generic-library/ffmpeg
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity
16,000 Commits 24 Branches 241 Tags 173 MiB
8b1fd554e1
Commit Graph

7777 Commits

Author SHA1 Message Date
Michael Niedermayer
8b1fd554e1 Make MBAFF temporal direct mode closer to the spec. 
Fixes at least:
CAMANL2_TOSHIBA_B
CVMANL2_TOSHIBA_B
camp_mot_mbaff0_full

Originally committed as revision 15375 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-09-21 02:39:09 +00:00
Michael Niedermayer
8f56e219b2 Move direct_dist_scale_factor() and direct_ref_list_init() after fill_mbaff_ref_list() 
as future changes will need the mbaff lists inited.

Originally committed as revision 15374 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-09-21 02:23:51 +00:00
Michael Niedermayer
078f42ddae Set poc for field references in MBAFF. 
Originally committed as revision 15373 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-09-21 02:19:45 +00:00
Stefano Sabatini
2c006f4380 Log an error message in case of invalid number of channels. 
Originally committed as revision 15371 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-09-20 11:13:14 +00:00
Michael Niedermayer
a85d0c84a4 Remove fixme that is based on the assumption of !direct_8x8_inference && MBAFF. 
Originally committed as revision 15370 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-09-19 18:53:31 +00:00
Michael Niedermayer
6a855aae94 Remove check and error message for "MBAFF + !direct_8x8_inference is not implemented" 
because this mode does not exist, H.264-2007 says "When frame_mbs_only_flag is
equal to 0, direct_8x8_inference_flag shall be equal to 1."

Originally committed as revision 15369 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-09-19 18:49:17 +00:00
Michael Niedermayer
96b52f9e04 Make pps/sps scaling_matrix initialization closer to the spec. 
This does not affect any of the conformance streams.

Originally committed as revision 15365 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-09-18 16:36:19 +00:00
Robert Swain
341b28c0d2 Zero 'idx' for each iteration of the coupling gain loop and as it is only used 
within this loop, we can move its declaration there too. This fixes bitstream
desync when decoding streams containing CCEs.

Based on a patch by Alex Converse (alex converse gmail com)

Originally committed as revision 15364 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-09-18 16:02:51 +00:00
Robert Swain
f80a8ca5cb Cosmetics: missing brackets 
Originally committed as revision 15363 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-09-18 15:48:37 +00:00
Alex Converse
c8947a5612 Cast the return value of get_bits() to signed integer such that the subtraction 
is conducted as a signed calculation.

Patch by Alex Converse (alex converse gmail com)

Originally committed as revision 15362 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-09-18 15:35:08 +00:00
Michael Niedermayer
d464bceff2 Ensure that extradata is not parsed twice. 
Fixes FRExt1_Panasonic.

Originally committed as revision 15361 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-09-18 15:13:15 +00:00
Kostya Shishkov
3a9a15c4f3 add newline at the end of message 
Originally committed as revision 15360 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-09-18 14:08:44 +00:00
Kostya Shishkov
e6ab862368 ZMBV encoder forgot to fill coded frame pointer 
Originally committed as revision 15359 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-09-18 13:17:44 +00:00
Kostya Shishkov
44aa9771c9 Factorize out code used for MS RLE format decoding in different decoders. 
Originally committed as revision 15356 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-09-18 05:20:54 +00:00
David Conrad
f7e5b0cc8f Remove duplication of TRANSPOSE8 macro, also removing some redundant load/stores 
Originally committed as revision 15355 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-09-18 03:08:33 +00:00
David Conrad
7b16a97969 Make a separate macro for writing an entire block from registers to memory 
Originally committed as revision 15354 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-09-18 03:08:13 +00:00
David Conrad
3f429d11da Factorize SSE2_(Row|Column)_IDCT into one macro 
Originally committed as revision 15353 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-09-18 03:07:41 +00:00
David Conrad
3c1ebfd56d Cosmetics: remove excessive blank lines 
Originally committed as revision 15351 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-09-17 20:52:54 +00:00
David Conrad
8cfd78ce8f Ensure MMX/SSE2 VP3 IDCT selection isn't disabled when only Theora is enabled 
Originally committed as revision 15350 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-09-17 19:49:31 +00:00
David Conrad
c3c5bba1e2 Cosmetics: reindent 
Originally committed as revision 15349 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-09-17 19:38:44 +00:00
David Conrad
f7fc9e499c Rewrite SSE2 VP3 IDCT in inline asm 
Originally committed as revision 15347 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-09-17 19:30:51 +00:00
David Conrad
437e3f4dbc Rewrite MMX VP3 IDCT in inline asm 
Originally committed as revision 15346 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-09-17 19:30:33 +00:00
David Conrad
ccd3ec82b8 MMX/SSE2 VP3 IDCT are bitexact now that the dequantization matrices are permutated correctly 
Originally committed as revision 15345 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-09-17 19:30:03 +00:00
David Conrad
7f065b36c8 Take idct permutation into account when initializing dequantization matrices 
Originally committed as revision 15344 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-09-17 19:29:40 +00:00
Vitor Sessak
faf6d6964b Use SAMPLE_FMT_FLT instead of doing the float->int conversion in the 
decoder. This changes the output very little and any difference should
be inaudible.

