KernelGen - GPU Technology Conference
KernelGen - GPU Technology Conference
KernelGen - GPU Technology Conference
- No tags were found...
You also want an ePaper? Increase the reach of your titles
YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.
<strong>KernelGen</strong>: smart pointers tracking.Pointer alias analysis is performed in runtime, assisted with addresses substitution:sincos.cvoid sincos(int nx, int ny, int nz, float* x, float* y, float* xy) {#pragma acc parallelfor (int k = 0; k < nz; k++)for (int j = 0; j < ny; j++)for (int i = 0; i < nx; i++) {int idx = i + nx * j + nx * ny * k;xy[idx] = sin(x[idx]) + cos(y[idx]);}}resultLaunching kernel __kernelgen_sincos_loop_10blockDim = { 32, 4, 4 }gridDim = { 16, 128, 16 }Finishing kernel __kernelgen_sincos_loop_10__kernelgen_sincos_loop_10 time = 2.300006e-02 sec15 / 75