Announcement

Collapse
No announcement yet.

Redox: A Rust-Written, Microkernel Open-Source OS

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

  • #21
    To me it looks more like a hobby/toy OS (nothing wrong with that btw) riding the Rust hype wave.

    Comment


    • #22
      One rather nice microkernel feature is filesystems in userspace, I hope they specifically implement the FUSE api, as that would give them a bunch of filesystems for free, including zfs, which they say they want to support, and sshfs, which is obligatory.

      I also hope they take some inspiration from Plan9, which in many ways was unix as it should have been.

      Btw, redox sounds like a chemical reaction — one that creates rust
      Last edited by andreano; 20 March 2016, 05:26 PM.

      Comment


      • #23
        Originally posted by zanny View Post

        That page is actually painful to read. They are actually trying to claim that proprietary forks that bleed the project dry are beneficial to end users, when exactly the opposite is true. Violating users software freedoms never gives them a better product.
        The end user could be a corporation. Maybe the authors themselves want to profit using the code? MIT/BSD provides the authors the highest available morale. They're morally superior compared to GPL zealots. They're giving away everything like Jesus Christ. This is the true form of love. GPL locks down the commercial users badly.

        Comment


        • #24
          Originally posted by log0 View Post
          the Rust hype wave.
          It is the best attempt at killing C/C++ I've seen so far. Most other candidates (e.g. Go) are too high level to rival C/C++ in their own game. I've seen that Go is actually used some places (e.g. Docker is written in Go), and I suppose Rust will become bigger.

          I am more doubtful about Redox for its choice of license, as others have pointed out.
          Last edited by andreano; 20 March 2016, 06:45 PM.

          Comment


          • #25
            Originally posted by andreano View Post
            Btw, redox sounds like a chemical reaction — one that creates rust
            In the Soviet Union Rust creates Redox... That makes Redox a truly capitalist operating system!

            Comment


            • #26
              Originally posted by pal666 View Post
              i.e. it will be as successfull as hurd
              Hurd isn't successful because it has no goals, is over 15 years old. and still ended up with shoddy design.
              There's nothing wrong with the concept of a microkernel... it actually has quite a few advantages over monolithic kernels. The main downside of the micro kernel from my understanding is the latency introduced due to the IPC between user space and kernel space (which exists in monolithic kernels as well if you ever write a user space USB driver for instance in Linux).

              Comment


              • #27
                Originally posted by Shimon View Post
                Redox is important for showing Rust off as a systems programming language. It's not going to try replacing Linux any time soon and its developers are sane people who never put such claims forward.
                No, they are not entirely sane when it comes to Linux. Don't read their FAQ, they are anti-Linux nuts.

                Comment


                • #28
                  Originally posted by carewolf View Post
                  No, they are not entirely sane when it comes to Linux. Don't read their FAQ, they are anti-Linux nuts.
                  "Will this replace Linux? No." <- right there in the FAQ.

                  I don't see why it's 'nuts' for them to spend much of the FAQ to explaining the problems they hope to solve. Step One of any project is "why will people want this over what's already there?".

                  Since we already have a couple of good open OSs, the rationale for a new one is obviously going to be a list of things they think are wrong with Linux or BSD and hope to improve on.
                  Last edited by FLHerne; 20 March 2016, 09:27 PM.

                  Comment


                  • #29
                    Originally posted by computerquip View Post

                    Hurd isn't successful because it has no goals, is over 15 years old. and still ended up with shoddy design.
                    There's nothing wrong with the concept of a microkernel... it actually has quite a few advantages over monolithic kernels. The main downside of the micro kernel from my understanding is the latency introduced due to the IPC between user space and kernel space (which exists in monolithic kernels as well if you ever write a user space USB driver for instance in Linux).
                    IIRC not kernel ⇔ userspace but userspace A ⇔ userspace B is the problem. In a mode switch (kernel ⇔ user) there are no TLB flushes, which are expensive, but when you remap the address space (context switch), there are. And with the ring-based security model of the x86 MMU, there is basically no way around giving every process its own address space to ensure isolation.

                    Comment


                    • #30
                      Originally posted by computerquip View Post
                      The main downside of the micro kernel from my understanding is the latency introduced due to the IPC between user space and kernel space (which exists in monolithic kernels as well if you ever write a user space USB driver for instance in Linux).
                      the main downside is not latency, but human effort needed to do ipc for everything
                      Last edited by pal666; 21 March 2016, 06:29 AM.

                      Comment

                      Working...
                      X