libcutlass-dev (3.4.1+ds-2build1)
Details
Core information at a glance
- Distribution
- ubuntu
- Origin
- ubuntu-stonking
- Repository
- https://nl.archive.ubuntu.com/ubuntu
- Codename
- stonking
- Component
- multiverse
- Source
- nvidia-cutlass
- Architecture
- amd64
- Section
- multiverse/libdevel
- Priority
- optional
- Maintainer
- Ubuntu Developers <[email protected]>
Size & integrity
Byte sizes and integrity verification
- Installed size
- 11.8 kB
- Size expected
- 662.5 kB
- Size actual
- 662.5 kB
- Size match
Dependencies
Required package dependencies
- None
Suggested packages
Recommended additional packages
- None
Description
CUDA Templates for Linear Algebra Subroutines
Tags
Package classification tags
None
Checksums
Hash values and integrity verification status
| Type | Actual | Match |
|---|---|---|
| MD5 | c35f41fb…348ced8c | |
| SHA-1 | 982fdc59…7c2afe65 | |
| SHA-256 | e394e9bc…cf99a452 | |
| SHA-512 | 20cea8fc…fadd64e5 |
Contents
Files and directories included
. usr usr/include usr/include/cute usr/include/cute/algorithm usr/include/cute/algorithm/axpby.hpp usr/include/cute/algorithm/clear.hpp usr/include/cute/algorithm/copy.hpp usr/include/cute/algorithm/fill.hpp usr/include/cute/algorithm/functional.hpp usr/include/cute/algorithm/gemm.hpp usr/include/cute/algorithm/prefer.hpp usr/include/cute/algorithm/tensor_algorithms.hpp usr/include/cute/algorithm/tuple_algorithms.hpp usr/include/cute/arch usr/include/cute/arch/cluster_sm90.hpp usr/include/cute/arch/copy.hpp usr/include/cute/arch/copy_sm75.hpp usr/include/cute/arch/copy_sm80.hpp usr/include/cute/arch/copy_sm90.hpp usr/include/cute/arch/copy_sm90_desc.hpp usr/include/cute/arch/copy_sm90_tma.hpp usr/include/cute/arch/mma.hpp usr/include/cute/arch/mma_sm61.hpp usr/include/cute/arch/mma_sm70.hpp usr/include/cute/arch/mma_sm75.hpp usr/include/cute/arch/mma_sm80.hpp usr/include/cute/arch/mma_sm90.hpp usr/include/cute/arch/mma_sm90_desc.hpp usr/include/cute/arch/mma_sm90_gmma.hpp usr/include/cute/arch/util.hpp usr/include/cute/atom usr/include/cute/atom/copy_atom.hpp usr/include/cute/atom/copy_traits.hpp usr/include/cute/atom/copy_traits_sm75.hpp usr/include/cute/atom/copy_traits_sm80.hpp usr/include/cute/atom/copy_traits_sm90.hpp usr/include/cute/atom/copy_traits_sm90_tma.hpp usr/include/cute/atom/copy_traits_sm90_tma_swizzle.hpp usr/include/cute/atom/mma_atom.hpp usr/include/cute/atom/mma_traits.hpp usr/include/cute/atom/mma_traits_sm61.hpp usr/include/cute/atom/mma_traits_sm70.hpp usr/include/cute/atom/mma_traits_sm75.hpp usr/include/cute/atom/mma_traits_sm80.hpp usr/include/cute/atom/mma_traits_sm90.hpp usr/include/cute/atom/mma_traits_sm90_gmma.hpp usr/include/cute/config.hpp usr/include/cute/container usr/include/cute/container/alignment.hpp usr/include/cute/container/array.hpp usr/include/cute/container/array_aligned.hpp usr/include/cute/container/array_subbyte.hpp usr/include/cute/container/bit_field.hpp usr/include/cute/container/cuda_types.hpp usr/include/cute/container/tuple.hpp usr/include/cute/container/type_list.hpp usr/include/cute/int_tuple.hpp usr/include/cute/layout.hpp usr/include/cute/layout_composed.hpp usr/include/cute/numeric usr/include/cute/numeric/arithmetic_tuple.hpp usr/include/cute/numeric/bfloat.hpp usr/include/cute/numeric/complex.hpp usr/include/cute/numeric/float8.hpp usr/include/cute/numeric/half.hpp usr/include/cute/numeric/int.hpp usr/include/cute/numeric/integer_sequence.hpp usr/include/cute/numeric/integer_subbyte.hpp usr/include/cute/numeric/integral_constant.hpp usr/include/cute/numeric/integral_ratio.hpp usr/include/cute/numeric/math.hpp usr/include/cute/numeric/real.hpp usr/include/cute/numeric/tfloat.hpp usr/include/cute/numeric/uint128.hpp usr/include/cute/pointer.hpp usr/include/cute/pointer_base.hpp usr/include/cute/pointer_flagged.hpp usr/include/cute/pointer_swizzle.hpp usr/include/cute/stride.hpp usr/include/cute/swizzle.hpp usr/include/cute/swizzle_layout.hpp usr/include/cute/tensor.hpp usr/include/cute/tensor_predicate.hpp usr/include/cute/tile.hpp usr/include/cute/underscore.hpp usr/include/cute/util usr/include/cute/util/debug.hpp usr/include/cute/util/print.hpp usr/include/cute/util/type_traits.hpp usr/include/cutlass usr/include/cutlass/aligned_buffer.h usr/include/cutlass/arch usr/include/cutlass/arch/arch.h usr/include/cutlass/arch/barrier.h usr/include/cutlass/arch/cache_operation.h usr/include/cutlass/arch/memory.h usr/include/cutlass/arch/memory_sm75.h usr/include/cutlass/arch/memory_sm80.h usr/include/cutlass/arch/mma.h usr/include/cutlass/arch/mma_sm50.h usr/include/cutlass/arch/mma_sm60.h usr/include/cutlass/arch/mma_sm61.h usr/include/cutlass/arch/mma_sm70.h usr/include/cutlass/arch/mma_sm75.h usr/include/cutlass/arch/mma_sm80.h usr/include/cutlass/arch/mma_sm90.h usr/include/cutlass/arch/mma_sparse_sm80.h usr/include/cutlass/arch/reg_reconfig.h usr/include/cutlass/arch/simd.h usr/include/cutlass/arch/simd_sm60.h usr/include/cutlass/arch/simd_sm61.h usr/include/cutlass/arch/wmma.h usr/include/cutlass/arch/wmma_sm70.h usr/include/cutlass/arch/wmma_sm72.h