[Cherry-Pick][BugFix] fix paddle optional get assert in sm103 (#7816)

[zoooo0820]

Motivation

💡 If this PR is a Cherry Pick, the PR title needs to follow the format by adding the [Cherry-Pick] label at the very beginning and appending the original PR ID at the end. For example, [Cherry-Pick][CI] Add check trigger and logic(#5191)

💡 如若此PR是Cherry Pick，PR标题需遵循格式，在最开始加上[Cherry-Pick]标签，以及最后面加上原PR ID，例如[Cherry-Pick][CI] Add check trigger and logic(#5191)

cherry-pick #7816 to release/2.6

Modifications

Usage or Command

Accuracy Tests

Checklist

• Add at least a tag in the PR title.
  • Tag list: [[FDConfig],[APIServer],[Engine], [Scheduler], [PD Disaggregation], [Executor], [Graph Optimization], [Speculative Decoding], [RL], [Models], [Quantization], [Loader], [OP], [KVCache], [DataProcessor], [BugFix], [Docs], [CI], [Optimization], [Feature], [Benchmark], [Others], [XPU], [HPU], [GCU], [DCU], [Iluvatar], [Metax]]
  • You can add new tags based on the PR content, but the semantics must be clear.
• Format your code, run pre-commit before commit.
• Add unit tests. Please write the reason in this PR if no unit tests.
• Provide accuracy results.
• If the current PR is submitting to the release branch, make sure the PR has been submitted to the develop branch, then cherry-pick it to the release branch with the [Cherry-Pick] PR tag.

[paddle-bot]

Thanks for your contribution!

[PaddlePaddle-bot]

🤖 Paddle-CI-Agent | pr_review | 2026-05-14 18:21:40

📋 Review 摘要

PR 概述：Cherry-pick #7816 到 release/2.6，修复在 SM103 GPU 上编译时触发的 optional get assert 错误，为 SM103 架构添加正确的 gencode 编译标志及 nvcc 版本门控。
变更范围：custom_ops/setup_ops.py
影响面 Tag：[OP]

📝 PR 规范检查

PR 标题格式符合 Cherry-Pick 规范（[Cherry-Pick][BugFix] + 原 PR 号 #7816）。描述中 ## Modifications 段落内容为空（仅留有模板注释），不符合描述模板要求。

PR 描述建议（可直接复制）：

## Motivation
cherry-pick #7816 到 release/2.6 分支，修复在 SM103 GPU 上编译时触发的 optional get assert 错误。

## Modifications
- `custom_ops/setup_ops.py`：在 `get_gencode_flags()` 中新增 `cc_val == 103` 分支，arch_code 设为 `103a`，生成 `-gencode arch=compute_103a,code=sm_103a` 编译参数
- `custom_ops/setup_ops.py`：新增 `has_sm103`（要求 nvcc >= 13.0）变量，将 SM103 从 `has_generic_fp8` 路径剔除，并纳入 Blackwell SM100/SM103 优化编译路径

## Usage or Command
N/A

## Accuracy Tests
N/A

## Checklist

- [x] Add at least a tag in the PR title.
  - Tag list: [`[FDConfig]`,`[APIServer]`,`[Engine]`, `[Scheduler]`, `[PD Disaggregation]`, `[Executor]`, `[Graph Optimization]`, `[Speculative Decoding]`, `[RL]`, `[Models]`, `[Quantization]`, `[Loader]`, `[OP]`, `[KVCache]`, `[DataProcessor]`, `[BugFix]`, `[Docs]`, `[CI]`, `[Optimization]`, `[Feature]`, `[Benchmark]`, `[Others]`, `[XPU]`, `[HPU]`, `[GCU]`, `[DCU]`, `[Iluvatar]`, `[Metax]`]
  - You can add new tags based on the PR content, but the semantics must be clear.
- [x] Format your code, run `pre-commit` before commit.
- [ ] Add unit tests. Please write the reason in this PR if no unit tests.
- [ ] Provide accuracy results.
- [x] If the current PR is submitting to the `release` branch, make sure the PR has been submitted to the `develop` branch, then cherry-pick it to the `release` branch with the `[Cherry-Pick]` PR tag.

问题

未发现阻塞性问题。

总体评价

代码变更逻辑清晰，SM103 架构支持遵循了 SM90/SM100 的既有模式，103a gencode 处理和 nvcc 版本门控实现正确。建议补全 PR 描述 Modifications 段落。

[PaddlePaddle-bot]

🤖 Paddle-CI-Agent | ci_status_monitor | 2026-05-14 18:27:57

CI报告基于以下代码生成（30分钟更新一次）:

• PR commit: 6e25cd5
• Merge base: df637af (branch: release/2.6)
• 查看完整 Diff
• CI 详情

---

1 任务总览

CI 仍在进行中，Required 任务暂无失败：4 个运行中，2 个等待中，4 个已通过。可选任务有 2 个失败（不阻塞合并），建议关注。

总执行（rerun次数）	总任务	✅ 通过	❌ 失败	⏳ 运行中	⏸️ 等待中	跳过
35(0)	35	25	2	5	3	0

---

2 任务状态汇总

2.1 Required任务 : 4/10 通过

必选任务阻塞合并，失败需优先处理。当前 CI 仍在进行中，暂无失败。

状态	任务	耗时	根因	修复建议	日志	重跑
⏳	Extracted partial CE model tasks to run in CI. / run_ce_cases	-	运行中	-	Job	-
⏳	Run Base Tests / base_tests	-	运行中	-	Job	-
⏳	xpu_4cards_case_test / run_xpu_4cards_cases	-	运行中	-	Job	-
⏳	xpu_8cards_case_test / run_xpu_8cards_cases	-	运行中	-	Job	-
⏸️	Run FastDeploy Unit Tests and Coverage / run_tests_with_coverage	-	等待中	-	-	-
⏸️	Run Four Cards Tests / run_4_cards_tests	-	等待中	-	-	-
✅	其余 4 个必选任务通过	-	-	-	-	-

2.2 可选任务 — 21/25 通过

可选任务不阻塞合并，失败仅供参考。

状态	任务	耗时	日志	重跑
❌	Check PR Template	10s	Job	-
❌	Trigger Jenkins for PR	24s	Job	-
⏳	Run iluvatar Tests / run_iluvatar_cases	-	-	-
⏸️	CI_HPU	-	-	-
✅	其余 21 个可选任务通过	-	-	-

---

3 失败详情（仅 required）

无 required 失败任务。

[codecov-commenter]

Codecov Report

✅ All modified and coverable lines are covered by tests.
⚠️ Please upload report for BASE (release/2.6@df637af). Learn more about missing BASE report.

Additional details and impacted files

@@              Coverage Diff               @@
##             release/2.6    #7820   +/-   ##
==============================================
  Coverage               ?   72.45%           
==============================================
  Files                  ?      381           
  Lines                  ?    54139           
  Branches               ?     8456           
==============================================
  Hits                   ?    39229           
  Misses                 ?    12153           
  Partials               ?     2757           

Flag	Coverage Δ
GPU	72.45% <ø> (?)

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.