* You may not use this file except in compliance with the License.
*
* You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE
- * or http://www.opensolaris.org/os/licensing.
+ * or https://opensource.org/licenses/CDDL-1.0.
* See the License for the specific language governing permissions
* and limitations under the License.
*
*/
/*
* Copyright (c) 2005, 2010, Oracle and/or its affiliates. All rights reserved.
- * Copyright (c) 2011, 2019 by Delphix. All rights reserved.
+ * Copyright (c) 2011, 2021 by Delphix. All rights reserved.
* Copyright 2011 Nexenta Systems, Inc. All rights reserved.
* Copyright (c) 2014 Spectra Logic Corporation, All rights reserved.
* Copyright 2013 Saso Kiselkov. All rights reserved.
* Copyright (c) 2014 Integros [integros.com]
* Copyright 2017 Joyent, Inc.
- * Copyright (c) 2017, 2019, Datto Inc. All rights reserved.
* Copyright (c) 2017, Intel Corporation.
* Copyright (c) 2019, Allan Jude
* Copyright (c) 2019, Klara Inc.
+ * Copyright (c) 2019, Datto Inc.
*/
#ifndef _SYS_SPA_H
typedef struct ddt ddt_t;
typedef struct ddt_entry ddt_entry_t;
typedef struct zbookmark_phys zbookmark_phys_t;
+typedef struct zbookmark_err_phys zbookmark_err_phys_t;
struct bpobj;
struct bplist;
struct dsl_dataset;
struct dsl_crypto_params;
-/*
- * We currently support block sizes from 512 bytes to 16MB.
- * The benefits of larger blocks, and thus larger IO, need to be weighed
- * against the cost of COWing a giant block to modify one byte, and the
- * large latency of reading or writing a large block.
- *
- * Note that although blocks up to 16MB are supported, the recordsize
- * property can not be set larger than zfs_max_recordsize (default 1MB).
- * See the comment near zfs_max_recordsize in dsl_dataset.c for details.
- *
- * Note that although the LSIZE field of the blkptr_t can store sizes up
- * to 32MB, the dnode's dn_datablkszsec can only store sizes up to
- * 32MB - 512 bytes. Therefore, we limit SPA_MAXBLOCKSIZE to 16MB.
- */
-#define SPA_MINBLOCKSHIFT 9
-#define SPA_OLD_MAXBLOCKSHIFT 17
-#define SPA_MAXBLOCKSHIFT 24
-#define SPA_MINBLOCKSIZE (1ULL << SPA_MINBLOCKSHIFT)
-#define SPA_OLD_MAXBLOCKSIZE (1ULL << SPA_OLD_MAXBLOCKSHIFT)
-#define SPA_MAXBLOCKSIZE (1ULL << SPA_MAXBLOCKSHIFT)
-
/*
* Alignment Shift (ashift) is an immutable, internal top-level vdev property
* which can only be set at vdev creation time. Physical writes are always done
#define BPE_SET_ETYPE(bp, t) do { \
ASSERT(BP_IS_EMBEDDED(bp)); \
BF64_SET((bp)->blk_prop, 40, 8, t); \
-_NOTE(CONSTCOND) } while (0)
+} while (0)
#define BPE_GET_LSIZE(bp) \
(ASSERT(BP_IS_EMBEDDED(bp)), \
#define BPE_SET_LSIZE(bp, x) do { \
ASSERT(BP_IS_EMBEDDED(bp)); \
BF64_SET_SB((bp)->blk_prop, 0, 25, 0, 1, x); \
-_NOTE(CONSTCOND) } while (0)
+} while (0)
#define BPE_GET_PSIZE(bp) \
(ASSERT(BP_IS_EMBEDDED(bp)), \
#define BPE_SET_PSIZE(bp, x) do { \
ASSERT(BP_IS_EMBEDDED(bp)); \
BF64_SET_SB((bp)->blk_prop, 25, 7, 0, 1, x); \
-_NOTE(CONSTCOND) } while (0)
+} while (0)
typedef enum bp_embedded_type {
BP_EMBEDDED_TYPE_DATA,
/*
* Macros to get and set fields in a bp or DVA.
*/
+
+/*
+ * Note, for gang blocks, DVA_GET_ASIZE() is the total space allocated for
+ * this gang DVA including its children BP's. The space allocated at this
+ * DVA's vdev/offset is vdev_gang_header_asize(vdev).
