[PATCH 9/12] ppc64: Misc fixups for non-zero KERNELBASE

[Hollis Blanchard]

On Aug 30, 2005, at 1:02 AM, Milton D. Miller II wrote:
>
> You took out the scheduling of putting independent ops between the
> clear and the ori of the lower half.
>
> This just cost the main kernel several wasted cycles.  (No, I didn't
> sim this to count).

Is that really true? I did an unofficial test (on 604 because it was 
handy) with this sequence:
         lis 3, 0x55
         lis 4, 0x55
         ori 3, 3, 0xaa55
vs
         lis 3, 0x55
         ori 3, 3, 0xaa55
         lis 4, 0x55

... run 100000000 iterations, a few times to allow for cache warming. I 
saw no difference in the run times.

Was this just a non-representative test case, or would it matter more 
on other processors?

-Hollis