Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs

Pull btrfs updates from Chris Mason:
 "The largest set of changes here come from Miao Xie.  He's cleaning up
  and improving read recovery/repair for raid, and has a number of
  related fixes.

  I've merged another set of fsync fixes from Filipe, and he's also
  improved the way we handle metadata write errors to make sure we force
  the FS readonly if things go wrong.

  Otherwise we have a collection of fixes and cleanups.  Dave Sterba
  gets a cookie for removing the most lines (thanks Dave)"

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs: (139 commits)
  btrfs: Fix compile error when CONFIG_SECURITY is not set.
  Btrfs: fix compiles when CONFIG_BTRFS_FS_RUN_SANITY_TESTS is off
  btrfs: Make btrfs handle security mount options internally to avoid losing security label.
  Btrfs: send, don't delay dir move if there's a new parent inode
  btrfs: add more superblock checks
  Btrfs: fix race in WAIT_SYNC ioctl
  Btrfs: be aware of btree inode write errors to avoid fs corruption
  Btrfs: remove redundant btrfs_verify_qgroup_counts declaration.
  btrfs: fix shadow warning on cmp
  Btrfs: fix compilation errors under DEBUG
  Btrfs: fix crash of btrfs_release_extent_buffer_page
  Btrfs: add missing end_page_writeback on submit_extent_page failure
  btrfs: Fix the wrong condition judgment about subset extent map
  Btrfs: fix build_backref_tree issue with multiple shared blocks
  Btrfs: cleanup error handling in build_backref_tree
  btrfs: move checks for DUMMY_ROOT into a helper
  btrfs: new define for the inline extent data start
  btrfs: kill extent_buffer_page helper
  btrfs: drop constant param from btrfs_release_extent_buffer_page
  btrfs: hide typecast to definition of BTRFS_SEND_TRANS_STUB
  ...

1 files changed, 110 insertions, 41 deletions

diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c
index ff1cc0399b9..a18ceabd99a 100644
--- a/fs/btrfs/file.c
+++ b/fs/btrfs/file.c
@@ -299,7 +299,7 @@ static int __btrfs_run_defrag_inode(struct btrfs_fs_info *fs_info,
 
 	/* get the inode */
 	key.objectid = defrag->root;
-	btrfs_set_key_type(&key, BTRFS_ROOT_ITEM_KEY);
+	key.type = BTRFS_ROOT_ITEM_KEY;
 	key.offset = (u64)-1;
 
 	index = srcu_read_lock(&fs_info->subvol_srcu);
@@ -311,7 +311,7 @@ static int __btrfs_run_defrag_inode(struct btrfs_fs_info *fs_info,
 	}
 
 	key.objectid = defrag->ino;
-	btrfs_set_key_type(&key, BTRFS_INODE_ITEM_KEY);
+	key.type = BTRFS_INODE_ITEM_KEY;
 	key.offset = 0;
 	inode = btrfs_iget(fs_info->sb, &key, inode_root, NULL);
 	if (IS_ERR(inode)) {
@@ -452,7 +452,7 @@ static noinline int btrfs_copy_from_user(loff_t pos, int num_pages,
 		if (unlikely(copied == 0))
 			break;
 
-		if (unlikely(copied < PAGE_CACHE_SIZE - offset)) {
+		if (copied < PAGE_CACHE_SIZE - offset) {
 			offset += copied;
 		} else {
 			pg++;
@@ -1481,9 +1481,8 @@ static noinline ssize_t __btrfs_buffered_write(struct file *file,
 	bool force_page_uptodate = false;
 	bool need_unlock;
 
