diff options
Diffstat (limited to 'fs')
147 files changed, 1264 insertions, 788 deletions
diff --git a/fs/affs/file.c b/fs/affs/file.c index a5a861dd5223..7a71018e3f67 100644 --- a/fs/affs/file.c +++ b/fs/affs/file.c @@ -596,7 +596,7 @@ affs_extent_file_ofs(struct inode *inode, u32 newsize) BUG_ON(tmp > bsize); AFFS_DATA_HEAD(bh)->ptype = cpu_to_be32(T_DATA); AFFS_DATA_HEAD(bh)->key = cpu_to_be32(inode->i_ino); - AFFS_DATA_HEAD(bh)->sequence = cpu_to_be32(bidx); + AFFS_DATA_HEAD(bh)->sequence = cpu_to_be32(bidx + 1); AFFS_DATA_HEAD(bh)->size = cpu_to_be32(tmp); affs_fix_checksum(sb, bh); bh->b_state &= ~(1UL << BH_New); @@ -724,7 +724,8 @@ static int affs_write_end_ofs(struct file *file, struct address_space *mapping, tmp = min(bsize - boff, to - from); BUG_ON(boff + tmp > bsize || tmp > bsize); memcpy(AFFS_DATA(bh) + boff, data + from, tmp); - be32_add_cpu(&AFFS_DATA_HEAD(bh)->size, tmp); + AFFS_DATA_HEAD(bh)->size = cpu_to_be32( + max(boff + tmp, be32_to_cpu(AFFS_DATA_HEAD(bh)->size))); affs_fix_checksum(sb, bh); mark_buffer_dirty_inode(bh, inode); written += tmp; @@ -746,7 +747,7 @@ static int affs_write_end_ofs(struct file *file, struct address_space *mapping, if (buffer_new(bh)) { AFFS_DATA_HEAD(bh)->ptype = cpu_to_be32(T_DATA); AFFS_DATA_HEAD(bh)->key = cpu_to_be32(inode->i_ino); - AFFS_DATA_HEAD(bh)->sequence = cpu_to_be32(bidx); + AFFS_DATA_HEAD(bh)->sequence = cpu_to_be32(bidx + 1); AFFS_DATA_HEAD(bh)->size = cpu_to_be32(bsize); AFFS_DATA_HEAD(bh)->next = 0; bh->b_state &= ~(1UL << BH_New); @@ -780,7 +781,7 @@ static int affs_write_end_ofs(struct file *file, struct address_space *mapping, if (buffer_new(bh)) { AFFS_DATA_HEAD(bh)->ptype = cpu_to_be32(T_DATA); AFFS_DATA_HEAD(bh)->key = cpu_to_be32(inode->i_ino); - AFFS_DATA_HEAD(bh)->sequence = cpu_to_be32(bidx); + AFFS_DATA_HEAD(bh)->sequence = cpu_to_be32(bidx + 1); AFFS_DATA_HEAD(bh)->size = cpu_to_be32(tmp); AFFS_DATA_HEAD(bh)->next = 0; bh->b_state &= ~(1UL << BH_New); diff --git a/fs/afs/server.c b/fs/afs/server.c index 038f9d0ae3af..4504e16b458c 100644 --- a/fs/afs/server.c +++ b/fs/afs/server.c @@ -163,6 +163,8 @@ static struct afs_server *afs_install_server(struct afs_cell *cell, rb_insert_color(&server->uuid_rb, &net->fs_servers); hlist_add_head_rcu(&server->proc_link, &net->fs_proc); + afs_get_cell(cell, afs_cell_trace_get_server); + added_dup: write_seqlock(&net->fs_addr_lock); estate = rcu_dereference_protected(server->endpoint_state, @@ -442,6 +444,7 @@ static void afs_server_rcu(struct rcu_head *rcu) atomic_read(&server->active), afs_server_trace_free); afs_put_endpoint_state(rcu_access_pointer(server->endpoint_state), afs_estate_trace_put_server); + afs_put_cell(server->cell, afs_cell_trace_put_server); kfree(server); } diff --git a/fs/afs/server_list.c b/fs/afs/server_list.c index 7e7e567a7f8a..d20cd902ef94 100644 --- a/fs/afs/server_list.c +++ b/fs/afs/server_list.c @@ -97,8 +97,8 @@ struct afs_server_list *afs_alloc_server_list(struct afs_volume *volume, break; if (j < slist->nr_servers) { if (slist->servers[j].server == server) { - afs_put_server(volume->cell->net, server, - afs_server_trace_put_slist_isort); + afs_unuse_server(volume->cell->net, server, + afs_server_trace_put_slist_isort); continue; } diff --git a/fs/bcachefs/Kconfig b/fs/bcachefs/Kconfig index 85eea7a4dea3..fc7efd0a7525 100644 --- a/fs/bcachefs/Kconfig +++ b/fs/bcachefs/Kconfig @@ -61,6 +61,13 @@ config BCACHEFS_DEBUG The resulting code will be significantly slower than normal; you probably shouldn't select this option unless you're a developer. +config BCACHEFS_INJECT_TRANSACTION_RESTARTS + bool "Randomly inject transaction restarts" + depends on BCACHEFS_DEBUG + help + Randomly inject transaction restarts in a few core paths - may have a + significant performance penalty + config BCACHEFS_TESTS bool "bcachefs unit and performance tests" depends on BCACHEFS_FS diff --git a/fs/bcachefs/alloc_background.c b/fs/bcachefs/alloc_background.c index fc2ef33b67b3..3ea809990ef1 100644 --- a/fs/bcachefs/alloc_background.c +++ b/fs/bcachefs/alloc_background.c @@ -1803,7 +1803,6 @@ struct discard_buckets_state { u64 open; u64 need_journal_commit; u64 discarded; - u64 need_journal_commit_this_dev; }; static int bch2_discard_one_bucket(struct btree_trans *trans, @@ -1827,11 +1826,11 @@ static int bch2_discard_one_bucket(struct btree_trans *trans, goto out; } - if (bch2_bucket_needs_journal_commit(&c->buckets_waiting_for_journal, - c->journal.flushed_seq_ondisk, - pos.inode, pos.offset)) { - s->need_journal_commit++; - s->need_journal_commit_this_dev++; + u64 seq_ready = bch2_bucket_journal_seq_ready(&c->buckets_waiting_for_journal, + pos.inode, pos.offset); + if (seq_ready > c->journal.flushed_seq_ondisk) { + if (seq_ready > c->journal.flushing_seq) + s->need_journal_commit++; goto out; } @@ -1865,23 +1864,24 @@ static int bch2_discard_one_bucket(struct btree_trans *trans, discard_locked = true; } - if (!bkey_eq(*discard_pos_done, iter.pos) && - ca->mi.discard && !c->opts.nochanges) { - /* - * This works without any other locks because this is the only - * thread that removes items from the need_discard tree - */ - bch2_trans_unlock_long(trans); - blkdev_issue_discard(ca->disk_sb.bdev, - k.k->p.offset * ca->mi.bucket_size, - ca->mi.bucket_size, - GFP_KERNEL); - *discard_pos_done = iter.pos; + if (!bkey_eq(*discard_pos_done, iter.pos)) { s->discarded++; + *discard_pos_done = iter.pos; - ret = bch2_trans_relock_notrace(trans); - if (ret) - goto out; + if (ca->mi.discard && !c->opts.nochanges) { + /* + * This works without any other locks because this is the only + * thread that removes items from the need_discard tree + */ + bch2_trans_unlock_long(trans); + blkdev_issue_discard(ca->disk_sb.bdev, + k.k->p.offset * ca->mi.bucket_size, + ca->mi.bucket_size, + GFP_KERNEL); + ret = bch2_trans_relock_notrace(trans); + if (ret) + goto out; + } } SET_BCH_ALLOC_V4_NEED_DISCARD(&a->v, false); @@ -1929,6 +1929,9 @@ static void bch2_do_discards_work(struct work_struct *work) POS(ca->dev_idx, U64_MAX), 0, k, bch2_discard_one_bucket(trans, ca, &iter, &discard_pos_done, &s, false))); + if (s.need_journal_commit > dev_buckets_available(ca, BCH_WATERMARK_normal)) + bch2_journal_flush_async(&c->journal, NULL); + trace_discard_buckets(c, s.seen, s.open, s.need_journal_commit, s.discarded, bch2_err_str(ret)); @@ -2024,7 +2027,7 @@ static void bch2_do_discards_fast_work(struct work_struct *work) break; } - trace_discard_buckets(c, s.seen, s.open, s.need_journal_commit, s.discarded, bch2_err_str(ret)); + trace_discard_buckets_fast(c, s.seen, s.open, s.need_journal_commit, s.discarded, bch2_err_str(ret)); bch2_trans_put(trans); percpu_ref_put(&ca->io_ref); diff --git a/fs/bcachefs/alloc_foreground.c b/fs/bcachefs/alloc_foreground.c index 6df41c331a52..5a781fb4c794 100644 --- a/fs/bcachefs/alloc_foreground.c +++ b/fs/bcachefs/alloc_foreground.c @@ -205,8 +205,12 @@ static inline bool may_alloc_bucket(struct bch_fs *c, return false; } - if (bch2_bucket_needs_journal_commit(&c->buckets_waiting_for_journal, - c->journal.flushed_seq_ondisk, bucket.inode, bucket.offset)) { + u64 journal_seq_ready = + bch2_bucket_journal_seq_ready(&c->buckets_waiting_for_journal, + bucket.inode, bucket.offset); + if (journal_seq_ready > c->journal.flushed_seq_ondisk) { + if (journal_seq_ready > c->journal.flushing_seq) + s->need_journal_commit++; s->skipped_need_journal_commit++; return false; } @@ -570,7 +574,7 @@ alloc: ? bch2_bucket_alloc_freelist(trans, ca, watermark, &s, cl) : bch2_bucket_alloc_early(trans, ca, watermark, &s, cl); - if (s.skipped_need_journal_commit * 2 > avail) + if (s.need_journal_commit * 2 > avail) bch2_journal_flush_async(&c->journal, NULL); if (!ob && s.btree_bitmap != BTREE_BITMAP_ANY) { diff --git a/fs/bcachefs/alloc_types.h b/fs/bcachefs/alloc_types.h index 9bbb28e90b93..4aa8ee026cb8 100644 --- a/fs/bcachefs/alloc_types.h +++ b/fs/bcachefs/alloc_types.h @@ -18,6 +18,7 @@ struct bucket_alloc_state { u64 buckets_seen; u64 skipped_open; u64 skipped_need_journal_commit; + u64 need_journal_commit; u64 skipped_nocow; u64 skipped_nouse; u64 skipped_mi_btree_bitmap; diff --git a/fs/bcachefs/btree_cache.c b/fs/bcachefs/btree_cache.c index ca755e8d1a37..1ec1f90e0eb3 100644 --- a/fs/bcachefs/btree_cache.c +++ b/fs/bcachefs/btree_cache.c @@ -203,7 +203,7 @@ struct btree *__bch2_btree_node_mem_alloc(struct bch_fs *c) return NULL; } - bch2_btree_lock_init(&b->c, 0); + bch2_btree_lock_init(&b->c, 0, GFP_KERNEL); __bch2_btree_node_to_freelist(bc, b); return b; @@ -795,17 +795,18 @@ struct btree *bch2_btree_node_mem_alloc(struct btree_trans *trans, bool pcpu_rea } b = __btree_node_mem_alloc(c, GFP_NOWAIT|__GFP_NOWARN); - if (!b) { + if (b) { + bch2_btree_lock_init(&b->c, pcpu_read_locks ? SIX_LOCK_INIT_PCPU : 0, GFP_NOWAIT); + } else { mutex_unlock(&bc->lock); bch2_trans_unlock(trans); b = __btree_node_mem_alloc(c, GFP_KERNEL); if (!b) goto err; + bch2_btree_lock_init(&b->c, pcpu_read_locks ? SIX_LOCK_INIT_PCPU : 0, GFP_KERNEL); mutex_lock(&bc->lock); } - bch2_btree_lock_init(&b->c, pcpu_read_locks ? SIX_LOCK_INIT_PCPU : 0); - BUG_ON(!six_trylock_intent(&b->c.lock)); BUG_ON(!six_trylock_write(&b->c.lock)); diff --git a/fs/bcachefs/btree_io.c b/fs/bcachefs/btree_io.c index e371e60e3133..dece27d9db04 100644 --- a/fs/bcachefs/btree_io.c +++ b/fs/bcachefs/btree_io.c @@ -996,7 +996,7 @@ drop_this_key: } got_good_key: le16_add_cpu(&i->u64s, -next_good_key); - memmove_u64s_down(k, bkey_p_next(k), (u64 *) vstruct_end(i) - (u64 *) k); + memmove_u64s_down(k, (u64 *) k + next_good_key, (u64 *) vstruct_end(i) - (u64 *) k); set_btree_node_need_rewrite(b); } fsck_err: diff --git a/fs/bcachefs/btree_iter.c b/fs/bcachefs/btree_iter.c index 5988219c6908..e32fce4fd258 100644 --- a/fs/bcachefs/btree_iter.c +++ b/fs/bcachefs/btree_iter.c @@ -2357,6 +2357,12 @@ struct bkey_s_c bch2_btree_iter_peek_max(struct btree_iter *iter, struct bpos en bch2_btree_iter_verify_entry_exit(iter); EBUG_ON((iter->flags & BTREE_ITER_filter_snapshots) && bkey_eq(end, POS_MAX)); + ret = trans_maybe_inject_restart(trans, _RET_IP_); + if (unlikely(ret)) { + k = bkey_s_c_err(ret); + goto out_no_locked; + } + if (iter->update_path) { bch2_path_put_nokeep(trans, iter->update_path, iter->flags & BTREE_ITER_intent); @@ -2622,6 +2628,12 @@ struct bkey_s_c bch2_btree_iter_peek_prev_min(struct btree_iter *iter, struct bp bch2_btree_iter_verify_entry_exit(iter); EBUG_ON((iter->flags & BTREE_ITER_filter_snapshots) && bpos_eq(end, POS_MIN)); + int ret = trans_maybe_inject_restart(trans, _RET_IP_); + if (unlikely(ret)) { + k = bkey_s_c_err(ret); + goto out_no_locked; + } + while (1) { k = __bch2_btree_iter_peek_prev(iter, search_key); if (unlikely(!k.k)) @@ -2749,6 +2761,12 @@ struct bkey_s_c bch2_btree_iter_peek_slot(struct btree_iter *iter) bch2_btree_iter_verify_entry_exit(iter); EBUG_ON(btree_iter_path(trans, iter)->level && (iter->flags & BTREE_ITER_with_key_cache)); + ret = trans_maybe_inject_restart(trans, _RET_IP_); + if (unlikely(ret)) { + k = bkey_s_c_err(ret); + goto out_no_locked; + } + /* extents can't span inode numbers: */ if ((iter->flags & BTREE_ITER_is_extents) && unlikely(iter->pos.offset == KEY_OFFSET_MAX)) { @@ -3106,6 +3124,10 @@ void *__bch2_trans_kmalloc(struct btree_trans *trans, size_t size) WARN_ON_ONCE(new_bytes > BTREE_TRANS_MEM_MAX); + ret = trans_maybe_inject_restart(trans, _RET_IP_); + if (ret) + return ERR_PTR(ret); + struct btree_transaction_stats *s = btree_trans_stats(trans); s->max_mem = max(s->max_mem, new_bytes); @@ -3163,7 +3185,8 @@ out_new_mem: if (old_bytes) { trace_and_count(c, trans_restart_mem_realloced, trans, _RET_IP_, new_bytes); - return ERR_PTR(btree_trans_restart(trans, BCH_ERR_transaction_restart_mem_realloced)); + return ERR_PTR(btree_trans_restart_ip(trans, + BCH_ERR_transaction_restart_mem_realloced, _RET_IP_)); } out_change_top: p = trans->mem + trans->mem_top; @@ -3271,6 +3294,14 @@ u32 bch2_trans_begin(struct btree_trans *trans) trans->last_begin_ip = _RET_IP_; +#ifdef CONFIG_BCACHEFS_INJECT_TRANSACTION_RESTARTS + if (trans->restarted) { + trans->restart_count_this_trans++; + } else { + trans->restart_count_this_trans = 0; + } +#endif + trans_set_locked(trans, false); if (trans->restarted) { diff --git a/fs/bcachefs/btree_iter.h b/fs/bcachefs/btree_iter.h index b9538e6e6d65..b96157f3dc9c 100644 --- a/fs/bcachefs/btree_iter.h +++ b/fs/bcachefs/btree_iter.h @@ -355,6 +355,18 @@ static int btree_trans_restart(struct btree_trans *trans, int err) return btree_trans_restart_ip(trans, err, _THIS_IP_); } +static inline int trans_maybe_inject_restart(struct btree_trans *trans, unsigned long ip) +{ +#ifdef CONFIG_BCACHEFS_INJECT_TRANSACTION_RESTARTS + if (!(ktime_get_ns() & ~(~0ULL << min(63, (10 + trans->restart_count_this_trans))))) { + trace_and_count(trans->c, trans_restart_injected, trans, ip); + return btree_trans_restart_ip(trans, + BCH_ERR_transaction_restart_fault_inject, ip); + } +#endif + return 0; +} + bool bch2_btree_node_upgrade(struct btree_trans *, struct btree_path *, unsigned); @@ -739,7 +751,7 @@ transaction_restart: \ if (!_ret2) \ bch2_trans_verify_not_restarted(_trans, _restart_count);\ \ - _ret2 ?: trans_was_restarted(_trans, _restart_count); \ + _ret2 ?: trans_was_restarted(_trans, _orig_restart_count); \ }) #define for_each_btree_key_max_continue(_trans, _iter, \ diff --git a/fs/bcachefs/btree_key_cache.c b/fs/bcachefs/btree_key_cache.c index c378b97ebeca..edce59433375 100644 --- a/fs/bcachefs/btree_key_cache.c +++ b/fs/bcachefs/btree_key_cache.c @@ -156,7 +156,7 @@ bkey_cached_alloc(struct btree_trans *trans, struct btree_path *path, unsigned k } if (ck) { - bch2_btree_lock_init(&ck->c, pcpu_readers ? SIX_LOCK_INIT_PCPU : 0); + bch2_btree_lock_init(&ck->c, pcpu_readers ? SIX_LOCK_INIT_PCPU : 0, GFP_KERNEL); ck->c.cached = true; goto lock; } @@ -748,7 +748,6 @@ void bch2_fs_btree_key_cache_exit(struct btree_key_cache *bc) rcu_read_unlock(); mutex_lock(&bc->table.mutex); mutex_unlock(&bc->table.mutex); - rcu_read_lock(); continue; } for (i = 0; i < tbl->size; i++) diff --git a/fs/bcachefs/btree_locking.c b/fs/bcachefs/btree_locking.c index 10b805a60f52..caef65adeae4 100644 --- a/fs/bcachefs/btree_locking.c +++ b/fs/bcachefs/btree_locking.c @@ -7,9 +7,10 @@ static struct lock_class_key bch2_btree_node_lock_key; void bch2_btree_lock_init(struct btree_bkey_cached_common *b, - enum six_lock_init_flags flags) + enum six_lock_init_flags flags, + gfp_t gfp) { - __six_lock_init(&b->lock, "b->c.lock", &bch2_btree_node_lock_key, flags); + __six_lock_init(&b->lock, "b->c.lock", &bch2_btree_node_lock_key, flags, gfp); lockdep_set_notrack_class(&b->lock); } diff --git a/fs/bcachefs/btree_locking.h b/fs/bcachefs/btree_locking.h index b54ef48eb8cc..b33ab7af8440 100644 --- a/fs/bcachefs/btree_locking.h +++ b/fs/bcachefs/btree_locking.h @@ -13,7 +13,7 @@ #include "btree_iter.h" #include "six.h" -void bch2_btree_lock_init(struct btree_bkey_cached_common *, enum six_lock_init_flags); +void bch2_btree_lock_init(struct btree_bkey_cached_common *, enum six_lock_init_flags, gfp_t gfp); void bch2_trans_unlock_noassert(struct btree_trans *); void bch2_trans_unlock_write(struct btree_trans *); diff --git a/fs/bcachefs/btree_trans_commit.c b/fs/bcachefs/btree_trans_commit.c index 2760dd9569ed..c4f524b2ca9a 100644 --- a/fs/bcachefs/btree_trans_commit.c +++ b/fs/bcachefs/btree_trans_commit.c @@ -999,6 +999,10 @@ int __bch2_trans_commit(struct btree_trans *trans, unsigned flags) bch2_trans_verify_not_unlocked_or_in_restart(trans); + ret = trans_maybe_inject_restart(trans, _RET_IP_); + if (unlikely(ret)) + goto out_reset; + if (!trans->nr_updates && !trans->journal_entries_u64s) goto out_reset; diff --git a/fs/bcachefs/btree_types.h b/fs/bcachefs/btree_types.h index a6f251eb4164..a09cbe9cd94f 100644 --- a/fs/bcachefs/btree_types.h +++ b/fs/bcachefs/btree_types.h @@ -509,6 +509,9 @@ struct btree_trans { bool notrace_relock_fail:1; enum bch_errcode restarted:16; u32 restart_count; +#ifdef CONFIG_BCACHEFS_INJECT_TRANSACTION_RESTARTS + u32 restart_count_this_trans; +#endif u64 last_begin_time; unsigned long last_begin_ip; diff --git a/fs/bcachefs/btree_update_interior.c b/fs/bcachefs/btree_update_interior.c index f4aeadbe53c1..e4e7c804625e 100644 --- a/fs/bcachefs/btree_update_interior.c +++ b/fs/bcachefs/btree_update_interior.c @@ -681,9 +681,11 @@ static void btree_update_nodes_written(struct btree_update *as) b = as->old_nodes[i]; + bch2_trans_begin(trans); btree_node_lock_nopath_nofail(trans, &b->c, SIX_LOCK_read); seq = b->data ? b->data->keys.seq : 0; six_unlock_read(&b->c.lock); + bch2_trans_unlock_long(trans); if (seq == as->old_nodes_seq[i]) wait_on_bit_io(&b->flags, BTREE_NODE_write_in_flight_inner, diff --git a/fs/bcachefs/btree_update_interior.h b/fs/bcachefs/btree_update_interior.h index 7930ffea3075..26d646e1275c 100644 --- a/fs/bcachefs/btree_update_interior.h +++ b/fs/bcachefs/btree_update_interior.h @@ -278,12 +278,12 @@ static inline struct btree_node_entry *want_new_bset(struct bch_fs *c, struct bt { struct bset_tree *t = bset_tree_last(b); struct btree_node_entry *bne = max(write_block(b), - (void *) btree_bkey_last(b, bset_tree_last(b))); + (void *) btree_bkey_last(b, t)); ssize_t remaining_space = __bch2_btree_u64s_remaining(b, bne->keys.start); if (unlikely(bset_written(b, bset(b, t)))) { - if (remaining_space > (ssize_t) (block_bytes(c) >> 3)) + if (b->written + block_sectors(c) <= btree_sectors(c)) return bne; } else { if (unlikely(bset_u64s(t) * sizeof(u64) > btree_write_set_buffer(b)) && diff --git a/fs/bcachefs/buckets_waiting_for_journal.c b/fs/bcachefs/buckets_waiting_for_journal.c index f9fb150eda70..c8a488e6b7b8 100644 --- a/fs/bcachefs/buckets_waiting_for_journal.c +++ b/fs/bcachefs/buckets_waiting_for_journal.c @@ -22,23 +22,21 @@ static void bucket_table_init(struct buckets_waiting_for_journal_table *t, size_ memset(t->d, 0, sizeof(t->d[0]) << t->bits); } -bool bch2_bucket_needs_journal_commit(struct buckets_waiting_for_journal *b, - u64 flushed_seq, - unsigned dev, u64 bucket) +u64 bch2_bucket_journal_seq_ready(struct buckets_waiting_for_journal *b, + unsigned dev, u64 bucket) { struct buckets_waiting_for_journal_table *t; u64 dev_bucket = (u64) dev << 56 | bucket; - bool ret = false; - unsigned i; + u64 ret = 0; mutex_lock(&b->lock); t = b->t; - for (i = 0; i < ARRAY_SIZE(t->hash_seeds); i++) { + for (unsigned i = 0; i < ARRAY_SIZE(t->hash_seeds); i++) { struct bucket_hashed *h = bucket_hash(t, i, dev_bucket); if (h->dev_bucket == dev_bucket) { - ret = h->journal_seq > flushed_seq; + ret = h->journal_seq; break; } } diff --git a/fs/bcachefs/buckets_waiting_for_journal.h b/fs/bcachefs/buckets_waiting_for_journal.h index d2ae19cbe18c..365619ca44c8 100644 --- a/fs/bcachefs/buckets_waiting_for_journal.h +++ b/fs/bcachefs/buckets_waiting_for_journal.h @@ -4,8 +4,8 @@ #include "buckets_waiting_for_journal_types.h" -bool bch2_bucket_needs_journal_commit(struct buckets_waiting_for_journal *, - u64, unsigned, u64); +u64 bch2_bucket_journal_seq_ready(struct buckets_waiting_for_journal *, + unsigned, u64); int bch2_set_bucket_needs_journal_commit(struct buckets_waiting_for_journal *, u64, unsigned, u64, u64); diff --git a/fs/bcachefs/data_update.c b/fs/bcachefs/data_update.c index 337494facac6..642fbc60ecab 100644 --- a/fs/bcachefs/data_update.c +++ b/fs/bcachefs/data_update.c @@ -340,6 +340,7 @@ restart_drop_extra_replicas: struct printbuf buf = PRINTBUF; prt_str(&buf, "about to insert invalid key in data update path"); + prt_printf(&buf, "\nop.nonce: %u", m->op.nonce); prt_str(&buf, "\nold: "); bch2_bkey_val_to_text(&buf, c, old); prt_str(&buf, "\nk: "); diff --git a/fs/bcachefs/dirent.h b/fs/bcachefs/dirent.h index a633f83c1ac7..362b3b2f2f2e 100644 --- a/fs/bcachefs/dirent.h +++ b/fs/bcachefs/dirent.h @@ -31,11 +31,6 @@ static inline unsigned dirent_val_u64s(unsigned len) sizeof(u64)); } -static inline unsigned int dirent_occupied_size(const struct qstr *name) -{ - return (BKEY_U64s + dirent_val_u64s(name->len)) * sizeof(u64); -} - int bch2_dirent_read_target(struct btree_trans *, subvol_inum, struct bkey_s_c_dirent, subvol_inum *); diff --git a/fs/bcachefs/disk_accounting.h b/fs/bcachefs/disk_accounting.h index 5360cbb3ec29..f4372cafea2e 100644 --- a/fs/bcachefs/disk_accounting.h +++ b/fs/bcachefs/disk_accounting.h @@ -210,11 +210,13 @@ static inline void bch2_accounting_mem_read_counters(struct bch_accounting_mem * static inline void bch2_accounting_mem_read(struct bch_fs *c, struct bpos p, u64 *v, unsigned nr) { + percpu_down_read(&c->mark_lock); struct bch_accounting_mem *acc = &c->accounting; unsigned idx = eytzinger0_find(acc->k.data, acc->k.nr, sizeof(acc->k.data[0]), accounting_pos_cmp, &p); bch2_accounting_mem_read_counters(acc, idx, v, nr, false); + percpu_up_read(&c->mark_lock); } static inline struct bversion journal_pos_to_bversion(struct journal_res *res, unsigned offset) diff --git a/fs/bcachefs/extents.h b/fs/bcachefs/extents.h index 620b284aa34f..204d765dd74c 100644 --- a/fs/bcachefs/extents.h +++ b/fs/bcachefs/extents.h @@ -704,7 +704,7 @@ static inline bool bch2_extent_ptr_eq(struct bch_extent_ptr ptr1, ptr1.unwritten == ptr2.unwritten && ptr1.offset == ptr2.offset && ptr1.dev == ptr2.dev && - ptr1.dev == ptr2.dev); + ptr1.gen == ptr2.gen); } void bch2_ptr_swab(struct bkey_s); diff --git a/fs/bcachefs/fs-common.c b/fs/bcachefs/fs-common.c index d70d9f634cea..2c3d46ac70c6 100644 --- a/fs/bcachefs/fs-common.c +++ b/fs/bcachefs/fs-common.c @@ -152,7 +152,6 @@ int bch2_create_trans(struct btree_trans *trans, if (is_subdir_for_nlink(new_inode)) dir_u->bi_nlink++; dir_u->bi_mtime = dir_u->bi_ctime = now; - dir_u->bi_size += dirent_occupied_size(name); ret = bch2_inode_write(trans, &dir_iter, dir_u); if (ret) @@ -221,7 +220,6 @@ int bch2_link_trans(struct btree_trans *trans, } dir_u->bi_mtime = dir_u->bi_ctime = now; - dir_u->bi_size += dirent_occupied_size(name); dir_hash = bch2_hash_info_init(c, dir_u); @@ -324,7 +322,6 @@ int bch2_unlink_trans(struct btree_trans *trans, dir_u->bi_mtime = dir_u->bi_ctime = inode_u->bi_ctime = now; dir_u->bi_nlink -= is_subdir_for_nlink(inode_u); - dir_u->bi_size -= dirent_occupied_size(name); ret = bch2_hash_delete_at(trans, bch2_dirent_hash_desc, &dir_hash, &dirent_iter, @@ -463,14 +460,6 @@ int bch2_rename_trans(struct btree_trans *trans, goto err; } - if (mode == BCH_RENAME) { - src_dir_u->bi_size -= dirent_occupied_size(src_name); - dst_dir_u->bi_size += dirent_occupied_size(dst_name); - } - - if (mode == BCH_RENAME_OVERWRITE) - src_dir_u->bi_size -= dirent_occupied_size(src_name); - if (src_inode_u->bi_parent_subvol) src_inode_u->bi_parent_subvol = dst_dir.subvol; diff --git a/fs/bcachefs/fs-io.c b/fs/bcachefs/fs-io.c index 94bf34b9b65f..717e7b94c66f 100644 --- a/fs/bcachefs/fs-io.c +++ b/fs/bcachefs/fs-io.c @@ -466,6 +466,7 @@ int bchfs_truncate(struct mnt_idmap *idmap, ret = bch2_truncate_folio(inode, iattr->ia_size); if (unlikely(ret < 0)) goto err; + ret = 0; truncate_setsize(&inode->v, iattr->ia_size); diff --git a/fs/bcachefs/fsck.c b/fs/bcachefs/fsck.c index 53a421ff136d..0e85131d0af8 100644 --- a/fs/bcachefs/fsck.c +++ b/fs/bcachefs/fsck.c @@ -823,6 +823,7 @@ struct inode_walker_entry { struct bch_inode_unpacked inode; u32 snapshot; u64 count; + u64 i_size; }; struct inode_walker { @@ -910,8 +911,9 @@ found: if (k.k->p.snapshot != i->snapshot && !is_whiteout) { struct inode_walker_entry new = *i; - new.snapshot = k.k->p.snapshot; - new.count = 0; + new.snapshot = k.k->p.snapshot; + new.count = 0; + new.i_size = 0; struct printbuf buf = PRINTBUF; bch2_bkey_val_to_text(&buf, c, k); @@ -1116,37 +1118,6 @@ err: return ret; } -static int check_directory_size(struct btree_trans *trans, - struct bch_inode_unpacked *inode_u, - struct bkey_s_c inode_k, bool *write_inode) -{ - struct btree_iter iter; - struct bkey_s_c k; - u64 new_size = 0; - int ret; - - for_each_btree_key_max_norestart(trans, iter, BTREE_ID_dirents, - SPOS(inode_k.k->p.offset, 0, inode_k.k->p.snapshot), - POS(inode_k.k->p.offset, U64_MAX), - 0, k, ret) { - if (k.k->type != KEY_TYPE_dirent) - continue; - - struct bkey_s_c_dirent dirent = bkey_s_c_to_dirent(k); - struct qstr name = bch2_dirent_get_name(dirent); - - new_size += dirent_occupied_size(&name); - } - bch2_trans_iter_exit(trans, &iter); - - if (!ret && inode_u->bi_size != new_size) { - inode_u->bi_size = new_size; - *write_inode = true; - } - - return ret; -} - static int check_inode(struct btree_trans *trans, struct btree_iter *iter, struct bkey_s_c k, @@ -1335,16 +1306,6 @@ static int check_inode(struct btree_trans *trans, u.bi_journal_seq = journal_cur_seq(&c->journal); do_update = true; } - - if (S_ISDIR(u.bi_mode)) { - ret = check_directory_size(trans, &u, k, &do_update); - - fsck_err_on(ret, - trans, directory_size_mismatch, - "directory inode %llu:%u with the mismatch directory size", - u.bi_inum, k.k->p.snapshot); - ret = 0; - } do_update: if (do_update) { ret = __bch2_fsck_write_inode(trans, &u); @@ -2017,7 +1978,7 @@ fsck_err: return ret; } -static int check_subdir_count(struct btree_trans *trans, struct inode_walker *w) +static int check_subdir_dirents_count(struct btree_trans *trans, struct inode_walker *w) { u32 restart_count = trans->restart_count; return check_subdir_count_notnested(trans, w) ?: @@ -2367,7 +2328,7 @@ static int check_dirent(struct btree_trans *trans, struct btree_iter *iter, goto out; if (dir->last_pos.inode != k.k->p.inode && dir->have_inodes) { - ret = check_subdir_count(trans, dir); + ret = check_subdir_dirents_count(trans, dir); if (ret) goto err; } @@ -2457,9 +2418,11 @@ static int check_dirent(struct btree_trans *trans, struct btree_iter *iter, if (ret) goto err; - if (d.v->d_type == DT_DIR) - for_each_visible_inode(c, s, dir, d.k->p.snapshot, i) + for_each_visible_inode(c, s, dir, d.k->p.snapshot, i) { + if (d.v->d_type == DT_DIR) i->count++; + i->i_size += bkey_bytes(d.k); + } out: err: fsck_err: diff --git a/fs/bcachefs/inode.h b/fs/bcachefs/inode.h index d2e134528f0e..428b9be6af34 100644 --- a/fs/bcachefs/inode.h +++ b/fs/bcachefs/inode.h @@ -285,12 +285,14 @@ void bch2_inode_opts_get(struct bch_io_opts *, struct bch_fs *, struct bch_inode_unpacked *); int bch2_inum_opts_get(struct btree_trans*, subvol_inum, struct bch_io_opts *); +#include "rebalance.h" + static inline struct bch_extent_rebalance bch2_inode_rebalance_opts_get(struct bch_fs *c, struct bch_inode_unpacked *inode) { struct bch_io_opts io_opts; bch2_inode_opts_get(&io_opts, c, inode); - return io_opts_to_rebalance_opts(&io_opts); + return io_opts_to_rebalance_opts(c, &io_opts); } int bch2_inode_rm_snapshot(struct btree_trans *, u64, u32); diff --git a/fs/bcachefs/io_write.c b/fs/bcachefs/io_write.c index dd508d93e9fc..03892388832b 100644 --- a/fs/bcachefs/io_write.c +++ b/fs/bcachefs/io_write.c @@ -411,6 +411,16 @@ void bch2_write_op_error(struct printbuf *out, struct bch_write_op *op) __bch2_write_op_error(out, op, op->pos.offset); } +static void bch2_write_op_error_trans(struct btree_trans *trans, struct printbuf *out, + struct bch_write_op *op, u64 offset) +{ + bch2_inum_offset_err_msg_trans(trans, out, + (subvol_inum) { op->subvol, op->pos.inode, }, + offset << 9); + prt_printf(out, "write error%s: ", + op->flags & BCH_WRITE_MOVE ? "(internal move)" : ""); +} + void bch2_submit_wbio_replicas(struct bch_write_bio *wbio, struct bch_fs *c, enum bch_data_type type, const struct bkey_i *k, @@ -1193,7 +1203,7 @@ static void bch2_nocow_write_convert_unwritten(struct bch_write_op *op) struct bkey_i *insert = bch2_keylist_front(&op->insert_keys); struct printbuf buf = PRINTBUF; - __bch2_write_op_error(&buf, op, bkey_start_offset(&insert->k)); + bch2_write_op_error_trans(trans, &buf, op, bkey_start_offset(&insert->k)); prt_printf(&buf, "btree update error: %s", bch2_err_str(ret)); bch_err_ratelimited(c, "%s", buf.buf); printbuf_exit(&buf); diff --git a/fs/bcachefs/journal.c b/fs/bcachefs/journal.c index cb2c3722f674..05b1250619ec 100644 --- a/fs/bcachefs/journal.c +++ b/fs/bcachefs/journal.c @@ -319,6 +319,16 @@ void bch2_journal_halt(struct journal *j) spin_unlock(&j->lock); } +void bch2_journal_halt_locked(struct journal *j) +{ + lockdep_assert_held(&j->lock); + + __journal_entry_close(j, JOURNAL_ENTRY_ERROR_VAL, true); + if (!j->err_seq) + j->err_seq = journal_cur_seq(j); + journal_wake(j); +} + static bool journal_entry_want_write(struct journal *j) { bool ret = !journal_entry_is_open(j) || @@ -381,9 +391,12 @@ static int journal_entry_open(struct journal *j) if (nr_unwritten_journal_entries(j) == ARRAY_SIZE(j->buf)) return JOURNAL_ERR_max_in_flight; - if (bch2_fs_fatal_err_on(journal_cur_seq(j) >= JOURNAL_SEQ_MAX, - c, "cannot start: journal seq overflow")) + if (journal_cur_seq(j) >= JOURNAL_SEQ_MAX) { + bch_err(c, "cannot start: journal seq overflow"); + if (bch2_fs_emergency_read_only_locked(c)) + bch_err(c, "fatal error - emergency read only"); return JOURNAL_ERR_insufficient_devices; /* -EROFS */ + } BUG_ON(!j->cur_entry_sectors); @@ -783,6 +796,7 @@ recheck_need_open: } buf->must_flush = true; + j->flushing_seq = max(j->flushing_seq, seq); if (parent && !closure_wait(&buf->wait, parent)) BUG(); @@ -1007,8 +1021,8 @@ struct journal_buf *bch2_next_write_buffer_flush_journal_buf(struct journal *j, /* allocate journal on a device: */ -static int __bch2_set_nr_journal_buckets(struct bch_dev *ca, unsigned nr, - bool new_fs, struct closure *cl) +static int bch2_set_nr_journal_buckets_iter(struct bch_dev *ca, unsigned nr, + bool new_fs, struct closure *cl) { struct bch_fs *c = ca->fs; struct journal_device *ja = &ca->journal; @@ -1136,26 +1150,20 @@ err_free: return ret; } -/* - * Allocate more journal space at runtime - not currently making use if it, but - * the code works: - */ -int bch2_set_nr_journal_buckets(struct bch_fs *c, struct bch_dev *ca, - unsigned nr) +static int bch2_set_nr_journal_buckets_loop(struct bch_fs *c, struct bch_dev *ca, + unsigned nr, bool new_fs) { struct journal_device *ja = &ca->journal; - struct closure cl; int ret = 0; + struct closure cl; closure_init_stack(&cl); - down_write(&c->state_lock); - /* don't handle reducing nr of buckets yet: */ if (nr < ja->nr) - goto unlock; + return 0; - while (ja->nr < nr) { + while (!ret && ja->nr < nr) { struct disk_reservation disk_res = { 0, 0, 0 }; /* @@ -1168,25 +1176,38 @@ int bch2_set_nr_journal_buckets(struct bch_fs *c, struct bch_dev *ca, * filesystem-wide allocation will succeed, this is a device * specific allocation - we can hang here: */ + if (!new_fs) { + ret = bch2_disk_reservation_get(c, &disk_res, + bucket_to_sector(ca, nr - ja->nr), 1, 0); + if (ret) + break; + } - ret = bch2_disk_reservation_get(c, &disk_res, - bucket_to_sector(ca, nr - ja->nr), 1, 0); - if (ret) - break; + ret = bch2_set_nr_journal_buckets_iter(ca, nr, new_fs, &cl); - ret = __bch2_set_nr_journal_buckets(ca, nr, false, &cl); + if (ret == -BCH_ERR_bucket_alloc_blocked || + ret == -BCH_ERR_open_buckets_empty) + ret = 0; /* wait and retry */ bch2_disk_reservation_put(c, &disk_res); - closure_sync(&cl); - - if (ret && ret != -BCH_ERR_bucket_alloc_blocked) - break; } - bch_err_fn(c, ret); -unlock: + return ret; +} + +/* + * Allocate more journal space at runtime - not currently making use if it, but + * the code works: + */ +int bch2_set_nr_journal_buckets(struct bch_fs *c, struct bch_dev *ca, + unsigned nr) +{ + down_write(&c->state_lock); + int ret = bch2_set_nr_journal_buckets_loop(c, ca, nr, false); up_write(&c->state_lock); + + bch_err_fn(c, ret); return ret; } @@ -1212,7 +1233,7 @@ int bch2_dev_journal_alloc(struct bch_dev *ca, bool new_fs) min(1 << 13, (1 << 24) / ca->mi.bucket_size)); - ret = __bch2_set_nr_journal_buckets(ca, nr, new_fs, NULL); + ret = bch2_set_nr_journal_buckets_loop(ca->fs, ca, nr, new_fs); err: bch_err_fn(ca, ret); return ret; diff --git a/fs/bcachefs/journal.h b/fs/bcachefs/journal.h index dccddd5420ad..107f7f901cd9 100644 --- a/fs/bcachefs/journal.h +++ b/fs/bcachefs/journal.h @@ -409,6 +409,7 @@ bool bch2_journal_noflush_seq(struct journal *, u64, u64); int bch2_journal_meta(struct journal *); void bch2_journal_halt(struct journal *); +void bch2_journal_halt_locked(struct journal *); static inline int bch2_journal_error(struct journal *j) { diff --git a/fs/bcachefs/journal_reclaim.c b/fs/bcachefs/journal_reclaim.c index 6a9cefb635d6..d373cd181a7f 100644 --- a/fs/bcachefs/journal_reclaim.c +++ b/fs/bcachefs/journal_reclaim.c @@ -384,12 +384,16 @@ void bch2_journal_pin_drop(struct journal *j, spin_unlock(&j->lock); } -static enum journal_pin_type journal_pin_type(journal_pin_flush_fn fn) +static enum journal_pin_type journal_pin_type(struct journal_entry_pin *pin, + journal_pin_flush_fn fn) { if (fn == bch2_btree_node_flush0 || - fn == bch2_btree_node_flush1) - return JOURNAL_PIN_TYPE_btree; - else if (fn == bch2_btree_key_cache_journal_flush) + fn == bch2_btree_node_flush1) { + unsigned idx = fn == bch2_btree_node_flush1; + struct btree *b = container_of(pin, struct btree, writes[idx].journal); + + return JOURNAL_PIN_TYPE_btree0 - b->c.level; + } else if (fn == bch2_btree_key_cache_journal_flush) return JOURNAL_PIN_TYPE_key_cache; else return JOURNAL_PIN_TYPE_other; @@ -441,7 +445,7 @@ void bch2_journal_pin_copy(struct journal *j, bool reclaim = __journal_pin_drop(j, dst); - bch2_journal_pin_set_locked(j, seq, dst, flush_fn, journal_pin_type(flush_fn)); + bch2_journal_pin_set_locked(j, seq, dst, flush_fn, journal_pin_type(dst, flush_fn)); if (reclaim) bch2_journal_reclaim_fast(j); @@ -465,7 +469,7 @@ void bch2_journal_pin_set(struct journal *j, u64 seq, bool reclaim = __journal_pin_drop(j, pin); - bch2_journal_pin_set_locked(j, seq, pin, flush_fn, journal_pin_type(flush_fn)); + bch2_journal_pin_set_locked(j, seq, pin, flush_fn, journal_pin_type(pin, flush_fn)); if (reclaim) bch2_journal_reclaim_fast(j); @@ -587,7 +591,7 @@ static size_t journal_flush_pins(struct journal *j, spin_lock(&j->lock); /* Pin might have been dropped or rearmed: */ if (likely(!err && !j->flush_in_progress_dropped)) - list_move(&pin->list, &journal_seq_pin(j, seq)->flushed[journal_pin_type(flush_fn)]); + list_move(&pin->list, &journal_seq_pin(j, seq)->flushed[journal_pin_type(pin, flush_fn)]); j->flush_in_progress = NULL; j->flush_in_progress_dropped = false; spin_unlock(&j->lock); @@ -869,18 +873,13 @@ static int journal_flush_done(struct journal *j, u64 seq_to_flush, mutex_lock(&j->reclaim_lock); - if (journal_flush_pins_or_still_flushing(j, seq_to_flush, - BIT(JOURNAL_PIN_TYPE_key_cache)| - BIT(JOURNAL_PIN_TYPE_other))) { - *did_work = true; - goto unlock; - } - - if (journal_flush_pins_or_still_flushing(j, seq_to_flush, - BIT(JOURNAL_PIN_TYPE_btree))) { - *did_work = true; - goto unlock; - } + for (int type = JOURNAL_PIN_TYPE_NR - 1; + type >= 0; + --type) + if (journal_flush_pins_or_still_flushing(j, seq_to_flush, BIT(type))) { + *did_work = true; + goto unlock; + } if (seq_to_flush > journal_cur_seq(j)) bch2_journal_entry_close(j); diff --git a/fs/bcachefs/journal_types.h b/fs/bcachefs/journal_types.h index 3ba433a48eb8..1ef3a28ed6ab 100644 --- a/fs/bcachefs/journal_types.h +++ b/fs/bcachefs/journal_types.h @@ -53,7 +53,10 @@ struct journal_buf { */ enum journal_pin_type { - JOURNAL_PIN_TYPE_btree, + JOURNAL_PIN_TYPE_btree3, + JOURNAL_PIN_TYPE_btree2, + JOURNAL_PIN_TYPE_btree1, + JOURNAL_PIN_TYPE_btree0, JOURNAL_PIN_TYPE_key_cache, JOURNAL_PIN_TYPE_other, JOURNAL_PIN_TYPE_NR, @@ -237,6 +240,7 @@ struct journal { /* seq, last_seq from the most recent journal entry successfully written */ u64 seq_ondisk; u64 flushed_seq_ondisk; + u64 flushing_seq; u64 last_seq_ondisk; u64 err_seq; u64 last_empty_seq; diff --git a/fs/bcachefs/movinggc.c b/fs/bcachefs/movinggc.c index 21805509ab9e..6718dc37c5a3 100644 --- a/fs/bcachefs/movinggc.c +++ b/fs/bcachefs/movinggc.c @@ -74,20 +74,14 @@ static int bch2_bucket_is_movable(struct btree_trans *trans, struct move_bucket *b, u64 time) { struct bch_fs *c = trans->c; - struct btree_iter iter; - struct bkey_s_c k; - struct bch_alloc_v4 _a; - const struct bch_alloc_v4 *a; - int ret; - if (bch2_bucket_is_open(trans->c, - b->k.bucket.inode, - b->k.bucket.offset)) + if (bch2_bucket_is_open(c, b->k.bucket.inode, b->k.bucket.offset)) return 0; - k = bch2_bkey_get_iter(trans, &iter, BTREE_ID_alloc, - b->k.bucket, BTREE_ITER_cached); - ret = bkey_err(k); + struct btree_iter iter; + struct bkey_s_c k = bch2_bkey_get_iter(trans, &iter, BTREE_ID_alloc, + b->k.bucket, BTREE_ITER_cached); + int ret = bkey_err(k); if (ret) return ret; @@ -95,13 +89,18 @@ static int bch2_bucket_is_movable(struct btree_trans *trans, if (!ca) goto out; - a = bch2_alloc_to_v4(k, &_a); + if (ca->mi.state != BCH_MEMBER_STATE_rw || + !bch2_dev_is_online(ca)) + goto out_put; + + struct bch_alloc_v4 _a; + const struct bch_alloc_v4 *a = bch2_alloc_to_v4(k, &_a); b->k.gen = a->gen; b->sectors = bch2_bucket_sectors_dirty(*a); u64 lru_idx = alloc_lru_idx_fragmentation(*a, ca); ret = lru_idx && lru_idx <= time; - +out_put: bch2_dev_put(ca); out: bch2_trans_iter_exit(trans, &iter); diff --git a/fs/bcachefs/opts.h b/fs/bcachefs/opts.h index a182b5d454ba..9d397fc2a1f0 100644 --- a/fs/bcachefs/opts.h +++ b/fs/bcachefs/opts.h @@ -659,18 +659,4 @@ static inline void bch2_io_opts_fixups(struct bch_io_opts *opts) struct bch_io_opts bch2_opts_to_inode_opts(struct bch_opts); bool bch2_opt_is_inode_opt(enum bch_opt_id); -/* rebalance opts: */ - -static inline struct bch_extent_rebalance io_opts_to_rebalance_opts(struct bch_io_opts *opts) -{ - return (struct bch_extent_rebalance) { - .type = BIT(BCH_EXTENT_ENTRY_rebalance), -#define x(_name) \ - ._name = opts->_name, \ - ._name##_from_inode = opts->_name##_from_inode, - BCH_REBALANCE_OPTS() -#undef x - }; -}; - #endif /* _BCACHEFS_OPTS_H */ diff --git a/fs/bcachefs/rebalance.c b/fs/bcachefs/rebalance.c index 4adc74cd3f70..d0a1f5cd5c2b 100644 --- a/fs/bcachefs/rebalance.c +++ b/fs/bcachefs/rebalance.c @@ -121,12 +121,10 @@ u64 bch2_bkey_sectors_need_rebalance(struct bch_fs *c, struct bkey_s_c k) } } incompressible: - if (opts->background_target && - bch2_target_accepts_data(c, BCH_DATA_user, opts->background_target)) { + if (opts->background_target) bkey_for_each_ptr_decode(k.k, ptrs, p, entry) if (!p.ptr.cached && !bch2_dev_in_target(c, p.ptr.dev, opts->background_target)) sectors += p.crc.compressed_size; - } return sectors; } @@ -140,7 +138,7 @@ static bool bch2_bkey_rebalance_needs_update(struct bch_fs *c, struct bch_io_opt const struct bch_extent_rebalance *old = bch2_bkey_rebalance_opts(k); if (k.k->type == KEY_TYPE_reflink_v || bch2_bkey_ptrs_need_rebalance(c, opts, k)) { - struct bch_extent_rebalance new = io_opts_to_rebalance_opts(opts); + struct bch_extent_rebalance new = io_opts_to_rebalance_opts(c, opts); return old == NULL || memcmp(old, &new, sizeof(new)); } else { return old != NULL; @@ -163,7 +161,7 @@ int bch2_bkey_set_needs_rebalance(struct bch_fs *c, struct bch_io_opts *opts, k.k->u64s += sizeof(*old) / sizeof(u64); } - *old = io_opts_to_rebalance_opts(opts); + *old = io_opts_to_rebalance_opts(c, opts); } else { if (old) extent_entry_drop(k, (union bch_extent_entry *) old); diff --git a/fs/bcachefs/rebalance.h b/fs/bcachefs/rebalance.h index 0a0821ab895d..62a3859d3823 100644 --- a/fs/bcachefs/rebalance.h +++ b/fs/bcachefs/rebalance.h @@ -4,8 +4,28 @@ #include "compress.h" #include "disk_groups.h" +#include "opts.h" #include "rebalance_types.h" +static inline struct bch_extent_rebalance io_opts_to_rebalance_opts(struct bch_fs *c, + struct bch_io_opts *opts) +{ + struct bch_extent_rebalance r = { + .type = BIT(BCH_EXTENT_ENTRY_rebalance), +#define x(_name) \ + ._name = opts->_name, \ + ._name##_from_inode = opts->_name##_from_inode, + BCH_REBALANCE_OPTS() +#undef x + }; + + if (r.background_target && + !bch2_target_accepts_data(c, BCH_DATA_user, r.background_target)) + r.background_target = 0; + + return r; +}; + u64 bch2_bkey_sectors_need_rebalance(struct bch_fs *, struct bkey_s_c); int bch2_bkey_set_needs_rebalance(struct bch_fs *, struct bch_io_opts *, struct bkey_i *); int bch2_get_update_rebalance_opts(struct btree_trans *, diff --git a/fs/bcachefs/reflink.c b/fs/bcachefs/reflink.c index 93ba4f4e47ca..441e648f28b5 100644 --- a/fs/bcachefs/reflink.c +++ b/fs/bcachefs/reflink.c @@ -172,7 +172,7 @@ static int bch2_indirect_extent_missing_error(struct btree_trans *trans, bool should_commit) { if (REFLINK_P_ERROR(p.v)) - return -BCH_ERR_missing_indirect_extent; + return 0; struct bch_fs *c = trans->c; u64 live_start = REFLINK_P_IDX(p.v); @@ -259,8 +259,6 @@ struct bkey_s_c bch2_lookup_indirect_extent(struct btree_trans *trans, return k; if (unlikely(!bkey_extent_is_reflink_data(k.k))) { - bch2_trans_iter_exit(trans, iter); - unsigned size = min((u64) k.k->size, REFLINK_P_IDX(p.v) + p.k->size + le32_to_cpu(p.v->back_pad) - reflink_offset); @@ -268,14 +266,16 @@ struct bkey_s_c bch2_lookup_indirect_extent(struct btree_trans *trans, int ret = bch2_indirect_extent_missing_error(trans, p, reflink_offset, k.k->p.offset, should_commit); - if (ret) + if (ret) { + bch2_trans_iter_exit(trans, iter); return bkey_s_c_err(ret); + } } else if (unlikely(REFLINK_P_ERROR(p.v))) { - bch2_trans_iter_exit(trans, iter); - int ret = bch2_indirect_extent_not_missing(trans, p, should_commit); - if (ret) + if (ret) { + bch2_trans_iter_exit(trans, iter); return bkey_s_c_err(ret); + } } *offset_into_extent = reflink_offset - bkey_start_offset(k.k); @@ -300,7 +300,7 @@ static int trans_trigger_reflink_p_segment(struct btree_trans *trans, if (ret) return ret; - if (bkey_deleted(k.k)) { + if (!bkey_refcount_c(k)) { if (!(flags & BTREE_TRIGGER_overwrite)) ret = -BCH_ERR_missing_indirect_extent; goto next; diff --git a/fs/bcachefs/sb-downgrade.c b/fs/bcachefs/sb-downgrade.c index 14f6b6a5fb38..051214fdc735 100644 --- a/fs/bcachefs/sb-downgrade.c +++ b/fs/bcachefs/sb-downgrade.c @@ -90,10 +90,7 @@ BIT_ULL(BCH_RECOVERY_PASS_check_allocations), \ BCH_FSCK_ERR_accounting_mismatch, \ BCH_FSCK_ERR_accounting_key_replicas_nr_devs_0, \ - BCH_FSCK_ERR_accounting_key_junk_at_end) \ - x(directory_size, \ - BIT_ULL(BCH_RECOVERY_PASS_check_inodes), \ - BCH_FSCK_ERR_directory_size_mismatch) \ + BCH_FSCK_ERR_accounting_key_junk_at_end) #define DOWNGRADE_TABLE() \ x(bucket_stripe_sectors, \ diff --git a/fs/bcachefs/sb-errors_format.h b/fs/bcachefs/sb-errors_format.h index ea0a18364751..b86ec013d7d7 100644 --- a/fs/bcachefs/sb-errors_format.h +++ b/fs/bcachefs/sb-errors_format.h @@ -180,9 +180,9 @@ enum bch_fsck_flags { x(ptr_crc_nonce_mismatch, 162, 0) \ x(ptr_stripe_redundant, 163, 0) \ x(reservation_key_nr_replicas_invalid, 164, 0) \ - x(reflink_v_refcount_wrong, 165, 0) \ + x(reflink_v_refcount_wrong, 165, FSCK_AUTOFIX) \ x(reflink_v_pos_bad, 292, 0) \ - x(reflink_p_to_missing_reflink_v, 166, 0) \ + x(reflink_p_to_missing_reflink_v, 166, FSCK_AUTOFIX) \ x(reflink_refcount_underflow, 293, 0) \ x(stripe_pos_bad, 167, 0) \ x(stripe_val_size_bad, 168, 0) \ diff --git a/fs/bcachefs/six.c b/fs/bcachefs/six.c index 7e7c66a1e1a6..7c403427fbdb 100644 --- a/fs/bcachefs/six.c +++ b/fs/bcachefs/six.c @@ -850,7 +850,8 @@ void six_lock_exit(struct six_lock *lock) EXPORT_SYMBOL_GPL(six_lock_exit); void __six_lock_init(struct six_lock *lock, const char *name, - struct lock_class_key *key, enum six_lock_init_flags flags) + struct lock_class_key *key, enum six_lock_init_flags flags, + gfp_t gfp) { atomic_set(&lock->state, 0); raw_spin_lock_init(&lock->wait_lock); @@ -873,7 +874,7 @@ void __six_lock_init(struct six_lock *lock, const char *name, * failure if they wish by checking lock->readers, but generally * will not want to treat it as an error. */ - lock->readers = alloc_percpu(unsigned); + lock->readers = alloc_percpu_gfp(unsigned, gfp); } #endif } diff --git a/fs/bcachefs/six.h b/fs/bcachefs/six.h index c142e06b7a3a..59b851cf8bac 100644 --- a/fs/bcachefs/six.h +++ b/fs/bcachefs/six.h @@ -164,18 +164,19 @@ enum six_lock_init_flags { }; void __six_lock_init(struct six_lock *lock, const char *name, - struct lock_class_key *key, enum six_lock_init_flags flags); + struct lock_class_key *key, enum six_lock_init_flags flags, + gfp_t gfp); /** * six_lock_init - initialize a six lock * @lock: lock to initialize * @flags: optional flags, i.e. SIX_LOCK_INIT_PCPU */ -#define six_lock_init(lock, flags) \ +#define six_lock_init(lock, flags, gfp) \ do { \ static struct lock_class_key __key; \ \ - __six_lock_init((lock), #lock, &__key, flags); \ + __six_lock_init((lock), #lock, &__key, flags, gfp); \ } while (0) /** diff --git a/fs/bcachefs/subvolume.c b/fs/bcachefs/subvolume.c index e3d0475232e5..b7b96283c316 100644 --- a/fs/bcachefs/subvolume.c +++ b/fs/bcachefs/subvolume.c @@ -428,7 +428,7 @@ static int __bch2_subvolume_delete(struct btree_trans *trans, u32 subvolid) bch2_bkey_get_iter_typed(trans, &snapshot_iter, BTREE_ID_snapshots, POS(0, snapid), 0, snapshot); - ret = bkey_err(subvol); + ret = bkey_err(snapshot); bch2_fs_inconsistent_on(bch2_err_matches(ret, ENOENT), trans->c, "missing snapshot %u", snapid); if (ret) @@ -440,6 +440,11 @@ static int __bch2_subvolume_delete(struct btree_trans *trans, u32 subvolid) bch2_bkey_get_iter_typed(trans, &snapshot_tree_iter, BTREE_ID_snapshot_trees, POS(0, treeid), 0, snapshot_tree); + ret = bkey_err(snapshot_tree); + bch2_fs_inconsistent_on(bch2_err_matches(ret, ENOENT), trans->c, + "missing snapshot tree %u", treeid); + if (ret) + goto err; if (le32_to_cpu(snapshot_tree.v->master_subvol) == subvolid) { struct bkey_i_snapshot_tree *snapshot_tree_mut = diff --git a/fs/bcachefs/super-io.c b/fs/bcachefs/super-io.c index 8037ccbacf6a..a81a7b6c0989 100644 --- a/fs/bcachefs/super-io.c +++ b/fs/bcachefs/super-io.c @@ -69,14 +69,20 @@ enum bcachefs_metadata_version bch2_latest_compatible_version(enum bcachefs_meta return v; } -void bch2_set_version_incompat(struct bch_fs *c, enum bcachefs_metadata_version version) +bool bch2_set_version_incompat(struct bch_fs *c, enum bcachefs_metadata_version version) { - mutex_lock(&c->sb_lock); - SET_BCH_SB_VERSION_INCOMPAT(c->disk_sb.sb, - max(BCH_SB_VERSION_INCOMPAT(c->disk_sb.sb), version)); - c->disk_sb.sb->features[0] |= cpu_to_le64(BCH_FEATURE_incompat_version_field); - bch2_write_super(c); - mutex_unlock(&c->sb_lock); + bool ret = (c->sb.features & BIT_ULL(BCH_FEATURE_incompat_version_field)) && + version <= c->sb.version_incompat_allowed; + + if (ret) { + mutex_lock(&c->sb_lock); + SET_BCH_SB_VERSION_INCOMPAT(c->disk_sb.sb, + max(BCH_SB_VERSION_INCOMPAT(c->disk_sb.sb), version)); + bch2_write_super(c); + mutex_unlock(&c->sb_lock); + } + + return ret; } const char * const bch2_sb_fields[] = { @@ -1219,9 +1225,11 @@ void bch2_sb_upgrade(struct bch_fs *c, unsigned new_version, bool incompat) c->disk_sb.sb->version = cpu_to_le16(new_version); c->disk_sb.sb->features[0] |= cpu_to_le64(BCH_SB_FEATURES_ALL); - if (incompat) + if (incompat) { SET_BCH_SB_VERSION_INCOMPAT_ALLOWED(c->disk_sb.sb, max(BCH_SB_VERSION_INCOMPAT_ALLOWED(c->disk_sb.sb), new_version)); + c->disk_sb.sb->features[0] |= cpu_to_le64(BCH_FEATURE_incompat_version_field); + } } static int bch2_sb_ext_validate(struct bch_sb *sb, struct bch_sb_field *f, diff --git a/fs/bcachefs/super-io.h b/fs/bcachefs/super-io.h index f1ab4f943720..b4cff9ebdebb 100644 --- a/fs/bcachefs/super-io.h +++ b/fs/bcachefs/super-io.h @@ -21,17 +21,14 @@ static inline bool bch2_version_compatible(u16 version) void bch2_version_to_text(struct printbuf *, enum bcachefs_metadata_version); enum bcachefs_metadata_version bch2_latest_compatible_version(enum bcachefs_metadata_version); -void bch2_set_version_incompat(struct bch_fs *, enum bcachefs_metadata_version); +bool bch2_set_version_incompat(struct bch_fs *, enum bcachefs_metadata_version); static inline bool bch2_request_incompat_feature(struct bch_fs *c, enum bcachefs_metadata_version version) { - if (unlikely(version > c->sb.version_incompat)) { - if (version > c->sb.version_incompat_allowed) - return false; - bch2_set_version_incompat(c, version); - } - return true; + return likely(version <= c->sb.version_incompat) + ? true + : bch2_set_version_incompat(c, version); } static inline size_t bch2_sb_field_bytes(struct bch_sb_field *f) diff --git a/fs/bcachefs/super.c b/fs/bcachefs/super.c index d97ea7bd1171..6d97d412fed9 100644 --- a/fs/bcachefs/super.c +++ b/fs/bcachefs/super.c @@ -411,6 +411,17 @@ bool bch2_fs_emergency_read_only(struct bch_fs *c) return ret; } +bool bch2_fs_emergency_read_only_locked(struct bch_fs *c) +{ + bool ret = !test_and_set_bit(BCH_FS_emergency_ro, &c->flags); + + bch2_journal_halt_locked(&c->journal); + bch2_fs_read_only_async(c); + + wake_up(&bch2_read_only_wait); + return ret; +} + static int bch2_fs_read_write_late(struct bch_fs *c) { int ret; diff --git a/fs/bcachefs/super.h b/fs/bcachefs/super.h index fa6d52216510..04f8287eff5c 100644 --- a/fs/bcachefs/super.h +++ b/fs/bcachefs/super.h @@ -29,6 +29,7 @@ int bch2_dev_resize(struct bch_fs *, struct bch_dev *, u64); struct bch_dev *bch2_dev_lookup(struct bch_fs *, const char *); bool bch2_fs_emergency_read_only(struct bch_fs *); +bool bch2_fs_emergency_read_only_locked(struct bch_fs *); void bch2_fs_read_only(struct bch_fs *); int bch2_fs_read_write(struct bch_fs *); diff --git a/fs/bcachefs/trace.h b/fs/bcachefs/trace.h index 56a5a7fbc0fd..c1b51009edf6 100644 --- a/fs/bcachefs/trace.h +++ b/fs/bcachefs/trace.h @@ -727,7 +727,7 @@ DEFINE_EVENT(fs_str, bucket_alloc_fail, TP_ARGS(c, str) ); -TRACE_EVENT(discard_buckets, +DECLARE_EVENT_CLASS(discard_buckets_class, TP_PROTO(struct bch_fs *c, u64 seen, u64 open, u64 need_journal_commit, u64 discarded, const char *err), TP_ARGS(c, seen, open, need_journal_commit, discarded, err), @@ -759,6 +759,18 @@ TRACE_EVENT(discard_buckets, __entry->err) ); +DEFINE_EVENT(discard_buckets_class, discard_buckets, + TP_PROTO(struct bch_fs *c, u64 seen, u64 open, + u64 need_journal_commit, u64 discarded, const char *err), + TP_ARGS(c, seen, open, need_journal_commit, discarded, err) +); + +DEFINE_EVENT(discard_buckets_class, discard_buckets_fast, + TP_PROTO(struct bch_fs *c, u64 seen, u64 open, + u64 need_journal_commit, u64 discarded, const char *err), + TP_ARGS(c, seen, open, need_journal_commit, discarded, err) +); + TRACE_EVENT(bucket_invalidate, TP_PROTO(struct bch_fs *c, unsigned dev, u64 bucket, u32 sectors), TP_ARGS(c, dev, bucket, sectors), diff --git a/fs/btrfs/ctree.c b/fs/btrfs/ctree.c index 92071ca0655f..3dc5a35dd19b 100644 --- a/fs/btrfs/ctree.c +++ b/fs/btrfs/ctree.c @@ -1496,6 +1496,7 @@ read_block_for_search(struct btrfs_root *root, struct btrfs_path *p, if (!p->skip_locking) { btrfs_unlock_up_safe(p, parent_level + 1); + btrfs_maybe_reset_lockdep_class(root, tmp); tmp_locked = true; btrfs_tree_read_lock(tmp); btrfs_release_path(p); @@ -1539,6 +1540,7 @@ read_block_for_search(struct btrfs_root *root, struct btrfs_path *p, if (!p->skip_locking) { ASSERT(ret == -EAGAIN); + btrfs_maybe_reset_lockdep_class(root, tmp); tmp_locked = true; btrfs_tree_read_lock(tmp); btrfs_release_path(p); diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c index a2cac9d0a1a9..b2fae67f8fa3 100644 --- a/fs/btrfs/extent_io.c +++ b/fs/btrfs/extent_io.c @@ -523,8 +523,6 @@ static void end_bbio_data_read(struct btrfs_bio *bbio) u64 end; u32 len; - /* For now only order 0 folios are supported for data. */ - ASSERT(folio_order(folio) == 0); btrfs_debug(fs_info, "%s: bi_sector=%llu, err=%d, mirror=%u", __func__, bio->bi_iter.bi_sector, bio->bi_status, @@ -552,7 +550,6 @@ static void end_bbio_data_read(struct btrfs_bio *bbio) if (likely(uptodate)) { loff_t i_size = i_size_read(inode); - pgoff_t end_index = i_size >> folio_shift(folio); /* * Zero out the remaining part if this range straddles @@ -561,9 +558,11 @@ static void end_bbio_data_read(struct btrfs_bio *bbio) * Here we should only zero the range inside the folio, * not touch anything else. * - * NOTE: i_size is exclusive while end is inclusive. + * NOTE: i_size is exclusive while end is inclusive and + * folio_contains() takes PAGE_SIZE units. */ - if (folio_index(folio) == end_index && i_size <= end) { + if (folio_contains(folio, i_size >> PAGE_SHIFT) && + i_size <= end) { u32 zero_start = max(offset_in_folio(folio, i_size), offset_in_folio(folio, start)); u32 zero_len = offset_in_folio(folio, end) + 1 - @@ -899,7 +898,6 @@ static struct extent_map *get_extent_map(struct btrfs_inode *inode, u64 len, struct extent_map **em_cached) { struct extent_map *em; - struct extent_state *cached_state = NULL; ASSERT(em_cached); @@ -915,14 +913,12 @@ static struct extent_map *get_extent_map(struct btrfs_inode *inode, *em_cached = NULL; } - btrfs_lock_and_flush_ordered_range(inode, start, start + len - 1, &cached_state); em = btrfs_get_extent(inode, folio, start, len); if (!IS_ERR(em)) { BUG_ON(*em_cached); refcount_inc(&em->refs); *em_cached = em; } - unlock_extent(&inode->io_tree, start, start + len - 1, &cached_state); return em; } @@ -956,7 +952,7 @@ static int btrfs_do_readpage(struct folio *folio, struct extent_map **em_cached, return ret; } - if (folio->index == last_byte >> folio_shift(folio)) { + if (folio_contains(folio, last_byte >> PAGE_SHIFT)) { size_t zero_offset = offset_in_folio(folio, last_byte); if (zero_offset) { @@ -1079,11 +1075,18 @@ static int btrfs_do_readpage(struct folio *folio, struct extent_map **em_cached, int btrfs_read_folio(struct file *file, struct folio *folio) { + struct btrfs_inode *inode = folio_to_inode(folio); + const u64 start = folio_pos(folio); + const u64 end = start + folio_size(folio) - 1; + struct extent_state *cached_state = NULL; struct btrfs_bio_ctrl bio_ctrl = { .opf = REQ_OP_READ }; struct extent_map *em_cached = NULL; int ret; + btrfs_lock_and_flush_ordered_range(inode, start, end, &cached_state); ret = btrfs_do_readpage(folio, &em_cached, &bio_ctrl, NULL); + unlock_extent(&inode->io_tree, start, end, &cached_state); + free_extent_map(em_cached); /* @@ -2380,12 +2383,20 @@ void btrfs_readahead(struct readahead_control *rac) { struct btrfs_bio_ctrl bio_ctrl = { .opf = REQ_OP_READ | REQ_RAHEAD }; struct folio *folio; + struct btrfs_inode *inode = BTRFS_I(rac->mapping->host); + const u64 start = readahead_pos(rac); + const u64 end = start + readahead_length(rac) - 1; + struct extent_state *cached_state = NULL; struct extent_map *em_cached = NULL; u64 prev_em_start = (u64)-1; + btrfs_lock_and_flush_ordered_range(inode, start, end, &cached_state); + while ((folio = readahead_folio(rac)) != NULL) btrfs_do_readpage(folio, &em_cached, &bio_ctrl, &prev_em_start); + unlock_extent(&inode->io_tree, start, end, &cached_state); + if (em_cached) free_extent_map(em_cached); submit_one_bio(&bio_ctrl); diff --git a/fs/btrfs/extent_map.c b/fs/btrfs/extent_map.c index 67ce85ff0ae2..7f46abbd6311 100644 --- a/fs/btrfs/extent_map.c +++ b/fs/btrfs/extent_map.c @@ -1128,6 +1128,8 @@ static long btrfs_scan_inode(struct btrfs_inode *inode, struct btrfs_em_shrink_c long nr_dropped = 0; struct rb_node *node; + lockdep_assert_held_write(&tree->lock); + /* * Take the mmap lock so that we serialize with the inode logging phase * of fsync because we may need to set the full sync flag on the inode, @@ -1139,28 +1141,12 @@ static long btrfs_scan_inode(struct btrfs_inode *inode, struct btrfs_em_shrink_c * to find new extents, which may not be there yet because ordered * extents haven't completed yet. * - * We also do a try lock because otherwise we could deadlock. This is - * because the shrinker for this filesystem may be invoked while we are - * in a path that is holding the mmap lock in write mode. For example in - * a reflink operation while COWing an extent buffer, when allocating - * pages for a new extent buffer and under memory pressure, the shrinker - * may be invoked, and therefore we would deadlock by attempting to read - * lock the mmap lock while we are holding already a write lock on it. + * We also do a try lock because we don't want to block for too long and + * we are holding the extent map tree's lock in write mode. */ if (!down_read_trylock(&inode->i_mmap_lock)) return 0; - /* - * We want to be fast so if the lock is busy we don't want to spend time - * waiting for it - either some task is about to do IO for the inode or - * we may have another task shrinking extent maps, here in this code, so - * skip this inode. - */ - if (!write_trylock(&tree->lock)) { - up_read(&inode->i_mmap_lock); - return 0; - } - node = rb_first(&tree->root); while (node) { struct rb_node *next = rb_next(node); @@ -1201,12 +1187,61 @@ next: break; node = next; } - write_unlock(&tree->lock); up_read(&inode->i_mmap_lock); return nr_dropped; } +static struct btrfs_inode *find_first_inode_to_shrink(struct btrfs_root *root, + u64 min_ino) +{ + struct btrfs_inode *inode; + unsigned long from = min_ino; + + xa_lock(&root->inodes); + while (true) { + struct extent_map_tree *tree; + + inode = xa_find(&root->inodes, &from, ULONG_MAX, XA_PRESENT); + if (!inode) + break; + + tree = &inode->extent_tree; + + /* + * We want to be fast so if the lock is busy we don't want to + * spend time waiting for it (some task is about to do IO for + * the inode). + */ + if (!write_trylock(&tree->lock)) + goto next; + + /* + * Skip inode if it doesn't have loaded extent maps, so we avoid + * getting a reference and doing an iput later. This includes + * cases like files that were opened for things like stat(2), or + * files with all extent maps previously released through the + * release folio callback (btrfs_release_folio()) or released in + * a previous run, or directories which never have extent maps. + */ + if (RB_EMPTY_ROOT(&tree->root)) { + write_unlock(&tree->lock); + goto next; + } + + if (igrab(&inode->vfs_inode)) + break; + + write_unlock(&tree->lock); +next: + from = btrfs_ino(inode) + 1; + cond_resched_lock(&root->inodes.xa_lock); + } + xa_unlock(&root->inodes); + + return inode; +} + static long btrfs_scan_root(struct btrfs_root *root, struct btrfs_em_shrink_ctx *ctx) { struct btrfs_fs_info *fs_info = root->fs_info; @@ -1214,21 +1249,21 @@ static long btrfs_scan_root(struct btrfs_root *root, struct btrfs_em_shrink_ctx long nr_dropped = 0; u64 min_ino = fs_info->em_shrinker_last_ino + 1; - inode = btrfs_find_first_inode(root, min_ino); + inode = find_first_inode_to_shrink(root, min_ino); while (inode) { nr_dropped += btrfs_scan_inode(inode, ctx); + write_unlock(&inode->extent_tree.lock); min_ino = btrfs_ino(inode) + 1; fs_info->em_shrinker_last_ino = btrfs_ino(inode); - btrfs_add_delayed_iput(inode); + iput(&inode->vfs_inode); - if (ctx->scanned >= ctx->nr_to_scan || - btrfs_fs_closing(inode->root->fs_info)) + if (ctx->scanned >= ctx->nr_to_scan || btrfs_fs_closing(fs_info)) break; cond_resched(); - inode = btrfs_find_first_inode(root, min_ino); + inode = find_first_inode_to_shrink(root, min_ino); } if (inode) { diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c index 36f51c311bb1..0b568c8d24cb 100644 --- a/fs/btrfs/file.c +++ b/fs/btrfs/file.c @@ -1039,7 +1039,6 @@ int btrfs_write_check(struct kiocb *iocb, size_t count) loff_t pos = iocb->ki_pos; int ret; loff_t oldsize; - loff_t start_pos; /* * Quickly bail out on NOWAIT writes if we don't have the nodatacow or @@ -1066,9 +1065,8 @@ int btrfs_write_check(struct kiocb *iocb, size_t count) inode_inc_iversion(inode); } - start_pos = round_down(pos, fs_info->sectorsize); oldsize = i_size_read(inode); - if (start_pos > oldsize) { + if (pos > oldsize) { /* Expand hole size to cover write data, preventing empty gap */ loff_t end_pos = round_up(pos + count, fs_info->sectorsize); @@ -1092,7 +1090,7 @@ ssize_t btrfs_buffered_write(struct kiocb *iocb, struct iov_iter *i) u64 lockend; size_t num_written = 0; ssize_t ret; - loff_t old_isize = i_size_read(inode); + loff_t old_isize; unsigned int ilock_flags = 0; const bool nowait = (iocb->ki_flags & IOCB_NOWAIT); unsigned int bdp_flags = (nowait ? BDP_ASYNC : 0); @@ -1105,6 +1103,13 @@ ssize_t btrfs_buffered_write(struct kiocb *iocb, struct iov_iter *i) if (ret < 0) return ret; + /* + * We can only trust the isize with inode lock held, or it can race with + * other buffered writes and cause incorrect call of + * pagecache_isize_extended() to overwrite existing data. + */ + old_isize = i_size_read(inode); + ret = generic_write_checks(iocb, i); if (ret <= 0) goto out; diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index a9322601ab5c..38756f8cef46 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -1382,8 +1382,13 @@ static noinline int cow_file_range(struct btrfs_inode *inode, continue; } if (done_offset) { - *done_offset = start - 1; - return 0; + /* + * Move @end to the end of the processed range, + * and exit the loop to unlock the processed extents. + */ + end = start - 1; + ret = 0; + break; } ret = -ENOSPC; } diff --git a/fs/btrfs/ordered-data.c b/fs/btrfs/ordered-data.c index 30eceaf829a7..4aca7475fd82 100644 --- a/fs/btrfs/ordered-data.c +++ b/fs/btrfs/ordered-data.c @@ -1229,6 +1229,18 @@ struct btrfs_ordered_extent *btrfs_split_ordered_extent( */ if (WARN_ON_ONCE(len >= ordered->num_bytes)) return ERR_PTR(-EINVAL); + /* + * If our ordered extent had an error there's no point in continuing. + * The error may have come from a transaction abort done either by this + * task or some other concurrent task, and the transaction abort path + * iterates over all existing ordered extents and sets the flag + * BTRFS_ORDERED_IOERR on them. + */ + if (unlikely(flags & (1U << BTRFS_ORDERED_IOERR))) { + const int fs_error = BTRFS_FS_ERROR(fs_info); + + return fs_error ? ERR_PTR(fs_error) : ERR_PTR(-EIO); + } /* We cannot split partially completed ordered extents. */ if (ordered->bytes_left) { ASSERT(!(flags & ~BTRFS_ORDERED_TYPE_FLAGS)); diff --git a/fs/btrfs/qgroup.c b/fs/btrfs/qgroup.c index b90fabe302e6..f9d3766c809b 100644 --- a/fs/btrfs/qgroup.c +++ b/fs/btrfs/qgroup.c @@ -1880,11 +1880,7 @@ int btrfs_qgroup_cleanup_dropped_subvolume(struct btrfs_fs_info *fs_info, u64 su * Commit current transaction to make sure all the rfer/excl numbers * get updated. */ - trans = btrfs_start_transaction(fs_info->quota_root, 0); - if (IS_ERR(trans)) - return PTR_ERR(trans); - - ret = btrfs_commit_transaction(trans); + ret = btrfs_commit_current_transaction(fs_info->quota_root); if (ret < 0) return ret; @@ -1897,8 +1893,11 @@ int btrfs_qgroup_cleanup_dropped_subvolume(struct btrfs_fs_info *fs_info, u64 su /* * It's squota and the subvolume still has numbers needed for future * accounting, in this case we can not delete it. Just skip it. + * + * Or the qgroup is already removed by a qgroup rescan. For both cases we're + * safe to ignore them. */ - if (ret == -EBUSY) + if (ret == -EBUSY || ret == -ENOENT) ret = 0; return ret; } diff --git a/fs/btrfs/sysfs.c b/fs/btrfs/sysfs.c index 53b846d99ece..14f53f757555 100644 --- a/fs/btrfs/sysfs.c +++ b/fs/btrfs/sysfs.c @@ -1330,13 +1330,13 @@ MODULE_PARM_DESC(read_policy, int btrfs_read_policy_to_enum(const char *str, s64 *value_ret) { - char param[32] = { 0 }; + char param[32]; char __maybe_unused *value_str; if (!str || strlen(str) == 0) return 0; - strncpy(param, str, sizeof(param) - 1); + strscpy(param, str); #ifdef CONFIG_BTRFS_EXPERIMENTAL /* Separate value from input in policy:value format. */ diff --git a/fs/btrfs/tests/delayed-refs-tests.c b/fs/btrfs/tests/delayed-refs-tests.c index 6558508c2ddf..265370e79a54 100644 --- a/fs/btrfs/tests/delayed-refs-tests.c +++ b/fs/btrfs/tests/delayed-refs-tests.c @@ -1009,6 +1009,7 @@ int btrfs_test_delayed_refs(u32 sectorsize, u32 nodesize) if (!ret) ret = select_delayed_refs_test(&trans); + kfree(transaction); out_free_fs_info: btrfs_free_dummy_fs_info(fs_info); return ret; diff --git a/fs/btrfs/transaction.c b/fs/btrfs/transaction.c index 15312013f2a3..aca83a98b75a 100644 --- a/fs/btrfs/transaction.c +++ b/fs/btrfs/transaction.c @@ -274,8 +274,10 @@ loop: cur_trans = fs_info->running_transaction; if (cur_trans) { if (TRANS_ABORTED(cur_trans)) { + const int abort_error = cur_trans->aborted; + spin_unlock(&fs_info->trans_lock); - return cur_trans->aborted; + return abort_error; } if (btrfs_blocked_trans_types[cur_trans->state] & type) { spin_unlock(&fs_info->trans_lock); diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c index 0a0776489055..3f8afbd1ebb5 100644 --- a/fs/btrfs/volumes.c +++ b/fs/btrfs/volumes.c @@ -7155,6 +7155,7 @@ static int read_one_chunk(struct btrfs_key *key, struct extent_buffer *leaf, btrfs_err(fs_info, "failed to add chunk map, start=%llu len=%llu: %d", map->start, map->chunk_len, ret); + btrfs_free_chunk_map(map); } return ret; @@ -7200,8 +7201,12 @@ static struct btrfs_fs_devices *open_seed_devices(struct btrfs_fs_info *fs_info, fs_devices = find_fsid(fsid, NULL); if (!fs_devices) { - if (!btrfs_test_opt(fs_info, DEGRADED)) + if (!btrfs_test_opt(fs_info, DEGRADED)) { + btrfs_err(fs_info, + "failed to find fsid %pU when attempting to open seed devices", + fsid); return ERR_PTR(-ENOENT); + } fs_devices = alloc_fs_devices(fsid); if (IS_ERR(fs_devices)) diff --git a/fs/coredump.c b/fs/coredump.c index 591700e1b2ce..4375c70144d0 100644 --- a/fs/coredump.c +++ b/fs/coredump.c @@ -63,6 +63,7 @@ static void free_vma_snapshot(struct coredump_params *cprm); static int core_uses_pid; static unsigned int core_pipe_limit; +static unsigned int core_sort_vma; static char core_pattern[CORENAME_MAX_SIZE] = "core"; static int core_name_size = CORENAME_MAX_SIZE; unsigned int core_file_note_size_limit = CORE_FILE_NOTE_SIZE_DEFAULT; @@ -1026,6 +1027,15 @@ static const struct ctl_table coredump_sysctls[] = { .extra1 = (unsigned int *)&core_file_note_size_min, .extra2 = (unsigned int *)&core_file_note_size_max, }, + { + .procname = "core_sort_vma", + .data = &core_sort_vma, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_douintvec_minmax, + .extra1 = SYSCTL_ZERO, + .extra2 = SYSCTL_ONE, + }, }; static int __init init_fs_coredump_sysctls(void) @@ -1256,8 +1266,9 @@ static bool dump_vma_snapshot(struct coredump_params *cprm) cprm->vma_data_size += m->dump_size; } - sort(cprm->vma_meta, cprm->vma_count, sizeof(*cprm->vma_meta), - cmp_vma_size, NULL); + if (core_sort_vma) + sort(cprm->vma_meta, cprm->vma_count, sizeof(*cprm->vma_meta), + cmp_vma_size, NULL); return true; } diff --git a/fs/dcache.c b/fs/dcache.c index 9cc0d47da321..e3634916ffb9 100644 --- a/fs/dcache.c +++ b/fs/dcache.c @@ -1700,7 +1700,7 @@ static struct dentry *__d_alloc(struct super_block *sb, const struct qstr *name) smp_store_release(&dentry->d_name.name, dname); /* ^^^ */ dentry->d_flags = 0; - lockref_init(&dentry->d_lockref, 1); + lockref_init(&dentry->d_lockref); seqcount_spinlock_init(&dentry->d_seq, &dentry->d_lock); dentry->d_inode = NULL; dentry->d_parent = dentry; @@ -2966,11 +2966,11 @@ static int __d_unalias(struct dentry *dentry, struct dentry *alias) goto out_err; m2 = &alias->d_parent->d_inode->i_rwsem; out_unalias: - if (alias->d_op->d_unalias_trylock && + if (alias->d_op && alias->d_op->d_unalias_trylock && !alias->d_op->d_unalias_trylock(alias)) goto out_err; __d_move(alias, dentry, false); - if (alias->d_op->d_unalias_unlock) + if (alias->d_op && alias->d_op->d_unalias_unlock) alias->d_op->d_unalias_unlock(alias); ret = 0; out_err: diff --git a/fs/efivarfs/file.c b/fs/efivarfs/file.c index cb1b6d0c3454..c294a8fc566d 100644 --- a/fs/efivarfs/file.c +++ b/fs/efivarfs/file.c @@ -57,10 +57,11 @@ static ssize_t efivarfs_file_write(struct file *file, if (bytes == -ENOENT) { /* - * zero size signals to release that the write deleted - * the variable + * FIXME: temporary workaround for fwupdate, signal + * failed write with a 1 to keep created but not + * written files */ - i_size_write(inode, 0); + i_size_write(inode, 1); } else { i_size_write(inode, datasize + sizeof(attributes)); inode_set_mtime_to_ts(inode, inode_set_ctime_current(inode)); @@ -124,7 +125,8 @@ static int efivarfs_file_release(struct inode *inode, struct file *file) struct efivar_entry *var = inode->i_private; inode_lock(inode); - var->removed = (--var->open_count == 0 && i_size_read(inode) == 0); + /* FIXME: temporary work around for fwupdate */ + var->removed = (--var->open_count == 0 && i_size_read(inode) == 1); inode_unlock(inode); if (var->removed) diff --git a/fs/efivarfs/super.c b/fs/efivarfs/super.c index 09fcf731e65d..6eae8cf655c1 100644 --- a/fs/efivarfs/super.c +++ b/fs/efivarfs/super.c @@ -367,6 +367,8 @@ static int efivarfs_fill_super(struct super_block *sb, struct fs_context *fc) if (err) return err; + register_pm_notifier(&sfi->pm_nb); + return