login
Search
Andrew Miklic recently announced progress on porting the Hurd gnumach microkernel to the Alpha processor. I approached him for more details on his efforts, and he explained:
"GNU Mach is the microkernel for the HURD. The rest of the HURD _should_ be platform-independent, so that the rest of the HURD/debian packages _should_ be functional on Alpha once the microkernel is up-and-running without requiring any more than a cross-compile.
"At this point, GNU Mach on Alpha is compiling, but not booting. To get this far was not that difficult for Alpha because some really old source code from the original CMU Mach 2.5 source was available, and whatever other code that exists for x86 but not on Alpha I just "stubbed" for the time being. To get it booting, I need to go back to the files that I "stubbed" and implement them properly."
From: Andrew M. Miklic
To: help dash hurd AT gnu.org
Subject: GNU Mach on Alpha
Date: Wed, 3 Apr 2002 14:23:59 -0700
I have a very preliminary port to Alpha (it compiles but does not completely
boot)--what (if any) requirements are there for such a submission to the
gnumach CVS tree, and how does one obtain such access?
At the moment, I have a directory gnumach/alpha (including its contents;
similar to the i386 directory)--gnumach/Makefile will eventually need to be
modified, but I don't want to potentially disturb x86...
Sincerely,
Andrew Miklic
From: Neal H Walfield
Subject: Re: GNU Mach on Alpha
Date: 03 Apr 2002 16:37:23 -0500
Submit your patches with comments to bug dash hurd AT gnu.org. There, we can
review and discuss them.
hmmm..
why bother with an alpha port when their base port (x86) is not even halfway decent yet?
Re: hmmm..
Quote from the original article:
"The rest of the HURD _should_ be platform-independent, so that the rest of the HURD/debian packages _should_ be functional on Alpha once the microkernel is up-and-running without requiring any more than a cross-compile."
So in theory should only need to port the micro-kernel and everything else should work as the micro-kernel is thought to be the only major arch-specific part of GNU/Hurd.
Practically you could be right, but at this point any help to GNU/Hurd is eagerly needed. So let him do his port and please don't try to ask "why code this...." question a lot a people asks these days instead of helping some project. No offense intended.
--
I used to have a sig until the great Kahuna of FOOness
told me to dump it and use /dev/urandom instead.
L4
better question is: why bother with gnumach Alpha port when much better thing would be a L4 port ?
L4 or OsKit?
Are the Hurd guys gonna use L4? I know there was some
developer discussion at a conference two or so
years ago and there is a hurd-l4@gnu mailing list
but I hear a lot more about moving to OsKit [mach].
L4 and OsKit mach are both microkernels, were they
planning on porting the Hurd to both to prove it's
portability?
In fact, yes, as I write this it makes sense, the Hurd
is going to use OsKit as its' primary microkernel but
there will also be an L4 port. L4 is hard to find
information on. I wonder will the fsf make there own
OsKit (GNU Oskit?) or will they use the existing one
in it's unmodified form?