Keyboard Controller woes

[7]

Hello There,
Happy Wednesday Yall.

      I had a business partner ask me a question about something that 
seriously rang a bell, but I could not find the reference after much 
searching. It's a problem I think I used to have on a LX based intel 
board a while ago, but I can't remember.

If he cold boots a certain machine without a keyboard (ps/2) plugged in, 
then tries to plug one in after it's up and running, it won't work. 
However if he already has a keyboard plugged in when he boots, then 
unplugs it and leaves it that way for days, then plugs the keyboard in 
it works fine. The whole point is he'd rather not have to be there to 
plug a keyboard into each of 50 1U rackmounts when they need rebooted to 
be sure they will work later. He's running the latest redhat on the 
exact same 1U's and they don't display the same behavior. I was *sure* 
this was something simple I read in LINT or on the mailing list at some 
point, but I'm not sure. Damn time and age and all that. Anyone remember 
what I'm talking about? It was something to do with a kernel config 
parameter if I'm not mistaken that fixed this issue with the keyboard 
controller. If I'm not mistaken, he's using 4.8 release updated with 
RELENG_4_8. If someone could just point me in the right direction, I'd 
be much appreciative. After searching all day I'm burnt.

THANKS IN ADVANCE!