0(zap_update(tx->tx_pool->dp_meta_objset,
+ ds->ds_object, DS_FIELD_RESUME_OBJECT, 8, 1,
+ &ds->ds_resume_object[tx->tx_txg & TXG_MASK], tx));
+ VERIFY0(zap_update(tx->tx_pool->dp_meta_objset,
+ ds->ds_object, DS_FIELD_RESUME_OFFSET, 8, 1,
+ &ds->ds_resume_offset[tx->tx_txg & TXG_MASK], tx));
+ VERIFY0(zap_update(tx->tx_pool->dp_meta_objset,
+ ds->ds_object, DS_FIELD_RESUME_BYTES, 8, 1,
+ &ds->ds_resume_bytes[tx->tx_txg & TXG_MASK], tx));
+ ds->ds_resume_object[tx->tx_txg & TXG_MASK] = 0;
+ ds->ds_resume_offset[tx->tx_txg & TXG_MASK] = 0;
+ ds->ds_resume_bytes[tx->tx_txg & TXG_MASK] = 0;
}
- if (ds->ds_dir->dd_phys->dd_head_dataset_obj == ds->ds_object) {
- /* Erase the link in the dir */
- dmu_buf_will_dirty(ds->ds_dir->dd_dbuf, tx);
- ds->ds_dir->dd_phys->dd_head_dataset_obj = 0;
- ASSERT(ds->ds_phys->ds_snapnames_zapobj != 0);
- err = zap_destroy(mos, ds->ds_phys->ds_snapnames_zapobj, tx);
- ASSERT(err == 0);
- } else {
- /* remove from snapshot namespace */
- dsl_dataset_t *ds_head;
- ASSERT(ds->ds_phys->ds_snapnames_zapobj == 0);
- VERIFY(0 == dsl_dataset_hold_obj(dp,
- ds->ds_dir->dd_phys->dd_head_dataset_obj, FTAG, &ds_head));
- VERIFY(0 == dsl_dataset_get_snapname(ds));
-#ifdef ZFS_DEBUG
- {
- uint64_t val;
+ dmu_objset_sync(ds->ds_objset, zio, tx);
- err = dsl_dataset_snap_lookup(ds_head,
- ds->ds_snapname, &val);
- ASSERT3U(err, ==, 0);
- ASSERT3U(val, ==, obj);
+ for (spa_feature_t f = 0; f < SPA_FEATURES; f++) {
+ if (zfeature_active(f, ds->ds_feature_activation[f])) {
+ if (zfeature_active(f, ds->ds_feature[f]))
+ continue;
+ dsl_dataset_activate_feature(ds->ds_object, f,
+ ds->ds_feature_activation[f], tx);
+ ds->ds_feature[f] = ds->ds_feature_activation[f];
}
-#endif
- err = dsl_dataset_snap_remove(ds_head, ds->ds_snapname, tx);
- ASSERT(err == 0);
- dsl_dataset_rele(ds_head, FTAG);
- }
-
- if (ds_prev && ds->ds_prev != ds_prev)
- dsl_dataset_rele(ds_prev, FTAG);
-
- spa_prop_clear_bootfs(dp->dp_spa, ds->ds_object, tx);
- spa_history_log_internal(LOG_DS_DESTROY, dp->dp_spa, tx,
- "dataset = %llu", ds->ds_object);
-
- if (ds->ds_phys->ds_next_clones_obj != 0) {
- ASSERTV(uint64_t count);
- ASSERT(0 == zap_count(mos,
- ds->ds_phys->ds_next_clones_obj, &count) && count == 0);
- VERIFY(0 == dmu_object_free(mos,
- ds->ds_phys->ds_next_clones_obj, tx));
- }
- if (ds->ds_phys->ds_props_obj != 0)
- VERIFY(0 == zap_destroy(mos, ds->ds_phys->ds_props_obj, tx));
- if (ds->ds_phys->ds_userrefs_obj != 0)
- VERIFY(0 == zap_destroy(mos, ds->ds_phys->ds_userrefs_obj, tx));
- dsl_dir_close(ds->ds_dir, ds);
- ds->ds_dir = NULL;
- dsl_dataset_drain_refs(ds, tag);
- VERIFY(0 == dmu_object_free(mos, obj, tx));
-
- if (dsda->rm_origin) {
- /*
- * Remove the origin of the clone we just destroyed.
- */
- struct dsl_ds_destroyarg ndsda = {0};
-
- ndsda.ds = dsda->rm_origin;
- dsl_dataset_destroy_sync(&ndsda, tag, tx);
}
}
static int
-dsl_dataset_snapshot_reserve_space(dsl_dataset_t *ds, dmu_tx_t *tx)
+deadlist_enqueue_cb(void *arg, const blkptr_t *bp, dmu_tx_t *tx)
{
- uint64_t asize;
-
- if (!dmu_tx_is_syncing(tx))
- return (0);
-
- /*
- * If there's an fs-only reservation, any blocks that might become
- * owned by the snapshot dataset must be accommodated by space
- * outside of the reservation.
- */
- ASSERT(ds->ds_reserved == 0 || DS_UNIQUE_IS_ACCURATE(ds));
- asize = MIN(ds->ds_phys->ds_unique_bytes, ds->ds_reserved);
- if (asize > dsl_dir_space_available(ds->ds_dir, NULL, 0, TRUE))
- return (ENOSPC);
-
- /*
- * Propogate any reserved space for this snapshot to other
- * snapshot checks in this sync group.
- */
- if (asize > 0)
- dsl_dir_willuse_space(ds->ds_dir, asize, tx);
-
- return (0);
-}
-
-int
-dsl_dataset_snapshot_check(void *arg1, void *arg2, dmu_tx_t *tx)
-{
- dsl_dataset_t *ds = arg1;
- const char *snapname = arg2;
- int err;
- uint64_t value;
-
- /*
- * We don't allow multiple snapshots of the same txg. If there
- * is already one, try again.
- */
- if (ds->ds_phys->ds_prev_snap_txg >= tx->tx_txg)
- return (EAGAIN);
-
- /*
- * Check for conflicting name snapshot name.
- */
- err = dsl_dataset_snap_lookup(ds, snapname, &value);
- if (err == 0)
- return (EEXIST);
- if (err != ENOENT)
- return (err);
-
- /*
- * Check that the dataset's name is not too long. Name consists
- * of the dataset's length + 1 for the @-sign + snapshot name's length
- */
- if (dsl_dataset_namelen(ds) + 1 + strlen(snapname) >= MAXNAMELEN)
- return (ENAMETOOLONG);
-
- err = dsl_dataset_snapshot_reserve_space(ds, tx);
- if (err)
- return (err);
-
- ds->ds_trysnap_txg = tx->tx_txg;
+ dsl_deadlist_t *dl = arg;
+ dsl_deadlist_insert(dl, bp, tx);
return (0);
}
void
-dsl_dataset_snapshot_sync(void *arg1, void *arg2, dmu_tx_t *tx)
+dsl_dataset_sync_done(dsl_dataset_t *ds, dmu_tx_t *tx)
{
- dsl_dataset_t *ds = arg1;
- const char *snapname = arg2;
- dsl_pool_t *dp = ds->ds_dir->dd_pool;
- dmu_buf_t *dbuf;
- dsl_dataset_phys_t *dsphys;
- uint64_t dsobj, crtxg;
- objset_t *mos = dp->dp_meta_objset;
- int err;
-
- ASSERT(RW_WRITE_HELD(&dp->dp_config_rwlock));
-
- /*
- * The origin's ds_creation_txg has to be < TXG_INITIAL
- */
- if (strcmp(snapname, ORIGIN_DIR_NAME) == 0)
- crtxg = 1;
- else
- crtxg = tx->tx_txg;
-
- dsobj = dmu_object_alloc(mos, DMU_OT_DSL_DATASET, 0,
- DMU_OT_DSL_DATASET, sizeof (dsl_dataset_phys_t), tx);
- VERIFY(0 == dmu_bonus_hold(mos, dsobj, FTAG, &dbuf));
- dmu_buf_will_dirty(dbuf, tx);
- dsphys = dbuf->db_data;
- bzero(dsphys, sizeof (dsl_dataset_phys_t));
- dsphys->ds_dir_obj = ds->ds_dir->dd_object;
- dsphys->ds_fsid_guid = unique_create();
- (void) random_get_pseudo_bytes((void*)&dsphys->ds_guid,
- sizeof (dsphys->ds_guid));
- dsphys->ds_prev_snap_obj = ds->ds_phys->ds_prev_snap_obj;
- dsphys->ds_prev_snap_txg = ds->ds_phys->ds_prev_snap_txg;
- dsphys->ds_next_snap_obj = ds->ds_object;
- dsphys->ds_num_children = 1;
- dsphys->ds_creation_time = gethrestime_sec();
- dsphys->ds_creation_txg = crtxg;
- dsphys->ds_deadlist_obj = ds->ds_phys->ds_deadlist_obj;
- dsphys->ds_referenced_bytes = ds->ds_phys->ds_referenced_bytes;
- dsphys->ds_compressed_bytes = ds->ds_phys->ds_compressed_bytes;
- dsphys->ds_uncompressed_bytes = ds->ds_phys->ds_uncompressed_bytes;
- dsphys->ds_flags = ds->ds_phys->ds_flags;
- dsphys->ds_bp = ds->ds_phys->ds_bp;
- dmu_buf_rele(dbuf, FTAG);
+ objset_t *os = ds->ds_objset;
- ASSERT3U(ds->ds_prev != 0, ==, ds->ds_phys->ds_prev_snap_obj != 0);
- if (ds->ds_prev) {
- uint64_t next_clones_obj =
- ds->ds_prev->ds_phys->ds_next_clones_obj;
- ASSERT(ds->ds_prev->ds_phys->ds_next_snap_obj ==
- ds->ds_object ||
- ds->ds_prev->ds_phys->ds_num_children > 1);
- if (ds->ds_prev->ds_phys->ds_next_snap_obj == ds->ds_object) {
- dmu_buf_will_dirty(ds->ds_prev->ds_dbuf, tx);
- ASSERT3U(ds->ds_phys->ds_prev_snap_txg, ==,
- ds->ds_prev->ds_phys->ds_creation_txg);
- ds->ds_prev->ds_phys->ds_next_snap_obj = dsobj;
- } else if (next_clones_obj != 0) {
- remove_from_next_clones(ds->ds_prev,
- dsphys->ds_next_snap_obj, tx);
- VERIFY3U(0, ==, zap_add_int(mos,
- next_clones_obj, dsobj, tx));
- }
- }
+ bplist_iterate(&ds->ds_pending_deadlist,
+ deadlist_enqueue_cb, &ds->ds_deadlist, tx);
- /*
- * If we have a reference-reservation on this dataset, we will
- * need to increase the amount of refreservation being charged
- * since our unique space is going to zero.
- */
- if (ds->ds_reserved) {
- int64_t delta;
- ASSERT(DS_UNIQUE_IS_ACCURATE(ds));
- delta = MIN(ds->ds_phys->ds_unique_bytes, ds->ds_reserved);
- dsl_dir_diduse_space(ds->ds_dir, DD_USED_REFRSRV,
- delta, 0, 0, tx);
+ if (os->os_synced_dnodes != NULL) {
+ multilist_destroy(os->os_synced_dnodes);
+ os->os_synced_dnodes = NULL;
}
- dmu_buf_will_dirty(ds->ds_dbuf, tx);
- zfs_dbgmsg("taking snapshot %s@%s/%llu; newkey=%llu",
- ds->ds_dir->dd_myname, snapname, dsobj,
- ds->ds_phys->ds_prev_snap_txg);
- ds->ds_phys->ds_deadlist_obj = dsl_deadlist_clone(&ds->ds_deadlist,
- UINT64_MAX, ds->ds_phys->ds_prev_snap_obj, tx);
- dsl_deadlist_close(&ds->ds_deadlist);
- dsl_deadlist_open(&ds->ds_deadlist, mos, ds->ds_phys->ds_deadlist_obj);
- dsl_deadlist_add_key(&ds->ds_deadlist,
- ds->ds_phys->ds_prev_snap_txg, tx);
-
- ASSERT3U(ds->ds_phys->ds_prev_snap_txg, <, tx->tx_txg);
- ds->ds_phys->ds_prev_snap_obj = dsobj;
- ds->ds_phys->ds_prev_snap_txg = crtxg;
- ds->ds_phys->ds_unique_bytes = 0;
- if (spa_version(dp->dp_spa) >= SPA_VERSION_UNIQUE_ACCURATE)
- ds->ds_phys->ds_flags |= DS_FLAG_UNIQUE_ACCURATE;
-
- err = zap_add(mos, ds->ds_phys->ds_snapnames_zapobj,
- snapname, 8, 1, &dsobj, tx);
- ASSERT(err == 0);
-
- if (ds->ds_prev)
- dsl_dataset_drop_ref(ds->ds_prev, ds);
- VERIFY(0 == dsl_dataset_get_ref(dp,
- ds->ds_phys->ds_prev_snap_obj, ds, &ds->ds_prev));
-
- dsl_scan_ds_snapshotted(ds, tx);
-
- dsl_dir_snap_cmtime_update(ds->ds_dir);
-
- spa_history_log_internal(LOG_DS_SNAPSHOT, dp->dp_spa, tx,
- "dataset = %llu", dsobj);
-}
-
-void
-dsl_dataset_sync(dsl_dataset_t *ds, zio_t *zio, dmu_tx_t *tx)
-{
- ASSERT(dmu_tx_is_syncing(tx));
- ASSERT(ds->ds_objset != NULL);
- ASSERT(ds->ds_phys->ds_next_snap_obj == 0);
+ if (os->os_encrypted)
+ os->os_next_write_raw[tx->tx_txg & TXG_MASK] = B_FALSE;
+ else
+ ASSERT0(os->os_next_write_raw[tx->tx_txg & TXG_MASK]);
- /*
- * in case we had to change ds_fsid_guid when we opened it,
- * sync it out now.
- */
- dmu_buf_will_dirty(ds->ds_dbuf, tx);
- ds->ds_phys->ds_fsid_guid = ds->ds_fsid_guid;
+ ASSERT(!dmu_objset_is_dirty(os, dmu_tx_get_txg(tx)));
- dmu_objset_sync(ds->ds_objset, zio, tx);
+ dmu_buf_rele(ds->ds_dbuf, ds);
}
-static void
-get_clones_stat(dsl_dataset_t *ds, nvlist_t *nv)
+int
+get_clones_stat_impl(dsl_dataset_t *ds, nvlist_t *val)
{
uint64_t count = 0;
objset_t *mos = ds->ds_dir->dd_pool->dp_meta_objset;
zap_cursor_t zc;
zap_attribute_t za;
- nvlist_t *propval;
- nvlist_t *val;
- rw_enter(&ds->ds_dir->dd_pool->dp_config_rwlock, RW_READER);
- VERIFY(nvlist_alloc(&propval, NV_UNIQUE_NAME, KM_SLEEP) == 0);
- VERIFY(nvlist_alloc(&val, NV_UNIQUE_NAME, KM_SLEEP) == 0);
+ ASSERT(dsl_pool_config_held(ds->ds_dir->dd_pool));
/*
- * There may me missing entries in ds_next_clones_obj
+ * There may be missing entries in ds_next_clones_obj
* due to a bug in a previous version of the code.
* Only trust it if it has the right number of entries.
*/
- if (ds->ds_phys->ds_next_clones_obj != 0) {
- ASSERT3U(0, ==, zap_count(mos, ds->ds_phys->ds_next_clones_obj,
+ if (dsl_dataset_phys(ds)->ds_next_clones_obj != 0) {
+ VERIFY0(zap_count(mos, dsl_dataset_phys(ds)->ds_next_clones_obj,
&count));
}
- if (count != ds->ds_phys->ds_num_children - 1) {
- goto fail;
+ if (count != dsl_dataset_phys(ds)->ds_num_children - 1) {
+ return (ENOENT);
}
- for (zap_cursor_init(&zc, mos, ds->ds_phys->ds_next_clones_obj);
+ for (zap_cursor_init(&zc, mos,
+ dsl_dataset_phys(ds)->ds_next_clones_obj);
zap_cursor_retrieve(&zc, &za) == 0;
zap_cursor_advance(&zc)) {
dsl_dataset_t *clone;
- char buf[ZFS_MAXNAMELEN];
- /*
- * Even though we hold the dp_config_rwlock, the dataset
- * may fail to open, returning ENOENT. If there is a
- * thread concurrently attempting to destroy this
- * dataset, it will have the ds_rwlock held for
- * RW_WRITER. Our call to dsl_dataset_hold_obj() ->
- * dsl_dataset_hold_ref() will fail its
- * rw_tryenter(&ds->ds_rwlock, RW_READER), drop the
- * dp_config_rwlock, and wait for the destroy progress
- * and signal ds_exclusive_cv. If the destroy was
- * successful, we will see that
- * DSL_DATASET_IS_DESTROYED(), and return ENOENT.
- */
- if (dsl_dataset_hold_obj(ds->ds_dir->dd_pool,
- za.za_first_integer, FTAG, &clone) != 0)
- continue;
+ char buf[ZFS_MAX_DATASET_NAME_LEN];
+ VERIFY0(dsl_dataset_hold_obj(ds->ds_dir->dd_pool,
+ za.za_first_integer, FTAG, &clone));
dsl_dir_name(clone->ds_dir, buf);
- VERIFY(nvlist_add_boolean(val, buf) == 0);
+ fnvlist_add_boolean(val, buf);
dsl_dataset_rele(clone, FTAG);
}
zap_cursor_fini(&zc);
- VERIFY(nvlist_add_nvlist(propval, ZPROP_VALUE, val) == 0);
- VERIFY(nvlist_add_nvlist(nv, zfs_prop_to_name(ZFS_PROP_CLONES),
- propval) == 0);
-fail:
- nvlist_free(val);
- nvlist_free(propval);
- rw_exit(&ds->ds_dir->dd_pool->dp_config_rwlock);
+ return (0);
}
void
-dsl_dataset_stats(dsl_dataset_t *ds, nvlist_t *nv)
+get_clones_stat(dsl_dataset_t *ds, nvlist_t *nv)
{
- uint64_t refd, avail, uobjs, aobjs, ratio;
+ nvlist_t *propval = fnvlist_alloc();
+ nvlist_t *val;
- dsl_dir_stats(ds->ds_dir, nv);
+ /*
+ * We use nvlist_alloc() instead of fnvlist_alloc() because the
+ * latter would allocate the list with NV_UNIQUE_NAME flag.
+ * As a result, every time a clone name is appended to the list
+ * it would be (linearly) searched for for a duplicate name.
+ * We already know that all clone names must be unique and we
+ * want avoid the quadratic complexity of double-checking that
+ * because we can have a large number of clones.
+ */
+ VERIFY0(nvlist_alloc(&val, 0, KM_SLEEP));
- dsl_dataset_space(ds, &refd, &avail, &uobjs, &aobjs);
- dsl_prop_nvlist_add_uint64(nv, ZFS_PROP_AVAILABLE, avail);
- dsl_prop_nvlist_add_uint64(nv, ZFS_PROP_REFERENCED, refd);
+ if (get_clones_stat_impl(ds, val) == 0) {
+ fnvlist_add_nvlist(propval, ZPROP_VALUE, val);
+ fnvlist_add_nvlist(nv, zfs_prop_to_name(ZFS_PROP_CLONES),
+ propval);
+ }
- dsl_prop_nvlist_add_uint64(nv, ZFS_PROP_CREATION,
- ds->ds_phys->ds_creation_time);
- dsl_prop_nvlist_add_uint64(nv, ZFS_PROP_CREATETXG,
- ds->ds_phys->ds_creation_txg);
- dsl_prop_nvlist_add_uint64(nv, ZFS_PROP_REFQUOTA,
- ds->ds_quota);
- dsl_prop_nvlist_add_uint64(nv, ZFS_PROP_REFRESERVATION,
- ds->ds_reserved);
- dsl_prop_nvlist_add_uint64(nv, ZFS_PROP_GUID,
- ds->ds_phys->ds_guid);
- dsl_prop_nvlist_add_uint64(nv, ZFS_PROP_UNIQUE,
- ds->ds_phys->ds_unique_bytes);
- dsl_prop_nvlist_add_uint64(nv, ZFS_PROP_OBJSETID,
- ds->ds_object);
- dsl_prop_nvlist_add_uint64(nv, ZFS_PROP_USERREFS,
- ds->ds_userrefs);
- dsl_prop_nvlist_add_uint64(nv, ZFS_PROP_DEFER_DESTROY,
- DS_IS_DEFER_DESTROY(ds) ? 1 : 0);
+ nvlist_free(val);
+ nvlist_free(propval);
+}
- if (ds->ds_phys->ds_prev_snap_obj != 0) {
- uint64_t written, comp, uncomp;
- dsl_pool_t *dp = ds->ds_dir->dd_pool;
- dsl_dataset_t *prev;
- int err;
+/*
+ * Returns a string that represents the receive resume stats token. It should
+ * be freed with strfree().
