GPU Kernel Information Aggregated by Name
| kernel_name | kernel_count | kernel_duration (us) | model_duration_percentage | kernel_flops | kernel_dram_read_bytes | kernel_dram_write_bytes | kernel_achieved_occupancy (%) | kernel_arithmetic_intensity (flops/byte) | kernel_arithmetic_throughput (GFlops) | kernel_memory_bound |
|---|
kernel_name | kernel_count | kernel_duration (us) | model_duration_percentage | kernel_flops | kernel_dram_read_bytes | kernel_dram_write_bytes | kernel_achieved_occupancy (%) | kernel_arithmetic_intensity (flops/byte) | kernel_arithmetic_throughput (GFlops) | kernel_memory_bound |
|---|---|---|---|---|---|---|---|---|---|---|
| cudnn::gemm::computeOffsetsKernel(cudnn::gemm::ComputeOffsetsParams) | 8 | 19.67 | 0.31 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true |
| void conv2d_c1_k1_nchw_hw_packed_kernel<float, float, 3>(cudnnTensorStruct, float const*, cudnnFilterStruct, float const*, cudnnConvolutionStruct, cudnnTensorStruct, float*, float, float, cudnn::reduced_divisor, cudnn::reduced_divisor, int) | 3 | 44.00 | 0.70 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true |
| void conv2d_grouped_direct_kernel<float, float, float, float, float, true, false, 0, 1, 3>(cudnnTensorStruct, float const*, cudnnFilterStruct, float const*, cudnnConvolutionStruct, cudnnTensorStruct, float*, float, float, cudnn::reduced_divisor, cudnn::reduced_divisor, cudnn::reduced_divisor, cudnn::reduced_divisor, cudnn::reduced_divisor, int, float const*, float const*, cudnnActivationStruct) | 12 | 67.67 | 1.08 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true |
| void cudnn::detail::bn_fw_inf_1C11_kernel_NCHW<float, float, true, 1>(float, float, cudnnTensorStruct, float const*, cudnnTensorStruct, float*, cudnnTensorStruct, float const*, float const*, float const*, float const*, float) | 52 | 204.66 | 3.27 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true |
| void cudnn::detail::implicit_convolve_sgemm<float, float, 1024, 5, 5, 3, 3, 3, 1, true, false, true>(int, int, int, float const*, int, float*, float*, kernel_conv_params, int, float, float, int, float*, float*, int, int) | 26 | 539.00 | 8.61 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true |
| void cudnn::detail::implicit_convolve_sgemm<float, float, 128, 5, 5, 3, 3, 3, 1, true, false, true>(int, int, int, float const*, int, float*, float*, kernel_conv_params, int, float, float, int, float*, float*, int, int) | 0 | 12.00 | 0.19 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true |
| void cudnn::detail::pooling_fw_4d_kernel<float, float, cudnn::detail::averpooling_func<float>, 1, false>(cudnnTensorStruct, float const*, cudnnTensorStruct, float*, cudnnPoolingStruct, float, float, int, cudnn::reduced_divisor, cudnn::reduced_divisor) | 0 | 6.33 | 0.10 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true |
| void mxnet::op::mxnet_op::mxnet_generic_kernel<mxnet::op::clip, float*, float*, float, float>(int, float*, float*, float, float) | 35 | 103.00 | 1.65 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true |
| void mxnet::op::mxnet_op::mxnet_generic_kernel<mxnet::op::mxnet_op::op_with_req<mxnet::op::mshadow_op::plus, 1>, float*, float*, float*>(int, float*, float*, float*) | 9 | 21.67 | 0.35 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true |
| volta_scudnn_128x32_relu_interior_nn_v1 | 1 | 26.00 | 0.42 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true |
| volta_scudnn_128x32_relu_small_nn_v1 | 0 | 14.00 | 0.22 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true |
| volta_scudnn_128x64_relu_interior_nn_v1 | 5 | 68.67 | 1.10 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true |
Showing 1 to 12 of 12 entries