[Cherry-Pick][OP][RL]update attn_mask_q(#7371)

[ckl117]

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!

[PaddlePaddle-bot]

🤖 AI Code Review | 2026-04-13 21:32 CST

📋 Review 摘要

PR 概述：优化 get_attn_mask_q 算子，将输出张量维度从 4 减少到 2，节省内存占用

变更范围：custom_ops/gpu_ops/get_attn_mask_q.cu

影响面 Tag：[OP]

📝 PR 规范检查

PR 标题格式正确，但描述部分未填写完整，建议补充：

Motivation（示例）：

优化内存使用：原实现中 startend_row_vec 数组有 2 个元素未被实际使用，将数组大小从 4 减少到 2，可降低约 50% 的输出张量内存占用。

Modifications（示例）：

• 将 startend_row_vec 数组大小从 4 改为 2
• 移除未使用的数组元素 [1] 和 [2] 的赋值
• 将输出张量 shape 从 {1, 1, kv_token_num, 4} 改为 {1, 1, kv_token_num, 2}
• 将 int4* 类型转换改为 int2* 以匹配新的数组大小

问题

级别	文件	概述
🟡 建议	custom_ops/gpu_ops/get_attn_mask_q.cu:25	VecSize 常量定义但未使用
🟡 建议	custom_ops/gpu_ops/get_attn_mask_q.cu:52	遗留被注释的代码

总体评价

变更逻辑正确，是一个合理的内存优化。测试覆盖充分（FA2/FA3/FA4 端到端测试）。代码中遗留了一些无用的常量和注释代码，建议一并清理。

[PaddlePaddle-bot]

🟡 建议 VecSize 常量在第 25 行定义但从未使用，建议移除以保持代码整洁。

[PaddlePaddle-bot]

🟡 建议 被注释的代码（第 52-53 行）可考虑移除，因为它们反映了之前的实现且已不再使用。

[codecov-commenter]

Codecov Report

✅ All modified and coverable lines are covered by tests.
⚠️ Please upload report for BASE (release/2.6@9823d63). Learn more about missing BASE report.

Additional details and impacted files

@@              Coverage Diff               @@
##             release/2.6    #7373   +/-   ##
==============================================
  Coverage               ?   73.72%           
==============================================
  Files                  ?      376           
  Lines                  ?    52968           
  Branches               ?     8271           
==============================================
  Hits                   ?    39052           
  Misses                 ?    11184           
  Partials               ?     2732           

Flag	Coverage Δ
GPU	73.72% <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.