1/*
2 * CDDL HEADER START
3 *
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.
7 *
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.
12 *
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]
18 *
19 * CDDL HEADER END
20 */
21
22/*
23 * Copyright (c) 2005, 2010, Oracle and/or its affiliates. All rights reserved.
24 * Copyright (c) 2012 by Delphix. All rights reserved.
25 */
26
27/*
28 * Virtual Device Labels
29 * ---------------------
30 *
31 * The vdev label serves several distinct purposes:
32 *
33 * 1. Uniquely identify this device as part of a ZFS pool and confirm its
34 * identity within the pool.
35 *
36 * 2. Verify that all the devices given in a configuration are present
37 * within the pool.
38 *
39 * 3. Determine the uberblock for the pool.
40 *
41 * 4. In case of an import operation, determine the configuration of the
42 * toplevel vdev of which it is a part.
43 *
44 * 5. If an import operation cannot find all the devices in the pool,
45 * provide enough information to the administrator to determine which
46 * devices are missing.
47 *
48 * It is important to note that while the kernel is responsible for writing the
49 * label, it only consumes the information in the first three cases. The
50 * latter information is only consumed in userland when determining the
51 * configuration to import a pool.
52 *
53 *
54 * Label Organization
55 * ------------------
56 *
57 * Before describing the contents of the label, it's important to understand how
58 * the labels are written and updated with respect to the uberblock.
59 *
60 * When the pool configuration is altered, either because it was newly created
61 * or a device was added, we want to update all the labels such that we can deal
62 * with fatal failure at any point. To this end, each disk has two labels which
63 * are updated before and after the uberblock is synced. Assuming we have
64 * labels and an uberblock with the following transaction groups:
65 *
66 * L1 UB L2
67 * +------+ +------+ +------+
68 * | | | | | |
69 * | t10 | | t10 | | t10 |
70 * | | | | | |
71 * +------+ +------+ +------+
72 *
73 * In this stable state, the labels and the uberblock were all updated within
74 * the same transaction group (10). Each label is mirrored and checksummed, so
75 * that we can detect when we fail partway through writing the label.
76 *
77 * In order to identify which labels are valid, the labels are written in the
78 * following manner:
79 *
80 * 1. For each vdev, update 'L1' to the new label
81 * 2. Update the uberblock
82 * 3. For each vdev, update 'L2' to the new label
83 *
84 * Given arbitrary failure, we can determine the correct label to use based on
85 * the transaction group. If we fail after updating L1 but before updating the
86 * UB, we will notice that L1's transaction group is greater than the uberblock,
87 * so L2 must be valid. If we fail after writing the uberblock but before
88 * writing L2, we will notice that L2's transaction group is less than L1, and
89 * therefore L1 is valid.
90 *
91 * Another added complexity is that not every label is updated when the config
92 * is synced. If we add a single device, we do not want to have to re-write
93 * every label for every device in the pool. This means that both L1 and L2 may
94 * be older than the pool uberblock, because the necessary information is stored
95 * on another vdev.
96 *
97 *
98 * On-disk Format
99 * --------------
100 *
101 * The vdev label consists of two distinct parts, and is wrapped within the
102 * vdev_label_t structure. The label includes 8k of padding to permit legacy
103 * VTOC disk labels, but is otherwise ignored.
104 *
105 * The first half of the label is a packed nvlist which contains pool wide
106 * properties, per-vdev properties, and configuration information. It is
107 * described in more detail below.
108 *
109 * The latter half of the label consists of a redundant array of uberblocks.
110 * These uberblocks are updated whenever a transaction group is committed,
111 * or when the configuration is updated. When a pool is loaded, we scan each
112 * vdev for the 'best' uberblock.
113 *
114 *
115 * Configuration Information
116 * -------------------------
117 *
118 * The nvlist describing the pool and vdev contains the following elements:
119 *
120 * version ZFS on-disk version
121 * name Pool name
122 * state Pool state
123 * txg Transaction group in which this label was written
124 * pool_guid Unique identifier for this pool
125 * vdev_tree An nvlist describing vdev tree.
126 * features_for_read
127 * An nvlist of the features necessary for reading the MOS.
128 *
129 * Each leaf device label also contains the following:
130 *
131 * top_guid Unique ID for top-level vdev in which this is contained
132 * guid Unique ID for the leaf vdev
133 *
134 * The 'vs' configuration follows the format described in 'spa_config.c'.
135 */
136
137#include <sys/zfs_context.h>
138#include <sys/spa.h>
139#include <sys/spa_impl.h>
140#include <sys/dmu.h>
141#include <sys/zap.h>
142#include <sys/vdev.h>
143#include <sys/vdev_impl.h>
144#include <sys/uberblock_impl.h>
145#include <sys/metaslab.h>
146#include <sys/zio.h>
147#include <sys/dsl_scan.h>
148#include <sys/fs/zfs.h>
149
150/*
151 * Basic routines to read and write from a vdev label.
152 * Used throughout the rest of this file.
153 */
154uint64_t
155vdev_label_offset(uint64_t psize, int l, uint64_t offset)
156{
157 ASSERT(offset < sizeof (vdev_label_t));
158 ASSERT(P2PHASE_TYPED(psize, sizeof (vdev_label_t), uint64_t) == 0);
159
160 return (offset + l * sizeof (vdev_label_t) + (l < VDEV_LABELS / 2 ?
161 0 : psize - VDEV_LABELS * sizeof (vdev_label_t)));
162}
163
164/*
165 * Returns back the vdev label associated with the passed in offset.
166 */
167int
168vdev_label_number(uint64_t psize, uint64_t offset)
169{
170 int l;
171
172 if (offset >= psize - VDEV_LABEL_END_SIZE) {
173 offset -= psize - VDEV_LABEL_END_SIZE;
174 offset += (VDEV_LABELS / 2) * sizeof (vdev_label_t);
175 }
176 l = offset / sizeof (vdev_label_t);
177 return (l < VDEV_LABELS ? l : -1);
178}
179
180static void
181vdev_label_read(zio_t *zio, vdev_t *vd, int l, void *buf, uint64_t offset,
182 uint64_t size, zio_done_func_t *done, void *private, int flags)
183{
184 ASSERT(spa_config_held(zio->io_spa, SCL_STATE_ALL, RW_WRITER) ==
185 SCL_STATE_ALL);
186 ASSERT(flags & ZIO_FLAG_CONFIG_WRITER);
187
188 zio_nowait(zio_read_phys(zio, vd,
189 vdev_label_offset(vd->vdev_psize, l, offset),
190 size, buf, ZIO_CHECKSUM_LABEL, done, private,
191 ZIO_PRIORITY_SYNC_READ, flags, B_TRUE));
192}
193
194static void
195vdev_label_write(zio_t *zio, vdev_t *vd, int l, void *buf, uint64_t offset,
196 uint64_t size, zio_done_func_t *done, void *private, int flags)
197{
198 ASSERT(spa_config_held(zio->io_spa, SCL_ALL, RW_WRITER) == SCL_ALL ||
199 (spa_config_held(zio->io_spa, SCL_CONFIG | SCL_STATE, RW_READER) ==
200 (SCL_CONFIG | SCL_STATE) &&
201 dsl_pool_sync_context(spa_get_dsl(zio->io_spa))));
202 ASSERT(flags & ZIO_FLAG_CONFIG_WRITER);
203
204 zio_nowait(zio_write_phys(zio, vd,
205 vdev_label_offset(vd->vdev_psize, l, offset),
206 size, buf, ZIO_CHECKSUM_LABEL, done, private,
207 ZIO_PRIORITY_SYNC_WRITE, flags, B_TRUE));
208}
209
210/*
211 * Generate the nvlist representing this vdev's config.
212 */
213nvlist_t *
214vdev_config_generate(spa_t *spa, vdev_t *vd, boolean_t getstats,
215 vdev_config_flag_t flags)
216{
217 nvlist_t *nv = NULL;
218
219 VERIFY(nvlist_alloc(&nv, NV_UNIQUE_NAME, KM_SLEEP) == 0);
220
221 VERIFY(nvlist_add_string(nv, ZPOOL_CONFIG_TYPE,
222 vd->vdev_ops->vdev_op_type) == 0);
223 if (!(flags & (VDEV_CONFIG_SPARE | VDEV_CONFIG_L2CACHE)))
224 VERIFY(nvlist_add_uint64(nv, ZPOOL_CONFIG_ID, vd->vdev_id)
225 == 0);
226 VERIFY(nvlist_add_uint64(nv, ZPOOL_CONFIG_GUID, vd->vdev_guid) == 0);
227
228 if (vd->vdev_path != NULL)
229 VERIFY(nvlist_add_string(nv, ZPOOL_CONFIG_PATH,
230 vd->vdev_path) == 0);
231
232 if (vd->vdev_devid != NULL)
233 VERIFY(nvlist_add_string(nv, ZPOOL_CONFIG_DEVID,
234 vd->vdev_devid) == 0);
235
236 if (vd->vdev_physpath != NULL)
237 VERIFY(nvlist_add_string(nv, ZPOOL_CONFIG_PHYS_PATH,
238 vd->vdev_physpath) == 0);
239
240 if (vd->vdev_fru != NULL)
241 VERIFY(nvlist_add_string(nv, ZPOOL_CONFIG_FRU,
242 vd->vdev_fru) == 0);
243
244 if (vd->vdev_nparity != 0) {
245 ASSERT(strcmp(vd->vdev_ops->vdev_op_type,
246 VDEV_TYPE_RAIDZ) == 0);
247
248 /*
249 * Make sure someone hasn't managed to sneak a fancy new vdev
250 * into a crufty old storage pool.
251 */
252 ASSERT(vd->vdev_nparity == 1 ||
253 (vd->vdev_nparity <= 2 &&
254 spa_version(spa) >= SPA_VERSION_RAIDZ2) ||
255 (vd->vdev_nparity <= 3 &&
256 spa_version(spa) >= SPA_VERSION_RAIDZ3));
257
258 /*
259 * Note that we'll add the nparity tag even on storage pools
260 * that only support a single parity device -- older software
261 * will just ignore it.
262 */
263 VERIFY(nvlist_add_uint64(nv, ZPOOL_CONFIG_NPARITY,
264 vd->vdev_nparity) == 0);
265 }
266
267 if (vd->vdev_wholedisk != -1ULL)
268 VERIFY(nvlist_add_uint64(nv, ZPOOL_CONFIG_WHOLE_DISK,
269 vd->vdev_wholedisk) == 0);
270
271 if (vd->vdev_not_present)
272 VERIFY(nvlist_add_uint64(nv, ZPOOL_CONFIG_NOT_PRESENT, 1) == 0);
273
274 if (vd->vdev_isspare)
275 VERIFY(nvlist_add_uint64(nv, ZPOOL_CONFIG_IS_SPARE, 1) == 0);
276
277 if (!(flags & (VDEV_CONFIG_SPARE | VDEV_CONFIG_L2CACHE)) &&
278 vd == vd->vdev_top) {
279 VERIFY(nvlist_add_uint64(nv, ZPOOL_CONFIG_METASLAB_ARRAY,
280 vd->vdev_ms_array) == 0);
281 VERIFY(nvlist_add_uint64(nv, ZPOOL_CONFIG_METASLAB_SHIFT,
282 vd->vdev_ms_shift) == 0);
283 VERIFY(nvlist_add_uint64(nv, ZPOOL_CONFIG_ASHIFT,
284 vd->vdev_ashift) == 0);
285 VERIFY(nvlist_add_uint64(nv, ZPOOL_CONFIG_ASIZE,
286 vd->vdev_asize) == 0);
287 VERIFY(nvlist_add_uint64(nv, ZPOOL_CONFIG_IS_LOG,
288 vd->vdev_islog) == 0);
289 if (vd->vdev_removing)
290 VERIFY(nvlist_add_uint64(nv, ZPOOL_CONFIG_REMOVING,
291 vd->vdev_removing) == 0);
292 }
293
294 if (vd->vdev_dtl_smo.smo_object != 0)
295 VERIFY(nvlist_add_uint64(nv, ZPOOL_CONFIG_DTL,
296 vd->vdev_dtl_smo.smo_object) == 0);
297
298 if (vd->vdev_crtxg)
299 VERIFY(nvlist_add_uint64(nv, ZPOOL_CONFIG_CREATE_TXG,
300 vd->vdev_crtxg) == 0);
301
302 if (getstats) {
303 vdev_stat_t vs;
304 pool_scan_stat_t ps;
305
306 vdev_get_stats(vd, &vs);
307 VERIFY(nvlist_add_uint64_array(nv, ZPOOL_CONFIG_VDEV_STATS,
308 (uint64_t *)&vs, sizeof (vs) / sizeof (uint64_t)) == 0);
309
310 /* provide either current or previous scan information */
311 if (spa_scan_get_stats(spa, &ps) == 0) {
312 VERIFY(nvlist_add_uint64_array(nv,
313 ZPOOL_CONFIG_SCAN_STATS, (uint64_t *)&ps,
314 sizeof (pool_scan_stat_t) / sizeof (uint64_t))
315 == 0);
316 }
317 }
318
319 if (!vd->vdev_ops->vdev_op_leaf) {
320 nvlist_t **child;
321 int c, idx;
322
323 ASSERT(!vd->vdev_ishole);
324
325 child = kmem_alloc(vd->vdev_children * sizeof (nvlist_t *),
326 KM_SLEEP);
327
328 for (c = 0, idx = 0; c < vd->vdev_children; c++) {
329 vdev_t *cvd = vd->vdev_child[c];
330
331 /*
332 * If we're generating an nvlist of removing
333 * vdevs then skip over any device which is
334 * not being removed.
