MCPcopy Create free account

hub / github.com/Tencent/hpc-ops / functions

Functions10,542 in github.com/Tencent/hpc-ops

⨍ Functions 10,542◇ Types & classes 8,025↳ Endpoints 13
↓ 8 callersMethodbegin_loop
Start of subtile store iteration
3rd/cutlass/include/cutlass/epilogue/fusion/sm90_visitor_tma_warpspecialized.hpp:195
↓ 8 callersMethodbegin_sync_needed
Is a thread sync needed after begin(). Allows chaining async copies across multiple nodes
3rd/cutlass/include/cutlass/epilogue/fusion/sm90_visitor_tma_warpspecialized.hpp:185
↓ 8 callersFunctioncall_fn
()
benchmark/attention_decode/bench_attention_decode_fp8.py:328
↓ 8 callersMethodcapacity
Compute the number of contiguous elements needed to store a tensor with the given size
3rd/cutlass/include/cutlass/layout/tensor_op_multiplicand_sm80.h:136
↓ 8 callersMethodclear_mask
Clears the predicate set efficiently
3rd/cutlass/include/cutlass/transform/threadblock/predicated_tile_access_iterator.h:257
↓ 8 callersFunctioncluster_arrive_relaxed
3rd/cutlass/include/cute/arch/cluster_sm90.hpp:48
↓ 8 callersFunctioncopy_tiles_and_advance
3rd/cutlass/include/cutlass/gemm/threadblock/mma_multistage.h:287
↓ 8 callersMethodell_add_mask
add mask for small tiles in ELL
3rd/cutlass/include/cutlass/transform/threadblock/ell_predicated_tile_iterator.h:865
↓ 8 callersMethodenable_mask
Clears the predicate set efficiently
3rd/cutlass/include/cutlass/transform/threadblock/predicated_tile_access_iterator.h:267
↓ 8 callersMethodend_loop
End of subtile store iteration
3rd/cutlass/include/cutlass/epilogue/fusion/sm90_visitor_tma_warpspecialized.hpp:265
↓ 8 callersFunctionfilter_tuple
3rd/cutlass/include/cute/algorithm/tuple_algorithms.hpp:345
↓ 8 callersFunctionflatten_to_tuple
3rd/cutlass/include/cute/algorithm/tuple_algorithms.hpp:544
↓ 8 callersMethodget
Returns a pointer
3rd/cutlass/include/cutlass/transform/threadblock/regular_tile_access_iterator_tensor_op_sm80.h:540
↓ 8 callersMethodget_k
3rd/cutlass/include/cutlass/transform/threadblock/ell_predicated_tile_access_iterator.h:888
↓ 8 callersMethodget_mask
Gets the mask
3rd/cutlass/include/cutlass/transform/threadblock/predicated_tile_access_iterator.h:286
↓ 8 callersMethodget_sk_block_idx
Obtains calling linear threadblock index of the first block to work on the given tile
3rd/cutlass/include/cutlass/gemm/threadblock/threadblock_swizzle_streamk.h:732
↓ 8 callersFunctionget_tiled_cta_shape_mnl
3rd/cutlass/include/cutlass/gemm/kernel/sm100_tile_scheduler_stream_k.hpp:233
↓ 8 callersFunctionispow2
Returns true if the argument is a power of 2
3rd/cutlass/include/cutlass/array.h:81
↓ 8 callersMethodload_ab
3rd/cutlass/include/cutlass/gemm/collective/sm100_mma_warpspecialized_blockwise_scaling.hpp:832
↓ 8 callersFunctionload_with_byte_offset
Loads a fragment from memory with additional logical offset
3rd/cutlass/include/cutlass/gemm/warp/mma_tensor_op_tile_iterator_sm80.h:270
↓ 8 callersMethodload_with_byte_offset
Loads a fragment from memory with additional logical offset
