Subject: Re: Real Time O.S vs 'conventional' one.
To: Marcin Jessa <email@example.com>
From: Stefan 'Kaishakunin' Schumacher <firstname.lastname@example.org>
Date: 10/26/2005 13:17:52
Content-Type: text/plain; charset=iso-8859-15
Also sprach Marcin Jessa (email@example.com)
> I have embedded boards which I want to use as acess points for WLAN
> The choice is between NetBSD and Real Time Linux.
> What I am wondering about is if there is any advantage using real
> time O.S on wireless enviroment.=20
A RTOS is an OS designed to meet time constraints and optimized
prediction of process behaviour. It is not necessarily designed to be
as fast as possible.=20
> Would one gain anything speedwise using RTOS ?
That can't usually be predicted. An RTOS doesn't need to be faster
than a normal one, it just has to meet it's timing constraints.=20
BTW: I had to use RT-Linux in a course on RT programming (scheduling,
semaphores, Dining philosophers problem and so on) and learned that
RTLinux is just a bunch of wild hacks to get some kind of RT into the
Kernel. It was running on a normal Linuxkernel and therefor, it wasn't a
real hard RTOS. As what I know from electrical engineering, all hard
RTOS are specifically designed for a task or device. Bringing hard
RTOS to a Unix is IMHO almost unsolvable.=20
So my advice is NetBSD, don't forget, it holds the Internet2 Land
Speed Record :-)
 at least in our Lab
PGP FPR: CF74 D5F2 4871 3E5C FFFE 0130 11F4 C41E B3FB AE33
Der Geist des Kriegers sollte mit Beginn des Neujahrstages bis zum Ende=20
des Jahres vom Gedanken an seinen Tod beherrscht werden.
Daijouji Shigesuke in "Budo Shoshin Shuu"
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.2 (NetBSD)
-----END PGP SIGNATURE-----