Commit Graph

9056 Commits

Author SHA1 Message Date
David Conrad
170e6b14f7 Remove long unused ALTIVEC_USE_REFERENCE_C_CODE ifdef; all other references
to it were removed in r6606

Originally committed as revision 17997 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-03-16 03:56:42 +00:00
Benjamin Larsson
b217024706 Add cook channel count function, part of multichannel cook
Originally committed as revision 17992 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-03-15 19:36:45 +00:00
Benjamin Larsson
0eec287596 Add max allowed subpackets in multichannel cook
Originally committed as revision 17991 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-03-15 19:34:52 +00:00
Benjamin Larsson
d4b3d0408f save avctx in cook decoder context and use it for av_log
Originally committed as revision 17988 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-03-15 14:11:11 +00:00
Benjamin Larsson
7c119ced99 memset when category is >=7, part of cook multichannel
Originally committed as revision 17987 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-03-15 14:04:25 +00:00
Reimar Döffinger
16473c1376 1 km^3 cola, timecode_frame_start is a 64 bit integer, this resulted in
complete nonsense if timecode_frame_start was used on big-endian systems.

Originally committed as revision 17986 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-03-15 13:38:36 +00:00
Diego Biurrun
60a847b5db cosmetics: Remove file name from file header.
Originally committed as revision 17984 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-03-15 12:46:07 +00:00
Stefano Sabatini
7867978374 Expand "lsb" to "least significant bits".
Originally committed as revision 17978 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-03-15 10:37:25 +00:00
Stefano Sabatini
e5681ba0f2 Rename the (yet not defined) macros:
PIX_FMT_RGB565_BE -> PIX_FMT_RGB565BE
PIX_FMT_RGB565_LE -> PIX_FMT_RGB565LE
for consistency with the other pixfmts macros.

Originally committed as revision 17977 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-03-15 10:33:51 +00:00
Kostya Shishkov
0304109df6 Correct MV prediction for B-frame blocks in RV4 decoder
Originally committed as revision 17974 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-03-15 08:21:42 +00:00
Diego Biurrun
6d79971e69 prettyprinting cosmetics
Originally committed as revision 17962 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-03-14 19:12:20 +00:00
Diego Biurrun
f5b2476fd3 Add av_uninit to vsrcBuc variable to work around some
'may be used uninitialized' warnings.

Originally committed as revision 17961 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-03-14 19:11:56 +00:00
Diego Biurrun
b91813b6da Improve description of next_free_data_block_num struct member.
Originally committed as revision 17960 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-03-14 17:20:28 +00:00
Daniel Verkamp
5cf59463a2 Add 1bpp, 8bpp, 15bpp, and 16bpp support to BMP encoder.
Patch by Daniel Verkamp, daniel drv nu

Originally committed as revision 17954 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-03-13 23:36:38 +00:00
Michael Niedermayer
7ed673af71 Print timing_info on -debug 1.
Originally committed as revision 17949 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-03-13 14:10:49 +00:00
Kostya Shishkov
923ed9748c Fix direct and skip MB motion compensation in RV4:
two conditions were incomplete and zeroing motion
vectors was performed only on half of them.

Originally committed as revision 17947 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-03-13 13:48:44 +00:00
Gwenole Beauchesne
cd823ff950 Use av_mallocz() to initialize hwaccel_data_private.
Originally committed as revision 17942 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-03-11 08:25:00 +00:00
Vitor Sessak
e40ee25c49 Fix bug when elbg->utility_inc[elbg->numCB-1] == 1
Originally committed as revision 17926 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-03-10 17:38:16 +00:00
Gwenole Beauchesne
eb2c24232f Add REGISTER_HWACCEL() for hardware accelerators.
Originally committed as revision 17925 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-03-10 14:35:55 +00:00
Måns Rullgård
f49568a467 Fix unaligned access in ff_copy_bits()
Originally committed as revision 17921 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-03-09 23:27:38 +00:00
Reynaldo H. Verdejo Pinochet
d2af569710 Cosmetics from Kenan Gillet's 'extract and share weighted_vector_sumf'
patchset. Just a reindent.

Originally committed as revision 17917 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-03-09 22:00:35 +00:00
Reynaldo H. Verdejo Pinochet
95e8325771 Functional part Kenan Gillet's 'extract and share weighted_vector_sumf'
patchset. Idea is to share this common code between the AMR and QCELP
decoders.

