Re: [Altosdev-kernel] Eh...

[Benjamin J. <bsj...@mi...>]

I agree.  I believe we should just get some fundamental ideas down about
what the basic implementation details of the kernel will be, and then
worry about the user later...we should still keep the user in mind while
we do it, but like Kernel Panic says, if we don't get started soon most o=
f
us will become bored and lose interest.

As for me, I just finished an OS class for my school where we had to
implement a kernel basically from scratch...since it was just in groups o=
f
2 in only 9 weeks, it wasn't anything great but we still had to do proces=
s
scheduling, memory management (physical and virtual), serial I/O, and
system calls like fork, exec, wait, getpid, ...

I'm ready to do some hardcore c programming (or whatever language).  I
have finals this week like I said earlier but those are over on friday.

If we need more programmers, I can probably get some but until we know
that we definitely do, I will hold off from asking them.

See ya,

Raz0rSharp

Kernel Panic wrote:

> Hello, Alexey.
>
> On Tuesday, March 12, 2002 at 9:46:26 PM you wrote:
>
> AM> No one answer about GUI organization. I wrote it several days ago.
> It's not in the Docs section. Do you mean you posted it here? ..Well,
> anyway, GUI will come later.. We first have to develop a kernel.. I've
> started many-many weeks ago.. I'm having minor problems, but that
> doesn't matter much..
>
> I mean, we can spend years on just thinking on it (as it is with
> another project I'm involved in now; not years.. but long enough to be
> irritating..).
>
> What about actually writing? Coding stuff? You know.. Compiling..
> Linking.. etc =3D) .
>
> The general idea is to get a simple kernel running now.. (wait, we
> have to write it yet..).. I've done a simple kernel with memory
> management routines (malloc, free..) and a console, but am now stuck
> with djgpp.. LD doesn't link the kernel as it should.. there appear to
> be overlapping segments there.. Well, that doesn't matter much, I hope
> I'll cope with that.. But I'm waiting for input from others (it's an
> ELF file built under DJGPP with GCC 3.0.4 and bintuils 2.11.2; it
> builds ok under FreeBSD 4.2)..
>
> You know, developing kernel is not just designing a streams model or
> something.. We should actually start writing!! Or it'll get boring
> very soon..
>
> I guess, a general approach will be keeping everything not very
> complicated.. Rather, simple.. We have a strategy (an OS with streams
> as an IPC method), now we have to develop the tactics. My approach now
> is doing everything as fast as possible, because if we concentrate on
> a certain part too much, we may get stuck in it and never go further.
> That means, eveything should be designed well, but not necessarilly
> implemented that well. We should get the system working, and only then
> think about some optimizations and stuff..
>
> Also, we should know who will write what.. At the moment, we have only
> heard a couple of developers here, out of 7 listed at our project's
> site. I would like to know what the others' skills are, and what they
> can do for the project. We have to somehow distribute our efforts, you
> know..
>
> I would love to write the OS-dependant stuff, but I can't do the whole
> thing on my own.. So.. anyone willing to help me? Well, anyway, kind
> of resum=E9s of all the members are welcome..
>
> Sorry for taking kind of administrating voice =3D) .. but we may be jus=
t
> discussing our great model of the OS forever, without writing..
>
> --
> Regards,
>  Kernel Panic                          mailto:rzh...@he...
>
> _______________________________________________
> altOSdev-kernel mailing list
> alt...@li...
> https://lists.sourceforge.net/lists/listinfo/altosdev-kernel