[bugfix] fix the complex and potentially problematic generate_kv_idx.#5957
Merged
wangxiyuan merged 1 commit intovllm-project:mainfrom Jan 21, 2026
Merged
[bugfix] fix the complex and potentially problematic generate_kv_idx.#5957wangxiyuan merged 1 commit intovllm-project:mainfrom
wangxiyuan merged 1 commit intovllm-project:mainfrom
Conversation
pisceskkk
requested review from
MengqingCao,
wangxiyuan and
weijinqian0
as code owners
Contributor
|
👋 Hi! Thank you for contributing to the vLLM Ascend project. The following points will speed up your PR merge:
If CI fails, you can run linting and testing checks locally according Contributing and Testing. |
![gemini-code-assist[bot]](https://avatars.githubusercontent.com/in/956858?s=60&v=4){kind=small}
Contributor
There was a problem hiding this comment.
Code Review
This pull request refactors the logic for generating KV cache indices in chunked prefill scenarios by removing the complex generate_kv_idx function. The new approach derives the necessary indices directly from existing metadata, which simplifies the code and fixes a dimension misalignment issue. The changes look good and improve maintainability. I've identified one area where the new implementation can be made more efficient by avoiding a redundant argsort operation.
weiguihua2
added
ready
Contributor
|
This pull request has conflicts, please resolve those before we can evaluate the pull request. |
pisceskkk
force-pushed
the
fix_cp
branch
6 times, most recently
from
eb48feb to
871ab62
Compare
Signed-off-by: QiuChunshuo <qiuchunshuo@huawei.com>
wangxiyuan
approved these changes
Jan 21, 2026
yiz-liu
pushed a commit
that referenced
this pull request
Jan 21, 2026
…atic generate_kv_idx. (#5955) ### What this PR does / why we need it? In long-sequence scenarios, the chunked-prefill component may encounter dimension misalignment issues, which previously occurred during precision testing on the code_generate_lite dataset. This PR removes redundant computations and instead derives the value using existing results and straightforward calculations. ref: #5957 Signed-off-by: QiuChunshuo <qiuchunshuo@huawei.com>
huangfeifei1995
pushed a commit
to huangfeifei1995/vllm-ascend
that referenced
this pull request
Jan 21, 2026
…vllm-project#5957) ### What this PR does / why we need it? In long-sequence scenarios, the chunked-prefill component may encounter dimension misalignment issues, which previously occurred during precision testing on the code_generate_lite dataset. This PR removes redundant computations and instead derives the value using existing results and straightforward calculations. - vLLM version: v0.13.0 - vLLM main: vllm-project/vllm@2c24bc6 Signed-off-by: QiuChunshuo <qiuchunshuo@huawei.com> Signed-off-by: huangning1995 <huangning12@huawei.com>
huangfeifei1995
added a commit
to huangfeifei1995/vllm-ascend
that referenced
this pull request
Jan 21, 2026
…_kv_idx. (vllm-project#5957)" This reverts commit 3d916f4.
845473182
pushed a commit
to 845473182/vllm-ascend
that referenced
this pull request
Jan 22, 2026
…to FIA_rebase * 'main' of https://github.com/vllm-project/vllm-ascend: [CI] Upgrade CANN to 8.5.0 (vllm-project#6070) Default enable MLAPO (vllm-project#5952) [Doc] Supplement PD separation parameters of DeepSeek V3.1 (vllm-project#6053) [Ascend] perf: optimize rope embedding with triton kernel for huge performance gain (vllm-project#5918) [Ops] update causal_conv1d_update (vllm-project#5984) [CI]Update triton ascend version in 3.2.0 (vllm-project#6067) [bugfix] fix the complex and potentially problematic generate_kv_idx. (vllm-project#5957)
starmountain1997
pushed a commit
to starmountain1997/vllm-ascend
that referenced
this pull request
Jan 31, 2026
…vllm-project#5957) ### What this PR does / why we need it? In long-sequence scenarios, the chunked-prefill component may encounter dimension misalignment issues, which previously occurred during precision testing on the code_generate_lite dataset. This PR removes redundant computations and instead derives the value using existing results and straightforward calculations. - vLLM version: v0.13.0 - vLLM main: vllm-project/vllm@2c24bc6 Signed-off-by: QiuChunshuo <qiuchunshuo@huawei.com>
starmountain1997
pushed a commit
to starmountain1997/vllm-ascend
that referenced
this pull request
Jan 31, 2026
…atic generate_kv_idx. (vllm-project#5955) ### What this PR does / why we need it? In long-sequence scenarios, the chunked-prefill component may encounter dimension misalignment issues, which previously occurred during precision testing on the code_generate_lite dataset. This PR removes redundant computations and instead derives the value using existing results and straightforward calculations. ref: vllm-project#5957 Signed-off-by: QiuChunshuo <qiuchunshuo@huawei.com>
