Pull Request已成功合入, 合并人@CANN-robot
（感谢 taochangmin 的贡献）
变更摘要
本 PR 修复了 SwigluGroup 与 SwigluGroupQuant 算子在输出全为空 Tensor 场景下校验失效的问题。原因是框架对输出全为空 Tensor 的场景不会下发 kernel 执行，导致原本依赖 kernel 执行的空 Tensor 校验无法生效；通过在这两个算子的定义中调用 AICore().LaunchWithZeroEleOutputTensors(true)，使该场景下仍会下发 kernel 执行，从而恢复空 Tensor 校验能力。
主要改动
swiglu_group_def.cpp开启零元素输出下发：在swiglu_group算子的定义中新增this->AICore().LaunchWithZeroEleOutputTensors(true);，使输出全为空 Tensor 时仍下发 kernel 执行，恢复空 Tensor 校验。swiglu_group_quant_def.cpp开启零元素输出下发：在swiglu_group_quant算子的定义中新增this->AICore().LaunchWithZeroEleOutputTensors(true);，同样解决输出全为空 Tensor 时校验失效的问题。
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 |
|---|---|---|
| */*/op_host/*_def.cpp | ✅ 汤磊, 钱泽洪 (2/2) | ✅ 汤磊 (1/1) |
| activation | ✅ 钱泽洪, 陈昊文 (2/2) | ✅ 钱泽洪, 陈昊文 (2/1) |
💡 Tip:
- Committer can comment
/approveor/lgtm- Commenting
/approveimplies both code review (lgtm) and intent to merge (approve)
CLA Signature Pass
taochangmin, thanks for your pull request. All authors of the commits have signed the CLA. 👍
CANN-robot/lgtm
/compile
流水线任务触发成功
任务链接 [081a441aec6c49f6bd4b50439d1594f8][流水线指导]
| 任务名称 | 状态 | 日志 | 下载链接 |
|---|---|---|---|
| Compile_Ascend_X86_ubuntu24 | ✅ SUCCESS | >>>>> | >>>>> |
| Compile_Ascend_X86_mobile_station | ✅ SUCCESS | >>>>> | >>>>> |
| Compile_Ascend_ARM | ✅ SUCCESS | >>>>> | >>>>> |
| Compile_Ascend_single | ✅ SUCCESS | >>>>> | >>>>> |
| Compile_Ascend_experimental | ✅ SUCCESS | >>>>> | >>>>> |
| Compile_Ascend_X86_950 | ✅ SUCCESS | >>>>> | >>>>> |
| Compile_Ascend_ARM_950 | ✅ SUCCESS | >>>>> | >>>>> |
| Compile_Pre | ✅ SUCCESS | >>>>> | |
| pre_comment | ✅ SUCCESS | >>>>> | |
| Compile_Ascend_X86 | ✅ SUCCESS | >>>>> | >>>>> |
| Compile_Ascend_X86_mobile_station_ubuntu24 | ✅ SUCCESS | >>>>> | >>>>> |
| Compile_Ascend_X86_950_ubuntu24 | ✅ SUCCESS | >>>>> | >>>>> |
| Compile_Ascend_ARM_ubuntu24 | ✅ SUCCESS | >>>>> | >>>>> |
| Compile_Ascend_single_ubuntu24 | ✅ SUCCESS | >>>>> | >>>>> |
| Compile_Ascend_experimental_ubuntu24 | ✅ SUCCESS | >>>>> | >>>>> |
| Compile_Ascend_ARM_950_ubuntu24 | ✅ SUCCESS | >>>>> | >>>>> |
| Compile_X86_monitor_910b | ✅ SUCCESS | >>>>> | >>>>> |
| Compile_X86_monitor_910c | ✅ SUCCESS | >>>>> | >>>>> |
| Compile_X86_monitor_950 | ✅ SUCCESS | >>>>> | >>>>> |
| Compile_classify | ✅ SUCCESS | >>>>> | |
| Compile_Ascend_X86_mobile_station_9030_ubuntu24 | ✅ SUCCESS | >>>>> | >>>>> |
| UT_Test_ophost | ✅ SUCCESS | ||
| UT_Test_opapi | ✅ SUCCESS | ||
| UT_Test_kernel | ✅ SUCCESS | ||
| UT_Test_opgraph | ✅ SUCCESS | ||
| PreSmoke_A900 | ✅ SUCCESS | >>>>> | |
| API_Check | ✅ SUCCESS | >>>>> | |
| PreSmoke_ATK_Test_A2 | ✅ SUCCESS | >>>>> | |
| UT_Test_report_lcov | ⚠️ WARNING | >>>>> |
[2026-08-31 15:35:47] CI执行结束
/compile
您好，pr合入已准备就绪，请尽快联系committer进行检视，谢谢！
/approve
/approve
/approve
/lgtm
描述
由于框架对输出全为空Tensor的场景不会下发kernel执行，导致空Tensor校验失效，开启LaunchWithZeroEleOutputTensors使该场景仍下发执行。
关联的Issue
https://gitcode.com/cann/ops-nn/issues/5363
测试
ut、st测试，冒烟测试
文档更新
类型标签
AI/Agent生成声明