Functions — github.com/bytedance/flux

CodeHub MCPcopy Create free account

hub / github.com/bytedance/flux / functions

Functions784 in github.com/bytedance/flux

⨍ Functions 784◇ Types & classes 309

Functionatomic_store_sys

include/flux/cuda/cuda_common_device.hpp:93

Methodauto_comm_spec

include/flux/gemm_hparams.h:289

Methodauto_gemm_kind

include/flux/gemm_hparams.h:325

Methodauto_mainloop_stage

include/flux/gemm_hparams.h:335

Methodauto_raster_order

include/flux/gemm_hparams.h:353

Methodauto_tile_shape

include/flux/gemm_hparams.h:295

Functionbarrier_all

pynvshmem/src/functions.cpp:68

Functionbarrier_on_stream

pynvshmem/src/functions.cpp:74

Methodbegin_step

src/reduce_scatter/epilogue_evt.hpp:234

Methodbegin_step

src/reduce_scatter/epilogue_evt_nvshmem.hpp:224

Methodbegin_step

src/reduce_scatter/gemmk_visitor_load.hpp:127

Functionbitwise_check

src/ths_op/helper_ops.cc:26

Functionbootstrap_c10_pg_allgather

pynvshmem/bootstrap/torch/bootstrap_torch.cpp:49

Functionbootstrap_c10_pg_alltoall

pynvshmem/bootstrap/torch/bootstrap_torch.cpp:78

Functionbootstrap_c10_pg_barrier

pynvshmem/bootstrap/torch/bootstrap_torch.cpp:36

Functionbootstrap_c10_pg_finalize

pynvshmem/bootstrap/torch/bootstrap_torch.cpp:114

Functionbootstrap_c10_pg_global_exit

pynvshmem/bootstrap/torch/bootstrap_torch.cpp:108

Methodbuffer_ptr

src/reduce_scatter/reduce_scatter_kernel.hpp:479

Methodbuffer_tile

src/reduce_scatter/reduce_scatter_kernel.hpp:336

Methodcacheline_align_up

Pad the given allocation size up to the nearest cache line

src/all_gather/sm80_all_gather_gemm.hpp:331

Methodcacheline_align_up

Pad the given allocation size up to the nearest cache line

src/reduce_scatter/epilogue_evt.hpp:169

Methodcacheline_align_up

Pad the given allocation size up to the nearest cache line

src/reduce_scatter/epilogue_evt_nvshmem.hpp:145

Methodcacheline_align_up

Pad the given allocation size up to the nearest cache line

src/reduce_scatter/gemmk_universal_with_visitor_streamk.h:278

Methodcan_implement

Determines whether the GEMM problem size satisfies this kernel's alignment requirements

src/all_gather/sm80_all_gather_gemm.hpp:560

Methodcan_implement

src/all_gather/sm90_all_gather_gemm_tma_warpspecialized_cooperative.hpp:291

Methodcan_implement

src/reduce_scatter/epilogue_reduce_scatter.hpp:136

Methodcan_implement

src/reduce_scatter/epilogue_vectorized_reduce_scatter.hpp:146

Methodcan_implement

src/reduce_scatter/epilogue_nvshmem_reduce_scatter.hpp:158

Methodcan_implement

Determines whether the GEMM problem size satisfies this kernel's alignment requirements

src/reduce_scatter/gemmk_universal_with_visitor_streamk.h:509

Methodcan_implement

src/reduce_scatter/sm90_gemm_tma_warpspecialized_cooperative_reduce_scatter.hpp:239

Methodcheck_length

include/flux/flux.h:960

Methodcheck_type

include/flux/gemm_meta.h:166

Methodcheck_type

include/flux/gemm_meta.h:252

Methodcheck_value

Uses thread[0] to wait for at least the specified count of signals on the given flag counter

