Originally committed as revision 21526 to svn://svn.ffmpeg.org/ffmpeg/trunk

This simplifies configuring for the most common remote testing
setup.

Originally committed as revision 21525 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 21524 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 21523 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 21522 to svn://svn.ffmpeg.org/ffmpeg/trunk

patch by Laurent Aimar  fenrir _at_ videolan _dot_ org

Originally committed as revision 21521 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 21520 to svn://svn.ffmpeg.org/ffmpeg/trunk

buffer size vs. frame size errors.

Originally committed as revision 21519 to svn://svn.ffmpeg.org/ffmpeg/trunk

Avoids a (incorrect) "possibly used uninitialized" warning, no measurable
speed difference.

Originally committed as revision 21518 to svn://svn.ffmpeg.org/ffmpeg/trunk

Patch by Rafaël Carré, rafael D carre A gmail

Originally committed as revision 21517 to svn://svn.ffmpeg.org/ffmpeg/trunk

5% faster on Intel Atom with gcc 4.4.1 (performance is unchanged
if using av_always_inline).

Originally committed as revision 21516 to svn://svn.ffmpeg.org/ffmpeg/trunk

This makes it easy to supply extra flags on the make command line
for testing purposes.

Originally committed as revision 21515 to svn://svn.ffmpeg.org/ffmpeg/trunk

smaller without any speed loss.

Originally committed as revision 21514 to svn://svn.ffmpeg.org/ffmpeg/trunk

since it is no longer required.

Originally committed as revision 21513 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 21512 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 21511 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 21510 to svn://svn.ffmpeg.org/ffmpeg/trunk

This makes it easy to use a replacement allocator instead of the
system default one.

Originally committed as revision 21509 to svn://svn.ffmpeg.org/ffmpeg/trunk

This allows freeing the parser memory in the thread_context freeing loop.

Patch by Rafaël Carré gmailify(rafael, carre)

Originally committed as revision 21508 to svn://svn.ffmpeg.org/ffmpeg/trunk

1% faster MBAFF decoding overall, maybe ~0.1% faster for the cathedral sample.

Originally committed as revision 21507 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 21506 to svn://svn.ffmpeg.org/ffmpeg/trunk

~20 cpu cycles faster loopfilter

Originally committed as revision 21505 to svn://svn.ffmpeg.org/ffmpeg/trunk

quite a bit faster

Originally committed as revision 21504 to svn://svn.ffmpeg.org/ffmpeg/trunk

~6% faster (slow path) loopfilter (should be ~2% overall)

Originally committed as revision 21503 to svn://svn.ffmpeg.org/ffmpeg/trunk

This is a hair slower (0.15% maybe) but i really dont want to have the
identical code duplicated 3 times because gcc adds odd threaded jumps with
register reshuffling and register safe/restore.

Originally committed as revision 21502 to svn://svn.ffmpeg.org/ffmpeg/trunk

Patch by Daniel Kristjansson: danielk cuymedia net

Originally committed as revision 21501 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 21500 to svn://svn.ffmpeg.org/ffmpeg/trunk

In the main per-compiler section, set speed_cflags and size_cflags
with suitable values.  Later one or the other is added to CFLAGS.

Originally committed as revision 21499 to svn://svn.ffmpeg.org/ffmpeg/trunk

This removes all but one MAX_STREAM from avienc.c.

Originally committed as revision 21498 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 21497 to svn://svn.ffmpeg.org/ffmpeg/trunk

to split out

Originally committed as revision 21496 to svn://svn.ffmpeg.org/ffmpeg/trunk

This prevents the issue with having to switch between slow and
fast code paths in each row.
0.5% faster loopfilter for cathedral

Originally committed as revision 21495 to svn://svn.ffmpeg.org/ffmpeg/trunk

a few cycles faster

Originally committed as revision 21494 to svn://svn.ffmpeg.org/ffmpeg/trunk

This allows many things to be simplified away.
h264 decoder is overall 1% faster with a mbaff sample and
0.1% slower with the cathedral sample, probably because the slow loop
filter code must be loaded into the code cache for each first MB of each
row but isnt used for the following MBs.

Originally committed as revision 21493 to svn://svn.ffmpeg.org/ffmpeg/trunk

slightly simpler and more robust.

Originally committed as revision 21492 to svn://svn.ffmpeg.org/ffmpeg/trunk

Patch sent to mailing list by Taylor Ralph, taylor D ralph A gmail

Originally committed as revision 21491 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 21490 to svn://svn.ffmpeg.org/ffmpeg/trunk

Patch by Rafaël Carré, rafael D carre A gmail

Originally committed as revision 21489 to svn://svn.ffmpeg.org/ffmpeg/trunk

These tables are accessed as uint32 so they must be suitably aligned.

Originally committed as revision 21488 to svn://svn.ffmpeg.org/ffmpeg/trunk

finding the 4-byte signature.
Add a check that ignores the signature if we do not have enough previous
data to go back at least 7 bytes.

Originally committed as revision 21487 to svn://svn.ffmpeg.org/ffmpeg/trunk