Advanced Computing Platform for Theoretical Physics

Commit 82646005 authored by mikeaclark's avatar mikeaclark
Browse files

Added initial blas benchmarking

git-svn-id: http://lattice.bu.edu/qcdalg/cuda/quda@487 be54200a-260c-0410-bdd7-ce6af2a381ab
parent 06609409
......@@ -1586,7 +1586,7 @@ template <typename Float, typename Float2>
#undef REDUCE_REAL_OPERATION
#undef REDUCE_IMAG_OPERATION
double2 xpayDotzyCuda(ParitySpinor x, double a, ParitySpinor y, ParitySpinor z) {
double2 xpaycDotzyCuda(ParitySpinor x, double a, ParitySpinor y, ParitySpinor z) {
blas_quda_flops += 6*x.length;
checkSpinor(x,y);
checkSpinor(x,z);
......
......@@ -82,7 +82,7 @@ double benchmark(int kernel) {
break;
case 4:
//mxpyQuda(x, y);
mxpyCuda(x, y);
break;
case 5:
......@@ -136,7 +136,7 @@ double benchmark(int kernel) {
break;
case 17:
//xpayDotzyCuda(x, a, y, z);
xpayDotzyCuda(x, a, y, z);
break;
// double3
......@@ -166,7 +166,7 @@ double benchmark(int kernel) {
int main(int argc, char** argv) {
init();
int kernels[] = {0, 1, 2, 3, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 18, 19, 20};
int kernels[] = {0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20};
nIters = 1;
// first do warmup run
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment