triton-operator-performance-eval
OfficialDiagnose Triton kernel performance on Ascend NPUs
AuthorAscend
Version1.0.0
Installs0
System Documentation
What problem does it solve?
通过使用 msprof 数据帮助开发者识别和诊断 Ascend NPU 上 Triton 核函数的性能瓶颈，从而实现有效优化。
Core Features & Use Cases
- 基于 msprof/ msprof op 的函数级与算子级性能采集，提供全面的硬件利用情况分析。
- 瓶颈诊断与优化建议，将 Memory-Bound/Compute-Bound 等分类输出，辅助定位改进方向。
- 适用场景广泛，可用于不同输入形状、精度和 Triton 内核实现的性能对比与调优。
Quick Start
Run msprof on your Triton kernel to collect function-level and operator-level performance data for Ascend NPUs.
Dependency Matrix
Required Modules
None requiredComponents
references
💻 Claude Code Installation
Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.
Please help me install this Skill: Name: triton-operator-performance-eval Download link: https://github.com/Ascend/agent-skills/archive/main.zip#triton-operator-performance-eval Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
Agent Skills Search Helper
Install a tiny helper to your Agent, search and equip skill from 471,000+ vetted skills library on demand.