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

Pull btrfs fixes from Chris Mason:
 "The biggest of these comes from Liu Bo, who tracked down a hang we've
  been hitting since moving to kernel workqueues (it's a btrfs bug, not
  in the generic code).  His patch needs backporting to 3.16 and 3.15
  stable, which I'll send once this is in.

  Otherwise these are assorted fixes.  Most were integrated last week
  during KS, but I wanted to give everyone the chance to test the
  result, so I waited for rc2 to come out before sending"

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs: (24 commits)
  Btrfs: fix task hang under heavy compressed write
  Btrfs: fix filemap_flush call in btrfs_file_release
  Btrfs: fix crash on endio of reading corrupted block
  btrfs: fix leak in qgroup_subtree_accounting() error path
  btrfs: Use right extent length when inserting overlap extent map.
  Btrfs: clone, don't create invalid hole extent map
  Btrfs: don't monopolize a core when evicting inode
  Btrfs: fix hole detection during file fsync
  Btrfs: ensure tmpfile inode is always persisted with link count of 0
  Btrfs: race free update of commit root for ro snapshots
  Btrfs: fix regression of btrfs device replace
  Btrfs: don't consider the missing device when allocating new chunks
  Btrfs: Fix wrong device size when we are resizing the device
  Btrfs: don't write any data into a readonly device when scrub
  Btrfs: Fix the problem that the replace destroys the seed filesystem
  btrfs: Return right extent when fiemap gives unaligned offset and len.
  Btrfs: fix wrong extent mapping for DirectIO
  Btrfs: fix wrong write range for filemap_fdatawrite_range()
  Btrfs: fix wrong missing device counter decrease
  Btrfs: fix unzeroed members in fs_devices when creating a fs from seed fs
  ...

1 files changed, 89 insertions, 20 deletions

diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index 03708ef3deef..9c194bd74d6e 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -1096,8 +1096,10 @@ static int cow_file_range_async(struct inode *inode, struct page *locked_page,
 		async_cow->end = cur_end;
 		INIT_LIST_HEAD(&async_cow->extents);
 
-		btrfs_init_work(&async_cow->work, async_cow_start,
-				async_cow_submit, async_cow_free);
+		btrfs_init_work(&async_cow->work,
+				btrfs_delalloc_helper,
+				async_cow_start, async_cow_submit,
+				async_cow_free);
 
 		nr_pages = (cur_end - start + PAGE_CACHE_SIZE) >>
 			PAGE_CACHE_SHIFT;
@@ -1881,7 +1883,8 @@ static int btrfs_writepage_start_hook(struct page *page, u64 start, u64 end)
 
 	SetPageChecked(page);
 	page_cache_get(page);
-	btrfs_init_work(&fixup->work, btrfs_writepage_fixup_worker, NULL, NULL);
+	btrfs_init_work(&fixup->work, btrfs_fixup_helper,
+			btrfs_writepage_fixup_worker, NULL, NULL);
 	fixup->page = page;
 	btrfs_queue_work(root->fs_info->fixup_workers, &fixup->work);
 	return -EBUSY;
@@ -2822,7 +2825,8 @@ static int btrfs_writepage_end_io_hook(struct page *page, u64 start, u64 end,
 	struct inode *inode = page->mapping->host;
 	struct btrfs_root *root = BTRFS_I(inode)->root;
 	struct btrfs_ordered_extent *ordered_extent = NULL;
-	struct btrfs_workqueue *workers;
+	struct btrfs_workqueue *wq;
+	btrfs_work_func_t func;
 
 	trace_btrfs_writepage_end_io_hook(page, start, end, uptodate);
 
@@ -2831,13 +2835,17 @@ static int btrfs_writepage_end_io_hook(struct page *page, u64 start, u64 end,
 					    end - start + 1, uptodate))
 		return 0;
 
-	btrfs_init_work(&ordered_extent->work, finish_ordered_fn, NULL, NULL);
+	if (btrfs_is_free_space_inode(inode)) {
+		wq = root->fs_info->endio_freespace_worker;
+		func = btrfs_freespace_write_helper;
+	} else {
+		wq = root->fs_info->endio_write_workers;
+		func = btrfs_endio_write_helper;
+	}
 
