4 * The contents of this file are subject to the terms of the
5 * Common Development and Distribution License (the "License").
6 * You may not use this file except in compliance with the License.
8 * You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE
9 * or http://www.opensolaris.org/os/licensing.
10 * See the License for the specific language governing permissions
11 * and limitations under the License.
13 * When distributing Covered Code, include this CDDL HEADER in each
14 * file and include the License file at usr/src/OPENSOLARIS.LICENSE.
15 * If applicable, add the following below this CDDL HEADER, with the
16 * fields enclosed by brackets "[]" replaced with your own identifying
17 * information: Portions Copyright [yyyy] [name of copyright owner]
22 * Copyright (c) 2005, 2010, Oracle and/or its affiliates. All rights reserved.
23 * Copyright (c) 2011, 2014 by Delphix. All rights reserved.
26 /* Portions Copyright 2010 Robert Milkowski */
28 #include <sys/zfs_context.h>
34 #include <sys/resource.h>
36 #include <sys/zil_impl.h>
37 #include <sys/dsl_dataset.h>
38 #include <sys/vdev_impl.h>
39 #include <sys/dmu_tx.h>
40 #include <sys/dsl_pool.h>
41 #include <sys/metaslab.h>
42 #include <sys/trace_zil.h>
45 * The zfs intent log (ZIL) saves transaction records of system calls
46 * that change the file system in memory with enough information
47 * to be able to replay them. These are stored in memory until
48 * either the DMU transaction group (txg) commits them to the stable pool
49 * and they can be discarded, or they are flushed to the stable log
50 * (also in the pool) due to a fsync, O_DSYNC or other synchronous
51 * requirement. In the event of a panic or power fail then those log
52 * records (transactions) are replayed.
54 * There is one ZIL per file system. Its on-disk (pool) format consists
61 * A log record holds a system call transaction. Log blocks can
62 * hold many log records and the blocks are chained together.
63 * Each ZIL block contains a block pointer (blkptr_t) to the next
64 * ZIL block in the chain. The ZIL header points to the first
65 * block in the chain. Note there is not a fixed place in the pool
66 * to hold blocks. They are dynamically allocated and freed as
67 * needed from the blocks available. Figure X shows the ZIL structure:
71 * See zil.h for more information about these fields.
73 zil_stats_t zil_stats = {
74 { "zil_commit_count", KSTAT_DATA_UINT64 },
75 { "zil_commit_writer_count", KSTAT_DATA_UINT64 },
76 { "zil_itx_count", KSTAT_DATA_UINT64 },
77 { "zil_itx_indirect_count", KSTAT_DATA_UINT64 },
78 { "zil_itx_indirect_bytes", KSTAT_DATA_UINT64 },
79 { "zil_itx_copied_count", KSTAT_DATA_UINT64 },
80 { "zil_itx_copied_bytes", KSTAT_DATA_UINT64 },
81 { "zil_itx_needcopy_count", KSTAT_DATA_UINT64 },
82 { "zil_itx_needcopy_bytes", KSTAT_DATA_UINT64 },
83 { "zil_itx_metaslab_normal_count", KSTAT_DATA_UINT64 },
84 { "zil_itx_metaslab_normal_bytes", KSTAT_DATA_UINT64 },
85 { "zil_itx_metaslab_slog_count", KSTAT_DATA_UINT64 },
86 { "zil_itx_metaslab_slog_bytes", KSTAT_DATA_UINT64 },
89 static kstat_t *zil_ksp;
92 * Disable intent logging replay. This global ZIL switch affects all pools.
94 int zil_replay_disable = 0;
97 * Tunable parameter for debugging or performance analysis. Setting
98 * zfs_nocacheflush will cause corruption on power loss if a volatile
99 * out-of-order write cache is enabled.
101 int zfs_nocacheflush = 0;
103 static kmem_cache_t *zil_lwb_cache;
105 static void zil_async_to_sync(zilog_t *zilog, uint64_t foid);
107 #define LWB_EMPTY(lwb) ((BP_GET_LSIZE(&lwb->lwb_blk) - \
108 sizeof (zil_chain_t)) == (lwb->lwb_sz - lwb->lwb_nused))
112 * ziltest is by and large an ugly hack, but very useful in
113 * checking replay without tedious work.
114 * When running ziltest we want to keep all itx's and so maintain
115 * a single list in the zl_itxg[] that uses a high txg: ZILTEST_TXG
116 * We subtract TXG_CONCURRENT_STATES to allow for common code.
118 #define ZILTEST_TXG (UINT64_MAX - TXG_CONCURRENT_STATES)
121 zil_bp_compare(const void *x1, const void *x2)
123 const dva_t *dva1 = &((zil_bp_node_t *)x1)->zn_dva;
124 const dva_t *dva2 = &((zil_bp_node_t *)x2)->zn_dva;
126 if (DVA_GET_VDEV(dva1) < DVA_GET_VDEV(dva2))
128 if (DVA_GET_VDEV(dva1) > DVA_GET_VDEV(dva2))
131 if (DVA_GET_OFFSET(dva1) < DVA_GET_OFFSET(dva2))
133 if (DVA_GET_OFFSET(dva1) > DVA_GET_OFFSET(dva2))
140 zil_bp_tree_init(zilog_t *zilog)
142 avl_create(&zilog->zl_bp_tree, zil_bp_compare,
143 sizeof (zil_bp_node_t), offsetof(zil_bp_node_t, zn_node));
147 zil_bp_tree_fini(zilog_t *zilog)
149 avl_tree_t *t = &zilog->zl_bp_tree;
153 while ((zn = avl_destroy_nodes(t, &cookie)) != NULL)
154 kmem_free(zn, sizeof (zil_bp_node_t));
160 zil_bp_tree_add(zilog_t *zilog, const blkptr_t *bp)
162 avl_tree_t *t = &zilog->zl_bp_tree;
167 if (BP_IS_EMBEDDED(bp))
170 dva = BP_IDENTITY(bp);
172 if (avl_find(t, dva, &where) != NULL)
173 return (SET_ERROR(EEXIST));
175 zn = kmem_alloc(sizeof (zil_bp_node_t), KM_SLEEP);
177 avl_insert(t, zn, where);
182 static zil_header_t *
183 zil_header_in_syncing_context(zilog_t *zilog)
185 return ((zil_header_t *)zilog->zl_header);
189 zil_init_log_chain(zilog_t *zilog, blkptr_t *bp)
191 zio_cksum_t *zc = &bp->blk_cksum;
193 zc->zc_word[ZIL_ZC_GUID_0] = spa_get_random(-1ULL);
194 zc->zc_word[ZIL_ZC_GUID_1] = spa_get_random(-1ULL);
195 zc->zc_word[ZIL_ZC_OBJSET] = dmu_objset_id(zilog->zl_os);
196 zc->zc_word[ZIL_ZC_SEQ] = 1ULL;
200 * Read a log block and make sure it's valid.
203 zil_read_log_block(zilog_t *zilog, const blkptr_t *bp, blkptr_t *nbp, void *dst,
206 enum zio_flag zio_flags = ZIO_FLAG_CANFAIL;
207 arc_flags_t aflags = ARC_FLAG_WAIT;
208 arc_buf_t *abuf = NULL;
212 if (zilog->zl_header->zh_claim_txg == 0)
213 zio_flags |= ZIO_FLAG_SPECULATIVE | ZIO_FLAG_SCRUB;
215 if (!(zilog->zl_header->zh_flags & ZIL_CLAIM_LR_SEQ_VALID))
216 zio_flags |= ZIO_FLAG_SPECULATIVE;
218 SET_BOOKMARK(&zb, bp->blk_cksum.zc_word[ZIL_ZC_OBJSET],
219 ZB_ZIL_OBJECT, ZB_ZIL_LEVEL, bp->blk_cksum.zc_word[ZIL_ZC_SEQ]);
221 error = arc_read(NULL, zilog->zl_spa, bp, arc_getbuf_func, &abuf,
222 ZIO_PRIORITY_SYNC_READ, zio_flags, &aflags, &zb);
225 zio_cksum_t cksum = bp->blk_cksum;
228 * Validate the checksummed log block.
230 * Sequence numbers should be... sequential. The checksum
231 * verifier for the next block should be bp's checksum plus 1.
233 * Also check the log chain linkage and size used.
235 cksum.zc_word[ZIL_ZC_SEQ]++;
237 if (BP_GET_CHECKSUM(bp) == ZIO_CHECKSUM_ZILOG2) {
238 zil_chain_t *zilc = abuf->b_data;
239 char *lr = (char *)(zilc + 1);
240 uint64_t len = zilc->zc_nused - sizeof (zil_chain_t);
242 if (bcmp(&cksum, &zilc->zc_next_blk.blk_cksum,
243 sizeof (cksum)) || BP_IS_HOLE(&zilc->zc_next_blk)) {
244 error = SET_ERROR(ECKSUM);
246 ASSERT3U(len, <=, SPA_OLD_MAXBLOCKSIZE);
248 *end = (char *)dst + len;
249 *nbp = zilc->zc_next_blk;
252 char *lr = abuf->b_data;
253 uint64_t size = BP_GET_LSIZE(bp);
254 zil_chain_t *zilc = (zil_chain_t *)(lr + size) - 1;
256 if (bcmp(&cksum, &zilc->zc_next_blk.blk_cksum,
257 sizeof (cksum)) || BP_IS_HOLE(&zilc->zc_next_blk) ||
258 (zilc->zc_nused > (size - sizeof (*zilc)))) {
259 error = SET_ERROR(ECKSUM);
261 ASSERT3U(zilc->zc_nused, <=,
262 SPA_OLD_MAXBLOCKSIZE);
263 bcopy(lr, dst, zilc->zc_nused);
264 *end = (char *)dst + zilc->zc_nused;
265 *nbp = zilc->zc_next_blk;
269 VERIFY(arc_buf_remove_ref(abuf, &abuf));
276 * Read a TX_WRITE log data block.
