[CI] Main2main upgrade to 0324 by leo-pony · Pull Request #7787 · vllm-project/vllm-ascend

leo-pony · 2026-03-28T08:33:02Z

What this PR does / why we need it?

main2main upgrade to vllm 0324.
fix breaks:

PR #37487 [V0 Deprecation] Refactor kv cache from list to element (c59a132) — self.kv_cache from list[tensor]（per virtual engine）changed to tensor
PR #37874 [KV Offload] Refactor CPU offloading: pluggable CachePolicy, remove Backend abstraction, restructure into cpu/ package (e3c6c10) — LRUOffloadingManager + CPUBackend been refactor to CPUOffloadingManager
PR #32951 [Async][Spec Decoding] Zero-bubble async scheduling + spec decoding (fafe76b) — a) changes self.positions and self.seq_lens from CpuGpuBuffer to plain GPU tensor; b) change _get_cumsum_and_arange output paramter. Another _prepare_input_ids add num_reqs.
PR #35007[Bugfix] Register VLLM_BATCH_INVARIANT in envs.py to fix spurious unknown env var warning (dc6908a) — delete vllm_is_batch_invariant() and const variable VLLM_BATCH_INVARIANT，replace with vllm.envs

Know issues:
1.310p Qwen3.5 test failed for qwen3.5 patch failure, see issue: #7976 @YangShuai52 is fixing.

Does this PR introduce any user-facing change?

As Zero Async Scheduler + spec decode needs _compute_slot_mapping_kernel of NPU and corresponding accepted draft token validation delaye suppots see PR [Performance]zero bubble async scheduling and spec decoding #7640 , this PR make this change: when in spec decode case close the async scheduler. In this way, the Main2Main can be developed in parallel with Spec Decode + Async scheduler, util next release version.

How was this patch tested?

CI

Co-Authored-By: zhaomingyu zhaomingyu13@h-partners.com
wangbj127 wangbj1207@126.com
SidaoY 1024863041@qq.com
22dimensions waitingwind@foxmail.com

vLLM main: vllm-project/vllm@35141a7

github-actions · 2026-03-28T08:33:15Z

👋 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.

gemini-code-assist · 2026-03-28T08:38:33Z

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 focuses on improving compatibility between vLLM Ascend and upstream vLLM changes. It introduces several abstraction layers to handle differences in buffer management and API usage, ensuring that the Ascend backend remains functional across different versions of the core vLLM library. The changes primarily affect model runners, attention mechanisms, and configuration handling.

Highlights

Compatibility Improvements: Introduced compatibility wrappers and checks to support both older vLLM versions (using CpuGpuBuffer) and newer versions (using direct Tensors) across multiple components.
Attention Mechanism Updates: Enhanced _get_fia_params and forward_fused_infer_attention to handle KV cache initialization and parameter passing more robustly.
Refactoring: Moved vllm_is_batch_invariant to a local utility and updated various model runners and patches to ensure compatibility with upstream vLLM refactoring.

🧠 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/** (6)
- .github/workflows/_e2e_test.yaml
- .github/workflows/bot_pr_create.yaml
- .github/workflows/dockerfiles/Dockerfile.lint
- .github/workflows/pr_test_full.yaml
- .github/workflows/pr_test_light.yaml
- .github/workflows/schedule_codecov_refresh.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:

[Attention][Ops][Worker][Misc] Enhance compatibility with upstream vLLM v0.18.0 and refactor buffer management

Suggested PR Summary:

### What this PR does / why we need it?
This PR updates the codebase to maintain compatibility with recent upstream vLLM refactoring, particularly version 0.18.0. It introduces compatibility wrappers in `NPUModelRunner` to handle the transition from `CpuGpuBuffer` to plain `Tensor` objects and provides a local implementation of `vllm_is_batch_invariant`. Additionally, it updates `kv_cache` indexing for multi-engine support and fixes potential aliasing bugs in position buffer management.

