Subject: | |
From: | |
Reply To: | |
Date: | Sat, 15 Dec 2001 01:41:31 -0500 |
Content-Type: | text/plain |
Parts/Attachments: |
|
|
I keep seeing "using the linux kernel"........is there any reason why a new
"MPE" kernel can't be written? Or is it just for convenience?
matt (definitely not a kernel writer) shade
> > Finally, some of us dreamers would like to see a new o.s. written
> > around MPE ideas, as an open source project, but using the Linux
> > kernel, perhaps with kernel modifications or leveraging HP's secure
> > Linux kernel. The hope here is that it would attract MPE vendors
> > and customers to it as it matures over the next five years, but it
> > would take *some* rewriting and recompiling to use it with existing
> > MPE applications. With MPE source and a good Pascal compiler, this
> > approach could leap ahead a lot faster, but there would still be a
> > lot of work to do, since (as I've been told) MPE is so monolithic
> > the source won't be as much help as if it were more modular.
> >
> > There is a view that all three make sense. The emulator merges
> > into the api model at some point, particularly if the approach
> > wanted an emulator for running existing MPE binaries. The new o.s.
> > approach would merge into the MPE api at some point, such as both
> > need an MPE (like) file system for Linux. It could be an evolutionary
> > thing, except the MPE api and the new o.s. approach needs the freedom
> > to break "pure" backward compatibility that is the MPE tradition.
* To join/leave the list, search archives, change list settings, *
* etc., please visit http://raven.utc.edu/archives/hp3000-l.html *
|
|
|