Let Claude Code write and test a Linux kernel speedup
- Level Intermediate
- Setup1–3 hrs (estimated)
- Evidence Some evidence
- CostNot stated in source
- No-codeNo
What they did
Linus Torvalds posted a page-cache patch series and credited Claude Code with the code, the test programs and the review. Small-read throughput on his laptop went from about 10 million to 33 million reads per second.
How it went
News write-up of a public LKML post (Oct 12, 2026). Specific numbers: ~10M to ~33M small reads/sec on Torvalds' own laptop benchmark
Worth knowing
- News write-up of a public LKML post (Oct 12, 2026). Specific numbers: ~10M to ~33M small reads/sec on Torvalds' own laptop benchmark.
Prompt
Cleaned version rewritten from the source — not a direct quote.
Work in [repo] on [the hot code path, e.g. how cached file reads are handled]. First write a small benchmark program that measures [the metric, e.g. reads per second] and record a baseline. Propose a change, implement it on a new branch, write tests for the edge cases, rerun the benchmark and report before/after numbers. Then review your own patch as a strict maintainer would and list every risk. Do not push anything.
Recommended
Try it in Grok Bot
- Copy the prompt (button below).
- Open Grok Bot chat and paste it. Replace any [bracketed] placeholders.
- Connect the listed tools (as needed) as connectors when Grok Bot asks.
- If the task repeats, ask: "Make this a daily routine."
Alternatives: the same prompt works in Claude Code, Claude, ChatGPT Agent, OpenClaw.
Set it up
Stack: Claude Code
- Read the original source for full context.
- Connect the tools your agent needs.
- Paste the prompt and replace [bracketed] placeholders.
- Run once manually and check the output before scheduling it.
Generic suggested steps, not from the source.
Evidence
News write-up of a public LKML post (Oct 12, 2026). Specific numbers: ~10M to ~33M small reads/sec on Torvalds' own laptop benchmark
Source: Web · Verified by us: not yet · Report broken or outdated