MCPcopy Create free account

hub / github.com/bytedance/flux / functions

Functions784 in github.com/bytedance/flux

↓ 1 callersFunctiontuple_for_each_impl
include/flux/flux.h:234
↓ 1 callersFunctiontuple_has_elem
include/flux/flux.h:446
↓ 1 callersFunctiontuple_transform_impl
include/flux/flux.h:218
↓ 1 callersFunctiontuple_unpack_cat_impl
include/flux/flux.h:412
MethodAGThreadblockSwizzleStreamKRankOffset
Constructor
src/all_gather/sm80_all_gather_gemm_threadblock_swizzle.hpp:48
MethodAppendCloseIpcMemHandleCallbackRAII
src/ths_op/ths_op.cc:97
MethodArguments
Default Constructor
src/all_gather/sm80_all_gather_gemm.hpp:155
MethodArguments
src/reduce_scatter/gemmk_universal_with_visitor_streamk.h:139
MethodCUTE_STATIC_ASSERT
TILE_N divides MMA_N bytedance::flux::StaticPrintInt<typename TiledCopyS2R::TiledNumThr{}>{}; bytedance::flux::StaticPrintInt<size<0>(typename TiledMm
src/reduce_scatter/epilogue_vectorized_reduce_scatter.hpp:250
MethodCUTE_STATIC_ASSERT
FIXME : l_coord
src/reduce_scatter/epilogue_nvshmem_reduce_scatter.hpp:302
MethodCheckFail
include/flux/flux.h:92
MethodCreateReduceScatterStream
src/reduce_scatter/ths_op/gemm_reduce_scatter.cc:199
MethodDistEnv
src/cuda/utils.cc:104
MethodDistEnvTP
src/ths_op/ths_op.cc:278
MethodDistEnvTPWithEP
src/ths_op/ths_op.cc:292
MethodDynamicLibrary
src/cuda/nvml_stub.cc:76
MethodDynamicLibrary
src/cuda/nvml_stub.cc:83
MethodDynamicLibrary
src/cuda/cuda_stub.cc:68
MethodDynamicLibrary
src/cuda/cuda_stub.cc:75
MethodFLUX_NAMED_TUPLE_DEFINE_FIELD
include/flux/gemm_meta.h:33
MethodFLUX_NAMED_TUPLE_DEFINE_FIELD
include/flux/gemm_meta.h:203
MethodFLUX_NAMED_TUPLE_DEFINE_FIELD
include/flux/gemm_meta.h:244
MethodFLUX_NAMED_TUPLE_DEFINE_FIELD
include/flux/gemm_hparams.h:49
MethodFLUX_NAMED_TUPLE_DEFINE_FIELD
include/flux/gemm_hparams.h:121
MethodFluxNamedTupleBase
include/flux/flux.h:998
MethodGpuTimer
Constructor
include/flux/cuda/cuda_common.h:170
MethodMoeArguments
src/ths_op/ths_op.cc:319
MethodOpRegistry
include/flux/op_registry.h:289
MethodOutputRankSwizzler
src/reduce_scatter/tile_scheduler/tile_mappings.hpp:53
FunctionPYBIND11_MODULE
pynvshmem/src/module.cpp:6
FunctionPYBIND11_MODULE
src/ths_op/ths_op.cc:420
MethodParams
Default constructor
src/all_gather/sm80_all_gather_gemm.hpp:361
MethodParams
Default constructor
src/reduce_scatter/gemmk_universal_with_visitor_streamk.h:308
MethodPersistentTileSchedulerSm90ReduceScatter
src/reduce_scatter/tile_scheduler/sm90_tile_scheduler_reduce_scatter.hpp:86
MethodPersistentTileSchedulerSm90ReduceScatterStreamK
src/reduce_scatter/tile_scheduler/sm90_tile_scheduler_reduce_scatter.hpp:154
MethodPyTuningRecord
src/ths_op/ths_op.cc:108
MethodReduceScatterRing1dPullGemmk
src/reduce_scatter/reduce_scatter_kernel.hpp:1230
MethodReduceScatterRing1dPushGemmk
src/reduce_scatter/reduce_scatter_kernel.hpp:1320
MethodReduceScatterRing2dPull
src/reduce_scatter/reduce_scatter_kernel.hpp:511
MethodReduceScatterRing2dPullGemmk
src/reduce_scatter/reduce_scatter_kernel.hpp:815
MethodReduceScatterRing2dPullPerWarp
src/reduce_scatter/reduce_scatter_kernel.hpp:640
MethodReduceScatterRing2dPullWithQueue
src/reduce_scatter/reduce_scatter_kernel.hpp:1081
MethodReduceScatterRing2dPushGemmk
src/reduce_scatter/reduce_scatter_kernel.hpp:939
MethodSm90AGKernelTileScheduler
src/all_gather/sm90_all_gather_gemm_tile_scheduler.hpp:128
MethodSm90AuxStoreReduceScatter
src/reduce_scatter/sm90_epilogue_evt.hpp:164
MethodSm90ReduceScatterDma
src/reduce_scatter/sm90_reduce_scatter_utils.hpp:209
MethodSymbol
src/cuda/nvml_stub.cc:88
MethodSymbol
src/cuda/cuda_stub.cc:84
MethodThreadblockSwizzleStreamKRankOffset
Constructor
src/reduce_scatter/tile_scheduler/threadblock_swizzle.hpp:39
MethodThreadblockSwizzleStreamKRankOffsetAcrossNode
Constructor
src/reduce_scatter/tile_scheduler/threadblock_swizzle_acrossnode.hpp:40
