Something of an fs infra library (Intel AVX2 accelerated) for C++17

• intrinsics.h: Type conversion, Bit-manipulation, Multi-precision arithmetic helper
• cpuid.h: CPU feature flag loader
• uint128.h: Unsigned 128 bit integer and bitwise-, arithmetic-, and comparing- operators
• xmm.h: Typed SSE/AVX integer vector operation wraper library (vu32x4, vu16x16, etc)
• lutgen.h: Compile-time LUT generation helper
• hexilit.h: Compile-time hexadecimal-integer parser
• base64.h: Compile-time base64 decoder

• camellia-ref.h: Reference implementation
• camellia-avx2.h: AVX2 LUT accelerated implementation (approx. 2x faster than ref-impl)
• camellia-avx2aesni.h: AVX2-AESNI accelerated implementation (based on "Block Ciphers: Fast Implementations on x86-64 Architecture" -- Oulu : J. Kivilinna, 2013) (approx. 6x faster than ref-impl)

• crc32-ref.h: Reference implementation
• crc32-ia32.h: IA32 loop-unrolling implementation (approx. 6x faster than ref-impl)
• crc32-avx2.h: AVX2 LUT accelerated implementation (approx. 7x faster than ref-impl)
• crc32-avx2clmul.h: pclmul accelerated implementation (based on "Fast CRC Computation for Generic Polynomials Using PCLMULQDQ Instruction" -- V. Gopal, E. Ozturk, J. Guilford, et al., 2009) (approx. 20x faster than ref-impl)

• sha2-ref.h: Reference implementation
• sha2-avx2.h: AVX2 accelerated implementation (based on "Fast SHA-256 Implementations on Intel® Architecture Processors" -- J. Guilford, K. Yap, V. Gopal, 2012)

• Tests for arkana.lib (using googletest)

1. Include arkana.lib.vcxproj into your solution.
2. Reference it from your project.

Do cmake.

MIT License Copyright (c) 2020-2022 ttsuki