What method do Unix-type systems use to prevent attackers from cracking passwords using pre- computed hashes?

Enforcing password complexity policies prevents weak password choices but does not technically defeat pre-computed hash attacks against whatever password is chosen.

CPU-intensive algorithms like bcrypt slow down brute-force and dictionary attacks but the question specifically asks about pre-computed hash attacks, which salting - not compute cost - defeats.

A salt is a randomly generated value stored alongside the hash in /etc/shadow. By concatenating the salt with the plaintext password before passing it to the hash function, each user's hash is unique even if two users share the same password. This invalidates pre-computed hash databases (rainbow tables) because an attacker would need to recompute hashes for every possible salt value.

One-way (irreversible) hashing is a fundamental property of all cryptographic hash functions and does not prevent an attacker from comparing stored hashes against a pre-computed rainbow table.

Unix password storage uses one-way hashing, not symmetric encryption; symmetric encryption is reversible and would be less secure for password storage than hashing with a salt.