MethodThsOpsInitRegistry
include/flux/ths_op/ths_op.h:120
MethodTimer
include/flux/utils.h:96
MethodTuningConfigGenerator
include/flux/op_registry.h:71
MethodTuningConfigRegistry
include/flux/op_registry.h:123
MethodVisitorAuxLoadGemmk
src/reduce_scatter/gemmk_visitor_load.hpp:97
MethodVisitorAuxStoreScatter
src/reduce_scatter/epilogue_evt.hpp:158
MethodVisitorAuxStoreScatterAccrossNode
src/reduce_scatter/epilogue_evt_nvshmem.hpp:131
MethodWorkIdxMSwizzler
src/reduce_scatter/tile_scheduler/tile_mappings.hpp:82
Method__init__
( self, tp_group: dist.ProcessGroup, nnodes: int, max_m: int, n_dim: i
python/flux/gemm_rs_sm80.py:87
Method__init__
(self, deterministic: bool = True)
python/flux/dist_utils.py:27
Method__init__
( self, weight: torch.Tensor, tp_group, full_m: int, nnodes: int = 1,
python/flux/ag_gemm.py:69
Method__init__
: ring_mode[int] -1 for auto: nvlink machine default to 0, non-nvlink machine default to 2 0: nvlink mode: all-to-all
python/flux/ag_kernel_crossnode.py:86
Method__init__
(self, name: str, output: torch.Tensor, gemm_time_ms: float)
test/test_gemm_only.py:41
Method__init__
( self, name: str, output: torch.Tensor, total_ms: float, time1: str,
test/test_ag_kernel_pyshmem.py:59
Method__init__
( self, name: str, output: torch.Tensor, gemm_time_ms: float, comm_time_ms: float )
test/test_gemm_rs.py:62
Method__init__
( self, name: str, output: torch.Tensor, total_ms: float, time1: str,
test/test_ag_kernel.py:60
Method__init__
( self, name: str, output: torch.Tensor, total_ms: float, time1: str,
test/test_ag_kernel_crossnode.py:61
Method__init__
(self, node: clang.cindex.Cursor)
pynvshmem/tools/autogen_pybind.py:120
Method__repr__
(self)
test/test_gemm_only.py:46
Method__repr__
(self)
test/test_ag_kernel_pyshmem.py:77
Method__repr__
(self)
test/test_gemm_rs.py:70
Method__repr__
(self)
test/test_ag_kernel.py:82
Method__repr__
(self)
test/test_ag_kernel_crossnode.py:84
Method_ensure_topo_initialized
src/all_gather/ths_op/all_gather_gemm_kernel_crossnode.cc:853
Method_ensure_topo_initialized
src/reduce_scatter/ths_op/gemm_reduce_scatter.cc:751
Function_get_text_from_extent
(extent)
pynvshmem/tools/autogen_pybind.py:254
Method_to_argument
(dtype, name)
pynvshmem/tools/autogen_pybind.py:227
Method_to_param
(dtype, name)
pynvshmem/tools/autogen_pybind.py:242
Methodacquire_accumulators
Acquire accumulators from peers
src/all_gather/sm80_all_gather_gemm.hpp:722
Methodacquire_accumulators
Acquire accumulators from peers
src/reduce_scatter/gemmk_universal_with_visitor_streamk.h:695
Functionadd
src/reduce_scatter/reduce_scatter_kernel.hpp:178
Methodadd
include/flux/op_registry.h:75
Methodadd
include/flux/op_registry.h:95
Functionadd_continous
src/reduce_scatter/reduce_scatter_kernel.hpp:270
Functionadd_tensor
src/reduce_scatter/epilogue_nvshmem_reduce_scatter.hpp:71
Methodall_gemm_hparams
include/flux/gemm_operator_base.h:57
Methodall_gemm_metas
include/flux/gemm_operator_base.h:52
Methodarrive_inc
include/flux/cuda/system_barrier.hpp:105
Methodarrive_inc_get
include/flux/cuda/system_barrier.hpp:116
Methodarrive_inc_get
include/flux/cuda/system_barrier.hpp:154
Methodas_tuple
include/flux/flux.h:971
Methodasync_load
include/flux/cuda/memory_utils.hpp:141
Methodasync_load
include/flux/cuda/memory_utils.hpp:155
Methodasync_load
include/flux/cuda/memory_utils.hpp:169
Functionatomic_add_dev
include/flux/cuda/cuda_common_device.hpp:149
Functionatomic_add_sys
include/flux/cuda/cuda_common_device.hpp:137
Functionatomic_load_acquire_dev
include/flux/cuda/cuda_common_device.hpp:132
Functionatomic_load_dev
include/flux/cuda/cuda_common_device.hpp:127
Functionatomic_store_release_dev
include/flux/cuda/cuda_common_device.hpp:121
Functionatomic_store_release_sys
include/flux/cuda/cuda_common_device.hpp:99
← previousnext →301–400 of 784, ranked by callers