3rd/cutlass/include/cutlass/gemm/warp/mma_complex_tensor_op_tile_iterator_sm80.h:1006
↓ 8 callersMethodload_with_byte_offset
Loads a fragment from memory with additional logical offset
3rd/cutlass/include/cutlass/gemm/warp/tile_iterator_planar_complex.h:164
↓ 8 callersMethodload_with_byte_offset
Loads a fragment from memory with additional logical offset
3rd/cutlass/include/cutlass/gemm/warp/mma_tensor_op_tile_iterator_sm70.h:1358
↓ 8 callersMethodload_with_byte_offset
Loads a fragment from memory with additional logical offset
3rd/cutlass/include/cutlass/gemm/warp/mma_tensor_op_tile_iterator.h:3386
↓ 8 callersMethodload_with_pointer_offset
Loads a fragment from memory with additional logical offset
3rd/cutlass/include/cutlass/gemm/warp/tile_iterator_planar_complex.h:176
↓ 8 callersFunctionmma
3rd/cutlass/include/cutlass/conv/collective/sm100_implicit_gemm_umma_warpspecialized.hpp:838
↓ 8 callersFunctionouter_partition
3rd/cutlass/include/cute/tensor_impl.hpp:1009
↓ 8 callersMethodreduction_subtile_idx
3rd/cutlass/include/cutlass/gemm/kernel/static_tile_scheduler.hpp:79
↓ 8 callersMethodscale_residue_if_needed
3rd/cutlass/include/cutlass/gemm/collective/fp8_accumulation.hpp:264
↓ 8 callersMethodset_iteration_index
Overrides the internal iteration index
3rd/cutlass/include/cutlass/transform/threadblock/regular_tile_access_iterator_tensor_op_sm80.h:525
↓ 8 callersMethodset_mask
Sets the predicate mask, overriding value stored in predicate iterator
3rd/cutlass/include/cutlass/transform/threadblock/predicated_tile_access_iterator.h:276
↓ 8 callersFunctionsize
3rd/cutlass/include/cute/atom/copy_atom.hpp:607
↓ 8 callersFunctionstore
Stores a fragment to memory at the location pointed to by the iterator
3rd/cutlass/include/cutlass/gemm/warp/mma_simt_tile_iterator.h:525
↓ 8 callersFunctionsum
3rd/cutlass/include/cute/int_tuple.hpp:286
↓ 8 callersMethodtransposed_problem
Returns arguments for the transposed problem
3rd/cutlass/include/cutlass/gemm/kernel/gemm_with_absmax.h:217
↓ 8 callersFunctionunzip_tensor
3rd/cutlass/include/cute/tensor_zip.hpp:166
↓ 7 callersFunctionLayoutAwareConvert
3rd/cutlass/include/cutlass/detail/collective/mixed_input_utils.hpp:456
↓ 7 callersFunctionabs_for_integer
3rd/cutlass/include/cutlass/fast_math.h:158
↓ 7 callersMethodadd_pointer_offset
Adds a pointer offset in units of Element
3rd/cutlass/include/cutlass/transform/threadblock/predicated_tile_iterator.h:1406
↓ 7 callersMethodarrive
Signal completion of Stage and move to the next stage (group_id) signals to (group_id+1)
3rd/cutlass/include/cutlass/pipeline/sm90_pipeline.hpp:1339
↓ 7 callersFunctionbit_width
3rd/cutlass/include/cute/numeric/math.hpp:145
↓ 7 callersFunctionbuild_a_scale
()
benchmark/fused_moe/backends/base.py:177
↓ 7 callersFunctionceil_div
3rd/cutlass/include/cute/layout.hpp:1592
↓ 7 callersMethodclear_mask
Clears the predicate set efficiently
3rd/cutlass/include/cutlass/transform/threadblock/predicated_tile_iterator.h:1435
↓ 7 callersFunctionconditional_return
3rd/cutlass/include/cute/numeric/integral_constant.hpp:432
