[linux-elitists] Freeing parts of Unix?

Aaron Sherman ajs@ajs.com
Fri Jun 15 08:28:29 PDT 2001


On Thu, Jun 14, 2001 at 03:17:05PM -0700, Aaron Lehmann wrote:
> On Thu, Jun 14, 2001 at 09:34:41AM -0400, Aaron Sherman wrote:
> > Go for it. No one has ever succeded in creating such a system that is
> > not painfully inefficient, but I welcome you to try.
> 
> Exokernels are amazing. http://www.pdos.lcs.mit.edu/exo/

MIT microkernel research has been getting some good press, but I've
yet to have the time to check them out.

No matter what, the end-result of all of this debate and research is
going to be better free/open source software, and that I'll take any
day.

> > Will we see ksmb, kbind, ksmtp and kgnome?
> 
> Look at exokernels instead. With that model, you move
> hardware-sensitive kernel code into libraries which the applications
> use, as I understand it.

The difference, then, is permission. Do you let the application
authenticate itself in some interesting way (always expensive, but
perhaps mitigatable) and then access hardware as it pleases? Or do you 
put the small amount of hardware-aware code into kernel space? Either
way, with loadable modules or shared libraries, you get pretty much
the same performance and footprint (after application startup, for
which shared libs will be more expensive). The key question is which
one offers more options to your developers?

I would think the library approach would be harder to get safe
and correct, but a bit more of a win for developers in the long run.

Then again, UNIX would likely not be the best model for such a
system. You'd probably want to make it more VMS-like, to get the
permissions per-installed-file and granular permissions concepts.

Interesting, though.

-- 
Aaron Sherman		
ajs@ajs.com		finger ajskey@b5.ajs.com for GPG info. Fingerprint:
www.ajs.com/~ajs	6DC1 F67A B9FB 2FBA D04C  619E FC35 5713 2676 CEAF
	"I want a drug store in The Twilight Zone." -Shriekback



More information about the linux-elitists mailing list