+ */
+char *
+get_receive_resume_stats_impl(dsl_dataset_t *ds)
+{
+ dsl_pool_t *dp = ds->ds_dir->dd_pool;
- rw_enter(&dp->dp_config_rwlock, RW_READER);
- err = dsl_dataset_hold_obj(dp,
- ds->ds_phys->ds_prev_snap_obj, FTAG, &prev);
- rw_exit(&dp->dp_config_rwlock);
- if (err == 0) {
- err = dsl_dataset_space_written(prev, ds, &written,
- &comp, &uncomp);
- dsl_dataset_rele(prev, FTAG);
- if (err == 0) {
- dsl_prop_nvlist_add_uint64(nv, ZFS_PROP_WRITTEN,
- written);
- }
+ if (dsl_dataset_has_resume_receive_state(ds)) {
+ char *str;
+ void *packed;
+ uint8_t *compressed;
+ uint64_t val;
+ nvlist_t *token_nv = fnvlist_alloc();
+ size_t packed_size, compressed_size;
+
+ if (zap_lookup(dp->dp_meta_objset, ds->ds_object,
+ DS_FIELD_RESUME_FROMGUID, sizeof (val), 1, &val) == 0) {
+ fnvlist_add_uint64(token_nv, "fromguid", val);
}
- }
+ if (zap_lookup(dp->dp_meta_objset, ds->ds_object,
+ DS_FIELD_RESUME_OBJECT, sizeof (val), 1, &val) == 0) {
+ fnvlist_add_uint64(token_nv, "object", val);
+ }
+ if (zap_lookup(dp->dp_meta_objset, ds->ds_object,
+ DS_FIELD_RESUME_OFFSET, sizeof (val), 1, &val) == 0) {
+ fnvlist_add_uint64(token_nv, "offset", val);
+ }
+ if (zap_lookup(dp->dp_meta_objset, ds->ds_object,
+ DS_FIELD_RESUME_BYTES, sizeof (val), 1, &val) == 0) {
+ fnvlist_add_uint64(token_nv, "bytes", val);
+ }
+ if (zap_lookup(dp->dp_meta_objset, ds->ds_object,
+ DS_FIELD_RESUME_TOGUID, sizeof (val), 1, &val) == 0) {
+ fnvlist_add_uint64(token_nv, "toguid", val);
+ }
+ char buf[MAXNAMELEN];
+ if (zap_lookup(dp->dp_meta_objset, ds->ds_object,
+ DS_FIELD_RESUME_TONAME, 1, sizeof (buf), buf) == 0) {
+ fnvlist_add_string(token_nv, "toname", buf);
+ }
+ if (zap_contains(dp->dp_meta_objset, ds->ds_object,
+ DS_FIELD_RESUME_LARGEBLOCK) == 0) {
+ fnvlist_add_boolean(token_nv, "largeblockok");
+ }
+ if (zap_contains(dp->dp_meta_objset, ds->ds_object,
+ DS_FIELD_RESUME_EMBEDOK) == 0) {
+ fnvlist_add_boolean(token_nv, "embedok");
+ }
+ if (zap_contains(dp->dp_meta_objset, ds->ds_object,
+ DS_FIELD_RESUME_COMPRESSOK) == 0) {
+ fnvlist_add_boolean(token_nv, "compressok");
+ }
+ if (zap_contains(dp->dp_meta_objset, ds->ds_object,
+ DS_FIELD_RESUME_RAWOK) == 0) {
+ fnvlist_add_boolean(token_nv, "rawok");
+ }
+ packed = fnvlist_pack(token_nv, &packed_size);
+ fnvlist_free(token_nv);
+ compressed = kmem_alloc(packed_size, KM_SLEEP);
+
+ compressed_size = gzip_compress(packed, compressed,
+ packed_size, packed_size, 6);
+
+ zio_cksum_t cksum;
+ fletcher_4_native_varsize(compressed, compressed_size, &cksum);
+
+ str = kmem_alloc(compressed_size * 2 + 1, KM_SLEEP);
+ for (int i = 0; i < compressed_size; i++) {
+ (void) sprintf(str + i * 2, "%02x", compressed[i]);
+ }
+ str[compressed_size * 2] = '\0';
+ char *propval = kmem_asprintf("%u-%llx-%llx-%s",
+ ZFS_SEND_RESUME_TOKEN_VERSION,
+ (longlong_t)cksum.zc_word[0],
+ (longlong_t)packed_size, str);
+ kmem_free(packed, packed_size);
+ kmem_free(str, compressed_size * 2 + 1);
+ kmem_free(compressed, packed_size);
+ return (propval);
+ }
+ return (strdup(""));
+}
+
+/*
+ * Returns a string that represents the receive resume stats token of the
+ * dataset's child. It should be freed with strfree().
+ */
+char *
+get_child_receive_stats(dsl_dataset_t *ds)
+{
+ char recvname[ZFS_MAX_DATASET_NAME_LEN + 6];
+ dsl_dataset_t *recv_ds;
+ dsl_dataset_name(ds, recvname);
+ if (strlcat(recvname, "/", sizeof (recvname)) <
+ sizeof (recvname) &&
+ strlcat(recvname, recv_clone_name, sizeof (recvname)) <
+ sizeof (recvname) &&
+ dsl_dataset_hold(ds->ds_dir->dd_pool, recvname, FTAG,
+ &recv_ds) == 0) {
+ char *propval = get_receive_resume_stats_impl(recv_ds);
+ dsl_dataset_rele(recv_ds, FTAG);
+ return (propval);
+ }
+ return (strdup(""));
+}
+
+static void
+get_receive_resume_stats(dsl_dataset_t *ds, nvlist_t *nv)
+{
+ char *propval = get_receive_resume_stats_impl(ds);
+ if (strcmp(propval, "") != 0) {
+ dsl_prop_nvlist_add_string(nv,
+ ZFS_PROP_RECEIVE_RESUME_TOKEN, propval);
+ } else {
+ char *childval = get_child_receive_stats(ds);
+ if (strcmp(childval, "") != 0) {
+ dsl_prop_nvlist_add_string(nv,
+ ZFS_PROP_RECEIVE_RESUME_TOKEN, childval);
+ }
+ strfree(childval);
+ }
+ strfree(propval);
+}
+
+uint64_t
+dsl_get_refratio(dsl_dataset_t *ds)
+{
+ uint64_t ratio = dsl_dataset_phys(ds)->ds_compressed_bytes == 0 ? 100 :
+ (dsl_dataset_phys(ds)->ds_uncompressed_bytes * 100 /
+ dsl_dataset_phys(ds)->ds_compressed_bytes);
+ return (ratio);
+}
+
+uint64_t
+dsl_get_logicalreferenced(dsl_dataset_t *ds)
+{
+ return (dsl_dataset_phys(ds)->ds_uncompressed_bytes);
+}
+
+uint64_t
+dsl_get_compressratio(dsl_dataset_t *ds)
+{
+ if (ds->ds_is_snapshot) {
+ return (dsl_get_refratio(ds));
+ } else {
+ dsl_dir_t *dd = ds->ds_dir;
+ mutex_enter(&dd->dd_lock);
+ uint64_t val = dsl_dir_get_compressratio(dd);
+ mutex_exit(&dd->dd_lock);
+ return (val);
+ }
+}
+
+uint64_t
+dsl_get_used(dsl_dataset_t *ds)
+{
+ if (ds->ds_is_snapshot) {
+ return (dsl_dataset_phys(ds)->ds_unique_bytes);
+ } else {
+ dsl_dir_t *dd = ds->ds_dir;
+ mutex_enter(&dd->dd_lock);
+ uint64_t val = dsl_dir_get_used(dd);
+ mutex_exit(&dd->dd_lock);
+ return (val);
+ }
+}
+
+uint64_t
+dsl_get_creation(dsl_dataset_t *ds)
+{
+ return (dsl_dataset_phys(ds)->ds_creation_time);
+}
+
+uint64_t
+dsl_get_creationtxg(dsl_dataset_t *ds)
+{
+ return (dsl_dataset_phys(ds)->ds_creation_txg);
+}
+
+uint64_t
+dsl_get_refquota(dsl_dataset_t *ds)
+{
+ return (ds->ds_quota);
+}
+
+uint64_t
+dsl_get_refreservation(dsl_dataset_t *ds)
+{
+ return (ds->ds_reserved);
+}
+
+uint64_t
+dsl_get_guid(dsl_dataset_t *ds)
+{
+ return (dsl_dataset_phys(ds)->ds_guid);
+}
+
+uint64_t
+dsl_get_unique(dsl_dataset_t *ds)
+{
+ return (dsl_dataset_phys(ds)->ds_unique_bytes);
+}
+
+uint64_t
+dsl_get_objsetid(dsl_dataset_t *ds)
+{
+ return (ds->ds_object);
+}
+
+uint64_t
+dsl_get_userrefs(dsl_dataset_t *ds)
+{
+ return (ds->ds_userrefs);
+}
+
+uint64_t
+dsl_get_defer_destroy(dsl_dataset_t *ds)
+{
+ return (DS_IS_DEFER_DESTROY(ds) ? 1 : 0);
+}
+
+uint64_t
+dsl_get_referenced(dsl_dataset_t *ds)
+{
+ return (dsl_dataset_phys(ds)->ds_referenced_bytes);
+}
+
+uint64_t
+dsl_get_numclones(dsl_dataset_t *ds)
+{
+ ASSERT(ds->ds_is_snapshot);
+ return (dsl_dataset_phys(ds)->ds_num_children - 1);
+}
+
+uint64_t
+dsl_get_inconsistent(dsl_dataset_t *ds)
+{
+ return ((dsl_dataset_phys(ds)->ds_flags & DS_FLAG_INCONSISTENT) ?
+ 1 : 0);
+}
+
+uint64_t
+dsl_get_available(dsl_dataset_t *ds)
+{
+ uint64_t refdbytes = dsl_get_referenced(ds);
+ uint64_t availbytes = dsl_dir_space_available(ds->ds_dir,
+ NULL, 0, TRUE);
+ if (ds->ds_reserved > dsl_dataset_phys(ds)->ds_unique_bytes) {
+ availbytes +=
+ ds->ds_reserved - dsl_dataset_phys(ds)->ds_unique_bytes;
+ }
+ if (ds->ds_quota != 0) {
+ /*
+ * Adjust available bytes according to refquota
+ */
+ if (refdbytes < ds->ds_quota) {
+ availbytes = MIN(availbytes,
+ ds->ds_quota - refdbytes);
+ } else {
+ availbytes = 0;
+ }
+ }
+ return (availbytes);
+}
+
+int
+dsl_get_written(dsl_dataset_t *ds, uint64_t *written)
+{
+ dsl_pool_t *dp = ds->ds_dir->dd_pool;
+ dsl_dataset_t *prev;
+ int err = dsl_dataset_hold_obj(dp,
+ dsl_dataset_phys(ds)->ds_prev_snap_obj, FTAG, &prev);
+ if (err == 0) {
+ uint64_t comp, uncomp;
+ err = dsl_dataset_space_written(prev, ds, written,
+ &comp, &uncomp);
+ dsl_dataset_rele(prev, FTAG);
+ }
+ return (err);
+}
+
+/*
+ * 'snap' should be a buffer of size ZFS_MAX_DATASET_NAME_LEN.
+ */
+int
+dsl_get_prev_snap(dsl_dataset_t *ds, char *snap)
+{
+ dsl_pool_t *dp = ds->ds_dir->dd_pool;
+ if (ds->ds_prev != NULL && ds->ds_prev != dp->dp_origin_snap) {
+ dsl_dataset_name(ds->ds_prev, snap);
+ return (0);
+ } else {
+ return (ENOENT);
+ }
+}
- ratio = ds->ds_phys->ds_compressed_bytes == 0 ? 100 :
- (ds->ds_phys->ds_uncompressed_bytes * 100 /
- ds->ds_phys->ds_compressed_bytes);
- dsl_prop_nvlist_add_uint64(nv, ZFS_PROP_REFRATIO, ratio);
+/*
+ * Returns the mountpoint property and source for the given dataset in the value
+ * and source buffers. The value buffer must be at least as large as MAXPATHLEN
+ * and the source buffer as least as large a ZFS_MAX_DATASET_NAME_LEN.
+ * Returns 0 on success and an error on failure.
+ */
+int
+dsl_get_mountpoint(dsl_dataset_t *ds, const char *dsname, char *value,
+ char *source)
+{
+ int error;
+ dsl_pool_t *dp = ds->ds_dir->dd_pool;
+
+ /* Retrieve the mountpoint value stored in the zap opbject */
+ error = dsl_prop_get_ds(ds, zfs_prop_to_name(ZFS_PROP_MOUNTPOINT), 1,
+ ZAP_MAXVALUELEN, value, source);
+ if (error != 0) {
+ return (error);
+ }
+
+ /*
+ * Process the dsname and source to find the full mountpoint string.
+ * Can be skipped for 'legacy' or 'none'.
+ */
+ if (value[0] == '/') {
+ char *buf = kmem_alloc(ZAP_MAXVALUELEN, KM_SLEEP);
+ char *root = buf;
+ const char *relpath;
+
+ /*
+ * If we inherit the mountpoint, even from a dataset
+ * with a received value, the source will be the path of
+ * the dataset we inherit from. If source is
+ * ZPROP_SOURCE_VAL_RECVD, the received value is not
+ * inherited.
+ */
+ if (strcmp(source, ZPROP_SOURCE_VAL_RECVD) == 0) {
+ relpath = "";
+ } else {
+ ASSERT0(strncmp(dsname, source, strlen(source)));
+ relpath = dsname + strlen(source);
+ if (relpath[0] == '/')
+ relpath++;
+ }
+
+ spa_altroot(dp->dp_spa, root, ZAP_MAXVALUELEN);
+
+ /*
+ * Special case an alternate root of '/'. This will
+ * avoid having multiple leading slashes in the
+ * mountpoint path.
+ */
+ if (strcmp(root, "/") == 0)
+ root++;
- if (ds->ds_phys->ds_next_snap_obj) {
/*
- * This is a snapshot; override the dd's space used with
- * our unique space and compression ratio.
+ * If the mountpoint is '/' then skip over this
+ * if we are obtaining either an alternate root or
+ * an inherited mountpoint.
*/
- dsl_prop_nvlist_add_uint64(nv, ZFS_PROP_USED,
- ds->ds_phys->ds_unique_bytes);
- dsl_prop_nvlist_add_uint64(nv, ZFS_PROP_COMPRESSRATIO, ratio);
+ char *mnt = value;
+ if (value[1] == '\0' && (root[0] != '\0' ||
+ relpath[0] != '\0'))
+ mnt = value + 1;
+
+ if (relpath[0] == '\0') {
+ (void) snprintf(value, ZAP_MAXVALUELEN, "%s%s",
+ root, mnt);
+ } else {
+ (void) snprintf(value, ZAP_MAXVALUELEN, "%s%s%s%s",
+ root, mnt, relpath[0] == '@' ? "" : "/",
+ relpath);
+ }
+ kmem_free(buf, ZAP_MAXVALUELEN);
+ }
+
+ return (0);
+}
+
+void
+dsl_dataset_stats(dsl_dataset_t *ds, nvlist_t *nv)
+{
+ dsl_pool_t *dp = ds->ds_dir->dd_pool;
+ ASSERT(dsl_pool_config_held(dp));
+
+ dsl_prop_nvlist_add_uint64(nv, ZFS_PROP_REFRATIO,
+ dsl_get_refratio(ds));
+ dsl_prop_nvlist_add_uint64(nv, ZFS_PROP_LOGICALREFERENCED,
+ dsl_get_logicalreferenced(ds));
+ dsl_prop_nvlist_add_uint64(nv, ZFS_PROP_COMPRESSRATIO,
+ dsl_get_compressratio(ds));
+ dsl_prop_nvlist_add_uint64(nv, ZFS_PROP_USED,
+ dsl_get_used(ds));
+
+ if (ds->ds_is_snapshot) {
get_clones_stat(ds, nv);
+ } else {
+ char buf[ZFS_MAX_DATASET_NAME_LEN];
+ if (dsl_get_prev_snap(ds, buf) == 0)
+ dsl_prop_nvlist_add_string(nv, ZFS_PROP_PREV_SNAP,
+ buf);
+ dsl_dir_stats(ds->ds_dir, nv);
+ }
+
+ dsl_prop_nvlist_add_uint64(nv, ZFS_PROP_AVAILABLE,
+ dsl_get_available(ds));
+ dsl_prop_nvlist_add_uint64(nv, ZFS_PROP_REFERENCED,
+ dsl_get_referenced(ds));
+ dsl_prop_nvlist_add_uint64(nv, ZFS_PROP_CREATION,
+ dsl_get_creation(ds));
+ dsl_prop_nvlist_add_uint64(nv, ZFS_PROP_CREATETXG,
+ dsl_get_creationtxg(ds));
+ dsl_prop_nvlist_add_uint64(nv, ZFS_PROP_REFQUOTA,
+ dsl_get_refquota(ds));
+ dsl_prop_nvlist_add_uint64(nv, ZFS_PROP_REFRESERVATION,
+ dsl_get_refreservation(ds));
+ dsl_prop_nvlist_add_uint64(nv, ZFS_PROP_GUID,
+ dsl_get_guid(ds));
+ dsl_prop_nvlist_add_uint64(nv, ZFS_PROP_UNIQUE,
+ dsl_get_unique(ds));
+ dsl_prop_nvlist_add_uint64(nv, ZFS_PROP_OBJSETID,
+ dsl_get_objsetid(ds));
+ dsl_prop_nvlist_add_uint64(nv, ZFS_PROP_USERREFS,
+ dsl_get_userrefs(ds));
+ dsl_prop_nvlist_add_uint64(nv, ZFS_PROP_DEFER_DESTROY,
+ dsl_get_defer_destroy(ds));
+ dsl_dataset_crypt_stats(ds, nv);
+
+ if (dsl_dataset_phys(ds)->ds_prev_snap_obj != 0) {
+ uint64_t written;
+ if (dsl_get_written(ds, &written) == 0) {
+ dsl_prop_nvlist_add_uint64(nv, ZFS_PROP_WRITTEN,
+ written);
+ }
+ }
+
+ if (!dsl_dataset_is_snapshot(ds)) {
+ /*
+ * A failed "newfs" (e.g. full) resumable receive leaves
+ * the stats set on this dataset. Check here for the prop.
+ */
+ get_receive_resume_stats(ds, nv);
+
+ /*
+ * A failed incremental resumable receive leaves the
+ * stats set on our child named "%recv". Check the child
+ * for the prop.
