Announcement

Collapse
No announcement yet.

Linux 5.7 Positioned To Retire ARM 32-bit KVM Virtualization Support

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

  • #11
    Originally posted by set135 View Post

    Internal changes are fine; the kernel has no fixed internal or driver ABI. Userspace breakage is verboten...
    Thanks for the condescending reply, but Linux's development model in the past did allow for userspace breakage.

    e.g. devsfs, ipfwadm and ipchains are no more, and were user-facing.

    Comment


    • #12
      I'm a bit surprised this is dropped so soon. 32 bit ARM chips are still commonly used today. They are on a steady decline but still common enough to remain relevant.
      Regardless, I don't especially care for my own sake.

      Comment


      • #13
        This is unrelated, but when does Michael publish a wrap-up article about kernel 5.7 features?

        Comment


        • #14
          Originally posted by Buntolo View Post
          This is unrelated, but when does Michael publish a wrap-up article about kernel 5.7 features?
          When the 5.7 merge window is about to close / 5.7-rc1, but usually before 5.6 is released I'll have a sort of 'preview' of the stuff I've noticed so far.
          Michael Larabel
          http://www.michaellarabel.com/

          Comment


          • #15
            The development (driver support) for many 32bit ARM devices have not even been finished and there are already plans to remove support.

            It's a shame, but I'm not surprised seeing as most ARM devices' lifetime are around 2 to 5 years.

            In my professional capacity I'm going to have to not update past Linux 5.7 for some integration tests VMs. By the time 5.7+ becomes LTS I'll hopefully have dropped support and replaced the devices. I hate even the idea of running legacy software.

            In my personal capacity I don't mind to tinker and mess around with compiler toolchains and custom kernels when I have time.

            I don't blame kernel devs, I blame short-sightedness of consumers and companies like raspberry pi foundation.

            Comment


            • #16
              Originally posted by Terrablit View Post

              Honestly, with those types of memory constraints, containers make more sense than VMs in the first place. Despite the feature being there, there's no consumer/prosumer VM images for ARM. Nobody's built the kind of images that would run and are useful. Most of it is for developers, and most who are developing on ARM either use the hardware they need, or they use qemu on a desktop.
              There actually are a bunch of VM images available for ARM. For example for FreeBSD they have qcow2, raw, vhd and vmdk VM images for arm64 aarch64. Here. Not sure anyone uses them, but they are available.

              Comment


              • #17
                Originally posted by AndyChow View Post

                There actually are a bunch of VM images available for ARM. For example for FreeBSD they have qcow2, raw, vhd and vmdk VM images for arm64 aarch64. Here. Not sure anyone uses them, but they are available.
                Those are 64-bit ARM images (aarch64). This post is about retiring KVM support on 32-bit ARM. 64-bit ARM is seeing a lot of use in server scenarios, but 32-bit ARM saw very little beyond the bespoke images for individual hardware and the various distros. With the combination of the hardware being more difficult to work with and consumer devices only offering limited RAM (rarely 4GB, but usually 512MB-2GB), virtualization still isn't really all that attractive on 32-bit ARM without a corporate sponsor. It's a technical loss, sure, but practically speaking, almost no end users were using it.

                Technically, 32-bit ARM had LPAE to give it 1TB of address space, but there wasn't really much hardware that used it.

                Comment

                Working...
                X