[Misc][Main2Main] Upgrade vLLM to 0429(DSV4/v0.20.0) by gcanlin · Pull Request #8856 · vllm-project/vllm-ascend

gcanlin · 2026-05-01T04:15:03Z

What this PR does / why we need it?

Cherry pick #8841 and then continue to upgrading vLLM.

#	Error	Category	Upstream Commit	Affected vllm-ascend Path	Fix
1	`encoder_compilation_time` AttributeError	Code Bug	`c08f3b2a6` (#39240)	`worker/worker.py:567`	`getattr` fallback
2	`AscendRMSNormGated activation` TypeError	Code Bug	`893611813` (#40245)	`ops/layernorm.py:160`, `_310p/ops/layernorm.py:43`	Accept `activation` kwarg
3	`EagleCudaGraphManager` ImportError	Code Bug	`4c7c69b4e` (#40410)	`worker/v2/spec_decode/eagle/{speculator,aclgraph}.py`	Version-guard import + patch new class names
4	`AscendFusedMoEMethod.apply topk_weights` TypeError	Code Bug	many (e.g., `5e584ce9e` (#35782), `809d83c2d` (#40560), `4d51588e2` (#40860))	`ops/fused_moe/fused_moe.py:107`	Major refactor — follow-up PR
5	`_all_lora_classes` is tuple	Code Bug	`a250f1bd5` (#35077)	`lora/utils.py:188`	Rebuild tuple instead of `.add()`
6	`ProfilingChunkScheduler hash_block_size` TypeError	Code Bug	`7b1bc0a3e` (#40946)	`core/scheduler_profiling_chunk.py:57`	Forward new kwarg
7	`_moe_C.topk_softmax` AttributeError	Code Bug	MoE router refactor	router dispatch override needed	Provide `torch_npu` topk-softmax (with Issue 4)
8	global experts shape mismatch	Code Bug	follow-on of Issue 4	`quantization/methods/w8a8_dynamic.py:198`	Resolve once Issue 4 is fixed

Does this PR introduce any user-facing change?

How was this patch tested?

vLLM version: v0.19.1
vLLM main: vllm-project/vllm@d886c26

Signed-off-by: wxsIcey <1790571317@qq.com>

Signed-off-by: Shanshan Shen <87969357+shen-shanshan@users.noreply.github.com>

Signed-off-by: gcanlin <canlinguosdu@gmail.com>

github-actions · 2026-05-01T04:15:26Z

👋 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.

Signed-off-by: gcanlin <canlinguosdu@gmail.com>

gemini-code-assist · 2026-05-01T04:18:29Z

Summary of Changes

Hello, I'm Gemini Code Assist¹! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed!

This pull request updates the vLLM integration to a newer version, necessitating significant refactoring of the Mixture-of-Experts (MoE) implementation. The changes streamline the MoE layer architecture by consolidating shared expert logic and updating internal APIs to match the upstream vLLM structure. Additionally, minor enhancements were made to the rejection sampler and worker compilation reporting to support new functionality and interface requirements.

Highlights

vLLM Upgrade: Updated the main vLLM commit hash in the documentation configuration to align with the latest version.
MoE Refactoring: Refactored AscendFusedMoE and removed AscendSharedFusedMoE, integrating shared expert logic directly into the main MoE class to improve maintainability and align with upstream vLLM changes.
Worker API Update: Updated the compile_or_warm_up_model method in the worker to return CompilationTimes including encoder compilation time, ensuring compatibility with updated vLLM interfaces.
Sampler Enhancement: Added support for synthetic mode and conditional rates to the rejection sampler.

🧠 New Feature in Public Preview: You can now enable Memory to help Gemini Code Assist learn from your team's feedback. This makes future code reviews more consistent and personalized to your project's style. Click here to enable Memory in your admin console.

Ignored Files

Ignored by pattern: .github/workflows/** (3)
- .github/workflows/_e2e_test.yaml
- .github/workflows/pr_test_full.yaml
- .github/workflows/pr_test_light.yaml

Using Gemini Code Assist

The full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips.

Invoking Gemini

You can request assistance from Gemini at any point by creating a comment using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands on the current page.

Feature	Command	Description
Code Review	`/gemini review`	Performs a code review for the current pull request in its current state.
Pull Request Summary	`/gemini summary`	Provides a summary of the current pull request in its current state.
Comment	@gemini-code-assist	Responds in comments when explicitly tagged, both in pull request comments and review comments.
Help	`/gemini help`	Displays a list of available commands.

