Much faster splitmv mode decoding.

Originally committed as revision 24680 to svn://svn.ffmpeg.org/ffmpeg/trunk

~50% faster MB mode decoding, plus eliminate a costly switch.

Originally committed as revision 24679 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 24678 to svn://svn.ffmpeg.org/ffmpeg/trunk

Always inline the arithmetic coder, except in the case of header-parsing stuff,
in which case don't inline it at all to save code size.

Originally committed as revision 24677 to svn://svn.ffmpeg.org/ffmpeg/trunk

This is especially important because classifs can be very large.

Originally committed as revision 24676 to svn://svn.ffmpeg.org/ffmpeg/trunk

If sizeof uint_fast8_t > 1 and sizeof size_t <= 4, the expression that mallocs
classifs  is susceptible to integer overflow.

Originally committed as revision 24675 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 24674 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 24673 to svn://svn.ffmpeg.org/ffmpeg/trunk

It is already selected through the H264DSP dependency of VP8.

Originally committed as revision 24672 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 24671 to svn://svn.ffmpeg.org/ffmpeg/trunk

A lot of the time the DC block is empty: don't do the WHT in this case.
A lot of the rest of the time, there's only one coefficient: make a special
DC-only transform for that case.
When the block is empty, don't incorrectly mark luma DCT blocks as having DC
coefficients.

Originally committed as revision 24670 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 24669 to svn://svn.ffmpeg.org/ffmpeg/trunk

Lets us do the zeroing in asm instead of C.
Also makes it consistent with the way the regular iDCT code does it.

Originally committed as revision 24668 to svn://svn.ffmpeg.org/ffmpeg/trunk

This is required when cross-testing on dos-like targets.

Originally committed as revision 24667 to svn://svn.ffmpeg.org/ffmpeg/trunk

Without this flag -Wall generates warnings that nobody wants to see fixed or
worked around. Disabling those warnings is thus the best solution.

Originally committed as revision 24666 to svn://svn.ffmpeg.org/ffmpeg/trunk

Increase readability and robustness, as the test result is not going
to differ if the order of the pixfmts codes changes.

Originally committed as revision 24665 to svn://svn.ffmpeg.org/ffmpeg/trunk

inter-frame behaviour unchanged.

Originally committed as revision 24664 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 24663 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 24662 to svn://svn.ffmpeg.org/ffmpeg/trunk

/**\u2264 => /**<

Originally committed as revision 24661 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 24660 to svn://svn.ffmpeg.org/ffmpeg/trunk

~1.5% faster decode_block_coeffs

Originally committed as revision 24659 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 24658 to svn://svn.ffmpeg.org/ffmpeg/trunk

The corresponding lavfi-pixfmts BE tests are not yet added, as there
are some bugs in the scaler (scaling rgba, argb, bgra, abgr, yuva420p)
which result in differences with the LE reference, and I cannot
visually check the generated files on BE.

Originally committed as revision 24657 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 24656 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 24655 to svn://svn.ffmpeg.org/ffmpeg/trunk

Introduce the function do_lavfi_pixfmts(), and use it for generating a
pixfmts test for each different filter.

Originally committed as revision 24654 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 24653 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 24652 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 24651 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 24650 to svn://svn.ffmpeg.org/ffmpeg/trunk

The 'tar' variable should be set to a command writing a tar archive
of the named files to stdout, typically "tar c" or "tar cf -"

Originally committed as revision 24649 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 24648 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 24647 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 24646 to svn://svn.ffmpeg.org/ffmpeg/trunk

Add the lavfi-showfiltfmts dependency in the Makefile, and correctly
use the $target_exec and $target_path variables for invoking the
lavfi-showfiltfmts tool.

Originally committed as revision 24645 to svn://svn.ffmpeg.org/ffmpeg/trunk

otherwise the test will be running whatever ffmpeg is installed on the
host system.

Originally committed as revision 24644 to svn://svn.ffmpeg.org/ffmpeg/trunk

test output files, and add a prefix with the name of the test.
Make per-filter grouping of the generated output files easier, which
is more useful than per-pixel-format grouping.

Originally committed as revision 24643 to svn://svn.ffmpeg.org/ffmpeg/trunk

Consistent with the lavfi pixdesc test code, and slightly improve
readability.

Originally committed as revision 24642 to svn://svn.ffmpeg.org/ffmpeg/trunk

test results and references depend on machine endianess.

Originally committed as revision 24641 to svn://svn.ffmpeg.org/ffmpeg/trunk