From 7d7ea89e756ea18a3b08cd396e2a4c0c12d473a8 Mon Sep 17 00:00:00 2001 From: Theodore Ts'o Date: Fri, 16 Aug 2013 21:20:41 -0400 Subject: ext4: refactor code to read the extent tree block Refactor out the code needed to read the extent tree block into a single read_extent_tree_block() function. In addition to simplifying the code, it also makes sure that we call the ext4_ext_load_extent tracepoint whenever we need to read an extent tree block from disk. Signed-off-by: "Theodore Ts'o" Reviewed-by: Zheng Liu --- fs/ext4/extents.c | 97 ++++++++++++++++++++++++------------------------------- 1 file changed, 43 insertions(+), 54 deletions(-) (limited to 'fs/ext4/extents.c') diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c index 72ba4705d4fa..a40be59ddce6 100644 --- a/fs/ext4/extents.c +++ b/fs/ext4/extents.c @@ -464,25 +464,39 @@ int ext4_ext_check_inode(struct inode *inode) return ext4_ext_check(inode, ext_inode_hdr(inode), ext_depth(inode)); } -static int __ext4_ext_check_block(const char *function, unsigned int line, - struct inode *inode, - struct ext4_extent_header *eh, - int depth, - struct buffer_head *bh) +static struct buffer_head * +__read_extent_tree_block(const char *function, unsigned int line, + struct inode *inode, ext4_fsblk_t pblk, int depth) { - int ret; + struct buffer_head *bh; + int err; + bh = sb_getblk(inode->i_sb, pblk); + if (unlikely(!bh)) + return ERR_PTR(-ENOMEM); + + if (!bh_uptodate_or_lock(bh)) { + trace_ext4_ext_load_extent(inode, pblk, _RET_IP_); + err = bh_submit_read(bh); + if (err < 0) + goto errout; + } if (buffer_verified(bh)) - return 0; - ret = ext4_ext_check(inode, eh, depth); - if (ret) - return ret; + return bh; + err = __ext4_ext_check(function, line, inode, + ext_block_hdr(bh), depth); + if (err) + goto errout; set_buffer_verified(bh); - return ret; + return bh; +errout: + put_bh(bh); + return ERR_PTR(err); + } -#define ext4_ext_check_block(inode, eh, depth, bh) \ - __ext4_ext_check_block(__func__, __LINE__, inode, eh, depth, bh) +#define read_extent_tree_block(inode, pblk, depth) \ + __read_extent_tree_block(__func__, __LINE__, (inode), (pblk), (depth)) #ifdef EXT_DEBUG static void ext4_ext_show_path(struct inode *inode, struct ext4_ext_path *path) @@ -748,20 +762,12 @@ ext4_ext_find_extent(struct inode *inode, ext4_lblk_t block, path[ppos].p_depth = i; path[ppos].p_ext = NULL; - bh = sb_getblk(inode->i_sb, path[ppos].p_block); - if (unlikely(!bh)) { - ret = -ENOMEM; + bh = read_extent_tree_block(inode, path[ppos].p_block, --i); + if (IS_ERR(bh)) { + ret = PTR_ERR(bh); goto err; } - if (!bh_uptodate_or_lock(bh)) { - trace_ext4_ext_load_extent(inode, block, - path[ppos].p_block); - ret = bh_submit_read(bh); - if (ret < 0) { - put_bh(bh); - goto err; - } - } + eh = ext_block_hdr(bh); ppos++; if (unlikely(ppos > depth)) { @@ -773,11 +779,6 @@ ext4_ext_find_extent(struct inode *inode, ext4_lblk_t block, } path[ppos].p_bh = bh; path[ppos].p_hdr = eh; - i--; - - ret = ext4_ext_check_block(inode, eh, i, bh); - if (ret < 0) - goto err; } path[ppos].p_depth = i; @@ -1412,29 +1413,21 @@ got_index: ix++; block = ext4_idx_pblock(ix); while (++depth < path->p_depth) { - bh = sb_bread(inode->i_sb, block); - if (bh == NULL) - return -EIO; - eh = ext_block_hdr(bh); /* subtract from p_depth to get proper eh_depth */ - if (ext4_ext_check_block(inode, eh, - path->p_depth - depth, bh)) { - put_bh(bh); - return -EIO; - } + bh = read_extent_tree_block(inode, block, + path->p_depth - depth); + if (IS_ERR(bh)) + return PTR_ERR(bh); + eh = ext_block_hdr(bh); ix = EXT_FIRST_INDEX(eh); block = ext4_idx_pblock(ix); put_bh(bh); } - bh = sb_bread(inode->i_sb, block); - if (bh == NULL) - return -EIO; + bh = read_extent_tree_block(inode, block, path->p_depth - depth); + if (IS_ERR(bh)) + return PTR_ERR(bh); eh = ext_block_hdr(bh); - if (ext4_ext_check_block(inode, eh, path->p_depth - depth, bh)) { - put_bh(bh); - return -EIO; - } ex = EXT_FIRST_EXTENT(eh); found_extent: *logical = le32_to_cpu(ex->ee_block); @@ -2829,10 +2822,11 @@ again: ext_debug("move to level %d (block %llu)\n", i + 1, ext4_idx_pblock(path[i].p_idx)); memset(path + i + 1, 0, sizeof(*path)); - bh = sb_bread(sb, ext4_idx_pblock(path[i].p_idx)); - if (!bh) { + bh = read_extent_tree_block(inode, + ext4_idx_pblock(path[i].p_idx), depth - i - 1); + if (IS_ERR(bh)) { /* should we reset i_size? */ - err = -EIO; + err = PTR_ERR(bh); break; } /* Yield here to deal with large extent trees. @@ -2842,11 +2836,6 @@ again: err = -EIO; break; } - if (ext4_ext_check_block(inode, ext_block_hdr(bh), - depth - i - 1, bh)) { - err = -EIO; - break; - } path[i + 1].p_bh = bh; /* save actual number of indexes since this -- cgit v1.2.3 From c349179b4808f7c8e1ff1b4dd967c047eefd24bc Mon Sep 17 00:00:00 2001 From: Theodore Ts'o Date: Fri, 16 Aug 2013 21:21:41 -0400 Subject: ext4: print the block number of invalid extent tree blocks When we find an invalid extent tree block, report the block number of the bad block for debugging purposes. Signed-off-by: "Theodore Ts'o" Reviewed-by: Zheng Liu --- fs/ext4/extents.c | 24 ++++++++++++------------ 1 file changed, 12 insertions(+), 12 deletions(-) (limited to 'fs/ext4/extents.c') diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c index a40be59ddce6..6e7b7d928f4a 100644 --- a/fs/ext4/extents.c +++ b/fs/ext4/extents.c @@ -407,7 +407,7 @@ static int ext4_valid_extent_entries(struct inode *inode, static int __ext4_ext_check(const char *function, unsigned int line, struct inode *inode, struct ext4_extent_header *eh, - int depth) + int depth, ext4_fsblk_t pblk) { const char *error_msg; int max = 0; @@ -447,21 +447,21 @@ static int __ext4_ext_check(const char *function, unsigned int line, corrupted: ext4_error_inode(inode, function, line, 0, - "bad header/extent: %s - magic %x, " - "entries %u, max %u(%u), depth %u(%u)", - error_msg, le16_to_cpu(eh->eh_magic), - le16_to_cpu(eh->eh_entries), le16_to_cpu(eh->eh_max), - max, le16_to_cpu(eh->eh_depth), depth); - + "pblk %llu bad header/extent: %s - magic %x, " + "entries %u, max %u(%u), depth %u(%u)", + (unsigned long long) pblk, error_msg, + le16_to_cpu(eh->eh_magic), + le16_to_cpu(eh->eh_entries), le16_to_cpu(eh->eh_max), + max, le16_to_cpu(eh->eh_depth), depth); return -EIO; } -#define ext4_ext_check(inode, eh, depth) \ - __ext4_ext_check(__func__, __LINE__, inode, eh, depth) +#define ext4_ext_check(inode, eh, depth, pblk) \ + __ext4_ext_check(__func__, __LINE__, (inode), (eh), (depth), (pblk)) int