335 */
336 if ((flags & VDEV_CONFIG_REMOVING) &&
337 !cvd->vdev_removing)
338 continue;
339
340 child[idx++] = vdev_config_generate(spa, cvd,
341 getstats, flags);
342 }
343
344 if (idx) {
345 VERIFY(nvlist_add_nvlist_array(nv,
346 ZPOOL_CONFIG_CHILDREN, child, idx) == 0);
347 }
348
349 for (c = 0; c < idx; c++)
350 nvlist_free(child[c]);
351
352 kmem_free(child, vd->vdev_children * sizeof (nvlist_t *));
353
354 } else {
355 const char *aux = NULL;
356
357 if (vd->vdev_offline && !vd->vdev_tmpoffline)
358 VERIFY(nvlist_add_uint64(nv, ZPOOL_CONFIG_OFFLINE,
359 B_TRUE) == 0);
360 if (vd->vdev_resilvering)
361 VERIFY(nvlist_add_uint64(nv, ZPOOL_CONFIG_RESILVERING,
362 B_TRUE) == 0);
363 if (vd->vdev_faulted)
364 VERIFY(nvlist_add_uint64(nv, ZPOOL_CONFIG_FAULTED,
365 B_TRUE) == 0);
366 if (vd->vdev_degraded)
367 VERIFY(nvlist_add_uint64(nv, ZPOOL_CONFIG_DEGRADED,
368 B_TRUE) == 0);
369 if (vd->vdev_removed)
370 VERIFY(nvlist_add_uint64(nv, ZPOOL_CONFIG_REMOVED,
371 B_TRUE) == 0);
372 if (vd->vdev_unspare)
373 VERIFY(nvlist_add_uint64(nv, ZPOOL_CONFIG_UNSPARE,
374 B_TRUE) == 0);
375 if (vd->vdev_ishole)
376 VERIFY(nvlist_add_uint64(nv, ZPOOL_CONFIG_IS_HOLE,
377 B_TRUE) == 0);
378
379 switch (vd->vdev_stat.vs_aux) {
380 case VDEV_AUX_ERR_EXCEEDED:
381 aux = "err_exceeded";
382 break;
383
384 case VDEV_AUX_EXTERNAL:
385 aux = "external";
386 break;
387 }
388
389 if (aux != NULL)
390 VERIFY(nvlist_add_string(nv, ZPOOL_CONFIG_AUX_STATE,
391 aux) == 0);
392
393 if (vd->vdev_splitting && vd->vdev_orig_guid != 0LL) {
394 VERIFY(nvlist_add_uint64(nv, ZPOOL_CONFIG_ORIG_GUID,
395 vd->vdev_orig_guid) == 0);
396 }
397 }
398
399 return (nv);
400}
401
402/*
403 * Generate a view of the top-level vdevs. If we currently have holes
404 * in the namespace, then generate an array which contains a list of holey
405 * vdevs. Additionally, add the number of top-level children that currently
406 * exist.
407 */
408void
409vdev_top_config_generate(spa_t *spa, nvlist_t *config)
410{
411 vdev_t *rvd = spa->spa_root_vdev;
412 uint64_t *array;
413 uint_t c, idx;
414
415 array = kmem_alloc(rvd->vdev_children * sizeof (uint64_t), KM_SLEEP);
416
417 for (c = 0, idx = 0; c < rvd->vdev_children; c++) {
418 vdev_t *tvd = rvd->vdev_child[c];
419
420 if (tvd->vdev_ishole)
421 array[idx++] = c;
422 }
423
424 if (idx) {
425 VERIFY(nvlist_add_uint64_array(config, ZPOOL_CONFIG_HOLE_ARRAY,
426 array, idx) == 0);
427 }
428
429 VERIFY(nvlist_add_uint64(config, ZPOOL_CONFIG_VDEV_CHILDREN,
430 rvd->vdev_children) == 0);
431
432 kmem_free(array, rvd->vdev_children * sizeof (uint64_t));
433}
434
435/*
436 * Returns the configuration from the label of the given vdev. For vdevs
437 * which don't have a txg value stored on their label (i.e. spares/cache)
438 * or have not been completely initialized (txg = 0) just return
439 * the configuration from the first valid label we find. Otherwise,
440 * find the most up-to-date label that does not exceed the specified
441 * 'txg' value.
442 */
443nvlist_t *
444vdev_label_read_config(vdev_t *vd, uint64_t txg)
445{
446 spa_t *spa = vd->vdev_spa;
447 nvlist_t *config = NULL;
448 vdev_phys_t *vp;
449 zio_t *zio;
450 uint64_t best_txg = 0;
451 int error = 0;
452 int flags = ZIO_FLAG_CONFIG_WRITER | ZIO_FLAG_CANFAIL |
453 ZIO_FLAG_SPECULATIVE;
454
455 ASSERT(spa_config_held(spa, SCL_STATE_ALL, RW_WRITER) == SCL_STATE_ALL);
456
457 if (!vdev_readable(vd))
458 return (NULL);
459
460 vp = zio_buf_alloc(sizeof (vdev_phys_t));
461
462retry:
463 for (int l = 0; l < VDEV_LABELS; l++) {
464 nvlist_t *label = NULL;
465
466 zio = zio_root(spa, NULL, NULL, flags);
467
468 vdev_label_read(zio, vd, l, vp,
469 offsetof(vdev_label_t, vl_vdev_phys),
470 sizeof (vdev_phys_t), NULL, NULL, flags);
471
472 if (zio_wait(zio) == 0 &&
473 nvlist_unpack(vp->vp_nvlist, sizeof (vp->vp_nvlist),
474 &label, 0) == 0) {
475 uint64_t label_txg = 0;
476
477 /*
478 * Auxiliary vdevs won't have txg values in their
479 * labels and newly added vdevs may not have been
480 * completely initialized so just return the
481 * configuration from the first valid label we
482 * encounter.
483 */
484 error = nvlist_lookup_uint64(label,
485 ZPOOL_CONFIG_POOL_TXG, &label_txg);
486 if ((error || label_txg == 0) && !config) {
487 config = label;
488 break;
489 } else if (label_txg <= txg && label_txg > best_txg) {
490 best_txg = label_txg;
491 nvlist_free(config);
492 config = fnvlist_dup(label);
493 }
494 }
495
496 if (label != NULL) {
497 nvlist_free(label);
498 label = NULL;
499 }
500 }
501
502 if (config == NULL && !(flags & ZIO_FLAG_TRYHARD)) {
503 flags |= ZIO_FLAG_TRYHARD;
504 goto retry;
505 }
506
507 zio_buf_free(vp, sizeof (vdev_phys_t));
508
509 return (config);
510}
511
512/*
513 * Determine if a device is in use. The 'spare_guid' parameter will be filled
514 * in with the device guid if this spare is active elsewhere on the system.
515 */
516static boolean_t
517vdev_inuse(vdev_t *vd, uint64_t crtxg, vdev_labeltype_t reason,
518 uint64_t *spare_guid, uint64_t *l2cache_guid)
519{
520 spa_t *spa = vd->vdev_spa;
521 uint64_t state, pool_guid, device_guid, txg, spare_pool;
522 uint64_t vdtxg = 0;
523 nvlist_t *label;
524
525 if (spare_guid)
526 *spare_guid = 0ULL;
527 if (l2cache_guid)
528 *l2cache_guid = 0ULL;
529
530 /*
531 * Read the label, if any, and perform some basic sanity checks.
532 */
533 if ((label = vdev_label_read_config(vd, -1ULL)) == NULL)
534 return (B_FALSE);
535
536 (void) nvlist_lookup_uint64(label, ZPOOL_CONFIG_CREATE_TXG,
537 &vdtxg);
538
539 if (nvlist_lookup_uint64(label, ZPOOL_CONFIG_POOL_STATE,
540 &state) != 0 ||
541 nvlist_lookup_uint64(label, ZPOOL_CONFIG_GUID,
542 &device_guid) != 0) {
543 nvlist_free(label);
544 return (B_FALSE);
545 }
546
547 if (state != POOL_STATE_SPARE && state != POOL_STATE_L2CACHE &&
548 (nvlist_lookup_uint64(label, ZPOOL_CONFIG_POOL_GUID,
549 &pool_guid) != 0 ||
550 nvlist_lookup_uint64(label, ZPOOL_CONFIG_POOL_TXG,
551 &txg) != 0)) {
552 nvlist_free(label);
553 return (B_FALSE);
554 }
555
556 nvlist_free(label);
557
558 /*
559 * Check to see if this device indeed belongs to the pool it claims to
560 * be a part of. The only way this is allowed is if the device is a hot
561 * spare (which we check for later on).
562 */
563 if (state != POOL_STATE_SPARE && state != POOL_STATE_L2CACHE &&
564 !spa_guid_exists(pool_guid, device_guid) &&
565 !spa_spare_exists(device_guid, NULL, NULL) &&
566 !spa_l2cache_exists(device_guid, NULL))
567 return (B_FALSE);
568
569 /*
570 * If the transaction group is zero, then this an initialized (but
571 * unused) label. This is only an error if the create transaction
572 * on-disk is the same as the one we're using now, in which case the
573 * user has attempted to add the same vdev multiple times in the same
574 * transaction.
575 */
576 if (state != POOL_STATE_SPARE && state != POOL_STATE_L2CACHE &&
577 txg == 0 && vdtxg == crtxg)
578 return (B_TRUE);
579
580 /*
581 * Check to see if this is a spare device. We do an explicit check for
582 * spa_has_spare() here because it may be on our pending list of spares
583 * to add. We also check if it is an l2cache device.
584 */
585 if (spa_spare_exists(device_guid, &spare_pool, NULL) ||
586 spa_has_spare(spa, device_guid)) {
587 if (spare_guid)
588 *spare_guid = device_guid;
589
590 switch (reason) {
591 case VDEV_LABEL_CREATE:
592 case VDEV_LABEL_L2CACHE:
593 return (B_TRUE);
594
595 case VDEV_LABEL_REPLACE:
596 return (!spa_has_spare(spa, device_guid) ||
597 spare_pool != 0ULL);
598
599 case VDEV_LABEL_SPARE:
600 return (spa_has_spare(spa, device_guid));
601 }
602 }
603
604 /*
605 * Check to see if this is an l2cache device.
606 */
607 if (spa_l2cache_exists(device_guid, NULL))
608 return (B_TRUE);
609
610 /*
611 * We can't rely on a pool's state if it's been imported
612 * read-only. Instead we look to see if the pools is marked
613 * read-only in the namespace and set the state to active.
614 */
615 if ((spa = spa_by_guid(pool_guid, device_guid)) != NULL &&
616 spa_mode(spa) == FREAD)
617 state = POOL_STATE_ACTIVE;
618
619 /*
620 * If the device is marked ACTIVE, then this device is in use by another
621 * pool on the system.
622 */
623 return (state == POOL_STATE_ACTIVE);
624}
625
626/*
627 * Initialize a vdev label. We check to make sure each leaf device is not in
628 * use, and writable. We put down an initial label which we will later
629 * overwrite with a complete label. Note that it's important to do this
630 * sequentially, not in parallel, so that we catch cases of multiple use of the
631 * same leaf vdev in the vdev we're creating -- e.g. mirroring a disk with
632 * itself.
