* 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, 2014 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 Datto Inc.
+ * Copyright (c) 2017, Intel Corporation.
+ * Copyright (c) 2019, Allan Jude
+ * Copyright (c) 2019, Klara Inc.
+ * Copyright (c) 2019, Datto Inc.
*/
#ifndef _SYS_SPA_H
#include <sys/fs/zfs.h>
#include <sys/spa_checksum.h>
#include <sys/dmu.h>
+#include <sys/space_map.h>
+#include <sys/bitops.h>
#ifdef __cplusplus
extern "C" {
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_pool;
struct dsl_dataset;
struct dsl_crypto_params;
-/*
- * General-purpose 32-bit and 64-bit bitfield encodings.
- */
-#define BF32_DECODE(x, low, len) P2PHASE((x) >> (low), 1U << (len))
-#define BF64_DECODE(x, low, len) P2PHASE((x) >> (low), 1ULL << (len))
-#define BF32_ENCODE(x, low, len) (P2PHASE((x), 1U << (len)) << (low))
-#define BF64_ENCODE(x, low, len) (P2PHASE((x), 1ULL << (len)) << (low))
-
-#define BF32_GET(x, low, len) BF32_DECODE(x, low, len)
-#define BF64_GET(x, low, len) BF64_DECODE(x, low, len)
-
-#define BF32_SET(x, low, len, val) do { \
- ASSERT3U(val, <, 1U << (len)); \
- ASSERT3U(low + len, <=, 32); \
- (x) ^= BF32_ENCODE((x >> low) ^ (val), low, len); \
-_NOTE(CONSTCOND) } while (0)
-
-#define BF64_SET(x, low, len, val) do { \
- ASSERT3U(val, <, 1ULL << (len)); \
- ASSERT3U(low + len, <=, 64); \
- ((x) ^= BF64_ENCODE((x >> low) ^ (val), low, len)); \
-_NOTE(CONSTCOND) } while (0)
-
-#define BF32_GET_SB(x, low, len, shift, bias) \
- ((BF32_GET(x, low, len) + (bias)) << (shift))
-#define BF64_GET_SB(x, low, len, shift, bias) \
- ((BF64_GET(x, low, len) + (bias)) << (shift))
-
-#define BF32_SET_SB(x, low, len, shift, bias, val) do { \
- ASSERT(IS_P2ALIGNED(val, 1U << shift)); \
- ASSERT3S((val) >> (shift), >=, bias); \
- BF32_SET(x, low, len, ((val) >> (shift)) - (bias)); \
-_NOTE(CONSTCOND) } while (0)
-#define BF64_SET_SB(x, low, len, shift, bias, val) do { \
- ASSERT(IS_P2ALIGNED(val, 1ULL << shift)); \
- ASSERT3S((val) >> (shift), >=, bias); \
- BF64_SET(x, low, len, ((val) >> (shift)) - (bias)); \
-_NOTE(CONSTCOND) } while (0)
-
-/*
- * 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 SPA_ASIZEBITS 24 /* ASIZE up to 64 times larger */
#define SPA_COMPRESSBITS 7
+#define SPA_VDEVBITS 24
+#define SPA_COMPRESSMASK ((1U << SPA_COMPRESSBITS) - 1)
/*
* All SPA data is represented by 128-bit data virtual addresses (DVAs).
*
* 64 56 48 40 32 24 16 8 0
* +-------+-------+-------+-------+-------+-------+-------+-------+
- * 0 | vdev1 | GRID | ASIZE |
+ * 0 | pad | vdev1 | GRID | ASIZE |
* +-------+-------+-------+-------+-------+-------+-------+-------+
* 1 |G| offset1 |
* +-------+-------+-------+-------+-------+-------+-------+-------+
- * 2 | vdev2 | GRID | ASIZE |
+ * 2 | pad | vdev2 | GRID | ASIZE |
* +-------+-------+-------+-------+-------+-------+-------+-------+
* 3 |G| offset2 |
* +-------+-------+-------+-------+-------+-------+-------+-------+
- * 4 | vdev3 | GRID | ASIZE |
+ * 4 | pad | vdev3 | GRID | ASIZE |
* +-------+-------+-------+-------+-------+-------+-------+-------+
* 5 |G| offset3 |
* +-------+-------+-------+-------+-------+-------+-------+-------+
* E blkptr_t contains embedded data (see below)
* lvl level of indirection
* type DMU object type
- * phys birth txg of block allocation; zero if same as logical birth txg
+ * phys birth txg when dva[0] was written; zero if same as logical birth txg
+ * note that typically all the dva's would be written in this
+ * txg, but they could be different if they were moved by
+ * device removal.
* log. birth transaction group in which the block was logically born
* fill count number of non-zero blocks under this bp
* checksum[4] 256-bit checksum of the data this bp describes
#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,
- BP_EMBEDDED_TYPE_RESERVED, /* Reserved for an unintegrated feature. */
- NUM_BP_EMBEDDED_TYPES = BP_EMBEDDED_TYPE_RESERVED
+ BP_EMBEDDED_TYPE_RESERVED, /* Reserved for Delphix byteswap feature. */
+ BP_EMBEDDED_TYPE_REDACTED,
+ NUM_BP_EMBEDDED_TYPES
} bp_embedded_type_t;
#define BPE_NUM_WORDS 14
#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
/*
* 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) \
#define DVA_GET_GRID(dva) BF64_GET((dva)->dva_word[0], 24, 8)
#define DVA_SET_GRID(dva, x) BF64_SET((dva)->dva_word[0], 24, 8, x)
-#define DVA_GET_VDEV(dva) BF64_GET((dva)->dva_word[0], 32, 32)
-#define DVA_SET_VDEV(dva, x) BF64_SET((dva)->dva_word[0], 32, 32, x)
+#define DVA_GET_VDEV(dva) BF64_GET((dva)->dva_word[0], 32, SPA_VDEVBITS)
+#define DVA_SET_VDEV(dva, x) \
+ BF64_SET((dva)->dva_word[0], 32, SPA_VDEVBITS, x)
#define DVA_GET_OFFSET(dva) \
BF64_GET_SB((dva)->dva_word[1], 0, 63, SPA_MINBLOCKSHIFT, 0)
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)
#define BP_GET_BYTEORDER(bp) BF64_GET((bp)->blk_prop, 63, 1)
#define BP_SET_BYTEORDER(bp, x) BF64_SET((bp)->blk_prop, 63, 1, x)
+#define BP_GET_FREE(bp) BF64_GET((bp)->blk_fill, 0, 1)
+#define BP_SET_FREE(bp, x) BF64_SET((bp)->blk_fill, 0, 1, x)
+
#define BP_PHYSICAL_BIRTH(bp) \
(BP_IS_EMBEDDED(bp) ? 0 : \
(bp)->blk_phys_birth ? (bp)->blk_phys_birth : (bp)->blk_birth)
#define BP_IS_HOLE(bp) \
(!BP_IS_EMBEDDED(bp) && DVA_IS_EMPTY(BP_IDENTITY(bp)))
+#define BP_SET_REDACTED(bp) \
+{ \
+ BP_SET_EMBEDDED(bp, B_TRUE); \
+ BPE_SET_ETYPE(bp, BP_EMBEDDED_TYPE_REDACTED); \
+}
+#define BP_IS_REDACTED(bp) \
+ (BP_IS_EMBEDDED(bp) && BPE_GET_ETYPE(bp) == BP_EMBEDDED_TYPE_REDACTED)
+
/* BP_IS_RAIDZ(bp) assumes no block compression */
#define BP_IS_RAIDZ(bp) (DVA_GET_ASIZE(&(bp)->blk_dva[0]) > \
BP_GET_PSIZE(bp))
ZIO_SET_CHECKSUM(&(bp)->blk_cksum, 0, 0, 0, 0); \
}
-#ifdef _BIG_ENDIAN
+#ifdef _ZFS_BIG_ENDIAN
#define ZFS_HOST_BYTEORDER (0ULL)
#else
#define ZFS_HOST_BYTEORDER (1ULL)
/*
* 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, crypt_type, \
- compress) \
+
+#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; \
- \
+ const char *crypt_type; \
+ if (bp != NULL) { \
+ if (BP_IS_ENCRYPTED(bp)) { \
+ crypt_type = "encrypted"; \
+ /* LINTED E_SUSPICIOUS_COMPARISON */ \
+ } else if (BP_IS_AUTHENTICATED(bp)) { \
+ crypt_type = "authenticated"; \
+ } else if (BP_HAS_INDIRECT_MAC_CKSUM(bp)) { \
+ crypt_type = "indirect-MAC"; \
+ } else { \
+ crypt_type = "unencrypted"; \
+ } \
+ } \
if (bp == NULL) { \
len += func(buf + len, size - len, "<NULL>"); \
} else if (BP_IS_HOLE(bp)) { \
(u_longlong_t)BPE_GET_LSIZE(bp), \
(u_longlong_t)BPE_GET_PSIZE(bp), \
(u_longlong_t)bp->blk_birth); \
+ } else if (BP_IS_REDACTED(bp)) { \
+ len += func(buf + len, size - len, \
+ "REDACTED [L%llu %s] size=%llxL birth=%lluL", \
+ (u_longlong_t)BP_GET_LEVEL(bp), \
+ type, \
+ (u_longlong_t)BP_GET_LSIZE(bp), \
+ (u_longlong_t)bp->blk_birth); \
} else { \
for (int d = 0; d < BP_GET_NDVAS(bp); d++) { \
const dva_t *dva = &bp->blk_dva[d]; \
(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, \
SPA_IMPORT_ASSEMBLE
} spa_import_type_t;
+typedef enum spa_mode {
+ SPA_MODE_UNINIT = 0,
+ SPA_MODE_READ = 1,
+ SPA_MODE_WRITE = 2,
+} spa_mode_t;
+
+/*
+ * Send TRIM commands in-line during normal pool operation while deleting.
