--- title: Linus Torvalds's first LLM-assisted patch series | daily.dev description: Linus Torvalds posted his first LLM-assisted patch series for Linux kernel review, a ten-patch set that speeds up folio lookup by using RCU to delay releasing... canonical: https://daily.dev/posts/linus-torvalds-s-first-llm-assisted-patch-series-ftwbshwip twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Linus Torvalds's first LLM-assisted patch series | daily.dev og:description: Linus Torvalds posted his first LLM-assisted patch series for Linux kernel review, a ten-patch set that speeds up folio lookup by using RCU to delay releasing... og:url: https://daily.dev/posts/linus-torvalds-s-first-llm-assisted-patch-series-ftwbshwip og:image: https://api.daily.dev/og/posts/FTwbShWip.png og:image:alt: Linus Torvalds's first LLM-assisted patch series og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # Linus Torvalds's first LLM-assisted patch series **[LWN.net](https://daily.dev/sources/lwn)** · 1 min read · 0 upvotes · 0 comments ## Summary Linus Torvalds posted his first LLM-assisted patch series for Linux kernel review, a ten-patch set that speeds up folio lookup by using RCU to delay releasing hot folios, allowing data access under RCU protection without refcount overhead in the fast path. The change produces an order-of-magnitude speedup on read()-heavy benchmarks, though reviewers have already found a race condition requiring further work. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### What does Linus Torvalds's new RCU-based folio lookup patch series do for read() performance? It delivers an order-of-magnitude speedup for read()-heavy benchmarks by marking hot folios for delayed release after an RCU grace period, then performing both the folio lookup and the actual data access itself under RCU protection. This avoids taking a refcount in the fast path when a user copy does not trigger a page fault. The ten-patch series is still under review, and reviewers have already found a race condition needing further fixes. _Developers tracking kernel performance work can follow this RCU folio optimization's review progress on daily.dev._ ### Did Linus Torvalds use an LLM to help write a Linux kernel patch? Yes, Torvalds mentioned at the Maintainers Summit that he used a large language model while working on a patch series addressing a longstanding kernel problem, then posted the resulting ten-patch series for review. The patches improve folio lookup performance using RCU-protected data access, though the series is still being reviewed and a race condition has already been found. _Developers curious about AI-assisted contributions to major open source projects can follow this story on daily.dev._ ## Community take How the wider developer community reacted, aggregated from 1 discussion and 78 comments across hackernews (as of 2026-10-15). **TL;DR:** Reaction centers less on the technical patch itself and more on what it symbolizes: a famously AI-skeptical figure using an LLM assistant, sparking a broad debate about AI's role in critical open-source infrastructure, copyright implications of machine-generated code, and whether AI-assisted coding is a slippery slope or just a useful tool. **Sentiment:** 40% positive · 35% mixed · 25% skeptical **The case for** - Several argue AI quality has improved dramatically and is now genuinely useful rather than producing unusable 'slop'. - Some see this as pragmatic validation: if someone as experienced as Linus finds it useful after reviewing it, that's meaningful. - The diffs themselves were described as moderately sized and readable, unlike typical bloated LLM commits. **The pushback** - Some worry machine-generated code may not be copyrightable, which could threaten the project's open-source licensing over time if LLM contributions scale up. - Critics note there's already a clear skill factor in prompting LLMs well, meaning 'anyone can do what Linus did' is misleading. - Some find the LLM-generated commit message text annoying/verbose compared to human-written commit messages. - A race condition was already found in the patch, used by skeptics as evidence AI-assisted code still needs heavy human review. **By community** - hackernews (heated): The thread sprawls into a broader, contentious debate about AI's legitimacy in core infrastructure, copyright of machine-written code, and whether anti-AI sentiment is outdated, with sharp disagreement throughout. **Hottest debate:** Whether AI-assisted contributions pose a long-term threat to the open-source/copyright status of projects versus being just another incremental tool like any other. **Open questions** - Does code significantly produced by an LLM remain copyrightable, and what does that mean for open-source licensing if such contributions grow? - Has the question of machine-generated code copyrightability actually been legally tested anywhere? **Highlights** > Linus's policy on patch acceptance covers the LLM problem: it must still be audited by a maintainer before acceptance (and thus, actually be decent code), but also, because of LLMs, be significantly produced by a human, as machine written code cannot be Copyrighted, thus, cannot be Open Source, and thus, cannot be in the kernel because it isn't open source. Nobody is really against LLMs producing, for example, exploits for novel security bugs to assist security researchers in patching those bugs. The exploit itself does not need to be Copyrightable nor Open Source. Nor is anybody against LLMs accelerating debugging, and helping find the location and root cause of the bug; again, that is not producing the end product code, thus does not need to be Copyrightable nor Open Source. > — [DiabloD3 on hackernews · 4 comments](https://news.ycombinator.com/item?id=50091555) > It will