[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Problem with IOB1 on CADR8
- To: (BUG LISPM) at MIT-AI, JLK at MIT-MC
- Subject: Re: Problem with IOB1 on CADR8
- From: moon@MIT-AI (Sent by MOON50@MIT-AI)
- Date: Wed ,8 Nov 79 13:11:20 EDT
This is actually a design bug. The computation of LOCKOUT INIT (on LMDETC)
has a race in it, because -CBLBSY A doesn't come on (drop) fast enough at
the beginning of a message, causing a narrow pulse that makes funny things
on GENCLK, which then inserts an extra bit at the front of the received message.
I removed the symptom by replacing the 74161 at B05 with a 74LS161, which proved
to be slightly faster. I don't see any obviously better ways to fix this.
The symptom from a software point of view is the bit count is 1 too high
and all the words in the message are shifted one bit to the left, with an
extra zero word at the front of the message and the CRC word missing at the
end (if your software doesn't round up when it divides the bit count by 16.)