7 Jan
2015
7 Jan
'15
10:19 p.m.
There’s a PS2 kernel directory in one of the files. This is a 3.x based kernel and runs on
286. There are some vestiges of 8086, but I don’t know how well it will work. It looked
fairly incomplete to me…
Warner
On Jan 7, 2015, at 12:02 PM, Michael Kerpan
<mjkerpan(a)kerpan.com> wrote:
My main interest is actually in pre-386 code. I'd love to see how a Unix-like system
could be made to work on an 8086 or a 286. Did any of that code survive to the 4.2 era or
were things 32-bit only by that point?
Mike
On Jan 7, 2015 1:46 PM, "Andrzej Popielewicz"
<andrzejpopielewicz(a)gmail.com> wrote:
* Warner Losh <imp(a)bsdimp.com> [2015-01-06 11:09:58]:
On Jan 6, 2015, at 10:04 AM, Dan Cross
<crossd(a)gmail.com> wrote:
On Tue, Jan 6, 2015 at 11:48 AM, Rico Pajarola <rp(a)servium.ch> wrote:
adding the list back
On Tue, Jan 6, 2015 at 10:42 AM, Michael Kerpan <mjkerpan(a)kerpan.com> wrote:
This is a cool development. Does this code build into a working version of Coherent or is
this mainly useful to study? Either way, it should be interesting to look at the code for
a clone specifically aimed at low-end hardware.
Unknown (to me, anyway). Steve said he had intended to organize and catalog the code at
some point, but that he hasn't gotten around to it (and not to hold one's
breath). I gathered that the tar ball he provided is a snapshot of (a subset of?) the MWC
development disks at the time he was asked to create the archive. To that end, I suspect
that if one were sufficiently motivated one *could* use it to build a distribution of
COHERENT, but I suspect you'd have to know quite a bit about their internal
development practices and release processes to do so successfully; knowledge that may very
well have been lost over time. Perhaps some motivated person will be able to reverse
engineer it, though I suspect it's more useful as a case study than as working code.
Looking at the tarballs and the tarballs inside, this is a mess. It looks like it is all
there, but there???s multiple copies of things that are almost identical, RCS files that
are mostly enough, but not completely enough, etc. Plus they were using gcc 2.5.1 for
compiling things, so using a more modern compiler likely will result in
???difficulties???. There???s some docs laying around, but I haven???t read through them
all. The collection needs curating TLC...
Warner
_______________________________________________
TUHS mailing list
TUHS(a)minnie.tuhs.org
https://minnie.tuhs.org/mailman/listinfo/tuhs
Hi,
They have used also gcc-2.5.6 , which is in TUHS archives , I believe.
I have ported gcc-2.8.1,2.95.3,3.2.3 and 4.2.x. Almost 95% of kernel sources
compiles well with these newer compilers.
Andrzej
_______________________________________________
TUHS mailing list
TUHS(a)minnie.tuhs.org
https://minnie.tuhs.org/mailman/listinfo/tuhs
_______________________________________________
TUHS mailing list
TUHS(a)minnie.tuhs.org
https://minnie.tuhs.org/mailman/listinfo/tuhs