Newsgroups: comp.realtime Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!zaphod.mps.ohio-state.edu!caen!hellgate.utah.edu!csn!everest!iplmail!iplmail!jans From: jans@iplmail.orl.mmc.com (jan schonhoff) Subject: vxWorks 4.0.2 Message-ID: <1991Apr26.151723.10444@iplmail.orl.mmc.com> Keywords: vxworks Sender: jans@iplmail (jan schonhoff) Organization: Martin Marietta Date: Fri, 26 Apr 1991 15:17:23 GMT We received the latest vxWorks release 4.0.2 for the target Sun1e board recently from Wind River Systems. We have had real difficulty in getting the target board to boot reliably. When it does boot (~1 out of 5 times), vxWorks seems okay. However, it (more often than not) seems to hang at the line "Loading ... # + # + #". ^ | The first number in this load message usually gets printed, then nothing else happens. We have used the "etherfind" command on the sun host to watch the network traffic, and up to this point, there is continuous "req", "ack" between the target and the sun. Once the "Loading ... #" is printed, all communications cease. The target Sun1e board is on a local net connected to the host Sun, which is connected to the wide area net. The wide net cannot see the target Sun1e, only the sun on the local net. We have used a previous version of vxWorks, Betad (from APLabs), which seems to have no problems. However, the Betad version seems to extract the ethernet address of the target board from EEPROM, while the 4.0.2 version uses a hard-coded address "5a,5a,5a,5a,00,00" plus the target board's inet address in the last 2 bytes of this hard-coded address. To add to the complication of things, we also have a version 4.0.2 of vxWorks from Verdix (VadsWorks). It also had a similar problem of booting up. To see if there was any difference, we changed the hard-coded ethernet address of "5a,5a,5a,5a,00,00" to the board's real ethernet address (gotten from EEPROM), and rebooted the board. It seems to work reliably with the board's real address. Yet our 4.0.2 software and proms from Wind River will not boot reliably with the "5a" address. Any clues as to this problem?? Any users of APLabs "RDbxWorks" source-level debugger for the Sparc 1E engine?? We are having difficulties using it due to our network setup, and APLabs has not been able to help us. I could sure use some suggestions?? Thanks. ************************************************************************* Jan Schonhoff E-Mail: jans@iplmail.orl.mmc.com Staff Engineer Tel: 407-356-9735 (Desk) 407-356-8839 (Lab) Martin Marietta ESC FAX #: 407-356-5651 US-Mail: Martin Marietta ESC P.O. Box 628007 Mail Point 1304 Orlando, Florida 32862-8007 *************************************************************************