### Does this PR introduce _any_ user-facing change?
No.

### How was this patch tested?
CI passed with existing tests.

I have no feedback to provide.

leo-pony · 2026-03-28T09:59:19Z

ClaudeCode fix summary:
Fixes Summary
7 files modified across 6 distinct issues:
1 patch_qwen3_next_mtp.py — Removed list wrapping of kv_cache ([kv_cache] → kv_cache). This was the ROOT CAUSE of garbage output/0 accuracy.
2 model_runner_v1.py — Three fixes:
◦ Added self._compute_prev_positions(num_reqs) call for async scheduling (fixes batch corruption where all decode input_ids were -1)
◦ Changed self.positions from CpuGpuBuffer to plain GPU tensor for PCP case (fixes TypeError: 'CpuGpuBuffer' object is not subscriptable in upstream _preprocess)
◦ Fixed _get_cumsum_and_arange to use self.query_pos.np as output buffer instead of self.arange_np (prevents aliasing/corruption)
3 attention_v1.py — Added tensor type detection for kv_cache (upstream changed from list [k,v] to tensor with shape (2, ...))
4 mla.py — Version-guarded self.mla_attn.kv_cache access (no longer indexed by virtual_engine in new vLLM)
5 npu.py — Updated import path from vllm.v1.kv_offload.backends.cpu.CPUBackend to vllm.v1.kv_offload.cpu.manager.CPUOffloadingManager
6 patch_qwen3_5.py & patch_qwen3_next.py — Version-guarded self.kv_cache access (same fix as mla.py)
Local test results:
• test_eager_mode_acc (Qwen3-0.6B): PASSED
• test_models (MiniCPM-2B, MiniCPM4-0.5B): PASSED
• test_sampler (3 tests): PASSED
• test_batch_invariant (user-confirmed): PASSED
• Batch inference smoke test (3 prompts): PASSED with correct output
Remaining CI failures (not code bugs):
• Memory issues (0.9 gpu_memory_utilization > available free memory)
• Model download failures (network/HF connectivity)
• DeepSeek-V2-Lite-W8A8 quantization accuracy (pre-existing)
• PCP NumPy boolean indexing (edge case in pcp_utils.py, likely scheduling-dependent)

Detail fix process:
[
async_sheduler_zero_bubble_spec_decoding.md
](url)

github-actions · 2026-03-29T05:24:44Z

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

Signed-off-by: wangbj127 <wangbj1207@126.com>

Signed-off-by: leo-pony <nengjunma@outlook.com>

…eam, see vllm pr:35007 Signed-off-by: leo-pony <nengjunma@outlook.com>

… PR:32951 Signed-off-by: leo-pony <nengjunma@outlook.com>

Signed-off-by: leo-pony <nengjunma@outlook.com>

MengqingCao · 2026-04-03T08:32:15Z

            discard_requests_mask = original_seq_lens_np < num_tokens_np
        else:
-            discard_requests_mask = self.seq_lens.np[:num_reqs] < num_tokens_np
+            discard_requests_mask = self.seq_lens.cpu().numpy()[:num_reqs] < num_tokens_np


plz use optimistic_seq_lens_cpu instead

Signed-off-by: leo-pony <nengjunma@outlook.com>

This reverts commit 811271d.