+ */
+ /* 6 extra bytes for /%recv */
+ char recvname[ZFS_MAX_DATASET_NAME_LEN + 6];
+ dsl_dataset_t *recv_ds;
+ dsl_dataset_name(ds, recvname);
+ if (strlcat(recvname, "/", sizeof (recvname)) <
+ sizeof (recvname) &&
+ strlcat(recvname, recv_clone_name, sizeof (recvname)) <
+ sizeof (recvname) &&
+ dsl_dataset_hold(dp, recvname, FTAG, &recv_ds) == 0) {
+ get_receive_resume_stats(recv_ds, nv);
+ dsl_dataset_rele(recv_ds, FTAG);
+ }
}
}
void
dsl_dataset_fast_stat(dsl_dataset_t *ds, dmu_objset_stats_t *stat)
{
- stat->dds_creation_txg = ds->ds_phys->ds_creation_txg;
- stat->dds_inconsistent = ds->ds_phys->ds_flags & DS_FLAG_INCONSISTENT;
- stat->dds_guid = ds->ds_phys->ds_guid;
- if (ds->ds_phys->ds_next_snap_obj) {
+ ASSERTV(dsl_pool_t *dp = ds->ds_dir->dd_pool);
+ ASSERT(dsl_pool_config_held(dp));
+
+ stat->dds_creation_txg = dsl_get_creationtxg(ds);
+ stat->dds_inconsistent = dsl_get_inconsistent(ds);
+ stat->dds_guid = dsl_get_guid(ds);
+ stat->dds_origin[0] = '\0';
+ if (ds->ds_is_snapshot) {
stat->dds_is_snapshot = B_TRUE;
- stat->dds_num_clones = ds->ds_phys->ds_num_children - 1;
+ stat->dds_num_clones = dsl_get_numclones(ds);
} else {
stat->dds_is_snapshot = B_FALSE;
stat->dds_num_clones = 0;
- }
- /* clone origin is really a dsl_dir thing... */
- rw_enter(&ds->ds_dir->dd_pool->dp_config_rwlock, RW_READER);
- if (dsl_dir_is_clone(ds->ds_dir)) {
- dsl_dataset_t *ods;
-
- VERIFY(0 == dsl_dataset_get_ref(ds->ds_dir->dd_pool,
- ds->ds_dir->dd_phys->dd_origin_obj, FTAG, &ods));
- dsl_dataset_name(ods, stat->dds_origin);
- dsl_dataset_drop_ref(ods, FTAG);
- } else {
- stat->dds_origin[0] = '\0';
+ if (dsl_dir_is_clone(ds->ds_dir)) {
+ dsl_dir_get_origin(ds->ds_dir, stat->dds_origin);
+ }
}
- rw_exit(&ds->ds_dir->dd_pool->dp_config_rwlock);
}
uint64_t
uint64_t *refdbytesp, uint64_t *availbytesp,
uint64_t *usedobjsp, uint64_t *availobjsp)
{
- *refdbytesp = ds->ds_phys->ds_referenced_bytes;
+ *refdbytesp = dsl_dataset_phys(ds)->ds_referenced_bytes;
*availbytesp = dsl_dir_space_available(ds->ds_dir, NULL, 0, TRUE);
- if (ds->ds_reserved > ds->ds_phys->ds_unique_bytes)
- *availbytesp += ds->ds_reserved - ds->ds_phys->ds_unique_bytes;
+ if (ds->ds_reserved > dsl_dataset_phys(ds)->ds_unique_bytes)
+ *availbytesp +=
+ ds->ds_reserved - dsl_dataset_phys(ds)->ds_unique_bytes;
if (ds->ds_quota != 0) {
/*
* Adjust available bytes according to refquota
else
*availbytesp = 0;
}
- *usedobjsp = ds->ds_phys->ds_bp.blk_fill;
+ rrw_enter(&ds->ds_bp_rwlock, RW_READER, FTAG);
+ *usedobjsp = BP_GET_FILL(&dsl_dataset_phys(ds)->ds_bp);
+ rrw_exit(&ds->ds_bp_rwlock, FTAG);
*availobjsp = DN_MAX_OBJECT - *usedobjsp;
}
boolean_t
-dsl_dataset_modified_since_lastsnap(dsl_dataset_t *ds)
+dsl_dataset_modified_since_snap(dsl_dataset_t *ds, dsl_dataset_t *snap)
{
ASSERTV(dsl_pool_t *dp = ds->ds_dir->dd_pool);
+ uint64_t birth;
- ASSERT(RW_LOCK_HELD(&dp->dp_config_rwlock) ||
- dsl_pool_sync_context(dp));
- if (ds->ds_prev == NULL)
+ ASSERT(dsl_pool_config_held(dp));
+ if (snap == NULL)
return (B_FALSE);
- if (ds->ds_phys->ds_bp.blk_birth >
- ds->ds_prev->ds_phys->ds_creation_txg) {
- objset_t *os, *os_prev;
+ rrw_enter(&ds->ds_bp_rwlock, RW_READER, FTAG);
+ birth = dsl_dataset_get_blkptr(ds)->blk_birth;
+ rrw_exit(&ds->ds_bp_rwlock, FTAG);
+ if (birth > dsl_dataset_phys(snap)->ds_creation_txg) {
+ objset_t *os, *os_snap;
/*
* It may be that only the ZIL differs, because it was
* reset in the head. Don't count that as being
*/
if (dmu_objset_from_ds(ds, &os) != 0)
return (B_TRUE);
- if (dmu_objset_from_ds(ds->ds_prev, &os_prev) != 0)
+ if (dmu_objset_from_ds(snap, &os_snap) != 0)
return (B_TRUE);
return (bcmp(&os->os_phys->os_meta_dnode,
- &os_prev->os_phys->os_meta_dnode,
+ &os_snap->os_phys->os_meta_dnode,
sizeof (os->os_phys->os_meta_dnode)) != 0);
}
return (B_FALSE);
}
+typedef struct dsl_dataset_rename_snapshot_arg {
+ const char *ddrsa_fsname;
+ const char *ddrsa_oldsnapname;
+ const char *ddrsa_newsnapname;
+ boolean_t ddrsa_recursive;
+ dmu_tx_t *ddrsa_tx;
+} dsl_dataset_rename_snapshot_arg_t;
+
/* ARGSUSED */
static int
-dsl_dataset_snapshot_rename_check(void *arg1, void *arg2, dmu_tx_t *tx)
+dsl_dataset_rename_snapshot_check_impl(dsl_pool_t *dp,
+ dsl_dataset_t *hds, void *arg)
{
- dsl_dataset_t *ds = arg1;
- char *newsnapname = arg2;
- dsl_dir_t *dd = ds->ds_dir;
- dsl_dataset_t *hds;
+ dsl_dataset_rename_snapshot_arg_t *ddrsa = arg;
+ int error;
uint64_t val;
- int err;
-
- err = dsl_dataset_hold_obj(dd->dd_pool,
- dd->dd_phys->dd_head_dataset_obj, FTAG, &hds);
- if (err)
- return (err);
- /* new name better not be in use */
- err = dsl_dataset_snap_lookup(hds, newsnapname, &val);
- dsl_dataset_rele(hds, FTAG);
+ error = dsl_dataset_snap_lookup(hds, ddrsa->ddrsa_oldsnapname, &val);
+ if (error != 0) {
+ /* ignore nonexistent snapshots */
+ return (error == ENOENT ? 0 : error);
+ }
- if (err == 0)
- err = EEXIST;
- else if (err == ENOENT)
- err = 0;
+ /* new name should not exist */
+ error = dsl_dataset_snap_lookup(hds, ddrsa->ddrsa_newsnapname, &val);
+ if (error == 0)
+ error = SET_ERROR(EEXIST);
+ else if (error == ENOENT)
+ error = 0;
/* dataset name + 1 for the "@" + the new snapshot name must fit */
- if (dsl_dir_namelen(ds->ds_dir) + 1 + strlen(newsnapname) >= MAXNAMELEN)
- err = ENAMETOOLONG;
+ if (dsl_dir_namelen(hds->ds_dir) + 1 +
+ strlen(ddrsa->ddrsa_newsnapname) >= ZFS_MAX_DATASET_NAME_LEN)
+ error = SET_ERROR(ENAMETOOLONG);
- return (err);
+ return (error);
}
-static void
-dsl_dataset_snapshot_rename_sync(void *arg1, void *arg2, dmu_tx_t *tx)
+static int
+dsl_dataset_rename_snapshot_check(void *arg, dmu_tx_t *tx)
{
- dsl_dataset_t *ds = arg1;
- const char *newsnapname = arg2;
- dsl_dir_t *dd = ds->ds_dir;
- objset_t *mos = dd->dd_pool->dp_meta_objset;
+ dsl_dataset_rename_snapshot_arg_t *ddrsa = arg;
+ dsl_pool_t *dp = dmu_tx_pool(tx);
dsl_dataset_t *hds;
- int err;
+ int error;
+
+ error = dsl_dataset_hold(dp, ddrsa->ddrsa_fsname, FTAG, &hds);
+ if (error != 0)
+ return (error);
+
+ if (ddrsa->ddrsa_recursive) {
+ error = dmu_objset_find_dp(dp, hds->ds_dir->dd_object,
+ dsl_dataset_rename_snapshot_check_impl, ddrsa,
+ DS_FIND_CHILDREN);
+ } else {
+ error = dsl_dataset_rename_snapshot_check_impl(dp, hds, ddrsa);
+ }
+ dsl_dataset_rele(hds, FTAG);
+ return (error);
+}
+
+static int
+dsl_dataset_rename_snapshot_sync_impl(dsl_pool_t *dp,
+ dsl_dataset_t *hds, void *arg)
+{
+ dsl_dataset_rename_snapshot_arg_t *ddrsa = arg;
+ dsl_dataset_t *ds;
+ uint64_t val;
+ dmu_tx_t *tx = ddrsa->ddrsa_tx;
+ int error;
+
+ error = dsl_dataset_snap_lookup(hds, ddrsa->ddrsa_oldsnapname, &val);
+ ASSERT(error == 0 || error == ENOENT);
+ if (error == ENOENT) {
+ /* ignore nonexistent snapshots */
+ return (0);
+ }
- ASSERT(ds->ds_phys->ds_next_snap_obj != 0);
+ VERIFY0(dsl_dataset_hold_obj(dp, val, FTAG, &ds));
- VERIFY(0 == dsl_dataset_hold_obj(dd->dd_pool,
- dd->dd_phys->dd_head_dataset_obj, FTAG, &hds));
+ /* log before we change the name */
+ spa_history_log_internal_ds(ds, "rename", tx,
+ "-> @%s", ddrsa->ddrsa_newsnapname);
- VERIFY(0 == dsl_dataset_get_snapname(ds));
- err = dsl_dataset_snap_remove(hds, ds->ds_snapname, tx);
- ASSERT3U(err, ==, 0);
+ VERIFY0(dsl_dataset_snap_remove(hds, ddrsa->ddrsa_oldsnapname, tx,
+ B_FALSE));
mutex_enter(&ds->ds_lock);
- (void) strcpy(ds->ds_snapname, newsnapname);
+ (void) strlcpy(ds->ds_snapname, ddrsa->ddrsa_newsnapname,
+ sizeof (ds->ds_snapname));
mutex_exit(&ds->ds_lock);
- err = zap_add(mos, hds->ds_phys->ds_snapnames_zapobj,
- ds->ds_snapname, 8, 1, &ds->ds_object, tx);
- ASSERT3U(err, ==, 0);
+ VERIFY0(zap_add(dp->dp_meta_objset,
+ dsl_dataset_phys(hds)->ds_snapnames_zapobj,
+ ds->ds_snapname, 8, 1, &ds->ds_object, tx));
+ zvol_rename_minors(dp->dp_spa, ddrsa->ddrsa_oldsnapname,
+ ddrsa->ddrsa_newsnapname, B_TRUE);
+
+ dsl_dataset_rele(ds, FTAG);
+ return (0);
+}
- spa_history_log_internal(LOG_DS_RENAME, dd->dd_pool->dp_spa, tx,
- "dataset = %llu", ds->ds_object);
+static void
+dsl_dataset_rename_snapshot_sync(void *arg, dmu_tx_t *tx)
+{
+ dsl_dataset_rename_snapshot_arg_t *ddrsa = arg;
+ dsl_pool_t *dp = dmu_tx_pool(tx);
+ dsl_dataset_t *hds = NULL;
+
+ VERIFY0(dsl_dataset_hold(dp, ddrsa->ddrsa_fsname, FTAG, &hds));
+ ddrsa->ddrsa_tx = tx;
+ if (ddrsa->ddrsa_recursive) {
+ VERIFY0(dmu_objset_find_dp(dp, hds->ds_dir->dd_object,
+ dsl_dataset_rename_snapshot_sync_impl, ddrsa,
+ DS_FIND_CHILDREN));
+ } else {
+ VERIFY0(dsl_dataset_rename_snapshot_sync_impl(dp, hds, ddrsa));
+ }
dsl_dataset_rele(hds, FTAG);
}
-struct renamesnaparg {
- dsl_sync_task_group_t *dstg;
- char failed[MAXPATHLEN];
- char *oldsnap;
- char *newsnap;
-};
+int
+dsl_dataset_rename_snapshot(const char *fsname,
+ const char *oldsnapname, const char *newsnapname, boolean_t recursive)
+{
+ dsl_dataset_rename_snapshot_arg_t ddrsa;
+
+ ddrsa.ddrsa_fsname = fsname;
+ ddrsa.ddrsa_oldsnapname = oldsnapname;
+ ddrsa.ddrsa_newsnapname = newsnapname;
+ ddrsa.ddrsa_recursive = recursive;
+
+ return (dsl_sync_task(fsname, dsl_dataset_rename_snapshot_check,
+ dsl_dataset_rename_snapshot_sync, &ddrsa,
+ 1, ZFS_SPACE_CHECK_RESERVED));
+}
+/*
+ * If we're doing an ownership handoff, we need to make sure that there is
+ * only one long hold on the dataset. We're not allowed to change anything here
+ * so we don't permanently release the long hold or regular hold here. We want
+ * to do this only when syncing to avoid the dataset unexpectedly going away
+ * when we release the long hold.
+ */
static int
-dsl_snapshot_rename_one(const char *name, void *arg)
+dsl_dataset_handoff_check(dsl_dataset_t *ds, void *owner, dmu_tx_t *tx)
{
- struct renamesnaparg *ra = arg;
- dsl_dataset_t *ds = NULL;
- char *snapname;
- int err;
+ boolean_t held;
- snapname = kmem_asprintf("%s@%s", name, ra->oldsnap);
- (void) strlcpy(ra->failed, snapname, sizeof (ra->failed));
+ if (!dmu_tx_is_syncing(tx))
+ return (0);
- /*
- * For recursive snapshot renames the parent won't be changing
- * so we just pass name for both the to/from argument.
- */
- err = zfs_secpolicy_rename_perms(snapname, snapname, CRED());
- if (err != 0) {
- strfree(snapname);
- return (err == ENOENT ? 0 : err);
+ if (owner != NULL) {
+ VERIFY3P(ds->ds_owner, ==, owner);
+ dsl_dataset_long_rele(ds, owner);
}
-#ifdef _KERNEL
- /*
- * For all filesystems undergoing rename, we'll need to unmount it.
- */
- (void) zfs_unmount_snap(snapname, NULL);
-#endif
- err = dsl_dataset_hold(snapname, ra->dstg, &ds);
- strfree(snapname);
- if (err != 0)
- return (err == ENOENT ? 0 : err);
+ held = dsl_dataset_long_held(ds);
+
+ if (owner != NULL)
+ dsl_dataset_long_hold(ds, owner);
- dsl_sync_task_create(ra->dstg, dsl_dataset_snapshot_rename_check,
- dsl_dataset_snapshot_rename_sync, ds, ra->newsnap, 0);
+ if (held)
+ return (SET_ERROR(EBUSY));
return (0);
}
-static int
-dsl_recursive_rename(char *oldname, const char *newname)
+int
+dsl_dataset_rollback_check(void *arg, dmu_tx_t *tx)
{
- int err;
- struct renamesnaparg *ra;
- dsl_sync_task_t *dst;
- spa_t *spa;
- char *cp, *fsname = spa_strdup(oldname);
- int len = strlen(oldname) + 1;
+ dsl_dataset_rollback_arg_t *ddra = arg;
+ dsl_pool_t *dp = dmu_tx_pool(tx);
+ dsl_dataset_t *ds;
+ int64_t unused_refres_delta;
+ int error;
- /* truncate the snapshot name to get the fsname */
- cp = strchr(fsname, '@');
- *cp = '\0';
+ error = dsl_dataset_hold(dp, ddra->ddra_fsname, FTAG, &ds);
+ if (error != 0)
+ return (error);
- err = spa_open(fsname, &spa, FTAG);
- if (err) {
- kmem_free(fsname, len);
- return (err);
+ /* must not be a snapshot */
+ if (ds->ds_is_snapshot) {
+ dsl_dataset_rele(ds, FTAG);
+ return (SET_ERROR(EINVAL));
+ }
+
+ /* must have a most recent snapshot */
+ if (dsl_dataset_phys(ds)->ds_prev_snap_txg < TXG_INITIAL) {
+ dsl_dataset_rele(ds, FTAG);
+ return (SET_ERROR(ESRCH));
+ }
+
+ /*
+ * No rollback to a snapshot created in the current txg, because
+ * the rollback may dirty the dataset and create blocks that are
+ * not reachable from the rootbp while having a birth txg that
+ * falls into the snapshot's range.
+ */
+ if (dmu_tx_is_syncing(tx) &&
+ dsl_dataset_phys(ds)->ds_prev_snap_txg >= tx->tx_txg) {
+ dsl_dataset_rele(ds, FTAG);
+ return (SET_ERROR(EAGAIN));
}
- ra = kmem_alloc(sizeof (struct renamesnaparg), KM_SLEEP);
- ra->dstg = dsl_sync_task_group_create(spa_get_dsl(spa));
- ra->oldsnap = strchr(oldname, '@') + 1;
- ra->newsnap = strchr(newname, '@') + 1;
- *ra->failed = '\0';
+ /*
+ * If the expected target snapshot is specified, then check that
+ * the latest snapshot is it.
+ */
+ if (ddra->ddra_tosnap != NULL) {
+ dsl_dataset_t *snapds;
- err = dmu_objset_find(fsname, dsl_snapshot_rename_one, ra,
- DS_FIND_CHILDREN);
- kmem_free(fsname, len);
+ /* Check if the target snapshot exists at all. */
+ error = dsl_dataset_hold(dp, ddra->ddra_tosnap, FTAG, &snapds);
+ if (error != 0) {
+ /*
+ * ESRCH is used to signal that the target snapshot does
+ * not exist, while ENOENT is used to report that
+ * the rolled back dataset does not exist.
+ * ESRCH is also used to cover other cases where the
+ * target snapshot is not related to the dataset being
+ * rolled back such as being in a different pool.
+ */
+ if (error == ENOENT || error == EXDEV)
+ error = SET_ERROR(ESRCH);
+ dsl_dataset_rele(ds, FTAG);
+ return (error);
+ }
+ ASSERT(snapds->ds_is_snapshot);
- if (err == 0) {
- err = dsl_sync_task_group_wait(ra->dstg);
+ /* Check if the snapshot is the latest snapshot indeed. */
+ if (snapds != ds->ds_prev) {
+ /*
+ * Distinguish between the case where the only problem
+ * is intervening snapshots (EEXIST) vs the snapshot
+ * not being a valid target for rollback (ESRCH).
+ */
+ if (snapds->ds_dir == ds->ds_dir ||
+ (dsl_dir_is_clone(ds->ds_dir) &&
+ dsl_dir_phys(ds->ds_dir)->dd_origin_obj ==
+ snapds->ds_object)) {
+ error = SET_ERROR(EEXIST);
+ } else {
+ error = SET_ERROR(ESRCH);
+ }
+ dsl_dataset_rele(snapds, FTAG);
+ dsl_dataset_rele(ds, FTAG);
+ return (error);
+ }
+ dsl_dataset_rele(snapds, FTAG);
}
- for (dst = list_head(&ra->dstg->dstg_tasks); dst;
- dst = list_next(&ra->dstg->dstg_tasks, dst)) {
- dsl_dataset_t *ds = dst->dst_arg1;
- if (dst->dst_err) {
- dsl_dir_name(ds->ds_dir, ra->failed);
- (void) strlcat(ra->failed, "@", sizeof (ra->failed));
- (void) strlcat(ra->failed, ra->newsnap,
- sizeof (ra->failed));
+ /* must not have any bookmarks after the most recent snapshot */
+ nvlist_t *proprequest = fnvlist_alloc();
+ fnvlist_add_boolean(proprequest, zfs_prop_to_name(ZFS_PROP_CREATETXG));
+ nvlist_t *bookmarks = fnvlist_alloc();
+ error = dsl_get_bookmarks_impl(ds, proprequest, bookmarks);
+ fnvlist_free(proprequest);
+ if (error != 0) {
+ dsl_dataset_rele(ds, FTAG);
+ return (error);
+ }
+ for (nvpair_t *pair = nvlist_next_nvpair(bookmarks, NULL);
+ pair != NULL; pair = nvlist_next_nvpair(bookmarks, pair)) {
+ nvlist_t *valuenv =
+ fnvlist_lookup_nvlist(fnvpair_value_nvlist(pair),
+ zfs_prop_to_name(ZFS_PROP_CREATETXG));
+ uint64_t createtxg = fnvlist_lookup_uint64(valuenv, "value");
+ if (createtxg > dsl_dataset_phys(ds)->ds_prev_snap_txg) {
+ fnvlist_free(bookmarks);
+ dsl_dataset_rele(ds, FTAG);
+ return (SET_ERROR(EEXIST));
}
- dsl_dataset_rele(ds, ra->dstg);
}
+ fnvlist_free(bookmarks);
- if (err)
- (void) strlcpy(oldname, ra->failed, sizeof (ra->failed));
+ error = dsl_dataset_handoff_check(ds, ddra->ddra_owner, tx);
+ if (error != 0) {
+ dsl_dataset_rele(ds, FTAG);
+ return (error);
+ }
- dsl_sync_task_group_destroy(ra->dstg);
- kmem_free(ra, sizeof (struct renamesnaparg));
- spa_close(spa, FTAG);
- return (err);
-}
+ /*
+ * Check if the snap we are rolling back to uses more than
+ * the refquota.
+ */
+ if (ds->ds_quota != 0 &&
+ dsl_dataset_phys(ds->ds_prev)->ds_referenced_bytes > ds->ds_quota) {
+ dsl_dataset_rele(ds, FTAG);
+ return (SET_ERROR(EDQUOT));
+ }
-static int
-dsl_valid_rename(const char *oldname, void *arg)
-{
- int delta = *(int *)arg;
+ /*
+ * When we do the clone swap, we will temporarily use more space
+ * due to the refreservation (the head will no longer have any
+ * unique space, so the entire amount of the refreservation will need
+ * to be free). We will immediately destroy the clone, freeing
+ * this space, but the freeing happens over many txg's.
+ */
+ unused_refres_delta = (int64_t)MIN(ds->ds_reserved,
+ dsl_dataset_phys(ds)->ds_unique_bytes);
- if (strlen(oldname) + delta >= MAXNAMELEN)
- return (ENAMETOOLONG);
+ if (unused_refres_delta > 0 &&
+ unused_refres_delta >
+ dsl_dir_space_available(ds->ds_dir, NULL, 0, TRUE)) {
+ dsl_dataset_rele(ds, FTAG);
+ return (SET_ERROR(ENOSPC));
+ }
+ dsl_dataset_rele(ds, FTAG);
return (0);
}
-#pragma weak dmu_objset_rename = dsl_dataset_rename
-int
-dsl_dataset_rename(char *oldname, const char *newname, boolean_t recursive)
+void
+dsl_dataset_rollback_sync(void *arg, dmu_tx_t *tx)
{
- dsl_dir_t *dd;
- dsl_dataset_t *ds;
- const char *tail;
- int err;
+ dsl_dataset_rollback_arg_t *ddra = arg;
+ dsl_pool_t *dp = dmu_tx_pool(tx);
+ dsl_dataset_t *ds, *clone;
+ uint64_t cloneobj;
+ char namebuf[ZFS_MAX_DATASET_NAME_LEN];
- err = dsl_dir_open(oldname, FTAG, &dd, &tail);
- if (err)
- return (err);
+ VERIFY0(dsl_dataset_hold(dp, ddra->ddra_fsname, FTAG, &ds));
- if (tail == NULL) {
- int delta = strlen(newname) - strlen(oldname);
+ dsl_dataset_name(ds->ds_prev, namebuf);
+ fnvlist_add_string(ddra->ddra_result, "target", namebuf);
- /* if we're growing, validate child name lengths */
- if (delta > 0)
- err = dmu_objset_find(oldname, dsl_valid_rename,
- &delta, DS_FIND_CHILDREN | DS_FIND_SNAPSHOTS);
+ cloneobj = dsl_dataset_create_sync(ds->ds_dir, "%rollback",
+ ds->ds_prev, DS_CREATE_FLAG_NODIRTY, kcred, NULL, tx);
- if (err == 0)
- err = dsl_dir_rename(dd, newname);
- dsl_dir_close(dd, FTAG);
- return (err);
- }
+ VERIFY0(dsl_dataset_hold_obj(dp, cloneobj, FTAG, &clone));
- if (tail[0] != '@') {
- /* the name ended in a nonexistent component */
- dsl_dir_close(dd, FTAG);
- return (ENOENT);
- }
-
- dsl_dir_close(dd, FTAG);
+ dsl_dataset_clone_swap_sync_impl(clone, ds, tx);
+ dsl_dataset_zero_zil(ds, tx);
- /* new name must be snapshot in same filesystem */
- tail = strchr(newname, '@');
- if (tail == NULL)
- return (EINVAL);
- tail++;
- if (strncmp(oldname, newname, tail - newname) != 0)
- return (EXDEV);
+ dsl_destroy_head_sync_impl(clone, tx);
- if (recursive) {
- err = dsl_recursive_rename(oldname, newname);
- } else {
- err = dsl_dataset_hold(oldname, FTAG, &ds);
- if (err)
- return (err);
+ dsl_dataset_rele(clone, FTAG);
+ dsl_dataset_rele(ds, FTAG);
+}
- err = dsl_sync_task_do(ds->ds_dir->dd_pool,
- dsl_dataset_snapshot_rename_check,
- dsl_dataset_snapshot_rename_sync, ds, (char *)tail, 1);
+/*
+ * Rolls back the given filesystem or volume to the most recent snapshot.
+ * The name of the most recent snapshot will be returned under key "target"
+ * in the result nvlist.
+ *
+ * If owner != NULL:
+ * - The existing dataset MUST be owned by the specified owner at entry
+ * - Upon return, dataset will still be held by the same owner, whether we
+ * succeed or not.
+ *
+ * This mode is required any time the existing filesystem is mounted. See
+ * notes above zfs_suspend_fs() for further details.
