5 Commits

Author	SHA1	Message	Date
Michael Niedermayer	9f36153415	dont trash mm0, its still used in the next iteration of the loop (fixes strange horizontal lines on some blocks in b frames on mmx only cpus) ... Originally committed as revision 855 to svn://svn.ffmpeg.org/ffmpeg/trunk	2002-08-20 19:23:06 +00:00
Zdenek Kabelac	d3cd07425a	* implemented remaing avg_ pixel functions (these are not used offen) ... Originally committed as revision 656 to svn://svn.ffmpeg.org/ffmpeg/trunk	2002-06-03 17:31:46 +00:00
Zdenek Kabelac	fca0f0e59f	* removed MANGLE from macros for setting constants ... * using MOVQ_WONE/MOVQ_BFE as two instruction instead of static memory value access as its always faster * PAVGB_MMX macro is using now mm6 -> mm7 is unmodified * replaced original pixels_xy2_mmx with new faster and equal implementation * replaced usage of mm7 for other then ZERO contstant in _rnd & _avg file with mm6 Originally committed as revision 632 to svn://svn.ffmpeg.org/ffmpeg/trunk	2002-05-30 15:14:56 +00:00
Zdenek Kabelac	ff9657069a	* using unrolled loops seems to help to MMX Celerons ... * calculating average in parallel to use two MMX units * disabled xy2 put code - it's producing rouding errors Originally committed as revision 626 to svn://svn.ffmpeg.org/ffmpeg/trunk	2002-05-29 20:00:10 +00:00
Zdenek Kabelac	91abb473fb	* code with new PAVGB for MMX only CPU splited into separate file ... and being compiled in the same way as _avg.h * PAVG_MMX macros accept also output parameter * implemented faster put_pixels_xy2, but it has slightly smaller precission. But there is not visible difference in the image quality - might be eventualy easily switched back (#if 0 #endif)- please check Originally committed as revision 624 to svn://svn.ffmpeg.org/ffmpeg/trunk	2002-05-29 17:16:22 +00:00