• Jordan_U@lemmy.ml
      link
      fedilink
      arrow-up
      2
      ·
      edit-2
      10 months ago

      Fun fact!

      The Asahi Linux drivers for the Apple M1 GPU were originally written in Python: https://asahilinux.org/2022/11/tales-of-the-m1-gpu/

      GPU drivers in Python?!

      Since getting all these structures right is critical for the GPU to work and the firmware to not crash, I needed a way of quickly experimenting with them while I reverse engineered things. Thankfully, the Asahi Linux project already has a tool for this: The m1n1 Python framework! Since I was already writing a GPU tracer for the m1n1 hypervisor and filling out structure definitions in Python, I decided to just flip it on its head and start writing a Python GPU kernel driver, using the same structure definitions. Python is great for this, since it is very easy to iterate with! Even better, it can already talk the basic RTKit protocols and parse crash logs, and I improved the tools for that so I could see exactly what the firmware was doing when it crashes. This is all done by running scripts on a development machine which connects to the M1 machine via USB, so you can easily reboot it every time you want to test something and the test cycle is very fast!

      • I Cast Fist@programming.dev
        link
        fedilink
        arrow-up
        2
        ·
        10 months ago

        Good for testing and iterating, but what about performance? Though I guess getting everything right is more important right now, translating it into another language will probably require less work that way

        • Jordan_U@lemmy.ml
          link
          fedilink
          arrow-up
          2
          ·
          10 months ago

          It has already been translated into rust. Python wasn’t ever intended to be used in the “real” driver, but I thought it was a fun anecdote none the less.

  • IndiBrony@lemmy.world
    link
    fedilink
    English
    arrow-up
    1
    ·
    edit-2
    10 months ago

    Fools haven’t even written it well! Translated:

    STOP WRITING

    • MEMORY WAS NEVER SUPPOSED TO BE AESSED DIRETLY

    • YEARS OF PROGRAMMING yet STILL ODE IS STILL WRITTEN with memory vulnerabilities

    • Wanted to aess memory diretly anyway? We had a tool for that: It was alled “ASSEMBLY”

    • “Yes please give me NULL of something. Please give me *&* of it” - Statements dreamed up by the utterly deranged

    LOOK at what Programmers have been demanding your Respet for all this time, with all of the omputers we built for them

    (These are REAL programs, written by REAL Programmers):

    ??? ??? ???

    They have played us for absolute fools

  • Ethanol@pawb.social
    link
    fedilink
    English
    arrow-up
    0
    ·
    10 months ago

    But what if I want a union struct to quickly interpret floats as ints and vice versa! I need my C hacks!

      • Jordan_U@lemmy.ml
        link
        fedilink
        arrow-up
        0
        ·
        10 months ago

        … Which compilers don’t consistently enforce, much like most undefined behavior in C.

          • Jordan_U@lemmy.ml
            link
            fedilink
            arrow-up
            0
            ·
            10 months ago

            By refusing to compile any code that has undefined behavior. This is what rust’s compiler does, and is simply not possible for a C compiler to do.

            • Jordan_U@lemmy.ml
              link
              fedilink
              arrow-up
              0
              ·
              10 months ago

              To put it another way:

              Strict aliasing is an invariant that C compilers assume you as a developer will not violate, and use that assumption to make optimization choices that, if you as the developer have failed to follow the strict aliasing rules, could lead to undefined behavior. So it’s a variant that the compiler expects, but doesn’t enforce at compile time.

              I guess it is possible to just disable all such optimizations to get a C compiler that doesn’t create UB just because strict aliasing rules were broken, but there are still many ways that you can trigger UB in C, while safe rust that compiles successfully theoretically has no UB at all.

              • uis@lemmy.world
                link
                fedilink
                arrow-up
                0
                ·
                10 months ago

                Strict aliasing exists not for optimization, but for type alignment. You may need more space on stack to save uint32_t than uint8_t[5] because former has 32-bit alignment.

                • Jordan_U@lemmy.ml
                  link
                  fedilink
                  arrow-up
                  0
                  ·
                  9 months ago

                  Either way, this is a rule that you as a human are required to follow, and if you fail the compiler is allowed to do anything, including killing your cat.

                  It’s not a rule that the compiler enforces by failing to build code with undefined behavior.

                  That is a fundamental, and extremely important, difference between C and rust.

