diff options
Diffstat (limited to 'fs/btrfs/tree-log.c')
| -rw-r--r-- | fs/btrfs/tree-log.c | 1547 |
1 files changed, 972 insertions, 575 deletions
diff --git a/fs/btrfs/tree-log.c b/fs/btrfs/tree-log.c index dcf75a8daa20..813986e38258 100644 --- a/fs/btrfs/tree-log.c +++ b/fs/btrfs/tree-log.c @@ -22,6 +22,8 @@ #include "zoned.h" #include "inode-item.h" +#define MAX_CONFLICT_INODES 10 + /* magic values for the inode_only field in btrfs_log_inode: * * LOG_INODE_ALL means to log everything @@ -31,8 +33,6 @@ enum { LOG_INODE_ALL, LOG_INODE_EXISTS, - LOG_OTHER_INODE, - LOG_OTHER_INODE_ALL, }; /* @@ -801,7 +801,7 @@ static noinline int replay_one_extent(struct btrfs_trans_handle *trans, ret = btrfs_lookup_csums_range(root->log_root, csum_start, csum_end - 1, - &ordered_sums, 0); + &ordered_sums, 0, false); if (ret) goto out; /* @@ -1063,8 +1063,7 @@ static inline int __add_inode_ref(struct btrfs_trans_handle *trans, struct btrfs_inode *dir, struct btrfs_inode *inode, u64 inode_objectid, u64 parent_objectid, - u64 ref_index, char *name, int namelen, - int *search_done) + u64 ref_index, char *name, int namelen) { int ret; char *victim_name; @@ -1126,19 +1125,12 @@ again: kfree(victim_name); if (ret) return ret; - *search_done = 1; goto again; } kfree(victim_name); ptr = (unsigned long)(victim_ref + 1) + victim_name_len; } - - /* - * NOTE: we have searched root tree and checked the - * corresponding ref, it does not need to check again. - */ - *search_done = 1; } btrfs_release_path(path); @@ -1146,7 +1138,9 @@ again: extref = btrfs_lookup_inode_extref(NULL, root, path, name, namelen, inode_objectid, parent_objectid, 0, 0); - if (!IS_ERR_OR_NULL(extref)) { + if (IS_ERR(extref)) { + return PTR_ERR(extref); + } else if (extref) { u32 item_size; u32 cur_offset = 0; unsigned long base; @@ -1200,14 +1194,12 @@ again: kfree(victim_name); if (ret) return ret; - *search_done = 1; goto again; } kfree(victim_name); next: cur_offset += victim_name_len + sizeof(*extref); } - *search_done = 1; } btrfs_release_path(path); @@ -1371,103 +1363,6 @@ again: return ret; } -static int btrfs_inode_ref_exists(struct inode *inode, struct inode *dir, - const u8 ref_type, const char *name, - const int namelen) -{ - struct btrfs_key key; - struct btrfs_path *path; - const u64 parent_id = btrfs_ino(BTRFS_I(dir)); - int ret; - - path = btrfs_alloc_path(); - if (!path) - return -ENOMEM; - - key.objectid = btrfs_ino(BTRFS_I(inode)); - key.type = ref_type; - if (key.type == BTRFS_INODE_REF_KEY) - key.offset = parent_id; - else - key.offset = btrfs_extref_hash(parent_id, name, namelen); - - ret = btrfs_search_slot(NULL, BTRFS_I(inode)->root, &key, path, 0, 0); - if (ret < 0) - goto out; - if (ret > 0) { - ret = 0; - goto out; - } - if (key.type == BTRFS_INODE_EXTREF_KEY) - ret = !!btrfs_find_name_in_ext_backref(path->nodes[0], - path->slots[0], parent_id, name, namelen); - else - ret = !!btrfs_find_name_in_backref(path->nodes[0], path->slots[0], - name, namelen); - -out: - btrfs_free_path(path); - return ret; -} - -static int add_link(struct btrfs_trans_handle *trans, - struct inode *dir, struct inode *inode, const char *name, - int namelen, u64 ref_index) -{ - struct btrfs_root *root = BTRFS_I(dir)->root; - struct btrfs_dir_item *dir_item; - struct btrfs_key key; - struct btrfs_path *path; - struct inode *other_inode = NULL; - int ret; - - path = btrfs_alloc_path(); - if (!path) - return -ENOMEM; - - dir_item = btrfs_lookup_dir_item(NULL, root, path, - btrfs_ino(BTRFS_I(dir)), - name, namelen, 0); - if (!dir_item) { - btrfs_release_path(path); - goto add_link; - } else if (IS_ERR(dir_item)) { - ret = PTR_ERR(dir_item); - goto out; - } - - /* - * Our inode's dentry collides with the dentry of another inode which is - * in the log but not yet processed since it has a higher inode number. - * So delete that other dentry. - */ - btrfs_dir_item_key_to_cpu(path->nodes[0], dir_item, &key); - btrfs_release_path(path); - other_inode = read_one_inode(root, key.objectid); - if (!other_inode) { - ret = -ENOENT; - goto out; - } - ret = unlink_inode_for_log_replay(trans, BTRFS_I(dir), BTRFS_I(other_inode), - name, namelen); - if (ret) - goto out; - /* - * If we dropped the link count to 0, bump it so that later the iput() - * on the inode will not free it. We will fixup the link count later. - */ - if (other_inode->i_nlink == 0) - inc_nlink(other_inode); -add_link: - ret = btrfs_add_link(trans, BTRFS_I(dir), BTRFS_I(inode), - name, namelen, 0, ref_index); -out: - iput(other_inode); - btrfs_free_path(path); - - return ret; -} - /* * replay one inode back reference item found in the log tree. * eb, slot and key refer to the buffer and key found in the log tree. @@ -1488,7 +1383,6 @@ static noinline int add_inode_ref(struct btrfs_trans_handle *trans, char *name = NULL; int namelen; int ret; - int search_done = 0; int log_ref_ver = 0; u64 parent_objectid; u64 inode_objectid; @@ -1563,51 +1457,19 @@ static noinline int add_inode_ref(struct btrfs_trans_handle *trans, * overwrite any existing back reference, and we don't * want to create dangling pointers in the directory. */ - - if (!search_done) { - ret = __add_inode_ref(trans, root, path, log, - BTRFS_I(dir), - BTRFS_I(inode), - inode_objectid, - parent_objectid, - ref_index, name, namelen, - &search_done); - if (ret) { - if (ret == 1) - ret = 0; - goto out; - } - } - - /* - * If a reference item already exists for this inode - * with the same parent and name, but different index, - * drop it and the corresponding directory index entries - * from the parent before adding the new reference item - * and dir index entries, otherwise we would fail with - * -EEXIST returned from btrfs_add_link() below. - */ - ret = btrfs_inode_ref_exists(inode, dir, key->type, - name, namelen); - if (ret > 0) { - ret = unlink_inode_for_log_replay(trans, - BTRFS_I(dir), - BTRFS_I(inode), - name, namelen); - /* - * If we dropped the link count to 0, bump it so - * that later the iput() on the inode will not - * free it. We will fixup the link count later. - */ - if (!ret && inode->i_nlink == 0) - inc_nlink(inode); - } - if (ret < 0) + ret = __add_inode_ref(trans, root, path, log, + BTRFS_I(dir), BTRFS_I(inode), + inode_objectid, parent_objectid, + ref_index, name, namelen); + if (ret) { + if (ret == 1) + ret = 0; goto out; + } /* insert our name */ - ret = add_link(trans, dir, inode, name, namelen, - ref_index); + ret = btrfs_add_link(trans, BTRFS_I(dir), BTRFS_I(inode), + name, namelen, 0, ref_index); if (ret) goto out; @@ -3873,6 +3735,11 @@ static int process_dir_items_leaf(struct btrfs_trans_handle *trans, *last_old_dentry_offset = key.offset; continue; } + + /* If we logged this dir index item before, we can skip it. */ + if (key.offset <= inode->last_dir_index_offset) + continue; + /* * We must make sure that when we log a directory entry, the * corresponding inode, after log replay, has a