+ */
+int
+dsl_dataset_rollback(const char *fsname, const char *tosnap, void *owner,
+ nvlist_t *result)
+{
+ dsl_dataset_rollback_arg_t ddra;
- dsl_dataset_rele(ds, FTAG);
- }
+ ddra.ddra_fsname = fsname;
+ ddra.ddra_tosnap = tosnap;
+ ddra.ddra_owner = owner;
+ ddra.ddra_result = result;
- return (err);
+ return (dsl_sync_task(fsname, dsl_dataset_rollback_check,
+ dsl_dataset_rollback_sync, &ddra,
+ 1, ZFS_SPACE_CHECK_RESERVED));
}
struct promotenode {
dsl_dataset_t *ds;
};
-struct promotearg {
- list_t shared_snaps, origin_snaps, clone_snaps;
- dsl_dataset_t *origin_origin;
- uint64_t used, comp, uncomp, unique, cloneusedsnap, originusedsnap;
- char *err_ds;
-};
-
static int snaplist_space(list_t *l, uint64_t mintxg, uint64_t *spacep);
+static int promote_hold(dsl_dataset_promote_arg_t *ddpa, dsl_pool_t *dp,
+ void *tag);
+static void promote_rele(dsl_dataset_promote_arg_t *ddpa, void *tag);
-static int
-dsl_dataset_promote_check(void *arg1, void *arg2, dmu_tx_t *tx)
+int
+dsl_dataset_promote_check(void *arg, dmu_tx_t *tx)
{
- dsl_dataset_t *hds = arg1;
- struct promotearg *pa = arg2;
- struct promotenode *snap = list_head(&pa->shared_snaps);
- dsl_dataset_t *origin_ds = snap->ds;
+ dsl_dataset_promote_arg_t *ddpa = arg;
+ dsl_pool_t *dp = dmu_tx_pool(tx);
+ dsl_dataset_t *hds;
+ struct promotenode *snap;
+ dsl_dataset_t *origin_ds;
int err;
uint64_t unused;
+ uint64_t ss_mv_cnt;
+ size_t max_snap_len;
+ boolean_t conflicting_snaps;
- /* Check that it is a real clone */
- if (!dsl_dir_is_clone(hds->ds_dir))
- return (EINVAL);
+ err = promote_hold(ddpa, dp, FTAG);
+ if (err != 0)
+ return (err);
- /* Since this is so expensive, don't do the preliminary check */
- if (!dmu_tx_is_syncing(tx))
- return (0);
+ hds = ddpa->ddpa_clone;
+ max_snap_len = MAXNAMELEN - strlen(ddpa->ddpa_clonename) - 1;
- if (hds->ds_phys->ds_flags & DS_FLAG_NOPROMOTE)
- return (EXDEV);
+ if (dsl_dataset_phys(hds)->ds_flags & DS_FLAG_NOPROMOTE) {
+ promote_rele(ddpa, FTAG);
+ return (SET_ERROR(EXDEV));
+ }
+
+ snap = list_head(&ddpa->shared_snaps);
+ if (snap == NULL) {
+ err = SET_ERROR(ENOENT);
+ goto out;
+ }
+ origin_ds = snap->ds;
+
+ /*
+ * Encrypted clones share a DSL Crypto Key with their origin's dsl dir.
+ * When doing a promote we must make sure the encryption root for
+ * both the target and the target's origin does not change to avoid
+ * needing to rewrap encryption keys
+ */
+ err = dsl_dataset_promote_crypt_check(hds->ds_dir, origin_ds->ds_dir);
+ if (err != 0)
+ goto out;
+
+ /*
+ * Compute and check the amount of space to transfer. Since this is
+ * so expensive, don't do the preliminary check.
+ */
+ if (!dmu_tx_is_syncing(tx)) {
+ promote_rele(ddpa, FTAG);
+ return (0);
+ }
/* compute origin's new unique space */
- snap = list_tail(&pa->clone_snaps);
- ASSERT3U(snap->ds->ds_phys->ds_prev_snap_obj, ==, origin_ds->ds_object);
+ snap = list_tail(&ddpa->clone_snaps);
+ ASSERT(snap != NULL);
+ ASSERT3U(dsl_dataset_phys(snap->ds)->ds_prev_snap_obj, ==,
+ origin_ds->ds_object);
dsl_deadlist_space_range(&snap->ds->ds_deadlist,
- origin_ds->ds_phys->ds_prev_snap_txg, UINT64_MAX,
- &pa->unique, &unused, &unused);
+ dsl_dataset_phys(origin_ds)->ds_prev_snap_txg, UINT64_MAX,
+ &ddpa->unique, &unused, &unused);
/*
* Walk the snapshots that we are moving
*
* Compute space to transfer. Consider the incremental changes
- * to used for each snapshot:
+ * to used by each snapshot:
* (my used) = (prev's used) + (blocks born) - (blocks killed)
* So each snapshot gave birth to:
* (blocks born) = (my used) - (prev's used) + (blocks killed)
* Note however, if we stop before we reach the ORIGIN we get:
* uN + kN + kN-1 + ... + kM - uM-1
*/
- pa->used = origin_ds->ds_phys->ds_referenced_bytes;
- pa->comp = origin_ds->ds_phys->ds_compressed_bytes;
- pa->uncomp = origin_ds->ds_phys->ds_uncompressed_bytes;
- for (snap = list_head(&pa->shared_snaps); snap;
- snap = list_next(&pa->shared_snaps, snap)) {
+ conflicting_snaps = B_FALSE;
+ ss_mv_cnt = 0;
+ ddpa->used = dsl_dataset_phys(origin_ds)->ds_referenced_bytes;
+ ddpa->comp = dsl_dataset_phys(origin_ds)->ds_compressed_bytes;
+ ddpa->uncomp = dsl_dataset_phys(origin_ds)->ds_uncompressed_bytes;
+ for (snap = list_head(&ddpa->shared_snaps); snap;
+ snap = list_next(&ddpa->shared_snaps, snap)) {
uint64_t val, dlused, dlcomp, dluncomp;
dsl_dataset_t *ds = snap->ds;
+ ss_mv_cnt++;
+
+ /*
+ * If there are long holds, we won't be able to evict
+ * the objset.
+ */
+ if (dsl_dataset_long_held(ds)) {
+ err = SET_ERROR(EBUSY);
+ goto out;
+ }
+
/* Check that the snapshot name does not conflict */
- VERIFY(0 == dsl_dataset_get_snapname(ds));
+ VERIFY0(dsl_dataset_get_snapname(ds));
+ if (strlen(ds->ds_snapname) >= max_snap_len) {
+ err = SET_ERROR(ENAMETOOLONG);
+ goto out;
+ }
err = dsl_dataset_snap_lookup(hds, ds->ds_snapname, &val);
if (err == 0) {
- err = EEXIST;
+ fnvlist_add_boolean(ddpa->err_ds,
+ snap->ds->ds_snapname);
+ conflicting_snaps = B_TRUE;
+ } else if (err != ENOENT) {
goto out;
}
- if (err != ENOENT)
- goto out;
/* The very first snapshot does not have a deadlist */
- if (ds->ds_phys->ds_prev_snap_obj == 0)
+ if (dsl_dataset_phys(ds)->ds_prev_snap_obj == 0)
continue;
dsl_deadlist_space(&ds->ds_deadlist,
&dlused, &dlcomp, &dluncomp);
- pa->used += dlused;
- pa->comp += dlcomp;
- pa->uncomp += dluncomp;
+ ddpa->used += dlused;
+ ddpa->comp += dlcomp;
+ ddpa->uncomp += dluncomp;
+ }
+
+ /*
+ * In order to return the full list of conflicting snapshots, we check
+ * whether there was a conflict after traversing all of them.
+ */
+ if (conflicting_snaps) {
+ err = SET_ERROR(EEXIST);
+ goto out;
}
/*
* If we are a clone of a clone then we never reached ORIGIN,
* so we need to subtract out the clone origin's used space.
*/
- if (pa->origin_origin) {
- pa->used -= pa->origin_origin->ds_phys->ds_referenced_bytes;
- pa->comp -= pa->origin_origin->ds_phys->ds_compressed_bytes;
- pa->uncomp -= pa->origin_origin->ds_phys->ds_uncompressed_bytes;
+ if (ddpa->origin_origin) {
+ ddpa->used -=
+ dsl_dataset_phys(ddpa->origin_origin)->ds_referenced_bytes;
+ ddpa->comp -=
+ dsl_dataset_phys(ddpa->origin_origin)->ds_compressed_bytes;
+ ddpa->uncomp -=
+ dsl_dataset_phys(ddpa->origin_origin)->
+ ds_uncompressed_bytes;
}
- /* Check that there is enough space here */
+ /* Check that there is enough space and limit headroom here */
err = dsl_dir_transfer_possible(origin_ds->ds_dir, hds->ds_dir,
- pa->used);
- if (err)
- return (err);
+ 0, ss_mv_cnt, ddpa->used, ddpa->cr);
+ if (err != 0)
+ goto out;
/*
* Compute the amounts of space that will be used by snapshots
* it is the amount of space that will be on all of their
* deadlists (that was not born before their new origin).
*/
- if (hds->ds_dir->dd_phys->dd_flags & DD_FLAG_USED_BREAKDOWN) {
+ if (dsl_dir_phys(hds->ds_dir)->dd_flags & DD_FLAG_USED_BREAKDOWN) {
uint64_t space;
/*
* calls will be fast because they do not have to
* iterate over all bps.
*/
- snap = list_head(&pa->origin_snaps);
- err = snaplist_space(&pa->shared_snaps,
- snap->ds->ds_dir->dd_origin_txg, &pa->cloneusedsnap);
- if (err)
- return (err);
+ snap = list_head(&ddpa->origin_snaps);
+ if (snap == NULL) {
+ err = SET_ERROR(ENOENT);
+ goto out;
+ }
+ err = snaplist_space(&ddpa->shared_snaps,
+ snap->ds->ds_dir->dd_origin_txg, &ddpa->cloneusedsnap);
+ if (err != 0)
+ goto out;
- err = snaplist_space(&pa->clone_snaps,
+ err = snaplist_space(&ddpa->clone_snaps,
snap->ds->ds_dir->dd_origin_txg, &space);
- if (err)
- return (err);
- pa->cloneusedsnap += space;
+ if (err != 0)
+ goto out;
+ ddpa->cloneusedsnap += space;
}
- if (origin_ds->ds_dir->dd_phys->dd_flags & DD_FLAG_USED_BREAKDOWN) {
- err = snaplist_space(&pa->origin_snaps,
- origin_ds->ds_phys->ds_creation_txg, &pa->originusedsnap);
- if (err)
- return (err);
+ if (dsl_dir_phys(origin_ds->ds_dir)->dd_flags &
+ DD_FLAG_USED_BREAKDOWN) {
+ err = snaplist_space(&ddpa->origin_snaps,
+ dsl_dataset_phys(origin_ds)->ds_creation_txg,
+ &ddpa->originusedsnap);
+ if (err != 0)
+ goto out;
}
- return (0);
out:
- pa->err_ds = snap->ds->ds_snapname;
+ promote_rele(ddpa, FTAG);
return (err);
}
-static void
-dsl_dataset_promote_sync(void *arg1, void *arg2, dmu_tx_t *tx)
+void
+dsl_dataset_promote_sync(void *arg, dmu_tx_t *tx)
{
- dsl_dataset_t *hds = arg1;
- struct promotearg *pa = arg2;
- struct promotenode *snap = list_head(&pa->shared_snaps);
- dsl_dataset_t *origin_ds = snap->ds;
+ dsl_dataset_promote_arg_t *ddpa = arg;
+ dsl_pool_t *dp = dmu_tx_pool(tx);
+ dsl_dataset_t *hds;
+ struct promotenode *snap;
+ dsl_dataset_t *origin_ds;
dsl_dataset_t *origin_head;
- dsl_dir_t *dd = hds->ds_dir;
- dsl_pool_t *dp = hds->ds_dir->dd_pool;
+ dsl_dir_t *dd;
dsl_dir_t *odd = NULL;
uint64_t oldnext_obj;
int64_t delta;
- ASSERT(0 == (hds->ds_phys->ds_flags & DS_FLAG_NOPROMOTE));
+ VERIFY0(promote_hold(ddpa, dp, FTAG));
+ hds = ddpa->ddpa_clone;
+
+ ASSERT0(dsl_dataset_phys(hds)->ds_flags & DS_FLAG_NOPROMOTE);
- snap = list_head(&pa->origin_snaps);
+ snap = list_head(&ddpa->shared_snaps);
+ origin_ds = snap->ds;
+ dd = hds->ds_dir;
+
+ snap = list_head(&ddpa->origin_snaps);
origin_head = snap->ds;
/*
* We need to explicitly open odd, since origin_ds's dd will be
* changing.
*/
- VERIFY(0 == dsl_dir_open_obj(dp, origin_ds->ds_dir->dd_object,
+ VERIFY0(dsl_dir_hold_obj(dp, origin_ds->ds_dir->dd_object,
NULL, FTAG, &odd));
+ dsl_dataset_promote_crypt_sync(hds->ds_dir, odd, tx);
+
/* change origin's next snap */
dmu_buf_will_dirty(origin_ds->ds_dbuf, tx);
- oldnext_obj = origin_ds->ds_phys->ds_next_snap_obj;
- snap = list_tail(&pa->clone_snaps);
- ASSERT3U(snap->ds->ds_phys->ds_prev_snap_obj, ==, origin_ds->ds_object);
- origin_ds->ds_phys->ds_next_snap_obj = snap->ds->ds_object;
+ oldnext_obj = dsl_dataset_phys(origin_ds)->ds_next_snap_obj;
+ snap = list_tail(&ddpa->clone_snaps);
+ ASSERT3U(dsl_dataset_phys(snap->ds)->ds_prev_snap_obj, ==,
+ origin_ds->ds_object);
+ dsl_dataset_phys(origin_ds)->ds_next_snap_obj = snap->ds->ds_object;
/* change the origin's next clone */
- if (origin_ds->ds_phys->ds_next_clones_obj) {
- remove_from_next_clones(origin_ds, snap->ds->ds_object, tx);
- VERIFY3U(0, ==, zap_add_int(dp->dp_meta_objset,
- origin_ds->ds_phys->ds_next_clones_obj,
+ if (dsl_dataset_phys(origin_ds)->ds_next_clones_obj) {
+ dsl_dataset_remove_from_next_clones(origin_ds,
+ snap->ds->ds_object, tx);
+ VERIFY0(zap_add_int(dp->dp_meta_objset,
+ dsl_dataset_phys(origin_ds)->ds_next_clones_obj,
oldnext_obj, tx));
}
/* change origin */
dmu_buf_will_dirty(dd->dd_dbuf, tx);
- ASSERT3U(dd->dd_phys->dd_origin_obj, ==, origin_ds->ds_object);
- dd->dd_phys->dd_origin_obj = odd->dd_phys->dd_origin_obj;
+ ASSERT3U(dsl_dir_phys(dd)->dd_origin_obj, ==, origin_ds->ds_object);
+ dsl_dir_phys(dd)->dd_origin_obj = dsl_dir_phys(odd)->dd_origin_obj;
dd->dd_origin_txg = origin_head->ds_dir->dd_origin_txg;
dmu_buf_will_dirty(odd->dd_dbuf, tx);
- odd->dd_phys->dd_origin_obj = origin_ds->ds_object;
+ dsl_dir_phys(odd)->dd_origin_obj = origin_ds->ds_object;
origin_head->ds_dir->dd_origin_txg =
- origin_ds->ds_phys->ds_creation_txg;
+ dsl_dataset_phys(origin_ds)->ds_creation_txg;
/* change dd_clone entries */
if (spa_version(dp->dp_spa) >= SPA_VERSION_DIR_CLONES) {
- VERIFY3U(0, ==, zap_remove_int(dp->dp_meta_objset,
- odd->dd_phys->dd_clones, hds->ds_object, tx));
- VERIFY3U(0, ==, zap_add_int(dp->dp_meta_objset,
- pa->origin_origin->ds_dir->dd_phys->dd_clones,
+ VERIFY0(zap_remove_int(dp->dp_meta_objset,
+ dsl_dir_phys(odd)->dd_clones, hds->ds_object, tx));
+ VERIFY0(zap_add_int(dp->dp_meta_objset,
+ dsl_dir_phys(ddpa->origin_origin->ds_dir)->dd_clones,
hds->ds_object, tx));
- VERIFY3U(0, ==, zap_remove_int(dp->dp_meta_objset,
- pa->origin_origin->ds_dir->dd_phys->dd_clones,
+ VERIFY0(zap_remove_int(dp->dp_meta_objset,
+ dsl_dir_phys(ddpa->origin_origin->ds_dir)->dd_clones,
origin_head->ds_object, tx));
- if (dd->dd_phys->dd_clones == 0) {
- dd->dd_phys->dd_clones = zap_create(dp->dp_meta_objset,
- DMU_OT_DSL_CLONES, DMU_OT_NONE, 0, tx);
+ if (dsl_dir_phys(dd)->dd_clones == 0) {
+ dsl_dir_phys(dd)->dd_clones =
+ zap_create(dp->dp_meta_objset, DMU_OT_DSL_CLONES,
+ DMU_OT_NONE, 0, tx);
}
- VERIFY3U(0, ==, zap_add_int(dp->dp_meta_objset,
- dd->dd_phys->dd_clones, origin_head->ds_object, tx));
-
+ VERIFY0(zap_add_int(dp->dp_meta_objset,
+ dsl_dir_phys(dd)->dd_clones, origin_head->ds_object, tx));
}
/* move snapshots to this dir */
- for (snap = list_head(&pa->shared_snaps); snap;
- snap = list_next(&pa->shared_snaps, snap)) {
+ for (snap = list_head(&ddpa->shared_snaps); snap;
+ snap = list_next(&ddpa->shared_snaps, snap)) {
dsl_dataset_t *ds = snap->ds;
- /* unregister props as dsl_dir is changing */
+ /*
+ * Property callbacks are registered to a particular
+ * dsl_dir. Since ours is changing, evict the objset
+ * so that they will be unregistered from the old dsl_dir.
+ */
if (ds->ds_objset) {
dmu_objset_evict(ds->ds_objset);
ds->ds_objset = NULL;
}
+
/* move snap name entry */
- VERIFY(0 == dsl_dataset_get_snapname(ds));
- VERIFY(0 == dsl_dataset_snap_remove(origin_head,
- ds->ds_snapname, tx));
- VERIFY(0 == zap_add(dp->dp_meta_objset,
- hds->ds_phys->ds_snapnames_zapobj, ds->ds_snapname,
+ VERIFY0(dsl_dataset_get_snapname(ds));
+ VERIFY0(dsl_dataset_snap_remove(origin_head,
+ ds->ds_snapname, tx, B_TRUE));
+ VERIFY0(zap_add(dp->dp_meta_objset,
+ dsl_dataset_phys(hds)->ds_snapnames_zapobj, ds->ds_snapname,
8, 1, &ds->ds_object, tx));
+ dsl_fs_ss_count_adjust(hds->ds_dir, 1,
+ DD_FIELD_SNAPSHOT_COUNT, tx);
/* change containing dsl_dir */
dmu_buf_will_dirty(ds->ds_dbuf, tx);
- ASSERT3U(ds->ds_phys->ds_dir_obj, ==, odd->dd_object);
- ds->ds_phys->ds_dir_obj = dd->dd_object;
+ ASSERT3U(dsl_dataset_phys(ds)->ds_dir_obj, ==, odd->dd_object);
+ dsl_dataset_phys(ds)->ds_dir_obj = dd->dd_object;
ASSERT3P(ds->ds_dir, ==, odd);
- dsl_dir_close(ds->ds_dir, ds);
- VERIFY(0 == dsl_dir_open_obj(dp, dd->dd_object,
+ dsl_dir_rele(ds->ds_dir, ds);
+ VERIFY0(dsl_dir_hold_obj(dp, dd->dd_object,
NULL, ds, &ds->ds_dir));
/* move any clone references */
- if (ds->ds_phys->ds_next_clones_obj &&
+ if (dsl_dataset_phys(ds)->ds_next_clones_obj &&
spa_version(dp->dp_spa) >= SPA_VERSION_DIR_CLONES) {
zap_cursor_t zc;
zap_attribute_t za;
for (zap_cursor_init(&zc, dp->dp_meta_objset,
- ds->ds_phys->ds_next_clones_obj);
+ dsl_dataset_phys(ds)->ds_next_clones_obj);
zap_cursor_retrieve(&zc, &za) == 0;
zap_cursor_advance(&zc)) {
dsl_dataset_t *cnds;
continue;
}
- VERIFY3U(0, ==, dsl_dataset_hold_obj(dp,
+ VERIFY0(dsl_dataset_hold_obj(dp,
za.za_first_integer, FTAG, &cnds));
- o = cnds->ds_dir->dd_phys->dd_head_dataset_obj;
+ o = dsl_dir_phys(cnds->ds_dir)->
+ dd_head_dataset_obj;
- VERIFY3U(zap_remove_int(dp->dp_meta_objset,
- odd->dd_phys->dd_clones, o, tx), ==, 0);
- VERIFY3U(zap_add_int(dp->dp_meta_objset,
- dd->dd_phys->dd_clones, o, tx), ==, 0);
+ VERIFY0(zap_remove_int(dp->dp_meta_objset,
+ dsl_dir_phys(odd)->dd_clones, o, tx));
+ VERIFY0(zap_add_int(dp->dp_meta_objset,
+ dsl_dir_phys(dd)->dd_clones, o, tx));
dsl_dataset_rele(cnds, FTAG);
- }
- zap_cursor_fini(&zc);
- }
-
- ASSERT3U(dsl_prop_numcb(ds), ==, 0);
- }
-
- /*
- * Change space accounting.
- * Note, pa->*usedsnap and dd_used_breakdown[SNAP] will either
- * both be valid, or both be 0 (resulting in delta == 0). This
- * is true for each of {clone,origin} independently.
- */
-
- delta = pa->cloneusedsnap -
- dd->dd_phys->dd_used_breakdown[DD_USED_SNAP];
- ASSERT3S(delta, >=, 0);
- ASSERT3U(pa->used, >=, delta);
- dsl_dir_diduse_space(dd, DD_USED_SNAP, delta, 0, 0, tx);
- dsl_dir_diduse_space(dd, DD_USED_HEAD,
- pa->used - delta, pa->comp, pa->uncomp, tx);
-
- delta = pa->originusedsnap -
- odd->dd_phys->dd_used_breakdown[DD_USED_SNAP];
- ASSERT3S(delta, <=, 0);
- ASSERT3U(pa->used, >=, -delta);
- dsl_dir_diduse_space(odd, DD_USED_SNAP, delta, 0, 0, tx);
- dsl_dir_diduse_space(odd, DD_USED_HEAD,
- -pa->used - delta, -pa->comp, -pa->uncomp, tx);
-
- origin_ds->ds_phys->ds_unique_bytes = pa->unique;
-
- /* log history record */
- spa_history_log_internal(LOG_DS_PROMOTE, dd->dd_pool->dp_spa, tx,
- "dataset = %llu", hds->ds_object);
-
- dsl_dir_close(odd, FTAG);
-}
-
-static char *snaplist_tag = "snaplist";
-/*
- * Make a list of dsl_dataset_t's for the snapshots between first_obj
- * (exclusive) and last_obj (inclusive). The list will be in reverse
- * order (last_obj will be the list_head()). If first_obj == 0, do all
- * snapshots back to this dataset's origin.
