6 ms·

> Fixed luna88k MULTIPROCESSOR kernels booting with CPU modules installed in arbitrary slots. An OS release capable of running on the latest hardware in 2022 i

> Fixed luna88k MULTIPROCESSOR kernels booting with CPU modules installed in arbitrary slots. An OS release capable of running on the latest hardware in 2022 is still making optimizations for a niche system from the late 80s/early 90s. There's something very beautiful about that.

jrumbut 4y ago
I searched for this luna88k and was taken to a Wikipedia page that I read until I realized it was for a computer called LUNA but it had a 68k processor. The reason it came up was a tiny quote from the NetBSD announcement of Luna 68k support which mentions its "rare cousin" the 88k LUNA. Now that's niche!
- reaperducer 4y ago
  I thought I knew a lot about 68k processors, but "88k" is new to me. It turns out to be the Motorola 88100¹, which was used in Unisys servers running Unix. ¹https://en.wikipedia.org/wiki/Motorola_88100 https://en.wikipedia.org/wiki/Motorola_88100
  - kalleboo 4y ago
    I only know about the 88k since Apple prototyped using it for their first RISC Mac back in the day before pushing for the PowerPC alliance
    - lukeh 4y ago
      NeXT as well I believe!
      - jkepler 4y ago
        M'y physics prof loved his NeXT machine back in the day.
  - cafard 4y ago
    Data General built a machine called the Aviion with the 88K.
  - johndoe0815 4y ago
    Motorola had a lot of VME bus boards with 88100/88110 CPUs. There's still a web site dedicated to the 88k CPUs at http://www.m88k.com http://www.m88k.com Fun fact - the 88k was designed by Mitch Alsup, who also designed SPARC CPUs for Ross, x86s for AMD and others. He is retired, but still active in the comp.arch Usenet newsgroup.
- SoftTalker 4y ago
  Also a good example of what I understand the OpenBSD development goals to be: developers work on the features and hardware support that interest them. It isn't a popularity contest; targeting the most popular platforms or growing the number of users of OpenBSD is not a project goal. So as long as there are developers who want to do the work to support the 88k LUNA (whatever it is) then it will happen. More specifically: https://www.openbsd.org/goals.html https://www.openbsd.org/goals.html
r12343a_19 4y ago
Yet, they dropped support for i386DX and some older (386) Cyrix?
- tremon 4y ago
  IIRC, the 386 did not have atomic load/store operations (XADD, CMPXCHG), so it's much harder to write reliable semaphores/mutexes on that platform. I'm surprised OpenBSD supported it for this long. (edit: it seems they didn't: OpenBSD/i386 hasn't actually supported running on [386sx/386dx] for some time. )
  - xxpor 4y ago
    Is there an SMP version of the 386? You may need locks on a non-SMP system (since you can still have concurrency), but without parallelism you don't need atomics... right?
    - tremon 4y ago
      The 386 still has hardware and software interrupts which might just happen between the load and store instructions, so it's not as simple as "we don't need atomics". If you were writing code specifically for the 386, I suppose you could implement all semaphores as critical sections to guard against external interrupts; I see no reason why that wouldn't work. But realistically, that code would be more of a museum piece than something you'd want to maintain in a 2022 operating system.
      - xxpor 4y ago
        Ah interrupts, of course. /me goes back to his polling-only application :) (we have an interrupt, but it's "the watchdog fired and you're about to die anyway", so not exactly complex to handle...)
    - my123 4y ago
      https://devblogs.microsoft.com/oldnewthing/20190404-00/?p=102383 https://devblogs.microsoft.com/oldnewthing/20190404-00/?p=10...
badsectoracula 4y ago
This is something i always felt was lost with what seems to be "commercialization of free software" - software doing things because it was technically possible regardless if that made commercial sense. Nowadays it seems most projects have commercial interests (or even if they don't they act like it) and processes that come against that, often ending up abandoning things because of usage metrics and agendas, as if they have some board of directors to please.
amatecha 4y ago
Yeah it's one of the many reasons I use OpenBSD - I can actually use my old, perfectly-functional hardware from literally decades ago and still run the latest software with latest security features and frequent updates. On numerous architectures, too!
mekster 4y ago
Or misallocation of limited resource. Who is funding OpenBSD these days?
- ben_bai 4y ago
  Development is done by volunteers, who of course can focus on their passion (i.e. https://twitter.com/ao_kenji https://twitter.com/ao_kenji). As for funding essentials like electricity, network, hackathons, and unpleasant development work, that would be the OpenBSD foundation (http://www.openbsdfoundation.org/activities.html http://www.openbsdfoundation.org/activities.html)
  - mekster 4y ago
    It's quite a question as how a team of very talented developers can keep on going with such little donation. $87.5k in 2022? That doesn't even pay for a single person after taking the rest of the expenses into account. Do they all have other jobs and using their spare time to devote to such a serious project?
    - ben_bai 4y ago
      > Do they all have other jobs and using their spare time to devote to such a serious project? Yes, except when <daywork> is also OpenBSD related. But that has nothing to do with the foundation. > $87.5k in 2022? The year isn't over... and there should be some left from bygone years. > That doesn't even pay for a single person after taking the rest of the expenses into account. I'm not aware of a single full-time paid developer, by the OpenBSD foundation. It's more a gig thing. In addition for Hackathons the foundation pays for the venue and network, and the developers usually fly themselves in. The foundation was founded b.c. the CD sales (back when there were still CDs sold) didn't even cover electricity and network costs of the main site anymore. It's all done like in the good old days of Linux, before big business stepped in.