matching link @@ -3903,51 +3770,6 @@ static int process_dir_items_leaf(struct btrfs_trans_handle *trans, ctx->log_new_dentries = true; } - if (!ctx->logged_before) - goto add_to_batch; - - /* - * If we were logged before and have logged dir items, we can skip - * checking if any item with a key offset larger than the last one - * we logged is in the log tree, saving time and avoiding adding - * contention on the log tree. We can only rely on the value of - * last_dir_index_offset when we know for sure that the inode was - * previously logged in the current transaction. - */ - if (key.offset > inode->last_dir_index_offset) - goto add_to_batch; - /* - * Check if the key was already logged before. If not we can add - * it to a batch for bulk insertion. - */ - ret = btrfs_search_slot(NULL, log, &key, dst_path, 0, 0); - if (ret < 0) { - return ret; - } else if (ret > 0) { - btrfs_release_path(dst_path); - goto add_to_batch; - } - - /* - * Item exists in the log. Overwrite the item in the log if it - * has different content or do nothing if it has exactly the same - * content. And then flush the current batch if any - do it after - * overwriting the current item, or we would deadlock otherwise, - * since we are holding a path for the existing item. - */ - ret = do_overwrite_item(trans, log, dst_path, src, i, &key); - if (ret < 0) - return ret; - - if (batch_size > 0) { - ret = flush_dir_items_batch(trans, log, src, dst_path, - batch_start, batch_size); - if (ret < 0) - return ret; - batch_size = 0; - } - continue; -add_to_batch: if (batch_size == 0) batch_start = i; batch_size++; @@ -4134,6 +3956,71 @@ done: } /* + * If the inode was logged before and it was evicted, then its + * last_dir_index_offset is (u64)-1, so we don't the value of the last index + * key offset. If that's the case, search for it and update the inode. This + * is to avoid lookups in the log tree every time we try to insert a dir index + * key from a leaf changed in the current transaction, and to allow us to always + * do batch insertions of dir index keys. + */ +static int update_last_dir_index_offset(struct btrfs_inode *inode, + struct btrfs_path *path, + const struct btrfs_log_ctx *ctx) +{ + const u64 ino = btrfs_ino(inode); + struct btrfs_key key; + int ret; + + lockdep_assert_held(&inode->log_mutex); + + if (inode->last_dir_index_offset != (u64)-1) + return 0; + + if (!ctx->logged_before) { + inode->last_dir_index_offset = BTRFS_DIR_START_INDEX - 1; + return 0; + } + + key.objectid = ino; + key.type = BTRFS_DIR_INDEX_KEY; + key.offset = (u64)-1; + + ret = btrfs_search_slot(NULL, inode->root->log_root, &key, path, 0, 0); + /* + * An error happened or we actually have an index key with an offset + * value of (u64)-1. Bail out, we're done. + */ + if (ret <= 0) + goto out; + + ret = 0; + inode->last_dir_index_offset = BTRFS_DIR_START_INDEX - 1; + + /* + * No dir index items, bail out and leave last_dir_index_offset with + * the value right before the first valid index value. + */ + if (path->slots[0] == 0) + goto out; + + /* + * btrfs_search_slot() left us at one slot beyond the slot with the last + * index key, or beyond the last key of the directory that is not an + * index key. If we have an index key before, set last_dir_index_offset + * to its offset value, otherwise leave it with a value right before the + * first valid index value, as it means we have an empty directory. + */ + btrfs_item_key_to_cpu(path->nodes[0], &key, path->slots[0] - 1); + if (key.objectid == ino && key.type == BTRFS_DIR_INDEX_KEY) + inode->last_dir_index_offset = key.offset; + +out: + btrfs_release_path(path); + + return ret; +} + +/* * logging directories is very similar to logging inodes, We find all the items * from the current transaction and write them to the log. * @@ -4155,6 +4042,10 @@ static noinline int log_directory_changes(struct btrfs_trans_handle *trans, u64 max_key; int ret; + ret = update_last_dir_index_offset(inode, path, ctx); + if (ret) + return ret; + min_key = BTRFS_DIR_START_INDEX; max_key = 0; ctx->last_dir_item_offset = inode->last_dir_index_offset; @@ -4380,8 +4271,8 @@ static int log_csums(struct btrfs_trans_handle *trans, * file which happens to refer to the same extent as well. Such races * can leave checksum items in the log with overlapping ranges. */ - ret = lock_extent_bits(&log_root->log_csum_range, sums->bytenr, - lock_end, &cached_state); + ret = lock_extent(&log_root->log_csum_range, sums->bytenr, lock_end, + &cached_state); if (ret) return ret; /* @@ -4397,8 +4288,8 @@ static int log_csums(struct btrfs_trans_handle *trans, if (!ret) ret = btrfs_csum_file_blocks(trans, log_root, sums); - unlock_extent_cached(&log_root->log_csum_range, sums->bytenr, lock_end, - &cached_state); + unlock_extent(&log_root->log_csum_range, sums->bytenr, lock_end, + &cached_state); return ret; } @@ -4511,7 +4402,7 @@ static noinline int copy_items(struct btrfs_trans_handle *trans, disk_bytenr += extent_offset; ret = btrfs_lookup_csums_range(csum_root, disk_bytenr, disk_bytenr + extent_num_bytes - 1, - &ordered_sums, 0); + &ordered_sums, 0, false); if (ret) goto out; @@ -4707,7 +4598,7 @@ static int log_extent_csums(struct btrfs_trans_handle *trans, ret = btrfs_lookup_csums_range(csum_root, em->block_start + csum_offset, em->block_start + csum_offset + - csum_len - 1, &ordered_sums, 0); + csum_len - 1, &ordered_sums, 0, false); if (ret) return ret; @@ -5219,10 +5110,9 @@ static int btrfs_log_holes(struct btrfs_trans_handle *trans, * leafs from the log root. */ btrfs_release_path(path); - ret = btrfs_insert_file_extent(trans, root->log_root, - ino, prev_extent_end, 0, - 0, hole_len, 0, hole_len, - 0, 0, 0); + ret = btrfs_insert_hole_extent(trans, root->log_root, + ino, prev_extent_end, + hole_len); if (ret < 0) return ret; @@ -5251,10 +5141,8 @@ static int btrfs_log_holes(struct btrfs_trans_handle *trans, btrfs_release_path(path); hole_len = ALIGN(i_size - prev_extent_end, fs_info->sectorsize); - ret = btrfs_insert_file_extent(trans, root->log_root, - ino, prev_extent_end, 0, 0, - hole_len, 0, hole_len, - 0, 0, 0); + ret = btrfs_insert_hole_extent(trans, root->log_root, ino, + prev_extent_end, hole_len); if (ret < 0) return ret; } @@ -5397,111 +5285,461 @@ out: return ret; } +/* + * Check if we need to log an inode. This is used in contexts where while + * logging an inode we need to log another inode (either that it exists or in + * full mode). This is used instead of btrfs_inode_in_log() because the later + * requires the inode to be in the log and have the log transaction committed, + * while here we do not care if the log transaction was already committed - our + * caller will commit the log later - and we want to avoid logging an inode + * multiple times when multiple tasks have joined the same log transaction. + */ +static bool need_log_inode(const struct btrfs_trans_handle *trans, + const struct btrfs_inode *inode) +{ + /* + * If a directory was not modified, no dentries added or removed, we can + * and should avoid logging it. + */ + if (S_ISDIR(inode->vfs_inode.i_mode) && inode->last_trans < trans->transid) + return false; + + /* + * If this inode does not have new/updated/deleted xattrs since the last + * time it was logged and is flagged as logged in the current transaction, + * we can skip logging it. As for new/deleted names, those are updated in + * the log by link/unlink/rename operations. + * In case the inode was logged and then evicted and reloaded, its + * logged_trans will be 0, in which case we have to fully log it since + * logged_trans is a transient field, not persisted. + */ + if (inode->logged_trans == trans->transid && + !test_bit(BTRFS_INODE_COPY_EVERYTHING, &inode->runtime_flags)) + return false; + + return true; +} + +struct btrfs_dir_list { + u64 ino; + struct list_head list; +}; + +/* + * Log the inodes of the new dentries of a directory. + * See process_dir_items_leaf() for details about why it is needed. + * This is a recursive operation - if an existing dentry corresponds to a + * directory, that directory's new entries are logged too (same behaviour as + * ext3/4, xfs, f2fs, reiserfs, nilfs2). Note that when logging the inodes + * the dentries point to we do not acquire their VFS lock, otherwise lockdep + * complains about the following circular lock dependency / possible deadlock: + * + * CPU0 CPU1 + * ---- ---- + * lock(&type->i_mutex_dir_key#3/2); + * lock(sb_internal#2); + * lock(&type->i_mutex_dir_key#3/2); + * lock(&sb->s_type->i_mutex_key#14); + * + * Where sb_internal is the lock (a counter that works as a lock) acquired by + * sb_start_intwrite() in btrfs_start_transaction(). + * Not acquiring the VFS lock of the inodes is still safe because: + * + * 1) For regular files we log with a mode of LOG_INODE_EXISTS. It's possible + * that while logging the inode new references (names) are added or removed + * from the inode, leaving the logged inode item with a link count that does + * not match the number of logged inode reference items. This is fine because + * at log replay time we compute the real number of links and correct the + * link count in the inode item (see replay_one_buffer() and + * link_to_fixup_dir()); + * + * 2) For directories we log with a mode of LOG_INODE_ALL. It's possible that + * while logging the inode's items new index items (key type + * BTRFS_DIR_INDEX_KEY) are added to fs/subvol tree and the logged inode item + * has a size that doesn't match the sum of the lengths of all the logged + * names - this is ok, not a problem, because at log replay time we set the + * directory's i_size to the correct value (see replay_one_name() and + * do_overwrite_item()). + */ +static int log_new_dir_dentries(struct btrfs_trans_handle *trans, + struct btrfs_inode *start_inode, + struct btrfs_log_ctx *ctx) +{ + struct btrfs_root *root = start_inode->root; + struct btrfs_fs_info *fs_info = root->fs_info; + struct btrfs_path *path; + LIST_HEAD(dir_list); + struct btrfs_dir_list *dir_elem; + u64 ino = btrfs_ino(start_inode); + int ret = 0; + + /* + * If we are logging a new name, as part of a link or rename operation, + * don't bother logging new dentries, as we just want to log the names + * of an inode and that any new parents exist. + */ + if (ctx->logging_new_name) + return 0; + + path = btrfs_alloc_path(); + if (!path) + return -ENOMEM; + + while (true) { + struct extent_buffer *leaf; + struct btrfs_key min_key; + bool continue_curr_inode = true; + int nritems; + int i; + + min_key.objectid = ino; + min_key.type = BTRFS_DIR_INDEX_KEY; + min_key.offset = 0; +again: + btrfs_release_path(path); + ret = btrfs_search_forward(root, &min_key, path, trans->transid); + if (ret < 0) { + break; + } else if (ret > 0) { + ret = 0; + goto next; + } + + leaf = path->nodes[0]; + nritems = btrfs_header_nritems(leaf); + for (i = path->slots[0]; i < nritems; i++) { + struct btrfs_dir_item *di; + struct btrfs_key di_key; + struct inode *di_inode; + int log_mode = LOG_INODE_EXISTS; + int type; + + btrfs_item_key_to_cpu(leaf, &min_key, i); + if (min_key.objectid != ino || + min_key.type != BTRFS_DIR_INDEX_KEY) { + continue_curr_inode = false; + break; + } + + di = btrfs_item_ptr(leaf, i, struct btrfs_dir_item); + type = btrfs_dir_type(leaf, di); + if (btrfs_dir_transid(leaf, di) < trans->transid) + continue; + btrfs_dir_item_key_to_cpu(leaf, di, &di_key); + if (di_key.type == BTRFS_ROOT_ITEM_KEY) + continue; + + btrfs_release_path(path); + di_inode = btrfs_iget(fs_info->sb, di_key.objectid, root); + if (IS_ERR(di_inode)) { + ret = PTR_ERR(di_inode); + goto out; + } + + if (!need_log_inode(trans, BTRFS_I(di_inode))) { + btrfs_add_delayed_iput(di_inode); + break; + } + + ctx->log_new_dentries = false; + if (type == BTRFS_FT_DIR) + log_mode = LOG_INODE_ALL; + ret = btrfs_log_inode(trans, BTRFS_I(di_inode), + log_mode, ctx); + btrfs_add_delayed_iput(di_inode); + if (ret) + goto out; + if (ctx->log_new_dentries) { + dir_elem = kmalloc(sizeof(*dir_elem), GFP_NOFS); + if (!dir_elem) { + ret = -ENOMEM; + goto out; + } + dir_elem->ino = di_key.objectid; + list_add_tail(&dir_elem->list, &dir_list); + } + break; + } + + if (continue_curr_inode && min_key.offset < (u64)-1) { + min_key.offset++; + goto again; + } + +next: + if (list_empty(&dir_list)) + break; + + dir_elem = list_first_entry(&dir_list, struct btrfs_dir_list, list); + ino = dir_elem->ino; + list_del(&dir_elem->list); + kfree(dir_elem); + } +out: + btrfs_free_path(path); + if (ret) { + struct btrfs_dir_list *next; + + list_for_each_entry_safe(dir_elem, next, &dir_list, list) + kfree(dir_elem); + } + + return ret; +} + struct btrfs_ino_list { u64 ino; u64 parent; struct list_head list; }; -static int log_conflicting_inodes(struct btrfs_trans_handle *trans, - struct btrfs_root *root, - struct btrfs_path *path, - struct btrfs_log_ctx *ctx, - u64 ino, u64 parent) +static void free_conflicting_inodes(struct btrfs_log_ctx *ctx) +{ + struct btrfs_ino_list *curr; + struct btrfs_ino_list *next; + + list_for_each_entry_safe(curr, next, &ctx->conflict_inodes, list) { + list_del(&curr->list); + kfree(curr); + } +} + +static int conflicting_inode_is_dir(struct btrfs_root *root, u64 ino, + struct btrfs_path *path) +{ + struct btrfs_key key; + int ret; + + key.objectid = ino; + key.type = BTRFS_INODE_ITEM_KEY; + key.offset = 0; + + path->search_commit_root = 1; + path->skip_locking = 1; + + ret = btrfs_search_slot(NULL, root, &key, path, 0, 0); + if (WARN_ON_ONCE(ret > 0)) { + /* + * We have previously found the inode through the commit root + * so this should not happen. If it does, just error out and + * fallback to a transaction commit. + */ + ret = -ENOENT; + } else if (ret == 0) { + struct btrfs_inode_item *item; + + item = btrfs_item_ptr(path->nodes[0], path->slots[0], + struct btrfs_inode_item); + if (S_ISDIR(btrfs_inode_mode(path->nodes[0], item))) + ret = 1; + } + + btrfs_release_path(path); + path->search_commit_root = 0; + path->skip_locking = 0; + + return ret; +} + +static int add_conflicting_inode(struct btrfs_trans_handle *trans, + struct btrfs_root *root, + struct btrfs_path *path, + u64 ino, u64 parent, + struct btrfs_log_ctx *ctx) { struct btrfs_ino_list *ino_elem; - LIST_HEAD(inode_list); - int ret = 0; + struct inode *inode; + + /* + * It's rare to have a lot of conflicting inodes, in practice it is not + * common to have more than 1 or 2. We don't want to collect too many, + * as we could end up logging too many inodes (even if only in + * LOG_INODE_EXISTS mode) and slow down other fsyncs or transaction + * commits. + */ + if (ctx->num_conflict_inodes >= MAX_CONFLICT_INODES) + return BTRFS_LOG_FORCE_COMMIT; + + inode = btrfs_iget(root->fs_info->sb, ino, root); + /* + * If the other inode that had a conflicting dir entry was deleted in + * the current transaction then we either: + * + * 1) Log the parent directory (later after adding it to the list) if + * the inode is a directory. This is because it may be a deleted + * subvolume/snapshot or it may be a regular directory that had + * deleted subvolumes/snapshots (or subdirectories that had them), + * and at the moment we can't deal with dropping subvolumes/snapshots + * during log replay. So we just log the parent, which will result in + * a fallback to a transaction commit if we are dealing with those + * cases (last_unlink_trans will match the current transaction); + * + * 2) Do nothing if it's not a directory. During log replay we simply + * unlink the conflicting dentry from the parent directory and then + * add the dentry for our inode. Like this we can avoid logging the + * parent directory (and maybe fallback to a transaction commit in + * case it has a last_unlink_trans == trans->transid, due to moving + * some inode from it to some other directory). + */ + if (IS_ERR(inode)) { + int ret = PTR_ERR(inode); + + if (ret != -ENOENT) + return ret; + + ret = conflicting_inode_is_dir(root, ino, path); + /* Not a directory or we got an error. */ + if (ret <= 0) + return ret; + + /* Conflicting inode is a directory, so we'll log its parent. */ + ino_elem = kmalloc(sizeof(*ino_elem), GFP_NOFS); + if (!ino_elem) + return -ENOMEM; + ino_elem->ino = ino; + ino_elem->parent = parent; + list_add_tail(&ino_elem->list, &ctx->conflict_inodes); + ctx->num_conflict_inodes++; + + return 0; + } + + /* + * If the inode was already logged skip it - otherwise we can hit an + * infinite loop. Example: + * + * From the commit root (previous transaction) we have the following + * inodes: + * + * inode 257 a directory + * inode 258 with references "zz" and "zz_link" on inode 257 + * inode 259 with reference "a" on inode 257 + * + * And in the current (uncommitted) transaction we have: + * + * inode 257 a directory, unchanged + * inode 258 with references "a" and "a2" on inode 257 + * inode 259 with reference "zz_link" on inode 257 + * inode 261 with reference "zz" on inode 257 + * + * When logging inode 261 the following infinite loop could + * happen if we don't skip already logged inodes: + * + * - we detect inode 258 as a conflicting inode, with inode 261 + * on reference "zz", and log it; + * + * - we detect inode 259 as a conflicting inode, with inode 258 + * on reference "a", and log it; + * + * - we detect inode 258 as a conflicting inode, with inode 259 + * on reference "zz_link", and log it - again! After this we + * repeat the above steps forever. + * + * Here we can use need_log_inode() because we only need to log the + * inode in LOG_INODE_EXISTS mode and rename operations update the log, + * so that the log ends up with the new name and without the old name. + */ + if (!need_log_inode(trans, BTRFS_I(inode))) { + btrfs_add_delayed_iput(inode); + return 0; + } + + btrfs_add_delayed_iput(inode); ino_elem = kmalloc(sizeof(*ino_elem), GFP_NOFS); if (!ino_elem) return -ENOMEM; ino_elem->ino = ino; ino_elem->parent = parent; - list_add_tail(&ino_elem->list, &inode_list); + list_add_tail(&ino_elem->list, &ctx->conflict_inodes); + ctx->num_conflict_inodes++; - while (!list_empty(&inode_list)) { - struct btrfs_fs_info *fs_info = root->fs_info; - struct btrfs_key key; - struct inode *inode; + return 0; +} - ino_elem = list_first_entry(&inode_list, struct btrfs_ino_list, - list); - ino = ino_elem->ino; - parent = ino_elem->parent; - list_del(&ino_elem->list); - kfree(ino_elem); - if (ret) - continue; +static int log_conflicting_inodes(struct btrfs_trans_handle *trans, + struct btrfs_root *root, + struct btrfs_log_ctx *ctx) +{ + struct btrfs_fs_info *fs_info = root->fs_info; + int ret = 0; - btrfs_release_path(path); + /* + * Conflicting inodes are logged by the first call to btrfs_log_inode(), + * otherwise we could have unbounded recursion of btrfs_log_inode() + * calls. This check guarantees we can have only 1 level of recursion. + */ + if (ctx->logging_conflict_inodes) + return 0; + + ctx->logging_conflict_inodes = true; + + /* + * New conflicting inodes may be found and added to the list while we + * are logging a conflicting inode, so keep iterating while the list is + * not empty. + */ + while (!list_empty(&ctx->conflict_inodes)) { + struct btrfs_ino_list *curr; + struct inode *inode; + u64 ino; + u64 parent; + + curr = list_first_entry(&ctx->conflict_inodes, + struct btrfs_ino_list, list); + ino = curr->ino; + parent = curr->parent; + list_del(&curr->list); + kfree(curr); inode = btrfs_iget(fs_info->sb, ino, root); /* * If the other inode that had a conflicting dir entry was * deleted in the current transaction, we need to log its parent - * directory. + * directory. See the comment at add_conflicting_inode(). */ if (IS_ERR(inode)) { ret = PTR_ERR(inode); - if (ret == -ENOENT) { - inode = btrfs_iget(fs_info->sb, parent, root); - if (IS_ERR(inode)) { - ret = PTR_ERR(inode); - } else { - ret = btrfs_log_inode(trans, - BTRFS_I(inode), - LOG_OTHER_INODE_ALL, - ctx); - btrfs_add_delayed_iput(inode); - } + if (ret != -ENOENT) + break; + + inode = btrfs_iget(fs_info->sb, parent, root); + if (IS_ERR(inode)) { + ret = PTR_ERR(inode); + break; } + + /* + * Always log the directory, we cannot make this + * conditional on need_log_inode() because the directory + * might have been logged in LOG_INODE_EXISTS mode or + * the dir index of the conflicting inode is not in a + * dir index key range logged for the directory. So we + * must make sure the deletion is recorded. + */ + ret = btrfs_log_inode(trans, BTRFS_I(inode), + LOG_INODE_ALL, ctx); + btrfs_add_delayed_iput(inode); + if (ret) + break; continue; } + /* - * If the inode was already logged skip it - otherwise we can - * hit an infinite loop. Example: - * - * From the commit root (previous transaction) we have the - * following inodes: + * Here we can use