usr/include/cutlass/arch/wmma_sm75.h usr/include/cutlass/array.h usr/include/cutlass/array_planar_complex.h usr/include/cutlass/array_subbyte.h usr/include/cutlass/barrier.h usr/include/cutlass/bfloat16.h usr/include/cutlass/blas3.h usr/include/cutlass/blas3_types.h usr/include/cutlass/block_striped.h usr/include/cutlass/cluster_launch.hpp usr/include/cutlass/complex.h usr/include/cutlass/constants.h usr/include/cutlass/conv usr/include/cutlass/conv/conv2d_problem_size.h usr/include/cutlass/conv/conv3d_problem_size.h usr/include/cutlass/conv/convolution.h usr/include/cutlass/conv/device usr/include/cutlass/conv/device/direct_convolution.h usr/include/cutlass/conv/device/implicit_gemm_convolution.h usr/include/cutlass/conv/device/implicit_gemm_convolution_fusion.h usr/include/cutlass/conv/kernel usr/include/cutlass/conv/kernel/default_conv2d.h usr/include/cutlass/conv/kernel/default_conv2d_dgrad.h usr/include/cutlass/conv/kernel/default_conv2d_fprop.h usr/include/cutlass/conv/kernel/default_conv2d_fprop_fusion.h usr/include/cutlass/conv/kernel/default_conv2d_fprop_with_broadcast.h usr/include/cutlass/conv/kernel/default_conv2d_fprop_with_reduction.h usr/include/cutlass/conv/kernel/default_conv2d_group_fprop.h usr/include/cutlass/conv/kernel/default_conv2d_wgrad.h usr/include/cutlass/conv/kernel/default_conv2d_wgrad_fusion.h usr/include/cutlass/conv/kernel/default_conv3d_dgrad.h usr/include/cutlass/conv/kernel/default_conv3d_fprop.h usr/include/cutlass/conv/kernel/default_conv3d_fprop_fusion.h usr/include/cutlass/conv/kernel/default_conv3d_wgrad.h usr/include/cutlass/conv/kernel/default_depthwise_fprop.h usr/include/cutlass/conv/kernel/direct_convolution.h usr/include/cutlass/conv/kernel/implicit_gemm_convolution.h usr/include/cutlass/conv/kernel/implicit_gemm_convolution_fusion.h usr/include/cutlass/conv/kernel/implicit_gemm_convolution_strided_dgrad.h usr/include/cutlass/conv/kernel/implicit_gemm_convolution_with_fused_epilogue.h usr/include/cutlass/conv/thread usr/include/cutlass/conv/thread/depthwise_mma.h usr/include/cutlass/conv/threadblock usr/include/cutlass/conv/threadblock/conv2d_dgrad_filter_tile_access_iterator_analytic.h usr/include/cutlass/conv/threadblock/conv2d_dgrad_filter_tile_access_iterator_optimized.h usr/include/cutlass/conv/threadblock/conv2d_dgrad_output_gradient_tile_access_iterator_analytic.h usr/include/cutlass/conv/threadblock/conv2d_dgrad_output_gradient_tile_access_iterator_optimized.h usr/include/cutlass/conv/threadblock/conv2d_fprop_activation_tile_access_iterator_analytic.h usr/include/cutlass/conv/threadblock/conv2d_fprop_activation_tile_access_iterator_few_channels.h usr/include/cutlass/conv/threadblock/conv2d_fprop_activation_tile_access_iterator_fixed_channels.h usr/include/cutlass/conv/threadblock/conv2d_fprop_activation_tile_access_iterator_optimized.h usr/include/cutlass/conv/threadblock/conv2d_fprop_filter_tile_access_iterator_analytic.h usr/include/cutlass/conv/threadblock/conv2d_fprop_filter_tile_access_iterator_few_channels.h usr/include/cutlass/conv/threadblock/conv2d_fprop_filter_tile_access_iterator_fixed_channels.h usr/include/cutlass/conv/threadblock/conv2d_fprop_filter_tile_access_iterator_optimized.h usr/include/cutlass/conv/threadblock/conv2d_params.h usr/include/cutlass/conv/threadblock/conv2d_tile_iterator.h usr/include/cutlass/conv/threadblock/conv2d_wgrad_activation_tile_access_iterator_analytic.h usr/include/cutlass/conv/threadblock/conv2d_wgrad_activation_tile_access_iterator_optimized.h usr/include/cutlass/conv/threadblock/conv2d_wgrad_output_gradient_tile_access_iterator_analytic.h usr/include/cutlass/conv/threadblock/conv2d_wgrad_output_gradient_tile_access_iterator_optimized.h usr/include/cutlass/conv/threadblock/conv3d_dgrad_filter_tile_access_iterator_analytic.h usr/include/cutlass/conv/threadblock/conv3d_dgrad_filter_tile_access_iterator_optimized.h usr/include/cutlass/conv/threadblock/conv3d_dgrad_output_gradient_tile_access_iterator_analytic.h usr/include/cutlass/conv/threadblock/conv3d_dgrad_output_gradient_tile_access_iterator_optimized.h usr/include/cutlass/conv/threadblock/conv3d_fprop_activation_tile_access_iterator_analytic.h usr/include/cutlass/conv/threadblock/conv3d_fprop_activation_tile_access_iterator_optimized.h usr/include/cutlass/conv/threadblock/conv3d_fprop_filter_tile_access_iterator_analytic.h usr/include/cutlass/conv/threadblock/conv3d_fprop_filter_tile_access_iterator_optimized.h usr/include/cutlass/conv/threadblock/conv3d_params.h usr/include/cutlass/conv/threadblock/conv3d_wgrad_activation_tile_access_iterator_analytic.h usr/include/cutlass/conv/threadblock/conv3d_wgrad_activation_tile_access_iterator_optimized.h usr/include/cutlass/conv/threadblock/conv3d_wgrad_output_gradient_tile_access_iterator_analytic.h usr/include/cutlass/conv/threadblock/conv3d_wgrad_output_gradient_tile_access_iterator_optimized.h usr/include/cutlass/conv/threadblock/depthwise_direct_conv_params.h usr/include/cutlass/conv/threadblock/depthwise_fprop_activation_tile_access_iterator_direct_conv_fixed_stride_dilation.h