diff options
Diffstat (limited to 'fs/btrfs/extent-tree.c')
-rw-r--r-- | fs/btrfs/extent-tree.c | 421 |
1 files changed, 222 insertions, 199 deletions
diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index 61b494e8e604..210c94ac8818 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -60,21 +60,6 @@ enum { CHUNK_ALLOC_FORCE = 2, }; -/* - * Control how reservations are dealt with. - * - * RESERVE_FREE - freeing a reservation. - * RESERVE_ALLOC - allocating space and we need to update bytes_may_use for - * ENOSPC accounting - * RESERVE_ALLOC_NO_ACCOUNT - allocating space and we should not update - * bytes_may_use as the ENOSPC accounting is done elsewhere - */ -enum { - RESERVE_FREE = 0, - RESERVE_ALLOC = 1, - RESERVE_ALLOC_NO_ACCOUNT = 2, -}; - static int update_block_group(struct btrfs_trans_handle *trans, struct btrfs_root *root, u64 bytenr, u64 num_bytes, int alloc); @@ -102,11 +87,13 @@ static int do_chunk_alloc(struct btrfs_trans_handle *trans, int force); static int find_next_key(struct btrfs_path *path, int level, struct btrfs_key *key); -static void dump_space_info(struct btrfs_space_info *info, u64 bytes, +static void dump_space_info(struct btrfs_fs_info *fs_info, + struct btrfs_space_info *info, u64 bytes, int dump_block_groups); -static int btrfs_update_reserved_bytes(struct btrfs_block_group_cache *cache, - u64 num_bytes, int reserve, - int delalloc); +static int btrfs_add_reserved_bytes(struct btrfs_block_group_cache *cache, + u64 ram_bytes, u64 num_bytes, int delalloc); +static int btrfs_free_reserved_bytes(struct btrfs_block_group_cache *cache, + u64 num_bytes, int delalloc); static int block_rsv_use_bytes(struct btrfs_block_rsv *block_rsv, u64 num_bytes); int btrfs_pin_extent(struct btrfs_root *root, @@ -280,9 +267,8 @@ static int exclude_super_stripes(struct btrfs_root *root, for (i = 0; i < BTRFS_SUPER_MIRROR_MAX; i++) { bytenr = btrfs_sb_offset(i); - ret = btrfs_rmap_block(&root->fs_info->mapping_tree, - cache->key.objectid, bytenr, - 0, &logical, &nr, &stripe_len); + ret = btrfs_rmap_block(root->fs_info, cache->key.objectid, + bytenr, 0, &logical, &nr, &stripe_len); if (ret) return ret; @@ -744,11 +730,7 @@ static int cache_block_group(struct btrfs_block_group_cache *cache, static struct btrfs_block_group_cache * btrfs_lookup_first_block_group(struct btrfs_fs_info *info, u64 bytenr) { - struct btrfs_block_group_cache *cache; - - cache = block_group_cache_tree_search(info, bytenr, 0); - - return cache; + return block_group_cache_tree_search(info, bytenr, 0); } /* @@ -758,11 +740,7 @@ struct btrfs_block_group_cache *btrfs_lookup_block_group( struct btrfs_fs_info *info, u64 bytenr) { - struct btrfs_block_group_cache *cache; - - cache = block_group_cache_tree_search(info, bytenr, 1); - - return cache; + return block_group_cache_tree_search(info, bytenr, 1); } static struct btrfs_space_info *__find_space_info(struct btrfs_fs_info *info, @@ -2374,7 +2352,13 @@ static int run_delayed_tree_ref(struct btrfs_trans_handle *trans, ins.type = BTRFS_EXTENT_ITEM_KEY; } - BUG_ON(node->ref_mod != 1); + if (node->ref_mod != 1) { + btrfs_err(root->fs_info, + "btree block(%llu) has %d references rather than 1: action %d ref_root %llu parent %llu", + node->bytenr, node->ref_mod, node->action, ref_root, + parent); + return -EIO; + } if (node->action == BTRFS_ADD_DELAYED_REF && insert_reserved) { BUG_ON(!extent_op || !extent_op->update_flags); ret = alloc_reserved_tree_block(trans, root, @@ -2604,7 +2588,9 @@ static noinline int __btrfs_run_delayed_refs(struct btrfs_trans_handle *trans, if (must_insert_reserved) locked_ref->must_insert_reserved = 1; locked_ref->processing = 0; - btrfs_debug(fs_info, "run_delayed_extent_op returned %d", ret); + btrfs_debug(fs_info, + "run_delayed_extent_op returned %d", + ret); btrfs_delayed_ref_unlock(locked_ref); return ret; } @@ -2664,7 +2650,8 @@ static noinline int __btrfs_run_delayed_refs(struct btrfs_trans_handle *trans, locked_ref->processing = 