From 91ef4caf800030fa6e5224b8a41f8c74787b303d Mon Sep 17 00:00:00 2001 From: Duane Griffin Date: Fri, 11 Jul 2008 19:27:31 -0400 Subject: ext4: handle corrupted orphan list at mount If the orphan node list includes valid, untruncatable nodes with nlink > 0 the ext4_orphan_cleanup loop which attempts to delete them will not do so, causing it to loop forever. Fix by checking for such nodes in the ext4_orphan_get function. This patch fixes the second case (image hdb.20000009.softlockup.gz) reported in http://bugzilla.kernel.org/show_bug.cgi?id=10882. Signed-off-by: Duane Griffin Signed-off-by: Theodore Ts'o --- fs/ext4/inode.c | 20 ++++++++++++++------ 1 file changed, 14 insertions(+), 6 deletions(-) (limited to 'fs/ext4/inode.c') diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 8d9707746413..269763b66361 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -2305,6 +2305,19 @@ static void ext4_free_branches(handle_t *handle, struct inode *inode, } } +int ext4_can_truncate(struct inode *inode) +{ + if (IS_APPEND(inode) || IS_IMMUTABLE(inode)) + return 0; + if (S_ISREG(inode->i_mode)) + return 1; + if (S_ISDIR(inode->i_mode)) + return 1; + if (S_ISLNK(inode->i_mode)) + return !ext4_inode_is_fast_symlink(inode); + return 0; +} + /* * ext4_truncate() * @@ -2349,12 +2362,7 @@ void ext4_truncate(struct inode *inode) unsigned blocksize = inode->i_sb->s_blocksize; struct page *page; - if (!(S_ISREG(inode->i_mode) || S_ISDIR(inode->i_mode) || - S_ISLNK(inode->i_mode))) - return; - if (ext4_inode_is_fast_symlink(inode)) - return; - if (IS_APPEND(inode) || IS_IMMUTABLE(inode)) + if (!ext4_can_truncate(inode)) return; /* -- cgit v1.2.2 From 71dc8fbcf5f6363342bd636a646eeac7cfef25c3 Mon Sep 17 00:00:00 2001 From: Duane Griffin Date: Fri, 11 Jul 2008 19:27:31 -0400 Subject: ext4: handle deleting corrupted indirect blocks While freeing indirect blocks we attach a journal head to the parent buffer head, free the blocks, then journal the parent. If the indirect block list is corrupted and points to the parent the journal head will be detached when the block is cleared, causing an OOPS. Check for that explicitly and handle it gracefully. This patch fixes the third case (image hdb.20000057.nullderef.gz) reported in http://bugzilla.kernel.org/show_bug.cgi?id=10882. Signed-off-by: Duane Griffin Signed-off-by: Theodore Ts'o --- fs/ext4/inode.c | 16 +++++++++++++++- 1 file changed, 15 insertions(+), 1 deletion(-) (limited to 'fs/ext4/inode.c') diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 269763b66361..7cce96a6935e 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -2179,7 +2179,21 @@ static void ext4_free_data(handle_t *handle, struct inode *inode, if (this_bh) { BUFFER_TRACE(this_bh, "call ext4_journal_dirty_metadata"); - ext4_journal_dirty_metadata(handle, this_bh); + + /* + * The buffer head should have an attached journal head at this + * point. However, if the data is corrupted and an indirect + * block pointed to itself, it would have been detached when + * the block was cleared. Check for this instead of OOPSing. + */ + if (bh2jh(this_bh)) + ext4_journal_dirty_metadata(handle, this_bh); + else + ext4_error(inode->i_sb, __func__, + "circular indirect block detected, " + "inode=%lu, block=%llu", + inode->i_ino, + (unsigned long long) this_bh->b_blocknr); } } -- cgit v1.2.2 From 7061eba75ceb0835ba61e7cbd757a6f9c1e4af92 Mon Sep 17 00:00:00 2001 From: "Aneesh Kumar K.V" Date: Fri, 11 Jul 2008 19:27:31 -0400 Subject: ext4: Use inode preallocation with -o noextents When mballoc is enabled, block allocation for old block-based files are allocated using mballoc allocator instead of old block-based allocator. The old ext3 block reservation is turned off when mballoc is turned on. However, the in-core preallocation is not enabled for block-based/ non-extent based file block allocation. This result in performance regression, as now we don't have "reservation" ore in-core preallocation to prevent interleaved fragmentation in multiple writes workload. This patch fix this by enable per inode in-core preallocation for non extent files when mballoc is used. Signed-off-by: Aneesh Kumar K.V Signed-off-by: Mingming Cao Signed-off-by: "Theodore Ts'o" --- fs/ext4/inode.c | 72 ++++++++++++++++++++++++++++++++++++++++++--------------- 1 file changed, 54 insertions(+), 18 deletions(-) (limited to 'fs/ext4/inode.c') diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 7cce96a6935e..bc950562b5ba 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -508,11 +508,12 @@ static int ext4_blks_to_allocate(Indirect *branch, int k, unsigned long blks, * direct blocks */ static int ext4_alloc_blocks(handle_t *handle, struct inode *inode, - ext4_fsblk_t goal, int indirect_blks, int blks, - ext4_fsblk_t new_blocks[4], int *err) + ext4_lblk_t iblock, ext4_fsblk_t goal, + int indirect_blks, int blks, + ext4_fsblk_t new_blocks[4], int *err) { int target, i; - unsigned long count = 0; + unsigned long count = 0, blk_allocated = 0; int index = 0; ext4_fsblk_t current_block = 0; int ret = 0; @@ -525,12 +526,13 @@ static int ext4_alloc_blocks(handle_t *handle, struct inode *inode, * the first direct block of this branch. That's the * minimum number of blocks need to allocate(required) */ - target = blks + indirect_blks; - - while (1) { + /* first we try to allocate the indirect blocks */ + target = indirect_blks; + while (target > 0) { count = target; /* allocating blocks for indirect blocks and direct blocks */ - current_block = ext4_new_blocks(handle,inode,goal,&count,err); + current_block = ext4_new_meta_blocks(handle, inode, + goal, &count, err); if (*err) goto failed_out; @@ -540,16 +542,48 @@ static int ext4_alloc_blocks(handle_t *handle, struct inode *inode, new_blocks[index++] = current_block++; count--; } - - if (count > 0) + if (count > 0) { + /* + * save the new block number + * for the first direct block + */ + new_blocks[index] = current_block; + printk(KERN_INFO "%s returned more blocks than " + "requested\n", __func__); + WARN_ON(1); break; + } } - /* save the new block number for the first direct block */ - new_blocks[index] = current_block; - + target = blks - count ; + blk_allocated = count; + if (!target) + goto allocated; + /* Now allocate data blocks */ + count = target; + /* allocating blocks for indirect blocks and direct blocks */ + current_block = ext4_new_blocks(handle, inode, iblock, + goal, &count, err); + if (*err && (target == blks)) { + /* + * if the allocation failed and we didn't allocate + * any blocks before + */ + goto failed_out; + } + if (!*err) { + if (target == blks) { + /* + * save the new block number + * for the first direct block + */ + new_blocks[index] = current_block; + } + blk_allocated += count; + } +allocated: /* total number of blocks allocated for direct blocks */ - ret = count; + ret = blk_allocated; *err = 0; return ret; failed_out: @@ -584,8 +618,9 @@ failed_out: * as described above and return 0. */ static int ext4_alloc_branch(handle_t *handle, struct inode *inode, - int indirect_blks, int *blks, ext4_fsblk_t goal, - ext4_lblk_t *offsets, Indirect *branch) + ext4_lblk_t iblock, int indirect_blks, + int *blks, ext4_fsblk_t goal, + ext4_lblk_t *offsets, Indirect *branch) { int blocksize = inode->i_sb->s_blocksize; int i, n = 0; @@ -595,7 +630,7 @@ static int ext4_alloc_branch(handle_t *handle, struct inode *inode, ext4_fsblk_t new_blocks[4]; ext4_fsblk_t current_block; - num = ext4_alloc_blocks(handle, inode, goal, indirect_blks, + num = ext4_alloc_blocks(handle, inode, iblock, goal, indirect_blks, *blks, new_blocks, &err); if (err) return err; @@ -855,8 +890,9 @@ int ext4_get_blocks_handle(handle_t *handle, struct inode *inode, /* * Block out ext4_truncate while we alter the tree */ - err = ext4_alloc_branch(handle, inode, indirect_blks, &count, goal, - offsets + (partial - chain), partial); + err = ext4_alloc_branch(handle, inode, iblock, indirect_blks, + &count, goal, + offsets + (partial - chain), partial); /* * The ext4_splice_branch call will free and forget any buffers -- cgit v1.2.2 From 654b4908bc17a6318d18f3036fecc5155de92f55 Mon Sep 17 00:00:00 2001 From: "Aneesh Kumar K.V" Date: Fri, 11 Jul 2008 19:27:31 -0400 Subject: ext4: cleanup block allocator Move the code related to block allocation to a single function and add helper funtions to differient allocation for data and meta data blocks Signed-off-by: Aneesh Kumar K.V Signed-off-by: Mingming Cao Signed-off-by: "Theodore Ts'o" --- fs/ext4/inode.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs/ext4/inode.c') diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index bc950562b5ba..c5d506dce8c5 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -561,7 +561,7 @@ static int ext4_alloc_blocks(handle_t *handle, struct inode *inode, goto allocated; /* Now allocate data blocks */ count = target; - /* allocating blocks for indirect blocks and direct blocks */ + /* allocating blocks for data blocks */ current_block = ext4_new_blocks(handle, inode, iblock, goal, &count, err); if (*err && (target == blks)) { -- cgit v1.2.2 From 2e9ee850355593e311d9a26542290fe51e152f74 Mon Sep 17 00:00:00 2001 From: "Aneesh Kumar K.V" Date: Fri, 11 Jul 2008 19:27:31 -0400 Subject: ext4: Use page_mkwrite vma_operations to get mmap write notification. We would like to get notified when we are doing a write on mmap section. This is needed with respect to preallocated area. We split the preallocated area into initialzed extent and uninitialzed extent in the call back. This let us handle ENOSPC better. Otherwise we get ENOSPC in the writepage and that would result in data loss. The changes are also needed to handle ENOSPC when writing to an mmap section of files with holes. Acked-by: Jan Kara Signed-off-by: Aneesh Kumar K.V Signed-off-by: Mingming Cao Signed-off-by: "Theodore Ts'o" --- fs/ext4/inode.c | 61 +++++++++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 61 insertions(+) (limited to 'fs/ext4/inode.c') diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index c5d506dce8c5..034fc544aa66 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -3564,3 +3564,64 @@ int ext4_change_inode_journal_flag(struct inode *inode, int val) return err; } + +static int ext4_bh_unmapped(handle_t *handle, struct buffer_head *bh) +{ + return !buffer_mapped(bh); +} + +int ext4_page_mkwrite(struct vm_area_struct *vma, struct page *page) +{ + loff_t size; + unsigned long len; + int ret = -EINVAL; + struct file *file = vma->vm_file; + struct inode *inode = file->f_path.dentry->d_inode; + struct address_space *mapping = inode->i_mapping; + + /* + * Get i_alloc_sem to stop truncates messing with the inode. We cannot + * get i_mutex because we are already holding mmap_sem. + */ + down_read(&inode->i_alloc_sem); + size = i_size_read(inode); + if (page->mapping != mapping || size <= page_offset(page) + || !PageUptodate(page)) { + /* page got truncated from under us? */ + goto out_unlock; + } + ret = 0; + if (PageMappedToDisk(page)) + goto out_unlock; + + if (page->index == size >> PAGE_CACHE_SHIFT) + len = size & ~PAGE_CACHE_MASK; + else + len = PAGE_CACHE_SIZE; + + if (page_has_buffers(page)) { + /* return if we have all the buffers mapped */ + if (!walk_page_buffers(NULL, page_buffers(page), 0, len, NULL, + ext4_bh_unmapped)) + goto out_unlock; + } + /* + * OK, we need to fill the hole... Do write_begin write_end + * to do block allocation/reservation.We are not holding + * inode.i__mutex here. That allow * parallel write_begin, + * write_end call. lock_page prevent this from happening + * on the same page though + */ + ret = mapping->a_ops->write_begin(file, mapping, page_offset(page), + len, AOP_FLAG_UNINTERRUPTIBLE, &page, NULL); + if (ret < 0) + goto out_unlock; + ret = mapping->a_ops->write_end(file, mapping, page_offset(page), + len, len, page, NULL); + if (ret < 0) + goto out_unlock; + ret = 0; +out_unlock: + up_read(&inode->i_alloc_sem); + return