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) | 0 | 2.67 | 0.05 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true |
| void cudnn::detail::pooling_fw_4d_kernel<float, float, cudnn::detail::maxpooling_func<float, (cudnnNanPropagation_t)0>, 0, false>(cudnnTensorStruct, float const*, cudnnTensorStruct, float*, cudnnPoolingStruct, float, float, int, cudnn::reduced_divisor, cudnn::reduced_divisor) | 4 | 60.33 | 1.06 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true |
| void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 11 | 334.00 | 5.86 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true |
| void gemv2T_kernel_val<int, int, float, float, float, 128, 16, 4, 4, false, cublasGemvParams<cublasGemvTensorStridedBatched<float const>, cublasGemvTensorStridedBatched<float>, float> >(cublasGemvParams<cublasGemvTensorStridedBatched<float const>, cublasGemvTensorStridedBatched<float>, float>, float, float) | 1 | 150.33 | 2.64 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true |
| void gemv2T_kernel_val<long, long, float, float, float, 128, 16, 2, 2, false, cublasGemvParams<cublasGemvTensorStridedBatched<float const>, cublasGemvTensorStridedBatched<float>, float> >(cublasGemvParams<cublasGemvTensorStridedBatched<float const>, cublasGemvTensorStridedBatched<float>, float>, float, float) | 0 | 703.33 | 12.34 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true |
| void mshadow::cuda::MapPlanKernel<mshadow::sv::plusto, 8, mshadow::expr::Plan<mshadow::Tensor<mshadow::gpu, 2, float>, float>, mshadow::expr::Plan<mshadow::expr::Broadcast1DExp<mshadow::Tensor<mshadow::gpu, 1, float>, float, 2, 1>, float> >(mshadow::expr::Plan<mshadow::Tensor<mshadow::gpu, 2, float>, float>, int, mshadow::Shape<2>, mshadow::expr::Plan<mshadow::expr::Broadcast1DExp<mshadow::Tensor<mshadow::gpu, 1, float>, float, 2, 1>, float>) | 2 | 7.33 | 0.13 | 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::identity, 1>, float*, float*>(int, float*, float*) | 1 | 4.00 | 0.07 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true |
| void op_generic_tensor_kernel<2, float, float, float, 256, (cudnnGenericOp_t)0, (cudnnNanPropagation_t)0, (cudnnDimOrder_t)0, 0>(cudnnTensorStruct, float*, cudnnTensorStruct, float const*, cudnnTensorStruct, float const*, float, float, float, float, dimArray, reducedDivisorArray, bool) | 12 | 210.33 | 3.69 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true |
| void op_generic_tensor_kernel<2, float, float, float, 256, (cudnnGenericOp_t)8, (cudnnNanPropagation_t)0, (cudnnDimOrder_t)0, 1>(cudnnTensorStruct, float*, cudnnTensorStruct, float const*, cudnnTensorStruct, float const*, float, float, float, float, dimArray, reducedDivisorArray, bool) | 14 | 180.67 | 3.17 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true |
| volta_scudnn_128x32_relu_small_nn_v1 | 0 | 39.67 | 0.70 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true |
| volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 11 | 2382.33 | 41.78 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true |
Showing 1 to 11 of 11 entries