xine-lib/src/post/deinterlace/plugins, branch 1-rc7

Ryan Tadlock's patch for compensating the field delay of some deinterlacing algorithms

2004-07-17T16:35:16+00:00

CVS patchset: 6801
CVS date: 2004/07/17 16:35:16

fix compiler warning (missing prototype)

2004-04-09T14:31:28+00:00

CVS patchset: 6359
CVS date: 2004/04/09 14:31:28

tvtime deinterlacing algorithms assumed top_field_first=1

2004-04-09T02:57:05+00:00

top_field_first=0 (aka bottom_field_first) should now work as expected

CVS patchset: 6356
CVS date: 2004/04/09 02:57:05

my gcc (partly 3.4 already) optimizes these away, because they are only used

2004-02-12T20:53:31+00:00

inside inline assembler (which the compiler does not recognize); so actually
the code is wrong (the asm parts should list these as inputs), but telling
the compiler to keep them is the easier fix

CVS patchset: 6144
CVS date: 2004/02/12 20:53:31

wonder why Mike isn't complaining about C++ style comments, any more...

2004-01-05T12:15:55+00:00

CVS patchset: 6003
CVS date: 2004/01/05 12:15:55

DOS/Win CRs are forbidden, verboten, interdit

2004-01-05T01:47:26+00:00

CVS patchset: 6001
CVS date: 2004/01/05 01:47:26

better MANGLE from ffmpeg

2004-01-02T20:53:43+00:00

CVS patchset: 5976
CVS date: 2004/01/02 20:53:43

my small contribution to the cygwin port ;-)

2004-01-02T20:47:03+00:00

CVS patchset: 5975
CVS date: 2004/01/02 20:47:03

prefetching seems to improve performance as well

2003-12-08T17:23:41+00:00

CVS patchset: 5876
CVS date: 2003/12/08 17:23:41

sfence is not a valid MMX instruction.

2003-12-08T17:06:53+00:00

provide a new MMXEXT version. about 15% faster here.

CVS patchset: 5875
CVS date: 2003/12/08 17:06:53