Originally committed as revision 17916 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-03-09 21:55:24 +00:00
Gwenole Beauchesne
68e5d5235e Add private HW accel data infrastructure.
Originally committed as revision 17899 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-03-09 08:04:41 +00:00
Vitor Sessak
9aa2ab4276 Use FLG pseudo-random number generator in RoQ and ELBG
Originally committed as revision 17881 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-03-08 17:43:12 +00:00
Kostya Shishkov
74e2a0784e Make Musepack decoders use LFG pseudorandom generator
Originally committed as revision 17880 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-03-08 16:37:57 +00:00
Vitor Sessak
616e899c43 VQF demuxer
Originally committed as revision 17866 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-03-07 22:42:09 +00:00
Alex Converse
7d87e2ceac Re-indent after last commit.
Originally committed as revision 17861 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-03-06 22:37:21 +00:00
Alex Converse
51741a82c3 If we get an error from ff_aac_parse_header() we should not trust the header info that it provides.
Originally committed as revision 17860 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-03-06 22:36:24 +00:00
Alex Converse
bb5c0988ea Fix the channel allocation bug/assumption (issue 800).
Approved by Rob on IRC.

Originally committed as revision 17856 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-03-06 19:47:01 +00:00
Stephen Warren
49caac2413 Fix H.264 bitstream field log2_max_pic_order_cnt_lsb_minus4
if sps.poc_type == 0.

Patch by Stephen Warren, swarren nvidia com

Originally committed as revision 17854 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-03-06 08:03:26 +00:00
Justin Ruggles
f48b9304cd flacdec: Parse the metadata header in the raw FLAC demuxer.
Originally committed as revision 17852 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-03-06 01:31:10 +00:00
Justin Ruggles
5b63d33d7d flacdec: Add a shared function for parsing a FLAC metadata block header.
Originally committed as revision 17851 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-03-06 01:25:11 +00:00
Justin Ruggles
5a92cc663c pnm: Add missing 'else'. Fixes decoding for 16-bit pgm.
Originally committed as revision 17848 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-03-06 00:54:49 +00:00
Diego Biurrun
1b0f4920b8 Mark variables "j" and "structure" in execute_ref_pic_marking() as av_uninit
to work around false positive 'may be used uninitialized' warnings.

Originally committed as revision 17837 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-03-05 23:38:52 +00:00
Alex Converse
cfd937b081 AAC: hoist dependent gain lookup out of the loop
Originally committed as revision 17835 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-03-05 23:10:22 +00:00
Diego Biurrun
27127202d6 Remove duplicated MUL16 macro definition, fixes the warning:
libavcodec/ac3enc.c:119:1: warning: "MUL16" redefined

Originally committed as revision 17832 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-03-05 22:17:12 +00:00
Måns Rullgård
35690321a7 ARM: disable inline asm for armcc
Originally committed as revision 17831 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-03-05 21:20:13 +00:00
Ivan Schreter
b283ba26ce Add handling of frame position in the parser.
Originally committed as revision 17823 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-03-05 07:35:06 +00:00
Justin Ruggles
8d1e885fc2 flacdec: cosmetics: Add a comment with the data of the smallest FLAC
frame for future reference.

Originally committed as revision 17822 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-03-05 04:40:42 +00:00
Justin Ruggles
629fb5c46a flacdec: There is an even smaller FLAC frame size possibility.
Originally committed as revision 17821 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-03-05 02:36:17 +00:00
Justin Ruggles
5756bc7b02 flacdec: 10l to me. I miscalculated the smallest FLAC frame. It is 16
bytes, not 24.

Originally committed as revision 17820 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-03-05 02:21:35 +00:00
Justin Ruggles
bf1d7e285c flacdec: cosmetics: move 2 statements to immediately after frame
decoding.

Originally committed as revision 17819 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-03-05 01:23:53 +00:00
Justin Ruggles
fd6fd470ce flacdec: Simplify frame sync by not using the bitstream reader.
Originally committed as revision 17818 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-03-05 01:15:38 +00:00
Justin Ruggles
a8ec12bc2e flacdec: Split the metadata_parse() function into 2 separate functions,
parse_streaminfo() and get_metadata_size().

Originally committed as revision 17817 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-03-05 00:56:09 +00:00
Justin Ruggles
5ef4fa87b3 flacdec: Add a check for small buffer size. This ensures reading as
much of the frame header as possible without excluding the smallest
possible FLAC frame.  It also fixes a false positive warning message
that was being emitted at the end of decoding.

Originally committed as revision 17816 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-03-04 23:55:10 +00:00
Justin Ruggles
c51997297a flacdec: cosmetics: Use a more descriptive variable name for the number
of bytes read, instead of reusing 'i'.

Originally committed as revision 17815 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-03-04 23:24:44 +00:00
Justin Ruggles
549bccdbe4 flacdec: Handle 32 bps data when checking allocated size.
Originally committed as revision 17814 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-03-04 23:16:30 +00:00
Justin Ruggles
a3d2379bdd flacdec: Check curr_bps after wasted bits are subtracted.
Originally committed as revision 17813 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-03-04 23:10:40 +00:00
Stefano Sabatini
d3f766124a Fix typo: "i" -> "in".
Originally committed as revision 17812 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-03-04 21:50:10 +00:00
Ivan Schreter
70e01da3bf Add support for ct_type to correctly detect interlaced flag
Originally committed as revision 17811 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-03-04 19:37:19 +00:00