diff options
Diffstat (limited to 'fs/ext3')
| -rw-r--r-- | fs/ext3/Kconfig | 1 | ||||
| -rw-r--r-- | fs/ext3/inode.c | 80 | ||||
| -rw-r--r-- | fs/ext3/namei.c | 3 | ||||
| -rw-r--r-- | fs/ext3/resize.c | 2 | ||||
| -rw-r--r-- | fs/ext3/super.c | 17 |
5 files changed, 49 insertions, 54 deletions
diff --git a/fs/ext3/Kconfig b/fs/ext3/Kconfig index 522b15498f4..e8c6ba0e4a3 100644 --- a/fs/ext3/Kconfig +++ b/fs/ext3/Kconfig @@ -31,6 +31,7 @@ config EXT3_FS config EXT3_DEFAULTS_TO_ORDERED bool "Default to 'data=ordered' in ext3" depends on EXT3_FS + default y help The journal mode options for ext3 have different tradeoffs between when data is guaranteed to be on disk and diff --git a/fs/ext3/inode.c b/fs/ext3/inode.c index 735f0190ec2..001eb0e2d48 100644 --- a/fs/ext3/inode.c +++ b/fs/ext3/inode.c @@ -1149,9 +1149,25 @@ static int walk_page_buffers( handle_t *handle, static int do_journal_get_write_access(handle_t *handle, struct buffer_head *bh) { + int dirty = buffer_dirty(bh); + int ret; + if (!buffer_mapped(bh) || buffer_freed(bh)) return 0; - return ext3_journal_get_write_access(handle, bh); + /* + * __block_prepare_write() could have dirtied some buffers. Clean + * the dirty bit as jbd2_journal_get_write_access() could complain + * otherwise about fs integrity issues. Setting of the dirty bit + * by __block_prepare_write() isn't a real problem here as we clear + * the bit before releasing a page lock and thus writeback cannot + * ever write the buffer. + */ + if (dirty) + clear_buffer_dirty(bh); + ret = ext3_journal_get_write_access(handle, bh); + if (!ret && dirty) + ret = ext3_journal_dirty_metadata(handle, bh); + return ret; } /* @@ -1625,10 +1641,7 @@ static int ext3_writeback_writepage(struct page *page, goto out_fail; } - 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); + ret = block_write_full_page(page, ext3_get_block, wbc); err = ext3_journal_stop(handle); if (!ret) @@ -1922,17 +1935,6 @@ static int ext3_block_truncate_page(handle_t *handle, struct page *page, length = blocksize - (offset & (blocksize - 1)); iblock = index << (PAGE_CACHE_SHIFT - inode->i_sb->s_blocksize_bits); - /* - * For "nobh" option, we can only work if we don't need to - * read-in the page - otherwise we create buffers to do the IO. - */ - if (!page_has_buffers(page) && test_opt(inode->i_sb, NOBH) && - ext3_should_writeback_data(inode) && PageUptodate(page)) { - zero_user(page, offset, length); - set_page_dirty(page); - goto unlock; - } - if (!page_has_buffers(page)) create_empty_buffers(page, blocksize, 0); @@ -2284,27 +2286,6 @@ static void ext3_free_branches(handle_t *handle, struct inode *inode, depth); /* - * We've probably journalled the indirect block several - * times during the truncate. But it's no longer - * needed and we now drop it from the transaction via - * journal_revoke(). - * - * That's easy if it's exclusively part of this - * transaction. But if it's part of the committing - * transaction then journal_forget() will simply - * brelse() it. That means that if the underlying - * block is reallocated in ext3_get_block(), - * unmap_underlying_metadata() will find this block - * and will try to get rid of it. damn, damn. - * - * If this block has already been committed to the - * journal, a revoke record will be written. And - * revoke records must be emitted *before* clearing - * this block's bit in the bitmaps. - */ - ext3_forget(handle, 1, inode, bh, bh->b_blocknr); - - /* * Everything below this this pointer has been * released. Now let this top-of-subtree go. * @@ -2327,6 +2308,31 @@ static void ext3_free_branches(handle_t *handle, struct inode *inode, truncate_restart_transaction(handle, inode); } + /* + * We've probably journalled the indirect block several + * times during the truncate. But it's no longer + * needed and we now drop it from the transaction via + * journal_revoke(). + * + * That's easy if it's exclusively part of this + * transaction. But if it's part of the committing + * transaction then journal_forget() will simply + * brelse() it. That means that if the underlying + * block is reallocated in ext3_get_block(), + * unmap_underlying_metadata() will find this block + * and will try to get rid of it. damn, damn. Thus + * we don't allow a block to be reallocated until + * a transaction freeing it has fully committed. + * + * We also have to make sure journal replay after a + * crash does not overwrite non-journaled data blocks + * with old metadata when the block got reallocated for + * data. Thus we have to store a revoke record for a + * block in the same transaction in which we free the + * block. + */ + ext3_forget(handle, 1, inode, bh, bh->b_blocknr); + ext3_free_blocks(handle, inode, nr, 1); if (parent_bh) { diff --git a/fs/ext3/namei.c b/fs/ext3/namei.c index ee184084ca4..2b35ddb70d6 100644 --- a/fs/ext3/namei.c +++ b/fs/ext3/namei.c @@ -1447,7 +1447,6 @@ static int ext3_add_entry (handle_t *handle, struct dentry *dentry, struct inode *inode) { struct inode *dir = dentry->d_parent->d_inode; - unsigned long offset; struct buffer_head * bh; struct ext3_dir_entry_2 *de; struct super_block * sb; @@ -1469,7 +1468,7 @@ static int ext3_add_entry (handle_t *handle, struct dentry *dentry, ext3_mark_inode_dirty(handle, dir); } blocks = dir->i_size >> sb->s_blocksize_bits; - for (block = 0, offset = 0; block < blocks; block++) { + for (block = 0; block < blocks; block++) { bh = ext3_bread(handle, dir, block, 0, &retval); if(!bh) return retval; diff --git a/fs/ext3/resize.c b/fs/ext3/resize.c index 54351ac7cef..0ccd7b12b73 100644 --- a/fs/ext3/resize.c +++ b/fs/ext3/resize.c @@ -964,7 +964,6 @@ int ext3_group_extend(struct super_block *sb, struct ext3_super_block *es, ext3_fsblk_t n_blocks_count) { ext3_fsblk_t o_blocks_count; - unsigned long o_groups_count; ext3_grpblk_t last; ext3_grpblk_t add; struct buffer_head * bh; @@ -976,7 +975,6 @@ int ext3_group_extend(struct super_block *sb, struct ext3_super_block *es, * yet: we're going to revalidate es->s_blocks_count after * taking the s_resize_lock below. */ o_blocks_count = le32_to_cpu(es->s_blocks_count); - o_groups_count = EXT3_SB(sb)->s_groups_count; if (test_opt(sb, DEBUG)) printk(KERN_DEBUG "EXT3-fs: extending last group from "E3FSBLK" uto "E3FSBLK" blocks\n", diff --git a/fs/ext3/super.c b/fs/ext3/super.c index 6c953bb255e..9650a956fd0 100644 --- a/fs/ext3/super.c +++ b/fs/ext3/super.c @@ -661,9 +661,6 @@ static int ext3_show_options(struct seq_file *seq, struct vfsmount *vfs) */ seq_puts(seq, ",barrier="); seq_puts(seq, test_opt(sb, BARRIER) ? "1" : "0"); - if (test_opt(sb, NOBH)) - seq_puts(seq, ",nobh"); - seq_printf(seq, ",data=%s", data_mode_string(test_opt(sb, DATA_FLAGS))); if (test_opt(sb, DATA_ERR_ABORT)) seq_puts(seq, ",data_err=abort"); @@ -1255,10 +1252,12 @@ set_qf_format: *n_blocks_count = option; break; case Opt_nobh: - set_opt(sbi->s_mount_opt, NOBH); + ext3_msg(sb, KERN_WARNING, + "warning: ignoring deprecated nobh option"); break; case Opt_bh: - clear_opt(sbi->s_mount_opt, NOBH); + ext3_msg(sb, KERN_WARNING, + "warning: ignoring deprecated bh option"); break; default: ext3_msg(sb, KERN_ERR, @@ -2001,14 +2000,6 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent) break; } - if (test_opt(sb, NOBH)) { - if (!(test_opt(sb, DATA_FLAGS) == EXT3_MOUNT_WRITEBACK_DATA)) { - ext3_msg(sb, KERN_WARNING, - "warning: ignoring nobh option - " - "it is supported only with writeback mode"); - clear_opt(sbi->s_mount_opt, NOBH); - } - } /* * The journal_load will have done any necessary log recovery, * so we can safely mount the rest of the filesystem now. |