kernel-debugging-advanced
Advanced kernel debugging skill for ftrace, trace-cmd, perf, kprobes, kgdb, and crash analysis. Use when tracing kernel functions, profiling syscalls in kernel, dynamic instrumentation, or analyzing vmcore. Activates on queries about ftrace, trace-cmd, kprobe, kgdb, kernel crash dump, or dyndbg.
Works with
---
name: kernel-debugging-advanced
description: Advanced kernel debugging skill for ftrace, trace-cmd, perf, kprobes, kgdb, and crash analysis. Use when tracing kernel functions, profiling syscalls in kernel, dynamic instrumentation, or analyzing vmcore. Activates on queries about ftrace, trace-cmd, kprobe, kgdb, kernel crash dump, or dyndbg.
license: MIT
---
# Advanced Kernel Debugging
## Purpose
Extend `skills/kernel/kernel-debugging` with production-grade tracing: `ftrace`, `trace-cmd`, kernel `perf`, `kprobes`/`kretprobes`, `kgdb`, crash dump analysis, and `printk` discipline. **Not merged** with `kernel-debugging` — that skill covers baseline kgdb/dyndbg; this one focuses on trace-cmd, function_graph, and live/post-mortem production workflows.
## When to Use
- Latency spikes in kernel driver without obvious bug
- Tracing function call graph in kernel
- Live inspection without recompiling (kprobes, bpf)
- Post-mortem `vmcore` after panic
## Workflow
### 1. ftrace function graph
```bash
# Enable function graph tracer
echo function_graph > /sys/kernel/debug/tracing/current_tracer
echo my_driver_probe > /sys/kernel/debug/tracing/set_graph_function
echo 1 > /sys/kernel/debug/tracing/tracing_on
# reproduce issue
cat /sys/kernel/debug/tracing/trace
echo 0 > /sys/kernel/debug/tracing/tracing_on
```
Requires `CONFIG_FUNCTION_GRAPH_TRACER` and debugfs mounted.
### 2. trace-cmd record
```bash
trace-cmd record -p function_graph -F my_probe
trace-cmd report
trace-cmd stat
```
Portable capture for sharing with others.
### 3. perf in kernel context
```bash
perf record -a -g -- sleep 10
perf report --stdio
perf probe --add my_driver:probe
```
See `skills/profilers/linux-perf` for userspace overlap.
### 4. kprobes (dynamic)
```c
#include <linux/kprobes.h>
static struct kprobe kp = {
.symbol_name = "do_sys_open",
.pre_handler = handler,
};
register_kprobe(&kp);
```
Use sparingly in production; prefer static tracepoints when available.
### 5. dyndbg and printk discipline
```bash
echo 'module mydriver +p' > /sys/kernel/debug/dynamic_debug/control
echo 'file drivers/foo/*.c +p' > /sys/kernel/debug/dynamic_debug/control
```
```c
pr_debug("state=%d\n", s); /* compile-time optional */
printk_ratelimited(KERN_WARNING "hw fault\n");
```
### 6. kgdb / kdb
```bash
# kernel cmdline: kgdboc=ttyS0,115200 kgdbwait
echo g > /proc/sysrq-trigger # break into debugger (when configured)
```
### 7. Crash dump (vmcore)
```bash
# after kdump capture
crash /usr/lib/debug/vmlinux vmcore
crash> bt
crash> dev -s
```
### 8. Agent usage
```
/kernel-debugging-advanced Trace my_driver_ioctl latency with ftrace function_graph
```
## Common Problems
| Symptom | Cause | Fix |
|---------|-------|-----|
| Empty trace | Tracer not enabled | `tracing_on`; check `current_tracer` |
| ftrace overhead | Graph all functions | Filter with `set_graph_function` |
| kprobe fail | Inlined symbol | Try nearby symbol or static key |
| kgdb no connect | Wrong `kgdboc` tty | Match serial/USB gadget |
| crash no symbols | Missing debuginfo | Install `linux-image-dbg` / vmlinux |
## Related Skills
- `skills/kernel/kernel-debugging` — baseline kgdb, dyndbg
- `skills/profilers/linux-perf` — perf record/report
- `skills/observability/ebpf` — BPF kprobe alternative
- `skills/kernel-dev/writing-char-drivers` — ioctl trace targets
- `skills/debuggers/core-dumps` — userspace core parallelsMore Debugging skills
diagnosing-bugs
mattpocock/skills
Diagnosis loop for hard bugs and performance regressions. Use when the user says "diagnose"/"debug this", or reports something broken/throwing/failing/slow.
explore-code
lllllllama/rigorpilot-skills
Rigor Improve implementation leaf skill for auditable candidate implementation in deep learning research repositories. Use when the researcher explicitly authorizes exploratory work on an isolated branch or worktree to transplant modules, adapt a backbone, add LoRA or adapter layers, replace a head, or stitch together meaningful low-risk migration ideas with rollback-aware records in `explore_outputs/`. Do not use for end-to-end exploration orchestration on top of `current_research`, trusted baseline reproduction, conservative debugging, environment setup, verified contribution claims, or default repository analysis.
safe-debug
lllllllama/rigorpilot-skills
Rigor Debug / Rigor Audit skill for deep learning research work. Use when the user pastes a traceback, terminal error, CUDA OOM, checkpoint load failure, shape mismatch, NaN loss symptom, or training failure and wants conservative diagnosis before any patching, with debug fixes clearly separated from research contributions. Do not use for broad refactoring, speculative adaptation, automatic exploratory patching, or general repository familiarization.

