qinghuazhou 2e73318c89 Revert "Clean up completed communicator receives (#804)" - thread-safety regression causing cudaErrorIllegalAddress on HT internode GB200

2026-05-27 02:15:39 +00:00

..

bootstrap

Torch integration (#692)

2026-01-21 20:32:24 -08:00

executor

Fix FP8 ROCm build/test issues and dtype naming (#792)

2026-04-28 15:02:22 -07:00

include

Revert "Clean up completed communicator receives (#804)" - thread-safety regression causing cudaErrorIllegalAddress on HT internode GB200

2026-05-27 02:15:39 +00:00

npkit

Updating NpKit (#785)

2026-04-13 13:36:42 -07:00

algorithm.cc

Fix FP8 ROCm build/test issues and dtype naming (#792)

2026-04-28 15:02:22 -07:00

atomicadd_kernel.cu

ext/ep: fix CUDA 11.8 build + apply clang-format/black

2026-05-18 21:44:20 +00:00

c_style_remnants.cc

Torch integration (#692)

2026-01-21 20:32:24 -08:00

CMakeLists.txt

Adding necessary macros for enabling mrc support (#797)

2026-05-05 17:17:41 -04:00

communicator.cc

Revert "Clean up completed communicator receives (#804)" - thread-safety regression causing cudaErrorIllegalAddress on HT internode GB200

2026-05-27 02:15:39 +00:00

connection.cc

Merge remote-tracking branch 'origin/main' into qinghuazhou/expert_parallel_merge_main_test

2026-05-18 22:02:50 +00:00

context.cc

Merge remote-tracking branch 'origin/main' into qinghuazhou/expert_parallel_merge_main_test

2026-05-18 22:02:50 +00:00

core.cc

Torch integration (#692)

2026-01-21 20:32:24 -08:00

debug.cc

Torch integration (#692)

2026-01-21 20:32:24 -08:00

endpoint.cc

Add MSCCLPP_IB_GID_INDEX env (#780)

2026-04-13 09:59:42 -07:00

env.cpp

Add MSCCLPP_IB_GID_INDEX env (#780)

2026-04-13 09:59:42 -07:00

errors.cc

Torch integration (#692)

2026-01-21 20:32:24 -08:00

fifo.cc

Handle PortChannel flush asynchronously from the host proxy (#802)

2026-05-15 11:50:43 -07:00

gdr.cc

GDRCopy status message to string (#793)

2026-04-27 10:32:20 -07:00

gpu_ipc_mem.cc

Enhance cross-node CudaIpc availability check (#803)

2026-05-14 14:06:12 -07:00

gpu_utils.cc

Reset GPU tokens before reuse (#795)

2026-05-04 15:11:47 -07:00

ib.cc

IB host-no-atomic: GDRCopy + mlx5dv Data Direct for memory-consistent low-latency signaling (#753)

2026-04-09 09:24:30 +00:00

ibverbs_wrapper.cc

Adding necessary macros for enabling mrc support (#797)

2026-05-05 17:17:41 -04:00

logger.cc

Torch integration (#692)

2026-01-21 20:32:24 -08:00

memory_channel.cc

Torch integration (#692)

2026-01-21 20:32:24 -08:00

mlx5dv_wrapper.cc

IB host-no-atomic: GDRCopy + mlx5dv Data Direct for memory-consistent low-latency signaling (#753)

2026-04-09 09:24:30 +00:00

numa.cc

Torch integration (#692)

2026-01-21 20:32:24 -08:00

port_channel.cc

Merge remote-tracking branch 'origin/main' into qinghuazhou/expert_parallel_merge_main_test

2026-05-18 22:02:50 +00:00

proxy.cc

Handle PortChannel flush asynchronously from the host proxy (#802)

2026-05-15 11:50:43 -07:00

registered_memory.cc

Fix multi-node H100 CI: CUDA compat, deploy improvements (#781)

2026-04-13 21:51:29 -07:00

semaphore.cc

IB host-no-atomic: GDRCopy + mlx5dv Data Direct for memory-consistent low-latency signaling (#753)

2026-04-09 09:24:30 +00:00

switch_channel.cc

Torch integration (#692)

2026-01-21 20:32:24 -08:00

unix_socket.cc

Torch integration (#692)

2026-01-21 20:32:24 -08:00

utils_internal.cc

Reset GPU tokens before reuse (#795)

2026-05-04 15:11:47 -07:00

utils.cc

Torch integration (#692)

2026-01-21 20:32:24 -08:00