X-Git-Url: https://git.sesse.net/?a=blobdiff_plain;f=libbcachefs%2Fbuckets.h;h=6387e039f7897534e27c207dd3818dc4b6afb3b7;hb=e5b2870d05753c1dedd79261ba5e321ce953f5ab;hp=399a853c80bbcb8c1f13c610c0e57b835c203eed;hpb=c598d91dcb0c7e95abdacb2711898ae14ab52ca1;p=bcachefs-tools-debian diff --git a/libbcachefs/buckets.h b/libbcachefs/buckets.h index 399a853..6387e03 100644 --- a/libbcachefs/buckets.h +++ b/libbcachefs/buckets.h @@ -1,3 +1,4 @@ +/* SPDX-License-Identifier: GPL-2.0 */ /* * Code for manipulating bucket marks for garbage collection. * @@ -8,62 +9,110 @@ #define _BUCKETS_H #include "buckets_types.h" -#include "super.h" +#include "extents.h" +#include "sb-members.h" + +static inline size_t sector_to_bucket(const struct bch_dev *ca, sector_t s) +{ + return div_u64(s, ca->mi.bucket_size); +} + +static inline sector_t bucket_to_sector(const struct bch_dev *ca, size_t b) +{ + return ((sector_t) b) * ca->mi.bucket_size; +} + +static inline sector_t bucket_remainder(const struct bch_dev *ca, sector_t s) +{ + u32 remainder; + + div_u64_rem(s, ca->mi.bucket_size, &remainder); + return remainder; +} + +static inline size_t sector_to_bucket_and_offset(const struct bch_dev *ca, sector_t s, + u32 *offset) +{ + return div_u64_rem(s, ca->mi.bucket_size, offset); +} #define for_each_bucket(_b, _buckets) \ for (_b = (_buckets)->b + (_buckets)->first_bucket; \ _b < (_buckets)->b + (_buckets)->nbuckets; _b++) -#define bucket_cmpxchg(g, new, expr) \ -({ \ - u64 _v = READ_ONCE((g)->_mark.counter); \ - struct bucket_mark _old; \ - \ - do { \ - (new).counter = _old.counter = _v; \ - expr; \ - } while ((_v = cmpxchg(&(g)->_mark.counter, \ - _old.counter, \ - (new).counter)) != _old.counter);\ - _old; \ -}) +/* + * Ugly hack alert: + * + * We need to cram a spinlock in a single byte, because that's what we have left + * in struct bucket, and we care about the size of these - during fsck, we need + * in memory state for every single bucket on every device. + * + * We used to do + * while (xchg(&b->lock, 1) cpu_relax(); + * but, it turns out not all architectures support xchg on a single byte. + * + * So now we use bit_spin_lock(), with fun games since we can't burn a whole + * ulong for this - we just need to make sure the lock bit always ends up in the + * first byte. + */ + +#if __BYTE_ORDER__ == __ORDER_LITTLE_ENDIAN__ +#define BUCKET_LOCK_BITNR 0 +#else +#define BUCKET_LOCK_BITNR (BITS_PER_LONG - 1) +#endif + +union ulong_byte_assert { + ulong ulong; + u8 byte; +}; + +static inline void bucket_unlock(struct bucket *b) +{ + BUILD_BUG_ON(!((union ulong_byte_assert) { .ulong = 1UL << BUCKET_LOCK_BITNR }).byte); + + clear_bit_unlock(BUCKET_LOCK_BITNR, (void *) &b->lock); + wake_up_bit((void *) &b->lock, BUCKET_LOCK_BITNR); +} + +static inline void bucket_lock(struct bucket *b) +{ + wait_on_bit_lock((void *) &b->lock, BUCKET_LOCK_BITNR, + TASK_UNINTERRUPTIBLE); +} -static inline struct bucket_array *bucket_array(struct bch_dev *ca) +static inline struct bucket_array *gc_bucket_array(struct