0; btrfs_delayed_ref_unlock(locked_ref); btrfs_put_delayed_ref(ref); - btrfs_debug(fs_info, "run_one_delayed_ref returned %d", ret); + btrfs_debug(fs_info, "run_one_delayed_ref returned %d", + ret); return ret; } @@ -2954,7 +2941,7 @@ int btrfs_run_delayed_refs(struct btrfs_trans_handle *trans, if (trans->aborted) return 0; - if (root->fs_info->creating_free_space_tree) + if (test_bit(BTRFS_FS_CREATING_FREE_SPACE_TREE, &root->fs_info->flags)) return 0; if (root == root->fs_info->extent_root) @@ -2985,7 +2972,6 @@ again: spin_unlock(&delayed_refs->lock); goto out; } - count = (unsigned long)-1; while (node) { head = rb_entry(node, struct btrfs_delayed_ref_head, @@ -3501,7 +3487,6 @@ again: dcs = BTRFS_DC_SETUP; else if (ret == -ENOSPC) set_bit(BTRFS_TRANS_CACHE_ENOSPC, &trans->transaction->flags); - btrfs_free_reserved_data_space(inode, 0, num_pages); out_put: iput(inode); @@ -3709,6 +3694,8 @@ again: goto again; } spin_unlock(&cur_trans->dirty_bgs_lock); + } else if (ret < 0) { + btrfs_cleanup_dirty_bgs(cur_trans, root); } btrfs_free_path(path); @@ -4286,13 +4273,10 @@ int btrfs_check_data_free_space(struct inode *inode, u64 start, u64 len) if (ret < 0) return ret; - /* - * Use new btrfs_qgroup_reserve_data to reserve precious data space - * - * TODO: Find a good method to avoid reserve data space for NOCOW - * range, but don't impact performance on quota disable case. - */ + /* Use new btrfs_qgroup_reserve_data to reserve precious data space. */ ret = btrfs_qgroup_reserve_data(inode, start, len); + if (ret) + btrfs_free_reserved_data_space_noquota(inode, start, len); return ret; } @@ -4447,7 +4431,7 @@ void check_system_chunk(struct btrfs_trans_handle *trans, if (left < thresh && btrfs_test_opt(root->fs_info, ENOSPC_DEBUG)) { btrfs_info(root->fs_info, "left=%llu, need=%llu, flags=%llu", left, thresh, type); - dump_space_info(info, 0, 0); + dump_space_info(root->fs_info, info, 0, 0); } if (left < thresh) { @@ -4472,6 +4456,15 @@ void check_system_chunk(struct btrfs_trans_handle *trans, } } +/* + * If force is CHUNK_ALLOC_FORCE: + * - return 1 if it successfully allocates a chunk, + * - return errors including -ENOSPC otherwise. + * If force is NOT CHUNK_ALLOC_FORCE: + * - return 0 if it doesn't need to allocate a new chunk, + * - return 1 if it successfully allocates a chunk, + * - return errors including -ENOSPC otherwise. + */ static int do_chunk_alloc(struct btrfs_trans_handle *trans, struct btrfs_root *extent_root, u64 flags, int force) { @@ -4882,7 +4875,7 @@ static int flush_space(struct btrfs_root *root, btrfs_get_alloc_profile(root, 0), CHUNK_ALLOC_NO_FORCE); btrfs_end_transaction(trans, root); - if (ret == -ENOSPC) + if (ret > 0 || ret == -ENOSPC) ret = 0; break; case COMMIT_TRANS: @@ -4907,11 +4900,6 @@ btrfs_calc_reclaim_metadata_size(struct btrfs_root *root, u64 expected; u64 to_reclaim = 0; - to_reclaim = min_t(u64, num_online_cpus() * SZ_1M, SZ_16M); - if (can_overcommit(root, space_info, to_reclaim, - BTRFS_RESERVE_FLUSH_ALL)) - return 0; - list_for_each_entry(ticket, &space_info->tickets, list) to_reclaim += ticket->bytes; list_for_each_entry(ticket, &space_info->priority_tickets, list) @@ -4919,6 +4907,11 @@ btrfs_calc_reclaim_metadata_size(struct btrfs_root *root, if (to_reclaim) return to_reclaim; + to_reclaim = min_t(u64, num_online_cpus() * SZ_1M, SZ_16M); + if (can_overcommit(root, space_info, to_reclaim, + BTRFS_RESERVE_FLUSH_ALL)) + return 0; + used = space_info->bytes_used + space_info->bytes_reserved + space_info->bytes_pinned + space_info->bytes_readonly + space_info->bytes_may_use; @@ -4972,12 +4965,12 @@ static void wake_all_tickets(struct list_head *head) */ static void btrfs_async_reclaim_metadata_space(struct work_struct *work) { - struct reserve_ticket *last_ticket = NULL; struct btrfs_fs_info *fs_info; struct btrfs_space_info *space_info; u64 to_reclaim; int flush_state; int commit_cycles = 0; + u64 last_tickets_id; fs_info = container_of(work, struct btrfs_fs_info, async_reclaim_work); space_info = __find_space_info(fs_info, BTRFS_BLOCK_GROUP_METADATA); @@ -4990,8 +4983,7 @@ static void btrfs_async_reclaim_metadata_space(struct work_struct *work) spin_unlock(&space_info->lock); return; } - last_ticket = list_first_entry(&space_info->tickets, - struct reserve_ticket, list); + last_tickets_id = space_info->tickets_id; spin_unlock(&space_info->lock); flush_state = FLUSH_DELAYED_ITEMS_NR; @@ -5011,10 +5003,10 @@ static void btrfs_async_reclaim_metadata_space(struct work_struct *work) space_info); ticket = list_first_entry(&space_info->tickets, struct reserve_ticket, list); - if (last_ticket == ticket) { + if (last_tickets_id == space_info->tickets_id) { flush_state++; } else { - last_ticket = ticket; + last_tickets_id = space_info->tickets_id; flush_state = FLUSH_DELAYED_ITEMS_NR; if (commit_cycles) commit_cycles--; @@ -5196,7 +5188,7 @@ static int __reserve_metadata_bytes(struct btrfs_root *root, * which means we won't have fs_info->fs_root set, so don't do * the async reclaim as we will panic. */ - if (!root->fs_info->log_root_recovering && + if (!test_bit(BTRFS_FS_LOG_RECOVERING, &root->fs_info->flags) && need_do_async_reclaim(space_info, root, used) && !work_busy(&root->fs_info->async_reclaim_work)) { trace_btrfs_trigger_flush(root->fs_info, @@ -5390,6 +5382,7 @@ again: list_del_init(&ticket->list); num_bytes -= ticket->bytes; ticket->bytes = 0; + space_info->tickets_id++; wake_up(&ticket->wait); } else { ticket->bytes -= num_bytes; @@ -5432,6 +5425,7 @@ again: num_bytes -= ticket->bytes; space_info->bytes_may_use += ticket->bytes; ticket->bytes = 0; + space_info->tickets_id++; wake_up(&ticket->wait); } else { trace_btrfs_space_reservation(fs_info, "space_info", @@ -5800,7 +5794,7 @@ int btrfs_subvolume_reserve_metadata(struct btrfs_root *root, int ret; struct btrfs_block_rsv *global_rsv = &root->fs_info->global_block_rsv; - if (root->fs_info->quota_enabled) { + if (test_bit(BTRFS_FS_QUOTA_ENABLED, &root->fs_info->flags)) { /* One for parent inode, two for dir entries */ num_bytes = 3 * root->nodesize; ret = btrfs_qgroup_reserve_meta(root, num_bytes); @@ -5978,7 +5972,7 @@ int btrfs_delalloc_reserve_metadata(struct inode *inode, u64 num_bytes) csum_bytes = BTRFS_I(inode)->csum_bytes; spin_unlock(&BTRFS_I(inode)->lock); - if (root->fs_info->quota_enabled) { + if (test_bit(BTRFS_FS_QUOTA_ENABLED, &root->fs_info->flags)) { ret = btrfs_qgroup_reserve_meta(root, nr_extents * root->nodesize); if (ret) @@ -6118,8 +6112,6 @@ void btrfs_delalloc_release_metadata(struct inode *inode, u64 num_bytes) * @start: start range we are writing to * @len: how long the range we are writing to * - * TODO: This function will finally replace old btrfs_delalloc_reserve_space() - * * This will do the following things * * o reserve space in data space info for num bytes @@ -6497,19 +6489,15 @@ void btrfs_wait_block_group_reservations(struct btrfs_block_group_cache *bg) } /** - * btrfs_update_reserved_bytes - update the block_group and space info counters + * btrfs_add_reserved_bytes - update the block_group and space info counters * @cache: The cache we are manipulating + * @ram_bytes: The number of bytes of file content, and will be same to + * @num_bytes except for the compress path. * @num_bytes: The number of bytes in question - * @reserve: One of the reservation enums * @delalloc: The blocks are allocated for the delalloc write * - * This is called by the allocator when it reserves space, or by somebody who is - * freeing space that was never actually used on disk. For example if you - * reserve some space for a new leaf in transaction A and before transaction A - * commits you free that leaf, you call this with reserve set to 0 in order to - * clear the reservation. - * - * Metadata reservations should be called with RESERVE_ALLOC so we do the proper + * This is called by the allocator when it reserves space. Metadata + * reservations should be called with RESERVE_ALLOC so we do the proper * ENOSPC accounting. For data we handle the reservation through clearing the * delalloc bits in the io_tree. We