Announcement

Collapse
No announcement yet.

Linux VGEM Driver Rewritten In Rust Sent Out For Review

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Linux VGEM Driver Rewritten In Rust Sent Out For Review

    Phoronix: Linux VGEM Driver Rewritten In Rust Sent Out For Review

    Besides the ongoing work around the reverse-engineered Apple Silicon graphics driver being brought up in the Rust programming language, the other notable Rust effort within the Direct Rendering Manager (DRM) kernel subsystem is a rewrite of the basic VGEM driver in this increasingly-used programming language. That Rust VGEM driver has now been sent out on the mailing list for review as part of a request for comments...

    Phoronix, Linux Hardware Reviews, Linux hardware benchmarks, Linux server benchmarks, Linux benchmarking, Desktop Linux, Linux performance, Open Source graphics, Linux How To, Ubuntu benchmarks, Ubuntu hardware, Phoronix Test Suite

  • #2
    I expect an upcoming avalanche of Rust code for Linux. Accepting Rust may be one of the maintainers' greatest choices: a higher minimum bar for code quality that's software-enforced gets more contributors that don't get the sting of human review for minor mistakes.

    Rust "levels up" the conversations.

    Comment


    • #3
      I really need to get better in Rust. In C I can mostly tell what code does, not necessarily in Rust. Certainly a challange for reviews and subsystem maintainers, at least I guess.

      Comment


      • #4
        Originally posted by STiAT View Post
        I really need to get better in Rust. In C I can mostly tell what code does, not necessarily in Rust. Certainly a challange for reviews and subsystem maintainers, at least I guess.
        I find that surprising since Rust's ownership system and other features like pattern matching makes things pretty easy to follow. I used to write in C# and would put in tremendous effort to basically make an ad-hoc runtime ownership system so I could track resources and execution paths easier. With Rust, it's baked right in. I don't even need to write tests or run the program, it's usually pretty obvious how things work just from how the code is laid out and annotated. Also helps that it has a culture of documenting absolutely everything.

        Comment


        • #5
          The "sent out for review" is the problem. How does "the public" review code when there are only so few individuals knowledgeable enough to conduct a meaningful review. "Trust the borrow-checker" is NOT a review.

          ... and don't tell me "rust is so simple anybody can do that". rust is clearly *not that simple*. It doesn't fail to amaze me to see how many developers that absolutely despise C++ welcome rust with open arms.

          Comment


          • #6
            Originally posted by lowflyer View Post
            rust is clearly *not that simple*. It doesn't fail to amaze me to see how many developers that absolutely despise C++ welcome rust with open arms.
            Are you claiming C++ is simpler than Rust??

            Comment


            • #7
              Originally posted by lowflyer View Post
              It doesn't fail to amaze me to see how many developers that absolutely despise C++ welcome rust with open arms.
              Maybe because rust learned from the many mistakes of C++? Maybe because C++ is even more complicated than rust?

              Disclaimer: I have done my stint of programming in C++, but have yet to truly put time into rust. But I am looking forward to it

              Comment


              • #8
                Originally posted by Veto View Post
                Maybe because rust learned from the many mistakes of C++? Maybe because C++ is even more complicated than rust?

                Disclaimer: I have done my stint of programming in C++, but have yet to truly put time into rust. But I am looking forward to it
                I welcome Rust precisely because I absolutely despise C++.

                Comment


                • #9
                  Originally posted by lowflyer View Post
                  The "sent out for review" is the problem. How does "the public" review code when there are only so few individuals knowledgeable enough to conduct a meaningful review. "Trust the borrow-checker" is NOT a review.

                  ... and don't tell me "rust is so simple anybody can do that". rust is clearly *not that simple*. It doesn't fail to amaze me to see how many developers that absolutely despise C++ welcome rust with open arms.
                  No, really. When you have clearly marked unsafe areas where pointer arithmetic is occurring and don't need to worry about data races because mutexes themselves control the data they guard, you remove a lot of time and effort using complex tools and manpower to track down bugs related to those, which are most bugs.
                  Not to mention, no worrying about the compiler introducing bugs due to a bad optimization, since Rust's syntax basically tells the compiler how to optimize it, instead of needing to employ strange and complex heuristics to attempt to find patterns in code that can be replaced or removed.
                  You know, believe it or not, Rust isn't complex if you actually bother to learn it. I find it hilarious people who don't bother to learn it think you have to be some kind of 4D chess grandmaster to comprehend it, it's probably one of the easiest languages to understand for real-world non-trivial programs. It reminds me of how people used to say C is hard because they didn't understand pointers.

                  Comment


                  • #10
                    Originally posted by Ironmask View Post
                    It reminds me of how people used to say C is hard because they didn't understand pointers.
                    Funny, because it's Rust proponents who hate pointers for being unsafe which means that they do not understand them themselves.

                    Comment

                    Working...
                    X