↓ 7 callersFunctioncrd2idx
3rd/cutlass/include/cute/layout.hpp:677
↓ 7 callersFunctioncrd2idx
3rd/cutlass/include/cute/stride.hpp:100
↓ 7 callersMethoddata
Returns a pointer to the shared memory buffer
3rd/cutlass/include/cutlass/epilogue/threadblock/epilogue_base.h:175
↓ 7 callersMethoddeterminant
Computes the determinant of a 2-by-2 matrix
3rd/cutlass/include/cutlass/matrix.h:3231
↓ 7 callersMethoddiv_cluster_size
Divides dividend by the cluster size
3rd/cutlass/include/cutlass/gemm/kernel/tile_scheduler_params.h:476
↓ 7 callersMethodenable_mask
Clears the predicate set efficiently
3rd/cutlass/include/cutlass/transform/threadblock/predicated_tile_iterator.h:1441
↓ 7 callersFunctionexplode_tuple
3rd/cutlass/include/cute/arch/util.hpp:284
↓ 7 callersFunctionfast_min
3rd/cutlass/include/cutlass/fast_math.h:763
↓ 7 callersFunctionfence_view_async_tmem_store
3rd/cutlass/include/cutlass/arch/barrier.h:936
↓ 7 callersFunctionfma
3rd/cutlass/include/cute/numeric/real.hpp:58
↓ 7 callersFunctionfp4_shift_A
3rd/cutlass/include/cute/atom/mma_traits_sm120.hpp:203
↓ 7 callersFunctionfp4_shift_B
3rd/cutlass/include/cute/atom/mma_traits_sm120.hpp:207
↓ 7 callersFunctionget
Returns a pointer
3rd/cutlass/include/cutlass/conv/threadblock/predicated_scale_bias_vector_access_iterator.h:240
↓ 7 callersMethodget_log_tile
Calculates optimal swizzle width
3rd/cutlass/include/cutlass/gemm/threadblock/threadblock_swizzle.h:116
↓ 7 callersMethodget_mask
Gets the mask
3rd/cutlass/include/cutlass/transform/threadblock/predicated_tile_iterator.h:1453
↓ 7 callersFunctionget_sm_count
src/utils/utils.cc:15
↓ 7 callersFunctionhas_single_bit
3rd/cutlass/include/cute/numeric/math.hpp:127
↓ 7 callersFunctioninit_masks
3rd/cutlass/include/cutlass/pipeline/sm100_pipeline.hpp:392
↓ 7 callersMethodlayout
Returns the layout object
3rd/cutlass/include/cutlass/tensor_ref.h:285
↓ 7 callersFunctionmake_basis_like
3rd/cutlass/include/cute/numeric/arithmetic_tuple.hpp:319
↓ 7 callersFunctionpartition_accumulator_shape
Construct A Single Stage's Accumulator Shape
3rd/cutlass/include/cutlass/gemm/collective/sm100_mma_array_warpspecialized.hpp:463
↓ 7 callersMethodpure
3rd/cutlass/include/cutlass/quaternion.h:201
↓ 7 callersMethodset_mask
Sets the predicate mask, overriding value stored in predicate iterator
3rd/cutlass/include/cutlass/transform/threadblock/predicated_tile_iterator.h:1447
↓ 7 callersFunctionshared_load
3rd/cutlass/include/cutlass/arch/memory_sm75.h:231
↓ 7 callersMethodsync
3rd/cutlass/include/cutlass/epilogue/collective/sm70_epilogue_vectorized.hpp:296
↓ 7 callersMethodtensormaps_perform_update
3rd/cutlass/include/cutlass/gemm/collective/sm90_mma_array_tma_gmma_ss_warpspecialized.hpp:714
↓ 7 callersFunctiontensormaps_replace_global_tensor_properties
3rd/cutlass/include/cutlass/gemm/collective/sm100_mma_array_warpspecialized.hpp:820
↓ 7 callersFunctionto_string
3rd/cutlass/include/cute/arch/mma_sm90_desc.hpp:55
↓ 7 callersMethodtry_wait
3rd/cutlass/include/cutlass/arch/barrier.h:365
