439   dim3 dimBlock(iNumThreadsPerBlock); /* (iNumThreadsPerBlock,1,1) */  in runStream()  local
440   dim3 dimGrid(N / dimBlock.x);       /* (N/dimBlock.x,1,1) */  in runStream()
441   if (N % dimBlock.x != 0) dimGrid.x += 1;  in runStream()
444   set_array<<<dimGrid, dimBlock>>>(d_a, 2.f, N);  in runStream()
445   set_array<<<dimGrid, dimBlock>>>(d_b, .5f, N);  in runStream()
446   set_array<<<dimGrid, dimBlock>>>(d_c, .5f, N);  in runStream()
454     STREAM_Copy<<<dimGrid, dimBlock>>>(d_a, d_c, N);  in runStream()
462     STREAM_Copy_Optimized<<<dimGrid, dimBlock>>>(d_a, d_c, N);  in runStream()
471     STREAM_Scale<<<dimGrid, dimBlock>>>(d_b, d_c, scalar, N);  in runStream()
480     STREAM_Scale_Optimized<<<dimGrid, dimBlock>>>(d_b, d_c, scalar, N);  in runStream()
490     STREAM_Add<<<dimGrid, dimBlock>>>(d_a, d_b, d_c, N);  in runStream()
504     STREAM_Add_Optimized<<<dimGrid, dimBlock>>>(d_a, d_b, d_c, N);  in runStream()
513     STREAM_Triad<<<dimGrid, dimBlock>>>(d_b, d_c, d_a, scalar, N);  in runStream()
522     STREAM_Triad_Optimized<<<dimGrid, dimBlock>>>(d_b, d_c, d_a, scalar, N);  in runStream()
554     set_array<<<dimGrid, dimBlock>>>(d_a, 2.f, N);  in runStream()
555     set_array<<<dimGrid, dimBlock>>>(d_b, .5f, N);  in runStream()
556     set_array<<<dimGrid, dimBlock>>>(d_c, .5f, N);  in runStream()
558     STREAM_Copy<<<dimGrid, dimBlock>>>(d_a, d_c, N);  in runStream()
568     set_array<<<dimGrid, dimBlock>>>(d_a, 2.f, N);  in runStream()
569     set_array<<<dimGrid, dimBlock>>>(d_b, .5f, N);  in runStream()
570     set_array<<<dimGrid, dimBlock>>>(d_c, .5f, N);  in runStream()
572     STREAM_Copy_Optimized<<<dimGrid, dimBlock>>>(d_a, d_c, N);  in runStream()
582     set_array<<<dimGrid, dimBlock>>>(d_a, 2.f, N);  in runStream()
583     set_array<<<dimGrid, dimBlock>>>(d_b, .5f, N);  in runStream()
584     set_array<<<dimGrid, dimBlock>>>(d_c, .5f, N);  in runStream()
586     STREAM_Scale<<<dimGrid, dimBlock>>>(d_b, d_c, scalar, N);  in runStream()
596     set_array<<<dimGrid, dimBlock>>>(d_a, 2.f, N);  in runStream()
597     set_array<<<dimGrid, dimBlock>>>(d_b, .5f, N);  in runStream()
598     set_array<<<dimGrid, dimBlock>>>(d_c, .5f, N);  in runStream()
600     STREAM_Add<<<dimGrid, dimBlock>>>(d_a, d_b, d_c, N);  in runStream()
611     set_array<<<dimGrid, dimBlock>>>(d_a, 2.f, N);  in runStream()
612     set_array<<<dimGrid, dimBlock>>>(d_b, .5f, N);  in runStream()
613     set_array<<<dimGrid, dimBlock>>>(d_c, .5f, N);  in runStream()
615     STREAM_Triad<<<dimGrid, dimBlock>>>(d_b, d_c, d_a, scalar, N);  in runStream()
654   dim3 dimBlock(iNumThreadsPerBlock); /* (iNumThreadsPerBlock,1,1) */  in runStreamDouble()  local
655   dim3 dimGrid(N / dimBlock.x);       /* (N/dimBlock.x,1,1) */  in runStreamDouble()
