Changho Hwang 253a1ba1a9 Use a stream pool for gpuCalloc*() (#509) ...

Previous `gpuCalloc*()` creates a new stream for each allocation, which
messes the timeline up in profiler traces. Now `GpuStreamPool` allows
reusing the temporal streams.

2025-06-04 10:07:20 -07:00

..

mscclpp

Use a stream pool for gpuCalloc*() (#509)

2025-06-04 10:07:20 -07:00

CMakeLists.txt

Add Python get_include() (#141)

2023-07-25 10:23:16 -07:00