Customization

To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a .gemini/ folder in the base of the repository. Detailed instructions can be found here.

Limitations & Feedback

Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for GitHub and other Google products, sign up here.

Review the Privacy Notices, Generative AI Prohibited Use Policy, Terms of Service, and learn how to configure Gemini Code Assist in GitHub here. Gemini can make mistakes, so double check it and use code with caution. ↩

gemini-code-assist

Code Review

Suggested PR Title:\n\n\n[Ops][Feature] Refactor Ascend Fused MoE and update worker compilation times\n\n\nSuggested PR Summary:\n\nmarkdown\n### What this PR does / why we need it?\nThis PR refactors the Ascend Fused MoE implementation by merging AscendSharedFusedMoE into AscendFusedMoE and updating the runner to inherit from MoERunner. It introduces shared expert consistency validation and multi-stream overlap configurations. Additionally, it updates the rejection sampler with synthetic mode parameters and modifies the worker to return structured CompilationTimes. Feedback includes fixing a function call bug where enable_sp was used as a boolean, removing placeholder comments, and cleaning up unreachable code in the worker.\n\n### Does this PR introduce _any_ user-facing change?\nNo significant user-facing API changes, though internal MoE logic and worker return types for compilation are updated.\n\n### How was this patch tested?\nThe PR includes a consistency check for shared expert computation.\n\nFixes #\n

Signed-off-by: gcanlin <canlinguosdu@gmail.com>

…n Ascend (vLLM PR #40860) vLLM PR #40860 ([Feat] DeepSeek V4 Rebased) introduced resolve_kv_cache_block_sizes() into engine/core.py and added a restriction that hybrid KV cache groups with multiple block sizes do not support context parallelism (dcp_world_size/pcp_world_size > 1), raising: ValueError: Hybrid KV cache groups with multiple block sizes do not support context parallelism (dcp_world_size/pcp_world_size > 1). This restriction is correct for CUDA (the CUDA MLA implementation cannot combine hybrid KV with CP), but Ascend has dedicated CP backends for MLA (mla_cp.py) and SFA (sfa_cp.py) that handle this combination. Fix by patching resolve_kv_cache_block_sizes() to skip the ValueError for multiple-groups + CP on Ascend, and instead compute scheduler_block_size as lcm(group_block_sizes) * dcp * pcp for proper alignment. Signed-off-by: gcanlin <canlinguosdu@gmail.com>

Signed-off-by: gcanlin <canlinguosdu@gmail.com>

gcanlin · 2026-05-02T17:11:51Z

+) -> tuple[int, int]:
+    """Ascend-compatible resolve_kv_cache_block_sizes.
+
+    vLLM PR #40860 added a restriction that hybrid KV cache groups with


cc @MengqingCao

Signed-off-by: gcanlin <canlinguosdu@gmail.com>

github-actions · 2026-05-06T01:30:33Z

This pull request has conflicts, please resolve those before we can evaluate the pull request.

github-actions · 2026-05-06T01:30:33Z

This pull request has conflicts, please resolve those before we can evaluate the pull request.