+ */
#define DVA_GET_ASIZE(dva) \
BF64_GET_SB((dva)->dva_word[0], 0, SPA_ASIZEBITS, SPA_MINBLOCKSHIFT, 0)
#define DVA_SET_ASIZE(dva, x) \
ASSERT(!BP_IS_EMBEDDED(bp)); \
BF64_SET_SB((bp)->blk_prop, \
0, SPA_LSIZEBITS, SPA_MINBLOCKSHIFT, 1, x); \
-_NOTE(CONSTCOND) } while (0)
+} while (0)
#define BP_GET_PSIZE(bp) \
(BP_IS_EMBEDDED(bp) ? 0 : \
ASSERT(!BP_IS_EMBEDDED(bp)); \
BF64_SET_SB((bp)->blk_prop, \
16, SPA_PSIZEBITS, SPA_MINBLOCKSHIFT, 1, x); \
-_NOTE(CONSTCOND) } while (0)
+} while (0)
#define BP_GET_COMPRESS(bp) \
BF64_GET((bp)->blk_prop, 32, SPA_COMPRESSBITS)
#define BP_SET_CHECKSUM(bp, x) do { \
ASSERT(!BP_IS_EMBEDDED(bp)); \
BF64_SET((bp)->blk_prop, 40, 8, x); \
-_NOTE(CONSTCOND) } while (0)
+} while (0)
#define BP_GET_TYPE(bp) BF64_GET((bp)->blk_prop, 48, 8)
#define BP_SET_TYPE(bp, x) BF64_SET((bp)->blk_prop, 48, 8, x)
/*
* This macro allows code sharing between zfs, libzpool, and mdb.
- * 'func' is either snprintf() or mdb_snprintf().
+ * 'func' is either kmem_scnprintf() or mdb_snprintf().
* 'ws' (whitespace) can be ' ' for single-line format, '\n' for multi-line.
*/
#define SNPRINTF_BLKPTR(func, ws, buf, size, bp, type, checksum, compress) \
{ \
- static const char *copyname[] = \
+ static const char *const copyname[] = \
{ "zero", "single", "double", "triple" }; \
int len = 0; \
int copies = 0; \
(u_longlong_t)DVA_GET_ASIZE(dva), \
ws); \
} \
+ ASSERT3S(copies, >, 0); \
if (BP_IS_ENCRYPTED(bp)) { \
len += func(buf + len, size - len, \
"salt=%llx iv=%llx:%llx%c", \
len += func(buf + len, size - len, \
"[L%llu %s] %s %s %s %s %s %s %s%c" \
"size=%llxL/%llxP birth=%lluL/%lluP fill=%llu%c" \
- "cksum=%llx:%llx:%llx:%llx", \
+ "cksum=%016llx:%016llx:%016llx:%016llx", \
(u_longlong_t)BP_GET_LEVEL(bp), \
type, \
checksum, \
* Send TRIM commands in-line during normal pool operation while deleting.
* OFF: no
* ON: yes
- * NB: IN_FREEBSD_BASE is defined within the FreeBSD sources.
*/
typedef enum {
SPA_AUTOTRIM_OFF = 0, /* default */
SPA_AUTOTRIM_ON,
-#ifdef IN_FREEBSD_BASE
- SPA_AUTOTRIM_DEFAULT = SPA_AUTOTRIM_ON,
-#else
- SPA_AUTOTRIM_DEFAULT = SPA_AUTOTRIM_OFF,
-#endif
} spa_autotrim_t;
/*
} trim_type_t;
/* state manipulation functions */
-extern int spa_open(const char *pool, spa_t **, void *tag);
-extern int spa_open_rewind(const char *pool, spa_t **, void *tag,
+extern int spa_open(const char *pool, spa_t **, const void *tag);
+extern int spa_open_rewind(const char *pool, spa_t **, const void *tag,
nvlist_t *policy, nvlist_t **config);
extern int spa_get_stats(const char *pool, nvlist_t **config, char *altroot,
size_t buflen);
extern int spa_import(char *pool, nvlist_t *config, nvlist_t *props,
uint64_t flags);
extern nvlist_t *spa_tryimport(nvlist_t *tryconfig);
-extern int spa_destroy(char *pool);
+extern int spa_destroy(const char *pool);
extern int spa_checkpoint(const char *pool);
extern int spa_checkpoint_discard(const char *pool);
-extern int spa_export(char *pool, nvlist_t **oldconfig, boolean_t force,
+extern int spa_export(const char *pool, nvlist_t **oldconfig, boolean_t force,
boolean_t hardforce);
-extern int spa_reset(char *pool);
+extern int spa_reset(const char *pool);
extern void spa_async_request(spa_t *spa, int flag);
extern void spa_async_unrequest(spa_t *spa, int flag);
extern void spa_async_suspend(spa_t *spa);
#define SPA_ASYNC_L2CACHE_REBUILD 0x800
