Function BM_Sum3DYReduceGPU

tensorflow/core/kernels/reduction_ops_test.cc:152–154

BENCHMARK(BM_Sum2DColumnReduceGPU)->RangePair(1, 8192, 1, 8192);

static void BM_Sum3DYReduceGPU(int iters, int num_x, int num_y) {
  Do3DYReduce(iters, "gpu", "Sum", num_x, num_y);
}
BENCHMARK(BM_Sum3DYReduceGPU)->RangePair(64, 4096, 64, 4096);

static void BM_Sum3DXZReduceGPU(int iters, int num_x, int num_y) {