utilize a bit reservoir. Also, make the lame encoder utilize this flag.

patch by Paul Kelly, paul a stjohnspoint d co d uk

Originally committed as revision 11903 to svn://svn.ffmpeg.org/ffmpeg/trunk

constants which are completely unneeded anyway. Note that 3 is a cheap
integer bound for log10(1<<CHAR_BIT).

Originally committed as revision 11902 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 11901 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 11900 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 11899 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 11898 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 11897 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 11896 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 11895 to svn://svn.ffmpeg.org/ffmpeg/trunk

Commited in SoC by Bobby Bingham

Originally committed as revision 11894 to svn://svn.ffmpeg.org/ffmpeg/trunk

Commited in SoC by Bobby Bingham

Originally committed as revision 11893 to svn://svn.ffmpeg.org/ffmpeg/trunk

reference.

Commited in SoC by Bobby Bingham

Originally committed as revision 11892 to svn://svn.ffmpeg.org/ffmpeg/trunk

Original log message:
Forgot to add in the big colorspace commit

Commited in SoC by Bobby Bingham

Originally committed as revision 11891 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 11890 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 25967 to svn://svn.mplayerhq.hu/mplayer/trunk/libswscale

Originally committed as revision 11889 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 11888 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 11887 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 11886 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 11885 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 11884 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 11883 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 11882 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 11881 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 11880 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 11879 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 11878 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 11877 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 11876 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 11875 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 11874 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 11873 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 11872 to svn://svn.ffmpeg.org/ffmpeg/trunk

25% faster tham mmx on core2, 35% if you discount fullpel, 4% overall decoding.

Originally committed as revision 11871 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 11870 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 11869 to svn://svn.ffmpeg.org/ffmpeg/trunk

clean up an ugliness introduced in r11826. this syntax will require fewer changes when adding future sse2 code.

Originally committed as revision 11868 to svn://svn.ffmpeg.org/ffmpeg/trunk

Correctly interleave ogg packets per granule
and set eos correctly, 2 packets buffering is needed.
It duplicates interleave_per_dts a bit,
if someone has a good solution, I'll implement it.

Originally committed as revision 11867 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 11866 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 11865 to svn://svn.ffmpeg.org/ffmpeg/trunk