Commit Graph

20631 Commits

Author SHA1 Message Date
Måns Rullgård 57c36bdcea intreadwrite: allow arch-specific 24-bit access macros
Originally committed as revision 19657 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-08-16 15:51:52 +00:00
Måns Rullgård e6956a6e48 ARM: first value loaded in AV_RN64 needs to be early-clobber
Originally committed as revision 19656 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-08-16 15:51:50 +00:00
Måns Rullgård 7f6d9b3089 version.sh: use standard sed syntax
Originally committed as revision 19655 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-08-16 15:51:47 +00:00
Ramiro Polla 9b734d44ab Indent libswscale:
- Use 4 spaces throughout for indentation;
- Fix inconsistent indentation;
- Indent function calls and declarations aligning arguments on multiple lines
  to the column after the opening parentheses;
- Align asm code to the column 4 spaces after the call to __asm__();
- Align cases in switch statements to the same column as "switch".

Originally committed as revision 29522 to svn://svn.mplayerhq.hu/mplayer/trunk/libswscale
2009-08-16 00:32:04 +00:00
Diego Biurrun 2513b2b436 Use pkg-config to generate the linker flags for libdirac.
We do the same for libschroedinger and already use pkg-config to generate
the CFLAGS for libdirac anyway, so there is no new dependency.
Thanks to Kovensky for reporting breakage on IRC.

Originally committed as revision 19654 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-08-15 12:35:15 +00:00
Anuradha Suraparaju d15f2e00c5 Fix bug caused by difference in stride and picture width.
When a frame is allocated using libschroedinger routines, the frame data size
does not match the actual frame size if the width is not a multiple of 16. So
we cannot do a straightforward memcpy of the frame returned by libschroedinger
into the FFmpeg picture as the stride differs from the width.

Fix this bug by allocating for the libschroedinger frame with the dimensions
in AVCodecContext within libavcodec and passing the frame to libschroedinger.

patch by Anuradha Suraparaju, anuradha rd.bbc.co uk

Originally committed as revision 19653 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-08-15 11:59:53 +00:00
Diego Biurrun 0ebe3b8e2b cosmetics: indentation, prettyprinting, K&R coding style
Originally committed as revision 19652 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-08-15 11:42:15 +00:00
Diego Biurrun c64380009d Drop some more useless braces around if/for constructs.
Originally committed as revision 19651 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-08-15 11:32:11 +00:00
Diego Biurrun e6cb49bfa9 Simplify another 'if' condition: Replace 'exp == 0' by '!exp'.
Originally committed as revision 19650 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-08-15 11:31:42 +00:00
Colin McQuillan bb937155ec Add a function that can apply an order 2 rational transfer function in-place.
This function will be used in the upcoming AMR-NB floating point decoder for
high-pass filtering.

Patch by Colin McQuillan ( m.niloc googlemail com )

Originally committed as revision 19649 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-08-15 11:22:55 +00:00
Diego Biurrun 735a38045a Remove useless braces around if/for/while expressions.
Originally committed as revision 19648 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-08-15 11:12:47 +00:00
Diego Biurrun 7c809dc3e2 Simplify 'if' condition statements.
Drop useless '!= 0' from 'exp != 0', replace 'exp == 0' by '!exp'.

Originally committed as revision 19647 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-08-15 11:02:50 +00:00
Reimar Döffinger cea0b5272d Fix start_display_time/end_display_time to be relative to packet pts in xsub decoder.
Makes no difference for the sample AVI files since they all have no packet pts values.

Originally committed as revision 19646 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-08-15 09:12:58 +00:00
Ramiro Polla ebe5dec22b Indent.
Originally committed as revision 29518 to svn://svn.mplayerhq.hu/mplayer/trunk/libswscale
2009-08-15 02:35:55 +00:00
Reimar Döffinger 4a3190ed40 Reindent.
Originally committed as revision 19645 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-08-15 00:58:26 +00:00
Reimar Döffinger fd35d6d097 Remove code that is now dead.
Originally committed as revision 19644 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-08-15 00:57:24 +00:00
Reimar Döffinger 90ea6fce1e Simplify initialization of AVSubtitle by using memset.
Originally committed as revision 19643 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-08-15 00:52:41 +00:00
Reimar Döffinger df47d24501 Make sure AVSubtitle is initialized, memset it to 0.
Originally committed as revision 19642 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-08-15 00:46:33 +00:00
Reimar Döffinger 266649a52f 100l, error out if subtitle encoder failed instead of writing an AVPacket
with ridiculous size and thus crashing.