#define SPA_ASYNC_L2CACHE_TRIM 0x1000
#define SPA_ASYNC_REBUILD_DONE 0x2000
+#define SPA_ASYNC_DETACH_SPARE 0x4000
/* device manipulation */
extern int spa_vdev_add(spa_t *spa, nvlist_t *nvroot);
int replacing, int rebuild);
extern int spa_vdev_detach(spa_t *spa, uint64_t guid, uint64_t pguid,
int replace_done);
-extern int spa_vdev_remove(spa_t *spa, uint64_t guid, boolean_t unspare);
+extern int spa_vdev_alloc(spa_t *spa, uint64_t guid);
+extern int spa_vdev_noalloc(spa_t *spa, uint64_t guid);
extern boolean_t spa_vdev_remove_active(spa_t *spa);
extern int spa_vdev_initialize(spa_t *spa, nvlist_t *nv, uint64_t cmd_type,
nvlist_t *vdev_errlist);
uint64_t rate, boolean_t partial, boolean_t secure, nvlist_t *vdev_errlist);
extern int spa_vdev_setpath(spa_t *spa, uint64_t guid, const char *newpath);
extern int spa_vdev_setfru(spa_t *spa, uint64_t guid, const char *newfru);
-extern int spa_vdev_split_mirror(spa_t *spa, char *newname, nvlist_t *config,
- nvlist_t *props, boolean_t exp);
+extern int spa_vdev_split_mirror(spa_t *spa, const char *newname,
+ nvlist_t *config, nvlist_t *props, boolean_t exp);
/* spare state (which is global across all pools) */
extern void spa_spare_add(vdev_t *vd);
extern void spa_sync(spa_t *spa, uint64_t txg); /* only for DMU use */
extern void spa_sync_allpools(void);
-extern int zfs_sync_pass_deferred_free;
+extern uint_t zfs_sync_pass_deferred_free;
+
+/* spa sync taskqueues */
+taskq_t *spa_sync_tq_create(spa_t *spa, const char *name);
+void spa_sync_tq_destroy(spa_t *spa);
+void spa_select_allocator(zio_t *zio);
/* spa namespace global mutex */
extern kmutex_t spa_namespace_lock;
#define SPA_CONFIG_UPDATE_POOL 0
#define SPA_CONFIG_UPDATE_VDEVS 1
-extern void spa_write_cachefile(spa_t *, boolean_t, boolean_t);
+extern void spa_write_cachefile(spa_t *, boolean_t, boolean_t, boolean_t);
extern void spa_config_load(void);
-extern nvlist_t *spa_all_configs(uint64_t *);
+extern int spa_all_configs(uint64_t *generation, nvlist_t **pools);
extern void spa_config_set(spa_t *spa, nvlist_t *config);
extern nvlist_t *spa_config_generate(spa_t *spa, vdev_t *vd, uint64_t txg,
int getstats);
extern spa_t *spa_next(spa_t *prev);
/* Refcount functions */
-extern void spa_open_ref(spa_t *spa, void *tag);
-extern void spa_close(spa_t *spa, void *tag);
-extern void spa_async_close(spa_t *spa, void *tag);
+extern void spa_open_ref(spa_t *spa, const void *tag);
+extern void spa_close(spa_t *spa, const void *tag);
+extern void spa_async_close(spa_t *spa, const void *tag);
extern boolean_t spa_refcount_zero(spa_t *spa);
#define SCL_NONE 0x00
spa_history_list_t read_history;
spa_history_list_t txg_history;
spa_history_kstat_t tx_assign_histogram;
- spa_history_kstat_t io_history;
spa_history_list_t mmp_history;
spa_history_kstat_t state; /* pool state */
+ spa_history_kstat_t guid; /* pool guid */
spa_history_kstat_t iostats;
} spa_stats_t;
spa_load_state_t spa_load_state);
/* Pool configuration locks */
-extern int spa_config_tryenter(spa_t *spa, int locks, void *tag, krw_t rw);
+extern int spa_config_tryenter(spa_t *spa, int locks, const void *tag,
+ krw_t rw);
extern void spa_config_enter(spa_t *spa, int locks, const void *tag, krw_t rw);
+extern void spa_config_enter_mmp(spa_t *spa, int locks, const void *tag,
+ krw_t rw);
extern void spa_config_exit(spa_t *spa, int locks, const void *tag);
