[comp.arch] NMI bug

lindsay@MATHOM.GANDALF.CS.CMU.EDU (Donald Lindsay) (08/25/90)

In article <2450@crdos1.crd.ge.COM> davidsen@crdos1.crd.ge.com
	 (bill davidsen) writes:
 [referring to a long-ago bug in the 8088 interrupt logic]
>  Was this problem only with NMI? I've run a lot of interrupts into an
>original XT and never seen a problem.

It was only the NMI, and since many boards didn't use NMI, the bug
escaped into the field.  Intel fixed the chip before the XT came out.

The reason I brought this up, was to illustrate a moral:  features
that aren't used, very possibly won't work and never did.

Much like Saltzer's Law: if a system is reliable enough, then the
outage procedures don't work any more.
-- 
Don		D.C.Lindsay