656   if (N % dimBlock.x != 0) dimGrid.x += 1;  in runStreamDouble()
659   set_array_double<<<dimGrid, dimBlock>>>(d_a, 2., N);  in runStreamDouble()
660   set_array_double<<<dimGrid, dimBlock>>>(d_b, .5, N);  in runStreamDouble()
661   set_array_double<<<dimGrid, dimBlock>>>(d_c, .5, N);  in runStreamDouble()
669     STREAM_Copy_double<<<dimGrid, dimBlock>>>(d_a, d_c, N);  in runStreamDouble()
680     STREAM_Copy_Optimized_double<<<dimGrid, dimBlock>>>(d_a, d_c, N);  in runStreamDouble()
691     STREAM_Scale_double<<<dimGrid, dimBlock>>>(d_b, d_c, scalar, N);  in runStreamDouble()
700     STREAM_Scale_Optimized_double<<<dimGrid, dimBlock>>>(d_b, d_c, scalar, N);  in runStreamDouble()
709     STREAM_Add_double<<<dimGrid, dimBlock>>>(d_a, d_b, d_c, N);  in runStreamDouble()
718     STREAM_Add_Optimized_double<<<dimGrid, dimBlock>>>(d_a, d_b, d_c, N);  in runStreamDouble()
727     STREAM_Triad_double<<<dimGrid, dimBlock>>>(d_b, d_c, d_a, scalar, N);  in runStreamDouble()
736     STREAM_Triad_Optimized_double<<<dimGrid, dimBlock>>>(d_b, d_c, d_a, scalar, N);  in runStreamDouble()
768     set_array_double<<<dimGrid, dimBlock>>>(d_a, 2., N);  in runStreamDouble()
769     set_array_double<<<dimGrid, dimBlock>>>(d_b, .5, N);  in runStreamDouble()
770     set_array_double<<<dimGrid, dimBlock>>>(d_c, .5, N);  in runStreamDouble()
772     STREAM_Copy_double<<<dimGrid, dimBlock>>>(d_a, d_c, N);  in runStreamDouble()
782     set_array_double<<<dimGrid, dimBlock>>>(d_a, 2., N);  in runStreamDouble()
783     set_array_double<<<dimGrid, dimBlock>>>(d_b, .5, N);  in runStreamDouble()
784     set_array_double<<<dimGrid, dimBlock>>>(d_c, .5, N);  in runStreamDouble()
786     STREAM_Copy_Optimized_double<<<dimGrid, dimBlock>>>(d_a, d_c, N);  in runStreamDouble()
796     set_array_double<<<dimGrid, dimBlock>>>(d_b, .5, N);  in runStreamDouble()
797     set_array_double<<<dimGrid, dimBlock>>>(d_c, .5, N);  in runStreamDouble()
799     STREAM_Scale_double<<<dimGrid, dimBlock>>>(d_b, d_c, scalar, N);  in runStreamDouble()
809     set_array_double<<<dimGrid, dimBlock>>>(d_a, 2., N);  in runStreamDouble()
810     set_array_double<<<dimGrid, dimBlock>>>(d_b, .5, N);  in runStreamDouble()
811     set_array_double<<<dimGrid, dimBlock>>>(d_c, .5, N);  in runStreamDouble()
813     STREAM_Add_double<<<dimGrid, dimBlock>>>(d_a, d_b, d_c, N);  in runStreamDouble()
824     set_array_double<<<dimGrid, dimBlock>>>(d_a, 2., N);  in runStreamDouble()
825     set_array_double<<<dimGrid, dimBlock>>>(d_b, .5, N);  in runStreamDouble()
826     set_array_double<<<dimGrid, dimBlock>>>(d_c, .5, N);  in runStreamDouble()
828     STREAM_Triad_double<<<dimGrid, dimBlock>>>(d_b, d_c, d_a, scalar, N);  in runStreamDouble()