@wxsIcey

@wxsIcey

@wxsIcey

### What this PR does / why we need it? Based on vllm-project#8856. Sync to vLLM `4d51588e2381018348f1022dfa3a7698899805b7`. Fix: --- - MoE refactor @wxsIcey, introduced by vllm-project/vllm#35782, vllm-project/vllm#35949, vllm-project/vllm#40560, vllm-project/vllm#40671. - `TypeError: rejection_sample() got an unexpected keyword argument 'synthetic_mode'` -> Add `synthetic_mode` and `synthetic_conditional_rates` param to ascend `rejection_sample()`. --- | # | Error | Category | Upstream Commit | Affected vllm-ascend Path | Fix | | :- | :------------------------------------------------- | :-------- | :----------------------------------------------- | :--------------------------------------------------------- | :----------------------------------------------- | | 1 | `encoder_compilation_time` AttributeError | Code Bug | `c08f3b2a6` ([#39240](vllm-project/vllm#39240)) | `worker/worker.py:567` | `getattr` fallback | | 2 | `AscendRMSNormGated activation` TypeError | Code Bug | `893611813` ([#40245](vllm-project/vllm#40245)) | `ops/layernorm.py:160`, `_310p/ops/layernorm.py:43` | Accept `activation` kwarg | | 3 | `AscendFusedMoEMethod.apply topk_weights` TypeError| Code Bug | many (e.g., `5e584ce9e` ([#35782](vllm-project/vllm#35782)), `809d83c2d` ([#40560](vllm-project/vllm#40560)), `4d51588e2` ([#40860](vllm-project/vllm#40860))) | `ops/fused_moe/fused_moe.py:107` | Major refactor — follow-up PR | | 4 | `_all_lora_classes` is tuple | Code Bug | `a250f1bd5` ([#35077](vllm-project/vllm#35077)) | `lora/utils.py:188` | Rebuild tuple instead of `.add()` | | 5 | `ProfilingChunkScheduler hash_block_size` TypeError| Code Bug | `7b1bc0a3e` ([#40946](vllm-project/vllm#40946)) | `core/scheduler_profiling_chunk.py:57` | Forward new kwarg | | 6 | `_moe_C.topk_softmax` AttributeError | Code Bug | MoE router refactor | router dispatch override needed | Provide `torch_npu` topk-softmax (with Issue 4) | | 7 | global experts shape mismatch | Code Bug | follow-on of Issue 4 | `quantization/methods/w8a8_dynamic.py:198` | Resolve once Issue 4 is fixed | - vLLM main: vllm-project/vllm@d886c26 --------- Signed-off-by: wxsIcey <1790571317@qq.com> Signed-off-by: Shanshan Shen <87969357+shen-shanshan@users.noreply.github.com> Signed-off-by: gcanlin <canlinguosdu@gmail.com> Signed-off-by: shen-shanshan <467638484@qq.com> Co-authored-by: wxsIcey <1790571317@qq.com> Co-authored-by: gcanlin <canlinguosdu@gmail.com> Signed-off-by: yangzhe-2026 <yangzhe@isrc.iscas.ac.cn>

@wxsIcey

@wxsIcey

### What this PR does / why we need it? Based on vllm-project#8856. Sync to vLLM `4d51588e2381018348f1022dfa3a7698899805b7`. Fix: --- - MoE refactor @wxsIcey, introduced by vllm-project/vllm#35782, vllm-project/vllm#35949, vllm-project/vllm#40560, vllm-project/vllm#40671. - `TypeError: rejection_sample() got an unexpected keyword argument 'synthetic_mode'` -> Add `synthetic_mode` and `synthetic_conditional_rates` param to ascend `rejection_sample()`. --- | # | Error | Category | Upstream Commit | Affected vllm-ascend Path | Fix | | :- | :------------------------------------------------- | :-------- | :----------------------------------------------- | :--------------------------------------------------------- | :----------------------------------------------- | | 1 | `encoder_compilation_time` AttributeError | Code Bug | `c08f3b2a6` ([#39240](vllm-project/vllm#39240)) | `worker/worker.py:567` | `getattr` fallback | | 2 | `AscendRMSNormGated activation` TypeError | Code Bug | `893611813` ([#40245](vllm-project/vllm#40245)) | `ops/layernorm.py:160`, `_310p/ops/layernorm.py:43` | Accept `activation` kwarg | | 3 | `AscendFusedMoEMethod.apply topk_weights` TypeError| Code Bug | many (e.g., `5e584ce9e` ([#35782](vllm-project/vllm#35782)), `809d83c2d` ([#40560](vllm-project/vllm#40560)), `4d51588e2` ([#40860](vllm-project/vllm#40860))) | `ops/fused_moe/fused_moe.py:107` | Major refactor — follow-up PR | | 4 | `_all_lora_classes` is tuple | Code Bug | `a250f1bd5` ([#35077](vllm-project/vllm#35077)) | `lora/utils.py:188` | Rebuild tuple instead of `.add()` | | 5 | `ProfilingChunkScheduler hash_block_size` TypeError| Code Bug | `7b1bc0a3e` ([#40946](vllm-project/vllm#40946)) | `core/scheduler_profiling_chunk.py:57` | Forward new kwarg | | 6 | `_moe_C.topk_softmax` AttributeError | Code Bug | MoE router refactor | router dispatch override needed | Provide `torch_npu` topk-softmax (with Issue 4) | | 7 | global experts shape mismatch | Code Bug | follow-on of Issue 4 | `quantization/methods/w8a8_dynamic.py:198` | Resolve once Issue 4 is fixed | - vLLM main: vllm-project/vllm@d886c26 --------- Signed-off-by: wxsIcey <1790571317@qq.com> Signed-off-by: Shanshan Shen <87969357+shen-shanshan@users.noreply.github.com> Signed-off-by: gcanlin <canlinguosdu@gmail.com> Signed-off-by: shen-shanshan <467638484@qq.com> Co-authored-by: wxsIcey <1790571317@qq.com> Co-authored-by: gcanlin <canlinguosdu@gmail.com> Signed-off-by: zhuqi <z00480217@china.huawei.com>

@wxsIcey

### What this PR does / why we need it? Based on vllm-project#8856. Sync to vLLM `4d51588e2381018348f1022dfa3a7698899805b7`. Fix: --- - MoE refactor @wxsIcey, introduced by vllm-project/vllm#35782, vllm-project/vllm#35949, vllm-project/vllm#40560, vllm-project/vllm#40671. - `TypeError: rejection_sample() got an unexpected keyword argument 'synthetic_mode'` -> Add `synthetic_mode` and `synthetic_conditional_rates` param to ascend `rejection_sample()`. --- | # | Error | Category | Upstream Commit | Affected vllm-ascend Path | Fix | | :- | :------------------------------------------------- | :-------- | :----------------------------------------------- | :--------------------------------------------------------- | :----------------------------------------------- | | 1 | `encoder_compilation_time` AttributeError | Code Bug | `c08f3b2a6` ([#39240](vllm-project/vllm#39240)) | `worker/worker.py:567` | `getattr` fallback | | 2 | `AscendRMSNormGated activation` TypeError | Code Bug | `893611813` ([#40245](vllm-project/vllm#40245)) | `ops/layernorm.py:160`, `_310p/ops/layernorm.py:43` | Accept `activation` kwarg | | 3 | `AscendFusedMoEMethod.apply topk_weights` TypeError| Code Bug | many (e.g., `5e584ce9e` ([#35782](vllm-project/vllm#35782)), `809d83c2d` ([#40560](vllm-project/vllm#40560)), `4d51588e2` ([#40860](vllm-project/vllm#40860))) | `ops/fused_moe/fused_moe.py:107` | Major refactor — follow-up PR | | 4 | `_all_lora_classes` is tuple | Code Bug | `a250f1bd5` ([#35077](vllm-project/vllm#35077)) | `lora/utils.py:188` | Rebuild tuple instead of `.add()` | | 5 | `ProfilingChunkScheduler hash_block_size` TypeError| Code Bug | `7b1bc0a3e` ([#40946](vllm-project/vllm#40946)) | `core/scheduler_profiling_chunk.py:57` | Forward new kwarg | | 6 | `_moe_C.topk_softmax` AttributeError | Code Bug | MoE router refactor | router dispatch override needed | Provide `torch_npu` topk-softmax (with Issue 4) | | 7 | global experts shape mismatch | Code Bug | follow-on of Issue 4 | `quantization/methods/w8a8_dynamic.py:198` | Resolve once Issue 4 is fixed | - vLLM main: vllm-project/vllm@d886c26 --------- Signed-off-by: wxsIcey <1790571317@qq.com> Signed-off-by: Shanshan Shen <87969357+shen-shanshan@users.noreply.github.com> Signed-off-by: gcanlin <canlinguosdu@gmail.com> Signed-off-by: shen-shanshan <467638484@qq.com> Co-authored-by: wxsIcey <1790571317@qq.com> Co-authored-by: gcanlin <canlinguosdu@gmail.com> Signed-off-by: zhuqi <z00480217@china.huawei.com> Signed-off-by: ZhuQi-seu <zhuqi12@huawei.com>

