Announcement

Collapse
No announcement yet.

Two Decades Late: Mainline Linux Kernel Getting Keyboard / Mouse Driver For SGI Octane

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

  • Two Decades Late: Mainline Linux Kernel Getting Keyboard / Mouse Driver For SGI Octane

    Phoronix: Two Decades Late: Mainline Linux Kernel Getting Keyboard / Mouse Driver For SGI Octane

    The MIPS-based SGI Octane IRIX workstations were first introduced in the late 90's while recently there has been a resurgence in the work on getting these vintage PCs running off a mainline Linux kernel...

    http://www.phoronix.com/scan.php?pag...eyboard-Driver

  • #2
    IMHO - Not Bloat.

    People scratching their itches and getting their skills up is not a bad thing. Hardware enablement is always a good thing.

    It might also lead to more restoration efforts for this bit of kit if it is easy to get working.

    Comment


    • #3
      Working code for systems being utilised is hardly bloat.

      Interestingly, how enabled are these machines currently? Are they producing anything?

      Comment


      • #4
        Originally posted by stiiixy View Post
        Working code for systems being utilised is hardly bloat.
        I don't think these systems are being utilized. They are on scrapyards and museums.

        Comment


        • #5
          Those are fascinating machines, from a era before X86 had totally dominated everything:



          Comment


          • #6
            But the entire appeal of those is that proprietary system and those super interesting tools, like 3D file browser. Why would you want to run linux on it?
            I think they are super cool but only with that system on them, not just linux like on any other machine.

            Comment


            • #7
              Originally posted by uid313 View Post
              I don't think these systems are being utilized. They are on scrapyards and museums.
              Yes, exactly. We have two in our basement, one is not working anymore at all. I do concider this bloat, because the software used to run on these systems only works on IRIX and has no real use anymore.

              Comment


              • #8
                For a code patch to be committed does it not need to be tested thoroughly? Surely not too many people have these devices still to do the testing?

                I have a broken Indy in the attic but even when they were modern they were fairly rare. They were almost as awkward to get hold of back then as a server-grade ARM board is today.

                I did like it with IRIX though. IrisGL was cool and the slightly quirky Motif implementation looked great. I can't believe there are no IRIX capable MIPS emulators around.

                Comment


                • #9
                  Is this a joke? by now do we even need this in the kernel at all? Is there anyone actually using this specific 2 decade old keyboard?

                  Comment


                  • #10
                    Originally posted by rabcor View Post
                    Is this a joke? by now do we even need this in the kernel at all? Is there anyone actually using this specific 2 decade old keyboard?
                    At least the developer is using it, and having the code in the mainline tree protects it from bitrot (the kernel doesn't have an internal API, refactorings are common).
                    I know you didn't specifically mentioned it, but a lot of people consider supporting old hardware that almost nobody uses "bloatware", but it's far from it. Sure, the kernel source code base gets larger, but only the people with the hardware will ever compile/run its specific code. In other words, you're not getting MIPS64 code on your x86-64 machine, nor even selecting1 drivers, at compile time, for hardware which can't be installed on a x86-64 machine.

                    1 I lied, sometimes it's useful, to maximise compiler exposure.

                    Comment

                    Working...
                    X