                  Also, C compilers do make optimization decisions by assuming that you as a human programmer have followed these strict aliasing rules.

                  https://gist.github.com/shafik/848ae25ee209f698763cffee272a58f8

                  Has a few examples where code runs “properly” without optimizations but “improperly” with optimizations.

                  I put “improperly” in quotes because the C spec says that a compiler can do whatever it wants if you as a human invoke undefined behavior. Safe rust does not have undefined behavior, because if you write code which would invoke UB, rustc will refuse to build it.

            • uis@lemmy.world
              link
              fedilink
              arrow-up
              0
              ·
              10 months ago

              Compiler knows what is undefined behaviour better than all lemmy experts

    • entropicdrift@lemmy.sdf.org
      link
      fedilink
      arrow-up
      1
      ·
      10 months ago

      Originally Windows was written in assembly and ran on top of DOS, but since Windows 2000 and XP, it’s been exclusively running on the NT kernel, which is written primarily in C, with some C++ in there as well.

      The actual userspace is mostly C++ and C#.

      • Skull giver@popplesburger.hilciferous.nl
        link
        fedilink
        arrow-up
        0
        ·
        10 months ago

        Microsoft is quickly writing more and more Rust code these days. They rolled out Rust kernel components even before Linux, and their efforts actually include rewrites rather than making the API available to developers.

        There’s decades of code in Windows, but the successful conversion for DirectWrite font parsing is probably a sign of things to come. MS seems to even be porting some COM modules to Rust, which would be the last thing I would start to port (so many pointer pointers!).

    • wreckedcarzz@lemmy.world
      link
      fedilink
      English
      arrow-up
      0
      ·
      10 months ago

      Well, you know those claims that Java runs on 18 trillion devices? How do you think they got there, hmmmmm?

      • Skull giver@popplesburger.hilciferous.nl
        link
        fedilink
        arrow-up
        0
        ·
        10 months ago

        Windows actually came with a Java runtime for a short while. Then Microsoft got sued, the Java VM was killed, and C# was created (which looks, feels, and behaves very similar to Java) as an alternative.

        Just imagine what would’ve happened if Microsoft and Sun worked together. Java would be in everything on Windows now, not just as a basis for modded Minecraft.

    • Jordan_U@lemmy.ml
      link
      fedilink
      arrow-up
      0
      ·
      10 months ago

      Leaks aren’t usually security critical though, and I’ve never heard of sudo triggering the OOM killer.

      Also, no general purpose language that I’m aware of can guarantee a lack of memory leaks.

      • simple@lemm.ee
        link
        fedilink
        arrow-up
        0
        ·
        10 months ago

        no general purposd language can guarantee a lack of memory leaks

        You’re going to summon every Rust enthusiast on the platform

      • kevincox@lemmy.ml
        link
        fedilink
        arrow-up
        0
        ·
        10 months ago

        Especially since sudo is generally quite short lived. Unless it is leaking a significant amount of memory waiting for authentication that never comes it is insignificant. It would actually be pretty easy to argue that sudo just shouldn’t free memory at all. This would be better for security (all pointers live forever) and possibly faster as upon exec the kernel can just wipe all state rather than having free carefully account for the releases.

  • MeanEYE@lemmy.world
    link
    fedilink
    arrow-up
    0
    ·
    10 months ago

    Do people still think, after all this time and so many different languages, that there will be one language to rule them all? I mean technically you can drive nails with a rock, but you don’t see a carpenter using one. Right tool for the job. Always was, always will be.

  • pinkdrunkenelephants@lemmy.cafe
    link
    fedilink
    English
    arrow-up
    0
    ·
    edit-2
    10 months ago

    Primitive, provincial knuckledragging Americans/Brits/Right-wing country’s citizens don’t want to learn C or understand how their computer truly works. 😆

    Yeah. That tracks

    • Pipoca@lemmy.world
      link
      fedilink
      arrow-up
      0
      ·
      edit-2
      10 months ago

      C is not how a computer truly works.

      If you want to know how computers work, learn assembly and circuit design. You can learn C without ever thinking about registers, register allocation, the program counter, etc.

      Although you can learn assembly without ever learning about e.g. branch prediction. There’s tons of levels of abstraction in computers, and many of the lower level ones try to pretend you’ve still got a computer from the 80s even though CPUs are a lot more complex than they used to be.

      As an aside, I’ve anecdotally heard of some schools teaching Rust instead of C as a systems language in courses. Rust has a different model than C, but will still teach you about static memory vs the stack vs the heap, pointers, etc.