633 */
634int
635vdev_label_init(vdev_t *vd, uint64_t crtxg, vdev_labeltype_t reason)
636{
637 spa_t *spa = vd->vdev_spa;
638 nvlist_t *label;
639 vdev_phys_t *vp;
640 char *pad2;
641 uberblock_t *ub;
642 zio_t *zio;
643 char *buf;
644 size_t buflen;
645 int error;
646 uint64_t spare_guid, l2cache_guid;
647 int flags = ZIO_FLAG_CONFIG_WRITER | ZIO_FLAG_CANFAIL;
648
649 ASSERT(spa_config_held(spa, SCL_ALL, RW_WRITER) == SCL_ALL);
650
651 for (int c = 0; c < vd->vdev_children; c++)
652 if ((error = vdev_label_init(vd->vdev_child[c],
653 crtxg, reason)) != 0)
654 return (error);
655
656 /* Track the creation time for this vdev */
657 vd->vdev_crtxg = crtxg;
658
659 if (!vd->vdev_ops->vdev_op_leaf)
660 return (0);
661
662 /*
663 * Dead vdevs cannot be initialized.
664 */
665 if (vdev_is_dead(vd))
666 return (EIO);
667
668 /*
669 * Determine if the vdev is in use.
670 */
671 if (reason != VDEV_LABEL_REMOVE && reason != VDEV_LABEL_SPLIT &&
672 vdev_inuse(vd, crtxg, reason, &spare_guid, &l2cache_guid))
673 return (EBUSY);
674
675 /*
676 * If this is a request to add or replace a spare or l2cache device
677 * that is in use elsewhere on the system, then we must update the
678 * guid (which was initialized to a random value) to reflect the
679 * actual GUID (which is shared between multiple pools).
680 */
681 if (reason != VDEV_LABEL_REMOVE && reason != VDEV_LABEL_L2CACHE &&
682 spare_guid != 0ULL) {
683 uint64_t guid_delta = spare_guid - vd->vdev_guid;
684
685 vd->vdev_guid += guid_delta;
686
687 for (vdev_t *pvd = vd; pvd != NULL; pvd = pvd->vdev_parent)
688 pvd->vdev_guid_sum += guid_delta;
689
690 /*
691 * If this is a replacement, then we want to fallthrough to the
692 * rest of the code. If we're adding a spare, then it's already
693 * labeled appropriately and we can just return.
694 */
695 if (reason == VDEV_LABEL_SPARE)
696 return (0);
697 ASSERT(reason == VDEV_LABEL_REPLACE ||
698 reason == VDEV_LABEL_SPLIT);
699 }
700
701 if (reason != VDEV_LABEL_REMOVE && reason != VDEV_LABEL_SPARE &&
702 l2cache_guid != 0ULL) {
703 uint64_t guid_delta = l2cache_guid - vd->vdev_guid;
704
705 vd->vdev_guid += guid_delta;
706
707 for (vdev_t *pvd = vd; pvd != NULL; pvd = pvd->vdev_parent)
708 pvd->vdev_guid_sum += guid_delta;
709
710 /*
711 * If this is a replacement, then we want to fallthrough to the
712 * rest of the code. If we're adding an l2cache, then it's
713 * already labeled appropriately and we can just return.
714 */
715 if (reason == VDEV_LABEL_L2CACHE)
716 return (0);
717 ASSERT(reason == VDEV_LABEL_REPLACE);
718 }
719
720 /*
721 * Initialize its label.
722 */
723 vp = zio_buf_alloc(sizeof (vdev_phys_t));
724 bzero(vp, sizeof (vdev_phys_t));
725
726 /*
727 * Generate a label describing the pool and our top-level vdev.
728 * We mark it as being from txg 0 to indicate that it's not
729 * really part of an active pool just yet. The labels will
730 * be written again with a meaningful txg by spa_sync().
731 */
732 if (reason == VDEV_LABEL_SPARE ||
733 (reason == VDEV_LABEL_REMOVE && vd->vdev_isspare)) {
734 /*
735 * For inactive hot spares, we generate a special label that
736 * identifies as a mutually shared hot spare. We write the
737 * label if we are adding a hot spare, or if we are removing an
738 * active hot spare (in which case we want to revert the
739 * labels).
740 */
741 VERIFY(nvlist_alloc(&label, NV_UNIQUE_NAME, KM_SLEEP) == 0);
742
743 VERIFY(nvlist_add_uint64(label, ZPOOL_CONFIG_VERSION,
744 spa_version(spa)) == 0);
745 VERIFY(nvlist_add_uint64(label, ZPOOL_CONFIG_POOL_STATE,
746 POOL_STATE_SPARE) == 0);
747 VERIFY(nvlist_add_uint64(label, ZPOOL_CONFIG_GUID,
748 vd->vdev_guid) == 0);
749 } else if (reason == VDEV_LABEL_L2CACHE ||
750 (reason == VDEV_LABEL_REMOVE && vd->vdev_isl2cache)) {
751 /*
752 * For level 2 ARC devices, add a special label.
753 */
754 VERIFY(nvlist_alloc(&label, NV_UNIQUE_NAME, KM_SLEEP) == 0);
755
756 VERIFY(nvlist_add_uint64(label, ZPOOL_CONFIG_VERSION,
757 spa_version(spa)) == 0);
758 VERIFY(nvlist_add_uint64(label, ZPOOL_CONFIG_POOL_STATE,
759 POOL_STATE_L2CACHE) == 0);
760 VERIFY(nvlist_add_uint64(label, ZPOOL_CONFIG_GUID,
761 vd->vdev_guid) == 0);
762 } else {
763 uint64_t txg = 0ULL;
764
765 if (reason == VDEV_LABEL_SPLIT)
766 txg = spa->spa_uberblock.ub_txg;
767 label = spa_config_generate(spa, vd, txg, B_FALSE);
768
769 /*
770 * Add our creation time. This allows us to detect multiple
771 * vdev uses as described above, and automatically expires if we
772 * fail.
773 */
774 VERIFY(nvlist_add_uint64(label, ZPOOL_CONFIG_CREATE_TXG,
775 crtxg) == 0);
776 }
777
778 buf = vp->vp_nvlist;
779 buflen = sizeof (vp->vp_nvlist);
780
781 error = nvlist_pack(label, &buf, &buflen, NV_ENCODE_XDR, KM_SLEEP);
782 if (error != 0) {
783 nvlist_free(label);
784 zio_buf_free(vp, sizeof (vdev_phys_t));
785 /* EFAULT means nvlist_pack ran out of room */
786 return (error == EFAULT ? ENAMETOOLONG : EINVAL);
787 }
788
789 /*
790 * Initialize uberblock template.
791 */
792 ub = zio_buf_alloc(VDEV_UBERBLOCK_RING);
793 bzero(ub, VDEV_UBERBLOCK_RING);
794 *ub = spa->spa_uberblock;
795 ub->ub_txg = 0;
796
797 /* Initialize the 2nd padding area. */
798 pad2 = zio_buf_alloc(VDEV_PAD_SIZE);
799 bzero(pad2, VDEV_PAD_SIZE);
800
801 /*
802 * Write everything in parallel.
803 */
804retry:
805 zio = zio_root(spa, NULL, NULL, flags);
806
807 for (int l = 0; l < VDEV_LABELS; l++) {
808
809 vdev_label_write(zio, vd, l, vp,
810 offsetof(vdev_label_t, vl_vdev_phys),
811 sizeof (vdev_phys_t), NULL, NULL, flags);
812
813 /*
814 * Skip the 1st padding area.
815 * Zero out the 2nd padding area where it might have
816 * left over data from previous filesystem format.
817 */
818 vdev_label_write(zio, vd, l, pad2,
819 offsetof(vdev_label_t, vl_pad2),
820 VDEV_PAD_SIZE, NULL, NULL, flags);
821
822 vdev_label_write(zio, vd, l, ub,
823 offsetof(vdev_label_t, vl_uberblock),
824 VDEV_UBERBLOCK_RING, NULL, NULL, flags);
825 }
826
827 error = zio_wait(zio);
828
829 if (error != 0 && !(flags & ZIO_FLAG_TRYHARD)) {
830 flags |= ZIO_FLAG_TRYHARD;
831 goto retry;
832 }
833
834 nvlist_free(label);
835 zio_buf_free(pad2, VDEV_PAD_SIZE);
836 zio_buf_free(ub, VDEV_UBERBLOCK_RING);
837 zio_buf_free(vp, sizeof (vdev_phys_t));
838
839 /*
840 * If this vdev hasn't been previously identified as a spare, then we
841 * mark it as such only if a) we are labeling it as a spare, or b) it
842 * exists as a spare elsewhere in the system. Do the same for
843 * level 2 ARC devices.
844 */
845 if (error == 0 && !vd->vdev_isspare &&
846 (reason == VDEV_LABEL_SPARE ||
847 spa_spare_exists(vd->vdev_guid, NULL, NULL)))
848 spa_spare_add(vd);
849
850 if (error == 0 && !vd->vdev_isl2cache &&
851 (reason == VDEV_LABEL_L2CACHE ||
852 spa_l2cache_exists(vd->vdev_guid, NULL)))
853 spa_l2cache_add(vd);
854
855 return (error);
856}
857
858/*
859 * ==========================================================================
860 * uberblock load/sync
861 * ==========================================================================
862 */
863
864/*
865 * Consider the following situation: txg is safely synced to disk. We've
866 * written the first uberblock for txg + 1, and then we lose power. When we
867 * come back up, we fail to see the uberblock for txg + 1 because, say,
868 * it was on a mirrored device and the replica to which we wrote txg + 1
869 * is now offline. If we then make some changes and sync txg + 1, and then
870 * the missing replica comes back, then for a few seconds we'll have two
871 * conflicting uberblocks on disk with the same txg. The solution is simple:
872 * among uberblocks with equal txg, choose the one with the latest timestamp.
873 */
874static int
875vdev_uberblock_compare(uberblock_t *ub1, uberblock_t *ub2)
876{
877 if (ub1->ub_txg < ub2->ub_txg)
878 return (-1);
879 if (ub1->ub_txg > ub2->ub_txg)
880 return (1);
881
882 if (ub1->ub_timestamp < ub2->ub_timestamp)
883 return (-1);
884 if (ub1->ub_timestamp > ub2->ub_timestamp)
885 return (1);
886
887 return (0);
888}
889
890struct ubl_cbdata {
891 uberblock_t *ubl_ubbest; /* Best uberblock */
892 vdev_t *ubl_vd; /* vdev associated with the above */
893};
894
895static void
896vdev_uberblock_load_done(zio_t *zio)
897{
898 vdev_t *vd = zio->io_vd;
899 spa_t *spa = zio->io_spa;
900 zio_t *rio = zio->io_private;
901 uberblock_t *ub = zio->io_data;
902 struct ubl_cbdata *cbp = rio->io_private;
903
904 ASSERT3U(zio->io_size, ==, VDEV_UBERBLOCK_SIZE(vd));
905
906 if (zio->io_error == 0 && uberblock_verify(ub) == 0) {
907 mutex_enter(&rio->io_lock);
908 if (ub->ub_txg <= spa->spa_load_max_txg &&
909 vdev_uberblock_compare(ub, cbp->ubl_ubbest) > 0) {
910 /*
911 * Keep track of the vdev in which this uberblock
912 * was found. We will use this information later
913 * to obtain the config nvlist associated with
914 * this uberblock.
915 */
916 *cbp->ubl_ubbest = *ub;
917 cbp->ubl_vd = vd;
918 }
919 mutex_exit(&rio->io_lock);
920 }
921
922 zio_buf_free(zio->io_data, zio->io_size);
923}
924
925static void
926vdev_uberblock_load_impl(zio_t *zio, vdev_t *vd, int flags,
927 struct ubl_cbdata *cbp)
928{
929 for (int c = 0; c < vd->vdev_children; c++)
930 vdev_uberblock_load_impl(zio, vd->vdev_child[c], flags, cbp);
931
932 if (vd->vdev_ops->vdev_op_leaf && vdev_readable(vd)) {
933 for (int l = 0; l < VDEV_LABELS; l++) {
934 for (int n = 0; n < VDEV_UBERBLOCK_COUNT(vd); n++) {
935 vdev_label_read(zio, vd, l,
936 zio_buf_alloc(VDEV_UBERBLOCK_SIZE(vd)),
937 VDEV_UBERBLOCK_OFFSET(vd, n),
938 VDEV_UBERBLOCK_SIZE(vd),
939 vdev_uberblock_load_done, zio, flags);
940 }
941 }
942 }
943}
944
945/*
946 * Reads the 'best' uberblock from disk along with its associated
947 * configuration. First, we read the uberblock array of each label of each
948 * vdev, keeping track of the uberblock with the highest txg in each array.
949 * Then, we read the configuration from the same vdev as the best uberblock.