- */
-static int
-snaplist_make(dsl_pool_t *dp, boolean_t own,
- uint64_t first_obj, uint64_t last_obj, list_t *l)
-{
- uint64_t obj = last_obj;
-
- ASSERT(RW_LOCK_HELD(&dp->dp_config_rwlock));
-
- list_create(l, sizeof (struct promotenode),
- offsetof(struct promotenode, link));
-
- while (obj != first_obj) {
- dsl_dataset_t *ds;
- struct promotenode *snap;
- int err;
-
- if (own) {
- err = dsl_dataset_own_obj(dp, obj,
- 0, snaplist_tag, &ds);
- if (err == 0)
- dsl_dataset_make_exclusive(ds, snaplist_tag);
- } else {
- err = dsl_dataset_hold_obj(dp, obj, snaplist_tag, &ds);
- }
- if (err == ENOENT) {
- /* lost race with snapshot destroy */
- struct promotenode *last = list_tail(l);
- ASSERT(obj != last->ds->ds_phys->ds_prev_snap_obj);
- obj = last->ds->ds_phys->ds_prev_snap_obj;
- continue;
- } else if (err) {
- return (err);
- }
-
- if (first_obj == 0)
- first_obj = ds->ds_dir->dd_phys->dd_origin_obj;
-
- snap = kmem_alloc(sizeof (struct promotenode), KM_SLEEP);
- snap->ds = ds;
- list_insert_tail(l, snap);
- obj = ds->ds_phys->ds_prev_snap_obj;
- }
-
- return (0);
-}
-
-static int
-snaplist_space(list_t *l, uint64_t mintxg, uint64_t *spacep)
-{
- struct promotenode *snap;
-
- *spacep = 0;
- for (snap = list_head(l); snap; snap = list_next(l, snap)) {
- uint64_t used, comp, uncomp;
- dsl_deadlist_space_range(&snap->ds->ds_deadlist,
- mintxg, UINT64_MAX, &used, &comp, &uncomp);
- *spacep += used;
- }
- return (0);
-}
-
-static void
-snaplist_destroy(list_t *l, boolean_t own)
-{
- struct promotenode *snap;
-
- if (!l || !list_link_active(&l->list_head))
- return;
-
- while ((snap = list_tail(l)) != NULL) {
- list_remove(l, snap);
- if (own)
- dsl_dataset_disown(snap->ds, snaplist_tag);
- else
- dsl_dataset_rele(snap->ds, snaplist_tag);
- kmem_free(snap, sizeof (struct promotenode));
- }
- list_destroy(l);
-}
-
-/*
- * Promote a clone. Nomenclature note:
- * "clone" or "cds": the original clone which is being promoted
- * "origin" or "ods": the snapshot which is originally clone's origin
- * "origin head" or "ohds": the dataset which is the head
- * (filesystem/volume) for the origin
- * "origin origin": the origin of the origin's filesystem (typically
- * NULL, indicating that the clone is not a clone of a clone).
- */
-int
-dsl_dataset_promote(const char *name, char *conflsnap)
-{
- dsl_dataset_t *ds;
- dsl_dir_t *dd;
- dsl_pool_t *dp;
- dmu_object_info_t doi;
- struct promotearg pa;
- struct promotenode *snap;
- int err;
-
- bzero(&pa, sizeof(struct promotearg));
- err = dsl_dataset_hold(name, FTAG, &ds);
- if (err)
- return (err);
- dd = ds->ds_dir;
- dp = dd->dd_pool;
-
- err = dmu_object_info(dp->dp_meta_objset,
- ds->ds_phys->ds_snapnames_zapobj, &doi);
- if (err) {
- dsl_dataset_rele(ds, FTAG);
- return (err);
- }
-
- if (dsl_dataset_is_snapshot(ds) || dd->dd_phys->dd_origin_obj == 0) {
- dsl_dataset_rele(ds, FTAG);
- return (EINVAL);
- }
-
- /*
- * We are going to inherit all the snapshots taken before our
- * origin (i.e., our new origin will be our parent's origin).
- * Take ownership of them so that we can rename them into our
- * namespace.
- */
- rw_enter(&dp->dp_config_rwlock, RW_READER);
-
- err = snaplist_make(dp, B_TRUE, 0, dd->dd_phys->dd_origin_obj,
- &pa.shared_snaps);
- if (err != 0)
- goto out;
-
- err = snaplist_make(dp, B_FALSE, 0, ds->ds_object, &pa.clone_snaps);
- if (err != 0)
- goto out;
-
- snap = list_head(&pa.shared_snaps);
- ASSERT3U(snap->ds->ds_object, ==, dd->dd_phys->dd_origin_obj);
- err = snaplist_make(dp, B_FALSE, dd->dd_phys->dd_origin_obj,
- snap->ds->ds_dir->dd_phys->dd_head_dataset_obj, &pa.origin_snaps);
- if (err != 0)
- goto out;
-
- if (snap->ds->ds_dir->dd_phys->dd_origin_obj != 0) {
- err = dsl_dataset_hold_obj(dp,
- snap->ds->ds_dir->dd_phys->dd_origin_obj,
- FTAG, &pa.origin_origin);
- if (err != 0)
- goto out;
- }
-
-out:
- rw_exit(&dp->dp_config_rwlock);
-
- /*
- * Add in 128x the snapnames zapobj size, since we will be moving
- * a bunch of snapnames to the promoted ds, and dirtying their
- * bonus buffers.
- */
- if (err == 0) {
- err = dsl_sync_task_do(dp, dsl_dataset_promote_check,
- dsl_dataset_promote_sync, ds, &pa,
- 2 + 2 * doi.doi_physical_blocks_512);
- if (err && pa.err_ds && conflsnap)
- (void) strncpy(conflsnap, pa.err_ds, MAXNAMELEN);
- }
-
- snaplist_destroy(&pa.shared_snaps, B_TRUE);
- snaplist_destroy(&pa.clone_snaps, B_FALSE);
- snaplist_destroy(&pa.origin_snaps, B_FALSE);
- if (pa.origin_origin)
- dsl_dataset_rele(pa.origin_origin, FTAG);
- dsl_dataset_rele(ds, FTAG);
- return (err);
-}
-
-struct cloneswaparg {
- dsl_dataset_t *cds; /* clone dataset */
- dsl_dataset_t *ohds; /* origin's head dataset */
- boolean_t force;
- int64_t unused_refres_delta; /* change in unconsumed refreservation */
-};
-
-/* ARGSUSED */
-static int
-dsl_dataset_clone_swap_check(void *arg1, void *arg2, dmu_tx_t *tx)
-{
- struct cloneswaparg *csa = arg1;
-
- /* they should both be heads */
- if (dsl_dataset_is_snapshot(csa->cds) ||
- dsl_dataset_is_snapshot(csa->ohds))
- return (EINVAL);
-
- /* the branch point should be just before them */
- if (csa->cds->ds_prev != csa->ohds->ds_prev)
- return (EINVAL);
-
- /* cds should be the clone (unless they are unrelated) */
- if (csa->cds->ds_prev != NULL &&
- csa->cds->ds_prev != csa->cds->ds_dir->dd_pool->dp_origin_snap &&
- csa->ohds->ds_object !=
- csa->cds->ds_prev->ds_phys->ds_next_snap_obj)
- return (EINVAL);
-
- /* the clone should be a child of the origin */
- if (csa->cds->ds_dir->dd_parent != csa->ohds->ds_dir)
- return (EINVAL);
-
- /* ohds shouldn't be modified unless 'force' */
- if (!csa->force && dsl_dataset_modified_since_lastsnap(csa->ohds))
- return (ETXTBSY);
-
- /* adjust amount of any unconsumed refreservation */
- csa->unused_refres_delta =
- (int64_t)MIN(csa->ohds->ds_reserved,
- csa->ohds->ds_phys->ds_unique_bytes) -
- (int64_t)MIN(csa->ohds->ds_reserved,
- csa->cds->ds_phys->ds_unique_bytes);
-
- if (csa->unused_refres_delta > 0 &&
- csa->unused_refres_delta >
- dsl_dir_space_available(csa->ohds->ds_dir, NULL, 0, TRUE))
- return (ENOSPC);
-
- if (csa->ohds->ds_quota != 0 &&
- csa->cds->ds_phys->ds_unique_bytes > csa->ohds->ds_quota)
- return (EDQUOT);
-
- return (0);
-}
-
-/* ARGSUSED */
-static void
-dsl_dataset_clone_swap_sync(void *arg1, void *arg2, dmu_tx_t *tx)
-{
- struct cloneswaparg *csa = arg1;
- dsl_pool_t *dp = csa->cds->ds_dir->dd_pool;
-
- ASSERT(csa->cds->ds_reserved == 0);
- ASSERT(csa->ohds->ds_quota == 0 ||
- csa->cds->ds_phys->ds_unique_bytes <= csa->ohds->ds_quota);
-
- dmu_buf_will_dirty(csa->cds->ds_dbuf, tx);
- dmu_buf_will_dirty(csa->ohds->ds_dbuf, tx);
-
- if (csa->cds->ds_objset != NULL) {
- dmu_objset_evict(csa->cds->ds_objset);
- csa->cds->ds_objset = NULL;
- }
-
- if (csa->ohds->ds_objset != NULL) {
- dmu_objset_evict(csa->ohds->ds_objset);
- csa->ohds->ds_objset = NULL;
- }
-
- /*
- * Reset origin's unique bytes, if it exists.
- */
- if (csa->cds->ds_prev) {
- dsl_dataset_t *origin = csa->cds->ds_prev;
- uint64_t comp, uncomp;
-
- dmu_buf_will_dirty(origin->ds_dbuf, tx);
- dsl_deadlist_space_range(&csa->cds->ds_deadlist,
- origin->ds_phys->ds_prev_snap_txg, UINT64_MAX,
- &origin->ds_phys->ds_unique_bytes, &comp, &uncomp);
- }
-
- /* swap blkptrs */
- {
- blkptr_t tmp;
- tmp = csa->ohds->ds_phys->ds_bp;
- csa->ohds->ds_phys->ds_bp = csa->cds->ds_phys->ds_bp;
- csa->cds->ds_phys->ds_bp = tmp;
- }
-
- /* set dd_*_bytes */
- {
- int64_t dused, dcomp, duncomp;
- uint64_t cdl_used, cdl_comp, cdl_uncomp;
- uint64_t odl_used, odl_comp, odl_uncomp;
-
- ASSERT3U(csa->cds->ds_dir->dd_phys->
- dd_used_breakdown[DD_USED_SNAP], ==, 0);
-
- dsl_deadlist_space(&csa->cds->ds_deadlist,
- &cdl_used, &cdl_comp, &cdl_uncomp);
- dsl_deadlist_space(&csa->ohds->ds_deadlist,
- &odl_used, &odl_comp, &odl_uncomp);
-
- dused = csa->cds->ds_phys->ds_referenced_bytes + cdl_used -
- (csa->ohds->ds_phys->ds_referenced_bytes + odl_used);
- dcomp = csa->cds->ds_phys->ds_compressed_bytes + cdl_comp -
- (csa->ohds->ds_phys->ds_compressed_bytes + odl_comp);
- duncomp = csa->cds->ds_phys->ds_uncompressed_bytes +
- cdl_uncomp -
- (csa->ohds->ds_phys->ds_uncompressed_bytes + odl_uncomp);
-
- dsl_dir_diduse_space(csa->ohds->ds_dir, DD_USED_HEAD,
- dused, dcomp, duncomp, tx);
- dsl_dir_diduse_space(csa->cds->ds_dir, DD_USED_HEAD,
- -dused, -dcomp, -duncomp, tx);
-
- /*
- * The difference in the space used by snapshots is the
- * difference in snapshot space due to the head's
- * deadlist (since that's the only thing that's
- * changing that affects the snapused).
- */
- dsl_deadlist_space_range(&csa->cds->ds_deadlist,
- csa->ohds->ds_dir->dd_origin_txg, UINT64_MAX,
- &cdl_used, &cdl_comp, &cdl_uncomp);
- dsl_deadlist_space_range(&csa->ohds->ds_deadlist,
- csa->ohds->ds_dir->dd_origin_txg, UINT64_MAX,
- &odl_used, &odl_comp, &odl_uncomp);
- dsl_dir_transfer_space(csa->ohds->ds_dir, cdl_used - odl_used,
- DD_USED_HEAD, DD_USED_SNAP, tx);
- }
-
- /* swap ds_*_bytes */
- SWITCH64(csa->ohds->ds_phys->ds_referenced_bytes,
- csa->cds->ds_phys->ds_referenced_bytes);
- SWITCH64(csa->ohds->ds_phys->ds_compressed_bytes,
- csa->cds->ds_phys->ds_compressed_bytes);
- SWITCH64(csa->ohds->ds_phys->ds_uncompressed_bytes,
- csa->cds->ds_phys->ds_uncompressed_bytes);
- SWITCH64(csa->ohds->ds_phys->ds_unique_bytes,
- csa->cds->ds_phys->ds_unique_bytes);
-
- /* apply any parent delta for change in unconsumed refreservation */
- dsl_dir_diduse_space(csa->ohds->ds_dir, DD_USED_REFRSRV,
- csa->unused_refres_delta, 0, 0, tx);
-
- /*
- * Swap deadlists.
- */
- dsl_deadlist_close(&csa->cds->ds_deadlist);
- dsl_deadlist_close(&csa->ohds->ds_deadlist);
- SWITCH64(csa->ohds->ds_phys->ds_deadlist_obj,
- csa->cds->ds_phys->ds_deadlist_obj);
- dsl_deadlist_open(&csa->cds->ds_deadlist, dp->dp_meta_objset,
- csa->cds->ds_phys->ds_deadlist_obj);
- dsl_deadlist_open(&csa->ohds->ds_deadlist, dp->dp_meta_objset,
- csa->ohds->ds_phys->ds_deadlist_obj);
-
- dsl_scan_ds_clone_swapped(csa->ohds, csa->cds, tx);
-}
-
-/*
- * Swap 'clone' with its origin head datasets. Used at the end of "zfs
- * recv" into an existing fs to swizzle the file system to the new
- * version, and by "zfs rollback". Can also be used to swap two
- * independent head datasets if neither has any snapshots.
- */
-int
-dsl_dataset_clone_swap(dsl_dataset_t *clone, dsl_dataset_t *origin_head,
- boolean_t force)
-{
- struct cloneswaparg csa;
- int error;
-
- ASSERT(clone->ds_owner);
- ASSERT(origin_head->ds_owner);
-retry:
- /*
- * Need exclusive access for the swap. If we're swapping these
- * datasets back after an error, we already hold the locks.
- */
- if (!RW_WRITE_HELD(&clone->ds_rwlock))
- rw_enter(&clone->ds_rwlock, RW_WRITER);
- if (!RW_WRITE_HELD(&origin_head->ds_rwlock) &&
- !rw_tryenter(&origin_head->ds_rwlock, RW_WRITER)) {
- rw_exit(&clone->ds_rwlock);
- rw_enter(&origin_head->ds_rwlock, RW_WRITER);
- if (!rw_tryenter(&clone->ds_rwlock, RW_WRITER)) {
- rw_exit(&origin_head->ds_rwlock);
- goto retry;
- }
- }
- csa.cds = clone;
- csa.ohds = origin_head;
- csa.force = force;
- error = dsl_sync_task_do(clone->ds_dir->dd_pool,
- dsl_dataset_clone_swap_check,
- dsl_dataset_clone_swap_sync, &csa, NULL, 9);
- return (error);
-}
-
-/*
- * Given a pool name and a dataset object number in that pool,
- * return the name of that dataset.
- */
-int
-dsl_dsobj_to_dsname(char *pname, uint64_t obj, char *buf)
-{
- spa_t *spa;
- dsl_pool_t *dp;
- dsl_dataset_t *ds;
- int error;
-
- if ((error = spa_open(pname, &spa, FTAG)) != 0)
- return (error);
- dp = spa_get_dsl(spa);
- rw_enter(&dp->dp_config_rwlock, RW_READER);
- if ((error = dsl_dataset_hold_obj(dp, obj, FTAG, &ds)) == 0) {
- dsl_dataset_name(ds, buf);
- dsl_dataset_rele(ds, FTAG);
- }
- rw_exit(&dp->dp_config_rwlock);
- spa_close(spa, FTAG);
-
- return (error);
-}
-
-int
-dsl_dataset_check_quota(dsl_dataset_t *ds, boolean_t check_quota,
- uint64_t asize, uint64_t inflight, uint64_t *used, uint64_t *ref_rsrv)
-{
- int error = 0;
-
- ASSERT3S(asize, >, 0);
-
- /*
- * *ref_rsrv is the portion of asize that will come from any
- * unconsumed refreservation space.
- */
- *ref_rsrv = 0;
-
- mutex_enter(&ds->ds_lock);
- /*
- * Make a space adjustment for reserved bytes.
- */
- if (ds->ds_reserved > ds->ds_phys->ds_unique_bytes) {
- ASSERT3U(*used, >=,
- ds->ds_reserved - ds->ds_phys->ds_unique_bytes);
- *used -= (ds->ds_reserved - ds->ds_phys->ds_unique_bytes);
- *ref_rsrv =
- asize - MIN(asize, parent_delta(ds, asize + inflight));
- }
+ }
+ zap_cursor_fini(&zc);
+ }
- if (!check_quota || ds->ds_quota == 0) {
- mutex_exit(&ds->ds_lock);
- return (0);
+ ASSERT(!dsl_prop_hascb(ds));
}
+
/*
- * If they are requesting more space, and our current estimate
- * is over quota, they get to try again unless the actual
- * on-disk is over quota and there are no pending changes (which
- * may free up space for us).
+ * Change space accounting.
+ * Note, pa->*usedsnap and dd_used_breakdown[SNAP] will either
+ * both be valid, or both be 0 (resulting in delta == 0). This
+ * is true for each of {clone,origin} independently.
*/
- if (ds->ds_phys->ds_referenced_bytes + inflight >= ds->ds_quota) {
- if (inflight > 0 ||
- ds->ds_phys->ds_referenced_bytes < ds->ds_quota)
- error = ERESTART;
- else
- error = EDQUOT;
- DMU_TX_STAT_BUMP(dmu_tx_quota);
- }
- mutex_exit(&ds->ds_lock);
+ delta = ddpa->cloneusedsnap -
+ dsl_dir_phys(dd)->dd_used_breakdown[DD_USED_SNAP];
+ ASSERT3S(delta, >=, 0);
+ ASSERT3U(ddpa->used, >=, delta);
+ dsl_dir_diduse_space(dd, DD_USED_SNAP, delta, 0, 0, tx);
+ dsl_dir_diduse_space(dd, DD_USED_HEAD,
+ ddpa->used - delta, ddpa->comp, ddpa->uncomp, tx);
- return (error);
+ delta = ddpa->originusedsnap -
+ dsl_dir_phys(odd)->dd_used_breakdown[DD_USED_SNAP];
+ ASSERT3S(delta, <=, 0);
+ ASSERT3U(ddpa->used, >=, -delta);
+ dsl_dir_diduse_space(odd, DD_USED_SNAP, delta, 0, 0, tx);
+ dsl_dir_diduse_space(odd, DD_USED_HEAD,
+ -ddpa->used - delta, -ddpa->comp, -ddpa->uncomp, tx);
+
+ dsl_dataset_phys(origin_ds)->ds_unique_bytes = ddpa->unique;
+
+ /* log history record */
+ spa_history_log_internal_ds(hds, "promote", tx, "");
+
+ dsl_dir_rele(odd, FTAG);
+ promote_rele(ddpa, FTAG);
}
-/* ARGSUSED */
+/*
+ * Make a list of dsl_dataset_t's for the snapshots between first_obj
+ * (exclusive) and last_obj (inclusive). The list will be in reverse
+ * order (last_obj will be the list_head()). If first_obj == 0, do all
+ * snapshots back to this dataset's origin.
+ */
static int
-dsl_dataset_set_quota_check(void *arg1, void *arg2, dmu_tx_t *tx)
+snaplist_make(dsl_pool_t *dp,
+ uint64_t first_obj, uint64_t last_obj, list_t *l, void *tag)
{
- dsl_dataset_t *ds = arg1;
- dsl_prop_setarg_t *psa = arg2;
- int err;
+ uint64_t obj = last_obj;
- if (spa_version(ds->ds_dir->dd_pool->dp_spa) < SPA_VERSION_REFQUOTA)
- return (ENOTSUP);
+ list_create(l, sizeof (struct promotenode),
+ offsetof(struct promotenode, link));
- if ((err = dsl_prop_predict_sync(ds->ds_dir, psa)) != 0)
- return (err);
+ while (obj != first_obj) {
+ dsl_dataset_t *ds;
+ struct promotenode *snap;
+ int err;
- if (psa->psa_effective_value == 0)
- return (0);
+ err = dsl_dataset_hold_obj(dp, obj, tag, &ds);
+ ASSERT(err != ENOENT);
+ if (err != 0)
+ return (err);
+
+ if (first_obj == 0)
+ first_obj = dsl_dir_phys(ds->ds_dir)->dd_origin_obj;
- if (psa->psa_effective_value < ds->ds_phys->ds_referenced_bytes ||
- psa->psa_effective_value < ds->ds_reserved)
- return (ENOSPC);
+ snap = kmem_alloc(sizeof (*snap), KM_SLEEP);
+ snap->ds = ds;
+ list_insert_tail(l, snap);
+ obj = dsl_dataset_phys(ds)->ds_prev_snap_obj;
+ }
return (0);
}
-extern void dsl_prop_set_sync(void *, void *, dmu_tx_t *);
+static int
+snaplist_space(list_t *l, uint64_t mintxg, uint64_t *spacep)
+{
+ struct promotenode *snap;
-void
-dsl_dataset_set_quota_sync(void *arg1, void *arg2, dmu_tx_t *tx)
+ *spacep = 0;
+ for (snap = list_head(l); snap; snap = list_next(l, snap)) {
+ uint64_t used, comp, uncomp;
+ dsl_deadlist_space_range(&snap->ds->ds_deadlist,
+ mintxg, UINT64_MAX, &used, &comp, &uncomp);
+ *spacep += used;
+ }
+ return (0);
+}
+
+static void
+snaplist_destroy(list_t *l, void *tag)
{
- dsl_dataset_t *ds = arg1;
- dsl_prop_setarg_t *psa = arg2;
- uint64_t effective_value = psa->psa_effective_value;
+ struct promotenode *snap;
- dsl_prop_set_sync(ds, psa, tx);
- DSL_PROP_CHECK_PREDICTION(ds->ds_dir, psa);
+ if (l == NULL || !list_link_active(&l->list_head))
+ return;
- if (ds->ds_quota != effective_value) {
- dmu_buf_will_dirty(ds->ds_dbuf, tx);
- ds->ds_quota = effective_value;
+ while ((snap = list_tail(l)) != NULL) {
+ list_remove(l, snap);
+ dsl_dataset_rele(snap->ds, tag);
+ kmem_free(snap, sizeof (*snap));
}
+ list_destroy(l);
}
-int
-dsl_dataset_set_quota(const char *dsname, zprop_source_t source, uint64_t quota)
+static int
+promote_hold(dsl_dataset_promote_arg_t *ddpa, dsl_pool_t *dp, void *tag)
{
- dsl_dataset_t *ds;
- dsl_prop_setarg_t psa;
- int err;
+ int error;
+ dsl_dir_t *dd;
+ struct promotenode *snap;
- dsl_prop_setarg_init_uint64(&psa, "refquota", source, "a);
+ error = dsl_dataset_hold(dp, ddpa->ddpa_clonename, tag,
+ &ddpa->ddpa_clone);
+ if (error != 0)
+ return (error);
+ dd = ddpa->ddpa_clone->ds_dir;
- err = dsl_dataset_hold(dsname, FTAG, &ds);
- if (err)
- return (err);
+ if (ddpa->ddpa_clone->ds_is_snapshot ||
+ !dsl_dir_is_clone(dd)) {
+ dsl_dataset_rele(ddpa->ddpa_clone, tag);
+ return (SET_ERROR(EINVAL));
+ }
- /*
- * If someone removes a file, then tries to set the quota, we
- * want to make sure the file freeing takes effect.
