[v0.9.1][Fix] Fix block table shape #1297
Merged
Conversation
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
Signed-off-by: Yizhou Liu <liu_yizhou@outlook.com>
yiz-liu
force-pushed
the
fix-prefix-cache
branch
from
a0ba5ef to
e8ba3e2
Compare
|
I have tested myself and observed no accuracy issues. However, I recommend adding accuracy tests for this scenario to catch any potential bugs. We should take a look into it @MengqingCao @ganyi1996ppo @wangxiyuan Please review and merge this at your earliest convenience, as this bug can be triggered easily and has a broad impact. |
yiz-liu
changed the title
wangxiyuan
approved these changes
Jun 26, 2025
wangxiyuan
pushed a commit
that referenced
this pull request
Jun 30, 2025
…heduler (#1446) ### What this PR does / why we need it? This fix the shape of block_table which was introduced by hybrid kv groups several weeks ago. Error will be raised when enable prefix-cache (eager or not) and Ascend Scheduler at the same time, just send two identical requests and it will reproduce. v0.9.1: #1297 ### Does this PR introduce _any_ user-facing change? No ### How was this patch tested? Test manually Signed-off-by: Yizhou Liu <liu_yizhou@outlook.com>
weijinqian0
pushed a commit
to weijinqian0/vllm-ascend
that referenced
this pull request
Jun 30, 2025
…heduler (vllm-project#1446) ### What this PR does / why we need it? This fix the shape of block_table which was introduced by hybrid kv groups several weeks ago. Error will be raised when enable prefix-cache (eager or not) and Ascend Scheduler at the same time, just send two identical requests and it will reproduce. v0.9.1: vllm-project#1297 ### Does this PR introduce _any_ user-facing change? No ### How was this patch tested? Test manually Signed-off-by: Yizhou Liu <liu_yizhou@outlook.com>
Yikun
added a commit
that referenced
this pull request
Jul 6, 2025
### What this PR does / why we need it? Fix the shape of the `npu_moe_init_routing` input parameters to support aclgraph mode on qwen3-moe In addition to this PR, resolving the `gatherv3` error might be necessary. See related PR #1297 #1446 Thanks to @yiz-liu for providing the idea ### Does this PR introduce _any_ user-facing change? No ### How was this patch tested? Tested on Qwen3-30B-A3B Closes: #1368 --------- Signed-off-by: ApsarasX <apsarax@outlook.com> Signed-off-by: Yikun Jiang <yikunkero@gmail.com> Co-authored-by: Yizhou Liu <liu_yizhou@outlook.com> Co-authored-by: Yikun Jiang <yikunkero@gmail.com>
chopper0126
pushed a commit
to chopper0126/vllm-ascend
that referenced
this pull request
Oct 16, 2025
…heduler (vllm-project#1446) ### What this PR does / why we need it? This fix the shape of block_table which was introduced by hybrid kv groups several weeks ago. Error will be raised when enable prefix-cache (eager or not) and Ascend Scheduler at the same time, just send two identical requests and it will reproduce. v0.9.1: vllm-project#1297 ### Does this PR introduce _any_ user-facing change? No ### How was this patch tested? Test manually Signed-off-by: Yizhou Liu <liu_yizhou@outlook.com>
chopper0126
pushed a commit
to chopper0126/vllm-ascend
that referenced
this pull request
Oct 16, 2025
### What this PR does / why we need it? Fix the shape of the `npu_moe_init_routing` input parameters to support aclgraph mode on qwen3-moe In addition to this PR, resolving the `gatherv3` error might be necessary. See related PR vllm-project#1297 vllm-project#1446 Thanks to @yiz-liu for providing the idea ### Does this PR introduce _any_ user-facing change? No ### How was this patch tested? Tested on Qwen3-30B-A3B Closes: vllm-project#1368 --------- Signed-off-by: ApsarasX <apsarax@outlook.com> Signed-off-by: Yikun Jiang <yikunkero@gmail.com> Co-authored-by: Yizhou Liu <liu_yizhou@outlook.com> Co-authored-by: Yikun Jiang <yikunkero@gmail.com>
Angazenn
pushed a commit
to Angazenn/vllm-ascend
that referenced
this pull request
Oct 21, 2025
…heduler (vllm-project#1446) ### What this PR does / why we need it? This fix the shape of block_table which was introduced by hybrid kv groups several weeks ago. Error will be raised when enable prefix-cache (eager or not) and Ascend Scheduler at the same time, just send two identical requests and it will reproduce. v0.9.1: vllm-project#1297 ### Does this PR introduce _any_ user-facing change? No ### How was this patch tested? Test manually Signed-off-by: Yizhou Liu <liu_yizhou@outlook.com>
Angazenn
pushed a commit
to Angazenn/vllm-ascend
that referenced
this pull request
Oct 21, 2025
### What this PR does / why we need it? Fix the shape of the `npu_moe_init_routing` input parameters to support aclgraph mode on qwen3-moe In addition to this PR, resolving the `gatherv3` error might be necessary. See related PR vllm-project#1297 vllm-project#1446 Thanks to @yiz-liu for providing the idea ### Does this PR introduce _any_ user-facing change? No ### How was this patch tested? Tested on Qwen3-30B-A3B Closes: vllm-project#1368 --------- Signed-off-by: ApsarasX <apsarax@outlook.com> Signed-off-by: Yikun Jiang <yikunkero@gmail.com> Co-authored-by: Yizhou Liu <liu_yizhou@outlook.com> Co-authored-by: Yikun Jiang <yikunkero@gmail.com>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 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.
What this PR does / why we need it?
This fix the shape of block_table which was introduced by hybrid kv groups several weeks ago.
Error will be raised when enable prefix-cache (eager or not) and Ascend Scheduler at the same time, just send two identical requests and it will reproduce.
Backported: #1446
Does this PR introduce any user-facing change?
How was this patch tested?