[PATCH] atomic_dec_if_positive sign extension fix

[Segher Boessenkool]

> What is this function supposed to do if it gets 0x80000000 as
> input btw?  The current code happily makes it 0x7fffffff as
> far as I can see?  The "rlwinm" thing would fix that ;-)  (Or
> unfix, if the current behaviour is intended).

...except it doesn't change behaviour on 32-bit of
course.  Forget it, do one of the other suggested
things instead please :-)


Segher