bch_dev *ca) { - return rcu_dereference_check(ca->buckets, + return rcu_dereference_check(ca->buckets_gc, !ca->fs || - lockdep_is_held(&ca->fs->usage_lock) || + percpu_rwsem_is_held(&ca->fs->mark_lock) || lockdep_is_held(&ca->fs->gc_lock) || lockdep_is_held(&ca->bucket_lock)); } -static inline struct bucket *bucket(struct bch_dev *ca, size_t b) +static inline struct bucket *gc_bucket(struct bch_dev *ca, size_t b) { - struct bucket_array *buckets = bucket_array(ca); + struct bucket_array *buckets = gc_bucket_array(ca); BUG_ON(b < buckets->first_bucket || b >= buckets->nbuckets); return buckets->b + b; } -static inline void bucket_io_clock_reset(struct bch_fs *c, struct bch_dev *ca, - size_t b, int rw) +static inline struct bucket_gens *bucket_gens(struct bch_dev *ca) { - bucket(ca, b)->io_time[rw] = c->bucket_clock[rw].hand; + return rcu_dereference_check(ca->bucket_gens, + !ca->fs || + percpu_rwsem_is_held(&ca->fs->mark_lock) || + lockdep_is_held(&ca->fs->gc_lock) || + lockdep_is_held(&ca->bucket_lock)); } -static inline u16 bucket_last_io(struct bch_fs *c, struct bucket *g, int rw) +static inline u8 *bucket_gen(struct bch_dev *ca, size_t b) { - return c->bucket_clock[rw].hand - g->io_time[rw]; -} - -/* - * bucket_gc_gen() returns the difference between the bucket's current gen and - * the oldest gen of any pointer into that bucket in the btree. - */ + struct bucket_gens *gens = bucket_gens(ca); -static inline u8 bucket_gc_gen(struct bch_dev *ca, size_t b) -{ - return bucket(ca, b)->mark.gen - ca->oldest_gens[b]; + BUG_ON(b < gens->first_bucket || b >= gens->nbuckets); + return gens->b + b; } static inline size_t PTR_BUCKET_NR(const struct bch_dev *ca, @@ -72,22 +121,46 @@ static inline size_t PTR_BUCKET_NR(const struct bch_dev *ca, return sector_to_bucket(ca, ptr->offset); } -static inline struct bucket *PTR_BUCKET(struct bch_dev *ca, - const struct bch_extent_ptr *ptr) +static inline struct bpos PTR_BUCKET_POS(const struct bch_fs *c, + const struct bch_extent_ptr *ptr) { - return bucket(ca, PTR_BUCKET_NR(ca, ptr)); + struct bch_dev *ca = bch_dev_bkey_exists(c, ptr->dev); + + return POS(ptr->dev, PTR_BUCKET_NR(ca, ptr)); } -static inline struct bucket_mark ptr_bucket_mark(struct bch_dev *ca, - const struct bch_extent_ptr *ptr) +static inline struct bpos PTR_BUCKET_POS_OFFSET(const struct bch_fs *c, + const struct bch_extent_ptr *ptr, + u32 *bucket_offset) { - struct bucket_mark m; + struct bch_dev *ca = bch_dev_bkey_exists(c, ptr->dev); - rcu_read_lock(); - m = READ_ONCE(bucket(ca, PTR_BUCKET_NR(ca, ptr))->mark); - rcu_read_unlock(); + return POS(ptr->dev, sector_to_bucket_and_offset(ca, ptr->offset, bucket_offset)); +} - return m; +static inline struct bucket *PTR_GC_BUCKET(struct bch_dev *ca, + const struct bch_extent_ptr *ptr) +{ + return gc_bucket(ca, PTR_BUCKET_NR(ca, ptr)); +} + +static inline enum bch_data_type ptr_data_type(const struct bkey *k, + const struct bch_extent_ptr *ptr) +{ + if (bkey_is_btree_ptr(k)) + return BCH_DATA_btree; + + return