diff options
Diffstat (limited to 'fs/bcachefs/bcachefs.h')
| -rw-r--r-- | fs/bcachefs/bcachefs.h | 430 |
1 files changed, 213 insertions, 217 deletions
diff --git a/fs/bcachefs/bcachefs.h b/fs/bcachefs/bcachefs.h index 91c3c1fef233..7824da2af9d0 100644 --- a/fs/bcachefs/bcachefs.h +++ b/fs/bcachefs/bcachefs.h @@ -203,20 +203,24 @@ #include <linux/types.h> #include <linux/workqueue.h> #include <linux/zstd.h> +#include <linux/unicode.h> #include "bcachefs_format.h" +#include "btree_journal_iter_types.h" +#include "disk_accounting_types.h" #include "errcode.h" +#include "fast_list.h" #include "fifo.h" #include "nocow_locking_types.h" #include "opts.h" -#include "recovery_passes_types.h" #include "sb-errors_types.h" #include "seqmutex.h" +#include "snapshot_types.h" #include "time_stats.h" #include "util.h" #ifdef CONFIG_BCACHEFS_DEBUG -#define BCH_WRITE_REF_DEBUG +#define ENUMERATED_REF_DEBUG #endif #ifndef dynamic_fault @@ -266,6 +270,9 @@ do { \ #define bch2_fmt(_c, fmt) bch2_log_msg(_c, fmt "\n") +void bch2_print_str(struct bch_fs *, const char *, const char *); +void bch2_print_str_nonblocking(struct bch_fs *, const char *, const char *); + __printf(2, 3) void bch2_print_opts(struct bch_opts *, const char *, ...); @@ -288,8 +295,20 @@ do { \ bch2_print(_c, __VA_ARGS__); \ } while (0) +#define bch2_print_str_ratelimited(_c, ...) \ +do { \ + static DEFINE_RATELIMIT_STATE(_rs, \ + DEFAULT_RATELIMIT_INTERVAL, \ + DEFAULT_RATELIMIT_BURST); \ + \ + if (__ratelimit(&_rs)) \ + bch2_print_str(_c, __VA_ARGS__); \ +} while (0) + #define bch_info(c, fmt, ...) \ bch2_print(c, KERN_INFO bch2_fmt(c, fmt), ##__VA_ARGS__) +#define bch_info_ratelimited(c, fmt, ...) \ + bch2_print_ratelimited(c, KERN_INFO bch2_fmt(c, fmt), ##__VA_ARGS__) #define bch_notice(c, fmt, ...) \ bch2_print(c, KERN_NOTICE bch2_fmt(c, fmt), ##__VA_ARGS__) #define bch_warn(c, fmt, ...) \ @@ -349,6 +368,12 @@ do { \ bch_info(c, fmt, ##__VA_ARGS__); \ } while (0) +#define bch_verbose_ratelimited(c, fmt, ...) \ +do { \ + if ((c)->opts.verbose) \ + bch_info_ratelimited(c, fmt, ##__VA_ARGS__); \ +} while (0) + #define pr_verbose_init(opts, fmt, ...) \ do { \ if (opt_get(opts, verbose)) \ @@ -359,6 +384,8 @@ do { \ #define BCH_DEBUG_PARAMS_ALWAYS() \ BCH_DEBUG_PARAM(key_merging_disabled, \ "Disables merging of extents") \ + BCH_DEBUG_PARAM(btree_node_merging_disabled, \ + "Disables merging of btree nodes") \ BCH_DEBUG_PARAM(btree_gc_always_rewrite, \ "Causes mark and sweep to compact and rewrite every " \ "btree node it traverses") \ @@ -375,17 +402,20 @@ do { \ "compare them") \ BCH_DEBUG_PARAM(backpointers_no_use_write_buffer, \ "Don't use the write buffer for backpointers, enabling "\ - "extra runtime checks") - -/* Parameters that should only be compiled in debug mode: */ -#define BCH_DEBUG_PARAMS_DEBUG() \ - BCH_DEBUG_PARAM(expensive_debug_checks, \ - "Enables various runtime debugging checks that " \ - "significantly affect performance") \ + "extra runtime checks") \ + BCH_DEBUG_PARAM(debug_check_btree_locking, \ + "Enable additional asserts for btree locking") \ BCH_DEBUG_PARAM(debug_check_iterators, \ "Enables extra verification for btree iterators") \ + BCH_DEBUG_PARAM(debug_check_bset_lookups, \ + "Enables extra verification for bset lookups") \ BCH_DEBUG_PARAM(debug_check_btree_accounting, \ "Verify btree accounting for keys within a node") \ + BCH_DEBUG_PARAM(debug_check_bkey_unpack, \ + "Enables extra verification for bkey unpack") + +/* Parameters that should only be compiled in debug mode: */ +#define BCH_DEBUG_PARAMS_DEBUG() \ BCH_DEBUG_PARAM(journal_seq_verify, \ "Store the journal sequence number in the version " \ "number of every btree key, and verify that btree " \ @@ -412,15 +442,9 @@ do { \ #define BCH_DEBUG_PARAMS() BCH_DEBUG_PARAMS_ALWAYS() #endif -#define BCH_DEBUG_PARAM(name, description) extern bool bch2_##name; -BCH_DEBUG_PARAMS() -#undef BCH_DEBUG_PARAM - -#ifndef CONFIG_BCACHEFS_DEBUG -#define BCH_DEBUG_PARAM(name, description) static const __maybe_unused bool bch2_##name; -BCH_DEBUG_PARAMS_DEBUG() +#define BCH_DEBUG_PARAM(name, description) extern struct static_key_false bch2_##name; +BCH_DEBUG_PARAMS_ALL() #undef BCH_DEBUG_PARAM -#endif #define BCH_TIME_STATS() \ x(btree_node_mem_alloc) \ @@ -428,12 +452,18 @@ BCH_DEBUG_PARAMS_DEBUG() x(btree_node_compact) \ x(btree_node_merge) \ x(btree_node_sort) \ + x(btree_node_get) \ x(btree_node_read) \ x(btree_node_read_done) \ + x(btree_node_write) \ x(btree_interior_update_foreground) \ x(btree_interior_update_total) \ x(btree_gc) \ x(data_write) \ + x(data_write_to_submit) \ + x(data_write_to_queue) \ + x(data_write_to_btree_update) \ + x(data_write_btree_update) \ x(data_read) \ x(data_promote) \ x(journal_flush_write) \ @@ -442,6 +472,8 @@ BCH_DEBUG_PARAMS_DEBUG() x(blocked_journal_low_on_space) \ x(blocked_journal_low_on_pin) \ x(blocked_journal_max_in_flight) \ + x(blocked_journal_max_open) \ + x(blocked_key_cache_flush) \ x(blocked_allocate) \ x(blocked_allocate_open_bucket) \ x(blocked_write_buffer_full) \ @@ -455,6 +487,8 @@ enum bch_time_stats { }; #include "alloc_types.h" +#include "async_objs_types.h" +#include "btree_gc_types.h" #include "btree_types.h" #include "btree_node_scan_types.h" #include "btree_write_buffer_types.h" @@ -463,11 +497,14 @@ enum bch_time_stats { #include "clock_types.h" #include "disk_groups_types.h" #include "ec_types.h" +#include "enumerated_ref_types.h" #include "journal_types.h" #include "keylist_types.h" #include "quota_types.h" #include "rebalance_types.h" +#include "recovery_passes_types.h" #include "replicas_types.h" +#include "sb-members_types.h" #include "subvolume_types.h" #include "super_types.h" #include "thread_with_file_types.h" @@ -485,59 +522,77 @@ enum bch_time_stats { struct btree; -enum gc_phase { - GC_PHASE_NOT_RUNNING, - GC_PHASE_START, - GC_PHASE_SB, - - GC_PHASE_BTREE_stripes, - GC_PHASE_BTREE_extents, - GC_PHASE_BTREE_inodes, - GC_PHASE_BTREE_dirents, - GC_PHASE_BTREE_xattrs, - GC_PHASE_BTREE_alloc, - GC_PHASE_BTREE_quotas, - GC_PHASE_BTREE_reflink, - GC_PHASE_BTREE_subvolumes, - GC_PHASE_BTREE_snapshots, - GC_PHASE_BTREE_lru, - GC_PHASE_BTREE_freespace, - GC_PHASE_BTREE_need_discard, - GC_PHASE_BTREE_backpointers, - GC_PHASE_BTREE_bucket_gens, - GC_PHASE_BTREE_snapshot_trees, - GC_PHASE_BTREE_deleted_inodes, - GC_PHASE_BTREE_logged_ops, - GC_PHASE_BTREE_rebalance_work, - GC_PHASE_BTREE_subvolume_children, - - GC_PHASE_PENDING_DELETE, +struct io_count { + u64 sectors[2][BCH_DATA_NR]; }; -struct gc_pos { - enum gc_phase phase; - struct