*/
/*
* Copyright (c) 2005, 2010, Oracle and/or its affiliates. All rights reserved.
- * Copyright (c) 2011, 2016 by Delphix. All rights reserved.
+ * Copyright (c) 2011, 2017 by Delphix. All rights reserved.
* Copyright (c) 2013 by Saso Kiselkov. All rights reserved.
* Copyright (c) 2013, Joyent, Inc. All rights reserved.
- * Copyright (c) 2014, Nexenta Systems, Inc. All rights reserved.
+ * Copyright (c) 2016, Nexenta Systems, Inc. All rights reserved.
* Copyright (c) 2015 by Chunwei Chen. All rights reserved.
*/
#include <sys/zio_compress.h>
#include <sys/sa.h>
#include <sys/zfeature.h>
+#include <sys/abd.h>
+#include <sys/trace_dmu.h>
+#include <sys/zfs_rlock.h>
#ifdef _KERNEL
#include <sys/vmsystm.h>
#include <sys/zfs_znode.h>
*/
int zfs_nopwrite_enabled = 1;
+/*
+ * Tunable to control percentage of dirtied blocks from frees in one TXG.
+ * After this threshold is crossed, additional dirty blocks from frees
+ * wait until the next TXG.
+ * A value of zero will disable this throttle.
+ */
+unsigned long zfs_per_txg_dirty_frees_percent = 30;
+
+/*
+ * Enable/disable forcing txg sync when dirty in dmu_offset_next.
+ */
+int zfs_dmu_offset_next_sync = 0;
+
+/*
+ * This can be used for testing, to ensure that certain actions happen
+ * while in the middle of a remap (which might otherwise complete too
+ * quickly).
+ */
+int zfs_object_remap_one_indirect_delay_ticks = 0;
+
const dmu_object_type_info_t dmu_ot[DMU_OT_NUMTYPES] = {
- { DMU_BSWAP_UINT8, TRUE, "unallocated" },
- { DMU_BSWAP_ZAP, TRUE, "object directory" },
- { DMU_BSWAP_UINT64, TRUE, "object array" },
- { DMU_BSWAP_UINT8, TRUE, "packed nvlist" },
- { DMU_BSWAP_UINT64, TRUE, "packed nvlist size" },
- { DMU_BSWAP_UINT64, TRUE, "bpobj" },
- { DMU_BSWAP_UINT64, TRUE, "bpobj header" },
- { DMU_BSWAP_UINT64, TRUE, "SPA space map header" },
- { DMU_BSWAP_UINT64, TRUE, "SPA space map" },
- { DMU_BSWAP_UINT64, TRUE, "ZIL intent log" },
- { DMU_BSWAP_DNODE, TRUE, "DMU dnode" },
- { DMU_BSWAP_OBJSET, TRUE, "DMU objset" },
- { DMU_BSWAP_UINT64, TRUE, "DSL directory" },
- { DMU_BSWAP_ZAP, TRUE, "DSL directory child map"},
- { DMU_BSWAP_ZAP, TRUE, "DSL dataset snap map" },
- { DMU_BSWAP_ZAP, TRUE, "DSL props" },
- { DMU_BSWAP_UINT64, TRUE, "DSL dataset" },
- { DMU_BSWAP_ZNODE, TRUE, "ZFS znode" },
- { DMU_BSWAP_OLDACL, TRUE, "ZFS V0 ACL" },
- { DMU_BSWAP_UINT8, FALSE, "ZFS plain file" },
- { DMU_BSWAP_ZAP, TRUE, "ZFS directory" },
- { DMU_BSWAP_ZAP, TRUE, "ZFS master node" },
- { DMU_BSWAP_ZAP, TRUE, "ZFS delete queue" },
- { DMU_BSWAP_UINT8, FALSE, "zvol object" },
- { DMU_BSWAP_ZAP, TRUE, "zvol prop" },
- { DMU_BSWAP_UINT8, FALSE, "other uint8[]" },
- { DMU_BSWAP_UINT64, FALSE, "other uint64[]" },
- { DMU_BSWAP_ZAP, TRUE, "other ZAP" },
- { DMU_BSWAP_ZAP, TRUE, "persistent error log" },
- { DMU_BSWAP_UINT8, TRUE, "SPA history" },
- { DMU_BSWAP_UINT64, TRUE, "SPA history offsets" },
- { DMU_BSWAP_ZAP, TRUE, "Pool properties" },
- { DMU_BSWAP_ZAP, TRUE, "DSL permissions" },
- { DMU_BSWAP_ACL, TRUE, "ZFS ACL" },
- { DMU_BSWAP_UINT8, TRUE, "ZFS SYSACL" },
- { DMU_BSWAP_UINT8, TRUE, "FUID table" },
- { DMU_BSWAP_UINT64, TRUE, "FUID table size" },
- { DMU_BSWAP_ZAP, TRUE, "DSL dataset next clones"},
- { DMU_BSWAP_ZAP, TRUE, "scan work queue" },
- { DMU_BSWAP_ZAP, TRUE, "ZFS user/group used" },
- { DMU_BSWAP_ZAP, TRUE, "ZFS user/group quota" },
- { DMU_BSWAP_ZAP, TRUE, "snapshot refcount tags"},
- { DMU_BSWAP_ZAP, TRUE, "DDT ZAP algorithm" },
- { DMU_BSWAP_ZAP, TRUE, "DDT statistics" },
- { DMU_BSWAP_UINT8, TRUE, "System attributes" },
- { DMU_BSWAP_ZAP, TRUE, "SA master node" },
- { DMU_BSWAP_ZAP, TRUE, "SA attr registration" },
- { DMU_BSWAP_ZAP, TRUE, "SA attr layouts" },
- { DMU_BSWAP_ZAP, TRUE, "scan translations" },
- { DMU_BSWAP_UINT8, FALSE, "deduplicated block" },
- { DMU_BSWAP_ZAP, TRUE, "DSL deadlist map" },
- { DMU_BSWAP_UINT64, TRUE, "DSL deadlist map hdr" },
- { DMU_BSWAP_ZAP, TRUE, "DSL dir clones" },
- { DMU_BSWAP_UINT64, TRUE, "bpobj subobj" }
+ { DMU_BSWAP_UINT8, TRUE, FALSE, "unallocated" },
+ { DMU_BSWAP_ZAP, TRUE, FALSE, "object directory" },
+ { DMU_BSWAP_UINT64, TRUE, FALSE, "object array" },
+ { DMU_BSWAP_UINT8, TRUE, FALSE, "packed nvlist" },
+ { DMU_BSWAP_UINT64, TRUE, FALSE, "packed nvlist size" },
+ { DMU_BSWAP_UINT64, TRUE, FALSE, "bpobj" },
+ { DMU_BSWAP_UINT64, TRUE, FALSE, "bpobj header" },
+ { DMU_BSWAP_UINT64, TRUE, FALSE, "SPA space map header" },
+ { DMU_BSWAP_UINT64, TRUE, FALSE, "SPA space map" },
+ { DMU_BSWAP_UINT64, TRUE, TRUE, "ZIL intent log" },
