Merge tag 'ext4_for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4

Pull ext4 updates from Ted Ts'o:
 "New features for 3.12:

   - Added aggressive extent caching using the extent status tree.  This
     can actually decrease memory usage in read-mostly workloads since
     the information is much more compactly stored in the extent status
     tree than if we had to keep the extent tree metadata blocks in the
     buffer cache.  This also improves Asynchronous I/O since it is it
     makes much less likely that we need to do metadata I/O to lookup
     the extent tree information.

   - Improve the recovery after corrupted allocation bitmaps are found
     when running in errors=ignore mode.

  Also fixed some writeback vs truncate races when using a blocksize
  less than the page size"

* tag 'ext4_for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4: (25 commits)
  ext4: allow specifying external journal by pathname mount option
  ext4: mark group corrupt on group descriptor checksum
  ext4: mark block group as corrupt on inode bitmap error
  ext4: mark block group as corrupt on block bitmap error
  ext4: fix type declaration of ext4_validate_block_bitmap
  ext4: error out if verifying the block bitmap fails
  jbd2: Fix endian mixing problems in the checksumming code
  ext4: isolate ext4_extents.h file
  ext4: Fix misspellings using 'codespell' tool
  ext4: convert write_begin methods to stable_page_writes semantics
  ext4: fix use of potentially uninitialized variables in debugging code
  ext4: fix lost truncate due to race with writeback
  ext4: simplify truncation code in ext4_setattr()
  ext4: fix ext4_writepages() in presence of truncate
  ext4: move test whether extent to map can be extended to one place
  ext4: fix warning in ext4_da_update_reserve_space()
  quota: provide interface for readding allocated space into reserved space
  ext4: avoid reusing recently deleted inodes in no journal mode
  ext4: allocate delayed allocation blocks before rename
  ext4: start handle at least possible moment when renaming files
  ...

1 files changed, 162 insertions, 131 deletions

diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c
index c2ca04e67a4..9115f280751 100644
--- a/fs/ext4/inode.c
+++ b/fs/ext4/inode.c
@@ -553,7 +553,7 @@ int ext4_map_blocks(handle_t *handle, struct inode *inode,
 	}
 	if (retval > 0) {
 		int ret;
-		unsigned long long status;
+		unsigned int status;
 
 		if (unlikely(retval != map->m_len)) {
 			ext4_warning(inode->i_sb,
@@ -653,7 +653,7 @@ found:
 
 	if (retval > 0) {
 		int ret;
-		unsigned long long status;
+		unsigned int status;
 
 		if (unlikely(retval != map->m_len)) {
 			ext4_warning(inode->i_sb,
@@ -969,7 +969,8 @@ retry_journal:
 		ext4_journal_stop(handle);
 		goto retry_grab;
 	}
-	wait_on_page_writeback(page);
+	/* In case writeback began while the page was unlocked */
+	wait_for_stable_page(page);
 
 	if (ext4_should_dioread_nolock(inode))
 		ret = __block_write_begin(page, pos, len, ext4_get_block_write);
@@ -1633,7 +1634,7 @@ add_delayed:
 		set_buffer_delay(bh);
 	} else if (retval > 0) {
 		int ret;
-		unsigned long long status;
+		unsigned int status;
 
 		if (unlikely(retval != map->m_len)) {
 			ext4_warning(inode->i_sb,
@@ -1890,12 +1891,32 @@ static int ext4_writepage(struct page *page,
 	return ret;
 }
 
+static int mpage_submit_page(struct mpage_da_data *mpd, struct page *page)
+{
+	int len;
+	loff_t size = i_size_read(mpd->inode);
+	int err;
+
+	BUG_ON(page->index != mpd->first_page);
+	if (page->index == size >> PAGE_CACHE_SHIFT)
+		len = size & ~PAGE_CACHE_MASK;
+	else
+		len = PAGE_CACHE_SIZE;
+	clear_page_dirty_for_io(page);
+	err = ext4_bio_write_page(&mpd->io_submit, page, len, mpd->wbc);
+	if (!err)
+		mpd->wbc->nr_to_write--;
+	mpd->first_page++;
+
+	return err;
+}
+
 #define BH_FLAGS ((1 << BH_Unwritten) | (1 << BH_Delay))
 