@YangShuai52

### What this PR does / why we need it? main2main upgrade to vllm 0324. fix breaks: 1. PR [#37487](vllm-project/vllm#37487) [V0 Deprecation] Refactor kv cache from list to element (c59a132) — self.kv_cache from list[tensor]（per virtual engine）changed to tensor 2. PR [#37874](vllm-project/vllm#37874) [KV Offload] Refactor CPU offloading: pluggable CachePolicy, remove Backend abstraction, restructure into cpu/ package (e3c6c10) — LRUOffloadingManager + CPUBackend been refactor to CPUOffloadingManager 3. PR [#32951](vllm-project/vllm#32951) [Async][Spec Decoding] Zero-bubble async scheduling + spec decoding (fafe76b) — a) changes self.positions and self.seq_lens from CpuGpuBuffer to plain GPU tensor; b) change _get_cumsum_and_arange output paramter. Another _prepare_input_ids add num_reqs. 5. PR [#35007](https://github.com/vllm-project/vllm/pull/35007)[Bugfix] Register VLLM_BATCH_INVARIANT in envs.py to fix spurious unknown env var warning (dc6908a) — delete vllm_is_batch_invariant() and const variable VLLM_BATCH_INVARIANT，replace with vllm.envs Know issues: 1.310p Qwen3.5 test failed for qwen3.5 patch failure, see issue: vllm-project#7976 @YangShuai52 is fixing. ### Does this PR introduce _any_ user-facing change? 1. As Zero Async Scheduler + spec decode needs _compute_slot_mapping_kernel of NPU and corresponding accepted draft token validation delaye suppots see PR vllm-project#7640 , this PR make this change: when in spec decode case close the async scheduler. In this way, the Main2Main can be developed in parallel with Spec Decode + Async scheduler, util next release version. Co-Authored-By: zhaomingyu <zhaomingyu13@h-partners.com> wangbj127 <wangbj1207@126.com> SidaoY <1024863041@qq.com> 22dimensions <waitingwind@foxmail.com> - vLLM main: vllm-project/vllm@35141a7 --------- Signed-off-by: 22dimensions <waitingwind@foxmail.com> Signed-off-by: leo-pony <nengjunma@outlook.com> Signed-off-by: Your Name <you@example.com> Signed-off-by: wangbj127 <wangbj1207@126.com> Co-authored-by: 22dimensions <waitingwind@foxmail.com> Co-authored-by: Claude Code <claude@anthropic.com> Co-authored-by: Claude Code <noreply@anthropic.com> Co-authored-by: Your Name <you@example.com> Co-authored-by: wangbj127 <wangbj1207@126.com> Signed-off-by: guxin108 <1252896542@qq.com>

@YangShuai52

### What this PR does / why we need it? main2main upgrade to vllm 0324. fix breaks: 1. PR [#37487](vllm-project/vllm#37487) [V0 Deprecation] Refactor kv cache from list to element (c59a132) — self.kv_cache from list[tensor]（per virtual engine）changed to tensor 2. PR [#37874](vllm-project/vllm#37874) [KV Offload] Refactor CPU offloading: pluggable CachePolicy, remove Backend abstraction, restructure into cpu/ package (e3c6c10) — LRUOffloadingManager + CPUBackend been refactor to CPUOffloadingManager 3. PR [#32951](vllm-project/vllm#32951) [Async][Spec Decoding] Zero-bubble async scheduling + spec decoding (fafe76b) — a) changes self.positions and self.seq_lens from CpuGpuBuffer to plain GPU tensor; b) change _get_cumsum_and_arange output paramter. Another _prepare_input_ids add num_reqs. 5. PR [#35007](https://github.com/vllm-project/vllm/pull/35007)[Bugfix] Register VLLM_BATCH_INVARIANT in envs.py to fix spurious unknown env var warning (dc6908a) — delete vllm_is_batch_invariant() and const variable VLLM_BATCH_INVARIANT，replace with vllm.envs Know issues: 1.310p Qwen3.5 test failed for qwen3.5 patch failure, see issue: vllm-project#7976 @YangShuai52 is fixing. ### Does this PR introduce _any_ user-facing change? 1. As Zero Async Scheduler + spec decode needs _compute_slot_mapping_kernel of NPU and corresponding accepted draft token validation delaye suppots see PR vllm-project#7640 , this PR make this change: when in spec decode case close the async scheduler. In this way, the Main2Main can be developed in parallel with Spec Decode + Async scheduler, util next release version. Co-Authored-By: zhaomingyu <zhaomingyu13@h-partners.com> wangbj127 <wangbj1207@126.com> SidaoY <1024863041@qq.com> 22dimensions <waitingwind@foxmail.com> - vLLM main: vllm-project/vllm@35141a7 --------- Signed-off-by: 22dimensions <waitingwind@foxmail.com> Signed-off-by: leo-pony <nengjunma@outlook.com> Signed-off-by: Your Name <you@example.com> Signed-off-by: wangbj127 <wangbj1207@126.com> Co-authored-by: 22dimensions <waitingwind@foxmail.com> Co-authored-by: Claude Code <claude@anthropic.com> Co-authored-by: Claude Code <noreply@anthropic.com> Co-authored-by: Your Name <you@example.com> Co-authored-by: wangbj127 <wangbj1207@126.com>

