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

name                        4x4-dgemm.log ns/iter  8x4-dgemm.log ns/iter    diff ns/iter   diff %
mat_mul_f64::m016           1,254                  1,103                            -151  -12.04%
mat_mul_f64::m064           47,573                 38,887                         -8,686  -18.26%
mat_mul_f64::m127           344,298                288,225                       -56,073  -16.29%
nonative_mat_mul_f64::m127  473,111                549,677                        76,566   16.18%