An implementation of the Lyra2 key derivation function targeting modern CPU architectures and compilers. This is an alternative to the reference implementation aiming at:
- Helping debug the specification with an independent implementation;
- Exploring the speedups provided by SIMD instruction sets;
- Being presented as my final graduation project at university ;)
Just type:
$ make
and you should have a binary named lyra2 in the current directory. When
executed, that binary will run some performance tests (see src/main.c) with
different parameters for the same input.
The ref directory contains an updated reference implementation for comparison
purposes. Use the benchmark.py script to build both implementations and
compare their speeds:
$ ./scripts/benchmark.py
This will perform the same tests as above for both this and the reference implementation and output their relative speed.