+ * OFF: no
+ * ON: yes
+ */
+typedef enum {
+ SPA_AUTOTRIM_OFF = 0, /* default */
+ SPA_AUTOTRIM_ON,
+} spa_autotrim_t;
+
+/*
+ * Reason TRIM command was issued, used internally for accounting purposes.
+ */
+typedef enum trim_type {
+ TRIM_TYPE_MANUAL = 0,
+ TRIM_TYPE_AUTO = 1,
+ TRIM_TYPE_SIMPLE = 2
+} 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_export(char *pool, nvlist_t **oldconfig, boolean_t force,
+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(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);
extern void spa_async_resume(spa_t *spa);
+extern int spa_async_tasks(spa_t *spa);
extern spa_t *spa_inject_addref(char *pool);
extern void spa_inject_delref(spa_t *spa);
extern void spa_scan_stat_init(spa_t *spa);
extern int spa_scan_get_stats(spa_t *spa, pool_scan_stat_t *ps);
-
-#define SPA_ASYNC_CONFIG_UPDATE 0x01
-#define SPA_ASYNC_REMOVE 0x02
-#define SPA_ASYNC_PROBE 0x04
-#define SPA_ASYNC_RESILVER_DONE 0x08
-#define SPA_ASYNC_RESILVER 0x10
-#define SPA_ASYNC_AUTOEXPAND 0x20
-#define SPA_ASYNC_REMOVE_DONE 0x40
-#define SPA_ASYNC_REMOVE_STOP 0x80
-
-/*
- * Controls the behavior of spa_vdev_remove().
- */
-#define SPA_REMOVE_UNSPARE 0x01
-#define SPA_REMOVE_DONE 0x02
+extern int bpobj_enqueue_alloc_cb(void *arg, const blkptr_t *bp, dmu_tx_t *tx);
+extern int bpobj_enqueue_free_cb(void *arg, const blkptr_t *bp, dmu_tx_t *tx);
+
+#define SPA_ASYNC_CONFIG_UPDATE 0x01
+#define SPA_ASYNC_REMOVE 0x02
+#define SPA_ASYNC_PROBE 0x04
+#define SPA_ASYNC_RESILVER_DONE 0x08
+#define SPA_ASYNC_RESILVER 0x10
+#define SPA_ASYNC_AUTOEXPAND 0x20
+#define SPA_ASYNC_REMOVE_DONE 0x40
+#define SPA_ASYNC_REMOVE_STOP 0x80
+#define SPA_ASYNC_INITIALIZE_RESTART 0x100
+#define SPA_ASYNC_TRIM_RESTART 0x200
+#define SPA_ASYNC_AUTOTRIM_RESTART 0x400
+#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);
extern int spa_vdev_attach(spa_t *spa, uint64_t guid, nvlist_t *nvroot,
- int replacing);
+ 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);
+extern int spa_vdev_trim(spa_t *spa, nvlist_t *nv, uint64_t cmd_type,
+ 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_config_sync(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 void spa_config_update(spa_t *spa, int what);
+extern int spa_config_parse(spa_t *spa, vdev_t **vdp, nvlist_t *nv,
+ vdev_t *parent, uint_t id, int atype);
+
/*
* Miscellaneous SPA routines in spa_misc.c
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
#define SCL_STATE_ALL (SCL_STATE | SCL_L2ARC | SCL_ZIO)
/* Historical pool statistics */
-typedef struct spa_stats_history {
+typedef struct spa_history_kstat {
kmutex_t lock;
uint64_t count;
uint64_t size;
kstat_t *kstat;
- void *private;
+ void *priv;
list_t list;
-} spa_stats_history_t;
+} spa_history_kstat_t;
+
+typedef struct spa_history_list {
+ uint64_t size;
+ procfs_list_t procfs_list;
+} spa_history_list_t;
typedef struct spa_stats {
- spa_stats_history_t read_history;
- spa_stats_history_t txg_history;
- spa_stats_history_t tx_assign_histogram;
- spa_stats_history_t io_history;
- spa_stats_history_t mmp_history;
+ spa_history_list_t read_history;
+ spa_history_list_t txg_history;
+ spa_history_kstat_t tx_assign_histogram;
+ 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;
typedef enum txg_state {
uint64_t ndirty;
} txg_stat_t;
+/* Assorted pool IO kstats */
+typedef struct spa_iostats {
+ kstat_named_t trim_extents_written;
+ kstat_named_t trim_bytes_written;
+ kstat_named_t trim_extents_skipped;
+ kstat_named_t trim_bytes_skipped;
+ kstat_named_t trim_extents_failed;