extern int spa_config_held(spa_t *spa, int locks, krw_t rw);
extern uint64_t spa_vdev_detach_enter(spa_t *spa, uint64_t guid);
extern uint64_t spa_vdev_config_enter(spa_t *spa);
extern void spa_vdev_config_exit(spa_t *spa, vdev_t *vd, uint64_t txg,
- int error, char *tag);
+ int error, const char *tag);
extern int spa_vdev_exit(spa_t *spa, vdev_t *vd, uint64_t txg, int error);
/* Pool vdev state change lock */
extern blkptr_t *spa_get_rootblkptr(spa_t *spa);
extern void spa_set_rootblkptr(spa_t *spa, const blkptr_t *bp);
extern void spa_altroot(spa_t *, char *, size_t);
-extern int spa_sync_pass(spa_t *spa);
+extern uint32_t spa_sync_pass(spa_t *spa);
extern char *spa_name(spa_t *spa);
extern uint64_t spa_guid(spa_t *spa);
extern uint64_t spa_load_guid(spa_t *spa);
extern boolean_t spa_deflate(spa_t *spa);
extern metaslab_class_t *spa_normal_class(spa_t *spa);
extern metaslab_class_t *spa_log_class(spa_t *spa);
+extern metaslab_class_t *spa_embedded_log_class(spa_t *spa);
extern metaslab_class_t *spa_special_class(spa_t *spa);
extern metaslab_class_t *spa_dedup_class(spa_t *spa);
extern metaslab_class_t *spa_preferred_class(spa_t *spa, uint64_t size,
extern uint64_t spa_deadman_ziotime(spa_t *spa);
extern uint64_t spa_dirty_data(spa_t *spa);
extern spa_autotrim_t spa_get_autotrim(spa_t *spa);
+extern int spa_get_allocator(spa_t *spa);
+extern void spa_set_allocator(spa_t *spa, const char *allocator);
/* Miscellaneous support routines */
-extern void spa_load_failed(spa_t *spa, const char *fmt, ...);
-extern void spa_load_note(spa_t *spa, const char *fmt, ...);
+extern void spa_load_failed(spa_t *spa, const char *fmt, ...)
+ __attribute__((format(printf, 2, 3)));
+extern void spa_load_note(spa_t *spa, const char *fmt, ...)
+ __attribute__((format(printf, 2, 3)));
extern void spa_activate_mos_feature(spa_t *spa, const char *feature,
dmu_tx_t *tx);
extern void spa_deactivate_mos_feature(spa_t *spa, const char *feature);
extern boolean_t spa_guid_exists(uint64_t pool_guid, uint64_t device_guid);
extern char *spa_strdup(const char *);
extern void spa_strfree(char *);
-extern uint64_t spa_get_random(uint64_t range);
extern uint64_t spa_generate_guid(spa_t *spa);
extern void snprintf_blkptr(char *buf, size_t buflen, const blkptr_t *bp);
extern void spa_freeze(spa_t *spa);
extern void spa_evict_all(void);
extern vdev_t *spa_lookup_by_guid(spa_t *spa, uint64_t guid,
boolean_t l2cache);
+extern boolean_t spa_has_l2cache(spa_t *, uint64_t guid);
extern boolean_t spa_has_spare(spa_t *, uint64_t guid);
extern uint64_t dva_get_dsize_sync(spa_t *spa, const dva_t *dva);
extern uint64_t bp_get_dsize_sync(spa_t *spa, const blkptr_t *bp);
/* error handling */
struct zbookmark_phys;
-extern void spa_log_error(spa_t *spa, const zbookmark_phys_t *zb);
+extern void spa_log_error(spa_t *spa, const zbookmark_phys_t *zb,
+ const uint64_t *birth);
+extern void spa_remove_error(spa_t *spa, zbookmark_phys_t *zb,
+ const uint64_t *birth);
extern int zfs_ereport_post(const char *clazz, spa_t *spa, vdev_t *vd,
- const zbookmark_phys_t *zb, zio_t *zio, uint64_t stateoroffset,
- uint64_t length);
+ const zbookmark_phys_t *zb, zio_t *zio, uint64_t state);
extern boolean_t zfs_ereport_is_valid(const char *clazz, spa_t *spa, vdev_t *vd,
zio_t *zio);
+extern void zfs_ereport_taskq_fini(void);