+ { DMU_BSWAP_DNODE, TRUE, TRUE, "DMU dnode" },
+ { DMU_BSWAP_OBJSET, TRUE, FALSE, "DMU objset" },
+ { DMU_BSWAP_UINT64, TRUE, FALSE, "DSL directory" },
+ { DMU_BSWAP_ZAP, TRUE, FALSE, "DSL directory child map"},
+ { DMU_BSWAP_ZAP, TRUE, FALSE, "DSL dataset snap map" },
+ { DMU_BSWAP_ZAP, TRUE, FALSE, "DSL props" },
+ { DMU_BSWAP_UINT64, TRUE, FALSE, "DSL dataset" },
+ { DMU_BSWAP_ZNODE, TRUE, FALSE, "ZFS znode" },
+ { DMU_BSWAP_OLDACL, TRUE, TRUE, "ZFS V0 ACL" },
+ { DMU_BSWAP_UINT8, FALSE, TRUE, "ZFS plain file" },
+ { DMU_BSWAP_ZAP, TRUE, TRUE, "ZFS directory" },
+ { DMU_BSWAP_ZAP, TRUE, FALSE, "ZFS master node" },
+ { DMU_BSWAP_ZAP, TRUE, TRUE, "ZFS delete queue" },
+ { DMU_BSWAP_UINT8, FALSE, TRUE, "zvol object" },
+ { DMU_BSWAP_ZAP, TRUE, FALSE, "zvol prop" },
+ { DMU_BSWAP_UINT8, FALSE, TRUE, "other uint8[]" },
+ { DMU_BSWAP_UINT64, FALSE, TRUE, "other uint64[]" },
+ { DMU_BSWAP_ZAP, TRUE, FALSE, "other ZAP" },
+ { DMU_BSWAP_ZAP, TRUE, FALSE, "persistent error log" },
+ { DMU_BSWAP_UINT8, TRUE, FALSE, "SPA history" },
+ { DMU_BSWAP_UINT64, TRUE, FALSE, "SPA history offsets" },
+ { DMU_BSWAP_ZAP, TRUE, FALSE, "Pool properties" },
+ { DMU_BSWAP_ZAP, TRUE, FALSE, "DSL permissions" },
+ { DMU_BSWAP_ACL, TRUE, TRUE, "ZFS ACL" },
+ { DMU_BSWAP_UINT8, TRUE, TRUE, "ZFS SYSACL" },
+ { DMU_BSWAP_UINT8, TRUE, TRUE, "FUID table" },
+ { DMU_BSWAP_UINT64, TRUE, FALSE, "FUID table size" },
+ { DMU_BSWAP_ZAP, TRUE, FALSE, "DSL dataset next clones"},
+ { DMU_BSWAP_ZAP, TRUE, FALSE, "scan work queue" },
+ { DMU_BSWAP_ZAP, TRUE, TRUE, "ZFS user/group/project used" },
+ { DMU_BSWAP_ZAP, TRUE, TRUE, "ZFS user/group/project quota"},
+ { DMU_BSWAP_ZAP, TRUE, FALSE, "snapshot refcount tags"},
+ { DMU_BSWAP_ZAP, TRUE, FALSE, "DDT ZAP algorithm" },
+ { DMU_BSWAP_ZAP, TRUE, FALSE, "DDT statistics" },
+ { DMU_BSWAP_UINT8, TRUE, TRUE, "System attributes" },
+ { DMU_BSWAP_ZAP, TRUE, TRUE, "SA master node" },
+ { DMU_BSWAP_ZAP, TRUE, TRUE, "SA attr registration" },
+ { DMU_BSWAP_ZAP, TRUE, TRUE, "SA attr layouts" },
+ { DMU_BSWAP_ZAP, TRUE, FALSE, "scan translations" },
+ { DMU_BSWAP_UINT8, FALSE, TRUE, "deduplicated block" },
+ { DMU_BSWAP_ZAP, TRUE, FALSE, "DSL deadlist map" },
+ { DMU_BSWAP_UINT64, TRUE, FALSE, "DSL deadlist map hdr" },
+ { DMU_BSWAP_ZAP, TRUE, FALSE, "DSL dir clones" },
+ { DMU_BSWAP_UINT64, TRUE, FALSE, "bpobj subobj" }
};
const dmu_object_byteswap_info_t dmu_ot_byteswap[DMU_BSWAP_NUMFUNCS] = {
if (flags & DMU_READ_NO_PREFETCH)
db_flags |= DB_RF_NOPREFETCH;
+ if (flags & DMU_READ_NO_DECRYPT)
+ db_flags |= DB_RF_NO_DECRYPT;
err = dmu_buf_hold_noread_by_dnode(dn, offset, tag, dbp);
if (err == 0) {
if (flags & DMU_READ_NO_PREFETCH)
db_flags |= DB_RF_NOPREFETCH;
+ if (flags & DMU_READ_NO_DECRYPT)
+ db_flags |= DB_RF_NO_DECRYPT;
err = dmu_buf_hold_noread(os, object, offset, tag, dbp);
if (err == 0) {
* returns ENOENT, EIO, or 0.
*/
int
-dmu_bonus_hold(objset_t *os, uint64_t object, void *tag, dmu_buf_t **dbp)
+dmu_bonus_hold_impl(objset_t *os, uint64_t object, void *tag, uint32_t flags,
+ dmu_buf_t **dbp)
{
dnode_t *dn;
dmu_buf_impl_t *db;
int error;
+ uint32_t db_flags = DB_RF_MUST_SUCCEED;
+
+ if (flags & DMU_READ_NO_PREFETCH)
+ db_flags |= DB_RF_NOPREFETCH;
+ if (flags & DMU_READ_NO_DECRYPT)
+ db_flags |= DB_RF_NO_DECRYPT;
error = dnode_hold(os, object, FTAG, &dn);
if (error)
dnode_rele(dn, FTAG);
- VERIFY(0 == dbuf_read(db, NULL, DB_RF_MUST_SUCCEED | DB_RF_NOPREFETCH));
+ error = dbuf_read(db, NULL, db_flags);
+ if (error) {
+ dnode_evict_bonus(dn);
+ dbuf_rele(db, tag);
+ *dbp = NULL;
+ return (error);
+ }
*dbp = &db->db;
return (0);
}
+int
+dmu_bonus_hold(objset_t *os, uint64_t obj, void *tag, dmu_buf_t **dbp)
+{
+ return (dmu_bonus_hold_impl(os, obj, tag, DMU_READ_NO_PREFETCH, dbp));
+}
+
/*
* returns ENOENT, EIO, or 0.
*
if ((flags & DB_RF_HAVESTRUCT) == 0)
rw_exit(&dn->dn_struct_rwlock);
- ASSERT(db != NULL);
+ if (db == NULL) {
+ *dbp = NULL;
+ return (SET_ERROR(EIO));
+ }
err = dbuf_read(db, NULL, flags);
if (err == 0)
*dbp = &db->db;
- else
+ else {
dbuf_rele(db, tag);
+ *dbp = NULL;
+ }
return (err);
}
* indirect blocks prefeteched will be those that point to the blocks containing
* the data starting at offset, and continuing to offset + len.
*
- * Note that if the indirect blocks above the blocks being prefetched are not in
- * cache, they will be asychronously read in.
+ * Note that if the indirect blocks above the blocks being prefetched are not
+ * in cache, they will be asychronously read in.
