diff options
Diffstat (limited to 'usr/src/uts/common/fs/zfs')
| -rw-r--r-- | usr/src/uts/common/fs/zfs/spa.c | 948 | ||||
| -rw-r--r-- | usr/src/uts/common/fs/zfs/spa_config.c | 12 | ||||
| -rw-r--r-- | usr/src/uts/common/fs/zfs/spa_misc.c | 27 | ||||
| -rw-r--r-- | usr/src/uts/common/fs/zfs/sys/spa.h | 6 | ||||
| -rw-r--r-- | usr/src/uts/common/fs/zfs/sys/spa_impl.h | 16 | ||||
| -rw-r--r-- | usr/src/uts/common/fs/zfs/sys/vdev.h | 9 | ||||
| -rw-r--r-- | usr/src/uts/common/fs/zfs/sys/vdev_impl.h | 1 | ||||
| -rw-r--r-- | usr/src/uts/common/fs/zfs/sys/vdev_removal.h | 2 | ||||
| -rw-r--r-- | usr/src/uts/common/fs/zfs/sys/zio.h | 11 | ||||
| -rw-r--r-- | usr/src/uts/common/fs/zfs/sys/zthr.h | 52 | ||||
| -rw-r--r-- | usr/src/uts/common/fs/zfs/vdev.c | 461 | ||||
| -rw-r--r-- | usr/src/uts/common/fs/zfs/vdev_indirect.c | 201 | ||||
| -rw-r--r-- | usr/src/uts/common/fs/zfs/vdev_label.c | 9 | ||||
| -rw-r--r-- | usr/src/uts/common/fs/zfs/vdev_mirror.c | 54 | ||||
| -rw-r--r-- | usr/src/uts/common/fs/zfs/vdev_root.c | 41 | ||||
| -rw-r--r-- | usr/src/uts/common/fs/zfs/zio.c | 98 | ||||
| -rw-r--r-- | usr/src/uts/common/fs/zfs/zthr.c | 319 |
17 files changed, 1630 insertions, 637 deletions
diff --git a/usr/src/uts/common/fs/zfs/spa.c b/usr/src/uts/common/fs/zfs/spa.c index 56d316a16d..a772c24e65 100644 --- a/usr/src/uts/common/fs/zfs/spa.c +++ b/usr/src/uts/common/fs/zfs/spa.c @@ -151,9 +151,8 @@ const zio_taskq_info_t zio_taskqs[ZIO_TYPES][ZIO_TASKQ_TYPES] = { static void spa_sync_version(void *arg, dmu_tx_t *tx); static void spa_sync_props(void *arg, dmu_tx_t *tx); static boolean_t spa_has_active_shared_spare(spa_t *spa); -static int spa_load_impl(spa_t *spa, uint64_t, nvlist_t *config, - spa_load_state_t state, spa_import_type_t type, boolean_t trust_config, - char **ereport); +static int spa_load_impl(spa_t *spa, spa_import_type_t type, char **ereport, + boolean_t reloading); static void spa_vdev_resilver_done(spa_t *spa); uint_t zio_taskq_batch_pct = 75; /* 1 thread per cpu in pset */ @@ -177,6 +176,54 @@ boolean_t spa_load_verify_dryrun = B_FALSE; #define TRYIMPORT_NAME "$import" /* + * For debugging purposes: print out vdev tree during pool import. + */ +boolean_t spa_load_print_vdev_tree = B_FALSE; + +/* + * A non-zero value for zfs_max_missing_tvds means that we allow importing + * pools with missing top-level vdevs. This is strictly intended for advanced + * pool recovery cases since missing data is almost inevitable. Pools with + * missing devices can only be imported read-only for safety reasons, and their + * fail-mode will be automatically set to "continue". + * + * With 1 missing vdev we should be able to import the pool and mount all + * datasets. User data that was not modified after the missing device has been + * added should be recoverable. This means that snapshots created prior to the + * addition of that device should be completely intact. + * + * With 2 missing vdevs, some datasets may fail to mount since there are + * dataset statistics that are stored as regular metadata. Some data might be + * recoverable if those vdevs were added recently. + * + * With 3 or more missing vdevs, the pool is severely damaged and MOS entries + * may be missing entirely. Chances of data recovery are very low. Note that + * there are also risks of performing an inadvertent rewind as we might be + * missing all the vdevs with the latest uberblocks. + */ +uint64_t zfs_max_missing_tvds = 0; + +/* + * The parameters below are similar to zfs_max_missing_tvds but are only + * intended for a preliminary open of the pool with an untrusted config which + * might be incomplete or out-dated. + * + * We are more tolerant for pools opened from a cachefile since we could have + * an out-dated cachefile where a device removal was not registered. + * We could have set the limit arbitrarily high but in the case where devices + * are really missing we would want to return the proper error codes; we chose + * SPA_DVAS_PER_BP - 1 so that some copies of the MOS would still be available + * and we get a chance to retrieve the trusted config. + */ +uint64_t zfs_max_missing_tvds_cachefile = SPA_DVAS_PER_BP - 1; +/* + * In the case where config was assembled by scanning device paths (/dev/dsks + * by default) we are less tolerant since all the existing devices should have + * been detected and we want spa_load to return the right error codes. + */ +uint64_t zfs_max_missing_tvds_scan = 0; + +/* * ========================================================================== * SPA properties routines * ========================================================================== @@ -1291,6 +1338,12 @@ spa_unload(spa_t *spa) spa->spa_vdev_removal = NULL; } + if (spa->spa_condense_zthr != NULL) { + ASSERT(!zthr_isrunning(spa->spa_condense_zthr)); + zthr_destroy(spa->spa_condense_zthr); + spa->spa_condense_zthr = NULL; + } + spa_condense_fini(spa); bpobj_close(&spa->spa_deferred_bpobj); @@ -1633,13 +1686,34 @@ load_nvlist(spa_t *spa, uint64_t obj, nvlist_t **value) } /* + * Concrete top-level vdevs that are not missing and are not logs. At every + * spa_sync we write new uberblocks to at least SPA_SYNC_MIN_VDEVS core tvds. + */ +static uint64_t +spa_healthy_core_tvds(spa_t *spa) +{ + vdev_t *rvd = spa->spa_root_vdev; + uint64_t tvds = 0; + + for (uint64_t i = 0; i < rvd->vdev_children; i++) { + vdev_t *vd = rvd->vdev_child[i]; + if (vd->vdev_islog) + continue; + if (vdev_is_concrete(vd) && !vdev_is_dead(vd)) + tvds++; + } + + return (tvds); +} + +/* * Checks to see if the given vdev could not be opened, in which case we post a * sysevent to notify the autoreplace code that the device has been removed. */ static void spa_check_removed(vdev_t *vd) { - for (int c = 0; c < vd->vdev_children; c++) + for (uint64_t c = 0; c < vd->vdev_children; c++) spa_check_removed(vd->vdev_child[c]); if (vd->vdev_ops->vdev_op_leaf && vdev_is_dead(vd) && @@ -1649,38 +1723,14 @@ spa_check_removed(vdev_t *vd) } } -static void -spa_config_valid_zaps(vdev_t *vd, vdev_t *mvd) -{ - ASSERT3U(vd->vdev_children, ==, mvd->vdev_children); - - vd->vdev_top_zap = mvd->vdev_top_zap; - vd->vdev_leaf_zap = mvd->vdev_leaf_zap; - - for (uint64_t i = 0; i < vd->vdev_children; i++) { - spa_config_valid_zaps(vd->vdev_child[i], mvd->vdev_child[i]); - } -} - -/* - * Validate the current config against the MOS config - */ -static boolean_t -spa_config_valid(spa_t *spa, nvlist_t *config) +static int +spa_check_for_missing_logs(spa_t *spa) { - vdev_t *mrvd, *rvd = spa->spa_root_vdev; - nvlist_t *nv; - - VERIFY(nvlist_lookup_nvlist(config, ZPOOL_CONFIG_VDEV_TREE, &nv) == 0); - - spa_config_enter(spa, SCL_ALL, FTAG, RW_WRITER); - VERIFY(spa_config_parse(spa, &mrvd, nv, NULL, 0, VDEV_ALLOC_LOAD) == 0); - - ASSERT3U(rvd->vdev_children, ==, mrvd->vdev_children); + vdev_t *rvd = spa->spa_root_vdev; /* * If we're doing a normal import, then build up any additional - * diagnostic information about missing devices in this config. + * diagnostic information about missing log devices. * We'll pass this up to the user for further processing. */ if (!(spa->spa_import_flags & ZFS_IMPORT_MISSING_LOG)) { @@ -1691,109 +1741,52 @@ spa_config_valid(spa_t *spa, nvlist_t *config) KM_SLEEP); VERIFY(nvlist_alloc(&nv, NV_UNIQUE_NAME, KM_SLEEP) == 0); - for (int c = 0; c < rvd->vdev_children; c++) { + for (uint64_t c = 0; c < rvd->vdev_children; c++) { vdev_t *tvd = rvd->vdev_child[c]; - vdev_t *mtvd = mrvd->vdev_child[c]; - if (tvd->vdev_ops == &vdev_missing_ops && - mtvd->vdev_ops != &vdev_missing_ops && - mtvd->vdev_islog) - child[idx++] = vdev_config_generate(spa, mtvd, - B_FALSE, 0); + /* + * We consider a device as missing only if it failed + * to open (i.e. offline or faulted is not considered + * as missing). + */ + if (tvd->vdev_islog && + tvd->vdev_state == VDEV_STATE_CANT_OPEN) { + child[idx++] = vdev_config_generate(spa, tvd, + B_FALSE, VDEV_CONFIG_MISSING); + } } - if (idx) { - VERIFY(nvlist_add_nvlist_array(nv, - ZPOOL_CONFIG_CHILDREN, child, idx) == 0); - VERIFY(nvlist_add_nvlist(spa->spa_load_info, - ZPOOL_CONFIG_MISSING_DEVICES, nv) == 0); + if (idx > 0) { + fnvlist_add_nvlist_array(nv, + ZPOOL_CONFIG_CHILDREN, child, idx); + fnvlist_add_nvlist(spa->spa_load_info, + ZPOOL_CONFIG_MISSING_DEVICES, nv); - for (int i = 0; i < idx; i++) + for (uint64_t i = 0; i < idx; i++) nvlist_free(child[i]); } nvlist_free(nv); kmem_free(child, rvd->vdev_children * sizeof (char **)); - } - - /* - * Compare the root vdev tree with the information we have - * from the MOS config (mrvd). Check each top-level vdev - * with the corresponding MOS config top-level (mtvd). - */ - for (int c = 0; c < rvd->vdev_children; c++) { - vdev_t *tvd = rvd->vdev_child[c]; - vdev_t *mtvd = mrvd->vdev_child[c]; - - /* - * Resolve any "missing" vdevs in the current configuration. - * Also trust the MOS config about any "indirect" vdevs. - * If we find that the MOS config has more accurate information - * about the top-level vdev then use that vdev instead. - */ - if ((tvd->vdev_ops == &vdev_missing_ops && - mtvd->vdev_ops != &vdev_missing_ops) || - (mtvd->vdev_ops == &vdev_indirect_ops && - tvd->vdev_ops != &vdev_indirect_ops)) { - /* - * Device specific actions. - */ - if (mtvd->vdev_islog) { - if (!(spa->spa_import_flags & - ZFS_IMPORT_MISSING_LOG)) { - continue; - } + if (idx > 0) { + spa_load_failed(spa, "some log devices are missing"); + return (SET_ERROR(ENXIO)); + } + } else { + for (uint64_t c = 0; c < rvd->vdev_children; c++) { + vdev_t *tvd = rvd->vdev_child[c]; + if (tvd->vdev_islog && + tvd->vdev_state == VDEV_STATE_CANT_OPEN) { spa_set_log_state(spa, SPA_LOG_CLEAR); - } else if (mtvd->vdev_ops != &vdev_indirect_ops) { - continue; - } - - /* - * Swap the missing vdev with the data we were - * able to obtain from the MOS config. - */ - vdev_remove_child(rvd, tvd); - vdev_remove_child(mrvd, mtvd); - - vdev_add_child(rvd, mtvd); - vdev_add_child(mrvd, tvd); - - vdev_reopen(rvd); - } else { - if (mtvd->vdev_islog) { - /* - * Load the slog device's state from the MOS - * config since it's possible that the label - * does not contain the most up-to-date - * information. - */ - vdev_load_log_state(tvd, mtvd); - vdev_reopen(tvd); + spa_load_note(spa, "some log devices are " + "missing, ZIL is dropped."); + break; } - - /* - * Per-vdev ZAP info is stored exclusively in the MOS. - */ - spa_config_valid_zaps(tvd, mtvd); } - - /* - * Never trust this info from userland; always use what's - * in the MOS. This prevents it from getting out of sync - * with the rest of the info in the MOS. - */ - tvd->vdev_removing = mtvd->vdev_removing; - tvd->vdev_indirect_config = mtvd->vdev_indirect_config; } - vdev_free(mrvd); - spa_config_exit(spa, SCL_ALL, FTAG); - - /* - * Ensure we were able to validate the config. - */ - return (rvd->vdev_guid_sum == spa->spa_uberblock.ub_guid_sum); + return (0); } /* @@ -2092,6 +2085,16 @@ spa_vdev_err(vdev_t *vdev, vdev_aux_t aux, int err) return (SET_ERROR(err)); } +static void +spa_spawn_aux_threads(spa_t *spa) +{ + ASSERT(spa_writeable(spa)); + + ASSERT(MUTEX_HELD(&spa_namespace_lock)); + + spa_start_indirect_condensing_thread(spa); +} + /* * Fix up config after a partly-completed split. This is done with the * ZPOOL_CONFIG_SPLIT nvlist. Both the splitting pool and the split-off @@ -2175,53 +2178,15 @@ spa_try_repair(spa_t *spa, nvlist_t *config) } static int -spa_load(spa_t *spa, spa_load_state_t state, spa_import_type_t type, - boolean_t trust_config) +spa_load(spa_t *spa, spa_load_state_t state, spa_import_type_t type) { - nvlist_t *config = spa->spa_config; char *ereport = FM_EREPORT_ZFS_POOL; - char *comment; int error; - uint64_t pool_guid; - nvlist_t *nvl; - - if (nvlist_lookup_uint64(config, ZPOOL_CONFIG_POOL_GUID, &pool_guid)) - return (SET_ERROR(EINVAL)); - - ASSERT(spa->spa_comment == NULL); - if (nvlist_lookup_string(config, ZPOOL_CONFIG_COMMENT, &comment) == 0) - spa->spa_comment = spa_strdup(comment); - - /* - * Versioning wasn't explicitly added to the label until later, so if - * it's not present treat it as the initial version. - */ - if (nvlist_lookup_uint64(config, ZPOOL_CONFIG_VERSION, - &spa->spa_ubsync.ub_version) != 0) - spa->spa_ubsync.ub_version = SPA_VERSION_INITIAL; - - (void) nvlist_lookup_uint64(config, ZPOOL_CONFIG_POOL_TXG, - &spa->spa_config_txg); - if ((state == SPA_LOAD_IMPORT || state == SPA_LOAD_TRYIMPORT) && - spa_guid_exists(pool_guid, 0)) { - error = SET_ERROR(EEXIST); - } else { - spa->spa_config_guid = pool_guid; - - if (nvlist_lookup_nvlist(config, ZPOOL_CONFIG_SPLIT, - &nvl) == 0) { - VERIFY(nvlist_dup(nvl, &spa->spa_config_splitting, - KM_SLEEP) == 0); - } - - nvlist_free(spa->spa_load_info); - spa->spa_load_info = fnvlist_alloc(); + spa->spa_load_state = state; - gethrestime(&spa->spa_loaded_ts); - error = spa_load_impl(spa, pool_guid, config, state, type, - trust_config, &ereport); - } + gethrestime(&spa->spa_loaded_ts); + error = spa_load_impl(spa, type, &ereport, B_FALSE); /* * Don't count references from objsets that are already closed @@ -2273,13 +2238,80 @@ vdev_count_verify_zaps(vdev_t *vd) } static int -spa_ld_parse_config(spa_t *spa, uint64_t