alex-breslow-amd bd614458c3 [gfx950] Turn On Single Node One Slice Optimization for gfx950 and MI300A (#2017) ...

* Internal benchmarking shows nice single-node performance uplift for MI300A and MI350

[ROCm/rccl commit: 56e0b4e445]

2025-11-06 12:12:45 -08:00

projects/rccl

[gfx950] Turn On Single Node One Slice Optimization for gfx950 and MI300A (#2017)

2025-11-06 12:12:45 -08:00

S

Описание

No description provided

283 MiB

0 Stars 2 Наблюдатели 0 Форки

Languages

C++ 67.5%

C 20.6%

Python 6.6%

CMake 3.4%

Shell 0.6%

Разное 1.1%