https://github.com/lostincompilation/hashme
Hardware accelerated hash library for C++ (SIMD intrinsics for ARM and x86)
https://github.com/lostincompilation/hashme
arm64 crypto hash hashing md5 md5-hash sha-hashes sha1 sha2 sha256 sha256-hash sha256sum sha3 sha512 simd simd-intrinsics
Last synced: 11 months ago
JSON representation
Hardware accelerated hash library for C++ (SIMD intrinsics for ARM and x86)
- Host: GitHub
- URL: https://github.com/lostincompilation/hashme
- Owner: LostInCompilation
- License: zlib
- Created: 2024-06-26T00:05:04.000Z (over 1 year ago)
- Default Branch: main
- Last Pushed: 2024-07-23T08:30:42.000Z (over 1 year ago)
- Last Synced: 2025-04-12T22:56:48.542Z (11 months ago)
- Topics: arm64, crypto, hash, hashing, md5, md5-hash, sha-hashes, sha1, sha2, sha256, sha256-hash, sha256sum, sha3, sha512, simd, simd-intrinsics
- Language: C++
- Homepage:
- Size: 297 KB
- Stars: 2
- Watchers: 2
- Forks: 1
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE.md
Awesome Lists containing this project
README
# HashMe - Hardware accelerated hash library for C++ (SIMD intrinsics)
   
Hardware acceleration for
## Contents
- [Description](#description)
- [Supported hash algorithms](#supported-hash-algorithms)
- [Additional algorithms](#additional-algorithms)
- [Performance and Benchmarks](#performance-and-benchmarks)
- [ARMv8 (M1 Pro AppleSilicon)](#armv8-m1-pro-applesilicon)
- [x86_64 (Intel)](#x86-64-intel)
- [Getting the library](#getting-the-library)
- [Compiling](#compiling)
- [CMake](#cmake)
- [Visual Studio](#visual-studio)
- [Xcode](#xcode)
- [How to use it](#how-to-use-it)
- [Hardware acceleration](#hardware-acceleration)
- [Included examples](#included-examples)
*See also: [License (zlib)](LICENSE.md)*
## Description
**Library is currently WIP! Will be updated frequently.**
A simple to use, fast and modern C++20 Hash library which supports **hardware accelerated** algorithms (SIMD intrinsics) for **x86 and ARMv8 (like AppleSilicon)**. See [Performance and Benchmarks](#performance-and-benchmarks) for speeds. The static library provides very easy integration in your projects and ultra high speeds for the most common hash algorithms like **MD5, MD6, SHA0, SHA1, SHA2 (SHA224, SHA256, SHA384, SHA512), SHA3 (SHA3-224, SHA3-256, SHA3-384, SHA3-512), CRC16, CRC32, CRC64, CRC128** and others. See [Supported hash algorithms](#supported-hash-algorithms) for details.
CMake build environment is supported, as well as generating projects for VisualStudio or Xcode.
TODO WRITE: STREAMING FOR LARGE FILES
### Supported hash algorithms
#### CRC
Algorithm | Supported | Hardware acceleration (SIMD) |
| -- | --------- | --------- |
| CRC16 | ✅ Yes | ❌ ARM \| ❌ x86 |
| CRC32 | ✅ Yes | ✅ ARM \| ❌ x86 |
| CRC64 (ECMA) | ✅ Yes | ❌ ARM \| ❌ x86 |
#### MD
Algorithm | Supported | Hardware acceleration (SIMD) |
| -- | --------- | --------- |
| MD5 | ✅ Yes | ❌ ARM \| ❌ x86 |
| MD6 | ❌ WIP | ❌ ARM \| ❌ x86 |
#### SHA
Algorithm | Supported | Hardware acceleration (SIMD) |
| -- | --------- | --------- |
| SHA1 | ❌ WIP | ❌ ARM \| ❌ x86 |
| SHA224 | ✅ Yes | ✅ ARM \| ⚠️ x86 |
| SHA256 | ✅ Yes | ✅ ARM \| ⚠️ x86 |
| SHA384 | ✅ Yes | ✅ ARM \| ❌ x86 |
| SHA512 | ✅ Yes | ✅ ARM \| ❌ x86 |
| SHA512/224 | ❌ WIP | ❌ ARM \| ❌ x86 |
| SHA512/256 | ❌ WIP | ❌ ARM \| ❌ x86 |
| SHA3-512 | ❌ WIP | ❌ ARM \| ❌ x86 |
*⚠️ = Work in progress*
*❌ = Not yet supported*
---
### Additional algorithms
Algorithm | Supported | Hardware acceleration |
| -- | --------- | --------- |
| Base64 | ❌ WIP | ❌ ARM \| ❌ x86 |
*⚠️ = Work in progress*
*❌ = Not yet supported*
---
### Performance and Benchmarks
#### ARMv8 (M1 Pro AppleSilicon)
*Apple Clang, Release build, -O3 optimization*
#### CRC
Algorithm | Speed (MB/s)
---------------|--------:
CRC16 Software | 375.54 MB/s
CRC16 Hardware (SIMD) | xxx MB/s
CRC32 Software | 376.49 MB/s
CRC32 Hardware (SIMD) | 8031.77 MB/s
CRC64 Software | 378.56 MB/s
CRC64 Hardware (SIMD) | xxx MB/s
#### MD
Algorithm | Speed (MB/s)
---------------|--------:
MD5 Software | 538.12 MB/s
MD5 Hardware (SIMD) | xxx MB/s
#### SHA
Algorithm | Speed (MB/s)
---------------|--------:
SHA224 Software | 218.50 MB/s
SHA224 Hardware (SIMD) | 1944.76 MB/s
SHA256 Software | 217.76 MB/s
SHA256 Hardware (SIMD) | 1944.79 MB/s
SHA384 Software | 317.07 MB/s
SHA384 Hardware (SIMD) | xxx MB/s
SHA512 Software | 317.10 MB/s
SHA512 Hardware (SIMD) | 1337.49 MB/s
#### x86_64 (Intel)
**TODO**
## Getting the library
To download the library simply clone the repository:
~~~
git clone https://github.com/LostInCompilation/HashMe
~~~
## Compiling
Compilation of the library and examples can be done using CMake and make. Additionally projects for VisualStudio and Xcode can be generated. Small Shell scripts for compiling and generating projects are included.
### CMake
TODO
### Visual Studio
TODO
### Xcode
TODO
## How to use it
Include the main header file of the library in your project:
```cpp
#include "HashMe.hpp"
```
LINKING
If you prefer, directly use the namespace for the library:
```cpp
using namespace HashMe;
```
A shorter namespace alias ```HM```is also available:
```cpp
using namespace HM;
```
TODO
### Hardware acceleration
TODO
## Included Examples
TODO