Re: Hurd general design question

[olafBuddenhagen]

Hi,

On Fri, Jul 25, 2008 at 11:45:36AM -0700, Power, Mike wrote:

> I had an observation; it seems that your support for native C++ is a
> limiting factor to the Hurd.
[...]
> My general design question, what if you moved C++ to emulation?  The
> native language would then be a programming language that did not have
> random access to the entire memory space.

I'm not sure what you are proposing here. Is it to write the operating
system in some safe language, and run existing C (and other) software
only in an emulator?

As for writing an operating system in a safe language, that is certainly
an option, but not a panacea. Not using the hardware-provided protection
mechanisms indeed safes the overhead of context switches; but doing the
protection in software means *everything* runs slower in general -- in
most situations it's not a net win.

Anyways, there is a lot of existing research in this area -- it's not
like you are suggesting anything new here...

As for running existing software in an emulation, this is totally out of
question for the Hurd. The Hurd, as the kernel of the GNU system, *must*
be able to run existing GNU and UNIX software well, not only as some
kind of second class citizen.

> I could go on with some other stuff, but I am not sure if I am crazy
> or if this is even the right audience.

Maybe the right audience; but I think it should be obvious that
esentially suggesting to throw everything away and work on something
completely different instead, is not likely to meet applause on a
project's developer mailing list -- even if it *was* a superior design
you are suggesting :-)

In any case, before making such suggestions, it would help a lot to get
familiar with the existing research in the area first...

-antrik-