279 zil_read_log_data(zilog_t *zilog, const lr_write_t *lr, void *wbuf)
281 enum zio_flag zio_flags = ZIO_FLAG_CANFAIL;
282 const blkptr_t *bp = &lr->lr_blkptr;
283 arc_flags_t aflags = ARC_FLAG_WAIT;
284 arc_buf_t *abuf = NULL;
288 if (BP_IS_HOLE(bp)) {
290 bzero(wbuf, MAX(BP_GET_LSIZE(bp), lr->lr_length));
294 if (zilog->zl_header->zh_claim_txg == 0)
295 zio_flags |= ZIO_FLAG_SPECULATIVE | ZIO_FLAG_SCRUB;
297 SET_BOOKMARK(&zb, dmu_objset_id(zilog->zl_os), lr->lr_foid,
298 ZB_ZIL_LEVEL, lr->lr_offset / BP_GET_LSIZE(bp));
300 error = arc_read(NULL, zilog->zl_spa, bp, arc_getbuf_func, &abuf,
301 ZIO_PRIORITY_SYNC_READ, zio_flags, &aflags, &zb);
305 bcopy(abuf->b_data, wbuf, arc_buf_size(abuf));
306 (void) arc_buf_remove_ref(abuf, &abuf);
313 * Parse the intent log, and call parse_func for each valid record within.
316 zil_parse(zilog_t *zilog, zil_parse_blk_func_t *parse_blk_func,
317 zil_parse_lr_func_t *parse_lr_func, void *arg, uint64_t txg)
319 const zil_header_t *zh = zilog->zl_header;
320 boolean_t claimed = !!zh->zh_claim_txg;
321 uint64_t claim_blk_seq = claimed ? zh->zh_claim_blk_seq : UINT64_MAX;
322 uint64_t claim_lr_seq = claimed ? zh->zh_claim_lr_seq : UINT64_MAX;
323 uint64_t max_blk_seq = 0;
324 uint64_t max_lr_seq = 0;
325 uint64_t blk_count = 0;
326 uint64_t lr_count = 0;
327 blkptr_t blk, next_blk;
331 bzero(&next_blk, sizeof (blkptr_t));
334 * Old logs didn't record the maximum zh_claim_lr_seq.
336 if (!(zh->zh_flags & ZIL_CLAIM_LR_SEQ_VALID))
337 claim_lr_seq = UINT64_MAX;
340 * Starting at the block pointed to by zh_log we read the log chain.
341 * For each block in the chain we strongly check that block to
342 * ensure its validity. We stop when an invalid block is found.
343 * For each block pointer in the chain we call parse_blk_func().
344 * For each record in each valid block we call parse_lr_func().
345 * If the log has been claimed, stop if we encounter a sequence
346 * number greater than the highest claimed sequence number.
348 lrbuf = zio_buf_alloc(SPA_OLD_MAXBLOCKSIZE);
349 zil_bp_tree_init(zilog);
351 for (blk = zh->zh_log; !BP_IS_HOLE(&blk); blk = next_blk) {
352 uint64_t blk_seq = blk.blk_cksum.zc_word[ZIL_ZC_SEQ];
356 if (blk_seq > claim_blk_seq)
358 if ((error = parse_blk_func(zilog, &blk, arg, txg)) != 0)
360 ASSERT3U(max_blk_seq, <, blk_seq);
361 max_blk_seq = blk_seq;
364 if (max_lr_seq == claim_lr_seq && max_blk_seq == claim_blk_seq)
367 error = zil_read_log_block(zilog, &blk, &next_blk, lrbuf, &end);
371 for (lrp = lrbuf; lrp < end; lrp += reclen) {
372 lr_t *lr = (lr_t *)lrp;
373 reclen = lr->lrc_reclen;
374 ASSERT3U(reclen, >=, sizeof (lr_t));
375 if (lr->lrc_seq > claim_lr_seq)
377 if ((error = parse_lr_func(zilog, lr, arg, txg)) != 0)
379 ASSERT3U(max_lr_seq, <, lr->lrc_seq);
380 max_lr_seq = lr->lrc_seq;
385 zilog->zl_parse_error = error;
386 zilog->zl_parse_blk_seq = max_blk_seq;
387 zilog->zl_parse_lr_seq = max_lr_seq;
388 zilog->zl_parse_blk_count = blk_count;
389 zilog->zl_parse_lr_count = lr_count;
391 ASSERT(!claimed || !(zh->zh_flags & ZIL_CLAIM_LR_SEQ_VALID) ||
392 (max_blk_seq == claim_blk_seq && max_lr_seq == claim_lr_seq));
394 zil_bp_tree_fini(zilog);
395 zio_buf_free(lrbuf, SPA_OLD_MAXBLOCKSIZE);
401 zil_claim_log_block(zilog_t *zilog, blkptr_t *bp, void *tx, uint64_t first_txg)
404 * Claim log block if not already committed and not already claimed.
405 * If tx == NULL, just verify that the block is claimable.
407 if (BP_IS_HOLE(bp) || bp->blk_birth < first_txg ||
408 zil_bp_tree_add(zilog, bp) != 0)
411 return (zio_wait(zio_claim(NULL, zilog->zl_spa,
412 tx == NULL ? 0 : first_txg, bp, spa_claim_notify, NULL,
413 ZIO_FLAG_CANFAIL | ZIO_FLAG_SPECULATIVE | ZIO_FLAG_SCRUB)));
417 zil_claim_log_record(zilog_t *zilog, lr_t *lrc, void *tx, uint64_t first_txg)
419 lr_write_t *lr = (lr_write_t *)lrc;
422 if (lrc->lrc_txtype != TX_WRITE)
426 * If the block is not readable, don't claim it. This can happen
427 * in normal operation when a log block is written to disk before
428 * some of the dmu_sync() blocks it points to. In this case, the
429 * transaction cannot have been committed to anyone (we would have
430 * waited for all writes to be stable first), so it is semantically
431 * correct to declare this the end of the log.
433 if (lr->lr_blkptr.blk_birth >= first_txg &&
434 (error = zil_read_log_data(zilog, lr, NULL)) != 0)
436 return (zil_claim_log_block(zilog, &lr->lr_blkptr, tx, first_txg));
441 zil_free_log_block(zilog_t *zilog, blkptr_t *bp, void *tx, uint64_t claim_txg)
443 zio_free_zil(zilog->zl_spa, dmu_tx_get_txg(tx), bp);
449 zil_free_log_record(zilog_t *zilog, lr_t *lrc, void *tx, uint64_t claim_txg)
451 lr_write_t *lr = (lr_write_t *)lrc;
452 blkptr_t *bp = &lr->lr_blkptr;
455 * If we previously claimed it, we need to free it.
457 if (claim_txg != 0 && lrc->lrc_txtype == TX_WRITE &&
458 bp->blk_birth >= claim_txg && zil_bp_tree_add(zilog, bp) == 0 &&
460 zio_free(zilog->zl_spa, dmu_tx_get_txg(tx), bp);
466 zil_alloc_lwb(zilog_t *zilog, blkptr_t *bp, uint64_t txg, boolean_t fastwrite)
470 lwb = kmem_cache_alloc(zil_lwb_cache, KM_SLEEP);
471 lwb->lwb_zilog = zilog;
473 lwb->lwb_fastwrite = fastwrite;
474 lwb->lwb_buf = zio_buf_alloc(BP_GET_LSIZE(bp));
475 lwb->lwb_max_txg = txg;
478 if (BP_GET_CHECKSUM(bp) == ZIO_CHECKSUM_ZILOG2) {
479 lwb->lwb_nused = sizeof (zil_chain_t);
480 lwb->lwb_sz = BP_GET_LSIZE(bp);
483 lwb->lwb_sz = BP_GET_LSIZE(bp) - sizeof (zil_chain_t);
486 mutex_enter(&zilog->zl_lock);
487 list_insert_tail(&zilog->zl_lwb_list, lwb);
488 mutex_exit(&zilog->zl_lock);
494 * Called when we create in-memory log transactions so that we know
495 * to cleanup the itxs at the end of spa_sync().
498 zilog_dirty(zilog_t *zilog, uint64_t txg)
500 dsl_pool_t *dp = zilog->zl_dmu_pool;
501 dsl_dataset_t *ds = dmu_objset_ds(zilog->zl_os);
503 if (ds->ds_is_snapshot)
504 panic("dirtying snapshot!");
506 if (txg_list_add(&dp->dp_dirty_zilogs, zilog, txg)) {
507 /* up the hold count until we can be written out */
508 dmu_buf_add_ref(ds->ds_dbuf, zilog);
513 zilog_is_dirty(zilog_t *zilog)
515 dsl_pool_t *dp = zilog->zl_dmu_pool;
518 for (t = 0; t < TXG_SIZE; t++) {
519 if (txg_list_member(&dp->dp_dirty_zilogs, zilog, t))
526 * Create an on-disk intent log.
529 zil_create(zilog_t *zilog)
531 const zil_header_t *zh = zilog->zl_header;
537 boolean_t fastwrite = FALSE;
540 * Wait for any previous destroy to complete.
542 txg_wait_synced(zilog->zl_dmu_pool, zilog->zl_destroy_txg);
544 ASSERT(zh->zh_claim_txg == 0);
545 ASSERT(zh->zh_replay_seq == 0);
550 * Allocate an initial log block if:
551 * - there isn't one already
552 * - the existing block is the wrong endianess
554 if (BP_IS_HOLE(&blk) || BP_SHOULD_BYTESWAP(&blk)) {
555 tx = dmu_tx_create(zilog->zl_os);
556 VERIFY(dmu_tx_assign(tx, TXG_WAIT) == 0);
557 dsl_dataset_dirty(dmu_objset_ds(zilog->zl_os), tx);
558 txg = dmu_tx_get_txg(tx);
560 if (!BP_IS_HOLE(&blk)) {
561 zio_free_zil(zilog->zl_spa, txg, &blk);
565 error = zio_alloc_zil(zilog->zl_spa, txg, &blk,
566 ZIL_MIN_BLKSZ, B_TRUE);
570 zil_init_log_chain(zilog, &blk);
574 * Allocate a log write buffer (lwb) for the first log block.
