Commit Graph

21 Commits

Author SHA1 Message Date
rfelker f713cb0faf begin moving const filter data to .text/.rodata sections
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@24997 b3059339-0415-0410-9bf9-f77b7e298cf2
2007-11-09 06:58:32 +00:00
reimar 6a09e8e2ed Replace implicit use of fast_memcpy via macro by explicit use to allow
for future optimization.


git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@23475 b3059339-0415-0410-9bf9-f77b7e298cf2
2007-06-05 14:27:54 +00:00
diego 653d16096e Conditionally compile vf_bmovl.c instead of enclosing all of the file
in #ifdef HAVE_POSIX_SELECT.


git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@22770 b3059339-0415-0410-9bf9-f77b7e298cf2
2007-03-21 08:29:46 +00:00
diego d6a63f429f Invert logic for HAVE_NO_POSIX_SELECT to HAVE_POSIX_SELECT to be in line
with the other HAVE_ #defines throughout the codebase.


git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@22769 b3059339-0415-0410-9bf9-f77b7e298cf2
2007-03-21 08:24:20 +00:00
reimar 8e7945155c Use libavutil's av_clip* instead of unreadable MIN/MAX chaos.
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@22395 b3059339-0415-0410-9bf9-f77b7e298cf2
2007-03-01 12:54:08 +00:00
gpoirier a759a80397 in some cases, vf_bmovl produces junk due to source and
destination stride mismatch. The attached patch fixes it by replacing
memcpy with memcpy_pic.
Patch by Jindrich Makovicka
Original thread:
Date: Jul 12, 2006 6:35 PM
Subject: [MPlayer-dev-eng] [PATCH] vf_bmovl: respect destination stride


git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@19171 b3059339-0415-0410-9bf9-f77b7e298cf2
2006-07-23 14:35:04 +00:00
michael 02b199e5e9 passing pts through the filter layer (lets see if pts or cola comes out at the end)
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@17907 b3059339-0415-0410-9bf9-f77b7e298cf2
2006-03-21 21:26:42 +00:00
iive 2431c2a2cb include config.h before includes that need it
based on patch by Alan Curry <pacman at theworld com>


git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@17631 b3059339-0415-0410-9bf9-f77b7e298cf2
2006-02-16 10:08:03 +00:00
diego 36e1974be0 Unify include path handling, -I.. is in CFLAGS.
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@17013 b3059339-0415-0410-9bf9-f77b7e298cf2
2005-11-18 14:39:25 +00:00
reimar 86ea3067e9 close stream_fd on uninit. Fixes bugzilla bug #400.
Modified patch from T. Dekker {t dekker [at] student utwente nl}.


git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@16887 b3059339-0415-0410-9bf9-f77b7e298cf2
2005-11-01 18:22:18 +00:00
attila b164638fba fix bug when bmovl can't read the whole pic at once
patch by Detlev Droege <droege@uni-koblenz.de>


git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@11621 b3059339-0415-0410-9bf9-f77b7e298cf2
2003-12-10 12:28:20 +00:00
alex b28363ebaa correct handling of select ret=0, patch by Jonas Jensen <jbj@knef.dk>
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@11016 b3059339-0415-0410-9bf9-f77b7e298cf2
2003-10-05 14:17:29 +00:00
albeu ba4aeeb473 Fix SIG11 if no args given
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@10338 b3059339-0415-0410-9bf9-f77b7e298cf2
2003-06-26 11:17:28 +00:00
faust3 425b08fcc6 disable vf_bmovl for systems without posix select
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@9833 b3059339-0415-0410-9bf9-f77b7e298cf2
2003-04-04 20:12:24 +00:00
albeu 33b62af947 Add the new -vf option wich is the same as vop in reverse order.
Syntax is we decided, so you can give the nomes or not with both
vop and vf. vf take precedence over vop.


git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@9594 b3059339-0415-0410-9bf9-f77b7e298cf2
2003-03-15 18:01:02 +00:00
arpi 7d11e17fb3 The code for converting RGB to YUV in bmovl is slow because it uses
floating point arithmetic. The attached patch changes it to use integers
instead, giving about a 2x performance boost on animations.
Jonas Jensen <jbj@knef.dk>


git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@9132 b3059339-0415-0410-9bf9-f77b7e298cf2
2003-01-28 00:27:59 +00:00
arpi 8755d76ce2 This is a simple patch to change the alpha blending code in bmovl to use
integers instead of floating point. It speeds it up by about 400%
meaning that it's now possible to apply alpha blending to a whole
1024x768 image on my 1266MHz PIII.
Jonas Jensen <jbj@knef.dk>


git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@9111 b3059339-0415-0410-9bf9-f77b7e298cf2
2003-01-26 17:42:24 +00:00
rfelker 399b33e974 may not be perfect but it's certainly a start. feel free to change
message levels or error behavior but DON'T exit() from filters!


git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@8879 b3059339-0415-0410-9bf9-f77b7e298cf2
2003-01-10 21:44:28 +00:00
arpi 76b9620138 updated comments
patch by Per Wigren <wigren@home.se>


git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@7896 b3059339-0415-0410-9bf9-f77b7e298cf2
2002-10-24 17:39:12 +00:00
arpi 044b2d550e time.h required for select()
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@7859 b3059339-0415-0410-9bf9-f77b7e298cf2
2002-10-23 11:20:07 +00:00
arpi a80ed8913a -vop bmovl - BitMap OVerLay videofilter for MPlayer
patch by Per Wigren <wigren@home.se>
TODO: -dr, MMX opt, fix alpha-only png, other colorspaces


git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@7856 b3059339-0415-0410-9bf9-f77b7e298cf2
2002-10-23 00:26:27 +00:00