remove unused code

[zhoutianzi666]

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)

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!

[codecov-commenter]

Codecov Report

✅ All modified and coverable lines are covered by tests.
⚠️ Please upload report for BASE (develop@791d77f). Learn more about missing BASE report.

Additional details and impacted files

@@            Coverage Diff             @@
##             develop    #7961   +/-   ##
==========================================
  Coverage           ?   67.63%           
==========================================
  Files              ?      467           
  Lines              ?    65191           
  Branches           ?    10008           
==========================================
  Hits               ?    44094           
  Misses             ?    18272           
  Partials           ?     2825           

Flag	Coverage Δ
GPU	77.89% <ø> (?)
XPU	7.08% <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.

[PaddlePaddle-bot]

🤖 Paddle-CI-Agent | ci_status_monitor | 2026-05-31 18:22:59

CI报告基于以下代码生成（30分钟更新一次）:

• PR commit: a521d9c
• Merge base: 791d77f (branch: develop)
• 查看完整 Diff
• CI 详情

---

1 任务总览

存在 1 个 Required 失败任务，需处理后方可合并。

总执行（rerun次数）	总任务	✅ 通过	❌ 失败	⏳ 运行中	⏸️ 等待中	跳过
42(0)	42	37	5	0	0	0

2 任务状态汇总

2.1 Required任务 : 9/10 通过

必选任务阻塞合并，失败需优先处理。

状态	任务	耗时	根因	修复建议	日志	重跑
❌	Run FastDeploy Unit Tests and Coverage / run_tests_with_coverage	1h31m	不稳定问题：IPC模式KV cache传输挂起，请求全部超时60s	与PR代码无关，请 rerun	Job	-
✅	其余 9 个必选任务通过	-	-	-	-	-

2.2 可选任务 — 28/32 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
❌	Run iluvatar Tests / run_iluvatar_cases	2m27s	Job	-
❌	Check PR Template	26s	Job	-
❌	CI_HPU	1h4m	Job	-
❌	Trigger Jenkins for PR	17s	Job	-
✅	其余 28 个可选任务通过	-	-	-

---

3 失败详情（仅 required）

Run FastDeploy Unit Tests and Coverage / run_tests_with_coverage — 测试失败（置信度: 中）

Run FastDeploy Unit Tests and Coverage / run_tests_with_coverage

• 状态: ❌ 失败
• 错误类型: 测试失败
• 置信度: 中
• 根因摘要: IPC模式KV cache传输挂起，请求全部超时60s
• 分析器: ci_analyze_unittest_fastdeploy

失败用例:

测试	错误	根因
e2e/test_ernie_03b_pd_router_v1_ipc.py::test_non_chat_usage_non_stream	AttributeError: 'NoneType' object has no attribute 'json'	send_request() 返回 None（HTTP请求60s超时）

根因详情:
日志显示 prefill worker 成功将任务通过 zmq 发送至 decode（splitwise_connector.log 记录 cache_sync 确认，dest_block_ids 已分配），但 IPC 模式下 KV cache 实际数据传输挂起，导致 decode 无法完成推理，每个请求均在 60 秒后被中止（Receive abort request）。此后 multiprocessing Manager 队列操作抛出 BrokenPipeError。

关键日志:

# splitwise_connector.log（prefill）
17:17:24 send_splitwise_tasks: protocol=rdma, addr=10.63.64.228:8639, task=chatcmpl-ccdec929
17:17:24 _process_message: cache_sync task: {'dest_block_ids': [0, 1, 2]}
# ↑ decode已确认，但60秒后请求被abort

# fastdeploy.log（decode）
17:18:24 Receive abort request, req_id: chatcmpl-ccdec929::n::0
17:21:27 ERROR common_engine.py BrokenPipeError: [Errno 32] Broken pipe

修复建议:

1. 本次失败与 PR 代码变更无关（PR 仅删除 CUDA 注意力 kernel 中未使用的模板参数 IS_SYSTEM/partition_kv/num_frags_y，不影响 IPC 传输逻辑），建议直接 rerun 该 Job
2. 若反复失败，排查 CI runner 的 /dev/shm IPC 共享内存可用空间及 PD 分离 splitwise 传输稳定性

修复建议摘要: 与PR无关，请 rerun；如反复失败查IPC共享内存

关联变更: PR 仅修改 custom_ops/gpu_ops/append_attn/*.cuh（CUDA kernel 模板参数），未涉及 IPC 传输代码
链接: 查看日志

[PaddlePaddle-bot]

🤖 Paddle-CI-Agent | pr_review | 2026-05-31 16:29:31

📋 Review 摘要

PR 概述：移除 mask_s 函数中未使用的模板参数 partition_kv、num_frags_y 和 IS_SYSTEM，消除对应的死代码分支。
变更范围：custom_ops/gpu_ops/append_attn/
影响面 Tag：[OP]

问题

未发现阻塞性问题。

📝 PR 规范检查

PR 标题和描述均不合规：标题缺少 Tag 且无实际描述；描述各段落为空。

标题建议（可直接复制）：

• [OP] Remove unused IS_SYSTEM template parameter from mask_s

PR 描述建议（点击展开，可直接复制）

## Motivation

Remove dead code in append attention kernel. The `IS_SYSTEM` template parameter in `mask_s` function was never instantiated with `true` by any caller, making the corresponding branch unreachable.

## Modifications

- Removed `IS_SYSTEM` template parameter (defaulted to `false`) from `mask_s` in `append_attention_func.cuh`
- Eliminated the `if constexpr (IS_SYSTEM)` dead code branch, keeping only the active `!IS_SYSTEM` logic
- Cleaned up redundant parentheses in `mask_check_iteration` calculation in `multiquery_attention_c16_impl.cuh`

## Usage or Command

N/A

## Accuracy Tests

N/A (pure refactoring, no behavioral change)

## 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.
- [x] 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.

总体评价

本 PR 是纯重构变更，移除了从未被实例化为 true 的 IS_SYSTEM 模板分支和两个未使用的模板参数（partition_kv、num_frags_y），逻辑等价且安全。建议补充 PR 标题 Tag 和描述以符合项目规范。