 /*
  * mballoc gives us at most this number of blocks...
  * XXX: That seems to be only a limitation of ext4_mb_normalize_request().
- * The rest of mballoc seems to handle chunks upto full group size.
+ * The rest of mballoc seems to handle chunks up to full group size.
  */
 #define MAX_WRITEPAGES_EXTENT_LEN 2048
 
@@ -1904,82 +1925,94 @@ static int ext4_writepage(struct page *page,
  *
  * @mpd - extent of blocks
  * @lblk - logical number of the block in the file
- * @b_state - b_state of the buffer head added
+ * @bh - buffer head we want to add to the extent
  *
- * the function is used to collect contig. blocks in same state
+ * The function is used to collect contig. blocks in the same state. If the
+ * buffer doesn't require mapping for writeback and we haven't started the
+ * extent of buffers to map yet, the function returns 'true' immediately - the
+ * caller can write the buffer right away. Otherwise the function returns true
+ * if the block has been added to the extent, false if the block couldn't be
+ * added.
  */
-static int mpage_add_bh_to_extent(struct mpage_da_data *mpd, ext4_lblk_t lblk,
-				  unsigned long b_state)
+static bool mpage_add_bh_to_extent(struct mpage_da_data *mpd, ext4_lblk_t lblk,
+				   struct buffer_head *bh)
 {
 	struct ext4_map_blocks *map = &mpd->map;
 
-	/* Don't go larger than mballoc is willing to allocate */
-	if (map->m_len >= MAX_WRITEPAGES_EXTENT_LEN)
-		return 0;
+	/* Buffer that doesn't need mapping for writeback? */
+	if (!buffer_dirty(bh) || !buffer_mapped(bh) ||
+	    (!buffer_delay(bh) && !buffer_unwritten(bh))) {
+		/* So far no extent to map => we write the buffer right away */
+		if (map->m_len == 0)
+			return true;
+		return false;
+	}
 
 	/* First block in the extent? */
 	if (map->m_len == 0) {
 		map->m_lblk = lblk;
 		map->m_len = 1;
-		map->m_flags = b_state & BH_FLAGS;
-		return 1;
+		map->m_flags = bh->b_state & BH_FLAGS;
+		return true;
 	}
 
+	/* Don't go larger than mballoc is willing to allocate */
+	if (map->m_len >= MAX_WRITEPAGES_EXTENT_LEN)
+		return false;
+
 	/* Can we merge the block to our big extent? */
 	if (lblk == map->m_lblk + map->m_len &&
-	    (b_state & BH_FLAGS) == map->m_flags) {
+	    (bh->b_state & BH_FLAGS) == map->m_flags) {
 		map->m_len++;
-		return 1;
+		return true;
 	}
-	return 0;
+	return false;
 }
 
-static bool add_page_bufs_to_extent(struct mpage_da_data *mpd,
-				    struct buffer_head *head,
-				    struct buffer_head *bh,
-				    ext4_lblk_t lblk)
+/*
+ * mpage_process_page_bufs - submit page buffers for IO or add them to extent
+ *
+ * @mpd - extent of blocks for mapping
+ * @head - the first buffer in the page
+ * @bh - buffer we should start processing from
+ * @lblk - logical number of the block in the file corresponding to @bh
+ *
+ * Walk through page buffers from @bh upto @head (exclusive) and either submit
+ * the page for IO if all buffers in this page were mapped and there's no
+ * accumulated extent of buffers to map or add buffers in the page to the
+ * extent of buffers to map. The function returns 1 if the caller can continue
+ * by processing the next page, 0 if it should stop adding buffers to the
+ * extent to map because we cannot extend it anymore. It can also return value
+ * < 0 in case of error during IO submission.
+ */
+static int mpage_process_page_bufs(struct mpage_da_data *mpd,
+				   struct buffer_head *head,
+				   struct buffer_head *bh,
+				   ext4_lblk_t lblk)
 {
 	struct inode *inode = mpd->inode;
+	int err;
 	ext4_lblk_t blocks = (i_size_read(inode) + (1 << inode->i_blkbits) - 1)
 							>> inode->i_blkbits;
 
 	do {
 		BUG_ON(buffer_locked(bh));
 