efivar_init(efivarfs_callback, sb, true); } @@ -552,7 +554,6 @@ static int efivarfs_init_fs_context(struct fs_context *fc) sfi->pm_nb.notifier_call = efivarfs_pm_notify; sfi->pm_nb.priority = 0; - register_pm_notifier(&sfi->pm_nb); return 0; } diff --git a/fs/erofs/zdata.c b/fs/erofs/zdata.c index 29f8963bb523..d771e06db738 100644 --- a/fs/erofs/zdata.c +++ b/fs/erofs/zdata.c @@ -726,7 +726,7 @@ static int z_erofs_register_pcluster(struct z_erofs_frontend *fe) if (IS_ERR(pcl)) return PTR_ERR(pcl); - lockref_init(&pcl->lockref, 1); /* one ref for this request */ + lockref_init(&pcl->lockref); /* one ref for this request */ pcl->algorithmformat = map->m_algorithmformat; pcl->length = 0; pcl->partial = true; diff --git a/fs/exfat/balloc.c b/fs/exfat/balloc.c index ce9be95c9172..9ff825f1502d 100644 --- a/fs/exfat/balloc.c +++ b/fs/exfat/balloc.c @@ -141,7 +141,7 @@ int exfat_set_bitmap(struct inode *inode, unsigned int clu, bool sync) return 0; } -void exfat_clear_bitmap(struct inode *inode, unsigned int clu, bool sync) +int exfat_clear_bitmap(struct inode *inode, unsigned int clu, bool sync) { int i, b; unsigned int ent_idx; @@ -150,13 +150,17 @@ void exfat_clear_bitmap(struct inode *inode, unsigned int clu, bool sync) struct exfat_mount_options *opts = &sbi->options; if (!is_valid_cluster(sbi, clu)) - return; + return -EIO; ent_idx = CLUSTER_TO_BITMAP_ENT(clu); i = BITMAP_OFFSET_SECTOR_INDEX(sb, ent_idx); b = BITMAP_OFFSET_BIT_IN_SECTOR(sb, ent_idx); + if (!test_bit_le(b, sbi->vol_amap[i]->b_data)) + return -EIO; + clear_bit_le(b, sbi->vol_amap[i]->b_data); + exfat_update_bh(sbi->vol_amap[i], sync); if (opts->discard) { @@ -171,6 +175,8 @@ void exfat_clear_bitmap(struct inode *inode, unsigned int clu, bool sync) opts->discard = 0; } } + + return 0; } /* diff --git a/fs/exfat/exfat_fs.h b/fs/exfat/exfat_fs.h index 78be6964a8a0..d30ce18a88b7 100644 --- a/fs/exfat/exfat_fs.h +++ b/fs/exfat/exfat_fs.h @@ -456,7 +456,7 @@ int exfat_count_num_clusters(struct super_block *sb, int exfat_load_bitmap(struct super_block *sb); void exfat_free_bitmap(struct exfat_sb_info *sbi); int exfat_set_bitmap(struct inode *inode, unsigned int clu, bool sync); -void exfat_clear_bitmap(struct inode *inode, unsigned int clu, bool sync); +int exfat_clear_bitmap(struct inode *inode, unsigned int clu, bool sync); unsigned int exfat_find_free_bitmap(struct super_block *sb, unsigned int clu); int exfat_count_used_clusters(struct super_block *sb, unsigned int *ret_count); int exfat_trim_fs(struct inode *inode, struct fstrim_range *range); diff --git a/fs/exfat/fatent.c b/fs/exfat/fatent.c index 9e5492ac409b..6f3651c6ca91 100644 --- a/fs/exfat/fatent.c +++ b/fs/exfat/fatent.c @@ -175,6 +175,7 @@ static int __exfat_free_cluster(struct inode *inode, struct exfat_chain *p_chain BITMAP_OFFSET_SECTOR_INDEX(sb, CLUSTER_TO_BITMAP_ENT(clu)); if (p_chain->flags == ALLOC_NO_FAT_CHAIN) { + int err; unsigned int last_cluster = p_chain->dir + p_chain->size - 1; do { bool sync = false; @@ -189,7 +190,9 @@ static int __exfat_free_cluster(struct inode *inode, struct exfat_chain *p_chain cur_cmap_i = next_cmap_i; } - exfat_clear_bitmap(inode, clu, (sync && IS_DIRSYNC(inode))); + err = exfat_clear_bitmap(inode, clu, (sync && IS_DIRSYNC(inode))); + if (err) + break; clu++; num_clusters++; } while (num_clusters < p_chain->size); @@ -210,12 +213,13 @@ static int __exfat_free_cluster(struct inode *inode, struct exfat_chain *p_chain cur_cmap_i = next_cmap_i; } - exfat_clear_bitmap(inode, clu, (sync && IS_DIRSYNC(inode))); + if (exfat_clear_bitmap(inode, clu, (sync && IS_DIRSYNC(inode)))) + break; clu = n_clu; num_clusters++; if (err) - goto dec_used_clus; + break; if (num_clusters >= sbi->num_clusters - EXFAT_FIRST_CLUSTER) { /* @@ -229,7 +233,6 @@ static int __exfat_free_cluster(struct inode *inode, struct exfat_chain *p_chain } while (clu != EXFAT_EOF_CLUSTER); } -dec_used_clus: sbi->used_clusters -= num_clusters; return 0; } diff --git a/fs/exfat/file.c b/fs/exfat/file.c index 05b51e721783..807349d8ea05 100644 --- a/fs/exfat/file.c +++ b/fs/exfat/file.c @@ -587,7 +587,7 @@ static ssize_t exfat_file_write_iter(struct kiocb *iocb, struct iov_iter *iter) valid_size = ei->valid_size; ret = generic_write_checks(iocb, iter); - if (ret < 0) + if (ret <= 0) goto unlock; if (iocb->ki_flags & IOCB_DIRECT) { diff --git a/fs/exfat/namei.c b/fs/exfat/namei.c index 691dd77b6ab5..8b30027d8251 100644 --- a/fs/exfat/namei.c +++ b/fs/exfat/namei.c @@ -232,7 +232,7 @@ static int exfat_search_empty_slot(struct super_block *sb, dentry = 0; } - while (dentry + num_entries < total_entries && + while (dentry + num_entries <= total_entries && clu.dir != EXFAT_EOF_CLUSTER) { i = dentry & (dentries_per_clu - 1); @@ -646,6 +646,11 @@ static int exfat_find(struct inode *dir, struct qstr *qname, info->valid_size = le64_to_cpu(ep2->dentry.stream.valid_size); info->size = le64_to_cpu(ep2->dentry.stream.size); + if (unlikely(EXFAT_B_TO_CLU_ROUND_UP(info->size, sbi) > sbi->used_clusters)) { + exfat_fs_error(sb, "data size is invalid(%lld)", info->size); + return -EIO; + } + info->start_clu = le32_to_cpu(ep2->dentry.stream.start_clu); if (!is_valid_cluster(sbi, info->start_clu) && info->size) { exfat_warn(sb, "start_clu is invalid cluster(0x%x)", diff --git a/fs/file_table.c b/fs/file_table.c index f0291a66f9db..5c00dc38558d 100644 --- a/fs/file_table.c +++ b/fs/file_table.c @@ -194,6 +194,11 @@ static int init_file(struct file *f, int flags, const struct cred *cred) * refcount bumps we should reinitialize the reused file first. */ file_ref_init(&f->f_ref, 1); + /* + * Disable permission and pre-content events for all files by default. + * They may be enabled later by file_set_fsnotify_mode_from_watchers(). + */ + file_set_fsnotify_mode(f, FMODE_NONOTIFY_PERM); return 0; } @@ -375,7 +380,13 @@ struct file *alloc_file_pseudo(struct inode *inode, struct vfsmount *mnt, if (IS_ERR(file)) { ihold(inode); path_put(&path); + return file; } + /* + * Disable all fsnotify events for pseudo files by default. + * They may be enabled by caller with file_set_fsnotify_mode(). + */ + file_set_fsnotify_mode(file, FMODE_NONOTIFY); return file; } EXPORT_SYMBOL(alloc_file_pseudo); @@ -400,6 +411,11 @@ struct file *alloc_file_pseudo_noaccount(struct inode *inode, return file; } file_init_path(file, &path, fops); + /* + * Disable all fsnotify events for pseudo files by default. + * They may be enabled by caller with file_set_fsnotify_mode(). + */ + file_set_fsnotify_mode(file, FMODE_NONOTIFY); return file; } EXPORT_SYMBOL_GPL(alloc_file_pseudo_noaccount); diff --git a/fs/fuse/dev.c b/fs/fuse/dev.c index 5b5f789b37eb..2c3a4d09e500 100644 --- a/fs/fuse/dev.c +++ b/fs/fuse/dev.c @@ -838,6 +838,12 @@ static int fuse_check_folio(struct folio *folio) return 0; } +/* + * Attempt to steal a page from the splice() pipe and move it into the + * pagecache. If successful, the pointer in @pagep will be updated. The + * folio that was originally in @pagep will lose a reference and the new + * folio returned in @pagep will carry a reference. + */ static int fuse_try_move_page(struct fuse_copy_state *cs, struct page **pagep) { int err; @@ -1451,7 +1457,7 @@ static ssize_t fuse_dev_splice_read(struct file *in, loff_t *ppos, if (ret < 0) goto out; - if (pipe_occupancy(pipe->head, pipe->tail) + cs.nr_segs > pipe->max_usage) { + if (pipe_buf_usage(pipe) + cs.nr_segs > pipe->max_usage) { ret = -EIO; goto out; } @@ -2101,7 +2107,7 @@ static ssize_t fuse_dev_splice_write(struct pipe_inode_info *pipe, struct file *out, loff_t *ppos, size_t len, unsigned int flags) { - unsigned int head, tail, mask, count; + unsigned int head, tail, count; unsigned nbuf; unsigned idx; struct pipe_buffer *bufs; @@ -2118,8 +2124,7 @@ static ssize_t fuse_dev_splice_write(struct pipe_inode_info *pipe, head = pipe->head; tail = pipe->tail; - mask = pipe->ring_size - 1; - count = head - tail; + count = pipe_occupancy(head, tail); bufs = kvmalloc_array(count, sizeof(struct pipe_buffer), GFP_KERNEL); if (!bufs) { @@ -2129,8 +2134,8 @@ static ssize_t fuse_dev_splice_write(struct pipe_inode_info *pipe, nbuf = 0; rem = 0; - for (idx = tail; idx != head && rem < len; idx++) - rem += pipe->bufs[idx & mask].len; + for (idx = tail; !pipe_empty(head, idx) && rem < len; idx++) + rem += pipe_buf(pipe, idx)->len; ret = -EINVAL; if (rem < len) @@ -2141,10 +2146,10 @@ static ssize_t fuse_dev_splice_write(struct pipe_inode_info *pipe, struct pipe_buffer *ibuf; struct pipe_buffer *obuf; - if (WARN_ON(nbuf >= count || tail == head)) + if (WARN_ON(nbuf >= count || pipe_empty(head, tail))) goto out_free; - ibuf = &pipe->bufs[tail & mask]; + ibuf = pipe_buf(pipe, tail); obuf = &bufs[nbuf]; if (rem >= ibuf->len) { diff --git a/fs/fuse/dir.c b/fs/fuse/dir.c index 198862b086ff..3805f9b06c9d 100644 --- a/fs/fuse/dir.c +++ b/fs/fuse/dir.c @@ -1636,7 +1636,7 @@ static const char *fuse_get_link(struct dentry *dentry, struct inode *inode, goto out_err; if (fc->cache_symlinks) - return page_get_link(dentry, inode, callback); + return page_get_link_raw(dentry, inode, callback); err = -ECHILD; if (!dentry) diff --git a/fs/fuse/file.c b/fs/fuse/file.c index 7d92a5479998..d63e56fd3dd2 100644 --- a/fs/fuse/file.c +++ b/fs/fuse/file.c @@ -955,8 +955,10 @@ static void fuse_readpages_end(struct fuse_mount *fm, struct fuse_args *args, fuse_invalidate_atime(inode); } - for (i = 0; i < ap->num_folios; i++) + for (i = 0; i < ap->num_folios; i++) { folio_end_read(ap->folios[i], !err); + folio_put(ap->folios[i]); + } if (ia->ff) fuse_file_put(ia->ff, false); @@ -1048,7 +1050,14 @@ static void fuse_readahead(struct readahead_control *rac) ap = &ia->ap; while (ap->num_folios < cur_pages) { - folio = readahead_folio(rac); + /* + * This returns a folio with a ref held on it. + * The ref needs to be held until the request is + * completed, since the splice case (see + * fuse_try_move_page()) drops the ref after it's + * replaced in the page cache. + */ + folio = __readahead_folio(rac); ap->folios[ap->num_folios] = folio; ap->descs[ap->num_folios].length = folio_size(folio); ap->num_folios++; diff --git a/fs/gfs2/glock.c b/fs/gfs2/glock.c index 8c4c1f871a88..65c07aa95718 100644 --- a/fs/gfs2/glock.c +++ b/fs/gfs2/glock.c @@ -1201,8 +1201,8 @@ int gfs2_glock_get(struct gfs2_sbd *sdp, u64 number, if (glops->go_instantiate) gl->gl_flags |= BIT(GLF_INSTANTIATE_NEEDED); gl->gl_name = name; + lockref_init(&gl->gl_lockref); lockdep_set_subclass(&gl->gl_lockref.lock, glops->go_subclass); - gl->gl_lockref.count = 1; gl->gl_state = LM_ST_UNLOCKED; gl->gl_target = LM_ST_UNLOCKED; gl->gl_demote_state = LM_ST_EXCLUSIVE; diff --git a/fs/gfs2/main.c b/fs/gfs2/main.c index 04cadc02e5a6..0727f60ad028 100644 --- a/fs/gfs2/main.c +++ b/fs/gfs2/main.c @@ -51,7 +51,6 @@ static void gfs2_init_glock_once(void *foo) { struct gfs2_glock *gl = foo; - spin_lock_init(&gl->gl_lockref.lock); INIT_LIST_HEAD(&gl->gl_holders); INIT_LIST_HEAD(&gl->gl_lru); INIT_LIST_HEAD(&gl->gl_ail_list); diff --git a/fs/gfs2/quota.c b/fs/gfs2/quota.c index 58bc5013ca49..2298e06797ac 100644 --- a/fs/gfs2/quota.c +++ b/fs/gfs2/quota.c @@ -236,7 +236,7 @@ static struct gfs2_quota_data *qd_alloc(unsigned hash, struct gfs2_sbd *sdp, str return NULL; qd->qd_sbd = sdp; - lockref_init(&qd->qd_lockref, 0); + lockref_init(&qd->qd_lockref); qd->qd_id = qid; qd->qd_slot = -1; INIT_LIST_HEAD(&qd->qd_lru); @@ -297,7 +297,6 @@ static int qd_get(struct gfs2_sbd *sdp, struct kqid qid, spin_lock_bucket(hash); *qdp = qd = gfs2_qd_search_bucket(hash, sdp, qid); if (qd == NULL) { - new_qd->qd_lockref.count++; *qdp = new_qd; list_add(&new_qd->qd_list, &sdp->sd_quota_list); hlist_bl_add_head_rcu(&new_qd->qd_hlist, &qd_hash_table[hash]); @@ -1450,6 +1449,7 @@ int gfs2_quota_init(struct gfs2_sbd *sdp) if (qd == NULL) goto fail_brelse; + qd->qd_lockref.count = 0; set_bit(QDF_CHANGE, &qd->qd_flags); qd->qd_change = qc_change; qd->qd_slot = slot; diff --git a/fs/iomap/direct-io.c b/fs/iomap/direct-io.c index b521eb15759e..0e47da82b0c2 100644 --- a/fs/iomap/direct-io.c +++ b/fs/iomap/direct-io.c @@ -427,12 +427,10 @@ static loff_t iomap_dio_bio_iter(const struct iomap_iter *iter, bio_put(bio); goto zero_tail; } - if (dio->flags & IOMAP_DIO_WRITE) { + if (dio->flags & IOMAP_DIO_WRITE) task_io_account_write(n); - } else { - if (dio->flags & IOMAP_DIO_DIRTY) - bio_set_pages_dirty(bio); - } + else if (dio->flags & IOMAP_DIO_DIRTY) + bio_set_pages_dirty(bio); dio->size += n; copied += n; diff --git a/fs/namei.c b/fs/namei.c index 3ab9440c5b93..ecb7b95c2ca3 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -5356,10 +5356,9 @@ const char *vfs_get_link(struct dentry *dentry, struct delayed_call *done) EXPORT_SYMBOL(vfs_get_link); /* get the link contents into pagecache */ -const char *page_get_link(struct dentry *dentry, struct inode *inode, - struct delayed_call *callback) +static char *__page_get_link(struct dentry *dentry, struct inode *inode, + struct delayed_call *callback) { - char *kaddr; struct page *page; struct address_space *mapping = inode->i_mapping; @@ -5378,8 +5377,23 @@ const char *page_get_link(struct dentry *dentry, struct inode *inode, } set_delayed_call(callback, page_put_link, page); BUG_ON(mapping_gfp_mask(mapping) & __GFP_HIGHMEM); - kaddr = page_address(page); - nd_terminate_link(kaddr, inode->i_size, PAGE_SIZE - 1); + return page_address(page); +} + +const char *page_get_link_raw(struct dentry *dentry, struct inode *inode, + struct delayed_call *callback) +{ + return __page_get_link(dentry, inode, callback); +} +EXPORT_SYMBOL_GPL(page_get_link_raw); + +const char *page_get_link(struct dentry *dentry, struct inode *inode, + struct delayed_call *callback) +{ + char *kaddr = __page_get_link(dentry, inode, callback); + + if (!IS_ERR(kaddr)) + nd_terminate_link(kaddr, inode->i_size, PAGE_SIZE - 1); return kaddr; } diff --git a/fs/namespace.c b/fs/namespace.c index a3ed3f2980cb..8f1000f9f3df 100644 --- a/fs/namespace.c +++ b/fs/namespace.c @@ -5087,30 +5087,29 @@ static int statmount_mnt_opts(struct kstatmount *s, struct seq_file *seq) { struct vfsmount *mnt = s->mnt; struct super_block *sb = mnt->mnt_sb; + size_t start = seq->count; int err; - if (sb->s_op->show_options) { - size_t start = seq->count; - - err = security_sb_show_options(seq, sb); - if (err) - return err; + err = security_sb_show_options(seq, sb); + if (err) + return err; + if (sb->s_op->show_options) { err = sb->s_op->show_options(seq, mnt->mnt_root); if (err) return err; + } - if (unlikely(seq_has_overflowed(seq))) - return -EAGAIN; + if (unlikely(seq_has_overflowed(seq))) + return -EAGAIN; - if (seq->count == start) - return 0; + if (seq->count == start) + return 0; - /* skip leading comma */ - memmove(seq->buf + start, seq->buf + start + 1, - seq->count - start - 1); - seq->count--; - } + /* skip leading comma */ + memmove(seq->buf + start, seq->buf + start + 1, + seq->count - start - 1); + seq->count--; return 0; } @@ -5191,39 +5190,45 @@ static int statmount_string(struct kstatmount *s, u64 flag) size_t kbufsize; struct seq_file *seq = &s->seq; struct statmount *sm = &s->sm; - u32 start = seq->count; + u32 start, *offp; + + /* Reserve an empty string at the beginning for any unset offsets */ + if (!seq->count) + seq_putc(seq, 0); + + start = seq->count; switch (flag) { case STATMOUNT_FS_TYPE: - sm->fs_type = start; + offp = &sm->fs_type; ret = statmount_fs_type(s, seq); break; case STATMOUNT_MNT_ROOT: - sm->mnt_root = start; + offp = &sm->mnt_root; ret = statmount_mnt_root(s, seq); break; case STATMOUNT_MNT_POINT: - sm->mnt_point = start; + offp = &sm->mnt_point; ret = statmount_mnt_point(s, seq); break; case STATMOUNT_MNT_OPTS: - sm->mnt_opts = start; + offp = &sm->mnt_opts; ret = statmount_mnt_opts(s, seq); break; case STATMOUNT_OPT_ARRAY: - sm->opt_array = start; + offp = &sm->opt_array; ret = statmount_opt_array(s, seq); break; case STATMOUNT_OPT_SEC_ARRAY: - sm->opt_sec_array = start; + offp = &sm->opt_sec_array; ret = statmount_opt_sec_array(s, seq); break; case STATMOUNT_FS_SUBTYPE: - sm->fs_subtype = start; + offp = &sm->fs_subtype; statmount_fs_subtype(s, seq); break; case STATMOUNT_SB_SOURCE: - sm->sb_source = start; + offp = &sm->sb_source; ret = statmount_sb_source(s, seq); break; default: @@ -5251,6 +5256,7 @@ static int statmount_string(struct kstatmount *s, u64 flag) seq->buf[seq->count++] = '\0'; sm->mask |= flag; + *offp = start; return 0; } diff --git a/fs/netfs/buffered_read.c b/fs/netfs/buffered_read.c index f761d44b3436..0d1b6d35ff3b 100644 --- a/fs/netfs/buffered_read.c +++ b/fs/netfs/buffered_read.c @@ -155,8 +155,9 @@ static void netfs_read_cache_to_pagecache(struct netfs_io_request *rreq, netfs_cache_read_terminated, subreq); } -static void netfs_issue_read(struct netfs_io_request *rreq, - struct netfs_io_subrequest *subreq) +static void netfs_queue_read(struct netfs_io_request *rreq, + struct netfs_io_subrequest *subreq, + bool last_subreq) { struct netfs_io_stream *stream = &rreq->io_streams[0]; @@ -177,8 +178,17 @@ static void netfs_issue_read(struct netfs_io_request *rreq, } } + if (last_subreq) { + smp_wmb(); /* Write lists before ALL_QUEUED. */ + set_bit(NETFS_RREQ_ALL_QUEUED, &rreq->flags); + } + spin_unlock(&rreq->lock); +} +static void netfs_issue_read(struct netfs_io_request *rreq, + struct netfs_io_subrequest *subreq) +{ switch (subreq->source) { case NETFS_DOWNLOAD_FROM_SERVER: rreq->netfs_ops->issue_read(subreq); @@ -293,11 +303,8 @@ static void netfs_read_to_pagecache(struct netfs_io_request *rreq) } size -= slice; start += slice; - if (size <= 0) { - smp_wmb(); /* Write lists before ALL_QUEUED. */ - set_bit(NETFS_RREQ_ALL_QUEUED, &rreq->flags); - } + netfs_queue_read(rreq, subreq, size <= 0); netfs_issue_read(rreq, subreq); cond_resched(); } while (size > 0); diff --git a/fs/netfs/internal.h b/fs/netfs/internal.h index eb76f98c894b..1c4f953c3d68 100644 --- a/fs/netfs/internal.h +++ b/fs/netfs/internal.h @@ -135,6 +135,8 @@ extern atomic_t netfs_n_rh_write_begin; extern atomic_t netfs_n_rh_write_done; extern atomic_t netfs_n_rh_write_failed; extern atomic_t netfs_n_rh_write_zskip; +extern atomic_t netfs_n_rh_retry_read_req; +extern atomic_t netfs_n_rh_retry_read_subreq; extern atomic_t netfs_n_wh_buffered_write; extern atomic_t netfs_n_wh_writethrough; extern atomic_t netfs_n_wh_dio_write; @@ -147,6 +149,8 @@ extern atomic_t netfs_n_wh_upload_failed; extern atomic_t netfs_n_wh_write; extern atomic_t netfs_n_wh_write_done; extern atomic_t netfs_n_wh_write_failed; +extern atomic_t netfs_n_wh_retry_write_req; +extern atomic_t netfs_n_wh_retry_write_subreq; extern atomic_t netfs_n_wb_lock_skip; extern atomic_t netfs_n_wb_lock_wait; extern atomic_t netfs_n_folioq; diff --git a/fs/netfs/read_collect.c b/fs/netfs/read_collect.c index f65affa5a9e4..636cc5a98ef5 100644 --- a/fs/netfs/read_collect.c +++ b/fs/netfs/read_collect.c @@ -470,7 +470,8 @@ void netfs_read_collection_worker(struct work_struct *work) */ void netfs_wake_read_collector(struct netfs_io_request *rreq) { - if (test_bit(NETFS_RREQ_OFFLOAD_COLLECTION, &rreq->flags)) { + if (test_bit(NETFS_RREQ_OFFLOAD_COLLECTION, &rreq->flags) && + !test_bit(NETFS_RREQ_RETRYING, &rreq->flags)) { if (!work_pending(&rreq->work)) { netfs_get_request(rreq, netfs_rreq_trace_get_work); if (!queue_work(system_unbound_wq, &rreq->work)) @@ -586,7 +587,8 @@ void netfs_read_subreq_terminated(struct netfs_io_subrequest *subreq) smp_mb__after_atomic(); /* Clear IN_PROGRESS before task state */ /* If we are at the head of the queue, wake up the collector. */ - if (list_is_first(&subreq->rreq_link, &stream->subrequests)) + if (list_is_first(&subreq->rreq_link, &stream->subrequests) || + test_bit(NETFS_RREQ_RETRYING, &rreq->flags)) netfs_wake_read_collector(rreq); netfs_put_subrequest(subreq, true, netfs_sreq_trace_put_terminated); diff --git a/fs/netfs/read_retry.c b/fs/netfs/read_retry.c index 2290af0d51ac..0f294b26e08c 100644 --- a/fs/netfs/read_retry.c +++ b/fs/netfs/read_retry.c @@ -14,7 +14,7 @@ static void netfs_reissue_read(struct netfs_io_request *rreq, { __clear_bit(NETFS_SREQ_MADE_PROGRESS, &subreq->flags); __set_bit(NETFS_SREQ_IN_PROGRESS, &subreq->flags); - netfs_get_subrequest(subreq, netfs_sreq_trace_get_resubmit); + netfs_stat(&netfs_n_rh_retry_read_subreq); subreq->rreq->netfs_ops->issue_read(subreq); } @@ -48,6 +48,7 @@ static void netfs_retry_read_subrequests(struct netfs_io_request *rreq) __clear_bit(NETFS_SREQ_MADE_PROGRESS, &subreq->flags); subreq->retry_count++; netfs_reset_iter(subreq); + netfs_get_subrequest(subreq, netfs_sreq_trace_get_resubmit); netfs_reissue_read(rreq, subreq); } } @@ -75,7 +76,7 @@ static void netfs_retry_read_subrequests(struct netfs_io_request *rreq) struct iov_iter source; unsigned long long start, len; size_t part; - bool boundary = false; + bool boundary = false, subreq_superfluous = false; /* Go through the subreqs and find the next span of contiguous * buffer that we then rejig (cifs, for example, needs the @@ -116,8 +117,10 @@ static void netfs_retry_read_subrequests(struct netfs_io_request *rreq) /* Work through the sublist. */ subreq = from; list_for_each_entry_from(subreq, &stream->subrequests, rreq_link) { - if (!len) + if (!len) { + subreq_superfluous = true; break; + } subreq->source = NETFS_DOWNLOAD_FROM_SERVER; subreq->start = start - subreq->transferred; subreq->len = len + subreq->transferred; @@ -154,19 +157,21 @@ static void netfs_retry_read_subrequests(struct netfs_io_request *rreq) netfs_get_subrequest(subreq, netfs_sreq_trace_get_resubmit); netfs_reissue_read(rreq, subreq); - if (subreq == to) + if (subreq == to) { + subreq_superfluous = false; break; + } } /* If we managed to use fewer subreqs, we can discard the * excess; if we used the same number, then we're done. */ if (!len) { - if (subreq == to) + if (!subreq_superfluous) continue; list_for_each_entry_safe_from(subreq, tmp, &stream->subrequests, rreq_link) { - trace_netfs_sreq(subreq, netfs_sreq_trace_discard); + trace_netfs_sreq(subreq, netfs_sreq_trace_superfluous); list_del(&subreq->rreq_link); netfs_put_subrequest(subreq, false, netfs_sreq_trace_put_done); if (subreq == to) @@ -187,14 +192,12 @@ static void netfs_retry_read_subrequests(struct netfs_io_request *rreq) subreq->source = NETFS_DOWNLOAD_FROM_SERVER; subreq->start = start; subreq->len = len; - subreq->debug_index = atomic_inc_return(&rreq->subreq_counter); subreq->stream_nr = stream->stream_nr; subreq->retry_count = 1; trace_netfs_sreq_ref(rreq->debug_id, subreq->debug_index, refcount_read(&subreq->ref), netfs_sreq_trace_new); - netfs_get_subrequest(subreq, netfs_sreq_trace_get_resubmit); list_add(&subreq->rreq_link, &to->rreq_link); to = list_next_entry(to, rreq_link); @@ -256,14 +259,34 @@ void netfs_retry_reads(struct netfs_io_request *rreq) { struct netfs_io_subrequest *subreq; struct netfs_io_stream *stream = &rreq->io_streams[0]; + DEFINE_WAIT(myself); + + netfs_stat(&netfs_n_rh_retry_read_req); + + set_bit(NETFS_RREQ_RETRYING, &rreq->flags); /* Wait for all outstanding I/O to quiesce before performing retries as * we may need to renegotiate the I/O sizes. */ list_for_each_entry(subreq, &stream->subrequests, rreq_link) { - wait_on_bit(&subreq->flags, NETFS_SREQ_IN_PROGRESS, - TASK_UNINTERRUPTIBLE); + if (!test_bit(NETFS_SREQ_IN_PROGRESS, &subreq->flags)) + continue; + + trace_netfs_rreq(rreq, netfs_rreq_trace_wait_queue); + for (;;) { + prepare_to_wait(&rreq->waitq, &myself, TASK_UNINTERRUPTIBLE); + + if (!test_bit(NETFS_SREQ_IN_PROGRESS, &subreq->flags)) + break; + + trace_netfs_sreq(subreq, netfs_sreq_trace_wait_for); + schedule(); + trace_netfs_rreq(rreq, netfs_rreq_trace_woke_queue); + } + + finish_wait(&rreq->waitq, &myself); } + clear_bit(NETFS_RREQ_RETRYING, &rreq->flags); trace_netfs_rreq(rreq, netfs_rreq_trace_resubmit); netfs_retry_read_subrequests(rreq); diff --git a/fs/netfs/stats.c b/fs/netfs/stats.c index f1af344266cc..ab6b916addc4 100644 --- a/fs/netfs/stats.c +++ b/fs/netfs/stats.c @@ -29,6 +29,8 @@ atomic_t netfs_n_rh_write_begin; atomic_t netfs_n_rh_write_done; atomic_t netfs_n_rh_write_failed; atomic_t netfs_n_rh_write_zskip; +atomic_t netfs_n_rh_retry_read_req; +atomic_t netfs_n_rh_retry_read_subreq; atomic_t netfs_n_wh_buffered_write; atomic_t netfs_n_wh_writethrough; atomic_t netfs_n_wh_dio_write; @@ -41,6 +43,8 @@ atomic_t netfs_n_wh_upload_failed; atomic_t netfs_n_wh_write; atomic_t netfs_n_wh_write_done; atomic_t netfs_n_wh_write_failed; +atomic_t netfs_n_wh_retry_write_req; +atomic_t netfs_n_wh_retry_write_subreq; atomic_t netfs_n_wb_lock_skip; atomic_t netfs_n_wb_lock_wait; atomic_t netfs_n_folioq; @@ -81,6 +85,11 @@ int netfs_stats_show(struct seq_file *m, void *v) atomic_read(&netfs_n_wh_write), atomic_read(&netfs_n_wh_write_done), atomic_read(&netfs_n_wh_write_failed)); + seq_printf(m, "Retries: rq=%u rs=%u wq=%u ws=%u\n", + atomic_read(&netfs_n_rh_retry_read_req), + atomic_read(&netfs_n_rh_retry_read_subreq), + atomic_read(&netfs_n_wh_retry_write_req), + atomic_read(&netfs_n_wh_retry_write_subreq)); seq_printf(m, "Objs : rr=%u sr=%u foq=%u wsc=%u\n", atomic_read(&netfs_n_rh_rreq), atomic_read(&netfs_n_rh_sreq), diff --git a/fs/netfs/write_issue.c b/fs/netfs/write_issue.c index 69727411683e..77279fc5b5a7 100644 --- a/fs/netfs/write_issue.c +++ b/fs/netfs/write_issue.c @@ -253,6 +253,7 @@ void netfs_reissue_write(struct netfs_io_stream *stream, subreq->retry_count++; __clear_bit(NETFS_SREQ_MADE_PROGRESS, &subreq->flags); __set_bit(NETFS_SREQ_IN_PROGRESS, &subreq->flags); + netfs_stat(&netfs_n_wh_retry_write_subreq); netfs_do_issue_write(stream, subreq); } diff --git a/fs/netfs/write_retry.c b/fs/netfs/write_retry.c index c841a851dd73..545d33079a77 100644 --- a/fs/netfs/write_retry.c +++ b/fs/netfs/write_retry.c @@ -203,6 +203,8 @@ void netfs_retry_writes(struct netfs_io_request *wreq) struct netfs_io_stream *stream; int s; + netfs_stat(&netfs_n_wh_retry_write_req); + /* Wait for all outstanding I/O to quiesce before performing retries as * we may need to renegotiate the I/O sizes. */ diff --git a/fs/nfs/delegation.c b/fs/nfs/delegation.c index 035ba52742a5..4db912f56230 100644 --- a/fs/nfs/delegation.c +++ b/fs/nfs/delegation.c @@ -781,6 +781,43 @@ int nfs4_inode_return_delegation(struct inode *inode) } /** + * nfs4_inode_set_return_delegation_on_close - asynchronously return a delegation + * @inode: inode to process + * + * This routine is called to request that the delegation be returned as soon + * as the file is closed. If the file is already closed, the delegation is + * immediately returned. + */ +void nfs4_inode_set_return_delegation_on_close(struct inode *inode) +{ + struct nfs_delegation *delegation; + struct nfs_delegation *ret = NULL; + + if (!inode) + return; + rcu_read_lock(); + delegation = nfs4_get_valid_delegation(inode); + if (!delegation) + goto out; + spin_lock(&delegation->lock); + if (!delegation->inode) + goto out_unlock; + if (list_empty(&NFS_I(inode)->open_files) && + !test_and_set_bit(NFS_DELEGATION_RETURNING, &delegation->flags)) { + /* Refcount matched in nfs_end_delegation_return() */ + ret = nfs_get_delegation(delegation); + } else + set_bit(NFS_DELEGATION_RETURN_IF_CLOSED, &delegation->flags); +out_unlock: + spin_unlock(&delegation->lock); + if (ret) + nfs_clear_verifier_delegated(inode); +out: + rcu_read_unlock(); + nfs_end_delegation_return(inode, ret, 0); +} + +/** * nfs4_inode_return_delegation_on_close - asynchronously return a delegation * @inode: inode to process * diff --git a/fs/nfs/delegation.h b/fs/nfs/delegation.h index 71524d34ed20..8ff5ab9c5c25 100644 --- a/fs/nfs/delegation.h +++ b/fs/nfs/delegation.h @@ -49,6 +49,7 @@ void nfs_inode_reclaim_delegation(struct inode *inode, const struct cred *cred, unsigned long pagemod_limit, u32 deleg_type); int nfs4_inode_return_delegation(struct inode *inode); void nfs4_inode_return_delegation_on_close(struct inode *inode); +void nfs4_inode_set_return_delegation_on_close(struct inode *inode); int nfs_async_inode_return_delegation(struct inode *inode, const nfs4_stateid *stateid); void nfs_inode_evict_delegation(struct inode *inode); diff --git a/fs/nfs/direct.c b/fs/nfs/direct.c index f45beea92d03..f32f8d7c9122 100644 --- a/fs/nfs/direct.c +++ b/fs/nfs/direct.c @@ -56,6 +56,7 @@ #include <linux/uaccess.h> #include <linux/atomic.h> +#include "delegation.h" #include "internal.h" #include "iostat.h" #include "pnfs.h" @@ -130,6 +131,20 @@ static void nfs_direct_truncate_request(struct nfs_direct_req *dreq, dreq->count = req_start; } +static void nfs_direct_file_adjust_size_locked(struct inode *inode, + loff_t offset, size_t count) +{ + loff_t newsize = offset + (loff_t)count; + loff_t oldsize = i_size_read(inode); + + if (newsize > oldsize) { + i_size_write(inode, newsize); + NFS_I(inode)->cache_validity &= ~NFS_INO_INVALID_SIZE; + trace_nfs_size_grow(inode, newsize); + nfs_inc_stats(inode, NFSIOS_EXTENDWRITE); + } +} + /** * nfs_swap_rw - NFS address space operation for swap I/O * @iocb: target I/O control block @@ -272,6 +287,8 @@ static void nfs_direct_read_completion(struct nfs_pgio_header *hdr) nfs_direct_count_bytes(dreq, hdr); spin_unlock(&dreq->lock); + nfs_update_delegated_atime(dreq->inode); + while (!list_empty(&hdr->pages)) { struct nfs_page *req = nfs_list_entry(hdr->pages.next); struct page *page = req->wb_page; @@ -741,6 +758,7 @@ static void nfs_direct_write_completion(struct nfs_pgio_header *hdr) struct nfs_direct_req *dreq = hdr->dreq; struct nfs_commit_info cinfo; struct nfs_page *req = nfs_list_entry(hdr->pages.next); + struct inode *inode = dreq->inode; int flags = NFS_ODIRECT_DONE; trace_nfs_direct_write_completion(dreq); @@ -762,6 +780,11 @@ static void nfs_direct_write_completion(struct nfs_pgio_header *hdr) } spin_unlock(&dreq->lock); + spin_lock(&inode->i_lock); + nfs_direct_file_adjust_size_locked(inode, dreq->io_start, dreq->count); + nfs_update_delegated_mtime_locked(dreq->inode); + spin_unlock(&inode->i_lock); + while (!list_empty(&hdr->pages)) { req = nfs_list_entry(hdr->pages.next); diff --git a/fs/nfs/file.c b/fs/nfs/file.c index 1bb646752e46..033feeab8c34 100644 --- a/fs/nfs/file.c +++ b/fs/nfs/file.c @@ -29,6 +29,7 @@ #include <linux/pagemap.h> #include <linux/gfp.h> #include <linux/swap.h> +#include <linux/compaction.h> #include <linux/uaccess.h> #include <linux/filelock.h> @@ -457,7 +458,7 @@ static bool nfs_release_folio(struct folio *folio, gfp_t gfp) /* If the private flag is set, then the folio is not freeable */ if (folio_test_private(folio)) { if ((current_gfp_context(gfp) & GFP_KERNEL) != GFP_KERNEL || - current_is_kswapd()) + current_is_kswapd() || current_is_kcompactd()) return false; if (nfs_wb_folio(folio->mapping->host, folio) < 0) return false; diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index df9669d4ded7..6e95db6c17e9 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -133,6 +133,7 @@ nfs4_label_init_security(struct inode *dir, struct dentry *dentry, if (err) return NULL; + label->lsmid = shim.id; label->label = shim.context; label->len = shim.len; return label; @@ -145,7 +146,7 @@ nfs4_label_release_security(struct nfs4_label *label) if (label) { shim.context = label->label; shim.len = label->len; - shim.id = LSM_ID_UNDEF; + shim.id = label->lsmid; security_release_secctx(&shim); } } @@ -3906,8 +3907,11 @@ nfs4_atomic_open(struct inode *dir, struct nfs_open_context *ctx, static void nfs4_close_context(struct nfs_open_context *ctx, int is_sync) { + struct dentry *dentry = ctx->dentry; if (ctx->state == NULL) return; + if (dentry->d_flags & DCACHE_NFSFS_RENAMED) + nfs4_inode_set_return_delegation_on_close(d_inode(dentry)); if (is_sync) nfs4_close_sync(ctx->state, _nfs4_ctx_to_openmode(ctx)); else @@ -6269,7 +6273,7 @@ static int _nfs4_get_security_label(struct inode *inode, void *buf, size_t buflen) { struct nfs_server *server = NFS_SERVER(inode); - struct nfs4_label label = {0, 0, buflen, buf}; + struct nfs4_label label = {0, 0, 0, buflen, buf}; u32 bitmask[3] = { 0, 0, FATTR4_WORD2_SECURITY_LABEL }; struct nfs_fattr fattr = { @@ -6374,7 +6378,7 @@ static int nfs4_do_set_security_label(struct inode *inode, static int nfs4_set_security_label(struct inode *inode, const void *buf, size_t buflen) { - struct nfs4_label ilabel = {0, 0, buflen, (char *)buf }; + struct nfs4_label ilabel = {0, 0, 0, buflen, (char *)buf }; struct nfs_fattr *fattr; int status; diff --git a/fs/nfsd/filecache.c b/fs/nfsd/filecache.c index 0e552d873eaa..fb9b1656a287 100644 --- a/fs/nfsd/filecache.c +++ b/fs/nfsd/filecache.c @@ -446,11 +446,20 @@ nfsd_file_dispose_list_delayed(struct list_head *dispose) struct nfsd_file, nf_gc); struct nfsd_net *nn = net_generic(nf->nf_net, nfsd_net_id); struct nfsd_fcache_disposal *l = nn->fcache_disposal; + struct svc_serv *serv; spin_lock(&l->lock); list_move_tail(&nf->nf_gc, &l->freeme); spin_unlock(&l->lock); - svc_wake_up(nn->nfsd_serv); + + /* + * The filecache laundrette is shut down after the + * nn->nfsd_serv pointer is cleared, but before the + * svc_serv is freed. + */ + serv = nn->nfsd_serv; + if (serv) + svc_wake_up(serv); } } diff --git a/fs/nfsd/nfs2acl.c b/fs/nfsd/nfs2acl.c index 4e3be7201b1c..5fb202acb0fd 100644 --- a/fs/nfsd/nfs2acl.c +++ b/fs/nfsd/nfs2acl.c @@ -84,6 +84,8 @@ out: fail: posix_acl_release(resp->acl_access); posix_acl_release(resp->acl_default); + resp->acl_access = NULL; + resp->acl_default = NULL; goto out; } diff --git a/fs/nfsd/nfs3acl.c b/fs/nfsd/nfs3acl.c index 5e34e98db969..7b5433bd3019 100644 --- a/fs/nfsd/nfs3acl.c +++ b/fs/nfsd/nfs3acl.c @@ -76,6 +76,8 @@ out: fail: posix_acl_release(resp->acl_access); posix_acl_release(resp->acl_default); + resp->acl_access = NULL; + resp->acl_default = NULL; goto out; } diff --git a/fs/nfsd/nfs4callback.c b/fs/nfsd/nfs4callback.c index 50e468bdb8d4..484077200c5d 100644 --- a/fs/nfsd/nfs4callback.c +++ b/fs/nfsd/nfs4callback.c @@ -679,7 +679,7 @@ static int nfs4_xdr_dec_cb_getattr(struct rpc_rqst *rqstp, return status; status = decode_cb_op_status(xdr, OP_CB_GETATTR, &cb->cb_status); - if (unlikely(status || cb->cb_seq_status)) + if (unlikely(status || cb->cb_status)) return status; if (xdr_stream_decode_uint32_array(xdr, bitmap, 3) < 0) return -NFSERR_BAD_XDR; @@ -1583,8 +1583,11 @@ nfsd4_run_cb_work(struct work_struct *work) nfsd4_process_cb_update(cb); clnt = clp->cl_cb_client; - if (!clnt) { - /* Callback channel broken, or client killed; give up: */ + if (!clnt || clp->cl_state == NFSD4_COURTESY) { + /* + * Callback channel broken, client killed or + * nfs4_client in courtesy state; give up. + */ nfsd41_destroy_cb(cb); return; } diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index b7a0cfd05401..153eeea2c7c9 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -4459,10 +4459,11 @@ nfsd4_sequence(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, } } while (slot && --cnt > 0); } + +out: seq->maxslots = max(session->se_target_maxslots, seq->maxslots); seq->target_maxslots = session->se_target_maxslots; -out: switch (clp->cl_cb_state) { case NFSD4_CB_DOWN: seq->status_flags = SEQ4_STATUS_CB_PATH_DOWN; diff --git a/fs/nfsd/nfsfh.c b/fs/nfsd/nfsfh.c index 32019751a41e..aef474f1b84b 100644 --- a/fs/nfsd/nfsfh.c +++ b/fs/nfsd/nfsfh.c @@ -380,8 +380,9 @@ __fh_verify(struct svc_rqst *rqstp, error = check_nfsd_access(exp, rqstp, may_bypass_gss); if (error) goto out; - - svc_xprt_set_valid(rqstp->rq_xprt); + /* During LOCALIO call to fh_verify will be called with a NULL rqstp */ + if (rqstp) + svc_xprt_set_valid(rqstp->rq_xprt); /* Finally, check access permissions. */ error = nfsd_permission(cred, exp, dentry, access); diff --git a/fs/notify/fsnotify.c b/fs/notify/fsnotify.c index 8ee495a58d0a..fae1b6d397ea 100644 --- a/fs/notify/fsnotify.c +++ b/fs/notify/fsnotify.c @@ -648,7 +648,7 @@ EXPORT_SYMBOL_GPL(fsnotify); * Later, fsnotify permission hooks do not check if there are permission event * watches, but that there were permission event watches at open time. */ -void file_set_fsnotify_mode(struct file *file) +void file_set_fsnotify_mode_from_watchers(struct file *file) { struct dentry *dentry = file->f_path.dentry, *parent; struct super_block *sb = dentry->d_sb; @@ -665,7 +665,7 @@ void file_set_fsnotify_mode(struct file *file) */ if (likely(!fsnotify_sb_has_priority_watchers(sb, FSNOTIFY_PRIO_CONTENT))) { - file->f_mode |= FMODE_NONOTIFY_PERM; + file_set_fsnotify_mode(file, FMODE_NONOTIFY_PERM); return; } @@ -676,7 +676,7 @@ void file_set_fsnotify_mode(struct file *file) if ((!d_is_dir(dentry) && !d_is_reg(dentry)) || likely(!fsnotify_sb_has_priority_watchers(sb, FSNOTIFY_PRIO_PRE_CONTENT))) { - file->f_mode |= FMODE_NONOTIFY | FMODE_NONOTIFY_PERM; + file_set_fsnotify_mode(file, FMODE_NONOTIFY | FMODE_NONOTIFY_PERM); return; } @@ -686,19 +686,25 @@ void file_set_fsnotify_mode(struct file *file) */ mnt_mask = READ_ONCE(real_mount(file->f_path.mnt)->mnt_fsnotify_mask); if (unlikely(fsnotify_object_watched(d_inode(dentry), mnt_mask, - FSNOTIFY_PRE_CONTENT_EVENTS))) + FSNOTIFY_PRE_CONTENT_EVENTS))) { + /* Enable pre-content events */ + file_set_fsnotify_mode(file, 0); return; + } /* Is parent watching for pre-content events on this file? */ if (dentry->d_flags & DCACHE_FSNOTIFY_PARENT_WATCHED) { parent = dget_parent(dentry); p_mask = fsnotify_inode_watches_children(d_inode(parent)); dput(parent); - if (p_mask & FSNOTIFY_PRE_CONTENT_EVENTS) + if (p_mask & FSNOTIFY_PRE_CONTENT_EVENTS) { + /* Enable pre-content events */ + file_set_fsnotify_mode(file, 0); return; + } } /* Nobody watching for pre-content events from this file */ - file->f_mode |= FMODE_NONOTIFY | FMODE_NONOTIFY_PERM; + file_set_fsnotify_mode(file, FMODE_NONOTIFY | FMODE_NONOTIFY_PERM); } #endif diff --git a/fs/nsfs.c b/fs/nsfs.c index 663f8656158d..f7fddf8ecf73 100644 --- a/fs/nsfs.c +++ b/fs/nsfs.c @@ -37,7 +37,6 @@ static char *ns_dname(struct dentry *dentry, char *buffer, int buflen) } const struct dentry_operations ns_dentry_operations = { - .d_delete = always_delete_dentry, .d_dname = ns_dname, .d_prune = stashed_dentry_prune, }; diff --git a/fs/open.c b/fs/open.c index 932e5a6de63b..1be20de9f283 100644 --- a/fs/open.c +++ b/fs/open.c @@ -905,7 +905,8 @@ static int do_dentry_open(struct file *f, f->f_sb_err = file_sample_sb_err(f); if (unlikely(f->f_flags & O_PATH)) { - f->f_mode = FMODE_PATH | FMODE_OPENED | FMODE_NONOTIFY; + f->f_mode = FMODE_PATH | FMODE_OPENED; + file_set_fsnotify_mode(f, FMODE_NONOTIFY); f->f_op = &empty_fops; return 0; } @@ -935,10 +936,10 @@ static int do_dentry_open(struct file *f, /* * Set FMODE_NONOTIFY_* bits according to existing permission watches. - * If FMODE_NONOTIFY was already set for an fanotify fd, this doesn't - * change anything. + * If FMODE_NONOTIFY mode was already set for an fanotify fd or for a + * pseudo file, this call will not change the mode. */ - file_set_fsnotify_mode(f); + file_set_fsnotify_mode_from_watchers(f); error = fsnotify_open_perm(f); if (error) goto cleanup_all; @@ -1122,7 +1123,7 @@ struct file *dentry_open_nonotify(const struct path *path, int flags, if (!IS_ERR(f)) { int error; - f->f_mode |= FMODE_NONOTIFY; + file_set_fsnotify_mode(f, FMODE_NONOTIFY); error = vfs_open(path, f); if (error) { fput(f); diff --git a/fs/overlayfs/copy_up.c b/fs/overlayfs/copy_up.c index 0c28e5fa3407..d7310fcf3888 100644 --- a/fs/overlayfs/copy_up.c +++ b/fs/overlayfs/copy_up.c @@ -618,7 +618,6 @@ static int ovl_link_up(struct ovl_copy_up_ctx *c) err = PTR_ERR(upper); if (!IS_ERR(upper)) { err = ovl_do_link(ofs, ovl_dentry_upper(c->dentry), udir, upper); - dput(upper); if (!err) { /* Restore timestamps on parent (best effort) */ @@ -626,6 +625,7 @@ static int ovl_link_up(struct ovl_copy_up_ctx *c) ovl_dentry_set_upper_alias(c->dentry); ovl_dentry_update_reval(c->dentry, upper); } + dput(upper); } inode_unlock(udir); if (err) diff --git a/fs/pidfs.c b/fs/pidfs.c index 049352f973de..c0478b3c55d9 100644 --- a/fs/pidfs.c +++ b/fs/pidfs.c @@ -287,7 +287,6 @@ static bool pidfs_ioctl_valid(unsigned int cmd) switch (cmd) { case FS_IOC_GETVERSION: case PIDFD_GET_CGROUP_NAMESPACE: - case PIDFD_GET_INFO: case PIDFD_GET_IPC_NAMESPACE: case PIDFD_GET_MNT_NAMESPACE: case PIDFD_GET_NET_NAMESPACE: @@ -300,6 +299,17 @@ static bool pidfs_ioctl_valid(unsigned int cmd) return true; } + /* Extensible ioctls require some more careful checks. */ + switch (_IOC_NR(cmd)) { + case _IOC_NR(PIDFD_GET_INFO): + /* + * Try to prevent performing a pidfd ioctl when someone + * erronously mistook the file descriptor for a pidfd. + * This is not perfect but will catch most cases. + */ + return (_IOC_TYPE(cmd) == _IOC_TYPE(PIDFD_GET_INFO)); + } + return false; } @@ -511,7 +521,6 @@ static char *pidfs_dname(struct dentry *dentry, char *buffer, int buflen) } const struct dentry_operations pidfs_dentry_operations = { - .d_delete = always_delete_dentry, .d_dname = pidfs_dname, .d_prune = stashed_dentry_prune, }; diff --git a/fs/pipe.c b/fs/pipe.c index 94b59045ab44..4d0799e4e719 100644 --- a/fs/pipe.c +++ b/fs/pipe.c @@ -210,11 +210,10 @@ static const struct pipe_buf_operations anon_pipe_buf_ops = { /* Done while waiting without holding the pipe lock - thus the READ_ONCE() */ static inline bool pipe_readable(const struct pipe_inode_info *pipe) { - unsigned int head = READ_ONCE(pipe->head); - unsigned int tail = READ_ONCE(pipe->tail); + union pipe_index idx = { .head_tail = READ_ONCE(pipe->head_tail) }; unsigned int writers = READ_ONCE(pipe->writers); - return !pipe_empty(head, tail) || !writers; + return !pipe_empty(idx.head, idx.tail) || !writers; } static inline unsigned int pipe_update_tail(struct pipe_inode_info *pipe, @@ -395,7 +394,7 @@ pipe_read(struct kiocb *iocb, struct iov_iter *to) wake_next_reader = true; mutex_lock(&pipe->mutex); } - if (pipe_empty(pipe->head, pipe->tail)) + if (pipe_is_empty(pipe)) wake_next_reader = false; mutex_unlock(&pipe->mutex); @@ -417,11 +416,10 @@ static inline int is_packetized(struct file *file) /* Done while waiting without holding the pipe lock - thus the READ_ONCE() */ static inline bool pipe_writable(const struct pipe_inode_info *pipe) { - unsigned int head = READ_ONCE(pipe->head); - unsigned int tail = READ_ONCE(pipe->tail); + union pipe_index idx = { .head_tail = READ_ONCE(pipe->head_tail) }; unsigned int max_usage = READ_ONCE(pipe->max_usage); - return !pipe_full(head, tail, max_usage) || + return !pipe_full(idx.head, idx.tail, max_usage) || !READ_ONCE(pipe->readers); } @@ -579,11 +577,11 @@ pipe_write(struct kiocb *iocb, struct iov_iter *from) kill_fasync(&pipe->fasync_readers, SIGIO, POLL_IN); wait_event_interruptible_exclusive(pipe->wr_wait, pipe_writable(pipe)); mutex_lock(&pipe->mutex); - was_empty = pipe_empty(pipe->head, pipe->tail); + was_empty = pipe_is_empty(pipe); wake_next_writer = true; } out: - if (pipe_full(pipe->head, pipe->tail, pipe->max_usage)) + if (pipe_is_full(pipe)) wake_next_writer = false; mutex_unlock(&pipe->mutex); @@ -616,7 +614,7 @@ out: static long pipe_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) { struct pipe_inode_info *pipe = filp->private_data; - unsigned int count, head, tail, mask; + unsigned int count, head, tail; switch (cmd) { case FIONREAD: @@ -624,10 +622,9 @@ static long pipe_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) count = 0; head = pipe->head; tail = pipe->tail; - mask = pipe->ring_size - 1; - while (tail != head) { - count += pipe->bufs[tail & mask].len; + while (!pipe_empty(head, tail)) { + count += pipe_buf(pipe, tail)->len; tail++; } mutex_unlock(&pipe->mutex); @@ -659,7 +656,7 @@ pipe_poll(struct file *filp, poll_table *wait) { __poll_t mask; struct pipe_inode_info *pipe = filp->private_data; - unsigned int head, tail; + union pipe_index idx; /* Epoll has some historical nasty semantics, this enables them */ WRITE_ONCE(pipe->poll_usage, true); @@ -680,19 +677,18 @@ pipe_poll(struct file *filp, poll_table *wait) * if something changes and you got it wrong, the poll * table entry will wake you up and fix it. */ - head = READ_ONCE(pipe->head); - tail = READ_ONCE(pipe->tail); + idx.head_tail = READ_ONCE(pipe->head_tail); mask = 0; if (filp->f_mode & FMODE_READ) { - if (!pipe_empty(head, tail)) + if (!pipe_empty(idx.head, idx.tail)) mask |= EPOLLIN | EPOLLRDNORM; if (!pipe->writers && filp->f_pipe != pipe->w_counter) mask |= EPOLLHUP; } if (filp->f_mode & FMODE_WRITE) { - if (!pipe_full(head, tail, pipe->max_usage)) + if (!pipe_full(idx.head, idx.tail, pipe->max_usage)) mask |= EPOLLOUT | EPOLLWRNORM; /* * Most Unices do not set EPOLLERR for FIFOs but on Linux they @@ -960,6 +956,12 @@ int create_pipe_files(struct file **res, int flags) res[1] = f; stream_open(inode, res[0]); stream_open(inode, res[1]); + /* + * Disable permission and pre-content events, but enable legacy + * inotify events for legacy users. + */ + file_set_fsnotify_mode(res[0], FMODE_NONOTIFY_PERM); + file_set_fsnotify_mode(res[1], FMODE_NONOTIFY_PERM); return 0; } diff --git a/fs/proc/vmcore.c b/fs/proc/vmcore.c index a00120a3c099..10d01eb09c43 100644 --- a/fs/proc/vmcore.c +++ b/fs/proc/vmcore.c @@ -1524,7 +1524,7 @@ int vmcore_add_device_dump(struct vmcoredd_data *data) pr_warn_once("Unexpected adding of device dump\n"); if (vmcore_open) { ret = -EBUSY; - goto out_err; + goto unlock; } list_add_tail(&dump->list, &vmcoredd_list); @@ -1532,6 +1532,9 @@ int vmcore_add_device_dump(struct vmcoredd_data *data) mutex_unlock(&vmcore_mutex); return 0; +unlock: + mutex_unlock(&vmcore_mutex); + out_err: vfree(buf); vfree(dump); diff --git a/fs/smb/client/cifsacl.c b/fs/smb/client/cifsacl.c index 699a3f76d083..64bd68f750f8 100644 --- a/fs/smb/client/cifsacl.c +++ b/fs/smb/client/cifsacl.c @@ -763,7 +763,7 @@ static void parse_dacl(struct smb_acl *pdacl, char *end_of_acl, struct cifs_fattr *fattr, bool mode_from_special_sid) { int i; - int num_aces = 0; + u16 num_aces = 0; int acl_size; char *acl_base; struct smb_ace **ppace; @@ -778,14 +778,15 @@ static void parse_dacl(struct smb_acl *pdacl, char *end_of_acl, } /* validate that we do not go past end of acl */ - if (end_of_acl < (char *)pdacl + le16_to_cpu(pdacl->size)) { + if (end_of_acl < (char *)pdacl + sizeof(struct smb_acl) || + end_of_acl < (char *)pdacl + le16_to_cpu(pdacl->size)) { cifs_dbg(VFS, "ACL too small to parse DACL\n"); return; } cifs_dbg(NOISY, "DACL revision %d size %d num aces %d\n", le16_to_cpu(pdacl->revision), le16_to_cpu(pdacl->size), - le32_to_cpu(pdacl->num_aces)); + le16_to_cpu(pdacl->num_aces)); /* reset rwx permissions for user/group/other. Also, if num_aces is 0 i.e. DACL has no ACEs, @@ -795,12 +796,15 @@ static void parse_dacl(struct smb_acl *pdacl, char *end_of_acl, acl_base = (char *)pdacl; acl_size = sizeof(struct smb_acl); - num_aces = le32_to_cpu(pdacl->num_aces); + num_aces = le16_to_cpu(pdacl->num_aces); if (num_aces > 0) { umode_t denied_mode = 0; - if (num_aces > ULONG_MAX / sizeof(struct smb_ace *)) + if (num_aces > (le16_to_cpu(pdacl->size) - sizeof(struct smb_acl)) / + (offsetof(struct smb_ace, sid) + + offsetof(struct smb_sid, sub_auth) + sizeof(__le16))) return; + ppace = kmalloc_array(num_aces, sizeof(struct smb_ace *), GFP_KERNEL); if (!ppace) @@ -937,12 +941,12 @@ unsigned int setup_special_user_owner_ACE(struct smb_ace *pntace) static void populate_new_aces(char *nacl_base, struct smb_sid *pownersid, struct smb_sid *pgrpsid, - __u64 *pnmode, u32 *pnum_aces, u16 *pnsize, + __u64 *pnmode, u16 *pnum_aces, u16 *pnsize, bool modefromsid, bool posix) { __u64 nmode; - u32 num_aces = 0; + u16 num_aces = 0; u16 nsize = 0; __u64 user_mode; __u64 group_mode; @@ -1050,7 +1054,7 @@ static __u16 replace_sids_and_copy_aces(struct smb_acl *pdacl, struct smb_acl *p u16 size = 0; struct smb_ace *pntace = NULL; char *acl_base = NULL; - u32 src_num_aces = 0; + u16 src_num_aces = 0; u16 nsize = 0; struct smb_ace *pnntace = NULL; char *nacl_base = NULL; @@ -1058,7 +1062,7 @@ static __u16 replace_sids_and_copy_aces(struct smb_acl *pdacl, struct smb_acl *p acl_base = (char *)pdacl; size = sizeof(struct smb_acl); - src_num_aces = le32_to_cpu(pdacl->num_aces); + src_num_aces = le16_to_cpu(pdacl->num_aces); nacl_base = (char *)pndacl; nsize = sizeof(struct smb_acl); @@ -1090,11 +1094,11 @@ static int set_chmod_dacl(struct smb_acl *pdacl, struct smb_acl *pndacl, u16 size = 0; struct smb_ace *pntace = NULL; char *acl_base = NULL; - u32 src_num_aces = 0; + u16 src_num_aces = 0; u16 nsize = 0; struct smb_ace *pnntace = NULL; char *nacl_base = NULL; - u32 num_aces = 0; + u16 num_aces = 0; bool new_aces_set = false; /* Assuming that pndacl and pnmode are never NULL */ @@ -1112,7 +1116,7 @@ static int set_chmod_dacl(struct smb_acl *pdacl, struct smb_acl *pndacl, acl_base = (char *)pdacl; size = sizeof(struct smb_acl); - src_num_aces = le32_to_cpu(pdacl->num_aces); + src_num_aces = le16_to_cpu(pdacl->num_aces); /* Retain old ACEs which we can retain */ for (i = 0; i < src_num_aces; ++i) { @@ -1158,7 +1162,7 @@ next_ace: } finalize_dacl: - pndacl->num_aces = cpu_to_le32(num_aces); + pndacl->num_aces = cpu_to_le16(num_aces); pndacl->size = cpu_to_le16(nsize); return 0; @@ -1293,7 +1297,7 @@ static int build_sec_desc(struct smb_ntsd *pntsd, struct smb_ntsd *pnntsd, dacloffset ? dacl_ptr->revision : cpu_to_le16(ACL_REVISION); ndacl_ptr->size = cpu_to_le16(0); - ndacl_ptr->num_aces = cpu_to_le32(0); + ndacl_ptr->num_aces = cpu_to_le16(0); rc = set_chmod_dacl(dacl_ptr, ndacl_ptr, owner_sid_ptr, group_sid_ptr, pnmode, mode_from_sid, posix); @@ -1653,7 +1657,7 @@ id_mode_to_cifs_acl(struct inode *inode, const char *path, __u64 *pnmode, dacl_ptr = (struct smb_acl *)((char *)pntsd + dacloffset); if (mode_from_sid) nsecdesclen += - le32_to_cpu(dacl_ptr->num_aces) * sizeof(struct smb_ace); + le16_to_cpu(dacl_ptr->num_aces) * sizeof(struct smb_ace); else /* cifsacl */ nsecdesclen += le16_to_cpu(dacl_ptr->size); } diff --git a/fs/smb/client/cifsglob.h b/fs/smb/client/cifsglob.h index a68434ad744a..cddeb2adbf4a 100644 --- a/fs/smb/client/cifsglob.h +++ b/fs/smb/client/cifsglob.h @@ -253,6 +253,7 @@ struct cifs_cred { struct cifs_open_info_data { bool adjust_tz; bool reparse_point; + bool contains_posix_file_info; struct { /* ioctl response buffer */ struct { @@ -357,7 +358,7 @@ struct smb_version_operations { int (*handle_cancelled_mid)(struct mid_q_entry *, struct TCP_Server_Info *); void (*downgrade_oplock)(struct TCP_Server_Info *server, struct cifsInodeInfo *cinode, __u32 oplock, - unsigned int epoch, bool *purge_cache); + __u16 epoch, bool *purge_cache); /* process transaction2 response */ bool (*check_trans2)(struct mid_q_entry *, struct TCP_Server_Info *, char *, int); @@ -552,12 +553,12 @@ struct smb_version_operations { /* if we can do cache read operations */ bool (*is_read_op)(__u32); /* set oplock level for the inode */ - void (*set_oplock_level)(struct cifsInodeInfo *, __u32, unsigned int, - bool *); + void (*set_oplock_level)(struct cifsInodeInfo *cinode, __u32 oplock, __u16 epoch, + bool *purge_cache); /* create lease context buffer for CREATE request */ char * (*create_lease_buf)(u8 *lease_key, u8 oplock); /* parse lease context buffer and return oplock/epoch info */ - __u8 (*parse_lease_buf)(void *buf, unsigned int *epoch, char *lkey); + __u8 (*parse_lease_buf)(void *buf, __u16 *epoch, char *lkey); ssize_t (*copychunk_range)(const unsigned int, struct cifsFileInfo *src_file, struct cifsFileInfo *target_file, @@ -1447,7 +1448,7 @@ struct cifs_fid { __u8 create_guid[16]; __u32 access; struct cifs_pending_open *pending_open; - unsigned int epoch; + __u16 epoch; #ifdef CONFIG_CIFS_DEBUG2 __u64 mid; #endif /* CIFS_DEBUG2 */ @@ -1480,7 +1481,7 @@ struct cifsFileInfo { bool oplock_break_cancelled:1; bool status_file_deleted:1; /* file has been deleted */ bool offload:1; /* offload final part of _put to a wq */ - unsigned int oplock_epoch; /* epoch from the lease break */ + __u16 oplock_epoch; /* epoch from the lease break */ __u32 oplock_level; /* oplock/lease level from the lease break */ int count; spinlock_t file_info_lock; /* protects four flag/count fields above */ @@ -1508,7 +1509,6 @@ struct cifs_io_parms { struct cifs_io_request { struct netfs_io_request rreq; struct cifsFileInfo *cfile; - struct TCP_Server_Info *server; pid_t pid; }; @@ -1577,7 +1577,7 @@ struct cifsInodeInfo { spinlock_t open_file_lock; /* protects openFileList */ __u32 cifsAttrs; /* e.g. DOS archive bit, sparse, compressed, system */ unsigned int oplock; /* oplock/lease level we have */ - unsigned int epoch; /* used to track lease state changes */ + __u16 epoch; /* used to track lease state changes */ #define CIFS_INODE_PENDING_OPLOCK_BREAK (0) /* oplock break in progress */ #define CIFS_INODE_PENDING_WRITERS (1) /* Writes in progress */ #define CIFS_INODE_FLAG_UNUSED (2) /* Unused flag */ @@ -2325,8 +2325,8 @@ struct smb2_compound_vars { struct kvec io_iov[SMB2_IOCTL_IOV_SIZE]; struct kvec si_iov[SMB2_SET_INFO_IOV_SIZE]; struct kvec close_iov; - struct smb2_file_rename_info rename_info; - struct smb2_file_link_info link_info; + struct smb2_file_rename_info_hdr rename_info; + struct smb2_file_link_info_hdr link_info; struct kvec ea_iov; }; diff --git a/fs/smb/client/cifssmb.c b/fs/smb/client/cifssmb.c index 3feaa0f68169..d07682020c64 100644 --- a/fs/smb/client/cifssmb.c +++ b/fs/smb/client/cifssmb.c @@ -1338,7 +1338,8 @@ cifs_readv_callback(struct mid_q_entry *mid) rdata->credits.value = 0; rdata->subreq.error = rdata->result; rdata->subreq.transferred += rdata->got_bytes; - queue_work(cifsiod_wq, &rdata->subreq.work); + trace_netfs_sreq(&rdata->subreq, netfs_sreq_trace_io_progress); + netfs_read_subreq_terminated(&rdata->subreq); release_mid(mid); add_credits(server, &credits, 0); } diff --git a/fs/smb/client/dfs.c b/fs/smb/client/dfs.c index dad521336b5e..f65a8a90ba27 100644 --- a/fs/smb/client/dfs.c +++ b/fs/smb/client/dfs.c @@ -150,25 +150,27 @@ again: if (rc) continue; - if (tgt.flags & DFSREF_STORAGE_SERVER) { - rc = cifs_mount_get_tcon(mnt_ctx); - if (!rc) - rc = cifs_is_path_remote(mnt_ctx); + rc = cifs_mount_get_tcon(mnt_ctx); + if (rc) { + if (tgt.server_type == DFS_TYPE_LINK && + DFS_INTERLINK(tgt.flags)) + rc = -EREMOTE; + } else { + rc = cifs_is_path_remote(mnt_ctx); if (!rc) { ref_walk_set_tgt_hint(rw); break; } - if (rc != -EREMOTE) - continue; } - - rc = ref_walk_advance(rw); - if (!rc) { - rc = setup_dfs_ref(&tgt, rw); - if (rc) - break; - ref_walk_mark_end(rw); - goto again; + if (rc == -EREMOTE) { + rc = ref_walk_advance(rw); + if (!rc) { + rc = setup_dfs_ref(&tgt, rw); + if (rc) + break; + ref_walk_mark_end(rw); + goto again; + } } } } while (rc && ref_walk_descend(rw)); diff --git a/fs/smb/client/dfs.h b/fs/smb/client/dfs.h index ed4cd7cf1ec6..e60f0a24a8a1 100644 --- a/fs/smb/client/dfs.h +++ b/fs/smb/client/dfs.h @@ -188,4 +188,11 @@ static inline void dfs_put_root_smb_sessions(struct list_head *head) } } +static inline const char *dfs_ses_refpath(struct cifs_ses *ses) +{ + const char *path = ses->server->leaf_fullpath; + + return path ? path + 1 : ERR_PTR(-ENOENT); +} + #endif /* _CIFS_DFS_H */ diff --git a/fs/smb/client/dfs_cache.c b/fs/smb/client/dfs_cache.c index 5022bb1f122a..4dada26d56b5 100644 --- a/fs/smb/client/dfs_cache.c +++ b/fs/smb/client/dfs_cache.c @@ -1136,33 +1136,19 @@ static bool is_ses_good(struct cifs_ses *ses) return ret; } -static char *get_ses_refpath(struct cifs_ses *ses) -{ - struct TCP_Server_Info *server = ses->server; - char *path = ERR_PTR(-ENOENT); - - if (server->leaf_fullpath) { - path = kstrdup(server->leaf_fullpath + 1, GFP_KERNEL); - if (!path) - path = ERR_PTR(-ENOMEM); - } - return path; -} - /* Refresh dfs referral of @ses */ static void refresh_ses_referral(struct cifs_ses *ses) { struct cache_entry *ce; unsigned int xid; - char *path; + const char *path; int rc = 0; xid = get_xid(); - path = get_ses_refpath(ses); + path = dfs_ses_refpath(ses); if (IS_ERR(path)) { rc = PTR_ERR(path); - path = NULL; goto out; } @@ -1181,7 +1167,6 @@ static void refresh_ses_referral(struct cifs_ses *ses) out: free_xid(xid); - kfree(path); } static int __refresh_tcon_referral(struct cifs_tcon *tcon, @@ -1231,19 +1216,18 @@ static void refresh_tcon_referral(struct cifs_tcon *tcon, bool force_refresh) struct dfs_info3_param *refs = NULL; struct cache_entry *ce; struct cifs_ses *ses; - unsigned int xid; bool needs_refresh; - char *path; + const char *path; + unsigned int xid; int numrefs = 0; int rc = 0; xid = get_xid(); ses = tcon->ses; - path = get_ses_refpath(ses); + path = dfs_ses_refpath(ses); if (IS_ERR(path)) { rc = PTR_ERR(path); - path = NULL; goto out; } @@ -1271,7 +1255,6 @@ static void refresh_tcon_referral(struct cifs_tcon *tcon, bool force_refresh) out: free_xid(xid); - kfree(path); free_dfs_info_array(refs, numrefs); } diff --git a/fs/smb/client/file.c b/fs/smb/client/file.c index 79de2f2f9c41..8582cf61242c 100644 --- a/fs/smb/client/file.c +++ b/fs/smb/client/file.c @@ -147,7 +147,7 @@ static int cifs_prepare_read(struct netfs_io_subrequest *subreq) struct netfs_io_request *rreq = subreq->rreq; struct cifs_io_subrequest *rdata = container_of(subreq, struct cifs_io_subrequest, subreq); struct cifs_io_request *req = container_of(subreq->rreq, struct cifs_io_request, rreq); - struct TCP_Server_Info *server = req->server; + struct TCP_Server_Info *server; struct cifs_sb_info *cifs_sb = CIFS_SB(rreq->inode->i_sb); size_t size; int rc = 0; @@ -156,6 +156,8 @@ static int cifs_prepare_read(struct netfs_io_subrequest *subreq) rdata->xid = get_xid(); rdata->have_xid = true; } + + server = cifs_pick_channel(tlink_tcon(req->cfile->tlink)->ses); rdata->server = server; if (cifs_sb->ctx->rsize == 0) @@ -198,7 +200,7 @@ static void cifs_issue_read(struct netfs_io_subrequest *subreq) struct netfs_io_request *rreq = subreq->rreq; struct cifs_io_subrequest *rdata = container_of(subreq, struct cifs_io_subrequest, subreq); struct cifs_io_request *req = container_of(subreq->rreq, struct cifs_io_request, rreq); - struct TCP_Server_Info *server = req->server; + struct TCP_Server_Info *server = rdata->server; int rc = 0; cifs_dbg(FYI, "%s: op=%08x[%x] mapping=%p len=%zu/%zu\n", @@ -266,7 +268,6 @@ static int cifs_init_request(struct netfs_io_request *rreq, struct file *file) open_file = file->private_data; rreq->netfs_priv = file->private_data; req->cfile = cifsFileInfo_get(open_file); - req->server = cifs_pick_channel(tlink_tcon(req->cfile->tlink)->ses); if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_RWPIDFORWARD) req->pid = req->cfile->pid; } else if (rreq->origin != NETFS_WRITEBACK) { diff --git a/fs/smb/client/inode.c b/fs/smb/client/inode.c index 9cc31cf6ebd0..616149c7f0a5 100644 --- a/fs/smb/client/inode.c +++ b/fs/smb/client/inode.c @@ -1215,6 +1215,19 @@ static int reparse_info_to_fattr(struct cifs_open_info_data *data, rc = server->ops->parse_reparse_point(cifs_sb, full_path, iov, data); + /* + * If the reparse point was not handled but it is the + * name surrogate which points to directory, then treat + * is as a new mount point. Name surrogate reparse point + * represents another named entity in the system. + */ + if (rc == -EOPNOTSUPP && + IS_REPARSE_TAG_NAME_SURROGATE(data->reparse.tag) && + (le32_to_cpu(data->fi.Attributes) & ATTR_DIRECTORY)) { + rc = 0; + cifs_create_junction_fattr(fattr, sb); + goto out; + } } if (data->reparse.tag == IO_REPARSE_TAG_SYMLINK && !rc) { @@ -1408,7 +1421,7 @@ int cifs_get_inode_info(struct inode **inode, struct cifs_fattr fattr = {}; int rc; - if (is_inode_cache_good(*inode)) { + if (!data && is_inode_cache_good(*inode)) { cifs_dbg(FYI, "No need to revalidate cached inode sizes\n"); return 0; } @@ -1507,7 +1520,7 @@ int smb311_posix_get_inode_info(struct inode **inode, struct cifs_fattr fattr = {}; int rc; - if (is_inode_cache_good(*inode)) { + if (!data && is_inode_cache_good(*inode)) { cifs_dbg(FYI, "No need to revalidate cached inode sizes\n"); return 0; } diff --git a/fs/smb/client/reparse.c b/fs/smb/client/reparse.c index 0a5a52a8a7dd..2b9e9885dc42 100644 --- a/fs/smb/client/reparse.c +++ b/fs/smb/client/reparse.c @@ -1088,13 +1088,12 @@ int parse_reparse_point(struct reparse_data_buffer *buf, le32_to_cpu(buf->ReparseTag)); return -EIO; } - break; + return 0; default: cifs_tcon_dbg(VFS | ONCE, "unhandled reparse tag: 0x%08x\n", le32_to_cpu(buf->ReparseTag)); - break; + return -EOPNOTSUPP; } - return 0; } int smb2_parse_reparse_point(struct cifs_sb_info *cifs_sb, diff --git a/fs/smb/client/reparse.h b/fs/smb/client/reparse.h index 5a753fec7e2c..c0be5ab45a78 100644 --- a/fs/smb/client/reparse.h +++ b/fs/smb/client/reparse.h @@ -99,14 +99,30 @@ static inline bool reparse_inode_match(struct inode *inode, static inline bool cifs_open_data_reparse(struct cifs_open_info_data *data) { - struct smb2_file_all_info *fi = &data->fi; - u32 attrs = le32_to_cpu(fi->Attributes); + u32 attrs; bool ret; - ret = data->reparse_point || (attrs & ATTR_REPARSE); - if (ret) - attrs |= ATTR_REPARSE; - fi->Attributes = cpu_to_le32(attrs); + if (data->contains_posix_file_info) { + struct smb311_posix_qinfo *fi = &data->posix_fi; + + attrs = le32_to_cpu(fi->DosAttributes); + if (data->reparse_point) { + attrs |= ATTR_REPARSE; + fi->DosAttributes = cpu_to_le32(attrs); + } + + } else { + struct smb2_file_all_info *fi = &data->fi; + + attrs = le32_to_cpu(fi->Attributes); + if (data->reparse_point) { + attrs |= ATTR_REPARSE; + fi->Attributes = cpu_to_le32(attrs); + } + } + + ret = attrs & ATTR_REPARSE; + return ret; } diff --git a/fs/smb/client/smb1ops.c b/fs/smb/client/smb1ops.c index 9756b876a75e..d6e2fb669c40 100644 --- a/fs/smb/client/smb1ops.c +++ b/fs/smb/client/smb1ops.c @@ -377,7 +377,7 @@ coalesce_t2(char *second_buf, struct smb_hdr *target_hdr) static void cifs_downgrade_oplock(struct TCP_Server_Info *server, struct cifsInodeInfo *cinode, __u32 oplock, - unsigned int epoch, bool *purge_cache) + __u16 epoch, bool *purge_cache) { cifs_set_oplock_level(cinode, oplock); } diff --git a/fs/smb/client/smb2inode.c b/fs/smb/client/smb2inode.c index 5dfb30b0a852..826b57a5a2a8 100644 --- a/fs/smb/client/smb2inode.c +++ b/fs/smb/client/smb2inode.c @@ -650,6 +650,7 @@ finished: switch (cmds[i]) { case SMB2_OP_QUERY_INFO: idata = in_iov[i].iov_base; + idata->contains_posix_file_info = false; if (rc == 0 && cfile && cfile->symlink_target) { idata->symlink_target = kstrdup(cfile->symlink_target, GFP_KERNEL); if (!idata->symlink_target) @@ -673,6 +674,7 @@ finished: break; case SMB2_OP_POSIX_QUERY_INFO: idata = in_iov[i].iov_base; + idata->contains_posix_file_info = true; if (rc == 0 && cfile && cfile->symlink_target) { idata->symlink_target = kstrdup(cfile->symlink_target, GFP_KERNEL); if (!idata->symlink_target) @@ -770,6 +772,7 @@ finished: idata = in_iov[i].iov_base; idata->reparse.io.iov = *iov; idata->reparse.io.buftype = resp_buftype[i + 1]; + idata->contains_posix_file_info = false; /* BB VERIFY */ rbuf = reparse_buf_ptr(iov); if (IS_ERR(rbuf)) { rc = PTR_ERR(rbuf); @@ -791,6 +794,7 @@ finished: case SMB2_OP_QUERY_WSL_EA: if (!rc) { idata = in_iov[i].iov_base; + idata->contains_posix_file_info = false; qi_rsp = rsp_iov[i + 1].iov_base; data[0] = (u8 *)qi_rsp + le16_to_cpu(qi_rsp->OutputBufferOffset); size[0] = le32_to_cpu(qi_rsp->OutputBufferLength); diff --git a/fs/smb/client/smb2ops.c b/fs/smb/client/smb2ops.c index 77309217dab4..4dd11eafb69d 100644 --- a/fs/smb/client/smb2ops.c +++ b/fs/smb/client/smb2ops.c @@ -1001,6 +1001,7 @@ static int smb2_query_file_info(const unsigned int xid, struct cifs_tcon *tcon, if (!data->symlink_target) return -ENOMEM; } + data->contains_posix_file_info = false; return SMB2_query_info(xid, tcon, fid->persistent_fid, fid->volatile_fid, &data->fi); } @@ -3904,22 +3905,22 @@ static long smb3_fallocate(struct file *file, struct cifs_tcon *tcon, int mode, static void smb2_downgrade_oplock(struct TCP_Server_Info *server, struct cifsInodeInfo *cinode, __u32 oplock, - unsigned int epoch, bool *purge_cache) + __u16 epoch, bool *purge_cache) { server->ops->set_oplock_level(cinode, oplock, 0, NULL); } static void smb21_set_oplock_level(struct cifsInodeInfo *cinode, __u32 oplock, - unsigned int epoch, bool *purge_cache); + __u16 epoch, bool *purge_cache); static void smb3_downgrade_oplock(struct TCP_Server_Info *server, struct cifsInodeInfo *cinode, __u32 oplock, - unsigned int epoch, bool *purge_cache) + __u16 epoch, bool *purge_cache) { unsigned int old_state = cinode->oplock; - unsigned int old_epoch = cinode->epoch; + __u16 old_epoch = cinode->epoch; unsigned int new_state; if (epoch > old_epoch) { @@ -3939,7 +3940,7 @@ smb3_downgrade_oplock(struct TCP_Server_Info *server, static void smb2_set_oplock_level(struct cifsInodeInfo *cinode, __u32 oplock, - unsigned int epoch, bool *purge_cache) + __u16 epoch, bool *purge_cache) { oplock &= 0xFF; cinode->lease_granted = false; @@ -3963,7 +3964,7 @@ smb2_set_oplock_level(struct cifsInodeInfo *cinode, __u32 oplock, static void smb21_set_oplock_level(struct cifsInodeInfo *cinode, __u32 oplock, - unsigned int epoch, bool *purge_cache) + __u16 epoch, bool *purge_cache) { char message[5] = {0}; unsigned int new_oplock = 0; @@ -4000,7 +4001,7 @@ smb21_set_oplock_level(struct cifsInodeInfo *cinode, __u32 oplock, static void smb3_set_oplock_level(struct cifsInodeInfo *cinode, __u32 oplock, - unsigned int epoch, bool *purge_cache) + __u16 epoch, bool *purge_cache) { unsigned int old_oplock = cinode->oplock; @@ -4114,7 +4115,7 @@ smb3_create_lease_buf(u8 *lease_key, u8 oplock) } static __u8 -smb2_parse_lease_buf(void *buf, unsigned int *epoch, char *lease_key) +smb2_parse_lease_buf(void *buf, __u16 *epoch, char *lease_key) { struct create_lease *lc = (struct create_lease *)buf; @@ -4125,7 +4126,7 @@ smb2_parse_lease_buf(void *buf, unsigned int *epoch, char *lease_key) } static __u8 -smb3_parse_lease_buf(void *buf, unsigned int *epoch, char *lease_key) +smb3_parse_lease_buf(void *buf, __u16 *epoch, char *lease_key) { struct create_lease_v2 *lc = (struct create_lease_v2 *)buf; @@ -4964,6 +4965,10 @@ one_more: next_buffer = (char *)cifs_buf_get(); else next_buffer = (char *)cifs_small_buf_get(); + if (!next_buffer) { + cifs_server_dbg(VFS, "No memory for (large) SMB response\n"); + return -1; + } memcpy(next_buffer, buf + next_cmd, pdu_length - next_cmd); } @@ -5146,7 +5151,7 @@ int __cifs_sfu_make_node(unsigned int xid, struct inode *inode, FILE_CREATE, CREATE_NOT_DIR | CREATE_OPTION_SPECIAL, ACL_NO_MODE); oparms.fid = &fid; - + idata.contains_posix_file_info = false; rc = server->ops->open(xid, &oparms, &oplock, &idata); if (rc) goto out; diff --git a/fs/smb/client/smb2pdu.c b/fs/smb/client/smb2pdu.c index 40ad9e79437a..ed7812247ebc 100644 --- a/fs/smb/client/smb2pdu.c +++ b/fs/smb/client/smb2pdu.c @@ -2169,7 +2169,7 @@ tcon_exit: tcon_error_exit: if (rsp && rsp->hdr.Status == STATUS_BAD_NETWORK_NAME) - cifs_tcon_dbg(VFS, "BAD_NETWORK_NAME: %s\n", tree); + cifs_dbg(VFS | ONCE, "BAD_NETWORK_NAME: %s\n", tree); goto tcon_exit; } @@ -2329,7 +2329,7 @@ parse_posix_ctxt(struct create_context *cc, struct smb2_file_all_info *info, int smb2_parse_contexts(struct TCP_Server_Info *server, struct kvec *rsp_iov, - unsigned int *epoch, + __u16 *epoch, char *lease_key, __u8 *oplock, struct smb2_file_all_info *buf, struct create_posix_rsp *posix) diff --git a/fs/smb/client/smb2proto.h b/fs/smb/client/smb2proto.h index 2336dfb23f36..4662c7e2d259 100644 --- a/fs/smb/client/smb2proto.h +++ b/fs/smb/client/smb2proto.h @@ -283,7 +283,7 @@ extern enum securityEnum smb2_select_sectype(struct TCP_Server_Info *, enum securityEnum); int smb2_parse_contexts(struct TCP_Server_Info *server, struct kvec *rsp_iov, - unsigned int *epoch, + __u16 *epoch, char *lease_key, __u8 *oplock, struct smb2_file_all_info *buf, struct create_posix_rsp *posix); diff --git a/fs/smb/common/smb2pdu.h b/fs/smb/common/smb2pdu.h index 3336df2ea5d4..c7a0efda4403 100644 --- a/fs/smb/common/smb2pdu.h +++ b/fs/smb/common/smb2pdu.h @@ -1707,23 +1707,33 @@ struct smb2_file_internal_info { } __packed; /* level 6 Query */ struct smb2_file_rename_info { /* encoding of request for level 10 */ - __u8 ReplaceIfExists; /* 1 = replace existing target with new */ - /* 0 = fail if target already exists */ - __u8 Reserved[7]; - __u64 RootDirectory; /* MBZ for network operations (why says spec?) */ - __le32 FileNameLength; + /* New members MUST be added within the struct_group() macro below. */ + __struct_group(smb2_file_rename_info_hdr, __hdr, __packed, + __u8 ReplaceIfExists; /* 1 = replace existing target with new */ + /* 0 = fail if target already exists */ + __u8 Reserved[7]; + __u64 RootDirectory; /* MBZ for network operations (why says spec?) */ + __le32 FileNameLength; + ); char FileName[]; /* New name to be assigned */ /* padding - overall struct size must be >= 24 so filename + pad >= 6 */ } __packed; /* level 10 Set */ +static_assert(offsetof(struct smb2_file_rename_info, FileName) == sizeof(struct smb2_file_rename_info_hdr), + "struct member likely outside of __struct_group()"); struct smb2_file_link_info { /* encoding of request for level 11 */ - __u8 ReplaceIfExists; /* 1 = replace existing link with new */ - /* 0 = fail if link already exists */ - __u8 Reserved[7]; - __u64 RootDirectory; /* MBZ for network operations (why says spec?) */ - __le32 FileNameLength; + /* New members MUST be added within the struct_group() macro below. */ + __struct_group(smb2_file_link_info_hdr, __hdr, __packed, + __u8 ReplaceIfExists; /* 1 = replace existing link with new */ + /* 0 = fail if link already exists */ + __u8 Reserved[7]; + __u64 RootDirectory; /* MBZ for network operations (why says spec?) */ + __le32 FileNameLength; + ); char FileName[]; /* Name to be assigned to new link */ } __packed; /* level 11 Set */ +static_assert(offsetof(struct smb2_file_link_info, FileName) == sizeof(struct smb2_file_link_info_hdr), + "struct member likely outside of __struct_group()"); /* * This level 18, although with struct with same name is different from cifs diff --git a/fs/smb/common/smbacl.h b/fs/smb/common/smbacl.h index 6a60698fc6f0..a624ec9e4a14 100644 --- a/fs/smb/common/smbacl.h +++ b/fs/smb/common/smbacl.h @@ -107,7 +107,8 @@ struct smb_sid { struct smb_acl { __le16 revision; /* revision level */ __le16 size; - __le32 num_aces; + __le16 num_aces; + __le16 reserved; } __attribute__((packed)); struct smb_ace { diff --git a/fs/smb/common/smbfsctl.h b/fs/smb/common/smbfsctl.h index 4b379e84c46b..3253a18ecb5c 100644 --- a/fs/smb/common/smbfsctl.h +++ b/fs/smb/common/smbfsctl.h @@ -159,6 +159,9 @@ #define IO_REPARSE_TAG_LX_CHR 0x80000025 #define IO_REPARSE_TAG_LX_BLK 0x80000026 +/* If Name Surrogate Bit is set, the file or directory represents another named entity in the system. */ +#define IS_REPARSE_TAG_NAME_SURROGATE(tag) (!!