Merged
fix(multi_add_rms_norm_dynamic_quant): insert graph example TensorList at input zero #13852
fix(multi_add_rms_norm_dynamic_quant): insert graph example TensorList at input zero #13852
Merged
zsq_888Add on 16 days ago
zsq_888
zsq_888Collaborators
16 days ago

描述

修复 MultiAddRmsNormDynamicQuant graph example 执行失败的问题。

算子原型将动态输入 x1 定义在第 0 个输入端口。原 example 使用
create_dynamic_input_x1 创建 TensorList,动态端口被追加到静态输入之后,
导致组图时的输入端口顺序与原型不一致,进而出现输出 shape 推导错误和
设备地址异常。

将创建方式改为 create_dynamic_input_byindex_x1(x1Num, 0),显式在第 0
个位置创建动态输入。同时增加 example 的输出检查,校验六个输出的
shape、dtype 和数值

关联的Issue

关联 Issue #8331:https://gitcode.com/cann/ops-nn/issues/8331

测试

在 Ascend950 环境下,使用原算子包运行 graph example。

  • x1 TensorList 长度为 1、2、5 时,图执行均成功。
  • 六个输出 y1、y2、x、y、scale1、scale2 的 shape、
    dtype 和数值均通过 example 内的确定性参考检查。

文档更新

类型标签

AI/Agent生成声明

likedislike
Pull Request has been successfully merged by CANN-robot.
(Thank you for zsq_888's contribution.)
zsq_888zsq_888Collaborators
16 days ago Created pull request, commit 495ca1c9
atomgit-bot
atomgit-bot
16 days ago Comments:

变更摘要

本 PR 修复了 multi_add_rms_norm_dynamic_quant 算子 graph example 执行失败的问题。算子的原型将动态输入 x1 定义在第 0 个输入端口,但原 example 通过 create_dynamic_input_x1 创建 TensorList 时,动态端口会被追加到静态输入端口之后,导致组图时输入端口顺序与算子原型不一致,进而引发输出 shape 推导错误和设备地址异常。改动将动态输入的创建方式改为显式指定索引的 create_dynamic_input_byindex_x1(x1Num, 0),使 x1 TensorList 位于第 0 个输入端口;并在 example 中新增 CheckExampleOutputs 校验函数,对六个输出的 shape、dtype 和数值做确定性比对。

主要改动

  • 动态输入端口位置修正:将 op1.create_dynamic_input_x1(x1Num) 替换为 op1.create_dynamic_input_byindex_x1(x1Num, 0),使 x1 TensorList 在第 0 个位置插入,与算子原型的输入端口顺序一致。
  • 新增输出校验函数:新增静态函数 CheckExampleOutputs,依据 g_x1Num 计算确定性参考值,逐个校验六个输出的 shape、dtype、数据大小及数值。
  • 接入校验流程:在 main 中图执行成功后调用 CheckExampleOutputs,校验失败时输出提示、释放 session 并 GEFinalize() 后返回 FAILED。
  • 新增头文件依赖:补充 <cmath> 与 <cstring>,用于有限性与数值比对以及按字节读取输出数据。

🧩 本次 Review 已使用自定义 Skill:

likedislike
Incorrect?
atomgit-bot
atomgit-bot
16 days ago Comments:

代码审查

✅ 未发现问题


🧩 本次 Review 已使用自定义 Skill:

likedislike
Incorrect?
CANN-robotCANN-robotCollaborators
16 days ago Added label: cann-cla/yes
CANN-robot
CANN-robotCollaborators
16 days ago Comments:

Thanks for your pull-request.
The full list of commands accepted by me can be found at here.
You can get sig-info at here.
You can self-configure the PR merge rules for this repository. For more details, please refer to Here.
For more, you also can visit HICANN.


PR Approval Progress

✅ Congratulations! All modules have met the lgtm and approve requirements.

Module Approval Details

module lgtm status approve status
norm ✅ 汤平川, 上官秦南 (2/2) ✅ 汤平川 (1/1)

💡 Tip:

  • Committer can comment /approve or /lgtm
  • Commenting /approve implies both code review (lgtm) and intent to merge (approve)

CLA Signature Pass

qq_57796320, thanks for your pull request. All authors of the commits have signed the CLA. 👍

likedislike
CANN-robotCANN-robotCollaborators
16 days ago Set zengjuan,u010470851,chaotang233,kevin_huang1234,crystalhu,yangyang016,fanqirui,wangrui_,renruhai,su-yueming,zhajianqing123,yourealize,chenqi317,wang-xing001,liubo75,tangweiwei2,chenxingyu18,liujie12345678,east_yang,pingchuantang,Chen_HaoWen,qianzehong as approver
CANN-robotCANN-robotCollaborators
16 days ago Set u010470851,kevin_huang1234,wangrui_,renruhai,su-yueming,zhajianqing123,yourealize,wang-xing001,chenxingyu18,liujie12345678,east_yang,pingchuantang,Chen_HaoWen,qianzehong as approver
zsq_888zsq_888Collaborators
16 days ago Associated issue: [Bug-Report|缺陷反馈]: MultiAddRmsNormDynamicQuant图模式示例动态输入端口顺序错误
zsq_888zsq_888Collaborators
16 days ago Modify description of pull request
zsq_888
zsq_888Collaborators
16 days ago Comments:

