diff options
Diffstat (limited to 'fs/ext3/inode.c')
-rw-r--r-- | fs/ext3/inode.c | 170 |
1 files changed, 86 insertions, 84 deletions
diff --git a/fs/ext3/inode.c b/fs/ext3/inode.c index 2edd7eec88f..dcf4f1dd108 100644 --- a/fs/ext3/inode.c +++ b/fs/ext3/inode.c @@ -13,11 +13,11 @@ * Copyright (C) 1991, 1992 Linus Torvalds * * Goal-directed block allocation by Stephen Tweedie - * (sct@redhat.com), 1993, 1998 + * (sct@redhat.com), 1993, 1998 * Big-endian to little-endian byte-swapping/bitmaps by * David S. Miller (davem@caip.rutgers.edu), 1995 * 64-bit file support on 64-bit platforms by Jakub Jelinek - * (jj@sunsite.ms.mff.cuni.cz) + * (jj@sunsite.ms.mff.cuni.cz) * * Assorted race fixes, rewrite of ext3_get_block() by Al Viro, 2000 */ @@ -55,14 +55,14 @@ static int ext3_inode_is_fast_symlink(struct inode *inode) /* * The ext3 forget function must perform a revoke if we are freeing data * which has been journaled. Metadata (eg. indirect blocks) must be - * revoked in all cases. + * revoked in all cases. * * "bh" may be NULL: a metadata block may have been freed from memory * but there may still be a record of it in the journal, and that record * still needs to be revoked. */ int ext3_forget(handle_t *handle, int is_metadata, struct inode *inode, - struct buffer_head *bh, int blocknr) + struct buffer_head *bh, ext3_fsblk_t blocknr) { int err; @@ -105,7 +105,7 @@ int ext3_forget(handle_t *handle, int is_metadata, struct inode *inode, * Work out how many blocks we need to proceed with the next chunk of a * truncate transaction. */ -static unsigned long blocks_for_truncate(struct inode *inode) +static unsigned long blocks_for_truncate(struct inode *inode) { unsigned long needed; @@ -122,13 +122,13 @@ static unsigned long blocks_for_truncate(struct inode *inode) /* But we need to bound the transaction so we don't overflow the * journal. */ - if (needed > EXT3_MAX_TRANS_DATA) + if (needed > EXT3_MAX_TRANS_DATA) needed = EXT3_MAX_TRANS_DATA; return EXT3_DATA_TRANS_BLOCKS(inode->i_sb) + needed; } -/* +/* * Truncate transactions can be complex and absolutely huge. So we need to * be able to restart the transaction at a conventient checkpoint to make * sure we don't overflow the journal. @@ -136,9 +136,9 @@ static unsigned long blocks_for_truncate(struct inode *inode) * start_transaction gets us a new handle for a truncate transaction, * and extend_transaction tries to extend the existing one a bit. If * extend fails, we need to propagate the failure up and restart the - * transaction in the top-level truncate loop. --sct + * transaction in the top-level truncate loop. --sct */ -static handle_t *start_transaction(struct inode *inode) +static handle_t *start_transaction(struct inode *inode) { handle_t *result; @@ -215,12 +215,12 @@ void ext3_delete_inode (struct inode * inode) ext3_orphan_del(handle, inode); EXT3_I(inode)->i_dtime = get_seconds(); - /* + /* * One subtle ordering requirement: if anything has gone wrong * (transaction abort, IO errors, whatever), then we can still * do these next steps (the fs will already have been marked as * having errors), but we can't free the inode if the mark_dirty - * fails. + * fails. */ if (ext3_mark_inode_dirty(handle, inode)) /* If that failed, just do the required in-core inode clear. */ @@ -398,7 +398,7 @@ no_block: * + if there is a block to the left of our position - allocate near it. * + if pointer will live in indirect block - allocate near