950 */
951void
952vdev_uberblock_load(vdev_t *rvd, uberblock_t *ub, nvlist_t **config)
953{
954 zio_t *zio;
955 spa_t *spa = rvd->vdev_spa;
956 struct ubl_cbdata cb;
957 int flags = ZIO_FLAG_CONFIG_WRITER | ZIO_FLAG_CANFAIL |
958 ZIO_FLAG_SPECULATIVE | ZIO_FLAG_TRYHARD;
959
960 ASSERT(ub);
961 ASSERT(config);
962
963 bzero(ub, sizeof (uberblock_t));
964 *config = NULL;
965
966 cb.ubl_ubbest = ub;
967 cb.ubl_vd = NULL;
968
969 spa_config_enter(spa, SCL_ALL, FTAG, RW_WRITER);
970 zio = zio_root(spa, NULL, &cb, flags);
971 vdev_uberblock_load_impl(zio, rvd, flags, &cb);
972 (void) zio_wait(zio);
973
974 /*
975 * It's possible that the best uberblock was discovered on a label
976 * that has a configuration which was written in a future txg.
977 * Search all labels on this vdev to find the configuration that
978 * matches the txg for our uberblock.
979 */
980 if (cb.ubl_vd != NULL)
981 *config = vdev_label_read_config(cb.ubl_vd, ub->ub_txg);
982 spa_config_exit(spa, SCL_ALL, FTAG);
983}
984
985/*
986 * On success, increment root zio's count of good writes.
987 * We only get credit for writes to known-visible vdevs; see spa_vdev_add().
988 */
989static void
990vdev_uberblock_sync_done(zio_t *zio)
991{
992 uint64_t *good_writes = zio->io_private;
993
994 if (zio->io_error == 0 && zio->io_vd->vdev_top->vdev_ms_array != 0)
995 atomic_add_64(good_writes, 1);
996}
997
998/*
999 * Write the uberblock to all labels of all leaves of the specified vdev.
1000 */
1001static void
1002vdev_uberblock_sync(zio_t *zio, uberblock_t *ub, vdev_t *vd, int flags)
1003{
1004 uberblock_t *ubbuf;
1005 int n;
1006
1007 for (int c = 0; c < vd->vdev_children; c++)
1008 vdev_uberblock_sync(zio, ub, vd->vdev_child[c], flags);
1009
1010 if (!vd->vdev_ops->vdev_op_leaf)
1011 return;
1012
1013 if (!vdev_writeable(vd))
1014 return;
1015
1016 n = ub->ub_txg & (VDEV_UBERBLOCK_COUNT(vd) - 1);
1017
1018 ubbuf = zio_buf_alloc(VDEV_UBERBLOCK_SIZE(vd));
1019 bzero(ubbuf, VDEV_UBERBLOCK_SIZE(vd));
1020 *ubbuf = *ub;
1021
1022 for (int l = 0; l < VDEV_LABELS; l++)
1023 vdev_label_write(zio, vd, l, ubbuf,
1024 VDEV_UBERBLOCK_OFFSET(vd, n), VDEV_UBERBLOCK_SIZE(vd),
1025 vdev_uberblock_sync_done, zio->io_private,
1026 flags | ZIO_FLAG_DONT_PROPAGATE);
1027
1028 zio_buf_free(ubbuf, VDEV_UBERBLOCK_SIZE(vd));
1029}
1030
1031int
1032vdev_uberblock_sync_list(vdev_t **svd, int svdcount, uberblock_t *ub, int flags)
1033{
1034 spa_t *spa = svd[0]->vdev_spa;
1035 zio_t *zio;
1036 uint64_t good_writes = 0;
1037
1038 zio = zio_root(spa, NULL, &good_writes, flags);
1039
1040 for (int v = 0; v < svdcount; v++)
1041 vdev_uberblock_sync(zio, ub, svd[v], flags);
1042
1043 (void) zio_wait(zio);
1044
1045 /*
1046 * Flush the uberblocks to disk. This ensures that the odd labels
1047 * are no longer needed (because the new uberblocks and the even
1048 * labels are safely on disk), so it is safe to overwrite them.
1049 */
1050 zio = zio_root(spa, NULL, NULL, flags);
1051
1052 for (int v = 0; v < svdcount; v++)
1053 zio_flush(zio, svd[v]);
1054
1055 (void) zio_wait(zio);
1056
1057 return (good_writes >= 1 ? 0 : EIO);
1058}
1059
1060/*
1061 * On success, increment the count of good writes for our top-level vdev.
1062 */
1063static void
1064vdev_label_sync_done(zio_t *zio)
1065{
1066 uint64_t *good_writes = zio->io_private;
1067
1068 if (zio->io_error == 0)
1069 atomic_add_64(good_writes, 1);
1070}
1071
1072/*
1073 * If there weren't enough good writes, indicate failure to the parent.
1074 */
1075static void
1076vdev_label_sync_top_done(zio_t *zio)
1077{
1078 uint64_t *good_writes = zio->io_private;
1079
1080 if (*good_writes == 0)
1081 zio->io_error = EIO;
1082
1083 kmem_free(good_writes, sizeof (uint64_t));
1084}
1085
1086/*
1087 * We ignore errors for log and cache devices, simply free the private data.
1088 */
1089static void
1090vdev_label_sync_ignore_done(zio_t *zio)
1091{
1092 kmem_free(zio->io_private, sizeof (uint64_t));
1093}
1094
1095/*
1096 * Write all even or odd labels to all leaves of the specified vdev.
1097 */
1098static void
1099vdev_label_sync(zio_t *zio, vdev_t *vd, int l, uint64_t txg, int flags)
1100{
1101 nvlist_t *label;
1102 vdev_phys_t *vp;
1103 char *buf;
1104 size_t buflen;
1105
1106 for (int c = 0; c < vd->vdev_children; c++)
1107 vdev_label_sync(zio, vd->vdev_child[c], l, txg, flags);
1108
1109 if (!vd->vdev_ops->vdev_op_leaf)
1110 return;
1111
1112 if (!vdev_writeable(vd))
1113 return;
1114
1115 /*
1116 * Generate a label describing the top-level config to which we belong.
1117 */
1118 label = spa_config_generate(vd->vdev_spa, vd, txg, B_FALSE);
1119
1120 vp = zio_buf_alloc(sizeof (vdev_phys_t));
1121 bzero(vp, sizeof (vdev_phys_t));
1122
1123 buf = vp->vp_nvlist;
1124 buflen = sizeof (vp->vp_nvlist);
1125
1126 if (nvlist_pack(label, &buf, &buflen, NV_ENCODE_XDR, KM_SLEEP) == 0) {
1127 for (; l < VDEV_LABELS; l += 2) {
1128 vdev_label_write(zio, vd, l, vp,
1129 offsetof(vdev_label_t, vl_vdev_phys),
1130 sizeof (vdev_phys_t),
1131 vdev_label_sync_done, zio->io_private,
1132 flags | ZIO_FLAG_DONT_PROPAGATE);
1133 }
1134 }
1135
1136 zio_buf_free(vp, sizeof (vdev_phys_t));
1137 nvlist_free(label);
1138}
1139
1140int
1141vdev_label_sync_list(spa_t *spa, int l, uint64_t txg, int flags)
1142{
1143 list_t *dl = &spa->spa_config_dirty_list;
1144 vdev_t *vd;
1145 zio_t *zio;
1146 int error;
1147
1148 /*
1149 * Write the new labels to disk.
1150 */
1151 zio = zio_root(spa, NULL, NULL, flags);
1152
1153 for (vd = list_head(dl); vd != NULL; vd = list_next(dl, vd)) {
1154 uint64_t *good_writes = kmem_zalloc(sizeof (uint64_t),
1155 KM_SLEEP);
1156
1157 ASSERT(!vd->vdev_ishole);
1158
1159 zio_t *vio = zio_null(zio, spa, NULL,
1160 (vd->vdev_islog || vd->vdev_aux != NULL) ?
1161 vdev_label_sync_ignore_done : vdev_label_sync_top_done,
1162 good_writes, flags);
1163 vdev_label_sync(vio, vd, l, txg, flags);
1164 zio_nowait(vio);
1165 }
1166
1167 error = zio_wait(zio);
1168
1169 /*
1170 * Flush the new labels to disk.
1171 */
1172 zio = zio_root(spa, NULL, NULL, flags);
1173
1174 for (vd = list_head(dl); vd != NULL; vd = list_next(dl, vd))
1175 zio_flush(zio, vd);
1176
1177 (void) zio_wait(zio);
1178
1179 return (error);
1180}
1181
1182/*
1183 * Sync the uberblock and any changes to the vdev configuration.
1184 *
1185 * The order of operations is carefully crafted to ensure that
1186 * if the system panics or loses power at any time, the state on disk
1187 * is still transactionally consistent. The in-line comments below
1188 * describe the failure semantics at each stage.
1189 *
1190 * Moreover, vdev_config_sync() is designed to be idempotent: if it fails
1191 * at any time, you can just call it again, and it will resume its work.
1192 */
1193int
1194vdev_config_sync(vdev_t **svd, int svdcount, uint64_t txg, boolean_t tryhard)
1195{
1196 spa_t *spa = svd[0]->vdev_spa;
1197 uberblock_t *ub = &spa->spa_uberblock;
1198 vdev_t *vd;
1199 zio_t *zio;
1200 int error;
1201 int flags = ZIO_FLAG_CONFIG_WRITER | ZIO_FLAG_CANFAIL;
1202
1203 /*
1204 * Normally, we don't want to try too hard to write every label and
1205 * uberblock. If there is a flaky disk, we don't want the rest of the
1206 * sync process to block while we retry. But if we can't write a
1207 * single label out, we should retry with ZIO_FLAG_TRYHARD before
1208 * bailing out and declaring the pool faulted.
1209 */
1210 if (tryhard)
1211 flags |= ZIO_FLAG_TRYHARD;
1212
1213 ASSERT(ub->ub_txg <= txg);
1214
1215 /*
1216 * If this isn't a resync due to I/O errors,
1217 * and nothing changed in this transaction group,
1218 * and the vdev configuration hasn't changed,
1219 * then there's nothing to do.
1220 */
1221 if (ub->ub_txg < txg &&
1222 uberblock_update(ub, spa->spa_root_vdev, txg) == B_FALSE &&
1223 list_is_empty(&spa->spa_config_dirty_list))
1224 return (0);
1225
1226 if (txg > spa_freeze_txg(spa))
1227 return (0);
1228
1229 ASSERT(txg <= spa->spa_final_txg);
1230
1231 /*
1232 * Flush the write cache of every disk that's been written to
1233 * in this transaction group. This ensures that all blocks
1234 * written in this txg will be committed to stable storage
1235 * before any uberblock that references them.
1236 */
1237 zio = zio_root(spa, NULL, NULL, flags);
1238
1239 for (vd = txg_list_head(&spa->spa_vdev_txg_list, TXG_CLEAN(txg)); vd;
1240 vd = txg_list_next(&spa->spa_vdev_txg_list, vd, TXG_CLEAN(txg)))
1241 zio_flush(zio, vd);
1242
1243 (void) zio_wait(zio);
1244
1245 /*
1246 * Sync out the even labels (L0, L2) for every dirty vdev. If the
1247 * system dies in the middle of this process, that's OK: all of the
1248 * even labels that made it to disk will be newer than any uberblock,
1249 * and will therefore be considered invalid. The odd labels (L1, L3),
1250 * which have not yet been touched, will still be valid. We flush
1251 * the new labels to disk to ensure that all even-label updates
1252 * are committed to stable storage before the uberblock update.
1253 */
1254 if ((error = vdev_label_sync_list(spa, 0, txg, flags)) != 0)
1255 return (error);
1256
1257 /*
1258 * Sync the uberblocks to all vdevs in svd[].
1259 * If the system dies in the middle of this step, there are two cases
1260 * to consider, and the on-disk state is consistent either way:
1261 *
1262 * (1) If none of the new uberblocks made it to disk, then the
1263 * previous uberblock will be the newest, and the odd labels
1264 * (which had not yet been touched) will be valid with respect
1265 * to that uberblock.
1266 *
1267 * (2) If one or more new uberblocks made it to disk, then they
1268 * will be the newest, and the even labels (which had all
1269 * been successfully committed) will be valid with respect
1270 * to the new uberblocks.
1271 */
1272 if ((error = vdev_uberblock_sync_list(svd, svdcount, ub, flags)) != 0)
1273 return (error);
1274
1275 /*
1276 * Sync out odd labels for every dirty vdev. If the system dies
1277 * in the middle of this process, the even labels and the new
1278 * uberblocks will suffice to open the pool. The next time
1279 * the pool is opened, the first thing we'll do -- before any
1280 * user data is modified -- is mark every vdev dirty so that
1281 * all labels will be brought up to date. We flush the new labels
1282 * to disk to ensure that all odd-label updates are committed to
1283 * stable storage before the next transaction group begins.
1284 */
1285 return (vdev_label_sync_list(spa, 1, txg, flags));
1286}
2 * CDDL HEADER START
3 *
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.