usr/include/cutlass/conv/threadblock/depthwise_fprop_activation_tile_access_iterator_direct_conv_optimized.h usr/include/cutlass/conv/threadblock/depthwise_fprop_direct_conv_multistage.h usr/include/cutlass/conv/threadblock/depthwise_fprop_filter_tile_access_iterator_direct_conv_optimized.h usr/include/cutlass/conv/threadblock/depthwise_fprop_pipelined.h usr/include/cutlass/conv/threadblock/depthwise_mma_base.h usr/include/cutlass/conv/threadblock/depthwise_mma_core_with_lane_access_size.h usr/include/cutlass/conv/threadblock/implicit_gemm_fprop_fusion_multistage.h usr/include/cutlass/conv/threadblock/implicit_gemm_multistage.h usr/include/cutlass/conv/threadblock/implicit_gemm_pipelined.h usr/include/cutlass/conv/threadblock/implicit_gemm_wgrad_fusion_multistage.h usr/include/cutlass/conv/threadblock/predicated_scale_bias_vector_access_iterator.h usr/include/cutlass/conv/threadblock/predicated_scale_bias_vector_iterator.h usr/include/cutlass/conv/threadblock/threadblock_swizzle.h usr/include/cutlass/conv/warp usr/include/cutlass/conv/warp/mma_depthwise_simt.h usr/include/cutlass/conv/warp/mma_depthwise_simt_tile_iterator.h usr/include/cutlass/conv/warp/scale_bias_relu_transform.h usr/include/cutlass/coord.h usr/include/cutlass/core_io.h usr/include/cutlass/cuda_host_adapter.hpp usr/include/cutlass/cutlass.h usr/include/cutlass/detail usr/include/cutlass/detail/collective.hpp usr/include/cutlass/detail/dependent_false.hpp usr/include/cutlass/detail/helper_macros.hpp usr/include/cutlass/detail/layout.hpp usr/include/cutlass/detail/mma.hpp usr/include/cutlass/device_kernel.h usr/include/cutlass/epilogue usr/include/cutlass/epilogue/collective usr/include/cutlass/epilogue/collective/builders usr/include/cutlass/epilogue/collective/builders/sm90_builder.inl usr/include/cutlass/epilogue/collective/collective_builder.hpp usr/include/cutlass/epilogue/collective/collective_epilogue.hpp usr/include/cutlass/epilogue/collective/default_epilogue.hpp usr/include/cutlass/epilogue/collective/default_epilogue_array.hpp usr/include/cutlass/epilogue/collective/detail.hpp usr/include/cutlass/epilogue/collective/epilogue_tensor_broadcast.hpp usr/include/cutlass/epilogue/collective/sm70_epilogue_vectorized.hpp usr/include/cutlass/epilogue/collective/sm90_epilogue_tma_warpspecialized.hpp usr/include/cutlass/epilogue/collective/sm90_epilogue_tma_warpspecialized_bias_elementwise.hpp usr/include/cutlass/epilogue/dispatch_policy.hpp usr/include/cutlass/epilogue/fusion usr/include/cutlass/epilogue/fusion/callbacks.hpp usr/include/cutlass/epilogue/fusion/operations.hpp usr/include/cutlass/epilogue/fusion/sm90_callbacks_tma_warpspecialized.hpp usr/include/cutlass/epilogue/fusion/sm90_visitor_compute_tma_warpspecialized.hpp usr/include/cutlass/epilogue/fusion/sm90_visitor_load_tma_warpspecialized.hpp usr/include/cutlass/epilogue/fusion/sm90_visitor_store_tma_warpspecialized.hpp usr/include/cutlass/epilogue/fusion/sm90_visitor_tma_warpspecialized.hpp usr/include/cutlass/epilogue/thread usr/include/cutlass/epilogue/thread/activation.h usr/include/cutlass/epilogue/thread/conversion_op.h usr/include/cutlass/epilogue/thread/detail.hpp usr/include/cutlass/epilogue/thread/linear_combination.h usr/include/cutlass/epilogue/thread/linear_combination_bias_elementwise.h usr/include/cutlass/epilogue/thread/linear_combination_bias_relu.h usr/include/cutlass/epilogue/thread/linear_combination_clamp.h usr/include/cutlass/epilogue/thread/linear_combination_dgelu.h usr/include/cutlass/epilogue/thread/linear_combination_drelu.h usr/include/cutlass/epilogue/thread/linear_combination_gelu.h usr/include/cutlass/epilogue/thread/linear_combination_generic.h usr/include/cutlass/epilogue/thread/linear_combination_hardswish.h usr/include/cutlass/epilogue/thread/linear_combination_leaky_relu.h usr/include/cutlass/epilogue/thread/linear_combination_params.h usr/include/cutlass/epilogue/thread/linear_combination_planar_complex.h usr/include/cutlass/epilogue/thread/linear_combination_relu.h usr/include/cutlass/epilogue/thread/linear_combination_relu0.h usr/include/cutlass/epilogue/thread/linear_combination_residual_block.h usr/include/cutlass/epilogue/thread/linear_combination_sigmoid.h usr/include/cutlass/epilogue/thread/linear_combination_silu.h usr/include/cutlass/epilogue/thread/linear_combination_tensor_broadcast.hpp usr/include/cutlass/epilogue/thread/linear_combination_with_elementwise.h usr/include/cutlass/epilogue/thread/reduction_op.h usr/include/cutlass/epilogue/thread/scale_type.h usr/include/cutlass/epilogue/threadblock usr/include/cutlass/epilogue/threadblock/default_epilogue_complex_tensor_op.h usr/include/cutlass/epilogue/threadblock/default_epilogue_complex_tensor_op_blas3.h usr/include/cutlass/epilogue/threadblock/default_epilogue_direct_store.h usr/include/cutlass/epilogue/threadblock/default_epilogue_planar_complex.h usr/include/cutlass/epilogue/threadblock/default_epilogue_simt.h usr/include/cutlass/epilogue/threadblock/default_epilogue_tensor_op.h usr/include/cutlass/epilogue/threadblock/default_epilogue_tensor_op_blas3.h