*/
void
dmu_prefetch(objset_t *os, uint64_t object, int64_t level, uint64_t offset,
}
if (nblks != 0) {
- int i;
-
blkid = dbuf_whichblock(dn, level, offset);
- for (i = 0; i < nblks; i++)
+ for (int i = 0; i < nblks; i++)
dbuf_prefetch(dn, level, blkid + i, pri, 0);
}
/* bytes of data covered by a level-1 indirect block */
uint64_t iblkrange =
dn->dn_datablksz * EPB(dn->dn_indblkshift, SPA_BLKPTRSHIFT);
- uint64_t blks;
ASSERT3U(minimum, <=, *start);
}
ASSERT(ISP2(iblkrange));
- for (blks = 0; *start > minimum && blks < maxblks; blks++) {
+ for (uint64_t blks = 0; *start > minimum && blks < maxblks; blks++) {
int err;
/*
return (0);
}
+/*
+ * If this objset is of type OST_ZFS return true if vfs's unmounted flag is set,
+ * otherwise return false.
+ * Used below in dmu_free_long_range_impl() to enable abort when unmounting
+ */
+/*ARGSUSED*/
+static boolean_t
+dmu_objset_zfs_unmounting(objset_t *os)
+{
+#ifdef _KERNEL
+ if (dmu_objset_type(os) == DMU_OST_ZFS)
+ return (zfs_get_vfs_flag_unmounted(os));
+#endif
+ return (B_FALSE);
+}
+
static int
dmu_free_long_range_impl(objset_t *os, dnode_t *dn, uint64_t offset,
- uint64_t length)
+ uint64_t length, boolean_t raw)
{
uint64_t object_size;
int err;
+ uint64_t dirty_frees_threshold;
+ dsl_pool_t *dp = dmu_objset_pool(os);
if (dn == NULL)
return (SET_ERROR(EINVAL));
if (offset >= object_size)
return (0);
+ if (zfs_per_txg_dirty_frees_percent <= 100)
+ dirty_frees_threshold =
+ zfs_per_txg_dirty_frees_percent * zfs_dirty_data_max / 100;
+ else
+ dirty_frees_threshold = zfs_dirty_data_max / 4;
+
if (length == DMU_OBJECT_END || offset + length > object_size)
length = object_size - offset;
while (length != 0) {
- uint64_t chunk_end, chunk_begin;
+ uint64_t chunk_end, chunk_begin, chunk_len;
+ uint64_t long_free_dirty_all_txgs = 0;
dmu_tx_t *tx;
+ if (dmu_objset_zfs_unmounting(dn->dn_objset))
+ return (SET_ERROR(EINTR));
+
chunk_end = chunk_begin = offset + length;
/* move chunk_begin backwards to the beginning of this chunk */
ASSERT3U(chunk_begin, >=, offset);
ASSERT3U(chunk_begin, <=, chunk_end);
+ chunk_len = chunk_end - chunk_begin;
+
+ mutex_enter(&dp->dp_lock);
+ for (int t = 0; t < TXG_SIZE; t++) {
+ long_free_dirty_all_txgs +=
+ dp->dp_long_free_dirty_pertxg[t];
+ }
+ mutex_exit(&dp->dp_lock);
+
+ /*
+ * To avoid filling up a TXG with just frees wait for
+ * the next TXG to open before freeing more chunks if
+ * we have reached the threshold of frees
+ */
+ if (dirty_frees_threshold != 0 &&
+ long_free_dirty_all_txgs >= dirty_frees_threshold) {
+ txg_wait_open(dp, 0);
+ continue;
+ }
+
tx = dmu_tx_create(os);
- dmu_tx_hold_free(tx, dn->dn_object,
- chunk_begin, chunk_end - chunk_begin);
+ dmu_tx_hold_free(tx, dn->dn_object, chunk_begin, chunk_len);
/*
* Mark this transaction as typically resulting in a net
dmu_tx_abort(tx);
return (err);
}
- dnode_free_range(dn, chunk_begin, chunk_end - chunk_begin, tx);
+
+ mutex_enter(&dp->dp_lock);
+ dp->dp_long_free_dirty_pertxg[dmu_tx_get_txg(tx) & TXG_MASK] +=
+ chunk_len;
+ mutex_exit(&dp->dp_lock);
+ DTRACE_PROBE3(free__long__range,
+ uint64_t, long_free_dirty_all_txgs, uint64_t, chunk_len,
+ uint64_t, dmu_tx_get_txg(tx));
+ dnode_free_range(dn, chunk_begin, chunk_len, tx);
+
+ /* if this is a raw free, mark the dirty record as such */
+ if (raw) {
+ dbuf_dirty_record_t *dr = dn->dn_dbuf->db_last_dirty;
+
+ while (dr != NULL && dr->dr_txg > tx->tx_txg)
+ dr = dr->dr_next;
+ if (dr != NULL && dr->dr_txg == tx->tx_txg) {
+ dr->dt.dl.dr_raw = B_TRUE;
+ dn->dn_objset->os_next_write_raw
+ [tx->tx_txg & TXG_MASK] = B_TRUE;
+ }
+ }
+
dmu_tx_commit(tx);
- length -= chunk_end - chunk_begin;
+ length -= chunk_len;
}
return (0);
}
err = dnode_hold(os, object, FTAG, &dn);
if (err != 0)
return (err);
- err = dmu_free_long_range_impl(os, dn, offset, length);
+ err = dmu_free_long_range_impl(os, dn, offset, length, B_FALSE);
/*
* It is important to zero out the maxblkid when freeing the entire
return (err);
}
+/*
+ * This function is equivalent to dmu_free_long_range(), but also
+ * marks the new dirty record as a raw write.
+ */
int
-dmu_free_long_object(objset_t *os, uint64_t object)
+dmu_free_long_range_raw(objset_t *os, uint64_t object,
+ uint64_t offset, uint64_t length)
+{
+ dnode_t *dn;
+ int err;
+
+ err = dnode_hold(os, object, FTAG, &dn);
+ if (err != 0)
+ return (err);
+ err = dmu_free_long_range_impl(os, dn, offset, length, B_TRUE);
+
+ /*
+ * It is important to zero out the maxblkid when freeing the entire
+ * file, so that (a) subsequent calls to dmu_free_long_range_impl()
+ * will take the fast path, and (b) dnode_reallocate() can verify
+ * that the entire file has been freed.
