>Date: Mon, 20 Jan 1997 00:09:12 -0500
>From: Michael Richardson <mcr@sandelman.ottawa.on.ca>

>  I had been running -current kernel with a PCVT console on it, and
>PPP running at 9600 baud on COM0. Due to the fact that I SUP on one
>machine and then use rsync to keep a second one up to date (no NFS
>allowed between them), I can't be certain what date of -current it
>was. I believe it was before Jan. 6.
>  I installed a new kernel via FTP, rebooted remotedly. I stupidly
>didn't save the old one (I had three working previous ones on the disk
>already...). I rebooted, and then discovered that my keyboard didn't
>work. The 1.2 kcadp install kernel works, as does a kernel from a
>production 1.2 machine. I have since rebuilt with pc0 instead of vt0
>(which i thought was the culprit) and have found the same problem..
>  I should note that the keyboard works at the boot loader. It doesn't
>work when you boot with -a to answer the root? prompt, so this isn't
>getty or something not working. I have not had DDB built in, or I'd
>have tried CTL-ALT-ESC.
>  With pcvt, I got complaints before the kernel banner of:	
>	pcvt: doreset() - timeout controller write command
>  
>  I'm beginning to get suspicious of perhaps the delay macros...

Try applying the patches in PR 2631.  They increase the length of the
delay, so if you use them you should at least find out if lack of
delay is your problem.
-- 
Mike Long <mike.long@analog.com>     <URL:http://www.shore.net/~mikel>
VLSI Design Engineer         finger mikel@shore.net for PGP public key
Analog Devices, CPD Division          CCBF225E7D3F7ECB2C8F7ABB15D9BE7B
Norwood, MA 02062 USA       (eq (opinion 'ADI) (opinion 'mike)) -> nil