577 lwb = zil_alloc_lwb(zilog, &blk, txg, fastwrite);
580 * If we just allocated the first log block, commit our transaction
581 * and wait for zil_sync() to stuff the block poiner into zh_log.
582 * (zh is part of the MOS, so we cannot modify it in open context.)
586 txg_wait_synced(zilog->zl_dmu_pool, txg);
589 ASSERT(bcmp(&blk, &zh->zh_log, sizeof (blk)) == 0);
595 * In one tx, free all log blocks and clear the log header.
596 * If keep_first is set, then we're replaying a log with no content.
597 * We want to keep the first block, however, so that the first
598 * synchronous transaction doesn't require a txg_wait_synced()
599 * in zil_create(). We don't need to txg_wait_synced() here either
600 * when keep_first is set, because both zil_create() and zil_destroy()
601 * will wait for any in-progress destroys to complete.
604 zil_destroy(zilog_t *zilog, boolean_t keep_first)
606 const zil_header_t *zh = zilog->zl_header;
612 * Wait for any previous destroy to complete.
614 txg_wait_synced(zilog->zl_dmu_pool, zilog->zl_destroy_txg);
616 zilog->zl_old_header = *zh; /* debugging aid */
618 if (BP_IS_HOLE(&zh->zh_log))
621 tx = dmu_tx_create(zilog->zl_os);
622 VERIFY(dmu_tx_assign(tx, TXG_WAIT) == 0);
623 dsl_dataset_dirty(dmu_objset_ds(zilog->zl_os), tx);
624 txg = dmu_tx_get_txg(tx);
626 mutex_enter(&zilog->zl_lock);
628 ASSERT3U(zilog->zl_destroy_txg, <, txg);
629 zilog->zl_destroy_txg = txg;
630 zilog->zl_keep_first = keep_first;
632 if (!list_is_empty(&zilog->zl_lwb_list)) {
633 ASSERT(zh->zh_claim_txg == 0);
635 while ((lwb = list_head(&zilog->zl_lwb_list)) != NULL) {
636 ASSERT(lwb->lwb_zio == NULL);
637 if (lwb->lwb_fastwrite)
638 metaslab_fastwrite_unmark(zilog->zl_spa,
640 list_remove(&zilog->zl_lwb_list, lwb);
641 if (lwb->lwb_buf != NULL)
642 zio_buf_free(lwb->lwb_buf, lwb->lwb_sz);
643 zio_free_zil(zilog->zl_spa, txg, &lwb->lwb_blk);
644 kmem_cache_free(zil_lwb_cache, lwb);
646 } else if (!keep_first) {
647 zil_destroy_sync(zilog, tx);
649 mutex_exit(&zilog->zl_lock);
655 zil_destroy_sync(zilog_t *zilog, dmu_tx_t *tx)
657 ASSERT(list_is_empty(&zilog->zl_lwb_list));
658 (void) zil_parse(zilog, zil_free_log_block,
659 zil_free_log_record, tx, zilog->zl_header->zh_claim_txg);
663 zil_claim(dsl_pool_t *dp, dsl_dataset_t *ds, void *txarg)
665 dmu_tx_t *tx = txarg;
666 uint64_t first_txg = dmu_tx_get_txg(tx);
672 error = dmu_objset_own_obj(dp, ds->ds_object,
673 DMU_OST_ANY, B_FALSE, FTAG, &os);
676 * EBUSY indicates that the objset is inconsistent, in which
677 * case it can not have a ZIL.
679 if (error != EBUSY) {
680 cmn_err(CE_WARN, "can't open objset for %llu, error %u",
681 (unsigned long long)ds->ds_object, error);
687 zilog = dmu_objset_zil(os);
688 zh = zil_header_in_syncing_context(zilog);
690 if (spa_get_log_state(zilog->zl_spa) == SPA_LOG_CLEAR) {
691 if (!BP_IS_HOLE(&zh->zh_log))
692 zio_free_zil(zilog->zl_spa, first_txg, &zh->zh_log);
693 BP_ZERO(&zh->zh_log);
694 dsl_dataset_dirty(dmu_objset_ds(os), tx);
695 dmu_objset_disown(os, FTAG);
700 * Claim all log blocks if we haven't already done so, and remember
701 * the highest claimed sequence number. This ensures that if we can
702 * read only part of the log now (e.g. due to a missing device),
703 * but we can read the entire log later, we will not try to replay
704 * or destroy beyond the last block we successfully claimed.
706 ASSERT3U(zh->zh_claim_txg, <=, first_txg);
707 if (zh->zh_claim_txg == 0 && !BP_IS_HOLE(&zh->zh_log)) {
708 (void) zil_parse(zilog, zil_claim_log_block,
709 zil_claim_log_record, tx, first_txg);
710 zh->zh_claim_txg = first_txg;
711 zh->zh_claim_blk_seq = zilog->zl_parse_blk_seq;
712 zh->zh_claim_lr_seq = zilog->zl_parse_lr_seq;
713 if (zilog->zl_parse_lr_count || zilog->zl_parse_blk_count > 1)
714 zh->zh_flags |= ZIL_REPLAY_NEEDED;
715 zh->zh_flags |= ZIL_CLAIM_LR_SEQ_VALID;
716 dsl_dataset_dirty(dmu_objset_ds(os), tx);
719 ASSERT3U(first_txg, ==, (spa_last_synced_txg(zilog->zl_spa) + 1));
720 dmu_objset_disown(os, FTAG);
725 * Check the log by walking the log chain.
726 * Checksum errors are ok as they indicate the end of the chain.
727 * Any other error (no device or read failure) returns an error.
731 zil_check_log_chain(dsl_pool_t *dp, dsl_dataset_t *ds, void *tx)
740 error = dmu_objset_from_ds(ds, &os);
742 cmn_err(CE_WARN, "can't open objset %llu, error %d",
743 (unsigned long long)ds->ds_object, error);
747 zilog = dmu_objset_zil(os);
748 bp = (blkptr_t *)&zilog->zl_header->zh_log;
751 * Check the first block and determine if it's on a log device
752 * which may have been removed or faulted prior to loading this
753 * pool. If so, there's no point in checking the rest of the log
754 * as its content should have already been synced to the pool.
756 if (!BP_IS_HOLE(bp)) {
758 boolean_t valid = B_TRUE;
760 spa_config_enter(os->os_spa, SCL_STATE, FTAG, RW_READER);
761 vd = vdev_lookup_top(os->os_spa, DVA_GET_VDEV(&bp->blk_dva[0]));
762 if (vd->vdev_islog && vdev_is_dead(vd))
763 valid = vdev_log_state_valid(vd);
764 spa_config_exit(os->os_spa, SCL_STATE, FTAG);
771 * Because tx == NULL, zil_claim_log_block() will not actually claim
772 * any blocks, but just determine whether it is possible to do so.
773 * In addition to checking the log chain, zil_claim_log_block()
774 * will invoke zio_claim() with a done func of spa_claim_notify(),
775 * which will update spa_max_claim_txg. See spa_load() for details.
777 error = zil_parse(zilog, zil_claim_log_block, zil_claim_log_record, tx,
778 zilog->zl_header->zh_claim_txg ? -1ULL : spa_first_txg(os->os_spa));
780 return ((error == ECKSUM || error == ENOENT) ? 0 : error);
784 zil_vdev_compare(const void *x1, const void *x2)
786 const uint64_t v1 = ((zil_vdev_node_t *)x1)->zv_vdev;
787 const uint64_t v2 = ((zil_vdev_node_t *)x2)->zv_vdev;
798 zil_add_block(zilog_t *zilog, const blkptr_t *bp)
800 avl_tree_t *t = &zilog->zl_vdev_tree;
802 zil_vdev_node_t *zv, zvsearch;
803 int ndvas = BP_GET_NDVAS(bp);
806 if (zfs_nocacheflush)
809 ASSERT(zilog->zl_writer);
812 * Even though we're zl_writer, we still need a lock because the
813 * zl_get_data() callbacks may have dmu_sync() done callbacks
814 * that will run concurrently.
816 mutex_enter(&zilog->zl_vdev_lock);
817 for (i = 0; i < ndvas; i++) {
818 zvsearch.zv_vdev = DVA_GET_VDEV(&bp->blk_dva[i]);
819 if (avl_find(t, &zvsearch, &where) == NULL) {
820 zv = kmem_alloc(sizeof (*zv), KM_SLEEP);
821 zv->zv_vdev = zvsearch.zv_vdev;
822 avl_insert(t, zv, where);
825 mutex_exit(&zilog->zl_vdev_lock);
829 zil_flush_vdevs(zilog_t *zilog)
831 spa_t *spa = zilog->zl_spa;
832 avl_tree_t *t = &zilog->zl_vdev_tree;
837 ASSERT(zilog->zl_writer);
840 * We don't need zl_vdev_lock here because we're the zl_writer,
841 * and all zl_get_data() callbacks are done.
843 if (avl_numnodes(t) == 0)
846 spa_config_enter(spa, SCL_STATE, FTAG, RW_READER);
848 zio = zio_root(spa, NULL, NULL, ZIO_FLAG_CANFAIL);
850 while ((zv = avl_destroy_nodes(t, &cookie)) != NULL) {
851 vdev_t *vd = vdev_lookup_top(spa, zv->zv_vdev);
854 kmem_free(zv, sizeof (*zv));
858 * Wait for all the flushes to complete. Not all devices actually
859 * support the DKIOCFLUSHWRITECACHE ioctl, so it's OK if it fails.