+extern void zfs_ereport_clear(spa_t *spa, vdev_t *vd);
extern nvlist_t *zfs_event_create(spa_t *spa, vdev_t *vd, const char *type,
const char *name, nvlist_t *aux);
extern void zfs_post_remove(spa_t *spa, vdev_t *vd);
extern void zfs_post_state_change(spa_t *spa, vdev_t *vd, uint64_t laststate);
extern void zfs_post_autoreplace(spa_t *spa, vdev_t *vd);
-extern uint64_t spa_get_errlog_size(spa_t *spa);
-extern int spa_get_errlog(spa_t *spa, void *uaddr, size_t *count);
+extern uint64_t spa_approx_errlog_size(spa_t *spa);
+extern int spa_get_errlog(spa_t *spa, void *uaddr, uint64_t *count);
+extern uint64_t spa_get_last_errlog_size(spa_t *spa);
extern void spa_errlog_rotate(spa_t *spa);
extern void spa_errlog_drain(spa_t *spa);
extern void spa_errlog_sync(spa_t *spa, uint64_t txg);
extern void spa_get_errlists(spa_t *spa, avl_tree_t *last, avl_tree_t *scrub);
-
-/* vdev cache */
-extern void vdev_cache_stat_init(void);
-extern void vdev_cache_stat_fini(void);
+extern void spa_delete_dataset_errlog(spa_t *spa, uint64_t ds, dmu_tx_t *tx);
+extern void spa_swap_errlog(spa_t *spa, uint64_t new_head_ds,
+ uint64_t old_head_ds, dmu_tx_t *tx);
+extern void sync_error_list(spa_t *spa, avl_tree_t *t, uint64_t *obj,
+ dmu_tx_t *tx);
+extern void spa_upgrade_errlog(spa_t *spa, dmu_tx_t *tx);
+extern int find_top_affected_fs(spa_t *spa, uint64_t head_ds,
+ zbookmark_err_phys_t *zep, uint64_t *top_affected_fs);
+extern int find_birth_txg(struct dsl_dataset *ds, zbookmark_err_phys_t *zep,
+ uint64_t *birth_txg);
+extern void zep_to_zb(uint64_t dataset, zbookmark_err_phys_t *zep,
+ zbookmark_phys_t *zb);
+extern void name_to_errphys(char *buf, zbookmark_err_phys_t *zep);
/* vdev mirror */
extern void vdev_mirror_stat_init(void);
/* asynchronous event notification */
extern void spa_event_notify(spa_t *spa, vdev_t *vdev, nvlist_t *hist_nvl,
const char *name);
+extern void zfs_ereport_zvol_post(const char *subclass, const char *name,
+ const char *device_name, const char *raw_name);
/* waiting for pool activities to complete */
extern int spa_wait(const char *pool, zpool_wait_activity_t activity,
extern void spa_notify_waiters(spa_t *spa);
extern void spa_wake_waiters(spa_t *spa);
+extern void spa_import_os(spa_t *spa);
+extern void spa_export_os(spa_t *spa);
+extern void spa_activate_os(spa_t *spa);
+extern void spa_deactivate_os(spa_t *spa);
+
/* module param call functions */
int param_set_deadman_ziotime(ZFS_MODULE_PARAM_ARGS);
int param_set_deadman_synctime(ZFS_MODULE_PARAM_ARGS);
int param_set_slop_shift(ZFS_MODULE_PARAM_ARGS);
int param_set_deadman_failmode(ZFS_MODULE_PARAM_ARGS);
+int param_set_active_allocator(ZFS_MODULE_PARAM_ARGS);
#ifdef ZFS_DEBUG
#define dprintf_bp(bp, fmt, ...) do { \
dprintf(fmt " %s\n", __VA_ARGS__, __blkbuf); \
kmem_free(__blkbuf, BP_SPRINTF_LEN); \
} \
-_NOTE(CONSTCOND) } while (0)
+} while (0)
#else
#define dprintf_bp(bp, fmt, ...)
#endif
extern spa_mode_t spa_mode_global;
extern int zfs_deadman_enabled;
-extern unsigned long zfs_deadman_synctime_ms;
-extern unsigned long zfs_deadman_ziotime_ms;
-extern unsigned long zfs_deadman_checktime_ms;
+extern uint64_t zfs_deadman_synctime_ms;
+extern uint64_t zfs_deadman_ziotime_ms;
+extern uint64_t zfs_deadman_checktime_ms;
+
+extern kmem_cache_t *zio_buf_cache[];
+extern kmem_cache_t *zio_data_buf_cache[];
#ifdef __cplusplus
}