matrixmultiply_mt 0.2.1

Multithreaded Fork of bluss's matrixmultiply crate. General matrix multiplication of f32 and f64 matrices in Rust. Supports matrices with general strides. Uses a microkernel strategy, so that the implementation is easy to parallelize and optimize. `RUSTFLAGS="-C target-cpu=native"` is your friend here.
Documentation 
1
2
3
4
5
6
7


test mat_mul_f32::mix10000         ... bench:  23,430,259 ns/iter (+/- 700,947)
test mat_mul_f64::mix10000         ... bench:  36,255,323 ns/iter (+/- 1,344,432)
test mat_mul_f32::m127             ... bench:     290,740 ns/iter (+/- 6,776)
test mat_mul_f64::m127             ... bench:     433,429 ns/iter (+/- 17,110)
test nonative_mat_mul_f32::m127             ... bench:     375,725 ns/iter (+/- 7,761)
test nonative_mat_mul_f64::m127             ... bench:     503,924 ns/iter (+/- 16,038)