Hi,
It is unclear what you are asking. Are you talking about the default hash code generation in AnyRef (aka Object)? Or are you talking about the hash code functions generated for case classes?
For case classes, the current implementation of Scala uses Murmur Hash 3. Although you should not rely on it, since it could change in future implementations.
For AnyRef.hashCode, that one is platform-dependent. On the JVM, it is whatever the JVM decides to use (e.g., Park-Miller, as you mentioned). On JavaScript, the current implementation always uses sequential numbers, although that could change in the future. In any case, you cannot rely on it.