pub fn hash_strict(bytes: &[u8]) -> HashMatrixExpand description
This procedure implements the same hash function as hash(), but
with a different performance tradeoff. The first time it’s invoked,
hash computes a 4MiB table of all the hashes for every pair of
bytes, which are then used to double hashing speed. For applications
that need to do a lot of hashing, this is nearly twice as fast as
hash(), but it also requires much more memory and initialization
time. As a rule of thumb, if you’re going to hash less than 100KiB
during your program’s execution, you should probably use
hash_strict.