ret; +} -- cgit v1.2.2 From cf108bca465dde0c015f32dd453b99457d31c7c7 Mon Sep 17 00:00:00 2001 From: Jan Kara Date: Fri, 11 Jul 2008 19:27:31 -0400 Subject: ext4: Invert the locking order of page_lock and transaction start This changes are needed to support data=ordered mode handling via inodes. This enables us to get rid of the journal heads and buffer heads for data buffers in the ordered mode. With the changes, during tranasaction commit we writeout the inode pages using the writepages()/writepage(). That implies we take page lock during transaction commit. This can cause a deadlock with the locking order page_lock -> jbd2_journal_start, since the jbd2_journal_start can wait for the journal_commit to happen and the journal_commit now needs to take the page lock. To avoid this dead lock reverse the locking order. Signed-off-by: Jan Kara Signed-off-by: Mingming Cao Signed-off-by: Aneesh Kumar K.V Signed-off-by: "Theodore Ts'o" --- fs/ext4/inode.c | 292 +++++++++++++++++++++++++++++--------------------------- 1 file changed, 149 insertions(+), 143 deletions(-) (limited to 'fs/ext4/inode.c') diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 034fc544aa66..320acb6c35bf 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -1239,19 +1239,20 @@ static int ext4_write_begin(struct file *file, struct address_space *mapping, to = from + len; retry: - page = __grab_cache_page(mapping, index); - if (!page) - return -ENOMEM; - *pagep = page; - handle = ext4_journal_start(inode, needed_blocks); if (IS_ERR(handle)) { - unlock_page(page); - page_cache_release(page); ret = PTR_ERR(handle); goto out; } + page = __grab_cache_page(mapping, index); + if (!page) { + ext4_journal_stop(handle); + ret = -ENOMEM; + goto out; + } + *pagep = page; + ret = block_write_begin(file, mapping, pos, len, flags, pagep, fsdata, ext4_get_block); @@ -1261,8 +1262,8 @@ retry: } if (ret) { - ext4_journal_stop(handle); unlock_page(page); + ext4_journal_stop(handle); page_cache_release(page); } @@ -1290,29 +1291,6 @@ static int write_end_fn(handle_t *handle, struct buffer_head *bh) return ext4_journal_dirty_metadata(handle, bh); } -/* - * Generic write_end handler for ordered and writeback ext4 journal modes. - * We can't use generic_write_end, because that unlocks the page and we need to - * unlock the page after ext4_journal_stop, but ext4_journal_stop must run - * after block_write_end. - */ -static int ext4_generic_write_end(struct file *file, - struct address_space *mapping, - loff_t pos, unsigned len, unsigned copied, - struct page *page, void *fsdata) -{ - struct inode *inode = file->f_mapping->host; - - copied = block_write_end(file, mapping, pos, len, copied, page, fsdata); - - if (pos+copied > inode->i_size) { - i_size_write(inode, pos+copied); - mark_inode_dirty(inode); - } - - return copied; -} - /* * We need to pick up the new inode size which generic_commit_write gave us * `file' can be NULL - eg, when called from page_symlink(). @@ -1326,7 +1304,7 @@ static int ext4_ordered_write_end(struct file *file, struct page *page, void *fsdata) { handle_t *handle = ext4_journal_current_handle(); - struct inode *inode = file->f_mapping->host; + struct inode *inode = mapping->host; unsigned from, to; int ret = 0, ret2; @@ -1347,7 +1325,7 @@ static int ext4_ordered_write_end(struct file *file, new_i_size = pos + copied; if (new_i_size > EXT4_I(inode)->i_disksize) EXT4_I(inode)->i_disksize = new_i_size; - ret2 = ext4_generic_write_end(file, mapping, pos, len, copied, + ret2 = generic_write_end(file, mapping, pos, len, copied, page, fsdata); copied = ret2; if (ret2 < 0) @@ -1356,8 +1334,6 @@ static int ext4_ordered_write_end(struct file *file, ret2 = ext4_journal_stop(handle); if (!ret) ret = ret2; - unlock_page(page); - page_cache_release(page); return ret ? ret : copied; } @@ -1368,7 +1344,7 @@ static int ext4_writeback_write_end(struct file *file, struct page *page, void *fsdata) { handle_t *handle = ext4_journal_current_handle(); - struct inode *inode = file->f_mapping->host; + struct inode *inode = mapping->host; int ret = 0, ret2; loff_t new_i_size; @@ -1376,7 +1352,7 @@ static int ext4_writeback_write_end(struct file *file, if (new_i_size > EXT4_I(inode)->i_disksize) EXT4_I(inode)->i_disksize = new_i_size; - ret2 = ext4_generic_write_end(file, mapping, pos, len, copied, + ret2 = generic_write_end(file, mapping, pos, len, copied, page, fsdata); copied = ret2; if (ret2 < 0) @@ -1385,8 +1361,6 @@ static int ext4_writeback_write_end(struct file *file, ret2 = ext4_journal_stop(handle); if (!ret) ret = ret2; - unlock_page(page); - page_cache_release(page); return ret ? ret : copied; } @@ -1425,10 +1399,10 @@ static int ext4_journalled_write_end(struct file *file, ret = ret2; } + unlock_page(page); ret2 = ext4_journal_stop(handle); if (!ret) ret = ret2; - unlock_page(page); page_cache_release(page); return ret ? ret : copied; @@ -1505,12 +1479,16 @@ static int jbd2_journal_dirty_data_fn(handle_t *handle, struct buffer_head *bh) return 0; } +static int ext4_bh_unmapped_or_delay(handle_t *handle, struct buffer_head *bh) +{ + return !buffer_mapped(bh) || buffer_delay(bh); +} + /* - * Note that we always start a transaction even if we're not journalling - * data. This is to preserve ordering: any hole instantiation within - * __block_write_full_page -> ext4_get_block() should be journalled - * along with the data so we don't crash and then get metadata which - * refers to old data. + * Note that we don't need to start a transaction unless we're journaling + * data because we should have holes filled from ext4_page_mkwrite(). If + * we are journaling data, we cannot start transaction directly because + * transaction start ranks above page lock so we have to do some magic... * * In all journalling modes block_write_full_page() will start the I/O. * @@ -1554,10 +1532,8 @@ static int jbd2_journal_dirty_data_fn(handle_t *handle, struct buffer_head *bh) * disastrous. Any write() or metadata operation will sync the fs for * us. * - * AKPM2: if all the page's buffers are mapped to disk and !data=journal, - * we don't need to open a transaction here. */ -static int ext4_ordered_writepage(struct page *page, +static int __ext4_ordered_writepage(struct page *page, struct writeback_control *wbc) { struct inode *inode = page->mapping->host; @@ -1566,22 +1542,6 @@ static int ext4_ordered_writepage(struct page *page, int ret = 0; int err; - J_ASSERT(PageLocked(page)); - - /* - * We give up here if we're reentered, because it might be for a - * different filesystem. - */ - if (ext4_journal_current_handle()) - goto out_fail; - - handle = ext4_journal_start(inode, ext4_writepage_trans_blocks(inode)); - - if (IS_ERR(handle)) { - ret = PTR_ERR(handle); - goto out_fail; - } - if (!page_has_buffers(page)) { create_empty_buffers(page, inode->i_sb->s_blocksize, (1 << BH_Dirty)|(1 << BH_Uptodate)); @@ -1605,54 +1565,135 @@ static int ext4_ordered_writepage(struct page *page, * and generally junk. */ if (ret == 0) { - err = walk_page_buffers(handle, page_bufs, 0, PAGE_CACHE_SIZE, + handle = ext4_journal_start(inode, + ext4_writepage_trans_blocks(inode)); + if (IS_ERR(handle)) { + ret = PTR_ERR(handle); + goto out_put; + } + + ret = walk_page_buffers(handle, page_bufs, 0, PAGE_CACHE_SIZE, NULL, jbd2_journal_dirty_data_fn); + err = ext4_journal_stop(handle); if (!ret) ret = err; } - walk_page_buffers(handle, page_bufs, 0, - PAGE_CACHE_SIZE, NULL, bput_one); - err = ext4_journal_stop(handle); - if (!ret) - ret = err; +out_put: + walk_page_buffers(handle, page_bufs, 0, PAGE_CACHE_SIZE, NULL, + bput_one); return ret; +} + +static int ext4_ordered_writepage(struct page *page, + struct writeback_control *wbc) +{ + struct inode *inode = page->mapping->host; + loff_t size = i_size_read(inode); + loff_t len; + + J_ASSERT(PageLocked(page)); + J_ASSERT(page_has_buffers(page)); + if (page->index == size >> PAGE_CACHE_SHIFT) + len = size & ~PAGE_CACHE_MASK; + else + len = PAGE_CACHE_SIZE; + BUG_ON(walk_page_buffers(NULL, page_buffers(page), 0, len, NULL, + ext4_bh_unmapped_or_delay)); + + /* + * We give up here if we're reentered, because it might be for a + * different filesystem. + */ + if (!ext4_journal_current_handle()) + return __ext4_ordered_writepage(page, wbc); -out_fail: redirty_page_for_writepage(wbc, page); unlock_page(page); - return ret; + return 0; } +static int __ext4_writeback_writepage(struct page *page, + struct writeback_control *wbc) +{ + struct inode *inode = page->mapping->host; + + if (test_opt(inode->i_sb, NOBH)) + return nobh_writepage(page, ext4_get_block, wbc); + else + return block_write_full_page(page, ext4_get_block, wbc); +} + + static int ext4_writeback_writepage(struct page *page, struct writeback_control *wbc) { struct inode *inode = page->mapping->host; + loff_t size = i_size_read(inode); + loff_t len; + + J_ASSERT(PageLocked(page)); + J_ASSERT(page_has_buffers(page)); + if (page->index == size >> PAGE_CACHE_SHIFT) + len = size & ~PAGE_CACHE_MASK; + else + len = PAGE_CACHE_SIZE; + BUG_ON(walk_page_buffers(NULL, page_buffers(page), 0, len, NULL, + ext4_bh_unmapped_or_delay)); + + if (!ext4_journal_current_handle()) + return __ext4_writeback_writepage(page, wbc); + + redirty_page_for_writepage(wbc, page); + unlock_page(page); + return 0; +} + +static int __ext4_journalled_writepage(struct page *page, + struct writeback_control *wbc) +{ + struct address_space *mapping = page->mapping; + struct inode *inode = mapping->host; + struct buffer_head *page_bufs; handle_t *handle = NULL; int ret = 0; int err; - if (ext4_journal_current_handle()) - goto out_fail; + ret = block_prepare_write(page, 0, PAGE_CACHE_SIZE, ext4_get_block); + if (ret != 0) + goto out_unlock; + + page_bufs = page_buffers(page); + walk_page_buffers(handle, page_bufs, 0, PAGE_CACHE_SIZE, NULL, + bget_one); + /* As soon as we unlock the page, it can go away, but we have + * references to buffers so we are safe */ + unlock_page(page); handle = ext4_journal_start(inode, ext4_writepage_trans_blocks(inode)); if (IS_ERR(handle)) { ret = PTR_ERR(handle); - goto out_fail; + goto out; } - if (test_opt(inode->i_sb, NOBH) && ext4_should_writeback_data(inode)) - ret = nobh_writepage(page, ext4_get_block, wbc); - else - ret = block_write_full_page(page, ext4_get_block, wbc); + ret = walk_page_buffers(handle, page_bufs, 0, + PAGE_CACHE_SIZE, NULL, do_journal_get_write_access); + err = walk_page_buffers(handle, page_bufs, 0, + PAGE_CACHE_SIZE, NULL, write_end_fn); + if (ret == 0) + ret = err; err = ext4_journal_stop(handle); if (!ret) ret = err; - return ret; -out_fail: - redirty_page_for_writepage(wbc, page); + walk_page_buffers(handle, page_bufs, 0, + PAGE_CACHE_SIZE, NULL, bput_one); + EXT4_I(inode)->i_state |= EXT4_STATE_JDATA; + goto out; + +out_unlock: unlock_page(page); +out: return ret; } @@ -1660,59 +1701,40 @@ static int ext4_journalled_writepage(struct page *page, struct writeback_control *wbc) { struct inode *inode = page->mapping->host; - handle_t *handle = NULL; - int ret = 0; - int err; + loff_t size = i_size_read(inode); + loff_t len; - if (ext4_journal_current_handle()) - goto no_write; + J_ASSERT(PageLocked(page)); + J_ASSERT(page_has_buffers(page)); + if (page->index == size >> PAGE_CACHE_SHIFT) + len = size & ~PAGE_CACHE_MASK; + else + len = PAGE_CACHE_SIZE; + BUG_ON(walk_page_buffers(NULL, page_buffers(page), 0, len, NULL, + ext4_bh_unmapped_or_delay)); - handle = ext4_journal_start(inode, ext4_writepage_trans_blocks(inode)); - if (IS_ERR(handle)) { - ret = PTR_ERR(handle); + if (ext4_journal_current_handle()) goto no_write; - } - if (!page_has_buffers(page) || PageChecked(page)) { + if (PageChecked(page)) { /* * It's mmapped pagecache. Add buffers and journal it. There * doesn't seem much point in redirtying the page here. */ ClearPageChecked(page); - ret = block_prepare_write(page, 0, PAGE_CACHE_SIZE, - ext4_get_block); - if (ret != 