+ */
+ if (err == 0 && offset == 0 && length == DMU_OBJECT_END)
+ dn->dn_maxblkid = 0;
+
+ dnode_rele(dn, FTAG);
+ return (err);
+}
+
+static int
+dmu_free_long_object_impl(objset_t *os, uint64_t object, boolean_t raw)
{
dmu_tx_t *tx;
int err;
dmu_tx_mark_netfree(tx);
err = dmu_tx_assign(tx, TXG_WAIT);
if (err == 0) {
- err = dmu_object_free(os, object, tx);
+ if (raw)
+ err = dmu_object_dirty_raw(os, object, tx);
+ if (err == 0)
+ err = dmu_object_free(os, object, tx);
+
dmu_tx_commit(tx);
} else {
dmu_tx_abort(tx);
return (err);
}
+int
+dmu_free_long_object(objset_t *os, uint64_t object)
+{
+ return (dmu_free_long_object_impl(os, object, B_FALSE));
+}
+
+int
+dmu_free_long_object_raw(objset_t *os, uint64_t object)
+{
+ return (dmu_free_long_object_impl(os, object, B_TRUE));
+}
+
+
int
dmu_free_range(objset_t *os, uint64_t object, uint64_t offset,
uint64_t size, dmu_tx_t *tx)
if (err)
return (err);
ASSERT(offset < UINT64_MAX);
- ASSERT(size == -1ULL || size <= UINT64_MAX - offset);
+ ASSERT(size == DMU_OBJECT_END || size <= UINT64_MAX - offset);
dnode_free_range(dn, offset, size, tx);
dnode_rele(dn, FTAG);
return (0);
}
-int
-dmu_read(objset_t *os, uint64_t object, uint64_t offset, uint64_t size,
+static int
+dmu_read_impl(dnode_t *dn, uint64_t offset, uint64_t size,
void *buf, uint32_t flags)
{
- dnode_t *dn;
dmu_buf_t **dbp;
- int numbufs, err;
-
- err = dnode_hold(os, object, FTAG, &dn);
- if (err)
- return (err);
+ int numbufs, err = 0;
/*
* Deal with odd block sizes, where there can't be data past the first
}
dmu_buf_rele_array(dbp, numbufs, FTAG);
}
- dnode_rele(dn, FTAG);
return (err);
}
-void
-dmu_write(objset_t *os, uint64_t object, uint64_t offset, uint64_t size,
- const void *buf, dmu_tx_t *tx)
+int
+dmu_read(objset_t *os, uint64_t object, uint64_t offset, uint64_t size,
+ void *buf, uint32_t flags)
{
- dmu_buf_t **dbp;
- int numbufs, i;
+ dnode_t *dn;
+ int err;
- if (size == 0)
- return;
+ err = dnode_hold(os, object, FTAG, &dn);
+ if (err != 0)
+ return (err);
- VERIFY0(dmu_buf_hold_array(os, object, offset, size,
- FALSE, FTAG, &numbufs, &dbp));
+ err = dmu_read_impl(dn, offset, size, buf, flags);
+ dnode_rele(dn, FTAG);
+ return (err);
+}
+
+int
+dmu_read_by_dnode(dnode_t *dn, uint64_t offset, uint64_t size, void *buf,
+ uint32_t flags)
+{
+ return (dmu_read_impl(dn, offset, size, buf, flags));
+}
+
+static void
+dmu_write_impl(dmu_buf_t **dbp, int numbufs, uint64_t offset, uint64_t size,
+ const void *buf, dmu_tx_t *tx)
+{
+ int i;
for (i = 0; i < numbufs; i++) {
uint64_t tocpy;
size -= tocpy;
buf = (char *)buf + tocpy;
}
+}
+
+void
+dmu_write(objset_t *os, uint64_t object, uint64_t offset, uint64_t size,
+ const void *buf, dmu_tx_t *tx)
+{
+ dmu_buf_t **dbp;
+ int numbufs;
+
+ if (size == 0)
+ return;
+
+ VERIFY0(dmu_buf_hold_array(os, object, offset, size,
+ FALSE, FTAG, &numbufs, &dbp));
+ dmu_write_impl(dbp, numbufs, offset, size, buf, tx);
+ dmu_buf_rele_array(dbp, numbufs, FTAG);
+}
+
+void
+dmu_write_by_dnode(dnode_t *dn, uint64_t offset, uint64_t size,
+ const void *buf, dmu_tx_t *tx)
+{
+ dmu_buf_t **dbp;
+ int numbufs;
+
+ if (size == 0)
+ return;
+
+ VERIFY0(dmu_buf_hold_array_by_dnode(dn, offset, size,
+ FALSE, FTAG, &numbufs, &dbp, DMU_READ_PREFETCH));
+ dmu_write_impl(dbp, numbufs, offset, size, buf, tx);
dmu_buf_rele_array(dbp, numbufs, FTAG);
}
+static int
+dmu_object_remap_one_indirect(objset_t *os, dnode_t *dn,
+ uint64_t last_removal_txg, uint64_t offset)
+{
+ uint64_t l1blkid = dbuf_whichblock(dn, 1, offset);
+ int err = 0;
+
+ rw_enter(&dn->dn_struct_rwlock, RW_READER);
+ dmu_buf_impl_t *dbuf = dbuf_hold_level(dn, 1, l1blkid, FTAG);
+ ASSERT3P(dbuf, !=, NULL);
+
+ /*
+ * If the block hasn't been written yet, this default will ensure
+ * we don't try to remap it.
+ */
+ uint64_t birth = UINT64_MAX;
+ ASSERT3U(last_removal_txg, !=, UINT64_MAX);
+ if (dbuf->db_blkptr != NULL)
+ birth = dbuf->db_blkptr->blk_birth;
+ rw_exit(&dn->dn_struct_rwlock);
+
+ /*
+ * If this L1 was already written after the last removal, then we've
+ * already tried to remap it.
+ */
+ if (birth <= last_removal_txg &&
+ dbuf_read(dbuf, NULL, DB_RF_MUST_SUCCEED) == 0 &&
+ dbuf_can_remap(dbuf)) {
+ dmu_tx_t *tx = dmu_tx_create(os);
+ dmu_tx_hold_remap_l1indirect(tx, dn->dn_object);
+ err = dmu_tx_assign(tx, TXG_WAIT);
+ if (err == 0) {
+ (void) dbuf_dirty(dbuf, tx);
+ dmu_tx_commit(tx);
+ } else {
+ dmu_tx_abort(tx);
+ }
+ }
+
+ dbuf_rele(dbuf, FTAG);
+
+ delay(zfs_object_remap_one_indirect_delay_ticks);
+
+ return (err);
+}
+
+/*
+ * Remap all blockpointers in the object, if possible, so that they reference
+ * only concrete vdevs.
+ *
+ * To do this, iterate over the L0 blockpointers and remap any that reference
+ * an indirect vdev. Note that we only examine L0 blockpointers; since we
+ * cannot guarantee that we can remap all blockpointer anyways (due to split
+ * blocks), we do not want to make the code unnecessarily complicated to
+ * catch the unlikely case that there is an L1 block on an indirect vdev that
+ * contains no indirect blockpointers.
+ */
+int
+dmu_object_remap_indirects(objset_t *os, uint64_t object,
+ uint64_t last_removal_txg)
+{
+ uint64_t offset, l1span;
+ int err;
+ dnode_t *dn;
+
+ err = dnode_hold(os, object, FTAG, &dn);
+ if (err != 0) {
+ return (err);
+ }
+
+ if (dn->dn_nlevels <= 1) {
+ if (issig(JUSTLOOKING) && issig(FORREAL)) {
+ err = SET_ERROR(EINTR);
+ }
+
+ /*
+ * If the dnode has no indirect blocks, we cannot dirty them.
+ * We still want to remap the blkptr(s) in the dnode if
+ * appropriate, so mark it as dirty.
+ */
+ if (err == 0 && dnode_needs_remap(dn)) {
+ dmu_tx_t *tx = dmu_tx_create(os);
+ dmu_tx_hold_bonus(tx, dn->dn_object);
+ if ((err = dmu_tx_assign(tx, TXG_WAIT)) == 0) {
+ dnode_setdirty(dn, tx);
+ dmu_tx_commit(tx);
+ } else {
+ dmu_tx_abort(tx);
+ }
+ }
+
+ dnode_rele(dn, FTAG);
+ return (err);
+ }
+
+ offset = 0;
+ l1span = 1ULL << (dn->dn_indblkshift - SPA_BLKPTRSHIFT +
+ dn->dn_datablkshift);
+ /*
+ * Find the next L1 indirect that is not a hole.