ext4_ext_check_inode(struct inode *inode) { - return ext4_ext_check(inode, ext_inode_hdr(inode), ext_depth(inode)); + return ext4_ext_check(inode, ext_inode_hdr(inode), ext_depth(inode), 0); } static struct buffer_head * @@ -484,7 +484,7 @@ __read_extent_tree_block(const char *function, unsigned int line, if (buffer_verified(bh)) return bh; err = __ext4_ext_check(function, line, inode, - ext_block_hdr(bh), depth); + ext_block_hdr(bh), depth, pblk); if (err) goto errout; set_buffer_verified(bh); @@ -2775,7 +2775,7 @@ again: path[0].p_hdr = ext_inode_hdr(inode); i = 0; - if (ext4_ext_check(inode, path[0].p_hdr, depth)) { + if (ext4_ext_check(inode, path[0].p_hdr, depth, 0)) { err = -EIO; goto out; } -- cgit v1.2.3 From 107a7bd31ac003e42c0f966aa8e5b26947de6024 Mon Sep 17 00:00:00 2001 From: Theodore Ts'o Date: Fri, 16 Aug 2013 21:23:41 -0400 Subject: ext4: cache all of an extent tree's leaf block upon reading When we read in an extent tree leaf block from disk, arrange to have all of its entries cached. In nearly all cases the in-memory representation will be more compact than the on-disk representation in the buffer cache, and it allows us to get the information without having to traverse the extent tree for successive extents. Signed-off-by: "Theodore Ts'o" Reviewed-by: Zheng Liu --- fs/ext4/extents.c | 87 ++++++++++++++++++++++++++++++++++++++----------------- 1 file changed, 60 insertions(+), 27 deletions(-) (limited to 'fs/ext4/extents.c') diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c index 6e7b7d928f4a..08c1ac976479 100644 --- a/fs/ext4/extents.c +++ b/fs/ext4/extents.c @@ -466,7 +466,8 @@ int ext4_ext_check_inode(struct inode *inode) static struct buffer_head * __read_extent_tree_block(const char *function, unsigned int line, - struct inode *inode, ext4_fsblk_t pblk, int depth) + struct inode *inode, ext4_fsblk_t pblk, int depth, + int flags) { struct buffer_head *bh; int err; @@ -488,6 +489,32 @@ __read_extent_tree_block(const char *function, unsigned int line, if (err) goto errout; set_buffer_verified(bh); + /* + * If this is a leaf block, cache all of its entries + */ + if (!(flags & EXT4_EX_NOCACHE) && depth == 0) { + struct ext4_extent_header *eh = ext_block_hdr(bh); + struct ext4_extent *ex = EXT_FIRST_EXTENT(eh); + ext4_lblk_t prev = 0; + int i; + + for (i = le16_to_cpu(eh->eh_entries); i > 0; i--, ex++) { + unsigned int status = EXTENT_STATUS_WRITTEN; + ext4_lblk_t lblk = le32_to_cpu(ex->ee_block); + int len = ext4_ext_get_actual_len(ex); + + if (prev && (prev != lblk)) + ext4_es_cache_extent(inode, prev, + lblk - prev, ~0, + EXTENT_STATUS_HOLE); + + if (ext4_ext_is_uninitialized(ex)) + status = EXTENT_STATUS_UNWRITTEN; + ext4_es_cache_extent(inode, lblk, len, + ext4_ext_pblock(ex), status); + prev = lblk + len; + } + } return bh; errout: put_bh(bh); @@ -495,8 +522,9 @@ errout: } -#define read_extent_tree_block(inode, pblk, depth) \ - __read_extent_tree_block(__func__, __LINE__, (inode), (pblk), (depth)) +#define read_extent_tree_block(inode, pblk, depth, flags) \ + __read_extent_tree_block(__func__, __LINE__, (inode), (pblk), \ + (depth), (flags)) #ifdef EXT_DEBUG static void ext4_ext_show_path(struct inode *inode, struct ext4_ext_path *path) @@ -730,7 +758,7 @@ int ext4_ext_tree_init(handle_t *handle, struct inode *inode) struct ext4_ext_path * ext4_ext_find_extent(struct inode *inode, ext4_lblk_t block, - struct ext4_ext_path *path) + struct ext4_ext_path *path, int flags) { struct ext4_extent_header *eh; struct buffer_head *bh; @@ -762,7 +790,8 @@ ext4_ext_find_extent(struct inode *inode, ext4_lblk_t block, path[ppos].p_depth = i; path[ppos].p_ext = NULL; - bh = read_extent_tree_block(inode, path[ppos].p_block, --i); + bh = read_extent_tree_block(inode, path[ppos].p_block, --i, + flags); if (IS_ERR(bh)) { ret = PTR_ERR(bh); goto err; @@ -1199,7 +1228,8 @@ out: * if no free index is found, then it requests in-depth growing. */ static int ext4_ext_create_new_leaf(handle_t *handle, struct inode *inode, - unsigned int flags, + unsigned int mb_flags, + unsigned int gb_flags, struct ext4_ext_path *path, struct ext4_extent *newext) { @@ -1221,7 +1251,7 @@ repeat: if (EXT_HAS_FREE_INDEX(curp)) { /* if we found index with free entry, then use that * entry: create all needed subtree and add new leaf */ - err = ext4_ext_split(handle, inode, flags, path, newext, i); + err = ext4_ext_split(handle, inode, mb_flags, path, newext, i); if (err) goto out; @@ -1229,12 +1259,12 @@ repeat: ext4_ext_drop_refs(path); path = ext4_ext_find_extent(inode, (ext4_lblk_t)le32_to_cpu(newext->ee_block), - path); + path, gb_flags); if (IS_ERR(path)) err = PTR_ERR(path); } else { /* tree is full, time to grow in depth */ - err = ext4_ext_grow_indepth(handle, inode, flags, newext); + err = ext4_ext_grow_indepth(handle, inode, mb_flags, newext); if (err) goto out; @@ -1242,7 +1272,7 @@ repeat: ext4_ext_drop_refs(path); path = ext4_ext_find_extent(inode, (ext4_lblk_t)le32_to_cpu(newext->ee_block), - path); + path, gb_flags); if (IS_ERR(path)) { err = PTR_ERR(path); goto out; @@ -1415,7 +1445,7 @@ got_index: while (++depth < path->p_depth) { /* subtract from p_depth to get proper eh_depth */ bh = read_extent_tree_block(inode, block, - path->p_depth - depth); + path->p_depth - depth, 0); if (IS_ERR(bh)) return PTR_ERR(bh); eh = ext_block_hdr(bh); @@ -1424,7 +1454,7 @@ got_index: put_bh(bh); } - bh = read_extent_tree_block(inode, block, path->p_depth - depth); + bh = read_extent_tree_block(inode, block, path->p_depth - depth, 0); if (IS_ERR(bh)) return PTR_ERR(bh); eh = ext_block_hdr(bh); @@ -1786,7 +1816,7 @@ out: */ int ext4_ext_insert_extent(handle_t *handle, struct inode *inode, struct ext4_ext_path *path, - struct ext4_extent *newext, int flag) + struct ext4_extent *newext, int gb_flags) { struct ext4_extent_header *eh; struct ext4_extent *ex, *fex; @@ -1795,7 +1825,7 @@ int ext4_ext_insert_extent(handle_t *handle, struct inode *inode, int depth, len, err; ext4_lblk_t next; unsigned uninitialized = 0; - int flags = 0; + int mb_flags = 0; if (unlikely(ext4_ext_get_actual_len(newext) == 0)) { EXT4_ERROR_INODE(inode, "ext4_ext_get_actual_len(newext) == 0"); @@ -1810,7 +1840,7 @@ int ext4_ext_insert_extent(handle_t *handle, struct inode *inode, } /* try to insert block into found extent and return */ - if (ex && !(flag & EXT4_GET_BLOCKS_PRE_IO)) { + if (ex && !