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)	44	155.00	0.34	0	5866.67	580650.67	5.92	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)	120	6214.00	13.70	3149537280	1937702154.67	2023561184.00	73.05	0.80	506.85	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)	15	1224.33	2.70	4832952320	8117653.33	19475125.33	6.98	175.15	3947.41	false
void cudnn::detail::implicit_convolve_sgemm<float, float, 128, 6, 7, 3, 3, 5, 1, true, false, true>(int, int, int, float const, int, float, float, kernel_conv_params, int, float, float, int, float, float*, int, int)	0	725.00	1.60	6579879936	832.00	3744.00	31.90	1437910.83	9075.70	false
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)	3	196.33	0.43	120310240	89293536.00	29608597.33	50.46	1.01	612.79	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	216.33	0.48	6422528	102768384.00	32166229.33	72.00	0.05	29.69	true
void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>)	45	214.67	0.47	10928128	6794453.33	6625728.00	6.20	0.81	50.91	true
void cudnn::winograd_nonfused::winogradForwardData4x4<float, float>(cudnn::winograd_nonfused::WinogradDataParams<float, float>)	11	766.00	1.69	578027520	152517248.00	350608853.33	42.88	1.15	754.61	true
void cudnn::winograd_nonfused::winogradForwardFilter4x4<float, float>(cudnn::winograd_nonfused::WinogradFilterParams<float, float>)	11	62.00	0.14	3981312	1777749.33	4575221.33	12.32	0.63	64.21	true
void cudnn::winograd_nonfused::winogradForwardOutput4x4<float, float>(cudnn::winograd_nonfused::WinogradOutputParams<float, float>)	11	219.00	0.48	198795264	68345685.33	40892938.67	36.31	1.82	907.74	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	4.33	0.01	32000	6442.67	4309.33	19.00	2.98	7.39	true
void mshadow::cuda::MapPlanKernel<mshadow::sv::saveto, 8, mshadow::expr::Plan<mshadow::expr::SliceExp<mshadow::Tensor<mshadow::gpu, 3, float>, mshadow::gpu, float, 3, 2>, float>, mshadow::expr::Plan<mshadow::Tensor<mshadow::gpu, 3, float>, float> >(mshadow::expr::Plan<mshadow::expr::SliceExp<mshadow::Tensor<mshadow::gpu, 3, float>, mshadow::gpu, float, 3, 2>, float>, int, mshadow::Shape<2>, mshadow::expr::Plan<mshadow::Tensor<mshadow::gpu, 3, float>, float>)	113	3544.66	7.81	0	1134910378.67	1140702688.00	77.20	0.00	0.00	true
void mshadow::cuda::MapPlanLargeKernel<mshadow::sv::saveto, 8, 1024, mshadow::expr::Plan<mshadow::expr::SliceExp<mshadow::Tensor<mshadow::gpu, 3, float>, mshadow::gpu, float, 3, 2>, float>, mshadow::expr::Plan<mshadow::Tensor<mshadow::gpu, 3, float>, float> >(mshadow::expr::Plan<mshadow::expr::SliceExp<mshadow::Tensor<mshadow::gpu, 3, float>, mshadow::gpu, float, 3, 2>, float>, int, mshadow::Shape<2>, mshadow::expr::Plan<mshadow::Tensor<mshadow::gpu, 3, float>, float>, int)	1	505.00	1.11	0	166998720.00	162589770.67	82.19	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)	120	5389.00	11.88	1002717184	1851847402.67	1939018549.34	93.98	0.26	186.07	true
volta_scudnn_128x64_relu_interior_nn_v1	43	8573.00	18.90	83133202432	1859480864.00	571192309.33	17.19	34.20	9697.10	false
volta_scudnn_128x64_relu_medium_nn_v1	0	670.67	1.48	7861174272	21.33	5205.33	24.30	1504051.39	11721.43	false
volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1	45	3685.00	8.12	32674152448	867243370.67	121673045.33	19.76	33.04	8866.79	false
volta_sgemm_128x64_nn	11	1064.33	2.35	11823611904	351070122.67	91420032.00	20.98	26.72	11108.94	false
volta_sgemm_32x32_sliced1x4_tn	0	28.67	0.06	67633152	4111626.67	929109.33	6.20	13.42	2359.27	true

Showing 1 to 19 of 19 entries

Download as CSV