+ */
+ while (dnode_next_offset(dn, 0, &offset, 2, 1, 0) == 0) {
+ if (issig(JUSTLOOKING) && issig(FORREAL)) {
+ err = SET_ERROR(EINTR);
+ break;
+ }
+ if ((err = dmu_object_remap_one_indirect(os, dn,
+ last_removal_txg, offset)) != 0) {
+ break;
+ }
+ offset += l1span;
+ }
+
+ dnode_rele(dn, FTAG);
+ return (err);
+}
+
void
dmu_prealloc(objset_t *os, uint64_t object, uint64_t offset, uint64_t size,
dmu_tx_t *tx)
atomic_add_64(&xuio_stats.stat.value.ui64, (val))
#define XUIOSTAT_BUMP(stat) XUIOSTAT_INCR(stat, 1)
+#ifdef HAVE_UIO_ZEROCOPY
int
dmu_xuio_init(xuio_t *xuio, int nblk)
{
ASSERT(i < priv->cnt);
priv->bufs[i] = NULL;
}
+#endif /* HAVE_UIO_ZEROCOPY */
static void
xuio_stat_init(void)
}
void
-xuio_stat_wbuf_copied()
+xuio_stat_wbuf_copied(void)
{
XUIOSTAT_BUMP(xuiostat_wbuf_copied);
}
void
-xuio_stat_wbuf_nocopy()
+xuio_stat_wbuf_nocopy(void)
{
XUIOSTAT_BUMP(xuiostat_wbuf_nocopy);
}
#ifdef _KERNEL
-static int
+int
dmu_read_uio_dnode(dnode_t *dn, uio_t *uio, uint64_t size)
{
dmu_buf_t **dbp;
int numbufs, i, err;
+#ifdef HAVE_UIO_ZEROCOPY
xuio_t *xuio = NULL;
+#endif
/*
* NB: we could do this block-at-a-time, but it's nice
bufoff = uio->uio_loffset - db->db_offset;
tocpy = MIN(db->db_size - bufoff, size);
+#ifdef HAVE_UIO_ZEROCOPY
if (xuio) {
dmu_buf_impl_t *dbi = (dmu_buf_impl_t *)db;
arc_buf_t *dbuf_abuf = dbi->db_buf;
XUIOSTAT_BUMP(xuiostat_rbuf_nocopy);
else
XUIOSTAT_BUMP(xuiostat_rbuf_copied);
- } else {
+ } else
+#endif
err = uiomove((char *)db->db_data + bufoff, tocpy,
UIO_READ, uio);
- }
if (err)
break;
return (err);
}
-static int
+int
dmu_write_uio_dnode(dnode_t *dn, uio_t *uio, uint64_t size, dmu_tx_t *tx)
{
dmu_buf_t **dbp;
arc_buf_destroy(buf, FTAG);
}
+int
+dmu_convert_mdn_block_to_raw(objset_t *os, uint64_t firstobj,
+ boolean_t byteorder, const uint8_t *salt, const uint8_t *iv,
+ const uint8_t *mac, dmu_tx_t *tx)
+{
+ int ret;
+ dmu_buf_t *handle = NULL;
+ dmu_buf_impl_t *db = NULL;
+ uint64_t offset = firstobj * DNODE_MIN_SIZE;
+ uint64_t dsobj = dmu_objset_id(os);
+
+ ret = dmu_buf_hold_by_dnode(DMU_META_DNODE(os), offset, FTAG, &handle,
+ DMU_READ_PREFETCH | DMU_READ_NO_DECRYPT);
+ if (ret != 0)
+ return (ret);
+
+ dmu_buf_will_change_crypt_params(handle, tx);
+
+ db = (dmu_buf_impl_t *)handle;
+ ASSERT3P(db->db_buf, !=, NULL);
+ ASSERT3U(dsobj, !=, 0);
+
+ /*
+ * This technically violates the assumption the dmu code makes
+ * that dnode blocks are only released in syncing context.
+ */
+ (void) arc_release(db->db_buf, db);
+ arc_convert_to_raw(db->db_buf, dsobj, byteorder, DMU_OT_DNODE,
+ salt, iv, mac);
+
+ dmu_buf_rele(handle, FTAG);
+
+ return (0);
+}
+
+void
+dmu_copy_from_buf(objset_t *os, uint64_t object, uint64_t offset,
+ dmu_buf_t *handle, dmu_tx_t *tx)
+{
+ dmu_buf_t *dst_handle;
+ dmu_buf_impl_t *dstdb;
+ dmu_buf_impl_t *srcdb = (dmu_buf_impl_t *)handle;
+ arc_buf_t *abuf;
+ uint64_t datalen;
+ boolean_t byteorder;
+ uint8_t salt[ZIO_DATA_SALT_LEN];
+ uint8_t iv[ZIO_DATA_IV_LEN];
+ uint8_t mac[ZIO_DATA_MAC_LEN];
+
+ ASSERT3P(srcdb->db_buf, !=, NULL);
+
+ /* hold the db that we want to write to */
+ VERIFY0(dmu_buf_hold(os, object, offset, FTAG, &dst_handle,
+ DMU_READ_NO_DECRYPT));
+ dstdb = (dmu_buf_impl_t *)dst_handle;
+ datalen = arc_buf_size(srcdb->db_buf);
+
+ /* allocated an arc buffer that matches the type of srcdb->db_buf */
+ if (arc_is_encrypted(srcdb->db_buf)) {
+ arc_get_raw_params(srcdb->db_buf, &byteorder, salt, iv, mac);
+ abuf = arc_loan_raw_buf(os->os_spa, dmu_objset_id(os),
+ byteorder, salt, iv, mac, DB_DNODE(dstdb)->dn_type,
+ datalen, arc_buf_lsize(srcdb->db_buf),
+ arc_get_compression(srcdb->db_buf));
+ } else {
+ /* we won't get a compressed db back from dmu_buf_hold() */
+ ASSERT3U(arc_get_compression(srcdb->db_buf),
+ ==, ZIO_COMPRESS_OFF);
+ abuf = arc_loan_buf(os->os_spa,
+ DMU_OT_IS_METADATA(DB_DNODE(dstdb)->dn_type), datalen);
+ }
+
+ ASSERT3U(datalen, ==, arc_buf_size(abuf));
+
+ /* copy the data to the new buffer and assign it to the dstdb */
+ bcopy(srcdb->db_buf->b_data, abuf->b_data, datalen);
+ dbuf_assign_arcbuf(dstdb, abuf, tx);
+ dmu_buf_rele(dst_handle, FTAG);
+}
+
/*
* When possible directly assign passed loaned arc buffer to a dbuf.
* If this is not possible copy the contents of passed arc buf via
* dmu_write().