bpos pos; - unsigned level; +struct discard_in_flight { + bool in_progress:1; + u64 bucket:63; }; -struct reflink_gc { - u64 offset; - u32 size; - u32 refcount; +#define BCH_DEV_READ_REFS() \ + x(bch2_online_devs) \ + x(trans_mark_dev_sbs) \ + x(read_fua_test) \ + x(sb_field_resize) \ + x(write_super) \ + x(journal_read) \ + x(fs_journal_alloc) \ + x(fs_resize_on_mount) \ + x(btree_node_read) \ + x(btree_node_read_all_replicas) \ + x(btree_node_scrub) \ + x(btree_node_write) \ + x(btree_node_scan) \ + x(btree_verify_replicas) \ + x(btree_node_ondisk_to_text) \ + x(io_read) \ + x(check_extent_checksums) \ + x(ec_block) + +enum bch_dev_read_ref { +#define x(n) BCH_DEV_READ_REF_##n, + BCH_DEV_READ_REFS() +#undef x + BCH_DEV_READ_REF_NR, }; -typedef GENRADIX(struct reflink_gc) reflink_gc_table; +#define BCH_DEV_WRITE_REFS() \ + x(journal_write) \ + x(journal_do_discards) \ + x(dev_do_discards) \ + x(discard_one_bucket_fast) \ + x(do_invalidates) \ + x(nocow_flush) \ + x(io_write) \ + x(ec_block) \ + x(ec_bucket_zero) + +enum bch_dev_write_ref { +#define x(n) BCH_DEV_WRITE_REF_##n, + BCH_DEV_WRITE_REFS() +#undef x + BCH_DEV_WRITE_REF_NR, +}; -struct io_count { - u64 sectors[2][BCH_DATA_NR]; +struct bucket_bitmap { + unsigned long *buckets; + u64 nr; + struct mutex lock; }; struct bch_dev { struct kobject kobj; +#ifdef CONFIG_BCACHEFS_DEBUG + atomic_long_t ref; + bool dying; + unsigned long last_put; +#else struct percpu_ref ref; +#endif struct completion ref_completion; - struct percpu_ref io_ref; - struct completion io_ref_completion; + struct enumerated_ref io_ref[2]; struct bch_fs *fs; @@ -548,6 +603,7 @@ struct bch_dev { */ struct bch_member_cpu mi; atomic64_t errors[BCH_MEMBER_ERROR_NR]; + unsigned long write_errors_start; __uuid_t uuid; char name[BDEVNAME_SIZE]; @@ -560,35 +616,34 @@ struct bch_dev { struct bch_devs_mask self; - /* biosets used in cloned bios for writing multiple replicas */ - struct bio_set replica_set; - /* * Buckets: - * Per-bucket arrays are protected by c->mark_lock, bucket_lock and - * gc_lock, for device resize - holding any is sufficient for access: - * Or rcu_read_lock(), but only for ptr_stale(): + * Per-bucket arrays are protected by either rcu_read_lock or + * state_lock, for device resize. */ - struct bucket_array __rcu *buckets_gc; + GENRADIX(struct bucket) buckets_gc; struct bucket_gens __rcu *bucket_gens; u8 *oldest_gen; unsigned long *buckets_nouse; - struct rw_semaphore bucket_lock; - struct bch_dev_usage *usage_base; - struct bch_dev_usage __percpu *usage[JOURNAL_BUF_NR]; - struct bch_dev_usage __percpu *usage_gc; + struct bucket_bitmap bucket_backpointer_mismatch; + struct bucket_bitmap bucket_backpointer_empty; + + struct bch_dev_usage_full __percpu + *usage; /* Allocator: */ - u64 new_fs_bucket_idx; - u64 alloc_cursor; + u64 alloc_cursor[3]; unsigned nr_open_buckets; + unsigned nr_partial_buckets; unsigned nr_btree_reserve; - size_t inc_gen_needs_gc; - size_t inc_gen_really_needs_gc; - size_t buckets_waiting_on_journal; + struct work_struct invalidate_work; + struct work_struct discard_work; + struct mutex discard_buckets_in_flight_lock; + DARRAY(struct discard_in_flight) discard_buckets_in_flight; + struct work_struct discard_fast_work; atomic64_t rebalance_work; @@ -617,22 +672,28 @@ struct bch_dev { #define BCH_FS_FLAGS() \ x(new_fs) \ x(started) \ + x(clean_recovery) \ + x(btree_running) \ + x(accounting_replay_done) \ x(may_go_rw) \ x(rw) \ + x(rw_init_done) \ x(was_rw) \ x(stopping) \ x(emergency_ro) \ x(going_ro) \ x(write_disable_complete) \ x(clean_shutdown) \ - x(fsck_running) \ + x(in_recovery) \ + x(in_fsck) \ x(initial_gc_unfixed) \ - x(need_another_gc) \ x(need_delete_dead_snapshots) \ x(error) \ x(topology_error) \ x(errors_fixed) \ - x(errors_not_fixed) + x(errors_not_fixed) \ + x(no_invalid_checks) \ + x(discard_mount_opt_set) \ enum bch_fs_flags { #define x(n) BCH_FS_##n, @@ -651,8 +712,10 @@ struct btree_transaction_stats { struct bch2_time_stats lock_hold_times; struct mutex lock; unsigned nr_max_paths; - unsigned journal_entries_size; unsigned max_mem; +#ifdef CONFIG_BCACHEFS_TRANS_KMALLOC_TRACE + darray_trans_kmalloc_trace trans_kmalloc_trace; +#endif char *max_paths_text; }; @@ -669,38 +732,12 @@ struct journal_seq_blacklist_table { } entries[]; }; -struct journal_keys { - /* must match layout in darray_types.h */ - size_t nr, size; - struct journal_key { - u64 journal_seq; - u32 journal_offset; - enum btree_id btree_id:8; - unsigned level:8; - bool allocated; - bool overwritten; - struct bkey_i *k; - } *data; - /* - * Gap buffer: instead of all the empty space in the array being at the - * end of the buffer - from @nr to @size - the empty space is at @gap. - * This means that sequential insertions are O(n) instead of O(n^2). - */ - size_t gap; - atomic_t ref; - bool initial_ref_held; -}; - struct btree_trans_buf { struct btree_trans *trans; }; -#define REPLICAS_DELTA_LIST_MAX (1U << 16) - -#define BCACHEFS_ROOT_SUBVOL_INUM \ - ((subvol_inum) { BCACHEFS_ROOT_SUBVOL, BCACHEFS_ROOT_INO }) - #define BCH_WRITE_REFS() \ + x(journal) \ x(trans) \ x(write) \ x(promote) \ @@ -713,11 +750,15 @@ struct btree_trans_buf { x(dio_write) \ x(discard) \ x(discard_fast) \ + x(check_discard_freespace_key) \ x(invalidate) \ x(delete_dead_snapshots) \ + x(gc_gens) \ x(snapshot_delete_pagecache) \ x(sysfs) \ - x(btree_write_buffer) + x(btree_write_buffer) \ + x(btree_node_scrub) \ + x(async_recovery_passes) enum bch_write_ref { #define x(n) BCH_WRITE_REF_##n, @@ -726,6 +767,8 @@ enum bch_write_ref { BCH_WRITE_REF_NR, }; +#define BCH_FS_DEFAULT_UTF8_ENCODING UNICODE_AGE(12, 1, 0) + struct bch_fs { struct closure cl; @@ -749,11 +792,13 @@ struct bch_fs { struct rw_semaphore state_lock; /* Counts outstanding writes, for clean transition to read-only */ -#ifdef BCH_WRITE_REF_DEBUG - atomic_long_t writes[BCH_WRITE_REF_NR]; -#else - struct percpu_ref writes; -#endif + struct enumerated_ref writes; + /* + * Certain operations are only allowed in single threaded mode, during + * recovery, and we want to assert that this is the case: + */ + struct task_struct *recovery_task; + /* * Analagous to c->writes, for asynchronous ops that don't necessarily * need fs to be read-write @@ -765,15 +810,14 @@ struct bch_fs { struct bch_dev __rcu *devs[BCH_SB_MEMBERS_MAX]; + struct bch_accounting_mem accounting; + struct bch_replicas_cpu replicas; struct bch_replicas_cpu replicas_gc; struct mutex replicas_gc_lock; - mempool_t replicas_delta_pool; struct journal_entry_res btree_root_journal_res; - struct