      Honestly, if I had to write some systems software, I’d be way more confident in any Rust code I wrote than C/C++ code. Nasal demons scare me.

  • MeanEYE@lemmy.world
    link
    fedilink
    arrow-up
    0
    ·
    10 months ago

    Just use brainfuck for everything. The entry barrier for the programming industry needs to be higher anyway.

      • dejected_warp_core@lemmy.world
        link
        fedilink
        arrow-up
        0
        ·
        edit-2
        10 months ago

        For the programmer? Very no.

        For saving space if run via interperter? No.

        For running compiled for conventional CPUs? No.

        Compared to CISC instruction sets? Absolutely no.

        BF might be highly efficient if crunched down to a bit-packed representation (3 bits per instruction) and run on an FPGA that understands it.

        • frezik@midwest.social
          link
          fedilink
          arrow-up
          0
          ·
          10 months ago

          For demonstrating to CS freshmen that Turing Completeness isn’t that remarkable of a language feature: very highly efficient.

      • MeanEYE@lemmy.world
        link
        fedilink
        arrow-up
        0
        ·
        10 months ago

        Can be compressed very efficiently. I do dread the thought of writing a driver in brainfuck.

  • slembcke@lemmy.ml
    link
    fedilink
    arrow-up
    0
    ·
    10 months ago

    I enjoy the selection bias in the comments for these sorts of posts. >_< There’s a few people saying “I kinda like C”, a few saying “use Python instead”, and a whole lot saying “Rust is my lord and savior”. Completely disjoint from the real world usage of the languages for whatever practical, pragmatic, or ideological measures they are used for.

    • MojoMcJojo@lemmy.world
      link
      fedilink
      arrow-up
      0
      ·
      10 months ago

      I know barely anything about programming languages and only ask as a fan, what are the real world usages of languages and what are their practical, pragmatic, or ideological measures that they are used for?

      • slembcke@lemmy.ml
        link
        fedilink
        arrow-up
        0
        ·
        10 months ago

        I guess by real world usage I mean what proportion of code is being made with them. You should be skeptical of their accuracy, but there are measures for that. Like there is this one: https://www.tiobe.com/tiobe-index/, but it describes it’s methodology as being about popularity based on articles, news, and other such things. Github publishes a very different chart as does RedMonk. Rust barely shows up on these charts, but Rust fans are very enthusiastic in threads like this. I like Rust well enough, but I also find the over-enthusiasm amusing.

        By practical/pragmatic I mean the ability to target a lot of hardware with C. Sometimes the tooling is crap, but it’s very universal. Being built on LLVM Rust can go onto plenty of hardware too, but it’s probably not the tooling given to you by a platform vendor. It’s also been around for a long time, so using Rust would mean a rewrite. Sometimes C is simply the choice. As for ideologically: Rust solves some pretty nasty programming issues, but sometimes I think it’s fans over-estimate the percentage of real world problems it actually solves while ignoring that Rust can be more expensive to write. (shrug) Sometimes there’s no such thing as a silver bullet.

  • RubberElectrons@lemmy.world
    link
    fedilink
    arrow-up
    0
    ·
    edit-2
    10 months ago

    I really like C because I can just get to the heart of an action and make it happen without much surrounding code.

    I could make classes and blah blah blah if I want to make a large, complex program but I’d rather write several small, simple to grok programs which pass information around so each program can do its one simple thing, quickly and easily. Chain the small programs together with bash or something, and bingo, you’ve got a modular high speed system.

    I’m not a programmer, actually a mechanical engineer. But the Unix philosophy of simple, modular tools is great, provided one properly checks and sanitizes inputs.

    • macaroni1556@lemmy.ca
      link
      fedilink
      arrow-up
      0
      ·
      10 months ago

      What you’re describing sounds like Python. Not really C’s strong suit.

      If you haven’t checked it out yet, you certainly should!

      • remotelove@lemmy.ca
        link
        fedilink
        arrow-up
        0
        ·
        10 months ago

        I agree with your main point. Python does a great job of replacing lots of tiny, chained scripts. Simple API calls with wget or curl have a place, but can spiral out of control quickly if you need to introduce any grain of control like with pagination, as an example.

        Maintaining one Python app (or “script”) can still adhere to the unix philosophy of simplicity but can bend some rules as far as monolithic design is concerned if you aren’t careful.

        It all boils down to whether you are introducing complexity or reducing it, IMHO.