@wxsIcey

### What this PR does / why we need it? Based on vllm-project#8856. Sync to vLLM `4d51588e2381018348f1022dfa3a7698899805b7`. Fix: --- - MoE refactor @wxsIcey, introduced by vllm-project/vllm#35782, vllm-project/vllm#35949, vllm-project/vllm#40560, vllm-project/vllm#40671. - `TypeError: rejection_sample() got an unexpected keyword argument 'synthetic_mode'` -> Add `synthetic_mode` and `synthetic_conditional_rates` param to ascend `rejection_sample()`. --- | # | Error | Category | Upstream Commit | Affected vllm-ascend Path | Fix | | :- | :------------------------------------------------- | :-------- | :----------------------------------------------- | :--------------------------------------------------------- | :----------------------------------------------- | | 1 | `encoder_compilation_time` AttributeError | Code Bug | `c08f3b2a6` ([#39240](vllm-project/vllm#39240)) | `worker/worker.py:567` | `getattr` fallback | | 2 | `AscendRMSNormGated activation` TypeError | Code Bug | `893611813` ([#40245](vllm-project/vllm#40245)) | `ops/layernorm.py:160`, `_310p/ops/layernorm.py:43` | Accept `activation` kwarg | | 3 | `AscendFusedMoEMethod.apply topk_weights` TypeError| Code Bug | many (e.g., `5e584ce9e` ([#35782](vllm-project/vllm#35782)), `809d83c2d` ([#40560](vllm-project/vllm#40560)), `4d51588e2` ([#40860](vllm-project/vllm#40860))) | `ops/fused_moe/fused_moe.py:107` | Major refactor — follow-up PR | | 4 | `_all_lora_classes` is tuple | Code Bug | `a250f1bd5` ([#35077](vllm-project/vllm#35077)) | `lora/utils.py:188` | Rebuild tuple instead of `.add()` | | 5 | `ProfilingChunkScheduler hash_block_size` TypeError| Code Bug | `7b1bc0a3e` ([#40946](vllm-project/vllm#40946)) | `core/scheduler_profiling_chunk.py:57` | Forward new kwarg | | 6 | `_moe_C.topk_softmax` AttributeError | Code Bug | MoE router refactor | router dispatch override needed | Provide `torch_npu` topk-softmax (with Issue 4) | | 7 | global experts shape mismatch | Code Bug | follow-on of Issue 4 | `quantization/methods/w8a8_dynamic.py:198` | Resolve once Issue 4 is fixed | - vLLM main: vllm-project/vllm@d886c26 --------- Signed-off-by: wxsIcey <1790571317@qq.com> Signed-off-by: Shanshan Shen <87969357+shen-shanshan@users.noreply.github.com> Signed-off-by: gcanlin <canlinguosdu@gmail.com> Signed-off-by: shen-shanshan <467638484@qq.com> Co-authored-by: wxsIcey <1790571317@qq.com> Co-authored-by: gcanlin <canlinguosdu@gmail.com> Signed-off-by: ZhuQi-seu <zhuqi12@huawei.com>

@wxsIcey

### What this PR does / why we need it? Based on vllm-project#8856. Sync to vLLM `4d51588e2381018348f1022dfa3a7698899805b7`. Fix: --- - MoE refactor @wxsIcey, introduced by vllm-project/vllm#35782, vllm-project/vllm#35949, vllm-project/vllm#40560, vllm-project/vllm#40671. - `TypeError: rejection_sample() got an unexpected keyword argument 'synthetic_mode'` -> Add `synthetic_mode` and `synthetic_conditional_rates` param to ascend `rejection_sample()`. --- | # | Error | Category | Upstream Commit | Affected vllm-ascend Path | Fix | | :- | :------------------------------------------------- | :-------- | :----------------------------------------------- | :--------------------------------------------------------- | :----------------------------------------------- | | 1 | `encoder_compilation_time` AttributeError | Code Bug | `c08f3b2a6` ([#39240](vllm-project/vllm#39240)) | `worker/worker.py:567` | `getattr` fallback | | 2 | `AscendRMSNormGated activation` TypeError | Code Bug | `893611813` ([#40245](vllm-project/vllm#40245)) | `ops/layernorm.py:160`, `_310p/ops/layernorm.py:43` | Accept `activation` kwarg | | 3 | `AscendFusedMoEMethod.apply topk_weights` TypeError| Code Bug | many (e.g., `5e584ce9e` ([#35782](vllm-project/vllm#35782)), `809d83c2d` ([#40560](vllm-project/vllm#40560)), `4d51588e2` ([#40860](vllm-project/vllm#40860))) | `ops/fused_moe/fused_moe.py:107` | Major refactor — follow-up PR | | 4 | `_all_lora_classes` is tuple | Code Bug | `a250f1bd5` ([#35077](vllm-project/vllm#35077)) | `lora/utils.py:188` | Rebuild tuple instead of `.add()` | | 5 | `ProfilingChunkScheduler hash_block_size` TypeError| Code Bug | `7b1bc0a3e` ([#40946](vllm-project/vllm#40946)) | `core/scheduler_profiling_chunk.py:57` | Forward new kwarg | | 6 | `_moe_C.topk_softmax` AttributeError | Code Bug | MoE router refactor | router dispatch override needed | Provide `torch_npu` topk-softmax (with Issue 4) | | 7 | global experts shape mismatch | Code Bug | follow-on of Issue 4 | `quantization/methods/w8a8_dynamic.py:198` | Resolve once Issue 4 is fixed | - vLLM main: vllm-project/vllm@d886c26 --------- Signed-off-by: wxsIcey <1790571317@qq.com> Signed-off-by: Shanshan Shen <87969357+shen-shanshan@users.noreply.github.com> Signed-off-by: gcanlin <canlinguosdu@gmail.com> Signed-off-by: shen-shanshan <467638484@qq.com> Co-authored-by: wxsIcey <1790571317@qq.com> Co-authored-by: gcanlin <canlinguosdu@gmail.com> Signed-off-by: zhuqi <z00480217@china.huawei.com>