pool_guid, nvlist_t *config, - spa_import_type_t type) +spa_verify_host(spa_t *spa, nvlist_t *mos_config) +{ + uint64_t hostid; + char *hostname; + uint64_t myhostid = 0; + + if (!spa_is_root(spa) && nvlist_lookup_uint64(mos_config, + ZPOOL_CONFIG_HOSTID, &hostid) == 0) { + hostname = fnvlist_lookup_string(mos_config, + ZPOOL_CONFIG_HOSTNAME); + + myhostid = zone_get_hostid(NULL); + + if (hostid != 0 && myhostid != 0 && hostid != myhostid) { + cmn_err(CE_WARN, "pool '%s' could not be " + "loaded as it was last accessed by " + "another system (host: %s hostid: 0x%llx). " + "See: http://illumos.org/msg/ZFS-8000-EY", + spa_name(spa), hostname, (u_longlong_t)hostid); + spa_load_failed(spa, "hostid verification failed: pool " + "last accessed by host: %s (hostid: 0x%llx)", + hostname, (u_longlong_t)hostid); + return (SET_ERROR(EBADF)); + } + } + + return (0); +} + +static int +spa_ld_parse_config(spa_t *spa, spa_import_type_t type) { int error = 0; - nvlist_t *nvtree = NULL; + nvlist_t *nvtree, *nvl, *config = spa->spa_config; int parse; vdev_t *rvd; + uint64_t pool_guid; + char *comment; + + /* + * Versioning wasn't explicitly added to the label until later, so if + * it's not present treat it as the initial version. + */ + if (nvlist_lookup_uint64(config, ZPOOL_CONFIG_VERSION, + &spa->spa_ubsync.ub_version) != 0) + spa->spa_ubsync.ub_version = SPA_VERSION_INITIAL; + + if (nvlist_lookup_uint64(config, ZPOOL_CONFIG_POOL_GUID, &pool_guid)) { + spa_load_failed(spa, "invalid config provided: '%s' missing", + ZPOOL_CONFIG_POOL_GUID); + return (SET_ERROR(EINVAL)); + } + + if ((spa->spa_load_state == SPA_LOAD_IMPORT || spa->spa_load_state == + SPA_LOAD_TRYIMPORT) && spa_guid_exists(pool_guid, 0)) { + spa_load_failed(spa, "a pool with guid %llu is already open", + (u_longlong_t)pool_guid); + return (SET_ERROR(EEXIST)); + } + + spa->spa_config_guid = pool_guid; + + nvlist_free(spa->spa_load_info); + spa->spa_load_info = fnvlist_alloc(); + + ASSERT(spa->spa_comment == NULL); + if (nvlist_lookup_string(config, ZPOOL_CONFIG_COMMENT, &comment) == 0) + spa->spa_comment = spa_strdup(comment); + + (void) nvlist_lookup_uint64(config, ZPOOL_CONFIG_POOL_TXG, + &spa->spa_config_txg); + + if (nvlist_lookup_nvlist(config, ZPOOL_CONFIG_SPLIT, &nvl) == 0) + spa->spa_config_splitting = fnvlist_dup(nvl); if (nvlist_lookup_nvlist(config, ZPOOL_CONFIG_VDEV_TREE, &nvtree)) { spa_load_failed(spa, "invalid config provided: '%s' missing", @@ -2287,9 +2319,6 @@ spa_ld_parse_config(spa_t *spa, uint64_t pool_guid, nvlist_t *config, return (SET_ERROR(EINVAL)); } - parse = (type == SPA_IMPORT_EXISTING ? - VDEV_ALLOC_LOAD : VDEV_ALLOC_SPLIT); - /* * Create "The Godfather" zio to hold all async IOs */ @@ -2307,6 +2336,8 @@ spa_ld_parse_config(spa_t *spa, uint64_t pool_guid, nvlist_t *config, * configuration requires knowing the version number. */ spa_config_enter(spa, SCL_ALL, FTAG, RW_WRITER); + parse = (type == SPA_IMPORT_EXISTING ? + VDEV_ALLOC_LOAD : VDEV_ALLOC_SPLIT); error = spa_config_parse(spa, &rvd, nvtree, NULL, 0, parse); spa_config_exit(spa, SCL_ALL, FTAG); @@ -2327,71 +2358,105 @@ spa_ld_parse_config(spa_t *spa, uint64_t pool_guid, nvlist_t *config, return (0); } +/* + * Recursively open all vdevs in the vdev tree. This function is called twice: + * first with the untrusted config, then with the trusted config. + */ static int spa_ld_open_vdevs(spa_t *spa) { int error = 0; + /* + * spa_missing_tvds_allowed defines how many top-level vdevs can be + * missing/unopenable for the root vdev to be still considered openable. + */ + if (spa->spa_trust_config) { + spa->spa_missing_tvds_allowed = zfs_max_missing_tvds; + } else if (spa->spa_config_source == SPA_CONFIG_SRC_CACHEFILE) { + spa->spa_missing_tvds_allowed = zfs_max_missing_tvds_cachefile; + } else if (spa->spa_config_source == SPA_CONFIG_SRC_SCAN) { + spa->spa_missing_tvds_allowed = zfs_max_missing_tvds_scan; + } else { + spa->spa_missing_tvds_allowed = 0; + } + + spa->spa_missing_tvds_allowed = + MAX(zfs_max_missing_tvds, spa->spa_missing_tvds_allowed); + spa_config_enter(spa, SCL_ALL, FTAG, RW_WRITER); error = vdev_open(spa->spa_root_vdev); spa_config_exit(spa, SCL_ALL, FTAG); + + if (spa->spa_missing_tvds != 0) { + spa_load_note(spa, "vdev tree has %lld missing top-level " + "vdevs.", (u_longlong_t)spa->spa_missing_tvds); + if (spa->spa_trust_config && (spa->spa_mode & FWRITE)) { + /* + * Although theoretically we could allow users to open + * incomplete pools in RW mode, we'd need to add a lot + * of extra logic (e.g. adjust pool space to account + * for missing vdevs). + * This limitation also prevents users from accidentally + * opening the pool in RW mode during data recovery and + * damaging it further. + */ + spa_load_note(spa, "pools with missing top-level " + "vdevs can only be opened in read-only mode."); + error = SET_ERROR(ENXIO); + } else { + spa_load_note(spa, "current settings allow for maximum " + "%lld missing top-level vdevs at this stage.", + (u_longlong_t)spa->spa_missing_tvds_allowed); + } + } if (error != 0) { spa_load_failed(spa, "unable to open vdev tree [error=%d]", error); } + if (spa->spa_missing_tvds != 0 || error != 0) + vdev_dbgmsg_print_tree(spa->spa_root_vdev, 2); return (error); } +/* + * We need to validate the vdev labels against the configuration that + * we have in hand. This function is called twice: first with an untrusted + * config, then with a trusted config. The validation is more strict when the + * config is trusted. + */ static int -spa_ld_validate_vdevs(spa_t *spa, spa_import_type_t type, - boolean_t trust_config) +spa_ld_validate_vdevs(spa_t *spa) { int error = 0; vdev_t *rvd = spa->spa_root_vdev; - /* - * We need to validate the vdev labels against the configuration that - * we have in hand, which is dependent on the setting of trust_config. - * If trust_config is true then we're validating the vdev labels based - * on that config. Otherwise, we're validating against the cached - * config (zpool.cache) that was read when we loaded the zfs module, and - * then later we will recursively call spa_load() and validate against - * the vdev config. - * - * If we're assembling a new pool that's been split off from an - * existing pool, the labels haven't yet been updated so we skip - * validation for now. - */ - if (type != SPA_IMPORT_ASSEMBLE) { - spa_config_enter(spa, SCL_ALL, FTAG, RW_WRITER); - error = vdev_validate(rvd, trust_config); - spa_config_exit(spa, SCL_ALL, FTAG); + spa_config_enter(spa, SCL_ALL, FTAG, RW_WRITER); + error = vdev_validate(rvd); + spa_config_exit(spa, SCL_ALL, FTAG); - if (error != 0) { - spa_load_failed(spa, "vdev_validate failed [error=%d]", - error); - return (error); - } + if (error != 0) { + spa_load_failed(spa, "vdev_validate failed [error=%d]", error); + return (error); + } - if (rvd->vdev_state <= VDEV_STATE_CANT_OPEN) { - spa_load_failed(spa, "cannot open vdev tree after " - "invalidating some vdevs"); - return (SET_ERROR(ENXIO)); - } + if (rvd->vdev_state <= VDEV_STATE_CANT_OPEN) { + spa_load_failed(spa, "cannot open vdev tree after invalidating " + "some vdevs"); + vdev_dbgmsg_print_tree(rvd, 2); + return (SET_ERROR(ENXIO)); } return (0); } static int -spa_ld_select_uberblock(spa_t *spa, nvlist_t *config, spa_import_type_t type, - boolean_t trust_config) +spa_ld_select_uberblock(spa_t *spa, spa_import_type_t type) { vdev_t *rvd = spa->spa_root_vdev; nvlist_t *label; uberblock_t *ub = &spa->spa_uberblock; - uint64_t children; /* * Find the best uberblock. @@ -2484,26 +2549,9 @@ spa_ld_select_uberblock(spa_t *spa, nvlist_t *config, spa_import_type_t type, nvlist_free(unsup_feat); } - /* - * If the vdev guid sum doesn't match the uberblock, we have an - * incomplete configuration. We first check to see if the pool - * is aware of the complete config (i.e ZPOOL_CONFIG_VDEV_CHILDREN). - * If it is, defer the vdev_guid_sum check till later so we - * can handle missing vdevs. - */ - if (nvlist_lookup_uint64(config, ZPOOL_CONFIG_VDEV_CHILDREN, - &children) != 0 && trust_config && type != SPA_IMPORT_ASSEMBLE && - rvd->vdev_guid_sum != ub->ub_guid_sum) { - spa_load_failed(spa, "guid sum in config doesn't match guid " - "sum in uberblock (%llu != %llu)", - (u_longlong_t)rvd->vdev_guid_sum, - (u_longlong_t)ub->ub_guid_sum); - return (spa_vdev_err(rvd, VDEV_AUX_BAD_GUID_SUM, ENXIO)); - } - if (type != SPA_IMPORT_ASSEMBLE && spa->spa_config_splitting) { spa_config_enter(spa, SCL_ALL, FTAG, RW_WRITER); - spa_try_repair(spa, config); + spa_try_repair(spa, spa->spa_config); spa_config_exit(spa, SCL_ALL, FTAG); nvlist_free(spa->spa_config_splitting); spa->spa_config_splitting = NULL; @@ -2542,49 +2590,167 @@ spa_ld_open_rootbp(spa_t *spa) } static int -spa_ld_validate_config(spa_t *spa, spa_import_type_t type) +spa_ld_load_trusted_config(spa_t *spa, spa_import_type_t type, + boolean_t reloading) { - vdev_t *rvd = spa->spa_root_vdev; + vdev_t *mrvd, *rvd = spa->spa_root_vdev; + nvlist_t *nv, *mos_config, *policy; + int error = 0, copy_error; + uint64_t healthy_tvds, healthy_tvds_mos; + uint64_t mos_config_txg; if (spa_dir_prop(spa, DMU_POOL_CONFIG, &spa->spa_config_object, B_TRUE) != 0) return (spa_vdev_err(rvd, VDEV_AUX_CORRUPT_DATA, EIO)); /* - * Validate the config, using the MOS config to fill in any - * information which might be missing. If we fail to validate - * the config then declare the pool unfit for use. If we're - * assembling a pool from a split, the log is not transferred - * over. + * If we're assembling a pool from a split, the config provided is + * already trusted so there is nothing to do. */ - if (type != SPA_IMPORT_ASSEMBLE) { - nvlist_t *mos_config; - if (load_nvlist(spa, spa->spa_config_object, &mos_config) - != 0) { - spa_load_failed(spa, "unable to retrieve MOS config"); - return (spa_vdev_err(rvd, VDEV_AUX_CORRUPT_DATA, EIO)); - } + if (type == SPA_IMPORT_ASSEMBLE) + return (0); + + healthy_tvds = spa_healthy_core_tvds(spa); + + if (load_nvlist(spa, spa->spa_config_object, &mos_config) + != 0) { + spa_load_failed(spa, "unable to retrieve MOS config"); + return (spa_vdev_err(rvd, VDEV_AUX_CORRUPT_DATA, EIO)); + } - if (!spa_config_valid(spa, mos_config)) { + /* + * If we are doing an open, pool owner wasn't verified yet, thus do + * the verification here. + */ + if (spa->spa_load_state == SPA_LOAD_OPEN) { + error = spa_verify_host(spa, mos_config); + if (error != 0) { nvlist_free(mos_config); - spa_load_failed(spa, "mismatch between config provided " - "and config stored in MOS"); - return (spa_vdev_err(rvd, VDEV_AUX_BAD_GUID_SUM, - ENXIO)); + return (error); } - nvlist_free(mos_config); + } + + nv = fnvlist_lookup_nvlist(mos_config, ZPOOL_CONFIG_VDEV_TREE); + + spa_config_enter(spa, SCL_ALL, FTAG, RW_WRITER); + /* + * Build a new vdev tree from the trusted config + */ + VERIFY(spa_config_parse(spa, &mrvd, nv, NULL, 0, VDEV_ALLOC_LOAD) == 0); + + /* + * Vdev paths in the MOS may be obsolete. If the untrusted config was + * obtained by scanning /dev/dsk, then it will have the right vdev + * paths. We update the trusted MOS config with this information. + * We first try to copy the paths with vdev_copy_path_strict, which + * succeeds only when both configs have exactly the same vdev tree. + * If that fails, we fall back to a more flexible method that has a + * best effort policy. + */ + copy_error = vdev_copy_path_strict(rvd, mrvd); + if (copy_error != 0 || spa_load_print_vdev_tree) { + spa_load_note(spa, "provided vdev tree:"); + vdev_dbgmsg_print_tree(rvd, 2); + spa_load_note(spa, "MOS vdev tree:"); + vdev_dbgmsg_print_tree(mrvd, 2); + } + if (copy_error != 0) { + spa_load_note(spa, "vdev_copy_path_strict failed, falling " + "back to vdev_copy_path_relaxed"); + vdev_copy_path_relaxed(rvd, mrvd); + } + + vdev_close(rvd); + vdev_free(rvd); + spa->spa_root_vdev = mrvd; + rvd = mrvd; + spa_config_exit(spa, SCL_ALL, FTAG); + + /* + * We will use spa_config if we decide to reload the spa or if spa_load + * fails and we rewind. We must thus regenerate the config using the + * MOS information with the updated paths. Rewind policy is an import + * setting and is not in the MOS. We copy it over to our new, trusted + * config. + */ + mos_config_txg = fnvlist_lookup_uint64(mos_config, + ZPOOL_CONFIG_POOL_TXG); + nvlist_free(mos_config); + mos_config = spa_config_generate(spa, NULL, mos_config_txg, B_FALSE); + if (nvlist_lookup_nvlist(spa->spa_config, ZPOOL_REWIND_POLICY, + &policy) == 0) + fnvlist_add_nvlist(mos_config, ZPOOL_REWIND_POLICY, policy); + spa_config_set(spa, mos_config); + spa->spa_config_source = SPA_CONFIG_SRC_MOS; + + /* + * Now that we got the config from the MOS, we should be more strict + * in checking blkptrs and can make assumptions about the consistency + * of the vdev tree. spa_trust_config must be set to true before opening + * vdevs in order for them to be writeable. + */ + spa->spa_trust_config = B_TRUE; + + /* + * Open and validate the new vdev tree + */ + error = spa_ld_open_vdevs(spa); + if (error != 0) + return (error); + + error = spa_ld_validate_vdevs(spa); + if (error != 0) + return (error); + + if (copy_error != 0 || spa_load_print_vdev_tree) { + spa_load_note(spa, "final