*/
void
-dmu_assign_arcbuf(dmu_buf_t *handle, uint64_t offset, arc_buf_t *buf,
+dmu_assign_arcbuf_by_dnode(dnode_t *dn, uint64_t offset, arc_buf_t *buf,
dmu_tx_t *tx)
{
- dmu_buf_impl_t *dbuf = (dmu_buf_impl_t *)handle;
- dnode_t *dn;
dmu_buf_impl_t *db;
+ objset_t *os = dn->dn_objset;
+ uint64_t object = dn->dn_object;
uint32_t blksz = (uint32_t)arc_buf_lsize(buf);
uint64_t blkid;
- DB_DNODE_ENTER(dbuf);
- dn = DB_DNODE(dbuf);
rw_enter(&dn->dn_struct_rwlock, RW_READER);
blkid = dbuf_whichblock(dn, 0, offset);
VERIFY((db = dbuf_hold(dn, blkid, FTAG)) != NULL);
rw_exit(&dn->dn_struct_rwlock);
- DB_DNODE_EXIT(dbuf);
/*
* We can only assign if the offset is aligned, the arc buf is the
dbuf_assign_arcbuf(db, buf, tx);
dbuf_rele(db, FTAG);
} else {
- objset_t *os;
- uint64_t object;
-
/* compressed bufs must always be assignable to their dbuf */
ASSERT3U(arc_get_compression(buf), ==, ZIO_COMPRESS_OFF);
ASSERT(!(buf->b_flags & ARC_BUF_FLAG_COMPRESSED));
- DB_DNODE_ENTER(dbuf);
- dn = DB_DNODE(dbuf);
- os = dn->dn_objset;
- object = dn->dn_object;
- DB_DNODE_EXIT(dbuf);
-
dbuf_rele(db, FTAG);
dmu_write(os, object, offset, blksz, buf->b_data, tx);
dmu_return_arcbuf(buf);
}
}
+void
+dmu_assign_arcbuf_by_dbuf(dmu_buf_t *handle, uint64_t offset, arc_buf_t *buf,
+ dmu_tx_t *tx)
+{
+ dmu_buf_impl_t *dbuf = (dmu_buf_impl_t *)handle;
+
+ DB_DNODE_ENTER(dbuf);
+ dmu_assign_arcbuf_by_dnode(DB_DNODE(dbuf), offset, buf, tx);
+ DB_DNODE_EXIT(dbuf);
+}
+
typedef struct {
dbuf_dirty_record_t *dsa_dr;
dmu_sync_cb_t *dsa_done;
BP_SET_LSIZE(bp, db->db_size);
} else if (!BP_IS_EMBEDDED(bp)) {
ASSERT(BP_GET_LEVEL(bp) == 0);
- bp->blk_fill = 1;
+ BP_SET_FILL(bp, 1);
}
}
}
if (zio->io_error == 0) {
dr->dt.dl.dr_nopwrite = !!(zio->io_flags & ZIO_FLAG_NOPWRITE);
if (dr->dt.dl.dr_nopwrite) {
- ASSERTV(blkptr_t *bp = zio->io_bp);
- ASSERTV(blkptr_t *bp_orig = &zio->io_bp_orig);
- ASSERTV(uint8_t chksum = BP_GET_CHECKSUM(bp_orig));
+ blkptr_t *bp = zio->io_bp;
+ blkptr_t *bp_orig = &zio->io_bp_orig;
+ uint8_t chksum = BP_GET_CHECKSUM(bp_orig);
ASSERT(BP_EQUAL(bp, bp_orig));
+ VERIFY(BP_EQUAL(bp, db->db_blkptr));
ASSERT(zio->io_prop.zp_compress != ZIO_COMPRESS_OFF);
- ASSERT(zio_checksum_table[chksum].ci_flags &
+ VERIFY(zio_checksum_table[chksum].ci_flags &
ZCHECKSUM_FLAG_NOPWRITE);
}
dr->dt.dl.dr_overridden_by = *zio->io_bp;
ASSERTV(blkptr_t *bp_orig = &zio->io_bp_orig);
if (zio->io_error == 0 && !BP_IS_HOLE(bp)) {
- /*
- * If we didn't allocate a new block (i.e. ZIO_FLAG_NOPWRITE)
- * then there is nothing to do here. Otherwise, free the
- * newly allocated block in this txg.
- */
- if (zio->io_flags & ZIO_FLAG_NOPWRITE) {
- ASSERT(BP_EQUAL(bp, bp_orig));
- } else {
- ASSERT(BP_IS_HOLE(bp_orig) || !BP_EQUAL(bp, bp_orig));
- ASSERT(zio->io_bp->blk_birth == zio->io_txg);
- ASSERT(zio->io_txg > spa_syncing_txg(zio->io_spa));
- zio_free(zio->io_spa, zio->io_txg, zio->io_bp);
- }
+ ASSERT(!(zio->io_flags & ZIO_FLAG_NOPWRITE));
+ ASSERT(BP_IS_HOLE(bp_orig) || !BP_EQUAL(bp, bp_orig));
+ ASSERT(zio->io_bp->blk_birth == zio->io_txg);
+ ASSERT(zio->io_txg > spa_syncing_txg(zio->io_spa));
+ zio_free(zio->io_spa, zio->io_txg, zio->io_bp);
}
dmu_tx_commit(dsa->dsa_tx);
dsa->dsa_done(dsa->dsa_zgd, zio->io_error);
+ abd_put(zio->io_abd);
kmem_free(dsa, sizeof (*dsa));
}
return (SET_ERROR(EIO));
}
+ /*
+ * In order to prevent the zgd's lwb from being free'd prior to
+ * dmu_sync_late_arrival_done() being called, we have to ensure
+ * the lwb's "max txg" takes this tx's txg into account.
+ */
+ zil_lwb_add_txg(zgd->zgd_lwb, dmu_tx_get_txg(tx));
+
dsa = kmem_alloc(sizeof (dmu_sync_arg_t), KM_SLEEP);
dsa->dsa_dr = NULL;
dsa->dsa_done = done;
dsa->dsa_zgd = zgd;
dsa->dsa_tx = tx;
- zio_nowait(zio_write(pio, os->os_spa, dmu_tx_get_txg(tx),
- zgd->zgd_bp, zgd->zgd_db->db_data, zgd->zgd_db->db_size,
- zgd->zgd_db->db_size, zp, dmu_sync_late_arrival_ready, NULL,
- NULL, dmu_sync_late_arrival_done, dsa, ZIO_PRIORITY_SYNC_WRITE,
- ZIO_FLAG_CANFAIL, zb));
+ /*
+ * Since we are currently syncing this txg, it's nontrivial to
+ * determine what BP to nopwrite against, so we disable nopwrite.
+ *
+ * When syncing, the db_blkptr is initially the BP of the previous
+ * txg. We can not nopwrite against it because it will be changed
+ * (this is similar to the non-late-arrival case where the dbuf is
+ * dirty in a future txg).
+ *
+ * Then dbuf_write_ready() sets bp_blkptr to the location we will write.
+ * We can not nopwrite against it because although the BP will not
+ * (typically) be changed, the data has not yet been persisted to this
+ * location.
+ *
+ * Finally, when dbuf_write_done() is called, it is theoretically
+ * possible to always nopwrite, because the data that was written in
+ * this txg is the same data that we are trying to write. However we
+ * would need to check that this dbuf is not dirty in any future
+ * txg's (as we do in the normal dmu_sync() path). For simplicity, we
+ * don't nopwrite in this case.