0) { - ext4_journal_stop(handle); - goto out_unlock; - } - ret = walk_page_buffers(handle, page_buffers(page), 0, - PAGE_CACHE_SIZE, NULL, do_journal_get_write_access); - - err = walk_page_buffers(handle, page_buffers(page), 0, - PAGE_CACHE_SIZE, NULL, write_end_fn); - if (ret == 0) - ret = err; - EXT4_I(inode)->i_state |= EXT4_STATE_JDATA; - unlock_page(page); + return __ext4_journalled_writepage(page, wbc); } else { /* * It may be a page full of checkpoint-mode buffers. We don't * really know unless we go poke around in the buffer_heads. * But block_write_full_page will do the right thing. */ - ret = block_write_full_page(page, ext4_get_block, wbc); + return block_write_full_page(page, ext4_get_block, wbc); } - err = ext4_journal_stop(handle); - if (!ret) - ret = err; -out: - return ret; - no_write: redirty_page_for_writepage(wbc, page); -out_unlock: unlock_page(page); - goto out; + return 0; } static int ext4_readpage(struct file *file, struct page *page) @@ -1909,7 +1931,7 @@ void ext4_set_aops(struct inode *inode) * This required during truncate. We need to physically zero the tail end * of that block so it doesn't yield old data if the file is later grown. */ -int ext4_block_truncate_page(handle_t *handle, struct page *page, +int ext4_block_truncate_page(handle_t *handle, struct address_space *mapping, loff_t from) { ext4_fsblk_t index = from >> PAGE_CACHE_SHIFT; @@ -1918,8 +1940,13 @@ int ext4_block_truncate_page(handle_t *handle, struct page *page, ext4_lblk_t iblock; struct inode *inode = mapping->host; struct buffer_head *bh; + struct page *page; int err = 0; + page = grab_cache_page(mapping, from >> PAGE_CACHE_SHIFT); + if (!page) + return -EINVAL; + blocksize = inode->i_sb->s_blocksize; length = blocksize - (offset & (blocksize - 1)); iblock = index << (PAGE_CACHE_SHIFT - inode->i_sb->s_blocksize_bits); @@ -2410,46 +2437,25 @@ void ext4_truncate(struct inode *inode) int n; ext4_lblk_t last_block; unsigned blocksize = inode->i_sb->s_blocksize; - struct page *page; if (!ext4_can_truncate(inode)) return; - /* - * We have to lock the EOF page here, because lock_page() nests - * outside jbd2_journal_start(). - */ - if ((inode->i_size & (blocksize - 1)) == 0) { - /* Block boundary? Nothing to do */ - page = NULL; - } else { - page = grab_cache_page(mapping, - inode->i_size >> PAGE_CACHE_SHIFT); - if (!page) - return; - } - if (EXT4_I(inode)->i_flags & EXT4_EXTENTS_FL) { - ext4_ext_truncate(inode, page); + ext4_ext_truncate(inode); return; } handle = start_transaction(inode); - if (IS_ERR(handle)) { - if (page) { - clear_highpage(page); - flush_dcache_page(page); - unlock_page(page); - page_cache_release(page); - } + if (IS_ERR(handle)) return; /* AKPM: return what? */ - } last_block = (inode->i_size + blocksize-1) >> EXT4_BLOCK_SIZE_BITS(inode->i_sb); - if (page) - ext4_block_truncate_page(handle, page, mapping, inode->i_size); + if (inode->i_size & (blocksize - 1)) + if (ext4_block_truncate_page(handle, mapping, inode->i_size)) + goto out_stop; n = ext4_block_to_path(inode, last_block, offsets, NULL); if (n == 0) -- cgit v1.2.2 From 678aaf481496b01473b778685eca231d6784098b Mon Sep 17 00:00:00 2001 From: Jan Kara Date: Fri, 11 Jul 2008 19:27:31 -0400 Subject: ext4: Use new framework for data=ordered mode in JBD2 This patch makes ext4 use inode-based implementation of data=ordered mode in JBD2. It allows us to unify some data=ordered and data=writeback paths (especially writepage since we don't have to start a transaction anymore) and remove some buffer walking. Updated fix from Aneesh Kumar K.V to fix file system hang due to corrupt jinode values. Signed-off-by: Jan Kara Signed-off-by: Aneesh Kumar K.V Signed-off-by: Mingming Cao Signed-off-by: "Theodore Ts'o" --- fs/ext4/inode.c | 158 ++++++++++++++++++-------------------------------------- 1 file changed, 49 insertions(+), 109 deletions(-) (limited to 'fs/ext4/inode.c') diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 320acb6c35bf..7b9569179fdf 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -39,6 +39,13 @@ #include "xattr.h" #include "acl.h" +static inline int ext4_begin_ordered_truncate(struct inode *inode, + loff_t new_size) +{ + return jbd2_journal_begin_ordered_truncate(&EXT4_I(inode)->jinode, + new_size); +} + /* * Test whether an inode is a fast symlink. */ @@ -181,6 +188,8 @@ void ext4_delete_inode (struct inode * inode) { handle_t *handle; + if (ext4_should_order_data(inode)) + ext4_begin_ordered_truncate(inode, 0); truncate_inode_pages(&inode->i_data, 0); if (is_bad_inode(inode)) @@ -1273,15 +1282,6 @@ out: return ret; } -int ext4_journal_dirty_data(handle_t *handle, struct buffer_head *bh) -{ - int err = jbd2_journal_dirty_data(handle, bh); - if (err) - ext4_journal_abort_handle(__func__, __func__, - bh, handle, err); - return err; -} - /* For write_end() in data=journal mode */ static int write_end_fn(handle_t *handle, struct buffer_head *bh) { @@ -1311,8 +1311,7 @@ static int ext4_ordered_write_end(struct file *file, from = pos & (PAGE_CACHE_SIZE - 1); to = from + len; - ret = walk_page_buffers(handle, page_buffers(page), - from, to, NULL, ext4_journal_dirty_data); + ret = ext4_jbd2_file_inode(handle, inode); if (ret == 0) { /* @@ -1472,25 +1471,22 @@ static int bput_one(handle_t *handle, struct buffer_head *bh) return 0; } -static int jbd2_journal_dirty_data_fn(handle_t *handle, struct buffer_head *bh) -{ - if (buffer_mapped(bh)) - return ext4_journal_dirty_data(handle, bh); - return 0; -} - static int ext4_bh_unmapped_or_delay(handle_t *handle, struct buffer_head *bh) { return !buffer_mapped(bh) || buffer_delay(bh); } /* - * Note that we don't need to start a transaction unless we're journaling - * data because we should have holes filled from ext4_page_mkwrite(). If - * we are journaling data, we cannot start transaction directly because - * transaction start ranks above page lock so we have to do some magic... + * Note that we don't need to start a transaction unless we're journaling data + * because we should have holes filled from ext4_page_mkwrite(). We even don't + * need to file the inode to the transaction's list in ordered mode because if + * we are writing back data added by write(), the inode is already there and if + * we are writing back data modified via mmap(), noone guarantees in which + * transaction the data will hit the disk. In case we are journaling data, we + * cannot start transaction directly because transaction start ranks above page + * lock so we have to do some magic. * - * In all journalling modes block_write_full_page() will start the I/O. + * In all journaling modes block_write_full_page() will start the I/O. * * Problem: * @@ -1533,86 +1529,7 @@ static int ext4_bh_unmapped_or_delay(handle_t *handle, struct buffer_head *bh) * us. * */ -static int __ext4_ordered_writepage(struct page *page, - struct writeback_control *wbc) -{ - struct inode *inode = page->mapping->host; - struct buffer_head *page_bufs; - handle_t *handle = NULL; - int ret = 0; - int err; - - if (!page_has_buffers(page)) { - create_empty_buffers(page, inode->i_sb->s_blocksize, - (1 << BH_Dirty)|(1 << BH_Uptodate)); - } - page_bufs = page_buffers(page); - walk_page_buffers(handle, page_bufs, 0, - PAGE_CACHE_SIZE, NULL, bget_one); - - ret = block_write_full_page(page, ext4_get_block, wbc); - - /* - * The page can become unlocked at any point now, and - * truncate can then come in and change things. So we - * can't touch *page from now on. But *page_bufs is - * safe due to elevated refcount. - */ - - /* - * And attach them to the current transaction. But only if - * block_write_full_page() succeeded. Otherwise they are unmapped, - * and generally junk. - */ - if (ret == 0) { - handle = ext4_journal_start(inode, - ext4_writepage_trans_blocks(inode)); - if (IS_ERR(handle)) { - ret = PTR_ERR(handle); - goto out_put; - } - - ret = walk_page_buffers(handle, page_bufs, 0, PAGE_CACHE_SIZE, - NULL, jbd2_journal_dirty_data_fn); - err = ext4_journal_stop(handle); - if (!ret) - ret = err; - } -out_put: - walk_page_buffers(handle, page_bufs, 0, PAGE_CACHE_SIZE, NULL, - bput_one); - return ret; -} - -static int ext4_ordered_writepage(struct page *page, - struct writeback_control *wbc) -{ - struct inode *inode = page->mapping->host; - loff_t size = i_size_read(inode); - loff_t len; - - J_ASSERT(PageLocked(page)); - J_ASSERT(page_has_buffers(page)); - if (page->index == size >> PAGE_CACHE_SHIFT) - len = size & ~PAGE_CACHE_MASK; - else - len = PAGE_CACHE_SIZE; - BUG_ON(walk_page_buffers(NULL, page_buffers(page), 0, len, NULL, - ext4_bh_unmapped_or_delay)); - - /* - * We give up here if we're reentered, because it might be for a - * different filesystem. - */ - if (!ext4_journal_current_handle()) - return __ext4_ordered_writepage(page, wbc); - - redirty_page_for_writepage(wbc, page); - unlock_page(page); - return 0; -} - -static int __ext4_writeback_writepage(struct page *page, +static int __ext4_normal_writepage(struct page *page, struct writeback_control *wbc) { struct inode *inode = page->mapping->host; @@ -1624,7 +1541,7 @@ static int __ext4_writeback_writepage(struct page *page, } -static int ext4_writeback_writepage(struct page *page, +static int ext4_normal_writepage(struct page *page, struct writeback_control *wbc) { struct inode *inode = page->mapping->host; @@ -1641,7 +1558,7 @@ static int ext4_writeback_writepage(struct page *page, ext4_bh_unmapped_or_delay)); if (!ext4_journal_current_handle()) - return __ext4_writeback_writepage(page, wbc); + return __ext4_normal_writepage(page, wbc); redirty_page_for_writepage(wbc, page); unlock_page(page); @@ -1877,7 +1794,7 @@ static int ext4_journalled_set_page_dirty(struct page *page) static const struct address_space_operations ext4_ordered_aops = { .readpage = ext4_readpage, .readpages = ext4_readpages, - .writepage = ext4_ordered_writepage, + .writepage = ext4_normal_writepage, .sync_page = block_sync_page, .write_begin = ext4_write_begin, .write_end = ext4_ordered_write_end, @@ -1891,7 +1808,7 @@ static const struct address_space_operations ext4_ordered_aops = { static const struct address_space_operations ext4_writeback_aops = { .readpage = ext4_readpage, .readpages = ext4_readpages, - .writepage = ext4_writeback_writepage, + .writepage = ext4_normal_writepage, .sync_page = block_sync_page, .write_begin = ext4_write_begin, .write_end = ext4_writeback_write_end, @@ -2019,7 +1936,7 @@ int ext4_block_truncate_page(handle_t *handle, err = ext4_journal_dirty_metadata(handle, bh); } else { if (ext4_should_order_data(inode)) - err = ext4_journal_dirty_data(handle, bh); + err = ext4_jbd2_file_inode(handle, inode); mark_buffer_dirty(bh); } @@ -3171,7 +3088,14 @@ int ext4_write_inode(struct inode *inode, int wait) * be freed, so we have a strong guarantee that no future commit will * leave these blocks visible to the user.) * - * Called with inode->sem down. + * Another thing we have to assure is that if we are in ordered mode + * and inode is still attached to the committing transaction, we must + * we start writeout of all the dirty pages which are being truncated. + * This way we are sure that all the data written in the previous + * transaction are already on disk (truncate waits for pages under + * writeback). + * + * Called with inode->i_mutex down. */ int ext4_setattr(struct dentry *dentry, struct iattr *attr) { @@ -3237,6 +3161,22 @@ int ext4_setattr(struct dentry *dentry, struct iattr *attr) if (!error) error = rc; ext4_journal_stop(handle); + + if (ext4_should_order_data(inode)) { + error = ext4_begin_ordered_truncate(inode, + attr->ia_size); + if (error) { + /* Do as much error cleanup as possible */ + handle = ext4_journal_start(inode, 3); + if (IS_ERR(handle)) { + ext4_orphan_del(NULL, inode); + goto err_out; + } + ext4_orphan_del(handle, inode); + ext4_journal_stop(handle); + goto err_out; + } + } } rc = inode_setattr(inode, attr); -- cgit v1.2.2 From 64769240bd07f446f83660bb143bb609d8ab4910 