vdev tree:"); + vdev_dbgmsg_print_tree(rvd, 2); + } + + if (spa->spa_load_state != SPA_LOAD_TRYIMPORT && + !spa->spa_extreme_rewind && zfs_max_missing_tvds == 0) { /* - * Now that we've validated the config, check the state of the - * root vdev. If it can't be opened, it indicates one or - * more toplevel vdevs are faulted. + * Sanity check to make sure that we are indeed loading the + * latest uberblock. If we missed SPA_SYNC_MIN_VDEVS tvds + * in the config provided and they happened to be the only ones + * to have the latest uberblock, we could involuntarily perform + * an extreme rewind. */ - if (rvd->vdev_state <= VDEV_STATE_CANT_OPEN) { - spa_load_failed(spa, "some top vdevs are unavailable"); - return (SET_ERROR(ENXIO)); + healthy_tvds_mos = spa_healthy_core_tvds(spa); + if (healthy_tvds_mos - healthy_tvds >= + SPA_SYNC_MIN_VDEVS) { + spa_load_note(spa, "config provided misses too many " + "top-level vdevs compared to MOS (%lld vs %lld). ", + (u_longlong_t)healthy_tvds, + (u_longlong_t)healthy_tvds_mos); + spa_load_note(spa, "vdev tree:"); + vdev_dbgmsg_print_tree(rvd, 2); + if (reloading) { + spa_load_failed(spa, "config was already " + "provided from MOS. Aborting."); + return (spa_vdev_err(rvd, + VDEV_AUX_CORRUPT_DATA, EIO)); + } + spa_load_note(spa, "spa must be reloaded using MOS " + "config"); + return (SET_ERROR(EAGAIN)); } } + error = spa_check_for_missing_logs(spa); + if (error != 0) + return (spa_vdev_err(rvd, VDEV_AUX_BAD_GUID_SUM, ENXIO)); + + if (rvd->vdev_guid_sum != spa->spa_uberblock.ub_guid_sum) { + spa_load_failed(spa, "uberblock guid sum doesn't match MOS " + "guid sum (%llu != %llu)", + (u_longlong_t)spa->spa_uberblock.ub_guid_sum, + (u_longlong_t)rvd->vdev_guid_sum); + return (spa_vdev_err(rvd, VDEV_AUX_BAD_GUID_SUM, + ENXIO)); + } + return (0); } @@ -2751,62 +2917,6 @@ spa_ld_load_special_directories(spa_t *spa) } static int -spa_ld_prepare_for_reload(spa_t *spa, int orig_mode) -{ - vdev_t *rvd = spa->spa_root_vdev; - - uint64_t hostid; - nvlist_t *policy = NULL; - nvlist_t *mos_config; - - if (load_nvlist(spa, spa->spa_config_object, &mos_config) != 0) { - spa_load_failed(spa, "unable to retrieve MOS config"); - return (spa_vdev_err(rvd, VDEV_AUX_CORRUPT_DATA, EIO)); - } - - if (!spa_is_root(spa) && nvlist_lookup_uint64(mos_config, - ZPOOL_CONFIG_HOSTID, &hostid) == 0) { - char *hostname; - unsigned long myhostid = 0; - - VERIFY(nvlist_lookup_string(mos_config, - ZPOOL_CONFIG_HOSTNAME, &hostname) == 0); - -#ifdef _KERNEL - myhostid = zone_get_hostid(NULL); -#else /* _KERNEL */ - /* - * We're emulating the system's hostid in userland, so - * we can't use zone_get_hostid(). - */ - (void) ddi_strtoul(hw_serial, NULL, 10, &myhostid); -#endif /* _KERNEL */ - if (hostid != 0 && myhostid != 0 && - hostid != myhostid) { - nvlist_free(mos_config); - cmn_err(CE_WARN, "pool '%s' could not be " - "loaded as it was last accessed by " - "another system (host: %s hostid: 0x%lx). " - "See: http://illumos.org/msg/ZFS-8000-EY", - spa_name(spa), hostname, - (unsigned long)hostid); - return (SET_ERROR(EBADF)); - } - } - if (nvlist_lookup_nvlist(spa->spa_config, - ZPOOL_REWIND_POLICY, &policy) == 0) - VERIFY(nvlist_add_nvlist(mos_config, - ZPOOL_REWIND_POLICY, policy) == 0); - - spa_config_set(spa, mos_config); - spa_unload(spa); - spa_deactivate(spa); - spa_activate(spa, orig_mode); - - return (0); -} - -static int spa_ld_get_props(spa_t *spa) { int error = 0; @@ -2933,6 +3043,19 @@ spa_ld_get_props(spa_t *spa) spa->spa_autoreplace = (autoreplace != 0); } + /* + * If we are importing a pool with missing top-level vdevs, + * we enforce that the pool doesn't panic or get suspended on + * error since the likelihood of missing data is extremely high. + */ + if (spa->spa_missing_tvds > 0 && + spa->spa_failmode != ZIO_FAILURE_MODE_CONTINUE && + spa->spa_load_state != SPA_LOAD_TRYIMPORT) { + spa_load_note(spa, "forcing failmode to 'continue' " + "as some top level vdevs are missing"); + spa->spa_failmode = ZIO_FAILURE_MODE_CONTINUE; + } + return (0); } @@ -3063,9 +3186,15 @@ spa_ld_verify_logs(spa_t *spa, spa_import_type_t type, char **ereport) if (type != SPA_IMPORT_ASSEMBLE && spa_writeable(spa)) { boolean_t missing = spa_check_logs(spa); if (missing) { - *ereport = FM_EREPORT_ZFS_LOG_REPLAY; - spa_load_failed(spa, "spa_check_logs failed"); - return (spa_vdev_err(rvd, VDEV_AUX_BAD_LOG, ENXIO)); + if (spa->spa_missing_tvds != 0) { + spa_load_note(spa, "spa_check_logs failed " + "so dropping the logs"); + } else { + *ereport = FM_EREPORT_ZFS_LOG_REPLAY; + spa_load_failed(spa, "spa_check_logs failed"); + return (spa_vdev_err(rvd, VDEV_AUX_BAD_LOG, + ENXIO)); + } } } @@ -3121,7 +3250,8 @@ spa_ld_claim_log_blocks(spa_t *spa) } static void -spa_ld_check_for_config_update(spa_t *spa, uint64_t config_cache_txg) +spa_ld_check_for_config_update(spa_t *spa, uint64_t config_cache_txg, + boolean_t reloading) { vdev_t *rvd = spa->spa_root_vdev; int need_update = B_FALSE; @@ -3133,7 +3263,7 @@ spa_ld_check_for_config_update(spa_t *spa, uint64_t config_cache_txg) * If this is a verbatim import, trust the current * in-core spa_config and update the disk labels. */ - if (config_cache_txg != spa->spa_config_txg || + if (reloading || config_cache_txg != spa->spa_config_txg || spa->spa_load_state == SPA_LOAD_IMPORT || spa->spa_load_state == SPA_LOAD_RECOVER || (spa->spa_import_flags & ZFS_IMPORT_VERBATIM)) @@ -3151,6 +3281,24 @@ spa_ld_check_for_config_update(spa_t *spa, uint64_t config_cache_txg) spa_async_request(spa, SPA_ASYNC_CONFIG_UPDATE); } +static void +spa_ld_prepare_for_reload(spa_t *spa) +{ + int mode = spa->spa_mode; + int async_suspended = spa->spa_async_suspended; + + spa_unload(spa); + spa_deactivate(spa); + spa_activate(spa, mode); + + /* + * We save the value of spa_async_suspended as it gets reset to 0 by + * spa_unload(). We want to restore it back to the original value before + * returning as we might be calling spa_async_resume() later. + */ + spa->spa_async_suspended = async_suspended; +} + /* * Load an existing storage pool, using the config provided. This config * describes which vdevs are part of the pool and is later validated against @@ -3158,32 +3306,35 @@ spa_ld_check_for_config_update(spa_t *spa, uint64_t config_cache_txg) * config stored in the MOS. */ static int -spa_load_impl(spa_t *spa, uint64_t pool_guid, nvlist_t *config, - spa_load_state_t state, spa_import_type_t type, boolean_t trust_config, - char **ereport) +spa_load_impl(spa_t *spa, spa_import_type_t type, char **ereport, + boolean_t reloading) { int error = 0; - uint64_t config_cache_txg = spa->spa_config_txg; - int orig_mode = spa->spa_mode; boolean_t missing_feat_write = B_FALSE; ASSERT(MUTEX_HELD(&spa_namespace_lock)); - - spa->spa_load_state = state; - spa_load_note(spa, "LOADING"); + ASSERT(spa->spa_config_source != SPA_CONFIG_SRC_NONE); /* - * If this is an untrusted config, first access the pool in read-only - * mode. We will then retrieve a trusted copy of the config from the MOS - * and use it to reopen the pool in read-write mode. + * Never trust the config that is provided unless we are assembling + * a pool following a split. + * This means don't trust blkptrs and the vdev tree in general. This + * also effectively puts the spa in read-only mode since + * spa_writeable() checks for spa_trust_config to be true. + * We will later load a trusted config from the MOS. */ - if (!trust_config) - spa->spa_mode = FREAD; + if (type != SPA_IMPORT_ASSEMBLE) + spa->spa_trust_config = B_FALSE; + + if (reloading) + spa_load_note(spa, "RELOADING"); + else + spa_load_note(spa, "LOADING"); /* * Parse the config provided to create a vdev tree. */ - error = spa_ld_parse_config(spa, pool_guid, config, type); + error = spa_ld_parse_config(spa, type); if (error != 0) return (error); @@ -3201,10 +3352,15 @@ spa_load_impl(spa_t *spa, uint64_t pool_guid, nvlist_t *config, /* * Read the label of each vdev and make sure that the GUIDs stored * there match the GUIDs in the config provided. + * If we're assembling a new pool that's been split off from an + * existing pool, the labels haven't yet been updated so we skip + * validation for now. */ - error = spa_ld_validate_vdevs(spa, type, trust_config); - if (error != 0) - return (error); + if (type != SPA_IMPORT_ASSEMBLE) { + error = spa_ld_validate_vdevs(spa); + if (error != 0) + return (error); + } /* * Read vdev labels to find the best uberblock (i.e. latest, unless @@ -3213,7 +3369,7 @@ spa_load_impl(spa_t *spa, uint64_t pool_guid, nvlist_t *config, * label with the best uberblock and verify that our version of zfs * supports them all. */ - error = spa_ld_select_uberblock(spa, config, type, trust_config); + error = spa_ld_select_uberblock(spa, type); if (error != 0) return (error); @@ -3227,13 +3383,21 @@ spa_load_impl(spa_t *spa, uint64_t pool_guid, nvlist_t *config, return (error); /* - * Retrieve the config stored in the MOS and use it to validate the - * config provided. Also extract some information from the MOS config - * to update our vdev tree. + * Retrieve the trusted config stored in the MOS and use it to create + * a new, exact version of the vdev tree, then reopen all vdevs. */ - error = spa_ld_validate_config(spa, type); - if (error != 0) + error = spa_ld_load_trusted_config(spa, type, reloading); + if (error == EAGAIN) { + VERIFY(!reloading); + /* + * Redo the loading process with the trusted config if it is + * too different from the untrusted config. + */ + spa_ld_prepare_for_reload(spa); + return (spa_load_impl(spa, type, ereport, B_TRUE)); + } else if (error != 0) { return (error); + } /* * Retrieve the mapping of indirect vdevs. Those vdevs were removed @@ -3264,19 +3428,6 @@ spa_load_impl(spa_t *spa, uint64_t pool_guid, nvlist_t *config, return (error); /* - * If the config provided is not trusted, discard it and use the config - * from the MOS to reload the pool. - */ - if (!trust_config) { - error = spa_ld_prepare_for_reload(spa, orig_mode); - if (error != 0) - return (error); - - spa_load_note(spa, "RELOADING"); - return (spa_load(spa, state, SPA_IMPORT_EXISTING, B_TRUE)); - } - - /* * Retrieve pool properties from the MOS. */ error = spa_ld_get_props(spa); @@ -3312,7 +3463,7 @@ spa_load_impl(spa_t *spa, uint64_t pool_guid, nvlist_t *config, return (error); if (missing_feat_write) { - ASSERT(state == SPA_LOAD_TRYIMPORT); + ASSERT(spa->spa_load_state == SPA_LOAD_TRYIMPORT); /* * At this point, we know that we can open the pool in @@ -3344,21 +3495,11 @@ spa_load_impl(spa_t *spa, uint64_t pool_guid, nvlist_t *config, * pool. If we are importing the pool in read-write mode, a few * additional steps must be performed to finish the import. */ - if (spa_writeable(spa) && (state == SPA_LOAD_RECOVER || + if (spa_writeable(spa) && (spa->spa_load_state == SPA_LOAD_RECOVER || spa->spa_load_max_txg == UINT64_MAX)) { - ASSERT(state != SPA_LOAD_TRYIMPORT); + uint64_t config_cache_txg = spa->spa_config_txg; - /* - * We must check this before we start the sync thread, because - * we only want to start a condense thread for condense - * operations that were in progress when the pool was - * imported. Once we start syncing, spa_sync() could - * initiate a condense (and start a thread for it). In - * that case it would be wrong to start a second - * condense thread. - */ - boolean_t condense_in_progress = - (spa->spa_condensing_indirect != NULL); + ASSERT(spa->spa_load_state != SPA_LOAD_TRYIMPORT); /* * Traverse the ZIL and claim all blocks. @@ -3385,7 +3526,8 @@ spa_load_impl(spa_t *spa, uint64_t pool_guid, nvlist_t *config, * next sync, we would update the config stored in vdev labels * and the cachefile (by default /etc/zfs/zpool.cache). */ - spa_ld_check_for_config_update(spa, config_cache_txg); + spa_ld_check_for_config_update(spa, config_cache_txg, + reloading); /* * Check all DTLs to see if anything needs resilvering. @@ -3411,15 +3553,9 @@ spa_load_impl(spa_t *spa, uint64_t pool_guid, nvlist_t *config, */ dsl_pool_clean_tmp_userrefs(spa->spa_dsl_pool); - /* - * Note: unlike condensing, we don't need an analogous - * "removal_in_progress" dance because no other thread - * can start a removal while we hold the spa_namespace_lock. - */ spa_restart_removal(spa); - if (condense_in_progress) - spa_condense_indirect_restart(spa); + spa_spawn_aux_threads(spa); } spa_load_note(spa, "LOADED"); @@ -3428,7 +3564,7 @@ spa_load_impl(spa_t *spa, uint64_t pool_guid, nvlist_t *config, } static int -spa_load_retry(spa_t *spa, spa_load_state_t state, int trust_config) +spa_load_retry(spa_t *spa, spa_load_state_t state) { int mode = spa->spa_mode; @@ -3443,7 +3579,7 @@ spa_load_retry(spa_t *spa, spa_load_state_t state, int trust_config) spa_load_note(spa, "spa_load_retry: rewind, max txg: %llu", (u_longlong_t)spa->spa_load_max_txg); - return (spa_load(spa, state, SPA_IMPORT_EXISTING, trust_config)); + return (spa_load(spa, state, SPA_IMPORT_EXISTING)); } /* @@ -3454,8 +3590,8 @@ spa_load_retry(spa_t *spa, spa_load_state_t state, int trust_config) * spa_load(). */ static int -spa_load_best(spa_t *spa, spa_load_state_t state, int trust_config, - uint64_t