861 (void) zio_wait(zio);
863 spa_config_exit(spa, SCL_STATE, FTAG);
867 * Function called when a log block write completes
870 zil_lwb_write_done(zio_t *zio)
872 lwb_t *lwb = zio->io_private;
873 zilog_t *zilog = lwb->lwb_zilog;
874 dmu_tx_t *tx = lwb->lwb_tx;
876 ASSERT(BP_GET_COMPRESS(zio->io_bp) == ZIO_COMPRESS_OFF);
877 ASSERT(BP_GET_TYPE(zio->io_bp) == DMU_OT_INTENT_LOG);
878 ASSERT(BP_GET_LEVEL(zio->io_bp) == 0);
879 ASSERT(BP_GET_BYTEORDER(zio->io_bp) == ZFS_HOST_BYTEORDER);
880 ASSERT(!BP_IS_GANG(zio->io_bp));
881 ASSERT(!BP_IS_HOLE(zio->io_bp));
882 ASSERT(BP_GET_FILL(zio->io_bp) == 0);
885 * Ensure the lwb buffer pointer is cleared before releasing
886 * the txg. If we have had an allocation failure and
887 * the txg is waiting to sync then we want want zil_sync()
888 * to remove the lwb so that it's not picked up as the next new
889 * one in zil_commit_writer(). zil_sync() will only remove
890 * the lwb if lwb_buf is null.
892 zio_buf_free(lwb->lwb_buf, lwb->lwb_sz);
893 mutex_enter(&zilog->zl_lock);
895 lwb->lwb_fastwrite = FALSE;
898 mutex_exit(&zilog->zl_lock);
901 * Now that we've written this log block, we have a stable pointer
902 * to the next block in the chain, so it's OK to let the txg in
903 * which we allocated the next block sync.
909 * Initialize the io for a log block.
912 zil_lwb_write_init(zilog_t *zilog, lwb_t *lwb)
916 SET_BOOKMARK(&zb, lwb->lwb_blk.blk_cksum.zc_word[ZIL_ZC_OBJSET],
917 ZB_ZIL_OBJECT, ZB_ZIL_LEVEL,
918 lwb->lwb_blk.blk_cksum.zc_word[ZIL_ZC_SEQ]);
920 if (zilog->zl_root_zio == NULL) {
921 zilog->zl_root_zio = zio_root(zilog->zl_spa, NULL, NULL,
925 /* Lock so zil_sync() doesn't fastwrite_unmark after zio is created */
926 mutex_enter(&zilog->zl_lock);
927 if (lwb->lwb_zio == NULL) {
928 if (!lwb->lwb_fastwrite) {
929 metaslab_fastwrite_mark(zilog->zl_spa, &lwb->lwb_blk);
930 lwb->lwb_fastwrite = 1;
932 lwb->lwb_zio = zio_rewrite(zilog->zl_root_zio, zilog->zl_spa,
933 0, &lwb->lwb_blk, lwb->lwb_buf, BP_GET_LSIZE(&lwb->lwb_blk),
934 zil_lwb_write_done, lwb, ZIO_PRIORITY_SYNC_WRITE,
935 ZIO_FLAG_CANFAIL | ZIO_FLAG_DONT_PROPAGATE |
936 ZIO_FLAG_FASTWRITE, &zb);
938 mutex_exit(&zilog->zl_lock);
942 * Define a limited set of intent log block sizes.
944 * These must be a multiple of 4KB. Note only the amount used (again
945 * aligned to 4KB) actually gets written. However, we can't always just
946 * allocate SPA_OLD_MAXBLOCKSIZE as the slog space could be exhausted.
948 uint64_t zil_block_buckets[] = {
949 4096, /* non TX_WRITE */
950 8192+4096, /* data base */
951 32*1024 + 4096, /* NFS writes */
956 * Use the slog as long as the current commit size is less than the
957 * limit or the total list size is less than 2X the limit. Limit
958 * checking is disabled by setting zil_slog_limit to UINT64_MAX.
960 unsigned long zil_slog_limit = 1024 * 1024;
961 #define USE_SLOG(zilog) (((zilog)->zl_cur_used < zil_slog_limit) || \
962 ((zilog)->zl_itx_list_sz < (zil_slog_limit << 1)))
965 * Start a log block write and advance to the next log block.
966 * Calls are serialized.
969 zil_lwb_write_start(zilog_t *zilog, lwb_t *lwb)
973 spa_t *spa = zilog->zl_spa;
977 uint64_t zil_blksz, wsz;
981 if (BP_GET_CHECKSUM(&lwb->lwb_blk) == ZIO_CHECKSUM_ZILOG2) {
982 zilc = (zil_chain_t *)lwb->lwb_buf;
983 bp = &zilc->zc_next_blk;
985 zilc = (zil_chain_t *)(lwb->lwb_buf + lwb->lwb_sz);
986 bp = &zilc->zc_next_blk;
989 ASSERT(lwb->lwb_nused <= lwb->lwb_sz);
992 * Allocate the next block and save its address in this block
993 * before writing it in order to establish the log chain.
994 * Note that if the allocation of nlwb synced before we wrote
995 * the block that points at it (lwb), we'd leak it if we crashed.
996 * Therefore, we don't do dmu_tx_commit() until zil_lwb_write_done().
997 * We dirty the dataset to ensure that zil_sync() will be called
998 * to clean up in the event of allocation failure or I/O failure.
1000 tx = dmu_tx_create(zilog->zl_os);
1001 VERIFY(dmu_tx_assign(tx, TXG_WAIT) == 0);
1002 dsl_dataset_dirty(dmu_objset_ds(zilog->zl_os), tx);
1003 txg = dmu_tx_get_txg(tx);
1008 * Log blocks are pre-allocated. Here we select the size of the next
1009 * block, based on size used in the last block.
1010 * - first find the smallest bucket that will fit the block from a
1011 * limited set of block sizes. This is because it's faster to write
1012 * blocks allocated from the same metaslab as they are adjacent or
1014 * - next find the maximum from the new suggested size and an array of
1015 * previous sizes. This lessens a picket fence effect of wrongly
1016 * guesssing the size if we have a stream of say 2k, 64k, 2k, 64k
1019 * Note we only write what is used, but we can't just allocate
1020 * the maximum block size because we can exhaust the available
1023 zil_blksz = zilog->zl_cur_used + sizeof (zil_chain_t);
1024 for (i = 0; zil_blksz > zil_block_buckets[i]; i++)
1026 zil_blksz = zil_block_buckets[i];
1027 if (zil_blksz == UINT64_MAX)
1028 zil_blksz = SPA_OLD_MAXBLOCKSIZE;
1029 zilog->zl_prev_blks[zilog->zl_prev_rotor] = zil_blksz;
1030 for (i = 0; i < ZIL_PREV_BLKS; i++)
1031 zil_blksz = MAX(zil_blksz, zilog->zl_prev_blks[i]);
1032 zilog->zl_prev_rotor = (zilog->zl_prev_rotor + 1) & (ZIL_PREV_BLKS - 1);
1035 use_slog = USE_SLOG(zilog);
1036 error = zio_alloc_zil(spa, txg, bp, zil_blksz,
1039 ZIL_STAT_BUMP(zil_itx_metaslab_slog_count);
1040 ZIL_STAT_INCR(zil_itx_metaslab_slog_bytes, lwb->lwb_nused);
1042 ZIL_STAT_BUMP(zil_itx_metaslab_normal_count);
1043 ZIL_STAT_INCR(zil_itx_metaslab_normal_bytes, lwb->lwb_nused);
1046 ASSERT3U(bp->blk_birth, ==, txg);
1047 bp->blk_cksum = lwb->lwb_blk.blk_cksum;
1048 bp->blk_cksum.zc_word[ZIL_ZC_SEQ]++;
1051 * Allocate a new log write buffer (lwb).
1053 nlwb = zil_alloc_lwb(zilog, bp, txg, TRUE);
1055 /* Record the block for later vdev flushing */
1056 zil_add_block(zilog, &lwb->lwb_blk);
1059 if (BP_GET_CHECKSUM(&lwb->lwb_blk) == ZIO_CHECKSUM_ZILOG2) {
1060 /* For Slim ZIL only write what is used. */
1061 wsz = P2ROUNDUP_TYPED(lwb->lwb_nused, ZIL_MIN_BLKSZ, uint64_t);
1062 ASSERT3U(wsz, <=, lwb->lwb_sz);
1063 zio_shrink(lwb->lwb_zio, wsz);
1070 zilc->zc_nused = lwb->lwb_nused;
1071 zilc->zc_eck.zec_cksum = lwb->lwb_blk.blk_cksum;
1074 * clear unused data for security
1076 bzero(lwb->lwb_buf + lwb->lwb_nused, wsz - lwb->lwb_nused);
1078 zio_nowait(lwb->lwb_zio); /* Kick off the write for the old log block */
1081 * If there was an allocation failure then nlwb will be null which
1082 * forces a txg_wait_synced().
1088 zil_lwb_commit(zilog_t *zilog, itx_t *itx, lwb_t *lwb)
1090 lr_t *lrc = &itx->itx_lr; /* common log record */
1091 lr_write_t *lrw = (lr_write_t *)lrc;
1093 uint64_t txg = lrc->lrc_txg;
1094 uint64_t reclen = lrc->lrc_reclen;
1100 ASSERT(lwb->lwb_buf != NULL);
1101 ASSERT(zilog_is_dirty(zilog) ||
1102 spa_freeze_txg(zilog->zl_spa) != UINT64_MAX);
1104 if (lrc->lrc_txtype == TX_WRITE && itx->itx_wr_state == WR_NEED_COPY)
1105 dlen = P2ROUNDUP_TYPED(
1106 lrw->lr_length, sizeof (uint64_t), uint64_t);
1108 zilog->zl_cur_used += (reclen + dlen);
1110 zil_lwb_write_init(zilog, lwb);
1113 * If this record won't fit in the current log block, start a new one.
