1. Forum
  2. Usenet
  3. COMP.OS.VMS

  • Microkernels

    From Simon Clubley@21:1/5 to Lawrence D'Oliveiro on Mon Jan 8 13:57:09 2024
    On 2024-01-05, Lawrence D'Oliveiro <ldo@nz.invalid> wrote:

    That can be blamed on the limitations of Mach. People still seem to think microkernels are somehow a good idea, but they really don?t help much, do they?


    The seL4 and QNX users would probably disagree with you. Strongly.

    Microkernels offer excellent security and isolation benefits that
    monolithic kernels cannot provide.

    It really is a pity Linux didn't go down the microkernel route. I suspect
    many of the kernel-level vulnerabilities would be far less severe in such
    an environment.

    In case you are unaware BTW, the monolithic versus microkernel design was
    a _major_ debating point in the early days of Linux.

    Simon.

    --
    Simon Clubley, clubley@remove_me.eisner.decus.org-Earth.UFP
    Walking destinations on a map are further away than they appear.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From =?UTF-8?Q?Arne_Vajh=C3=B8j?=@21:1/5 to Simon Clubley on Mon Jan 8 09:48:00 2024
    On 1/8/2024 8:57 AM, Simon Clubley wrote:
    Microkernels offer excellent security and isolation benefits that
    monolithic kernels cannot provide.

    That is sort of by definition.

    And there are use cases where that requirement is
    everything.

    But for traditional server and desktop it seems like
    either monolith or hybrid is the preference.

    Key question must be whether that is just for historic
    reasons or there is a technical reason - aka that a
    pure microkernel design does not perform good enough.

    Arne

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Lawrence D'Oliveiro@21:1/5 to Simon Clubley on Tue Jan 9 01:31:55 2024
    On Mon, 8 Jan 2024 13:57:09 -0000 (UTC), Simon Clubley wrote:

    It really is a pity Linux didn't go down the microkernel route.

    Andy Tanenbaum already had an argument with Linus Torvalds about that, and lost.

    Today, Tanenbaum has (unsuccessfully) been trying to commercialize MINIX,
    while Linux essentially dominates the computing world.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Scott Dorsey@21:1/5 to ldo@nz.invalid on Tue Jan 9 14:54:30 2024
    In article <uni7mb$1o066$1@dont-email.me>,
    Lawrence D'Oliveiro <ldo@nz.invalid> wrote:
    On Mon, 8 Jan 2024 13:57:09 -0000 (UTC), Simon Clubley wrote:

    It really is a pity Linux didn't go down the microkernel route.

    Andy Tanenbaum already had an argument with Linus Torvalds about that, and >lost.

    Today, Tanenbaum has (unsuccessfully) been trying to commercialize MINIX, >while Linux essentially dominates the computing world.

    Although at the same time QNX continues plugging along with a small but dedicated customer base.
    --scott
    --
    "C'est un Nagra. C'est suisse, et tres, tres precis."

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Lawrence D'Oliveiro@21:1/5 to Scott Dorsey on Tue Jan 9 20:11:32 2024
    On 9 Jan 2024 14:54:30 -0000, Scott Dorsey wrote:

    In article <uni7mb$1o066$1@dont-email.me>,
    Lawrence D'Oliveiro <ldo@nz.invalid> wrote:

    Today, Tanenbaum has (unsuccessfully) been trying to commercialize
    MINIX, while Linux essentially dominates the computing world.

    Although at the same time QNX continues plugging along with a small but dedicated customer base.

    So, would you say that microkernels have a better than 50% success rate,
    or not?

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)