ptr->cached ? BCH_DATA_cached : BCH_DATA_user; +} + +static inline s64 ptr_disk_sectors(s64 sectors, struct extent_ptr_decoded p) +{ + EBUG_ON(sectors < 0); + + return crc_is_compressed(p.crc) + ? DIV_ROUND_UP_ULL(sectors * p.crc.compressed_size, + p.crc.uncompressed_size) + : sectors; } static inline int gen_cmp(u8 a, u8 b) @@ -109,139 +182,264 @@ static inline int gen_after(u8 a, u8 b) static inline u8 ptr_stale(struct bch_dev *ca, const struct bch_extent_ptr *ptr) { - return gen_after(ptr_bucket_mark(ca, ptr).gen, ptr->gen); -} - -/* bucket gc marks */ + u8 ret; -/* The dirty and cached sector counts saturate. If this occurs, - * reference counting alone will not free the bucket, and a btree - * GC must be performed. */ -#define GC_MAX_SECTORS_USED ((1U << 15) - 1) + rcu_read_lock(); + ret = gen_after(*bucket_gen(ca, PTR_BUCKET_NR(ca, ptr)), ptr->gen); + rcu_read_unlock(); -static inline unsigned bucket_sectors_used(struct bucket_mark mark) -{ - return mark.dirty_sectors + mark.cached_sectors; + return ret; } -static inline bool bucket_unused(struct bucket_mark mark) +/* Device usage: */ + +void bch2_dev_usage_read_fast(struct bch_dev *, struct bch_dev_usage *); +static inline struct bch_dev_usage bch2_dev_usage_read(struct bch_dev *ca) { - return !mark.owned_by_allocator && - !mark.data_type && - !bucket_sectors_used(mark); -} + struct bch_dev_usage ret; -/* Device usage: */ + bch2_dev_usage_read_fast(ca, &ret); + return ret; +} -struct bch_dev_usage __bch2_dev_usage_read(struct bch_dev *); -struct bch_dev_usage bch2_dev_usage_read(struct bch_fs *, struct bch_dev *); +void bch2_dev_usage_init(struct bch_dev *); +void bch2_dev_usage_to_text(struct printbuf *, struct bch_dev_usage *); -static inline u64 __dev_buckets_available(struct bch_dev *ca, - struct bch_dev_usage stats) +static inline u64 bch2_dev_buckets_reserved(struct bch_dev *ca, enum bch_watermark watermark) { - u64 total = ca->mi.nbuckets - ca->mi.first_bucket; + s64 reserved = 0; - if (WARN_ONCE(stats.buckets_unavailable > total, - "buckets_unavailable overflow\n")) - return 0; + switch (watermark) { + case BCH_WATERMARK_NR: + BUG(); + case BCH_WATERMARK_stripe: + reserved += ca->mi.nbuckets >> 6; + fallthrough; + case BCH_WATERMARK_normal: + reserved += ca->mi.nbuckets >> 6; + fallthrough; + case BCH_WATERMARK_copygc: + reserved += ca->nr_btree_reserve; + fallthrough; + case BCH_WATERMARK_btree: + reserved += ca->nr_btree_reserve; + fallthrough; + case BCH_WATERMARK_btree_copygc: + case BCH_WATERMARK_reclaim: + break; + } - return total - stats.buckets_unavailable; + return reserved; } -/* - * Number of reclaimable buckets - only for use by the allocator thread: - */ -static inline u64 dev_buckets_available(struct bch_fs *c, struct bch_dev *ca) +static inline u64 dev_buckets_free(struct bch_dev *ca, + struct bch_dev_usage usage, + enum bch_watermark watermark) { - return __dev_buckets_available(ca, bch2_dev_usage_read(c, ca)); + return max_t(s64, 0, + usage.d[BCH_DATA_free].buckets - + ca->nr_open_buckets - + bch2_dev_buckets_reserved(ca, watermark)); } -static inline u64 __dev_buckets_free(struct bch_dev *ca, - struct bch_dev_usage stats) +static inline u64 __dev_buckets_available(struct bch_dev *ca, + struct bch_dev_usage usage, + enum bch_watermark watermark) { - return __dev_buckets_available(ca, stats) + - fifo_used(&ca->free[RESERVE_NONE]) + - fifo_used(&ca->free_inc); + return max_t(s64, 0, + usage.d[BCH_DATA_free].buckets + + usage.d[BCH_DATA_cached].buckets + + usage.d[BCH_DATA_need_gc_gens].buckets + + usage.d[BCH_DATA_need_discard].buckets + - ca->nr_open_buckets + - bch2_dev_buckets_reserved(ca, watermark)); } -static inline u64 dev_buckets_free(struct bch_fs *c, struct bch_dev *ca) +static inline u64 dev_buckets_available(struct bch_dev *ca, + enum bch_watermark watermark) { - return __dev_buckets_free(ca, bch2_dev_usage_read(c, ca)); + return __dev_buckets_available(ca, bch2_dev_usage_read(ca), watermark); } /* Filesystem usage: */ -static inline enum bch_data_type s_alloc_to_data_type(enum s_alloc s) +static inline unsigned __fs_usage_u64s(unsigned nr_replicas) { - switch (s) { - case S_META: - return BCH_DATA_BTREE; - case S_DIRTY: - return BCH_DATA_USER; - default: - BUG(); - } + return sizeof(struct bch_fs_usage) / sizeof(u64) + nr_replicas; } -struct bch_fs_usage __bch2_fs_usage_read(struct bch_fs *); -struct bch_fs_usage bch2_fs_usage_read(struct bch_fs *); -void bch2_fs_usage_apply(struct bch_fs *, struct bch_fs_usage *, - struct disk_reservation *, struct gc_pos); +static inline unsigned fs_usage_u64s(struct bch_fs *c) +{ + return __fs_usage_u64s(READ_ONCE(c->replicas.nr)); +} -u64 __bch2_fs_sectors_used(struct bch_fs *, struct bch_fs_usage); -u64 bch2_fs_sectors_used(struct bch_fs *, struct bch_fs_usage); -u64 bch2_fs_sectors_free(struct bch_fs *, struct bch_fs_usage); +static inline unsigned __fs_usage_online_u64s(unsigned nr_replicas) +{ + return sizeof(struct bch_fs_usage_online) / sizeof(u64) + nr_replicas; +} -static inline bool is_available_bucket(struct bucket_mark mark) +static inline unsigned fs_usage_online_u64s(struct bch_fs *c) { - return (!mark.owned_by_allocator && - !mark.dirty_sectors && - !mark.nouse); + return __fs_usage_online_u64s(READ_ONCE(c->replicas.nr)); } -static inline bool bucket_needs_journal_commit(struct bucket_mark m, - u16 last_seq_ondisk) +static inline unsigned dev_usage_u64s(void) +{ + return sizeof(struct bch_dev_usage) / sizeof(u64); +} + +u64 bch2_fs_usage_read_one(struct bch_fs *, u64 *); + +struct bch_fs_usage_online *bch2_fs_usage_read(struct bch_fs *); + +void bch2_fs_usage_acc_to_base(struct bch_fs *, unsigned); + +void bch2_fs_usage_to_text(struct printbuf *, + struct bch_fs *, struct bch_fs_usage_online *); + +u64 bch2_fs_sectors_used(struct bch_fs *, struct bch_fs_usage_online *); + +struct bch_fs_usage_short +bch2_fs_usage_read_short(struct bch_fs *); + +void bch2_dev_usage_update(struct bch_fs *, struct bch_dev *, + const struct bch_alloc_v4 *, + const struct bch_alloc_v4 *, u64, bool); +void