max_request, int rewind_flags) +spa_load_best(spa_t *spa, spa_load_state_t state, uint64_t max_request, + int rewind_flags) { nvlist_t *loadinfo = NULL; nvlist_t *config = NULL; @@ -3472,8 +3608,7 @@ spa_load_best(spa_t *spa, spa_load_state_t state, int trust_config, spa->spa_extreme_rewind = B_TRUE; } - load_error = rewind_error = spa_load(spa, state, SPA_IMPORT_EXISTING, - trust_config); + load_error = rewind_error = spa_load(spa, state, SPA_IMPORT_EXISTING); if (load_error == 0) return (0); @@ -3514,7 +3649,7 @@ spa_load_best(spa_t *spa, spa_load_state_t state, int trust_config, spa->spa_uberblock.ub_txg <= spa->spa_load_max_txg) { if (spa->spa_load_max_txg < safe_rewind_txg) spa->spa_extreme_rewind = B_TRUE; - rewind_error = spa_load_retry(spa, state, trust_config); + rewind_error = spa_load_retry(spa, state); } spa->spa_extreme_rewind = B_FALSE; @@ -3593,9 +3728,10 @@ spa_open_common(const char *pool, spa_t **spapp, void *tag, nvlist_t *nvpolicy, if (state != SPA_LOAD_RECOVER) spa->spa_last_ubsync_txg = spa->spa_load_txg = 0; + spa->spa_config_source = SPA_CONFIG_SRC_CACHEFILE; zfs_dbgmsg("spa_open_common: opening %s", pool); - error = spa_load_best(spa, state, B_FALSE, policy.zrp_txg, + error = spa_load_best(spa, state, policy.zrp_txg, policy.zrp_request); if (error == EBADF) { @@ -4328,6 +4464,8 @@ spa_create(const char *pool, nvlist_t *nvroot, nvlist_t *props, */ txg_wait_synced(spa->spa_dsl_pool, txg); + spa_spawn_aux_threads(spa); + spa_write_cachefile(spa, B_FALSE, B_TRUE); spa_event_notify(spa, NULL, NULL, ESC_ZFS_POOL_CREATE); @@ -4616,18 +4754,16 @@ spa_import(const char *pool, nvlist_t *config, nvlist_t *props, uint64_t flags) if (policy.zrp_request & ZPOOL_DO_REWIND) state = SPA_LOAD_RECOVER; - /* - * Pass off the heavy lifting to spa_load(). Pass TRUE for trust_config - * because the user-supplied config is actually the one to trust when - * doing an import. - */ - if (state != SPA_LOAD_RECOVER) - spa->spa_last_ubsync_txg = spa->spa_load_txg = 0; + spa->spa_config_source = SPA_CONFIG_SRC_TRYIMPORT; - zfs_dbgmsg("spa_import: importing %s%s", pool, - (state == SPA_LOAD_RECOVER) ? " (RECOVERY MODE)" : ""); - error = spa_load_best(spa, state, B_TRUE, policy.zrp_txg, - policy.zrp_request); + if (state != SPA_LOAD_RECOVER) { + spa->spa_last_ubsync_txg = spa->spa_load_txg = 0; + zfs_dbgmsg("spa_import: importing %s", pool); + } else { + zfs_dbgmsg("spa_import: importing %s, max_txg=%lld " + "(RECOVERY MODE)", pool, (longlong_t)policy.zrp_txg); + } + error = spa_load_best(spa, state, policy.zrp_txg, policy.zrp_request); /* * Propagate anything learned while loading the pool and pass it @@ -4745,10 +4881,11 @@ nvlist_t * spa_tryimport(nvlist_t *tryconfig) { nvlist_t *config = NULL; - char *poolname; + char *poolname, *cachefile; spa_t *spa; uint64_t state; int error; + zpool_rewind_policy_t policy; if (nvlist_lookup_string(tryconfig, ZPOOL_CONFIG_POOL_NAME, &poolname)) return (NULL); @@ -4763,14 +4900,30 @@ spa_tryimport(nvlist_t *tryconfig) spa = spa_add(TRYIMPORT_NAME, tryconfig, NULL); spa_activate(spa, FREAD); - zfs_dbgmsg("spa_tryimport: importing %s", poolname); - /* - * Pass off the heavy lifting to spa_load(). - * Pass TRUE for trust_config because the user-supplied config - * is actually the one to trust when doing an import. + * Rewind pool if a max txg was provided. Note that even though we + * retrieve the complete rewind policy, only the rewind txg is relevant + * for tryimport. */ - error = spa_load(spa, SPA_LOAD_TRYIMPORT, SPA_IMPORT_EXISTING, B_TRUE); + zpool_get_rewind_policy(spa->spa_config, &policy); + if (policy.zrp_txg != UINT64_MAX) { + spa->spa_load_max_txg = policy.zrp_txg; + spa->spa_extreme_rewind = B_TRUE; + zfs_dbgmsg("spa_tryimport: importing %s, max_txg=%lld", + poolname, (longlong_t)policy.zrp_txg); + } else { + zfs_dbgmsg("spa_tryimport: importing %s", poolname); + } + + if (nvlist_lookup_string(tryconfig, ZPOOL_CONFIG_CACHEFILE, &cachefile) + == 0) { + zfs_dbgmsg("spa_tryimport: using cachefile '%s'", cachefile); + spa->spa_config_source = SPA_CONFIG_SRC_CACHEFILE; + } else { + spa->spa_config_source = SPA_CONFIG_SRC_SCAN; + } + + error = spa_load(spa, SPA_LOAD_TRYIMPORT, SPA_IMPORT_EXISTING); /* * If 'tryconfig' was at least parsable, return the current config. @@ -5775,8 +5928,10 @@ spa_vdev_split_mirror(spa_t *spa, char *newname, nvlist_t *config, spa_activate(newspa, spa_mode_global); spa_async_suspend(newspa); + newspa->spa_config_source = SPA_CONFIG_SRC_SPLIT; + /* create the new pool from the disks of the original pool */ - error = spa_load(newspa, SPA_LOAD_IMPORT, SPA_IMPORT_ASSEMBLE, B_TRUE); + error = spa_load(newspa, SPA_LOAD_IMPORT, SPA_IMPORT_ASSEMBLE); if (error) goto out; @@ -6250,12 +6405,15 @@ spa_async_suspend(spa_t *spa) { mutex_enter(&spa->spa_async_lock); spa->spa_async_suspended++; - while (spa->spa_async_thread != NULL || - spa->spa_condense_thread != NULL) + while (spa->spa_async_thread != NULL) cv_wait(&spa->spa_async_cv, &spa->spa_async_lock); mutex_exit(&spa->spa_async_lock); spa_vdev_remove_suspend(spa); + + zthr_t *condense_thread = spa->spa_condense_zthr; + if (condense_thread != NULL && zthr_isrunning(condense_thread)) + VERIFY0(zthr_cancel(condense_thread)); } void @@ -6266,6 +6424,10 @@ spa_async_resume(spa_t *spa) spa->spa_async_suspended--; mutex_exit(&spa->spa_async_lock); spa_restart_removal(spa); + + zthr_t *condense_thread = spa->spa_condense_zthr; + if (condense_thread != NULL && !zthr_isrunning(condense_thread)) + zthr_resume(condense_thread); } static boolean_t @@ -7079,7 +7241,7 @@ spa_sync(spa_t *spa, uint64_t txg) spa_config_enter(spa, SCL_STATE, FTAG, RW_READER); if (list_is_empty(&spa->spa_config_dirty_list)) { - vdev_t *svd[SPA_DVAS_PER_BP]; + vdev_t *svd[SPA_SYNC_MIN_VDEVS]; int svdcount = 0; int children = rvd->vdev_children; int c0 = spa_get_random(children); @@ -7090,7 +7252,7 @@ spa_sync(spa_t *spa, uint64_t txg) !vdev_is_concrete(vd)) continue; svd[svdcount++] = vd; - if (svdcount == SPA_DVAS_PER_BP) + if (svdcount == SPA_SYNC_MIN_VDEVS) break; } error = vdev_config_sync(svd, svdcount, txg); diff --git a/usr/src/uts/common/fs/zfs/spa_config.c b/usr/src/uts/common/fs/zfs/spa_config.c index e118317adb..1fe675337f 100644 --- a/usr/src/uts/common/fs/zfs/spa_config.c +++ b/usr/src/uts/common/fs/zfs/spa_config.c @@ -339,7 +339,8 @@ void spa_config_set(spa_t *spa, nvlist_t *config) { mutex_enter(&spa->spa_props_lock); - nvlist_free(spa->spa_config); + if (spa->spa_config != NULL && spa->spa_config != config) + nvlist_free(spa->spa_config); spa->spa_config = config; mutex_exit(&spa->spa_props_lock); } @@ -386,15 +387,8 @@ spa_config_generate(spa_t *spa, vdev_t *vd, uint64_t txg, int getstats) spa->spa_comment); } -#ifdef _KERNEL hostid = zone_get_hostid(NULL); -#else /* _KERNEL */ - /* - * We're emulating the system's hostid in userland, so we can't use - * zone_get_hostid(). - */ - (void) ddi_strtoul(hw_serial, NULL, 10, &hostid); -#endif /* _KERNEL */ + if (hostid != 0) { fnvlist_add_uint64(config, ZPOOL_CONFIG_HOSTID, hostid); } diff --git a/usr/src/uts/common/fs/zfs/spa_misc.c b/usr/src/uts/common/fs/zfs/spa_misc.c index 33857db7ac..e79664dd33 100644 --- a/usr/src/uts/common/fs/zfs/spa_misc.c +++ b/usr/src/uts/common/fs/zfs/spa_misc.c @@ -365,7 +365,8 @@ spa_load_failed(spa_t *spa, const char *fmt, ...) (void) vsnprintf(buf, sizeof (buf), fmt, adx); va_end(adx); - zfs_dbgmsg("spa_load(%s): FAILED: %s", spa->spa_name, buf); + zfs_dbgmsg("spa_load(%s, config %s): FAILED: %s", spa->spa_name, + spa->spa_trust_config ? "trusted" : "untrusted", buf); } /*PRINTFLIKE2*/ @@ -379,7 +380,8 @@ spa_load_note(spa_t *spa, const char *fmt, ...) (void) vsnprintf(buf, sizeof (buf), fmt, adx); va_end(adx); - zfs_dbgmsg("spa_load(%s): %s", spa->spa_name, buf); + zfs_dbgmsg("spa_load(%s, config %s): %s", spa->spa_name, + spa->spa_trust_config ? "trusted" : "untrusted", buf); } /* @@ -620,6 +622,7 @@ spa_add(const char *name, nvlist_t *config, const char *altroot) spa->spa_load_max_txg = UINT64_MAX; spa->spa_proc = &p0; spa->spa_proc_state = SPA_PROC_NONE; + spa->spa_trust_config = B_TRUE; hdlr.cyh_func = spa_deadman; hdlr.cyh_arg = spa; @@ -2013,7 +2016,7 @@ spa_is_root(spa_t *spa) boolean_t spa_writeable(spa_t *spa) { - return (!!(spa->spa_mode & FWRITE)); + return (!!(spa->spa_mode & FWRITE) && spa->spa_trust_config); } /* @@ -2161,3 +2164,21 @@ spa_get_last_removal_txg(spa_t *spa) return (ret); } + +boolean_t +spa_trust_config(spa_t *spa) +{ + return (spa->spa_trust_config); +} + +uint64_t +spa_missing_tvds_allowed(spa_t *spa) +{ + return (spa->spa_missing_tvds_allowed); +} + +void +spa_set_missing_tvds(spa_t *spa, uint64_t missing) +{ + spa->spa_missing_tvds = missing; +} diff --git a/usr/src/uts/common/fs/zfs/sys/spa.h b/usr/src/uts/common/fs/zfs/sys/spa.h index aff251b80e..d6d48ddfbf 100644 --- a/usr/src/uts/common/fs/zfs/sys/spa.h +++ b/usr/src/uts/common/fs/zfs/sys/spa.h @@ -322,6 +322,7 @@ typedef enum bp_embedded_type { #define SPA_BLKPTRSHIFT 7 /* blkptr_t is 128 bytes */ #define SPA_DVAS_PER_BP 3 /* Number of DVAs in a bp */ +#define SPA_SYNC_MIN_VDEVS 3 /* min vdevs to update during sync */ /* * A block is a hole when it has either 1) never been written to, or @@ -836,11 +837,16 @@ extern boolean_t spa_writeable(spa_t *spa); extern boolean_t spa_has_pending_synctask(spa_t *spa); extern int spa_maxblocksize(spa_t *spa); extern void zfs_blkptr_verify(spa_t *spa, const blkptr_t *bp); +extern boolean_t zfs_dva_valid(spa_t *spa, const dva_t *dva, + const blkptr_t *bp); typedef void (*spa_remap_cb_t)(uint64_t vdev, uint64_t offset, uint64_t size, void *arg); extern boolean_t spa_remap_blkptr(spa_t *spa, blkptr_t *bp, spa_remap_cb_t callback, void *arg); extern uint64_t spa_get_last_removal_txg(spa_t *spa); +extern boolean_t spa_trust_config(spa_t *spa); +extern uint64_t spa_missing_tvds_allowed(spa_t *spa); +extern void spa_set_missing_tvds(spa_t *spa, uint64_t missing); extern int spa_mode(spa_t *spa); extern uint64_t zfs_strtonum(const char *str, char **nptr); diff --git a/usr/src/uts/common/fs/zfs/sys/spa_impl.h b/usr/src/uts/common/fs/zfs/sys/spa_impl.h index 955568c05b..1e440a6767 100644 --- a/usr/src/uts/common/fs/zfs/sys/spa_impl.h +++ b/usr/src/uts/common/fs/zfs/sys/spa_impl.h @@ -43,6 +43,7 @@ #include <sys/bplist.h> #include <sys/bpobj.h> #include <sys/zfeature.h> +#include <sys/zthr.h> #include <zfeature_common.h> #ifdef __cplusplus @@ -181,6 +182,15 @@ typedef enum spa_all_vdev_zap_action { AVZ_ACTION_INITIALIZE } spa_avz_action_t; +typedef enum spa_config_source { + SPA_CONFIG_SRC_NONE = 0, + SPA_CONFIG_SRC_SCAN, /* scan of path (default: /dev/dsk) */ + SPA_CONFIG_SRC_CACHEFILE, /* any cachefile */ + SPA_CONFIG_SRC_TRYIMPORT, /* returned from call to tryimport */ + SPA_CONFIG_SRC_SPLIT, /* new pool in a pool split */ + SPA_CONFIG_SRC_MOS /* MOS, but not always from right txg */ +} spa_config_source_t; + struct spa { /* * Fields protected by spa_namespace_lock. @@ -199,6 +209,8 @@ struct spa { uint8_t spa_sync_on; /* sync threads are running */ spa_load_state_t spa_load_state; /* current load operation */ boolean_t spa_indirect_vdevs_loaded; /* mappings loaded? */ + boolean_t spa_trust_config; /* do we trust vdev tree? */ + spa_config_source_t spa_config_source; /* where config comes from? */ uint64_t spa_import_flags; /* import specific flags */ spa_taskqs_t spa_zio_taskq[ZIO_TYPES][ZIO_TASKQ_TYPES]; dsl_pool_t *spa_dsl_pool; @@ -259,13 +271,15 @@ struct spa { int spa_async_suspended; /* async tasks suspended */ kcondvar_t spa_async_cv; /* wait for thread_exit() */ uint16_t spa_async_tasks; /* async task mask */ + uint64_t spa_missing_tvds; /* unopenable tvds on load */ + uint64_t spa_missing_tvds_allowed; /* allow loading spa? */ spa_removing_phys_t spa_removing_phys; spa_vdev_removal_t *spa_vdev_removal; spa_condensing_indirect_phys_t spa_condensing_indirect_phys; spa_condensing_indirect_t *spa_condensing_indirect; - kthread_t *spa_condense_thread; /* thread doing condense. */ + zthr_t *spa_condense_zthr; /* zthr doing condense. */ char *spa_root; /* alternate root directory */ uint64_t spa_ena; /* spa-wide ereport ENA */ diff --git a/usr/src/uts/common/fs/zfs/sys/vdev.h b/usr/src/uts/common/fs/zfs/sys/vdev.h index 1297590eb3..91d4db62f9 100644 --- a/usr/src/uts/common/fs/zfs/sys/vdev.h +++ b/usr/src/uts/common/fs/zfs/sys/vdev.h @@ -48,10 +48,13 @@ typedef enum vdev_dtl_type { extern boolean_t zfs_nocacheflush; extern void vdev_dbgmsg(vdev_t *vd, const char *fmt, ...); +extern void vdev_dbgmsg_print_tree(vdev_t *, int); extern int vdev_open(vdev_t *); extern void vdev_open_children(vdev_t *); extern boolean_t vdev_uses_zvols(vdev_t *); -extern int vdev_validate(vdev_t *, boolean_t); +extern int vdev_validate(vdev_t *); +extern int vdev_copy_path_strict(vdev_t *, vdev_t *); +extern void vdev_copy_path_relaxed(vdev_t *, vdev_t *); extern void vdev_close(vdev_t *); extern int vdev_create(vdev_t *, uint64_t txg, boolean_t isreplace); extern void vdev_reopen(vdev_t *); @@ -99,6 +102,7 @@ extern void vdev_scan_stat_init(vdev_t *vd); extern void vdev_propagate_state(vdev_t *vd); extern void vdev_set_state(vdev_t *vd, boolean_t isopen, vdev_state_t state, vdev_aux_t aux); +extern boolean_t vdev_children_are_offline(vdev_t *vd); extern void vdev_space_update(vdev_t *vd, int64_t alloc_delta, int64_t defer_delta, int64_t space_delta); @@ -140,7 +144,8 @@ typedef enum vdev_config_flag { VDEV_CONFIG_SPARE = 1 << 0, VDEV_CONFIG_L2CACHE = 1 << 1, VDEV_CONFIG_REMOVING = 1 << 2, - VDEV_CONFIG_MOS = 1 << 3 + VDEV_CONFIG_MOS = 1 << 3, + VDEV_CONFIG_MISSING = 1 << 4 } vdev_config_flag_t; extern void vdev_top_config_generate(spa_t *spa, nvlist_t *config); diff --git a/usr/src/uts/common/fs/zfs/sys/vdev_impl.h b/usr/src/uts/common/fs/zfs/sys/vdev_impl.h index 12d8df93af..5035e6ab9d 100644 --- a/usr/src/uts/common/fs/zfs/sys/vdev_impl.h +++ b/usr/src/uts/common/fs/zfs/sys/vdev_impl.h @@ -406,7 +406,6 @@ extern void vdev_remove_parent(vdev_t *cvd); /* * vdev sync load and sync */ -extern void vdev_load_log_state(vdev_t *nvd, vdev_t *ovd); extern boolean_t vdev_log_state_valid(vdev_t *vd); extern int vdev_load(vdev_t *vd); extern int vdev_dtl_load(vdev_t *vd); diff --git a/usr/src/uts/common/fs/zfs/sys/vdev_removal.h b/usr/src/uts/common/fs/zfs/sys/vdev_removal.h index 5b1e3056be..45cf4d8ec5 100644 --- a/usr/src/uts/common/fs/zfs/sys/vdev_removal.h +++ b/usr/src/uts/common/fs/zfs/sys/vdev_removal.h @@ -76,7 +76,7 @@ extern int spa_remove_init(spa_t *); extern void spa_restart_removal(spa_t *); extern int spa_condense_init(spa_t *); extern void spa_condense_fini(spa_t *); -extern void spa_condense_indirect_restart(spa_t *); +extern void spa_start_indirect_condensing_thread(spa_t *); extern void spa_vdev_condense_suspend(spa_t *); extern int spa_vdev_remove(spa_t *, uint64_t, boolean_t); extern void free_from_removing_vdev(vdev_t *, uint64_t, uint64_t, uint64_t); diff --git a/usr/src/uts/common/fs/zfs/sys/zio.h b/usr/src/uts/common/fs/zfs/sys/zio.h index 48121d4c50..828cc2a92b 100644 --- a/usr/src/uts/common/fs/zfs/sys/zio.h +++ b/usr/src/uts/common/fs/zfs/sys/zio.h @@ -208,6 +208,9 @@ enum zio_flag { (((zio)->io_flags & ZIO_FLAG_VDEV_INHERIT) | \ ZIO_FLAG_DONT_PROPAGATE | ZIO_FLAG_CANFAIL) +#define ZIO_CHILD_BIT(x) (1 << (x)) +#define ZIO_CHILD_BIT_IS_SET(val, x) ((val) & (1 << (x))) + enum zio_child { ZIO_CHILD_VDEV = 0, ZIO_CHILD_GANG, @@ -216,6 +219,14 @@ enum zio_child { ZIO_CHILD_TYPES }; +#define ZIO_CHILD_VDEV_BIT ZIO_CHILD_BIT(ZIO_CHILD_VDEV) +#define ZIO_CHILD_GANG_BIT ZIO_CHILD_BIT(ZIO_CHILD_GANG) +#define ZIO_CHILD_DDT_BIT ZIO_CHILD_BIT(ZIO_CHILD_DDT) +#define ZIO_CHILD_LOGICAL_BIT ZIO_CHILD_BIT(ZIO_CHILD_LOGICAL) +#define ZIO_CHILD_ALL_BITS \ + (ZIO_CHILD_VDEV_BIT | ZIO_CHILD_GANG_BIT | \ + ZIO_CHILD_DDT_BIT | ZIO_CHILD_LOGICAL_BIT) + enum zio_wait_type { ZIO_WAIT_READY = 0, ZIO_WAIT_DONE, diff --git a/usr/src/uts/common/fs/zfs/sys/zthr.h b/usr/src/uts/common/fs/zfs/sys/zthr.h new file mode 100644 index 0000000000..6bfb6b6c0d --- /dev/null +++ b/usr/src/uts/common/fs/zfs/sys/zthr.h @@ -0,0 +1,52 @@ +/* + * CDDL HEADER START + * + * This file and its contents are supplied under the terms of the + * Common Development and Distribution License ("CDDL"), version 1.0. + * You may only use this file in accordance with the terms of version + * 1.0 of the CDDL. + * + * A full copy of the text of the CDDL should have accompanied this + * source. A copy of the CDDL is also available via the Internet at + * http://www.illumos.org/license/CDDL. + * + * CDDL HEADER END + */ + + +/* + * Copyright (c) 2017 by Delphix. All rights reserved. + */ + +#ifndef _SYS_ZTHR_H +#define _SYS_ZTHR_H + +typedef struct zthr zthr_t; +typedef int (zthr_func_t)(void *, zthr_t *); +typedef boolean_t (zthr_checkfunc_t)(void *, zthr_t *); + +struct zthr { + kthread_t *zthr_thread; + kmutex_t zthr_lock; + kcondvar_t zthr_cv; + boolean_t zthr_cancel; + + zthr_checkfunc_t *zthr_checkfunc; + zthr_func_t *zthr_func; + void *zthr_arg; + int zthr_rc; +}; + +extern zthr_t *zthr_create(zthr_checkfunc_t checkfunc, + zthr_func_t *func, void *arg); +extern void zthr_exit(zthr_t *t, int rc); +extern void zthr_destroy(zthr_t *t); + +extern void zthr_wakeup(zthr_t *t); +extern int zthr_cancel(zthr_t *t); +extern void zthr_resume(zthr_t *t); + +extern boolean_t zthr_iscancelled(zthr_t *t); +extern boolean_t zthr_isrunning(zthr_t *t); + +#endif /* _SYS_ZTHR_H */ diff --git a/usr/src/uts/common/fs/zfs/vdev.c b/usr/src/uts/common/fs/zfs/vdev.c index 656008bf9e..73306aec85 100644 --- a/usr/src/uts/common/fs/zfs/vdev.c +++ b/usr/src/uts/common/fs/zfs/vdev.c @@ -77,6 +77,8 @@ int zfs_scrub_limit = 10; */ int metaslabs_per_vdev = 200; +boolean_t vdev_validate_skip = B_FALSE; + /*PRINTFLIKE2*/ void vdev_dbgmsg(vdev_t *vd, const char *fmt, ...) @@ -99,6 +101,57 @@ vdev_dbgmsg(vdev_t *vd, const char *fmt, ...) } } +void +vdev_dbgmsg_print_tree(vdev_t *vd, int indent) +{ + char state[20]; + + if (vd->vdev_ishole || vd->vdev_ops == &vdev_missing_ops) { + zfs_dbgmsg("%*svdev %u: %s", indent, "", vd->vdev_id, + vd->vdev_ops->vdev_op_type); + return; + } + + switch (vd->vdev_state) { + case VDEV_STATE_UNKNOWN: + (void) snprintf(state, sizeof (state), "unknown"); + break; + case VDEV_STATE_CLOSED: + (void) snprintf(state, sizeof (state), "closed"); + break; + case VDEV_STATE_OFFLINE: + (void) snprintf(state, sizeof (state), "offline"); + break; + case VDEV_STATE_REMOVED: + (void) snprintf(state, sizeof (state), "removed"); + break; + case VDEV_STATE_CANT_OPEN: + (void) snprintf(state, sizeof (state), "can't open"); + break; + case VDEV_STATE_FAULTED: + (void) snprintf(state, sizeof (state), "faulted"); + break; + case VDEV_STATE_DEGRADED: + (void) snprintf(state, sizeof (state), "degraded"); + break; + case VDEV_STATE_HEALTHY: + (void) snprintf(state, sizeof (state), "healthy"); + break; + default: + (void) snprintf(state, sizeof (state), "<state %u>", + (uint_t)vd->vdev_state); + } + + zfs_dbgmsg("%*svdev %u: %s%s, guid: %llu, path: %s, %s", indent, + "", vd->vdev_id, vd->vdev_ops->vdev_op_type, + vd->vdev_islog ? " (log)" : "", + (u_longlong_t)vd->vdev_guid, + vd->vdev_path ? vd->vdev_path : "N/A", state); + + for (uint64_t i = 0; i < vd->vdev_children; i++) + vdev_dbgmsg_print_tree(vd->vdev_child[i], indent + 2); +} + /* * Given a vdev type, return the appropriate ops vector. */ @@ -1287,8 +1340,13 @@ vdev_open(vdev_t *vd) vd->vdev_stat.vs_aux != VDEV_AUX_OPEN_FAILED) vd->vdev_removed = B_FALSE; - vdev_set_state(vd, B_TRUE, VDEV_STATE_CANT_OPEN, - vd->vdev_stat.vs_aux); + if (vd->vdev_stat.vs_aux == VDEV_AUX_CHILDREN_OFFLINE) { + vdev_set_state(vd, B_TRUE, VDEV_STATE_OFFLINE, + vd->vdev_stat.vs_aux); + } else { + vdev_set_state(vd, B_TRUE, VDEV_STATE_CANT_OPEN, + vd->vdev_stat.vs_aux); + } return (error); } @@ -1453,29 +1511,29 @@ vdev_open(vdev_t *vd) /* * Called once the vdevs are all opened, this routine validates the label - * contents. This needs to be done before vdev_load() so that we don't + * contents. This needs to be done before vdev_load() so that we don't * inadvertently do repair I/Os to the wrong device. * - * If 'strict' is false ignore the spa guid check. This is necessary because - * if the machine crashed during a re-guid the new guid might have been written - * to all of the vdev labels, but not the cached config. The strict check - * will be performed when the pool is opened again using the mos config. - * * This function will only return failure if one of the vdevs indicates that it * has since been destroyed or exported. This is only possible if * /etc/zfs/zpool.cache was readonly at the time. Otherwise, the vdev state * will be updated but the function will return 0. */ int -vdev_validate(vdev_t *vd, boolean_t strict) +vdev_validate(vdev_t *vd) { spa_t *spa = vd->vdev_spa; nvlist_t *label; - uint64_t guid = 0, top_guid; + uint64_t guid = 0, aux_guid = 0, top_guid; uint64_t state; + nvlist_t *nvl; + uint64_t txg; - for (int c = 0; c < vd->vdev_children; c++) - if (vdev_validate(vd->vdev_child[c], strict) != 0) + if (vdev_validate_skip) + return (0); + + for (uint64_t c = 0; c < vd->vdev_children; c++) + if (vdev_validate(vd->vdev_child[c]) != 0) return (SET_ERROR(EBADF)); /* @@ -1483,115 +1541,276 @@ vdev_validate(vdev_t *vd, boolean_t strict) * any further validation. Otherwise, label I/O will fail and we will * overwrite the previous state. */ - if (vd->vdev_ops->vdev_op_leaf && vdev_readable(vd)) { - uint64_t aux_guid = 0; - nvlist_t *nvl; - uint64_t txg = spa_last_synced_txg(spa) != 0 ? - spa_last_synced_txg(spa) : -1ULL; + if (!vd->vdev_ops->vdev_op_leaf || !vdev_readable(vd)) + return (0); - if ((label = vdev_label_read_config(vd, txg)) == NULL) { - vdev_set_state(vd, B_TRUE, VDEV_STATE_CANT_OPEN, - VDEV_AUX_BAD_LABEL); - vdev_dbgmsg(vd, "vdev_validate: failed reading config"); - return (0); - } + /* + * If we are performing an extreme rewind, we allow for a label that + * was modified at a point after the current txg. + */ + if (spa->spa_extreme_rewind || spa_last_synced_txg(spa) == 0) + txg = UINT64_MAX; + else + txg = spa_last_synced_txg(spa); - /* - * Determine if this vdev has been split off into another - * pool. If so, then refuse to open it. - */ - if (nvlist_lookup_uint64(label, ZPOOL_CONFIG_SPLIT_GUID, - &aux_guid) == 0 && aux_guid == spa_guid(spa)) { - vdev_set_state(vd, B_FALSE, VDEV_STATE_CANT_OPEN, - VDEV_AUX_SPLIT_POOL); - nvlist_free(label); - vdev_dbgmsg(vd, "vdev_validate: vdev split into other " - "pool"); - return (0); - } + if ((label = vdev_label_read_config(vd, txg)) == NULL) { + vdev_set_state(vd, B_TRUE, VDEV_STATE_CANT_OPEN, + VDEV_AUX_BAD_LABEL); + vdev_dbgmsg(vd, "vdev_validate: failed reading config"); + return (0); + } - if (strict && (nvlist_lookup_uint64(label, - ZPOOL_CONFIG_POOL_GUID, &guid) != 0 || - guid != spa_guid(spa))) { - vdev_set_state(vd, B_FALSE, VDEV_STATE_CANT_OPEN, - VDEV_AUX_CORRUPT_DATA); - nvlist_free(label); - vdev_dbgmsg(vd, "vdev_validate: vdev label pool_guid " - "doesn't match config (%llu != %llu)", - (u_longlong_t)guid, - (u_longlong_t)spa_guid(spa)); - return (0); - } + /* + * Determine if this vdev has been split off into another + * pool. If so, then refuse to open it. + */ + if (nvlist_lookup_uint64(label, ZPOOL_CONFIG_SPLIT_GUID, + &aux_guid) == 0 && aux_guid == spa_guid(spa)) { + vdev_set_state(vd, B_FALSE, VDEV_STATE_CANT_OPEN, + VDEV_AUX_SPLIT_POOL); + nvlist_free(label); + vdev_dbgmsg(vd, "vdev_validate: vdev split into other pool"); + return (0); + } - if (nvlist_lookup_nvlist(label, ZPOOL_CONFIG_VDEV_TREE, &nvl) - != 0 || nvlist_lookup_uint64(nvl, ZPOOL_CONFIG_ORIG_GUID, - &aux_guid) != 0) - aux_guid = 0; + if (nvlist_lookup_uint64(label, ZPOOL_CONFIG_POOL_GUID, &guid) != 0) { + vdev_set_state(vd, B_FALSE, VDEV_STATE_CANT_OPEN, + VDEV_AUX_CORRUPT_DATA); + nvlist_free(label); + vdev_dbgmsg(vd, "vdev_validate: '%s' missing from label", + ZPOOL_CONFIG_POOL_GUID); + return (0); + } - /* - * If this vdev just became a top-level vdev because its - * sibling was detached, it will have adopted the parent's - * vdev guid -- but the label may or may not be on disk yet. - * Fortunately, either version of the label will have the - * same top guid, so if we're a top-level vdev, we can - * safely compare to that instead. - * - * If we split this vdev off instead, then we also check the - * original pool's guid. We don't want to consider the vdev - * corrupt if it is partway through a split operation. - */ - if (nvlist_lookup_uint64(label, ZPOOL_CONFIG_GUID, - &guid) != 0 || - nvlist_lookup_uint64(label, ZPOOL_CONFIG_TOP_GUID, - &top_guid) != 0 || - ((vd->vdev_guid != guid && vd->vdev_guid != aux_guid) && - (vd->vdev_guid != top_guid || vd != vd->vdev_top))) { - vdev_set_state(vd, B_FALSE, VDEV_STATE_CANT_OPEN, - VDEV_AUX_CORRUPT_DATA); - nvlist_free(label); - vdev_dbgmsg(vd, "vdev_validate: config guid doesn't " - "match label guid (%llu != %llu)", - (u_longlong_t)vd->vdev_guid, (u_longlong_t)guid); - return (0); + /* + * If config is not trusted then ignore the spa guid check. This is + * necessary because if the machine crashed during a re-guid the new + * guid might have been written to all of the vdev labels, but not the + * cached config. The check will be performed again once we have the + * trusted config from the MOS. + */ + if (spa->spa_trust_config && guid != spa_guid(spa)) { + vdev_set_state(vd, B_FALSE, VDEV_STATE_CANT_OPEN, + VDEV_AUX_CORRUPT_DATA); + nvlist_free(label); + vdev_dbgmsg(vd, "vdev_validate: vdev label pool_guid doesn't " + "match config (%llu != %llu)", (u_longlong_t)guid, + (u_longlong_t)spa_guid(spa)); + return (0); + } + + if (nvlist_lookup_nvlist(label, ZPOOL_CONFIG_VDEV_TREE, &nvl) + != 0 || nvlist_lookup_uint64(nvl, ZPOOL_CONFIG_ORIG_GUID, + &aux_guid) != 0) + aux_guid = 0; + + if (nvlist_lookup_uint64(label, ZPOOL_CONFIG_GUID, &guid) != 0) { + vdev_set_state(vd, B_FALSE, VDEV_STATE_CANT_OPEN, + VDEV_AUX_CORRUPT_DATA); + nvlist_free(label); + vdev_dbgmsg(vd, "vdev_validate: '%s' missing from label", + ZPOOL_CONFIG_GUID); + return (0); + } + + if (nvlist_lookup_uint64(label, ZPOOL_CONFIG_TOP_GUID, &top_guid) + != 0) { + vdev_set_state(vd, B_FALSE, VDEV_STATE_CANT_OPEN, + VDEV_AUX_CORRUPT_DATA); + nvlist_free(label); + vdev_dbgmsg(vd, "vdev_validate: '%s' missing from label", + ZPOOL_CONFIG_TOP_GUID); + return (0); + } + + /* + * If this vdev just became a top-level vdev because its sibling was + * detached, it will have adopted the parent's vdev guid -- but the + * label may or may not be on disk yet. Fortunately, either version + * of the label will have the same top guid, so if we're a top-level + * vdev, we can safely compare to that instead. + * However, if the config comes from a cachefile that failed to update + * after the detach, a top-level vdev will appear as a non top-level + * vdev in the config. Also relax the constraints if we perform an + * extreme rewind. + * + * If we split this vdev off instead, then we also check the + * original pool's guid. We don't want to consider the vdev + * corrupt if it is partway through a split operation. + */ + if (vd->vdev_guid != guid && vd->vdev_guid != aux_guid) { + boolean_t mismatch = B_FALSE; + if (spa->spa_trust_config && !spa->spa_extreme_rewind) { + if (vd != vd->vdev_top || vd->vdev_guid != top_guid) + mismatch = B_TRUE; + } else { + if (vd->vdev_guid != top_guid && + vd->vdev_top->vdev_guid != guid) + mismatch = B_TRUE; } - if (nvlist_lookup_uint64(label, ZPOOL_CONFIG_POOL_STATE, - &state) != 0) { + if (mismatch) { vdev_set_state(vd, B_FALSE, VDEV_STATE_CANT_OPEN, VDEV_AUX_CORRUPT_DATA); nvlist_free(label); - vdev_dbgmsg(vd, "vdev_validate: '%s' missing", - ZPOOL_CONFIG_POOL_STATE); + vdev_dbgmsg(vd, "vdev_validate: config guid " + "doesn't match label guid"); + vdev_dbgmsg(vd, "CONFIG: guid %llu, top_guid %llu", + (u_longlong_t)vd->vdev_guid, + (u_longlong_t)vd->vdev_top->vdev_guid); + vdev_dbgmsg(vd, "LABEL: guid %llu, top_guid %llu, " + "aux_guid %llu", (u_longlong_t)guid, + (u_longlong_t)top_guid, (u_longlong_t)aux_guid); return (0); } + } + if (nvlist_lookup_uint64(label, ZPOOL_CONFIG_POOL_STATE, + &state) != 0) { + vdev_set_state(vd, B_FALSE, VDEV_STATE_CANT_OPEN, + VDEV_AUX_CORRUPT_DATA); nvlist_free(label); + vdev_dbgmsg(vd, "vdev_validate: '%s' missing from label", + ZPOOL_CONFIG_POOL_STATE); + return (0); + } - /* - * If this is a verbatim import, no need to check the - * state of the pool. - */ - if (!(spa->spa_import_flags & ZFS_IMPORT_VERBATIM) && - spa_load_state(spa) == SPA_LOAD_OPEN && - state != POOL_STATE_ACTIVE) { - vdev_dbgmsg(vd, "vdev_validate: invalid pool state " - "(%llu) for spa %s", (u_longlong_t)state, - spa->spa_name); - return (SET_ERROR(EBADF)); + nvlist_free(label); + + /* + * If this is a verbatim import, no need to check the + * state of the pool. + */ + if (!(spa->spa_import_flags & ZFS_IMPORT_VERBATIM) && + spa_load_state(spa) == SPA_LOAD_OPEN && + state != POOL_STATE_ACTIVE) { + vdev_dbgmsg(vd, "vdev_validate: invalid pool state (%llu) " + "for spa %s", (u_longlong_t)state, spa->spa_name); + return (SET_ERROR(EBADF)); + } + + /* + * If we were able to open and validate a vdev that was + * previously marked permanently unavailable, clear that state + * now. + */ + if (vd->vdev_not_present) + vd->vdev_not_present = 0; + + return (0); +} + +static void +vdev_copy_path_impl(vdev_t *svd, vdev_t *dvd) +{ + if (svd->vdev_path != NULL && dvd->vdev_path != NULL) { + if (strcmp(svd->vdev_path, dvd->vdev_path) != 0) { + zfs_dbgmsg("vdev_copy_path: vdev %llu: path changed " + "from '%s' to '%s'", (u_longlong_t)dvd->vdev_guid, + dvd->vdev_path, svd->vdev_path); + spa_strfree(dvd->vdev_path); + dvd->vdev_path = spa_strdup(svd->vdev_path); } + } else if (svd->vdev_path != NULL) { + dvd->vdev_path = spa_strdup(svd->vdev_path); + zfs_dbgmsg("vdev_copy_path: vdev %llu: path set to '%s'", + (u_longlong_t)dvd->vdev_guid, dvd->vdev_path); + } +} - /* - * If we were able to open and validate a vdev that was - * previously marked permanently unavailable, clear that state - * now. - */ - if (vd->vdev_not_present) - vd->vdev_not_present = 0; +/* + * Recursively copy vdev paths from one vdev to another. Source and destination + * vdev trees must have same geometry otherwise return error. Intended to copy + * paths from userland config into MOS config. + */ +int +vdev_copy_path_strict(vdev_t *svd, vdev_t *dvd) +{ + if ((svd->vdev_ops == &vdev_missing_ops) || + (svd->vdev_ishole && dvd->vdev_ishole) || + (dvd->vdev_ops == &vdev_indirect_ops)) + return (0); + + if (svd->vdev_ops != dvd->vdev_ops) { + vdev_dbgmsg(svd, "vdev_copy_path: vdev type mismatch: %s != %s", + svd->vdev_ops->vdev_op_type, dvd->vdev_ops->vdev_op_type); + return (SET_ERROR(EINVAL)); + } + + if (svd->vdev_guid != dvd->vdev_guid) { + vdev_dbgmsg(svd, "vdev_copy_path: guids mismatch (%llu != " + "%llu)", (u_longlong_t)svd->vdev_guid, + (u_longlong_t)dvd->vdev_guid); + return (SET_ERROR(EINVAL)); + } + + if (svd->vdev_children != dvd->vdev_children) { + vdev_dbgmsg(svd, "vdev_copy_path: children count mismatch: " + "%llu != %llu", (u_longlong_t)svd->vdev_children, + (u_longlong_t)dvd->vdev_children); + return (SET_ERROR(EINVAL)); } + for (uint64_t i = 0; i < svd->vdev_children; i++) { + int error = vdev_copy_path_strict(svd->vdev_child[i], + dvd->vdev_child[i]); + if (error != 0) + return (error); + } + + if (svd->vdev_ops->vdev_op_leaf) + vdev_copy_path_impl(svd, dvd); + return (0); } +static void +vdev_copy_path_search(vdev_t *stvd, vdev_t *dvd) +{ + ASSERT(stvd->vdev_top == stvd); + ASSERT3U(stvd->vdev_id, ==, dvd->vdev_top->vdev_id); + + for (uint64_t i = 0; i < dvd->vdev_children; i++) { + vdev_copy_path_search(stvd, dvd->vdev_child[i]); + } + + if (!dvd->vdev_ops->vdev_op_leaf || !vdev_is_concrete(dvd)) + return; + + /* + * The idea here is that while a vdev can shift positions within + * a top vdev (when replacing, attaching mirror, etc.) it cannot + * step outside of it. + */ + vdev_t *vd = vdev_lookup_by_guid(stvd, dvd->vdev_guid); + + if (vd == NULL || vd->vdev_ops != dvd->vdev_ops) + return; + + ASSERT(vd->vdev_ops->vdev_op_leaf); + + vdev_copy_path_impl(vd, dvd); +} + +/* + * Recursively copy vdev paths from one root vdev to another. Source and + * destination vdev trees may differ in geometry. For each destination leaf + * vdev, search a vdev with the same guid and top vdev id in the source. + * Intended to copy paths from userland config into MOS config. + */ +void +vdev_copy_path_relaxed(vdev_t *srvd, vdev_t *drvd) +{ + uint64_t children = MIN(srvd->vdev_children, drvd->vdev_children); + ASSERT(srvd->vdev_ops == &vdev_root_ops); + ASSERT(drvd->vdev_ops == &vdev_root_ops); + + for (uint64_t i = 0; i < children; i++) { + vdev_copy_path_search(srvd->vdev_child[i], + drvd->vdev_child[i]); + } +} + /* * Close a virtual device. */ @@ -1687,7 +1906,7 @@ vdev_reopen(vdev_t *vd) !l2arc_vdev_present(vd)) l2arc_add_vdev(spa, vd); } else { - (void) vdev_validate(vd, B_TRUE); + (void) vdev_validate(vd); } /* @@ -3551,6 +3770,19 @@ vdev_set_state(vdev_t *vd, boolean_t isopen, vdev_state_t state, vdev_aux_t aux) vdev_propagate_state(vd->vdev_parent); } +boolean_t +vdev_children_are_offline(vdev_t *vd) +{ + ASSERT(!vd->vdev_ops->vdev_op_leaf); + + for (uint64_t i = 0; i < vd->vdev_children; i++) { + if (vd->vdev_child[i]->vdev_state != VDEV_STATE_OFFLINE) + return (B_FALSE); + } + + return (B_TRUE); +} + /* * Check the vdev configuration to ensure that it's capable of supporting * a root pool. We do not support partial configuration. @@ -3591,35 +3823,6 @@ vdev_is_concrete(vdev_t *vd) } /* - * Load the state from the original vdev tree (ovd) which - * we've retrieved from the MOS config object. If the original - * vdev was offline or faulted then we transfer that state to the - * device in the current vdev tree (nvd). - */ -void -vdev_load_log_state(vdev_t *nvd, vdev_t *ovd) -{ - spa_t *spa = nvd->vdev_spa; - - ASSERT(nvd->vdev_top->vdev_islog); - ASSERT(spa_config_held(spa, SCL_STATE_ALL, RW_WRITER) == SCL_STATE_ALL); - ASSERT3U(nvd->vdev_guid, ==, ovd->vdev_guid); - - for (int c = 0; c < nvd->vdev_children; c++) - vdev_load_log_state(nvd->vdev_child[c], ovd->vdev_child[c]); - - if (nvd->vdev_ops->vdev_op_leaf) { - /* - * Restore the persistent vdev state - */ - nvd->vdev_offline = ovd->vdev_offline; - nvd->vdev_faulted = ovd->vdev_faulted; - nvd->vdev_degraded = ovd->vdev_degraded; - nvd->vdev_removed = ovd->vdev_removed; - } -} - -/* * Determine if a log device has valid content. If the vdev was * removed or faulted in the MOS config then we know that * the content on the log device has already been written to the pool. diff --git a/usr/src/uts/common/fs/zfs/vdev_indirect.c b/usr/src/uts/common/fs/zfs/vdev_indirect.c index 1025c8090e..5c4ea02047 100644 --- a/usr/src/uts/common/fs/zfs/vdev_indirect.c +++ b/usr/src/uts/common/fs/zfs/vdev_indirect.c @@ -14,7 +14,7 @@ */ /* - * Copyright (c) 2014, 2015 by Delphix. All rights reserved. + * Copyright (c) 2014, 2017 by Delphix. All rights reserved. */ #include <sys/zfs_context.h> @@ -30,6 +30,8 @@ #include <sys/dmu_tx.h> #include <sys/dsl_synctask.h> #include <sys/zap.h> +#include <sys/abd.h> +#include <sys/zthr.h> /* * An indirect vdev corresponds to a vdev that has been removed. Since @@ -475,7 +477,7 @@ spa_condense_indirect_commit_entry(spa_t *spa, static void spa_condense_indirect_generate_new_mapping(vdev_t *vd, - uint32_t *obsolete_counts, uint64_t start_index) + uint32_t *obsolete_counts, uint64_t start_index, zthr_t *zthr) { spa_t *spa = vd->vdev_spa; uint64_t mapi = start_index; @@ -490,7 +492,15 @@ spa_condense_indirect_generate_new_mapping(vdev_t *vd, (u_longlong_t)vd->vdev_id, (u_longlong_t)mapi); - while (mapi < old_num_entries && !spa_shutting_down(spa)) { + while (mapi < old_num_entries) { + + if (zthr_iscancelled(zthr)) { + zfs_dbgmsg("pausing condense of vdev %llu " + "at index %llu", (u_longlong_t)vd->vdev_id, + (u_longlong_t)mapi); + break; + } + vdev_indirect_mapping_entry_phys_t *entry = &old_mapping->vim_entries[mapi]; uint64_t entry_size = DVA_GET_ASIZE(&entry->vimep_dst); @@ -508,18 +518,30 @@ spa_condense_indirect_generate_new_mapping(vdev_t *vd, mapi++; } - if (spa_shutting_down(spa)) { - zfs_dbgmsg("pausing condense of vdev %llu at index %llu", - (u_longlong_t)vd->vdev_id, - (u_longlong_t)mapi); - } } -static void -spa_condense_indirect_thread(void *arg) +/* ARGSUSED */ +static boolean_t +spa_condense_indirect_thread_check(void *arg, zthr_t *zthr) { - vdev_t *vd = arg; - spa_t *spa = vd->vdev_spa; + spa_t *spa = arg; + + return (spa->spa_condensing_indirect != NULL); +} + +/* ARGSUSED */ +static int +spa_condense_indirect_thread(void *arg, zthr_t *zthr) +{ + spa_t *spa = arg; + vdev_t *vd; + + ASSERT3P(spa->spa_condensing_indirect, !=, NULL); + spa_config_enter(spa, SCL_VDEV, FTAG, RW_READER); + vd = vdev_lookup_top(spa, spa->spa_condensing_indirect_phys.scip_vdev); + ASSERT3P(vd, !=, NULL); + spa_config_exit(spa, SCL_VDEV, FTAG); + spa_condensing_indirect_t *sci = spa->spa_condensing_indirect; spa_condensing_indirect_phys_t *scip = &spa->spa_condensing_indirect_phys; @@ -593,25 +615,24 @@ spa_condense_indirect_thread(void *arg) } } - spa_condense_indirect_generate_new_mapping(vd, counts, start_index); + spa_condense_indirect_generate_new_mapping(vd, counts, + start_index, zthr); vdev_indirect_mapping_free_obsolete_counts(old_mapping, counts); /* - * We may have bailed early from generate_new_mapping(), if - * the spa is shutting down. In this case, do not complete - * the condense. + * If the zthr has received a cancellation signal while running + * in generate_new_mapping() or at any point after that, then bail + * early. We don't want to complete the condense if the spa is + * shutting down. */ - if (!spa_shutting_down(spa)) { - VERIFY0(dsl_sync_task(spa_name(spa), NULL, - spa_condense_indirect_complete_sync, sci, 0, - ZFS_SPACE_CHECK_NONE)); - } + if (zthr_iscancelled(zthr)) + return (0); + + VERIFY0(dsl_sync_task(spa_name(spa), NULL, + spa_condense_indirect_complete_sync, sci, 0, ZFS_SPACE_CHECK_NONE)); - mutex_enter(&spa->spa_async_lock); - spa->spa_condense_thread = NULL; - cv_broadcast(&spa->spa_async_cv); - mutex_exit(&spa->spa_async_lock); + return (0); } /* @@ -664,9 +685,7 @@ spa_condense_indirect_start_sync(vdev_t *vd, dmu_tx_t *tx) (u_longlong_t)scip->scip_prev_obsolete_sm_object, (u_longlong_t)scip->scip_next_mapping_object); - ASSERT3P(spa->spa_condense_thread, ==, NULL); - spa->spa_condense_thread = thread_create(NULL, 0, - spa_condense_indirect_thread, vd, 0, &p0, TS_RUN, minclsyspri); + zthr_wakeup(spa->spa_condense_zthr); } /* @@ -743,24 +762,12 @@ spa_condense_fini(spa_t *spa) } } -/* - * Restart the condense - called when the pool is opened. - */ void -spa_condense_indirect_restart(spa_t *spa) +spa_start_indirect_condensing_thread(spa_t *spa) { - vdev_t *vd; - ASSERT(spa->spa_condensing_indirect != NULL); - spa_config_enter(spa, SCL_VDEV, FTAG, RW_READER); - vd = vdev_lookup_top(spa, - spa->spa_condensing_indirect_phys.scip_vdev); - ASSERT(vd != NULL); - spa_config_exit(spa, SCL_VDEV, FTAG); - - ASSERT3P(spa->spa_condense_thread, ==, NULL); - spa->spa_condense_thread = thread_create(NULL, 0, - spa_condense_indirect_thread, vd, 0, &p0, TS_RUN, - minclsyspri); + ASSERT3P(spa->spa_condense_zthr, ==, NULL); + spa->spa_condense_zthr = zthr_create(spa_condense_indirect_thread_check, + spa_condense_indirect_thread, spa); } /* @@ -845,6 +852,57 @@ rs_alloc(vdev_t *vd, uint64_t offset, uint64_t asize, uint64_t split_offset) } /* + * Given an indirect vdev and an extent on that vdev, it duplicates the + * physical entries of the indirect mapping that correspond to the extent + * to a new array and returns a pointer to it. In addition, copied_entries + * is populated with the number of mapping entries that were duplicated. + * + * Note that the function assumes that the caller holds vdev_indirect_rwlock. + * This ensures that the mapping won't change due to condensing as we + * copy over its contents. + * + * Finally, since we are doing an allocation, it is up to the caller to + * free the array allocated in this function. + */ +vdev_indirect_mapping_entry_phys_t * +vdev_indirect_mapping_duplicate_adjacent_entries(vdev_t *vd, uint64_t offset, + uint64_t asize, uint64_t *copied_entries) +{ + vdev_indirect_mapping_entry_phys_t *duplicate_mappings = NULL; + vdev_indirect_mapping_t *vim = vd->vdev_indirect_mapping; + uint64_t entries = 0; + + ASSERT(RW_READ_HELD(&vd->vdev_indirect_rwlock)); + + vdev_indirect_mapping_entry_phys_t *first_mapping = + vdev_indirect_mapping_entry_for_offset(vim, offset); + ASSERT3P(first_mapping, !