that block. * + if pointer will live in inode - allocate in the same - * cylinder group. + * cylinder group. * * In the latter case we colour the starting block by the callers PID to * prevent it from clashing with concurrent allocations for a different inode @@ -407,13 +407,13 @@ no_block: * * Caller must make sure that @ind is valid and will stay that way. */ -static unsigned long ext3_find_near(struct inode *inode, Indirect *ind) +static ext3_fsblk_t ext3_find_near(struct inode *inode, Indirect *ind) { struct ext3_inode_info *ei = EXT3_I(inode); __le32 *start = ind->bh ? (__le32*) ind->bh->b_data : ei->i_data; __le32 *p; - unsigned long bg_start; - unsigned long colour; + ext3_fsblk_t bg_start; + ext3_grpblk_t colour; /* Try to find previous block */ for (p = ind->p - 1; p >= start; p--) { @@ -429,8 +429,7 @@ static unsigned long ext3_find_near(struct inode *inode, Indirect *ind) * It is going to be referred to from the inode itself? OK, just put it * into the same cylinder group then. */ - bg_start = (ei->i_block_group * EXT3_BLOCKS_PER_GROUP(inode->i_sb)) + - le32_to_cpu(EXT3_SB(inode->i_sb)->s_es->s_first_data_block); + bg_start = ext3_group_first_block_no(inode->i_sb, ei->i_block_group); colour = (current->pid % 16) * (EXT3_BLOCKS_PER_GROUP(inode->i_sb) / 16); return bg_start + colour; @@ -448,7 +447,7 @@ static unsigned long ext3_find_near(struct inode *inode, Indirect *ind) * stores it in *@goal and returns zero. */ -static unsigned long ext3_find_goal(struct inode *inode, long block, +static ext3_fsblk_t ext3_find_goal(struct inode *inode, long block, Indirect chain[4], Indirect *partial) { struct ext3_block_alloc_info *block_i; @@ -471,7 +470,7 @@ static unsigned long ext3_find_goal(struct inode *inode, long block, * ext3_blks_to_allocate: Look up the block map and count the number * of direct blocks need to be allocated for the given branch. * - * @branch: chain of indirect blocks + * @branch: chain of indirect blocks * @k: number of blocks need for indirect blocks * @blks: number of data blocks to be mapped. * @blocks_to_boundary: the offset in the indirect block @@ -516,13 +515,13 @@ static int ext3_blks_to_allocate(Indirect *branch, int k, unsigned long blks, * direct blocks */ static int ext3_alloc_blocks(handle_t *handle, struct inode *inode, - unsigned long goal, int indirect_blks, int blks, - unsigned long long new_blocks[4], int *err) + ext3_fsblk_t goal, int indirect_blks, int blks, + ext3_fsblk_t new_blocks[4], int *err) { int target, i; unsigned long count = 0; int index = 0; - unsigned long current_block = 0; + ext3_fsblk_t current_block = 0; int ret = 0; /* @@ -592,7 +591,7 @@ failed_out: * as described above and return 0. */ static int ext3_alloc_branch(handle_t *handle, struct inode *inode, - int indirect_blks, int *blks, unsigned long goal, + int indirect_blks, int *blks, ext3_fsblk_t goal, int *offsets, Indirect *branch) { int blocksize = inode->i_sb->s_blocksize; @@ -600,8 +599,8 @@ static int ext3_alloc_branch(handle_t *handle, struct inode *inode, int err = 0; struct buffer_head *bh; int num; - unsigned long long new_blocks[4]; - unsigned long long current_block; + ext3_fsblk_t new_blocks[4]; + ext3_fsblk_t current_block; num = ext3_alloc_blocks(handle, inode, goal, indirect_blks, *blks, new_blocks, &err); @@ -688,7 +687,7 @@ static int ext3_splice_branch(handle_t *handle, struct inode *inode, int i; int err = 0; struct ext3_block_alloc_info *block_i; - unsigned long current_block; + ext3_fsblk_t current_block; block_i = EXT3_I(inode)->i_block_alloc_info; /* @@ -745,7 +744,7 @@ static int ext3_splice_branch(handle_t *handle, struct inode *inode, jbd_debug(5, "splicing indirect only\n"); BUFFER_TRACE(where->bh, "call ext3_journal_dirty_metadata"); err = ext3_journal_dirty_metadata(handle, where->bh); - if (err) + if (err) goto err_out; } else { /* @@ -795,13 +794,13 @@ int ext3_get_blocks_handle(handle_t *handle, struct inode *inode, int offsets[4]; Indirect chain[4]; Indirect *partial; - unsigned long goal; + ext3_fsblk_t goal; int indirect_blks; int blocks_to_boundary = 0; int depth; struct ext3_inode_info *ei = EXT3_I(inode); int count = 0; - unsigned long first_block = 0; + ext3_fsblk_t first_block = 0; J_ASSERT(handle != NULL || create == 0); @@ -819,7 +818,7 @@ int ext3_get_blocks_handle(handle_t *handle, struct inode *inode, count++; /*map more blocks*/ while (count < maxblocks && count <= blocks_to_boundary) { - unsigned long blk; + ext3_fsblk_t blk; if (!verify_chain(chain, partial)) { /* @@ -926,7 +925,7 @@ int ext3_get_blocks_handle(handle_t *handle, struct inode *inode, set_buffer_new(bh_result); got_it: map_bh(bh_result, inode->i_sb, le32_to_cpu(chain[depth-1].key)); - if (blocks_to_boundary == 0) + if (count > blocks_to_boundary) set_buffer_boundary(bh_result); err = count; /* Clean up and exit */ @@ -1010,11 +1009,14 @@ struct buffer_head *ext3_getblk(handle_t *handle, struct inode *inode, buffer_trace_init(&dummy.b_history); err = ext3_get_blocks_handle(handle, inode, block, 1, &dummy, create, 1); - if (err == 1) { + /* + * ext3_get_blocks_handle() returns number of blocks + * mapped. 0 in case of a HOLE. + */ + if (err > 0) { + if (err > 1) + WARN_ON(1); err = 0; - } else if (err >= 0) { - WARN_ON(1); - err = -EIO; } *errp = err; if (!err && buffer_mapped(&dummy)) { @@ -1096,7 +1098,7 @@ static int walk_page_buffers( handle_t *handle, for ( bh = head, block_start = 0; ret == 0 && (bh != head || !block_start); - block_start = block_end, bh = next) + block_start = block_end, bh = next) { next = bh->b_this_page; block_end = block_start + blocksize; @@ -1135,7 +1137,7 @@ static int walk_page_buffers( handle_t *handle, * So what we do is to rely on the fact that journal_stop/journal_start * will _not_ run commit under these circumstances because handle->h_ref * is elevated. We'll still have enough credits for the tiny quotafile - * write. + * write. */ static int do_journal_get_write_access(handle_t *handle, struct buffer_head *bh) @@ -1159,7 +1161,7 @@ retry: ret = PTR_ERR(handle); goto out; } - if (test_opt(inode->i_sb, NOBH)) + if (test_opt(inode->i_sb, NOBH) && ext3_should_writeback_data(inode)) ret = nobh_prepare_write(page, from, to, ext3_get_block); else ret = block_prepare_write(page, from, to, ext3_get_block); @@ -1245,7 +1247,7 @@ static int ext3_writeback_commit_write(struct file *file, struct page *page, if (new_i_size > EXT3_I(inode)->i_disksize) EXT3_I(inode)->i_disksize = new_i_size; - if (test_opt(inode->i_sb, NOBH)) + if (test_opt(inode->i_sb, NOBH) && ext3_should_writeback_data(inode)) ret = nobh_commit_write(file, page, from, to); else ret = generic_commit_write(file, page, from, to); @@ -1280,7 +1282,7 @@ static int ext3_journalled_commit_write(struct file *file, if (inode->i_size > EXT3_I(inode)->i_disksize) { EXT3_I(inode)->i_disksize = inode->i_size; ret2 = ext3_mark_inode_dirty(handle, inode); - if (!ret) + if (!ret) ret = ret2; } ret2 = ext3_journal_stop(handle); @@ -1289,7 +1291,7 @@ static int ext3_journalled_commit_write(struct file *file, return ret; } -/* +/* * bmap() is special. It gets used by applications such as lilo and by * the swapper to find the on-disk block of a specific piece of data. * @@ -1298,10 +1300,10 @@ static int ext3_journalled_commit_write(struct file *file, * filesystem and enables swap, then they may get a nasty shock when the * data getting swapped to that swapfile suddenly gets overwritten by * the original zero's written out previously to the journal and - * awaiting writeback in the kernel's buffer cache. + * awaiting writeback in the kernel's buffer cache. * * So, if we see any bmap calls here on a modified, data-journaled file, - * take extra steps to flush any blocks which might be in the cache. + * take extra steps to flush any blocks which might be in the cache. */ static sector_t ext3_bmap(struct address_space *mapping, sector_t block) { @@ -1310,16 +1312,16 @@ static sector_t ext3_bmap(struct address_space *mapping, sector_t block) int err; if (EXT3_I(inode)->i_state & EXT3_STATE_JDATA) { - /* + /* * This is a REALLY heavyweight approach, but the use of * bmap on dirty files is expected to be extremely rare: * only if we run lilo or swapon on a freshly made file - * do we expect this to happen. + * do we expect this to happen. * * (bmap requires CAP_SYS_RAWIO so this does not * represent an unprivileged user DOS attack --- we'd be * in trouble if mortal users could trigger this path at - * will.) + * will.) * * NB. EXT3_STATE_JDATA is not set on files other than * regular files. If somebody wants to bmap a directory @@ -1455,7 +1457,7 @@ static int ext3_ordered_writepage(struct page *page, */ /* - * And attach them to the current transaction. But only if + * And attach them to the current transaction. But only if * block_write_full_page() succeeded. Otherwise they are unmapped, * and generally junk. */ @@ -1495,7 +1497,7 @@ static int ext3_writeback_writepage(struct page *page, goto out_fail; } - if (test_opt(inode->i_sb, NOBH)) + if (test_opt(inode->i_sb, NOBH) && ext3_should_writeback_data(inode)) ret = nobh_writepage(page, ext3_get_block, wbc); else ret = block_write_full_page(page, ext3_get_block, wbc); @@ -1642,7 +1644,7 @@ static ssize_t ext3_direct_IO(int rw, struct kiocb *iocb, } } - ret = blockdev_direct_IO(rw, iocb, inode, inode->i_sb->s_bdev, iov, + ret = blockdev_direct_IO(rw, iocb, inode, inode->i_sb->s_bdev, iov, offset, nr_segs, ext3_get_block, NULL); @@ -1699,7 +1701,7 @@ static int ext3_journalled_set_page_dirty(struct page *page) return __set_page_dirty_nobuffers(page); } -static struct address_space_operations ext3_ordered_aops = { +static const struct address_space_operations ext3_ordered_aops = { .readpage = ext3_readpage, .readpages = ext3_readpages, .writepage = ext3_ordered_writepage, @@ -1713,7 +1715,7 @@ static struct address_space_operations ext3_ordered_aops = { .migratepage = buffer_migrate_page, }; -static struct address_space_operations ext3_writeback_aops = { +static const struct address_space_operations ext3_writeback_aops = { .readpage = ext3_readpage, .readpages = ext3_readpages, .writepage = ext3_writeback_writepage, @@ -1727,7 +1729,7 @@ static struct address_space_operations ext3_writeback_aops = { .migratepage = buffer_migrate_page, }; -static struct address_space_operations