poison it if you replace all of your code with public domain code. A small patch here and there of uncopyrightable code doesn't really matter. Years of small patches, on the other hand, could end up replacing most or all the code. If you have the policy that all contributions have to be significantly written by a human, and thus copyrightable, you don't have to worry about whether or not certain files or modules or the entire project are copyrightable in the future. > — [Aloisius on hackernews](https://news.ycombinator.com/item?id=50092628) > I am not surprised - the general pattern with Torvalds seems to be that he's maximally pragmatic. His problems with say Nvidia drivers aren't the fact that they're proprietary, it's that they are difficult to interface with the kernel. I have a strong impression that what matters to Linus is software that works well, so he cares a lot about the quality of the kernel and ideally wants other software to be similarly reliable. If he judges that Claude does the best job, it'd be very much like him to use Claude. It would be unlike him to use an open program if he considers it inferior. > — [ACS\_Solver on hackernews](https://news.ycombinator.com/item?id=50092915) > I looked at the diffs, they are moderately sized and well readable. Nothing like the 2k loc people tend to commit with LLMs. One can work with commits and MRs like that. > — [niemandhier on hackernews](https://news.ycombinator.com/item?id=50092905) > Didn't Linus say he stopped coding a while ago, or at least coding the kernel, and mainly reviews and manages the codebase now? Whether the code comes from a machine or a human, perhaps it matters little to how he's been working for a while. It's interesting that although I guess he reviewed this LLM generated code, the human review found a race condition. > — [tarkin2 on hackernews · 2 comments](https://news.ycombinator.com/item?id=50091497) **Source threads** - [hackernews](https://news.ycombinator.com/item?id=50090842) · 78 points · 78 comments ## Similar posts on daily.dev - [Linus Torvalds "Did What I Preach", Used AI To Make Code Decipherable & Uncovered More Bugs](https://daily.dev/posts/linus-torvalds-did-what-i-preach-used-ai-to-make-code-decipherable-uncovered-more-bugs-tsqqcmos3) · Phoronix · 1 upvotes · 0 comments - [Kernel prepatch 7.1-rc5](https://daily.dev/posts/kernel-prepatch-7-1-rc5-vzqd1yxpu) · LWN.net · 0 upvotes · 0 comments - [MGLRU-FG Delivering Up To 10~40% Higher Performance For Linux In Some Tests](https://daily.dev/posts/mglru-fg-delivering-up-to-10-40-higher-performance-for-linux-in-some-tests-vi17swap9) · Phoronix · 1 upvotes · 0 comments --- Tags: [#linux](https://daily.dev/tags/linux), [#performance](https://daily.dev/tags/performance) [View this post on daily.dev](https://daily.dev/posts/linus-torvalds-s-first-llm-assisted-patch-series-ftwbshwip) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"Linus Torvalds's first LLM-assisted patch series","url":"https://daily.dev/posts/linus-torvalds-s-first-llm-assisted-patch-series-ftwbshwip","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/linus-torvalds-s-first-llm-assisted-patch-series-ftwbshwip"},"datePublished":"2026-10-14T13:55:43.817Z","dateModified":"2026-10-15T10:15:12.498Z","description":"Linus Torvalds posted his first LLM-assisted patch series for Linux kernel review, a ten-patch set that speeds up folio lookup by using RCU to delay releasing...","image":"https://media.daily.dev/image/upload/s--0_ODbtD2--/f_auto/v1722860399/public/Placeholder%2008","thumbnailUrl":"https://media.daily.dev/image/upload/s--0_ODbtD2--/f_auto/v1722860399/public/Placeholder%2008","isAccessibleForFree":true,"articleSection":"LWN.net","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"LWN.net","logo":"https://media.daily.dev/image/upload/s--4jjE7GIK--/f_auto,q_auto/v1774960833/logos/lwn?_a=BAMAMiWQ0","url":"https://daily.dev/sources/lwn"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/linus-torvalds-s-first-llm-assisted-patch-series-ftwbshwip","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"linux,performance","timeRequired":"PT1M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"LWN.net","item":"https://daily.dev/sources/lwn"},{"@type":"ListItem","position":3,"name":"Linus Torvalds's first LLM-assisted patch series"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/linus-torvalds-s-first-llm-assisted-patch-series-ftwbshwip#faq","mainEntity":[{"@type":"Question","name":"What does Linus Torvalds's new RCU-based folio lookup patch series do for read() performance?","acceptedAnswer":{"@type":"Answer","text":"It delivers an order-of-magnitude speedup for read()-heavy benchmarks by marking hot folios for delayed release after an RCU grace period, then performing both the folio lookup and the actual data access itself under RCU protection. This avoids taking a refcount in the fast path when a user copy does not trigger a page fault. The ten-patch series is still under review, and reviewers have already found a race condition needing further fixes. Developers tracking kernel performance work can follow this RCU folio optimization's review progress on daily.dev."}},{"@type":"Question","name":"Did Linus Torvalds use an LLM to help write a Linux kernel patch?","acceptedAnswer":{"@type":"Answer","text":"Yes, Torvalds mentioned at the Maintainers Summit that he used a large language model while working on a patch series addressing a longstanding kernel problem, then posted the resulting ten-patch series for review. The patches improve folio lookup performance using RCU-protected data access, though the series is still being reviewed and a race condition has already been found. Developers curious about AI-assisted contributions to major open source projects can follow this story on daily.dev."}}]} ```