bch2_dev_usage_update_m(struct bch_fs *, struct bch_dev *, + struct bucket *, struct bucket *); + +/* key/bucket marking: */ + +static inline struct bch_fs_usage *fs_usage_ptr(struct bch_fs *c, + unsigned journal_seq, + bool gc) { - return m.journal_seq_valid && - ((s16) m.journal_seq - (s16) last_seq_ondisk > 0); + percpu_rwsem_assert_held(&c->mark_lock); + BUG_ON(!gc && !journal_seq); + + return this_cpu_ptr(gc + ? c->usage_gc + : c->usage[journal_seq & JOURNAL_BUF_MASK]); } -void bch2_bucket_seq_cleanup(struct bch_fs *); +int bch2_update_replicas(struct bch_fs *, struct bkey_s_c, + struct bch_replicas_entry_v1 *, s64, + unsigned, bool); +int bch2_update_replicas_list(struct btree_trans *, + struct bch_replicas_entry_v1 *, s64); +int bch2_update_cached_sectors_list(struct btree_trans *, unsigned, s64); +int bch2_replicas_deltas_realloc(struct btree_trans *, unsigned); + +void bch2_fs_usage_initialize(struct bch_fs *); + +int bch2_check_bucket_ref(struct btree_trans *, struct bkey_s_c, + const struct bch_extent_ptr *, + s64, enum bch_data_type, u8, u8, u32); + +int bch2_mark_metadata_bucket(struct bch_fs *, struct bch_dev *, + size_t, enum bch_data_type, unsigned, + struct gc_pos, unsigned); + +int bch2_trigger_extent(struct btree_trans *, enum btree_id, unsigned, + struct bkey_s_c, struct bkey_s, unsigned); +int bch2_trigger_reservation(struct btree_trans *, enum btree_id, unsigned, + struct bkey_s_c, struct bkey_s, unsigned); + +#define trigger_run_overwrite_then_insert(_fn, _trans, _btree_id, _level, _old, _new, _flags)\ +({ \ + int ret = 0; \ + \ + if (_old.k->type) \ + ret = _fn(_trans, _btree_id, _level, _old, _flags & ~BTREE_TRIGGER_INSERT); \ + if (!ret && _new.k->type) \ + ret = _fn(_trans, _btree_id, _level, _new.s_c, _flags & ~BTREE_TRIGGER_OVERWRITE);\ + ret; \ +}) + +void bch2_trans_account_disk_usage_change(struct btree_trans *); + +void bch2_trans_fs_usage_revert(struct btree_trans *, struct replicas_delta_list *); +int bch2_trans_fs_usage_apply(struct btree_trans *, struct replicas_delta_list *); -bool bch2_invalidate_bucket(struct bch_fs *, struct bch_dev *, - size_t, struct bucket_mark *); -void bch2_mark_alloc_bucket(struct bch_fs *, struct bch_dev *, - size_t, bool, struct gc_pos, unsigned); -void bch2_mark_metadata_bucket(struct bch_fs *, struct bch_dev *, - size_t, enum bch_data_type, unsigned, - struct gc_pos, unsigned); +int bch2_trans_mark_metadata_bucket(struct btree_trans *, struct bch_dev *, + size_t, enum bch_data_type, unsigned); +int bch2_trans_mark_dev_sb(struct bch_fs *, struct bch_dev *); +int bch2_trans_mark_dev_sbs(struct bch_fs *); -#define BCH_BUCKET_MARK_NOATOMIC (1 << 0) -#define BCH_BUCKET_MARK_MAY_MAKE_UNAVAILABLE (1 << 1) -#define BCH_BUCKET_MARK_GC_WILL_VISIT (1 << 2) -#define BCH_BUCKET_MARK_GC_LOCK_HELD (1 << 3) +static inline bool is_superblock_bucket(struct bch_dev *ca, u64 b) +{ + struct bch_sb_layout *layout = &ca->disk_sb.sb->layout; + u64 b_offset = bucket_to_sector(ca, b); + u64 b_end = bucket_to_sector(ca, b + 1); + unsigned i; + + if (!b) + return true; -void bch2_mark_key(struct bch_fs *, struct bkey_s_c, s64, bool, struct gc_pos, - struct bch_fs_usage *, u64, unsigned); + for (i = 0; i < layout->nr_superblocks; i++) { + u64 offset = le64_to_cpu(layout->sb_offset[i]); + u64 end = offset + (1 << layout->sb_max_size_bits); -void bch2_recalc_sectors_available(struct bch_fs *); + if (!