1115 if (lwb->lwb_nused + reclen + dlen > lwb->lwb_sz) {
1116 lwb = zil_lwb_write_start(zilog, lwb);
1119 zil_lwb_write_init(zilog, lwb);
1120 ASSERT(LWB_EMPTY(lwb));
1121 if (lwb->lwb_nused + reclen + dlen > lwb->lwb_sz) {
1122 txg_wait_synced(zilog->zl_dmu_pool, txg);
1127 lr_buf = lwb->lwb_buf + lwb->lwb_nused;
1128 bcopy(lrc, lr_buf, reclen);
1129 lrc = (lr_t *)lr_buf;
1130 lrw = (lr_write_t *)lrc;
1132 ZIL_STAT_BUMP(zil_itx_count);
1135 * If it's a write, fetch the data or get its blkptr as appropriate.
1137 if (lrc->lrc_txtype == TX_WRITE) {
1138 if (txg > spa_freeze_txg(zilog->zl_spa))
1139 txg_wait_synced(zilog->zl_dmu_pool, txg);
1140 if (itx->itx_wr_state == WR_COPIED) {
1141 ZIL_STAT_BUMP(zil_itx_copied_count);
1142 ZIL_STAT_INCR(zil_itx_copied_bytes, lrw->lr_length);
1148 ASSERT(itx->itx_wr_state == WR_NEED_COPY);
1149 dbuf = lr_buf + reclen;
1150 lrw->lr_common.lrc_reclen += dlen;
1151 ZIL_STAT_BUMP(zil_itx_needcopy_count);
1152 ZIL_STAT_INCR(zil_itx_needcopy_bytes,
1155 ASSERT(itx->itx_wr_state == WR_INDIRECT);
1157 ZIL_STAT_BUMP(zil_itx_indirect_count);
1158 ZIL_STAT_INCR(zil_itx_indirect_bytes,
1161 error = zilog->zl_get_data(
1162 itx->itx_private, lrw, dbuf, lwb->lwb_zio);
1164 txg_wait_synced(zilog->zl_dmu_pool, txg);
1168 ASSERT(error == ENOENT || error == EEXIST ||
1176 * We're actually making an entry, so update lrc_seq to be the
1177 * log record sequence number. Note that this is generally not
1178 * equal to the itx sequence number because not all transactions
1179 * are synchronous, and sometimes spa_sync() gets there first.
1181 lrc->lrc_seq = ++zilog->zl_lr_seq; /* we are single threaded */
1182 lwb->lwb_nused += reclen + dlen;
1183 lwb->lwb_max_txg = MAX(lwb->lwb_max_txg, txg);
1184 ASSERT3U(lwb->lwb_nused, <=, lwb->lwb_sz);
1185 ASSERT0(P2PHASE(lwb->lwb_nused, sizeof (uint64_t)));
1191 zil_itx_create(uint64_t txtype, size_t lrsize)
1195 lrsize = P2ROUNDUP_TYPED(lrsize, sizeof (uint64_t), size_t);
1197 itx = zio_data_buf_alloc(offsetof(itx_t, itx_lr) + lrsize);
1198 itx->itx_lr.lrc_txtype = txtype;
1199 itx->itx_lr.lrc_reclen = lrsize;
1200 itx->itx_sod = lrsize; /* if write & WR_NEED_COPY will be increased */
1201 itx->itx_lr.lrc_seq = 0; /* defensive */
1202 itx->itx_sync = B_TRUE; /* default is synchronous */
1203 itx->itx_callback = NULL;
1204 itx->itx_callback_data = NULL;
1210 zil_itx_destroy(itx_t *itx)
1212 zio_data_buf_free(itx, offsetof(itx_t, itx_lr)+itx->itx_lr.lrc_reclen);
1216 * Free up the sync and async itxs. The itxs_t has already been detached
1217 * so no locks are needed.
1220 zil_itxg_clean(itxs_t *itxs)
1226 itx_async_node_t *ian;
1228 list = &itxs->i_sync_list;
1229 while ((itx = list_head(list)) != NULL) {
1230 if (itx->itx_callback != NULL)
1231 itx->itx_callback(itx->itx_callback_data);
1232 list_remove(list, itx);
1233 zil_itx_destroy(itx);
1237 t = &itxs->i_async_tree;
1238 while ((ian = avl_destroy_nodes(t, &cookie)) != NULL) {
1239 list = &ian->ia_list;
1240 while ((itx = list_head(list)) != NULL) {
1241 if (itx->itx_callback != NULL)
1242 itx->itx_callback(itx->itx_callback_data);
1243 list_remove(list, itx);
1244 zil_itx_destroy(itx);
1247 kmem_free(ian, sizeof (itx_async_node_t));
1251 kmem_free(itxs, sizeof (itxs_t));
1255 zil_aitx_compare(const void *x1, const void *x2)
1257 const uint64_t o1 = ((itx_async_node_t *)x1)->ia_foid;
1258 const uint64_t o2 = ((itx_async_node_t *)x2)->ia_foid;
1269 * Remove all async itx with the given oid.
1272 zil_remove_async(zilog_t *zilog, uint64_t oid)
1275 itx_async_node_t *ian;
1282 list_create(&clean_list, sizeof (itx_t), offsetof(itx_t, itx_node));
1284 if (spa_freeze_txg(zilog->zl_spa) != UINT64_MAX) /* ziltest support */
1287 otxg = spa_last_synced_txg(zilog->zl_spa) + 1;
1289 for (txg = otxg; txg < (otxg + TXG_CONCURRENT_STATES); txg++) {
1290 itxg_t *itxg = &zilog->zl_itxg[txg & TXG_MASK];
1292 mutex_enter(&itxg->itxg_lock);
1293 if (itxg->itxg_txg != txg) {
1294 mutex_exit(&itxg->itxg_lock);
1299 * Locate the object node and append its list.
1301 t = &itxg->itxg_itxs->i_async_tree;
1302 ian = avl_find(t, &oid, &where);
1304 list_move_tail(&clean_list, &ian->ia_list);
1305 mutex_exit(&itxg->itxg_lock);
1307 while ((itx = list_head(&clean_list)) != NULL) {
1308 if (itx->itx_callback != NULL)
1309 itx->itx_callback(itx->itx_callback_data);
1310 list_remove(&clean_list, itx);
1311 zil_itx_destroy(itx);
1313 list_destroy(&clean_list);
1317 zil_itx_assign(zilog_t *zilog, itx_t *itx, dmu_tx_t *tx)
1321 itxs_t *itxs, *clean = NULL;
1324 * Object ids can be re-instantiated in the next txg so
1325 * remove any async transactions to avoid future leaks.
1326 * This can happen if a fsync occurs on the re-instantiated
1327 * object for a WR_INDIRECT or WR_NEED_COPY write, which gets
1328 * the new file data and flushes a write record for the old object.
1330 if ((itx->itx_lr.lrc_txtype & ~TX_CI) == TX_REMOVE)
1331 zil_remove_async(zilog, itx->itx_oid);
1334 * Ensure the data of a renamed file is committed before the rename.
1336 if ((itx->itx_lr.lrc_txtype & ~TX_CI) == TX_RENAME)
1337 zil_async_to_sync(zilog, itx->itx_oid);
1339 if (spa_freeze_txg(zilog->zl_spa) != UINT64_MAX)
1342 txg = dmu_tx_get_txg(tx);
1344 itxg = &zilog->zl_itxg[txg & TXG_MASK];
1345 mutex_enter(&itxg->itxg_lock);
1346 itxs = itxg->itxg_itxs;
1347 if (itxg->itxg_txg != txg) {
1350 * The zil_clean callback hasn't got around to cleaning
1351 * this itxg. Save the itxs for release below.
1352 * This should be rare.
1354 atomic_add_64(&zilog->zl_itx_list_sz, -itxg->itxg_sod);
1356 clean = itxg->itxg_itxs;
1358 ASSERT(itxg->itxg_sod == 0);
1359 itxg->itxg_txg = txg;
1360 itxs = itxg->itxg_itxs = kmem_zalloc(sizeof (itxs_t),
1363 list_create(&itxs->i_sync_list, sizeof (itx_t),
1364 offsetof(itx_t, itx_node));
1365 avl_create(&itxs->i_async_tree, zil_aitx_compare,
1366 sizeof (itx_async_node_t),
1367 offsetof(itx_async_node_t, ia_node));
1369 if (itx->itx_sync) {
1370 list_insert_tail(&itxs->i_sync_list, itx);
1371 atomic_add_64(&zilog->zl_itx_list_sz, itx->itx_sod);
1372 itxg->itxg_sod += itx->itx_sod;
1374 avl_tree_t *t = &itxs->i_async_tree;
1376 LR_FOID_GET_OBJ(((lr_ooo_t *)&itx->itx_lr)->lr_foid);
1377 itx_async_node_t *ian;
1380 ian = avl_find(t, &foid, &where);
1382 ian = kmem_alloc(sizeof (itx_async_node_t),
1384 list_create(&ian->ia_list, sizeof (itx_t),
1385 offsetof(itx_t, itx_node));
1386 ian->ia_foid = foid;
1387 avl_insert(t, ian, where);
1389 list_insert_tail(&ian->ia_list, itx);
1392 itx->itx_lr.lrc_txg = dmu_tx_get_txg(tx);
1393 zilog_dirty(zilog, txg);
1394 mutex_exit(&itxg->itxg_lock);
1396 /* Release the old itxs now we've dropped the lock */
1398 zil_itxg_clean(clean);
1402 * If there are any in-memory intent log transactions which have now been
1403 * synced then start up a taskq to free them. We should only do this after we
1404 * have written out the uberblocks (i.e. txg has been comitted) so that
1405 * don't inadvertently clean out in-memory log records that would be required
1409 zil_clean(zilog_t *zilog, uint64_t synced_txg)
1411 itxg_t *itxg = &zilog->zl_itxg[synced_txg & TXG_MASK];
1414 mutex_enter(&itxg->itxg_lock);
1415 if (itxg->itxg_itxs == NULL || itxg->itxg_txg == ZILTEST_TXG) {
1416 mutex_exit(&itxg->itxg_lock);
1419 ASSERT3U(itxg->itxg_txg, <=, synced_txg);
1420 ASSERT(itxg->itxg_txg != 0);
1421 ASSERT(zilog->zl_clean_taskq != NULL);
1422 atomic_add_64(&zilog->zl_itx_list_sz, -itxg->itxg_sod);
1424 clean_me = itxg->itxg_itxs;
1425 itxg->itxg_itxs = NULL;
1427 mutex_exit(&itxg->itxg_lock);
1429 * Preferably start a task queue to free up the old itxs but
1430 * if taskq_dispatch can't allocate resources to do that then
1431 * free it in-line. This should be rare. Note, using TQ_SLEEP
1432 * created a bad performance problem.
