Skip to content

Revert "[feat] enable hierarchical mc2 ops on A2 by default (#5300)"#5434

Merged
realliujiaxu merged 2 commits intovllm-project:mainfrom
realliujiaxu:revert
Dec 27, 2025
Merged

Revert "[feat] enable hierarchical mc2 ops on A2 by default (#5300)"#5434
realliujiaxu merged 2 commits intovllm-project:mainfrom
realliujiaxu:revert

Conversation

@realliujiaxu
Copy link
Copy Markdown
Collaborator

@realliujiaxu realliujiaxu commented Dec 27, 2025

We'll release 0.13.0 soon. The main branch is freeze. Let's revert the newest change and redo it once 0.13.0 is released.

…ject#5300)"

This reverts commit 12da9f9.

Signed-off-by: realliujiaxu <realliujiaxu@163.com>
@github-actions
Copy link
Copy Markdown
Contributor

👋 Hi! Thank you for contributing to the vLLM Ascend project. The following points will speed up your PR merge:‌‌

  • A PR should do only one thing, smaller PRs enable faster reviews.
  • Every PR should include unit tests and end-to-end tests ‌to ensure it works and is not broken by other future PRs.
  • Write the commit message by fulfilling the PR description to help reviewer and future developers understand.

If CI fails, you can run linting and testing checks locally according Contributing and Testing.

Copy link
Copy Markdown
Contributor

@gemini-code-assist gemini-code-assist bot left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Code Review

This pull request appears to be a revert of a previous feature, likely to stabilize the main branch for an upcoming release. The changes correctly undo the addition of hierarchical communication for MoE operations. However, if my understanding is incorrect and this is actually adding the feature, there is a potential issue in vllm_ascend/utils.py where a function's implementation doesn't match its accompanying comment, which I've detailed in a specific comment. Assuming this is a revert as titled, the changes look correct and complete.

Comment thread vllm_ascend/utils.py
Comment on lines +964 to +966
def is_hierarchical_communication_enabled():
return (os.getenv("HCCL_INTRA_ROCE_ENABLE", "") == "0"
and os.getenv("HCCL_INTRA_PCIE_ENABLE", "") == "1")
Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

high

The comment for this function specifies that this hierarchical communication optimization is for A2 devices. However, the function implementation does not include a check for the device type. This could lead to this communication path being enabled on other device types if the environment variables are set, which might cause unexpected behavior or performance issues if not supported. To align the implementation with the documentation and prevent potential issues, it's recommended to add a device type check.

Suggested change
def is_hierarchical_communication_enabled():
return (os.getenv("HCCL_INTRA_ROCE_ENABLE", "") == "0"
and os.getenv("HCCL_INTRA_PCIE_ENABLE", "") == "1")
def is_hierarchical_communication_enabled():
return (get_ascend_device_type() == AscendDeviceType.A2 and
os.getenv("HCCL_INTRA_ROCE_ENABLE", "") == "0" and
os.getenv("HCCL_INTRA_PCIE_ENABLE", "") == "1")

@realliujiaxu realliujiaxu merged commit 09f71c1 into vllm-project:main Dec 27, 2025
7 of 9 checks passed
845473182 pushed a commit to 845473182/vllm-ascend that referenced this pull request Dec 29, 2025
…to eplb_refactor

* 'main' of https://github.com/vllm-project/vllm-ascend: (46 commits)
  [Feature] Support to use fullgraph with eagle (vllm-project#5118)
  [EPLB][refactor] Modification of the initialization logic for expert_map and log2phy(depend on pr5285) (vllm-project#5311)
  [Refactor]6/N Extract common code of class AscendMLAImpl (vllm-project#5314)
  [Refactor] cache cos/sin in mla & remove parameter model in builder. (vllm-project#5277)
  update vllm pin to 12.27 (vllm-project#5412)
  [ReleaseNote] Add release note for v0.13.0rc1 (vllm-project#5334)
  [Bugfix] Correctly handle the output shape in multimodal attention (vllm-project#5443)
  Fix nightly (vllm-project#5413)
  [bugfix] fix typo of _skip_all_reduce_across_dp_group (vllm-project#5435)
  [Doc]modify pcp tutorial doc (vllm-project#5440)
  [Misc] fast fail for exiting if tools/install_flash_infer_attention_score_ops_a2.sh (vllm-project#5422)
  [Doc] Update DeepSeek V3.1/R1 2P1D doc (vllm-project#5387)
  [DOC]Fix model weight download links (vllm-project#5436)
  [Doc] Modify DeepSeek-R1/V3.1 documentation (vllm-project#5426)
  Revert "[feat] enable hierarchical mc2 ops on A2 by default (vllm-project#5300)" (vllm-project#5434)
  [Bugfix] fix greedy temperature detection (vllm-project#5417)
  [doc] Update Qwen3-235B doc for reproducing latest performance (vllm-project#5323)
  [feat] enable hierarchical mc2 ops on A2 by default (vllm-project#5300)
  [Doc] delete environment variable HCCL_OP_EXPANSION_MODE in DeepSeekV3.1/R1 (vllm-project#5419)
  [Doc] add long_sequence feature user guide (vllm-project#5343)
  ...
ZRJ026 pushed a commit to ZRJ026/vllm-ascend that referenced this pull request Feb 28, 2026
…ject#5300)" (vllm-project#5434)

We'll release 0.13.0 soon. The main branch is freeze. Let's revert the
newest change and redo it once 0.13.0 is released.

- vLLM version: release/v0.13.0
- vLLM main:
vllm-project/vllm@81786c8

Signed-off-by: realliujiaxu <realliujiaxu@163.com>
Signed-off-by: zrj026 <zhangrunjiang026@gmail.com>
maoxx241 pushed a commit to maoxx241/vllm-ascend that referenced this pull request Mar 2, 2026
…ject#5300)" (vllm-project#5434)

We'll release 0.13.0 soon. The main branch is freeze. Let's revert the
newest change and redo it once 0.13.0 is released.

- vLLM version: release/v0.13.0
- vLLM main:
vllm-project/vllm@81786c8

Signed-off-by: realliujiaxu <realliujiaxu@163.com>
ZRJ026 pushed a commit to ZRJ026/vllm-ascend that referenced this pull request Mar 4, 2026
…ject#5300)" (vllm-project#5434)

We'll release 0.13.0 soon. The main branch is freeze. Let's revert the
newest change and redo it once 0.13.0 is released.

- vLLM version: release/v0.13.0
- vLLM main:
vllm-project/vllm@81786c8

Signed-off-by: realliujiaxu <realliujiaxu@163.com>
Signed-off-by: zrj026 <zhangrunjiang026@gmail.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants