[BugFix] Address PrefillCacheHit state to fix prefix cache accuracy bug#1498
Merged
wangxiyuan merged 1 commit intovllm-project:mainfrom Jun 30, 2025
Merged
Conversation
Signed-off-by: whx-sjtu <2952154980@qq.com>
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## main #1498 +/- ##
==========================================
+ Coverage 27.39% 31.65% +4.25%
==========================================
Files 56 60 +4
Lines 6191 6638 +447
==========================================
+ Hits 1696 2101 +405
- Misses 4495 4537 +42
Flags with carried forward coverage won't be shown. Click here to find out more. ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
Member
|
Thanks for fixing it, how can we prevent this precision regresion break again? And please fulfill commit msg |
Collaborator
|
Add e2e test in #1505 |
Yikun
changed the title
wangxiyuan
approved these changes
Jun 30, 2025
Yikun
pushed a commit
that referenced
this pull request
Jul 2, 2025
…uler (#1505) ### What this PR does / why we need it? Add test for chunked prefill and prefix cache on v1/AscendScheduler Covered scenarios: - `Qwen/Qwen3-0.6B-Base` and `deepseek-ai/DeepSeek-V2-Lite-Chat` --- multicard CI time increased by 19 min - `V1 + default scheduler` vs `V1 + default scheduler + enable prefix cache` - `V1 + Ascend scheduler` vs `V1 + Ascend scheduler + enable prefix cache` vs `V1 + Ascend scheduler + enable prefix cache + enable chunked prefill` - `Qwen/Qwen3-0.6B-Base` --- singlecard CI time increased by 8 min - `V1 + Ascend scheduler` vs `V1 + Ascend scheduler + enable chunked prefill` should rebase after #1498 and #1446 ### Does this PR introduce _any_ user-facing change? N/A ### How was this patch tested? CI passed with new added test. Signed-off-by: MengqingCao <cmq0113@163.com>
ZhengWG
pushed a commit
to ZhengWG/vllm-ascend
that referenced
this pull request
Jul 3, 2025
…uler (vllm-project#1505) ### What this PR does / why we need it? Add test for chunked prefill and prefix cache on v1/AscendScheduler Covered scenarios: - `Qwen/Qwen3-0.6B-Base` and `deepseek-ai/DeepSeek-V2-Lite-Chat` --- multicard CI time increased by 19 min - `V1 + default scheduler` vs `V1 + default scheduler + enable prefix cache` - `V1 + Ascend scheduler` vs `V1 + Ascend scheduler + enable prefix cache` vs `V1 + Ascend scheduler + enable prefix cache + enable chunked prefill` - `Qwen/Qwen3-0.6B-Base` --- singlecard CI time increased by 8 min - `V1 + Ascend scheduler` vs `V1 + Ascend scheduler + enable chunked prefill` should rebase after vllm-project#1498 and vllm-project#1446 ### Does this PR introduce _any_ user-facing change? N/A ### How was this patch tested? CI passed with new added test. Signed-off-by: MengqingCao <cmq0113@163.com> Signed-off-by: ZhengWG <zwg0606@gmail.com>
ZhengWG
pushed a commit
to ZhengWG/vllm-ascend
that referenced
this pull request
Jul 3, 2025
…uler (vllm-project#1505) ### What this PR does / why we need it? Add test for chunked prefill and prefix cache on v1/AscendScheduler Covered scenarios: - `Qwen/Qwen3-0.6B-Base` and `deepseek-ai/DeepSeek-V2-Lite-Chat` --- multicard CI time increased by 19 min - `V1 + default scheduler` vs `V1 + default scheduler + enable prefix cache` - `V1 + Ascend scheduler` vs `V1 + Ascend scheduler + enable prefix cache` vs `V1 + Ascend scheduler + enable prefix cache + enable chunked prefill` - `Qwen/Qwen3-0.6B-Base` --- singlecard CI time increased by 8 min - `V1 + Ascend scheduler` vs `V1 + Ascend scheduler + enable chunked prefill` should rebase after vllm-project#1498 and vllm-project#1446 ### Does this PR introduce _any_ user-facing change? N/A ### How was this patch tested? CI passed with new added test. Signed-off-by: MengqingCao <cmq0113@163.com> Signed-off-by: ZhengWG <zwg0606@gmail.com>
chopper0126
pushed a commit
to chopper0126/vllm-ascend
that referenced
this pull request
Oct 16, 2025
…ug (vllm-project#1498) When use AscendScheduler with prefix-cache enabled and chunk-prefill disabled, there will be accuray problem because there is no branch in mla_v1 to process this scenario. This PR fixes it. Signed-off-by: whx-sjtu <2952154980@qq.com>
chopper0126
pushed a commit
to chopper0126/vllm-ascend
that referenced
this pull request
Oct 16, 2025
…uler (vllm-project#1505) ### What this PR does / why we need it? Add test for chunked prefill and prefix cache on v1/AscendScheduler Covered scenarios: - `Qwen/Qwen3-0.6B-Base` and `deepseek-ai/DeepSeek-V2-Lite-Chat` --- multicard CI time increased by 19 min - `V1 + default scheduler` vs `V1 + default scheduler + enable prefix cache` - `V1 + Ascend scheduler` vs `V1 + Ascend scheduler + enable prefix cache` vs `V1 + Ascend scheduler + enable prefix cache + enable chunked prefill` - `Qwen/Qwen3-0.6B-Base` --- singlecard CI time increased by 8 min - `V1 + Ascend scheduler` vs `V1 + Ascend scheduler + enable chunked prefill` should rebase after vllm-project#1498 and vllm-project#1446 ### Does this PR introduce _any_ user-facing change? N/A ### How was this patch tested? CI passed with new added test. Signed-off-by: MengqingCao <cmq0113@163.com>
Angazenn
pushed a commit
to Angazenn/vllm-ascend
that referenced
this pull request
Oct 21, 2025
…ug (vllm-project#1498) When use AscendScheduler with prefix-cache enabled and chunk-prefill disabled, there will be accuray problem because there is no branch in mla_v1 to process this scenario. This PR fixes it. Signed-off-by: whx-sjtu <2952154980@qq.com>
Angazenn
pushed a commit
to Angazenn/vllm-ascend
that referenced
this pull request
Oct 21, 2025
…uler (vllm-project#1505) ### What this PR does / why we need it? Add test for chunked prefill and prefix cache on v1/AscendScheduler Covered scenarios: - `Qwen/Qwen3-0.6B-Base` and `deepseek-ai/DeepSeek-V2-Lite-Chat` --- multicard CI time increased by 19 min - `V1 + default scheduler` vs `V1 + default scheduler + enable prefix cache` - `V1 + Ascend scheduler` vs `V1 + Ascend scheduler + enable prefix cache` vs `V1 + Ascend scheduler + enable prefix cache + enable chunked prefill` - `Qwen/Qwen3-0.6B-Base` --- singlecard CI time increased by 8 min - `V1 + Ascend scheduler` vs `V1 + Ascend scheduler + enable chunked prefill` should rebase after vllm-project#1498 and vllm-project#1446 ### Does this PR introduce _any_ user-facing change? N/A ### How was this patch tested? CI passed with new added test. Signed-off-by: MengqingCao <cmq0113@163.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
When use AscendScheduler with prefix-cache enabled and chunk-prefill disabled, there will be accuray problem because there is no branch in mla_v1 to process this scenario. This PR fixes it.