↓ 7 callersFunctionvalid
Returns whether access is valid or not
3rd/cutlass/include/cutlass/conv/threadblock/predicated_scale_bias_vector_access_iterator.h:274
↓ 7 callersFunctionzip
3rd/cutlass/include/cute/layout.hpp:1515
↓ 6 callersMethodRecv
src/communicator/channel.cc:52
↓ 6 callersMethodSend
src/communicator/channel.cc:33
↓ 6 callersFunctionSmId
Computes SM number the thread is running on
3rd/cutlass/include/cutlass/arch/arch.h:58
↓ 6 callersFunction_gumbel0_like
External Gumbel(0) noise matching the kernel convention: g = -log(-log(U)), U ~ Uniform(0, 1]. Always float32 with the logits' [B, V] shape —
tests/test_sampler.py:35
↓ 6 callersFunctionadd_pointer_offset
Adds a pointer offset in units of Element
3rd/cutlass/include/cutlass/transform/threadblock/regular_tile_iterator_tensor_op_sm70.h:166
↓ 6 callersMethodcall
3rd/cutlass/include/cute/atom/mma_atom.hpp:92
↓ 6 callersFunctioncan_implement
Determines whether the Implicit GEMM can execute the given problem.
3rd/cutlass/include/cutlass/conv/threadblock/conv3d_fprop_filter_tile_access_iterator_analytic.h:239
↓ 6 callersMethodcapacity
Compute the number of contiguous elements needed to store a tensor with the given size
3rd/cutlass/include/cutlass/layout/tensor_op_multiplicand_sm70.h:188
↓ 6 callersMethodcapacity
Compute the number of contiguous elements needed to store a tensor with the given size
3rd/cutlass/include/cutlass/layout/tensor_op_multiplicand_sm75.h:216
↓ 6 callersFunctioncompatible
3rd/cutlass/include/cute/int_tuple.hpp:486
↓ 6 callersFunctioncomplement
3rd/cutlass/include/cute/layout.hpp:1178
↓ 6 callersFunctioncompute_lower_corner_whd
3rd/cutlass/include/cutlass/conv/collective/detail.hpp:188
↓ 6 callersFunctioncompute_source_not_needed_
Streams the result to global memory
3rd/cutlass/include/cutlass/epilogue/threadblock/epilogue_direct_store.h:275
↓ 6 callersFunctioncompute_upper_corner_whd
3rd/cutlass/include/cutlass/conv/collective/detail.hpp:212
↓ 6 callersMethodconsumer_wait
3rd/cutlass/include/cutlass/pipeline/sm100_pipeline.hpp:1160
↓ 6 callersFunctioncopy_tiles_and_advance
3rd/cutlass/include/cutlass/conv/threadblock/implicit_gemm_multistage.h:84
↓ 6 callersFunctiondice
3rd/cutlass/include/cute/layout.hpp:706
↓ 6 callersMethodexponent_biased
Returns the biased exponent
3rd/cutlass/include/cutlass/tfloat32.h:173
↓ 6 callersMethodexponent_biased
Returns the biased exponent
3rd/cutlass/include/cutlass/half.h:475
↓ 6 callersMethodexponent_biased
Returns the biased exponent
3rd/cutlass/include/cutlass/bfloat16.h:211
↓ 6 callersFunctionfast_tanh
3rd/cutlass/include/cutlass/fast_math.h:921
↓ 6 callersFunctionget_block_shape
3rd/cutlass/include/cutlass/transform/kernel/sm90_sparse_gemm_compressor.hpp:203
↓ 6 callersMethodget_current_work
3rd/cutlass/include/cutlass/gemm/kernel/static_tile_scheduler.hpp:163
↓ 6 callersMethodget_grid_shape
Computes the grid shape
3rd/cutlass/include/cutlass/conv/device/conv_universal_adapter.hpp:174
← previousnext →401–500 of 10,542, ranked by callers