dshbase

插件目录 / Developer / dsh-plugin-amdgpu-inspect

dsh-plugin-amdgpu-inspect

已验证 · 实测可装 yujiaoliang

✓ 持续维护 基于 2 个官方 DSH 包

查看 GitHub ↗ ← 返回插件目录

0Stars
0Forks
0未关闭 issue
—语言
2026-08-15最近推送
跨平台平台

功能简介

Structured AMDGPU code-object and ISA inspection for DeepSeek Harness.

✅
我们的评价
可用 — 实测通过,早期项目

Structured AMDGPU code-object and ISA inspection for DeepSeek Harness. 实测能干净安装、正常启动。早期项目,但功能可用。

「已验证」表示我们的自动化 CI 在干净 profile 里实际执行了 dsh plugin add 并启动成功——仅此而已。功能描述与版本兼容性均为作者声明。这不是安全审计,也不代表对第三方代码的背书。

README

AMDGPU Inspect

Structured AMDGPU code-object and ISA inspection for DeepSeek Harness.

The plugin turns an existing HSACO or AMDGPU ELF file into bounded JSON that an agent can query. It reports compiler artifacts without compiling source, executing a kernel, predicting performance, or prescribing an optimization.

AMDGPU ELF -> llvm-readelf + llvm-objdump -> normalized object model -> DSH tools

Tools

amdgpu_object_inspect

List kernels and report public code-object metadata plus instruction-category counts:

{ "path": "/tmp/vector-add.hsaco" }

amdgpu_isa_query

Query one kernel's normalized instruction stream without placing the complete disassembly in model context:

{
  "path": "/tmp/vector-add.hsaco",
  "kernel": "vector_add",
  "opcodes": ["scratch_*", "s_waitcnt"],
  "cursor": 0,
  "limit": 40
}

The result contains addresses, opcodes, operands, categories, and a nextCursor when more matches remain.

Install

Requirements:

  • Linux
  • DeepSeek Harness developer preview
  • ROCm LLVM tools in /opt/rocm/llvm/bin or on PATH

Install the bundle from GitHub:

dsh plugin --profile web add github:yujiaoliang/dsh-plugin-amdgpu-inspect
dsh --profile web

The package ships plain ESM JavaScript and needs no install-time build script.

Kernel X-Ray Skill

skills/kernel-xray/SKILL.md is the opinionated layer: it tells an agent how to compile before/after artifacts, query this plugin, form hypotheses, and require a workload measurement before claiming a speedup. The plugin itself remains a low-level facts API.

Design

  • llvm-readelf --notes supplies public AMDGPU metadata.
  • llvm-objdump --disassemble --demangle supplies ISA.
  • A one-artifact cache is keyed by absolute path, size, and modification time.
  • ISA queries are capped at 200 records and support deterministic cursors.
  • Child processes run without a shell and honor tool-call cancellation.
  • Inputs are limited to regular files no larger than 256 MiB.

Security boundary

This version invokes LLVM and reads the artifact directly on the DSH host. It does not inherit a deployment's virtual filesystem or subprocess provider, so treat access as host-local rather than sandboxed. It never loads the artifact into an HSA runtime or executes GPU code. LLVM diagnostics can expose local paths.

Limitations

  • Parsing follows the text emitted by current ROCm LLVM tools; fixture coverage cannot guarantee every historical or future output spelling.
  • Kernel arguments, symbols, relocations, control-flow graphs, and binary encodings are not exposed yet.
  • The plugin reports static facts only. Resource counts and instruction counts do not establish performance.

Development

Tests use synthetic public-format metadata and disassembly; no GPU or ROCm installation is required:

npm test

License

MIT. ROCm and LLVM are external tools and are not redistributed.

安装

🧩 让 Agent 自动装(推荐)

装一次目录插件,之后本站所有插件都能让 DeepSeek Harness 自动找、自动装:

dsh plugin add dshbase-catalog

然后对 agent 说「帮我装 dsh-plugin-amdgpu-inspect」,它会在目录里找到并自动安装。文档:dshbase-catalog · 已验证场景包。

该插件是 GitHub 源码(未发 npm)——直接从仓库装:

Web profile:

dsh plugin --profile web add github:yujiaoliang/dsh-plugin-amdgpu-inspect

Headless(CLI)profile:

dsh plugin --profile headless add github:yujiaoliang/dsh-plugin-amdgpu-inspect

实测报告

验证通过:从 GitHub 源码完成 L1 安装 + L2 加载 + L3 运行(dsh 0.1.0-rc.6)。

安全:尚未扫描——我们的每日静态扫描将很快覆盖它。

分享徽章

Developer 里更多

浏览全部 7819 个插件 →