include/flux/cuda/system_barrier.hpp:57

FunctioncloseIpcMemHandleCallback

src/ths_op/ths_op.cc:84

Methodconstexpr

include/flux/cuda/cutlass_v3_builder.hpp:345

src/reduce_scatter/reduce_scatter_kernel.hpp:453

include/flux/cuda/memory_utils.hpp:259

include/flux/cuda/memory_utils.hpp:282

include/flux/cuda/memory_utils.hpp:306

include/flux/cuda/memory_utils.hpp:337

Methodcopy_2d_ring

src/all_gather/ths_op/all_gather_gemm_kernel_crossnode.cc:565

Methodcopy_all2all

src/all_gather/ths_op/all_gather_gemm_kernel_crossnode.cc:422

Methodcopy_all_to_all

src/all_gather/ths_op/all_gather_gemm_kernel.cc:817

Methodcopy_local

src/all_gather/ths_op/all_gather_gemm_kernel_crossnode.cc:393

Methodcopy_local_ring

src/all_gather/ths_op/all_gather_gemm_kernel_crossnode.cc:672

Methodcopy_remote

src/reduce_scatter/epilogue_evt_nvshmem.hpp:300

Methodcopy_ring_1d_push

src/all_gather/ths_op/all_gather_gemm_kernel_crossnode.cc:734

Methodcopy_ring_pull

src/all_gather/ths_op/all_gather_gemm_kernel.cc:837

Methodcopy_ring_push_1d

src/all_gather/ths_op/all_gather_gemm_kernel.cc:863

Methodcopy_ring_push_2d_pcie

src/all_gather/ths_op/all_gather_gemm_kernel.cc:885

Methodcopy_to_next_node

src/all_gather/ths_op/all_gather_gemm_kernel_crossnode.cc:539

Methodcopy_unpack

include/flux/cuda/memory_utils.hpp:411

Functioncudaipc_barrier_all_on_stream

src/ths_op/helper_ops.cc:45

Methoddata_tile

src/reduce_scatter/reduce_scatter_kernel.hpp:408

Methoddebug_print_v

src/reduce_scatter/sm90_epilogue_evt.hpp:238

Methoddebug_print_v

src/reduce_scatter/sm90_reduce_scatter_utils.hpp:217

Methoddefault_gemm_kernel

include/flux/cuda/gemm_impls/gemm_grouped_impl.hpp:108

Methoddefault_tile_scheduler

include/flux/cuda/gemm_impls/gemm_v3_impl.hpp:56

include/flux/cuda/gemm_impls/gemm_operator_base_default_impl.hpp:46

Methoddo_epilogue

Perform epilogue computations and output

src/all_gather/sm80_all_gather_gemm.hpp:739

Methoddo_epilogue

Perform epilogue computations and output

src/reduce_scatter/gemmk_universal_with_visitor_streamk.h:712

Methodelapsed_millis

Return the elapsed time (in milliseconds)

include/flux/cuda/cuda_common.h:195

Methodend_epilogue

src/reduce_scatter/epilogue_evt.hpp:328

Methodend_epilogue

src/reduce_scatter/epilogue_evt_nvshmem.hpp:306

src/reduce_scatter/epilogue_evt.hpp:257

src/reduce_scatter/epilogue_evt_nvshmem.hpp:264

Methodenumerate_split_meta_hparams_pairs

src/comm_none/gemm_v2_comm_none.hpp:210

Methodenumerate_split_meta_hparams_pairs

include/flux/gemm_operator_base.h:71

include/flux/flux.h:905

Functionestimate_gemm_sol_time_ms

refer to this: https://arnon.dk/matching-sm-architectures-arch-and-gencode-for-various-nvidia-cards/

python/flux/util.py:44

Functionexec_in_rank_order

(group: torch.distributed.ProcessGroup, func: Callable)

python/flux/dist_utils.py:79

Methodfetch_next_work

src/all_gather/sm90_all_gather_gemm_tma_warpspecialized_cooperative.hpp:773

Methodfetch_next_work

src/reduce_scatter/sm90_gemm_tma_warpspecialized_cooperative_reduce_scatter.hpp:735

include/flux/flux.h:955

Methodfilter_fast_accum

include/flux/gemm_meta.h:386

Functionfilter_kernel_schedule

include/flux/gemm_hparams.h:433

Methodfilter_layout

include/flux/gemm_meta.h:401

src/reduce_scatter/reduce_scatter_kernel.hpp:310

Functionflux_create_tensor

src/ths_op/flux_shm.cc:170

( self, input: torch.Tensor, weight: torch.Tensor, bias: torch.Tensor )

python/flux/gemm_rs_sm80.py:135

src/all_gather/ths_op/all_gather_gemm_kernel_crossnode.cc:376

src/reduce_scatter/ths_op/gemm_reduce_scatter.cc:628

Methodforward_allgather

src/all_gather/ths_op/all_gather_gemm_kernel.cc:529

Methodforward_barrier

src/reduce_scatter/ths_op/gemm_reduce_scatter.cc:607

Methodforward_gemm

src/all_gather/ths_op/all_gather_gemm_kernel.cc:592

Methodforward_gemm

src/reduce_scatter/ths_op/gemm_reduce_scatter.cc:587

Methodforward_gemm_impl

src/all_gather/ths_op/all_gather_gemm_kernel.cc:440

Methodforward_gemm_impl

src/reduce_scatter/ths_op/gemm_reduce_scatter.cc:345

Methodforward_impl

src/all_gather/ths_op/all_gather_gemm_kernel_crossnode.cc:299

Methodforward_reduce_scatter

src/reduce_scatter/ths_op/gemm_reduce_scatter.cc:621

Methodforward_reduce_scatter_impl

src/reduce_scatter/ths_op/gemm_reduce_scatter.cc:481

Methodfrom_tuple

include/flux/flux.h:966

Methodfrom_tuple

include/flux/cuda/cutlass_v3_builder.hpp:96

Executes one GEMM

src/all_gather/sm80_all_gather_gemm.hpp:848

Executes one GEMM

src/reduce_scatter/gemmk_universal_with_visitor_streamk.h:829

Methodgemm_device

///////////////////// CRTP functions /////////////////////

include/flux/cuda/gemm_impls/gemm_grouped_impl.hpp:138

Methodgemm_device

///////////////////// CRTP functions /////////////////////

include/flux/cuda/gemm_impls/gemm_v2_impl.hpp:348

Methodgemm_kernel

src/comm_none/gemm_v2_comm_none.hpp:48

Methodgemm_kernel

src/all_gather/gemm_v2_ag_kernel.hpp:64

← previous next →401–500 of 784, ranked by callers