ext3_journalled_aops = { +static const struct address_space_operations ext3_journalled_aops = { .readpage = ext3_readpage, .readpages = ext3_readpages, .writepage = ext3_journalled_writepage, @@ -1759,7 +1761,7 @@ void ext3_set_aops(struct inode *inode) static int ext3_block_truncate_page(handle_t *handle, struct page *page, struct address_space *mapping, loff_t from) { - unsigned long index = from >> PAGE_CACHE_SHIFT; + ext3_fsblk_t index = from >> PAGE_CACHE_SHIFT; unsigned offset = from & (PAGE_CACHE_SIZE-1); unsigned blocksize, iblock, length, pos; struct inode *inode = mapping->host; @@ -1960,7 +1962,7 @@ no_top: * than `count' because there can be holes in there. */ static void ext3_clear_blocks(handle_t *handle, struct inode *inode, - struct buffer_head *bh, unsigned long block_to_free, + struct buffer_head *bh, ext3_fsblk_t block_to_free, unsigned long count, __le32 *first, __le32 *last) { __le32 *p; @@ -2022,12 +2024,12 @@ static void ext3_free_data(handle_t *handle, struct inode *inode, struct buffer_head *this_bh, __le32 *first, __le32 *last) { - unsigned long block_to_free = 0; /* Starting block # of a run */ - unsigned long count = 0; /* Number of blocks in the run */ + ext3_fsblk_t block_to_free = 0; /* Starting block # of a run */ + unsigned long count = 0; /* Number of blocks in the run */ __le32 *block_to_free_p = NULL; /* Pointer into inode/ind corresponding to block_to_free */ - unsigned long nr; /* Current block # */ + ext3_fsblk_t nr; /* Current block # */ __le32 *p; /* Pointer into inode/ind for current block */ int err; @@ -2052,7 +2054,7 @@ static void ext3_free_data(handle_t *handle, struct inode *inode, } else if (nr == block_to_free + count) { count++; } else { - ext3_clear_blocks(handle, inode, this_bh, + ext3_clear_blocks(handle, inode, this_bh, block_to_free, count, block_to_free_p, p); block_to_free = nr; @@ -2089,7 +2091,7 @@ static void ext3_free_branches(handle_t *handle, struct inode *inode, struct buffer_head *parent_bh, __le32 *first, __le32 *last, int depth) { - unsigned long nr; + ext3_fsblk_t nr; __le32 *p; if (is_handle_aborted(handle)) @@ -2113,7 +2115,7 @@ static void ext3_free_branches(handle_t *handle, struct inode *inode, */ if (!bh) { ext3_error(inode->i_sb, "ext3_free_branches", - "Read failure, inode=%ld, block=%ld", + "Read failure, inode=%lu, block="E3FSBLK, inode->i_ino, nr); continue; } @@ -2182,7 +2184,7 @@ static void ext3_free_branches(handle_t *handle, struct inode *inode, *p = 0; BUFFER_TRACE(parent_bh, "call ext3_journal_dirty_metadata"); - ext3_journal_dirty_metadata(handle, + ext3_journal_dirty_metadata(handle, parent_bh); } } @@ -2394,22 +2396,24 @@ out_stop: ext3_journal_stop(handle); } -static unsigned long ext3_get_inode_block(struct super_block *sb, +static ext3_fsblk_t ext3_get_inode_block(struct super_block *sb, unsigned long ino, struct ext3_iloc *iloc) { unsigned long desc, group_desc, block_group; - unsigned long offset, block; + unsigned long offset; + ext3_fsblk_t block; struct buffer_head *bh; struct ext3_group_desc * gdp; - - if ((ino != EXT3_ROOT_INO && ino != EXT3_JOURNAL_INO && - ino != EXT3_RESIZE_INO && ino < EXT3_FIRST_INO(sb)) || - ino > le32_to_cpu(EXT3_SB(sb)->s_es->s_inodes_count)) { - ext3_error(sb, "ext3_get_inode_block", - "bad inode number: %lu", ino); + if (!ext3_valid_inum(sb, ino)) { + /* + * This error is already checked for in namei.c unless we are + * looking at an NFS filehandle, in which case