pub const GEMM_BLOCK_ROWS: usize = 256;Expand description
Rows of C and A handled per block in NaiveDense::gemm_sub and
NaiveDense::syrk_sub (a cache-blocking size, not a numerical
threshold): GEMM_BLOCK_ROWS × GEMM_BLOCK_K entries of A — 128 KiB
in f64 — stay in cache while the columns of C pass over them.