usr/include/cutlass/epilogue/threadblock/default_epilogue_volta_tensor_op.h usr/include/cutlass/epilogue/threadblock/default_epilogue_with_broadcast.h usr/include/cutlass/epilogue/threadblock/default_epilogue_with_reduction.h usr/include/cutlass/epilogue/threadblock/default_epilogue_wmma_tensor_op.h usr/include/cutlass/epilogue/threadblock/default_thread_map_simt.h usr/include/cutlass/epilogue/threadblock/default_thread_map_tensor_op.h usr/include/cutlass/epilogue/threadblock/default_thread_map_volta_tensor_op.h usr/include/cutlass/epilogue/threadblock/default_thread_map_wmma_tensor_op.h usr/include/cutlass/epilogue/threadblock/direct_store_epilogue_iterator.h usr/include/cutlass/epilogue/threadblock/epilogue.h usr/include/cutlass/epilogue/threadblock/epilogue_base.h usr/include/cutlass/epilogue/threadblock/epilogue_base_streamk.h usr/include/cutlass/epilogue/threadblock/epilogue_depthwise.h usr/include/cutlass/epilogue/threadblock/epilogue_direct_store.h usr/include/cutlass/epilogue/threadblock/epilogue_gemm_k_reduction.h usr/include/cutlass/epilogue/threadblock/epilogue_planar_complex.h usr/include/cutlass/epilogue/threadblock/epilogue_smem_accumulator.h usr/include/cutlass/epilogue/threadblock/epilogue_streamk_with_broadcast.h usr/include/cutlass/epilogue/threadblock/epilogue_visitor_with_softmax.h usr/include/cutlass/epilogue/threadblock/epilogue_with_broadcast.h usr/include/cutlass/epilogue/threadblock/epilogue_with_reduction.h usr/include/cutlass/epilogue/threadblock/epilogue_with_visitor.h usr/include/cutlass/epilogue/threadblock/epilogue_with_visitor_callbacks.h usr/include/cutlass/epilogue/threadblock/epilogue_workspace.h usr/include/cutlass/epilogue/threadblock/fusion usr/include/cutlass/epilogue/threadblock/fusion/visitor_2x.hpp usr/include/cutlass/epilogue/threadblock/fusion/visitor_compute.hpp usr/include/cutlass/epilogue/threadblock/fusion/visitor_load.hpp usr/include/cutlass/epilogue/threadblock/fusion/visitor_store.hpp usr/include/cutlass/epilogue/threadblock/fusion/visitors.hpp usr/include/cutlass/epilogue/threadblock/interleaved_epilogue.h usr/include/cutlass/epilogue/threadblock/output_iterator_parameter.h usr/include/cutlass/epilogue/threadblock/output_tile_thread_map.h usr/include/cutlass/epilogue/threadblock/predicated_tile_iterator.h usr/include/cutlass/epilogue/threadblock/predicated_tile_iterator_affine.h usr/include/cutlass/epilogue/threadblock/predicated_tile_iterator_affine_layout_params.h usr/include/cutlass/epilogue/threadblock/predicated_tile_iterator_blas3.h usr/include/cutlass/epilogue/threadblock/predicated_tile_iterator_direct_conv.h usr/include/cutlass/epilogue/threadblock/predicated_tile_iterator_params.h usr/include/cutlass/epilogue/threadblock/predicated_tile_iterator_predicates.h usr/include/cutlass/epilogue/threadblock/predicated_tile_iterator_strided_dgrad.h usr/include/cutlass/epilogue/threadblock/shared_load_iterator.h usr/include/cutlass/epilogue/threadblock/shared_load_iterator_mixed.h usr/include/cutlass/epilogue/threadblock/shared_load_iterator_pitch_liner.h usr/include/cutlass/epilogue/warp usr/include/cutlass/epilogue/warp/fragment_iterator_complex_tensor_op.h usr/include/cutlass/epilogue/warp/fragment_iterator_gaussian_complex_tensor_op.h usr/include/cutlass/epilogue/warp/fragment_iterator_simt.h usr/include/cutlass/epilogue/warp/fragment_iterator_tensor_op.h usr/include/cutlass/epilogue/warp/fragment_iterator_volta_tensor_op.h usr/include/cutlass/epilogue/warp/fragment_iterator_wmma_tensor_op.h usr/include/cutlass/epilogue/warp/simt_policy.h usr/include/cutlass/epilogue/warp/tensor_op_policy.h usr/include/cutlass/epilogue/warp/tile_iterator_simt.h usr/include/cutlass/epilogue/warp/tile_iterator_tensor_op.h usr/include/cutlass/epilogue/warp/tile_iterator_tensor_op_mixed.h usr/include/cutlass/epilogue/warp/tile_iterator_volta_tensor_op.h usr/include/cutlass/epilogue/warp/tile_iterator_wmma_tensor_op.h usr/include/cutlass/epilogue/warp/volta_tensor_op_policy.h usr/include/cutlass/epilogue/warp/wmma_tensor_op_policy.h usr/include/cutlass/fast_math.h usr/include/cutlass/float8.h usr/include/cutlass/floating_point_nvrtc.h usr/include/cutlass/functional.h usr/include/cutlass/gemm usr/include/cutlass/gemm/collective usr/include/cutlass/gemm/collective/builders usr/include/cutlass/gemm/collective/builders/sm90_common.inl usr/include/cutlass/gemm/collective/builders/sm90_gmma_builder.inl usr/include/cutlass/gemm/collective/collective_builder.hpp usr/include/cutlass/gemm/collective/collective_mma.hpp usr/include/cutlass/gemm/collective/fp8_accumulation.hpp usr/include/cutlass/gemm/collective/sm70_mma_twostage.hpp usr/include/cutlass/gemm/collective/sm80_mma_multistage.hpp usr/include/cutlass/gemm/collective/sm90_mma_array_tma_gmma_ss_warpspecialized.hpp usr/include/cutlass/gemm/collective/sm90_mma_multistage_gmma_rs_warpspecialized.hpp usr/include/cutlass/gemm/collective/sm90_mma_multistage_gmma_ss_warpspecialized.hpp usr/include/cutlass/gemm/collective/sm90_mma_tma_gmma_rs_warpspecialized.hpp