-	if (btrfs_is_free_space_inode(inode))
-		workers = root->fs_info->endio_freespace_worker;
-	else
-		workers = root->fs_info->endio_write_workers;
-	btrfs_queue_work(workers, &ordered_extent->work);
+	btrfs_init_work(&ordered_extent->work, func, finish_ordered_fn, NULL,
+			NULL);
+	btrfs_queue_work(wq, &ordered_extent->work);
 
 	return 0;
 }
@@ -4674,6 +4682,11 @@ static void evict_inode_truncate_pages(struct inode *inode)
 		clear_bit(EXTENT_FLAG_LOGGING, &em->flags);
 		remove_extent_mapping(map_tree, em);
 		free_extent_map(em);
+		if (need_resched()) {
+			write_unlock(&map_tree->lock);
+			cond_resched();
+			write_lock(&map_tree->lock);
+		}
 	}
 	write_unlock(&map_tree->lock);
 
@@ -4696,6 +4709,7 @@ static void evict_inode_truncate_pages(struct inode *inode)
 				 &cached_state, GFP_NOFS);
 		free_extent_state(state);
 
+		cond_resched();
 		spin_lock(&io_tree->lock);
 	}
 	spin_unlock(&io_tree->lock);
@@ -5181,6 +5195,42 @@ struct inode *btrfs_lookup_dentry(struct inode *dir, struct dentry *dentry)
 			iput(inode);
 			inode = ERR_PTR(ret);
 		}
+		/*
+		 * If orphan cleanup did remove any orphans, it means the tree
+		 * was modified and therefore the commit root is not the same as
+		 * the current root anymore. This is a problem, because send
+		 * uses the commit root and therefore can see inode items that
+		 * don't exist in the current root anymore, and for example make
+		 * calls to btrfs_iget, which will do tree lookups based on the
+		 * current root and not on the commit root. Those lookups will
+		 * fail, returning a -ESTALE error, and making send fail with
+		 * that error. So make sure a send does not see any orphans we
+		 * have just removed, and that it will see the same inodes
+		 * regardless of whether a transaction commit happened before
+		 * it started (meaning that the commit root will be the same as
+		 * the current root) or not.
+		 */
+		if (sub_root->node != sub_root->commit_root) {
+			u64 sub_flags = btrfs_root_flags(&sub_root->root_item);
+
+			if (sub_flags & BTRFS_ROOT_SUBVOL_RDONLY) {
+				struct extent_buffer *eb;
+
+				/*
+				 * Assert we can't have races between dentry
+				 * lookup called through the snapshot creation
+				 * ioctl and the VFS.
+				 */
+				ASSERT(mutex_is_locked(&dir->i_mutex));
+
+				down_write(&root->fs_info->commit_root_sem);
+				eb = sub_root->commit_root;
+				sub_root->commit_root =
+					btrfs_root_node(sub_root);
+				up_write(&root->fs_info->commit_root_sem);
+				free_extent_buffer(eb);
+			}
+		}
 	}
 
 	return inode;
@@ -5606,6 +5656,13 @@ static struct inode *btrfs_new_inode(struct btrfs_trans_handle *trans,
 	}
 
 	/*
+	 * O_TMPFILE, set link count to 0, so that after this point,
+	 * we fill in an inode item with the correct link count.
+	 */
+	if (!name)
+		set_nlink(inode, 0);
+
+	/*
 	 * we have to initialize this early, so we can reclaim the inode
 	 * number if we fail afterwards in this function.
 	 */
@@ -6097,14 +6154,14 @@ out_fail:
 static int merge_extent_mapping(struct extent_map_tree *em_tree,
 				struct extent_map *existing,
 				struct extent_map *em,
-				u64 map_start, u64 map_len)
+				u64 map_start)
 {
 	u64 start_diff;
 
