class: center, middle, inverse # Rust Coreutils in Ubuntu ## One troll, one argument, one answer ## Sylvestre Ledru ## FOSDEM 2026 --- # What are the Coreutils? -- Who doesn't know about it? -- First implementation in 1970! -- Some examples: -- * ls, cp, mv, ln, rm, etc -- * chmod, chown, runcon, etc -- * vdir, tac, pr -- * More than 100 commands... --- # Why? - I did two presentations at FOSDEM about it: - <a href="https://archive.fosdem.org/2023/schedule/event/rust_coreutils/">in 2023</a> - <a href="https://archive.fosdem.org/2025/schedule/event/fosdem-2025-6196-rewriting-the-future-of-the-linux-essential-packages-in-rust-/">in 2025</a> -- The TLDR is basically: - It is super interesting - learning Rust, low level & Coreutils, -- - C is a language from the past, -- - New generations don't want to do C, -- - It isn't about license, -- - Harder, Better, Faster, Stronger --- # Ubuntu's Approach "Carefully but purposefully oxidising" -- - We have been coordinating closely -- - Not a rewrite frenzy -- - Not "Rust everywhere" -- - Gradual introduction -- - Explicit rollback paths -- - They think that Rust is a future for Linux -- https://discourse.ubuntu.com/t/carefully-but-purposefully-oxidising-ubuntu/56995 -- It isn't the first major software written in Rust in Ubuntu -- Firefox was, years ago! --- class: center, middle, inverse template: inverse <img src="duty_calls.png" /><br /> https://xkcd.com/386/ -- # Who reads -- Hacker News ? -- Phoronix ? -- Reddit ? -- And ... who cried reading some of the comments ? --- class: center, middle, inverse template: inverse # One Troll, One Argument, One Answer --- # Troll #1 <img src="msg1.png" width="100%" alt="That's the worst idea I've seen in a while. People rely on the behaviors of the GNU coreutils, this is going to break a lot of stuff. The uutils are not there yet at all." /> – Hacker News commenter https://news.ycombinator.com/item?id=43779457 --- # Argument #1 Coreutils exposes: - undocumented behaviour - accidental APIs - semantics scripts depend on Replacing them *will* surface breakage. --- # Answer #1 Yes, indeed -- But... And we found those behaviours but we fixed most of them thanks to: - Pre-release testing - 95% GNU tests pass - 87% code coverage - Differential fuzzing - Strong community testing - Ubuntu Pre release testing - CI rebuilds - Users testing --- # Troll #2 <img src="msg5.png" width="100%" alt="It might be a bit of bad publicity for those who want to rewrite as much as possible in Rust. While Rust is not to blame, it shows that just rewriting something in Rust doesn't magically make it better (as some Rust hype might suggest). Maybe Ubuntu was a bit too eager in adopting the Rust Coreutils, caring more about that hype than about stability."/> – Hacker News commenter https://news.ycombinator.com/item?id=45388054 --- # Argument #2 - Rust hype over practical benefits - Rewriting doesn't automatically improve software - Ubuntu prioritized trendy tech over stability - Language choice ≠ better outcomes --- # Answer #2 We weren't chasing hype. We chose Rust for **concrete benefits**: - Memory safety (eliminates entire bug classes) - Performance improvements - Better tooling and maintainability - Modern development experience - Easy portability --- # Troll #3 <img src="msg3.png" alt="Absolutely fuck this. I hope coreutils will still be available for sane people along with sudo." /> - Phoronix forum user https://www.phoronix.com/forums/forum/software/distributions/1574244-ubuntu-25-10-proceeding-with-its-rust-coreutils-transition --- # Argument #3 Fear of: - forced replacement - no rollback - ideological push --- # Answer #3 No hard cutover. - GNU coreutils remained available - Easy rollback - Potentially important for some users - Reversibility was a requirement, not an afterthought - We contribute to the GNU project, they contribute to ours --- # Troll #4 <img src="msg4.png" width="100%" alt="If these tools make their way into Linux distributions, as seems to be the goal, it's going to be war. We're not going to allow technofascists to steal Linux away from us, one Rust rewrite at a time. The whole Rust-in-the-Linux-kernel debate just got serious. We need to move against this development forcefully and decisively so that Rust development in Linux can move forward with confidence." /> https://users.rust-lang.org/t/im-shocked-the-rust-community-is-pushing-an-mit-licensed-rust-rewrite-of-gnu-coreutils/126110 --- # Argument #4 - Ideological strategy - Coordinated plan --- # Answer #4 **There is no conspiracy.** - Purely technical decisions - No coordinated takeover plan - Individual projects, individual choices - No connection to Linux kernel politics --- # Troll #5 <img src="memes/so-whats-our-next-move-after-coreutils-v0-t2736xnglopf1.jpeg" width="100%" /> --- # Answer #5 Yes, we have bugs, but: -- - The 17x slower has been fixed in a week - One long line - fixed a month --- # What These Bugs Taught Us - Rust has never been a silver bullet (we knew!) -- - Coreutils can be weird -- - It improved the GNU implementation! -- - Rollout strategy matters as much as code --- # Community Contributions <img src="pr_timeline.png" width="100%" /> -- in 2025: -- 2054 pull requests - 4004 commits -- by 206 Contributors --- # Community Contributions - Growing contributor base -- - Cross-platform improvements -- - Upstream collaboration with GNU -- - Identified and fixed --- # Next? - Reach 100% compat with GNU - Replicate this with other programs: -- - sed - diffutils - procps - util linux - .. etc -- Lot of good first bugs to learn Rust or improve your skills --- # Final Takeaway We shipped with 25.10 for most of the programs. -- 26.04 in a few weeks -- The trolls weren't wrong about *risk* -- They were wrong about *process* & *mitigation*. --- class: center, middle, inverse template: inverse # Rewriting `/bin/true` is easy # Rewriting 30 years of expectations is not --- class: center, middle # Questions?