Re: Porting the Hurd to L4 (glibc dependencies, dropping glibc?)

[Farid Hajji]

> Please excuse if this is a trivial non-issue (but let me know so I learn 
> something :). My impression was that the translators depended upon some bits 
> in the filesystem that the Hurd sets differently (and which no other OS 
> really understands yet). Would this not cause problems porting sub-hurds in 
> other operating systems? I suppose it could be done under a loopback 
> filesystem. I haven't played with those enough to know how sturdy they are 
> when it comes to compiling or other intense use. But the Hurd would have to 
> be tucked away somewhere out of reach otherwise everytime they fsck things 
> break. I seem to remember hearing that tar did not keep track of translators 
> yet, so archiving them wouldn't solve this. 
A sub-hurd running under a host-os would have access to either a raw partition
where it could put its enhanced ext2fs/ufs if it so wishes. The same holds
true for virtual filesystems located on a file (what you mean by loopback
filesystem). I don't see any (major) problem here.

-Farid.

-- 
Farid Hajji -- Unix Systems and Network Admin | Phone: +49-2131-67-555
Broicherdorfstr. 83, D-41564 Kaarst, Germany  | farid.hajji@ob.kamp.net
- - - - - - - - - - - - - - - - - - - - - - - + - - - - - - - - - - - -
Murphy's Law fails only when you try to demonstrate it, and thus succeeds.