no error + * report is needed + */ return 0; } + block_group = (ino - 1) / EXT3_INODES_PER_GROUP(sb); if (block_group >= EXT3_SB(sb)->s_groups_count) { ext3_error(sb,"ext3_get_inode_block","group >= groups count"); @@ -2448,7 +2452,7 @@ static unsigned long ext3_get_inode_block(struct super_block *sb, static int __ext3_get_inode_loc(struct inode *inode, struct ext3_iloc *iloc, int in_mem) { - unsigned long block; + ext3_fsblk_t block; struct buffer_head *bh; block = ext3_get_inode_block(inode->i_sb, inode->i_ino, iloc); @@ -2459,7 +2463,8 @@ static int __ext3_get_inode_loc(struct inode *inode, if (!bh) { ext3_error (inode->i_sb, "ext3_get_inode_loc", "unable to read inode block - " - "inode=%lu, block=%lu", inode->i_ino, block); + "inode=%lu, block="E3FSBLK, + inode->i_ino, block); return -EIO; } if (!buffer_uptodate(bh)) { @@ -2540,7 +2545,7 @@ make_io: if (!buffer_uptodate(bh)) { ext3_error(inode->i_sb, "ext3_get_inode_loc", "unable to read inode block - " - "inode=%lu, block=%lu", + "inode=%lu, block="E3FSBLK, inode->i_ino, block); brelse(bh); return -EIO; @@ -2627,9 +2632,6 @@ void ext3_read_inode(struct inode * inode) * recovery code: that's fine, we're about to complete * the process of deleting those. */ } - inode->i_blksize = PAGE_SIZE; /* This is the optimal IO size - * (for stat), not the fs block - * size */ inode->i_blocks = le32_to_cpu(raw_inode->i_blocks); ei->i_flags = le32_to_cpu(raw_inode->i_flags); #ifdef EXT3_FRAGMENTS @@ -2699,7 +2701,7 @@ void ext3_read_inode(struct inode * inode) if (raw_inode->i_block[0]) init_special_inode(inode, inode->i_mode, old_decode_dev(le32_to_cpu(raw_inode->i_block[0]))); - else + else init_special_inode(inode, inode->i_mode, new_decode_dev(le32_to_cpu(raw_inode->i_block[1]))); } @@ -2719,8 +2721,8 @@ bad_inode: * * The caller must have write access to iloc->bh. */ -static int ext3_do_update_inode(handle_t *handle, - struct inode *inode, +static int ext3_do_update_inode(handle_t *handle, + struct inode *inode, struct ext3_iloc *iloc) { struct ext3_inode *raw_inode = ext3_raw_inode(iloc); @@ -2895,7 +2897,7 @@ int ext3_write_inode(struct inode *inode, int wait) * commit will leave the blocks being flushed in an unused state on * disk. (On recovery, the inode will get truncated and the blocks will * be freed, so we have a strong guarantee that no future commit will - * leave these blocks visible to the user.) + * leave these blocks visible to the user.) * * Called with inode->sem down. */ @@ -3038,13 +3040,13 @@ int ext3_mark_iloc_dirty(handle_t *handle, return err; } -/* +/* * On success, We end up with an outstanding reference count against - * iloc->bh. This _must_ be cleaned up later. + * iloc->bh. This _must_ be cleaned up later. */ int -ext3_reserve_inode_write(handle_t *handle, struct inode *inode, +ext3_reserve_inode_write(handle_t *handle, struct inode *inode, struct ext3_iloc *iloc) { int err = 0; @@ -3134,7 +3136,7 @@ out: } #if 0 -/* +/* * Bind an inode's backing buffer_head into this transaction, to prevent * it from being flushed to disk early. Unlike * ext3_reserve_inode_write, this leaves behind no bh reference and @@ -3152,7 +3154,7 @@ static int ext3_pin_inode(handle_t *handle, struct inode *inode) BUFFER_TRACE(iloc.bh, "get_write_access"); err = journal_get_write_access(handle, iloc.bh); if (!err) - err = ext3_journal_dirty_metadata(handle, + err = ext3_journal_dirty_metadata(handle, iloc.bh); brelse(iloc.bh); } |