Btrfs: fix regression in re-setting a large xattr

Recently I changed the xattr stuff to unconditionally set the xattr first in
case the xattr didn't exist yet.  This has introduced a regression when setting
an xattr that already exists with a large value.  If we find the key we are
looking for split_leaf will assume that we're extending that item.  The problem
is the size we pass down to btrfs_search_slot includes the size of the item
already, so if we have the largest xattr we can possibly have plus the size of
the xattr item plus the xattr item that btrfs_search_slot we'd overflow the
leaf.  Thankfully this is not what we're doing, but split_leaf doesn't know this
so it just returns EOVERFLOW.  So in the xattr code we need to check and see if
we got back EOVERFLOW and treat it like EEXIST since that's really what
happened.  Thanks,

Signed-off-by: Josef Bacik <josef@redhat.com>

fs/btrfs/xattr.c

@@ -127,6 +127,17 @@ static int do_setxattr(struct btrfs_trans_handle *trans,
 again:
 	ret = btrfs_insert_xattr_item(trans, root, path, btrfs_ino(inode),
 				      name, name_len, value, size);
+	/*
+	 * If we're setting an xattr to a new value but the new value is say
+	 * exactly BTRFS_MAX_XATTR_SIZE, we could end up with EOVERFLOW getting
+	 * back from split_leaf.  This is because it thinks we'll be extending
+	 * the existing item size, but we're asking for enough space to add the
+	 * item itself.  So if we get EOVERFLOW just set ret to EEXIST and let
+	 * the rest of the function figure it out.
+	 */
+	if (ret == -EOVERFLOW)
+		ret = -EEXIST;
+
 	if (ret == -EEXIST) {
 		if (flags & XATTR_CREATE)
 			goto out;