(gb_flags & EXT4_GET_BLOCKS_PRE_IO)) { /* * Try to see whether we should rather test the extent on @@ -1913,7 +1943,7 @@ prepend: if (next != EXT_MAX_BLOCKS) { ext_debug("next leaf block - %u\n", next); BUG_ON(npath != NULL); - npath = ext4_ext_find_extent(inode, next, NULL); + npath = ext4_ext_find_extent(inode, next, NULL, 0); if (IS_ERR(npath)) return PTR_ERR(npath); BUG_ON(npath->p_depth != path->p_depth); @@ -1932,9 +1962,10 @@ prepend: * There is no free space in the found leaf. * We're gonna add a new leaf in the tree. */ - if (flag & EXT4_GET_BLOCKS_METADATA_NOFAIL) - flags = EXT4_MB_USE_RESERVED; - err = ext4_ext_create_new_leaf(handle, inode, flags, path, newext); + if (gb_flags & EXT4_GET_BLOCKS_METADATA_NOFAIL) + mb_flags = EXT4_MB_USE_RESERVED; + err = ext4_ext_create_new_leaf(handle, inode, mb_flags, gb_flags, + path, newext); if (err) goto cleanup; depth = ext_depth(inode); @@ -2000,7 +2031,7 @@ has_space: merge: /* try to merge extents */ - if (!(flag & EXT4_GET_BLOCKS_PRE_IO)) + if (!(gb_flags & EXT4_GET_BLOCKS_PRE_IO)) ext4_ext_try_to_merge(handle, inode, path, nearex); @@ -2043,7 +2074,7 @@ static int ext4_fill_fiemap_extents(struct inode *inode, path = NULL; } - path = ext4_ext_find_extent(inode, block, path); + path = ext4_ext_find_extent(inode, block, path, 0); if (IS_ERR(path)) { up_read(&EXT4_I(inode)->i_data_sem); err = PTR_ERR(path); @@ -2705,7 +2736,7 @@ again: ext4_lblk_t ee_block; /* find extent for this block */ - path = ext4_ext_find_extent(inode, end, NULL); + path = ext4_ext_find_extent(inode, end, NULL, EXT4_EX_NOCACHE); if (IS_ERR(path)) { ext4_journal_stop(handle); return PTR_ERR(path); @@ -2747,6 +2778,7 @@ again: */ err = ext4_split_extent_at(handle, inode, path, end + 1, split_flag, + EXT4_EX_NOCACHE | EXT4_GET_BLOCKS_PRE_IO | EXT4_GET_BLOCKS_METADATA_NOFAIL); @@ -2823,7 +2855,8 @@ again: i + 1, ext4_idx_pblock(path[i].p_idx)); memset(path + i + 1, 0, sizeof(*path)); bh = read_extent_tree_block(inode, - ext4_idx_pblock(path[i].p_idx), depth - i - 1); + ext4_idx_pblock(path[i].p_idx), depth - i - 1, + EXT4_EX_NOCACHE); if (IS_ERR(bh)) { /* should we reset i_size? */ err = PTR_ERR(bh); @@ -3170,7 +3203,7 @@ static int ext4_split_extent(handle_t *handle, * result in split of original leaf or extent zeroout. */ ext4_ext_drop_refs(path); - path = ext4_ext_find_extent(inode, map->m_lblk, path); + path = ext4_ext_find_extent(inode, map->m_lblk, path, 0); if (IS_ERR(path)) return PTR_ERR(path); depth = ext_depth(inode); @@ -3554,7 +3587,7 @@ static int ext4_convert_unwritten_extents_endio(handle_t *handle, if (err < 0) goto out; ext4_ext_drop_refs(path); - path = ext4_ext_find_extent(inode, map->m_lblk, path); + path = ext4_ext_find_extent(inode, map->m_lblk, path, 0); if (IS_ERR(path)) { err = PTR_ERR(path); goto out; @@ -4041,7 +4074,7 @@ int ext4_ext_map_blocks(handle_t *handle, struct inode *inode, trace_ext4_ext_map_blocks_enter(inode, map->m_lblk, map->m_len, flags); /* find extent for this block */ - path = ext4_ext_find_extent(inode, map->m_lblk, NULL); + path = ext4_ext_find_extent(inode, map->m_lblk, NULL, 0); if (IS_ERR(path)) { err = PTR_ERR(path); path = NULL; @@ -4760,6 +4793,6 @@ int ext4_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo, error = ext4_fill_fiemap_extents(inode, start_blk, len_blks, fieinfo); } - + ext4_es_lru_add(inode); return