Originally committed as revision 15343 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-09-17 19:26:37 +00:00
Vladimir Voroshilov
00cc0ce91e Second-stage gain and Moving Average Predictor codebooks. 
Originally committed as revision 15342 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-09-17 06:13:07 +00:00
Vladimir Voroshilov
1a18912e49 Another set of approved changes for G.729: includes, formats structure, 
several parts of main decoding routine and initialization code.

Originally committed as revision 15341 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-09-17 06:05:12 +00:00
Alex Converse
aac0eda407 Validate pulse position and error out if an invalid position is encountered. 
Patch by Alex Converse (alex converse gmail com)

Originally committed as revision 15340 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-09-16 15:59:43 +00:00
Michael Niedermayer
1d4113d053 Only print "invalid new backstep" when it is really invalid. 
Fixes issue541.

Originally committed as revision 15339 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-09-16 01:49:32 +00:00
Michael Niedermayer
740c14dbd7 Fix regression test failure caused by golomb limit not being considered 
in r15334.

Originally committed as revision 15338 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-09-16 01:24:12 +00:00
Baptiste Coudurier
0bc4c436aa sync to latest x264 git, field was renamed 
Originally committed as revision 15337 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-09-16 01:13:55 +00:00
Michael Niedermayer
bf3a971c12 Make decoder more robust by default against broken encoders. 
Fixes issue540.

Originally committed as revision 15336 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-09-16 00:31:32 +00:00
Michael Niedermayer
631e8afb8c Correct the threshold of get_ur_golomb_jpegls() at which the optimized 
case is used.
Fix issue245.

Originally committed as revision 15334 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-09-15 20:30:22 +00:00
Justin Ruggles
8fd64adf08 cosmetics: fix indentation after r15321 
Originally committed as revision 15332 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-09-14 22:27:47 +00:00
Justin Ruggles
6682ae42a4 write actual min and max frame size to FLAC header. update regression test checksum. 
Originally committed as revision 15331 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-09-14 22:25:50 +00:00
Diego Biurrun
dd857bfdf7 cosmetics: Fix indentation. 
Originally committed as revision 15330 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-09-14 22:20:14 +00:00
Diego Biurrun
3fe5ae2b3a Remove useless #ifdef around #include. 
Originally committed as revision 15329 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-09-14 22:18:50 +00:00
Justin Ruggles
a9f8587e15 write MD5 checksum to FLAC header. update regression tests. 
Originally committed as revision 15326 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-09-14 21:39:54 +00:00
Justin Ruggles
e1a12934c5 write number of samples in FLAC extradata. 
based on a patch by Mathieu Velten (matmaul gmail com).

Originally committed as revision 15324 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-09-14 20:00:36 +00:00
Justin Ruggles
0605f5c86b Return error when trying to decode non-grayscale 16-bit PNM images. 
Fixes issue 566.

Originally committed as revision 15321 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-09-14 15:50:59 +00:00
Robert Swain
70735a3f9e Reindent after last commit 
Originally committed as revision 15315 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-09-13 20:52:57 +00:00
Alex Converse
5143684891 Apply pulses to unscaled coefficients rather than scaled. 
Slight reworking of a patch by Alex Converse (alex converse gmail com)

Originally committed as revision 15314 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-09-13 20:51:36 +00:00
Robert Swain
408992bae2 Operands of '+' are not necessarily evaluated in order so r15311 was pointless. 
Splitting the assignment onto two lines should resolve the order issue.

Originally committed as revision 15313 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-09-13 20:22:31 +00:00
Alex Converse
19645d742e Correct order of parsing for pulse scalefactor band and offset to match the 
specification.

Patch by Alex Converse (alex converse gmail com)

Originally committed as revision 15311 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-09-13 18:47:43 +00:00
Vitor Sessak
9547cadb8d Cosmetics: line breaks 
Originally committed as revision 15308 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-09-13 16:56:58 +00:00
Vitor Sessak
111734de09 Simplify: use a single history buffer for gain and a single one for 
speech instead of having two for each in the context.

Originally committed as revision 15307 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-09-13 16:49:05 +00:00
David Conrad
7f0d242b14 Fix compilation on Mac OS X 10.4: Defining _POSIX_C_SOURCE hides the u_char & 
similar typedefs that sysctl.h needs. Since sysctl() itself isn't POSIX
undefining _POSIX_C_SOURCE for check_altivec.c seems the best way to fix this.
patch by David Conrad lessen42 at gmail com

Originally committed as revision 15306 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-09-13 13:18:35 +00:00
Robert Swain
35445d29f5 Reindent after last commit 
Originally committed as revision 15304 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-09-12 14:33:35 +00:00
Alex Converse
5167364777 Only read some of TNS bitstream data in the case that the TNS filter order is 
non-zero as per the specification.

Patch by Alex Converse (alex converse gmail com)
Fixes another issue pertaining to issue632

Originally committed as revision 15303 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-09-12 14:32:49 +00:00
Vitor Sessak
9c10ab6cd1 Cosmetics: align spec references 
Originally committed as revision 15300 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-09-11 19:10:26 +00:00