4277b5aa882a26784adf9a762f465debc89a4bfa
Using a single slice rather than the typical two provides about 5% speedup (sometimes more or less) on some GFX9 systems for single node.
[ROCm/rccl commit: 2f6b20c00a]
Описание
No description provided
Languages
C++
67.5%
C
20.6%
Python
6.6%
CMake
3.4%
Shell
0.6%
Разное
1.1%