@YangShuai52

### What this PR does / why we need it? main2main upgrade to vllm 0324. fix breaks: 1. PR [#37487](vllm-project/vllm#37487) [V0 Deprecation] Refactor kv cache from list to element (c59a132) — self.kv_cache from list[tensor]（per virtual engine）changed to tensor 2. PR [#37874](vllm-project/vllm#37874) [KV Offload] Refactor CPU offloading: pluggable CachePolicy, remove Backend abstraction, restructure into cpu/ package (e3c6c10) — LRUOffloadingManager + CPUBackend been refactor to CPUOffloadingManager 3. PR [#32951](vllm-project/vllm#32951) [Async][Spec Decoding] Zero-bubble async scheduling + spec decoding (fafe76b) — a) changes self.positions and self.seq_lens from CpuGpuBuffer to plain GPU tensor; b) change _get_cumsum_and_arange output paramter. Another _prepare_input_ids add num_reqs. 5. PR [#35007](https://github.com/vllm-project/vllm/pull/35007)[Bugfix] Register VLLM_BATCH_INVARIANT in envs.py to fix spurious unknown env var warning (dc6908a) — delete vllm_is_batch_invariant() and const variable VLLM_BATCH_INVARIANT，replace with vllm.envs Know issues: 1.310p Qwen3.5 test failed for qwen3.5 patch failure, see issue: vllm-project#7976 @YangShuai52 is fixing. ### Does this PR introduce _any_ user-facing change? 1. As Zero Async Scheduler + spec decode needs _compute_slot_mapping_kernel of NPU and corresponding accepted draft token validation delaye suppots see PR vllm-project#7640 , this PR make this change: when in spec decode case close the async scheduler. In this way, the Main2Main can be developed in parallel with Spec Decode + Async scheduler, util next release version. Co-Authored-By: zhaomingyu <zhaomingyu13@h-partners.com> wangbj127 <wangbj1207@126.com> SidaoY <1024863041@qq.com> 22dimensions <waitingwind@foxmail.com> - vLLM main: vllm-project/vllm@35141a7 --------- Signed-off-by: 22dimensions <waitingwind@foxmail.com> Signed-off-by: leo-pony <nengjunma@outlook.com> Signed-off-by: Your Name <you@example.com> Signed-off-by: wangbj127 <wangbj1207@126.com> Co-authored-by: 22dimensions <waitingwind@foxmail.com> Co-authored-by: Claude Code <claude@anthropic.com> Co-authored-by: Claude Code <noreply@anthropic.com> Co-authored-by: Your Name <you@example.com> Co-authored-by: wangbj127 <wangbj1207@126.com> Signed-off-by: zouyida2052 <zouyida2002@gmail.com>

@YangShuai52

### What this PR does / why we need it? main2main upgrade to vllm 0324. fix breaks: 1. PR [#37487](vllm-project/vllm#37487) [V0 Deprecation] Refactor kv cache from list to element (c59a132) — self.kv_cache from list[tensor]（per virtual engine）changed to tensor 2. PR [#37874](vllm-project/vllm#37874) [KV Offload] Refactor CPU offloading: pluggable CachePolicy, remove Backend abstraction, restructure into cpu/ package (e3c6c10) — LRUOffloadingManager + CPUBackend been refactor to CPUOffloadingManager 3. PR [#32951](vllm-project/vllm#32951) [Async][Spec Decoding] Zero-bubble async scheduling + spec decoding (fafe76b) — a) changes self.positions and self.seq_lens from CpuGpuBuffer to plain GPU tensor; b) change _get_cumsum_and_arange output paramter. Another _prepare_input_ids add num_reqs. 5. PR [#35007](https://github.com/vllm-project/vllm/pull/35007)[Bugfix] Register VLLM_BATCH_INVARIANT in envs.py to fix spurious unknown env var warning (dc6908a) — delete vllm_is_batch_invariant() and const variable VLLM_BATCH_INVARIANT，replace with vllm.envs Know issues: 1.310p Qwen3.5 test failed for qwen3.5 patch failure, see issue: vllm-project#7976 @YangShuai52 is fixing. ### Does this PR introduce _any_ user-facing change? 1. As Zero Async Scheduler + spec decode needs _compute_slot_mapping_kernel of NPU and corresponding accepted draft token validation delaye suppots see PR vllm-project#7640 , this PR make this change: when in spec decode case close the async scheduler. In this way, the Main2Main can be developed in parallel with Spec Decode + Async scheduler, util next release version. Co-Authored-By: zhaomingyu <zhaomingyu13@h-partners.com> wangbj127 <wangbj1207@126.com> SidaoY <1024863041@qq.com> 22dimensions <waitingwind@foxmail.com> - vLLM main: vllm-project/vllm@35141a7 --------- Signed-off-by: 22dimensions <waitingwind@foxmail.com> Signed-off-by: leo-pony <nengjunma@outlook.com> Signed-off-by: Your Name <you@example.com> Signed-off-by: wangbj127 <wangbj1207@126.com> Co-authored-by: 22dimensions <waitingwind@foxmail.com> Co-authored-by: Claude Code <claude@anthropic.com> Co-authored-by: Claude Code <noreply@anthropic.com> Co-authored-by: Your Name <you@example.com> Co-authored-by: wangbj127 <wangbj1207@126.com>

