Consistently apply this rule: the guard name is obtained from the
filename by stripping the leading "lib", converting '/' and '.'  to
'_' and uppercasing the resulting name. Guard names in the root
directory have to be prefixed by "FFMPEG_".

Originally committed as revision 15120 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 14961 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 14935 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 14771 to svn://svn.ffmpeg.org/ffmpeg/trunk

256, 512, 1024 and 2048) to mdct.c. Make them accessible via dsputil.h. Make
wma.c use these shared tables.

Originally committed as revision 14758 to svn://svn.ffmpeg.org/ffmpeg/trunk

20% faster ac3 if downmixing, 15% if not

Originally committed as revision 14743 to svn://svn.ffmpeg.org/ffmpeg/trunk

13% faster ac3 if downmixing

Originally committed as revision 14742 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 14703 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 14702 to svn://svn.ffmpeg.org/ffmpeg/trunk

remove tmp buffer.
skip fft reinterleave pass, leaving data in a format more convenient for simd.
merge post-rotate with post-reorder.

Originally committed as revision 14700 to svn://svn.ffmpeg.org/ffmpeg/trunk

c is 1.9x faster than previous c (on various x86 cpus), sse is 1.6x faster than previous sse.

Originally committed as revision 14698 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 14466 to svn://svn.ffmpeg.org/ffmpeg/trunk

fixed at least:
CVFI1_Sony_D.jsv
CVFI1_SVA_C.264
MR6_BT_B.h264

Originally committed as revision 14310 to svn://svn.ffmpeg.org/ffmpeg/trunk

this has no immediate effect, but will allow it to be used in more codecs.

Originally committed as revision 14252 to svn://svn.ffmpeg.org/ffmpeg/trunk

2% faster vorbis on conroe, k8. 7% on celeron.

Originally committed as revision 14207 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 14205 to svn://svn.ffmpeg.org/ffmpeg/trunk

a multiple of 16. Reflect that fact in the documentation for them.

Originally committed as revision 14148 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 14103 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 14081 to svn://svn.ffmpeg.org/ffmpeg/trunk

Make sure we don't try to set them if that's the case.
This lets dct-test and fft-test build with --disable-mmx on x86, but not yet on ARM.

Originally committed as revision 13957 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 13888 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 12842 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 12428 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 12410 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 12322 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 12318 to svn://svn.ffmpeg.org/ffmpeg/trunk

DECLARE_ALIGNED_8 is not defined at that point, but this code is powerpc
only, so it's really equivalent to DECLARE_ALIGNED(16...

Originally committed as revision 12313 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 12311 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 12309 to svn://svn.ffmpeg.org/ffmpeg/trunk

(as far as that is possible) with the Sun C compiler.

Originally committed as revision 12188 to svn://svn.ffmpeg.org/ffmpeg/trunk

cycles per 1000 pixels on core2:
left: 9211->5170
top: 9283->2138
avg: 12215->7611
paeth: 64024->17360
overall rgb png decoding speed: +45%
overall greyscale png decoding speed: +6%

Originally committed as revision 12164 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 12126 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 11732 to svn://svn.ffmpeg.org/ffmpeg/trunk

generation. Hard code Bessel I0 approximation iterations to 50.

See thread for discussion:
[FFmpeg-devel] [PATCH] Move Kaiser-Bessel Derived window to mdct.c
Started on the 2008/01/10

Originally committed as revision 11520 to svn://svn.ffmpeg.org/ffmpeg/trunk

Patch by Robert Swain, robert d swain a gmail d com

Originally committed as revision 11514 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 11296 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 11188 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 10993 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 10971 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 10765 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 10640 to svn://svn.ffmpeg.org/ffmpeg/trunk

4x faster than c (somehow, even though doubles only allow 2x simd).
overal flac encoding: 15-50% faster on core2, 4-11% on k8, 3-13% on p4.

Originally committed as revision 10621 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 10602 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 10559 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 10239 to svn://svn.ffmpeg.org/ffmpeg/trunk

the old 32bit code)
disable mmx/sse2 optimizations as they need a rewrite now

Originally committed as revision 10218 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 10190 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 10023 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 9605 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 9603 to svn://svn.ffmpeg.org/ffmpeg/trunk