Is it a bad idea to make redundant rows when 2 columns reference to the same table?

Question

Suppose we have this database design

Users table:
- id
- name

Friends table:
- id
- user_id
- friend_user_id

Let's say USER 1 befriended USER 2, this is what my application logic will execute:

INSERT INTO friends (user_id, friend_user_id) VALUES (1, 2);
INSERT INTO friends (user_id, friend_user_id) VALUES (2, 1);

For the retrieval of the USER 1 friends, we simply query this:

SELECT * FROM friends WHERE user_id = 1

Is it a bad idea to make redundant rows when friending users?

What I am trying to avoid here is not to store the friendship in a single row and keep guessing in every query which column the user is in:

SELECT * FROM friends WHERE user_id = 1 OR friend_user_id = 1

The latter made my application logic very complicated in some cases.