diff options
| author | Romain Dolbeau <dolbeau@irisa.fr> | 2004-05-28 13:31:38 +0000 |
|---|---|---|
| committer | Michael Niedermayer <michaelni@gmx.at> | 2004-05-28 13:31:38 +0000 |
| commit | a7b2871cd1401ce7be59b153eed3f25565b0bb23 (patch) | |
| tree | 68c0753c1e00e531fba09d0edca351322fd12083 /libavcodec/libpostproc/postprocess.c | |
| parent | 39d89b69666b1f8596c9edc8bfdc29a70610fb68 (diff) | |
| download | ffmpeg-a7b2871cd1401ce7be59b153eed3f25565b0bb23.tar.gz | |
Newer version, using a vectorized version of the
new organisation of code in doVertLowPass. it
seems to be faster in AltiVec also...
Also includes a compile fix for the new do_a_deblock
when using AltiVec.
patch by (Romain Dolbeau <dolbeau at irisa dot fr>)
Originally committed as revision 3167 to svn://svn.ffmpeg.org/ffmpeg/trunk
Diffstat (limited to 'libavcodec/libpostproc/postprocess.c')
| -rw-r--r-- | libavcodec/libpostproc/postprocess.c | 4 |
1 files changed, 4 insertions, 0 deletions
diff --git a/libavcodec/libpostproc/postprocess.c b/libavcodec/libpostproc/postprocess.c index 2ab85071bc..e669c13e00 100644 --- a/libavcodec/libpostproc/postprocess.c +++ b/libavcodec/libpostproc/postprocess.c @@ -92,6 +92,10 @@ try to unroll inner for(x=0 ... loop to avoid these damn if(x ... checks #include "mangle.h" //FIXME should be supressed +#ifdef HAVE_ALTIVEC_H +#include <altivec.h> +#endif + #ifndef HAVE_MEMALIGN #define memalign(a,b) malloc(b) #endif |