Qinghua Zhou 43980da455 Use maximum threads (1024) for best bandwidth utilization

2026-02-18 03:00:29 +00:00

..

allgather_test.cu

connect() APIs changed to return an instance instead of a shared_ptr (#680)

2025-11-15 11:40:40 -08:00

allreduce_test.cu

Revised MemoryChannel interfaces (#508)

2025-04-25 00:02:56 +00:00

alltoall_test.cu

Fixed cpp linter (#619)

2025-08-25 12:15:45 -07:00

alltoallv_test.cu

Use maximum threads (1024) for best bandwidth utilization

2026-02-18 03:00:29 +00:00

check_perf_result.py

Add __launch_bounds__ for mscclpp-test (#273)

2024-03-25 15:55:37 -07:00

CMakeLists.txt

Move the alltoallv kernel to the src directory; Utilize the kernel in mscclpp-test

2026-02-06 02:57:34 +00:00

common.cc

connect() APIs changed to return an instance instead of a shared_ptr (#680)

2025-11-15 11:40:40 -08:00

common.hpp

connect() APIs changed to return an instance instead of a shared_ptr (#680)

2025-11-15 11:40:40 -08:00

sendrecv_test.cu

Use smart pointer for IB structure (#585)

2025-08-06 10:01:58 -07:00