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)	73	256.00	0.07	0	16256.00	490794.66	5.79	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)	101	42948.66	11.48	24246484992	6432225685.33	8775772096.00	82.59	1.59	564.55	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	238.33	0.06	37017088	176778400.00	5639744.00	60.50	0.20	155.32	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	1668.00	0.45	51380224	177097909.33	140263498.67	72.30	0.16	30.80	true
void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>)	2	16.00	0.00	712704	449600.00	814037.33	6.20	0.56	44.54	true
void cudnn::winograd_nonfused::winogradForwardData4x4<float, float>(cudnn::winograd_nonfused::WinogradDataParams<float, float>)	23	6361.00	1.70	5788139520	1181142656.00	3463662453.33	46.88	1.25	909.94	true
void cudnn::winograd_nonfused::winogradForwardFilter4x4<float, float>(cudnn::winograd_nonfused::WinogradFilterParams<float, float>)	23	499.00	0.13	159252480	70793557.33	280135434.67	42.99	0.45	319.14	true
void cudnn::winograd_nonfused::winogradForwardOutput4x4<float, float>(cudnn::winograd_nonfused::WinogradOutputParams<float, float>)	23	5979.33	1.60	6795821056	3481293248.00	1251392522.67	48.22	1.44	1136.55	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	1005.67	0.27	2113929216	428159850.67	331822080.00	47.53	2.78	2102.02	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	1079.67	0.29	2219311104	310766389.33	428015200.00	47.87	3.00	2055.55	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	328.67	0.09	1109655552	1810314.67	212381034.67	46.63	5.18	3376.23	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	6.00	0.00	256000	5760.00	0.00	59.90	44.44	42.67	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	3.33	0.00	0	1504.00	5717.33	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)	32	33009.33	8.83	2286419968	2085898250.67	2717975296.00	96.29	0.48	69.27	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)	100	40772.00	10.90	7642808320	6657250165.33	8660278709.33	98.46	0.50	187.45	true
volta_gcgemm_32x32_nt	2	3522.33	0.94	42180280320	138921717.33	187178709.33	16.97	129.35	11975.10	false
volta_scudnn_128x128_relu_interior_nn_v1	2	13031.67	3.48	158199709696	638549717.33	308768778.67	16.46	167.00	12139.64	false
volta_scudnn_128x128_relu_small_nn_v1	1	10076.00	2.69	118444261376	55551594.67	48108864.00	15.26	1142.62	11755.09	false
volta_scudnn_128x64_relu_interior_nn_v1	66	146443.66	39.16	1828134060032	9904047786.67	8717622581.33	24.22	98.17	12483.53	false
volta_scudnn_128x64_relu_medium_nn_v1	0	5133.00	1.37	62889394176	11424256.00	288808714.67	12.70	209.47	12251.98	false
volta_scudnn_128x64_relu_small_nn_v1	0	4675.67	1.25	59215708160	67374090.67	20237280.00	14.70	675.89	12664.65	false
volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1	2	8874.67	2.37	95567216640	228331552.00	431761536.00	24.80	144.78	10768.54	false
volta_sgemm_128x32_sliced1x4_tn	0	138.67	0.04	1077936128	10723221.33	208629.33	12.50	98.61	7773.56	false
volta_sgemm_128x64_nn	23	36295.00	9.70	464724688896	1971264778.66	1781654058.67	24.70	123.83	12804.10	false

Showing 1 to 24 of 24 entries

Download as CSV