usr/include/cutlass/gemm/collective/sm90_mma_tma_gmma_rs_warpspecialized_mixed_input.hpp usr/include/cutlass/gemm/collective/sm90_mma_tma_gmma_ss.hpp usr/include/cutlass/gemm/collective/sm90_mma_tma_gmma_ss_warpspecialized.hpp usr/include/cutlass/gemm/collective/sm90_mma_tma_gmma_ss_warpspecialized_fp8.hpp usr/include/cutlass/gemm/device usr/include/cutlass/gemm/device/base_grouped.h usr/include/cutlass/gemm/device/default_gemm_configuration.h usr/include/cutlass/gemm/device/ell_gemm.h usr/include/cutlass/gemm/device/gemm.h usr/include/cutlass/gemm/device/gemm_array.h usr/include/cutlass/gemm/device/gemm_batched.h usr/include/cutlass/gemm/device/gemm_complex.h usr/include/cutlass/gemm/device/gemm_grouped.h usr/include/cutlass/gemm/device/gemm_layernorm_mainloop_fusion.h usr/include/cutlass/gemm/device/gemm_sparse.h usr/include/cutlass/gemm/device/gemm_sparse_with_visitor.h usr/include/cutlass/gemm/device/gemm_splitk_parallel.h usr/include/cutlass/gemm/device/gemm_universal.h usr/include/cutlass/gemm/device/gemm_universal_adapter.h usr/include/cutlass/gemm/device/gemm_universal_base.h usr/include/cutlass/gemm/device/gemm_universal_streamk_with_broadcast.h usr/include/cutlass/gemm/device/gemm_universal_with_broadcast.h usr/include/cutlass/gemm/device/gemm_with_k_reduction.h usr/include/cutlass/gemm/device/gemv.h usr/include/cutlass/gemm/device/rank_2k.h usr/include/cutlass/gemm/device/rank_2k_grouped.h usr/include/cutlass/gemm/device/rank_k.h usr/include/cutlass/gemm/device/symm.h usr/include/cutlass/gemm/device/trmm.h usr/include/cutlass/gemm/dispatch_policy.hpp usr/include/cutlass/gemm/gemm.h usr/include/cutlass/gemm/gemm_enumerated_types.h usr/include/cutlass/gemm/group_array_problem_shape.hpp usr/include/cutlass/gemm/kernel usr/include/cutlass/gemm/kernel/default_ell_gemm.h usr/include/cutlass/gemm/kernel/default_gemm.h usr/include/cutlass/gemm/kernel/default_gemm_complex.h usr/include/cutlass/gemm/kernel/default_gemm_grouped.h usr/include/cutlass/gemm/kernel/default_gemm_grouped_softmax_mainloop_fusion.h usr/include/cutlass/gemm/kernel/default_gemm_layernorm_mainloop_fusion.h usr/include/cutlass/gemm/kernel/default_gemm_planar_complex_universal.h usr/include/cutlass/gemm/kernel/default_gemm_sparse.h usr/include/cutlass/gemm/kernel/default_gemm_sparse_with_visitor.h usr/include/cutlass/gemm/kernel/default_gemm_splitk_parallel.h usr/include/cutlass/gemm/kernel/default_gemm_streamk_with_broadcast.h usr/include/cutlass/gemm/kernel/default_gemm_universal.h usr/include/cutlass/gemm/kernel/default_gemm_universal_with_visitor.h usr/include/cutlass/gemm/kernel/default_gemm_with_broadcast.h usr/include/cutlass/gemm/kernel/default_gemm_with_k_reduction.h usr/include/cutlass/gemm/kernel/default_gemm_with_reduction.h usr/include/cutlass/gemm/kernel/default_gemv.h usr/include/cutlass/gemm/kernel/default_rank_2k.h usr/include/cutlass/gemm/kernel/default_rank_2k_complex.h usr/include/cutlass/gemm/kernel/default_rank_2k_grouped.h usr/include/cutlass/gemm/kernel/default_rank_2k_universal.h usr/include/cutlass/gemm/kernel/default_rank_k.h usr/include/cutlass/gemm/kernel/default_rank_k_complex.h usr/include/cutlass/gemm/kernel/default_rank_k_universal.h usr/include/cutlass/gemm/kernel/default_symm.h usr/include/cutlass/gemm/kernel/default_symm_complex.h usr/include/cutlass/gemm/kernel/default_symm_universal.h usr/include/cutlass/gemm/kernel/default_trmm.h usr/include/cutlass/gemm/kernel/default_trmm_complex.h usr/include/cutlass/gemm/kernel/default_trmm_universal.h usr/include/cutlass/gemm/kernel/ell_gemm.h usr/include/cutlass/gemm/kernel/gemm.h usr/include/cutlass/gemm/kernel/gemm_array.h usr/include/cutlass/gemm/kernel/gemm_batched.h usr/include/cutlass/gemm/kernel/gemm_grouped.h usr/include/cutlass/gemm/kernel/gemm_grouped_problem_visitor.h usr/include/cutlass/gemm/kernel/gemm_grouped_softmax_mainloop_fusion.h usr/include/cutlass/gemm/kernel/gemm_layernorm_mainloop_fusion.h usr/include/cutlass/gemm/kernel/gemm_params.h usr/include/cutlass/gemm/kernel/gemm_pipelined.h usr/include/cutlass/gemm/kernel/gemm_planar_complex.h usr/include/cutlass/gemm/kernel/gemm_planar_complex_array.h usr/include/cutlass/gemm/kernel/gemm_splitk_parallel.h usr/include/cutlass/gemm/kernel/gemm_streamk_with_fused_epilogue.h usr/include/cutlass/gemm/kernel/gemm_transpose_operands.h usr/include/cutlass/gemm/kernel/gemm_universal.h usr/include/cutlass/gemm/kernel/gemm_universal.hpp usr/include/cutlass/gemm/kernel/gemm_universal_streamk.h usr/include/cutlass/gemm/kernel/gemm_universal_with_visitor.h usr/include/cutlass/gemm/kernel/gemm_universal_with_visitor_streamk.h usr/include/cutlass/gemm/kernel/gemm_with_fused_epilogue.h usr/include/cutlass/gemm/kernel/gemm_with_k_reduction.h usr/include/cutlass/gemm/kernel/gemv.h usr/include/cutlass/gemm/kernel/gemv_batched_strided.h usr/include/cutlass/gemm/kernel/grouped_problem_visitor.h usr/include/cutlass/gemm/kernel/params_sparse_base.h usr/include/cutlass/gemm/kernel/params_universal_base.h usr/include/cutlass/gemm/kernel/rank_2k_grouped.h usr/include/cutlass/gemm/kernel/rank_2k_grouped_problem_visitor.h