-	nrptrs = min((iov_iter_count(i) + PAGE_CACHE_SIZE - 1) /
-		     PAGE_CACHE_SIZE, PAGE_CACHE_SIZE /
-		     (sizeof(struct page *)));
+	nrptrs = min(DIV_ROUND_UP(iov_iter_count(i), PAGE_CACHE_SIZE),
+			PAGE_CACHE_SIZE / (sizeof(struct page *)));
 	nrptrs = min(nrptrs, current->nr_dirtied_pause - current->nr_dirtied);
 	nrptrs = max(nrptrs, 8);
 	pages = kmalloc(nrptrs * sizeof(struct page *), GFP_KERNEL);
@@ -1497,8 +1496,8 @@ static noinline ssize_t __btrfs_buffered_write(struct file *file,
 		size_t write_bytes = min(iov_iter_count(i),
 					 nrptrs * (size_t)PAGE_CACHE_SIZE -
 					 offset);
-		size_t num_pages = (write_bytes + offset +
-				    PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT;
+		size_t num_pages = DIV_ROUND_UP(write_bytes + offset,
+						PAGE_CACHE_SIZE);
 		size_t reserve_bytes;
 		size_t dirty_pages;
 		size_t copied;
@@ -1526,9 +1525,8 @@ static noinline ssize_t __btrfs_buffered_write(struct file *file,
 				 * our prealloc extent may be smaller than
 				 * write_bytes, so scale down.
 				 */
-				num_pages = (write_bytes + offset +
-					     PAGE_CACHE_SIZE - 1) >>
-					PAGE_CACHE_SHIFT;
+				num_pages = DIV_ROUND_UP(write_bytes + offset,
+							 PAGE_CACHE_SIZE);
 				reserve_bytes = num_pages << PAGE_CACHE_SHIFT;
 				ret = 0;
 			} else {
@@ -1590,9 +1588,8 @@ again:
 			dirty_pages = 0;
 		} else {
 			force_page_uptodate = false;
-			dirty_pages = (copied + offset +
-				       PAGE_CACHE_SIZE - 1) >>
-				       PAGE_CACHE_SHIFT;
+			dirty_pages = DIV_ROUND_UP(copied + offset,
+						   PAGE_CACHE_SIZE);
 		}
 
 		/*
@@ -1653,7 +1650,7 @@ again:
 		cond_resched();
 
 		balance_dirty_pages_ratelimited(inode->i_mapping);
-		if (dirty_pages < (root->leafsize >> PAGE_CACHE_SHIFT) + 1)
+		if (dirty_pages < (root->nodesize >> PAGE_CACHE_SHIFT) + 1)
 			btrfs_btree_balance_dirty(root);
 
 		pos += copied;
@@ -1795,7 +1792,7 @@ static ssize_t btrfs_file_write_iter(struct kiocb *iocb,
 	if (sync)
 		atomic_inc(&BTRFS_I(inode)->sync_writers);
 
-	if (unlikely(file->f_flags & O_DIRECT)) {
+	if (file->f_flags & O_DIRECT) {
 		num_written = __btrfs_direct_write(iocb, from, pos);
 	} else {
 		num_written = __btrfs_buffered_write(file, from, pos);
@@ -1852,6 +1849,20 @@ int btrfs_release_file(struct inode *inode, struct file *filp)
 	return 0;
 }
 
+static int start_ordered_ops(struct inode *inode, loff_t start, loff_t end)
+{
+	int ret;
+
+	atomic_inc(&BTRFS_I(inode)->sync_writers);
+	ret = filemap_fdatawrite_range(inode->i_mapping, start, end);
+	if (!ret && test_bit(BTRFS_INODE_HAS_ASYNC_EXTENT,
+			     &BTRFS_I(inode)->runtime_flags))
+		ret = filemap_fdatawrite_range(inode->i_mapping, start, end);
+	atomic_dec(&BTRFS_I(inode)->sync_writers);
+
+	return ret;
+}
+
 /*
  * fsync call for both files and directories.  This logs the inode into
  * the tree log instead of forcing full commits whenever possible.
@@ -1881,30 +1892,64 @@ int btrfs_sync_file(struct file *file, loff_t start, loff_t end, int datasync)
 	 * multi-task, and make the performance up.  See
 	 * btrfs_wait_ordered_range for an explanation of the ASYNC check.
 	 */
-	atomic_inc(&BTRFS_I(inode)->sync_writers);
-	ret = filemap_fdatawrite_range(inode->i_mapping, start, end);
-	if (!ret && test_bit(BTRFS_INODE_HAS_ASYNC_EXTENT,
-			     &BTRFS_I(inode)->runtime_flags))
-		ret = filemap_fdatawrite_range(inode->i_mapping, start, end);
-	atomic_dec(&BTRFS_I(inode)->sync_writers);
+	ret = start_ordered_ops(inode, start, end);
 	if (ret)
 		return ret;
 