have to do this since we could end up * allocating less disk space for the amount of data we have reserved in the @@ -6519,44 +6507,63 @@ void btrfs_wait_block_group_reservations(struct btrfs_block_group_cache *bg) * make the reservation and return -EAGAIN, otherwise this function always * succeeds. */ -static int btrfs_update_reserved_bytes(struct btrfs_block_group_cache *cache, - u64 num_bytes, int reserve, int delalloc) +static int btrfs_add_reserved_bytes(struct btrfs_block_group_cache *cache, + u64 ram_bytes, u64 num_bytes, int delalloc) { struct btrfs_space_info *space_info = cache->space_info; int ret = 0; spin_lock(&space_info->lock); spin_lock(&cache->lock); - if (reserve != RESERVE_FREE) { - if (cache->ro) { - ret = -EAGAIN; - } else { - cache->reserved += num_bytes; - space_info->bytes_reserved += num_bytes; - if (reserve == RESERVE_ALLOC) { - trace_btrfs_space_reservation(cache->fs_info, - "space_info", space_info->flags, - num_bytes, 0); - space_info->bytes_may_use -= num_bytes; - } - - if (delalloc) - cache->delalloc_bytes += num_bytes; - } + if (cache->ro) { + ret = -EAGAIN; } else { - if (cache->ro) - space_info->bytes_readonly += num_bytes; - cache->reserved -= num_bytes; - space_info->bytes_reserved -= num_bytes; + cache->reserved += num_bytes; + space_info->bytes_reserved += num_bytes; + trace_btrfs_space_reservation(cache->fs_info, + "space_info", space_info->flags, + ram_bytes, 0); + space_info->bytes_may_use -= ram_bytes; if (delalloc) - cache->delalloc_bytes -= num_bytes; + cache->delalloc_bytes += num_bytes; } spin_unlock(&cache->lock); spin_unlock(&space_info->lock); return ret; } +/** + * btrfs_free_reserved_bytes - update the block_group and space info counters + * @cache: The cache we are manipulating + * @num_bytes: The number of bytes in question + * @delalloc: The blocks are allocated for the delalloc write + * + * This is called by somebody who is freeing space that was never actually used + * on disk. For example if you reserve some space for a new leaf in transaction + * A and before transaction A commits you free that leaf, you call this with + * reserve set to 0 in order to clear the reservation. + */ + +static int btrfs_free_reserved_bytes(struct btrfs_block_group_cache *cache, + u64 num_bytes, int delalloc) +{ + struct btrfs_space_info *space_info = cache->space_info; + int ret = 0; + + spin_lock(&space_info->lock); + spin_lock(&cache->lock); + if (cache->ro) + space_info->bytes_readonly += num_bytes; + cache->reserved -= num_bytes; + space_info->bytes_reserved -= num_bytes; + + if (delalloc) + cache->delalloc_bytes -= num_bytes; + spin_unlock(&cache->lock); + spin_unlock(&space_info->lock); + return ret; +} void btrfs_prepare_extent_commit(struct btrfs_trans_handle *trans, struct btrfs_root *root) { @@ -6923,8 +6930,9 @@ static int __btrfs_free_extent(struct btrfs_trans_handle *trans, } if (ret) { - btrfs_err(info, "umm, got %d back from search, was looking for %llu", - ret, bytenr); + btrfs_err(info, + "umm, got %d back from search, was looking for %llu", + ret, bytenr); if (ret > 0) btrfs_print_leaf(extent_root, path->nodes[0]); @@ -6970,7 +6978,8 @@ static int __btrfs_free_extent(struct btrfs_trans_handle *trans, ret = btrfs_search_slot(trans, extent_root, &key, path, -1, 1); if (ret) { - btrfs_err(info, "umm, got %d back from search, was looking for %llu", + btrfs_err(info, + "umm, got %d back from search, was looking for %llu", ret, bytenr); btrfs_print_leaf(extent_root, path->nodes[0]); } @@ -6997,8 +7006,9 @@ static int __btrfs_free_extent(struct btrfs_trans_handle *trans, refs = btrfs_extent_refs(leaf, ei); if (refs < refs_to_drop) { - btrfs_err(info, "trying to drop %d refs but we only have %Lu " - "for bytenr %Lu", refs_to_drop, refs, bytenr); + btrfs_err(info, + "trying to drop %d refs but we only have %Lu for bytenr %Lu", + refs_to_drop, refs, bytenr); ret = -EINVAL; btrfs_abort_transaction(trans, ret); goto out; @@ -7191,7 +7201,7 @@ void btrfs_free_tree_block(struct