usr/include/cutlass/gemm/kernel/rank_2k_transpose_operands.h usr/include/cutlass/gemm/kernel/rank_2k_universal.h usr/include/cutlass/gemm/kernel/rank_k_universal.h usr/include/cutlass/gemm/kernel/sm70_gemm.hpp usr/include/cutlass/gemm/kernel/sm90_gemm_array_tma_warpspecialized_cooperative.hpp usr/include/cutlass/gemm/kernel/sm90_gemm_tma.hpp usr/include/cutlass/gemm/kernel/sm90_gemm_tma_warpspecialized.hpp usr/include/cutlass/gemm/kernel/sm90_gemm_tma_warpspecialized_cooperative.hpp usr/include/cutlass/gemm/kernel/sm90_gemm_tma_warpspecialized_pingpong.hpp usr/include/cutlass/gemm/kernel/sm90_gemm_warpspecialized.hpp usr/include/cutlass/gemm/kernel/sm90_gemm_warpspecialized_cooperative.hpp usr/include/cutlass/gemm/kernel/sm90_gemm_warpspecialized_pingpong.hpp usr/include/cutlass/gemm/kernel/sm90_tile_scheduler.hpp usr/include/cutlass/gemm/kernel/sm90_tile_scheduler_group.hpp usr/include/cutlass/gemm/kernel/sm90_tile_scheduler_stream_k.hpp usr/include/cutlass/gemm/kernel/sparse_gemm.h usr/include/cutlass/gemm/kernel/sparse_gemm_with_visitor.h usr/include/cutlass/gemm/kernel/static_tile_scheduler.hpp usr/include/cutlass/gemm/kernel/symm_universal.h usr/include/cutlass/gemm/kernel/tile_scheduler.hpp usr/include/cutlass/gemm/kernel/tile_scheduler_params.h usr/include/cutlass/gemm/kernel/trmm_universal.h usr/include/cutlass/gemm/thread usr/include/cutlass/gemm/thread/mma.h usr/include/cutlass/gemm/thread/mma_sm50.h usr/include/cutlass/gemm/thread/mma_sm60.h usr/include/cutlass/gemm/thread/mma_sm61.h usr/include/cutlass/gemm/threadblock usr/include/cutlass/gemm/threadblock/default_ell_mma.h usr/include/cutlass/gemm/threadblock/default_gemv_core.h usr/include/cutlass/gemm/threadblock/default_mma.h usr/include/cutlass/gemm/threadblock/default_mma_core.h usr/include/cutlass/gemm/threadblock/default_mma_core_simt.h usr/include/cutlass/gemm/threadblock/default_mma_core_sm70.h usr/include/cutlass/gemm/threadblock/default_mma_core_sm75.h usr/include/cutlass/gemm/threadblock/default_mma_core_sm80.h usr/include/cutlass/gemm/threadblock/default_mma_core_sparse_sm80.h usr/include/cutlass/gemm/threadblock/default_mma_core_with_access_size.h usr/include/cutlass/gemm/threadblock/default_mma_core_with_reduction.h usr/include/cutlass/gemm/threadblock/default_mma_core_wmma.h usr/include/cutlass/gemm/threadblock/default_mma_layernorm_mainloop_fusion.h usr/include/cutlass/gemm/threadblock/default_mma_planar_complex_multistage.h usr/include/cutlass/gemm/threadblock/default_mma_planar_complex_pipelined.h usr/include/cutlass/gemm/threadblock/default_mma_softmax_mainloop_fusion.h usr/include/cutlass/gemm/threadblock/default_mma_with_reduction.h usr/include/cutlass/gemm/threadblock/default_multistage_mma_complex.h usr/include/cutlass/gemm/threadblock/default_multistage_mma_complex_core.h usr/include/cutlass/gemm/threadblock/default_multistage_mma_complex_core_sm80.h usr/include/cutlass/gemm/threadblock/default_multistage_trmm_complex.h usr/include/cutlass/gemm/threadblock/default_sparse_mma.h usr/include/cutlass/gemm/threadblock/default_trmm.h usr/include/cutlass/gemm/threadblock/ell_mma_multistage.h usr/include/cutlass/gemm/threadblock/ell_mma_pipelined.h usr/include/cutlass/gemm/threadblock/gemv.h usr/include/cutlass/gemm/threadblock/index_remat.h usr/include/cutlass/gemm/threadblock/mma_base.h usr/include/cutlass/gemm/threadblock/mma_blas3_multistage.h usr/include/cutlass/gemm/threadblock/mma_layernorm_mainloop_fusion_multistage.h usr/include/cutlass/gemm/threadblock/mma_multistage.h usr/include/cutlass/gemm/threadblock/mma_pipelined.h usr/include/cutlass/gemm/threadblock/mma_planar_complex_base.h usr/include/cutlass/gemm/threadblock/mma_planar_complex_multistage.h usr/include/cutlass/gemm/threadblock/mma_planar_complex_pipelined.h usr/include/cutlass/gemm/threadblock/mma_singlestage.h usr/include/cutlass/gemm/threadblock/mma_softmax_mainloop_fusion_multistage.h usr/include/cutlass/gemm/threadblock/mma_sparse_base.h usr/include/cutlass/gemm/threadblock/mma_sparse_multistage.h usr/include/cutlass/gemm/threadblock/mma_with_reduction_multistage.h usr/include/cutlass/gemm/threadblock/threadblock_swizzle.h usr/include/cutlass/gemm/threadblock/threadblock_swizzle_streamk.h usr/include/cutlass/gemm/warp usr/include/cutlass/gemm/warp/default_mma_complex_tensor_op.h usr/include/cutlass/gemm/warp/default_mma_sparse_tensor_op.h usr/include/cutlass/gemm/warp/default_mma_tensor_op.h usr/include/cutlass/gemm/warp/default_mma_tensor_op_sm80.h usr/include/cutlass/gemm/warp/default_mma_with_reduction_tensor_op.h usr/include/cutlass/gemm/warp/default_mma_wmma_tensor_op.h usr/include/cutlass/gemm/warp/layernorm_scale_bias_transform.h usr/include/cutlass/gemm/warp/mma.h usr/include/cutlass/gemm/warp/mma_complex_tensor_op.h usr/include/cutlass/gemm/warp/mma_complex_tensor_op_fast_f32.h usr/include/cutlass/gemm/warp/mma_complex_tensor_op_tile_iterator_sm80.h usr/include/cutlass/gemm/warp/mma_gaussian_complex_tensor_op.h usr/include/cutlass/gemm/warp/mma_gaussian_complex_tensor_op_tile_iterator_sm80.h