Plain C, x86-32 and -64 have been tested and should work, other
archs that had asm optmizations in swscale likely will need some fixes
to either fall back on C if SWS_BITEXACT is set or make the asm match C.
This also disables the PAL8 test as neither swscale nor the old scaler
really support PAL8 output, imgconvert supported a fixed 666 palette
as output and swscale supports fixed 884 and 422.

Originally committed as revision 15305 to svn://svn.ffmpeg.org/ffmpeg/trunk

bitexact ones.

Originally committed as revision 27597 to svn://svn.mplayerhq.hu/mplayer/trunk/libswscale

Originally committed as revision 27596 to svn://svn.mplayerhq.hu/mplayer/trunk/libswscale

are 0 if bitexact mode is requested.

Originally committed as revision 27595 to svn://svn.mplayerhq.hu/mplayer/trunk/libswscale

bitexactness to C.

Originally committed as revision 27594 to svn://svn.mplayerhq.hu/mplayer/trunk/libswscale

this decreases the difference between C and MMX, its also faster.

Originally committed as revision 27593 to svn://svn.mplayerhq.hu/mplayer/trunk/libswscale

Originally committed as revision 27592 to svn://svn.mplayerhq.hu/mplayer/trunk/libswscale

thus revert the respective optimization. The table generator code
has to be rewritten anyway one day by some volunteer because its
not LGPL, fixing the GPL table generator thus seems like wasted time.

Originally committed as revision 27591 to svn://svn.mplayerhq.hu/mplayer/trunk/libswscale

Originally committed as revision 27590 to svn://svn.mplayerhq.hu/mplayer/trunk/libswscale

Originally committed as revision 27589 to svn://svn.mplayerhq.hu/mplayer/trunk/libswscale

Thats what happens when one does not do the full set of tests before each commit
and just quickly goes over the diff thinking, "hey its a trivial change".

Originally committed as revision 27588 to svn://svn.mplayerhq.hu/mplayer/trunk/libswscale

Originally committed as revision 27587 to svn://svn.mplayerhq.hu/mplayer/trunk/libswscale

Originally committed as revision 27586 to svn://svn.mplayerhq.hu/mplayer/trunk/libswscale

Originally committed as revision 15304 to svn://svn.ffmpeg.org/ffmpeg/trunk

non-zero as per the specification.

Patch by Alex Converse (alex converse gmail com)
Fixes another issue pertaining to issue632

Originally committed as revision 15303 to svn://svn.ffmpeg.org/ffmpeg/trunk

user asks for accurate rounding.

Originally committed as revision 27585 to svn://svn.mplayerhq.hu/mplayer/trunk/libswscale

because they do not accurately round.

Originally committed as revision 27584 to svn://svn.mplayerhq.hu/mplayer/trunk/libswscale

no need to duplicate it

Originally committed as revision 15302 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 15301 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 27583 to svn://svn.mplayerhq.hu/mplayer/trunk/libswscale

Originally committed as revision 15300 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 15299 to svn://svn.ffmpeg.org/ffmpeg/trunk

remove duplicate code" thread on ML.

Originally committed as revision 15298 to svn://svn.ffmpeg.org/ffmpeg/trunk

SDP and RTSP into a new function. See discussion on ML in "[PATCH] rtsp
cleanup part 1: remove duplicate code" thread.

Originally committed as revision 15297 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 27582 to svn://svn.mplayerhq.hu/mplayer/trunk/libswscale

Originally committed as revision 27581 to svn://svn.mplayerhq.hu/mplayer/trunk/libswscale

Originally committed as revision 27580 to svn://svn.mplayerhq.hu/mplayer/trunk/libswscale

Originally committed as revision 15296 to svn://svn.ffmpeg.org/ffmpeg/trunk

Unlike H.264 it does not affect picture quality on any files I've watched.

Originally committed as revision 15295 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 27579 to svn://svn.mplayerhq.hu/mplayer/trunk/libswscale

Originally committed as revision 27578 to svn://svn.mplayerhq.hu/mplayer/trunk/libswscale

Originally committed as revision 27577 to svn://svn.mplayerhq.hu/mplayer/trunk/libswscale

pulse is subtracted, else it is added. Also avoid a divide by 0.

Based on a patch by Alex Converse (alex converse gmail com)
Fixes part of issue632

Originally committed as revision 15294 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 15293 to svn://svn.ffmpeg.org/ffmpeg/trunk

only affects the C code while mmx uses different tables.

Originally committed as revision 27576 to svn://svn.mplayerhq.hu/mplayer/trunk/libswscale

completely sure why this was at such incorrect value, but i could
not find any problems when it was set correctly.

Originally committed as revision 27575 to svn://svn.mplayerhq.hu/mplayer/trunk/libswscale

nearly never occuring 0.0, 1.0.

Originally committed as revision 27574 to svn://svn.mplayerhq.hu/mplayer/trunk/libswscale

Originally committed as revision 27573 to svn://svn.mplayerhq.hu/mplayer/trunk/libswscale

Originally committed as revision 15292 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 27569 to svn://svn.mplayerhq.hu/mplayer/trunk/libswscale