Re: Tasks list for GNU Mach

[Marcus Brinkmann]

At Thu, 22 Dec 2005 13:45:05 +0100,
<olafBuddenhagen@gmx.net> wrote:
> A couple of years ago, Marcus (and some others), more and more exposed
> to the problems in Mach, made the claim that Hurd/Mach has no future,
> and further developement should be directed to a port to a modern
> microkernel like L4. This wasn't really surprising, as today it's widely
> known that Mach in it's existing form is severly flawed.

Credit where credit is due: It was Neal Walfield who started to
seriously analyze and explain Mach's architectural problems while I
was still porting packages and fixing bugs and thought that we only
need a bit more cleanup to get the boat a-rockin'.

> This, or maybe just unrelated loss of interest, made any work on GNUMach
> practically stop for the last few years. So additionally to the design
> flaws, it was also getting severly outdated.

In fairness, Mach was never worked on much by anybody (in public)
after its last release.  The substantial things that have been done in
the past were: Linux driver framework update, page mechanism fine
tuning, then Roland's OSKit work, followed by my work on the I/O port
protection stuff.  Also, I have worked on memory object wrappers which
are needed for security reasons, but this patch is not in (although it
should be), but that's just a small thing, as I never aimed for a
complete implementation.

I have done, at some point, an eyeball comparison of GNU Mach against
xMach (which was supposed to be the core for a BSD-based project), and
found no significant differences.  In fact, GNU Mach had a couple of
minor bug fixes that xMach didn't have.  We can only conclude that our
Mach is as (not) up-to-date as anybody else's (except internal IBM,
Apple and Microsoft stuff).

The rest of your mail contains a quite subjective view on the events.
My view has been expressed elsewhere and discussed at length, and is
not really important here, so I won't reply on the individual points.

I do agree that there is room for substantial improvements, and I
think that justifies the motivation behind any revival efforts, for
some type of cost-benefit calculation.

I might be misremembering, but I believe we had this running gag that
whoever started to work on Mach would soon drop from the earth and
never be seen again (in the Hurd project).  May a luckier star shine
on the brave people who try to attack this beast this time around :)

Thanks,
Marcus