usr/include/cutlass/gemm/warp/mma_mixed_input_tensor_op.h usr/include/cutlass/gemm/warp/mma_planar_complex.h usr/include/cutlass/gemm/warp/mma_simt.h usr/include/cutlass/gemm/warp/mma_simt_policy.h usr/include/cutlass/gemm/warp/mma_simt_tile_iterator.h usr/include/cutlass/gemm/warp/mma_sparse_tensor_op.h usr/include/cutlass/gemm/warp/mma_tensor_op.h usr/include/cutlass/gemm/warp/mma_tensor_op_fast_f32.h usr/include/cutlass/gemm/warp/mma_tensor_op_fragment_iterator.h usr/include/cutlass/gemm/warp/mma_tensor_op_policy.h usr/include/cutlass/gemm/warp/mma_tensor_op_sm70.h usr/include/cutlass/gemm/warp/mma_tensor_op_tile_access_iterator.h usr/include/cutlass/gemm/warp/mma_tensor_op_tile_iterator.h usr/include/cutlass/gemm/warp/mma_tensor_op_tile_iterator_sm70.h usr/include/cutlass/gemm/warp/mma_tensor_op_tile_iterator_sm80.h usr/include/cutlass/gemm/warp/mma_tensor_op_tile_iterator_sparse.h usr/include/cutlass/gemm/warp/mma_tensor_op_tile_iterator_wmma.h usr/include/cutlass/gemm/warp/mma_tensor_op_wmma.h usr/include/cutlass/gemm/warp/mma_with_reduction_tensor_op.h usr/include/cutlass/gemm/warp/scale_bias_tile_iterator.h usr/include/cutlass/gemm/warp/softmax_scale_bias_transform.h usr/include/cutlass/gemm/warp/tile_iterator_planar_complex.h usr/include/cutlass/gemm_coord.h usr/include/cutlass/gemm_coord.hpp usr/include/cutlass/half.h usr/include/cutlass/integer_subbyte.h usr/include/cutlass/kernel_hardware_info.h usr/include/cutlass/kernel_hardware_info.hpp usr/include/cutlass/kernel_launch.h usr/include/cutlass/layout usr/include/cutlass/layout/layout.h usr/include/cutlass/layout/matrix.h usr/include/cutlass/layout/permute.h usr/include/cutlass/layout/pitch_linear.h usr/include/cutlass/layout/tensor.h usr/include/cutlass/layout/tensor_op_multiplicand_sm70.h usr/include/cutlass/layout/tensor_op_multiplicand_sm75.h usr/include/cutlass/layout/tensor_op_multiplicand_sm80.h usr/include/cutlass/layout/vector.h usr/include/cutlass/matrix.h usr/include/cutlass/matrix_coord.h usr/include/cutlass/matrix_shape.h usr/include/cutlass/numeric_conversion.h usr/include/cutlass/numeric_size.h usr/include/cutlass/numeric_types.h usr/include/cutlass/pipeline usr/include/cutlass/pipeline/pipeline.hpp usr/include/cutlass/pipeline/sm90_pipeline.hpp usr/include/cutlass/pitch_linear_coord.h usr/include/cutlass/platform usr/include/cutlass/platform/platform.h usr/include/cutlass/predicate_vector.h usr/include/cutlass/quaternion.h usr/include/cutlass/real.h usr/include/cutlass/reduction usr/include/cutlass/reduction/device usr/include/cutlass/reduction/device/reduce_split_k.h usr/include/cutlass/reduction/device/tensor_reduce.h usr/include/cutlass/reduction/device/tensor_reduce_affine_contiguous.h usr/include/cutlass/reduction/device/tensor_reduce_affine_strided.h usr/include/cutlass/reduction/kernel usr/include/cutlass/reduction/kernel/reduce_softmax_final.h usr/include/cutlass/reduction/kernel/reduce_split_k.h usr/include/cutlass/reduction/kernel/tensor_reduce_affine_contiguous.h usr/include/cutlass/reduction/kernel/tensor_reduce_affine_strided.h usr/include/cutlass/reduction/thread usr/include/cutlass/reduction/thread/reduce.h usr/include/cutlass/reduction/thread/reduction_operators.h usr/include/cutlass/reduction/threadblock_swizzle.h usr/include/cutlass/relatively_equal.h usr/include/cutlass/semaphore.h usr/include/cutlass/subbyte_reference.h usr/include/cutlass/tensor_coord.h usr/include/cutlass/tensor_ref.h usr/include/cutlass/tensor_ref_planar_complex.h usr/include/cutlass/tensor_view.h usr/include/cutlass/tensor_view_planar_complex.h usr/include/cutlass/tfloat32.h usr/include/cutlass/thread usr/include/cutlass/thread/matrix.h usr/include/cutlass/trace.h usr/include/cutlass/transform usr/include/cutlass/transform/collective usr/include/cutlass/transform/collective/sm90_wgmma_transpose.hpp usr/include/cutlass/transform/pitch_linear_thread_map.h usr/include/cutlass/transform/thread usr/include/cutlass/transform/thread/transpose.h usr/include/cutlass/transform/thread/unary_op.h usr/include/cutlass/transform/threadblock usr/include/cutlass/transform/threadblock/ell_iterator.h usr/include/cutlass/transform/threadblock/ell_predicated_tile_access_iterator.h usr/include/cutlass/transform/threadblock/ell_predicated_tile_iterator.h usr/include/cutlass/transform/threadblock/predicated_scale_bias_vector_access_iterator.h usr/include/cutlass/transform/threadblock/predicated_scale_bias_vector_iterator.h usr/include/cutlass/transform/threadblock/predicated_tile_access_iterator.h usr/include/cutlass/transform/threadblock/predicated_tile_access_iterator_2dthreadtile.h usr/include/cutlass/transform/threadblock/predicated_tile_access_iterator_params.h usr/include/cutlass/transform/threadblock/predicated_tile_access_iterator_triangular_matrix.h usr/include/cutlass/transform/threadblock/predicated_tile_iterator.h usr/include/cutlass/transform/threadblock/predicated_tile_iterator_2dthreadtile.h usr/include/cutlass/transform/threadblock/predicated_tile_iterator_triangular_matrix.h usr/include/cutlass/transform/threadblock/predicated_vector_access_iterator.h