journal_entry_res replicas_journal_res; struct journal_entry_res clock_journal_res; - struct journal_entry_res dev_usage_journal_res; struct bch_disk_groups_cpu __rcu *disk_groups; @@ -785,11 +829,14 @@ struct bch_fs { __uuid_t user_uuid; u16 version; + u16 version_incompat; + u16 version_incompat_allowed; u16 version_min; u16 version_upgrade_complete; u8 nr_devices; u8 clean; + bool multi_device; /* true if we've ever had more than one device */ u8 encryption_type; @@ -799,10 +846,16 @@ struct bch_fs { unsigned nsec_per_time_unit; u64 features; u64 compat; - unsigned long errors_silent[BITS_TO_LONGS(BCH_SB_ERR_MAX)]; + u64 recovery_passes_required; + unsigned long errors_silent[BITS_TO_LONGS(BCH_FSCK_ERR_MAX)]; u64 btrees_lost_data; } sb; + DARRAY(enum bcachefs_metadata_version) + incompat_versions_requested; +#ifdef CONFIG_UNICODE + struct unicode_map *cf_encoding; +#endif struct bch_sb_handle disk_sb; @@ -818,14 +871,15 @@ struct bch_fs { struct mutex snapshot_table_lock; struct rw_semaphore snapshot_create_lock; - struct work_struct snapshot_delete_work; + struct snapshot_delete snapshot_delete; struct work_struct snapshot_wait_for_pagecache_and_delete_work; snapshot_id_list snapshots_unlinked; struct mutex snapshots_unlinked_lock; /* BTREE CACHE */ struct bio_set btree_bio; - struct workqueue_struct *io_complete_wq; + struct workqueue_struct *btree_read_complete_wq; + struct workqueue_struct *btree_write_submit_wq; struct btree_root btree_roots_known[BTREE_ID_NR]; DARRAY(struct btree_root) btree_roots_extra; @@ -854,9 +908,10 @@ struct bch_fs { struct work_struct btree_interior_update_work; struct workqueue_struct *btree_node_rewrite_worker; - - struct list_head pending_node_rewrites; - struct mutex pending_node_rewrites_lock; + struct list_head btree_node_rewrites; + struct list_head btree_node_rewrites_pending; + spinlock_t btree_node_rewrites_lock; + struct closure_waitlist btree_node_rewrites_wait; /* btree_io.c: */ spinlock_t btree_write_error_lock; @@ -881,7 +936,7 @@ struct bch_fs { struct btree_write_buffer btree_write_buffer; struct workqueue_struct *btree_update_wq; - struct workqueue_struct *btree_io_complete_wq; + struct workqueue_struct *btree_write_complete_wq; /* copygc needs its own workqueue for index updates.. */ struct workqueue_struct *copygc_wq; /* @@ -892,9 +947,12 @@ struct bch_fs { struct workqueue_struct *write_ref_wq; /* ALLOCATION */ + struct bch_devs_mask online_devs; struct bch_devs_mask rw_devs[BCH_DATA_NR]; + unsigned long rw_devs_change_count; u64 capacity; /* sectors */ + u64 reserved; /* sectors */ /* * When capacity _decreases_ (due to a disk being removed), we @@ -912,21 +970,16 @@ struct bch_fs { struct percpu_rw_semaphore mark_lock; seqcount_t usage_lock; - struct bch_fs_usage *usage_base; - struct bch_fs_usage __percpu *usage[JOURNAL_BUF_NR]; - struct bch_fs_usage __percpu *usage_gc; + struct bch_fs_usage_base __percpu *usage; u64 __percpu *online_reserved; - /* single element mempool: */ - struct mutex usage_scratch_lock; - struct bch_fs_usage_online *usage_scratch; + unsigned long allocator_last_stuck; struct io_clock io_clock[2]; /* JOURNAL SEQ BLACKLIST */ struct journal_seq_blacklist_table * journal_seq_blacklist_table; - struct work_struct journal_seq_blacklist_gc_work; /* ALLOCATOR */ spinlock_t freelist_lock; @@ -950,15 +1003,9 @@ struct bch_fs { unsigned write_points_nr; struct buckets_waiting_for_journal buckets_waiting_for_journal; - struct work_struct invalidate_work; - struct work_struct discard_work; - struct mutex discard_buckets_in_flight_lock; - DARRAY(struct bpos) discard_buckets_in_flight; - struct work_struct discard_fast_work; /* GARBAGE COLLECTION */ - struct task_struct *gc_thread; - atomic_t kick_gc; + struct work_struct gc_gens_work; unsigned long gc_count; enum btree_id gc_gens_btree; @@ -988,20 +1035,23 @@ struct bch_fs { struct bio_set bio_read; struct bio_set bio_read_split; struct bio_set bio_write; + struct bio_set replica_set; struct mutex bio_bounce_pages_lock; mempool_t bio_bounce_pages; struct bucket_nocow_lock_table nocow_locks; struct rhashtable promote_table; +#ifdef CONFIG_BCACHEFS_ASYNC_OBJECT_LISTS + struct async_obj_list async_objs[BCH_ASYNC_OBJ_NR]; +#endif + mempool_t compression_bounce[2]; - mempool_t compress_workspace[BCH_COMPRESSION_TYPE_NR]; - mempool_t decompress_workspace; + mempool_t compress_workspace[BCH_COMPRESSION_OPT_NR]; size_t zstd_workspace_size; - struct crypto_shash *sha256; - struct crypto_sync_skcipher *chacha20; - struct crypto_shash *poly1305; + struct bch_key chacha20_key; + bool chacha20_key_set; atomic64_t key_version; @@ -1023,15 +1073,11 @@ struct bch_fs { wait_queue_head_t copygc_running_wq; /* STRIPES: */ - GENRADIX(struct stripe) stripes; GENRADIX(struct gc_stripe) gc_stripes; struct hlist_head ec_stripes_new[32]; spinlock_t ec_stripes_new_lock; - ec_stripes_heap ec_stripes_heap; - struct mutex ec_stripes_heap_lock; - /* ERASURE CODING */ struct list_head ec_stripe_head_list; struct mutex ec_stripe_head_lock; @@ -1054,6 +1100,8 @@ struct bch_fs { /* fs.c */ struct list_head vfs_inodes_list; struct mutex vfs_inodes_lock; + struct rhashtable vfs_inodes_table; + struct rhltable vfs_inodes_by_inum_table; /* VFS IO PATH - fs-io.c */ struct bio_set writepage_bioset; @@ -1067,33 +1115,17 @@ struct bch_fs { /* RECOVERY */ u64 journal_replay_seq_start; u64 journal_replay_seq_end; - /* - * Two different uses: - * "Has this fsck pass?" - i.e. should this type of error be an - * emergency read-only - * And, in certain situations fsck will rewind to an earlier pass: used - * for signaling to the toplevel code which pass we want to run now. - */ - enum bch_recovery_pass curr_recovery_pass; - /* bitmap of explicitly enabled recovery passes: */ - u64 recovery_passes_explicit; - /* bitmask of recovery passes that we actually ran */ - u64 recovery_passes_complete; - /* never rewinds version of curr_recovery_pass */ - enum bch_recovery_pass recovery_pass_done; - struct semaphore online_fsck_mutex; + struct bch_fs_recovery recovery; /* DEBUG JUNK */ struct dentry *fs_debug_dir; struct dentry *btree_debug_dir; + struct dentry *async_obj_dir; struct btree_debug btree_debug[BTREE_ID_NR]; struct btree *verify_data; struct btree_node *verify_ondisk; struct mutex verify_lock; - u64 *unused_inode_hints; - unsigned inode_shard_bits; - /* * A btree node on disk could have too many bsets for an iterator to fit * on the stack - have to dynamically allocate them @@ -1115,10 +1147,6 @@ struct bch_fs { u64 counters_on_mount[BCH_COUNTER_NR]; u64 __percpu *counters; - unsigned btree_gc_periodic:1; - unsigned copy_gc_enabled:1; - bool promote_whole_extents; - struct bch2_time_stats times[BCH_TIME_STAT_NR]; struct btree_transaction_stats btree_transaction_stats[BCH_TRANSACTIONS_NR]; @@ -1134,54 +1162,6 @@ struct bch_fs { extern struct wait_queue_head bch2_read_only_wait; -static inline void bch2_write_ref_get(struct bch_fs *c, enum bch_write_ref ref) -{ -#ifdef BCH_WRITE_REF_DEBUG - atomic_long_inc(&c->writes[ref]); -#else - percpu_ref_get(&c->writes); -#endif -} - -static inline bool __bch2_write_ref_tryget(struct bch_fs *c, enum bch_write_ref ref) -{ -#ifdef BCH_WRITE_REF_DEBUG - return !test_bit(BCH_FS_going_ro, &c->flags) && - atomic_long_inc_not_zero(&c->writes[ref]); -#else - return percpu_ref_tryget(&c->writes); -#endif -} - -static inline bool bch2_write_ref_tryget(struct bch_fs *c, enum bch_write_ref ref) -{ -#ifdef BCH_WRITE_REF_DEBUG - return !test_bit(BCH_FS_going_ro, &c->flags) && - atomic_long_inc_not_zero(&c->writes[ref]); -#else - return percpu_ref_tryget_live(&c->writes); -#endif -} - -static inline void bch2_write_ref_put(struct bch_fs *c, enum bch_write_ref ref) -{ -#ifdef BCH_WRITE_REF_DEBUG - long v = atomic_long_dec_return(&c->writes[ref]); - - BUG_ON(v < 0); - if (v) - return; - for (unsigned i = 0; i < BCH_WRITE_REF_NR; i++) - if (atomic_long_read(&c->writes[i])) - return; - - set_bit(BCH_FS_write_disable_complete, &c->flags); - wake_up(&bch2_read_only_wait); -#else - percpu_ref_put(&c->writes); -#endif -} - static inline bool bch2_ro_ref_tryget(struct bch_fs *c) { if (test_bit(BCH_FS_stopping, &c->flags)) @@ -1227,12 +1207,15 @@ static inline bool btree_id_cached(const struct bch_fs *c, enum btree_id btree) static inline struct timespec64 bch2_time_to_timespec(const struct bch_fs *c, s64 time) { struct timespec64 t; + s64 sec; s32 rem; time += c->sb.time_base_lo; - t.tv_sec = div_s64_rem(time, c->sb.time_units_per_sec, &rem); - t.tv_nsec = rem * c->sb.nsec_per_time_unit; + sec = div_s64_rem(time, c->sb.time_units_per_sec, &rem); + + set_normalized_timespec64(&t, sec, rem * (s64)c->sb.nsec_per_time_unit); + return t; } @@ -1250,9 +1233,9 @@ static inline s64 bch2_current_time(const struct bch_fs *c) return timespec_to_bch2_time(c, now); } -static inline bool bch2_dev_exists2(const struct bch_fs *c, unsigned dev) +static inline u64 bch2_current_io_time(const struct bch_fs *c, int rw) { - return dev < c->sb.nr_devices && c->devs[dev]; + return max(1ULL, (u64) atomic64_read(&c->io_clock[rw].now) & LRU_TIME_MAX); } static inline struct stdio_redirect *bch2_fs_stdio_redirect(struct bch_fs *c) @@ -1279,4 +1262,17 @@ static inline unsigned data_replicas_required(struct bch_fs *c) #define BKEY_PADDED_ONSTACK(key, pad) \ struct { struct bkey_i key; __u64 key ## _pad[pad]; } +/* + * This is needed because discard is both a filesystem option and a device + * option, and mount options are supposed to apply to that mount and not be + * persisted, i.e. if it's set as a mount option we can't propagate it to the + * device. + */ +static inline bool bch2_discard_opt_enabled(struct bch_fs *c, struct bch_dev *ca) +{ + return test_bit(BCH_FS_discard_mount_opt_set, &c->flags) + ? c->opts.discard + : ca->mi.discard; +} + #endif /* _BCACHEFS_H */ |