home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.sys.sgi
- Path: sparky!uunet!paladin.american.edu!darwin.sura.net!sgiblab!sgigate!odin!twilight!zuni!anchor!olson
- From: olson@anchor.esd.sgi.com (Dave Olson)
- Subject: Re: DEADLOCK
- Message-ID: <pmdm6po@zuni.esd.sgi.com>
- Sender: news@zuni.esd.sgi.com (Net News)
- Organization: Silicon Graphics, Inc. Mountain View, CA
- References: <1992Sep9.155859.5492@asf.com>
- Distribution: usa
- Date: Thu, 10 Sep 92 19:50:37 GMT
- Lines: 39
-
- In <1992Sep9.155859.5492@asf.com> zum@asf.com (Craig Zumwinkle) writes:
- | We have two programs running concurrently. The one that is visible
- | is a user interface for command and control ( i.e. lots of buttons, sliders,
- | and bitmap graphics, etc) and the other is a communications program that
- | sends and receives data via a serial port connection using the RS422
- | protocol. The communications program receives packets of data that
- | average about 50 bytes @ 9600 baud at about 7 to 10 times a second.
- | These two programs exchange data with each other through shared memory.
- | SYMPTOMS:
- | =========
- |
- | Our machine appears to deadlock at unpredictable periods are not
- | reliably reproducible. However, it has deadlocked when we were
- | not running our programs; in fact one time I was just using vi when
- | it locked up. It has locked up when running our software in the
- | debugger. Now when I say locked up I mean it is impossible to interact
- | with the machine. No cursor movement, unable to ping it from another
- | host, no magic key sequences seem to free it up. My only alternative
- | is to reboot the sucker. Upon system boot fsck reports no problems
- | on the /usr file system and there aren't any new error or warning
- | messages in the SYSLOG. So far we have completely replaced the graphics
- | cards twice and the cpu cards once. We also replaced one bad simm (A4).
-
- How are you doing EIA 422? If it is a seperate board (as I assume it
- most likely is, since we don't have full 422 support), perhaps that
- board is your problem?
-
- | HELP! HELP! We are at the end of our rope here. Do we have a problem
- | because of the latest release of the OS coupled with this new venture
- | into 6U VMEbus computer???
-
- I don't know a whole lot about the VME based systems, but I would
- certainly call and ask the support folks; there were some VME interface
- problems on the regular 4D/30 and 35, but I thought they were fixed
- for the VME based systems.
- --
- Let no one tell me that silence gives consent, | Dave Olson
- because whoever is silent dissents. | Silicon Graphics, Inc.
- Maria Isabel Barreno | olson@sgi.com
-