Re: Hurd on Minix3 or other kernels?

[Damien THEBAULT]

Le mercredi 16 novembre 2005 à 16:43 -0500, B. Douglas Hilton a écrit :
> Hi all. I'm a long-time lurker here and used to do a bit of high-level 
> porting and tinkering on the old Gnumach/Hurd. I tried to help debug the 
> OSkit/Mach kernel, mostly without much success, as unfortunately I'm not 
> really very knowledgeable about kernel level programming.
> 
> It was always my hope that Hurd would become more microkernel neutral 
> than really specified for one in particular. Look at the problems 
> getting Mach unhooked for instance.
> 
> A while back, when I was reading about the recent release of Minix3, I 
> had this probably stupid thought that porting the Hurd to the Minix3 
> kernel might be advantageous for academic purposes at least. I realize 
> that the L4/Hurd effort is sailing into some extremely advanced areas 
> and is all about fixing the performance and other problems stemming 
> mostly from the Mach legacy, and thats cool, but it looks like the 
> usability horizon for L4/Hurd is speeding away into the future.
> 
> I guess what I'm trying to get at is that I would really like to see an 
> "academic version of Hurd" with a well documented and somewhat 
> understandable kernel for laymen like myself. Mach seemed to me like it 
> was terribly convoluted, and I even purchased a book on Mach System 
> Programming in an attempt to get a handle on what was going on. I spent 
> a fair amount of time poking around in OSkit/Mach 's guts and never 
> really got to where I had much of a clue as to what was really 
> happening. I believe that the Minix3 kernel would be suitable 
> license-wise and that Dr. Tannenbaum probably wouldn't mind if somebody 
> tried to build a Hurd userland against his microkernel. In any event, 
> the Minix3 kernel could be presumed to be relatively stable, somewhat 
> recent, reasonably well supplied with hardware drivers, moderately well 
> performing, and most importantly well documented and written with 
> students of computer science in mind.
> 
> Is there any merit to my idea of trying to fork a branch of Hurd over to 
> something like Minix3 just for mainly academic purposes? I realize that 
> it wouldn't be cutting edge or anything, but I'm lucky I can still even 
> boot Gnumach/Hurd on my new(ish) system, most people won't be able to 
> even do that soon. I think a lot of interested student types could 
> potentially benefit if it were possible. I know that MIG is deeply 
> intertwined with the Hurd userland and I really don't have a clue as to 
> what would be involved in creating a MIG-like tool for the Minix3 
> microkernel, but I assume it would not be easy.
> 
> My apologies for being slightly off-topic from L4-Hurd, but since all 
> the discussion about microkernels lately, I thought I'd add my two cents 
> and see what the rest of you all thought. I know there are a lot of  
> people like myself who would like to be more involved with L4/Hurd but 
> are really in way over their heads with the current state of affairs and 
> can't really make much meaningful contribution at the present time (and 
> near future).
> 
> Cheers!
> B. Douglas Hilton
> Mechanical Engineer
> 

I think we should port Hurd to another faster microkernel than Mach too.
Some work has already be done with the L4 microkernel, and this
microkernel may be good enough to be used this way.

Some people may create a new "Hurd-ng" project with advanced features
(persistance, etc...). Many posts on this list were saying that L4 lacks
some features and can't be used as the "Hurd-ng" microkernel.

This way we could have a (fast) GNU OS quickly, and later "Hurd-ng".

-- 
Damien Thebault
public keys on http://pgp.mit.edu

signature.asc
Description: Ceci est une partie de message numériquement signée