I have a uniquely soft spot for SPARC, having written and disassembled a bunch of SPARC early in my career. If this is its swan song, I'll take the moment to share some code the takes advantage of the odd (today) delay slot architecture to implement instruction picking:<p><a href="https://github.com/illumos/illumos-gate/blob/master/usr/src/uts/sparc/dtrace/dtrace_asm.s#L430" rel="nofollow">https://github.com/illumos/illumos-gate/blob/master/usr/src/...</a><p>The trick uses a branch in the delay slot of a jmp--a decidedly unusual construction. At the time I found this to be extremely clever and elegant... but apparently not so clever as to warrant a comment.
With Linux having caught up with key Solaris features in recent years (DTrace -> eBPF, Zones -> Namespaces, ZFS -> ZFS on Linux), I always thought that the main reason to use Illumos would be first-class SPARC support. With that now dropped, I'm concerned that Illumos soon become irrelevant. Are there any compelling reasons left to use Illumos, other than being something for those who just want a free Solaris alternative?
This is really sad. The world is heading to a duopoly x86 - arm. Alpha is dead, Mips is almost dead, PA-RISC is dead, POWER is too expensive and RISC-V is mostly nice to have.
One thing I've wondered (randomly) and I could be way off the mark here, but does Illumos have any kind of place at Oxide Computer? The author of the link and the CTO of Oxide both have strong links to Illumos in one way or another but on the other hand some of their team are Linux kernel developers, or is the work they are doing not at this level in the stack?
While the primary issue is likely developer time and hardware availability to test on, there are other OSs like OpenBSD which supports much newer SPARC64 hardware: <a href="https://www.openbsd.org/sparc64.html" rel="nofollow">https://www.openbsd.org/sparc64.html</a>
Why don't they just upgrade GCC to a more recent version. GCC still actively supports SPARC to this date and Rust support is also present and while not perfect, it definitely works.<p>So, while I don't really have a problem with removing SPARC support from Illumos which I wouldn't be using on SPARC systems anyway, the reasons mentioned in the document aren't convincing me at all.<p>FWIW, we still support sparc64 in Debian Ports:<p>> <a href="https://cdimage.debian.org/cdimage/ports/current/" rel="nofollow">https://cdimage.debian.org/cdimage/ports/current/</a>
A fun way to make Oracle donate a machine would be to make an official POWER port.<p>That's a lot of work and I don't see IBM making a machine available.
> Without ready access to build machines, one might consider cross compilation. Though we have some support for cross-architecture software generation in the tools, the operating system does not currently support being cross compiled in full.<p>SPARC or not SPARC, I would love to help with that!