Navigation :

GPU Kernel Information Aggregated by Name

Search:

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
compute_gemm_pointers(float2*, float2 const, int, float2 const, int, float2 const, int, int)	5	22.33	0.04	0	2176.00	78282.67	2.63	0.00	0.00	true
cudnn::gemm::computeOffsetsKernel(cudnn::gemm::ComputeOffsetsParams)	41	144.67	0.27	0	10218.67	91882.67	5.62	0.00	0.00	true
void cudnn::detail::pooling_fw_4d_kernel<float, float, cudnn::detail::averpooling_func<float>, 2, false>(cudnnTensorStruct, float const, cudnnTensorStruct, float, cudnnPoolingStruct, float, float, int, cudnn::reduced_divisor, cudnn::reduced_divisor)	0	81.33	0.15	9240832	54464576.00	3549568.00	63.00	0.16	113.62	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)	12	4928.00	9.24	181436416	633789152.00	521068042.67	66.94	0.16	36.82	true
void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>)	6	42.33	0.08	5374976	715946.67	4305408.00	13.37	1.07	126.97	true
void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long)	29	4177.67	7.83	0	926342730.67	997180938.67	96.39	0.00	0.00	true
void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 3, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorBroadcastingOp<Eigen::array<long, 3ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 3, 1, int>, 16, Eigen::MakePointer> const> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 3ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 3, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 3, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorBroadcastingOp<Eigen::array<long, 3ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 3, 1, int>, 16, Eigen::MakePointer> const> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 3ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 3, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int)	56	4344.33	8.14	412948480	1078396373.33	1150848000.00	48.68	0.19	95.05	true
void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorSlicingOp<Eigen::array<int, 2ul> const, Eigen::array<int, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer> >, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const, Eigen::GpuDevice>, int>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorSlicingOp<Eigen::array<int, 2ul> const, Eigen::array<int, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer> >, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const, Eigen::GpuDevice>, int)	35	1652.09	3.10	0	576977002.66	578189600.00	95.52	0.00	0.00	true
void fft2d_c2r_16x16<float, false>(float, float2, int, int, int, int, int, int, int, int, int, int, float, float, int, float, float)	5	573.00	1.07	842096640	209564949.33	157340352.00	69.36	2.30	1469.63	true
void fft2d_c2r_32x32<float, false, false, 0u, false, false>(float, float2 const, int, int, int, int, int, int, int, int, int, float, float, cudnn::reduced_divisor, bool, float, float, int2, int, int)	2	593.00	1.11	1145044992	231536832.00	180083029.33	47.15	2.78	1930.93	true
void fft2d_r2c_16x16<float>(float2, float const, int, int, int, int, int, int, int, int)	11	569.29	1.07	1148760064	73668053.33	295973162.67	64.59	3.11	2017.90	true
void fft2d_r2c_32x32<float, false, 0u, false>(float2, float const, int, int, int, int, int, int, int, int, int, cudnn::reduced_divisor, bool, int2, int, int)	2	395.00	0.74	739770368	103695008.00	142780714.67	46.53	3.00	1872.84	true
void fft2d_r2c_32x32<float, false, 1u, true>(float2, float const, int, int, int, int, int, int, int, int, int, cudnn::reduced_divisor, bool, int2, int, int)	2	298.67	0.56	901595136	1446890.67	170368341.33	46.05	5.25	3018.74	true
void flip_filter<float, float>(float, float const, int, int, int, int)	5	305.67	0.57	0	278858.67	6384960.00	4.12	0.00	0.00	true
void tensorflow::(anonymous namespace)::GenerateNormalizedProb<float, float>(float const, float const, float const, float, int, int, bool)	0	6.00	0.01	3075072	2560.00	170.67	37.80	1126.12	512.51	false
void tensorflow::BiasNCHWKernel<float>(int, float const, float const, float*, int, int)	0	6.00	0.01	128128	6080.00	74.67	44.30	20.82	21.35	false
void tensorflow::functor::PadInputCustomKernelNCHW<float, 4>(int, float const, tensorflow::functor::Dimension<4>, float, tensorflow::functor::Dimension<4>, tensorflow::functor::Dimension<(4)-(2)>)	0	322.00	0.60	0	79864416.00	78048373.33	47.60	0.00	0.00	true
void tensorflow::functor::RowReduceKernel<cub::TransformInputIterator<float, tensorflow::(anonymous namespace)::SubtractAndExpFunctor<float, float>, cub::CountingInputIterator<int, long>, long>, float, cub::Sum>(cub::TransformInputIterator<float, tensorflow::(anonymous namespace)::SubtractAndExpFunctor<float, float>, cub::CountingInputIterator<int, long>, long>, float, int, int, cub::Sum, std::iterator_traits<cub::TransformInputIterator<float, tensorflow::(anonymous namespace)::SubtractAndExpFunctor<float, float>, cub::CountingInputIterator<int, long>, long> >::value_type)	0	9.67	0.02	1291776	6656.00	85.33	6.20	191.62	133.63	false
void tensorflow::functor::RowReduceKernel<float const, float, cub::Max>(float const, float, int, int, cub::Max, std::iterator_traits<float const*>::value_type)	0	5.00	0.01	0	3840.00	213.33	6.20	0.00	0.00	true
void tensorflow::functor::ShuffleInTensor3Simple<float, 2, 1, 0, false>(int, float const, tensorflow::functor::Dimension<3>, float)	57	404.00	0.76	0	26664810.67	25462656.00	44.31	0.00	0.00	true
void tensorflow::functor::SwapDimension1And2InTensor3UsingTiles<unsigned int, 1024, 1024, 2, false>(unsigned int const, tensorflow::functor::Dimension<3>, unsigned int)	0	275.33	0.52	0	77077600.00	79710400.00	93.30	0.00	0.00	true
volta_cgemm_32x32_tn	5	2197.33	4.12	26130382848	184809898.67	143112096.00	23.56	79.68	11891.86	false
volta_gcgemm_32x32_nt	2	1510.67	2.83	17199104000	117622304.00	120662517.33	16.88	72.18	11385.11	false
volta_scudnn_128x32_relu_interior_nn_v1	5	1569.67	2.94	14811955200	677660000.00	74096192.00	20.27	19.70	9436.37	false
volta_scudnn_128x64_relu_interior_nn_v1	31	8993.33	16.86	89883836416	2078775658.66	409815509.33	19.83	36.12	9994.50	false
volta_scudnn_128x64_relu_medium_nn_v1	0	2582.67	4.84	31444697088	11255360.00	262971498.67	24.90	114.67	12175.28	false
volta_scudnn_128x64_relu_small_nn_v1	2	778.67	1.46	6943555584	1257888.00	8415125.33	16.13	717.83	8917.25	false
volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1	6	5586.00	10.47	60254322688	61305589.33	195721738.67	17.65	234.43	10786.67	false

Showing 1 to 28 of 28 entries

Download as CSV