+ kstat_named_t trim_bytes_failed;
+ kstat_named_t autotrim_extents_written;
+ kstat_named_t autotrim_bytes_written;
+ kstat_named_t autotrim_extents_skipped;
+ kstat_named_t autotrim_bytes_skipped;
+ kstat_named_t autotrim_extents_failed;
+ kstat_named_t autotrim_bytes_failed;
+ kstat_named_t simple_trim_extents_written;
+ kstat_named_t simple_trim_bytes_written;
+ kstat_named_t simple_trim_extents_skipped;
+ kstat_named_t simple_trim_bytes_skipped;
+ kstat_named_t simple_trim_extents_failed;
+ kstat_named_t simple_trim_bytes_failed;
+} spa_iostats_t;
+
extern void spa_stats_init(spa_t *spa);
extern void spa_stats_destroy(spa_t *spa);
extern void spa_read_history_add(spa_t *spa, const zbookmark_phys_t *zb,
extern int spa_mmp_history_set_skip(spa_t *spa, uint64_t mmp_kstat_id);
extern int spa_mmp_history_set(spa_t *spa, uint64_t mmp_kstat_id, int io_error,
hrtime_t duration);
-extern void *spa_mmp_history_add(spa_t *spa, uint64_t txg, uint64_t timestamp,
+extern void spa_mmp_history_add(spa_t *spa, uint64_t txg, uint64_t timestamp,
uint64_t mmp_delay, vdev_t *vd, int label, uint64_t mmp_kstat_id,
int error);
+extern void spa_iostats_trim_add(spa_t *spa, trim_type_t type,
+ uint64_t extents_written, uint64_t bytes_written,
+ uint64_t extents_skipped, uint64_t bytes_skipped,
+ uint64_t extents_failed, uint64_t bytes_failed);
+extern void spa_import_progress_add(spa_t *spa);
+extern void spa_import_progress_remove(uint64_t spa_guid);
+extern int spa_import_progress_set_mmp_check(uint64_t pool_guid,
+ uint64_t mmp_sec_remaining);
+extern int spa_import_progress_set_max_txg(uint64_t pool_guid,
+ uint64_t max_txg);
+extern int spa_import_progress_set_state(uint64_t pool_guid,
+ 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 void spa_config_enter(spa_t *spa, int locks, void *tag, krw_t rw);
-extern void spa_config_exit(spa_t *spa, int locks, void *tag);
+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);
/* Pool vdev add/remove lock */
extern uint64_t spa_vdev_enter(spa_t *spa);
+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 spa_log_state_t spa_get_log_state(spa_t *spa);
extern void spa_set_log_state(spa_t *spa, spa_log_state_t state);
-extern int spa_offline_log(spa_t *spa);
+extern int spa_reset_logs(spa_t *spa);
/* Log claim callback */
extern void spa_claim_notify(zio_t *zio);
extern boolean_t spa_shutting_down(spa_t *spa);
extern struct dsl_pool *spa_get_dsl(spa_t *spa);
extern boolean_t spa_is_initializing(spa_t *spa);
+extern boolean_t spa_indirect_vdevs_loaded(spa_t *spa);
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 uint64_t spa_freeze_txg(spa_t *spa);
extern uint64_t spa_get_worst_case_asize(spa_t *spa, uint64_t lsize);
extern uint64_t spa_get_dspace(spa_t *spa);
+extern uint64_t spa_get_checkpoint_space(spa_t *spa);
extern uint64_t spa_get_slop_space(spa_t *spa);
extern void spa_update_dspace(spa_t *spa);
extern uint64_t spa_version(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,
+ dmu_object_type_t objtype, uint_t level, uint_t special_smallblk);
+
extern void spa_evicting_os_register(spa_t *, objset_t *os);
extern void spa_evicting_os_deregister(spa_t *, objset_t *os);
extern void spa_evicting_os_wait(spa_t *spa);
extern uint64_t spa_bootfs(spa_t *spa);
extern uint64_t spa_delegation(spa_t *spa);
extern objset_t *spa_meta_objset(spa_t *spa);
+extern space_map_t *spa_syncing_log_sm(spa_t *spa);
extern uint64_t spa_deadman_synctime(spa_t *spa);
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, ...)
