From: NeilBrown <neilb@cse.unsw.edu.au>

From: Josh Litherland <josh@emperorlinux.com>

The XOR routine selection process is often confusing as it often doesn't
choose the routine that measures the fastest, as sometime one routine is
known to have better cache performance.

This patch avoids the comparison when there is a known best choice, and
makes the report more meaningful.
Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>