From: Akira Fujita Subject: [RFC][PATCH 3/5] Move the file data to the new blocks Date: Thu, 27 Dec 2007 20:11:45 +0900 Message-ID: <200712271111.AA00288@TNESG9526.rs.jp.nec.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii To: linux-ext4@vger.kernel.org, linux-fsdevel@vger.kernel.org Return-path: Sender: linux-fsdevel-owner@vger.kernel.org List-Id: linux-ext4.vger.kernel.org Move the blocks on the temporary inode to the original inode by a page. 1. Read the file data from the old blocks to the page 2. Move the block on the temporary inode to the original inode 3. Write the file data on the page into the new blocks *This patch is applied on the top of ext4 git tree(linux-2.6.24-rc5). http://repo.or.cz/r/ext4-patch-queue.git Signed-off-by: Takashi Sato Signed-off-by: Akira Fujita --- diff -X linux-2.6.24-rc5-defrag/Documentation/dontdiff -upNr linux-2.6.24-rc5-move-data/fs/ext4/defrag.c linux-2.6.24-rc5-rf-option/fs/ext4/defrag.c --- linux-2.6.24-rc5-move-data/fs/ext4/defrag.c 2007-12-25 20:47:03.000000000 +0900 +++ linux-2.6.24-rc5-rf-option/fs/ext4/defrag.c 2007-12-25 20:28:16.000000000 +0900 @@ -13,6 +13,24 @@ #include #include "group.h" +/* Will go away */ +ext4_fsblk_t idx_pblock(struct ext4_extent_idx *ix) +{ + ext4_fsblk_t block; + + block = le32_to_cpu(ix->ei_leaf_lo); + block |= ((ext4_fsblk_t) le16_to_cpu(ix->ei_leaf_hi) << 31) << 1; + return block; +} + +/* Will go away */ +static void ext4_ext_store_pblock(struct ext4_extent *ex, ext4_fsblk_t pb) +{ + ex->ee_start_lo = cpu_to_le32((unsigned long) (pb & 0xffffffff)); + ex->ee_start_hi = + cpu_to_le16((unsigned long) ((pb >> 31) >> 1) & 0xffff); +} + /* * this structure is used to gather extents from the tree via ioctl */ @@ -33,6 +51,112 @@ struct ext4_extent_tree_stats { int leaf_num; }; +int ext4_ext_walk_space(struct inode *inode, unsigned long block, + unsigned long num, ext_prepare_callback func, + void *cbdata) +{ + struct ext4_ext_path *path = NULL; + struct ext4_ext_cache cbex; + struct ext4_extent *ex; + unsigned long next, start = 0, end = 0; + unsigned long last = block + num; + int depth, exists, err = 0; + + BUG_ON(func == NULL); + BUG_ON(inode == NULL); + + while (block < last && block != EXT_MAX_BLOCK) { + num = last - block; + /* find extent for this block */ + path = ext4_ext_find_extent(inode, block, path); + if (IS_ERR(path)) { + err = PTR_ERR(path); + path = NULL; + break; + } + + depth = ext_depth(inode); + BUG_ON(path[depth].p_hdr == NULL); + ex = path[depth].p_ext; + next = ext4_ext_next_allocated_block(path); + + exists = 0; + if (!ex) { + /* there is no extent yet, so try to allocate + * all requested space */ + start = block; + end = block + num; + } else if (le32_to_cpu(ex->ee_block) > block) { + /* need to allocate space before found extent */ + start = block; + end = le32_to_cpu(ex->ee_block); + if (block + num < end) + end = block + num; + } else if (block >= le32_to_cpu(ex->ee_block) + + ext4_ext_get_actual_len(ex)) { + /* need to allocate space after found extent */ + start = block; + end = block + num; + if (end >= next) + end = next; + } else if (block >= le32_to_cpu(ex->ee_block)) { + /* + * some part of requested space is covered + * by found extent + */ + start = block; + end = le32_to_cpu(ex->ee_block) + + ext4_ext_get_actual_len(ex); + if (block + num < end) + end = block + num; + exists = 1; + } else { + BUG(); + } + BUG_ON(end <= start); + + if (!exists) { + cbex.ec_block = start; + cbex.ec_len = end - start; + cbex.ec_start = 0; + cbex.ec_type = EXT4_EXT_CACHE_GAP; + } else { + cbex.ec_block = le32_to_cpu(ex->ee_block); + cbex.ec_len = ext4_ext_get_actual_len(ex); + cbex.ec_start = ext_pblock(ex); + cbex.ec_type = EXT4_EXT_CACHE_EXTENT; + } + + BUG_ON(cbex.ec_len == 0); + err = func(inode, path, &cbex, cbdata); + ext4_ext_drop_refs(path); + + if (err < 0) + break; + if (err == EXT_REPEAT) + continue; + else if (err == EXT_BREAK) { + err = 0; + break; + } + + if (ext_depth(inode) != depth) { + /* depth was changed. we have to realloc path */ + kfree(path); + path = NULL; + } + + block = cbex.ec_block + cbex.ec_len; + } + + if (path) { + ext4_ext_drop_refs(path); + kfree(path); + } + + return err; +} + static int ext4_ext_store_extent_cb(struct inode *inode, struct ext4_ext_path *path, @@ -212,6 +336,621 @@ int ext4_ext_ioctl(struct inode *inode, } /** + * ext4_ext_merge_across - merge extents across leaf block + * + * @handle journal handle + * @inode target file's inode + * @o_start first original extent to be defraged + * @o_end last original extent to be defraged + * @start_ext first new extent to be merged + * @new_ext middle of new extent to be merged + * @end_ext last new extent to be merged + * @flag defrag mode (e.g. -f) + * + * This function returns 0 if succeed, otherwise returns error value. + */ +static int +ext4_ext_merge_across_blocks(handle_t *handle, struct inode *inode, + struct ext4_extent *o_start, + struct ext4_extent *o_end, struct ext4_extent *start_ext, + struct ext4_extent *new_ext, struct ext4_extent *end_ext, + int flag) +{ + struct ext4_ext_path *org_path = NULL; + unsigned long eblock = 0; + int err = 0; + int new_flag = 0; + int end_flag = 0; + int defrag_flag; + + if (flag == DEFRAG_FORCE_VICTIM) + defrag_flag = 1; + else + defrag_flag = 0; + + if (le16_to_cpu(start_ext->ee_len) && + le16_to_cpu(new_ext->ee_len) && + le16_to_cpu(end_ext->ee_len)) { + + if ((o_start) == (o_end)) { + + /* start_ext new_ext end_ext + * dest |---------|-----------|--------| + * org |------------------------------| + */ + + end_flag = 1; + } else { + + /* start_ext new_ext end_ext + * dest |---------|----------|---------| + * org |---------------|--------------| + */ + + o_end->ee_block = end_ext->ee_block; + o_end->ee_len = end_ext->ee_len; + ext4_ext_store_pblock(o_end, ext_pblock(end_ext)); + } + + o_start->ee_len = start_ext->ee_len; + new_flag = 1; + + } else if ((le16_to_cpu(start_ext->ee_len)) && + (le16_to_cpu(new_ext->ee_len)) && + (!le16_to_cpu(end_ext->ee_len)) && + ((o_start) == (o_end))) { + + /* start_ext new_ext + * dest |--------------|---------------| + * org |------------------------------| + */ + + o_start->ee_len = start_ext->ee_len; + new_flag = 1; + + } else if ((!le16_to_cpu(start_ext->ee_len)) && + (le16_to_cpu(new_ext->ee_len)) && + (le16_to_cpu(end_ext->ee_len)) && + ((o_start) == (o_end))) { + + /* new_ext end_ext + * dest |--------------|---------------| + * org |------------------------------| + */ + + o_end->ee_block = end_ext->ee_block; + o_end->ee_len = end_ext->ee_len; + ext4_ext_store_pblock(o_end, ext_pblock(end_ext)); + + /* If new_ext was first block */ + if (!new_ext->ee_block) + eblock = 0; + else + eblock = le32_to_cpu(new_ext->ee_block); + + new_flag = 1; + } else { + printk(KERN_ERR "Unexpected case \n"); + return -EIO; + } + + if (new_flag) { + org_path = ext4_ext_find_extent(inode, eblock, NULL); + if (IS_ERR(org_path)) { + err = PTR_ERR(org_path); + org_path = NULL; + goto ERR; + } + err = ext4_ext_insert_extent_defrag(handle, inode, + org_path, new_ext, defrag_flag); + if (err) + goto ERR; + } + + if (end_flag) { + org_path = ext4_ext_find_extent(inode, + le32_to_cpu(end_ext->ee_block) - 1, org_path); + if (IS_ERR(org_path)) { + err = PTR_ERR(org_path); + org_path = NULL; + goto ERR; + } + err = ext4_ext_insert_extent_defrag(handle, inode, + org_path, end_ext, defrag_flag); + if (err) + goto ERR; + } +ERR: + if (org_path) { + ext4_ext_drop_refs(org_path); + kfree(org_path); + } + + return err; + +} + +/** + * ext4_ext_merge_inside_block - merge new extent to the extent block + * + * @handle journal handle + * @inode target file's inode + * @o_start first original extent to be defraged + * @o_end last original extent to be merged + * @start_ext first new extent to be merged + * @new_ext middle of new extent to be merged + * @end_ext last new extent to be merged + * @eh extent header of target leaf block + * @replaced the number of blocks which will be replaced with new_ext + * @range_to_move used to dicide how to merge + * + * This function always returns 0. +*/ +static int +ext4_ext_merge_inside_block(handle_t *handle, struct inode *inode, + struct ext4_extent *o_start, struct ext4_extent *o_end, + struct ext4_extent *start_ext, struct ext4_extent *new_ext, + struct ext4_extent *end_ext, struct ext4_extent_header *eh, + ext4_fsblk_t replaced, int range_to_move) +{ + int i = 0; + unsigned len; + + /* Move the existing extents */ + if (range_to_move && o_end < EXT_LAST_EXTENT(eh)) { + len = EXT_LAST_EXTENT(eh) - (o_end + 1) + 1; + len = len * sizeof(struct ext4_extent); + memmove(o_end + 1 + range_to_move, o_end + 1, len); + } + + /* Insert start entry */ + if (le16_to_cpu(start_ext->ee_len)) + o_start[i++].ee_len = start_ext->ee_len; + + /* Insert new entry */ + if (le16_to_cpu(new_ext->ee_len)) { + o_start[i].ee_block = new_ext->ee_block; + o_start[i].ee_len = cpu_to_le16(replaced); + ext4_ext_store_pblock(&o_start[i++], ext_pblock(new_ext)); + } + + /* Insert end entry */ + if (end_ext->ee_len) + o_start[i] = *end_ext; + + /* Increment the total entries counter on the extent block */ + eh->eh_entries + = cpu_to_le16(le16_to_cpu(eh->eh_entries) + range_to_move); + + return 0; +} + +/** + * ext4_ext_merge_extents - merge new extent + * + * @handle journal handle + * @inode target file's inode + * @org_path path indicates first extent to be defraged + * @o_start first original extent to be defraged + * @o_end last original extent to be defraged + * @start_ext first new extent to be merged + * @new_ext middle of new extent to be merged + * @end_ext last new extent to be merged + * @replaced the number of blocks which will be replaced with new_ext + * @flag defrag mode (e.g. -f) + * + * This function returns 0 if succeed, otherwise returns error value. + */ +static int +ext4_ext_merge_extents(handle_t *handle, struct inode *inode, + struct ext4_ext_path *org_path, + struct ext4_extent *o_start, struct ext4_extent *o_end, + struct ext4_extent *start_ext, struct ext4_extent *new_ext, + struct ext4_extent *end_ext, ext4_fsblk_t replaced, int flag) +{ + struct ext4_extent_header *eh; + unsigned need_slots, slots_range; + int range_to_move, depth, ret; + + /* The extents need to be inserted + * start_extent + new_extent + end_extent + */ + need_slots = (le16_to_cpu(start_ext->ee_len) ? 1 : 0) + + (le16_to_cpu(end_ext->ee_len) ? 1 : 0) + + (le16_to_cpu(new_ext->ee_len) ? 