need_log_inode() because we only need to log + * the inode in LOG_INODE_EXISTS mode and rename operations + * update the log, so that the log ends up with the new name and + * without the old name. * - * inode 257 a directory - * inode 258 with references "zz" and "zz_link" on inode 257 - * inode 259 with reference "a" on inode 257 - * - * And in the current (uncommitted) transaction we have: - * - * inode 257 a directory, unchanged - * inode 258 with references "a" and "a2" on inode 257 - * inode 259 with reference "zz_link" on inode 257 - * inode 261 with reference "zz" on inode 257 - * - * When logging inode 261 the following infinite loop could - * happen if we don't skip already logged inodes: - * - * - we detect inode 258 as a conflicting inode, with inode 261 - * on reference "zz", and log it; - * - * - we detect inode 259 as a conflicting inode, with inode 258 - * on reference "a", and log it; - * - * - we detect inode 258 as a conflicting inode, with inode 259 - * on reference "zz_link", and log it - again! After this we - * repeat the above steps forever. + * We did this check at add_conflicting_inode(), but here we do + * it again because if some other task logged the inode after + * that, we can avoid doing it again. */ - spin_lock(&BTRFS_I(inode)->lock); - /* - * Check the inode's logged_trans only instead of - * btrfs_inode_in_log(). This is because the last_log_commit of - * the inode is not updated when we only log that it exists (see - * btrfs_log_inode()). - */ - if (BTRFS_I(inode)->logged_trans == trans->transid) { - spin_unlock(&BTRFS_I(inode)->lock); + if (!need_log_inode(trans, BTRFS_I(inode))) { btrfs_add_delayed_iput(inode); continue; } - spin_unlock(&BTRFS_I(inode)->lock); + /* * We are safe logging the other inode without acquiring its * lock as long as we log with the LOG_INODE_EXISTS mode. We @@ -5509,67 +5747,16 @@ static int log_conflicting_inodes(struct btrfs_trans_handle *trans, * well because during a rename we pin the log and update the * log with the new name before we unpin it. */ - ret = btrfs_log_inode(trans, BTRFS_I(inode), LOG_OTHER_INODE, ctx); - if (ret) { - btrfs_add_delayed_iput(inode); - continue; - } - - key.objectid = ino; - key.type = BTRFS_INODE_REF_KEY; - key.offset = 0; - ret = btrfs_search_slot(NULL, root, &key, path, 0, 0); - if (ret < 0) { - btrfs_add_delayed_iput(inode); - continue; - } - - while (true) { - struct extent_buffer *leaf = path->nodes[0]; - int slot = path->slots[0]; - u64 other_ino = 0; - u64 other_parent = 0; - - if (slot >= btrfs_header_nritems(leaf)) { - ret = btrfs_next_leaf(root, path); - if (ret < 0) { - break; - } else if (ret > 0) { - ret = 0; - break; - } - continue; - } - - btrfs_item_key_to_cpu(leaf, &key, slot); - if (key.objectid != ino || - (key.type != BTRFS_INODE_REF_KEY && - key.type != BTRFS_INODE_EXTREF_KEY)) { - ret = 0; - break; - } - - ret = btrfs_check_ref_name_override(leaf, slot, &key, - BTRFS_I(inode), &other_ino, - &other_parent); - if (ret < 0) - break; - if (ret > 0) { - ino_elem = kmalloc(sizeof(*ino_elem), GFP_NOFS); - if (!ino_elem) { - ret = -ENOMEM; - break; - } - ino_elem->ino = other_ino; - ino_elem->parent = other_parent; - list_add_tail(&ino_elem->list, &inode_list); - ret = 0; - } - path->slots[0]++; - } + ret = btrfs_log_inode(trans, BTRFS_I(inode), LOG_INODE_EXISTS, ctx); btrfs_add_delayed_iput(inode); + if (ret) + break; } + ctx->logging_conflict_inodes = false; + if (ret) + free_conflicting_inodes(ctx); + return ret; } @@ -5580,7 +5767,6 @@ static int copy_inode_items_to_log(struct btrfs_trans_handle *trans, struct btrfs_path *path, struct btrfs_path *dst_path, const u64 logged_isize, - const bool recursive_logging, const int inode_only, struct btrfs_log_ctx *ctx, bool *need_log_inode_item) @@ -5619,8 +5805,8 @@ again: break; } else if ((min_key->type == BTRFS_INODE_REF_KEY || min_key->type == BTRFS_INODE_EXTREF_KEY) && - inode->generation == trans->transid && - !recursive_logging) { + (inode->generation == trans->transid || + ctx->logging_conflict_inodes)) { u64 other_ino = 0; u64 other_parent = 0; @@ -5644,11 +5830,12 @@ again: return ret; ins_nr = 0; - ret = log_conflicting_inodes(trans, root, path, - ctx, other_ino, other_parent); + btrfs_release_path(path); + ret = add_conflicting_inode(trans, root, path, + other_ino, + other_parent, ctx); if (ret) return ret; - btrfs_release_path(path); goto next_key; } } else if (min_key->type == BTRFS_XATTR_ITEM_KEY) { @@ -5731,6 +5918,371 @@ next_key: return ret; } +static int insert_delayed_items_batch(struct btrfs_trans_handle *trans, + struct btrfs_root *log, + struct btrfs_path *path, + const struct btrfs_item_batch *batch, + const struct btrfs_delayed_item *first_item) +{ + const struct btrfs_delayed_item *curr = first_item; + int ret; + + ret = btrfs_insert_empty_items(trans, log, path, batch); + if (ret) + return ret; + + for (int i = 0; i < batch->nr; i++) { + char *data_ptr; + + data_ptr = btrfs_item_ptr(path->nodes[0], path->slots[0], char); + write_extent_buffer(path->nodes[0], &curr->data, + (unsigned long)data_ptr, curr->data_len); + curr = list_next_entry(curr, log_list); + path->slots[0]++; + } + + btrfs_release_path(path); + + return 0; +} + +static int log_delayed_insertion_items(struct btrfs_trans_handle *trans, + struct btrfs_inode *inode, + struct btrfs_path *path, + const struct list_head *delayed_ins_list, + struct btrfs_log_ctx *ctx) +{ + /* 195 (4095 bytes of keys and sizes) fits in a single 4K page. */ + const int max_batch_size = 195; + const int leaf_data_size = BTRFS_LEAF_DATA_SIZE(trans->fs_info); + const u64 ino = btrfs_ino(inode); + struct btrfs_root *log = inode->root->log_root; + struct btrfs_item_batch batch = { + .nr = 0, + .total_data_size = 0, + }; + const struct btrfs_delayed_item *first = NULL; + const struct btrfs_delayed_item *curr; + char *ins_data; + struct btrfs_key *ins_keys; + u32 *ins_sizes; + u64 curr_batch_size = 0; + int batch_idx = 0; + int ret; + + /* We are adding dir index items to the log tree. */ + lockdep_assert_held(&inode->log_mutex); + + /* + * We collect delayed items before copying index keys from the subvolume + * to the log tree. However just after we collected them, they may have + * been flushed (all of them or just some of them), and therefore we + * could have copied them from the subvolume tree to the log tree. + * So find the first delayed item that was not yet logged (they are + * sorted by index number). + */ + list_for_each_entry(curr, delayed_ins_list, log_list) { + if (curr->index > inode->last_dir_index_offset) { + first = curr; + break; + } + } + + /* Empty list or all delayed items were already logged. */ + if (!first) + return 0; + + ins_data = kmalloc(max_batch_size * sizeof(u32) + + max_batch_size * sizeof(struct btrfs_key), GFP_NOFS); + if (!ins_data) + return -ENOMEM; + ins_sizes = (u32 *)ins_data; + batch.data_sizes = ins_sizes; + ins_keys = (struct btrfs_key *)(ins_data + max_batch_size * sizeof(u32)); + batch.keys = ins_keys; + + curr = first; + while (!list_entry_is_head(curr, delayed_ins_list, log_list)) { + const u32 curr_size = curr->data_len + sizeof(struct btrfs_item); + + if (curr_batch_size + curr_size > leaf_data_size || + batch.nr == max_batch_size) { + ret = insert_delayed_items_batch(trans, log, path, + &batch, first); + if (ret) + goto out; + batch_idx = 0; + batch.nr = 0; + batch.total_data_size = 0; + curr_batch_size = 0; + first = curr; + } + + ins_sizes[batch_idx] = curr->data_len; + ins_keys[batch_idx].objectid = ino; + ins_keys[batch_idx].type = BTRFS_DIR_INDEX_KEY; + ins_keys[batch_idx].offset = curr->index; + curr_batch_size += curr_size; + batch.total_data_size += curr->data_len; + batch.nr++; + batch_idx++; + curr = list_next_entry(curr, log_list); + } + + ASSERT(batch.nr >= 1); + ret = insert_delayed_items_batch(trans, log, path, &batch, first); + + curr = list_last_entry(delayed_ins_list, struct btrfs_delayed_item, + log_list); + inode->last_dir_index_offset = curr->index; +out: + kfree(ins_data); + + return ret; +} + +static int log_delayed_deletions_full(struct btrfs_trans_handle *trans, + struct btrfs_inode *inode, + struct btrfs_path *path, + const struct list_head *delayed_del_list, + struct btrfs_log_ctx *ctx) +{ + const u64 ino = btrfs_ino(inode); + const struct btrfs_delayed_item *curr; + + curr = list_first_entry(delayed_del_list, struct btrfs_delayed_item, + log_list); + + while (!list_entry_is_head(curr, delayed_del_list, log_list)) { + u64 first_dir_index = curr->index; + u64 last_dir_index; + const struct btrfs_delayed_item *next; + int ret; + + /* + * Find a range of consecutive dir index items to delete. Like + * this we log a single dir range item spanning several contiguous + * dir items instead of logging one range item per dir index item. + */ + next = list_next_entry(curr, log_list); + while (!list_entry_is_head(next, delayed_del_list, log_list)) { + if (next->index != curr->index + 1) + break; + curr = next; + next = list_next_entry(next, log_list); + } + + last_dir_index = curr->index; + ASSERT(last_dir_index >= first_dir_index); + + ret = insert_dir_log_key(trans, inode->root->log_root, path, + ino, first_dir_index, last_dir_index); + if (ret) + return ret; + curr = list_next_entry(curr, log_list); + } + + return 0; +} + +static int batch_delete_dir_index_items(struct btrfs_trans_handle *trans, + struct btrfs_inode *inode, + struct btrfs_path *path, + struct btrfs_log_ctx *ctx, + const struct list_head *delayed_del_list, + const struct btrfs_delayed_item *first, + const struct btrfs_delayed_item **last_ret) +{ + const struct btrfs_delayed_item *next; + struct extent_buffer *leaf = path->nodes[0]; + const int last_slot = btrfs_header_nritems(leaf) - 1; + int slot = path->slots[0] + 1; + const u64 ino = btrfs_ino(inode); + + next = list_next_entry(first, log_list); + + while (slot < last_slot && + !list_entry_is_head(next, delayed_del_list, log_list)) { + struct btrfs_key key; + + btrfs_item_key_to_cpu(leaf, &key, slot); + if (key.objectid != ino || + key.type != BTRFS_DIR_INDEX_KEY || + key.offset != next->index) + break; + + slot++; + *last_ret = next; + next = list_next_entry(next, log_list); + } + + return btrfs_del_items(trans, inode->root->log_root, path, + path->slots[0], slot - path->slots[0]); +} + +static int log_delayed_deletions_incremental(struct btrfs_trans_handle *trans, + struct btrfs_inode *inode, + struct btrfs_path *path, + const struct list_head *delayed_del_list, + struct btrfs_log_ctx *ctx) +{ + struct btrfs_root *log = inode->root->log_root; + const struct btrfs_delayed_item *curr; + u64 last_range_start; + u64 last_range_end = 0; + struct btrfs_key key; + + key.objectid = btrfs_ino(inode); + key.type = BTRFS_DIR_INDEX_KEY; + curr = list_first_entry(delayed_del_list, struct btrfs_delayed_item, + log_list); + + while (!list_entry_is_head(curr, delayed_del_list, log_list)) { + const struct btrfs_delayed_item *last = curr; + u64 first_dir_index = curr->index; + u64 last_dir_index; + bool deleted_items = false; + int ret; + + key.offset = curr->index; + ret = btrfs_search_slot(trans, log, &key, path, -1, 1); + if (ret < 0) { + return ret; + } else if (ret == 0) { + ret = batch_delete_dir_index_items(trans, inode, path, ctx, + delayed_del_list, curr, + &last); + if (ret) + return ret; + deleted_items = true; + } + + btrfs_release_path(path); + + /* + * If we deleted items from the leaf, it means we have a range + * item logging their range, so no need to add one or update an + * existing one. Otherwise we have to log a dir range item. + */ + if (deleted_items) + goto next_batch; + + last_dir_index = last->index; + ASSERT(last_dir_index >= first_dir_index); + /* + * If this range starts right after where the previous one ends, + * then we want to reuse the previous range item and change its + * end offset to the end of this range. This is just to minimize + * leaf space usage, by avoiding adding a new range item. + */ + if (last_range_end != 0 && first_dir_index == last_range_end + 1) + first_dir_index = last_range_start; + + ret = insert_dir_log_key(trans, log, path, key.objectid, + first_dir_index, last_dir_index); + if (ret) + return ret; + + last_range_start = first_dir_index; + last_range_end = last_dir_index; +next_batch: + curr = list_next_entry(last, log_list); + } + + return 0; +} + +static int log_delayed_deletion_items(struct btrfs_trans_handle *trans, + struct btrfs_inode *inode, + struct btrfs_path *path, + const struct list_head *delayed_del_list, + struct btrfs_log_ctx *ctx) +{ + /* + * We are deleting dir index items from the log tree or adding range + * items to it. + */ + lockdep_assert_held(&inode->log_mutex); + + if (list_empty(delayed_del_list)) + return 0; + + if (ctx->logged_before) + return log_delayed_deletions_incremental(trans, inode, path, + delayed_del_list, ctx); + + return log_delayed_deletions_full(trans, inode, path, delayed_del_list, + ctx); +} + +/* + * Similar logic as for log_new_dir_dentries(), but it iterates over the delayed + * items instead of the subvolume tree. + */ +static int log_new_delayed_dentries(struct btrfs_trans_handle *trans, + struct btrfs_inode *inode, + const struct list_head *delayed_ins_list, + struct btrfs_log_ctx *ctx) +{ + const bool orig_log_new_dentries = ctx->log_new_dentries; + struct btrfs_fs_info *fs_info = trans->fs_info; + struct btrfs_delayed_item *item; + int ret = 0; + + /* + * No need for the log mutex, plus to avoid potential deadlocks or + * lockdep annotations due to nesting of delayed inode mutexes and log + * mutexes. + */ + lockdep_assert_not_held(&inode->log_mutex); + + ASSERT(!ctx->logging_new_delayed_dentries); + ctx->logging_new_delayed_dentries = true; + + list_for_each_entry(item, delayed_ins_list, log_list) { + struct btrfs_dir_item *dir_item; + struct