=, NULL); + + vdev_indirect_mapping_entry_phys_t *m = first_mapping; + while (asize > 0) { + uint64_t size = DVA_GET_ASIZE(&m->vimep_dst); + + ASSERT3U(offset, >=, DVA_MAPPING_GET_SRC_OFFSET(m)); + ASSERT3U(offset, <, DVA_MAPPING_GET_SRC_OFFSET(m) + size); + + uint64_t inner_offset = offset - DVA_MAPPING_GET_SRC_OFFSET(m); + uint64_t inner_size = MIN(asize, size - inner_offset); + + offset += inner_size; + asize -= inner_size; + entries++; + m++; + } + + size_t copy_length = entries * sizeof (*first_mapping); + duplicate_mappings = kmem_alloc(copy_length, KM_SLEEP); + bcopy(first_mapping, duplicate_mappings, copy_length); + *copied_entries = entries; + + return (duplicate_mappings); +} + +/* * Goes through the relevant indirect mappings until it hits a concrete vdev * and issues the callback. On the way to the concrete vdev, if any other * indirect vdevs are encountered, then the callback will also be called on @@ -884,24 +942,42 @@ vdev_indirect_remap(vdev_t *vd, uint64_t offset, uint64_t asize, for (remap_segment_t *rs = rs_alloc(vd, offset, asize, 0); rs != NULL; rs = list_remove_head(&stack)) { vdev_t *v = rs->rs_vd; + uint64_t num_entries = 0; + + ASSERT(spa_config_held(spa, SCL_ALL, RW_READER) != 0); + ASSERT(rs->rs_asize > 0); /* - * Note: this can be called from open context - * (eg. zio_read()), so we need the rwlock to prevent - * the mapping from being changed by condensing. + * Note: As this function can be called from open context + * (e.g. zio_read()), we need the following rwlock to + * prevent the mapping from being changed by condensing. + * + * So we grab the lock and we make a copy of the entries + * that are relevant to the extent that we are working on. + * Once that is done, we drop the lock and iterate over + * our copy of the mapping. Once we are done with the with + * the remap segment and we free it, we also free our copy + * of the indirect mapping entries that are relevant to it. + * + * This way we don't need to wait until the function is + * finished with a segment, to condense it. In addition, we + * don't need a recursive rwlock for the case that a call to + * vdev_indirect_remap() needs to call itself (through the + * codepath of its callback) for the same vdev in the middle + * of its execution. */ rw_enter(&v->vdev_indirect_rwlock, RW_READER); vdev_indirect_mapping_t *vim = v->vdev_indirect_mapping; ASSERT3P(vim, !=, NULL); - ASSERT(spa_config_held(spa, SCL_ALL, RW_READER) != 0); - ASSERT(rs->rs_asize > 0); - vdev_indirect_mapping_entry_phys_t *mapping = - vdev_indirect_mapping_entry_for_offset(vim, rs->rs_offset); + vdev_indirect_mapping_duplicate_adjacent_entries(v, + rs->rs_offset, rs->rs_asize, &num_entries); ASSERT3P(mapping, !=, NULL); + ASSERT3U(num_entries, >, 0); + rw_exit(&v->vdev_indirect_rwlock); - while (rs->rs_asize > 0) { + for (uint64_t i = 0; i < num_entries; i++) { /* * Note: the vdev_indirect_mapping can not change * while we are running. It only changes while the @@ -910,20 +986,23 @@ vdev_indirect_remap(vdev_t *vd, uint64_t offset, uint64_t asize, * function is only called for frees, which also only * happen from syncing context. */ + vdev_indirect_mapping_entry_phys_t *m = &mapping[i]; + + ASSERT3P(m, !=, NULL); + ASSERT3U(rs->rs_asize, >, 0); - uint64_t size = DVA_GET_ASIZE(&mapping->vimep_dst); - uint64_t dst_offset = - DVA_GET_OFFSET(&mapping->vimep_dst); - uint64_t dst_vdev = DVA_GET_VDEV(&mapping->vimep_dst); + uint64_t size = DVA_GET_ASIZE(&m->vimep_dst); + uint64_t dst_offset = DVA_GET_OFFSET(&m->vimep_dst); + uint64_t dst_vdev = DVA_GET_VDEV(&m->vimep_dst); ASSERT3U(rs->rs_offset, >=, - DVA_MAPPING_GET_SRC_OFFSET(mapping)); + DVA_MAPPING_GET_SRC_OFFSET(m)); ASSERT3U(rs->rs_offset, <, - DVA_MAPPING_GET_SRC_OFFSET(mapping) + size); + DVA_MAPPING_GET_SRC_OFFSET(m) + size); ASSERT3U(dst_vdev, !=, v->vdev_id); uint64_t inner_offset = rs->rs_offset - - DVA_MAPPING_GET_SRC_OFFSET(mapping); + DVA_MAPPING_GET_SRC_OFFSET(m); uint64_t inner_size = MIN(rs->rs_asize, size - inner_offset); @@ -964,10 +1043,10 @@ vdev_indirect_remap(vdev_t *vd, uint64_t offset, uint64_t asize, rs->rs_offset += inner_size; rs->rs_asize -= inner_size; rs->rs_split_offset += inner_size; - mapping++; } + VERIFY0(rs->rs_asize); - rw_exit(&v->vdev_indirect_rwlock); + kmem_free(mapping, num_entries * sizeof (*mapping)); kmem_free(rs, sizeof (remap_segment_t)); } list_destroy(&stack); diff --git a/usr/src/uts/common/fs/zfs/vdev_label.c b/usr/src/uts/common/fs/zfs/vdev_label.c index be3b89ee37..7ccc672437 100644 --- a/usr/src/uts/common/fs/zfs/vdev_label.c +++ b/usr/src/uts/common/fs/zfs/vdev_label.c @@ -265,7 +265,7 @@ vdev_config_generate(spa_t *spa, vdev_t *vd, boolean_t getstats, fnvlist_add_uint64(nv, ZPOOL_CONFIG_WHOLE_DISK, vd->vdev_wholedisk); - if (vd->vdev_not_present) + if (vd->vdev_not_present && !(flags & VDEV_CONFIG_MISSING)) fnvlist_add_uint64(nv, ZPOOL_CONFIG_NOT_PRESENT, 1); if (vd->vdev_isspare) @@ -1062,6 +1062,11 @@ vdev_uberblock_load(vdev_t *rvd, uberblock_t *ub, nvlist_t **config) "txg %llu", spa->spa_name, (u_longlong_t)ub->ub_txg); *config = vdev_label_read_config(cb.ubl_vd, ub->ub_txg); + if (*config == NULL && spa->spa_extreme_rewind) { + vdev_dbgmsg(cb.ubl_vd, "failed to read label config. " + "Trying again without txg restrictions."); + *config = vdev_label_read_config(cb.ubl_vd, UINT64_MAX); + } if (*config == NULL) { vdev_dbgmsg(cb.ubl_vd, "failed to read label config"); } @@ -1088,7 +1093,7 @@ vdev_uberblock_sync_done(zio_t *zio) static void vdev_uberblock_sync(zio_t *zio, uberblock_t *ub, vdev_t *vd, int flags) { - for (int c = 0; c < vd->vdev_children; c++) + for (uint64_t c = 0; c < vd->vdev_children; c++) vdev_uberblock_sync(zio, ub, vd->vdev_child[c], flags); if (!vd->vdev_ops->vdev_op_leaf) diff --git a/usr/src/uts/common/fs/zfs/vdev_mirror.c b/usr/src/uts/common/fs/zfs/vdev_mirror.c index a2a4925942..9d181a874e 100644 --- a/usr/src/uts/common/fs/zfs/vdev_mirror.c +++ b/usr/src/uts/common/fs/zfs/vdev_mirror.c @@ -84,9 +84,32 @@ vdev_mirror_map_alloc(zio_t *zio) if (vd == NULL) { dva_t *dva = zio->io_bp->blk_dva; spa_t *spa = zio->io_spa; + dva_t dva_copy[SPA_DVAS_PER_BP]; c = BP_GET_NDVAS(zio->io_bp); + /* + * If we do not trust the pool config, some DVAs might be + * invalid or point to vdevs that do not exist. We skip them. + */ + if (!spa_trust_config(spa)) { + ASSERT3U(zio->io_type, ==, ZIO_TYPE_READ); + int j = 0; + for (int i = 0; i < c; i++) { + if (zfs_dva_valid(spa, &dva[i], zio->io_bp)) + dva_copy[j++] = dva[i]; + } + if (j == 0) { + zio->io_vsd = NULL; + zio->io_error = ENXIO; + return (NULL); + } + if (j < c) { + dva = dva_copy; + c = j; + } + } + mm = kmem_zalloc(offsetof(mirror_map_t, mm_child[c]), KM_SLEEP); mm->mm_children = c; mm->mm_resilvering = B_FALSE; @@ -201,7 +224,10 @@ vdev_mirror_open(vdev_t *vd, uint64_t *asize, uint64_t *max_asize, } if (numerrors == vd->vdev_children) { - vd->vdev_stat.vs_aux = VDEV_AUX_NO_REPLICAS; + if (vdev_children_are_offline(vd)) + vd->vdev_stat.vs_aux = VDEV_AUX_CHILDREN_OFFLINE; + else + vd->vdev_stat.vs_aux = VDEV_AUX_NO_REPLICAS; return (lasterror); } @@ -311,6 +337,13 @@ vdev_mirror_io_start(zio_t *zio) mm = vdev_mirror_map_alloc(zio); + if (mm == NULL) { + ASSERT(!spa_trust_config(zio->io_spa)); + ASSERT(zio->io_type == ZIO_TYPE_READ); + zio_execute(zio); + return; + } + if (zio->io_type == ZIO_TYPE_READ) { if ((zio->io_flags & ZIO_FLAG_SCRUB) && !mm->mm_resilvering) { /* @@ -381,6 +414,9 @@ vdev_mirror_io_done(zio_t *zio) int good_copies = 0; int unexpected_errors = 0; + if (mm == NULL) + return; + for (c = 0; c < mm->mm_children; c++) { mc = &mm->mm_child[c]; @@ -486,13 +522,19 @@ vdev_mirror_io_done(zio_t *zio) static void vdev_mirror_state_change(vdev_t *vd, int faulted, int degraded) { - if (faulted == vd->vdev_children) - vdev_set_state(vd, B_FALSE, VDEV_STATE_CANT_OPEN, - VDEV_AUX_NO_REPLICAS); - else if (degraded + faulted != 0) + if (faulted == vd->vdev_children) { + if (vdev_children_are_offline(vd)) { + vdev_set_state(vd, B_FALSE, VDEV_STATE_OFFLINE, + VDEV_AUX_CHILDREN_OFFLINE); + } else { + vdev_set_state(vd, B_FALSE, VDEV_STATE_CANT_OPEN, + VDEV_AUX_NO_REPLICAS); + } + } else if (degraded + faulted != 0) { vdev_set_state(vd, B_FALSE, VDEV_STATE_DEGRADED, VDEV_AUX_NONE); - else + } else { vdev_set_state(vd, B_FALSE, VDEV_STATE_HEALTHY, VDEV_AUX_NONE); + } } vdev_ops_t vdev_mirror_ops = { diff --git a/usr/src/uts/common/fs/zfs/vdev_root.c b/usr/src/uts/common/fs/zfs/vdev_root.c index 6971b942f0..b3433c2424 100644 --- a/usr/src/uts/common/fs/zfs/vdev_root.c +++ b/usr/src/uts/common/fs/zfs/vdev_root.c @@ -37,6 +37,23 @@ * Virtual device vector for the pool's root vdev. */ +static uint64_t +vdev_root_core_tvds(vdev_t *vd) +{ + uint64_t tvds = 0; + + for (uint64_t c = 0; c < vd->vdev_children; c++) { + vdev_t *cvd = vd->vdev_child[c]; + + if (!cvd->vdev_ishole && !cvd->vdev_islog && + cvd->vdev_ops != &vdev_indirect_ops) { + tvds++; + } + } + + return (tvds); +} + /* * We should be able to tolerate one failure with absolutely no damage * to our metadata. Two failures will take out space maps, a bunch of @@ -46,17 +63,28 @@ * probably fine. Adding bean counters during alloc/free can make this * future guesswork more accurate. */ -static int -too_many_errors(vdev_t *vd, int numerrors) +static boolean_t +too_many_errors(vdev_t *vd, uint64_t numerrors) { - ASSERT3U(numerrors, <=, vd->vdev_children); - return (numerrors > 0); + uint64_t tvds; + + if (numerrors == 0) + return (B_FALSE); + + tvds = vdev_root_core_tvds(vd); + ASSERT3U(numerrors, <=, tvds); + + if (numerrors == tvds) + return (B_TRUE); + + return (numerrors > spa_missing_tvds_allowed(vd->vdev_spa)); } static int vdev_root_open(vdev_t *vd, uint64_t *asize, uint64_t *max_asize, uint64_t *ashift) { + spa_t *spa = vd->vdev_spa; int lasterror = 0; int numerrors = 0; @@ -76,6 +104,9 @@ vdev_root_open(vdev_t *vd, uint64_t *asize, uint64_t *max_asize, } } + if (spa_load_state(spa) != SPA_LOAD_NONE) + spa_set_missing_tvds(spa, numerrors); + if (too_many_errors(vd, numerrors)) { vd->vdev_stat.vs_aux = VDEV_AUX_NO_REPLICAS; return (lasterror); @@ -101,7 +132,7 @@ vdev_root_state_change(vdev_t *vd, int faulted, int degraded) if (too_many_errors(vd, faulted)) { vdev_set_state(vd, B_FALSE, VDEV_STATE_CANT_OPEN, VDEV_AUX_NO_REPLICAS); - } else if (degraded) { + } else if (degraded || faulted) { vdev_set_state(vd, B_FALSE, VDEV_STATE_DEGRADED, VDEV_AUX_NONE); } else { vdev_set_state(vd, B_FALSE, VDEV_STATE_HEALTHY, VDEV_AUX_NONE); diff --git a/usr/src/uts/common/fs/zfs/zio.c b/usr/src/uts/common/fs/zfs/zio.c index 7bde5a8391..29cdf08d52 100644 --- a/usr/src/uts/common/fs/zfs/zio.c +++ b/usr/src/uts/common/fs/zfs/zio.c @@ -444,21 +444,26 @@ zio_remove_child(zio_t *pio, zio_t *cio, zio_link_t *zl) } static boolean_t -zio_wait_for_children(zio_t *zio, enum zio_child child, enum zio_wait_type wait) +zio_wait_for_children(zio_t *zio, uint8_t childbits, enum zio_wait_type wait) { - uint64_t *countp = &zio->io_children[child][wait]; boolean_t waiting = B_FALSE; mutex_enter(&zio->io_lock); ASSERT(zio->io_stall == NULL); - if (*countp != 0) { - zio->io_stage >>= 1; - ASSERT3U(zio->io_stage, !=, ZIO_STAGE_OPEN); - zio->io_stall = countp; - waiting = B_TRUE; + for (int c = 0; c < ZIO_CHILD_TYPES; c++) { + if (!(ZIO_CHILD_BIT_IS_SET(childbits, c))) + continue; + + uint64_t *countp = &zio->io_children[c][wait]; + if (*countp != 0) { + zio->io_stage >>= 1; + ASSERT3U(zio->io_stage, !