Crate simdeez[][src]

A library that abstracts over SIMD instruction sets, including ones with differing widths. SIMDeez is designed to allow you to write a function one time and produce scalar, SSE2, SSE41, and AVX2 versions of the function. You can either have the version you want selected automatically at runtime, at compiletime, or select yourself by hand.

SIMDeez is currently in Beta, if there are intrinsics you need that are not currently implemented, create an issue and I’ll add them. PRs to add more intrinsics are welcome. Currently things are well fleshed out for i32, i64, f32, and f64 types.

As Rust stabilizes support for Neon and AVX-512 I plan to add those as well.

Refer to the excellent Intel Intrinsics Guide for documentation on these functions.


  • SSE2, SSE41, and AVX2 and scalar fallback
  • Can be used with compile time or run time selection
  • No runtime overhead
  • Uses familiar intel intrinsic naming conventions, easy to port.
    • _mm_add_ps(a,b) becomes add_ps(a,b)
  • Fills in missing intrinsics in older APIs with fast SIMD workarounds.
    • ceil, floor, round,blend, etc
  • Can be used by #[no_std] projects
  • Operator overloading: let sum = va + vb or s *= s
  • Extract or set a single lane with the index operator: let v1 = v[1];

Trig Functions via Sleef-sys

A number of trigonometric and other common math functions are provided in vectorized form via the Sleef-sys crate. This is an optional feature sleef that you can enable. Doing so currently requires nightly, as well as having CMake and Clang installed.

Compared to stdsimd

  • SIMDeez can abstract over differing simd widths. stdsimd does not
  • SIMDeez builds on stable rust now, stdsimd does not

Compared to Faster

  • SIMDeez can be used with runtime selection, Faster cannot.
  • SIMDeez has faster fallbacks for some functions
  • SIMDeez does not currently work with iterators, Faster does.
  • SIMDeez uses more idiomatic intrinsic syntax while Faster uses more idomatic Rust syntax
  • SIMDeez can be used by #[no_std] projects
  • SIMDeez builds on stable rust now, Faster does not.

All of the above could change! Faster seems to generally have the same performance as long as you don’t run into some of the slower fallback functions.


    use simdeez::*;
    use simdeez::scalar::*;
    use simdeez::sse2::*;
    use simdeez::sse41::*;
    use simdeez::avx2::*;
    // If you want your SIMD function to use use runtime feature detection to call
    // the fastest available version, use the simd_runtime_generate macro:
    fn distance(
        x1: &[f32],
        y1: &[f32],
        x2: &[f32],
        y2: &[f32]) -> Vec<f32> {
        let mut result: Vec<f32> = Vec::with_capacity(x1.len());
        result.set_len(x1.len()); // for efficiency
        /// Set each slice to the same length for iteration efficiency
        let mut x1 = &x1[..x1.len()];
        let mut y1 = &y1[..x1.len()];
        let mut x2 = &x2[..x1.len()];
        let mut y2 = &y2[..x1.len()];
        let mut res = &mut result[..x1.len()];
        // Operations have to be done in terms of the vector width
        // so that it will work with any size vector.
        // the width of a vector type is provided as a constant
        // so the compiler is free to optimize it more.
        // S::VF32_WIDTH is a constant, 4 when using SSE, 8 when using AVX2, etc
        while x1.len() >= S::VF32_WIDTH {
            //load data from your vec into an SIMD value
            let xv1 = S::loadu_ps(&x1[0]);
            let yv1 = S::loadu_ps(&y1[0]);
            let xv2 = S::loadu_ps(&x2[0]);
            let yv2 = S::loadu_ps(&y2[0]);
            // Use the usual intrinsic syntax if you prefer
            let mut xdiff = S::sub_ps(xv1, xv2);
            // Or use operater overloading if you like
            let mut ydiff = yv1 - yv2;
            xdiff *= xdiff;
            ydiff *= ydiff;
            let distance = S::sqrt_ps(xdiff + ydiff);
            // Store the SIMD value into the result vec
            S::storeu_ps(&mut res[0], distance);
            // Move each slice to the next position
            x1 = &x1[S::VF32_WIDTH..];
            y1 = &y1[S::VF32_WIDTH..];
            x2 = &x2[S::VF32_WIDTH..];
            y2 = &y2[S::VF32_WIDTH..];
            res = &mut res[S::VF32_WIDTH..];
        // (Optional) Compute the remaining elements. Not necessary if you are sure the length
        // of your data is always a multiple of the maximum S::VF32_WIDTH you compile for (4 for SSE, 8 for AVX2, etc).
        // This can be asserted by putting `assert_eq!(x1.len(), 0);` here
        for i in 0..x1.len() {
            let mut xdiff = x1[i] - x2[i];
            let mut ydiff = y1[i] - y2[i];
            xdiff *= xdiff;
            ydiff *= ydiff;
            let distance = (xdiff + ydiff).sqrt();
            res[i] = distance;

This will generate 5 functions for you:

  • distance<S:Simd> the generic version of your function
  • distance_scalar a scalar fallback
  • distance_sse2 SSE2 version
  • distance_sse41 SSE41 version
  • distance_avx AVX version
  • distance_avx2 AVX2 version
  • distance_runtime_select picks the fastest of the above at runtime

You can use any of these you wish, though typically you would use the runtime_select version unless you want to force an older instruction set to avoid throttling or for other arcane reasons.

Optionally you can use the simd_compiletime_generate! macro in the same way. This will produce 2 active functions via the cfg attribute feature:

  • distance<S:Simd> the generic version of your function
  • distance_compiletime the fastest instruction set availble for the given compile time feature set

You may also forgoe the macros if you know what you are doing, just keep in mind there are lots of arcane subtleties with inlining and target_features that must be managed. See how the macros expand for more detail.


pub extern crate paste;



libm in pure Rust




Generates a generic version of your function (fn_name) And the fastest version supported by your rust compilation settings (fn_name_compiletime)


Generates a generic version of your function (fn_name), and versions for:



The abstract SIMD trait which is implemented by Avx2, Sse41, etc


Grouping all the constraints shared by associated types in the Simd trait into this marker trait drastically reduces compile time.


f32 and f64 share these constraints, grouping them here speeds up compile times considerably


16 and 32 bit int types share all of these constraints, grouping them here speeds up compile times considerably