inode *di_inode; + struct btrfs_key key; + int log_mode = LOG_INODE_EXISTS; + + dir_item = (struct btrfs_dir_item *)item->data; + btrfs_disk_key_to_cpu(&key, &dir_item->location); + + if (key.type == BTRFS_ROOT_ITEM_KEY) + continue; + + di_inode = btrfs_iget(fs_info->sb, key.objectid, inode->root); + if (IS_ERR(di_inode)) { + ret = PTR_ERR(di_inode); + break; + } + + if (!need_log_inode(trans, BTRFS_I(di_inode))) { + btrfs_add_delayed_iput(di_inode); + continue; + } + + if (btrfs_stack_dir_type(dir_item) == BTRFS_FT_DIR) + log_mode = LOG_INODE_ALL; + + ctx->log_new_dentries = false; + ret = btrfs_log_inode(trans, BTRFS_I(di_inode), log_mode, ctx); + + if (!ret && ctx->log_new_dentries) + ret = log_new_dir_dentries(trans, BTRFS_I(di_inode), ctx); + + btrfs_add_delayed_iput(di_inode); + + if (ret) + break; + } + + ctx->log_new_dentries = orig_log_new_dentries; + ctx->logging_new_delayed_dentries = false; + + return ret; +} + /* log a single inode in the tree log. * At least one parent directory for this inode must exist in the tree * or be logged already. @@ -5762,9 +6314,10 @@ static int btrfs_log_inode(struct btrfs_trans_handle *trans, u64 logged_isize = 0; bool need_log_inode_item = true; bool xattrs_logged = false; - bool recursive_logging = false; bool inode_item_dropped = true; - const bool orig_logged_before = ctx->logged_before; + bool full_dir_logging = false; + LIST_HEAD(delayed_ins_list); + LIST_HEAD(delayed_del_list); path = btrfs_alloc_path(); if (!path) @@ -5792,27 +6345,46 @@ static int btrfs_log_inode(struct btrfs_trans_handle *trans, max_key.type = (u8)-1; max_key.offset = (u64)-1; + if (S_ISDIR(inode->vfs_inode.i_mode) && inode_only == LOG_INODE_ALL) + full_dir_logging = true; + /* - * Only run delayed items if we are a directory. We want to make sure - * all directory indexes hit the fs/subvolume tree so we can find them - * and figure out which index ranges have to be logged. + * If we are logging a directory while we are logging dentries of the + * delayed items of some other inode, then we need to flush the delayed + * items of this directory and not log the delayed items directly. This + * is to prevent more than one level of recursion into btrfs_log_inode() + * by having something like this: + * + * $ mkdir -p a/b/c/d/e/f/g/h/... + * $ xfs_io -c "fsync" a + * + * Where all directories in the path did not exist before and are + * created in the current transaction. + * So in such a case we directly log the delayed items of the main + * directory ("a") without flushing them first, while for each of its + * subdirectories we flush their delayed items before logging them. + * This prevents a potential unbounded recursion like this: + * + * btrfs_log_inode() + * log_new_delayed_dentries() + * btrfs_log_inode() + * log_new_delayed_dentries() + * btrfs_log_inode() + * log_new_delayed_dentries() + * (...) + * + * We have thresholds for the maximum number of delayed items to have in + * memory, and once they are hit, the items are flushed asynchronously. + * However the limit is quite high, so lets prevent deep levels of + * recursion to happen by limiting the maximum depth to be 1. */ - if (S_ISDIR(inode->vfs_inode.i_mode)) { + if (full_dir_logging && ctx->logging_new_delayed_dentries) { ret = btrfs_commit_inode_delayed_items(trans, inode); if (ret) goto out; } - if (inode_only == LOG_OTHER_INODE || inode_only == LOG_OTHER_INODE_ALL) { - recursive_logging = true; - if (inode_only == LOG_OTHER_INODE) - inode_only = LOG_INODE_EXISTS; - else - inode_only = LOG_INODE_ALL; - mutex_lock_nested(&inode->log_mutex, SINGLE_DEPTH_NESTING); - } else { - mutex_lock(&inode->log_mutex); - } + mutex_lock(&inode->log_mutex); /* * For symlinks, we must always log their content, which is stored in an @@ -5844,9 +6416,7 @@ static int btrfs_log_inode(struct btrfs_trans_handle *trans, * to known the file was moved from A to B, so logging just A would * result in losing the file after a log replay. */ - if (S_ISDIR(inode->vfs_inode.i_mode) && - inode_only == LOG_INODE_ALL && - inode->last_unlink_trans >= trans->transid) { + if (full_dir_logging && inode->last_unlink_trans >= trans->transid) { btrfs_set_log_full_commit(trans); ret = BTRFS_LOG_FORCE_COMMIT; goto out_unlock; @@ -5857,14 +6427,10 @@ static int btrfs_log_inode(struct btrfs_trans_handle *trans, * copies of everything. */ if (S_ISDIR(inode->vfs_inode.i_mode)) { - int max_key_type = BTRFS_DIR_LOG_INDEX_KEY; - clear_bit(BTRFS_INODE_COPY_EVERYTHING, &inode->runtime_flags); - if (inode_only == LOG_INODE_EXISTS) - max_key_type = BTRFS_XATTR_ITEM_KEY; if (ctx->logged_before) ret = drop_inode_items(trans, log, path, inode, - max_key_type); + BTRFS_XATTR_ITEM_KEY); } else { if (inode_only == LOG_INODE_EXISTS && ctx->logged_before) { /* @@ -5920,9 +6486,19 @@ static int btrfs_log_inode(struct btrfs_trans_handle *trans, if (ret) goto out_unlock; + /* + * If we are logging a directory in full mode, collect the delayed items + * before iterating the subvolume tree, so that we don't miss any new + * dir index items in case they get flushed while or right after we are + * iterating the subvolume tree. + */ + if (full_dir_logging && !ctx->logging_new_delayed_dentries) + btrfs_log_get_delayed_items(inode, &delayed_ins_list, + &delayed_del_list); + ret = copy_inode_items_to_log(trans, inode, &min_key, &max_key, path, dst_path, logged_isize, - recursive_logging, inode_only, ctx, + inode_only, ctx, &need_log_inode_item); if (ret) goto out_unlock; @@ -5975,10 +6551,18 @@ log_extents: write_unlock(&em_tree->lock); } - if (inode_only == LOG_INODE_ALL && S_ISDIR(inode->vfs_inode.i_mode)) { + if (full_dir_logging) { ret = log_directory_changes(trans, inode, path, dst_path, ctx); if (ret) goto out_unlock; + ret = log_delayed_insertion_items(trans, inode, path, + &delayed_ins_list, ctx); + if (ret) + goto out_unlock; + ret = log_delayed_deletion_items(trans, inode, path, + &delayed_del_list, ctx); + if (ret) + goto out_unlock; } spin_lock(&inode->lock); @@ -6031,208 +6615,20 @@ out: btrfs_free_path(path); btrfs_free_path(dst_path); - if (recursive_logging) - ctx->logged_before = orig_logged_before; - - return ret; -} - -/* - * Check if we need to log an inode. This is used in contexts where while - * logging an inode we need to log another inode (either that it exists or in - * full mode). This is used instead of btrfs_inode_in_log() because the later - * requires the inode to be in the log and have the log transaction committed, - * while here we do not care if the log transaction was already committed - our - * caller will commit the log later - and we want to avoid logging an inode - * multiple times when multiple tasks have joined the same log transaction. - */ -static bool need_log_inode(struct btrfs_trans_handle *trans, - struct btrfs_inode *inode) -{ - /* - * If a directory was not modified, no dentries added or removed, we can - * and