Path: utzoo!attcan!uunet!cs.utexas.edu!csd4.csd.uwm.edu!mailrus!sharkey!itivax!umich!crim.eecs.umich.edu!projoe
From: projoe@crim.eecs.umich.edu (Joseph A. Dionise)
Newsgroups: comp.realtime
Subject: vxWorks backplane driver problem
Message-ID: <234@zip.eecs.umich.edu>
Date: 18 Aug 89 04:15:19 GMT
Sender: news@zippy.eecs.umich.edu
Reply-To: projoe@crim.eecs.umich.edu.UUCP (Joseph A. Dionise)
Organization: University of Michigan EECS Dept., Ann Arbor, MI
Lines: 25

We are experiencing problems with the backplane driver for vxWorks 
(version 4.0.1).  Our setup includes an MVME-147 (25 Mhz 68030) and 
two MVME-133A's (20 Mhz 68020).  The 147 is the backplane master and 
is connected to our Ethernet.  The processors boot correctly when both 
133 are in polling mode.  However, we would like the backplane network
to be interrupt driven.  Since the 133 boards do not support mailbox 
interrupts, we have chosen VME interrupts IRQ3 & IRQ4 for the pair.  
In this configuration, the 147 boots fine, but the 133 boards hang at
various points while booting up.  Running bpShow on the 147 shows that 
their packets are simply queued up.  

Now for the strange!  If we include only a single (interrupt-driven)
133 everything works fine.  But adding the second 133 is trouble.  We 
feel that the problem lies with the 147 software.  Any help would be
appreciated.  Thanks in advance.

By the way, has anyone compiled a list of bugs for the vxWorks kernel ?

################################################################################
#  Joseph A. Dionise                                                           #
#  Robot Systems Division    Internet : projoe@crim.eecs.umich.edu             #
#  University of Michigan    uucp : {..}!umich.uucp!crim.eecs.umich.edu!projoe #
#  1101 Beal Avenue          BIX  : jdionise                                   #
#  Ann Arbor, MI 48109       (313) 936-2830                                    #
################################################################################