RE: sick 3640

[gooch@sw-sw.dialnet.symbolics.com (William D. Gooch)]

    Date: Tue, 18 Feb 1992 18:12 CST
    From: drstrip@cs.sandia.gov (David R. Strip)

    I have a '40 that was just moved to a new lab, and now
    won't boot. I can bring up the FEP, issue the boot instruction
    and watch it go all the way through to the start instruction.
    At that point things get wierd. The usual disk seek noise
    is way too short, but if I wait (and wait) and then h-c-function
    out I get a long wait with the message
    "Waiting for lisp to stop itself...

    If you then do show status you get
    FEP Lbus control: Doorbel Int Enb, Use Corrected Data, Ignore Double ECC error
		    Task 3 req
    FEP Proc Control: Fep int enable
    Seqn misc status: TSK-STOP

    3600 program ctrs
    CPC/2
    all the pc's are 40002

    Attempting to use the FEP debugger is a dismal failure:
    you get
    <error printing object: the arg given to the CAR inst, #<DTP-0 0>, was
    not a list, locative, or nil>
    Then you are totally lost.

Did you do a c-m-s (switch to main stack)?  Sometimes the first stack
that you get in the FEP debugger is not very helpful.  However, I suspect
that you may not get much useful information there, anyway.

    I have switched the DP, XSQ (yeah, I suppose it's really '45), MC for an IFU,
    all from another machine, switched the memory too, and still get the same symptoms,
    though I did not do the status checks on all the combinations.

Sounds to me like it could be a bad FEP (first guess), or maybe an IO.
Paddle cards are perhaps suspect also, but I would first try a new FEP.

    Any suggestions besides a merciful death? 
    Thanks.
    David Strip
    drstrip@cs.sandia.gov