- */
- txg_wait_open(ds->ds_dir->dd_pool, 0);
+ error = snaplist_make(dp, 0, dsl_dir_phys(dd)->dd_origin_obj,
+ &ddpa->shared_snaps, tag);
+ if (error != 0)
+ goto out;
- err = dsl_sync_task_do(ds->ds_dir->dd_pool,
- dsl_dataset_set_quota_check, dsl_dataset_set_quota_sync,
- ds, &psa, 0);
+ error = snaplist_make(dp, 0, ddpa->ddpa_clone->ds_object,
+ &ddpa->clone_snaps, tag);
+ if (error != 0)
+ goto out;
- dsl_dataset_rele(ds, FTAG);
- return (err);
+ snap = list_head(&ddpa->shared_snaps);
+ ASSERT3U(snap->ds->ds_object, ==, dsl_dir_phys(dd)->dd_origin_obj);
+ error = snaplist_make(dp, dsl_dir_phys(dd)->dd_origin_obj,
+ dsl_dir_phys(snap->ds->ds_dir)->dd_head_dataset_obj,
+ &ddpa->origin_snaps, tag);
+ if (error != 0)
+ goto out;
+
+ if (dsl_dir_phys(snap->ds->ds_dir)->dd_origin_obj != 0) {
+ error = dsl_dataset_hold_obj(dp,
+ dsl_dir_phys(snap->ds->ds_dir)->dd_origin_obj,
+ tag, &ddpa->origin_origin);
+ if (error != 0)
+ goto out;
+ }
+out:
+ if (error != 0)
+ promote_rele(ddpa, tag);
+ return (error);
}
-static int
-dsl_dataset_set_reservation_check(void *arg1, void *arg2, dmu_tx_t *tx)
+static void
+promote_rele(dsl_dataset_promote_arg_t *ddpa, void *tag)
{
- dsl_dataset_t *ds = arg1;
- dsl_prop_setarg_t *psa = arg2;
- uint64_t effective_value;
- uint64_t unique;
- int err;
+ snaplist_destroy(&ddpa->shared_snaps, tag);
+ snaplist_destroy(&ddpa->clone_snaps, tag);
+ snaplist_destroy(&ddpa->origin_snaps, tag);
+ if (ddpa->origin_origin != NULL)
+ dsl_dataset_rele(ddpa->origin_origin, tag);
+ dsl_dataset_rele(ddpa->ddpa_clone, tag);
+}
- if (spa_version(ds->ds_dir->dd_pool->dp_spa) <
- SPA_VERSION_REFRESERVATION)
- return (ENOTSUP);
+/*
+ * Promote a clone.
+ *
+ * If it fails due to a conflicting snapshot name, "conflsnap" will be filled
+ * in with the name. (It must be at least ZFS_MAX_DATASET_NAME_LEN bytes long.)
+ */
+int
+dsl_dataset_promote(const char *name, char *conflsnap)
+{
+ dsl_dataset_promote_arg_t ddpa = { 0 };
+ uint64_t numsnaps;
+ int error;
+ nvpair_t *snap_pair;
+ objset_t *os;
- if (dsl_dataset_is_snapshot(ds))
- return (EINVAL);
+ /*
+ * We will modify space proportional to the number of
+ * snapshots. Compute numsnaps.
+ */
+ error = dmu_objset_hold(name, FTAG, &os);
+ if (error != 0)
+ return (error);
+ error = zap_count(dmu_objset_pool(os)->dp_meta_objset,
+ dsl_dataset_phys(dmu_objset_ds(os))->ds_snapnames_zapobj,
+ &numsnaps);
+ dmu_objset_rele(os, FTAG);
+ if (error != 0)
+ return (error);
- if ((err = dsl_prop_predict_sync(ds->ds_dir, psa)) != 0)
- return (err);
+ ddpa.ddpa_clonename = name;
+ ddpa.err_ds = fnvlist_alloc();
+ ddpa.cr = CRED();
- effective_value = psa->psa_effective_value;
+ error = dsl_sync_task(name, dsl_dataset_promote_check,
+ dsl_dataset_promote_sync, &ddpa,
+ 2 + numsnaps, ZFS_SPACE_CHECK_RESERVED);
/*
- * If we are doing the preliminary check in open context, the
- * space estimates may be inaccurate.
+ * Return the first conflicting snapshot found.
*/
- if (!dmu_tx_is_syncing(tx))
- return (0);
+ snap_pair = nvlist_next_nvpair(ddpa.err_ds, NULL);
+ if (snap_pair != NULL && conflsnap != NULL)
+ (void) strcpy(conflsnap, nvpair_name(snap_pair));
- mutex_enter(&ds->ds_lock);
- if (!DS_UNIQUE_IS_ACCURATE(ds))
- dsl_dataset_recalc_head_uniq(ds);
- unique = ds->ds_phys->ds_unique_bytes;
- mutex_exit(&ds->ds_lock);
+ fnvlist_free(ddpa.err_ds);
+ return (error);
+}
- if (MAX(unique, effective_value) > MAX(unique, ds->ds_reserved)) {
- uint64_t delta = MAX(unique, effective_value) -
- MAX(unique, ds->ds_reserved);
+int
+dsl_dataset_clone_swap_check_impl(dsl_dataset_t *clone,
+ dsl_dataset_t *origin_head, boolean_t force, void *owner, dmu_tx_t *tx)
+{
+ /*
+ * "slack" factor for received datasets with refquota set on them.
+ * See the bottom of this function for details on its use.
+ */
+ uint64_t refquota_slack = (uint64_t)DMU_MAX_ACCESS *
+ spa_asize_inflation;
+ int64_t unused_refres_delta;
- if (delta > dsl_dir_space_available(ds->ds_dir, NULL, 0, TRUE))
- return (ENOSPC);
- if (ds->ds_quota > 0 &&
- effective_value > ds->ds_quota)
- return (ENOSPC);
- }
+ /* they should both be heads */
+ if (clone->ds_is_snapshot ||
+ origin_head->ds_is_snapshot)
+ return (SET_ERROR(EINVAL));
+
+ /* if we are not forcing, the branch point should be just before them */
+ if (!force && clone->ds_prev != origin_head->ds_prev)
+ return (SET_ERROR(EINVAL));
+
+ /* clone should be the clone (unless they are unrelated) */
+ if (clone->ds_prev != NULL &&
+ clone->ds_prev != clone->ds_dir->dd_pool->dp_origin_snap &&
+ origin_head->ds_dir != clone->ds_prev->ds_dir)
+ return (SET_ERROR(EINVAL));
+
+ /* the clone should be a child of the origin */
+ if (clone->ds_dir->dd_parent != origin_head->ds_dir)
+ return (SET_ERROR(EINVAL));
+
+ /* origin_head shouldn't be modified unless 'force' */
+ if (!force &&
+ dsl_dataset_modified_since_snap(origin_head, origin_head->ds_prev))
+ return (SET_ERROR(ETXTBSY));
+
+ /* origin_head should have no long holds (e.g. is not mounted) */
+ if (dsl_dataset_handoff_check(origin_head, owner, tx))
+ return (SET_ERROR(EBUSY));
+
+ /* check amount of any unconsumed refreservation */
+ unused_refres_delta =
+ (int64_t)MIN(origin_head->ds_reserved,
+ dsl_dataset_phys(origin_head)->ds_unique_bytes) -
+ (int64_t)MIN(origin_head->ds_reserved,
+ dsl_dataset_phys(clone)->ds_unique_bytes);
+
+ if (unused_refres_delta > 0 &&
+ unused_refres_delta >
+ dsl_dir_space_available(origin_head->ds_dir, NULL, 0, TRUE))
+ return (SET_ERROR(ENOSPC));
+
+ /*
+ * The clone can't be too much over the head's refquota.
+ *
+ * To ensure that the entire refquota can be used, we allow one
+ * transaction to exceed the the refquota. Therefore, this check
+ * needs to also allow for the space referenced to be more than the
+ * refquota. The maximum amount of space that one transaction can use
+ * on disk is DMU_MAX_ACCESS * spa_asize_inflation. Allowing this
+ * overage ensures that we are able to receive a filesystem that
+ * exceeds the refquota on the source system.
+ *
+ * So that overage is the refquota_slack we use below.
+ */
+ if (origin_head->ds_quota != 0 &&
+ dsl_dataset_phys(clone)->ds_referenced_bytes >
+ origin_head->ds_quota + refquota_slack)
+ return (SET_ERROR(EDQUOT));
return (0);
}
static void
-dsl_dataset_set_reservation_sync(void *arg1, void *arg2, dmu_tx_t *tx)
+dsl_dataset_swap_remap_deadlists(dsl_dataset_t *clone,
+ dsl_dataset_t *origin, dmu_tx_t *tx)
{
- dsl_dataset_t *ds = arg1;
- dsl_prop_setarg_t *psa = arg2;
- uint64_t effective_value = psa->psa_effective_value;
- uint64_t unique;
- int64_t delta;
+ uint64_t clone_remap_dl_obj, origin_remap_dl_obj;
+ dsl_pool_t *dp = dmu_tx_pool(tx);
- dsl_prop_set_sync(ds, psa, tx);
- DSL_PROP_CHECK_PREDICTION(ds->ds_dir, psa);
+ ASSERT(dsl_pool_sync_context(dp));
- dmu_buf_will_dirty(ds->ds_dbuf, tx);
+ clone_remap_dl_obj = dsl_dataset_get_remap_deadlist_object(clone);
+ origin_remap_dl_obj = dsl_dataset_get_remap_deadlist_object(origin);
- mutex_enter(&ds->ds_dir->dd_lock);
- mutex_enter(&ds->ds_lock);
- ASSERT(DS_UNIQUE_IS_ACCURATE(ds));
- unique = ds->ds_phys->ds_unique_bytes;
- delta = MAX(0, (int64_t)(effective_value - unique)) -
- MAX(0, (int64_t)(ds->ds_reserved - unique));
- ds->ds_reserved = effective_value;
- mutex_exit(&ds->ds_lock);
+ if (clone_remap_dl_obj != 0) {
+ dsl_deadlist_close(&clone->ds_remap_deadlist);
+ dsl_dataset_unset_remap_deadlist_object(clone, tx);
+ }
+ if (origin_remap_dl_obj != 0) {
+ dsl_deadlist_close(&origin->ds_remap_deadlist);
+ dsl_dataset_unset_remap_deadlist_object(origin, tx);
+ }
- dsl_dir_diduse_space(ds->ds_dir, DD_USED_REFRSRV, delta, 0, 0, tx);
- mutex_exit(&ds->ds_dir->dd_lock);
+ if (clone_remap_dl_obj != 0) {
+ dsl_dataset_set_remap_deadlist_object(origin,
+ clone_remap_dl_obj, tx);
+ dsl_deadlist_open(&origin->ds_remap_deadlist,
+ dp->dp_meta_objset, clone_remap_dl_obj);
+ }
+ if (origin_remap_dl_obj != 0) {
+ dsl_dataset_set_remap_deadlist_object(clone,
+ origin_remap_dl_obj, tx);
+ dsl_deadlist_open(&clone->ds_remap_deadlist,
+ dp->dp_meta_objset, origin_remap_dl_obj);
+ }
}
-int
-dsl_dataset_set_reservation(const char *dsname, zprop_source_t source,
- uint64_t reservation)
+void
+dsl_dataset_clone_swap_sync_impl(dsl_dataset_t *clone,
+ dsl_dataset_t *origin_head, dmu_tx_t *tx)
{
- dsl_dataset_t *ds;
- dsl_prop_setarg_t psa;
- int err;
+ dsl_pool_t *dp = dmu_tx_pool(tx);
+ int64_t unused_refres_delta;
- dsl_prop_setarg_init_uint64(&psa, "refreservation", source,
- &reservation);
+ ASSERT(clone->ds_reserved == 0);
+ /*
+ * NOTE: On DEBUG kernels there could be a race between this and
+ * the check function if spa_asize_inflation is adjusted...
+ */
+ ASSERT(origin_head->ds_quota == 0 ||
+ dsl_dataset_phys(clone)->ds_unique_bytes <= origin_head->ds_quota +
+ DMU_MAX_ACCESS * spa_asize_inflation);
+ ASSERT3P(clone->ds_prev, ==, origin_head->ds_prev);
- err = dsl_dataset_hold(dsname, FTAG, &ds);
- if (err)
- return (err);
+ /*
+ * Swap per-dataset feature flags.
+ */
+ for (spa_feature_t f = 0; f < SPA_FEATURES; f++) {
+ if (!(spa_feature_table[f].fi_flags &
+ ZFEATURE_FLAG_PER_DATASET)) {
+ ASSERT(!dsl_dataset_feature_is_active(clone, f));
+ ASSERT(!dsl_dataset_feature_is_active(origin_head, f));
+ continue;
+ }
- err = dsl_sync_task_do(ds->ds_dir->dd_pool,
- dsl_dataset_set_reservation_check,
- dsl_dataset_set_reservation_sync, ds, &psa, 0);
+ boolean_t clone_inuse = dsl_dataset_feature_is_active(clone, f);
+ void *clone_feature = clone->ds_feature[f];
+ boolean_t origin_head_inuse =
+ dsl_dataset_feature_is_active(origin_head, f);
+ void *origin_head_feature = origin_head->ds_feature[f];
+
+ if (clone_inuse)
+ dsl_dataset_deactivate_feature_impl(clone, f, tx);
+ if (origin_head_inuse)
+ dsl_dataset_deactivate_feature_impl(origin_head, f, tx);
+
+ if (clone_inuse) {
+ dsl_dataset_activate_feature(origin_head->ds_object, f,
+ clone_feature, tx);
+ origin_head->ds_feature[f] = clone_feature;
+ }
+ if (origin_head_inuse) {
+ dsl_dataset_activate_feature(clone->ds_object, f,
+ origin_head_feature, tx);
+ clone->ds_feature[f] = origin_head_feature;
+ }
+ }
- dsl_dataset_rele(ds, FTAG);
- return (err);
-}
+ dmu_buf_will_dirty(clone->ds_dbuf, tx);
+ dmu_buf_will_dirty(origin_head->ds_dbuf, tx);
-typedef struct zfs_hold_cleanup_arg {
- dsl_pool_t *dp;
- uint64_t dsobj;
- char htag[MAXNAMELEN];
-} zfs_hold_cleanup_arg_t;
+ if (clone->ds_objset != NULL) {
+ dmu_objset_evict(clone->ds_objset);
+ clone->ds_objset = NULL;
+ }
-static void
-dsl_dataset_user_release_onexit(void *arg)
-{
- zfs_hold_cleanup_arg_t *ca = arg;
+ if (origin_head->ds_objset != NULL) {
+ dmu_objset_evict(origin_head->ds_objset);
+ origin_head->ds_objset = NULL;
+ }
- (void) dsl_dataset_user_release_tmp(ca->dp, ca->dsobj, ca->htag,
- B_TRUE);
- kmem_free(ca, sizeof (zfs_hold_cleanup_arg_t));
-}
+ unused_refres_delta =
+ (int64_t)MIN(origin_head->ds_reserved,
+ dsl_dataset_phys(origin_head)->ds_unique_bytes) -
+ (int64_t)MIN(origin_head->ds_reserved,
+ dsl_dataset_phys(clone)->ds_unique_bytes);
-void
-dsl_register_onexit_hold_cleanup(dsl_dataset_t *ds, const char *htag,
- minor_t minor)
-{
- zfs_hold_cleanup_arg_t *ca;
-
- ca = kmem_alloc(sizeof (zfs_hold_cleanup_arg_t), KM_SLEEP);
- ca->dp = ds->ds_dir->dd_pool;
- ca->dsobj = ds->ds_object;
- (void) strlcpy(ca->htag, htag, sizeof (ca->htag));
- VERIFY3U(0, ==, zfs_onexit_add_cb(minor,
- dsl_dataset_user_release_onexit, ca, NULL));
-}
+ /*
+ * Reset origin's unique bytes, if it exists.
+ */
+ if (clone->ds_prev) {
+ dsl_dataset_t *origin = clone->ds_prev;
+ uint64_t comp, uncomp;
-/*
- * If you add new checks here, you may need to add
- * additional checks to the "temporary" case in
- * snapshot_check() in dmu_objset.c.
- */
-static int
-dsl_dataset_user_hold_check(void *arg1, void *arg2, dmu_tx_t *tx)
-{
- dsl_dataset_t *ds = arg1;
- struct dsl_ds_holdarg *ha = arg2;
- char *htag = ha->htag;
- objset_t *mos = ds->ds_dir->dd_pool->dp_meta_objset;
- int error = 0;
+ dmu_buf_will_dirty(origin->ds_dbuf, tx);
+ dsl_deadlist_space_range(&clone->ds_deadlist,
+ dsl_dataset_phys(origin)->ds_prev_snap_txg, UINT64_MAX,
+ &dsl_dataset_phys(origin)->ds_unique_bytes, &comp, &uncomp);
+ }
- if (spa_version(ds->ds_dir->dd_pool->dp_spa) < SPA_VERSION_USERREFS)
- return (ENOTSUP);
+ /* swap blkptrs */
+ {
+ rrw_enter(&clone->ds_bp_rwlock, RW_WRITER, FTAG);
+ rrw_enter(&origin_head->ds_bp_rwlock, RW_WRITER, FTAG);
+ blkptr_t tmp;
+ tmp = dsl_dataset_phys(origin_head)->ds_bp;
+ dsl_dataset_phys(origin_head)->ds_bp =
+ dsl_dataset_phys(clone)->ds_bp;
+ dsl_dataset_phys(clone)->ds_bp = tmp;
+ rrw_exit(&origin_head->ds_bp_rwlock, FTAG);
+ rrw_exit(&clone->ds_bp_rwlock, FTAG);
+ }
- if (!dsl_dataset_is_snapshot(ds))
- return (EINVAL);
+ /* set dd_*_bytes */
+ {
+ int64_t dused, dcomp, duncomp;
+ uint64_t cdl_used, cdl_comp, cdl_uncomp;
+ uint64_t odl_used, odl_comp, odl_uncomp;
- /* tags must be unique */
- mutex_enter(&ds->ds_lock);
- if (ds->ds_phys->ds_userrefs_obj) {
- error = zap_lookup(mos, ds->ds_phys->ds_userrefs_obj, htag,
- 8, 1, tx);
- if (error == 0)
- error = EEXIST;
- else if (error == ENOENT)
- error = 0;
- }
- mutex_exit(&ds->ds_lock);
+ ASSERT3U(dsl_dir_phys(clone->ds_dir)->
+ dd_used_breakdown[DD_USED_SNAP], ==, 0);
- if (error == 0 && ha->temphold &&
- strlen(htag) + MAX_TAG_PREFIX_LEN >= MAXNAMELEN)
- error = E2BIG;
+ dsl_deadlist_space(&clone->ds_deadlist,
+ &cdl_used, &cdl_comp, &cdl_uncomp);
+ dsl_deadlist_space(&origin_head->ds_deadlist,
+ &odl_used, &odl_comp, &odl_uncomp);
- return (error);
-}
+ dused = dsl_dataset_phys(clone)->ds_referenced_bytes +
+ cdl_used -
+ (dsl_dataset_phys(origin_head)->ds_referenced_bytes +
+ odl_used);
+ dcomp = dsl_dataset_phys(clone)->ds_compressed_bytes +
+ cdl_comp -
+ (dsl_dataset_phys(origin_head)->ds_compressed_bytes +
+ odl_comp);
+ duncomp = dsl_dataset_phys(clone)->ds_uncompressed_bytes +
+ cdl_uncomp -
+ (dsl_dataset_phys(origin_head)->ds_uncompressed_bytes +
+ odl_uncomp);
-void
-dsl_dataset_user_hold_sync(void *arg1, void *arg2, dmu_tx_t *tx)
-{
- dsl_dataset_t *ds = arg1;
- struct dsl_ds_holdarg *ha = arg2;
- char *htag = ha->htag;
- dsl_pool_t *dp = ds->ds_dir->dd_pool;
- objset_t *mos = dp->dp_meta_objset;
- uint64_t now = gethrestime_sec();
- uint64_t zapobj;
+ dsl_dir_diduse_space(origin_head->ds_dir, DD_USED_HEAD,
+ dused, dcomp, duncomp, tx);
+ dsl_dir_diduse_space(clone->ds_dir, DD_USED_HEAD,
+ -dused, -dcomp, -duncomp, tx);
- mutex_enter(&ds->ds_lock);
- if (ds->ds_phys->ds_userrefs_obj == 0) {
/*
- * This is the first user hold for this dataset. Create
- * the userrefs zap object.
