mscclpp

mirror of https://github.com/microsoft/mscclpp.git synced 2026-03-30 12:06:02 +00:00

Files

Changho Hwang b4dde38db8 FIFO improvements (#557 )

* Revert `MSCCLPP_FIFO_USE_TAIL_REPLICA=1` back to the default.
* Optimize `FifoDeviceHandle`.
* Do not use `cudaHostAllocWriteCombined` that increases latency.
* Pin host memory for `Host2DeviceSemaphore::outboundSemaphore_`.
* Fix proxy NUMA binding issues.
* Prevent graph capture inside proxy threads.
* Now `CudaIpcConnection` skips stream sync when unnecessary.
* Now any type of connection needs to hold a shared pointer to the
context for memory safety.
* Now a context should be always managed by a shared pointer for memory
safety.
* Minor docs & interface improvements.
* Minor fix in `mscclpp-test` correctness test.

2025-06-24 09:50:28 -07:00

_cpp

FIFO improvements (#557 )

2025-06-24 09:50:28 -07:00

configs

Add multi-nodes example & update doc (#455 )

2025-01-31 17:52:15 -08:00

__init__.py

Pytest (#162 )

2023-09-01 21:22:11 +08:00

CMakeLists.txt

Fix bootstrapping mechanism (#278 )