[FFmpeg-cvslog] r25259 - trunk/libavcodec/x86/dsputil_mmx.c

rbultje subversion
Wed Sep 29 17:34:44 CEST 2010


Author: rbultje
Date: Wed Sep 29 17:34:43 2010
New Revision: 25259

Log:
Use sse2 variant of put_pixels16() for no_rnd also. Provides a minor speed
increase to e.g. vc1, snow and mpeg decoding.

Patch by Eli Friedman <eli dot friedman gmail com>.

Modified:
   trunk/libavcodec/x86/dsputil_mmx.c

Modified: trunk/libavcodec/x86/dsputil_mmx.c
==============================================================================
--- trunk/libavcodec/x86/dsputil_mmx.c	Wed Sep 29 17:09:38 2010	(r25258)
+++ trunk/libavcodec/x86/dsputil_mmx.c	Wed Sep 29 17:34:43 2010	(r25259)
@@ -2820,6 +2820,7 @@ void dsputil_init_mmx(DSPContext* c, AVC
         if((mm_flags & AV_CPU_FLAG_SSE2) && !(mm_flags & AV_CPU_FLAG_3DNOW)){
             // these functions are slower than mmx on AMD, but faster on Intel
             c->put_pixels_tab[0][0] = put_pixels16_sse2;
+            c->put_no_rnd_pixels_tab[0][0] = put_pixels16_sse2;
             c->avg_pixels_tab[0][0] = avg_pixels16_sse2;
             H264_QPEL_FUNCS(0, 0, sse2);
         }



More information about the ffmpeg-cvslog mailing list