/compile

likedislike
zsq_888zsq_888Collaborators
16 days ago Pre merge succeeded (commit_id: dbf4ae332bf60ccd27e5ddcc4859cbd9aaf77be0)
CANN-robot
CANN-robotCollaborators
16 days ago Comments:
🚀 CI 流水线已启动
📋 执行详情: 点击查看流水线
likedislike
CANN-robotCANN-robotCollaborators
16 days ago Added label: ci-pipeline-running
CANN-robotCANN-robotCollaborators
16 days ago Removed label: ci-pipeline-running
CANN-robotCANN-robotCollaborators
16 days ago Added label: ci-pipeline-failed
CANN-robot
CANN-robotCollaborators
16 days ago Comments:

流水线分析报告 — ops-nn run#17861

流水线: https://gitcode.com/CANN/ops-nn/actions/runs/b609a449b11d476db1a4c109606f049d | PR#13852 | 2026-11-06 16:50:27 ~ 16:54:29 | 失败 Job 1 个

定界结论: 代码问题 / 代码检查错误 — clang-format 格式不符
  • 失败类型: 代码问题(提交者修复)
  • 根因: CodeCheck_precommit 中 clang-format hook 检测到 1 个 C++ 文件未按规范格式化,hook 已自动改写该文件并判定 Failed:
    norm/multi_add_rms_norm_dynamic_quant/examples/test_geir_multi_add_rms_norm_dynamic_quant.cpp
  • 处置建议: 本地执行 pre-commit run clang-format --files norm/multi_add_rms_norm_dynamic_quant/examples/test_geir_multi_add_rms_norm_dynamic_quant.cpp,将格式化结果一并提交后重新触发流水线。参考: 本地执行 pre-commit 指导
  • 排除项: 同 run 其余 hook(large files / merge conflicts / private key / codespell / OAT)均 Passed;ruff、check json 无文件跳过;无平台/环境异常,无下游级联失败。

失败根因分析

根因 #1: 代码问题 / 代码检查错误 — clang-format(1 个文件格式不合规)
  • 判据: pre-commit hook clang-format 返回 Failed(files were modified by this hook)
  • 定位步骤: CodeCheck_precommit → 3_precommit.log
  • 影响 Job: CodeCheck_precommit(唯一失败 Job,无级联)
  • 根因证据(日志原文):
clang-format.............................................................Failed
- hook id: clang-format
- files were modified by this hook

Formatting [1/1] norm/multi_add_rms_norm_dynamic_quant/examples/test_geir_multi_add_rms_norm_dynamic_quant.cpp
  • 判定依据: 属确定性失败——重跑流水线不会自动消失,必须先把 clang-format 的格式化改动提交到 PR。
结构化定界摘要
[{"repo": "ops-nn", "run_number": 17861, "Failed_Type": "代码问题",
  "Failed_Sub_Type": "代码检查错误",
  "Failed_Reason": "CodeCheck_precommit 的 clang-format hook 检测到 norm/multi_add_rms_norm_dynamic_quant/examples/test_geir_multi_add_rms_norm_dynamic_quant.cpp 格式不合规,hook 已改写该文件并判定失败",
  "evidence": "clang-format.............................................................Failed / - files were modified by this hook / Formatting [1/1] norm/multi_add_rms_norm_dynamic_quant/examples/test_geir_multi_add_rms_norm_dynamic_quant.cpp",
  "suggestion": "本地执行 pre-commit run clang-format 修复该文件并提交格式化结果后重新触发流水线",
  "jobs": "CodeCheck_precommit"}]

由门禁失败定界分析自动生成 | repo: ops-nn | run: b609a449b11d476db1a4c109606f049d | PR#13852

likedislike
zsq_888zsq_888Collaborators
16 days ago Pre merge succeeded (commit_id: 99c13e19ee4bf34c9ec8d2ce015bc12de9e0a4d2)
zsq_888zsq_888Collaborators
16 days ago Force-pushed 1 commits:d4ea66e0-fix(multi_add_rms_norm_dynamic_quant): insert graph example TensorList at input zero
zsq_888zsq_888Collaborators
16 days ago Pre merge succeeded (commit_id: 436336e20eabba58fa5cb6bf6b80d693519c8585)
CANN-robotCANN-robotCollaborators
16 days ago Removed label: cann-cla/yes
CANN-robotCANN-robotCollaborators
16 days ago Added label: cann-cla/yes
zsq_888
zsq_888Collaborators
16 days ago Comments:

/compile

likedislike
zsq_888zsq_888Collaborators
16 days ago Pre merge succeeded (commit_id: e980bef0540aa5debcca012b24656c59f6859736)
CANN-robot
CANN-robotCollaborators
16 days ago Comments:
🚀 CI 流水线已启动
📋 执行详情: 点击查看流水线
likedislike
CANN-robotCANN-robotCollaborators
16 days ago Removed label: ci-pipeline-failed
CANN-robotCANN-robotCollaborators
16 days ago Added label: ci-pipeline-running
CANN-robotCANN-robotCollaborators
16 days ago Removed label: ci-pipeline-running
CANN-robotCANN-robotCollaborators
16 days ago Added label: ci-pipeline-passed
TangPC
TangPCCollaborators
16 days ago Comments:

/lgtm
/approve

likedislike
CANN-robotCANN-robotCollaborators
16 days ago Added label: approved
sgqt12
sgqt12Collaborators
16 days ago Comments:

/lgtm

likedislike
CANN-robotCANN-robotCollaborators
16 days ago Added label: lgtm
CANN-robotCANN-robotCollaborators
16 days ago Closed associated issue
CANN-robotCANN-robotCollaborators
16 days ago Merged pull request