in a library.
Should not change anything for PPC, the autodetection is currently pointless due
to other code being compiled with -maltivec as well (and detection for OSX and
AmigaOS remains in place).
SPARC binaries built with VIS support can now only run on systems with VIS.
Originally committed as revision 10648 to svn://svn.ffmpeg.org/ffmpeg/trunk
optimizations and no general PPC optimizations. Instead make the file be
compiled conditional to HAVE_ALTIVEC.
Originally committed as revision 10637 to svn://svn.ffmpeg.org/ffmpeg/trunk
interpolations, since they are always 16-bytes aligned in practice.
Add asserts to ease narrowing down potential image corructions on exotic plateforms
Based on a patch by Mauricio Alvarez % lokifo A gmail P com %
Original thread:
Date: Jun 26, 2007 1:07 PM
Subject: Re: [FFmpeg-devel] [PATCH] h264 luma interpolation 8x8 for altivec
Originally committed as revision 10238 to svn://svn.ffmpeg.org/ffmpeg/trunk
part 1 of h264 luma interpolation 8x8 for altivec contributed by
Mauricio Alvarez % lokifo A gmail P com %
Original thread:
Date: Jun 26, 2007 8:15 PM
Subject: Re: [FFmpeg-devel] [PATCH] h264 luma interpolation 8x8 for altivec
Originally committed as revision 10090 to svn://svn.ffmpeg.org/ffmpeg/trunk
In h264_deblock_q1, the result of the deblock needs to be kept to
be used in future deblocks, so return this value now.
Also change the sign of tc0 vector: It is really a signed value, so
treat it as such until after the >=0 check;
then, at that point, after being masked, it can be treated as unsigned.
Patch by Graham Booker % gbooker A tamu P edu%
Originally committed as revision 9349 to svn://svn.ffmpeg.org/ffmpeg/trunk
all 255s, and then doing the subtraction, nor of the vector with itself: saves
one instruction and a register.
Patch by Graham Booker % gbooker A tamu P edu%
Originally committed as revision 9340 to svn://svn.ffmpeg.org/ffmpeg/trunk
( http://www.pennfans.net/files/videos/Penn&Teller.on.The.View.mp4 )
with current Altivec implementation of loopfilter, while others are fine.
Let's disable it until we iron this bug out.
Originally committed as revision 9317 to svn://svn.ffmpeg.org/ffmpeg/trunk
(there's still 2 more, but there's burried into several levels of macros, so it's hard to narrow them down)
Originally committed as revision 9265 to svn://svn.ffmpeg.org/ffmpeg/trunk
patch by Graham Booker % perian A cod3r P com% with some minor fixes by me.
historic of the patch: http://trac.perian.org/ticket/113
Original thread:
Date: May 11, 2007 9:45 PM
Subject: [FFmpeg-devel] [PATCH] Altivec version of-altivec h264_h-v_loop_filter_luma
Originally committed as revision 9264 to svn://svn.ffmpeg.org/ffmpeg/trunk
instead of compiler-dependent __attribute__((aligned(16)))
Origiginal thread:
Date: May 17, 2007 12:30 AM
Subject: [PATCH] Use DECLARE_ALIGNED_16 in libavcodec/ppc/
Originally committed as revision 9047 to svn://svn.ffmpeg.org/ffmpeg/trunk
include paths in the source files.
mostly from a patch by Ronald S. Bultje, rbultje ronald.bitfreak net
Originally committed as revision 9034 to svn://svn.ffmpeg.org/ffmpeg/trunk
Reimar Döffinger
Diego Biurrun
Luca Barbato
Guillaume Poirier
Måns Rullgård
Graham Booker
Sigbjorn Skjaeret
Ronald S. Bultje