error; } -- cgit v1.2.3 From 7869a4a6c5caa7b2e5c41ccaf46eb3371f88eea7 Mon Sep 17 00:00:00 2001 From: Theodore Ts'o Date: Fri, 16 Aug 2013 22:05:14 -0400 Subject: ext4: add support for extent pre-caching Add a new fiemap flag which forces the all of the extents in an inode to be cached in the extent_status tree. This is critically important when using AIO to a preallocated file, since if we need to read in blocks from the extent tree, the io_submit(2) system call becomes synchronous, and the AIO is no longer "A", which is bad. In addition, for most files which have an external leaf tree block, the cost of caching the information in the extent status tree will be less than caching the entire 4k block in the buffer cache. So it is generally a win to keep the extent information cached. Signed-off-by: "Theodore Ts'o" --- fs/ext4/extents.c | 73 ++++++++++++++++++++++++++++++++++++++++++++++++++++++- 1 file changed, 72 insertions(+), 1 deletion(-) (limited to 'fs/ext4/extents.c') diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c index 08c1ac976479..01838875fcaf 100644 --- a/fs/ext4/extents.c +++ b/fs/ext4/extents.c @@ -482,7 +482,7 @@ __read_extent_tree_block(const char *function, unsigned int line, if (err < 0) goto errout; } - if (buffer_verified(bh)) + if (buffer_verified(bh) && !(flags & EXT4_EX_FORCE_CACHE)) return bh; err = __ext4_ext_check(function, line, inode, ext_block_hdr(bh), depth, pblk); @@ -526,6 +526,71 @@ errout: __read_extent_tree_block(__func__, __LINE__, (inode), (pblk), \ (depth), (flags)) +/* + * This function is called to cache a file's extent information in the + * extent status tree + */ +int ext4_ext_precache(struct inode *inode) +{ + struct ext4_inode_info *ei = EXT4_I(inode); + struct ext4_ext_path *path = NULL; + struct buffer_head *bh; + int i = 0, depth, ret = 0; + + if (!ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS)) + return 0; /* not an extent-mapped inode */ + + down_read(&ei->i_data_sem); + depth = ext_depth(inode); + + path = kzalloc(sizeof(struct ext4_ext_path) * (depth + 1), + GFP_NOFS); + if (path == NULL) { + up_read(&ei->i_data_sem); + return -ENOMEM; + } + + /* Don't cache anything if there are no external extent blocks */ + if (depth == 0) + goto out; + path[0].p_hdr = ext_inode_hdr(inode); + ret = ext4_ext_check(inode, path[0].p_hdr, depth, 0); + if (ret) + goto out; + path[0].p_idx = EXT_FIRST_INDEX(path[0].p_hdr); + while (i >= 0) { + /* + * If this is a leaf block or we've reached the end of + * the index block, go up + */ + if ((i == depth) || + path[i].p_idx > EXT_LAST_INDEX(path[i].p_hdr)) { + brelse(path[i].p_bh); + path[i].p_bh = NULL; + i--; + continue; + } + bh = read_extent_tree_block(inode, + ext4_idx_pblock(path[i].p_idx++), + depth - i - 1, + EXT4_EX_FORCE_CACHE); + if (IS_ERR(bh)) { + ret = PTR_ERR(bh); + break; + } + i++; + path[i].p_bh = bh; + path[i].p_hdr = ext_block_hdr(bh); + path[i].p_idx = EXT_FIRST_INDEX(path[i].p_hdr); + } + ext4_set_inode_state(inode, EXT4_STATE_EXT_PRECACHED); +out: + up_read(&ei->i_data_sem); + ext4_ext_drop_refs(path); + kfree(path); + return ret; +} + #ifdef EXT_DEBUG static void ext4_ext_show_path(struct inode *inode, struct ext4_ext_path *path) { @@ -4766,6 +4831,12 @@ int ext4_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo, return error; } + if (fieinfo->fi_flags & FIEMAP_FLAG_CACHE) { + error = ext4_ext_precache(inode); + if (error) + return error; + } + /* fallback to generic here if not in extents fmt */ if (!(ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS))) return generic_block_fiemap(inode, fieinfo, start, len, -- cgit v1.2.3 From 7d7345322d60edb0fa49a64a89b31360f01d09cb Mon Sep 17 00:00:00 2001 From: Jan Kara Date: Sat, 17 Aug 2013 09:36:54 -0400 Subject: ext4: fix warning in ext4_da_update_reserve_space() reaim workfile.dbase test easily triggers warning in ext4_da_update_reserve_space(): EXT4-fs warning (device ram0): ext4_da_update_reserve_space:365: ino 12, allocated 1 with only 0 reserved metadata blocks (releasing 1 blocks with reserved 9 data blocks) The problem is that (one of) tests creates file and then randomly writes to it with O_SYNC. That results in writing back pages of the file in random order so we create extents for written blocks say 0, 2, 4, 6, 8 - this last allocation also allocates new block for extents. Then we writeout block 1 so we have extents 0-2, 4, 6, 8 and we release indirect extent block because extents fit in the inode again. Then we writeout block 10 and we need to allocate indirect extent block again which triggers the warning because we don't have the reservation anymore. Fix the problem by giving back freed metadata blocks resulting from extent merging into inode's reservation pool. Signed-off-by: Jan Kara --- fs/ext4/extents.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'fs/ext4/extents.c') diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c index 01838875fcaf..62b21ccea882 100644 --- a/fs/ext4/extents.c +++ b/fs/ext4/extents.c @@ -1793,7 +1793,8 @@ static void ext4_ext_try_to_merge_up(handle_t *handle, brelse(path[1].p_bh); ext4_free_blocks(handle, inode, NULL, blk, 1, - EXT4_FREE_BLOCKS_METADATA | EXT4_FREE_BLOCKS_FORGET); + EXT4_FREE_BLOCKS_METADATA | EXT4_FREE_BLOCKS_FORGET | + EXT4_FREE_BLOCKS_RESERVE); } /* -- cgit v1.2.3 From 27b1b22882d32aa711ab4801700dad997440d940 Mon Sep 17 00:00:00 2001 From: Andi Shyti Date: Wed, 28 Aug 2013 14:00:00 -0400 Subject: ext4: fix use of potentially uninitialized variables in debugging code If ext_debugging is enabled and path[depth].p_ext is NULL, len and lblock are printed non initialized Signed-off-by: Andi Shyti Signed-off-by: "Theodore Ts'o" --- fs/ext4/extents.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) (limited to 'fs/ext4/extents.c') diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c index 62b21ccea882..e7580aeb6fbc 100644 --- a/fs/ext4/extents.c +++ b/fs/ext4/extents.c @@ -2285,8 +2285,8 @@ ext4_ext_put_gap_in_cache(struct inode *inode, struct ext4_ext_path *path, ext4_lblk_t block) { int depth = ext_depth(inode); - unsigned long len; - ext4_lblk_t lblock; + unsigned long len = 0; + ext4_lblk_t lblock = 0; struct ext4_extent *ex; ex = path[depth].p_ext; @@ -2323,7 +2323,6 @@ ext4_ext_put_gap_in_cache(struct inode *inode, struct ext4_ext_path *path, ext4_es_insert_extent(inode, lblock, len, ~0, EXTENT_STATUS_HOLE); } else { - lblock = len = 0; BUG(); } -- cgit v1.2.3 From 70261f568f3c08552f034742e3d5cb78c3877766 Mon Sep 17 00:00:00 2001 From: Anatol Pomozov Date: Wed, 28 Aug 2013 14:40:12 -0400 Subject: ext4: Fix misspellings using 'codespell' tool Signed-off-by: Anatol Pomozov Signed-off-by: "Theodore Ts'o" --- fs/ext4/extents.