=, ZIO_STAGE_OPEN); + zio->io_stall = countp; + waiting = B_TRUE; + break; + } } mutex_exit(&zio->io_lock); - return (waiting); } @@ -693,6 +698,13 @@ zfs_blkptr_verify(spa_t *spa, const blkptr_t *bp) } /* + * Do not verify individual DVAs if the config is not trusted. This + * will be done once the zio is executed in vdev_mirror_map_alloc. + */ + if (!spa->spa_trust_config) + return; + + /* * Pool-specific checks. * * Note: it would be nice to verify that the blk_birth and @@ -741,6 +753,36 @@ zfs_blkptr_verify(spa_t *spa, const blkptr_t *bp) } } +boolean_t +zfs_dva_valid(spa_t *spa, const dva_t *dva, const blkptr_t *bp) +{ + uint64_t vdevid = DVA_GET_VDEV(dva); + + if (vdevid >= spa->spa_root_vdev->vdev_children) + return (B_FALSE); + + vdev_t *vd = spa->spa_root_vdev->vdev_child[vdevid]; + if (vd == NULL) + return (B_FALSE); + + if (vd->vdev_ops == &vdev_hole_ops) + return (B_FALSE); + + if (vd->vdev_ops == &vdev_missing_ops) { + return (B_FALSE); + } + + uint64_t offset = DVA_GET_OFFSET(dva); + uint64_t asize = DVA_GET_ASIZE(dva); + + if (BP_IS_GANG(bp)) + asize = vdev_psize_to_asize(vd, SPA_GANGBLOCKSIZE); + if (offset + asize > vd->vdev_asize) + return (B_FALSE); + + return (B_TRUE); +} + zio_t * zio_read(zio_t *pio, spa_t *spa, const blkptr_t *bp, abd_t *data, uint64_t size, zio_done_func_t *done, void *private, @@ -1267,9 +1309,10 @@ zio_write_compress(zio_t *zio) * If our children haven't all reached the ready stage, * wait for them and then repeat this pipeline stage. */ - if (zio_wait_for_children(zio, ZIO_CHILD_GANG, ZIO_WAIT_READY) || - zio_wait_for_children(zio, ZIO_CHILD_LOGICAL, ZIO_WAIT_READY)) + if (zio_wait_for_children(zio, ZIO_CHILD_LOGICAL_BIT | + ZIO_CHILD_GANG_BIT, ZIO_WAIT_READY)) { return (ZIO_PIPELINE_STOP); + } if (!IO_IS_ALLOCATING(zio)) return (ZIO_PIPELINE_CONTINUE); @@ -2112,8 +2155,9 @@ zio_gang_issue(zio_t *zio) { blkptr_t *bp = zio->io_bp; - if (zio_wait_for_children(zio, ZIO_CHILD_GANG, ZIO_WAIT_DONE)) + if (zio_wait_for_children(zio, ZIO_CHILD_GANG_BIT, ZIO_WAIT_DONE)) { return (ZIO_PIPELINE_STOP); + } ASSERT(BP_IS_GANG(bp) && zio->io_gang_leader == zio); ASSERT(zio->io_child_type > ZIO_CHILD_GANG); @@ -2434,8 +2478,9 @@ zio_ddt_read_done(zio_t *zio) { blkptr_t *bp = zio->io_bp; - if (zio_wait_for_children(zio, ZIO_CHILD_DDT, ZIO_WAIT_DONE)) + if (zio_wait_for_children(zio, ZIO_CHILD_DDT_BIT, ZIO_WAIT_DONE)) { return (ZIO_PIPELINE_STOP); + } ASSERT(BP_GET_DEDUP(bp)); ASSERT(BP_GET_PSIZE(bp) == zio->io_size); @@ -3038,10 +3083,14 @@ zio_vdev_io_start(zio_t *zio) } ASSERT3P(zio->io_logical, !=, zio); - if (zio->io_type == ZIO_TYPE_WRITE && zio->io_vd->vdev_removing) { - ASSERT(zio->io_flags & - (ZIO_FLAG_PHYSICAL | ZIO_FLAG_SELF_HEAL | - ZIO_FLAG_INDUCE_DAMAGE)); + if (zio->io_type == ZIO_TYPE_WRITE) { + ASSERT(spa->spa_trust_config); + + if (zio->io_vd->vdev_removing) { + ASSERT(zio->io_flags & + (ZIO_FLAG_PHYSICAL | ZIO_FLAG_SELF_HEAL | + ZIO_FLAG_INDUCE_DAMAGE)); + } } /* @@ -3146,8 +3195,9 @@ zio_vdev_io_done(zio_t *zio) vdev_ops_t *ops = vd ? vd->vdev_ops : &vdev_mirror_ops; boolean_t unexpected_error = B_FALSE; - if (zio_wait_for_children(zio, ZIO_CHILD_VDEV, ZIO_WAIT_DONE)) + if (zio_wait_for_children(zio, ZIO_CHILD_VDEV_BIT, ZIO_WAIT_DONE)) { return (ZIO_PIPELINE_STOP); + } ASSERT(zio->io_type == ZIO_TYPE_READ || zio->io_type == ZIO_TYPE_WRITE); @@ -3213,8 +3263,9 @@ zio_vdev_io_assess(zio_t *zio) { vdev_t *vd = zio->io_vd; - if (zio_wait_for_children(zio, ZIO_CHILD_VDEV, ZIO_WAIT_DONE)) + if (zio_wait_for_children(zio, ZIO_CHILD_VDEV_BIT, ZIO_WAIT_DONE)) { return (ZIO_PIPELINE_STOP); + } if (vd == NULL && !(zio->io_flags & ZIO_FLAG_CONFIG_WRITER)) spa_config_exit(zio->io_spa, SCL_ZIO, zio); @@ -3429,9 +3480,10 @@ zio_ready(zio_t *zio) zio_t *pio, *pio_next; zio_link_t *zl = NULL; - if (zio_wait_for_children(zio, ZIO_CHILD_GANG, ZIO_WAIT_READY) || - zio_wait_for_children(zio, ZIO_CHILD_DDT, ZIO_WAIT_READY)) + if (zio_wait_for_children(zio, ZIO_CHILD_GANG_BIT | ZIO_CHILD_DDT_BIT, + ZIO_WAIT_READY)) { return (ZIO_PIPELINE_STOP); + } if (zio->io_ready) { ASSERT(IO_IS_ALLOCATING(zio)); @@ -3571,11 +3623,9 @@ zio_done(zio_t *zio) * If our children haven't all completed, * wait for them and then repeat this pipeline stage. */ - if (zio_wait_for_children(zio, ZIO_CHILD_VDEV, ZIO_WAIT_DONE) || - zio_wait_for_children(zio, ZIO_CHILD_GANG, ZIO_WAIT_DONE) || - zio_wait_for_children(zio, ZIO_CHILD_DDT, ZIO_WAIT_DONE) || - zio_wait_for_children(zio, ZIO_CHILD_LOGICAL, ZIO_WAIT_DONE)) + if (zio_wait_for_children(zio, ZIO_CHILD_ALL_BITS, ZIO_WAIT_DONE)) { return (ZIO_PIPELINE_STOP); + } /* * If the allocation throttle is enabled, then update the accounting. diff --git a/usr/src/uts/common/fs/zfs/zthr.c b/usr/src/uts/common/fs/zfs/zthr.c new file mode 100644 index 0000000000..9beb7e128f --- /dev/null +++ b/usr/src/uts/common/fs/zfs/zthr.c @@ -0,0 +1,319 @@ +/* + * CDDL HEADER START + * + * This file and its contents are supplied under the terms of the + * Common Development and Distribution License ("CDDL"), version 1.0. + * You may only use this file in accordance with the terms of version + * 1.0 of the CDDL. + * + * A full copy of the text of the CDDL should have accompanied this + * source. A copy of the CDDL is also available via the Internet at + * http://www.illumos.org/license/CDDL. + * + * CDDL HEADER END + */ + +/* + * Copyright (c) 2017 by Delphix. All rights reserved. + */ + +/* + * ZTHR Infrastructure + * =================== + * + * ZTHR threads are used for isolated operations that span multiple txgs + * within a SPA. They generally exist from SPA creation/loading and until + * the SPA is exported/destroyed. The ideal requirements for an operation + * to be modeled with a zthr are the following: + * + * 1] The operation needs to run over multiple txgs. + * 2] There is be a single point of reference in memory or on disk that + * indicates whether the operation should run/is running or is + * stopped. + * + * If the operation satisfies the above then the following rules guarantee + * a certain level of correctness: + * + * 1] Any thread EXCEPT the zthr changes the work indicator from stopped + * to running but not the opposite. + * 2] Only the zthr can change the work indicator from running to stopped + * (e.g. when it is done) but not the opposite. + * + * This way a normal zthr cycle should go like this: + * + * 1] An external thread changes the work indicator from stopped to + * running and wakes up the zthr. + * 2] The zthr wakes up, checks the indicator and starts working. + * 3] When the zthr is done, it changes the indicator to stopped, allowing + * a new cycle to start. + * + * == ZTHR creation + * + * Every zthr needs three inputs to start running: + * + * 1] A user-defined checker function (checkfunc) that decides whether + * the zthr should start working or go to sleep. The function should + * return TRUE when the zthr needs to work or FALSE to let it sleep, + * and should adhere to the following signature: + * boolean_t checkfunc_name(void *args, zthr_t *t); + * + * 2] A user-defined ZTHR function (func) which the zthr executes when + * it is not sleeping. The function should adhere to the following + * signature type: + * int func_name(void *args, zthr_t *t); + * + * 3] A void args pointer that will be passed to checkfunc and func + * implicitly by the infrastructure. + * + * The reason why the above API needs two different functions, + * instead of one that both checks and does the work, has to do with + * the zthr's internal lock (zthr_lock) and the allowed cancellation + * windows. We want to hold the zthr_lock while running checkfunc + * but not while running func. This way the zthr can be cancelled + * while doing work and not while checking for work. + * + * To start a zthr: + * zthr_t *zthr_pointer = zthr_create(checkfunc, func, args); + * + * After that you should be able to wakeup, cancel, and resume the + * zthr from another thread using zthr_pointer. + * + * NOTE: ZTHR threads could potentially wake up spuriously and the + * user should take this into account when writing a checkfunc. + * [see ZTHR state transitions] + * + * == ZTHR cancellation + * + * ZTHR threads must be cancelled when their SPA is being exported + * or when they need to be paused so they don't interfere with other + * operations. + * + * To cancel a zthr: + * zthr_cancel(zthr_pointer); + * + * To resume it: + * zthr_resume(zthr_pointer); + * + * A zthr will implicitly check if it has received a cancellation + * signal every time func returns and everytime it wakes up [see ZTHR + * state transitions below]. + * + * At times, waiting for the zthr's func to finish its job may take + * time. This may be very time-consuming for some operations that + * need to cancel the SPA's zthrs (e.g spa_export). For this scenario + * the user can explicitly make their ZTHR function aware of incoming + * cancellation signals using zthr_iscancelled(). A common pattern for + * that looks like this: + * + * int + * func_name(void *args, zthr_t *t) + * { + * ... <unpack args> ... + * while (!work_done && !zthr_iscancelled(t)) { + * ... <do more work> ... + * } + * return (0); + * } + * + * == ZTHR exit + * + * For the rare cases where the zthr wants to stop running voluntarily + * while running its ZTHR function (func), we provide zthr_exit(). + * When a zthr has voluntarily stopped running, it can be resumed with + * zthr_resume(), just like it would if it was cancelled by some other + * thread. + * + * == ZTHR cleanup + * + * Cancelling a zthr doesn't clean up its metadata (internal locks, + * function pointers to func and checkfunc, etc..). This is because + * we want to keep them around in case we want to resume the execution + * of the zthr later. Similarly for zthrs that exit themselves. + * + * To completely cleanup a zthr, cancel it first to ensure that it + * is not running and then use zthr_destroy(). + * + * == ZTHR state transitions + * + * zthr creation + * + + * | + * | woke up + * | +--------------+ sleep + * | | ^ + * | | | + * | | | FALSE + * | | | + * v v FALSE + + * cancelled? +---------> checkfunc? + * + ^ + + * | | | + * | | | TRUE + * | | | + * | | func returned v + * | +---------------+ func + * | + * | TRUE + * | + * v + * zthr stopped running + * + */ + +#include <sys/zfs_context.h> +#include <sys/zthr.h> + +void +zthr_exit(zthr_t *t, int rc) +{ + ASSERT3P(t->zthr_thread, ==, curthread); + mutex_enter(&t->zthr_lock); + t->zthr_thread = NULL; + t->zthr_rc = rc; + cv_broadcast(&t->zthr_cv); + mutex_exit(&t->zthr_lock); + thread_exit(); +} + +static void +zthr_procedure(void *arg) +{ + zthr_t *t = arg; + int rc = 0; + + mutex_enter(&t->zthr_lock); + while (!t->zthr_cancel) { + if (t->zthr_checkfunc(t->zthr_arg, t)) { + mutex_exit(&t->zthr_lock); + rc = t->zthr_func(t->zthr_arg, t); + mutex_enter(&t->zthr_lock); + } else { + /* go to sleep */ + cv_wait(&t->zthr_cv, &t->zthr_lock); + } + } + mutex_exit(&t->zthr_lock); + + zthr_exit(t, rc); +} + +zthr_t * +zthr_create(zthr_checkfunc_t *checkfunc, zthr_func_t *func, void *arg) +{ + zthr_t *t = kmem_zalloc(sizeof (*t), KM_SLEEP); + mutex_init(&t->zthr_lock, NULL, MUTEX_DEFAULT, NULL); + cv_init(&t->zthr_cv, NULL, CV_DEFAULT, NULL); + + mutex_enter(&t->zthr_lock); + t->zthr_checkfunc = checkfunc; + t->zthr_func = func; + t->zthr_arg = arg; + + t->zthr_thread = thread_create(NULL, 0, zthr_procedure, t, + 0, &p0, TS_RUN, minclsyspri); + mutex_exit(&t->zthr_lock); + + return (t); +} + +void +zthr_destroy(zthr_t *t) +{ + VERIFY3P(t->zthr_thread, ==, NULL); + mutex_destroy(&t->zthr_lock); + cv_destroy(&t->zthr_cv); + kmem_free(t, sizeof (*t)); +} + +/* + * Note: If the zthr is not sleeping and misses the wakeup + * (e.g it is running its ZTHR function), it will check if + * there is work to do before going to sleep using its checker + * function [see ZTHR state transition in ZTHR block comment]. + * Thus, missing the wakeup still yields the expected behavior. + */ +void +zthr_wakeup(zthr_t *t) +{ + ASSERT3P(t->zthr_thread, !=, NULL); + + mutex_enter(&t->zthr_lock); + cv_broadcast(&t->zthr_cv); + mutex_exit(&t->zthr_lock); +} + +/* + * Note: If the zthr is not running (e.g. has been cancelled + * already), this is a no-op. + */ +int +zthr_cancel(zthr_t *t) +{ + int rc = 0; + + mutex_enter(&t->zthr_lock); + + /* broadcast in case the zthr is sleeping */ + cv_broadcast(&t->zthr_cv); + + t->zthr_cancel = B_TRUE; + while (t->zthr_thread != NULL) + cv_wait(&t->zthr_cv, &t->zthr_lock); + t->zthr_cancel = B_FALSE; + rc = t->zthr_rc; + mutex_exit(&t->zthr_lock); + + return (rc); +} + +void +zthr_resume(zthr_t *t) +{ + ASSERT3P(t->zthr_thread, ==, NULL); + + mutex_enter(&t->zthr_lock); + + ASSERT3P(&t->zthr_checkfunc, !=, NULL); + ASSERT3P(&t->zthr_func, !=, NULL); + ASSERT(!t->zthr_cancel); + + t->zthr_thread = thread_create(NULL, 0, zthr_procedure, t, + 0, &p0, TS_RUN, minclsyspri); + + mutex_exit(&t->zthr_lock); +} + +/* + * This function is intended to be used by the zthr itself + * to check if another thread has signal it to stop running. + * + * returns TRUE if we are in the middle of trying to cancel + * this thread. + * + * returns FALSE otherwise. + */ +boolean_t +zthr_iscancelled(zthr_t *t) +{ + boolean_t cancelled; + + ASSERT3P(t->zthr_thread, ==, curthread); + + mutex_enter(&t->zthr_lock); + cancelled = t->zthr_cancel; + mutex_exit(&t->zthr_lock); + + return (cancelled); +} + +boolean_t +zthr_isrunning(zthr_t *t) +{ + boolean_t running; + + mutex_enter(&t->zthr_lock); + running = (t->zthr_thread != NULL); + mutex_exit(&t->zthr_lock); + + return (running); +} |