Merge tag 'for-5.12-rc3-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux

Pull btrfs fixes from David Sterba:
 "There are still regressions being found and fixed in the zoned mode
  and subpage code, the rest are fixes for bugs reported by users.

  Regressions:

   - subpage block support:
      - readahead works on the proper block size
      - fix last page zeroing

   - zoned mode:
      - linked list corruption for tree log

  Fixes:

   - qgroup leak after falloc failure

   - tree mod log and backref resolving:
      - extent buffer cloning race when resolving backrefs
      - pin deleted leaves with active tree mod log users

   - drop debugging flag from slab cache"

* tag 'for-5.12-rc3-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux:
  btrfs: always pin deleted leaves when there are active tree mod log users
  btrfs: fix race when cloning extent buffer during rewind of an old root
  btrfs: fix slab cache flags for free space tree bitmap
  btrfs: subpage: make readahead work properly
  btrfs: subpage: fix wild pointer access during metadata read failure
  btrfs: zoned: fix linked list corruption after log root tree allocation failure
  btrfs: fix qgroup data rsv leak caused by falloc failure
  btrfs: track qgroup released data in own variable in insert_prealloc_file_extent
  btrfs: fix wrong offset to zero out range beyond i_size

1 files changed, 26 insertions, 11 deletions

diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index 35bfa0533f23..7cdf65be3707 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -9008,7 +9008,7 @@ int __init btrfs_init_cachep(void)
 
 	btrfs_free_space_bitmap_cachep = kmem_cache_create("btrfs_free_space_bitmap",
 							PAGE_SIZE, PAGE_SIZE,
-							SLAB_RED_ZONE, NULL);
+							SLAB_MEM_SPREAD, NULL);
 	if (!btrfs_free_space_bitmap_cachep)
 		goto fail;
 
@@ -9877,6 +9877,7 @@ static struct btrfs_trans_handle *insert_prealloc_file_extent(
 	struct btrfs_path *path;
 	u64 start = ins->objectid;
 	u64 len = ins->offset;
+	int qgroup_released;
 	int ret;
 
 	memset(&stack_fi, 0, sizeof(stack_fi));
@@ -9889,16 +9890,16 @@ static struct btrfs_trans_handle *insert_prealloc_file_extent(
 	btrfs_set_stack_file_extent_compression(&stack_fi, BTRFS_COMPRESS_NONE);
 	/* Encryption and other encoding is reserved and all 0 */
 
-	ret = btrfs_qgroup_release_data(inode, file_offset, len);
-	if (ret < 0)
-		return ERR_PTR(ret);
+	qgroup_released = btrfs_qgroup_release_data(inode, file_offset, len);
+	if (qgroup_released < 0)
+		return ERR_PTR(qgroup_released);
 
 	if (trans) {
 		ret = insert_reserved_file_extent(trans, inode,
 						  file_offset, &stack_fi,
-						  true, ret);
+						  true, qgroup_released);
 		if (ret)
-			return ERR_PTR(ret);
+			goto free_qgroup;
 		return trans;
 	}
 
@@ -9909,21 +9910,35 @@ static struct btrfs_trans_handle *insert_prealloc_file_extent(
 	extent_info.file_offset = file_offset;
 	extent_info.extent_buf = (char *)&stack_fi;
 	extent_info.is_new_extent = true;
-	extent_info.qgroup_reserved = ret;
+	extent_info.qgroup_reserved = qgroup_released;
 	extent_info.insertions = 0;
 
 	path = btrfs_alloc_path();
-	if (!path)
-		return ERR_PTR(-ENOMEM);
+	if (!path) {
+		ret = -ENOMEM;
+		goto free_qgroup;
+	}
 
 	ret = btrfs_replace_file_extents(&inode->vfs_inode, path, file_offset,
 				     file_offset + len - 1, &extent_info,
 				     &trans);
 	btrfs_free_path(path);
 	if (ret)
-		return ERR_PTR(ret);
-
+		goto free_qgroup;
 	return trans;
+
+free_qgroup:
+	/*
+	 * We have released qgroup data range at the beginning of the function,
+	 * and normally qgroup_released bytes will be freed when committing
+	 * transaction.
+	 * But if we error out early, we have to free what we have released
+	 * or we leak qgroup data reservation.
+	 */
+	btrfs_qgroup_free_refroot(inode->root->fs_info,
+			inode->root->root_key.objectid, qgroup_released,
+			BTRFS_QGROUP_RSV_DATA);
+	return ERR_PTR(ret);
 }
 
 static int __btrfs_prealloc_file_range(struct inode *inode, int mode,