1434 if (taskq_dispatch(zilog->zl_clean_taskq,
1435 (void (*)(void *))zil_itxg_clean, clean_me, TQ_NOSLEEP) == 0)
1436 zil_itxg_clean(clean_me);
1440 * Get the list of itxs to commit into zl_itx_commit_list.
1443 zil_get_commit_list(zilog_t *zilog)
1446 list_t *commit_list = &zilog->zl_itx_commit_list;
1447 uint64_t push_sod = 0;
1449 if (spa_freeze_txg(zilog->zl_spa) != UINT64_MAX) /* ziltest support */
1452 otxg = spa_last_synced_txg(zilog->zl_spa) + 1;
1454 for (txg = otxg; txg < (otxg + TXG_CONCURRENT_STATES); txg++) {
1455 itxg_t *itxg = &zilog->zl_itxg[txg & TXG_MASK];
1457 mutex_enter(&itxg->itxg_lock);
1458 if (itxg->itxg_txg != txg) {
1459 mutex_exit(&itxg->itxg_lock);
1463 list_move_tail(commit_list, &itxg->itxg_itxs->i_sync_list);
1464 push_sod += itxg->itxg_sod;
1467 mutex_exit(&itxg->itxg_lock);
1469 atomic_add_64(&zilog->zl_itx_list_sz, -push_sod);
1473 * Move the async itxs for a specified object to commit into sync lists.
1476 zil_async_to_sync(zilog_t *zilog, uint64_t foid)
1479 itx_async_node_t *ian;
1483 if (spa_freeze_txg(zilog->zl_spa) != UINT64_MAX) /* ziltest support */
1486 otxg = spa_last_synced_txg(zilog->zl_spa) + 1;
1488 for (txg = otxg; txg < (otxg + TXG_CONCURRENT_STATES); txg++) {
1489 itxg_t *itxg = &zilog->zl_itxg[txg & TXG_MASK];
1491 mutex_enter(&itxg->itxg_lock);
1492 if (itxg->itxg_txg != txg) {
1493 mutex_exit(&itxg->itxg_lock);
1498 * If a foid is specified then find that node and append its
1499 * list. Otherwise walk the tree appending all the lists
1500 * to the sync list. We add to the end rather than the
1501 * beginning to ensure the create has happened.
1503 t = &itxg->itxg_itxs->i_async_tree;
1505 ian = avl_find(t, &foid, &where);
1507 list_move_tail(&itxg->itxg_itxs->i_sync_list,
1511 void *cookie = NULL;
1513 while ((ian = avl_destroy_nodes(t, &cookie)) != NULL) {
1514 list_move_tail(&itxg->itxg_itxs->i_sync_list,
1516 list_destroy(&ian->ia_list);
1517 kmem_free(ian, sizeof (itx_async_node_t));
1520 mutex_exit(&itxg->itxg_lock);
1525 zil_commit_writer(zilog_t *zilog)
1530 spa_t *spa = zilog->zl_spa;
1533 ASSERT(zilog->zl_root_zio == NULL);
1535 mutex_exit(&zilog->zl_lock);
1537 zil_get_commit_list(zilog);
1540 * Return if there's nothing to commit before we dirty the fs by
1541 * calling zil_create().
1543 if (list_head(&zilog->zl_itx_commit_list) == NULL) {
1544 mutex_enter(&zilog->zl_lock);
1548 if (zilog->zl_suspend) {
1551 lwb = list_tail(&zilog->zl_lwb_list);
1553 lwb = zil_create(zilog);
1556 DTRACE_PROBE1(zil__cw1, zilog_t *, zilog);
1557 for (itx = list_head(&zilog->zl_itx_commit_list); itx != NULL;
1558 itx = list_next(&zilog->zl_itx_commit_list, itx)) {
1559 txg = itx->itx_lr.lrc_txg;
1562 if (txg > spa_last_synced_txg(spa) || txg > spa_freeze_txg(spa))
1563 lwb = zil_lwb_commit(zilog, itx, lwb);
1565 DTRACE_PROBE1(zil__cw2, zilog_t *, zilog);
1567 /* write the last block out */
1568 if (lwb != NULL && lwb->lwb_zio != NULL)
1569 lwb = zil_lwb_write_start(zilog, lwb);
1571 zilog->zl_cur_used = 0;
1574 * Wait if necessary for the log blocks to be on stable storage.
1576 if (zilog->zl_root_zio) {
1577 error = zio_wait(zilog->zl_root_zio);
1578 zilog->zl_root_zio = NULL;
1579 zil_flush_vdevs(zilog);
1582 if (error || lwb == NULL)
1583 txg_wait_synced(zilog->zl_dmu_pool, 0);
1585 while ((itx = list_head(&zilog->zl_itx_commit_list))) {
1586 txg = itx->itx_lr.lrc_txg;
1589 if (itx->itx_callback != NULL)
1590 itx->itx_callback(itx->itx_callback_data);
1591 list_remove(&zilog->zl_itx_commit_list, itx);
1592 zil_itx_destroy(itx);
1595 mutex_enter(&zilog->zl_lock);
1598 * Remember the highest committed log sequence number for ztest.
1599 * We only update this value when all the log writes succeeded,
1600 * because ztest wants to ASSERT that it got the whole log chain.
1602 if (error == 0 && lwb != NULL)
1603 zilog->zl_commit_lr_seq = zilog->zl_lr_seq;
1607 * Commit zfs transactions to stable storage.
1608 * If foid is 0 push out all transactions, otherwise push only those
1609 * for that object or might reference that object.
1611 * itxs are committed in batches. In a heavily stressed zil there will be
1612 * a commit writer thread who is writing out a bunch of itxs to the log
1613 * for a set of committing threads (cthreads) in the same batch as the writer.
1614 * Those cthreads are all waiting on the same cv for that batch.
1616 * There will also be a different and growing batch of threads that are
1617 * waiting to commit (qthreads). When the committing batch completes
1618 * a transition occurs such that the cthreads exit and the qthreads become
1619 * cthreads. One of the new cthreads becomes the writer thread for the
1620 * batch. Any new threads arriving become new qthreads.
1622 * Only 2 condition variables are needed and there's no transition
1623 * between the two cvs needed. They just flip-flop between qthreads
1626 * Using this scheme we can efficiently wakeup up only those threads
1627 * that have been committed.
1630 zil_commit(zilog_t *zilog, uint64_t foid)
1634 if (zilog->zl_sync == ZFS_SYNC_DISABLED)
1637 ZIL_STAT_BUMP(zil_commit_count);
1639 /* move the async itxs for the foid to the sync queues */
1640 zil_async_to_sync(zilog, foid);
1642 mutex_enter(&zilog->zl_lock);
1643 mybatch = zilog->zl_next_batch;
1644 while (zilog->zl_writer) {
1645 cv_wait(&zilog->zl_cv_batch[mybatch & 1], &zilog->zl_lock);
1646 if (mybatch <= zilog->zl_com_batch) {
1647 mutex_exit(&zilog->zl_lock);
1652 zilog->zl_next_batch++;
1653 zilog->zl_writer = B_TRUE;
1654 ZIL_STAT_BUMP(zil_commit_writer_count);
1655 zil_commit_writer(zilog);
1656 zilog->zl_com_batch = mybatch;
1657 zilog->zl_writer = B_FALSE;
1659 /* wake up one thread to become the next writer */
1660 cv_signal(&zilog->zl_cv_batch[(mybatch+1) & 1]);
1662 /* wake up all threads waiting for this batch to be committed */
1663 cv_broadcast(&zilog->zl_cv_batch[mybatch & 1]);
1665 mutex_exit(&zilog->zl_lock);
1669 * Called in syncing context to free committed log blocks and update log header.
1672 zil_sync(zilog_t *zilog, dmu_tx_t *tx)
1674 zil_header_t *zh = zil_header_in_syncing_context(zilog);
1675 uint64_t txg = dmu_tx_get_txg(tx);
1676 spa_t *spa = zilog->zl_spa;
1677 uint64_t *replayed_seq = &zilog->zl_replayed_seq[txg & TXG_MASK];
1681 * We don't zero out zl_destroy_txg, so make sure we don't try
1682 * to destroy it twice.
1684 if (spa_sync_pass(spa) != 1)
1687 mutex_enter(&zilog->zl_lock);
1689 ASSERT(zilog->zl_stop_sync == 0);
1691 if (*replayed_seq != 0) {
1692 ASSERT(zh->zh_replay_seq < *replayed_seq);
1693 zh->zh_replay_seq = *replayed_seq;
1697 if (zilog->zl_destroy_txg == txg) {
1698 blkptr_t blk = zh->zh_log;
1700 ASSERT(list_head(&zilog->zl_lwb_list) == NULL);
1702 bzero(zh, sizeof (zil_header_t));
1703 bzero(zilog->zl_replayed_seq, sizeof (zilog->zl_replayed_seq));
1705 if (zilog->zl_keep_first) {
1707 * If this block was part of log chain that couldn't
1708 * be claimed because a device was missing during
1709 * zil_claim(), but that device later returns,
1710 * then this block could erroneously appear valid.
1711 * To guard against this, assign a new GUID to the new
1712 * log chain so it doesn't matter what blk points to.
