[vlc-devel] commit: Do not use a macro to clip integer to?uint8 (2x faster YUVA->RV32) (Laurent Aimar )

[Rémi Denis-Courmont]

Le mercredi 23 juillet 2008 22:30:07 Laurent Aimar, vous avez écrit :
> blend.c would be way faster with mmx/sse2 but I will not write gcc inline
> ASM. (It is way too ugly)

AFAIK, you don't need inline assembly to use MMX. gcc provides MMX through 
__builtin_*() functions.

-- 
Rémi Denis-Courmont
http://www.remlab.net/