@YangShuai52

### What this PR does / why we need it? main2main upgrade to vllm 0324. fix breaks: 1. PR [#37487](vllm-project/vllm#37487) [V0 Deprecation] Refactor kv cache from list to element (c59a132) — self.kv_cache from list[tensor]（per virtual engine）changed to tensor 2. PR [#37874](vllm-project/vllm#37874) [KV Offload] Refactor CPU offloading: pluggable CachePolicy, remove Backend abstraction, restructure into cpu/ package (e3c6c10) — LRUOffloadingManager + CPUBackend been refactor to CPUOffloadingManager 3. PR [#32951](vllm-project/vllm#32951) [Async][Spec Decoding] Zero-bubble async scheduling + spec decoding (fafe76b) — a) changes self.positions and self.seq_lens from CpuGpuBuffer to plain GPU tensor; b) change _get_cumsum_and_arange output paramter. Another _prepare_input_ids add num_reqs. 5. PR [#35007](https://github.com/vllm-project/vllm/pull/35007)[Bugfix] Register VLLM_BATCH_INVARIANT in envs.py to fix spurious unknown env var warning (dc6908a) — delete vllm_is_batch_invariant() and const variable VLLM_BATCH_INVARIANT，replace with vllm.envs Know issues: 1.310p Qwen3.5 test failed for qwen3.5 patch failure, see issue: vllm-project#7976 @YangShuai52 is fixing. ### Does this PR introduce _any_ user-facing change? 1. As Zero Async Scheduler + spec decode needs _compute_slot_mapping_kernel of NPU and corresponding accepted draft token validation delaye suppots see PR vllm-project#7640 , this PR make this change: when in spec decode case close the async scheduler. In this way, the Main2Main can be developed in parallel with Spec Decode + Async scheduler, util next release version. Co-Authored-By: zhaomingyu <zhaomingyu13@h-partners.com> wangbj127 <wangbj1207@126.com> SidaoY <1024863041@qq.com> 22dimensions <waitingwind@foxmail.com> - vLLM main: vllm-project/vllm@35141a7 --------- Signed-off-by: 22dimensions <waitingwind@foxmail.com> Signed-off-by: leo-pony <nengjunma@outlook.com> Signed-off-by: Your Name <you@example.com> Signed-off-by: wangbj127 <wangbj1207@126.com> Co-authored-by: 22dimensions <waitingwind@foxmail.com> Co-authored-by: Claude Code <claude@anthropic.com> Co-authored-by: Claude Code <noreply@anthropic.com> Co-authored-by: Your Name <you@example.com> Co-authored-by: wangbj127 <wangbj1207@126.com> Signed-off-by: nanxing <1014662416@qq.com>

leo-pony requested review from LCAIZJ, MengqingCao, Yikun, nalinaly, realliujiaxu, wangxiyuan, weijinqian0, whx-sjtu and zzzzwwjj as code owners March 28, 2026 08:33

github-actions Bot added documentation Improvements or additions to documentation ci/build module:ops module:core labels Mar 28, 2026

leo-pony 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 Mar 28, 2026

gemini-code-assist Bot reviewed Mar 28, 2026

View reviewed changes

github-actions Bot added the merge-conflicts label Mar 29, 2026

leo-pony force-pushed the main2main_0324_qz branch from 453b20c to dfc18b8 Compare March 31, 2026 06:21