btrfs_trans_handle *trans, WARN_ON(test_bit(EXTENT_BUFFER_DIRTY, &buf->bflags)); btrfs_add_free_space(cache, buf->start, buf->len); - btrfs_update_reserved_bytes(cache, buf->len, RESERVE_FREE, 0); + btrfs_free_reserved_bytes(cache, buf->len, 0); btrfs_put_block_group(cache); trace_btrfs_reserved_extent_free(root, buf->start, buf->len); pin = 0; @@ -7416,9 +7426,9 @@ btrfs_release_block_group(struct btrfs_block_group_cache *cache, * the free space extent currently. */ static noinline int find_free_extent(struct btrfs_root *orig_root, - u64 num_bytes, u64 empty_size, - u64 hint_byte, struct btrfs_key *ins, - u64 flags, int delalloc) + u64 ram_bytes, u64 num_bytes, u64 empty_size, + u64 hint_byte, struct btrfs_key *ins, + u64 flags, int delalloc) { int ret = 0; struct btrfs_root *root = orig_root->fs_info->extent_root; @@ -7430,8 +7440,6 @@ static noinline int find_free_extent(struct btrfs_root *orig_root, struct btrfs_space_info *space_info; int loop = 0; int index = __get_raid_index(flags); - int alloc_type = (flags & BTRFS_BLOCK_GROUP_DATA) ? - RESERVE_ALLOC_NO_ACCOUNT : RESERVE_ALLOC; bool failed_cluster_refill = false; bool failed_alloc = false; bool use_cluster = true; @@ -7763,8 +7771,8 @@ checks: search_start - offset); BUG_ON(offset > search_start); - ret = btrfs_update_reserved_bytes(block_group, num_bytes, - alloc_type, delalloc); + ret = btrfs_add_reserved_bytes(block_group, ram_bytes, + num_bytes, delalloc); if (ret == -EAGAIN) { btrfs_add_free_space(block_group, offset, num_bytes); goto loop; @@ -7896,23 +7904,24 @@ out: return ret; } -static void dump_space_info(struct btrfs_space_info *info, u64 bytes, +static void dump_space_info(struct btrfs_fs_info *fs_info, + struct btrfs_space_info *info, u64 bytes, int dump_block_groups) { struct btrfs_block_group_cache *cache; int index = 0; spin_lock(&info->lock); - printk(KERN_INFO "BTRFS: space_info %llu has %llu free, is %sfull\n", - info->flags, - info->total_bytes - info->bytes_used - info->bytes_pinned - - info->bytes_reserved - info->bytes_readonly - - info->bytes_may_use, (info->full) ? "" : "not "); - printk(KERN_INFO "BTRFS: space_info total=%llu, used=%llu, pinned=%llu, " - "reserved=%llu, may_use=%llu, readonly=%llu\n", - info->total_bytes, info->bytes_used, info->bytes_pinned, - info->bytes_reserved, info->bytes_may_use, - info->bytes_readonly); + btrfs_info(fs_info, "space_info %llu has %llu free, is %sfull", + info->flags, + info->total_bytes - info->bytes_used - info->bytes_pinned - + info->bytes_reserved - info->bytes_readonly - + info->bytes_may_use, (info->full) ? "" : "not "); + btrfs_info(fs_info, + "space_info total=%llu, used=%llu, pinned=%llu, reserved=%llu, may_use=%llu, readonly=%llu", + info->total_bytes, info->bytes_used, info->bytes_pinned, + info->bytes_reserved, info->bytes_may_use, + info->bytes_readonly); spin_unlock(&info->lock); if (!dump_block_groups) @@ -7922,12 +7931,11 @@ static void dump_space_info(struct btrfs_space_info *info, u64 bytes, again: list_for_each_entry(cache, &info->block_groups[index], list) { spin_lock(&cache->lock); - printk(KERN_INFO "BTRFS: " - "block group %llu has %llu bytes, " - "%llu used %llu pinned %llu reserved %s\n", - cache->key.objectid, cache->key.offset, - btrfs_block_group_used(&cache->item), cache->pinned, - cache->reserved, cache->ro ? "[readonly]" : ""); + btrfs_info(fs_info, + "block group %llu has %llu bytes, %llu used %llu pinned %llu reserved %s", + cache->key.objectid, cache->key.offset, + btrfs_block_group_used(&cache->item), cache->pinned, + cache->reserved, cache->ro ? "[readonly]" : ""); btrfs_dump_free_space(cache, bytes); spin_unlock(&cache->lock); } @@ -7936,11 +7944,12 @@ again: up_read(&info->groups_sem); } -int btrfs_reserve_extent(struct btrfs_root *root, +int btrfs_reserve_extent(struct btrfs_root *root, u64 ram_bytes, u64 num_bytes, u64 min_alloc_size, u64 empty_size, u64 hint_byte, struct btrfs_key *ins, int is_data, int delalloc) { + struct btrfs_fs_info *fs_info = root->fs_info; bool final_tried = num_bytes == min_alloc_size; u64 flags; int ret; @@ -7948,27 +7957,28 @@ int btrfs_reserve_extent(struct btrfs_root *root, flags = btrfs_get_alloc_profile(root, is_data); again: WARN_ON(num_bytes < root->sectorsize); - ret = find_free_extent(root, num_bytes, empty_size, hint_byte, ins, - flags, delalloc); + ret = find_free_extent(root, ram_bytes, num_bytes, empty_size, + hint_byte, ins, flags, delalloc); if (!ret && !is_data) { - btrfs_dec_block_group_reservations(root->fs_info, - ins->objectid); + btrfs_dec_block_group_reservations(fs_info, ins->objectid); } else if (ret == -ENOSPC) { if (!final_tried && ins->offset) { num_bytes = min(num_bytes >> 1, ins->offset); num_bytes = round_down(num_bytes, root->sectorsize); num_bytes = max(num_bytes, min_alloc_size); + ram_bytes = num_bytes; if (num_bytes == min_alloc_size) final_tried = true; goto again; - } else if (btrfs_test_opt(root->fs_info, ENOSPC_DEBUG)) { + } else if (btrfs_test_opt(fs_info, ENOSPC_DEBUG)) { struct btrfs_space_info *sinfo; - sinfo = __find_space_info(root->fs_info, flags); - btrfs_err(root->fs_info, "allocation failed flags %llu, wanted %llu", - flags, num_bytes); + sinfo = __find_space_info(fs_info, flags); + btrfs_err(root->fs_info, + "allocation failed flags %llu, wanted %llu", + flags, num_bytes); if (sinfo) - dump_space_info(sinfo, num_bytes, 1); + dump_space_info(fs_info, sinfo, num_bytes, 1); } } @@ -7995,7 +8005,7 @@ static int __btrfs_free_reserved_extent(struct btrfs_root *root, if (btrfs_test_opt(root->fs_info, DISCARD)) ret = btrfs_discard_extent(root, start, len, NULL); btrfs_add_free_space(cache, start, len); - btrfs_update_reserved_bytes(cache, len, RESERVE_FREE, delalloc); + btrfs_free_reserved_bytes(cache, len, delalloc); trace_btrfs_reserved_extent_free(root, start, len); } @@ -8208,6 +8218,7 @@ int btrfs_alloc_logged_file_extent(struct btrfs_trans_handle *trans, { int ret; struct btrfs_block_group_cache *block_group; + struct btrfs_space_info *space_info; /* * Mixed block groups will exclude before processing the log so we only @@ -8223,9 +8234,14 @@ int btrfs_alloc_logged_file_extent(struct btrfs_trans_handle *trans, if (!block_group) return -EINVAL; - ret = btrfs_update_reserved_bytes(block_group, ins->offset, - RESERVE_ALLOC_NO_ACCOUNT, 0); - BUG_ON(ret); /* logic error */ + space_info = block_group->space_info; + spin_lock(&space_info->lock); + spin_lock(&block_group->lock); + space_info->bytes_reserved += ins->offset; + block_group->reserved += ins->offset; + spin_unlock(&block_group->lock); + spin_unlock(&space_info->lock); + ret = alloc_reserved_file_extent(trans, root, 0, root_objectid, 0, owner, offset, ins, 1); btrfs_put_block_group(block_group); @@ -8368,7 +8384,7 @@ struct extent_buffer *btrfs_alloc_tree_block(struct btrfs_trans_handle *trans, if (IS_ERR(block_rsv)) return ERR_CAST(block_rsv); - ret = btrfs_reserve_extent(root, blocksize, blocksize, + ret = btrfs_reserve_extent(root, blocksize, blocksize, blocksize, empty_size, hint, &ins, 0, 0); if (ret) goto out_unuse; @@ -8450,7 +8466,6 @@ static noinline void reada_walk_down(struct btrfs_trans_handle *trans, u64 refs; u64 flags; u32 nritems; - u32 blocksize; struct btrfs_key key; struct extent_buffer *eb; int ret; @@ -8468,7 +8483,6 @@ static noinline void reada_walk_down(struct btrfs_trans_handle *trans, eb = path->nodes[wc->level]; nritems = btrfs_header_nritems(eb); - blocksize = root->nodesize; for (slot = path->slots[wc->level]; slot < nritems; slot++) { if (nread >= wc->reada_count) @@ -8521,35 +8535,6 @@ reada: wc->reada_slot = slot; } -/* - * These may not be seen by the usual inc/dec ref code so we have to - * add them here. - */ -static int record_one_subtree_extent(struct btrfs_trans_handle *trans, - struct btrfs_root *root, u64 bytenr, - u64 num_bytes) -{ - struct btrfs_qgroup_extent_record *qrecord; - struct btrfs_delayed_ref_root *delayed_refs; - - qrecord = kmalloc(sizeof(*qrecord), GFP_NOFS); - if (!qrecord) - return -ENOMEM; - - qrecord->bytenr = bytenr; - qrecord->num_bytes = num_bytes; - qrecord->old_roots = NULL; - - delayed_refs = &trans->transaction->delayed_refs; - spin_lock(&delayed_refs->lock); - if (btrfs_qgroup_insert_dirty_extent(trans->fs_info, - delayed_refs, qrecord)) - kfree(qrecord); - spin_unlock(&delayed_refs->lock); - - return 0; -} - static int account_leaf_items(struct btrfs_trans_handle *trans, struct btrfs_root *root, struct extent_buffer *eb) @@ -8561,7 +8546,7 @@ static int account_leaf_items(struct btrfs_trans_handle *trans, u64 bytenr, num_bytes; /* We can be called directly from walk_up_proc() */ - if (!root->fs_info->quota_enabled) + if (!test_bit(BTRFS_FS_QUOTA_ENABLED, &root->fs_info->flags)) return 0; for (i = 0; i < nr; i++) { @@ -8583,7 +8568,8 @@ static int account_leaf_items(struct btrfs_trans_handle *trans, num_bytes = btrfs_file_extent_disk_num_bytes(eb, fi); - ret = record_one_subtree_extent(trans, root, bytenr, num_bytes); + ret = btrfs_qgroup_insert_dirty_extent(trans, root->fs_info, + bytenr, num_bytes, GFP_NOFS); if (ret) return ret; } @@ -8669,7 +8655,7 @@ static int account_shared_subtree(struct btrfs_trans_handle *trans, BUG_ON(root_level < 0 || root_level > BTRFS_MAX_LEVEL); BUG_ON(root_eb == NULL); - if (!root->fs_info->quota_enabled) + if (!test_bit(BTRFS_FS_QUOTA_ENABLED, &root->fs_info->flags)) return 0; if (!extent_buffer_uptodate(root_eb)) { @@ -8732,8 +8718,9 @@ walk_down: btrfs_set_lock_blocking_rw(eb, BTRFS_READ_LOCK); path->locks[level] = BTRFS_READ_LOCK_BLOCKING; - ret = record_one_subtree_extent(trans, root, child_bytenr, - root->nodesize); + ret = btrfs_qgroup_insert_dirty_extent(trans, + root->fs_info, child_bytenr, + root->nodesize, GFP_NOFS); if (ret) goto out; } @@ -8899,14 +8886,13 @@ static noinline int do_walk_down(struct btrfs_trans_handle *trans, ret = btrfs_lookup_extent_info(trans, root, bytenr, level - 1, 1, &wc->refs[level - 1], &wc->flags[level - 1]); - if (ret < 0) { - btrfs_tree_unlock(next); - return ret; - } + if (ret < 0) + goto out_unlock; if (unlikely(wc->refs[level - 1] == 0)) { btrfs_err(root->fs_info, "Missing references."); - BUG(); + ret = -EIO; + goto out_unlock; } *lookup_info = 0; @@ -8958,7 +8944,12 @@ static noinline int do_walk_down(struct btrfs_trans_handle *trans, } level--; - BUG_ON(level != btrfs_header_level(next)); + ASSERT(level == btrfs_header_level(next)); + if (level != btrfs_header_level(next)) { + btrfs_err(root->fs_info, "mismatched level"); + ret = -EIO; + goto out_unlock; + } path->nodes[level] = next; path->slots[level] = 0; path->locks[level] = BTRFS_WRITE_LOCK_BLOCKING; @@ -8973,8 +8964,15 @@ skip: if (wc->flags[level] & BTRFS_BLOCK_FLAG_FULL_BACKREF) { parent = path->nodes[level]->start; } else { - BUG_ON(root->root_key.objectid != + ASSERT(root->root_key.objectid == btrfs_header_owner(path->nodes[level])); + if (root->root_key.objectid != + btrfs_header_owner(path->nodes[level])) { + btrfs_err(root->fs_info, + "mismatched block owner"); + ret = -EIO; + goto out_unlock; + } parent = 0; } @@ -8983,20 +8981,24 @@ skip: generation, level - 1); if (ret) { btrfs_err_rl(root->fs_info, - "Error " - "%d accounting shared subtree. Quota " - "is out of sync, rescan required.", - ret); + "Error %d accounting shared subtree. Quota is out of sync, rescan required.", + ret); } } ret = btrfs_free_extent(trans, root, bytenr, blocksize, parent, root->root_key.objectid, level - 1, 0); - BUG_ON(ret); /* -ENOMEM */ + if (ret) + goto out_unlock; } + + *lookup_info = 1; + ret = 1; + +out_unlock: btrfs_tree_unlock(next); free_extent_buffer(next); - *lookup_info = 1; - return 1; + + return ret; } /* @@ -9076,10 +9078,8 @@ static noinline int walk_up_proc(struct btrfs_trans_handle *trans, ret = account_leaf_items(trans, root, eb); if (ret) { btrfs_err_rl(root->fs_info, - "error " - "%d accounting leaf items. Quota " - "is out of sync, rescan required.", - ret); + "error %d accounting leaf items. Quota is out of sync, rescan required.", + ret); } } /* make block locked assertion in clean_tree_block happy */ @@ -9195,9 +9195,10 @@ int btrfs_drop_snapshot(struct btrfs_root *root, struct btrfs_block_rsv *block_rsv, int update_ref, int for_reloc) { + struct btrfs_fs_info *fs_info = root->fs_info; struct btrfs_path *path; struct btrfs_trans_handle *trans; - struct btrfs_root *tree_root = root->fs_info->tree_root; + struct btrfs_root *tree_root = fs_info->tree_root; struct btrfs_root_item *root_item = &root->root_item; struct walk_control *wc; struct btrfs_key key; @@ -9206,7 +9207,7 @@ int btrfs_drop_snapshot(struct btrfs_root *root, int level; bool root_dropped = false; - btrfs_debug(root->fs_info, "Drop subvolume %llu", root->objectid); + btrfs_debug(fs_info, "Drop subvolume %llu", root->objectid); path = btrfs_alloc_path(); if (!path) { @@ -9335,7 +9336,8 @@ int btrfs_drop_snapshot(struct btrfs_root *root, btrfs_end_transaction_throttle(trans, tree_root); if (!for_reloc && btrfs_need_cleaner_sleep(root)) { - pr_debug("BTRFS: drop snapshot early exit\n"); + btrfs_debug(fs_info, + "drop snapshot early exit"); err = -EAGAIN; goto out_free; } @@ -9401,7 +9403,7 @@ out: if (!for_reloc && root_dropped == false) btrfs_add_dead_root(root); if (err && err != -EAGAIN) - btrfs_handle_fs_error(root->fs_info, err, NULL); + btrfs_handle_fs_error(fs_info, err, NULL); return err; } @@ -9906,6 +9908,7 @@ static int find_first_block_group(struct btrfs_root *root, } else { ret = 0; } + free_extent_map(em); goto out; } path->slots[0]++; @@ -9942,6 +9945,7 @@ void btrfs_put_block_group_cache(struct btrfs_fs_info *info) block_group->iref = 0; block_group->inode = NULL; spin_unlock(&block_group->lock); + ASSERT(block_group->io_ctl.inode == NULL); iput(inode); last = block_group->key.objectid + block_group->key.offset; btrfs_put_block_group(block_group); @@ -9999,6 +10003,10 @@ int btrfs_free_block_groups(struct btrfs_fs_info *info) free_excluded_extents(info->extent_root, block_group); btrfs_remove_free_space_cache(block_group); + ASSERT(list_empty(&block_group->dirty_list)); + ASSERT(list_empty(&block_group->io_list)); + ASSERT(list_empty(&block_group->bg_list)); + ASSERT(atomic_read(&block_group->count) == 1); btrfs_put_block_group(block_group); spin_lock(&info->block_group_cache_lock); @@ -10029,7 +10037,7 @@ int btrfs_free_block_groups(struct btrfs_fs_info *info) if (WARN_ON(space_info->bytes_pinned > 0 || space_info->bytes_reserved > 0 || space_info->bytes_may_use > 0)) - dump_space_info(space_info, 0, 0); + dump_space_info(info, space_info, 0, 0); list_del(&space_info->list); for (i = 0; i < BTRFS_NR_RAID_TYPES; i++) { struct kobject *kobj; @@ -10078,7 +10086,8 @@ static void __link_block_group(struct btrfs_space_info *space_info, return; out_err: - pr_warn("BTRFS: failed to add kobject for block cache. ignoring.\n"); + btrfs_warn(cache->fs_info, + "failed to add kobject for block cache, ignoring"); } static struct btrfs_block_group_cache * @@ -10136,6 +10145,11 @@ int btrfs_read_block_groups(struct btrfs_root *root) struct extent_buffer *leaf; int need_clear = 0; u64 cache_gen; + u64 feature; + int mixed; + + feature = btrfs_super_incompat_flags(info->super_copy); + mixed = !!(feature & BTRFS_FEATURE_INCOMPAT_MIXED_GROUPS); root = info->extent_root; key.objectid = 0; @@ -10189,6 +10203,15 @@ int btrfs_read_block_groups(struct btrfs_root *root) btrfs_item_ptr_offset(leaf, path->slots[0]), sizeof(cache->item)); cache->flags = btrfs_block_group_flags(&cache->item); + if (!mixed && + ((cache->flags & BTRFS_BLOCK_GROUP_METADATA) && + (cache->flags & BTRFS_BLOCK_GROUP_DATA))) { + btrfs_err(info, +"bg %llu is a mixed block group but filesystem hasn't enabled mixed block groups", + cache->key.objectid); + ret = -EINVAL; + goto error; + } key.objectid = found_key.objectid + found_key.offset; btrfs_release_path(path); @@ -10798,7 +10821,7 @@ void btrfs_delete_unused_bgs(struct btrfs_fs_info *fs_info) struct btrfs_trans_handle *trans; int ret = 0; - if (!fs_info->open) + if (!test_bit(BTRFS_FS_OPEN, &fs_info->flags)) return; spin_lock(&fs_info->unused_bgs_lock); |