(offset >= b_end || end <= b_offset)) + return true; + } -void __bch2_disk_reservation_put(struct bch_fs *, struct disk_reservation *); + return false; +} + +static inline const char *bch2_data_type_str(enum bch_data_type type) +{ + return type < BCH_DATA_NR + ? __bch2_data_types[type] + : "(invalid data type)"; +} + +static inline void bch2_prt_data_type(struct printbuf *out, enum bch_data_type type) +{ + if (type < BCH_DATA_NR) + prt_str(out, __bch2_data_types[type]); + else + prt_printf(out, "(invalid data type %u)", type); +} + +/* disk reservations: */ static inline void bch2_disk_reservation_put(struct bch_fs *c, struct disk_reservation *res) { - if (res->sectors) - __bch2_disk_reservation_put(c, res); + if (res->sectors) { + this_cpu_sub(*c->online_reserved, res->sectors); + res->sectors = 0; + } } #define BCH_DISK_RESERVATION_NOFAIL (1 << 0) -#define BCH_DISK_RESERVATION_GC_LOCK_HELD (1 << 1) -#define BCH_DISK_RESERVATION_BTREE_LOCKS_HELD (1 << 2) -int bch2_disk_reservation_add(struct bch_fs *, - struct disk_reservation *, - unsigned, int); +int __bch2_disk_reservation_add(struct bch_fs *, + struct disk_reservation *, + u64, int); + +static inline int bch2_disk_reservation_add(struct bch_fs *c, struct disk_reservation *res, + u64 sectors, int flags) +{ +#ifdef __KERNEL__ + u64 old, new; + + do { + old = this_cpu_read(c->pcpu->sectors_available); + if (sectors > old) + return __bch2_disk_reservation_add(c, res, sectors, flags); + + new = old - sectors; + } while (this_cpu_cmpxchg(c->pcpu->sectors_available, old, new) != old); + + this_cpu_add(*c->online_reserved, sectors); + res->sectors += sectors; + return 0; +#else + return __bch2_disk_reservation_add(c, res, sectors, flags); +#endif +} static inline struct disk_reservation bch2_disk_reservation_init(struct bch_fs *c, unsigned nr_replicas) @@ -258,8 +456,7 @@ bch2_disk_reservation_init(struct bch_fs *c, unsigned nr_replicas) static inline int bch2_disk_reservation_get(struct bch_fs *c, struct disk_reservation *res, - unsigned sectors, - unsigned nr_replicas, + u64 sectors, unsigned nr_replicas, int flags) { *res = bch2_disk_reservation_init(c, nr_replicas); @@ -267,6 +464,13 @@ static inline int bch2_disk_reservation_get(struct bch_fs *c, return bch2_disk_reservation_add(c, res, sectors * nr_replicas, flags); } +#define RESERVE_FACTOR 6 + +static inline u64 avail_factor(u64 r) +{ + return div_u64(r << RESERVE_FACTOR, (1 << RESERVE_FACTOR) + 1); +} + int bch2_dev_buckets_resize(struct bch_fs *, struct bch_dev *, u64); void bch2_dev_buckets_free(struct bch_dev *); int bch2_dev_buckets_alloc(struct bch_fs *, struct bch_dev *);