Re: Current state of the Linux kernel on SPARC

John Paul Adrian Glaubitz <[email protected]>
Newsgroups gmane.linux.debian.ports.sparc
Message-ID <f7cb953034ff955867274a23b8cfe3ae1a8f9ce4.camel@physik.fu-berlin.de>
On Fri, 2025-08-29 at 18:41 -0400, Dennis Clarke wrote:
> I will hunt around and see what I have laying about in a back room.
> I know I have old Netra X1 stuff and maybe a few other units with very
> old UltraSPARC processors in them. They will be horrifically slow.
> The SPARC M7 or M8 will be a unicorn. Heard of only in myth.

While I don't assume you have a SPARC M7-compatible machine, I think
chances are good you might have a T1 machine. As a bonus, a test on
an UltraSPARC I or II is very welcome as well.

To clarify this:

The kernel code in question contains optimizations for different CPUs,
namely:

- UltraSPARC I
- UltraSPARC III
- Niagara 1
- Niagara 2
- Niagara 4
- SPARC M7

All but Niagara 2 had bugs which Michael Karcher fixed.

So far, I successfully tested on UltraSPARC III and Niagara 4 as I have
access to these machines. I also own a Niagara 2 machine, but there is
no testing necessary as the code was not touched.

I also tested on QEMU which emulates an UltraSPARC II, meaning that the
UltraSPARC I code was tested (there is no UltraSPARC II code). I used a
patched kernel to force Niagara 4 to use the Niagara 1 code which worked
without problems.

Still, it would be nice to have testing on real machines of these types:

- UltraSPARC I or II
- Niagara 1
- SPARC M7

Adrian

-- 
 .''`.  John Paul Adrian Glaubitz
: :' :  Debian Developer
`. `'   Physicist
  `-    GPG: 62FF 8A75 84E0 2956 9546  0006 7426 3B37 F5B5 F913
lmpx.com only provides a reader for public news (NNTP) servers. It is not affiliated with the servers or forums shown here and is not responsible for the content of articles, which is written by their respective authors.