 	BUG_ON(map_start < em->start || map_start >= extent_map_end(em));
 	start_diff = map_start - em->start;
 	em->start = map_start;
-	em->len = map_len;
+	em->len = existing->start - em->start;
 	if (em->block_start < EXTENT_MAP_LAST_BYTE &&
 	    !test_bit(EXTENT_FLAG_COMPRESSED, &em->flags)) {
 		em->block_start += start_diff;
@@ -6275,6 +6332,8 @@ next:
 			goto not_found;
 		if (start + len <= found_key.offset)
 			goto not_found;
+		if (start > found_key.offset)
+			goto next;
 		em->start = start;
 		em->orig_start = start;
 		em->len = found_key.offset - start;
@@ -6390,8 +6449,7 @@ insert:
 							 em->len);
 			if (existing) {
 				err = merge_extent_mapping(em_tree, existing,
-							   em, start,
-							   root->sectorsize);
+							   em, start);
 				free_extent_map(existing);
 				if (err) {
 					free_extent_map(em);
@@ -7158,7 +7216,8 @@ again:
 	if (!ret)
 		goto out_test;
 
-	btrfs_init_work(&ordered->work, finish_ordered_fn, NULL, NULL);
+	btrfs_init_work(&ordered->work, btrfs_endio_write_helper,
+			finish_ordered_fn, NULL, NULL);
 	btrfs_queue_work(root->fs_info->endio_write_workers,
 			 &ordered->work);
 out_test:
@@ -7306,10 +7365,8 @@ static int btrfs_submit_direct_hook(int rw, struct btrfs_dio_private *dip,
 	map_length = orig_bio->bi_iter.bi_size;
 	ret = btrfs_map_block(root->fs_info, rw, start_sector << 9,
 			      &map_length, NULL, 0);
-	if (ret) {
-		bio_put(orig_bio);
+	if (ret)
 		return -EIO;
-	}
 
 	if (map_length >= orig_bio->bi_iter.bi_size) {
 		bio = orig_bio;
@@ -7326,6 +7383,7 @@ static int btrfs_submit_direct_hook(int rw, struct btrfs_dio_private *dip,
 	bio = btrfs_dio_bio_alloc(orig_bio->bi_bdev, start_sector, GFP_NOFS);
 	if (!bio)
 		return -ENOMEM;
+
 	bio->bi_private = dip;
 	bio->bi_end_io = btrfs_end_dio_bio;
 	atomic_inc(&dip->pending_bios);
@@ -7534,7 +7592,8 @@ static ssize_t btrfs_direct_IO(int rw, struct kiocb *iocb,
 	count = iov_iter_count(iter);
 	if (test_bit(BTRFS_INODE_HAS_ASYNC_EXTENT,
 		     &BTRFS_I(inode)->runtime_flags))
-		filemap_fdatawrite_range(inode->i_mapping, offset, count);
+		filemap_fdatawrite_range(inode->i_mapping, offset,
+					 offset + count - 1);
 
 	if (rw & WRITE) {
 		/*
@@ -8495,7 +8554,9 @@ struct btrfs_delalloc_work *btrfs_alloc_delalloc_work(struct inode *inode,
 	work->inode = inode;
 	work->wait = wait;
 	work->delay_iput = delay_iput;
-	btrfs_init_work(&work->work, btrfs_run_delalloc_work, NULL, NULL);
+	WARN_ON_ONCE(!inode);
+	btrfs_init_work(&work->work, btrfs_flush_delalloc_helper,
+			btrfs_run_delalloc_work, NULL, NULL);
 
 	return work;
 }
@@ -8979,6 +9040,14 @@ static int btrfs_tmpfile(struct inode *dir, struct dentry *dentry, umode_t mode)
 	if (ret)
 		goto out;
 
+	/*
+	 * We set number of links to 0 in btrfs_new_inode(), and here we set
+	 * it to 1 because d_tmpfile() will issue a warning if the count is 0,
+	 * through:
+	 *
+	 *    d_tmpfile() -> inode_dec_link_count() -> drop_nlink()
+	 */
+	set_nlink(inode, 1);
 	d_tmpfile(dentry, inode);
 	mark_inode_dirty(inode);