变更摘要

本 PR 完成 DivCustomTemplate 自定义算子的工程化实现：在 host 侧实现 Tiling 参数计算与 Shape/DataType 推导，在 kernel 侧基于 AscendC 完成除法计算，并配套算子定义、TF 插件注册、CMake 构建脚本、一键构建/安装/运行脚本以及基于 ACLNN 接口的测试用例。算子支持 FP16/FP32 两种数据类型，按输入规模动态生成 Tiling 参数并遵循 32B 对齐，同时支持 ascend310p 与 ascend910b 两种编译目标。

主要改动

• 算子定义与注册: 新增 div_custom_template.json 定义算子输入 x/y 与输出 z（ND 格式，float16/float），并在 tensorflow_div_custom_template_plugin.cc 中通过 REGISTER_CUSTOM_OP("DivCustomTemplate") 完成 TF 框架插件注册。

• Host 侧 Tiling 与推导逻辑: op_host/div_custom_template.cpp 实现 TilingFunc（校验输入 shape/数据类型一致性，按 32B 对齐计算 blockDim/blockLength/tileLength 并写入 DivCustomTemplateTilingData）、InferShape/InferDataType，并通过 AICore().AddConfig 注册 ascend910b 与 ascend310p 配置。

• Kernel 侧算子实现: op_kernel/div_custom_template.cpp 使用 TPipe/TQue 双缓冲流水，经 DataCopy 搬入、Div 计算、DataCopy 搬出完成逐 tile 处理，并根据 tilingData.dataType 在 float 与 half 之间分发。

• Tiling 数据结构: op_kernel/div_custom_template_tiling.h 通过 BEGIN_TILING_DATA_DEF 定义 totalLength/blockLength/tileLength/dataType 四个字段并注册 DivCustomTemplateTilingData。

• 构建、运行与验证流程: 新增 CMakePresets.json/build.sh 构建配置与脚本、run.sh 一键加载 CANN 环境并完成编译、安装、环境 source 与测试执行，test/main.cpp 通过 aclnnDivCustomTemplateGetWorkspaceSize/aclnnDivCustomTemplate 验证 FP16/FP32 除法结果，并实现资源申请与异常路径的释放清理。

Hi @iss_liqinghua, welcome to submitting your first PR to cann-outreach!

PR Merge Steps

1. CLA Signing

If the current PR label includes cann-cla/yes, it means you have signed the CLA and can proceed to the next step. If the label includes cann-cla/no, please sign the CLA first. If you have any questions, please refer to the FAQ.

2. CI Check

Please comment /compile to trigger the CI pipeline check. If the CI run is successful, the PR will be tagged with ci-pipeline-passed and you can proceed to the next step. If the CI run fails, the PR will be tagged with ci-pipeline-failed, please check the CI logs to fix the issues in the PR. If you have any questions, please refer to the FAQ.

3. Code Review

After CI passes, please refer to the PR Approval Progress and proactively @ the committers in the table to review the code. After approval, committers will comment /lgtm and /approve. Once the lgtm and approved labels are successfully added, the PR will be merged automatically.

代码审查

✅ 未发现问题

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.
For more, you also can visit HICANN.

PR Approval Progress

⚠️ This PR does not yet meet the following requirements:lgtm (requires ≥ 2 person(s) per module)、approve (requires ≥ 1 person(s) per module)

Module Approval Details

💡 Tip:

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

CLA Signature Pass

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

/compile

/compile

/compile