Btrfs: fix reservations in btrfs_page_mkwrite

Josef fixed btrfs_page_mkwrite to properly release reserved
extents if there was an error.  But if we fail to get a reservation
and we fail to dirty the inode (for ENOSPC reasons), we'll end up
trying to release a reservation we never had.

This makes sure we only release if we were able to reserve.
Signed-off-by: NChris Mason <chris.mason@oracle.com>

fs/btrfs/inode.c

@@ -6401,18 +6401,23 @@ int btrfs_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf)
 	unsigned long zero_start;
 	loff_t size;
 	int ret;
+	int reserved = 0;
 	u64 page_start;
 	u64 page_end;
 
 	ret  = btrfs_delalloc_reserve_space(inode, PAGE_CACHE_SIZE);
-	if (!ret)
+	if (!ret) {
 		ret = btrfs_update_time(vma->vm_file);
+		reserved = 1;
+	}
 	if (ret) {
 		if (ret == -ENOMEM)
 			ret = VM_FAULT_OOM;
 		else /* -ENOSPC, -EIO, etc */
 			ret = VM_FAULT_SIGBUS;
-		goto out;
+		if (reserved)
+			goto out;
+		goto out_noreserve;
 	}
 
 	ret = VM_FAULT_NOPAGE; /* make the VM retry the fault */
@@ -6495,6 +6500,7 @@ int btrfs_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf)
 	unlock_page(page);
 out:
 	btrfs_delalloc_release_space(inode, PAGE_CACHE_SIZE);
+out_noreserve:
 	return ret;
 }