Gregory Shtrasberg 61a3431613 [Bugfix][ROCm] Fixing trying to import non-existent symbols from libnccl.so (#25605) ...

Signed-off-by: Gregory Shtrasberg <Gregory.Shtrasberg@amd.com>

2025-09-29 17:01:50 -04:00

..

__init__.py

[Core][Refactor] move parallel_utils into vllm/distributed (#3950)

2024-04-10 15:33:30 -07:00

all2all.py

[Bugfix][WideEP] Apply TP Attn + EP MoE fix to other models (#24982)

2025-09-27 14:22:28 +00:00

all_reduce_utils.py

[core] add nccl symmetric memory for all reduce (#24532)

2025-09-23 14:33:06 -04:00

base_device_communicator.py

[Bugfix][WideEP] Apply TP Attn + EP MoE fix to other models (#24982)

2025-09-27 14:22:28 +00:00

cpu_communicator.py

[CPU] Enable shared-memory based pipeline parallel for CPU backend (#21289)

2025-07-21 09:07:08 -07:00

cuda_communicator.py

[Bugfix][WideEP] Apply TP Attn + EP MoE fix to other models (#24982)

2025-09-27 14:22:28 +00:00

cuda_wrapper.py

[Misc] Add SPDX-FileCopyrightText (#19100)

2025-06-03 11:20:17 -07:00

custom_all_reduce.py

[Kernels] Enable Torch Symmetric Memory All-Reduce By Default (#24111)

2025-09-11 09:45:31 -07:00

mnnvl_compat.py

Support mnnvl all2allv from Flashinfer (#21003)

2025-09-24 14:38:16 -04:00

pynccl_allocator.py

[core] add nccl symmetric memory for all reduce (#24532)

2025-09-23 14:33:06 -04:00

pynccl_wrapper.py

[Bugfix][ROCm] Fixing trying to import non-existent symbols from libnccl.so (#25605)

2025-09-29 17:01:50 -04:00

pynccl.py

Remove redundant mutates_args and dispatch_key for direct_register_custom_op (#25512)

2025-09-23 22:48:40 +00:00

quick_all_reduce.py

[Doc]: fix typos in Python comments (#24042)

2025-09-01 19:07:45 -07:00

ray_communicator.py

[mypy] Fix wrong type annotations related to tuple (#25660)

2025-09-25 13:00:45 +00:00

shm_broadcast.py

[Misc] Remove unnecessary memoryviews in shm_broadcast.py (#25721)

2025-09-26 03:11:44 +00:00

shm_object_storage.py

[Docs] Fix warnings in mkdocs build (continued) (#25163)

2025-09-18 08:23:26 -07:00

symm_mem.py

[Kernels] Enable Torch Symmetric Memory All-Reduce By Default (#24111)

2025-09-11 09:45:31 -07:00

tpu_communicator.py

[TPU] make ptxla not imported when using tpu_commons (#23081)

2025-08-19 11:46:42 +08:00

xpu_communicator.py

[Bugfix][WideEP] Apply TP Attn + EP MoE fix to other models (#24982)

2025-09-27 14:22:28 +00:00