Re: Roadmap?

[Farid Hajji]

> >    It's not yet clear wether we should stick to Mach's device API
> >    or if the oskit-mach people are considering a totally new approach.
> >    Basically, the drivers must be implemented in user-space (L4 API
> >    sends INTs to driver threads through IPC, much like in Mach).
> 
> The Mach device interface is open, read, write, close, I don't see how you
> can have anything else.  The details are almost insignificant, it's trivial
> to go from one interface to another.  With one exception, and that is the
> terminal interface.  Luckily, the braindead Mach terminal interface is
> completely encapsulated in term/device.c (or something like that), and you
> just need to rewrite this file if you use a saner semantics.

Right.

I didn't mean the Mach <-> user-space dev_*() interface you mentioned
above (that is easy), but rather the (future) interface between L4
(and Mach?) and the user-space drivers. Things like passing INTs to
the drivers or clearing them, locking w. cli/sti etc...

> I suggest to work on putting OSKit drivers into user space now, and try to
> get it working with Mach and/or L4.  Waiting for L4Env is probably not so
> good an idea before we know more about it (it might after all never really
> mature, while OSKit has a lot of drivers already and works quite well).

That is a very good idea indeed. We could start with a few essential
drivers like: serial, keyboard, vga, and probably atapi/disk. Any takers?

-Farid.

-- 
Farid Hajji -- Unix Systems and Network Admin | Phone: +49-2131-67-555
Broicherdorfstr. 83, D-41564 Kaarst, Germany  | address@hidden
- - - - - - - - - - - - - - - - - - - - - - - + - - - - - - - - - - - -
One OS To Rule Them All And In The Darkness Bind Them... --Bill Gates.