Path: utzoo!attcan!uunet!cs.utexas.edu!tut.cis.ohio-state.edu!pt.cs.cmu.edu!andrew.cmu.edu!cb1p+
From: cb1p+@andrew.cmu.edu (Chris Beasley)
Newsgroups: comp.realtime
Subject: Re: real time operating systems for 68xxx
Message-ID: <8abRihm00WB9EVo18y@andrew.cmu.edu>
Date: 13 Jul 90 14:33:17 GMT
Organization: Mathematics, Carnegie Mellon, Pittsburgh, PA
Lines: 30


Here at RedZone robotics we have been using VxWorks with the Wind kernel
from WRS for about a year.  We have several complaints although we havn't
found anything better yet (we are looking into os9).
1) Select:  This is the single largest problem we have hd to date
with VxWorks.  Select polls instead of blocking like on Unix.
When asked to fix this WRS said it was in the works.  The fix they
came up with was more efficient polling.  This of course is not a fix.
Also, select only works with sockets and not with any old FD as on Unix.
In the first version we had pipes, stdio and other FD's worked even
though the manuals said they wouldn't.  The more efficient polling
version now lives up to the documentation and only works on sockets.
2) IO system:  The IO system is single threaded and somewhat buggy.
The shell interface is sonewhat inconsistent.  The whole system is supposed
to be multitasking and has some limited support for multiprocessing
but the IO system is single threaded and this has caused us no end of
headaches, aside from the fact that it renders the system effectively
single user.  This is a pain during the development cycle.  The shell
also uses a non reentrant parser so it's not really going to be easy
to make a multi threaded shell.  STDIN and OUT are global so the IO
system can only point one place at one time.

This discussion seems to be losing clarity so I'll stop now...
If you have more specific questions lemme know.

Beez
Employed by but not officially representing
RedZone Robotics Inc.

louie!beez@sunpitt.east.sun.com