intel-vtune-amd-uprof

What problem does it solve?

This Skill helps you diagnose performance bottlenecks in your code by analyzing CPU microarchitecture, identifying pipeline stalls, and optimizing memory access patterns.

Core Features & Use Cases

• Hotspot Analysis: Pinpoint CPU-bound functions consuming the most time.
• Microarchitecture Exploration: Understand IPC, pipeline stalls, and execution bottlenecks.
• Memory Access Profiling: Detect cache misses and DRAM bandwidth limitations.
• AMD uProf Integration: Provides a free alternative for AMD CPUs.
• Roofline Model: Visualize performance against hardware limits to identify compute vs. memory bounds.
• Use Case: Your application runs slower than expected despite a high cache hit rate. Use this Skill to analyze pipeline stalls and identify if the bottleneck is in the CPU's execution units or memory subsystem.

Quick Start

Use the intel-vtune-amd-uprof skill to collect and report hotspots for the executable 'my_program'.