+ * The difference in the space used by snapshots is the
+ * difference in snapshot space due to the head's
+ * deadlist (since that's the only thing that's
+ * changing that affects the snapused).
*/
- dmu_buf_will_dirty(ds->ds_dbuf, tx);
- zapobj = ds->ds_phys->ds_userrefs_obj =
- zap_create(mos, DMU_OT_USERREFS, DMU_OT_NONE, 0, tx);
- } else {
- zapobj = ds->ds_phys->ds_userrefs_obj;
+ dsl_deadlist_space_range(&clone->ds_deadlist,
+ origin_head->ds_dir->dd_origin_txg, UINT64_MAX,
+ &cdl_used, &cdl_comp, &cdl_uncomp);
+ dsl_deadlist_space_range(&origin_head->ds_deadlist,
+ origin_head->ds_dir->dd_origin_txg, UINT64_MAX,
+ &odl_used, &odl_comp, &odl_uncomp);
+ dsl_dir_transfer_space(origin_head->ds_dir, cdl_used - odl_used,
+ DD_USED_HEAD, DD_USED_SNAP, tx);
}
- ds->ds_userrefs++;
- mutex_exit(&ds->ds_lock);
- VERIFY(0 == zap_add(mos, zapobj, htag, 8, 1, &now, tx));
+ /* swap ds_*_bytes */
+ SWITCH64(dsl_dataset_phys(origin_head)->ds_referenced_bytes,
+ dsl_dataset_phys(clone)->ds_referenced_bytes);
+ SWITCH64(dsl_dataset_phys(origin_head)->ds_compressed_bytes,
+ dsl_dataset_phys(clone)->ds_compressed_bytes);
+ SWITCH64(dsl_dataset_phys(origin_head)->ds_uncompressed_bytes,
+ dsl_dataset_phys(clone)->ds_uncompressed_bytes);
+ SWITCH64(dsl_dataset_phys(origin_head)->ds_unique_bytes,
+ dsl_dataset_phys(clone)->ds_unique_bytes);
- if (ha->temphold) {
- VERIFY(0 == dsl_pool_user_hold(dp, ds->ds_object,
- htag, &now, tx));
- }
+ /* apply any parent delta for change in unconsumed refreservation */
+ dsl_dir_diduse_space(origin_head->ds_dir, DD_USED_REFRSRV,
+ unused_refres_delta, 0, 0, tx);
- spa_history_log_internal(LOG_DS_USER_HOLD,
- dp->dp_spa, tx, "<%s> temp = %d dataset = %llu", htag,
- (int)ha->temphold, ds->ds_object);
+ /*
+ * Swap deadlists.
+ */
+ dsl_deadlist_close(&clone->ds_deadlist);
+ dsl_deadlist_close(&origin_head->ds_deadlist);
+ SWITCH64(dsl_dataset_phys(origin_head)->ds_deadlist_obj,
+ dsl_dataset_phys(clone)->ds_deadlist_obj);
+ dsl_deadlist_open(&clone->ds_deadlist, dp->dp_meta_objset,
+ dsl_dataset_phys(clone)->ds_deadlist_obj);
+ dsl_deadlist_open(&origin_head->ds_deadlist, dp->dp_meta_objset,
+ dsl_dataset_phys(origin_head)->ds_deadlist_obj);
+ dsl_dataset_swap_remap_deadlists(clone, origin_head, tx);
+
+ dsl_scan_ds_clone_swapped(origin_head, clone, tx);
+
+ spa_history_log_internal_ds(clone, "clone swap", tx,
+ "parent=%s", origin_head->ds_dir->dd_myname);
}
-static int
-dsl_dataset_user_hold_one(const char *dsname, void *arg)
+/*
+ * Given a pool name and a dataset object number in that pool,
+ * return the name of that dataset.
+ */
+int
+dsl_dsobj_to_dsname(char *pname, uint64_t obj, char *buf)
{
- struct dsl_ds_holdarg *ha = arg;
+ dsl_pool_t *dp;
dsl_dataset_t *ds;
int error;
- char *name;
-
- /* alloc a buffer to hold dsname@snapname plus terminating NULL */
- name = kmem_asprintf("%s@%s", dsname, ha->snapname);
- error = dsl_dataset_hold(name, ha->dstg, &ds);
- strfree(name);
- if (error == 0) {
- ha->gotone = B_TRUE;
- dsl_sync_task_create(ha->dstg, dsl_dataset_user_hold_check,
- dsl_dataset_user_hold_sync, ds, ha, 0);
- } else if (error == ENOENT && ha->recursive) {
- error = 0;
- } else {
- (void) strlcpy(ha->failed, dsname, sizeof (ha->failed));
- }
- return (error);
-}
-int
-dsl_dataset_user_hold_for_send(dsl_dataset_t *ds, char *htag,
- boolean_t temphold)
-{
- struct dsl_ds_holdarg *ha;
- int error;
+ error = dsl_pool_hold(pname, FTAG, &dp);
+ if (error != 0)
+ return (error);
- ha = kmem_zalloc(sizeof (struct dsl_ds_holdarg), KM_SLEEP);
- ha->htag = htag;
- ha->temphold = temphold;
- error = dsl_sync_task_do(ds->ds_dir->dd_pool,
- dsl_dataset_user_hold_check, dsl_dataset_user_hold_sync,
- ds, ha, 0);
- kmem_free(ha, sizeof (struct dsl_ds_holdarg));
+ error = dsl_dataset_hold_obj(dp, obj, FTAG, &ds);
+ if (error == 0) {
+ dsl_dataset_name(ds, buf);
+ dsl_dataset_rele(ds, FTAG);
+ }
+ dsl_pool_rele(dp, FTAG);
return (error);
}
int
-dsl_dataset_user_hold(char *dsname, char *snapname, char *htag,
- boolean_t recursive, boolean_t temphold, int cleanup_fd)
+dsl_dataset_check_quota(dsl_dataset_t *ds, boolean_t check_quota,
+ uint64_t asize, uint64_t inflight, uint64_t *used, uint64_t *ref_rsrv)
{
- struct dsl_ds_holdarg *ha;
- dsl_sync_task_t *dst;
- spa_t *spa;
- int error;
- minor_t minor = 0;
-
- if (cleanup_fd != -1) {
- /* Currently we only support cleanup-on-exit of tempholds. */
- if (!temphold)
- return (EINVAL);
- error = zfs_onexit_fd_hold(cleanup_fd, &minor);
- if (error)
- return (error);
- }
+ int error = 0;
- ha = kmem_zalloc(sizeof (struct dsl_ds_holdarg), KM_SLEEP);
+ ASSERT3S(asize, >, 0);
- (void) strlcpy(ha->failed, dsname, sizeof (ha->failed));
+ /*
+ * *ref_rsrv is the portion of asize that will come from any
+ * unconsumed refreservation space.
+ */
+ *ref_rsrv = 0;
- error = spa_open(dsname, &spa, FTAG);
- if (error) {
- kmem_free(ha, sizeof (struct dsl_ds_holdarg));
- if (cleanup_fd != -1)
- zfs_onexit_fd_rele(cleanup_fd);
- return (error);
+ mutex_enter(&ds->ds_lock);
+ /*
+ * Make a space adjustment for reserved bytes.
+ */
+ if (ds->ds_reserved > dsl_dataset_phys(ds)->ds_unique_bytes) {
+ ASSERT3U(*used, >=,
+ ds->ds_reserved - dsl_dataset_phys(ds)->ds_unique_bytes);
+ *used -=
+ (ds->ds_reserved - dsl_dataset_phys(ds)->ds_unique_bytes);
+ *ref_rsrv =
+ asize - MIN(asize, parent_delta(ds, asize + inflight));
}
- ha->dstg = dsl_sync_task_group_create(spa_get_dsl(spa));
- ha->htag = htag;
- ha->snapname = snapname;
- ha->recursive = recursive;
- ha->temphold = temphold;
-
- if (recursive) {
- error = dmu_objset_find(dsname, dsl_dataset_user_hold_one,
- ha, DS_FIND_CHILDREN);
- } else {
- error = dsl_dataset_user_hold_one(dsname, ha);
+ if (!check_quota || ds->ds_quota == 0) {
+ mutex_exit(&ds->ds_lock);
+ return (0);
}
- if (error == 0)
- error = dsl_sync_task_group_wait(ha->dstg);
-
- for (dst = list_head(&ha->dstg->dstg_tasks); dst;
- dst = list_next(&ha->dstg->dstg_tasks, dst)) {
- dsl_dataset_t *ds = dst->dst_arg1;
-
- if (dst->dst_err) {
- dsl_dataset_name(ds, ha->failed);
- *strchr(ha->failed, '@') = '\0';
- } else if (error == 0 && minor != 0 && temphold) {
- /*
- * If this hold is to be released upon process exit,
- * register that action now.
- */
- dsl_register_onexit_hold_cleanup(ds, htag, minor);
- }
- dsl_dataset_rele(ds, ha->dstg);
+ /*
+ * If they are requesting more space, and our current estimate
+ * is over quota, they get to try again unless the actual
+ * on-disk is over quota and there are no pending changes (which
+ * may free up space for us).
+ */
+ if (dsl_dataset_phys(ds)->ds_referenced_bytes + inflight >=
+ ds->ds_quota) {
+ if (inflight > 0 ||
+ dsl_dataset_phys(ds)->ds_referenced_bytes < ds->ds_quota)
+ error = SET_ERROR(ERESTART);
+ else
+ error = SET_ERROR(EDQUOT);
}
+ mutex_exit(&ds->ds_lock);
- if (error == 0 && recursive && !ha->gotone)
- error = ENOENT;
-
- if (error)
- (void) strlcpy(dsname, ha->failed, sizeof (ha->failed));
-
- dsl_sync_task_group_destroy(ha->dstg);
-
- kmem_free(ha, sizeof (struct dsl_ds_holdarg));
- spa_close(spa, FTAG);
- if (cleanup_fd != -1)
- zfs_onexit_fd_rele(cleanup_fd);
return (error);
}
-struct dsl_ds_releasearg {
- dsl_dataset_t *ds;
- const char *htag;
- boolean_t own; /* do we own or just hold ds? */
-};
+typedef struct dsl_dataset_set_qr_arg {
+ const char *ddsqra_name;
+ zprop_source_t ddsqra_source;
+ uint64_t ddsqra_value;
+} dsl_dataset_set_qr_arg_t;
+
+/* ARGSUSED */
static int
-dsl_dataset_release_might_destroy(dsl_dataset_t *ds, const char *htag,
- boolean_t *might_destroy)
+dsl_dataset_set_refquota_check(void *arg, dmu_tx_t *tx)
{
- objset_t *mos = ds->ds_dir->dd_pool->dp_meta_objset;
- uint64_t zapobj;
- uint64_t tmp;
+ dsl_dataset_set_qr_arg_t *ddsqra = arg;
+ dsl_pool_t *dp = dmu_tx_pool(tx);
+ dsl_dataset_t *ds;
int error;
+ uint64_t newval;
- *might_destroy = B_FALSE;
+ if (spa_version(dp->dp_spa) < SPA_VERSION_REFQUOTA)
+ return (SET_ERROR(ENOTSUP));
- mutex_enter(&ds->ds_lock);
- zapobj = ds->ds_phys->ds_userrefs_obj;
- if (zapobj == 0) {
- /* The tag can't possibly exist */
- mutex_exit(&ds->ds_lock);
- return (ESRCH);
+ error = dsl_dataset_hold(dp, ddsqra->ddsqra_name, FTAG, &ds);
+ if (error != 0)
+ return (error);
+
+ if (ds->ds_is_snapshot) {
+ dsl_dataset_rele(ds, FTAG);
+ return (SET_ERROR(EINVAL));
}
- /* Make sure the tag exists */
- error = zap_lookup(mos, zapobj, htag, 8, 1, &tmp);
- if (error) {
- mutex_exit(&ds->ds_lock);
- if (error == ENOENT)
- error = ESRCH;
+ error = dsl_prop_predict(ds->ds_dir,
+ zfs_prop_to_name(ZFS_PROP_REFQUOTA),
+ ddsqra->ddsqra_source, ddsqra->ddsqra_value, &newval);
+ if (error != 0) {
+ dsl_dataset_rele(ds, FTAG);
return (error);
}
- if (ds->ds_userrefs == 1 && ds->ds_phys->ds_num_children == 1 &&
- DS_IS_DEFER_DESTROY(ds))
- *might_destroy = B_TRUE;
+ if (newval == 0) {
+ dsl_dataset_rele(ds, FTAG);
+ return (0);
+ }
+
+ if (newval < dsl_dataset_phys(ds)->ds_referenced_bytes ||
+ newval < ds->ds_reserved) {
+ dsl_dataset_rele(ds, FTAG);
+ return (SET_ERROR(ENOSPC));
+ }
- mutex_exit(&ds->ds_lock);
+ dsl_dataset_rele(ds, FTAG);
return (0);
}
-static int
-dsl_dataset_user_release_check(void *arg1, void *tag, dmu_tx_t *tx)
+static void
+dsl_dataset_set_refquota_sync(void *arg, dmu_tx_t *tx)
{
- struct dsl_ds_releasearg *ra = arg1;
- dsl_dataset_t *ds = ra->ds;
- boolean_t might_destroy;
- int error;
+ dsl_dataset_set_qr_arg_t *ddsqra = arg;
+ dsl_pool_t *dp = dmu_tx_pool(tx);
+ dsl_dataset_t *ds = NULL;
+ uint64_t newval;
- if (spa_version(ds->ds_dir->dd_pool->dp_spa) < SPA_VERSION_USERREFS)
- return (ENOTSUP);
+ VERIFY0(dsl_dataset_hold(dp, ddsqra->ddsqra_name, FTAG, &ds));
- error = dsl_dataset_release_might_destroy(ds, ra->htag, &might_destroy);
- if (error)
- return (error);
+ dsl_prop_set_sync_impl(ds,
+ zfs_prop_to_name(ZFS_PROP_REFQUOTA),
+ ddsqra->ddsqra_source, sizeof (ddsqra->ddsqra_value), 1,
+ &ddsqra->ddsqra_value, tx);
- if (might_destroy) {
- struct dsl_ds_destroyarg dsda = {0};
+ VERIFY0(dsl_prop_get_int_ds(ds,
+ zfs_prop_to_name(ZFS_PROP_REFQUOTA), &newval));
- if (dmu_tx_is_syncing(tx)) {
- /*
- * If we're not prepared to remove the snapshot,
- * we can't allow the release to happen right now.
- */
- if (!ra->own)
- return (EBUSY);
- }
- dsda.ds = ds;
- dsda.releasing = B_TRUE;
- return (dsl_dataset_destroy_check(&dsda, tag, tx));
+ if (ds->ds_quota != newval) {
+ dmu_buf_will_dirty(ds->ds_dbuf, tx);
+ ds->ds_quota = newval;
}
-
- return (0);
+ dsl_dataset_rele(ds, FTAG);
}
-static void
-dsl_dataset_user_release_sync(void *arg1, void *tag, dmu_tx_t *tx)
+int
+dsl_dataset_set_refquota(const char *dsname, zprop_source_t source,
+ uint64_t refquota)
{
- struct dsl_ds_releasearg *ra = arg1;
- dsl_dataset_t *ds = ra->ds;
- dsl_pool_t *dp = ds->ds_dir->dd_pool;
- objset_t *mos = dp->dp_meta_objset;
- uint64_t zapobj;
- uint64_t dsobj = ds->ds_object;
- uint64_t refs;
- int error;
+ dsl_dataset_set_qr_arg_t ddsqra;
- mutex_enter(&ds->ds_lock);
- ds->ds_userrefs--;
- refs = ds->ds_userrefs;
- mutex_exit(&ds->ds_lock);
- error = dsl_pool_user_release(dp, ds->ds_object, ra->htag, tx);
- VERIFY(error == 0 || error == ENOENT);
- zapobj = ds->ds_phys->ds_userrefs_obj;
- VERIFY(0 == zap_remove(mos, zapobj, ra->htag, tx));
-
- spa_history_log_internal(LOG_DS_USER_RELEASE,
- dp->dp_spa, tx, "<%s> %lld dataset = %llu",
- ra->htag, (longlong_t)refs, dsobj);
-
- if (ds->ds_userrefs == 0 && ds->ds_phys->ds_num_children == 1 &&
- DS_IS_DEFER_DESTROY(ds)) {
- struct dsl_ds_destroyarg dsda = {0};
+ ddsqra.ddsqra_name = dsname;
+ ddsqra.ddsqra_source = source;
+ ddsqra.ddsqra_value = refquota;
- ASSERT(ra->own);
- dsda.ds = ds;
- dsda.releasing = B_TRUE;
- /* We already did the destroy_check */
- dsl_dataset_destroy_sync(&dsda, tag, tx);
- }
+ return (dsl_sync_task(dsname, dsl_dataset_set_refquota_check,
+ dsl_dataset_set_refquota_sync, &ddsqra, 0,
+ ZFS_SPACE_CHECK_EXTRA_RESERVED));
}
static int
-dsl_dataset_user_release_one(const char *dsname, void *arg)
+dsl_dataset_set_refreservation_check(void *arg, dmu_tx_t *tx)
{
- struct dsl_ds_holdarg *ha = arg;
- struct dsl_ds_releasearg *ra;
+ dsl_dataset_set_qr_arg_t *ddsqra = arg;
+ dsl_pool_t *dp = dmu_tx_pool(tx);
dsl_dataset_t *ds;
int error;
- void *dtag = ha->dstg;
- char *name;
- boolean_t own = B_FALSE;
- boolean_t might_destroy;
-
- /* alloc a buffer to hold dsname@snapname, plus the terminating NULL */
- name = kmem_asprintf("%s@%s", dsname, ha->snapname);
- error = dsl_dataset_hold(name, dtag, &ds);
- strfree(name);
- if (error == ENOENT && ha->recursive)
- return (0);
- (void) strlcpy(ha->failed, dsname, sizeof (ha->failed));
- if (error)
- return (error);
-
- ha->gotone = B_TRUE;
+ uint64_t newval, unique;
- ASSERT(dsl_dataset_is_snapshot(ds));
+ if (spa_version(dp->dp_spa) < SPA_VERSION_REFRESERVATION)
+ return (SET_ERROR(ENOTSUP));
- error = dsl_dataset_release_might_destroy(ds, ha->htag, &might_destroy);
- if (error) {
- dsl_dataset_rele(ds, dtag);
+ error = dsl_dataset_hold(dp, ddsqra->ddsqra_name, FTAG, &ds);
+ if (error != 0)
return (error);
- }
- if (might_destroy) {
-#ifdef _KERNEL
- name = kmem_asprintf("%s@%s", dsname, ha->snapname);
- error = zfs_unmount_snap(name, NULL);
- strfree(name);
- if (error) {
- dsl_dataset_rele(ds, dtag);
- return (error);
- }
-#endif
- if (!dsl_dataset_tryown(ds, B_TRUE, dtag)) {
- dsl_dataset_rele(ds, dtag);
- return (EBUSY);
- } else {
- own = B_TRUE;
- dsl_dataset_make_exclusive(ds, dtag);
- }
+ if (ds->ds_is_snapshot) {
+ dsl_dataset_rele(ds, FTAG);
+ return (SET_ERROR(EINVAL));
}
- ra = kmem_alloc(sizeof (struct dsl_ds_releasearg), KM_SLEEP);
- ra->ds = ds;
- ra->htag = ha->htag;
- ra->own = own;
- dsl_sync_task_create(ha->dstg, dsl_dataset_user_release_check,
- dsl_dataset_user_release_sync, ra, dtag, 0);
-
- return (0);
-}
-
-int
-dsl_dataset_user_release(char *dsname, char *snapname, char *htag,
- boolean_t recursive)
-{
- struct dsl_ds_holdarg *ha;
- dsl_sync_task_t *dst;
- spa_t *spa;
- int error;
-
-top:
- ha = kmem_zalloc(sizeof (struct dsl_ds_holdarg), KM_SLEEP);
-
- (void) strlcpy(ha->failed, dsname, sizeof (ha->failed));
-
- error = spa_open(dsname, &spa, FTAG);
- if (error) {
- kmem_free(ha, sizeof (struct dsl_ds_holdarg));
+ error = dsl_prop_predict(ds->ds_dir,
+ zfs_prop_to_name(ZFS_PROP_REFRESERVATION),
+ ddsqra->ddsqra_source, ddsqra->ddsqra_value, &newval);
+ if (error != 0) {
+ dsl_dataset_rele(ds, FTAG);
return (error);
}
- ha->dstg = dsl_sync_task_group_create(spa_get_dsl(spa));
- ha->htag = htag;
- ha->snapname = snapname;
- ha->recursive = recursive;
- if (recursive) {
- error = dmu_objset_find(dsname, dsl_dataset_user_release_one,
- ha, DS_FIND_CHILDREN);
- } else {
- error = dsl_dataset_user_release_one(dsname, ha);
+ /*
+ * If we are doing the preliminary check in open context, the
+ * space estimates may be inaccurate.
+ */
+ if (!dmu_tx_is_syncing(tx)) {
+ dsl_dataset_rele(ds, FTAG);
+ return (0);
}
- if (error == 0)
- error = dsl_sync_task_group_wait(ha->dstg);
-
- for (dst = list_head(&ha->dstg->dstg_tasks); dst;
- dst = list_next(&ha->dstg->dstg_tasks, dst)) {
- struct dsl_ds_releasearg *ra = dst->dst_arg1;
- dsl_dataset_t *ds = ra->ds;
- if (dst->dst_err)
- dsl_dataset_name(ds, ha->failed);
+ mutex_enter(&ds->ds_lock);
+ if (!DS_UNIQUE_IS_ACCURATE(ds))
+ dsl_dataset_recalc_head_uniq(ds);
+ unique = dsl_dataset_phys(ds)->ds_unique_bytes;
+ mutex_exit(&ds->ds_lock);
- if (ra->own)
- dsl_dataset_disown(ds, ha->dstg);
- else
- dsl_dataset_rele(ds, ha->dstg);
+ if (MAX(unique, newval) > MAX(unique, ds->ds_reserved)) {
+ uint64_t delta = MAX(unique, newval) -
+ MAX(unique, ds->ds_reserved);
- kmem_free(ra, sizeof (struct dsl_ds_releasearg));
+ if (delta >
+ dsl_dir_space_available(ds->ds_dir, NULL, 0, B_TRUE) ||
+ (ds->ds_quota > 0 && newval > ds->ds_quota)) {
+ dsl_dataset_rele(ds, FTAG);
+ return (SET_ERROR(ENOSPC));
+ }
}
- if (error == 0 && recursive && !ha->gotone)
- error = ENOENT;
-
- if (error && error != EBUSY)
- (void) strlcpy(dsname, ha->failed, sizeof (ha->failed));
-
- dsl_sync_task_group_destroy(ha->dstg);
- kmem_free(ha, sizeof (struct dsl_ds_holdarg));
- spa_close(spa, FTAG);
-
- /*
- * We can get EBUSY if we were racing with deferred destroy and
- * dsl_dataset_user_release_check() hadn't done the necessary
- * open context setup. We can also get EBUSY if we're racing
- * with destroy and that thread is the ds_owner. Either way
- * the busy condition should be transient, and we should retry
- * the release operation.
