Originally committed as revision 19745 to svn://svn.ffmpeg.org/ffmpeg/trunk

architectures since stride is multiple of 4 and not of 8, so split
fill_rectangle() calls to operate on 32-bit words instead of 64-bit ones.

Originally committed as revision 19744 to svn://svn.ffmpeg.org/ffmpeg/trunk

inside it.

Originally committed as revision 19743 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 29587 to svn://svn.mplayerhq.hu/mplayer/trunk/libswscale

Originally committed as revision 29586 to svn://svn.mplayerhq.hu/mplayer/trunk/libswscale

Originally committed as revision 29585 to svn://svn.mplayerhq.hu/mplayer/trunk/libswscale

Originally committed as revision 29584 to svn://svn.mplayerhq.hu/mplayer/trunk/libswscale

It works the same as CHECKED_ALLOCZ except that it does not zero the allocated
memory.

Originally committed as revision 19742 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 19741 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 19740 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 19739 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 19738 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 29583 to svn://svn.mplayerhq.hu/mplayer/trunk/libswscale

Originally committed as revision 29582 to svn://svn.mplayerhq.hu/mplayer/trunk/libswscale

The DECLARE_ALIGNED_8 macro is defined to align to 16 bytes instead
the 8 suggested by the name on some CPUs.  None of the uses of this
macro ever need 16-byte alignment, cases which once did having been
changed to always specify 16 bytes explicitly.

Originally committed as revision 19737 to svn://svn.ffmpeg.org/ffmpeg/trunk

decoded.

Originally committed as revision 19736 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 19735 to svn://svn.ffmpeg.org/ffmpeg/trunk

a packet of Speex frames as a single frame.

Originally committed as revision 19734 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 19733 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 19732 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 19731 to svn://svn.ffmpeg.org/ffmpeg/trunk

multiply the buffer size by the number of channels.

Originally committed as revision 19730 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 19729 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 19728 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 19727 to svn://svn.ffmpeg.org/ffmpeg/trunk

1.3x and 3% on G4.
Though I think only part of this speedup is due to my optimizations per se;
some of it is that I got a better roll on the GCC random code generator.
Trivial reorderings of this function have a disproportionate effect on speed.

Originally committed as revision 19726 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 19725 to svn://svn.ffmpeg.org/ffmpeg/trunk

The first member of struct ScanTable is a simple pointer, extra alignment
of which serves no purpose.  The alignment specifier was added along with
some Altivec optimisations also adding a 16-byte-aligned array at the end
of struct ScanTable.  Presumably the redundant, outer alignment was added
by mistake.  The inner one is clearly sufficient.

Originally committed as revision 19724 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 19723 to svn://svn.ffmpeg.org/ffmpeg/trunk

Patch by Christian Schmidt, schmidt digadd de

Originally committed as revision 19722 to svn://svn.ffmpeg.org/ffmpeg/trunk

Patch by Lars Täuber, lars D taeuber A gmx D net

Originally committed as revision 19720 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 19719 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 19718 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 19717 to svn://svn.ffmpeg.org/ffmpeg/trunk

These arrays are passed to functions for which we promise 16-byte
alignment, so make sure we fulfil that promise.

Originally committed as revision 19716 to svn://svn.ffmpeg.org/ffmpeg/trunk

None of these arrays are used in ways requiring extra alignment.

Originally committed as revision 19715 to svn://svn.ffmpeg.org/ffmpeg/trunk

Bluray subtitle decoder instead of as small as possible.
This avoids completely unnecessary issues with e.g. libswscale.
(See r19673)

Patch by Stephen Backway, stev391 A exemail D com D au

Originally committed as revision 19714 to svn://svn.ffmpeg.org/ffmpeg/trunk

Fixes "dvd_subtitle too big" errors when encoding HD video subtitles
to DVD format.

Originally committed as revision 19713 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 19712 to svn://svn.ffmpeg.org/ffmpeg/trunk

(which uses movdqa to xmm1 on inv_zigzag_direct16).

Originally committed as revision 19711 to svn://svn.ffmpeg.org/ffmpeg/trunk