Neil Roberts 1b2ff7eff7 cogl: Use SSE2 when possible for premultiplying
This adds a fast path for premultiplying an RGBA image using SSE2
instructions. SSE registers are 128-bit and we need at least 16-bits
per component for the intermediate result of the multiplication so we
can do two pixels in parallel with one register. The function
interleaves 2 SSE registers to multiply 4 pixels in one function call
with the hope that this will pipeline better.

http://bugzilla.openedhand.com/show_bug.cgi?id=1939
Signed-off-by: Emmanuele Bassi <ebassi@linux.intel.com>
2010-02-09 11:03:21 +00:00
..
2009-11-06 11:48:42 +00:00
2009-01-20 18:47:50 +00:00
2010-02-03 16:34:27 +00:00
2010-02-01 10:40:34 +00:00
2009-10-14 11:31:26 +01:00
2010-01-20 00:38:07 +00:00
2009-07-10 11:38:42 +01:00
2010-02-04 21:10:02 +00:00
2010-02-03 14:38:13 +00:00
2010-02-01 10:40:34 +00:00
2010-02-01 10:40:34 +00:00
2010-02-07 18:25:19 +00:00
2009-10-16 15:55:05 +01:00
2010-02-07 18:25:19 +00:00
2010-02-07 18:25:19 +00:00