@wxsIcey

### What this PR does / why we need it? Based on vllm-project#8856. Sync to vLLM `4d51588e2381018348f1022dfa3a7698899805b7`. Fix: --- - MoE refactor @wxsIcey, introduced by vllm-project/vllm#35782, vllm-project/vllm#35949, vllm-project/vllm#40560, vllm-project/vllm#40671. - `TypeError: rejection_sample() got an unexpected keyword argument 'synthetic_mode'` -> Add `synthetic_mode` and `synthetic_conditional_rates` param to ascend `rejection_sample()`. --- | # | Error | Category | Upstream Commit | Affected vllm-ascend Path | Fix | | :- | :------------------------------------------------- | :-------- | :----------------------------------------------- | :--------------------------------------------------------- | :----------------------------------------------- | | 1 | `encoder_compilation_time` AttributeError | Code Bug | `c08f3b2a6` ([#39240](vllm-project/vllm#39240)) | `worker/worker.py:567` | `getattr` fallback | | 2 | `AscendRMSNormGated activation` TypeError | Code Bug | `893611813` ([#40245](vllm-project/vllm#40245)) | `ops/layernorm.py:160`, `_310p/ops/layernorm.py:43` | Accept `activation` kwarg | | 3 | `AscendFusedMoEMethod.apply topk_weights` TypeError| Code Bug | many (e.g., `5e584ce9e` ([#35782](vllm-project/vllm#35782)), `809d83c2d` ([#40560](vllm-project/vllm#40560)), `4d51588e2` ([#40860](vllm-project/vllm#40860))) | `ops/fused_moe/fused_moe.py:107` | Major refactor — follow-up PR | | 4 | `_all_lora_classes` is tuple | Code Bug | `a250f1bd5` ([#35077](vllm-project/vllm#35077)) | `lora/utils.py:188` | Rebuild tuple instead of `.add()` | | 5 | `ProfilingChunkScheduler hash_block_size` TypeError| Code Bug | `7b1bc0a3e` ([#40946](vllm-project/vllm#40946)) | `core/scheduler_profiling_chunk.py:57` | Forward new kwarg | | 6 | `_moe_C.topk_softmax` AttributeError | Code Bug | MoE router refactor | router dispatch override needed | Provide `torch_npu` topk-softmax (with Issue 4) | | 7 | global experts shape mismatch | Code Bug | follow-on of Issue 4 | `quantization/methods/w8a8_dynamic.py:198` | Resolve once Issue 4 is fixed | - vLLM main: vllm-project/vllm@d886c26 --------- Signed-off-by: wxsIcey <1790571317@qq.com> Signed-off-by: Shanshan Shen <87969357+shen-shanshan@users.noreply.github.com> Signed-off-by: gcanlin <canlinguosdu@gmail.com> Signed-off-by: shen-shanshan <467638484@qq.com> Co-authored-by: wxsIcey <1790571317@qq.com> Co-authored-by: gcanlin <canlinguosdu@gmail.com> Signed-off-by: zhuqi <z00480217@china.huawei.com>

@wxsIcey

### What this PR does / why we need it? Based on vllm-project#8856. Sync to vLLM `4d51588e2381018348f1022dfa3a7698899805b7`. Fix: --- - MoE refactor @wxsIcey, introduced by vllm-project/vllm#35782, vllm-project/vllm#35949, vllm-project/vllm#40560, vllm-project/vllm#40671. - `TypeError: rejection_sample() got an unexpected keyword argument 'synthetic_mode'` -> Add `synthetic_mode` and `synthetic_conditional_rates` param to ascend `rejection_sample()`. --- | # | Error | Category | Upstream Commit | Affected vllm-ascend Path | Fix | | :- | :------------------------------------------------- | :-------- | :----------------------------------------------- | :--------------------------------------------------------- | :----------------------------------------------- | | 1 | `encoder_compilation_time` AttributeError | Code Bug | `c08f3b2a6` ([#39240](vllm-project/vllm#39240)) | `worker/worker.py:567` | `getattr` fallback | | 2 | `AscendRMSNormGated activation` TypeError | Code Bug | `893611813` ([#40245](vllm-project/vllm#40245)) | `ops/layernorm.py:160`, `_310p/ops/layernorm.py:43` | Accept `activation` kwarg | | 3 | `AscendFusedMoEMethod.apply topk_weights` TypeError| Code Bug | many (e.g., `5e584ce9e` ([#35782](vllm-project/vllm#35782)), `809d83c2d` ([#40560](vllm-project/vllm#40560)), `4d51588e2` ([#40860](vllm-project/vllm#40860))) | `ops/fused_moe/fused_moe.py:107` | Major refactor — follow-up PR | | 4 | `_all_lora_classes` is tuple | Code Bug | `a250f1bd5` ([#35077](vllm-project/vllm#35077)) | `lora/utils.py:188` | Rebuild tuple instead of `.add()` | | 5 | `ProfilingChunkScheduler hash_block_size` TypeError| Code Bug | `7b1bc0a3e` ([#40946](vllm-project/vllm#40946)) | `core/scheduler_profiling_chunk.py:57` | Forward new kwarg | | 6 | `_moe_C.topk_softmax` AttributeError | Code Bug | MoE router refactor | router dispatch override needed | Provide `torch_npu` topk-softmax (with Issue 4) | | 7 | global experts shape mismatch | Code Bug | follow-on of Issue 4 | `quantization/methods/w8a8_dynamic.py:198` | Resolve once Issue 4 is fixed | - vLLM main: vllm-project/vllm@d886c26 --------- Signed-off-by: wxsIcey <1790571317@qq.com> Signed-off-by: Shanshan Shen <87969357+shen-shanshan@users.noreply.github.com> Signed-off-by: gcanlin <canlinguosdu@gmail.com> Signed-off-by: shen-shanshan <467638484@qq.com> Co-authored-by: wxsIcey <1790571317@qq.com> Co-authored-by: gcanlin <canlinguosdu@gmail.com> Signed-off-by: zhuqi <z00480217@china.huawei.com> Signed-off-by: ZhuQi-seu <zhuqi12@huawei.com>

wxsIcey and others added 4 commits April 30, 2026 14:49

remove shared_fused_moe

edeeb6f

Signed-off-by: wxsIcey <1790571317@qq.com>

remove shared FusedMoE

67a1011

Signed-off-by: wxsIcey <1790571317@qq.com>

main2main

0852239

Signed-off-by: Shanshan Shen <87969357+shen-shanshan@users.noreply.github.com>

fix lint

109bb18

Signed-off-by: gcanlin <canlinguosdu@gmail.com>

gcanlin requested review from LCAIZJ, MengqingCao, Yikun, realliujiaxu, wangxiyuan, whx-sjtu and zzzzwwjj as code owners May 1, 2026 04:15

github-actions Bot added documentation Improvements or additions to documentation ci/build module:ops module:core labels May 1, 2026

gcanlin added ready read for review ready-for-test start test by label for PR and removed documentation Improvements or additions to documentation ci/build module:ops module:core labels May 1, 2026

gcanlin changed the title ~~[Main2Main] Upgrade vLLM to 0429~~ [Misc][Main2Main] Upgrade vLLM to 0429 May 1, 2026