Originally committed as revision 19641 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-08-15 00:37:31 +00:00
Reimar Döffinger ffbe087b58 Fix cmd_pos bounds check to avoid the overflow case.
Originally committed as revision 19640 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-08-15 00:02:42 +00:00
Carl Eugen Hoyos 84c1b149b2 Fix typo.
Originally committed as revision 19639 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-08-14 16:41:21 +00:00
Ramiro Polla 52154148f4 Vertical yuv -> yuv16 scaler.
Originally committed as revision 29517 to svn://svn.mplayerhq.hu/mplayer/trunk/libswscale
2009-08-14 15:52:40 +00:00
Peter Schlaile 562f685307 BGR32 MMX special convertor.
Patch by Peter Schlaile < peter at schlaile dot de >

Originally committed as revision 29515 to svn://svn.mplayerhq.hu/mplayer/trunk/libswscale
2009-08-14 02:15:28 +00:00
Måns Rullgård 9ae443f420 10L: remove instruction left over from debugging
Originally committed as revision 19638 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-08-14 01:05:48 +00:00
Måns Rullgård e814015d69 ARM: NEON optimised vorbis_inverse_coupling
12% faster Vorbis decoding on Cortex-A8.

Originally committed as revision 19637 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-08-14 01:02:06 +00:00
Ramiro Polla 737cbcde08 Indent.
Originally committed as revision 29512 to svn://svn.mplayerhq.hu/mplayer/trunk/libswscale
2009-08-13 19:03:14 +00:00
Ramiro Polla f51fb686de Introduce av_clip_uint16().
Originally committed as revision 19636 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-08-13 18:47:13 +00:00
Ramiro Polla d426ea20c9 swscale-example: Don't check for chroma planes in mono formats.
Originally committed as revision 29511 to svn://svn.mplayerhq.hu/mplayer/trunk/libswscale
2009-08-13 18:39:39 +00:00
Ramiro Polla 48f4c61252 MMX2 horizontal scaler: Determine code size at runtime.
Originally committed as revision 29510 to svn://svn.mplayerhq.hu/mplayer/trunk/libswscale
2009-08-13 18:31:07 +00:00
Ramiro Polla 0728778885 Protect mmx2 filter code buffers so they are not executable and writeable at
the same time (only mmap for now).

Originally committed as revision 29509 to svn://svn.mplayerhq.hu/mplayer/trunk/libswscale
2009-08-13 18:29:54 +00:00
Ramiro Polla b0e1343b85 Remove duplicate define (it is the same in the lum code).
Originally committed as revision 29508 to svn://svn.mplayerhq.hu/mplayer/trunk/libswscale
2009-08-13 18:28:55 +00:00
Ramiro Polla bcdedf675f Rename "funny" code to "mmx2 filter" code.
Originally committed as revision 29507 to svn://svn.mplayerhq.hu/mplayer/trunk/libswscale
2009-08-13 18:28:17 +00:00
Ramiro Polla 475f361308 Indent.
Originally committed as revision 29506 to svn://svn.mplayerhq.hu/mplayer/trunk/libswscale
2009-08-13 13:18:59 +00:00
Ramiro Polla e55ed689a2 swscale-example: Remove hack to end loop by setting variables to break.
Originally committed as revision 29505 to svn://svn.mplayerhq.hu/mplayer/trunk/libswscale
2009-08-13 11:48:06 +00:00
Ramiro Polla d7cf0a6ff7 swscale-example: Don't return random value from main().
Originally committed as revision 29504 to svn://svn.mplayerhq.hu/mplayer/trunk/libswscale
2009-08-13 11:34:27 +00:00
Ramiro Polla f5a9c4ee50 swscale-example: Check for Gray formats.
Originally committed as revision 29503 to svn://svn.mplayerhq.hu/mplayer/trunk/libswscale
2009-08-13 11:33:18 +00:00
Colin McQuillan d4d6ae1603 Add ff_celp_circ_addf() function to be used for sparse vector circular
convolution in the upcoming AMR-NB floating point decoder. The function scales
and adds a vector, that is lagged by some offset, to another vector with the
same number of elements.