starmountain1997
pushed a commit
to starmountain1997/vllm-ascend
that referenced
this pull request
Jan 31, 2026
…vllm-project#5957) ### What this PR does / why we need it? In long-sequence scenarios, the chunked-prefill component may encounter dimension misalignment issues, which previously occurred during precision testing on the code_generate_lite dataset. This PR removes redundant computations and instead derives the value using existing results and straightforward calculations. - vLLM version: v0.13.0 - vLLM main: vllm-project/vllm@2c24bc6 Signed-off-by: QiuChunshuo <qiuchunshuo@huawei.com>
tangtiangu
pushed a commit
to tangtiangu/jiusi-vllm-ascend
that referenced
this pull request
Feb 24, 2026
…atic generate_kv_idx. (vllm-project#5955) ### What this PR does / why we need it? In long-sequence scenarios, the chunked-prefill component may encounter dimension misalignment issues, which previously occurred during precision testing on the code_generate_lite dataset. This PR removes redundant computations and instead derives the value using existing results and straightforward calculations. ref: vllm-project#5957 Signed-off-by: QiuChunshuo <qiuchunshuo@huawei.com>
tangtiangu
pushed a commit
to tangtiangu/jiusi-vllm-ascend
that referenced
this pull request
Feb 24, 2026
…atic generate_kv_idx. (vllm-project#5955) ### What this PR does / why we need it? In long-sequence scenarios, the chunked-prefill component may encounter dimension misalignment issues, which previously occurred during precision testing on the code_generate_lite dataset. This PR removes redundant computations and instead derives the value using existing results and straightforward calculations. ref: vllm-project#5957 Signed-off-by: QiuChunshuo <qiuchunshuo@huawei.com>
ZRJ026
pushed a commit
to ZRJ026/vllm-ascend
that referenced
this pull request
Feb 28, 2026
…vllm-project#5957) ### What this PR does / why we need it? In long-sequence scenarios, the chunked-prefill component may encounter dimension misalignment issues, which previously occurred during precision testing on the code_generate_lite dataset. This PR removes redundant computations and instead derives the value using existing results and straightforward calculations. - vLLM version: v0.13.0 - vLLM main: vllm-project/vllm@2c24bc6 Signed-off-by: QiuChunshuo <qiuchunshuo@huawei.com> Signed-off-by: zrj026 <zhangrunjiang026@gmail.com>
maoxx241
pushed a commit
to maoxx241/vllm-ascend
that referenced
this pull request
Mar 2, 2026
…vllm-project#5957) ### What this PR does / why we need it? In long-sequence scenarios, the chunked-prefill component may encounter dimension misalignment issues, which previously occurred during precision testing on the code_generate_lite dataset. This PR removes redundant computations and instead derives the value using existing results and straightforward calculations. - vLLM version: v0.13.0 - vLLM main: vllm-project/vllm@2c24bc6 Signed-off-by: QiuChunshuo <qiuchunshuo@huawei.com>
ZRJ026
pushed a commit
to ZRJ026/vllm-ascend
that referenced
this pull request
Mar 4, 2026
…vllm-project#5957) ### What this PR does / why we need it? In long-sequence scenarios, the chunked-prefill component may encounter dimension misalignment issues, which previously occurred during precision testing on the code_generate_lite dataset. This PR removes redundant computations and instead derives the value using existing results and straightforward calculations. - vLLM version: v0.13.0 - vLLM main: vllm-project/vllm@2c24bc6 Signed-off-by: QiuChunshuo <qiuchunshuo@huawei.com> Signed-off-by: zrj026 <zhangrunjiang026@gmail.com>
LCAIZJ
pushed a commit
to LCAIZJ/vllm-ascend
that referenced
this pull request
Mar 7, 2026
…vllm-project#5957) ### What this PR does / why we need it? In long-sequence scenarios, the chunked-prefill component may encounter dimension misalignment issues, which previously occurred during precision testing on the code_generate_lite dataset. This PR removes redundant computations and instead derives the value using existing results and straightforward calculations. - vLLM version: v0.13.0 - vLLM main: vllm-project/vllm@2c24bc6 Signed-off-by: QiuChunshuo <qiuchunshuo@huawei.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
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?
In long-sequence scenarios, the chunked-prefill component may encounter dimension misalignment issues, which previously occurred during precision testing on the code_generate_lite dataset. This PR removes redundant computations and instead derives the value using existing results and straightforward calculations.