Mon Sep 17 00:00:00 2001 From: Alex Tomas Date: Fri, 11 Jul 2008 19:27:31 -0400 Subject: ext4: Add delayed allocation support in data=writeback mode Updated with fixes from Mingming Cao to unlock and release the page from page cache if the delalloc write_begin failed, and properly handle preallocated blocks. Also added a fix to clear buffer_delay in block_write_full_page() after allocating a delayed buffer. Updated with fixes from Aneesh Kumar K.V to update i_disksize properly and to add bmap support for delayed allocation. Updated with a fix from Valerie Clement to avoid filesystem corruption when the filesystem is mounted with the delalloc option and blocksize < pagesize. Signed-off-by: Alex Tomas Signed-off-by: Mingming Cao Signed-off-by: Dave Kleikamp Signed-off-by: "Theodore Ts'o" Signed-off-by: Aneesh Kumar K.V --- fs/ext4/inode.c | 699 +++++++++++++++++++++++++++++++++++++++++++++++++++++++- 1 file changed, 694 insertions(+), 5 deletions(-) (limited to 'fs/ext4/inode.c') diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 7b9569179fdf..2bef4f879e4b 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -32,6 +32,7 @@ #include #include #include +#include #include #include #include @@ -46,6 +47,8 @@ static inline int ext4_begin_ordered_truncate(struct inode *inode, new_size); } +static void ext4_invalidatepage(struct page *page, unsigned long offset); + /* * Test whether an inode is a fast symlink. */ @@ -1407,6 +1410,669 @@ static int ext4_journalled_write_end(struct file *file, return ret ? ret : copied; } +/* + * Delayed allocation stuff + */ + +struct mpage_da_data { + struct inode *inode; + struct buffer_head lbh; /* extent of blocks */ + unsigned long first_page, next_page; /* extent of pages */ + get_block_t *get_block; + struct writeback_control *wbc; +}; + +/* + * mpage_da_submit_io - walks through extent of pages and try to write + * them with __mpage_writepage() + * + * @mpd->inode: inode + * @mpd->first_page: first page of the extent + * @mpd->next_page: page after the last page of the extent + * @mpd->get_block: the filesystem's block mapper function + * + * By the time mpage_da_submit_io() is called we expect all blocks + * to be allocated. this may be wrong if allocation failed. + * + * As pages are already locked by write_cache_pages(), we can't use it + */ +static int mpage_da_submit_io(struct mpage_da_data *mpd) +{ + struct address_space *mapping = mpd->inode->i_mapping; + struct mpage_data mpd_pp = { + .bio = NULL, + .last_block_in_bio = 0, + .get_block = mpd->get_block, + .use_writepage = 1, + }; + int ret = 0, err, nr_pages, i; + unsigned long index, end; + struct pagevec pvec; + + BUG_ON(mpd->next_page <= mpd->first_page); + + pagevec_init(&pvec, 0); + index = mpd->first_page; + end = mpd->next_page - 1; + + while (index <= end) { + /* XXX: optimize tail */ + nr_pages = pagevec_lookup(&pvec, mapping, index, PAGEVEC_SIZE); + if (nr_pages == 0) + break; + for (i = 0; i < nr_pages; i++) { + struct page *page = pvec.pages[i]; + + index = page->index; + if (index > end) + break; + index++; + + err = __mpage_writepage(page, mpd->wbc, &mpd_pp); + + /* + * In error case, we have to continue because + * remaining pages are still locked + * XXX: unlock and re-dirty them? + */ + if (ret == 0) + ret = err; + } + pagevec_release(&pvec); + } + if (mpd_pp.bio) + mpage_bio_submit(WRITE, mpd_pp.bio); + + return ret; +} + +/* + * mpage_put_bnr_to_bhs - walk blocks and assign them actual numbers + * + * @mpd->inode - inode to walk through + * @exbh->b_blocknr - first block on a disk + * @exbh->b_size - amount of space in bytes + * @logical - first logical block to start assignment with + * + * the function goes through all passed space and put actual disk + * block numbers into buffer heads, dropping BH_Delay + */ +static void mpage_put_bnr_to_bhs(struct mpage_da_data *mpd, sector_t logical, + struct buffer_head *exbh) +{ + struct inode *inode = mpd->inode; + struct address_space *mapping = inode->i_mapping; + int blocks = exbh->b_size >> inode->i_blkbits; + sector_t pblock = exbh->b_blocknr, cur_logical; + struct buffer_head *head, *bh; + unsigned long index, end; + struct pagevec pvec; + int nr_pages, i; + + index = logical >> (PAGE_CACHE_SHIFT - inode->i_blkbits); + end = (logical + blocks - 1) >> (PAGE_CACHE_SHIFT - inode->i_blkbits); + cur_logical = index << (PAGE_CACHE_SHIFT - inode->i_blkbits); + + pagevec_init(&pvec, 0); + + while (index <= end) { + /* XXX: optimize tail */ + nr_pages = pagevec_lookup(&pvec, mapping, index, PAGEVEC_SIZE); + if (nr_pages == 0) + break; + for (i = 0; i < nr_pages; i++) { + struct page *page = pvec.pages[i]; + + index = page->index; + if (index > end) + break; + index++; + + BUG_ON(!PageLocked(page)); + BUG_ON(PageWriteback(page)); + BUG_ON(!page_has_buffers(page)); + + bh = page_buffers(page); + head = bh; + + /* skip blocks out of the range */ + do { + if (cur_logical >= logical) + break; + cur_logical++; + } while ((bh = bh->b_this_page) != head); + + do { + if (cur_logical >= logical + blocks) + break; + + if (buffer_delay(bh)) { + bh->b_blocknr = pblock; + clear_buffer_delay(bh); + } else if (buffer_mapped(bh)) { + BUG_ON(bh->b_blocknr != pblock); + } + + cur_logical++; + pblock++; + } while ((bh = bh->b_this_page) != head); + } + pagevec_release(&pvec); + } +} + + +/* + * __unmap_underlying_blocks - just a helper function to unmap + * set of blocks described by @bh + */ +static inline void __unmap_underlying_blocks(struct inode *inode, + struct buffer_head *bh) +{ + struct block_device *bdev = inode->i_sb->s_bdev; + int blocks, i; + + blocks = bh->b_size >> inode->i_blkbits; + for (i = 0; i < blocks; i++) + unmap_underlying_metadata(bdev, bh->b_blocknr + i); +} + +/* + * mpage_da_map_blocks - go through given space + * + * @mpd->lbh - bh describing space + * @mpd->get_block - the filesystem's block mapper function + * + * The function skips space we know is already mapped to disk blocks. + * + * The function ignores errors ->get_block() returns, thus real + * error handling is postponed to __mpage_writepage() + */ +static void mpage_da_map_blocks(struct mpage_da_data *mpd) +{ + struct buffer_head *lbh = &mpd->lbh; + int err = 0, remain = lbh->b_size; + sector_t next = lbh->b_blocknr; + struct buffer_head new; + + /* + * We consider only non-mapped and non-allocated blocks + */ + if (buffer_mapped(lbh) && !buffer_delay(lbh)) + return; + + while (remain) { + new.b_state = lbh->b_state; + new.b_blocknr = 0; + new.b_size = remain; + err = mpd->get_block(mpd->inode, next, &new, 1); + if (err) { + /* + * Rather than implement own error handling + * here, we just leave remaining blocks + * unallocated and try again with ->writepage() + */ + break; + } + BUG_ON(new.b_size == 0); + + if (buffer_new(&new)) + __unmap_underlying_blocks(mpd->inode, &new); + + /* + * If blocks are delayed marked, we need to + * put actual blocknr and drop delayed bit + */ + if (buffer_delay(lbh)) + mpage_put_bnr_to_bhs(mpd, next, &new); + + /* go for the remaining blocks */ + next += new.b_size >> mpd->inode->i_blkbits; + remain -= new.b_size; + } +} + +#define BH_FLAGS ((1 << BH_Uptodate) | (1 << BH_Mapped) | (1 << BH_Delay)) + +/* + * mpage_add_bh_to_extent - try to add one more block to extent of blocks + * + * @mpd->lbh - extent of blocks + * @logical - logical number of the block in the file + * @bh - bh of the block (used to access block's state) + * + * the function is used to collect contig. blocks in same state + */ +static void mpage_add_bh_to_extent(struct mpage_da_data *mpd, + sector_t logical, struct buffer_head *bh) +{ + struct buffer_head *lbh = &mpd->lbh; + sector_t next; + + next = lbh->b_blocknr + (lbh->b_size >> mpd->inode->i_blkbits); + + /* + * First block in the extent + */ + if (lbh->b_size == 0) { + lbh->b_blocknr = logical; + lbh->b_size = bh->b_size; + lbh->b_state = bh->b_state & BH_FLAGS; + return; + } + + /* + * Can we merge the block to our big extent? + */ + if (logical == next && (bh->b_state & BH_FLAGS) == lbh->b_state) { + lbh->b_size += bh->b_size; + return; + } + + /* + * We couldn't merge the block to our extent, so we + * need to flush current extent and start new one + */ + mpage_da_map_blocks(mpd); + + /* + * Now start a new extent + */ + lbh->b_size = bh->b_size; + lbh->b_state = bh->b_state & BH_FLAGS; + lbh->b_blocknr = logical; +} + +/* + * __mpage_da_writepage - finds extent of pages and blocks + * + * @page: page to consider + * @wbc: not used, we just follow rules + * @data: context + * + * The function finds extents of pages and scan them for all blocks. + */ +static int __mpage_da_writepage(struct page *page, + struct writeback_control *wbc, void *data) +{ + struct mpage_da_data *mpd = data; + struct inode *inode = mpd->inode; + struct buffer_head *bh, *head, fake; + sector_t logical; + + /* + * Can we merge this page to current extent? + */ + if (mpd->next_page != page->index) { + /* + * Nope, we can't. So, we map non-allocated blocks + * and start IO on them using __mpage_writepage() + */ + if (mpd->next_page != mpd->first_page) { + mpage_da_map_blocks(mpd); + mpage_da_submit_io(mpd); + } + + /* + * Start next extent of pages ... + */ + mpd->first_page = page->index; + + /* + * ... and blocks + */ + mpd->lbh.b_size = 0; + mpd->lbh.b_state = 0; + mpd->lbh.b_blocknr = 0; + } + + mpd->next_page = page->index + 1; + logical = (sector_t) page->index << + (PAGE_CACHE_SHIFT - inode->i_blkbits); + + if (!page_has_buffers(page)) { + /* + * There is no attached buffer heads yet (mmap?) + * we treat the page asfull of dirty blocks + */ + bh = &fake; + bh->b_size = PAGE_CACHE_SIZE; + bh->b_state = 0; + set_buffer_dirty(bh); + set_buffer_uptodate(bh); + mpage_add_bh_to_extent(mpd, logical, bh); + } else { + /* + * Page with regular buffer heads, just add all dirty ones + */ + head = page_buffers(page); + bh = head; + do { + BUG_ON(buffer_locked(bh)); + if (buffer_dirty(bh)) + mpage_add_bh_to_extent(mpd, logical, bh); + logical++; + } while ((bh = bh->b_this_page) != head); + } + + return 0; +} + +/* + * mpage_da_writepages - walk the list of dirty pages of the given + * address space, allocates non-allocated blocks, maps newly-allocated + * blocks to existing bhs and issue IO them + * + * @mapping: address space structure to write + * @wbc: subtract the number of written pages from *@wbc->nr_to_write + * @get_block: the filesystem's block mapper function. + * + * This is a library function, which implements the writepages() + * address_space_operation. + * + * In order to avoid duplication of logic that deals with partial pages, + * multiple bio per page, etc, we find non-allocated blocks, allocate + * them with minimal calls to ->get_block() and re-use __mpage_writepage() + * + * It's important that we call __mpage_writepage() only once for each + * involved page, otherwise we'd have to implement more complicated logic + * to deal with pages w/o PG_lock or w/ PG_writeback and so on. + * + * See comments to mpage_writepages() + */ +static int mpage_da_writepages(struct address_space *mapping, + struct writeback_control *wbc, + get_block_t get_block) +{ + struct mpage_da_data mpd; + int ret; + + if (!get_block) + return generic_writepages(mapping, wbc); + + mpd.wbc = wbc; + mpd.inode = mapping->host; + mpd.lbh.b_size = 0; + mpd.lbh.b_state = 0; + mpd.lbh.b_blocknr = 0; + mpd.first_page = 0; + mpd.next_page = 0; + mpd.get_block = get_block; + + ret = write_cache_pages(mapping, wbc, __mpage_da_writepage, &mpd); + + /* + * Handle last extent of pages + */ + if (mpd.next_page != mpd.first_page) { + mpage_da_map_blocks(&mpd); + mpage_da_submit_io(&mpd); + } + + return ret; +} + +/* + * this is a special callback for ->write_begin() only + * it's intention is to return mapped block or reserve space + */ +static int ext4_da_get_block_prep(struct inode *inode, sector_t iblock, + struct buffer_head *bh_result, int