-		if (!buffer_dirty(bh) || !buffer_mapped(bh) ||
-		    (!buffer_delay(bh) && !buffer_unwritten(bh)) ||
-		    lblk >= blocks) {
+		if (lblk >= blocks || !mpage_add_bh_to_extent(mpd, lblk, bh)) {
 			/* Found extent to map? */
 			if (mpd->map.m_len)
-				return false;
-			if (lblk >= blocks)
-				return true;
-			continue;
+				return 0;
+			/* Everything mapped so far and we hit EOF */
+			break;
 		}
-		if (!mpage_add_bh_to_extent(mpd, lblk, bh->b_state))
-			return false;
 	} while (lblk++, (bh = bh->b_this_page) != head);
-	return true;
-}
-
-static int mpage_submit_page(struct mpage_da_data *mpd, struct page *page)
-{
-	int len;
-	loff_t size = i_size_read(mpd->inode);
-	int err;
-
-	BUG_ON(page->index != mpd->first_page);
-	if (page->index == size >> PAGE_CACHE_SHIFT)
-		len = size & ~PAGE_CACHE_MASK;
-	else
-		len = PAGE_CACHE_SIZE;
-	clear_page_dirty_for_io(page);
-	err = ext4_bio_write_page(&mpd->io_submit, page, len, mpd->wbc);
-	if (!err)
-		mpd->wbc->nr_to_write--;
-	mpd->first_page++;
-
-	return err;
+	/* So far everything mapped? Submit the page for IO. */
+	if (mpd->map.m_len == 0) {
+		err = mpage_submit_page(mpd, head->b_page);
+		if (err < 0)
+			return err;
+	}
+	return lblk < blocks;
 }
 
 /*
@@ -2003,8 +2036,6 @@ static int mpage_map_and_submit_buffers(struct mpage_da_data *mpd)
 	struct inode *inode = mpd->inode;
 	struct buffer_head *head, *bh;
 	int bpp_bits = PAGE_CACHE_SHIFT - inode->i_blkbits;
-	ext4_lblk_t blocks = (i_size_read(inode) + (1 << inode->i_blkbits) - 1)
-							>> inode->i_blkbits;
 	pgoff_t start, end;
 	ext4_lblk_t lblk;
 	sector_t pblock;
@@ -2026,7 +2057,7 @@ static int mpage_map_and_submit_buffers(struct mpage_da_data *mpd)
 
 			if (page->index > end)
 				break;
-			/* Upto 'end' pages must be contiguous */
+			/* Up to 'end' pages must be contiguous */
 			BUG_ON(page->index != start);
 			bh = head = page_buffers(page);
 			do {
@@ -2039,18 +2070,26 @@ static int mpage_map_and_submit_buffers(struct mpage_da_data *mpd)
 					 */
 					mpd->map.m_len = 0;
 					mpd->map.m_flags = 0;
-					add_page_bufs_to_extent(mpd, head, bh,
-								lblk);
+					/*
+					 * FIXME: If dioread_nolock supports
+					 * blocksize < pagesize, we need to make
+					 * sure we add size mapped so far to
+					 * io_end->size as the following call
+					 * can submit the page for IO.
+					 */
+					err = mpage_process_page_bufs(mpd, head,
+								      bh, lblk);
 					pagevec_release(&pvec);
-					return 0;
+					if (err > 0)
+						err = 0;
+					return err;
 				}
 				if (buffer_delay(bh)) {
 					clear_buffer_delay(bh);
 					bh->b_blocknr = pblock++;
 				}
 				clear_buffer_unwritten(bh);
-			} while (++lblk < blocks &&
-				 (bh = bh->b_this_page) != head);
+			} while (lblk++, (bh = bh->b_this_page) != head);
 