1 : 0); + + /* The number of slots between start and end */ + slots_range = o_end - o_start + 1; + + /* Range to move the end of extent */ + range_to_move = need_slots - slots_range; + depth = org_path->p_depth; + org_path += depth; + eh = org_path->p_hdr; + + if (depth) { + /* Register to journal */ + ret = ext4_journal_get_write_access(handle, org_path->p_bh); + if (ret) + return ret; + } + + /* expansion */ + if ((range_to_move > 0) && + (range_to_move > le16_to_cpu(eh->eh_max) + - le16_to_cpu(eh->eh_entries))) { + + ret = ext4_ext_merge_across_blocks(handle, inode, o_start, + o_end, start_ext, new_ext, + end_ext, flag); + if (ret < 0) + return ret; + } else { + ret = ext4_ext_merge_inside_block(handle, inode, o_start, + o_end, start_ext, new_ext, end_ext, + eh, replaced, range_to_move); + if (ret < 0) + return ret; + } + + if (depth) { + ret = ext4_journal_dirty_metadata(handle, org_path->p_bh); + if (ret) + return ret; + } else { + ret = ext4_mark_inode_dirty(handle, inode); + if (ret < 0) + return ret; + } + + return 0; + +} + +/** + * ext4_ext_defrag_leaf_block - Defragmentation for one leaf extent block. + * @handle journal handle + * @org_inode target inode + * @org_path path indicates first extent to be defraged + * @dext destination extent + * @from start offset on the target file + * @flag defrag mode (e.g. -f) + * + * This function returns 0 if succeed, otherwise returns error value. + */ +static int +ext4_ext_defrag_leaf_block(handle_t *handle, struct inode *org_inode, + struct ext4_ext_path *org_path, struct ext4_extent *dext, + ext4_fsblk_t *from, int flag) +{ + unsigned long depth; + ext4_fsblk_t replaced = 0; + struct ext4_extent *oext, *o_start = NULL, *o_end = NULL, *prev_ext; + struct ext4_extent new_ext, start_ext, end_ext; + ext4_fsblk_t new_end; + ext4_fsblk_t lblock; + unsigned short len; + ext4_fsblk_t new_phys_end; + int ret; + + depth = ext_depth(org_inode); + start_ext.ee_len = end_ext.ee_len = 0; + o_start = o_end = oext = org_path[depth].p_ext; + ext4_ext_store_pblock(&new_ext, ext_pblock(dext)); + new_ext.ee_len = dext->ee_len; + len = le16_to_cpu(new_ext.ee_len); + new_ext.ee_block = cpu_to_le32(*from); + lblock = le32_to_cpu(oext->ee_block); + new_end = le32_to_cpu(new_ext.ee_block) + + le16_to_cpu(new_ext.ee_len) - 1; + new_phys_end = ext_pblock(&new_ext) + + le16_to_cpu(new_ext.ee_len) - 1; + + /* First original extent + * dest |---------------| + * org |---------------| + */ + if (le32_to_cpu(new_ext.ee_block) > + le32_to_cpu(oext->ee_block) && + le32_to_cpu(new_ext.ee_block) < + le32_to_cpu(oext->ee_block) + + le16_to_cpu(oext->ee_len)) { + start_ext.ee_len = cpu_to_le32(le32_to_cpu(new_ext.ee_block) + - le32_to_cpu(oext->ee_block)); + replaced += le16_to_cpu(oext->ee_len) + - le16_to_cpu(start_ext.ee_len); + } else if (oext > EXT_FIRST_EXTENT(org_path[depth].p_hdr)) { + /* We can merge previous extent. */ + prev_ext = oext - 1; + if (((ext_pblock(prev_ext) + le32_to_cpu(prev_ext->ee_len)) + == ext_pblock(&new_ext)) + && (le32_to_cpu(prev_ext->ee_block) + + le32_to_cpu(prev_ext->ee_len) + == le32_to_cpu(new_ext.ee_block))) { + o_start = prev_ext; + start_ext.ee_len = cpu_to_le32( + le16_to_cpu(prev_ext->ee_len) + + le16_to_cpu(new_ext.ee_len)); + new_ext.ee_len = 0; + } + } + for (;;) { + /* The extent for destination must be found. */ + BUG_ON(!oext || lblock != le32_to_cpu(oext->ee_block)); + lblock += le16_to_cpu(oext->ee_len); + + /* Middle of original extent + * dest |-------------------| + * org |-----------------| + */ + if (le32_to_cpu(new_ext.ee_block) <= + le32_to_cpu(oext->ee_block) && + new_end >= le32_to_cpu(oext->ee_block) + + le16_to_cpu(oext->ee_len) - 1) + replaced += le16_to_cpu(oext->ee_len); + + /* Last original extent + * dest |----------------| + * org |---------------| + */ + if (new_end >= le32_to_cpu(oext->ee_block) && + new_end < le32_to_cpu(oext->ee_block) + + le16_to_cpu(oext->ee_len) - 1) { + end_ext.ee_len + = cpu_to_le16(le32_to_cpu(oext->ee_block) + + le16_to_cpu(oext->ee_len) - 1 - new_end); + ext4_ext_store_pblock(&end_ext, (ext_pblock(o_end) + + cpu_to_le16(oext->ee_len) + - cpu_to_le16(end_ext.ee_len))); + end_ext.ee_block + = cpu_to_le32(le32_to_cpu(o_end->ee_block) + + le16_to_cpu(oext->ee_len) + - le16_to_cpu(end_ext.ee_len)); + replaced += le16_to_cpu(oext->ee_len) + - le16_to_cpu(end_ext.ee_len); + } + + /* Detected the block end, reached the number of replaced + * blocks to dext->ee_len. Then, merge the extent. + */ + if (oext == EXT_LAST_EXTENT(org_path[depth].p_hdr) || + new_end <= le32_to_cpu(oext->ee_block) + + le16_to_cpu(oext->ee_len) - 1) { + ret = ext4_ext_merge_extents(handle, org_inode, + org_path, o_start, o_end, &start_ext, + &new_ext, &end_ext, replaced, flag); + if (ret < 0) + return ret; + + /* All expected blocks are replaced */ + if (le16_to_cpu(new_ext.ee_len) <= 0) { + if (DQUOT_ALLOC_BLOCK(org_inode, len)) + return -EDQUOT; + return 0; + } + + /* re-calculate new_ext */ + new_ext.ee_len = cpu_to_le32(le16_to_cpu(new_ext.ee_len) + - replaced); + new_ext.ee_block = + cpu_to_le32(le32_to_cpu(new_ext.ee_block) + + replaced); + ext4_ext_store_pblock(&new_ext, ext_pblock(&new_ext) + + replaced); + replaced = 0; + start_ext.ee_len = end_ext.ee_len = 0; + o_start = NULL; + + /* All expected blocks are replaced */ + if (le16_to_cpu(new_ext.ee_len) <= 0) { + if (DQUOT_ALLOC_BLOCK(org_inode, len)) + return -EDQUOT; + return 0; + } + } + + /* Get next extent for original. */ + if (org_path) + ext4_ext_drop_refs(org_path); + org_path = ext4_ext_find_extent(org_inode, lblock, org_path); + if (IS_ERR(org_path)) { + ret = PTR_ERR(org_path); + org_path = NULL; + return ret; + } + depth = ext_depth(org_inode); + oext = org_path[depth].p_ext; + if (oext->ee_block + oext->ee_len <= lblock) + return -ENOENT; + + o_end = oext; + if (!o_start) + o_start = oext; + } +} + +/** + * ext4_ext_replace_branches - replace original extents with new extents. + * @org_inode Original inode + * @dest_inode temporary inode + * @from_page Page offset + * @count_page Page count to be replaced + * @flag defrag mode (e.g. -f) + * + * This function returns 0 if succeed, otherwise returns error value. + * Replace extents for blocks from "from" to "from+count-1". + */ +static int +ext4_ext_replace_branches(struct inode *org_inode, struct inode *dest_inode, + pgoff_t from_page, pgoff_t dest_from_page, + pgoff_t count_page, int flag) +{ + struct ext4_ext_path *org_path = NULL; + struct ext4_ext_path *dest_path = NULL; + struct ext4_extent *swap_ext = NULL; + struct ext4_extent *oext, *dext; + struct ext4_extent tmp_ext; + struct ext4_extent tmp_ext2; + int err = 0; + int depth; + ext4_fsblk_t from, count, dest_off, diff, org_diff, replaced_count = 0; + handle_t *handle = NULL; + unsigned jnum; + + from = (ext4_fsblk_t)from_page << + (PAGE_CACHE_SHIFT - dest_inode->i_blkbits); + count = (ext4_fsblk_t)count_page << + (PAGE_CACHE_SHIFT - dest_inode->i_blkbits); + dest_off = (ext4_fsblk_t)dest_from_page << + (PAGE_CACHE_SHIFT - dest_inode->i_blkbits); + jnum = ext4_ext_writepage_trans_blocks(org_inode, count) + 3; + handle = ext4_journal_start(org_inode, jnum); + if (IS_ERR(handle)) { + err = PTR_ERR(handle); + goto out; + } + + /* Get the original extent for the block "from" */ + org_path = ext4_ext_find_extent(org_inode, from, NULL); + if (IS_ERR(org_path)) { + err = PTR_ERR(org_path); + org_path = NULL; + goto out; + } + + /* Get the destination extent for the head */ + dest_path = ext4_ext_find_extent(dest_inode, dest_off, NULL); + if (IS_ERR(dest_path)) { + err = PTR_ERR(dest_path); + dest_path = NULL; + goto out; + } + depth = ext_depth(dest_inode); + dext = dest_path[depth].p_ext; + /* When dext is too large, pick up the target range. */ + diff = dest_off - le32_to_cpu(dext->ee_block); + ext4_ext_store_pblock(&tmp_ext, ext_pblock(dext) + diff); + tmp_ext.ee_block = cpu_to_le32(le32_to_cpu(dext->ee_block) + diff); + tmp_ext.ee_len = cpu_to_le16(le16_to_cpu(dext->ee_len) - diff); + if (count < le16_to_cpu(tmp_ext.ee_len)) + tmp_ext.ee_len = cpu_to_le16(count); + dext = &tmp_ext; + + depth = ext_depth(org_inode); + oext = org_path[depth].p_ext; + org_diff = from - le32_to_cpu(oext->ee_block); + ext4_ext_store_pblock(&tmp_ext2, ext_pblock(oext) + org_diff); + tmp_ext2.ee_block = tmp_ext.ee_block; + + /* adjust extent length when blocksize != pagesize */ + if (tmp_ext.ee_len <= (oext->ee_len - org_diff)) { + tmp_ext2.ee_len = tmp_ext.ee_len; + } else { + tmp_ext2.ee_len = oext->ee_len - org_diff; + tmp_ext.ee_len = tmp_ext2.ee_len; + } + swap_ext = &tmp_ext2; + + /* loop for the destination extents */ + while (1) { + /* The extent for destination must be found. */ + BUG_ON(!dext || dest_off != le32_to_cpu(dext->ee_block)); + + /* loop for the original extent blocks */ + err = ext4_ext_defrag_leaf_block(handle, org_inode, + org_path, dext, &from, flag); + if (err < 0) + goto out; + + /* We need the function which fixes extent information for + * inserting. + * e.g. ext4_ext_merge_extents(). + */ + err = ext4_ext_defrag_leaf_block(handle, dest_inode, + dest_path, swap_ext, &dest_off, -1); + if (err < 0) + goto out; + + replaced_count += le16_to_cpu(dext->ee_len); + dest_off += le16_to_cpu(dext->ee_len); + from += le16_to_cpu(dext->ee_len); + + /* Already moved the expected blocks */ + if (replaced_count >= count) + break; + + if (org_path) + ext4_ext_drop_refs(org_path); + org_path = ext4_ext_find_extent(org_inode, from, NULL); + if (IS_ERR(org_path)) { + err = PTR_ERR(org_path); + org_path = NULL; + goto out; + } + depth = ext_depth(org_inode); + oext = org_path[depth].p_ext; + if (oext->ee_block + oext->ee_len <= from) { + err = 0; + goto out; + } + + if (dest_path) + ext4_ext_drop_refs(dest_path); + dest_path = ext4_ext_find_extent(dest_inode, dest_off, NULL); + if (IS_ERR(dest_path)) { + err = PTR_ERR(dest_path); + dest_path = NULL; + goto out; + } + depth = ext_depth(dest_inode); + dext = dest_path[depth].p_ext; + if (dext->ee_block + dext->ee_len <= dest_off) { + err = 0; + goto out; + } + + /* When dext is too large, pick up the target range. */ + diff = dest_off - le32_to_cpu(dext->ee_block); + ext4_ext_store_pblock(&tmp_ext, ext_pblock(dext) + diff); + tmp_ext.ee_block = + cpu_to_le32(le32_to_cpu(dext->ee_block) + diff); + tmp_ext.ee_len = cpu_to_le16(le16_to_cpu(dext->ee_len) - diff); + + if ((count - replaced_count) < le16_to_cpu(tmp_ext.ee_len)) + tmp_ext.ee_len = count - replaced_count ; + + dext = &tmp_ext; + + org_diff = from - le32_to_cpu(oext->ee_block); + ext4_ext_store_pblock(&tmp_ext2, ext_pblock(oext) + org_diff); + tmp_ext2.ee_block = tmp_ext.ee_block; + + /* adjust extent length when blocksize != pagesize */ + if (tmp_ext.ee_len <= (oext->ee_len - org_diff)) { + tmp_ext2.ee_len = tmp_ext.ee_len; + } else { + tmp_ext2.ee_len = oext->ee_len - org_diff; + tmp_ext.ee_len = tmp_ext2.ee_len; + } + swap_ext = &tmp_ext2; + } + +out: + if (handle) + ext4_journal_stop(handle); + if (org_path) { + ext4_ext_drop_refs(org_path); + kfree(org_path); + } + if (dest_path) { + ext4_ext_drop_refs(dest_path); + kfree(dest_path); + } + + return err; +} + +/** * ext4_ext_alloc_blocks - allocate contiguous blocks to temporary inode * @dest_inode temporary inode for multiple block allocation * @org_inode original inode @@ -412,6 +1151,188 @@ out2: } /** + * ext4_ext_defrag_partial - defrag original file partially + * @filp: pointer to file + * @org_offset: page index on original file + * @dest_offset: page index on temporary file + * @flag: defrag mode (e.g. -f) + * + * This function returns 0 if succeeded, otherwise returns error value + */ +static int +ext4_ext_defrag_partial(struct inode *tmp_inode, struct file *filp, + pgoff_t org_offset, pgoff_t dest_offset, int flag) +{ + struct inode *inode = filp->f_dentry->d_inode; + struct address_space *mapping = inode->i_mapping; + struct page *page; + pgoff_t offset_in_page = PAGE_SIZE; + int ret = 0; + + mutex_unlock(&EXT4_I(inode)->truncate_mutex); + page = read_cache_page(inode->i_mapping, org_offset, + (filler_t *)inode->i_mapping->a_ops->readpage, NULL); + mutex_lock(&EXT4_I(inode)->truncate_mutex); + + if (IS_ERR(page)) { + ret = PTR_ERR(page); + return ret; + } + + lock_page(page); + + /* + * try_to_release_page() doesn't call relasepage in writeback mode. + * We should care about the order of writing to the same file + * by multiple defrag processes. + * It needs to call wait_on_page_writeback() to wait for the + * writeback of the page. + */ + if (PageWriteback(page)) + wait_on_page_writeback(page); + + /* release old bh and drop refs */ + try_to_release_page(page, 0); + ret = ext4_ext_replace_branches(inode, tmp_inode, org_offset, + dest_offset, 1, flag); + if (ret < 0) + goto ERR; + + /* Clear the inode cache not to refer to the old data. */ + ext4_ext_invalidate_cache(inode); + + if (org_offset == ((inode->i_size - 1) >> PAGE_CACHE_SHIFT)) { + offset_in_page = (inode->i_size & (PAGE_CACHE_SIZE - 1)); + /* + * If org_offset is the last page and i_size is + * multiples of PAGE_CACHE_SIZE, set PAGE_CACHE_SIZE to + * offset_in_page not to be 0. + */ + if (offset_in_page == 0) + offset_in_page = PAGE_CACHE_SIZE; + } + + mutex_unlock(&EXT4_I(inode)->truncate_mutex); + ret = mapping->a_ops->prepare_write(filp, page, + 0, offset_in_page); + mutex_lock(&EXT4_I(inode)->truncate_mutex); + if (ret) + goto ERR; + + ret = mapping->a_ops->commit_write(filp, page, + 0, offset_in_page); +ERR: + unlock_page(page); + page_cache_release(page); + + return (ret < 0 ? ret : 0); +} + +/** + * ext4_ext_defrag_partial2 - defrag_partial with write_{begin, end} + * @filp: pointer to file + * @org_offset: page index on original file + * @dest_offset: page index on temporary file + * @flag: defrag mode (e.g. -f) + * + * This function returns 0 if succeeded, otherwise returns error value + */ +static int +ext4_ext_defrag_partial2(struct inode *tmp_inode, struct file *filp, + pgoff_t org_offset, pgoff_t dest_offset, int flag) +{ + struct inode *inode = filp->f_dentry->d_inode; + struct address_space *mapping = inode->i_mapping; + struct buffer_head *bh; + struct page *page; + const struct address_space_operations *a_ops = mapping->a_ops; + pgoff_t offset_in_page = PAGE_SIZE; + int ret = 0; + int blocksize = inode->i_sb->s_blocksize; + int blocks_per_page = 0; + int i = 0; + long long offs = org_offset << PAGE_CACHE_SHIFT; + unsigned long blk_off = 0; + unsigned int w_flags = 0; + void *fsdata; + + if (segment_eq(get_fs(), KERNEL_DS)) + w_flags |= AOP_FLAG_UNINTERRUPTIBLE; + + if (org_offset == ((inode->i_size - 1) >> PAGE_CACHE_SHIFT)) { + offset_in_page = (inode->i_size & (PAGE_CACHE_SIZE - 1)); + /* + * If org_offset is the last page and i_size is + * multiples of PAGE_CACHE_SIZE, set PAGE_CACHE_SIZE to + * offset_in_page not to be 0. + */ + if (offset_in_page == 0) + offset_in_page = PAGE_CACHE_SIZE; + } + + mutex_unlock(&EXT4_I(inode)->truncate_mutex); + ret = a_ops->write_begin(filp, mapping, offs, + offset_in_page, w_flags, &page, &fsdata); + mutex_lock(&EXT4_I(inode)->truncate_mutex); + + if (unlikely(ret < 0)) + goto ERR; + + if (!PageUptodate(page)) { + mapping->a_ops->readpage(filp, page); + lock_page(page); + } + + /* + * try_to_release_page() doesn't call relasepage in writeback mode. + * We should care about the order of writing to the same file + * by multiple defrag processes. + * It needs to call wait_on_page_writeback() to wait for the + * writeback of the page. + */ + if (PageWriteback(page)) + wait_on_page_writeback(page); + + /* release old bh and drop refs */ + try_to_release_page(page, 0); + ret = ext4_ext_replace_branches(inode, tmp_inode, org_offset, + dest_offset, 1, flag); + + if (ret < 0) + goto ERR; + + /* Clear the inode cache not to refer to the old data. */ + ext4_ext_invalidate_cache(inode); + + if (!page_has_buffers(page)) + create_empty_buffers(page, 1 << inode->i_blkbits, 0); + + blocks_per_page = PAGE_SIZE / blocksize; + blk_off = org_offset * blocks_per_page; + + bh = page_buffers(page); + for (i = 0; i < blocks_per_page; i++) { + mutex_unlock(&EXT4_I(inode)->truncate_mutex); + ret = ext4_get_block(inode, blk_off++, bh, 0); + mutex_lock(&EXT4_I(inode)->truncate_mutex); + + if (ret < 0) + goto ERR; + + if (bh->b_this_page != NULL) + bh = bh->b_this_page; + } + + ret = a_ops->write_end(filp, mapping, offs, offset_in_page, + offset_in_page, page, fsdata); + + if (unlikely(ret < 0)) + goto ERR; +ERR: + return (ret < 0 ? ret : 0); +} + +/** * ext4_ext_new_extent_tree - allocate contiguous blocks * @inode: inode of the original file * @tmp_inode: inode of the temporary file diff -X linux-2.6.24-rc5-defrag/Documentation/dontdiff -upNr linux-2.6.24-rc5-move-data/fs/ext4/extents.c linux-2.6.24-rc5-rf-option/fs/ext4/extents.c --- linux-2.6.24-rc5-move-data/fs/ext4/extents.c 2007-12-25 20:47:33.000000000 +0900 +++ linux-2.6.24-rc5-rf-option/fs/ext4/extents.c 2007-12-25 20:33:18.000000000 +0900 @@ -349,7 +349,7 @@ static void ext4_ext_show_leaf(struct in #define ext4_ext_show_leaf(inode,path) #endif -static void ext4_ext_drop_refs(struct ext4_ext_path *path) +void ext4_ext_drop_refs(struct ext4_ext_path *path) { int depth = path->p_depth; int i; @@ -1165,7 +1165,7 @@ ext4_ext_search_right(struct inode *inod * allocated block. Thus, index entries have to be consistent * with leaves. */ -static ext4_lblk_t +ext4_lblk_t ext4_ext_next_allocated_block(struct ext4_ext_path *path) { int depth; diff -X linux-2.6.24-rc5-defrag/Documentation/dontdiff -upNr linux-2.6.24-rc5-move-data/fs/ext4/inode.c linux-2.6.24-rc5-rf-option/fs/ext4/inode.c --- linux-2.6.24-rc5-move-data/fs/ext4/inode.c 2007-12-25 20:47:54.000000000 +0900 +++ linux-2.6.24-rc5-rf-option/fs/ext4/inode.c 2007-12-25 20:22:56.000000000 +0900 @@ -896,7 +896,7 @@ out: #define DIO_CREDITS (EXT4_RESERVE_TRANS_BLOCKS + 32) -static int ext4_get_block(struct inode *inode, sector_t iblock, +int ext4_get_block(struct inode *inode, sector_t iblock, struct buffer_head *bh_result, int create) { handle_t *handle = ext4_journal_current_handle();