home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!zaphod.mps.ohio-state.edu!magnus.acs.ohio-state.edu!usenet.ins.cwru.edu!agate!doc.ic.ac.uk!mrccrc!warwick!pipex!logica!whiskerp
- From: whiskerp@logica.co.uk
- Newsgroups: vmsnet.pdp-11
- Subject: SCSI upsets UDA50 problem
- Message-ID: <1992Nov17.104126.331@logica.co.uk>
- Date: 17 Nov 92 10:41:26 GMT
- Lines: 64
- Summary: SCSI upsets UDA50 problem
- Reply-To: whiskerp@logica.co.uk
- Followup-To: whiskerp@logica.co.uk
- Keywords: pdp-11
-
- I have a problem with a UDA-50 !
-
- The current system is a PDP-11/84 with :
- 1024KW memory
- UDA-50 + RA81 (CSR/VEC = 172150/154)
- TU80
- DZ11
- KMS11-BD
- VT30H
-
- The initial (successful) piece of work involved connecting two SCSI disks
- and a SCSI DAT tape via a CMD CU712 UNIBUS controller. The existing RSX-11M
- V4.5 had a single controller DUDRV. To add the second controller, I created
- a new driver table for DU and MU by doing a partial PREPGEN and extracted
- them from the SYSTB.MAC thus created. I sorted out the assembly and build
- files and built PUCOM, DUDRV and MUDRV for the correct number of units of
- MSCP and TMSCP devices. I then zapped the existing DU: DCB in RSX11M.TSK to
- change U.DNAM to "EU" so that the VMR LOA DU: would not find any existing
- data structures. I loaded DUDRV, MUDRV and PUCOM. This all worked fine, and
- we ran a system with the RA81, 2xSCSI disks and a SCSI MU0:.
-
- Now the difficult bit.
-
- We wanted to drive a QBUS rack with a QBUS SCSI disk and tape controller
- (same SCSI drives as before). The following equipment was in the QBUS rack:
-
- CMD QBUS card CQD-220 (CSR/VEC=160334/150)
- Able Qniverter (Revision E)
- Bus terminator resistors as required.
- 2xSCSI disks (as before) (SCSI addresses 1 and 2)
- 1xDAT tape (as before)
-
- Using exactly the same VMR of RSX11M, the system BOOTS from DU0: (the RA81)
- and then hangs before the >RED DU0:=SY: line. It is not dead completely,
- but seems to lose contact with the RA81 and thus can't read MCR.
-
- Tests using BRUSYS and RT11 show that the system can access the SCSI disks
- and tapes , however it seems that accessing the SCSI disks clobbers the
- UDA50. We have succesfully used BRUSYS to copy from disk to DAT tape, so it
- is not simply the controller or Qniverter. If I had not seen the system
- working with the UNIBUS SCSI card without any problem, I would obviously
- thought that it was the software. We tried it also with the Viking QDT
- controller (with autoconfigure disabled) and had similar but not identical
- symptoms.
-
- The QBUS devices had the DMA mapped to the lower 18-bit segment of the
- UNIBUS. We disabled the clever bits in the QBUS SCSI controller which allowed
- adaptive DMA etc but it made no difference. The QBUS controller was also
- tried with IRP of 4 and 5 (the UDA-50 is 5). The UDA50 is the last thing on
- the bus, followed by the bus terminator which gets replaced by the bus
- cable which goes to the Qniverter (thus all the QBUS devices follow the
- UDA50).
-
- If anybody has any ideas, please email me or put it here and I will
- summarise.
-
- Many thanks
- Peter Whisker (still fixing PDP's after 13 years!)
-
- --
-
- Peter Whisker | X400: <Peter Whisker@OR=LG@PRMD=LOGICA@ADMD=TMAILUK@C=GB>
- Logica DCG Ltd. | Internet: <whiskerp@logica.co.uk>
- Cobham, Surrey, UK |
-