 			/*
 			 * FIXME: This is going to break if dioread_nolock
@@ -2199,12 +2238,10 @@ static int mpage_map_and_submit_extent(handle_t *handle,
 
 	/* Update on-disk size after IO is submitted */
 	disksize = ((loff_t)mpd->first_page) << PAGE_CACHE_SHIFT;
-	if (disksize > i_size_read(inode))
-		disksize = i_size_read(inode);
 	if (disksize > EXT4_I(inode)->i_disksize) {
 		int err2;
 
-		ext4_update_i_disksize(inode, disksize);
+		ext4_wb_update_i_disksize(inode, disksize);
 		err2 = ext4_mark_inode_dirty(handle, inode);
 		if (err2)
 			ext4_error(inode->i_sb,
@@ -2219,7 +2256,7 @@ static int mpage_map_and_submit_extent(handle_t *handle,
 /*
  * Calculate the total number of credits to reserve for one writepages
  * iteration. This is called from ext4_writepages(). We map an extent of
- * upto MAX_WRITEPAGES_EXTENT_LEN blocks and then we go on and finish mapping
+ * up to MAX_WRITEPAGES_EXTENT_LEN blocks and then we go on and finish mapping
  * the last partial page. So in total we can map MAX_WRITEPAGES_EXTENT_LEN +
  * bpp - 1 blocks in bpp different extents.
  */
@@ -2319,14 +2356,10 @@ static int mpage_prepare_extent_to_map(struct mpage_da_data *mpd)
 			lblk = ((ext4_lblk_t)page->index) <<
 				(PAGE_CACHE_SHIFT - blkbits);
 			head = page_buffers(page);
-			if (!add_page_bufs_to_extent(mpd, head, head, lblk))
+			err = mpage_process_page_bufs(mpd, head, head, lblk);
+			if (err <= 0)
 				goto out;
-			/* So far everything mapped? Submit the page for IO. */
-			if (mpd->map.m_len == 0) {
-				err = mpage_submit_page(mpd, page);
-				if (err < 0)
-					goto out;
-			}
+			err = 0;
 
 			/*
 			 * Accumulated enough dirty pages? This doesn't apply
@@ -2410,7 +2443,7 @@ static int ext4_writepages(struct address_space *mapping,
 
 	if (ext4_should_dioread_nolock(inode)) {
 		/*
-		 * We may need to convert upto one extent per block in
+		 * We may need to convert up to one extent per block in
 		 * the page and we may dirty the inode.
 		 */
 		rsv_blocks = 1 + (PAGE_CACHE_SIZE >> inode->i_blkbits);
@@ -2646,7 +2679,7 @@ retry_journal:
 		goto retry_grab;
 	}
 	/* In case writeback began while the page was unlocked */
-	wait_on_page_writeback(page);
+	wait_for_stable_page(page);
 
 	ret = __block_write_begin(page, pos, len, ext4_da_get_block_prep);
 	if (ret < 0) {
@@ -4566,7 +4599,9 @@ int ext4_setattr(struct dentry *dentry, struct iattr *attr)
 		ext4_journal_stop(handle);
 	}
 
-	if (attr->ia_valid & ATTR_SIZE) {
+	if (attr->ia_valid & ATTR_SIZE && attr->ia_size != inode->i_size) {
+		handle_t *handle;
+		loff_t oldsize = inode->i_size;
 