+ */
+ zp->zp_nopwrite = B_FALSE;
+
+ zio_nowait(zio_write(pio, os->os_spa, dmu_tx_get_txg(tx), zgd->zgd_bp,
+ abd_get_from_buf(zgd->zgd_db->db_data, zgd->zgd_db->db_size),
+ zgd->zgd_db->db_size, zgd->zgd_db->db_size, zp,
+ dmu_sync_late_arrival_ready, NULL, NULL, dmu_sync_late_arrival_done,
+ dsa, ZIO_PRIORITY_SYNC_WRITE, ZIO_FLAG_CANFAIL, zb));
return (0);
}
int
dmu_sync(zio_t *pio, uint64_t txg, dmu_sync_cb_t *done, zgd_t *zgd)
{
- blkptr_t *bp = zgd->zgd_bp;
dmu_buf_impl_t *db = (dmu_buf_impl_t *)zgd->zgd_db;
objset_t *os = db->db_objset;
dsl_dataset_t *ds = os->os_dsl_dataset;
ASSERT(pio != NULL);
ASSERT(txg != 0);
+ /* dbuf is within the locked range */
+ ASSERT3U(db->db.db_offset, >=, zgd->zgd_rl->r_off);
+ ASSERT3U(db->db.db_offset + db->db.db_size, <=,
+ zgd->zgd_rl->r_off + zgd->zgd_rl->r_len);
+
SET_BOOKMARK(&zb, ds->ds_object,
db->db.db_object, db->db_level, db->db_blkid);
DB_DNODE_ENTER(db);
dn = DB_DNODE(db);
- dmu_write_policy(os, dn, db->db_level, WP_DMU_SYNC,
- ZIO_COMPRESS_INHERIT, &zp);
+ dmu_write_policy(os, dn, db->db_level, WP_DMU_SYNC, &zp);
DB_DNODE_EXIT(db);
/*
ASSERT(dr->dr_next == NULL || dr->dr_next->dr_txg < txg);
+ if (db->db_blkptr != NULL) {
+ /*
+ * We need to fill in zgd_bp with the current blkptr so that
+ * the nopwrite code can check if we're writing the same
+ * data that's already on disk. We can only nopwrite if we
+ * are sure that after making the copy, db_blkptr will not
+ * change until our i/o completes. We ensure this by
+ * holding the db_mtx, and only allowing nopwrite if the
+ * block is not already dirty (see below). This is verified
+ * by dmu_sync_done(), which VERIFYs that the db_blkptr has
+ * not changed.
+ */
+ *zgd->zgd_bp = *db->db_blkptr;
+ }
+
/*
* Assume the on-disk data is X, the current syncing data (in
* txg - 1) is Y, and the current in-memory data is Z (currently
dsa->dsa_tx = NULL;
zio_nowait(arc_write(pio, os->os_spa, txg,
- bp, dr->dt.dl.dr_data, DBUF_IS_L2CACHEABLE(db),
+ zgd->zgd_bp, dr->dt.dl.dr_data, DBUF_IS_L2CACHEABLE(db),
&zp, dmu_sync_ready, NULL, NULL, dmu_sync_done, dsa,
ZIO_PRIORITY_SYNC_WRITE, ZIO_FLAG_CANFAIL, &zb));
return (0);
}
+int
+dmu_object_set_nlevels(objset_t *os, uint64_t object, int nlevels, dmu_tx_t *tx)
+{
+ dnode_t *dn;
+ int err;
+
+ err = dnode_hold(os, object, FTAG, &dn);
+ if (err)
+ return (err);
+ err = dnode_set_nlevels(dn, nlevels, tx);
+ dnode_rele(dn, FTAG);
+ return (err);
+}
+
int
dmu_object_set_blocksize(objset_t *os, uint64_t object, uint64_t size, int ibs,
- dmu_tx_t *tx)
+ dmu_tx_t *tx)
{
dnode_t *dn;
int err;
return (err);
}
+int
+dmu_object_set_maxblkid(objset_t *os, uint64_t object, uint64_t maxblkid,
+ dmu_tx_t *tx)
+{
+ dnode_t *dn;
+ int err;
+
+ err = dnode_hold(os, object, FTAG, &dn);
+ if (err)
+ return (err);
+ rw_enter(&dn->dn_struct_rwlock, RW_WRITER);
+ dnode_new_blkid(dn, maxblkid, tx, B_FALSE);
+ rw_exit(&dn->dn_struct_rwlock);
+ dnode_rele(dn, FTAG);
+ return (0);
+}
+
void
dmu_object_set_checksum(objset_t *os, uint64_t object, uint8_t checksum,
- dmu_tx_t *tx)
+ dmu_tx_t *tx)
{
dnode_t *dn;
void
dmu_object_set_compress(objset_t *os, uint64_t object, uint8_t compress,
- dmu_tx_t *tx)
+ dmu_tx_t *tx)
{
dnode_t *dn;
dnode_rele(dn, FTAG);
}
-int zfs_mdcomp_disable = 0;
+/*
+ * Dirty an object and set the dirty record's raw flag. This is used
+ * when writing raw data to an object that will not effect the
+ * encryption parameters, specifically during raw receives.
+ */
+int
+dmu_object_dirty_raw(objset_t *os, uint64_t object, dmu_tx_t *tx)
+{
+ dnode_t *dn;
+ int err;
+
+ err = dnode_hold(os, object, FTAG, &dn);
+ if (err)
+ return (err);
+ dmu_buf_will_change_crypt_params((dmu_buf_t *)dn->dn_dbuf, tx);
+ dnode_rele(dn, FTAG);
+ return (err);
+}
/*
* When the "redundant_metadata" property is set to "most", only indirect
int zfs_redundant_metadata_most_ditto_level = 2;
void
-dmu_write_policy(objset_t *os, dnode_t *dn, int level, int wp,
- enum zio_compress override_compress, zio_prop_t *zp)
+dmu_write_policy(objset_t *os, dnode_t *dn, int level, int wp, zio_prop_t *zp)
{
dmu_object_type_t type = dn ? dn->dn_type : DMU_OT_OBJSET;
boolean_t ismd = (level > 0 || DMU_OT_IS_METADATA(type) ||
boolean_t dedup = B_FALSE;
boolean_t nopwrite = B_FALSE;
boolean_t dedup_verify = os->os_dedup_verify;
+ boolean_t encrypt = B_FALSE;
int copies = os->os_copies;
/*
* 3. all other level 0 blocks
*/
if (ismd) {
- if (zfs_mdcomp_disable) {
- compress = ZIO_COMPRESS_EMPTY;
- } else {
- /*
- * XXX -- we should design a compression algorithm
- * that specializes in arrays of bps.
- */
- compress = zio_compress_select(os->os_spa,
- ZIO_COMPRESS_ON, ZIO_COMPRESS_ON);
- }
+ /*
+ * XXX -- we should design a compression algorithm
+ * that specializes in arrays of bps.
+ */
+ compress = zio_compress_select(os->os_spa,
+ ZIO_COMPRESS_ON, ZIO_COMPRESS_ON);
/*
* Metadata always gets checksummed. If the data
compress != ZIO_COMPRESS_OFF && zfs_nopwrite_enabled);
}
- zp->zp_checksum = checksum;
-
/*
- * If we're writing a pre-compressed buffer, the compression type we use
- * must match the data. If it hasn't been compressed yet, then we should
- * use the value dictated by the policies above.
+ * All objects in an encrypted objset are protected from modification
+ * via a MAC. Encrypted objects store their IV and salt in the last DVA
+ * in the bp, so we cannot use all copies. Encrypted objects are also
+ * not subject to nopwrite since writing the same data will still
+ * result in a new ciphertext. Only encrypted blocks can be dedup'd
+ * to avoid ambiguity in the dedup code since the DDT does not store
+ * object types.