7 *
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.
12 *
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]
18 *
19 * CDDL HEADER END
20 */
21
22/*
23 * Copyright (c) 2005, 2010, Oracle and/or its affiliates. All rights reserved.
24 * Copyright (c) 2012 by Delphix. All rights reserved.
25 */
26
27/*
28 * Virtual Device Labels
29 * ---------------------
30 *
31 * The vdev label serves several distinct purposes:
32 *
33 * 1. Uniquely identify this device as part of a ZFS pool and confirm its
34 * identity within the pool.
35 *
36 * 2. Verify that all the devices given in a configuration are present
37 * within the pool.
38 *
39 * 3. Determine the uberblock for the pool.
40 *
41 * 4. In case of an import operation, determine the configuration of the
42 * toplevel vdev of which it is a part.
43 *
44 * 5. If an import operation cannot find all the devices in the pool,
45 * provide enough information to the administrator to determine which
46 * devices are missing.
47 *
48 * It is important to note that while the kernel is responsible for writing the
49 * label, it only consumes the information in the first three cases. The
50 * latter information is only consumed in userland when determining the
51 * configuration to import a pool.
52 *
53 *
54 * Label Organization
55 * ------------------
56 *
57 * Before describing the contents of the label, it's important to understand how
58 * the labels are written and updated with respect to the uberblock.
59 *
60 * When the pool configuration is altered, either because it was newly created
61 * or a device was added, we want to update all the labels such that we can deal
62 * with fatal failure at any point. To this end, each disk has two labels which
63 * are updated before and after the uberblock is synced. Assuming we have
64 * labels and an uberblock with the following transaction groups:
65 *
66 * L1 UB L2
67 * +------+ +------+ +------+
68 * | | | | | |
69 * | t10 | | t10 | | t10 |
70 * | | | | | |
71 * +------+ +------+ +------+
72 *
73 * In this stable state, the labels and the uberblock were all updated within
74 * the same transaction group (10). Each label is mirrored and checksummed, so
75 * that we can detect when we fail partway through writing the label.
76 *
77 * In order to identify which labels are valid, the labels are written in the
78 * following manner:
79 *
80 * 1. For each vdev, update 'L1' to the new label
81 * 2. Update the uberblock
82 * 3. For each vdev, update 'L2' to the new label
83 *
84 * Given arbitrary failure, we can determine the correct label to use based on
85 * the transaction group. If we fail after updating L1 but before updating the
86 * UB, we will notice that L1's transaction group is greater than the uberblock,
87 * so L2 must be valid. If we fail after writing the uberblock but before
88 * writing L2, we will notice that L2's transaction group is less than L1, and
89 * therefore L1 is valid.
90 *
91 * Another added complexity is that not every label is updated when the config
92 * is synced. If we add a single device, we do not want to have to re-write
93 * every label for every device in the pool. This means that both L1 and L2 may
94 * be older than the pool uberblock, because the necessary information is stored
95 * on another vdev.
96 *
97 *
98 * On-disk Format
99 * --------------
100 *
101 * The vdev label consists of two distinct parts, and is wrapped within the
102 * vdev_label_t structure. The label includes 8k of padding to permit legacy
103 * VTOC disk labels, but is otherwise ignored.
104 *
105 * The first half of the label is a packed nvlist which contains pool wide
106 * properties, per-vdev properties, and configuration information. It is
107 * described in more detail below.
108 *
109 * The latter half of the label consists of a redundant array of uberblocks.
110 * These uberblocks are updated whenever a transaction group is committed,
111 * or when the configuration is updated. When a pool is loaded, we scan each
112 * vdev for the 'best' uberblock.
113 *
114 *
115 * Configuration Information
116 * -------------------------
117 *
118 * The nvlist describing the pool and vdev contains the following elements:
119 *
120 * version ZFS on-disk version
121 * name Pool name
122 * state Pool state
123 * txg Transaction group in which this label was written
124 * pool_guid Unique identifier for this pool
125 * vdev_tree An nvlist describing vdev tree.
126 * features_for_read
127 * An nvlist of the features necessary for reading the MOS.
128 *
129 * Each leaf device label also contains the following:
130 *
131 * top_guid Unique ID for top-level vdev in which this is contained
132 * guid Unique ID for the leaf vdev
133 *
134 * The 'vs' configuration follows the format described in 'spa_config.c'.
135 */
136
137#include <sys/zfs_context.h>
138#include <sys/spa.h>
139#include <sys/spa_impl.h>
140#include <sys/dmu.h>
141#include <sys/zap.h>
142#include <sys/vdev.h>
143#include <sys/vdev_impl.h>
144#include <sys/uberblock_impl.h>
145#include <sys/metaslab.h>
146#include <sys/zio.h>
147#include <sys/dsl_scan.h>
148#include <sys/fs/zfs.h>
149
150/*
151 * Basic routines to read and write from a vdev label.
152 * Used throughout the rest of this file.
153 */
154uint64_t
155vdev_label_offset(uint64_t psize, int l, uint64_t offset)
156{
157 ASSERT(offset < sizeof (vdev_label_t));
158 ASSERT(P2PHASE_TYPED(psize, sizeof (vdev_label_t), uint64_t) == 0);
159
160 return (offset + l * sizeof (vdev_label_t) + (l < VDEV_LABELS / 2 ?
161 0 : psize - VDEV_LABELS * sizeof (vdev_label_t)));
162}
163
164/*
165 * Returns back the vdev label associated with the passed in offset.
166 */
167int
168vdev_label_number(uint64_t psize, uint64_t offset)
169{
170 int l;
171
172 if (offset >= psize - VDEV_LABEL_END_SIZE) {
173 offset -= psize - VDEV_LABEL_END_SIZE;
174 offset += (VDEV_LABELS / 2) * sizeof (vdev_label_t);
175 }
176 l = offset / sizeof (vdev_label_t);
177 return (l < VDEV_LABELS ? l : -1);
178}
179
180static void
181vdev_label_read(zio_t *zio, vdev_t *vd, int l, void *buf, uint64_t offset,
182 uint64_t size, zio_done_func_t *done, void *private, int flags)
183{
184 ASSERT(spa_config_held(zio->io_spa, SCL_STATE_ALL, RW_WRITER) ==
185 SCL_STATE_ALL);
186 ASSERT(flags & ZIO_FLAG_CONFIG_WRITER);
187
188 zio_nowait(zio_read_phys(zio, vd,
189 vdev_label_offset(vd->vdev_psize, l, offset),
190 size, buf, ZIO_CHECKSUM_LABEL, done, private,
191 ZIO_PRIORITY_SYNC_READ, flags, B_TRUE));
192}
193
194static void
195vdev_label_write(zio_t *zio, vdev_t *vd, int l, void *buf, uint64_t offset,
196 uint64_t size, zio_done_func_t *done, void *private, int flags)
197{
198 ASSERT(spa_config_held(zio->io_spa, SCL_ALL, RW_WRITER) == SCL_ALL ||
199 (spa_config_held(zio->io_spa, SCL_CONFIG | SCL_STATE, RW_READER) ==
200 (SCL_CONFIG | SCL_STATE) &&
201 dsl_pool_sync_context(spa_get_dsl(zio->io_spa))));
202 ASSERT(flags & ZIO_FLAG_CONFIG_WRITER);
203
204 zio_nowait(zio_write_phys(zio, vd,
205 vdev_label_offset(vd->vdev_psize, l, offset),
206 size, buf, ZIO_CHECKSUM_LABEL, done, private,
207 ZIO_PRIORITY_SYNC_WRITE, flags, B_TRUE));
208}
209
210/*
211 * Generate the nvlist representing this vdev's config.
212 */
213nvlist_t *
214vdev_config_generate(spa_t *spa, vdev_t *vd, boolean_t getstats,
215 vdev_config_flag_t flags)
216{
217 nvlist_t *nv = NULL;
218
219 VERIFY(nvlist_alloc(&nv, NV_UNIQUE_NAME, KM_SLEEP) == 0);
220
221 VERIFY(nvlist_add_string(nv, ZPOOL_CONFIG_TYPE,
222 vd->vdev_ops->vdev_op_type) == 0);
223 if (!(flags & (VDEV_CONFIG_SPARE | VDEV_CONFIG_L2CACHE)))
224 VERIFY(nvlist_add_uint64(nv, ZPOOL_CONFIG_ID, vd->vdev_id)
225 == 0);
226 VERIFY(nvlist_add_uint64(nv, ZPOOL_CONFIG_GUID, vd->vdev_guid) == 0);
227
228 if (vd->vdev_path != NULL)
229 VERIFY(nvlist_add_string(nv, ZPOOL_CONFIG_PATH,
230 vd->vdev_path) == 0);
231
232 if (vd->vdev_devid != NULL)
233 VERIFY(nvlist_add_string(nv, ZPOOL_CONFIG_DEVID,
234 vd->vdev_devid) == 0);
235
236 if (vd->vdev_physpath != NULL)
237 VERIFY(nvlist_add_string(nv, ZPOOL_CONFIG_PHYS_PATH,
238 vd->vdev_physpath) == 0);
239
240 if (vd->vdev_fru != NULL)
241 VERIFY(nvlist_add_string(nv, ZPOOL_CONFIG_FRU,
242 vd->vdev_fru) == 0);
243
244 if (vd->vdev_nparity != 0) {
245 ASSERT(strcmp(vd->vdev_ops->vdev_op_type,
246 VDEV_TYPE_RAIDZ) == 0);
247
248 /*
249 * Make sure someone hasn't managed to sneak a fancy new vdev
250 * into a crufty old storage pool.
251 */
252 ASSERT(vd->vdev_nparity == 1 ||
253 (vd->vdev_nparity <= 2 &&
254 spa_version(spa) >= SPA_VERSION_RAIDZ2) ||
255 (vd->vdev_nparity <= 3 &&
256 spa_version(spa) >= SPA_VERSION_RAIDZ3));
257
258 /*
259 * Note that we'll add the nparity tag even on storage pools
260 * that only support a single parity device -- older software
261 * will just ignore it.
262 */
263 VERIFY(nvlist_add_uint64(nv, ZPOOL_CONFIG_NPARITY,
264 vd->vdev_nparity) == 0);
265 }
266
267 if (vd->vdev_wholedisk != -1ULL)
268 VERIFY(nvlist_add_uint64(nv, ZPOOL_CONFIG_WHOLE_DISK,
269 vd->vdev_wholedisk) == 0);
270
271 if (vd->vdev_not_present)
272 VERIFY(nvlist_add_uint64(nv, ZPOOL_CONFIG_NOT_PRESENT, 1) == 0);
273
274 if (vd->vdev_isspare)
275 VERIFY(nvlist_add_uint64(nv, ZPOOL_CONFIG_IS_SPARE, 1) == 0);
276
277 if (!(flags & (VDEV_CONFIG_SPARE | VDEV_CONFIG_L2CACHE)) &&
278 vd == vd->vdev_top) {
279 VERIFY(nvlist_add_uint64(nv, ZPOOL_CONFIG_METASLAB_ARRAY,
280 vd->vdev_ms_array) == 0);
281 VERIFY(nvlist_add_uint64(nv, ZPOOL_CONFIG_METASLAB_SHIFT,
282 vd->vdev_ms_shift) == 0);
283 VERIFY(nvlist_add_uint64(nv, ZPOOL_CONFIG_ASHIFT,
284 vd->vdev_ashift) == 0);
285 VERIFY(nvlist_add_uint64(nv, ZPOOL_CONFIG_ASIZE,
286 vd->vdev_asize) == 0);
287 VERIFY(nvlist_add_uint64(nv, ZPOOL_CONFIG_IS_LOG,
288 vd->vdev_islog) == 0);
289 if (vd->vdev_removing)
290 VERIFY(nvlist_add_uint64(nv, ZPOOL_CONFIG_REMOVING,
291 vd->vdev_removing) == 0);
292 }
293
294 if (vd->vdev_dtl_smo.smo_object != 0)
295 VERIFY(nvlist_add_uint64(nv, ZPOOL_CONFIG_DTL,
296 vd->vdev_dtl_smo.smo_object) == 0);
297
298 if (vd->vdev_crtxg)
299 VERIFY(nvlist_add_uint64(nv, ZPOOL_CONFIG_CREATE_TXG,
300 vd->vdev_crtxg) == 0);
301
302 if (getstats) {
303 vdev_stat_t vs;
304 pool_scan_stat_t ps;
305
306 vdev_get_stats(vd, &vs);
307 VERIFY(nvlist_add_uint64_array(nv, ZPOOL_CONFIG_VDEV_STATS,
308 (uint64_t *)&vs, sizeof (vs) / sizeof (uint64_t)) == 0);
309
310 /* provide either current or previous scan information */
311 if (spa_scan_get_stats(spa, &ps) == 0) {
312 VERIFY(nvlist_add_uint64_array(nv,
313 ZPOOL_CONFIG_SCAN_STATS, (uint64_t *)&ps,
314 sizeof (pool_scan_stat_t) / sizeof (uint64_t))
315 == 0);
316 }
317 }
318
319 if (!vd->vdev_ops->vdev_op_leaf) {
320 nvlist_t **child;
321 int c, idx;
322
323 ASSERT(!vd->vdev_ishole);
324
325 child = kmem_alloc(vd->vdev_children * sizeof (nvlist_t *),
326 KM_SLEEP);
327
328 for (c = 0, idx = 0; c < vd->vdev_children; c++) {
329 vdev_t *cvd = vd->vdev_child[c];
330
331 /*
332 * If we're generating an nvlist of removing
333 * vdevs then skip over any device which is
334 * not being removed.
