Skip to content

bugfix: make DSV4 SWA capacity PD-aware. - #2265

Open
Kang-Meng wants to merge 8 commits into
xLLM-AI:mainfrom
Kang-Meng:bugfix_swa
Open

bugfix: make DSV4 SWA capacity PD-aware.#2265
Kang-Meng wants to merge 8 commits into
xLLM-AI:mainfrom
Kang-Meng:bugfix_swa

Conversation

@Kang-Meng

Copy link
Copy Markdown
Collaborator

Description

Related Issues

Change Type

  • Bug fix
  • New feature
  • Performance improvement
  • Refactor
  • Documentation
  • Test
  • Build or CI

Pull Request Checklist

Thank you for contributing to xLLM. Before requesting review, please make sure the following items are complete.

PR Title and Commit Messages

  • The PR title and each commit message follow the xLLM commit format: <type>: <subject>.

Allowed types: feat, bugfix, docs, test, refactor, chore, style, revert, perf, model, build, release.
The subject should use clear English, start with a verb, include at least 4 words, and end with ..

Pre-commit Checks

  • I have installed pre-commit by running pip install pre-commit or an equivalent command.
  • I have installed the hooks with pre-commit install.
  • I have run pre-commit run --all-files and fixed any reported issues.

If you are unsure how to set up pre-commit, see the pre-commit documentation.

Self Review

  • I have self-reviewed the code according to .agents/skills/code-review/references/custom-code-style.md, especially code written or assisted by AI.
  • I have rebased this PR onto the latest main branch.

Build and Test Coverage

  • Tests have been added or updated as needed.
  • CUDA: python setup.py build test has passed on a CUDA machine.
  • NPU: python setup.py build test has passed on an NPU machine.
  • MLU: python setup.py build test has passed on an MLU machine.

Reviewer Notes

ShuweiShen772 and others added 5 commits August 21, 2026 11:56
* test: derive cache shapes from topology.

* feat: add decode preemption kv restore waiting.
* fix: avoid duplicate NPU fake registrations

* Merge PR 2241 DeepSeek V4 NPU kernel bindings

* test: remove redundant NPU fake contract test

* test: preserve NPU DSV4 fake contracts

* Merge branch 'main' into dsv4_kernels_split

* Merge origin/main and align Python kernel tests

* fix: register internal Python modules lazily

* test: scope DSV4 operator probes to A3

* test: make DSV4 NPU probes SoC-aware

* test: stabilize platform-specific kernel checks

* style: group NPU fake registrations

* fix: address DeepSeek V4 kernel review feedback

* style: format DeepSeek V4 kernel bindings

* feat: add selected-expert grouped MoE NPU path

* feat: add DeepSeek V4 NPU kernel Python bindings
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

4 participants