gcanlin added 2 commits May 1, 2026 04:17

test

c0102e5

Signed-off-by: gcanlin <canlinguosdu@gmail.com>

test

a3dda42

Signed-off-by: gcanlin <canlinguosdu@gmail.com>

gemini-code-assist Bot reviewed May 1, 2026

View reviewed changes

Comment thread vllm_ascend/ops/fused_moe/fused_moe.py Outdated

Comment thread vllm_ascend/ops/fused_moe/fused_moe.py Outdated

Comment thread vllm_ascend/ops/fused_moe/fused_moe.py Outdated

Comment thread vllm_ascend/worker/worker.py

fix

9a42bbb

Signed-off-by: gcanlin <canlinguosdu@gmail.com>

gcanlin added the e2e-310p-test label May 1, 2026

fix

c8c5013

Signed-off-by: gcanlin <canlinguosdu@gmail.com>

gcanlin added 3 commits May 2, 2026 14:50

revert to 4d51588e2

6391bbb

Signed-off-by: gcanlin <canlinguosdu@gmail.com>

revert 40410

6e8e7f9

Signed-off-by: gcanlin <canlinguosdu@gmail.com>

gcanlin commented May 2, 2026

View reviewed changes

gcanlin changed the title ~~[Misc][Main2Main] Upgrade vLLM to 0429~~ [Misc][Main2Main] Upgrade vLLM to 0429(DSV4) May 2, 2026

gcanlin added 3 commits May 2, 2026 17:23

remove 0.19.1 in CI tests

325e72e

Signed-off-by: gcanlin <canlinguosdu@gmail.com>

remove chinese comments

2d208fa

Signed-off-by: gcanlin <canlinguosdu@gmail.com>

fix conf.py

cb70a85

Signed-off-by: gcanlin <canlinguosdu@gmail.com>

gcanlin changed the title ~~[Misc][Main2Main] Upgrade vLLM to 0429(DSV4)~~ [Misc][Main2Main] Upgrade vLLM to 0429(DSV4/v0.20.0) May 2, 2026

gcanlin added 3 commits May 2, 2026 19:23

fix 310p shared fused moe

60424fc

Signed-off-by: gcanlin <canlinguosdu@gmail.com>

fix base_quant_method bug

f5523b9

Signed-off-by: gcanlin <canlinguosdu@gmail.com>

patch kv_utils

a0ae6cd

Signed-off-by: gcanlin <canlinguosdu@gmail.com>

gcanlin mentioned this pull request May 3, 2026

[RFC]: NPU release v0.20.0 checklist vllm-project/vllm-omni#3324

Closed

8 tasks

github-actions Bot added the merge-conflicts label May 6, 2026

shen-shanshan mentioned this pull request May 6, 2026

[Misc][Main2Main] Upgrade vLLM to 0.20.1 #8880

Closed

wxsIcey mentioned this pull request May 6, 2026

[Misc][Main2Main] Upgrade vLLM to v0.20.1 and 0506 #8888

Closed

shen-shanshan mentioned this pull request May 6, 2026

[Misc][Main2Main] Upgrade vLLM to 0427 #8899

Merged

gcanlin closed this May 11, 2026

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

[Misc][Main2Main] Upgrade vLLM to 0429(DSV4/v0.20.0)#8856

[Misc][Main2Main] Upgrade vLLM to 0429(DSV4/v0.20.0)#8856
gcanlin wants to merge 30 commits into
vllm-project:mainfrom
gcanlin:pr-8841

gcanlin commented May 1, 2026 •

edited

Loading

Uh oh!

github-actions Bot commented May 1, 2026

Uh oh!

gemini-code-assist Bot commented May 1, 2026

Uh oh!

gemini-code-assist Bot left a comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

gcanlin May 2, 2026

Uh oh!

github-actions Bot commented May 6, 2026

Uh oh!

github-actions Bot commented May 6, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

Conversation

gcanlin commented May 1, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

What this PR does / why we need it?

Does this PR introduce any user-facing change?

How was this patch tested?

Uh oh!

github-actions Bot commented May 1, 2026

Uh oh!

gemini-code-assist Bot commented May 1, 2026

Summary of Changes

Highlights

Footnotes

Uh oh!

gemini-code-assist Bot left a comment

Choose a reason for hiding this comment

Code Review

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

gcanlin May 2, 2026

Choose a reason for hiding this comment

Uh oh!

github-actions Bot commented May 6, 2026

Uh oh!

github-actions Bot commented May 6, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

gcanlin commented May 1, 2026 •

edited

Loading