[FFmpeg-cvslog] r21736 - trunk/libavcodec/x86/dsputil_mmx.c
conrad
subversion
Wed Feb 10 03:02:06 CET 2010
Author: conrad
Date: Wed Feb 10 03:02:06 2010
New Revision: 21736
Log:
Enable SSE2 (put|avg)_pixels_16_sse2
SVQ1 chroma has been special-cased aligned to 16-bytes since at least r15466
Other architectures also assume 16-byte alignment here too but set STRIDE_ALIGN
to 16.
Modified:
trunk/libavcodec/x86/dsputil_mmx.c
Modified: trunk/libavcodec/x86/dsputil_mmx.c
==============================================================================
--- trunk/libavcodec/x86/dsputil_mmx.c Tue Feb 9 23:55:16 2010 (r21735)
+++ trunk/libavcodec/x86/dsputil_mmx.c Wed Feb 10 03:02:06 2010 (r21736)
@@ -2840,10 +2840,8 @@ void dsputil_init_mmx(DSPContext* c, AVC
c->avg_h264_qpel_pixels_tab[1][x+y*4] = avg_h264_qpel8_mc##x##y##_##CPU;
if((mm_flags & FF_MM_SSE2) && !(mm_flags & FF_MM_3DNOW)){
// these functions are slower than mmx on AMD, but faster on Intel
-/* FIXME works in most codecs, but crashes svq1 due to unaligned chroma
c->put_pixels_tab[0][0] = put_pixels16_sse2;
c->avg_pixels_tab[0][0] = avg_pixels16_sse2;
-*/
H264_QPEL_FUNCS(0, 0, sse2);
}
if(mm_flags & FF_MM_SSE2){
More information about the ffmpeg-cvslog
mailing list