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs/ext4/extents.c') diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c index e7580aeb6fbc..916e88498c87 100644 --- a/fs/ext4/extents.c +++ b/fs/ext4/extents.c @@ -3220,7 +3220,7 @@ fix_extent_len: * ext4_split_extents() splits an extent and mark extent which is covered * by @map as split_flags indicates * - * It may result in splitting the extent into multiple extents (upto three) + * It may result in splitting the extent into multiple extents (up to three) * There are three possibilities: * a> There is no split required * b> Splits in two extents: Split is happening at either end of the extent -- cgit v1.2.3 From d7b2a00c2e2eedf460ce2a15237f28de40412d86 Mon Sep 17 00:00:00 2001 From: Zheng Liu Date: Wed, 28 Aug 2013 14:47:06 -0400 Subject: ext4: isolate ext4_extents.h file After applied the commit (4a092d73), we have reduced the number of source files that need to #include ext4_extents.h. But we can do better. This commit defines ext4_zeroout_es() in extents.c and move EXT_MAX_BLOCKS into ext4.h in order not to include ext4_extents.h in indirect.c and ioctl.c. Meanwhile we just need to include this file in extent_status.c when ES_AGGRESSIVE_TEST is defined. Otherwise, this commit removes a duplicated declaration in trace/events/ext4.h. After applied this patch, we just need to include ext4_extents.h file in {super,migrate,move_extents,extents}.c, and it is easy for us to define a new extent disk layout. Signed-off-by: Zheng Liu Signed-off-by: "Theodore Ts'o" --- fs/ext4/extents.c | 21 +++++++++++++++++++-- 1 file changed, 19 insertions(+), 2 deletions(-) (limited to 'fs/ext4/extents.c') diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c index 916e88498c87..54d52afcdb19 100644 --- a/fs/ext4/extents.c +++ b/fs/ext4/extents.c @@ -3048,6 +3048,23 @@ void ext4_ext_release(struct super_block *sb) #endif } +static int ext4_zeroout_es(struct inode *inode, struct ext4_extent *ex) +{ + ext4_lblk_t ee_block; + ext4_fsblk_t ee_pblock; + unsigned int ee_len; + + ee_block = le32_to_cpu(ex->ee_block); + ee_len = ext4_ext_get_actual_len(ex); + ee_pblock = ext4_ext_pblock(ex); + + if (ee_len == 0) + return 0; + + return ext4_es_insert_extent(inode, ee_block, ee_len, ee_pblock, + EXTENT_STATUS_WRITTEN); +} + /* FIXME!! we need to try to merge to left or right after zero-out */ static int ext4_ext_zeroout(struct inode *inode, struct ext4_extent *ex) { @@ -3200,7 +3217,7 @@ static int ext4_split_extent_at(handle_t *handle, goto fix_extent_len; /* update extent status tree */ - err = ext4_es_zeroout(inode, &zero_ex); + err = ext4_zeroout_es(inode, &zero_ex); goto out; } else if (err) @@ -3551,7 +3568,7 @@ static int ext4_ext_convert_to_initialized(handle_t *handle, out: /* If we have gotten a failure, don't zero out status tree */ if (!err) - err = ext4_es_zeroout(inode, &zero_ex); + err = ext4_zeroout_es(inode, &zero_ex); return err ? err : allocated; } -- cgit v1.2.3