+ __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 int spa_rename(const char *oldname, const char *newname);
extern spa_t *spa_by_guid(uint64_t pool_guid, uint64_t device_guid);
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);
extern boolean_t spa_has_pending_synctask(spa_t *spa);
extern int spa_maxblocksize(spa_t *spa);
extern int spa_maxdnodesize(spa_t *spa);
-extern void zfs_blkptr_verify(spa_t *spa, const blkptr_t *bp);
+extern boolean_t spa_has_checkpoint(spa_t *spa);
+extern boolean_t spa_importing_readonly_checkpoint(spa_t *spa);
+extern boolean_t spa_suspend_async_destroy(spa_t *spa);
+extern uint64_t spa_min_claim_txg(spa_t *spa);
+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 boolean_t spa_top_vdevs_spacemap_addressable(spa_t *spa);
+extern uint64_t spa_total_metaslabs(spa_t *spa);
extern boolean_t spa_multihost(spa_t *spa);
-extern unsigned long spa_get_hostid(void);
+extern uint32_t spa_get_hostid(spa_t *spa);
+extern void spa_activate_allocation_classes(spa_t *, dmu_tx_t *);
+extern boolean_t spa_livelist_delete_check(spa_t *spa);
-extern int spa_mode(spa_t *spa);
+extern spa_mode_t spa_mode(spa_t *spa);
extern uint64_t zfs_strtonum(const char *str, char **nptr);
extern char *spa_his_ievent_table[];
extern void spa_history_log_version(spa_t *spa, const char *operation,
dmu_tx_t *tx);
extern void spa_history_log_internal(spa_t *spa, const char *operation,
- dmu_tx_t *tx, const char *fmt, ...);
+ dmu_tx_t *tx, const char *fmt, ...) __printflike(4, 5);
extern void spa_history_log_internal_ds(struct dsl_dataset *ds, const char *op,
- dmu_tx_t *tx, const char *fmt, ...);
+ dmu_tx_t *tx, const char *fmt, ...) __printflike(4, 5);
extern void spa_history_log_internal_dd(dsl_dir_t *dd, const char *operation,
- dmu_tx_t *tx, const char *fmt, ...);
+ dmu_tx_t *tx, const char *fmt, ...) __printflike(4, 5);
+
+extern const char *spa_state_to_name(spa_t *spa);
/* error handling */
struct zbookmark_phys;
-extern void spa_log_error(spa_t *spa, const zbookmark_phys_t *zb);
-extern void zfs_ereport_post(const char *class, spa_t *spa, vdev_t *vd,
- zbookmark_phys_t *zb, zio_t *zio, uint64_t stateoroffset, uint64_t length);
+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 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);
extern void vdev_mirror_stat_fini(void);
/* Initialization and termination */
-extern void spa_init(int flags);
+extern void spa_init(spa_mode_t mode);
extern void spa_fini(void);
extern void spa_boot_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,
+ boolean_t *waited);
+extern int spa_wait_tag(const char *name, zpool_wait_activity_t activity,
+ uint64_t tag, boolean_t *waited);
+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 boolean_t spa_debug_enabled(spa_t *spa);
-#define spa_dbgmsg(spa, ...) \
-{ \
- if (spa_debug_enabled(spa)) \
- zfs_dbgmsg(__VA_ARGS__); \
-}
-
-extern int spa_mode_global; /* mode, e.g. FREAD | FWRITE */
+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
}
projects / mirror_zfs.git / blobdiff