*/
- zp->zp_compress = override_compress != ZIO_COMPRESS_INHERIT
- ? override_compress : compress;
- ASSERT3U(zp->zp_compress, !=, ZIO_COMPRESS_INHERIT);
+ if (os->os_encrypted && (wp & WP_NOFILL) == 0) {
+ encrypt = B_TRUE;
+
+ if (DMU_OT_IS_ENCRYPTED(type)) {
+ copies = MIN(copies, SPA_DVAS_PER_BP - 1);
+ nopwrite = B_FALSE;
+ } else {
+ dedup = B_FALSE;
+ }
+
+ if (level <= 0 &&
+ (type == DMU_OT_DNODE || type == DMU_OT_OBJSET)) {
+ compress = ZIO_COMPRESS_EMPTY;
+ }
+ }
+ zp->zp_compress = compress;
+ zp->zp_checksum = checksum;
zp->zp_type = (wp & WP_SPILL) ? dn->dn_bonustype : type;
zp->zp_level = level;
zp->zp_copies = MIN(copies, spa_max_replication(os->os_spa));
zp->zp_dedup = dedup;
zp->zp_dedup_verify = dedup && dedup_verify;
zp->zp_nopwrite = nopwrite;
+ zp->zp_encrypt = encrypt;
+ zp->zp_byteorder = ZFS_HOST_BYTEORDER;
+ bzero(zp->zp_salt, ZIO_DATA_SALT_LEN);
+ bzero(zp->zp_iv, ZIO_DATA_IV_LEN);
+ bzero(zp->zp_mac, ZIO_DATA_MAC_LEN);
+
+ ASSERT3U(zp->zp_compress, !=, ZIO_COMPRESS_INHERIT);
}
+/*
+ * This function is only called from zfs_holey_common() for zpl_llseek()
+ * in order to determine the location of holes. In order to accurately
+ * report holes all dirty data must be synced to disk. This causes extremely
+ * poor performance when seeking for holes in a dirty file. As a compromise,
+ * only provide hole data when the dnode is clean. When a dnode is dirty
+ * report the dnode as having no holes which is always a safe thing to do.
+ */
int
dmu_offset_next(objset_t *os, uint64_t object, boolean_t hole, uint64_t *off)
{
dnode_t *dn;
int i, err;
+ boolean_t clean = B_TRUE;
err = dnode_hold(os, object, FTAG, &dn);
if (err)
return (err);
+
/*
- * Sync any current changes before
- * we go trundling through the block pointers.
+ * Check if dnode is dirty
*/
for (i = 0; i < TXG_SIZE; i++) {
- if (list_link_active(&dn->dn_dirty_link[i]))
+ if (multilist_link_active(&dn->dn_dirty_link[i])) {
+ clean = B_FALSE;
break;
+ }
}
- if (i != TXG_SIZE) {
+
+ /*
+ * If compatibility option is on, sync any current changes before
+ * we go trundling through the block pointers.
+ */
+ if (!clean && zfs_dmu_offset_next_sync) {
+ clean = B_TRUE;
dnode_rele(dn, FTAG);
txg_wait_synced(dmu_objset_pool(os), 0);
err = dnode_hold(os, object, FTAG, &dn);
return (err);
}
- err = dnode_next_offset(dn, (hole ? DNODE_FIND_HOLE : 0), off, 1, 1, 0);
+ if (clean)
+ err = dnode_next_offset(dn,
+ (hole ? DNODE_FIND_HOLE : 0), off, 1, 1, 0);
+ else
+ err = SET_ERROR(EBUSY);
+
dnode_rele(dn, FTAG);
return (err);
__dmu_object_info_from_dnode(dnode_t *dn, dmu_object_info_t *doi)
{
dnode_phys_t *dnp = dn->dn_phys;
- int i;
doi->doi_data_block_size = dn->dn_datablksz;
doi->doi_metadata_block_size = dn->dn_indblkshift ?
doi->doi_physical_blocks_512 = (DN_USED_BYTES(dnp) + 256) >> 9;
doi->doi_max_offset = (dn->dn_maxblkid + 1) * dn->dn_datablksz;
doi->doi_fill_count = 0;
- for (i = 0; i < dnp->dn_nblkptr; i++)
+ for (int i = 0; i < dnp->dn_nblkptr; i++)
doi->doi_fill_count += BP_GET_FILL(&dnp->dn_blkptr[i]);
}
void
dmu_init(void)
{
+ abd_init();
zfs_dbgmsg_init();
sa_cache_init();
xuio_stat_init();
xuio_stat_fini();
sa_cache_fini();
zfs_dbgmsg_fini();
+ abd_fini();
}
#if defined(_KERNEL) && defined(HAVE_SPL)
EXPORT_SYMBOL(dmu_prefetch);
EXPORT_SYMBOL(dmu_free_range);
EXPORT_SYMBOL(dmu_free_long_range);
+EXPORT_SYMBOL(dmu_free_long_range_raw);
EXPORT_SYMBOL(dmu_free_long_object);
+EXPORT_SYMBOL(dmu_free_long_object_raw);
EXPORT_SYMBOL(dmu_read);
+EXPORT_SYMBOL(dmu_read_by_dnode);
EXPORT_SYMBOL(dmu_write);
+EXPORT_SYMBOL(dmu_write_by_dnode);
EXPORT_SYMBOL(dmu_prealloc);
EXPORT_SYMBOL(dmu_object_info);
EXPORT_SYMBOL(dmu_object_info_from_dnode);
EXPORT_SYMBOL(dmu_object_info_from_db);
EXPORT_SYMBOL(dmu_object_size_from_db);
EXPORT_SYMBOL(dmu_object_dnsize_from_db);
+EXPORT_SYMBOL(dmu_object_set_nlevels);
EXPORT_SYMBOL(dmu_object_set_blocksize);
+EXPORT_SYMBOL(dmu_object_set_maxblkid);
EXPORT_SYMBOL(dmu_object_set_checksum);
EXPORT_SYMBOL(dmu_object_set_compress);
EXPORT_SYMBOL(dmu_write_policy);
EXPORT_SYMBOL(dmu_sync);
EXPORT_SYMBOL(dmu_request_arcbuf);
EXPORT_SYMBOL(dmu_return_arcbuf);
-EXPORT_SYMBOL(dmu_assign_arcbuf);
+EXPORT_SYMBOL(dmu_assign_arcbuf_by_dnode);
+EXPORT_SYMBOL(dmu_assign_arcbuf_by_dbuf);
EXPORT_SYMBOL(dmu_buf_hold);
EXPORT_SYMBOL(dmu_ot);
-module_param(zfs_mdcomp_disable, int, 0644);
-MODULE_PARM_DESC(zfs_mdcomp_disable, "Disable meta data compression");
-
+/* BEGIN CSTYLED */
module_param(zfs_nopwrite_enabled, int, 0644);
MODULE_PARM_DESC(zfs_nopwrite_enabled, "Enable NOP writes");
+module_param(zfs_per_txg_dirty_frees_percent, ulong, 0644);
+MODULE_PARM_DESC(zfs_per_txg_dirty_frees_percent,
+ "percentage of dirtied blocks from frees in one TXG");
+
+module_param(zfs_dmu_offset_next_sync, int, 0644);
+MODULE_PARM_DESC(zfs_dmu_offset_next_sync,
+ "Enable forcing txg sync to find holes");
+
+/* END CSTYLED */
+
#endif