github-actions Bot removed the merge-conflicts label Mar 31, 2026

leo-pony changed the title ~~Main2main 0324~~ [CI] Main2main 0324 Mar 31, 2026

leo-pony force-pushed the main2main_0324_qz branch 2 times, most recently from 8264094 to 14a1449 Compare March 31, 2026 09:49

wangbj127 and others added 2 commits April 2, 2026 18:07

Fix model_runner_v1.py and cancel changes in attention_v1.py

bdee2e2

Signed-off-by: wangbj127 <wangbj1207@126.com>

Optmize the unneeded dtype translate

5719e82

Signed-off-by: leo-pony <nengjunma@outlook.com>

leo-pony changed the title ~~[CI] Main2main 0324~~ [CI] Main2main upgrade to 0324 Apr 2, 2026

leo-pony removed ready read for review ready-for-test start test by label for PR labels Apr 3, 2026

Replace vllm_is_batch_invariant with envs, to keep consist with upstr…

7fef40e

…eam, see vllm pr:35007 Signed-off-by: leo-pony <nengjunma@outlook.com>

leo-pony force-pushed the main2main_0324_qz branch from a0dffc1 to 7fef40e Compare April 3, 2026 02:26

leo-pony added ready read for review ready-for-test start test by label for PR and removed ready read for review ready-for-test start test by label for PR labels Apr 3, 2026

fix 310P _prepare_input_ids less num_reqs param, break import in vllm…

43ee1a3

… PR:32951 Signed-off-by: leo-pony <nengjunma@outlook.com>

leo-pony force-pushed the main2main_0324_qz branch from 8c6ccbe to 43ee1a3 Compare April 3, 2026 03:33

Skip the 310P qwen3.5 test case and make test case more stable

2c6d795

Signed-off-by: leo-pony <nengjunma@outlook.com>

leo-pony force-pushed the main2main_0324_qz branch from fa3cd4d to 2c6d795 Compare April 3, 2026 04:48

leo-pony added ready read for review ready-for-test start test by label for PR labels Apr 3, 2026

This was referenced Apr 3, 2026

[Bug]: Main2main 310p Qwen3.5 test failed for qwen3.5 patch failure #7976

Closed

[Bug]: Main2main 0324 Spec Decode + Async Scheduler failure #7979

Closed

MengqingCao reviewed Apr 3, 2026

View reviewed changes

Optimize the performance: remove seqs_len device to host

e56fd9c

Signed-off-by: leo-pony <nengjunma@outlook.com>

MengqingCao self-assigned this Apr 3, 2026

MengqingCao approved these changes Apr 3, 2026

View reviewed changes

MengqingCao merged commit 811271d into vllm-project:main Apr 3, 2026
39 checks passed

csoulnd added a commit to csoulnd/vllm-ascend that referenced this pull request Apr 7, 2026

Revert "[CI] Main2main upgrade to 0324 (vllm-project#7787)"

025339e

This reverts commit 811271d.

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

[CI] Main2main upgrade to 0324 #7787

[CI] Main2main upgrade to 0324 #7787
MengqingCao merged 21 commits into
vllm-project:mainfrom
leo-pony:main2main_0324_qz

leo-pony commented Mar 28, 2026 •

edited

Loading

Uh oh!

github-actions Bot commented Mar 28, 2026

Uh oh!

gemini-code-assist Bot commented Mar 28, 2026

Uh oh!

gemini-code-assist Bot left a comment

Uh oh!

leo-pony commented Mar 28, 2026 •

edited

Loading

Uh oh!

github-actions Bot commented Mar 29, 2026

Uh oh!

MengqingCao Apr 3, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

5 participants

Conversation

leo-pony commented Mar 28, 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 Mar 28, 2026

Uh oh!

gemini-code-assist Bot commented Mar 28, 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!

leo-pony commented Mar 28, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

github-actions Bot commented Mar 29, 2026

Uh oh!

MengqingCao Apr 3, 2026

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

5 participants

leo-pony commented Mar 28, 2026 •

edited

Loading

leo-pony commented Mar 28, 2026 •

edited

Loading