mirror of
https://github.com/microsoft/mscclpp.git
synced 2026-05-13 09:46:00 +00:00
4f3638b60d
## Summary - Replace the two-step `signal()` implementation (`incOutbound()` + `atomicStore()`) with a single fire-and-forget PTX `red.release.sys.global.add.u64` instruction - This eliminates one local atomic fetch-add and replaces a remote store with a remote atomic add that has no return value — more efficient on both NVIDIA (PTX `red`) and AMD (compiler optimizes `(void)fetch_add` to fire-and-forget `flat_atomic_add_x2`) - Add a C++ perf test (`PERF_TEST`) in `mp_unit` for signal+wait ping-pong latency ### Performance (H100, 2 ranks, signal+wait round-trip) ``` SemaphorePerfTest.SignalPingPong: Store-based (old): 2.595 us/iter Red-based (new): 2.345 us/iter Speedup: 1.11x ``` ## Test plan - [x] Builds successfully (`make mp_unit_tests`) - [x] `mpirun -np 2 ./build/bin/mp_unit_tests --filter "SemaphorePerfTest"` — 1.11x speedup 🤖 Generated with [Claude Code](https://claude.com/claude-code) --------- Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
15 lines
331 B
CMake
15 lines
331 B
CMake
# Copyright (c) Microsoft Corporation.
|
|
# Licensed under the MIT license.
|
|
|
|
target_sources(mp_unit_tests PRIVATE
|
|
mp_unit_tests.cc
|
|
bootstrap_tests.cc
|
|
ib_tests.cu
|
|
communicator_tests.cu
|
|
port_channel_tests.cu
|
|
memory_channel_tests.cu
|
|
semaphore_perf_tests.cu
|
|
switch_channel_tests.cu
|
|
executor_tests.cc
|
|
)
|