tilelang-npu-repair-guide
Heuristic fixes for TileLang Ascend compile/JIT/kernel errors and numerical precision mismatches when editing or converting the operator. Does not replace npu-gen-test, npu-gen-bench, or harness specs.
- License 许可
- Apache-2.0
- 所属仓库
- Piagari/helix
- 最近更新
- 2026-07-16
- Stars
- ★ 0
- Forks
- 0
01 前言 Front Matter
- name
- tilelang-npu-repair-guide
- description
- Heuristic fixes for TileLang Ascend compile/JIT/kernel errors and numerical precision mismatches when editing or converting the operator. Does not replace npu-gen-test, npu-gen-bench, or harness specs.
02 资源文件 2
- 参考文档output.md274 B
- 参考文档references/repair-experience.md9.1 KB
03 SKILL.md 正文
TileLang Repair Experience (Ascend)
Use this skill when repairing the operator after TileLang/Ascend compilation,
JIT, kernel-side failures, or numerical mismatches vs the torch baseline
— especially during npu-gen-eval-suite, convert, optimize, or any flow
that exercises the real TileLang path.
Content (to be populated)
references/repair-experience.md— TileLang-specific compile/JIT/kernel error heuristics.output.md— append-only log of new fixes discovered.
Relationship to other skills
- Does not replace
npu-gen-test,npu-gen-bench, or normative harness specs. - Complements
npu-run-eval(re-validate after applying a heuristic).
How to apply
- Open
references/repair-experience.mdand match error text or symptom. - Apply the smallest change; re-run validation through
npu-run-eval. - If nothing fits, fall back to logs, IR skills, or deeper debugging.
04 同仓库更多 Skills
triton-npu-repair-guide
Heuristic fixes for Ascend Triton compile/JIT/kernel errors and numerical precision mismatches when editing or converting the operator. Does not replace ascend-npu-gen-test, ascend-npu-gen-bench, or harness specs.
triton-npu-optimize
Iteratively optimize a Triton Ascend NPU operator with correctness and performance gates. Use for operator optimization tasks that need repeated correctness validation, benchmark validation, multi-round experiment tracking, reusable optimization notes, and profiler-backed performance analysis when benchmark results need deeper explanation.
triton-npu-optimize-knowledge
Generic reference-only optimize knowledge for pattern triage and evidence-backed symptom routing. This skill does not define optimize workflow or own round artifacts.
triton-npu-optimize-knowledge-v3
Generic reference-only optimize knowledge (v3 working copy) for pattern triage and evidence-backed symptom routing. This skill does not define optimize workflow or own round artifacts; prefer this tree for ongoing knowledge updates.
triton-npu-optimize-knowledge
Generic reference-only optimize knowledge for pattern triage and evidence-backed symptom routing. This skill does not define optimize workflow or own round artifacts.
triton-npu-convert-pytorch-operator
Convert one PyTorch operator into a Triton NPU-backed PyTorch operator, preserve the trailing input-helper block, and validate the converted output through standalone or differential testing.