jrumbut 4y ago

I searched for this luna88k and was taken to a Wikipedia page that I read until I realized it was for a computer called LUNA but it had a 68k processor. The reason it came up was a tiny quote from the NetBSD announcement of Luna 68k support which mentions its "rare cousin" the 88k LUNA. Now that's niche!

reaperducer 4y ago
I thought I knew a lot about 68k processors, but "88k" is new to me. It turns out to be the Motorola 88100¹, which was used in Unisys servers running Unix. ¹https://en.wikipedia.org/wiki/Motorola_88100 https://en.wikipedia.org/wiki/Motorola_88100
- kalleboo 4y ago
  I only know about the 88k since Apple prototyped using it for their first RISC Mac back in the day before pushing for the PowerPC alliance
  - lukeh 4y ago
    NeXT as well I believe!
    - jkepler 4y ago
      M'y physics prof loved his NeXT machine back in the day.
- cafard 4y ago
  Data General built a machine called the Aviion with the 88K.
- johndoe0815 4y ago
  Motorola had a lot of VME bus boards with 88100/88110 CPUs. There's still a web site dedicated to the 88k CPUs at http://www.m88k.com http://www.m88k.com Fun fact - the 88k was designed by Mitch Alsup, who also designed SPARC CPUs for Ross, x86s for AMD and others. He is retired, but still active in the comp.arch Usenet newsgroup.
SoftTalker 4y ago
Also a good example of what I understand the OpenBSD development goals to be: developers work on the features and hardware support that interest them. It isn't a popularity contest; targeting the most popular platforms or growing the number of users of OpenBSD is not a project goal. So as long as there are developers who want to do the work to support the 88k LUNA (whatever it is) then it will happen. More specifically: https://www.openbsd.org/goals.html https://www.openbsd.org/goals.html

r12343a_19 4y ago

Yet, they dropped support for i386DX and some older (386) Cyrix?

tremon 4y ago
IIRC, the 386 did not have atomic load/store operations (XADD, CMPXCHG), so it's much harder to write reliable semaphores/mutexes on that platform. I'm surprised OpenBSD supported it for this long. (edit: it seems they didn't: OpenBSD/i386 hasn't actually supported running on [386sx/386dx] for some time. )
- xxpor 4y ago
  Is there an SMP version of the 386? You may need locks on a non-SMP system (since you can still have concurrency), but without parallelism you don't need atomics... right?
  - tremon 4y ago
    The 386 still has hardware and software interrupts which might just happen between the load and store instructions, so it's not as simple as "we don't need atomics". If you were writing code specifically for the 386, I suppose you could implement all semaphores as critical sections to guard against external interrupts; I see no reason why that wouldn't work. But realistically, that code would be more of a museum piece than something you'd want to maintain in a 2022 operating system.
    - xxpor 4y ago
      Ah interrupts, of course. /me goes back to his polling-only application :) (we have an interrupt, but it's "the watchdog fired and you're about to die anyway", so not exactly complex to handle...)
  - my123 4y ago
    https://devblogs.microsoft.com/oldnewthing/20190404-00/?p=102383 https://devblogs.microsoft.com/oldnewthing/20190404-00/?p=10...

badsectoracula 4y ago

This is something i always felt was lost with what seems to be "commercialization of free software" - software doing things because it was technically possible regardless if that made commercial sense. Nowadays it seems most projects have commercial interests (or even if they don't they act like it) and processes that come against that, often ending up abandoning things because of usage metrics and agendas, as if they have some board of directors to please.

amatecha 4y ago

Yeah it's one of the many reasons I use OpenBSD - I can actually use my old, perfectly-functional hardware from literally decades ago and still run the latest software with latest security features and frequent updates. On numerous architectures, too!

mekster 4y ago

Or misallocation of limited resource. Who is funding OpenBSD these days?

ben_bai 4y ago
Development is done by volunteers, who of course can focus on their passion (i.e. https://twitter.com/ao_kenji https://twitter.com/ao_kenji). As for funding essentials like electricity, network, hackathons, and unpleasant development work, that would be the OpenBSD foundation (http://www.openbsdfoundation.org/activities.html http://www.openbsdfoundation.org/activities.html)
- mekster 4y ago
  It's quite a question as how a team of very talented developers can keep on going with such little donation. $87.5k in 2022? That doesn't even pay for a single person after taking the rest of the expenses into account. Do they all have other jobs and using their spare time to devote to such a serious project?
  - ben_bai 4y ago
    > Do they all have other jobs and using their spare time to devote to such a serious project? Yes, except when <daywork> is also OpenBSD related. But that has nothing to do with the foundation. > $87.5k in 2022? The year isn't over... and there should be some left from bygone years. > That doesn't even pay for a single person after taking the rest of the expenses into account. I'm not aware of a single full-time paid developer, by the OpenBSD foundation. It's more a gig thing. In addition for Hackathons the foundation pays for the venue and network, and the developers usually fly themselves in. The foundation was founded b.c. the CD sales (back when there were still CDs sold) didn't even cover electricity and network costs of the main site anymore. It's all done like in the good old days of Linux, before big business stepped in.