Hello,
It is always exciting to find somebody with an interest in our
project. We appreciate your support and spreading the word about
XOmB. We very much hope that it will be useful and a success overall.
The use of D is an interesting one for many reasons. D's goal is
primarily to provide a general purpose language. In many ways this is
counter to the specific nature and goal of XOmB. C, on the other
hand, is a systems language primarily. However, C was written for
UNIX, and we consider such things a form of 'legacy' and these
entities must be abolished completely in order to facilitate a better
prospect for newer technologies. The D language inherits, pretty much
directly, the systems capabilities from C, however (inline assembly,
compatibility with the C ABI). Overall, it suits our interests very
well by providing at least a good basis for not being legacy and being
extremely useful, especially in the future when user-space programs
are executed, while allowing for all of our sweet low-level magical
stuff. It offers readability over C (mostly due to the lack of the
macro preprocessor) and if offers writeability over C (through D's
sophisticated functional template engine; metaprogramming (without C+
+) is extremely useful!). Of course, the prospect of a new language
(much like C is to UNIX) for modern systems work (multiprocessor
issues mostly) is not out of the question. We have compiler writers
considering the prospects of such a language currently.
Also, our googlecode pages are embarrassingly out of date. We moved
our trunk to github to allow for the use of git and its more powerful
branching features. Also the googlecode wiki is horribly lacking for
our needs (to have sophisticated and detailed documentation on XOmB
and the actual hardware architecture), so we got our own mediawiki set
up on our website. So, to update your linking:
Our master branch in github for XOmB:
http://github.com/untwisted/xomb/tree/master
Our wiki:
http://wiki.xomb.org/
In terms of testing, XOmB is very picky. It should, in theory, heh,
work well on any modern multiprocessor x86 environment. The
requirements right now are basically that. Dual-core x86 or better.
Extremely soon, you will need HPET support, which should be available
on all modern mid-range dual-core chips.
This may help if you would like to try out the current stable build.
Embarrassingly, it has not updated for git, but it will be within the
next 7 days. Basically you can get information on git and pulling our
repository in general web searches and off of github's site. Also
information about building a cross compiler. We can also help within
this discussion group if you have a specific problem in this process.
(it is tricky!)
http://wiki.xomb.org/index.php?title=Getting_Started
Thank you for your interest,
Dave Wilkinson II