[Pixman] [PATCH 00/32] (Mostly) ARMv6 speedups

[Ben Avison]

No comments at all on the patch series I posted on 2014-08-07?

I do have a few tweaks and additions, but rather than sending them all again
I'll just repost the changed and new ones:

Patch 03: Corrected bounds calculations so that cover versions of bilinear
  scaled plots are used (which was the original intention).
Patch 31: I was contacted off-list to say that some versions of GCC choked
  on the unused macro argument "no_mask", so it has been changed to "0".
Patches 33-36: ARMv6 versions of bilinear scaled fetchers, up to 7x faster
  than generic code.
Patch 37: Correction to calculation of the two flags that determine that a
  plot can use a cover fast path or fetcher. This one should benefit all
  architectures.

Ben