should avoid logging it. - */ - if (S_ISDIR(inode->vfs_inode.i_mode) && inode->last_trans < trans->transid) - return false; - - /* - * If this inode does not have new/updated/deleted xattrs since the last - * time it was logged and is flagged as logged in the current transaction, - * we can skip logging it. As for new/deleted names, those are updated in - * the log by link/unlink/rename operations. - * In case the inode was logged and then evicted and reloaded, its - * logged_trans will be 0, in which case we have to fully log it since - * logged_trans is a transient field, not persisted. - */ - if (inode->logged_trans == trans->transid && - !test_bit(BTRFS_INODE_COPY_EVERYTHING, &inode->runtime_flags)) - return false; - - return true; -} - -struct btrfs_dir_list { - u64 ino; - struct list_head list; -}; - -/* - * Log the inodes of the new dentries of a directory. See log_dir_items() for - * details about the why it is needed. - * This is a recursive operation - if an existing dentry corresponds to a - * directory, that directory's new entries are logged too (same behaviour as - * ext3/4, xfs, f2fs, reiserfs, nilfs2). Note that when logging the inodes - * the dentries point to we do not lock their i_mutex, otherwise lockdep - * complains about the following circular lock dependency / possible deadlock: - * - * CPU0 CPU1 - * ---- ---- - * lock(&type->i_mutex_dir_key#3/2); - * lock(sb_internal#2); - * lock(&type->i_mutex_dir_key#3/2); - * lock(&sb->s_type->i_mutex_key#14); - * - * Where sb_internal is the lock (a counter that works as a lock) acquired by - * sb_start_intwrite() in btrfs_start_transaction(). - * Not locking i_mutex of the inodes is still safe because: - * - * 1) For regular files we log with a mode of LOG_INODE_EXISTS. It's possible - * that while logging the inode new references (names) are added or removed - * from the inode, leaving the logged inode item with a link count that does - * not match the number of logged inode reference items. This is fine because - * at log replay time we compute the real number of links and correct the - * link count in the inode item (see replay_one_buffer() and - * link_to_fixup_dir()); - * - * 2) For directories we log with a mode of LOG_INODE_ALL. It's possible that - * while logging the inode's items new index items (key type - * BTRFS_DIR_INDEX_KEY) are added to fs/subvol tree and the logged inode item - * has a size that doesn't match the sum of the lengths of all the logged - * names - this is ok, not a problem, because at log replay time we set the - * directory's i_size to the correct value (see replay_one_name() and - * do_overwrite_item()). - */ -static int log_new_dir_dentries(struct btrfs_trans_handle *trans, - struct btrfs_root *root, - struct btrfs_inode *start_inode, - struct btrfs_log_ctx *ctx) -{ - struct btrfs_fs_info *fs_info = root->fs_info; - struct btrfs_path *path; - LIST_HEAD(dir_list); - struct btrfs_dir_list *dir_elem; - int ret = 0; - - /* - * If we are logging a new name, as part of a link or rename operation, - * don't bother logging new dentries, as we just want to log the names - * of an inode and that any new parents exist. - */ - if (ctx->logging_new_name) - return 0; - - path = btrfs_alloc_path(); - if (!path) - return -ENOMEM; - - dir_elem = kmalloc(sizeof(*dir_elem), GFP_NOFS); - if (!dir_elem) { - btrfs_free_path(path); - return -ENOMEM; - } - dir_elem->ino = btrfs_ino(start_inode); - list_add_tail(&dir_elem->list, &dir_list); - - while (!list_empty(&dir_list)) { - struct extent_buffer *leaf; - struct btrfs_key min_key; - int nritems; - int i; - - dir_elem = list_first_entry(&dir_list, struct btrfs_dir_list, - list); - if (ret) - goto next_dir_inode; - - min_key.objectid = dir_elem->ino; - min_key.type = BTRFS_DIR_INDEX_KEY; - min_key.offset = 0; -again: - btrfs_release_path(path); - ret = btrfs_search_forward(root, &min_key, path, trans->transid); - if (ret < 0) { - goto next_dir_inode; - } else if (ret > 0) { - ret = 0; - goto next_dir_inode; - } - - leaf = path->nodes[0]; - nritems = btrfs_header_nritems(leaf); - for (i = path->slots[0]; i < nritems; i++) { - struct btrfs_dir_item *di; - struct btrfs_key di_key; - struct inode *di_inode; - struct btrfs_dir_list *new_dir_elem; - int log_mode = LOG_INODE_EXISTS; - int type; - - btrfs_item_key_to_cpu(leaf, &min_key, i); - if (min_key.objectid != dir_elem->ino || - min_key.type != BTRFS_DIR_INDEX_KEY) - goto next_dir_inode; - - di = btrfs_item_ptr(leaf, i, struct btrfs_dir_item); - type = btrfs_dir_type(leaf, di); - if (btrfs_dir_transid(leaf, di) < trans->transid) - continue; - btrfs_dir_item_key_to_cpu(leaf, di, &di_key); - if (di_key.type == BTRFS_ROOT_ITEM_KEY) - continue; - - btrfs_release_path(path); - di_inode = btrfs_iget(fs_info->sb, di_key.objectid, root); - if (IS_ERR(di_inode)) { - ret = PTR_ERR(di_inode); - goto next_dir_inode; - } + if (ret) + free_conflicting_inodes(ctx); + else + ret = log_conflicting_inodes(trans, inode->root, ctx); - if (!need_log_inode(trans, BTRFS_I(di_inode))) { - btrfs_add_delayed_iput(di_inode); - break; - } + if (full_dir_logging && !ctx->logging_new_delayed_dentries) { + if (!ret) + ret = log_new_delayed_dentries(trans, inode, + &delayed_ins_list, ctx); - ctx->log_new_dentries = false; - if (type == BTRFS_FT_DIR) - log_mode = LOG_INODE_ALL; - ret = btrfs_log_inode(trans, BTRFS_I(di_inode), - log_mode, ctx); - btrfs_add_delayed_iput(di_inode); - if (ret) - goto next_dir_inode; - if (ctx->log_new_dentries) { - new_dir_elem = kmalloc(sizeof(*new_dir_elem), - GFP_NOFS); - if (!new_dir_elem) { - ret = -ENOMEM; - goto next_dir_inode; - } - new_dir_elem->ino = di_key.objectid; - list_add_tail(&new_dir_elem->list, &dir_list); - } - break; - } - if (min_key.offset < (u64)-1) { - min_key.offset++; - goto again; - } -next_dir_inode: - list_del(&dir_elem->list); - kfree(dir_elem); + btrfs_log_put_delayed_items(inode, &delayed_ins_list, + &delayed_del_list); } - btrfs_free_path(path); return ret; } @@ -6344,7 +6740,7 @@ static int btrfs_log_all_parents(struct btrfs_trans_handle *trans, ret = btrfs_log_inode(trans, BTRFS_I(dir_inode), LOG_INODE_ALL, ctx); if (!ret && ctx->log_new_dentries) - ret = log_new_dir_dentries(trans, root, + ret = log_new_dir_dentries(trans, BTRFS_I(dir_inode), ctx); btrfs_add_delayed_iput(dir_inode); if (ret) @@ -6659,7 +7055,7 @@ static int btrfs_log_inode_parent(struct btrfs_trans_handle *trans, goto end_trans; if (log_dentries) - ret = log_new_dir_dentries(trans, root, inode, ctx); + ret = log_new_dir_dentries(trans, inode, ctx); else ret = 0; end_trans: @@ -7086,6 +7482,7 @@ void btrfs_log_new_name(struct btrfs_trans_handle *trans, * inconsistent state after a rename operation. */ btrfs_log_inode_parent(trans, inode, parent, LOG_INODE_EXISTS, &ctx); + ASSERT(list_empty(&ctx.conflict_inodes)); out: /* * If an error happened mark the log for a full commit because it's not |