1714 zil_init_log_chain(zilog, &blk);
1719 while ((lwb = list_head(&zilog->zl_lwb_list)) != NULL) {
1720 zh->zh_log = lwb->lwb_blk;
1721 if (lwb->lwb_buf != NULL || lwb->lwb_max_txg > txg)
1724 ASSERT(lwb->lwb_zio == NULL);
1726 list_remove(&zilog->zl_lwb_list, lwb);
1727 zio_free_zil(spa, txg, &lwb->lwb_blk);
1728 kmem_cache_free(zil_lwb_cache, lwb);
1731 * If we don't have anything left in the lwb list then
1732 * we've had an allocation failure and we need to zero
1733 * out the zil_header blkptr so that we don't end
1734 * up freeing the same block twice.
1736 if (list_head(&zilog->zl_lwb_list) == NULL)
1737 BP_ZERO(&zh->zh_log);
1741 * Remove fastwrite on any blocks that have been pre-allocated for
1742 * the next commit. This prevents fastwrite counter pollution by
1743 * unused, long-lived LWBs.
1745 for (; lwb != NULL; lwb = list_next(&zilog->zl_lwb_list, lwb)) {
1746 if (lwb->lwb_fastwrite && !lwb->lwb_zio) {
1747 metaslab_fastwrite_unmark(zilog->zl_spa, &lwb->lwb_blk);
1748 lwb->lwb_fastwrite = 0;
1752 mutex_exit(&zilog->zl_lock);
1758 zil_lwb_cache = kmem_cache_create("zil_lwb_cache",
1759 sizeof (struct lwb), 0, NULL, NULL, NULL, NULL, NULL, 0);
1761 zil_ksp = kstat_create("zfs", 0, "zil", "misc",
1762 KSTAT_TYPE_NAMED, sizeof (zil_stats) / sizeof (kstat_named_t),
1763 KSTAT_FLAG_VIRTUAL);
1765 if (zil_ksp != NULL) {
1766 zil_ksp->ks_data = &zil_stats;
1767 kstat_install(zil_ksp);
1774 kmem_cache_destroy(zil_lwb_cache);
1776 if (zil_ksp != NULL) {
1777 kstat_delete(zil_ksp);
1783 zil_set_sync(zilog_t *zilog, uint64_t sync)
1785 zilog->zl_sync = sync;
1789 zil_set_logbias(zilog_t *zilog, uint64_t logbias)
1791 zilog->zl_logbias = logbias;
1795 zil_alloc(objset_t *os, zil_header_t *zh_phys)
1800 zilog = kmem_zalloc(sizeof (zilog_t), KM_SLEEP);
1802 zilog->zl_header = zh_phys;
1804 zilog->zl_spa = dmu_objset_spa(os);
1805 zilog->zl_dmu_pool = dmu_objset_pool(os);
1806 zilog->zl_destroy_txg = TXG_INITIAL - 1;
1807 zilog->zl_logbias = dmu_objset_logbias(os);
1808 zilog->zl_sync = dmu_objset_syncprop(os);
1809 zilog->zl_next_batch = 1;
1811 mutex_init(&zilog->zl_lock, NULL, MUTEX_DEFAULT, NULL);
1813 for (i = 0; i < TXG_SIZE; i++) {
1814 mutex_init(&zilog->zl_itxg[i].itxg_lock, NULL,
1815 MUTEX_DEFAULT, NULL);
1818 list_create(&zilog->zl_lwb_list, sizeof (lwb_t),
1819 offsetof(lwb_t, lwb_node));
1821 list_create(&zilog->zl_itx_commit_list, sizeof (itx_t),
1822 offsetof(itx_t, itx_node));
1824 mutex_init(&zilog->zl_vdev_lock, NULL, MUTEX_DEFAULT, NULL);
1826 avl_create(&zilog->zl_vdev_tree, zil_vdev_compare,
1827 sizeof (zil_vdev_node_t), offsetof(zil_vdev_node_t, zv_node));
1829 cv_init(&zilog->zl_cv_writer, NULL, CV_DEFAULT, NULL);
1830 cv_init(&zilog->zl_cv_suspend, NULL, CV_DEFAULT, NULL);
1831 cv_init(&zilog->zl_cv_batch[0], NULL, CV_DEFAULT, NULL);
1832 cv_init(&zilog->zl_cv_batch[1], NULL, CV_DEFAULT, NULL);
1838 zil_free(zilog_t *zilog)
1842 zilog->zl_stop_sync = 1;
1844 ASSERT0(zilog->zl_suspend);
1845 ASSERT0(zilog->zl_suspending);
1847 ASSERT(list_is_empty(&zilog->zl_lwb_list));
1848 list_destroy(&zilog->zl_lwb_list);
1850 avl_destroy(&zilog->zl_vdev_tree);
1851 mutex_destroy(&zilog->zl_vdev_lock);
1853 ASSERT(list_is_empty(&zilog->zl_itx_commit_list));
1854 list_destroy(&zilog->zl_itx_commit_list);
1856 for (i = 0; i < TXG_SIZE; i++) {
1858 * It's possible for an itx to be generated that doesn't dirty
1859 * a txg (e.g. ztest TX_TRUNCATE). So there's no zil_clean()
1860 * callback to remove the entry. We remove those here.
1862 * Also free up the ziltest itxs.
1864 if (zilog->zl_itxg[i].itxg_itxs)
1865 zil_itxg_clean(zilog->zl_itxg[i].itxg_itxs);
1866 mutex_destroy(&zilog->zl_itxg[i].itxg_lock);
1869 mutex_destroy(&zilog->zl_lock);
1871 cv_destroy(&zilog->zl_cv_writer);
1872 cv_destroy(&zilog->zl_cv_suspend);
1873 cv_destroy(&zilog->zl_cv_batch[0]);
1874 cv_destroy(&zilog->zl_cv_batch[1]);
1876 kmem_free(zilog, sizeof (zilog_t));
1880 * Open an intent log.
1883 zil_open(objset_t *os, zil_get_data_t *get_data)
1885 zilog_t *zilog = dmu_objset_zil(os);
1887 ASSERT(zilog->zl_clean_taskq == NULL);
1888 ASSERT(zilog->zl_get_data == NULL);
1889 ASSERT(list_is_empty(&zilog->zl_lwb_list));
1891 zilog->zl_get_data = get_data;
1892 zilog->zl_clean_taskq = taskq_create("zil_clean", 1, defclsyspri,
1893 2, 2, TASKQ_PREPOPULATE);
1899 * Close an intent log.
1902 zil_close(zilog_t *zilog)
1907 zil_commit(zilog, 0); /* commit all itx */
1910 * The lwb_max_txg for the stubby lwb will reflect the last activity
1911 * for the zil. After a txg_wait_synced() on the txg we know all the
1912 * callbacks have occurred that may clean the zil. Only then can we
1913 * destroy the zl_clean_taskq.
1915 mutex_enter(&zilog->zl_lock);
1916 lwb = list_tail(&zilog->zl_lwb_list);
1918 txg = lwb->lwb_max_txg;
1919 mutex_exit(&zilog->zl_lock);
1921 txg_wait_synced(zilog->zl_dmu_pool, txg);
1922 if (txg < spa_freeze_txg(zilog->zl_spa))
1923 ASSERT(!zilog_is_dirty(zilog));
1925 taskq_destroy(zilog->zl_clean_taskq);
1926 zilog->zl_clean_taskq = NULL;
1927 zilog->zl_get_data = NULL;
1930 * We should have only one LWB left on the list; remove it now.
1932 mutex_enter(&zilog->zl_lock);
1933 lwb = list_head(&zilog->zl_lwb_list);
1935 ASSERT(lwb == list_tail(&zilog->zl_lwb_list));
1936 ASSERT(lwb->lwb_zio == NULL);
1937 if (lwb->lwb_fastwrite)
1938 metaslab_fastwrite_unmark(zilog->zl_spa, &lwb->lwb_blk);
1939 list_remove(&zilog->zl_lwb_list, lwb);
1940 zio_buf_free(lwb->lwb_buf, lwb->lwb_sz);
1941 kmem_cache_free(zil_lwb_cache, lwb);
1943 mutex_exit(&zilog->zl_lock);
1946 static char *suspend_tag = "zil suspending";
1949 * Suspend an intent log. While in suspended mode, we still honor
1950 * synchronous semantics, but we rely on txg_wait_synced() to do it.
1951 * On old version pools, we suspend the log briefly when taking a
1952 * snapshot so that it will have an empty intent log.
1954 * Long holds are not really intended to be used the way we do here --
1955 * held for such a short time. A concurrent caller of dsl_dataset_long_held()
1956 * could fail. Therefore we take pains to only put a long hold if it is
1957 * actually necessary. Fortunately, it will only be necessary if the
1958 * objset is currently mounted (or the ZVOL equivalent). In that case it
1959 * will already have a long hold, so we are not really making things any worse.
1961 * Ideally, we would locate the existing long-holder (i.e. the zfsvfs_t or
1962 * zvol_state_t), and use their mechanism to prevent their hold from being
1963 * dropped (e.g. VFS_HOLD()). However, that would be even more pain for
1966 * if cookiep == NULL, this does both the suspend & resume.
1967 * Otherwise, it returns with the dataset "long held", and the cookie
1968 * should be passed into zil_resume().
1971 zil_suspend(const char *osname, void **cookiep)
1975 const zil_header_t *zh;
1978 error = dmu_objset_hold(osname, suspend_tag, &os);
1981 zilog = dmu_objset_zil(os);
1983 mutex_enter(&zilog->zl_lock);
1984 zh = zilog->zl_header;
1986 if (zh->zh_flags & ZIL_REPLAY_NEEDED) { /* unplayed log */
1987 mutex_exit(&zilog->zl_lock);
1988 dmu_objset_rele(os, suspend_tag);
1989 return (SET_ERROR(EBUSY));
1993 * Don't put a long hold in the cases where we can avoid it. This
1994 * is when there is no cookie so we are doing a suspend & resume
1995 * (i.e. called from zil_vdev_offline()), and there's nothing to do
1996 * for the suspend because it's already suspended, or there's no ZIL.