- */
- if (error == EBUSY)
- goto top;
-
- return (error);
+ dsl_dataset_rele(ds, FTAG);
+ return (0);
}
-/*
- * Called at spa_load time (with retry == B_FALSE) to release a stale
- * temporary user hold. Also called by the onexit code (with retry == B_TRUE).
- */
-int
-dsl_dataset_user_release_tmp(dsl_pool_t *dp, uint64_t dsobj, char *htag,
- boolean_t retry)
+void
+dsl_dataset_set_refreservation_sync_impl(dsl_dataset_t *ds,
+ zprop_source_t source, uint64_t value, dmu_tx_t *tx)
{
- dsl_dataset_t *ds;
- char *snap;
- char *name;
- int namelen;
- int error;
+ uint64_t newval;
+ uint64_t unique;
+ int64_t delta;
- do {
- rw_enter(&dp->dp_config_rwlock, RW_READER);
- error = dsl_dataset_hold_obj(dp, dsobj, FTAG, &ds);
- rw_exit(&dp->dp_config_rwlock);
- if (error)
- return (error);
- namelen = dsl_dataset_namelen(ds)+1;
- name = kmem_alloc(namelen, KM_SLEEP);
- dsl_dataset_name(ds, name);
- dsl_dataset_rele(ds, FTAG);
+ dsl_prop_set_sync_impl(ds, zfs_prop_to_name(ZFS_PROP_REFRESERVATION),
+ source, sizeof (value), 1, &value, tx);
- snap = strchr(name, '@');
- *snap = '\0';
- ++snap;
- error = dsl_dataset_user_release(name, snap, htag, B_FALSE);
- kmem_free(name, namelen);
+ VERIFY0(dsl_prop_get_int_ds(ds,
+ zfs_prop_to_name(ZFS_PROP_REFRESERVATION), &newval));
- /*
- * The object can't have been destroyed because we have a hold,
- * but it might have been renamed, resulting in ENOENT. Retry
- * if we've been requested to do so.
- *
- * It would be nice if we could use the dsobj all the way
- * through and avoid ENOENT entirely. But we might need to
- * unmount the snapshot, and there's currently no way to lookup
- * a vfsp using a ZFS object id.
- */
- } while ((error == ENOENT) && retry);
+ dmu_buf_will_dirty(ds->ds_dbuf, tx);
+ mutex_enter(&ds->ds_dir->dd_lock);
+ mutex_enter(&ds->ds_lock);
+ ASSERT(DS_UNIQUE_IS_ACCURATE(ds));
+ unique = dsl_dataset_phys(ds)->ds_unique_bytes;
+ delta = MAX(0, (int64_t)(newval - unique)) -
+ MAX(0, (int64_t)(ds->ds_reserved - unique));
+ ds->ds_reserved = newval;
+ mutex_exit(&ds->ds_lock);
- return (error);
+ dsl_dir_diduse_space(ds->ds_dir, DD_USED_REFRSRV, delta, 0, 0, tx);
+ mutex_exit(&ds->ds_dir->dd_lock);
}
-int
-dsl_dataset_get_holds(const char *dsname, nvlist_t **nvp)
+static void
+dsl_dataset_set_refreservation_sync(void *arg, dmu_tx_t *tx)
{
- dsl_dataset_t *ds;
- int err;
-
- err = dsl_dataset_hold(dsname, FTAG, &ds);
- if (err)
- return (err);
+ dsl_dataset_set_qr_arg_t *ddsqra = arg;
+ dsl_pool_t *dp = dmu_tx_pool(tx);
+ dsl_dataset_t *ds = NULL;
- VERIFY(0 == nvlist_alloc(nvp, NV_UNIQUE_NAME, KM_SLEEP));
- if (ds->ds_phys->ds_userrefs_obj != 0) {
- zap_attribute_t *za;
- zap_cursor_t zc;
-
- za = kmem_alloc(sizeof (zap_attribute_t), KM_SLEEP);
- for (zap_cursor_init(&zc, ds->ds_dir->dd_pool->dp_meta_objset,
- ds->ds_phys->ds_userrefs_obj);
- zap_cursor_retrieve(&zc, za) == 0;
- zap_cursor_advance(&zc)) {
- VERIFY(0 == nvlist_add_uint64(*nvp, za->za_name,
- za->za_first_integer));
- }
- zap_cursor_fini(&zc);
- kmem_free(za, sizeof (zap_attribute_t));
- }
+ VERIFY0(dsl_dataset_hold(dp, ddsqra->ddsqra_name, FTAG, &ds));
+ dsl_dataset_set_refreservation_sync_impl(ds,
+ ddsqra->ddsqra_source, ddsqra->ddsqra_value, tx);
dsl_dataset_rele(ds, FTAG);
- return (0);
}
-/*
- * Note, this function is used as the callback for dmu_objset_find(). We
- * always return 0 so that we will continue to find and process
- * inconsistent datasets, even if we encounter an error trying to
- * process one of them.
- */
-/* ARGSUSED */
int
-dsl_destroy_inconsistent(const char *dsname, void *arg)
+dsl_dataset_set_refreservation(const char *dsname, zprop_source_t source,
+ uint64_t refreservation)
{
- dsl_dataset_t *ds;
+ dsl_dataset_set_qr_arg_t ddsqra;
- if (dsl_dataset_own(dsname, B_TRUE, FTAG, &ds) == 0) {
- if (DS_IS_INCONSISTENT(ds))
- (void) dsl_dataset_destroy(ds, FTAG, B_FALSE);
- else
- dsl_dataset_disown(ds, FTAG);
- }
- return (0);
-}
+ ddsqra.ddsqra_name = dsname;
+ ddsqra.ddsqra_source = source;
+ ddsqra.ddsqra_value = refreservation;
+ return (dsl_sync_task(dsname, dsl_dataset_set_refreservation_check,
+ dsl_dataset_set_refreservation_sync, &ddsqra, 0,
+ ZFS_SPACE_CHECK_EXTRA_RESERVED));
+}
/*
* Return (in *usedp) the amount of space written in new that is not
uint64_t snapobj;
dsl_pool_t *dp = new->ds_dir->dd_pool;
+ ASSERT(dsl_pool_config_held(dp));
+
*usedp = 0;
- *usedp += new->ds_phys->ds_referenced_bytes;
- *usedp -= oldsnap->ds_phys->ds_referenced_bytes;
+ *usedp += dsl_dataset_phys(new)->ds_referenced_bytes;
+ *usedp -= dsl_dataset_phys(oldsnap)->ds_referenced_bytes;
*compp = 0;
- *compp += new->ds_phys->ds_compressed_bytes;
- *compp -= oldsnap->ds_phys->ds_compressed_bytes;
+ *compp += dsl_dataset_phys(new)->ds_compressed_bytes;
+ *compp -= dsl_dataset_phys(oldsnap)->ds_compressed_bytes;
*uncompp = 0;
- *uncompp += new->ds_phys->ds_uncompressed_bytes;
- *uncompp -= oldsnap->ds_phys->ds_uncompressed_bytes;
+ *uncompp += dsl_dataset_phys(new)->ds_uncompressed_bytes;
+ *uncompp -= dsl_dataset_phys(oldsnap)->ds_uncompressed_bytes;
- rw_enter(&dp->dp_config_rwlock, RW_READER);
snapobj = new->ds_object;
while (snapobj != oldsnap->ds_object) {
dsl_dataset_t *snap;
break;
}
- if (snap->ds_phys->ds_prev_snap_txg ==
- oldsnap->ds_phys->ds_creation_txg) {
+ if (dsl_dataset_phys(snap)->ds_prev_snap_txg ==
+ dsl_dataset_phys(oldsnap)->ds_creation_txg) {
/*
* The blocks in the deadlist can not be born after
* ds_prev_snap_txg, so get the whole deadlist space,
&used, &comp, &uncomp);
} else {
dsl_deadlist_space_range(&snap->ds_deadlist,
- 0, oldsnap->ds_phys->ds_creation_txg,
+ 0, dsl_dataset_phys(oldsnap)->ds_creation_txg,
&used, &comp, &uncomp);
}
*usedp += used;
* (ds_prev_snap_obj == 0) before oldsnap, then oldsnap
* was not a snapshot of/before new.
*/
- snapobj = snap->ds_phys->ds_prev_snap_obj;
+ snapobj = dsl_dataset_phys(snap)->ds_prev_snap_obj;
if (snap != new)
dsl_dataset_rele(snap, FTAG);
if (snapobj == 0) {
- err = EINVAL;
+ err = SET_ERROR(EINVAL);
break;
}
}
- rw_exit(&dp->dp_config_rwlock);
return (err);
}
uint64_t snapobj;
dsl_pool_t *dp = firstsnap->ds_dir->dd_pool;
- ASSERT(dsl_dataset_is_snapshot(firstsnap));
- ASSERT(dsl_dataset_is_snapshot(lastsnap));
+ ASSERT(firstsnap->ds_is_snapshot);
+ ASSERT(lastsnap->ds_is_snapshot);
/*
* Check that the snapshots are in the same dsl_dir, and firstsnap
* is before lastsnap.
*/
if (firstsnap->ds_dir != lastsnap->ds_dir ||
- firstsnap->ds_phys->ds_creation_txg >
- lastsnap->ds_phys->ds_creation_txg)
- return (EINVAL);
+ dsl_dataset_phys(firstsnap)->ds_creation_txg >
+ dsl_dataset_phys(lastsnap)->ds_creation_txg)
+ return (SET_ERROR(EINVAL));
*usedp = *compp = *uncompp = 0;
- rw_enter(&dp->dp_config_rwlock, RW_READER);
- snapobj = lastsnap->ds_phys->ds_next_snap_obj;
+ snapobj = dsl_dataset_phys(lastsnap)->ds_next_snap_obj;
while (snapobj != firstsnap->ds_object) {
dsl_dataset_t *ds;
uint64_t used, comp, uncomp;
break;
dsl_deadlist_space_range(&ds->ds_deadlist,
- firstsnap->ds_phys->ds_prev_snap_txg, UINT64_MAX,
+ dsl_dataset_phys(firstsnap)->ds_prev_snap_txg, UINT64_MAX,
&used, &comp, &uncomp);
*usedp += used;
*compp += comp;
*uncompp += uncomp;
- snapobj = ds->ds_phys->ds_prev_snap_obj;
+ snapobj = dsl_dataset_phys(ds)->ds_prev_snap_obj;
ASSERT3U(snapobj, !=, 0);
dsl_dataset_rele(ds, FTAG);
}
- rw_exit(&dp->dp_config_rwlock);
return (err);
}
-#if defined(_KERNEL) && defined(HAVE_SPL)
-EXPORT_SYMBOL(dmu_snapshots_destroy_nvl);
+/*
+ * Return TRUE if 'earlier' is an earlier snapshot in 'later's timeline.
+ * For example, they could both be snapshots of the same filesystem, and
+ * 'earlier' is before 'later'. Or 'earlier' could be the origin of
+ * 'later's filesystem. Or 'earlier' could be an older snapshot in the origin's
+ * filesystem. Or 'earlier' could be the origin's origin.
+ *
+ * If non-zero, earlier_txg is used instead of earlier's ds_creation_txg.
+ */
+boolean_t
+dsl_dataset_is_before(dsl_dataset_t *later, dsl_dataset_t *earlier,
+ uint64_t earlier_txg)
+{
+ dsl_pool_t *dp = later->ds_dir->dd_pool;
+ int error;
+ boolean_t ret;
+
+ ASSERT(dsl_pool_config_held(dp));
+ ASSERT(earlier->ds_is_snapshot || earlier_txg != 0);
+
+ if (earlier_txg == 0)
+ earlier_txg = dsl_dataset_phys(earlier)->ds_creation_txg;
+
+ if (later->ds_is_snapshot &&
+ earlier_txg >= dsl_dataset_phys(later)->ds_creation_txg)
+ return (B_FALSE);
+
+ if (later->ds_dir == earlier->ds_dir)
+ return (B_TRUE);
+ if (!dsl_dir_is_clone(later->ds_dir))
+ return (B_FALSE);
+
+ if (dsl_dir_phys(later->ds_dir)->dd_origin_obj == earlier->ds_object)
+ return (B_TRUE);
+ dsl_dataset_t *origin;
+ error = dsl_dataset_hold_obj(dp,
+ dsl_dir_phys(later->ds_dir)->dd_origin_obj, FTAG, &origin);
+ if (error != 0)
+ return (B_FALSE);
+ ret = dsl_dataset_is_before(origin, earlier, earlier_txg);
+ dsl_dataset_rele(origin, FTAG);
+ return (ret);
+}
+
+void
+dsl_dataset_zapify(dsl_dataset_t *ds, dmu_tx_t *tx)
+{
+ objset_t *mos = ds->ds_dir->dd_pool->dp_meta_objset;
+ dmu_object_zapify(mos, ds->ds_object, DMU_OT_DSL_DATASET, tx);
+}
+
+boolean_t
+dsl_dataset_is_zapified(dsl_dataset_t *ds)
+{
+ dmu_object_info_t doi;
+
+ dmu_object_info_from_db(ds->ds_dbuf, &doi);
+ return (doi.doi_type == DMU_OTN_ZAP_METADATA);
+}
+
+boolean_t
+dsl_dataset_has_resume_receive_state(dsl_dataset_t *ds)
+{
+ return (dsl_dataset_is_zapified(ds) &&
+ zap_contains(ds->ds_dir->dd_pool->dp_meta_objset,
+ ds->ds_object, DS_FIELD_RESUME_TOGUID) == 0);
+}
+
+uint64_t
+dsl_dataset_get_remap_deadlist_object(dsl_dataset_t *ds)
+{
+ uint64_t remap_deadlist_obj;
+ int err;
+
+ if (!dsl_dataset_is_zapified(ds))
+ return (0);
+
+ err = zap_lookup(ds->ds_dir->dd_pool->dp_meta_objset, ds->ds_object,
+ DS_FIELD_REMAP_DEADLIST, sizeof (remap_deadlist_obj), 1,
+ &remap_deadlist_obj);
+
+ if (err != 0) {
+ VERIFY3S(err, ==, ENOENT);
+ return (0);
+ }
+
+ ASSERT(remap_deadlist_obj != 0);
+ return (remap_deadlist_obj);
+}
+
+boolean_t
+dsl_dataset_remap_deadlist_exists(dsl_dataset_t *ds)
+{
+ EQUIV(dsl_deadlist_is_open(&ds->ds_remap_deadlist),
+ dsl_dataset_get_remap_deadlist_object(ds) != 0);
+ return (dsl_deadlist_is_open(&ds->ds_remap_deadlist));
+}
+
+static void
+dsl_dataset_set_remap_deadlist_object(dsl_dataset_t *ds, uint64_t obj,
+ dmu_tx_t *tx)
+{
+ ASSERT(obj != 0);
+ dsl_dataset_zapify(ds, tx);
+ VERIFY0(zap_add(ds->ds_dir->dd_pool->dp_meta_objset, ds->ds_object,
+ DS_FIELD_REMAP_DEADLIST, sizeof (obj), 1, &obj, tx));
+}
+
+static void
+dsl_dataset_unset_remap_deadlist_object(dsl_dataset_t *ds, dmu_tx_t *tx)
+{
+ VERIFY0(zap_remove(ds->ds_dir->dd_pool->dp_meta_objset,
+ ds->ds_object, DS_FIELD_REMAP_DEADLIST, tx));
+}
+
+void
+dsl_dataset_destroy_remap_deadlist(dsl_dataset_t *ds, dmu_tx_t *tx)
+{
+ uint64_t remap_deadlist_object;
+ spa_t *spa = ds->ds_dir->dd_pool->dp_spa;
+
+ ASSERT(dmu_tx_is_syncing(tx));
+ ASSERT(dsl_dataset_remap_deadlist_exists(ds));
+
+ remap_deadlist_object = ds->ds_remap_deadlist.dl_object;
+ dsl_deadlist_close(&ds->ds_remap_deadlist);
+ dsl_deadlist_free(spa_meta_objset(spa), remap_deadlist_object, tx);
+ dsl_dataset_unset_remap_deadlist_object(ds, tx);
+ spa_feature_decr(spa, SPA_FEATURE_OBSOLETE_COUNTS, tx);
+}
+
+void
+dsl_dataset_create_remap_deadlist(dsl_dataset_t *ds, dmu_tx_t *tx)
+{
+ uint64_t remap_deadlist_obj;
+ spa_t *spa = ds->ds_dir->dd_pool->dp_spa;
+
+ ASSERT(dmu_tx_is_syncing(tx));
+ ASSERT(MUTEX_HELD(&ds->ds_remap_deadlist_lock));
+ /*
+ * Currently we only create remap deadlists when there are indirect
+ * vdevs with referenced mappings.
+ */
+ ASSERT(spa_feature_is_active(spa, SPA_FEATURE_DEVICE_REMOVAL));
+
+ remap_deadlist_obj = dsl_deadlist_clone(
+ &ds->ds_deadlist, UINT64_MAX,
+ dsl_dataset_phys(ds)->ds_prev_snap_obj, tx);
+ dsl_dataset_set_remap_deadlist_object(ds,
+ remap_deadlist_obj, tx);
+ dsl_deadlist_open(&ds->ds_remap_deadlist, spa_meta_objset(spa),
+ remap_deadlist_obj);
+ spa_feature_incr(spa, SPA_FEATURE_OBSOLETE_COUNTS, tx);
+}
+
+#if defined(_KERNEL)
+#if defined(_LP64)
+module_param(zfs_max_recordsize, int, 0644);
+MODULE_PARM_DESC(zfs_max_recordsize, "Max allowed record size");
+#else
+/* Limited to 1M on 32-bit platforms due to lack of virtual address space */
+module_param(zfs_max_recordsize, int, 0444);
+MODULE_PARM_DESC(zfs_max_recordsize, "Max allowed record size");
+#endif
+
EXPORT_SYMBOL(dsl_dataset_hold);
+EXPORT_SYMBOL(dsl_dataset_hold_flags);
EXPORT_SYMBOL(dsl_dataset_hold_obj);
+EXPORT_SYMBOL(dsl_dataset_hold_obj_flags);
EXPORT_SYMBOL(dsl_dataset_own);
EXPORT_SYMBOL(dsl_dataset_own_obj);
EXPORT_SYMBOL(dsl_dataset_name);
EXPORT_SYMBOL(dsl_dataset_rele);
+EXPORT_SYMBOL(dsl_dataset_rele_flags);
EXPORT_SYMBOL(dsl_dataset_disown);
-EXPORT_SYMBOL(dsl_dataset_drop_ref);
EXPORT_SYMBOL(dsl_dataset_tryown);
-EXPORT_SYMBOL(dsl_dataset_make_exclusive);
EXPORT_SYMBOL(dsl_dataset_create_sync);
EXPORT_SYMBOL(dsl_dataset_create_sync_dd);
-EXPORT_SYMBOL(dsl_dataset_destroy);
-EXPORT_SYMBOL(dsl_dataset_destroy_check);
-EXPORT_SYMBOL(dsl_dataset_destroy_sync);
EXPORT_SYMBOL(dsl_dataset_snapshot_check);
EXPORT_SYMBOL(dsl_dataset_snapshot_sync);
-EXPORT_SYMBOL(dsl_dataset_rename);
EXPORT_SYMBOL(dsl_dataset_promote);
-EXPORT_SYMBOL(dsl_dataset_clone_swap);
EXPORT_SYMBOL(dsl_dataset_user_hold);
EXPORT_SYMBOL(dsl_dataset_user_release);
-EXPORT_SYMBOL(dsl_dataset_user_release_tmp);
EXPORT_SYMBOL(dsl_dataset_get_holds);
EXPORT_SYMBOL(dsl_dataset_get_blkptr);
-EXPORT_SYMBOL(dsl_dataset_set_blkptr);
EXPORT_SYMBOL(dsl_dataset_get_spa);
-EXPORT_SYMBOL(dsl_dataset_modified_since_lastsnap);
+EXPORT_SYMBOL(dsl_dataset_modified_since_snap);
EXPORT_SYMBOL(dsl_dataset_space_written);
EXPORT_SYMBOL(dsl_dataset_space_wouldfree);
EXPORT_SYMBOL(dsl_dataset_sync);
EXPORT_SYMBOL(dsl_dataset_block_born);
EXPORT_SYMBOL(dsl_dataset_block_kill);
-EXPORT_SYMBOL(dsl_dataset_block_freeable);
-EXPORT_SYMBOL(dsl_dataset_prev_snap_txg);
EXPORT_SYMBOL(dsl_dataset_dirty);
EXPORT_SYMBOL(dsl_dataset_stats);
EXPORT_SYMBOL(dsl_dataset_fast_stat);
EXPORT_SYMBOL(dsl_dataset_fsid_guid);
EXPORT_SYMBOL(dsl_dsobj_to_dsname);
EXPORT_SYMBOL(dsl_dataset_check_quota);
-EXPORT_SYMBOL(dsl_dataset_set_quota);
-EXPORT_SYMBOL(dsl_dataset_set_quota_sync);
-EXPORT_SYMBOL(dsl_dataset_set_reservation);
-EXPORT_SYMBOL(dsl_destroy_inconsistent);
+EXPORT_SYMBOL(dsl_dataset_clone_swap_check_impl);
+EXPORT_SYMBOL(dsl_dataset_clone_swap_sync_impl);
#endif