Slide 4 of 8

Libcsmath R2.8 enhancements

• C = A*B case where the number of columns of C (n) is small.

• All DGEMM cases where K is small

• K=24 GEMM cases

• Kᡸ GEMM cases

• GEMM cases where the number of rows (M) is 2.

• More prefetching done on columns of B

• Enhancements to other level-3 kernels

• Faster handling of smaller BLAS

Acknowledgement and Disclaimer