Patch by Colin McQuillan ( m.niloc googlemail com )

Originally committed as revision 19634 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-08-12 19:54:28 +00:00
Shehzad Salim 51fdb6f08d Fix a memleak with win32 threads: the handle returned by _beginthreadex
must be closed (this differs from _beginthread).
Patch by Shehzad Salim (shehzadsalim gmail com)

Originally committed as revision 19633 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-08-12 13:25:37 +00:00
Peter Ross c0b88f1920 Moves the display of metadata to dump_format()
Originally committed as revision 19632 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-08-12 13:17:39 +00:00
Matti Hamalainen 5f71844e78 Parse ID3 tags in Musepack SV7 files
Patch by Matti Hamalainen (mhamalai@students.oamk.+358)
Thread: [PATCH] 6/6 Add parsing of ID3v2 tags in Musepack SV7 demuxer

Originally committed as revision 19631 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-08-11 17:28:56 +00:00
Matti Hamalainen 311f96a8cc Parse APE metadata tags in Musepack SV7 files
Patch by Matti Hamalainen (to get his mail address remove common endings from
 "mhamalainen@studentsnen.oamknen.finen")
Thread: [PATCH]5/6 Add APE tag metadata reading support in Musepack SV7 demuxer

Originally committed as revision 19630 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-08-11 17:18:10 +00:00
Kostya Shishkov 191e34cd67 Move APE tag parsing into separate module.
Based on patch by Matti Hamalainen (mhamalai<mot>students<punkt>oamk<punkt>)

Originally committed as revision 19629 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-08-11 17:08:09 +00:00
Matti Hamalainen 6984380c61 Do not use internals of ByteIOContext during APE tags parsing.
Patch by Matti Hamalainen, mail = reverse("if.kmao.stneduts@ialamahm")
Thread: [PATCH] 2/6 Fix broken APE tag key handling

Originally committed as revision 19628 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-08-11 16:55:49 +00:00
Matti Hamalainen 22667e40be Remove incorrect APE tag size adjustment based on flags.
Patch by Matti Hamalainen (mhamalai <luona> students <piste> oamk <piste> fi)
Thread: [PATCH] 1/6 Remove incorrect APE tag size adjustment based on flags

Originally committed as revision 19627 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-08-11 16:43:44 +00:00
Alexis Ballier ffd38d8b64 Use RGB5x5LE, RGB5x5BE, BGR5x5LE and BGR5x5BE instead of their native
endian counterparts.

Patch by Alexis Ballier, alexis D ballier gmail

Originally committed as revision 19626 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-08-11 16:18:55 +00:00
Loren Merritt d094052c8e fix a crash in SVQ1 with cmp!=sad
Originally committed as revision 19625 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-08-11 07:04:33 +00:00
Aurelien Jacobs 7a9c86e9f8 avidec: simplify, using av_rescale_q() instead of av_rescale()
Originally committed as revision 19624 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-08-10 20:53:32 +00:00
Aurelien Jacobs 2bf74f4955 actually define AVSEEK_FLAG_FRAME as described in avformat_seek_file() doc
Originally committed as revision 19623 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-08-10 20:48:05 +00:00
Aurelien Jacobs 8306be9c3b matroskadec: disable h264 parser as long as it totally mess up frame type
Originally committed as revision 19622 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-08-10 18:12:02 +00:00
Aurelien Jacobs 194d4b494b matroskadec: correctly parse flags for simpleblock frames
Originally committed as revision 19621 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-08-10 18:06:14 +00:00