Re: [PATCH] moving F0 0F bug check to bugs.h

[Alan Cox]

> > We only check for the bug once so we might as well move it to check
> > the boot cpu only in bugs.h.
> 
> Whilst not an ideal solution, some people do silly things like
> putting a P150 and a P166 clocked to 150 into SMP boxes.
> It could be possible for 1 CPU to have the bug whilst another doesn't.

Perhaps we should add the bugs as bit flags to the capability masks - but
inverted (ie "no foof bug" etc) that way they'll come out with the rest of
the SMP bug handling logic