usr/include/cutlass/transform/threadblock/regular_scale_bias_vector_access_iterator.h usr/include/cutlass/transform/threadblock/regular_tile_access_iterator.h usr/include/cutlass/transform/threadblock/regular_tile_access_iterator_pitch_linear.h usr/include/cutlass/transform/threadblock/regular_tile_access_iterator_pitch_linear_direct_conv.h usr/include/cutlass/transform/threadblock/regular_tile_access_iterator_tensor_op.h usr/include/cutlass/transform/threadblock/regular_tile_access_iterator_tensor_op_sm80.h usr/include/cutlass/transform/threadblock/regular_tile_iterator.h usr/include/cutlass/transform/threadblock/regular_tile_iterator_pitch_linear.h usr/include/cutlass/transform/threadblock/regular_tile_iterator_pitch_linear_2dthreadtile.h usr/include/cutlass/transform/threadblock/regular_tile_iterator_tensor_op.h usr/include/cutlass/transform/threadblock/regular_tile_iterator_tensor_op_sm70.h usr/include/cutlass/transform/threadblock/vector_iterator.h usr/include/cutlass/transform/warp usr/include/cutlass/transform/warp/vector_fragment_iterator.h usr/include/cutlass/uint128.h usr/include/cutlass/util usr/include/cutlass/util/GPU_Clock.hpp usr/include/cutlass/util/command_line.h usr/include/cutlass/util/cublas_wrappers.hpp usr/include/cutlass/util/debug.h usr/include/cutlass/util/device_dump.h usr/include/cutlass/util/device_groupnorm.h usr/include/cutlass/util/device_layernorm.h usr/include/cutlass/util/device_memory.h usr/include/cutlass/util/device_nchw_to_nhwc.h usr/include/cutlass/util/device_nhwc_padding.h usr/include/cutlass/util/device_nhwc_pooling.h usr/include/cutlass/util/device_nhwc_to_nchw.h usr/include/cutlass/util/device_rmsnorm.h usr/include/cutlass/util/device_utils.h usr/include/cutlass/util/distribution.h usr/include/cutlass/util/exceptions.h usr/include/cutlass/util/gett_commandline.hpp usr/include/cutlass/util/helper_cuda.hpp usr/include/cutlass/util/host_reorder.h usr/include/cutlass/util/host_tensor.h usr/include/cutlass/util/host_tensor_planar_complex.h usr/include/cutlass/util/host_uncompress.h usr/include/cutlass/util/index_sequence.h usr/include/cutlass/util/packed_stride.hpp usr/include/cutlass/util/print_error.hpp usr/include/cutlass/util/reference usr/include/cutlass/util/reference/detail usr/include/cutlass/util/reference/detail/inner_product.h usr/include/cutlass/util/reference/detail/linear_to_coordinate.h usr/include/cutlass/util/reference/device usr/include/cutlass/util/reference/device/convolution.h usr/include/cutlass/util/reference/device/gemm.h usr/include/cutlass/util/reference/device/gemm_complex.h usr/include/cutlass/util/reference/device/gemm_planar_complex.h usr/include/cutlass/util/reference/device/gett.hpp usr/include/cutlass/util/reference/device/kernel usr/include/cutlass/util/reference/device/kernel/gemm.h usr/include/cutlass/util/reference/device/kernel/tensor_elementwise.h usr/include/cutlass/util/reference/device/kernel/tensor_foreach.h usr/include/cutlass/util/reference/device/rank_2k_complex.h usr/include/cutlass/util/reference/device/tensor_compare.h usr/include/cutlass/util/reference/device/tensor_fill.h usr/include/cutlass/util/reference/device/tensor_foreach.h usr/include/cutlass/util/reference/device/tensor_reduce.h usr/include/cutlass/util/reference/device/tensor_relu.h usr/include/cutlass/util/reference/device/thread usr/include/cutlass/util/reference/device/thread/gemm.h usr/include/cutlass/util/reference/host usr/include/cutlass/util/reference/host/convolution.h usr/include/cutlass/util/reference/host/error_metrics.h usr/include/cutlass/util/reference/host/gemm.h usr/include/cutlass/util/reference/host/gemm_complex.h usr/include/cutlass/util/reference/host/gemm_planar_complex.h usr/include/cutlass/util/reference/host/gett.hpp usr/include/cutlass/util/reference/host/rank_2k.h usr/include/cutlass/util/reference/host/rank_2k_complex.h usr/include/cutlass/util/reference/host/rank_k_complex.h usr/include/cutlass/util/reference/host/symm.h usr/include/cutlass/util/reference/host/symm_complex.h usr/include/cutlass/util/reference/host/tensor_compare.h usr/include/cutlass/util/reference/host/tensor_compare.hpp usr/include/cutlass/util/reference/host/tensor_copy.h usr/include/cutlass/util/reference/host/tensor_elementwise.h usr/include/cutlass/util/reference/host/tensor_fill.h usr/include/cutlass/util/reference/host/tensor_fill.hpp usr/include/cutlass/util/reference/host/tensor_foreach.h usr/include/cutlass/util/reference/host/tensor_norm.h usr/include/cutlass/util/reference/host/tensor_reduce.h usr/include/cutlass/util/reference/host/tensor_reduce.hpp usr/include/cutlass/util/reference/host/trmm.h usr/include/cutlass/util/reference/host/trmm_complex.h usr/include/cutlass/util/tensor_view_io.h usr/include/cutlass/util/type_traits.h usr/include/cutlass/version.h usr/include/cutlass/version_extended.h usr/include/cutlass/wmma_array.h usr/include/cutlass/workspace.h usr/share usr/share/doc usr/share/doc/libcutlass-dev usr/share/doc/libcutlass-dev/changelog.Debian.gz usr/share/doc/libcutlass-dev/copyright