| /* 2D */ | |
| KOKKOS_INLINE_FUNCTION | |
| void index2coord(int index, int &i, int &j, int Nx, int Ny) | |
| { | |
| UNUSED(Nx); | |
| UNUSED(Ny); | |
| #ifdef KOKKOS_ENABLE_CUDA | |
| j = index / Nx; |
This is a short post that explains how to write a high-performance matrix multiplication program on modern processors. In this tutorial I will use a single core of the Skylake-client CPU with AVX2, but the principles in this post also apply to other processors with different instruction sets (such as AVX512).
Matrix multiplication is a mathematical operation that defines the product of
| #!/bin/bash | |
| tmpfile=".tmp.pdf" | |
| if [ $# -ge 1 ]; then | |
| if [ $1 != "-h" ]; then | |
| outfile=`dirname $1`/`basename "$1" \.agr`.pdf | |
| LANG=en xmgrace -hardcopy -hdevice PDF $1 -printfile $tmpfile | |
| pdfcrop $tmpfile $outfile | |
| rm -f $tmpfile | |
| exit | |
| fi |
