The audit of _p usage shows various drivers assume inb_p is somehow atomic.
 Of course it isn't and the delay can be split from the I/O cycle causing a
timing violation on chips that matter (eg this one)

With the proposed use of udelay() for some _p delays this will cease to be
a mostly theoretical bug (as the delay stall is unsplittable) and wants
fixing.

Lots of other drivers need fixing this way too.
Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Wim Van Sebroeck <wim@iguana.be>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>