 		if (!(ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS))) {
 			struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb);
@@ -4574,73 +4609,69 @@ int ext4_setattr(struct dentry *dentry, struct iattr *attr)
 			if (attr->ia_size > sbi->s_bitmap_maxbytes)
 				return -EFBIG;
 		}
-	}
-
-	if (S_ISREG(inode->i_mode) &&
-	    attr->ia_valid & ATTR_SIZE &&
-	    (attr->ia_size < inode->i_size)) {
-		handle_t *handle;
-
-		handle = ext4_journal_start(inode, EXT4_HT_INODE, 3);
-		if (IS_ERR(handle)) {
-			error = PTR_ERR(handle);
-			goto err_out;
-		}
-		if (ext4_handle_valid(handle)) {
-			error = ext4_orphan_add(handle, inode);
-			orphan = 1;
-		}
-		EXT4_I(inode)->i_disksize = attr->ia_size;
-		rc = ext4_mark_inode_dirty(handle, inode);
-		if (!error)
-			error = rc;
-		ext4_journal_stop(handle);
-
-		if (ext4_should_order_data(inode)) {
-			error = ext4_begin_ordered_truncate(inode,
+		if (S_ISREG(inode->i_mode) &&
+		    (attr->ia_size < inode->i_size)) {
+			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,
-							    EXT4_HT_INODE, 3);
-				if (IS_ERR(handle)) {
-					ext4_orphan_del(NULL, inode);
+				if (error)
 					goto err_out;
-				}
-				ext4_orphan_del(handle, inode);
-				orphan = 0;
-				ext4_journal_stop(handle);
+			}
+			handle = ext4_journal_start(inode, EXT4_HT_INODE, 3);
+			if (IS_ERR(handle)) {
+				error = PTR_ERR(handle);
 				goto err_out;
 			}
-		}
-	}
-
-	if (attr->ia_valid & ATTR_SIZE) {
-		if (attr->ia_size != inode->i_size) {
-			loff_t oldsize = inode->i_size;
-
-			i_size_write(inode, attr->ia_size);
-			/*
-			 * Blocks are going to be removed from the inode. Wait
-			 * for dio in flight.  Temporarily disable
-			 * dioread_nolock to prevent livelock.
-			 */
-			if (orphan) {
-				if (!ext4_should_journal_data(inode)) {
-					ext4_inode_block_unlocked_dio(inode);
-					inode_dio_wait(inode);
-					ext4_inode_resume_unlocked_dio(inode);
-				} else
-					ext4_wait_for_tail_page_commit(inode);
+			if (ext4_handle_valid(handle)) {
+				error = ext4_orphan_add(handle, inode);
+				orphan = 1;
 			}
+			down_write(&EXT4_I(inode)->i_data_sem);
+			EXT4_I(inode)->i_disksize = attr->ia_size;
+			rc = ext4_mark_inode_dirty(handle, inode);
+			if (!error)
+				error = rc;
 			/*
-			 * Truncate pagecache after we've waited for commit
-			 * in data=journal mode to make pages freeable.
+			 * We have to update i_size under i_data_sem together
+			 * with i_disksize to avoid races with writeback code
+			 * running ext4_wb_update_i_disksize().
 			 */
-			truncate_pagecache(inode, oldsize, inode->i_size);
+			if (!error)
+				i_size_write(inode, attr->ia_size);
+			up_write(&EXT4_I(inode)->i_data_sem);
+			ext4_journal_stop(handle);
+			if (error) {
+				ext4_orphan_del(NULL, inode);
+				goto err_out;
+			}
+		} else
+			i_size_write(inode, attr->ia_size);
+
+		/*
+		 * Blocks are going to be removed from the inode. Wait
+		 * for dio in flight.  Temporarily disable
+		 * dioread_nolock to prevent livelock.
+		 */
+		if (orphan) {
+			if (!ext4_should_journal_data(inode)) {
+				ext4_inode_block_unlocked_dio(inode);
+				inode_dio_wait(inode);
+				ext4_inode_resume_unlocked_dio(inode);
+			} else
+				ext4_wait_for_tail_page_commit(inode);
 		}
-		ext4_truncate(inode);
+		/*
+		 * Truncate pagecache after we've waited for commit
+		 * in data=journal mode to make pages freeable.
+		 */
+		truncate_pagecache(inode, oldsize, inode->i_size);
 	}
+	/*
+	 * We want to call ext4_truncate() even if attr->ia_size ==
+	 * inode->i_size for cases like truncation of fallocated space
+	 */
+	if (attr->ia_valid & ATTR_SIZE)
+		ext4_truncate(inode);
 
 	if (!rc) {
 		setattr_copy(inode, attr);