perf-prof

Profile Linux runtime events in memory for real-time system performance analysis.

119|27|Updated Dec 21, 2021
One-click install
npx skills add https://github.com/OpenCloudOS/perf-prof --skill perf-prof
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: perf-prof
Source: https://github.com/OpenCloudOS/perf-prof/tree/main/skills/perf-prof
Command: npx skills add https://github.com/OpenCloudOS/perf-prof --skill perf-prof

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

perf-prof enables engineers to diagnose Linux system problems by profiling events in memory, delivering real-time insights without data loss to disk.

Core Features & Use Cases

  • Real-time CPU profiling with profile for hotspots
  • Scheduling and process-state analyses via task-state and rundelay
  • Delay and event-correlation analyses with multi-trace
  • Memory leak detection with kmemleak
  • IO latency tracing with blktrace
  • High-level aggregations using sql/top
  • Virtualization delay analysis with kvm-exit
  • Flexible workflows that combine profiling, filtering, and output interpretation

Quick Start

Run perf-prof profile to collect CPU hotspots, then extend to multi-trace or kmemleak for deeper analysis.

Frequently Asked Questions about perf-prof

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I perform real-time Linux system profiling without losing data to disk?

Real-time Linux system profiling captures runtime events directly in memory to provide immediate performance insights without disk I/O bottlenecks. This approach prevents data loss during high-throughput tracing and enables instantaneous analysis.

What is the best way to trace CPU hotspots and scheduling delays on a Linux kernel?

Tracing CPU hotspots and scheduling delays involves using dedicated profilers to sample runtime events across kernel and user space. These tools identify execution bottlenecks and process-state delays to pinpoint exact performance degradation sources.

How do I detect memory leaks and IO latency in a virtualization environment?

Detecting memory leaks and IO latency requires applying specific profilers like kmemleak for memory tracking and blktrace for IO operations. These tools isolate virtualization overhead and block device delays within kernel and user space.

Can I analyze virtualization overhead and event correlations using multi-trace?

Multi-trace enables delay and event-correlation analyses across virtualized environments. It captures kvm-exit events and combines them with scheduling traces to isolate virtualization overhead from underlying kernel delays.

How do I select and filter profiling events for high-level aggregations?

Selecting and filtering profiling events requires configuring specific profilers to target kernel and user space operations. High-level aggregations are then generated using sql/top workflows to summarize the filtered runtime data.