lcamtuf :verified: :verified: :verified: (@lcamtuf@infosec.exchange)
infosec.exchange
The coreutils Rust rewrite story is pretty funny. Coreutils are tools like rm, mv, mkdir, etc. Unlike binutils, this isn't a fertile ground for memory safety bugs. But, the rewrite was completed, and in the spirit of progress, Canonical decided to switch. But do you know what coreutils are a fertile ground for? Race conditions around file creation, deletion, permission setting, and so on. The original code accounted for decades of hard-learned lessons in that space. The Rust rewrite did not: https://seclists.org/oss-sec/2026/q2/332 PS. I'm not dunking on Rust. It's just that... starting over from scratch has its hidden costs.
  • Sanctus ( Sanctus@anarchist.nexus ) English
    5·
    5 months ago

    Rust is so obviously a cuck licensing grift for corpos that it even does the corpo thing! Dont worry about why it is just remake it so I dont have to pay or recognize anybody.

    • TehPers ( TehPers@beehaw.org ) English
      9·
      5 months ago

      You’re right. The only reason someone would write something in Rust is to change the license. Nobody would use Rust for any other reason, nor would they use any other language for that purpose.

          • Lucy :3 ( 30p87@feddit.org ) 
            1·
            5 months ago

            It’s not, and it’s especially not to use resulting software. 90% of software I have problems with, let it be compilation errors or missing missing features due to bad libraries, is Rust

            • TehPers ( TehPers@beehaw.org ) English
              9·
              5 months ago

              The anti-Rust circlejerk is strong, wow.

              I have loads of software written in Rust installed, many of which I compile myself. Not once has the language choice been an issue that I’ve had with the software. Some of that software may be missing features, but that has little to do with the language and more to do with my decision to install v0.1.4 of the software.

              Python, on the other hand…

              And while we’re on the topic of coreutils, why does every C and C++ project I look at require me to install dozens of random libraries myself? I guess because they’re dynamically linked, which is cool and all, except the page to install each dependency has an encyclopedia of a README written in 1984 in a text file. It’s not usually an issue, just annoying.

  • spectrums_coherence ( spectrums_coherence@piefed.social ) English
    5·
    5 months ago

    I am not happy about the license change for the rewrite. Theoretically, affine type system can be used to reason about deadlock and race conditions (although with caveats https://dl.acm.org/doi/abs/10.1145/3571229).

    That being said, race conditions are logical bugs, and a programming language is only as powerful as the specification provided to it. It is possible there are condition that developers are simply not aware of.

    It seems like specifying the race behavior of coreutil in an language agnostic way would be a fruitful endeavor.

    • Snarwin ( Snarwin@fedia.io ) 
      3·
      5 months ago

      The real underlying issue here is that the OS interfaces these programs rely on are extremely error-prone and difficult to use correctly. A naive implementation of these system utilities in any language is basically guaranteed to stumble into bugs and vulnerabilities like the ones described in the linked message.

    • FizzyOrange ( FizzyOrange@programming.dev ) 
      1·
      5 months ago

      Tbh while DST (or just “testing” as hardware people would call it) is very obviously a great idea, I’m not sure it would have helped here - in order to detect these TOCTOU bugs you would need stimulus that triggers it and some kind of checker/model that has the correct behaviour.

      That’s totally possible but it’s pretty hardcore testing for a software project and it’s difficult to imagine doing that without realising that you have a TOCTOU issue just by inspection.