Qinghua Zhou eb1e0f33ee Fix alltoallv build and Python import errors ...

- Add missing accumDtype and symmetricMemory params to NativeAlgorithm
  lambda signatures in alltoallv_fullmesh.cu to match KernelFunc and
  ContextKeyGenFunc typedefs
- Use Cpp-prefixed binding names in alltoallv_single.py imports
  (CppCommunicator, CppDataType, etc.)
- Add missing symmetric_memory=False arg to algo.execute() call
- Fix test imports to use mscclpp public API instead of raw _mscclpp

2026-04-13 22:43:42 +00:00

..

csrc

Merge main branch

2026-04-13 22:06:38 +00:00

mscclpp

Fix alltoallv build and Python import errors

2026-04-13 22:43:42 +00:00

mscclpp_benchmark

Torch integration (#692)

2026-01-21 20:32:24 -08:00

test

Fix alltoallv build and Python import errors

2026-04-13 22:43:42 +00:00

CMakeLists.txt

Disabling Nanobind Memory Leak Warnings in Release Builds (#745)

2026-02-23 11:55:17 -08:00

requirements_cuda11.txt

Torch integration (#692)

2026-01-21 20:32:24 -08:00

requirements_cuda12.txt

Torch integration (#692)

2026-01-21 20:32:24 -08:00

requirements_cuda13.txt

create CI pipeline for rocm (#718)

2026-02-09 16:55:16 -08:00

requirements_rocm6.txt

Add ROCm FP8 E4M3B15 support (#774)

2026-04-08 09:53:45 -07:00