((tag) & 0x20000000)) + /* fsctl flags */ /* If Flags is set to this value, the request is an FSCTL not ioctl request */ #define SMB2_0_IOCTL_IS_FSCTL 0x00000001 diff --git a/fs/smb/server/smb2pdu.c b/fs/smb/server/smb2pdu.c index f1efcd027475..c53121538990 100644 --- a/fs/smb/server/smb2pdu.c +++ b/fs/smb/server/smb2pdu.c @@ -7458,17 +7458,17 @@ out_check_cl: } no_check_cl: + flock = smb_lock->fl; + list_del(&smb_lock->llist); + if (smb_lock->zero_len) { err = 0; goto skip; } - - flock = smb_lock->fl; - list_del(&smb_lock->llist); retry: rc = vfs_lock_file(filp, smb_lock->cmd, flock, NULL); skip: - if (flags & SMB2_LOCKFLAG_UNLOCK) { + if (smb_lock->flags & SMB2_LOCKFLAG_UNLOCK) { if (!rc) { ksmbd_debug(SMB, "File unlocked\n"); } else if (rc == -ENOENT) { diff --git a/fs/smb/server/smbacl.c b/fs/smb/server/smbacl.c index d39d3e553366..49b128698670 100644 --- a/fs/smb/server/smbacl.c +++ b/fs/smb/server/smbacl.c @@ -333,7 +333,7 @@ void posix_state_to_acl(struct posix_acl_state *state, pace->e_perm = state->other.allow; } -int init_acl_state(struct posix_acl_state *state, int cnt) +int init_acl_state(struct posix_acl_state *state, u16 cnt) { int alloc; @@ -368,7 +368,7 @@ static void parse_dacl(struct mnt_idmap *idmap, struct smb_fattr *fattr) { int i, ret; - int num_aces = 0; + u16 num_aces = 0; unsigned int acl_size; char *acl_base; struct smb_ace **ppace; @@ -389,16 +389,18 @@ static void parse_dacl(struct mnt_idmap *idmap, ksmbd_debug(SMB, "DACL revision %d size %d num aces %d\n", le16_to_cpu(pdacl->revision), le16_to_cpu(pdacl->size), - le32_to_cpu(pdacl->num_aces)); + le16_to_cpu(pdacl->num_aces)); acl_base = (char *)pdacl; acl_size = sizeof(struct smb_acl); - num_aces = le32_to_cpu(pdacl->num_aces); + num_aces = le16_to_cpu(pdacl->num_aces); if (num_aces <= 0) return; - if (num_aces > ULONG_MAX / sizeof(struct smb_ace *)) + if (num_aces > (le16_to_cpu(pdacl->size) - sizeof(struct smb_acl)) / + (offsetof(struct smb_ace, sid) + + offsetof(struct smb_sid, sub_auth) + sizeof(__le16))) return; ret = init_acl_state(&acl_state, num_aces); @@ -432,6 +434,7 @@ static void parse_dacl(struct mnt_idmap *idmap, offsetof(struct smb_sid, sub_auth); if (end_of_acl - acl_base < acl_size || + ppace[i]->sid.num_subauth == 0 || ppace[i]->sid.num_subauth > SID_MAX_SUB_AUTHORITIES || (end_of_acl - acl_base < acl_size + sizeof(__le32) * ppace[i]->sid.num_subauth) || @@ -580,7 +583,7 @@ static void parse_dacl(struct mnt_idmap *idmap, static void set_posix_acl_entries_dacl(struct mnt_idmap *idmap, struct smb_ace *pndace, - struct smb_fattr *fattr, u32 *num_aces, + struct smb_fattr *fattr, u16 *num_aces, u16 *size, u32 nt_aces_num) { struct posix_acl_entry *pace; @@ -701,7 +704,7 @@ static void set_ntacl_dacl(struct mnt_idmap *idmap, struct smb_fattr *fattr) { struct smb_ace *ntace, *pndace; - int nt_num_aces = le32_to_cpu(nt_dacl->num_aces), num_aces = 0; + u16 nt_num_aces = le16_to_cpu(nt_dacl->num_aces), num_aces = 0; unsigned short size = 0; int i; @@ -728,7 +731,7 @@ static void set_ntacl_dacl(struct mnt_idmap *idmap, set_posix_acl_entries_dacl(idmap, pndace, fattr, &num_aces, &size, nt_num_aces); - pndacl->num_aces = cpu_to_le32(num_aces); + pndacl->num_aces = cpu_to_le16(num_aces); pndacl->size = cpu_to_le16(le16_to_cpu(pndacl->size) + size); } @@ -736,7 +739,7 @@ static void set_mode_dacl(struct mnt_idmap *idmap, struct smb_acl *pndacl, struct smb_fattr *fattr) { struct smb_ace *pace, *pndace; - u32 num_aces = 0; + u16 num_aces = 0; u16 size = 0, ace_size = 0; uid_t uid; const struct smb_sid *sid; @@ -792,7 +795,7 @@ static void set_mode_dacl(struct mnt_idmap *idmap, fattr->cf_mode, 0007); out: - pndacl->num_aces = cpu_to_le32(num_aces); + pndacl->num_aces = cpu_to_le16(num_aces); pndacl->size = cpu_to_le16(le16_to_cpu(pndacl->size) + size); } @@ -807,6 +810,13 @@ static int parse_sid(struct smb_sid *psid, char *end_of_acl) return -EINVAL; } + if (!psid->num_subauth) + return 0; + + if (psid->num_subauth > SID_MAX_SUB_AUTHORITIES || + end_of_acl < (char *)psid + 8 + sizeof(__le32) * psid->num_subauth) + return -EINVAL; + return 0; } @@ -848,6 +858,9 @@ int parse_sec_desc(struct mnt_idmap *idmap, struct smb_ntsd *pntsd, pntsd->type = cpu_to_le16(DACL_PRESENT); if (pntsd->osidoffset) { + if (le32_to_cpu(pntsd->osidoffset) < sizeof(struct smb_ntsd)) + return -EINVAL; + rc = parse_sid(owner_sid_ptr, end_of_acl); if (rc) { pr_err("%s: Error %d parsing Owner SID\n", __func__, rc); @@ -863,6 +876,9 @@ int parse_sec_desc(struct mnt_idmap *idmap, struct smb_ntsd *pntsd, } if (pntsd->gsidoffset) { + if (le32_to_cpu(pntsd->gsidoffset) < sizeof(struct smb_ntsd)) + return -EINVAL; + rc = parse_sid(group_sid_ptr, end_of_acl); if (rc) { pr_err("%s: Error %d mapping Owner SID to gid\n", @@ -884,6 +900,9 @@ int parse_sec_desc(struct mnt_idmap *idmap, struct smb_ntsd *pntsd, pntsd->type |= cpu_to_le16(DACL_PROTECTED); if (dacloffset) { + if (dacloffset < sizeof(struct smb_ntsd)) + return -EINVAL; + parse_dacl(idmap, dacl_ptr, end_of_acl, owner_sid_ptr, group_sid_ptr, fattr); } @@ -1006,8 +1025,9 @@ int smb_inherit_dacl(struct ksmbd_conn *conn, struct smb_sid owner_sid, group_sid; struct dentry *parent = path->dentry->d_parent; struct mnt_idmap *idmap = mnt_idmap(path->mnt); - int inherited_flags = 0, flags = 0, i, ace_cnt = 0, nt_size = 0, pdacl_size; - int rc = 0, num_aces, dacloffset, pntsd_type, pntsd_size, acl_len, aces_size; + int inherited_flags = 0, flags = 0, i, nt_size = 0, pdacl_size; + int rc = 0, dacloffset, pntsd_type, pntsd_size, acl_len, aces_size; + u16 num_aces, ace_cnt = 0; char *aces_base; bool is_dir = S_ISDIR(d_inode(path->dentry)->i_mode); @@ -1023,7 +1043,7 @@ int smb_inherit_dacl(struct ksmbd_conn *conn, parent_pdacl = (struct smb_acl *)((char *)parent_pntsd + dacloffset); acl_len = pntsd_size - dacloffset; - num_aces = le32_to_cpu(parent_pdacl->num_aces); + num_aces = le16_to_cpu(parent_pdacl->num_aces); pntsd_type = le16_to_cpu(parent_pntsd->type); pdacl_size = le16_to_cpu(parent_pdacl->size); @@ -1183,7 +1203,7 @@ pass: pdacl = (struct smb_acl *)((char *)pntsd + le32_to_cpu(pntsd->dacloffset)); pdacl->revision = cpu_to_le16(2); pdacl->size = cpu_to_le16(sizeof(struct smb_acl) + nt_size); - pdacl->num_aces = cpu_to_le32(ace_cnt); + pdacl->num_aces = cpu_to_le16(ace_cnt); pace = (struct smb_ace *)((char *)pdacl + sizeof(struct smb_acl)); memcpy(pace, aces_base, nt_size); pntsd_size += sizeof(struct smb_acl) + nt_size; @@ -1264,7 +1284,7 @@ int smb_check_perm_dacl(struct ksmbd_conn *conn, const struct path *path, ace = (struct smb_ace *)((char *)pdacl + sizeof(struct smb_acl)); aces_size = acl_size - sizeof(struct smb_acl); - for (i = 0; i < le32_to_cpu(pdacl->num_aces); i++) { + for (i = 0; i < le16_to_cpu(pdacl->num_aces); i++) { if (offsetof(struct smb_ace, access_req) > aces_size) break; ace_size = le16_to_cpu(ace->size); @@ -1285,7 +1305,7 @@ int smb_check_perm_dacl(struct ksmbd_conn *conn, const struct path *path, ace = (struct smb_ace *)((char *)pdacl + sizeof(struct smb_acl)); aces_size = acl_size - sizeof(struct smb_acl); - for (i = 0; i < le32_to_cpu(pdacl->num_aces); i++) { + for (i = 0; i < le16_to_cpu(pdacl->num_aces); i++) { if (offsetof(struct smb_ace, access_req) > aces_size) break; ace_size = le16_to_cpu(ace->size); diff --git a/fs/smb/server/smbacl.h b/fs/smb/server/smbacl.h index 24ce576fc292..355adaee39b8 100644 --- a/fs/smb/server/smbacl.h +++ b/fs/smb/server/smbacl.h @@ -86,7 +86,7 @@ int parse_sec_desc(struct mnt_idmap *idmap, struct smb_ntsd *pntsd, int build_sec_desc(struct mnt_idmap *idmap, struct smb_ntsd *pntsd, struct smb_ntsd *ppntsd, int ppntsd_size, int addition_info, __u32 *secdesclen, struct smb_fattr *fattr); -int init_acl_state(struct posix_acl_state *state, int cnt); +int init_acl_state(struct posix_acl_state *state, u16 cnt); void free_acl_state(struct posix_acl_state *state); void posix_state_to_acl(struct posix_acl_state *state, struct posix_acl_entry *pace); diff --git a/fs/smb/server/transport_ipc.c b/fs/smb/server/transport_ipc.c index 0460ebea6ff0..3f185ae60dc5 100644 --- a/fs/smb/server/transport_ipc.c +++ b/fs/smb/server/transport_ipc.c @@ -281,6 +281,7 @@ static int handle_response(int type, void *payload, size_t sz) if (entry->type + 1 != type) { pr_err("Waiting for IPC type %d, got %d. Ignore.\n", entry->type + 1, type); + continue; } entry->response = kvzalloc(sz, KSMBD_DEFAULT_GFP); diff --git a/fs/splice.c b/fs/splice.c index 28cfa63aa236..23fa5561b944 100644 --- a/fs/splice.c +++ b/fs/splice.c @@ -331,7 +331,7 @@ ssize_t copy_splice_read(struct file *in, loff_t *ppos, int i; /* Work out how much data we can actually add into the pipe */ - used = pipe_occupancy(pipe->head, pipe->tail); + used = pipe_buf_usage(pipe); npages = max_t(ssize_t, pipe->max_usage - used, 0); len = min_t(size_t, len, npages * PAGE_SIZE); npages = DIV_ROUND_UP(len, PAGE_SIZE); @@ -527,7 +527,7 @@ static int splice_from_pipe_next(struct pipe_inode_info *pipe, struct splice_des return -ERESTARTSYS; repeat: - while (pipe_empty(pipe->head, pipe->tail)) { + while (pipe_is_empty(pipe)) { if (!pipe->writers) return 0; @@ -820,7 +820,7 @@ ssize_t splice_to_socket(struct pipe_inode_info *pipe, struct file *out, if (signal_pending(current)) break; - while (pipe_empty(pipe->head, pipe->tail)) { + while (pipe_is_empty(pipe)) { ret = 0; if (!pipe->writers) goto out; @@ -968,7 +968,7 @@ static ssize_t do_splice_read(struct file *in, loff_t *ppos, return 0; /* Don't try to read more the pipe has space for. */ - p_space = pipe->max_usage - pipe_occupancy(pipe->head, pipe->tail); + p_space = pipe->max_usage - pipe_buf_usage(pipe); len = min_t(size_t, len, p_space << PAGE_SHIFT); if (unlikely(len > MAX_RW_COUNT)) @@ -1080,7 +1080,7 @@ ssize_t splice_direct_to_actor(struct file *in, struct splice_desc *sd, more = sd->flags & SPLICE_F_MORE; sd->flags |= SPLICE_F_MORE; - WARN_ON_ONCE(!pipe_empty(pipe->head, pipe->tail)); + WARN_ON_ONCE(!pipe_is_empty(pipe)); while (len) { size_t read_len; @@ -1268,7 +1268,7 @@ static int wait_for_space(struct pipe_inode_info *pipe, unsigned flags) send_sig(SIGPIPE, current, 0); return -EPIPE; } - if (!pipe_full(pipe->head, pipe->tail, pipe->max_usage)) + if (!pipe_is_full(pipe)) return 0; if (flags & SPLICE_F_NONBLOCK) return -EAGAIN; @@ -1652,13 +1652,13 @@ static int ipipe_prep(struct pipe_inode_info *pipe, unsigned int flags) * Check the pipe occupancy without the inode lock first. This function * is speculative anyways, so missing one is ok. */ - if (!pipe_empty(pipe->head, pipe->tail)) + if (!pipe_is_empty(pipe)) return 0; ret = 0; pipe_lock(pipe); - while (pipe_empty(pipe->head, pipe->tail)) { + while (pipe_is_empty(pipe)) { if (signal_pending(current)) { ret = -ERESTARTSYS; break; @@ -1688,13 +1688,13 @@ static int opipe_prep(struct pipe_inode_info *pipe, unsigned int flags) * Check pipe occupancy without the inode lock first. This function * is speculative anyways, so missing one is ok. */ - if (!pipe_full(pipe->head, pipe->tail, pipe->max_usage)) + if (!pipe_is_full(pipe)) return 0; ret = 0; pipe_lock(pipe); - while (pipe_full(pipe->head, pipe->tail, pipe->max_usage)) { + while (pipe_is_full(pipe)) { if (!pipe->readers) { send_sig(SIGPIPE, current, 0); ret = -EPIPE; diff --git a/fs/stat.c b/fs/stat.c index 2c0e111a098a..f13308bfdc98 100644 --- a/fs/stat.c +++ b/fs/stat.c @@ -281,6 +281,8 @@ static int vfs_statx_path(struct path *path, int flags, struct kstat *stat, u32 request_mask) { int error = vfs_getattr(path, stat, request_mask, flags); + if (error) + return error; if (request_mask & STATX_MNT_ID_UNIQUE) { stat->mnt_id = real_mount(path->mnt)->mnt_id_unique; @@ -302,7 +304,7 @@ static int vfs_statx_path(struct path *path, int flags, struct kstat *stat, if (S_ISBLK(stat->mode)) bdev_statx(path, stat, request_mask); - return error; + return 0; } static int vfs_statx_fd(int fd, int flags, struct kstat *stat, diff --git a/fs/vboxsf/super.c b/fs/vboxsf/super.c index e95b8a48d8a0..1d94bb784108 100644 --- a/fs/vboxsf/super.c +++ b/fs/vboxsf/super.c @@ -21,7 +21,8 @@ #define VBOXSF_SUPER_MAGIC 0x786f4256 /* 'VBox' little endian */ -static const unsigned char VBSF_MOUNT_SIGNATURE[4] = "\000\377\376\375"; +static const unsigned char VBSF_MOUNT_SIGNATURE[4] = { '\000', '\377', '\376', + '\375' }; static int follow_symlinks; module_param(follow_symlinks, int, 0444); diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c index 40ad22fb808b..0ef19f1469ec 100644 --- a/fs/xfs/libxfs/xfs_bmap.c +++ b/fs/xfs/libxfs/xfs_bmap.c @@ -3563,12 +3563,12 @@ xfs_bmap_btalloc_at_eof( int error; /* - * If there are already extents in the file, try an exact EOF block - * allocation to extend the file as a contiguous extent. If that fails, - * or it's the first allocation in a file, just try for a stripe aligned - * allocation. + * If there are already extents in the file, and xfs_bmap_adjacent() has + * given a better blkno, try an exact EOF block allocation to extend the + * file as a contiguous extent. If that fails, or it's the first + * allocation in a file, just try for a stripe aligned allocation. */ - if (ap->offset) { + if (ap->eof) { xfs_extlen_t nextminlen = 0; /* @@ -3736,7 +3736,8 @@ xfs_bmap_btalloc_best_length( int error; ap->blkno = XFS_INO_TO_FSB(args->mp, ap->ip->i_ino); - xfs_bmap_adjacent(ap); + if (!xfs_bmap_adjacent(ap)) + ap->eof = false; /* * Search for an allocation group with a single extent large enough for diff --git a/fs/xfs/scrub/common.h b/fs/xfs/scrub/common.h index bdcd40f0ec74..19877d99f255 100644 --- a/fs/xfs/scrub/common.h +++ b/fs/xfs/scrub/common.h @@ -224,7 +224,6 @@ static inline bool xchk_skip_xref(struct xfs_scrub_metadata *sm) bool xchk_dir_looks_zapped(struct xfs_inode *dp); bool xchk_pptr_looks_zapped(struct xfs_inode *ip); -#ifdef CONFIG_XFS_ONLINE_REPAIR /* Decide if a repair is required. */ static inline bool xchk_needs_repair(const struct xfs_scrub_metadata *sm) { @@ -244,10 +243,6 @@ static inline bool xchk_could_repair(const struct xfs_scrub *sc) return (sc->sm->sm_flags & XFS_SCRUB_IFLAG_REPAIR) && !(sc->flags & XREP_ALREADY_FIXED); } -#else -# define xchk_needs_repair(sc) (false) -# define xchk_could_repair(sc) (false) -#endif /* CONFIG_XFS_ONLINE_REPAIR */ int xchk_metadata_inode_forks(struct xfs_scrub *sc); diff --git a/fs/xfs/scrub/inode_repair.c b/fs/xfs/scrub/inode_repair.c index 2f641b6d663e..13ff1c933cb8 100644 --- a/fs/xfs/scrub/inode_repair.c +++ b/fs/xfs/scrub/inode_repair.c @@ -1055,9 +1055,17 @@ xrep_dinode_check_dfork( return true; break; case S_IFREG: - if (fmt == XFS_DINODE_FMT_LOCAL) + switch (fmt) { + case XFS_DINODE_FMT_LOCAL: return true; - fallthrough; + case XFS_DINODE_FMT_EXTENTS: + case XFS_DINODE_FMT_BTREE: + case XFS_DINODE_FMT_META_BTREE: + break; + default: + return true; + } + break; case S_IFLNK: case S_IFDIR: switch (fmt) { diff --git a/fs/xfs/scrub/repair.h b/fs/xfs/scrub/repair.h index 823c00d1a502..af0a3a9e5ed9 100644 --- a/fs/xfs/scrub/repair.h +++ b/fs/xfs/scrub/repair.h @@ -191,7 +191,16 @@ int xrep_reset_metafile_resv(struct xfs_scrub *sc); #else #define xrep_ino_dqattach(sc) (0) -#define xrep_will_attempt(sc) (false) + +/* + * When online repair is not built into the kernel, we still want to attempt + * the repair so that the stub xrep_attempt below will return EOPNOTSUPP. + */ +static inline bool xrep_will_attempt(const struct xfs_scrub *sc) +{ + return (sc->sm->sm_flags & XFS_SCRUB_IFLAG_FORCE_REBUILD) || + xchk_needs_repair(sc->sm); +} static inline int xrep_attempt( diff --git a/fs/xfs/scrub/scrub.c b/fs/xfs/scrub/scrub.c index 7567dd5cad14..6fa9e3e5bab7 100644 --- a/fs/xfs/scrub/scrub.c +++ b/fs/xfs/scrub/scrub.c @@ -149,6 +149,18 @@ xchk_probe( if (xchk_should_terminate(sc, &error)) return error; + /* + * If the caller is probing to see if repair works but repair isn't + * built into the kernel, return EOPNOTSUPP because that's the signal + * that userspace expects. If online repair is built in, set the + * CORRUPT flag (without any of the usual tracing/logging) to force us + * into xrep_probe. + */ + if (xchk_could_repair(sc)) { + if (!IS_ENABLED(CONFIG_XFS_ONLINE_REPAIR)) + return -EOPNOTSUPP; + sc->sm->sm_flags |= XFS_SCRUB_OFLAG_CORRUPT; + } return 0; } diff --git a/fs/xfs/xfs_aops.c b/fs/xfs/xfs_aops.c index 67877c36ed11..6d9965b546cb 100644 --- a/fs/xfs/xfs_aops.c +++ b/fs/xfs/xfs_aops.c @@ -19,6 +19,7 @@ #include "xfs_reflink.h" #include "xfs_errortag.h" #include "xfs_error.h" +#include "xfs_icache.h" struct xfs_writepage_ctx { struct iomap_writepage_ctx ctx; @@ -528,12 +529,44 @@ xfs_vm_readahead( } static int -xfs_iomap_swapfile_activate( +xfs_vm_swap_activate( struct swap_info_struct *sis, struct file *swap_file, sector_t *span) { - sis->bdev = xfs_inode_buftarg(XFS_I(file_inode(swap_file)))->bt_bdev; + struct xfs_inode *ip = XFS_I(file_inode(swap_file)); + + /* + * Swap file activation can race against concurrent shared extent + * removal in files that have been cloned. If this happens, + * iomap_swapfile_iter() can fail because it encountered a shared + * extent even though an operation is in progress to remove those + * shared extents. + * + * This race becomes problematic when we defer extent removal + * operations beyond the end of a syscall (i.e. use async background + * processing algorithms). Users think the extents are no longer + * shared, but iomap_swapfile_iter() still sees them as shared + * because the refcountbt entries for the extents being removed have + * not yet been updated. Hence the swapon call fails unexpectedly. + * + * The race condition is currently most obvious from the unlink() + * operation as extent removal is deferred until after the last + * reference to the inode goes away. We then process the extent + * removal asynchronously, hence triggers the "syscall completed but + * work not done" condition mentioned above. To close this race + * window, we need to flush any pending inodegc operations to ensure + * they have updated the refcountbt records before we try to map the + * swapfile. + */ + xfs_inodegc_flush(ip->i_mount); + + /* + * Direct the swap code to the correct block device when this file + * sits on the RT device. + */ + sis->bdev = xfs_inode_buftarg(ip)->bt_bdev; + return iomap_swapfile_activate(sis, swap_file, span, &xfs_read_iomap_ops); } @@ -549,11 +582,11 @@ const struct address_space_operations xfs_address_space_operations = { .migrate_folio = filemap_migrate_folio, .is_partially_uptodate = iomap_is_partially_uptodate, .error_remove_folio = generic_error_remove_folio, - .swap_activate = xfs_iomap_swapfile_activate, + .swap_activate = xfs_vm_swap_activate, }; const struct address_space_operations xfs_dax_aops = { .writepages = xfs_dax_writepages, .dirty_folio = noop_dirty_folio, - .swap_activate = xfs_iomap_swapfile_activate, + .swap_activate = xfs_vm_swap_activate, }; diff --git a/fs/xfs/xfs_buf.c b/fs/xfs/xfs_buf.c index d1d4a0a22e13..5d560e9073f4 100644 --- a/fs/xfs/xfs_buf.c +++ b/fs/xfs/xfs_buf.c @@ -29,11 +29,6 @@ struct kmem_cache *xfs_buf_cache; /* * Locking orders * - * xfs_buf_ioacct_inc: - * xfs_buf_ioacct_dec: - * b_sema (caller holds) - * b_lock - * * xfs_buf_stale: * b_sema (caller holds) * b_lock @@ -41,8 +36,7 @@ struct kmem_cache *xfs_buf_cache; * * xfs_buf_rele: * b_lock - * pag_buf_lock - * lru_lock + * lru_lock * * xfs_buftarg_drain_rele * lru_lock @@ -83,51 +77,6 @@ xfs_buf_vmap_len( } /* - * Bump the I/O in flight count on the buftarg if we haven't yet done so for - * this buffer. The count is incremented once per buffer (per hold cycle) - * because the corresponding decrement is deferred to buffer release. Buffers - * can undergo I/O multiple times in a hold-release cycle and per buffer I/O - * tracking adds unnecessary overhead. This is used for sychronization purposes - * with unmount (see xfs_buftarg_drain()), so all we really need is a count of - * in-flight buffers. - * - * Buffers that are never released (e.g., superblock, iclog buffers) must set - * the XBF_NO_IOACCT flag before I/O submission. Otherwise, the buftarg count - * never reaches zero and unmount hangs indefinitely. - */ -static inline void -xfs_buf_ioacct_inc( - struct xfs_buf *bp) -{ - if (bp->b_flags & XBF_NO_IOACCT) - return; - - ASSERT(bp->b_flags & XBF_ASYNC); - spin_lock(&bp->b_lock); - if (!