MJPEG Tools Svn

Process vertical filter first (changes results but enables future optimizations)

Convolve Y completely before U, V instead of doing all horiz Y, U, V, then vert Y, U, V

13.5% faster for 4 elem filters on x86
(extract convolve code to convolveitem, manually increment pointers instead of using indexing)

More fixes for gcc-4 (I hope).

removed defined(NASM) check, is not needed anymore

fix comments

remove mblock_sad_mmx.s as it was converted to inline asm

converted mblock_sad_mmx.s to inline asm

minor improvements -- use mov ins instead of shuf or extra adds whenever possible; some reordering to improve out of order execution

Do something about the horrible indenting here