Function BM_Sum2DRowReduceGPU

tensorflow/core/kernels/reduction_ops_test.cc:142–144

BENCHMARK(BM_Sum2DToScalarGPUHalf)->RangePair(1, 8192, 1, 8192);

static void BM_Sum2DRowReduceGPU(int iters, int num_x, int num_y) {
  DoRowReduce(iters, "gpu", "Sum", num_x, num_y);
}
BENCHMARK(BM_Sum2DRowReduceGPU)->RangePair(1, 8192, 1, 8192);

static void BM_Sum2DColumnReduceGPU(int iters, int num_x, int num_y) {