Originally committed as revision 21194 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 21193 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 21192 to svn://svn.ffmpeg.org/ffmpeg/trunk

Up to 6% faster overall on i7, no change on A8.

Originally committed as revision 21191 to svn://svn.ffmpeg.org/ffmpeg/trunk

GCC should be able to do this, but doesn't.  Slightly faster code.

Originally committed as revision 21190 to svn://svn.ffmpeg.org/ffmpeg/trunk

Simplify cur_band_type, group_len, and coef/offset calculations.  This
makes the code easier to read and slightly faster.

Originally committed as revision 21189 to svn://svn.ffmpeg.org/ffmpeg/trunk

The codebooks each consist of small number of values repeated in
groups of 2 or 4.  Storing the codebooks as a packed list of 2- or
4-bit indexes into a table reduces their size substantially (from 7.5k
to 1.5k), resulting in less cache pressure.

For the band types with sign bits in the bitstream, storing the number
and position of non-zero codebook values using a few bits avoids
multiple get_bits() calls and floating-point comparisons which gcc
handles miserably.

Some float/int type punning also avoids gcc brain damage.

Overall speedup 20-35% on Cortex-A8, 20% on Core i7.

Originally committed as revision 21188 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 21187 to svn://svn.ffmpeg.org/ffmpeg/trunk

Patch by Tomas Härdin: $(name) punto hardin chez codemill dot se

Originally committed as revision 21186 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 21185 to svn://svn.ffmpeg.org/ffmpeg/trunk

the file for future 16k mode decoder code.

Originally committed as revision 21184 to svn://svn.ffmpeg.org/ffmpeg/trunk

Two of these are in fact constant size, so use the constant instead of
a variable in the declarations.  The remaining one is small enough
that always using the maximum size is acceptable.

Originally committed as revision 21183 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 21182 to svn://svn.ffmpeg.org/ffmpeg/trunk

not slower according to benchmarks.

Originally committed as revision 21181 to svn://svn.ffmpeg.org/ffmpeg/trunk

Yes i did compile&test, no svq3.c was not recompiled.

Originally committed as revision 21180 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 21179 to svn://svn.ffmpeg.org/ffmpeg/trunk

Seems to speed the code up a little...
The placement of many generic functions between h264.c and h264.h is still open
Currently they are a little randomly placed between them.

Originally committed as revision 21178 to svn://svn.ffmpeg.org/ffmpeg/trunk

which it inserts the new filter.

Originally committed as revision 21177 to svn://svn.ffmpeg.org/ffmpeg/trunk

filter cannot convert between input and output formats.

Originally committed as revision 21176 to svn://svn.ffmpeg.org/ffmpeg/trunk

Patch by myself and Yuriy Kaminskiy [yumkam mail ru]

Originally committed as revision 21175 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 21174 to svn://svn.ffmpeg.org/ffmpeg/trunk

called once per MB in worst case and doesnt seem to benefit from static inline.
Actually the code might be a hair faster now (0.1% according to my benchmark but
this could be random noise)

Originally committed as revision 21173 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 21172 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 21171 to svn://svn.ffmpeg.org/ffmpeg/trunk

in the context to increase compatibility.

Originally committed as revision 21170 to svn://svn.ffmpeg.org/ffmpeg/trunk

no speedloss meassured, also its really not touching anything that is speed relevant.

Originally committed as revision 21169 to svn://svn.ffmpeg.org/ffmpeg/trunk

but not here since it was unused before now.

Originally committed as revision 30289 to svn://svn.mplayerhq.hu/mplayer/trunk/libswscale

Originally committed as revision 21168 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 21167 to svn://svn.ffmpeg.org/ffmpeg/trunk

<$firstname.$lastname()codemill,se>.

Originally committed as revision 21166 to svn://svn.ffmpeg.org/ffmpeg/trunk

No speedloss meassured (its slightly faster here but that may be random fluctuations)

Originally committed as revision 21165 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 21164 to svn://svn.ffmpeg.org/ffmpeg/trunk

Patch by Martin Storsjo (martin AT martin DOT st)

Originally committed as revision 21163 to svn://svn.ffmpeg.org/ffmpeg/trunk

instead.  Avoid a crash on MinGW.

Originally committed as revision 21162 to svn://svn.ffmpeg.org/ffmpeg/trunk

Patch by Sergiy (gmail(piratfm))

Originally committed as revision 21161 to svn://svn.ffmpeg.org/ffmpeg/trunk

This fixes issue streaming from Red5 server.
Patch by Art Clarke (aclarke@`echo xyzzy|sed s/y/u/|sed s/y/le/|tr z g`.com)

Originally committed as revision 21160 to svn://svn.ffmpeg.org/ffmpeg/trunk

No meassureable speed difference on pentium dual & cathedral sample.

Originally committed as revision 21159 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 21158 to svn://svn.ffmpeg.org/ffmpeg/trunk

functions called more than per mb are moved into the header, scan8 is also
as it must be known at compiletime.
The code after this patch duplicates h264data.h, this has been done to minimize
the changes in this step and allow more fine grained benchmarking.
Speedwise this is 1% faster on my pentium dual core with diegos cursed cathedral
sample.

Originally committed as revision 21157 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 21156 to svn://svn.ffmpeg.org/ffmpeg/trunk