335 */
336 if ((flags & VDEV_CONFIG_REMOVING) &&
337 !cvd->vdev_removing)
338 continue;
339
340 child[idx++] = vdev_config_generate(spa, cvd,
341 getstats, flags);
342 }
343
344 if (idx) {
345 VERIFY(nvlist_add_nvlist_array(nv,
346 ZPOOL_CONFIG_CHILDREN, child, idx) == 0);
347 }
348
349 for (c = 0; c < idx; c++)
350 nvlist_free(child[c]);
351
352 kmem_free(child, vd->vdev_children * sizeof (nvlist_t *));
353
354 } else {
355 const char *aux = NULL;
356
357 if (vd->vdev_offline && !vd->vdev_tmpoffline)
358 VERIFY(nvlist_add_uint64(nv, ZPOOL_CONFIG_OFFLINE,
359 B_TRUE) == 0);
360 if (vd->vdev_resilvering)
361 VERIFY(nvlist_add_uint64(nv, ZPOOL_CONFIG_RESILVERING,
362 B_TRUE) == 0);
363 if (vd->vdev_faulted)
364 VERIFY(nvlist_add_uint64(nv, ZPOOL_CONFIG_FAULTED,
365 B_TRUE) == 0);
366 if (vd->vdev_degraded)
367 VERIFY(nvlist_add_uint64(nv, ZPOOL_CONFIG_DEGRADED,
368 B_TRUE) == 0);
369 if (vd->vdev_removed)
370 VERIFY(nvlist_add_uint64(nv, ZPOOL_CONFIG_REMOVED,
371 B_TRUE) == 0);
372 if (vd->vdev_unspare)
373 VERIFY(nvlist_add_uint64(nv, ZPOOL_CONFIG_UNSPARE,
374 B_TRUE) == 0);
375 if (vd->vdev_ishole)
376 VERIFY(nvlist_add_uint64(nv, ZPOOL_CONFIG_IS_HOLE,
377 B_TRUE) == 0);
378
379 switch (vd->vdev_stat.vs_aux) {
380 case VDEV_AUX_ERR_EXCEEDED:
381 aux = "err_exceeded";
382 break;
383
384 case VDEV_AUX_EXTERNAL:
385 aux = "external";
386 break;
387 }
388
389 if (aux != NULL)
390 VERIFY(nvlist_add_string(nv, ZPOOL_CONFIG_AUX_STATE,
391 aux) == 0);
392
393 if (vd->vdev_splitting && vd->vdev_orig_guid != 0LL) {
394 VERIFY(nvlist_add_uint64(nv, ZPOOL_CONFIG_ORIG_GUID,
395 vd->vdev_orig_guid) == 0);
396 }
397 }
398
399 return (nv);
400}
401
402/*
403 * Generate a view of the top-level vdevs. If we currently have holes
404 * in the namespace, then generate an array which contains a list of holey
405 * vdevs. Additionally, add the number of top-level children that currently
406 * exist.
407 */
408void
409vdev_top_config_generate(spa_t *spa, nvlist_t *config)
410{
411 vdev_t *rvd = spa->spa_root_vdev;
412 uint64_t *array;
413 uint_t c, idx;
414
415 array = kmem_alloc(rvd->vdev_children * sizeof (uint64_t), KM_SLEEP);
416
417 for (c = 0, idx = 0; c < rvd->vdev_children; c++) {
418 vdev_t *tvd = rvd->vdev_child[c];
419
420 if (tvd->vdev_ishole)
421 array[idx++] = c;
422 }
423
424 if (idx) {
425 VERIFY(nvlist_add_uint64_array(config, ZPOOL_CONFIG_HOLE_ARRAY,
426 array, idx) == 0);
427 }
428
429 VERIFY(nvlist_add_uint64(config, ZPOOL_CONFIG_VDEV_CHILDREN,
430 rvd->vdev_children) == 0);
431
432 kmem_free(array, rvd->vdev_children * sizeof (uint64_t));
433}
434
435/*
436 * Returns the configuration from the label of the given vdev. For vdevs
437 * which don't have a txg value stored on their label (i.e. spares/cache)
438 * or have not been completely initialized (txg = 0) just return
439 * the configuration from the first valid label we find. Otherwise,
440 * find the most up-to-date label that does not exceed the specified
441 * 'txg' value.
442 */
443nvlist_t *
444vdev_label_read_config(vdev_t *vd, uint64_t txg)
445{
446 spa_t *spa = vd->vdev_spa;
447 nvlist_t *config = NULL;
448 vdev_phys_t *vp;
449 zio_t *zio;
450 uint64_t best_txg = 0;
451 int error = 0;
452 int flags = ZIO_FLAG_CONFIG_WRITER | ZIO_FLAG_CANFAIL |
453 ZIO_FLAG_SPECULATIVE;
454
455 ASSERT(spa_config_held(spa, SCL_STATE_ALL, RW_WRITER) == SCL_STATE_ALL);
456
457 if (!vdev_readable(vd))
458 return (NULL);
459
460 vp = zio_buf_alloc(sizeof (vdev_phys_t));
461
462retry:
463 for (int l = 0; l < VDEV_LABELS; l++) {
464 nvlist_t *label = NULL;
465
466 zio = zio_root(spa, NULL, NULL, flags);
467
468 vdev_label_read(zio, vd, l, vp,
469 offsetof(vdev_label_t, vl_vdev_phys),
470 sizeof (vdev_phys_t), NULL, NULL, flags);
471
472 if (zio_wait(zio) == 0 &&
473 nvlist_unpack(vp->vp_nvlist, sizeof (vp->vp_nvlist),
474 &label, 0) == 0) {
475 uint64_t label_txg = 0;
476
477 /*
478 * Auxiliary vdevs won't have txg values in their
479 * labels and newly added vdevs may not have been
480 * completely initialized so just return the
481 * configuration from the first valid label we
482 * encounter.
483 */
484 error = nvlist_lookup_uint64(label,
485 ZPOOL_CONFIG_POOL_TXG, &label_txg);
486 if ((error || label_txg == 0) && !config) {
487 config = label;
488 break;
489 } else if (label_txg <= txg && label_txg > best_txg) {
490 best_txg = label_txg;
491 nvlist_free(config);
492 config = fnvlist_dup(label);
493 }
494 }
495
496 if (label != NULL) {
497 nvlist_free(label);
498 label = NULL;
499 }
500 }
501
502 if (config == NULL && !(flags & ZIO_FLAG_TRYHARD)) {
503 flags |= ZIO_FLAG_TRYHARD;
504 goto retry;
505 }
506
507 zio_buf_free(vp, sizeof (vdev_phys_t));
508
509 return (config);
510}
511
512/*
513 * Determine if a device is in use. The 'spare_guid' parameter will be filled
514 * in with the device guid if this spare is active elsewhere on the system.
515 */
516static boolean_t
517vdev_inuse(vdev_t *vd, uint64_t crtxg, vdev_labeltype_t reason,
518 uint64_t *spare_guid, uint64_t *l2cache_guid)
519{
520 spa_t *spa = vd->vdev_spa;
521 uint64_t state, pool_guid, device_guid, txg, spare_pool;
522 uint64_t vdtxg = 0;
523 nvlist_t *label;
524
525 if (spare_guid)
526 *spare_guid = 0ULL;
527 if (l2cache_guid)
528 *l2cache_guid = 0ULL;
529
530 /*
531 * Read the label, if any, and perform some basic sanity checks.
532 */
533 if ((label = vdev_label_read_config(vd, -1ULL)) == NULL)
534 return (B_FALSE);
535
536 (void) nvlist_lookup_uint64(label, ZPOOL_CONFIG_CREATE_TXG,
537 &vdtxg);
538
539 if (nvlist_lookup_uint64(label, ZPOOL_CONFIG_POOL_STATE,
540 &state) != 0 ||
541 nvlist_lookup_uint64(label, ZPOOL_CONFIG_GUID,
542 &device_guid) != 0) {
543 nvlist_free(label);
544 return (B_FALSE);
545 }
546
547 if (state != POOL_STATE_SPARE && state != POOL_STATE_L2CACHE &&
548 (nvlist_lookup_uint64(label, ZPOOL_CONFIG_POOL_GUID,
549 &pool_guid) != 0 ||
550 nvlist_lookup_uint64(label, ZPOOL_CONFIG_POOL_TXG,
551 &txg) != 0)) {
552 nvlist_free(label);
553 return (B_FALSE);
554 }
555
556 nvlist_free(label);
557
558 /*
559 * Check to see if this device indeed belongs to the pool it claims to
560 * be a part of. The only way this is allowed is if the device is a hot
561 * spare (which we check for later on).
562 */
563 if (state != POOL_STATE_SPARE && state != POOL_STATE_L2CACHE &&
564 !spa_guid_exists(pool_guid, device_guid) &&
565 !spa_spare_exists(device_guid, NULL, NULL) &&
566 !spa_l2cache_exists(device_guid, NULL))
567 return (B_FALSE);
568
569 /*
570 * If the transaction group is zero, then this an initialized (but
571 * unused) label. This is only an error if the create transaction
572 * on-disk is the same as the one we're using now, in which case the
573 * user has attempted to add the same vdev multiple times in the same
574 * transaction.
575 */
576 if (state != POOL_STATE_SPARE && state != POOL_STATE_L2CACHE &&
577 txg == 0 && vdtxg == crtxg)
578 return (B_TRUE);
579
580 /*
581 * Check to see if this is a spare device. We do an explicit check for
582 * spa_has_spare() here because it may be on our pending list of spares
583 * to add. We also check if it is an l2cache device.
584 */
585 if (spa_spare_exists(device_guid, &spare_pool, NULL) ||
586 spa_has_spare(spa, device_guid)) {
587 if (spare_guid)
588 *spare_guid = device_guid;
589
590 switch (reason) {
591 case VDEV_LABEL_CREATE:
592 case VDEV_LABEL_L2CACHE:
593 return (B_TRUE);
594
595 case VDEV_LABEL_REPLACE:
596 return (!spa_has_spare(spa, device_guid) ||
597 spare_pool != 0ULL);
598
599 case VDEV_LABEL_SPARE:
600 return (spa_has_spare(spa, device_guid));
601 }
602 }
603
604 /*
605 * Check to see if this is an l2cache device.
606 */
607 if (spa_l2cache_exists(device_guid, NULL))
608 return (B_TRUE);
609
610 /*
611 * We can't rely on a pool's state if it's been imported
612 * read-only. Instead we look to see if the pools is marked
613 * read-only in the namespace and set the state to active.
614 */
615 if ((spa = spa_by_guid(pool_guid, device_guid)) != NULL &&
616 spa_mode(spa) == FREAD)
617 state = POOL_STATE_ACTIVE;
618
619 /*
620 * If the device is marked ACTIVE, then this device is in use by another
621 * pool on the system.
622 */
623 return (state == POOL_STATE_ACTIVE);
624}
625
626/*
627 * Initialize a vdev label. We check to make sure each leaf device is not in
628 * use, and writable. We put down an initial label which we will later
629 * overwrite with a complete label. Note that it's important to do this
630 * sequentially, not in parallel, so that we catch cases of multiple use of the
631 * same leaf vdev in the vdev we're creating -- e.g. mirroring a disk with
632 * itself.