create) +{ + int ret = 0; + + BUG_ON(create == 0); + BUG_ON(bh_result->b_size != inode->i_sb->s_blocksize); + + /* + * first, we need to know whether the block is allocated already + * preallocated blocks are unmapped but should treated + * the same as allocated blocks. + */ + ret = ext4_get_blocks_wrap(NULL, inode, iblock, 1, bh_result, 0, 0); + if (ret == 0) { + /* the block isn't allocated yet, let's reserve space */ + /* XXX: call reservation here */ + /* + * XXX: __block_prepare_write() unmaps passed block, + * is it OK? + */ + map_bh(bh_result, inode->i_sb, 0); + set_buffer_new(bh_result); + set_buffer_delay(bh_result); + } else if (ret > 0) { + bh_result->b_size = (ret << inode->i_blkbits); + ret = 0; + } + + return ret; +} + +static int ext4_da_get_block_write(struct inode *inode, sector_t iblock, + struct buffer_head *bh_result, int create) +{ + int ret, needed_blocks = ext4_writepage_trans_blocks(inode); + unsigned max_blocks = bh_result->b_size >> inode->i_blkbits; + loff_t disksize = EXT4_I(inode)->i_disksize; + handle_t *handle = NULL; + + if (create) { + handle = ext4_journal_start(inode, needed_blocks); + if (IS_ERR(handle)) { + ret = PTR_ERR(handle); + goto out; + } + } + + ret = ext4_get_blocks_wrap(handle, inode, iblock, max_blocks, + bh_result, create, 0); + if (ret > 0) { + bh_result->b_size = (ret << inode->i_blkbits); + + /* + * Update on-disk size along with block allocation + * we don't use 'extend_disksize' as size may change + * within already allocated block -bzzz + */ + disksize = ((loff_t) iblock + ret) << inode->i_blkbits; + if (disksize > i_size_read(inode)) + disksize = i_size_read(inode); + if (disksize > EXT4_I(inode)->i_disksize) { + /* + * XXX: replace with spinlock if seen contended -bzzz + */ + down_write(&EXT4_I(inode)->i_data_sem); + if (disksize > EXT4_I(inode)->i_disksize) + EXT4_I(inode)->i_disksize = disksize; + up_write(&EXT4_I(inode)->i_data_sem); + + if (EXT4_I(inode)->i_disksize == disksize) { + if (handle == NULL) + handle = ext4_journal_start(inode, 1); + if (!IS_ERR(handle)) + ext4_mark_inode_dirty(handle, inode); + } + } + + ret = 0; + } + +out: + if (handle && !IS_ERR(handle)) + ext4_journal_stop(handle); + + return ret; +} +/* FIXME!! only support data=writeback mode */ +static int ext4_da_writepage(struct page *page, + struct writeback_control *wbc) +{ + struct inode *inode = page->mapping->host; + handle_t *handle = NULL; + int ret = 0; + int err; + + if (ext4_journal_current_handle()) + goto out_fail; + + handle = ext4_journal_start(inode, ext4_writepage_trans_blocks(inode)); + if (IS_ERR(handle)) { + ret = PTR_ERR(handle); + goto out_fail; + } + + if (test_opt(inode->i_sb, NOBH) && ext4_should_writeback_data(inode)) + ret = nobh_writepage(page, ext4_get_block, wbc); + else + ret = block_write_full_page(page, ext4_get_block, wbc); + + if (!ret && inode->i_size > EXT4_I(inode)->i_disksize) { + EXT4_I(inode)->i_disksize = inode->i_size; + ext4_mark_inode_dirty(handle, inode); + } + + err = ext4_journal_stop(handle); + if (!ret) + ret = err; + return ret; + +out_fail: + redirty_page_for_writepage(wbc, page); + unlock_page(page); + return ret; +} + +static int ext4_da_writepages(struct address_space *mapping, + struct writeback_control *wbc) +{ + return mpage_da_writepages(mapping, wbc, ext4_da_get_block_write); +} + +static int ext4_da_write_begin(struct file *file, struct address_space *mapping, + loff_t pos, unsigned len, unsigned flags, + struct page **pagep, void **fsdata) +{ + int ret; + struct page *page; + pgoff_t index; + unsigned from, to; + struct inode *inode = mapping->host; + handle_t *handle; + + index = pos >> PAGE_CACHE_SHIFT; + from = pos & (PAGE_CACHE_SIZE - 1); + to = from + len; + + /* + * With delayed allocation, we don't log the i_disksize update + * if there is delayed block allocation. But we still need + * to journalling the i_disksize update if writes to the end + * of file which has an already mapped buffer. + */ + handle = ext4_journal_start(inode, 1); + if (IS_ERR(handle)) { + ret = PTR_ERR(handle); + goto out; + } + + page = __grab_cache_page(mapping, index); + if (!page) + return -ENOMEM; + *pagep = page; + + ret = block_write_begin(file, mapping, pos, len, flags, pagep, fsdata, + ext4_da_get_block_prep); + if (ret < 0) { + unlock_page(page); + ext4_journal_stop(handle); + page_cache_release(page); + } + +out: + return ret; +} + +static int ext4_bh_unmapped_or_delay(handle_t *handle, struct buffer_head *bh) +{ + return !buffer_mapped(bh) || buffer_delay(bh); +} + +static int ext4_da_write_end(struct file *file, + struct address_space *mapping, + loff_t pos, unsigned len, unsigned copied, + struct page *page, void *fsdata) +{ + struct inode *inode = mapping->host; + int ret = 0, ret2; + handle_t *handle = ext4_journal_current_handle(); + loff_t new_i_size; + + /* + * generic_write_end() will run mark_inode_dirty() if i_size + * changes. So let's piggyback the i_disksize mark_inode_dirty + * into that. + */ + + new_i_size = pos + copied; + if (new_i_size > EXT4_I(inode)->i_disksize) + if (!walk_page_buffers(NULL, page_buffers(page), + 0, len, NULL, ext4_bh_unmapped_or_delay)){ + /* + * Updating i_disksize when extending file without + * needing block allocation + */ + if (ext4_should_order_data(inode)) + ret = ext4_jbd2_file_inode(handle, inode); + + EXT4_I(inode)->i_disksize = new_i_size; + } + ret2 = generic_write_end(file, mapping, pos, len, copied, + page, fsdata); + copied = ret2; + if (ret2 < 0) + ret = ret2; + ret2 = ext4_journal_stop(handle); + if (!ret) + ret = ret2; + + return ret ? ret : copied; +} + +static void ext4_da_invalidatepage(struct page *page, unsigned long offset) +{ + struct buffer_head *head, *bh; + unsigned int curr_off = 0; + + /* + * Drop reserved blocks + */ + BUG_ON(!PageLocked(page)); + if (!page_has_buffers(page)) + goto out; + + head = page_buffers(page); + bh = head; + do { + unsigned int next_off = curr_off + bh->b_size; + + /* + * is this block fully invalidated? + */ + if (offset <= curr_off && buffer_delay(bh)) { + clear_buffer_delay(bh); + /* XXX: add real stuff here */ + } + curr_off = next_off; + bh = bh->b_this_page; + } while (bh != head); + +out: + ext4_invalidatepage(page, offset); + + return; +} + + /* * 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. @@ -1427,6 +2093,16 @@ static sector_t ext4_bmap(struct address_space *mapping, sector_t block) journal_t *journal; int err; + if (mapping_tagged(mapping, PAGECACHE_TAG_DIRTY) && + test_opt(inode->i_sb, DELALLOC)) { + /* + * With delalloc we want to sync the file + * so that we can make sure we allocate + * blocks for file + */ + filemap_write_and_wait(mapping); + } + if (EXT4_I(inode)->i_state & EXT4_STATE_JDATA) { /* * This is a REALLY heavyweight approach, but the use of @@ -1471,11 +2147,6 @@ static int bput_one(handle_t *handle, struct buffer_head *bh) return 0; } -static int ext4_bh_unmapped_or_delay(handle_t *handle, struct buffer_head *bh) -{ - return !buffer_mapped(bh) || buffer_delay(bh); -} - /* * Note that we don't need to start a transaction unless we're journaling data * because we should have holes filled from ext4_page_mkwrite(). We even don't @@ -1832,10 +2503,28 @@ static const struct address_space_operations ext4_journalled_aops = { .releasepage = ext4_releasepage, }; +static const struct address_space_operations ext4_da_aops = { + .readpage = ext4_readpage, + .readpages = ext4_readpages, + .writepage = ext4_da_writepage, + .writepages = ext4_da_writepages, + .sync_page = block_sync_page, + .write_begin = ext4_da_write_begin, + .write_end = ext4_da_write_end, + .bmap = ext4_bmap, + .invalidatepage = ext4_da_invalidatepage, + .releasepage = ext4_releasepage, + .direct_IO = ext4_direct_IO, + .migratepage = buffer_migrate_page, +}; + void ext4_set_aops(struct inode *inode) { if (ext4_should_order_data(inode)) inode->i_mapping->a_ops = &ext4_ordered_aops; + else if (ext4_should_writeback_data(inode) && + test_opt(inode->i_sb, DELALLOC)) + inode->i_mapping->a_ops = &ext4_da_aops; else if (ext4_should_writeback_data(inode)) inode->i_mapping->a_ops = &ext4_writeback_aops; else -- cgit v1.2.2 From d2a1763791a634e315ec926b62829c1e88842c86 Mon Sep 17 00:00:00 2001 From: Mingming Cao Date: Mon, 14 Jul 2008 17:52:37 -0400 Subject: ext4: delayed allocation ENOSPC handling This patch does block reservation for delayed allocation, to avoid ENOSPC later at page flush time. Blocks(data and metadata) are reserved at da_write_begin() time, the freeblocks counter is updated by then, and the number of reserved blocks is store in per inode counter. At the writepage time, the unused reserved meta blocks are returned back. At unlink/truncate time, reserved blocks are properly released. Updated fix from Aneesh Kumar K.V to fix the oldallocator block reservation accounting with delalloc, added lock to guard the counters and also fix the reservation for meta blocks. Signed-off-by: Aneesh Kumar K.V Signed-off-by: Mingming Cao Signed-off-by: Theodore Ts'o --- fs/ext4/inode.c | 184 +++++++++++++++++++++++++++++++++++++++++++++++--------- 1 file changed, 156 insertions(+), 28 deletions(-) (limited to 'fs/ext4/inode.c') diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 2bef4f879e4b..a6b800c58474 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -39,6 +39,7 @@ #include "ext4_jbd2.h" #include "xattr.h" #include "acl.h" +#include "ext4_extents.h" static inline int ext4_begin_ordered_truncate(struct inode *inode, loff_t new_size) @@ -982,7 +983,7 @@ out: */ int ext4_get_blocks_wrap(handle_t *handle, struct inode *inode, sector_t block, unsigned long max_blocks, struct buffer_head *bh, - int create, int extend_disksize) + int create, int extend_disksize, int flag) { int retval; @@ -1023,6 +1024,15 @@ int ext4_get_blocks_wrap(handle_t *handle, struct inode *inode, sector_t block, * with create == 1 flag. */ down_write((&EXT4_I(inode)->i_data_sem)); + + /* + * if the caller is from delayed allocation writeout path + * we have already reserved fs blocks for allocation + * let the underlying get_block() function know to + * avoid double accounting + */ + if (flag) + EXT4_I(inode)->i_delalloc_reserved_flag = 1; /* * We need to check for EXT4 here because migrate * could have changed the inode type in between @@ -1044,6 +1054,18 @@ int ext4_get_blocks_wrap(handle_t *handle, struct inode *inode, sector_t block, ~EXT4_EXT_MIGRATE; } } + + if (flag) { + EXT4_I(inode)->i_delalloc_reserved_flag = 0; + /* + * Update reserved blocks/metadata blocks + * after successful block allocation + * which were deferred till now + */ + if ((retval > 0) && buffer_delay(bh)) + ext4_da_release_space(inode, retval, 0); + } + up_write((&EXT4_I(inode)->i_data_sem)); return retval; } @@ -1069,7 +1091,7 @@ static int ext4_get_block(struct inode *inode, sector_t iblock, } ret = ext4_get_blocks_wrap(handle, inode, iblock, - max_blocks, bh_result, create, 0); + max_blocks, bh_result, create, 0, 0); if (ret > 0) { bh_result->b_size = (ret << inode->i_blkbits); ret = 0; @@ -1095,7 +1117,7 @@ struct buffer_head *ext4_getblk(handle_t *handle, struct inode *inode, dummy.b_blocknr = -1000; buffer_trace_init(&dummy.b_history); err = ext4_get_blocks_wrap(handle, inode, block, 1, - &dummy, create, 1); + &dummy, create, 1, 0); /* * ext4_get_blocks_handle() returns number of blocks * mapped. 