 	mutex_lock(&inode->i_mutex);
-
-	/*
-	 * We flush the dirty pages again to avoid some dirty pages in the
-	 * range being left.
-	 */
 	atomic_inc(&root->log_batch);
 	full_sync = test_bit(BTRFS_INODE_NEEDS_FULL_SYNC,
 			     &BTRFS_I(inode)->runtime_flags);
+	/*
+	 * We might have have had more pages made dirty after calling
+	 * start_ordered_ops and before acquiring the inode's i_mutex.
+	 */
 	if (full_sync) {
+		/*
+		 * For a full sync, we need to make sure any ordered operations
+		 * start and finish before we start logging the inode, so that
+		 * all extents are persisted and the respective file extent
+		 * items are in the fs/subvol btree.
+		 */
 		ret = btrfs_wait_ordered_range(inode, start, end - start + 1);
-		if (ret) {
-			mutex_unlock(&inode->i_mutex);
-			goto out;
-		}
+	} else {
+		/*
+		 * Start any new ordered operations before starting to log the
+		 * inode. We will wait for them to finish in btrfs_sync_log().
+		 *
+		 * Right before acquiring the inode's mutex, we might have new
+		 * writes dirtying pages, which won't immediately start the
+		 * respective ordered operations - that is done through the
+		 * fill_delalloc callbacks invoked from the writepage and
+		 * writepages address space operations. So make sure we start
+		 * all ordered operations before starting to log our inode. Not
+		 * doing this means that while logging the inode, writeback
+		 * could start and invoke writepage/writepages, which would call
+		 * the fill_delalloc callbacks (cow_file_range,
+		 * submit_compressed_extents). These callbacks add first an
+		 * extent map to the modified list of extents and then create
+		 * the respective ordered operation, which means in
+		 * tree-log.c:btrfs_log_inode() we might capture all existing
+		 * ordered operations (with btrfs_get_logged_extents()) before
+		 * the fill_delalloc callback adds its ordered operation, and by
+		 * the time we visit the modified list of extent maps (with
+		 * btrfs_log_changed_extents()), we see and process the extent
+		 * map they created. We then use the extent map to construct a
+		 * file extent item for logging without waiting for the
+		 * respective ordered operation to finish - this file extent
+		 * item points to a disk location that might not have yet been
+		 * written to, containing random data - so after a crash a log
+		 * replay will make our inode have file extent items that point
+		 * to disk locations containing invalid data, as we returned
+		 * success to userspace without waiting for the respective
+		 * ordered operation to finish, because it wasn't captured by
+		 * btrfs_get_logged_extents().
+		 */
+		ret = start_ordered_ops(inode, start, end);
+	}
+	if (ret) {
+		mutex_unlock(&inode->i_mutex);
+		goto out;
 	}
 	atomic_inc(&root->log_batch);
 
@@ -1984,6 +2029,25 @@ int btrfs_sync_file(struct file *file, loff_t start, loff_t end, int datasync)
 	 */
 	mutex_unlock(&inode->i_mutex);
 
+	/*
+	 * If any of the ordered extents had an error, just return it to user
+	 * space, so that the application knows some writes didn't succeed and
+	 * can take proper action (retry for e.g.). Blindly committing the
+	 * transaction in this case, would fool userspace that everything was
+	 * successful. And we also want to make sure our log doesn't contain
+	 * file extent items pointing to extents that weren't fully written to -
+	 * just like in the non fast fsync path, where we check for the ordered
+	 * operation's error flag before writing to the log tree and return -EIO
+	 * if any of them had this flag set (btrfs_wait_ordered_range) -
+	 * therefore we need to check for errors in the ordered operations,
+	 * which are indicated by ctx.io_err.
+	 */
+	if (ctx.io_err) {
+		btrfs_end_transaction(trans, root);
+		ret = ctx.io_err;
+		goto out;
+	}
+
 	if (ret != BTRFS_NO_LOG_SYNC) {
 		if (!ret) {
 			ret = btrfs_sync_log(trans, root, &ctx);
@@ -2621,23 +2685,28 @@ static int find_desired_extent(struct inode *inode, loff_t *offset, int whence)
 	struct btrfs_root *root = BTRFS_I(inode)->root;
 	struct extent_map *em = NULL;
 	struct extent_state *cached_state = NULL;
-	u64 lockstart = *offset;
-	u64 lockend = i_size_read(inode);
-	u64 start = *offset;
-	u64 len = i_size_read(inode);
+	u64 lockstart;
+	u64 lockend;
+	u64 start;
+	u64 len;
 	int ret = 0;
 
-	lockend = max_t(u64, root->sectorsize, lockend);
+	if (inode->i_size == 0)
+		return -ENXIO;
+
+	/*
+	 * *offset can be negative, in this case we start finding DATA/HOLE from
+	 * the very start of the file.
+	 */
+	start = max_t(loff_t, 0, *offset);
+
+	lockstart = round_down(start, root->sectorsize);
+	lockend = round_up(i_size_read(inode), root->sectorsize);
 	if (lockend <= lockstart)
 		lockend = lockstart + root->sectorsize;
-
 	lockend--;
 	len = lockend - lockstart + 1;
 
-	len = max_t(u64, len, root->sectorsize);
-	if (inode->i_size == 0)
-		return -ENXIO;
-
 	lock_extent_bits(&BTRFS_I(inode)->io_tree, lockstart, lockend, 0,
 			 &cached_state);