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
cudnn::gemm::computeOffsetsKernel(cudnn::gemm::ComputeOffsetsParams)	107	360.33	0.07	0	18026.67	626837.33	5.80	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)	152	60124.66	11.26	34485829632	9624158144.00	12402388000.00	83.97	1.57	573.57	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	238.67	0.04	37040128	177505333.33	5618496.00	60.60	0.20	155.20	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)	0	1663.00	0.31	51380224	179586197.33	140975904.00	72.30	0.16	30.90	true
void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>)	2	16.67	0.00	712704	449344.00	812714.67	6.20	0.56	42.76	true
void cudnn::winograd_nonfused::winogradForwardData4x4<float, float>(cudnn::winograd_nonfused::WinogradDataParams<float, float>)	36	9927.33	1.86	9059696640	1849671989.34	5421199040.00	46.96	1.25	912.60	true
void cudnn::winograd_nonfused::winogradForwardFilter4x4<float, float>(cudnn::winograd_nonfused::WinogradFilterParams<float, float>)	36	710.34	0.13	228261888	101468416.00	401535690.67	39.24	0.45	321.34	true
void cudnn::winograd_nonfused::winogradForwardOutput4x4<float, float>(cudnn::winograd_nonfused::WinogradOutputParams<float, float>)	36	9452.33	1.77	10299113472	5448045610.66	1955122090.66	48.20	1.39	1089.58	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)	6	2330.33	0.44	4932501504	998925269.33	774036448.00	47.46	2.78	2116.65	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)	6	2512.00	0.47	5178392576	724868832.00	998594677.33	47.89	3.00	2061.46	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)	6	756.00	0.14	2589196288	4192906.67	495577749.33	46.57	5.18	3424.86	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>)	0	5.67	0.00	256000	6016.00	1024.00	59.90	36.36	45.17	false
void mshadow::cuda::MapPlanKernel<mshadow::sv::saveto, 8, mshadow::expr::Plan<mshadow::Tensor<mshadow::gpu, 1, float>, float>, mshadow::expr::Plan<mshadow::expr::ScalarExp<float>, float> >(mshadow::expr::Plan<mshadow::Tensor<mshadow::gpu, 1, float>, float>, int, mshadow::Shape<2>, mshadow::expr::Plan<mshadow::expr::ScalarExp<float>, float>)	0	3.67	0.00	0	1504.00	5632.00	12.10	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)	49	48541.00	9.09	3365404672	2941065098.67	3604765034.66	96.17	0.51	69.33	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)	151	57978.33	10.86	10879762432	9958465162.67	12238789845.34	98.34	0.49	187.65	true
volta_gcgemm_32x32_nt	6	8207.00	1.54	98420654080	320192160.00	429414741.33	16.84	131.30	11992.28	false
volta_scudnn_128x128_relu_interior_nn_v1	2	12989.00	2.43	158199709696	642222805.33	307160608.00	16.50	166.63	12179.51	false
volta_scudnn_128x128_relu_small_nn_v1	1	10074.33	1.89	118444261376	55888181.33	47991925.33	15.32	1140.20	11757.03	false
volta_scudnn_128x64_relu_interior_nn_v1	100	217689.00	40.78	2725258461184	15127680778.66	12556348576.00	24.43	98.44	12519.05	false
volta_scudnn_128x64_relu_medium_nn_v1	0	5133.00	0.96	62889394176	11323178.67	287966005.33	12.70	210.13	12251.98	false
volta_scudnn_128x64_relu_small_nn_v1	0	4678.00	0.88	59215708160	69098794.67	20239242.67	14.60	662.83	12658.34	false
volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1	2	8873.33	1.66	95567216640	229661141.33	432125066.67	24.80	144.41	10770.16	false
volta_sgemm_128x32_sliced1x4_tn	0	139.00	0.03	1077936128	10693504.00	325760.00	12.50	97.82	7754.94	false
volta_sgemm_128x64_nn	36	55811.67	10.46	716471009280	3086978218.67	2793039125.33	24.70	121.85	12837.30	false

Showing 1 to 24 of 24 entries

Download as CSV