(bp->b_state & XFS_BSTATE_IN_FLIGHT)) { - bp->b_state |= XFS_BSTATE_IN_FLIGHT; - percpu_counter_inc(&bp->b_target->bt_io_count); - } - spin_unlock(&bp->b_lock); -} - -/* - * Clear the in-flight state on a buffer about to be released to the LRU or - * freed and unaccount from the buftarg. - */ -static inline void -__xfs_buf_ioacct_dec( - struct xfs_buf *bp) -{ - lockdep_assert_held(&bp->b_lock); - - if (bp->b_state & XFS_BSTATE_IN_FLIGHT) { - bp->b_state &= ~XFS_BSTATE_IN_FLIGHT; - percpu_counter_dec(&bp->b_target->bt_io_count); - } -} - -/* * When we mark a buffer stale, we remove the buffer from the LRU and clear the * b_lru_ref count so that the buffer is freed immediately when the buffer * reference count falls to zero. If the buffer is already on the LRU, we need @@ -150,15 +99,7 @@ xfs_buf_stale( */ bp->b_flags &= ~_XBF_DELWRI_Q; - /* - * Once the buffer is marked stale and unlocked, a subsequent lookup - * could reset b_flags. There is no guarantee that the buffer is - * unaccounted (released to LRU) before that occurs. Drop in-flight - * status now to preserve accounting consistency. - */ spin_lock(&bp->b_lock); - __xfs_buf_ioacct_dec(bp); - atomic_set(&bp->b_lru_ref, 0); if (!(bp->b_state & XFS_BSTATE_DISPOSE) && (list_lru_del_obj(&bp->b_target->bt_lru, &bp->b_lru))) @@ -220,23 +161,25 @@ _xfs_buf_alloc( */ flags &= ~(XBF_UNMAPPED | XBF_TRYLOCK | XBF_ASYNC | XBF_READ_AHEAD); - spin_lock_init(&bp->b_lock); + /* + * A new buffer is held and locked by the owner. This ensures that the + * buffer is owned by the caller and racing RCU lookups right after + * inserting into the hash table are safe (and will have to wait for + * the unlock to do anything non-trivial). + */ bp->b_hold = 1; + sema_init(&bp->b_sema, 0); /* held, no waiters */ + + spin_lock_init(&bp->b_lock); atomic_set(&bp->b_lru_ref, 1); init_completion(&bp->b_iowait); INIT_LIST_HEAD(&bp->b_lru); INIT_LIST_HEAD(&bp->b_list); INIT_LIST_HEAD(&bp->b_li_list); - sema_init(&bp->b_sema, 0); /* held, no waiters */ bp->b_target = target; bp->b_mount = target->bt_mount; bp->b_flags = flags; - /* - * Set length and io_length to the same value initially. - * I/O routines should use io_length, which will be the same in - * most cases but may be reset (e.g. XFS recovery). - */ error = xfs_buf_get_maps(bp, nmaps); if (error) { kmem_cache_free(xfs_buf_cache, bp); @@ -502,7 +445,6 @@ int xfs_buf_cache_init( struct xfs_buf_cache *bch) { - spin_lock_init(&bch->bc_lock); return rhashtable_init(&bch->bc_hash, &xfs_buf_hash_params); } @@ -652,17 +594,20 @@ xfs_buf_find_insert( if (error) goto out_free_buf; - spin_lock(&bch->bc_lock); + /* The new buffer keeps the perag reference until it is freed. */ + new_bp->b_pag = pag; + + rcu_read_lock(); bp = rhashtable_lookup_get_insert_fast(&bch->bc_hash, &new_bp->b_rhash_head, xfs_buf_hash_params); if (IS_ERR(bp)) { + rcu_read_unlock(); error = PTR_ERR(bp); - spin_unlock(&bch->bc_lock); goto out_free_buf; } if (bp && xfs_buf_try_hold(bp)) { /* found an existing buffer */ - spin_unlock(&bch->bc_lock); + rcu_read_unlock(); error = xfs_buf_find_lock(bp, flags); if (error) xfs_buf_rele(bp); @@ -670,10 +615,8 @@ xfs_buf_find_insert( *bpp = bp; goto out_free_buf; } + rcu_read_unlock(); - /* The new buffer keeps the perag reference until it is freed. */ - new_bp->b_pag = pag; - spin_unlock(&bch->bc_lock); *bpp = new_bp; return 0; @@ -793,18 +736,13 @@ out_put_perag: int _xfs_buf_read( - struct xfs_buf *bp, - xfs_buf_flags_t flags) + struct xfs_buf *bp) { - ASSERT(!(flags & XBF_WRITE)); ASSERT(bp->b_maps[0].bm_bn != XFS_BUF_DADDR_NULL); bp->b_flags &= ~(XBF_WRITE | XBF_ASYNC | XBF_READ_AHEAD | XBF_DONE); - bp->b_flags |= flags & (XBF_READ | XBF_ASYNC | XBF_READ_AHEAD); - + bp->b_flags |= XBF_READ; xfs_buf_submit(bp); - if (flags & XBF_ASYNC) - return 0; return xfs_buf_iowait(bp); } @@ -856,6 +794,8 @@ xfs_buf_read_map( struct xfs_buf *bp; int error; + ASSERT(!(flags & (XBF_WRITE | XBF_ASYNC | XBF_READ_AHEAD))); + flags |= XBF_READ; *bpp = NULL; @@ -869,21 +809,11 @@ xfs_buf_read_map( /* Initiate the buffer read and wait. */ XFS_STATS_INC(target->bt_mount, xb_get_read); bp->b_ops = ops; - error = _xfs_buf_read(bp, flags); - - /* Readahead iodone already dropped the buffer, so exit. */ - if (flags & XBF_ASYNC) - return 0; + error = _xfs_buf_read(bp); } else { /* Buffer already read; all we need to do is check it. */ error = xfs_buf_reverify(bp, ops); - /* Readahead already finished; drop the buffer and exit. */ - if (flags & XBF_ASYNC) { - xfs_buf_relse(bp); - return 0; - } - /* We do not want read in the flags */ bp->b_flags &= ~XBF_READ; ASSERT(bp->b_ops != NULL || ops == NULL); @@ -935,6 +865,7 @@ xfs_buf_readahead_map( int nmaps, const struct xfs_buf_ops *ops) { + const xfs_buf_flags_t flags = XBF_READ | XBF_ASYNC | XBF_READ_AHEAD; struct xfs_buf *bp; /* @@ -944,9 +875,21 @@ xfs_buf_readahead_map( if (xfs_buftarg_is_mem(target)) return; - xfs_buf_read_map(target, map, nmaps, - XBF_TRYLOCK | XBF_ASYNC | XBF_READ_AHEAD, &bp, ops, - __this_address); + if (xfs_buf_get_map(target, map, nmaps, flags | XBF_TRYLOCK, &bp)) + return; + trace_xfs_buf_readahead(bp, 0, _RET_IP_); + + if (bp->b_flags & XBF_DONE) { + xfs_buf_reverify(bp, ops); + xfs_buf_relse(bp); + return; + } + XFS_STATS_INC(target->bt_mount, xb_get_read); + bp->b_ops = ops; + bp->b_flags &= ~(XBF_WRITE | XBF_DONE); + bp->b_flags |= flags; + percpu_counter_inc(&target->bt_readahead_count); + xfs_buf_submit(bp); } /* @@ -1002,10 +945,12 @@ xfs_buf_get_uncached( struct xfs_buf *bp; DEFINE_SINGLE_BUF_MAP(map, XFS_BUF_DADDR_NULL, numblks); + /* there are currently no valid flags for xfs_buf_get_uncached */ + ASSERT(flags == 0); + *bpp = NULL; - /* flags might contain irrelevant bits, pass only what we care about */ - error = _xfs_buf_alloc(target, &map, 1, flags & XBF_NO_IOACCT, &bp); + error = _xfs_buf_alloc(target, &map, 1, flags, &bp); if (error) return error; @@ -1059,7 +1004,6 @@ xfs_buf_rele_uncached( spin_unlock(&bp->b_lock); return; } - __xfs_buf_ioacct_dec(bp); spin_unlock(&bp->b_lock); xfs_buf_free(bp); } @@ -1078,21 +1022,12 @@ xfs_buf_rele_cached( spin_lock(&bp->b_lock); ASSERT(bp->b_hold >= 1); if (bp->b_hold > 1) { - /* - * Drop the in-flight state if the buffer is already on the LRU - * and it holds the only reference. This is racy because we - * haven't acquired the pag lock, but the use of _XBF_IN_FLIGHT - * ensures the decrement occurs only once per-buf. - */ - if (--bp->b_hold == 1 && !list_empty(&bp->b_lru)) - __xfs_buf_ioacct_dec(bp); + bp->b_hold--; goto out_unlock; } /* we are asked to drop the last reference */ - spin_lock(&bch->bc_lock); - __xfs_buf_ioacct_dec(bp); - if (!(bp->b_flags & XBF_STALE) && atomic_read(&bp->b_lru_ref)) { + if (atomic_read(&bp->b_lru_ref)) { /* * If the buffer is added to the LRU, keep the reference to the * buffer for the LRU and clear the (now stale) dispose list @@ -1102,7 +1037,6 @@ xfs_buf_rele_cached( bp->b_state &= ~XFS_BSTATE_DISPOSE; else bp->b_hold--; - spin_unlock(&bch->bc_lock); } else { bp->b_hold--; /* @@ -1120,7 +1054,6 @@ xfs_buf_rele_cached( ASSERT(!(bp->b_flags & _XBF_DELWRI_Q)); rhashtable_remove_fast(&bch->bc_hash, &bp->b_rhash_head, xfs_buf_hash_params); - spin_unlock(&bch->bc_lock); if (pag) xfs_perag_put(pag); freebuf = true; @@ -1347,6 +1280,7 @@ xfs_buf_ioend_handle_error( resubmit: xfs_buf_ioerror(bp, 0); bp->b_flags |= (XBF_DONE | XBF_WRITE_FAIL); + reinit_completion(&bp->b_iowait); xfs_buf_submit(bp); return true; out_stale: @@ -1357,8 +1291,9 @@ out_stale: return false; } -static void -xfs_buf_ioend( +/* returns false if the caller needs to resubmit the I/O, else true */ +static bool +__xfs_buf_ioend( struct xfs_buf *bp) { trace_xfs_buf_iodone(bp, _RET_IP_); @@ -1371,6 +1306,8 @@ xfs_buf_ioend( bp->b_ops->verify_read(bp); if (!bp->b_error) bp->b_flags |= XBF_DONE; + if (bp->b_flags & XBF_READ_AHEAD) + percpu_counter_dec(&bp->b_target->bt_readahead_count); } else { if (!bp->b_error) { bp->b_flags &= ~XBF_WRITE_FAIL; @@ -1378,7 +1315,7 @@ xfs_buf_ioend( } if (unlikely(bp->b_error) && xfs_buf_ioend_handle_error(bp)) - return; + return false; /* clear the retry state */ bp->b_last_error = 0; @@ -1399,7 +1336,15 @@ xfs_buf_ioend( bp->b_flags &= ~(XBF_READ | XBF_WRITE | XBF_READ_AHEAD | _XBF_LOGRECOVERY); + return true; +} +static void +xfs_buf_ioend( + struct xfs_buf *bp) +{ + if (!__xfs_buf_ioend(bp)) + return; if (bp->b_flags & XBF_ASYNC) xfs_buf_relse(bp); else @@ -1413,15 +1358,8 @@ xfs_buf_ioend_work( struct xfs_buf *bp = container_of(work, struct xfs_buf, b_ioend_work); - xfs_buf_ioend(bp); -} - -static void -xfs_buf_ioend_async( - struct xfs_buf *bp) -{ - INIT_WORK(&bp->b_ioend_work, xfs_buf_ioend_work); - queue_work(bp->b_mount->m_buf_workqueue, &bp->b_ioend_work); + if (__xfs_buf_ioend(bp)) + xfs_buf_relse(bp); } void @@ -1493,7 +1431,13 @@ xfs_buf_bio_end_io( XFS_TEST_ERROR(false, bp->b_mount, XFS_ERRTAG_BUF_IOERROR)) xfs_buf_ioerror(bp, -EIO); - xfs_buf_ioend_async(bp); + if (bp->b_flags & XBF_ASYNC) { + INIT_WORK(&bp->b_ioend_work, xfs_buf_ioend_work); + queue_work(bp->b_mount->m_buf_workqueue, &bp->b_ioend_work); + } else { + complete(&bp->b_iowait); + } + bio_put(bio); } @@ -1570,9 +1514,11 @@ xfs_buf_iowait( { ASSERT(!(bp->b_flags & XBF_ASYNC)); - trace_xfs_buf_iowait(bp, _RET_IP_); - wait_for_completion(&bp->b_iowait); - trace_xfs_buf_iowait_done(bp, _RET_IP_); + do { + trace_xfs_buf_iowait(bp, _RET_IP_); + wait_for_completion(&bp->b_iowait); + trace_xfs_buf_iowait_done(bp, _RET_IP_); + } while (!__xfs_buf_ioend(bp)); return bp->b_error; } @@ -1650,9 +1596,6 @@ xfs_buf_submit( */ bp->b_error = 0; - if (bp->b_flags & XBF_ASYNC) - xfs_buf_ioacct_inc(bp); - if ((bp->b_flags & XBF_WRITE) && !xfs_buf_verify_write(bp)) { xfs_force_shutdown(bp->b_mount, SHUTDOWN_CORRUPT_INCORE); xfs_buf_ioend(bp); @@ -1778,9 +1721,8 @@ xfs_buftarg_wait( struct xfs_buftarg *btp) { /* - * First wait on the buftarg I/O count for all in-flight buffers to be - * released. This is critical as new buffers do not make the LRU until - * they are released. + * First wait for all in-flight readahead buffers to be released. This is + * critical as new buffers do not make the LRU until they are released. * * Next, flush the buffer workqueue to ensure all completion processing * has finished. Just waiting on buffer locks is not sufficient for @@ -1789,7 +1731,7 @@ xfs_buftarg_wait( * all reference counts have been dropped before we start walking the * LRU list. */ - while (percpu_counter_sum(&btp->bt_io_count)) + while (percpu_counter_sum(&btp->bt_readahead_count)) delay(100); flush_workqueue(btp->bt_mount->m_buf_workqueue); } @@ -1906,8 +1848,8 @@ xfs_destroy_buftarg( struct xfs_buftarg *btp) { shrinker_free(btp->bt_shrinker); - ASSERT(percpu_counter_sum(&btp->bt_io_count) == 0); - percpu_counter_destroy(&btp->bt_io_count); + ASSERT(percpu_counter_sum(&btp->bt_readahead_count) == 0); + percpu_counter_destroy(&btp->bt_readahead_count); list_lru_destroy(&btp->bt_lru); } @@ -1961,7 +1903,7 @@ xfs_init_buftarg( if (list_lru_init(&btp->bt_lru)) return -ENOMEM; - if (percpu_counter_init(&btp->bt_io_count, 0, GFP_KERNEL)) + if (percpu_counter_init(&btp->bt_readahead_count, 0, GFP_KERNEL)) goto out_destroy_lru; btp->bt_shrinker = @@ -1975,7 +1917,7 @@ xfs_init_buftarg( return 0; out_destroy_io_count: - percpu_counter_destroy(&btp->bt_io_count); + percpu_counter_destroy(&btp->bt_readahead_count); out_destroy_lru: list_lru_destroy(&btp->bt_lru); return -ENOMEM; diff --git a/fs/xfs/xfs_buf.h b/fs/xfs/xfs_buf.h index 7e73663c5d4a..80e06eecaf56 100644 --- a/fs/xfs/xfs_buf.h +++ b/fs/xfs/xfs_buf.h @@ -27,7 +27,6 @@ struct xfs_buf; #define XBF_READ (1u << 0) /* buffer intended for reading from device */ #define XBF_WRITE (1u << 1) /* buffer intended for writing to device */ #define XBF_READ_AHEAD (1u << 2) /* asynchronous read-ahead */ -#define XBF_NO_IOACCT (1u << 3) /* bypass I/O accounting (non-LRU bufs) */ #define XBF_ASYNC (1u << 4) /* initiator will not wait for completion */ #define XBF_DONE (1u << 5) /* all pages in the buffer uptodate */ #define XBF_STALE (1u << 6) /* buffer has been staled, do not find it */ @@ -58,7 +57,6 @@ typedef unsigned int xfs_buf_flags_t; { XBF_READ, "READ" }, \ { XBF_WRITE, "WRITE" }, \ { XBF_READ_AHEAD, "READ_AHEAD" }, \ - { XBF_NO_IOACCT, "NO_IOACCT" }, \ { XBF_ASYNC, "ASYNC" }, \ { XBF_DONE, "DONE" }, \ { XBF_STALE, "STALE" }, \ @@ -77,10 +75,8 @@ typedef unsigned int xfs_buf_flags_t; * Internal state flags. */ #define XFS_BSTATE_DISPOSE (1 << 0) /* buffer being discarded */ -#define XFS_BSTATE_IN_FLIGHT (1 << 1) /* I/O in flight */ struct xfs_buf_cache { - spinlock_t bc_lock; struct rhashtable bc_hash; }; @@ -117,7 +113,7 @@ struct xfs_buftarg { struct shrinker *bt_shrinker; struct list_lru bt_lru; - struct percpu_counter bt_io_count; + struct percpu_counter bt_readahead_count; struct ratelimit_state bt_ioerror_rl; /* Atomic write unit values */ @@ -292,7 +288,7 @@ int xfs_buf_get_uncached(struct xfs_buftarg *target, size_t numblks, int xfs_buf_read_uncached(struct xfs_buftarg *target, xfs_daddr_t daddr, size_t numblks, xfs_buf_flags_t flags, struct xfs_buf **bpp, const struct xfs_buf_ops *ops); -int _xfs_buf_read(struct xfs_buf *bp, xfs_buf_flags_t flags); +int _xfs_buf_read(struct xfs_buf *bp); void xfs_buf_hold(struct xfs_buf *bp); /* Releasing Buffers */ diff --git a/fs/xfs/xfs_buf_mem.c b/fs/xfs/xfs_buf_mem.c index 07bebbfb16ee..5b64a2b3b113 100644 --- a/fs/xfs/xfs_buf_mem.c +++ b/fs/xfs/xfs_buf_mem.c @@ -117,7 +117,7 @@ xmbuf_free( struct xfs_buftarg *btp) { ASSERT(xfs_buftarg_is_mem(btp)); - ASSERT(percpu_counter_sum(&btp->bt_io_count) == 0); + ASSERT(percpu_counter_sum(&btp->bt_readahead_count) == 0); trace_xmbuf_free(btp); diff --git a/fs/xfs/xfs_exchrange.c b/fs/xfs/xfs_exchrange.c index f340a2015c4c..0b41bdfecdfb 100644 --- a/fs/xfs/xfs_exchrange.c +++ b/fs/xfs/xfs_exchrange.c @@ -329,22 +329,6 @@ out_trans_cancel: * successfully but before locks are dropped. */ -/* Verify that we have security clearance to perform this operation. */ -static int -xfs_exchange_range_verify_area( - struct xfs_exchrange *fxr) -{ - int ret; - - ret = remap_verify_area(fxr->file1, fxr->file1_offset, fxr->length, - true); - if (ret) - return ret; - - return remap_verify_area(fxr->file2, fxr->file2_offset, fxr->length, - true); -} - /* * Performs necessary checks before doing a range exchange, having stabilized * mutable inode attributes via i_rwsem. @@ -355,11 +339,13 @@ xfs_exchange_range_checks( unsigned int alloc_unit) { struct inode *inode1 = file_inode(fxr->file1); + loff_t size1 = i_size_read(inode1); struct inode *inode2 = file_inode(fxr->file2); + loff_t size2 = i_size_read(inode2); uint64_t allocmask = alloc_unit - 1; int64_t test_len; uint64_t blen; - loff_t size1, size2, tmp; + loff_t tmp; int error; /* Don't touch certain kinds of inodes */ @@ -368,24 +354,25 @@ xfs_exchange_range_checks( if (IS_SWAPFILE(inode1) || IS_SWAPFILE(inode2)) return -ETXTBSY; - size1 = i_size_read(inode1); - size2 = i_size_read(inode2); - /* Ranges cannot start after EOF. */ if (fxr->file1_offset > size1 || fxr->file2_offset > size2) return -EINVAL; - /* - * If the caller said to exchange to EOF, we set the length of the - * request large enough to cover everything to the end of both files. - */ if (fxr->flags & XFS_EXCHANGE_RANGE_TO_EOF) { + /* + * If the caller said to exchange to EOF, we set the length of + * the request large enough to cover everything to the end of + * both files. + */ fxr->length = max_t(int64_t, size1 - fxr->file1_offset, size2 - fxr->file2_offset); - - error = xfs_exchange_range_verify_area(fxr); - if (error) - return error; + } else { + /* + * Otherwise we require both ranges to end within EOF. + */ + if (fxr->file1_offset + fxr->length > size1 || + fxr->file2_offset + fxr->length > size2) + return -EINVAL; } /* @@ -402,15 +389,6 @@ xfs_exchange_range_checks( return -EINVAL; /* - * We require both ranges to end within EOF, unless we're exchanging - * to EOF. - */ - if (!(fxr->flags & XFS_EXCHANGE_RANGE_TO_EOF) && - (fxr->file1_offset + fxr->length > size1 || - fxr->file2_offset + fxr->length > size2)) - return -EINVAL; - - /* * Make sure we don't hit any file size limits. If we hit any size * limits such that test_length was adjusted, we abort the whole * operation. @@ -747,6 +725,7 @@ xfs_exchange_range( { struct inode *inode1 = file_inode(fxr->file1); struct inode *inode2 = file_inode(fxr->file2); + loff_t check_len = fxr->length; int ret; BUILD_BUG_ON(XFS_EXCHANGE_RANGE_ALL_FLAGS & @@ -779,14 +758,18 @@ xfs_exchange_range( return -EBADF; /* - * If we're not exchanging to EOF, we can check the areas before - * stabilizing both files' i_size. + * If we're exchanging to EOF we can't calculate the length until taking + * the iolock. Pass a 0 length to remap_verify_area similar to the + * FICLONE and FICLONERANGE ioctls that support cloning to EOF as well. */ - if (!(fxr->flags & XFS_EXCHANGE_RANGE_TO_EOF)) { - ret = xfs_exchange_range_verify_area(fxr); - if (ret) - return ret; - } + if (fxr->flags & XFS_EXCHANGE_RANGE_TO_EOF) + check_len = 0; + ret = remap_verify_area(fxr->file1, fxr->file1_offset, check_len, true); + if (ret) + return ret; + ret = remap_verify_area(fxr->file2, fxr->file2_offset, check_len, true); + if (ret) + return ret; /* Update cmtime if the fd/inode don't forbid it. */ if (!(fxr->file1->f_mode & FMODE_NOCMTIME) && !IS_NOCMTIME(inode1)) diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c index c95fe1b1de4e..b1f9f156ec88 100644 --- a/fs/xfs/xfs_inode.c +++ b/fs/xfs/xfs_inode.c @@ -1404,8 +1404,11 @@ xfs_inactive( goto out; /* Try to clean out the cow blocks if there are any. */ - if (xfs_inode_has_cow_data(ip)) - xfs_reflink_cancel_cow_range(ip, 0, NULLFILEOFF, true); + if (xfs_inode_has_cow_data(ip)) { + error = xfs_reflink_cancel_cow_range(ip, 0, NULLFILEOFF, true); + if (error) + goto out; + } if (VFS_I(ip)->i_nlink != 0) { /* diff --git a/fs/xfs/xfs_iomap.c b/fs/xfs/xfs_iomap.c index 50fa3ef89f6c..d61460309a78 100644 --- a/fs/xfs/xfs_iomap.c +++ b/fs/xfs/xfs_iomap.c @@ -976,10 +976,8 @@ xfs_dax_write_iomap_end( if (!xfs_is_cow_inode(ip)) return 0; - if (!written) { - xfs_reflink_cancel_cow_range(ip, pos, length, true); - return 0; - } + if (!written) + return xfs_reflink_cancel_cow_range(ip, pos, length, true); return xfs_reflink_end_cow(ip, pos, written); } diff --git a/fs/xfs/xfs_log_recover.c b/fs/xfs/xfs_log_recover.c index b3c27dbccce8..2f76531842f8 100644 --- a/fs/xfs/xfs_log_recover.c +++ b/fs/xfs/xfs_log_recover.c @@ -3380,7 +3380,7 @@ xlog_do_recover( */ xfs_buf_lock(bp); xfs_buf_hold(bp); - error = _xfs_buf_read(bp, XBF_READ); + error = _xfs_buf_read(bp); if (error) { if (!xlog_is_shutdown(log)) { xfs_buf_ioerror_alert(bp, __this_address); diff --git a/fs/xfs/xfs_mount.c b/fs/xfs/xfs_mount.c index 477c5262cf91..b69356582b86 100644 --- a/fs/xfs/xfs_mount.c +++ b/fs/xfs/xfs_mount.c @@ -181,14 +181,11 @@ xfs_readsb( /* * Allocate a (locked) buffer to hold the superblock. This will be kept - * around at all times to optimize access to the superblock. Therefore, - * set XBF_NO_IOACCT to make sure it doesn't hold the buftarg count - * elevated. + * around at all times to optimize access to the superblock. */ reread: error = xfs_buf_read_uncached(mp->m_ddev_targp, XFS_SB_DADDR, - BTOBB(sector_size), XBF_NO_IOACCT, &bp, - buf_ops); + BTOBB(sector_size), 0, &bp, buf_ops); if (error) { if (loud) xfs_warn(mp, "SB validate failed with error %d.", error); diff --git a/fs/xfs/xfs_qm_bhv.c b/fs/xfs/xfs_qm_bhv.c index 37f1230e7584..245d754f382a 100644 --- a/fs/xfs/xfs_qm_bhv.c +++ b/fs/xfs/xfs_qm_bhv.c @@ -78,6 +78,28 @@ xfs_qm_statvfs( } } +STATIC int +xfs_qm_validate_state_change( + struct xfs_mount *mp, + uint uqd, + uint gqd, + uint pqd) +{ + int state; + + /* Is quota state changing? */ + state = ((uqd && !XFS_IS_UQUOTA_ON(mp)) || + (!uqd && XFS_IS_UQUOTA_ON(mp)) || + (gqd && !XFS_IS_GQUOTA_ON(mp)) || + (!gqd && XFS_IS_GQUOTA_ON(mp)) || + (pqd && !XFS_IS_PQUOTA_ON(mp)) || + (!pqd && XFS_IS_PQUOTA_ON(mp))); + + return state && + (xfs_dev_is_read_only(mp, "changing quota state") || + xfs_has_norecovery(mp)); +} + int xfs_qm_newmount( xfs_mount_t *mp, @@ -97,24 +119,25 @@ xfs_qm_newmount( } /* - * If the device itself is read-only, we can't allow - * the user to change the state of quota on the mount - - * this would generate a transaction on the ro device, - * which would lead to an I/O error and shutdown + * If the device itself is read-only and/or in norecovery + * mode, we can't allow the user to change the state of + * quota on the mount - this would generate a transaction + * on the ro device, which would lead to an I/O error and + * shutdown. */ - if (((uquotaondisk && !XFS_IS_UQUOTA_ON(mp)) || - (!uquotaondisk && XFS_IS_UQUOTA_ON(mp)) || - (gquotaondisk && !XFS_IS_GQUOTA_ON(mp)) || - (!gquotaondisk && XFS_IS_GQUOTA_ON(mp)) || - (pquotaondisk && !XFS_IS_PQUOTA_ON(mp)) || - (!pquotaondisk && XFS_IS_PQUOTA_ON(mp))) && - xfs_dev_is_read_only(mp, "changing quota state")) { - xfs_warn(mp, "please mount with%s%s%s%s.", - (!quotaondisk ? "out quota" : ""), - (uquotaondisk ? " usrquota" : ""), - (gquotaondisk ? " grpquota" : ""), - (pquotaondisk ? " prjquota" : "")); + if (xfs_qm_validate_state_change(mp, uquotaondisk, + gquotaondisk, pquotaondisk)) { + + if (xfs_has_metadir(mp)) + xfs_warn(mp, + "metadir enabled, please mount without any quota mount options"); + else + xfs_warn(mp, "please mount with%s%s%s%s.", + (!quotaondisk ? "out quota" : ""), + (uquotaondisk ? " usrquota" : ""), + (gquotaondisk ? " grpquota" : ""), + (pquotaondisk ? " prjquota" : "")); return -EPERM; } diff --git a/fs/xfs/xfs_rtalloc.c b/fs/xfs/xfs_rtalloc.c index d8e6d073d64d..57bef567e011 100644 --- a/fs/xfs/xfs_rtalloc.c +++ b/fs/xfs/xfs_rtalloc.c @@ -1407,7 +1407,7 @@ xfs_rtmount_readsb( /* m_blkbb_log is not set up yet */ error = xfs_buf_read_uncached(mp->m_rtdev_targp, XFS_RTSB_DADDR, - mp->m_sb.sb_blocksize >> BBSHIFT, XBF_NO_IOACCT, &bp, + mp->m_sb.sb_blocksize >> BBSHIFT, 0, &bp, &xfs_rtsb_buf_ops); if (error) { xfs_warn(mp, "rt sb validate failed with error %d.", error); diff --git a/fs/xfs/xfs_super.c b/fs/xfs/xfs_super.c index d92d7a07ea89..0055066fb1d9 100644 --- a/fs/xfs/xfs_super.c +++ b/fs/xfs/xfs_super.c @@ -1661,8 +1661,12 @@ xfs_fs_fill_super( #endif } - /* Filesystem claims it needs repair, so refuse the mount. */ - if (xfs_has_needsrepair(mp)) { + /* + * Filesystem claims it needs repair, so refuse the mount unless + * norecovery is also specified, in which case the filesystem can + * be mounted with no risk of further damage. + */ + if (xfs_has_needsrepair(mp) && !xfs_has_norecovery(mp)) { xfs_warn(mp, "Filesystem needs repair. Please run xfs_repair."); error = -EFSCORRUPTED; goto out_free_sb; diff --git a/fs/xfs/xfs_trace.h b/fs/xfs/xfs_trace.h index b29462363b81..bfc2f1249022 100644 --- a/fs/xfs/xfs_trace.h +++ b/fs/xfs/xfs_trace.h @@ -593,6 +593,7 @@ DEFINE_EVENT(xfs_buf_flags_class, name, \ DEFINE_BUF_FLAGS_EVENT(xfs_buf_find); DEFINE_BUF_FLAGS_EVENT(xfs_buf_get); DEFINE_BUF_FLAGS_EVENT(xfs_buf_read); +DEFINE_BUF_FLAGS_EVENT(xfs_buf_readahead); TRACE_EVENT(xfs_buf_ioerror, TP_PROTO(struct xfs_buf *bp, int error, xfs_failaddr_t caller_ip), |