agentleFS
D-RoboticsD-Robotics/dsh-plugin-rdk/skills/oe-skills-x5/x5-runtime-perf-eval/SKILL.md

x5-runtime-perf-eval

Skill1 starsChanged 3 months agoApache-2.0View on GitHub

在 X5 板端评测模型延时、吞吐和稳定性;当功能正确性已通过,需要使用 hrt_model_exec 或 ai_benchmark 得到可复现性能报告时使用。先读本地工具参数,不猜命令;不修改频率或停止业务进程而未确认。

The file

Files
SKILL.md

X5 Runtime 性能评测

目标与边界

将静态估计、Runtime 实测和板端资源放到同一测试矩阵。性能评测不能替代正确性验证,也不能在热降频或后台负载未知时直接下结论。

输入合同

  • 已通过功能验证的模型和输入。
  • 板卡/Runtime 版本、测试次数、warmup、batch/core/线程等矩阵。
  • 目标延时/吞吐和允许的设备影响。

前置检查

  1. 确认板卡空闲程度、温度、频率和电源模式可记录。
  2. 从 _sources/runtime/source/tool_introduction/hrt_model_exec.rst.txt 读取当前版本参数。
  3. 固定输入、后处理范围和计时边界;不同方案必须使用同一口径。

执行步骤

  1. 先 warmup,再进行有界次数的重复测量;禁止无界 while true 压测。
  2. 保存原始工具日志、每轮延时和错误计数。
  3. 同步或分阶段调用 x5-board-monitor 记录温度、频率和 BPU ratio。
  4. 报告 p50/p95/平均/最小/最大、吞吐、测试次数和异常样本。
  5. 与 hb_perf 静态估计比较,说明计时范围、I/O 和 CPU 前后处理差异。

产物与完成标准

  • 测试矩阵、原始日志、统计报告和板端资源快照。
  • 每个数字可追溯到命令、模型哈希、输入和板端版本。
  • 性能结论区分 BPU 执行、CPU 前后处理、I/O 和系统负载。

风险与确认

短时只读评测为中风险;长时间压测、频率/电源修改、停止业务进程为高风险并需确认。

失败与交接

结果波动或不达标时交接 x5-performance-diagnostics;先保持测试条件不变,不自动改编译配置。

按需参考

  • _sources/runtime/source/tool_introduction/hrt_model_exec.rst.txt
  • _sources/runtime/source/ai_benchmark/ai_benchmark.rst.txt

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.