Originally committed as revision 744 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 743 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 742 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 741 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 740 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 739 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 738 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 737 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 736 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 735 to svn://svn.ffmpeg.org/ffmpeg/trunk

* generic optimization:
  - enable ALT and ALIGNED stream
  - move "if(const)" out from loops
  - supress duplicated computing
* first attempt to fix problem with strides

Originally committed as revision 734 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 733 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 732 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 731 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 730 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 729 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 728 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 727 to svn://svn.ffmpeg.org/ffmpeg/trunk

fixing sinc filter (seems the problem was caused by rounding in pmulhw -> solution use shorter filter, its long and slow enough anyway)
some typos
some warning fixes

Originally committed as revision 6680 to svn://svn.mplayerhq.hu/mplayer/trunk/postproc

Originally committed as revision 726 to svn://svn.ffmpeg.org/ffmpeg/trunk

new bitstream reader API (old get_bits() based one is emulated and will still be supported in the future cuz its simpler)
minor optimizations to get_vlc

Originally committed as revision 725 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 724 to svn://svn.ffmpeg.org/ffmpeg/trunk

encoding of slice_height != mb_height
1bit shorter wmv1 headers if bit_rate<50 !? ROTFL M$ is even more stupid than i thought
using dc of non intra blocks for dc prediction of wmv1 if bitrate < 129 && res < 320x240

Originally committed as revision 723 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 722 to svn://svn.ffmpeg.org/ffmpeg/trunk

* Don't include config.h in dsputil.c

Originally committed as revision 721 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 6638 to svn://svn.mplayerhq.hu/mplayer/trunk/postproc

Originally committed as revision 720 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 719 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 718 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 717 to svn://svn.ffmpeg.org/ffmpeg/trunk

* Use "restrict" in some dsputil routines (yields large speedup)

Originally committed as revision 716 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 715 to svn://svn.ffmpeg.org/ffmpeg/trunk

for some improvements.

Originally committed as revision 714 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 713 to svn://svn.ffmpeg.org/ffmpeg/trunk

be a noticeable win (at least for me :)

Originally committed as revision 712 to svn://svn.ffmpeg.org/ffmpeg/trunk

takes the most time, and it allows for more efficient unaligned access
and better control over memory latencies.

Originally committed as revision 711 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 710 to svn://svn.ffmpeg.org/ffmpeg/trunk

allows better scheduling of the memory accesses, and is portable among
all compilers.

Originally committed as revision 709 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 6617 to svn://svn.mplayerhq.hu/mplayer/trunk/postproc

Originally committed as revision 6616 to svn://svn.mplayerhq.hu/mplayer/trunk/postproc