simd_vector

simd_vector::Vector<T> represents an immutable vector with a size that's automatically determined by your CPU architecture and is designed as a building block for vectorizing large algorithms.

Features

• 📦 Built on Rust's portable_simd
• 🚀 Automatic selection of optimal SIMD lane count based on hardware architecture
• 🔒 Type-safe operations with compile-time size checking
• ⚡ Zero-cost abstractions over hardware SIMD instructions
• 💪 Support for common numeric types (f32, f64, i8, i16, i32, i64, u8, u16, u32, u64)
• 🔄 Cross-platform support (x86_64, aarch64)

Installation

Add this to your Cargo.toml:

[dependencies]
simd_vector = "0.1.0"

⚠️ Note: This crate requires a nightly Rust compiler due to the use of portable_simd.

rustup override set nightly  # Set nightly for this project
# or
rustup default nightly      # Set nightly as your default toolchain

Usage

use simd_vector::Vector;

// Create vectors
let a = Vector::<f32>::new(1.0);
let b = Vector::<f32>::new(2.0);

// Basic arithmetic
let sum = a + b;
let diff = a - b;
let product = a * b;
let quotient = a / b;

// Reduction operations
let total = sum.reduce_sum();

// Element-wise operations
let abs_values = a.abs();
let min_values = Vector::min(a, b);
let max_values = Vector::max(a, b);

Lane Counts

The number of lanes in a vector depends on both the data type and the target architecture.

Type	x86_64 (256-bit AVX)	aarch64 (128-bit NEON)	Other (128-bit)
f32	8 lanes	4 lanes	4 lanes
f64	4 lanes	2 lanes	2 lanes
i8	32 lanes	16 lanes	16 lanes
i16	16 lanes	8 lanes	8 lanes
i32	8 lanes	4 lanes	4 lanes
i64	4 lanes	2 lanes	2 lanes
u8	32 lanes	16 lanes	16 lanes
u16	16 lanes	8 lanes	8 lanes
u32	8 lanes	4 lanes	4 lanes
u64	4 lanes	2 lanes	2 lanes

Contributing

1. Fork the Repository: Start by forking the repository to your own GitHub account.
2. Clone the Forked Repository: Clone the fork to your local machine.
3. Create a New Branch: Always create a new branch for your changes.
4. Make Your Changes: Implement your changes.
5. Run Tests: Make sure to test your changes locally.
6. Submit a Pull Request: Commit and push your changes, then create a pull request against the main branch.