0 in case of a HOLE. @@ -1409,6 +1431,122 @@ static int ext4_journalled_write_end(struct file *file, return ret ? ret : copied; } +/* + * Calculate the number of metadata blocks need to reserve + * to allocate @blocks for non extent file based file + */ +static int ext4_indirect_calc_metadata_amount(struct inode *inode, int blocks) +{ + int icap = EXT4_ADDR_PER_BLOCK(inode->i_sb); + int ind_blks, dind_blks, tind_blks; + + /* number of new indirect blocks needed */ + ind_blks = (blocks + icap - 1) / icap; + + dind_blks = (ind_blks + icap - 1) / icap; + + tind_blks = 1; + + return ind_blks + dind_blks + tind_blks; +} + +/* + * Calculate the number of metadata blocks need to reserve + * to allocate given number of blocks + */ +static int ext4_calc_metadata_amount(struct inode *inode, int blocks) +{ + if (EXT4_I(inode)->i_flags & EXT4_EXTENTS_FL) + return ext4_ext_calc_metadata_amount(inode, blocks); + + return ext4_indirect_calc_metadata_amount(inode, blocks); +} + +static int ext4_da_reserve_space(struct inode *inode, int nrblocks) +{ + struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb); + unsigned long md_needed, mdblocks, total = 0; + + /* + * recalculate the amount of metadata blocks to reserve + * in order to allocate nrblocks + * worse case is one extent per block + */ + spin_lock(&EXT4_I(inode)->i_block_reservation_lock); + total = EXT4_I(inode)->i_reserved_data_blocks + nrblocks; + mdblocks = ext4_calc_metadata_amount(inode, total); + BUG_ON(mdblocks < EXT4_I(inode)->i_reserved_meta_blocks); + + md_needed = mdblocks - EXT4_I(inode)->i_reserved_meta_blocks; + total = md_needed + nrblocks; + + if (ext4_has_free_blocks(sbi, total) < total) { + spin_unlock(&EXT4_I(inode)->i_block_reservation_lock); + return -ENOSPC; + } + + /* reduce fs free blocks counter */ + percpu_counter_sub(&sbi->s_freeblocks_counter, total); + + EXT4_I(inode)->i_reserved_data_blocks += nrblocks; + EXT4_I(inode)->i_reserved_meta_blocks = mdblocks; + + spin_unlock(&EXT4_I(inode)->i_block_reservation_lock); + return 0; /* success */ +} + +void ext4_da_release_space(struct inode *inode, int used, int to_free) +{ + struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb); + int total, mdb, mdb_free, release; + + spin_lock(&EXT4_I(inode)->i_block_reservation_lock); + /* recalculate the number of metablocks still need to be reserved */ + total = EXT4_I(inode)->i_reserved_data_blocks - used - to_free; + mdb = ext4_calc_metadata_amount(inode, total); + + /* figure out how many metablocks to release */ + BUG_ON(mdb > EXT4_I(inode)->i_reserved_meta_blocks); + mdb_free = EXT4_I(inode)->i_reserved_meta_blocks - mdb; + + /* Account for allocated meta_blocks */ + mdb_free -= EXT4_I(inode)->i_allocated_meta_blocks; + + release = to_free + mdb_free; + + /* update fs free blocks counter for truncate case */ + percpu_counter_add(&sbi->s_freeblocks_counter, release); + + /* update per-inode reservations */ + BUG_ON(used + to_free > EXT4_I(inode)->i_reserved_data_blocks); + EXT4_I(inode)->i_reserved_data_blocks -= (used + to_free); + + BUG_ON(mdb > EXT4_I(inode)->i_reserved_meta_blocks); + EXT4_I(inode)->i_reserved_meta_blocks = mdb; + EXT4_I(inode)->i_allocated_meta_blocks = 0; + spin_unlock(&EXT4_I(inode)->i_block_reservation_lock); +} + +static void ext4_da_page_release_reservation(struct page *page, + unsigned long offset) +{ + int to_release = 0; + struct buffer_head *head, *bh; + unsigned int curr_off = 0; + + head = page_buffers(page); + bh = head; + do { + unsigned int next_off = curr_off + bh->b_size; + + if ((offset <= curr_off) && (buffer_delay(bh))) { + to_release++; + clear_buffer_delay(bh); + } + curr_off = next_off; + } while ((bh = bh->b_this_page) != head); + ext4_da_release_space(page->mapping->host, 0, to_release); +} /* * Delayed allocation stuff @@ -1829,14 +1967,18 @@ static int ext4_da_get_block_prep(struct inode *inode, sector_t iblock, * preallocated blocks are unmapped but should treated * the same as allocated blocks. */ - ret = ext4_get_blocks_wrap(NULL, inode, iblock, 1, bh_result, 0, 0); - if (ret == 0) { - /* the block isn't allocated yet, let's reserve space */ - /* XXX: call reservation here */ + ret = ext4_get_blocks_wrap(NULL, inode, iblock, 1, bh_result, 0, 0, 0); + if ((ret == 0) && !buffer_delay(bh_result)) { + /* the block isn't (pre)allocated yet, let's reserve space */ /* * XXX: __block_prepare_write() unmaps passed block, * is it OK? */ + ret = ext4_da_reserve_space(inode, 1); + if (ret) + /* not enough space to reserve */ + return ret; + map_bh(bh_result, inode->i_sb, 0); set_buffer_new(bh_result); set_buffer_delay(bh_result); @@ -1847,7 +1989,7 @@ static int ext4_da_get_block_prep(struct inode *inode, sector_t iblock, return ret; } - +#define EXT4_DELALLOC_RSVED 1 static int ext4_da_get_block_write(struct inode *inode, sector_t iblock, struct buffer_head *bh_result, int create) { @@ -1865,7 +2007,7 @@ static int ext4_da_get_block_write(struct inode *inode, sector_t iblock, } ret = ext4_get_blocks_wrap(handle, inode, iblock, max_blocks, - bh_result, create, 0); + bh_result, create, 0, EXT4_DELALLOC_RSVED); if (ret > 0) { bh_result->b_size = (ret << inode->i_blkbits); @@ -1952,7 +2094,7 @@ static int ext4_da_write_begin(struct file *file, struct address_space *mapping, loff_t pos, unsigned len, unsigned flags, struct page **pagep, void **fsdata) { - int ret; + int ret, retries = 0; struct page *page; pgoff_t index; unsigned from, to; @@ -1963,6 +2105,7 @@ static int ext4_da_write_begin(struct file *file, struct address_space *mapping, from = pos & (PAGE_CACHE_SIZE - 1); to = from + len; +retry: /* * With delayed allocation, we don't log the i_disksize update * if there is delayed block allocation. But we still need @@ -1988,6 +2131,8 @@ static int ext4_da_write_begin(struct file *file, struct address_space *mapping, page_cache_release(page); } + if (ret == -ENOSPC && ext4_should_retry_alloc(inode->i_sb, &retries)) + goto retry; out: return ret; } @@ -2040,9 +2185,6 @@ static int ext4_da_write_end(struct file *file, static void ext4_da_invalidatepage(struct page *page, unsigned long offset) { - struct buffer_head *head, *bh; - unsigned int curr_off = 0; - /* * Drop reserved blocks */ @@ -2050,21 +2192,7 @@ static void ext4_da_invalidatepage(struct page *page, unsigned long offset) if (!page_has_buffers(page)) goto out; - head = page_buffers(page); - bh = head; - do { - unsigned int next_off = curr_off + bh->b_size; - - /* - * is this block fully invalidated? - */ - if (offset <= curr_off && buffer_delay(bh)) { - clear_buffer_delay(bh); - /* XXX: add real stuff here */ - } - curr_off = next_off; - bh = bh->b_this_page; - } while (bh != head); + ext4_da_page_release_reservation(page, offset); out: ext4_invalidatepage(page, offset); -- cgit v1.2.2 From 61628a3f3a37af2bf25daf8e26fd6b76a78c4f76 Mon Sep 17 00:00:00 2001 From: Mingming Cao Date: Fri, 11 Jul 2008 19:27:31 -0400 Subject: ext4: Invert lock ordering of page_lock and transaction start in delalloc With the reverse locking, we need to start a transation before taking the page lock, so in ext4_da_writepages() we need to break the write-out into chunks, and restart the journal for each chunck to ensure the write-out fits in a single transaction. Updated patch from Aneesh Kumar K.V which fixes delalloc sync hang with journal lock inversion, and address the performance regression issue. Signed-off-by: Mingming Cao Signed-off-by: Aneesh Kumar K.V Signed-off-by: Jan Kara Signed-off-by: "Theodore Ts'o" --- fs/ext4/inode.c | 201 ++++++++++++++++++++++++++++++++++++++++---------------- 1 file changed, 144 insertions(+), 57 deletions(-) (limited to 'fs/ext4/inode.c') diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index a6b800c58474..7923336ecf94 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -847,6 +847,7 @@ int ext4_get_blocks_handle(handle_t *handle, struct inode *inode, struct ext4_inode_info *ei = EXT4_I(inode); int count = 0; ext4_fsblk_t first_block = 0; + loff_t disksize; J_ASSERT(!(EXT4_I(inode)->i_flags & EXT4_EXTENTS_FL)); @@ -922,8 +923,13 @@ int ext4_get_blocks_handle(handle_t *handle, struct inode *inode, * protect it if you're about to implement concurrent * ext4_get_block() -bzzz */ - if (!err && extend_disksize && inode->i_size > ei->i_disksize) - ei->i_disksize = inode->i_size; + if (!err && extend_disksize) { + disksize = ((loff_t) iblock + count) << inode->i_blkbits; + if (disksize > i_size_read(inode)) + disksize = i_size_read(inode); + if (disksize > ei->i_disksize) + ei->i_disksize = disksize; + } if (err) goto cleanup; @@ -1683,13 +1689,11 @@ static void mpage_put_bnr_to_bhs(struct mpage_da_data *mpd, sector_t logical, do { if (cur_logical >= logical + blocks) break; - if (buffer_delay(bh)) { bh->b_blocknr = pblock; clear_buffer_delay(bh); - } else if (buffer_mapped(bh)) { + } else if (buffer_mapped(bh)) BUG_ON(bh->b_blocknr != pblock); - } cur_logical++; pblock++; @@ -1764,10 +1768,10 @@ static void mpage_da_map_blocks(struct mpage_da_data *mpd) if (buffer_delay(lbh)) mpage_put_bnr_to_bhs(mpd, next, &new); - /* go for the remaining blocks */ - next += new.b_size >> mpd->inode->i_blkbits; - remain -= new.b_size; - } + /* go for the remaining blocks */ + next += new.b_size >> mpd->inode->i_blkbits; + remain -= new.b_size; + } } #define BH_FLAGS ((1 << BH_Uptodate) | (1 << BH_Mapped) | (1 << BH_Delay)) @@ -1993,18 +1997,14 @@ static int ext4_da_get_block_prep(struct inode *inode, sector_t iblock, static int ext4_da_get_block_write(struct inode *inode, sector_t iblock, struct buffer_head *bh_result, int create) { - int ret, needed_blocks = ext4_writepage_trans_blocks(inode); + int ret; unsigned max_blocks = bh_result->b_size >> inode->i_blkbits; loff_t disksize = EXT4_I(inode)->i_disksize; handle_t *handle = NULL; - if (create) { - handle = ext4_journal_start(inode, needed_blocks); - if (IS_ERR(handle)) { - ret = PTR_ERR(handle); - goto out; - } - } + handle = ext4_journal_current_handle(); + BUG_ON(handle == NULL); + BUG_ON(create == 0); ret = ext4_get_blocks_wrap(handle, inode, iblock, max_blocks, bh_result, create, 0, EXT4_DELALLOC_RSVED); @@ -2029,65 +2029,157 @@ static int ext4_da_get_block_write(struct inode *inode, sector_t iblock, up_write(&EXT4_I(inode)->i_data_sem); if (EXT4_I(inode)->i_disksize == disksize) { - if (handle == NULL) - handle = ext4_journal_start(inode, 1); - if (!IS_ERR(handle)) - ext4_mark_inode_dirty(handle, inode); + ret = ext4_mark_inode_dirty(handle, inode); + return ret; } } - ret = 0; } - -out: - if (handle && !IS_ERR(handle)) - ext4_journal_stop(handle); - return ret; } + +static int ext4_bh_unmapped_or_delay(handle_t *handle, struct buffer_head *bh) +{ + return !buffer_mapped(bh) || buffer_delay(bh); +} + /* FIXME!! only support data=writeback mode */ +/* + * get called vi ext4_da_writepages after taking page lock + * We may end up doing block allocation here in case + * mpage_da_map_blocks failed to allocate blocks. + */ static int ext4_da_writepage(struct page *page, struct writeback_control *wbc) { - struct inode *inode = page->mapping->host; - handle_t *handle = NULL; int ret = 0; - int err; + loff_t size; + unsigned long len; + handle_t *handle = NULL; + struct buffer_head *page_bufs; + struct inode *inode = page->mapping->host; - if (ext4_journal_current_handle()) - goto out_fail; + handle = ext4_journal_current_handle(); + if (!handle) { + /* + * This can happen when we aren't called via + * ext4_da_writepages() but directly (shrink_page_list). + * We cannot easily start a transaction here so we just skip + * writing the page in case we would have to do so. + */ + size = i_size_read(inode); - handle = ext4_journal_start(inode, ext4_writepage_trans_blocks(inode)); - if (IS_ERR(handle)) { - ret = PTR_ERR(handle); - goto out_fail; + page_bufs = page_buffers(page); + if (page->index == size >> PAGE_CACHE_SHIFT) + len = size & ~PAGE_CACHE_MASK; + else + len = PAGE_CACHE_SIZE; + + if (walk_page_buffers(NULL, page_bufs, 0, + len, NULL, ext4_bh_unmapped_or_delay)) { + /* + * We can't do block allocation under + * page lock without a handle . So redirty + * the page and return + */ + BUG_ON(wbc->sync_mode != WB_SYNC_NONE); + redirty_page_for_writepage(wbc, page); + unlock_page(page); + return 0; + } } if (test_opt(inode->i_sb, NOBH) && ext4_should_writeback_data(inode)) - ret = nobh_writepage(page, ext4_get_block, wbc); + ret = nobh_writepage(page, ext4_da_get_block_write, wbc); else - ret = block_write_full_page(page, ext4_get_block, wbc); - - if (!ret && inode->i_size > EXT4_I(inode)->i_disksize) { - EXT4_I(inode)->i_disksize = inode->i_size; - ext4_mark_inode_dirty(handle, inode); - } + ret = block_write_full_page(page, ext4_da_get_block_write, wbc); - err = ext4_journal_stop(handle); - if (!ret) - ret = err; - return ret; - -out_fail: - redirty_page_for_writepage(wbc, page); - unlock_page(page); return ret; } + +/* + * For now just follow the DIO way to estimate the max credits + * needed to write out EXT4_MAX_WRITEBACK_PAGES. + * todo: need to calculate the max credits need for + * extent based files, currently the DIO credits is based on + * indirect-blocks mapping way. + * + * Probably should have a generic way to calculate credits + * for DIO, writepages, and truncate + */ +#define EXT4_MAX_WRITEBACK_PAGES DIO_MAX_BLOCKS +#define EXT4_MAX_WRITEBACK_CREDITS DIO_CREDITS + static int ext4_da_writepages(struct address_space *mapping, struct writeback_control *wbc) { - return mpage_da_writepages(mapping, wbc, ext4_da_get_block_write); + struct inode *inode = mapping->host; + handle_t *handle = NULL; + int needed_blocks; + int ret = 0; + long to_write; + loff_t range_start = 0; + + /* + * No pages to write? This is mainly a kludge to avoid starting + * a transaction for special inodes like journal inode on last iput() + * because that could violate lock ordering on umount + */ + if (!mapping->nrpages) + return 0; + + /* + * Estimate the worse case needed credits to write out + * EXT4_MAX_BUF_BLOCKS pages + */ + needed_blocks = EXT4_MAX_WRITEBACK_CREDITS; + + to_write = wbc->nr_to_write; + if (!wbc->range_cyclic) { + /* + * If range_cyclic is not set force range_cont + * and save the old writeback_index + */ + wbc->range_cont = 1; + range_start = wbc->range_start; + } + + while (!ret && to_write) { + /* start a new transaction*/ + handle = ext4_journal_start(inode, needed_blocks); + if (IS_ERR(handle)) { + ret = PTR_ERR(handle); + goto out_writepages; + } + /* + * set the max dirty pages could be write at a time + * to fit into the reserved transaction credits + */ + if (wbc->nr_to_write > EXT4_MAX_WRITEBACK_PAGES) + wbc->nr_to_write = EXT4_MAX_WRITEBACK_PAGES; + + to_write -= wbc->nr_to_write; + ret = mpage_da_writepages(mapping, wbc, + ext4_da_get_block_write); + ext4_journal_stop(handle); + if (wbc->nr_to_write) { + /* + * There is no more writeout needed + * or we requested for a noblocking writeout + * and we found the device congested + */ + to_write += wbc->nr_to_write; + break; + } + wbc->nr_to_write = to_write; + } + +out_writepages: + wbc->nr_to_write = to_write; + if (range_start) + wbc->range_start = range_start; + return ret; } static int ext4_da_write_begin(struct file *file, struct address_space *mapping, @@ -2137,11 +2229,6 @@ out: return ret; } -static int ext4_bh_unmapped_or_delay(handle_t *handle, struct buffer_head *bh) -{ - return !buffer_mapped(bh) || buffer_delay(bh); -} - static int ext4_da_write_end(struct file *file, struct address_space *mapping, loff_t pos, unsigned len, unsigned copied, -- cgit v1.2.2 From cd1aac32923a9c8adcc0ae85e33c1ca0c5855838 Mon Sep 17 00:00:00 2001 From: "Aneesh Kumar K.V" Date: Fri, 11 Jul 2008 19:27:31 -0400 Subject: ext4: Add ordered mode support for delalloc This provides a new ordered mode implementation which gets rid of using buffer heads to enforce the ordering between metadata change with the related data chage. Instead, in the new ordering mode, it keeps track of all of the inodes touched by each transaction on a list, and when that transaction is committed, it flushes all of the dirty pages for those inodes. In addition, the new ordered mode reverses the lock ordering of the page lock and transaction lock, which provides easier support for delayed allocation. Signed-off-by: Aneesh Kumar K.V Signed-off-by: Mingming Cao Signed-off-by: "Theodore Ts'o" --- fs/ext4/inode.c | 30 +++++++++++++++++++++++++----- 1 file changed, 25 insertions(+), 5 deletions(-) (limited to 'fs/ext4/inode.c') diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 7923336ecf94..24518b57733e 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -2043,11 +2043,12 @@ static int ext4_bh_unmapped_or_delay(handle_t *handle, struct buffer_head *bh) return !buffer_mapped(bh) || buffer_delay(bh); } -/* FIXME!! only support data=writeback mode */ /* * get called vi ext4_da_writepages after taking page lock * We may end up doing block allocation here in case * mpage_da_map_blocks failed to allocate blocks. + * + * We also get called via journal_submit_inode_data_buffers */ static int ext4_da_writepage(struct page *page, struct writeback_control *wbc) @@ -2066,6 +2067,7 @@ static int ext4_da_writepage(struct page *page, * ext4_da_writepages() but directly (shrink_page_list). * We cannot easily start a transaction here so we just skip * writing the page in case we would have to do so. + * We reach here also via journal_submit_inode_data_buffers */ size = i_size_read(inode); @@ -2081,8 +2083,11 @@ static int ext4_da_writepage(struct page *page, * We can't do block allocation under * page lock without a handle . So redirty * the page and return + * We may reach here when we do a journal commit + * via journal_submit_inode_data_buffers. + * If we don't have mapping block we just ignore + * them */ - BUG_ON(wbc->sync_mode != WB_SYNC_NONE); redirty_page_for_writepage(wbc, page); unlock_page(page); return 0; @@ -2097,7 +2102,6 @@ static int ext4_da_writepage(struct page *page, return ret; } - /* * For now just follow the DIO way to estimate the max credits * needed to write out EXT4_MAX_WRITEBACK_PAGES. @@ -2130,7 +2134,7 @@ static int ext4_da_writepages(struct address_space *mapping, return 0; /* - * Estimate the worse case needed credits to write out + * Estimate the worse case needed credits to write out * EXT4_MAX_BUF_BLOCKS pages */ needed_blocks = EXT4_MAX_WRITEBACK_CREDITS; @@ -2152,6 +2156,19 @@ static int ext4_da_writepages(struct address_space *mapping, ret = PTR_ERR(handle); goto out_writepages; } + if (ext4_should_order_data(inode)) { + /* + * With ordered mode we need to add + * the inode to the journal handle + * when we do block allocation. + */ + ret = ext4_jbd2_file_inode(handle, inode); + if (ret) { + ext4_journal_stop(handle); + goto out_writepages; + } + + } /* * set the max dirty pages could be write at a time * to fit into the reserved transaction credits @@ -2735,7 +2752,10 @@ static const struct address_space_operations ext4_da_aops = { void ext4_set_aops(struct inode *inode) { - if (ext4_should_order_data(inode)) + if (ext4_should_order_data(inode) && + test_opt(inode->i_sb, DELALLOC)) + inode->i_mapping->a_ops = &ext4_da_aops; + else if (ext4_should_order_data(inode)) inode->i_mapping->a_ops = &ext4_ordered_aops; else if (ext4_should_writeback_data(inode) && test_opt(inode->i_sb, DELALLOC)) -- cgit v1.2.2 From f0e6c98593eb8a77edb7dd0edb22bb9f9368c567 Mon Sep 17 00:00:00 2001 From: "Aneesh Kumar K.V" Date: Fri, 11 Jul 2008 19:27:31 -0400 Subject: ext4: Handle page without buffers in ext4_*_writepage() It can happen that buffers are removed from the page before it gets marked dirty and then is passed to writepage(). In writepage() we just initialize the buffers and check whether they are mapped and non delay. If they are mapped and non delay we write the page. Otherwise we mark them dirty. With this change we don't do block allocation at all in ext4_*_write_page. writepage() can get called under many condition and with a locking order of journal_start -> lock_page, we should not try to allocate blocks in writepage() which get called after taking page lock. writepage() can get called via shrink_page_list even with a journal handle which was created for doing inode update. For example when doing ext4_da_write_begin we create a journal handle with credit 1 expecting a i_disksize update for the inode. But ext4_da_write_begin can cause shrink_page_list via _grab_page_cache. So having a valid handle via ext4_journal_current_handle is not a guarantee that we can use the handle for block allocation in writepage, since we shouldn't be using credits that had been reserved for other updates. That it could result in we running out of credits when we update inodes. Signed-off-by: Aneesh Kumar K.V Signed-off-by: Mingming Cao Signed-off-by: "Theodore Ts'o" --- fs/ext4/inode.c | 169 +++++++++++++++++++++++++++++++++++++++++--------------- 1 file changed, 124 insertions(+), 45 deletions(-) (limited to 'fs/ext4/inode.c') diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 24518b57733e..ce47847bb37d 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -2003,11 +2003,15 @@ static int ext4_da_get_block_write(struct inode *inode, sector_t iblock, handle_t *handle = NULL; handle = ext4_journal_current_handle(); - BUG_ON(handle == NULL); - BUG_ON(create == 0); - - ret = ext4_get_blocks_wrap(handle, inode, iblock, max_blocks, + if (!handle) { + ret = ext4_get_blocks_wrap(handle, inode, iblock, max_blocks, + bh_result, 0, 0, 0); + BUG_ON(!ret); + } else { + ret = ext4_get_blocks_wrap(handle, inode, iblock, max_blocks, bh_result, create, 0, EXT4_DELALLOC_RSVED); + } + if (ret > 0) { bh_result->b_size = (ret << inode->i_blkbits); @@ -2040,15 +2044,37 @@ static int ext4_da_get_block_write(struct inode *inode, sector_t iblock, static int ext4_bh_unmapped_or_delay(handle_t *handle, struct buffer_head *bh) { - return !buffer_mapped(bh) || buffer_delay(bh); + /* + * unmapped buffer is possible for holes. + * delay buffer is possible with delayed allocation + */ + return ((!buffer_mapped(bh) || buffer_delay(bh)) && buffer_dirty(bh)); +} + +static int ext4_normal_get_block_write(struct inode *inode, sector_t iblock, + struct buffer_head *bh_result, int create) +{ + int ret = 0; + unsigned max_blocks = bh_result->b_size >> inode->i_blkbits; + + /* + * we don't want to do block allocation in writepage + * so call get_block_wrap with create = 0 + */ + ret = ext4_get_blocks_wrap(NULL, inode, iblock, max_blocks, + bh_result, 0, 0, 0); + if (ret > 0) { + bh_result->b_size = (ret << inode->i_blkbits); + ret = 0; + } + return ret; } /* - * get called vi ext4_da_writepages after taking page lock - * We may end up doing block allocation here in case - * mpage_da_map_blocks failed to allocate blocks. - * - * We also get called via journal_submit_inode_data_buffers + * get called vi ext4_da_writepages after taking page lock (have journal handle) + * get called via journal_submit_inode_data_buffers (no journal handle) + * get called via shrink_page_list via pdflush (no journal handle) + * or grab_page_cache when doing write_begin (have journal handle) */ static int ext4_da_writepage(struct page *page, struct writeback_control *wbc) @@ -2056,37 +2082,61 @@ static int ext4_da_writepage(struct page *page, int ret = 0; loff_t size; unsigned long len; - handle_t *handle = NULL; struct buffer_head *page_bufs; struct inode *inode = page->mapping->host; - handle = ext4_journal_current_handle(); - if (!handle) { - /* - * This can happen when we aren't called via - * ext4_da_writepages() but directly (shrink_page_list). - * We cannot easily start a transaction here so we just skip - * writing the page in case we would have to do so. - * We reach here also via journal_submit_inode_data_buffers - */ - size = i_size_read(inode); + size = i_size_read(inode); + if (page->index == size >> PAGE_CACHE_SHIFT) + len = size & ~PAGE_CACHE_MASK; + else + len = PAGE_CACHE_SIZE; + if (page_has_buffers(page)) { page_bufs = page_buffers(page); - if (page->index == size >> PAGE_CACHE_SHIFT) - len = size & ~PAGE_CACHE_MASK; - else - len = PAGE_CACHE_SIZE; - - if (walk_page_buffers(NULL, page_bufs, 0, - len, NULL, ext4_bh_unmapped_or_delay)) { + if (walk_page_buffers(NULL, page_bufs, 0, len, NULL, + ext4_bh_unmapped_or_delay)) { /* - * We can't do block allocation under - * page lock without a handle . So redirty - * the page and return + * We don't want to do block allocation + * So redirty the page and return * We may reach here when we do a journal commit * via journal_submit_inode_data_buffers. * If we don't have mapping block we just ignore - * them + * them. We can also reach here via shrink_page_list + */ + redirty_page_for_writepage(wbc, page); + unlock_page(page); + return 0; + } + } else { + /* + * The test for page_has_buffers() is subtle: + * We know the page is dirty but it lost buffers. That means + * that at some moment in time after write_begin()/write_end() + * has been called all buffers have been clean and thus they + * must have been written at least once. So they are all + * mapped and we can happily proceed with mapping them + * and writing the page. + * + * Try to initialize the buffer_heads and check whether + * all are mapped and non delay. We don't want to + * do block allocation here. + */ + ret = block_prepare_write(page, 0, PAGE_CACHE_SIZE, + ext4_normal_get_block_write); + if (!ret) { + page_bufs = page_buffers(page); + /* check whether all are mapped and non delay */ + if (walk_page_buffers(NULL, page_bufs, 0, len, NULL, + ext4_bh_unmapped_or_delay)) { + redirty_page_for_writepage(wbc, page); + unlock_page(page); + return 0; + } + } else { + /* + * We can't do block allocation here + * so just redity the page and unlock + * and return */ redirty_page_for_writepage(wbc, page); unlock_page(page); @@ -2095,9 +2145,11 @@ static int ext4_da_writepage(struct page *page, } if (test_opt(inode->i_sb, NOBH) && ext4_should_writeback_data(inode)) - ret = nobh_writepage(page, ext4_da_get_block_write, wbc); + ret = nobh_writepage(page, ext4_normal_get_block_write, wbc); else - ret = block_write_full_page(page, ext4_da_get_block_write, wbc); + ret = block_write_full_page(page, + ext4_normal_get_block_write, + wbc); return ret; } @@ -2438,12 +2490,14 @@ static int __ext4_normal_writepage(struct page *page, struct inode *inode = page->mapping->host; if (test_opt(inode->i_sb, NOBH)) - return nobh_writepage(page, ext4_get_block, wbc); + return nobh_writepage(page, + ext4_normal_get_block_write, wbc); else - return block_write_full_page(page, ext4_get_block, wbc); + return block_write_full_page(page, + ext4_normal_get_block_write, + wbc); } - static int ext4_normal_writepage(struct page *page, struct writeback_control *wbc) { @@ -2452,13 +2506,24 @@ static int ext4_normal_writepage(struct page *page, loff_t len; J_ASSERT(PageLocked(page)); - J_ASSERT(page_has_buffers(page)); if (page->index == size >> PAGE_CACHE_SHIFT) len = size & ~PAGE_CACHE_MASK; else len = PAGE_CACHE_SIZE; - BUG_ON(walk_page_buffers(NULL, page_buffers(page), 0, len, NULL, - ext4_bh_unmapped_or_delay)); + + if (page_has_buffers(page)) { + /* if page has buffers it should all be mapped + * and allocated. If there are not buffers attached + * to the page we know the page is dirty but it lost + * buffers. That means that at some moment in time + * after write_begin() / write_end() has been called + * all buffers have been clean and thus they must have been + * written at least once. So they are all mapped and we can + * happily proceed with mapping them and writing the page. + */ + BUG_ON(walk_page_buffers(NULL, page_buffers(page), 0, len, NULL, + ext4_bh_unmapped_or_delay)); + } if (!ext4_journal_current_handle()) return __ext4_normal_writepage(page, wbc); @@ -2478,7 +2543,8 @@ static int __ext4_journalled_writepage(struct page *page, int ret = 0; int err; - ret = block_prepare_write(page, 0, PAGE_CACHE_SIZE, ext4_get_block); + ret = block_prepare_write(page, 0, PAGE_CACHE_SIZE, + ext4_normal_get_block_write); if (ret != 0) goto out_unlock; @@ -2525,13 +2591,24 @@ static int ext4_journalled_writepage(struct page *page, loff_t len; J_ASSERT(PageLocked(page)); - J_ASSERT(page_has_buffers(page)); if (page->index == size >> PAGE_CACHE_SHIFT) len = size & ~PAGE_CACHE_MASK; else len = PAGE_CACHE_SIZE; - BUG_ON(walk_page_buffers(NULL, page_buffers(page), 0, len, NULL, - ext4_bh_unmapped_or_delay)); + + if (page_has_buffers(page)) { + /* if page has buffers it should all be mapped + * and allocated. If there are not buffers attached + * to the page we know the page is dirty but it lost + * buffers. That means that at some moment in time + * after write_begin() / write_end() has been called + * all buffers have been clean and thus they must have been + * written at least once. So they are all mapped and we can + * happily proceed with mapping them and writing the page. + */ + BUG_ON(walk_page_buffers(NULL, page_buffers(page), 0, len, NULL, + ext4_bh_unmapped_or_delay)); + } if (ext4_journal_current_handle()) goto no_write; @@ -2549,7 +2626,9 @@ static int ext4_journalled_writepage(struct page *page, * really know unless we go poke around in the buffer_heads. * But block_write_full_page will do the right thing. */ - return block_write_full_page(page, ext4_get_block, wbc); + return block_write_full_page(page, + ext4_normal_get_block_write, + wbc); } no_write: redirty_page_for_writepage(wbc, page); -- cgit v1.2.2 From 632eaeab1feb5d78c1e2bfb1d2dfc0ebb8ac187f Mon Sep 17 00:00:00 2001 From: Mingming Cao Date: Fri, 11 Jul 2008 19:27:31 -0400 Subject: ext4: fix delalloc i_disksize early update issue Ext4_da_write_end() used walk_page_buffers() with a callback function of ext4_bh_unmapped_or_delay() to check if it extended the file size without allocating any blocks (since in this case i_disksize needs to be updated). However, this is didn't work proprely because the buffer head has not been marked dirty yet --- this is done later in block_commit_write() --- which caused ext4_bh_unmapped_or_delay() to always return false. In addition, walk_page_buffers() checks all of the buffer heads covering the page, and the only buffer_head that should be checked is the one covering the end of the write. Otherwise, given a 1k blocksize filesystem and a 4k page size, the buffer head covering the first 1k stripe of the file could be unmapped (because it was a sparse file), and the second or third buffer_head covering that page could be mapped, and using walk_page_buffers() would fail in this case since it would stop at the first unmapped buffer_head and return true. The core problem is that walk_page_buffers() was intended to do work in a callback function, and a non-zero return value indicated a failure, which termined the walk of the buffer heads covering the page. It was not intended to be used with a boolean function, such as ext4_bh_unmapped_or_delay(). Add addtional fix from Aneesh to protect i_disksize update rave with truncate. Signed-off-by: Mingming Cao Signed-off-by: Aneesh Kumar K.V Signed-off-by: "Theodore Ts'o" --- fs/ext4/inode.c | 63 ++++++++++++++++++++++++++++++++++++++++++--------------- 1 file changed, 47 insertions(+), 16 deletions(-) (limited to 'fs/ext4/inode.c') diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index ce47847bb37d..0fbe678d40bb 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -2298,6 +2298,29 @@ out: return ret; } +/* + * Check if we should update i_disksize + * when write to the end of file but not require block allocation + */ +static int ext4_da_should_update_i_disksize(struct page *page, + unsigned long offset) +{ + struct buffer_head *bh; + struct inode *inode = page->mapping->host; + unsigned int idx; + int i; + + bh = page_buffers(page); + idx = offset >> inode->i_blkbits; + + for (i=0; i < idx; i++) + bh = bh->b_this_page; + + if (!buffer_mapped(bh) || (buffer_delay(bh))) + return 0; + return 1; +} + static int ext4_da_write_end(struct file *file, struct address_space *mapping, loff_t pos, unsigned len, unsigned copied, @@ -2307,6 +2330,10 @@ static int ext4_da_write_end(struct file *file, int ret = 0, ret2; handle_t *handle = ext4_journal_current_handle(); loff_t new_i_size; + unsigned long start, end; + + start = pos & (PAGE_CACHE_SIZE - 1); + end = start + copied -1; /* * generic_write_end() will run mark_inode_dirty() if i_size @@ -2315,18 +2342,23 @@ static int ext4_da_write_end(struct file *file, */ new_i_size = pos + copied; - if (new_i_size > EXT4_I(inode)->i_disksize) - if (!walk_page_buffers(NULL, page_buffers(page), - 0, len, NULL, ext4_bh_unmapped_or_delay)){ - /* - * Updating i_disksize when extending file without - * needing block allocation - */ - if (ext4_should_order_data(inode)) - ret = ext4_jbd2_file_inode(handle, inode); + if (new_i_size > EXT4_I(inode)->i_disksize) { + if (ext4_da_should_update_i_disksize(page, end)) { + down_write(&EXT4_I(inode)->i_data_sem); + if (new_i_size > EXT4_I(inode)->i_disksize) { + /* + * Updating i_disksize when extending file + * without needing block allocation + */ + if (ext4_should_order_data(inode)) + ret = ext4_jbd2_file_inode(handle, + inode); - EXT4_I(inode)->i_disksize = new_i_size; + EXT4_I(inode)->i_disksize = new_i_size; + } + up_write(&EXT4_I(inode)->i_data_sem); } + } ret2 = generic_write_end(file, mapping, pos, len, copied, page, fsdata); copied = ret2; @@ -3393,6 +3425,11 @@ void ext4_truncate(struct inode *inode) if (ext4_orphan_add(handle, inode)) goto out_stop; + /* + * From here we block out all ext4_get_block() callers who want to + * modify the block allocation tree. + */ + down_write(&ei->i_data_sem); /* * The orphan list entry will now protect us from any crash which * occurs before the truncate completes, so it is now safe to propagate @@ -3402,12 +3439,6 @@ void ext4_truncate(struct inode *inode) */ ei->i_disksize = inode->i_size; - /* - * From here we block out all ext4_get_block() callers who want to - * modify the block allocation tree. - */ - down_write(&ei->i_data_sem); - if (n == 1) { /* direct blocks */ ext4_free_data(handle, inode, NULL, i_data+offsets[0], i_data + EXT4_NDIR_BLOCKS); -- cgit v1.2.2 From 3e3398a08d6e516675d5af853d625dc7dd90eab1 Mon Sep 17 00:00:00 2001 From: Mingming Cao Date: Fri, 11 Jul 2008 19:27:31 -0400 Subject: ext4: delayed allocation i_blocks fix for stat Right now i_blocks is not getting updated until the blocks are actually allocaed on disk. This means with delayed allocation, right after files are copied, "ls -sF" shoes the file as taking 0 blocks on disk. "du" also shows the files taking zero space, which is highly confusing to the user. Since delayed allocation already keeps track of per-inode total number of blocks that are subject to delayed allocation, this patch fix this by using that to adjust the value returned by stat(2). When real block allocation is done, the i_blocks will get updated. Since the reserved blocks for delayed allocation will be decreased, this will be keep value returned by stat(2) consistent. Signed-off-by: Mingming Cao Signed-off-by: "Theodore Ts'o" --- fs/ext4/inode.c | 26 ++++++++++++++++++++++++++ 1 file changed, 26 insertions(+) (limited to 'fs/ext4/inode.c') diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 0fbe678d40bb..8ca2763df091 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -4231,6 +4231,32 @@ err_out: return error; } +int ext4_getattr(struct vfsmount *mnt, struct dentry *dentry, + struct kstat *stat) +{ + struct inode *inode; + unsigned long delalloc_blocks; + + inode = dentry->d_inode; + generic_fillattr(inode, stat); + + /* + * We can't update i_blocks if the block allocation is delayed + * otherwise in the case of system crash before the real block + * allocation is done, we will have i_blocks inconsistent with + * on-disk file blocks. + * We always keep i_blocks updated together with real + * allocation. But to not confuse with user, stat + * will return the blocks that include the delayed allocation + * blocks for this file. + */ + spin_lock(&EXT4_I(inode)->i_block_reservation_lock); + delalloc_blocks = EXT4_I(inode)->i_reserved_data_blocks; + spin_unlock(&EXT4_I(inode)->i_block_reservation_lock); + + stat->blocks += (delalloc_blocks << inode->i_sb->s_blocksize_bits)>>9; + return 0; +} /* * How many blocks doth make a writepage()? -- cgit v1.2.2