633 */
634int
635vdev_label_init(vdev_t *vd, uint64_t crtxg, vdev_labeltype_t reason)
636{
637 spa_t *spa = vd->vdev_spa;
638 nvlist_t *label;
639 vdev_phys_t *vp;
640 char *pad2;
641 uberblock_t *ub;
642 zio_t *zio;
643 char *buf;
644 size_t buflen;
645 int error;
646 uint64_t spare_guid, l2cache_guid;
647 int flags = ZIO_FLAG_CONFIG_WRITER | ZIO_FLAG_CANFAIL;
648
649 ASSERT(spa_config_held(spa, SCL_ALL, RW_WRITER) == SCL_ALL);
650
651 for (int c = 0; c < vd->vdev_children; c++)
652 if ((error = vdev_label_init(vd->vdev_child[c],
653 crtxg, reason)) != 0)
654 return (error);
655
656 /* Track the creation time for this vdev */
657 vd->vdev_crtxg = crtxg;
658
659 if (!vd->vdev_ops->vdev_op_leaf)
660 return (0);
661
662 /*
663 * Dead vdevs cannot be initialized.
664 */
665 if (vdev_is_dead(vd))
666 return (EIO);
667
668 /*
669 * Determine if the vdev is in use.
670 */
671 if (reason != VDEV_LABEL_REMOVE && reason != VDEV_LABEL_SPLIT &&
672 vdev_inuse(vd, crtxg, reason, &spare_guid, &l2cache_guid))
673 return (EBUSY);
674
675 /*
676 * If this is a request to add or replace a spare or l2cache device
677 * that is in use elsewhere on the system, then we must update the
678 * guid (which was initialized to a random value) to reflect the
679 * actual GUID (which is shared between multiple pools).
680 */
681 if (reason != VDEV_LABEL_REMOVE && reason != VDEV_LABEL_L2CACHE &&
682 spare_guid != 0ULL) {
683 uint64_t guid_delta = spare_guid - vd->vdev_guid;
684
685 vd->vdev_guid += guid_delta;
686
687 for (vdev_t *pvd = vd; pvd != NULL; pvd = pvd->vdev_parent)
688 pvd->vdev_guid_sum += guid_delta;
689
690 /*
691 * If this is a replacement, then we want to fallthrough to the
692 * rest of the code. If we're adding a spare, then it's already
693 * labeled appropriately and we can just return.
694 */
695 if (reason == VDEV_LABEL_SPARE)
696 return (0);
697 ASSERT(reason == VDEV_LABEL_REPLACE ||
698 reason == VDEV_LABEL_SPLIT);
699 }
700
701 if (reason != VDEV_LABEL_REMOVE && reason != VDEV_LABEL_SPARE &&
702 l2cache_guid != 0ULL) {
703 uint64_t guid_delta = l2cache_guid - vd->vdev_guid;
704
705 vd->vdev_guid += guid_delta;
706
707 for (vdev_t *pvd = vd; pvd != NULL; pvd = pvd->vdev_parent)
708 pvd->vdev_guid_sum += guid_delta;
709
710 /*
711 * If this is a replacement, then we want to fallthrough to the
712 * rest of the code. If we're adding an l2cache, then it's
713 * already labeled appropriately and we can just return.
714 */
715 if (reason == VDEV_LABEL_L2CACHE)
716 return (0);
717 ASSERT(reason == VDEV_LABEL_REPLACE);
718 }
719
720 /*
721 * Initialize its label.
722 */
723 vp = zio_buf_alloc(sizeof (vdev_phys_t));
724 bzero(vp, sizeof (vdev_phys_t));
725
726 /*
727 * Generate a label describing the pool and our top-level vdev.
728 * We mark it as being from txg 0 to indicate that it's not
729 * really part of an active pool just yet. The labels will
730 * be written again with a meaningful txg by spa_sync().
731 */
732 if (reason == VDEV_LABEL_SPARE ||
733 (reason == VDEV_LABEL_REMOVE && vd->vdev_isspare)) {
734 /*
735 * For inactive hot spares, we generate a special label that
736 * identifies as a mutually shared hot spare. We write the
737 * label if we are adding a hot spare, or if we are removing an
738 * active hot spare (in which case we want to revert the
739 * labels).
740 */
741 VERIFY(nvlist_alloc(&label, NV_UNIQUE_NAME, KM_SLEEP) == 0);
742
743 VERIFY(nvlist_add_uint64(label, ZPOOL_CONFIG_VERSION,
744 spa_version(spa)) == 0);
745 VERIFY(nvlist_add_uint64(label, ZPOOL_CONFIG_POOL_STATE,
746 POOL_STATE_SPARE) == 0);
747 VERIFY(nvlist_add_uint64(label, ZPOOL_CONFIG_GUID,
748 vd->vdev_guid) == 0);
749 } else if (reason == VDEV_LABEL_L2CACHE ||
750 (reason == VDEV_LABEL_REMOVE && vd->vdev_isl2cache)) {
751 /*
752 * For level 2 ARC devices, add a special label.
753 */
754 VERIFY(nvlist_alloc(&label, NV_UNIQUE_NAME, KM_SLEEP) == 0);
755
756 VERIFY(nvlist_add_uint64(label, ZPOOL_CONFIG_VERSION,
757 spa_version(spa)) == 0);
758 VERIFY(nvlist_add_uint64(label, ZPOOL_CONFIG_POOL_STATE,
759 POOL_STATE_L2CACHE) == 0);
760 VERIFY(nvlist_add_uint64(label, ZPOOL_CONFIG_GUID,
761 vd->vdev_guid) == 0);
762 } else {
763 uint64_t txg = 0ULL;
764
765 if (reason == VDEV_LABEL_SPLIT)
766 txg = spa->spa_uberblock.ub_txg;
767 label = spa_config_generate(spa, vd, txg, B_FALSE);
768
769 /*
770 * Add our creation time. This allows us to detect multiple
771 * vdev uses as described above, and automatically expires if we
772 * fail.
773 */
774 VERIFY(nvlist_add_uint64(label, ZPOOL_CONFIG_CREATE_TXG,
775 crtxg) == 0);
776 }
777
778 buf = vp->vp_nvlist;
779 buflen = sizeof (vp->vp_nvlist);
780
781 error = nvlist_pack(label, &buf, &buflen, NV_ENCODE_XDR, KM_SLEEP);
782 if (error != 0) {
783 nvlist_free(label);
784 zio_buf_free(vp, sizeof (vdev_phys_t));
785 /* EFAULT means nvlist_pack ran out of room */
786 return (error == EFAULT ? ENAMETOOLONG : EINVAL);
787 }
788
789 /*
790 * Initialize uberblock template.
791 */
792 ub = zio_buf_alloc(VDEV_UBERBLOCK_RING);
793 bzero(ub, VDEV_UBERBLOCK_RING);
794 *ub = spa->spa_uberblock;
795 ub->ub_txg = 0;
796
797 /* Initialize the 2nd padding area. */
798 pad2 = zio_buf_alloc(VDEV_PAD_SIZE);
799 bzero(pad2, VDEV_PAD_SIZE);
800
801 /*
802 * Write everything in parallel.
803 */
804retry:
805 zio = zio_root(spa, NULL, NULL, flags);
806
807 for (int l = 0; l < VDEV_LABELS; l++) {
808
809 vdev_label_write(zio, vd, l, vp,
810 offsetof(vdev_label_t, vl_vdev_phys),
811 sizeof (vdev_phys_t), NULL, NULL, flags);
812
813 /*
814 * Skip the 1st padding area.
815 * Zero out the 2nd padding area where it might have
816 * left over data from previous filesystem format.
817 */
818 vdev_label_write(zio, vd, l, pad2,
819 offsetof(vdev_label_t, vl_pad2),
820 VDEV_PAD_SIZE, NULL, NULL, flags);
821
822 vdev_label_write(zio, vd, l, ub,
823 offsetof(vdev_label_t, vl_uberblock),
824 VDEV_UBERBLOCK_RING, NULL, NULL, flags);
825 }
826
827 error = zio_wait(zio);
828
829 if (error != 0 && !(flags & ZIO_FLAG_TRYHARD)) {
830 flags |= ZIO_FLAG_TRYHARD;
831 goto retry;
832 }
833
834 nvlist_free(label);
835 zio_buf_free(pad2, VDEV_PAD_SIZE);
836 zio_buf_free(ub, VDEV_UBERBLOCK_RING);
837 zio_buf_free(vp, sizeof (vdev_phys_t));
838
839 /*
840 * If this vdev hasn't been previously identified as a spare, then we
841 * mark it as such only if a) we are labeling it as a spare, or b) it
842 * exists as a spare elsewhere in the system. Do the same for
843 * level 2 ARC devices.
844 */
845 if (error == 0 && !vd->vdev_isspare &&
846 (reason == VDEV_LABEL_SPARE ||
847 spa_spare_exists(vd->vdev_guid, NULL, NULL)))
848 spa_spare_add(vd);
849
850 if (error == 0 && !vd->vdev_isl2cache &&
851 (reason == VDEV_LABEL_L2CACHE ||
852 spa_l2cache_exists(vd->vdev_guid, NULL)))
853 spa_l2cache_add(vd);
854
855 return (error);
856}
857
858/*
859 * ==========================================================================
860 * uberblock load/sync
861 * ==========================================================================
862 */
863
864/*
865 * Consider the following situation: txg is safely synced to disk. We've
866 * written the first uberblock for txg + 1, and then we lose power. When we
867 * come back up, we fail to see the uberblock for txg + 1 because, say,
868 * it was on a mirrored device and the replica to which we wrote txg + 1
869 * is now offline. If we then make some changes and sync txg + 1, and then
870 * the missing replica comes back, then for a few seconds we'll have two
871 * conflicting uberblocks on disk with the same txg. The solution is simple:
872 * among uberblocks with equal txg, choose the one with the latest timestamp.
873 */
874static int
875vdev_uberblock_compare(uberblock_t *ub1, uberblock_t *ub2)
876{
877 if (ub1->ub_txg < ub2->ub_txg)
878 return (-1);
879 if (ub1->ub_txg > ub2->ub_txg)
880 return (1);
881
882 if (ub1->ub_timestamp < ub2->ub_timestamp)
883 return (-1);
884 if (ub1->ub_timestamp > ub2->ub_timestamp)
885 return (1);
886
887 return (0);
888}
889
890struct ubl_cbdata {
891 uberblock_t *ubl_ubbest; /* Best uberblock */
892 vdev_t *ubl_vd; /* vdev associated with the above */
893};
894
895static void
896vdev_uberblock_load_done(zio_t *zio)
897{
898 vdev_t *vd = zio->io_vd;
899 spa_t *spa = zio->io_spa;
900 zio_t *rio = zio->io_private;
901 uberblock_t *ub = zio->io_data;
902 struct ubl_cbdata *cbp = rio->io_private;
903
904 ASSERT3U(zio->io_size, ==, VDEV_UBERBLOCK_SIZE(vd));
905
906 if (zio->io_error == 0 && uberblock_verify(ub) == 0) {
907 mutex_enter(&rio->io_lock);
908 if (ub->ub_txg <= spa->spa_load_max_txg &&
909 vdev_uberblock_compare(ub, cbp->ubl_ubbest) > 0) {
910 /*
911 * Keep track of the vdev in which this uberblock
912 * was found. We will use this information later
913 * to obtain the config nvlist associated with
914 * this uberblock.
915 */
916 *cbp->ubl_ubbest = *ub;
917 cbp->ubl_vd = vd;
918 }
919 mutex_exit(&rio->io_lock);
920 }
921
922 zio_buf_free(zio->io_data, zio->io_size);
923}
924
925static void
926vdev_uberblock_load_impl(zio_t *zio, vdev_t *vd, int flags,
927 struct ubl_cbdata *cbp)
928{
929 for (int c = 0; c < vd->vdev_children; c++)
930 vdev_uberblock_load_impl(zio, vd->vdev_child[c], flags, cbp);
931
932 if (vd->vdev_ops->vdev_op_leaf && vdev_readable(vd)) {
933 for (int l = 0; l < VDEV_LABELS; l++) {
934 for (int n = 0; n < VDEV_UBERBLOCK_COUNT(vd); n++) {
935 vdev_label_read(zio, vd, l,
936 zio_buf_alloc(VDEV_UBERBLOCK_SIZE(vd)),
937 VDEV_UBERBLOCK_OFFSET(vd, n),
938 VDEV_UBERBLOCK_SIZE(vd),
939 vdev_uberblock_load_done, zio, flags);
940 }
941 }
942 }
943}
944
945/*
946 * Reads the 'best' uberblock from disk along with its associated
947 * configuration. First, we read the uberblock array of each label of each
948 * vdev, keeping track of the uberblock with the highest txg in each array.
949 * Then, we read the configuration from the same vdev as the best uberblock.
