skills.rest
ProgMastermind
ProgMastermindCOMMUNITY
Explore all skills by this author →
Website𝕏@PrakashCollymo1GitHub

debug-agent-locate-kernel

Diagnoses faulty GPU kernels and silent hangs on ATOM platforms using rocm-debug-agent and rocgdb.

Updated Jul 15, 2026
Search Tags:#atom#gpu debugging#kernel diagnosis#rocm-debug-agent#rocgdb
One-click install
npx skills add https://github.com/ProgMastermind/ATOM --skill debug-agent-locate-kernel
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: debug-agent-locate-kernel
Source: https://github.com/ProgMastermind/ATOM/tree/main/.claude/skills/debug-agent-locate-kernel
Command: npx skills add https://github.com/ProgMastermind/ATOM --skill debug-agent-locate-kernel

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires rocm-debug-agent, rocgdb, and includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill helps diagnose and locate the cause of GPU kernel faults and hangs in the ATOM platform by utilizing rocm-debug-agent or rocgdb.

Core Features & Use Cases

  • Kernel Fault Identification: Locates and identifies faulty kernels causing issues in the ATOM environment.
  • Kernel Hang Diagnosis: Determines the cause of silent kernel hangs that are not causing fault logs.
  • Debug Agent vs. rocgdb Selection: Assists in choosing the appropriate tool based on symptoms and kernel behavior.
  • Use Case: If your ATOM server crashes with a memory access fault or hangs with 100% GPU utilization without output, this Skill can help diagnose and debug the issue.

Quick Start

Use the debug-agent-locate-kernel skill to identify which GPU kernel is causing issues in your ATOM environment.

Frequently Asked Questions about debug-agent-locate-kernel

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

FAQPage Schema
How do I diagnose GPU kernel faults and hangs in the ATOM platform?▼

Diagnose GPU kernel faults and hangs in the ATOM platform by utilizing rocm-debug-agent for memory access faults or rocgdb for silent livelocks to locate faulty kernels.

What is the best way to debug a silent GPU kernel livelock with 100% utilization?▼

Debug a silent GPU kernel livelock with 100% utilization by using rocgdb to diagnose the hang when no fault logs are generated in the ATOM environment.

When should I use rocm-debug-agent versus rocgdb for kernel diagnosis?▼

Choose rocm-debug-agent for diagnosing kernel faults and asserts, whereas use rocgdb when dealing with silent kernel hangs that do not produce fault logs in the ATOM system.

Do I need knowledge of ATOM architecture to locate faulty GPU kernels?▼

Locating faulty GPU kernels requires knowledge of the ATOM architecture, alongside familiarity with rocm-debug-agent and rocgdb to effectively diagnose kernel faults and hangs.

Why does my ATOM server crash with a memory access fault during GPU debugging?▼

An ATOM server crashes with a memory access fault due to faulty GPU kernels, which can be identified and located using rocm-debug-agent to diagnose the issue.

Can I use rocgdb to identify GPU kernel issues on ATOM systems?▼

You can use rocgdb to identify GPU kernel issues on ATOM systems, specifically for diagnosing silent kernel hangs and livelocks that lack explicit fault logs.

Related Skills

Explore More Software Engineering Skills→
fsx950223fsx950223
2

kernel-trace-analysis

Profile GPU kernels with rocprofv3 and generate optimization plans.

Community
Advanced
trymiraitrymirai
1.7k

metal-gpu-debug

Automate Metal GPU trace collection, parsing, and analysis on macOS.

Official
Advanced
forecast-bioforecast-bio
27

rust-gpu-discipline

Verify Rust GPU compute paths execute on GPU without silent CPU fallbacks.

Official
Advanced
Binh230199Binh230199
1

linux-debugging

Diagnose embedded Linux kernel crashes, hangs, and driver faults on automotive targets.

Community
Advanced