1998 if (cookiep == NULL && !zilog->zl_suspending &&
1999 (zilog->zl_suspend > 0 || BP_IS_HOLE(&zh->zh_log))) {
2000 mutex_exit(&zilog->zl_lock);
2001 dmu_objset_rele(os, suspend_tag);
2005 dsl_dataset_long_hold(dmu_objset_ds(os), suspend_tag);
2006 dsl_pool_rele(dmu_objset_pool(os), suspend_tag);
2008 zilog->zl_suspend++;
2010 if (zilog->zl_suspend > 1) {
2012 * Someone else is already suspending it.
2013 * Just wait for them to finish.
2016 while (zilog->zl_suspending)
2017 cv_wait(&zilog->zl_cv_suspend, &zilog->zl_lock);
2018 mutex_exit(&zilog->zl_lock);
2020 if (cookiep == NULL)
2028 * If there is no pointer to an on-disk block, this ZIL must not
2029 * be active (e.g. filesystem not mounted), so there's nothing
2032 if (BP_IS_HOLE(&zh->zh_log)) {
2033 ASSERT(cookiep != NULL); /* fast path already handled */
2036 mutex_exit(&zilog->zl_lock);
2040 zilog->zl_suspending = B_TRUE;
2041 mutex_exit(&zilog->zl_lock);
2043 zil_commit(zilog, 0);
2045 zil_destroy(zilog, B_FALSE);
2047 mutex_enter(&zilog->zl_lock);
2048 zilog->zl_suspending = B_FALSE;
2049 cv_broadcast(&zilog->zl_cv_suspend);
2050 mutex_exit(&zilog->zl_lock);
2052 if (cookiep == NULL)
2060 zil_resume(void *cookie)
2062 objset_t *os = cookie;
2063 zilog_t *zilog = dmu_objset_zil(os);
2065 mutex_enter(&zilog->zl_lock);
2066 ASSERT(zilog->zl_suspend != 0);
2067 zilog->zl_suspend--;
2068 mutex_exit(&zilog->zl_lock);
2069 dsl_dataset_long_rele(dmu_objset_ds(os), suspend_tag);
2070 dsl_dataset_rele(dmu_objset_ds(os), suspend_tag);
2073 typedef struct zil_replay_arg {
2074 zil_replay_func_t *zr_replay;
2076 boolean_t zr_byteswap;
2081 zil_replay_error(zilog_t *zilog, lr_t *lr, int error)
2083 char name[ZFS_MAX_DATASET_NAME_LEN];
2085 zilog->zl_replaying_seq--; /* didn't actually replay this one */
2087 dmu_objset_name(zilog->zl_os, name);
2089 cmn_err(CE_WARN, "ZFS replay transaction error %d, "
2090 "dataset %s, seq 0x%llx, txtype %llu %s\n", error, name,
2091 (u_longlong_t)lr->lrc_seq,
2092 (u_longlong_t)(lr->lrc_txtype & ~TX_CI),
2093 (lr->lrc_txtype & TX_CI) ? "CI" : "");
2099 zil_replay_log_record(zilog_t *zilog, lr_t *lr, void *zra, uint64_t claim_txg)
2101 zil_replay_arg_t *zr = zra;
2102 const zil_header_t *zh = zilog->zl_header;
2103 uint64_t reclen = lr->lrc_reclen;
2104 uint64_t txtype = lr->lrc_txtype;
2107 zilog->zl_replaying_seq = lr->lrc_seq;
2109 if (lr->lrc_seq <= zh->zh_replay_seq) /* already replayed */
2112 if (lr->lrc_txg < claim_txg) /* already committed */
2115 /* Strip case-insensitive bit, still present in log record */
2118 if (txtype == 0 || txtype >= TX_MAX_TYPE)
2119 return (zil_replay_error(zilog, lr, EINVAL));
2122 * If this record type can be logged out of order, the object
2123 * (lr_foid) may no longer exist. That's legitimate, not an error.
2125 if (TX_OOO(txtype)) {
2126 error = dmu_object_info(zilog->zl_os,
2127 LR_FOID_GET_OBJ(((lr_ooo_t *)lr)->lr_foid), NULL);
2128 if (error == ENOENT || error == EEXIST)
2133 * Make a copy of the data so we can revise and extend it.
2135 bcopy(lr, zr->zr_lr, reclen);
2138 * If this is a TX_WRITE with a blkptr, suck in the data.
2140 if (txtype == TX_WRITE && reclen == sizeof (lr_write_t)) {
2141 error = zil_read_log_data(zilog, (lr_write_t *)lr,
2142 zr->zr_lr + reclen);
2144 return (zil_replay_error(zilog, lr, error));
2148 * The log block containing this lr may have been byteswapped
2149 * so that we can easily examine common fields like lrc_txtype.
2150 * However, the log is a mix of different record types, and only the
2151 * replay vectors know how to byteswap their records. Therefore, if
2152 * the lr was byteswapped, undo it before invoking the replay vector.
2154 if (zr->zr_byteswap)
2155 byteswap_uint64_array(zr->zr_lr, reclen);
2158 * We must now do two things atomically: replay this log record,
2159 * and update the log header sequence number to reflect the fact that
2160 * we did so. At the end of each replay function the sequence number
2161 * is updated if we are in replay mode.
2163 error = zr->zr_replay[txtype](zr->zr_arg, zr->zr_lr, zr->zr_byteswap);
2166 * The DMU's dnode layer doesn't see removes until the txg
2167 * commits, so a subsequent claim can spuriously fail with
2168 * EEXIST. So if we receive any error we try syncing out
2169 * any removes then retry the transaction. Note that we
2170 * specify B_FALSE for byteswap now, so we don't do it twice.
2172 txg_wait_synced(spa_get_dsl(zilog->zl_spa), 0);
2173 error = zr->zr_replay[txtype](zr->zr_arg, zr->zr_lr, B_FALSE);
2175 return (zil_replay_error(zilog, lr, error));
2182 zil_incr_blks(zilog_t *zilog, blkptr_t *bp, void *arg, uint64_t claim_txg)
2184 zilog->zl_replay_blks++;
2190 * If this dataset has a non-empty intent log, replay it and destroy it.
2193 zil_replay(objset_t *os, void *arg, zil_replay_func_t replay_func[TX_MAX_TYPE])
2195 zilog_t *zilog = dmu_objset_zil(os);
2196 const zil_header_t *zh = zilog->zl_header;
2197 zil_replay_arg_t zr;
2199 if ((zh->zh_flags & ZIL_REPLAY_NEEDED) == 0) {
2200 zil_destroy(zilog, B_TRUE);
2204 zr.zr_replay = replay_func;
2206 zr.zr_byteswap = BP_SHOULD_BYTESWAP(&zh->zh_log);
2207 zr.zr_lr = vmem_alloc(2 * SPA_MAXBLOCKSIZE, KM_SLEEP);
2210 * Wait for in-progress removes to sync before starting replay.
2212 txg_wait_synced(zilog->zl_dmu_pool, 0);
2214 zilog->zl_replay = B_TRUE;
2215 zilog->zl_replay_time = ddi_get_lbolt();
2216 ASSERT(zilog->zl_replay_blks == 0);
2217 (void) zil_parse(zilog, zil_incr_blks, zil_replay_log_record, &zr,
2219 vmem_free(zr.zr_lr, 2 * SPA_MAXBLOCKSIZE);
2221 zil_destroy(zilog, B_FALSE);
2222 txg_wait_synced(zilog->zl_dmu_pool, zilog->zl_destroy_txg);
2223 zilog->zl_replay = B_FALSE;
2227 zil_replaying(zilog_t *zilog, dmu_tx_t *tx)
2229 if (zilog->zl_sync == ZFS_SYNC_DISABLED)
2232 if (zilog->zl_replay) {
2233 dsl_dataset_dirty(dmu_objset_ds(zilog->zl_os), tx);
2234 zilog->zl_replayed_seq[dmu_tx_get_txg(tx) & TXG_MASK] =
2235 zilog->zl_replaying_seq;
2244 zil_vdev_offline(const char *osname, void *arg)
2248 error = zil_suspend(osname, NULL);
2250 return (SET_ERROR(EEXIST));
2254 #if defined(_KERNEL) && defined(HAVE_SPL)
2255 EXPORT_SYMBOL(zil_alloc);
2256 EXPORT_SYMBOL(zil_free);
2257 EXPORT_SYMBOL(zil_open);
2258 EXPORT_SYMBOL(zil_close);
2259 EXPORT_SYMBOL(zil_replay);
2260 EXPORT_SYMBOL(zil_replaying);
2261 EXPORT_SYMBOL(zil_destroy);
2262 EXPORT_SYMBOL(zil_destroy_sync);
2263 EXPORT_SYMBOL(zil_itx_create);
2264 EXPORT_SYMBOL(zil_itx_destroy);
2265 EXPORT_SYMBOL(zil_itx_assign);
2266 EXPORT_SYMBOL(zil_commit);
2267 EXPORT_SYMBOL(zil_vdev_offline);
2268 EXPORT_SYMBOL(zil_claim);
2269 EXPORT_SYMBOL(zil_check_log_chain);
2270 EXPORT_SYMBOL(zil_sync);
2271 EXPORT_SYMBOL(zil_clean);
2272 EXPORT_SYMBOL(zil_suspend);
2273 EXPORT_SYMBOL(zil_resume);
2274 EXPORT_SYMBOL(zil_add_block);
2275 EXPORT_SYMBOL(zil_bp_tree_add);
2276 EXPORT_SYMBOL(zil_set_sync);
2277 EXPORT_SYMBOL(zil_set_logbias);
2279 module_param(zil_replay_disable, int, 0644);
2280 MODULE_PARM_DESC(zil_replay_disable, "Disable intent logging replay");
2282 module_param(zfs_nocacheflush, int, 0644);
2283 MODULE_PARM_DESC(zfs_nocacheflush, "Disable cache flushes");
2285 module_param(zil_slog_limit, ulong, 0644);
2286 MODULE_PARM_DESC(zil_slog_limit, "Max commit bytes to separate log device");