Announcement

Collapse
No announcement yet.

Rust-Based Redox OS Had A Busy Year With Rewriting Its Kernel, Writing A File-System

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

  • Rust-Based Redox OS Had A Busy Year With Rewriting Its Kernel, Writing A File-System

    Phoronix: Rust-Based Redox OS Had A Busy Year With Rewriting Its Kernel, Writing A File-System

    Redox OS started development mid-way through last year while this year things really took off for this Rust-written operating system from scratch. The project has provided a recap of all of their OS accomplishments for 2016...

    http://www.phoronix.com/scan.php?pag...-OS-2016-State

  • #2
    No, please no, not another FS. Are there not enough FOSS-Filesystems that they can implement?

    Comment


    • #3
      Originally posted by Nille View Post
      No, please no, not another FS. Are there not enough FOSS-Filesystems that they can implement?
      They said they tried ZFS. Reading was working (if i remember correct) but writing was just not working out to well, seems to bee pretty complicated in ZFS. I don't see a huge problem in this – it is not that one should install Redox in late April 2017 for production use. Let them experiment what works and what not. There is no such thing as the single best Filesystem to simply choose. They did anyway. ZFS is pretty cool but it turns out not working for them and it looks like – in the process of integrating it they seem to realized how to change it to better fit their requirements.

      Comment


      • #4
        Originally posted by Nille View Post
        No, please no, not another FS. Are there not enough FOSS-Filesystems that they can implement?
        If they were proposing another FS for Linux I'd maybe agree but no; they want to implement a FS written in Rust for their OS written in Rust. It is a fun and interesting project and I wish them all the best.

        Comment


        • #5
          Originally posted by Nille View Post
          No, please no, not another FS. Are there not enough FOSS-Filesystems that they can implement?
          They are aiming at next-gen filesystems, so it seems TFS will be competing with btrfs or ZFS (or with bcachefs, lol), not with every other filesystem.

          The sad part is that if it manages to be cool and done relatively fast due to Rust, it won't be merged EVAR in Linux because it's not written in C.

          Comment


          • #6
            The overall design seems right (the best off plan9 & minix) but targeting the x86 is already costing them in kernel purity with no real good reason other then the "Runs on Real Hardware" sticker... Well, whatever. Best of luck I guess.

            Comment


            • #7
              Originally posted by Nille View Post
              No, please no, not another FS. Are there not enough FOSS-Filesystems that they can implement?
              The purpose of writing an entire OS in rust is to write an entire OS in rust. They started with writing a ZFS clone in rust and it turned out not work out. I believe reading was working but writing not. On the way to implement ZFS they figured how to adjust ZFS to better fit the needs of Redox.

              Comment


              • #8
                Redox is currently looking to further fund development. He's got a lot of ambitions (see the link), and it's really exciting what could be accomplished. Rust is a powerful language and could open up a lot of opportunities for developers to get into OS/systems development, and this project is one of the flagships of that goal. I'd love to see development on Redox grow!
                I am [Jackpot51](https://github.com/jackpot51) the creator, lead programmer, and maintainer (BDFL) of [Redox OS](https://github.com/redox-os). I...

                Comment


                • #9
                  Originally posted by c117152 View Post
                  The overall design seems right (the best off plan9 & minix) but targeting the x86 is already costing them in kernel purity with no real good reason other then the "Runs on Real Hardware" sticker... Well, whatever. Best of luck I guess.
                  I'm not sure why you think that. I'd say a good reason to support that architecture is because most desktops support x86. What would you expect them to support?

                  In addition, you can just add other architectures at a later time. I would put money on the design allowing and expecting that to happen.

                  Comment


                  • #10
                    If you want to support the long-term development, you can fund this project on Patreon. One developer quit his job in order to work as a full-time developer on Redox.

                    https://www.patreon.com/redox_os

                    Comment

                    Working...
                    X