950 */
951void
952vdev_uberblock_load(vdev_t *rvd, uberblock_t *ub, nvlist_t **config)
953{
954 zio_t *zio;
955 spa_t *spa = rvd->vdev_spa;
956 struct ubl_cbdata cb;
957 int flags = ZIO_FLAG_CONFIG_WRITER | ZIO_FLAG_CANFAIL |
958 ZIO_FLAG_SPECULATIVE | ZIO_FLAG_TRYHARD;
959
960 ASSERT(ub);
961 ASSERT(config);
962
963 bzero(ub, sizeof (uberblock_t));
964 *config = NULL;
965
966 cb.ubl_ubbest = ub;
967 cb.ubl_vd = NULL;
968
969 spa_config_enter(spa, SCL_ALL, FTAG, RW_WRITER);
970 zio = zio_root(spa, NULL, &cb, flags);
971 vdev_uberblock_load_impl(zio, rvd, flags, &cb);
972 (void) zio_wait(zio);
973
974 /*
975 * It's possible that the best uberblock was discovered on a label
976 * that has a configuration which was written in a future txg.
977 * Search all labels on this vdev to find the configuration that
978 * matches the txg for our uberblock.
979 */
980 if (cb.ubl_vd != NULL)
981 *config = vdev_label_read_config(cb.ubl_vd, ub->ub_txg);
982 spa_config_exit(spa, SCL_ALL, FTAG);
983}
984
985/*
986 * On success, increment root zio's count of good writes.
987 * We only get credit for writes to known-visible vdevs; see spa_vdev_add().
988 */
989static void
990vdev_uberblock_sync_done(zio_t *zio)
991{
992 uint64_t *good_writes = zio->io_private;
993
994 if (zio->io_error == 0 && zio->io_vd->vdev_top->vdev_ms_array != 0)
995 atomic_add_64(good_writes, 1);
996}
997
998/*
999 * Write the uberblock to all labels of all leaves of the specified vdev.
1000 */
1001static void
1002vdev_uberblock_sync(zio_t *zio, uberblock_t *ub, vdev_t *vd, int flags)
1003{
1004 uberblock_t *ubbuf;
1005 int n;
1006
1007 for (int c = 0; c < vd->vdev_children; c++)
1008 vdev_uberblock_sync(zio, ub, vd->vdev_child[c], flags);
1009
1010 if (!vd->vdev_ops->vdev_op_leaf)
1011 return;
1012
1013 if (!vdev_writeable(vd))
1014 return;
1015
1016 n = ub->ub_txg & (VDEV_UBERBLOCK_COUNT(vd) - 1);
1017
1018 ubbuf = zio_buf_alloc(VDEV_UBERBLOCK_SIZE(vd));
1019 bzero(ubbuf, VDEV_UBERBLOCK_SIZE(vd));
1020 *ubbuf = *ub;
1021
1022 for (int l = 0; l < VDEV_LABELS; l++)
1023 vdev_label_write(zio, vd, l, ubbuf,
1024 VDEV_UBERBLOCK_OFFSET(vd, n), VDEV_UBERBLOCK_SIZE(vd),
1025 vdev_uberblock_sync_done, zio->io_private,
1026 flags | ZIO_FLAG_DONT_PROPAGATE);
1027
1028 zio_buf_free(ubbuf, VDEV_UBERBLOCK_SIZE(vd));
1029}
1030
1031int
1032vdev_uberblock_sync_list(vdev_t **svd, int svdcount, uberblock_t *ub, int flags)
1033{
1034 spa_t *spa = svd[0]->vdev_spa;
1035 zio_t *zio;
1036 uint64_t good_writes = 0;
1037
1038 zio = zio_root(spa, NULL, &good_writes, flags);
1039
1040 for (int v = 0; v < svdcount; v++)
1041 vdev_uberblock_sync(zio, ub, svd[v], flags);
1042
1043 (void) zio_wait(zio);
1044
1045 /*
1046 * Flush the uberblocks to disk. This ensures that the odd labels
1047 * are no longer needed (because the new uberblocks and the even
1048 * labels are safely on disk), so it is safe to overwrite them.
1049 */
1050 zio = zio_root(spa, NULL, NULL, flags);
1051
1052 for (int v = 0; v < svdcount; v++)
1053 zio_flush(zio, svd[v]);
1054
1055 (void) zio_wait(zio);
1056
1057 return (good_writes >= 1 ? 0 : EIO);
1058}
1059
1060/*
1061 * On success, increment the count of good writes for our top-level vdev.
1062 */
1063static void
1064vdev_label_sync_done(zio_t *zio)
1065{
1066 uint64_t *good_writes = zio->io_private;
1067
1068 if (zio->io_error == 0)
1069 atomic_add_64(good_writes, 1);
1070}
1071
1072/*
1073 * If there weren't enough good writes, indicate failure to the parent.
1074 */
1075static void
1076vdev_label_sync_top_done(zio_t *zio)
1077{
1078 uint64_t *good_writes = zio->io_private;
1079
1080 if (*good_writes == 0)
1081 zio->io_error = EIO;
1082
1083 kmem_free(good_writes, sizeof (uint64_t));
1084}
1085
1086/*
1087 * We ignore errors for log and cache devices, simply free the private data.
1088 */
1089static void
1090vdev_label_sync_ignore_done(zio_t *zio)
1091{
1092 kmem_free(zio->io_private, sizeof (uint64_t));
1093}
1094
1095/*
1096 * Write all even or odd labels to all leaves of the specified vdev.
1097 */
1098static void
1099vdev_label_sync(zio_t *zio, vdev_t *vd, int l, uint64_t txg, int flags)
1100{
1101 nvlist_t *label;
1102 vdev_phys_t *vp;
1103 char *buf;
1104 size_t buflen;
1105
1106 for (int c = 0; c < vd->vdev_children; c++)
1107 vdev_label_sync(zio, vd->vdev_child[c], l, txg, flags);
1108
1109 if (!vd->vdev_ops->vdev_op_leaf)
1110 return;
1111
1112 if (!vdev_writeable(vd))
1113 return;
1114
1115 /*
1116 * Generate a label describing the top-level config to which we belong.
1117 */
1118 label = spa_config_generate(vd->vdev_spa, vd, txg, B_FALSE);
1119
1120 vp = zio_buf_alloc(sizeof (vdev_phys_t));
1121 bzero(vp, sizeof (vdev_phys_t));
1122
1123 buf = vp->vp_nvlist;
1124 buflen = sizeof (vp->vp_nvlist);
1125
1126 if (nvlist_pack(label, &buf, &buflen, NV_ENCODE_XDR, KM_SLEEP) == 0) {
1127 for (; l < VDEV_LABELS; l += 2) {
1128 vdev_label_write(zio, vd, l, vp,
1129 offsetof(vdev_label_t, vl_vdev_phys),
1130 sizeof (vdev_phys_t),
1131 vdev_label_sync_done, zio->io_private,
1132 flags | ZIO_FLAG_DONT_PROPAGATE);
1133 }
1134 }
1135
1136 zio_buf_free(vp, sizeof (vdev_phys_t));
1137 nvlist_free(label);
1138}
1139
1140int
1141vdev_label_sync_list(spa_t *spa, int l, uint64_t txg, int flags)
1142{
1143 list_t *dl = &spa->spa_config_dirty_list;
1144 vdev_t *vd;
1145 zio_t *zio;
1146 int error;
1147
1148 /*
1149 * Write the new labels to disk.
1150 */
1151 zio = zio_root(spa, NULL, NULL, flags);
1152
1153 for (vd = list_head(dl); vd != NULL; vd = list_next(dl, vd)) {
1154 uint64_t *good_writes = kmem_zalloc(sizeof (uint64_t),
1155 KM_SLEEP);
1156
1157 ASSERT(!vd->vdev_ishole);
1158
1159 zio_t *vio = zio_null(zio, spa, NULL,
1160 (vd->vdev_islog || vd->vdev_aux != NULL) ?
1161 vdev_label_sync_ignore_done : vdev_label_sync_top_done,
1162 good_writes, flags);
1163 vdev_label_sync(vio, vd, l, txg, flags);
1164 zio_nowait(vio);
1165 }
1166
1167 error = zio_wait(zio);
1168
1169 /*
1170 * Flush the new labels to disk.
1171 */
1172 zio = zio_root(spa, NULL, NULL, flags);
1173
1174 for (vd = list_head(dl); vd != NULL; vd = list_next(dl, vd))
1175 zio_flush(zio, vd);
1176
1177 (void) zio_wait(zio);
1178
1179 return (error);
1180}
1181
1182/*
1183 * Sync the uberblock and any changes to the vdev configuration.
1184 *
1185 * The order of operations is carefully crafted to ensure that
1186 * if the system panics or loses power at any time, the state on disk
1187 * is still transactionally consistent. The in-line comments below
1188 * describe the failure semantics at each stage.
1189 *
1190 * Moreover, vdev_config_sync() is designed to be idempotent: if it fails
1191 * at any time, you can just call it again, and it will resume its work.
1192 */
1193int
1194vdev_config_sync(vdev_t **svd, int svdcount, uint64_t txg, boolean_t tryhard)
1195{
1196 spa_t *spa = svd[0]->vdev_spa;
1197 uberblock_t *ub = &spa->spa_uberblock;
1198 vdev_t *vd;
1199 zio_t *zio;
1200 int error;
1201 int flags = ZIO_FLAG_CONFIG_WRITER | ZIO_FLAG_CANFAIL;
1202
1203 /*
1204 * Normally, we don't want to try too hard to write every label and
1205 * uberblock. If there is a flaky disk, we don't want the rest of the
1206 * sync process to block while we retry. But if we can't write a
1207 * single label out, we should retry with ZIO_FLAG_TRYHARD before
1208 * bailing out and declaring the pool faulted.
1209 */
1210 if (tryhard)
1211 flags |= ZIO_FLAG_TRYHARD;
1212
1213 ASSERT(ub->ub_txg <= txg);
1214
1215 /*
1216 * If this isn't a resync due to I/O errors,
1217 * and nothing changed in this transaction group,
1218 * and the vdev configuration hasn't changed,
1219 * then there's nothing to do.
1220 */
1221 if (ub->ub_txg < txg &&
1222 uberblock_update(ub, spa->spa_root_vdev, txg) == B_FALSE &&
1223 list_is_empty(&spa->spa_config_dirty_list))
1224 return (0);
1225
1226 if (txg > spa_freeze_txg(spa))
1227 return (0);
1228
1229 ASSERT(txg <= spa->spa_final_txg);
1230
1231 /*
1232 * Flush the write cache of every disk that's been written to
1233 * in this transaction group. This ensures that all blocks
1234 * written in this txg will be committed to stable storage
1235 * before any uberblock that references them.
1236 */
1237 zio = zio_root(spa, NULL, NULL, flags);
1238
1239 for (vd = txg_list_head(&spa->spa_vdev_txg_list, TXG_CLEAN(txg)); vd;
1240 vd = txg_list_next(&spa->spa_vdev_txg_list, vd, TXG_CLEAN(txg)))
1241 zio_flush(zio, vd);
1242
1243 (void) zio_wait(zio);
1244
1245 /*
1246 * Sync out the even labels (L0, L2) for every dirty vdev. If the
1247 * system dies in the middle of this process, that's OK: all of the
1248 * even labels that made it to disk will be newer than any uberblock,
1249 * and will therefore be considered invalid. The odd labels (L1, L3),
1250 * which have not yet been touched, will still be valid. We flush
1251 * the new labels to disk to ensure that all even-label updates
1252 * are committed to stable storage before the uberblock update.
1253 */
1254 if ((error = vdev_label_sync_list(spa, 0, txg, flags)) != 0)
1255 return (error);
1256
1257 /*
1258 * Sync the uberblocks to all vdevs in svd[].
1259 * If the system dies in the middle of this step, there are two cases
1260 * to consider, and the on-disk state is consistent either way:
1261 *
1262 * (1) If none of the new uberblocks made it to disk, then the
1263 * previous uberblock will be the newest, and the odd labels
1264 * (which had not yet been touched) will be valid with respect
1265 * to that uberblock.
1266 *
1267 * (2) If one or more new uberblocks made it to disk, then they
1268 * will be the newest, and the even labels (which had all
1269 * been successfully committed) will be valid with respect
1270 * to the new uberblocks.
1271 */
1272 if ((error = vdev_uberblock_sync_list(svd, svdcount, ub, flags)) != 0)
1273 return (error);
1274
1275 /*
1276 * Sync out odd labels for every dirty vdev. If the system dies
1277 * in the middle of this process, the even labels and the new
1278 * uberblocks will suffice to open the pool. The next time
1279 * the pool is opened, the first thing we'll do -- before any
1280 * user data is modified -- is mark every vdev dirty so that
1281 * all labels will be brought up to date. We flush the new labels
1282 * to disk to ensure that all odd-label updates are committed to
1283 * stable storage before the next transaction group begins.
1284 */
1285 return (vdev_label_sync_list(spa, 1, txg, flags));
1286}