generic-library/ffmpeg
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity
17,076 Commits 24 Branches 241 Tags 173 MiB
c66216ed5d
Commit Graph

8390 Commits

Author SHA1 Message Date
Guillaume Poirier
337e3fd990 Disable usage of ff_h264_idct_add_altivec since AltiVec versions of h264_idct_add16, 
h264_idct_add16intra, h264_idct_add8 need to be implemented.

Add C version of ff_h264_idct8_dc_add in AltiVec so that ff_h264_idct8_add_altivec
can be used.

Originally committed as revision 16311 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-25 18:27:49 +00:00
Diego Biurrun
0fa5f24c3b Change mplayerhq.hu references to ffmpeg.org where appropriate. 
Originally committed as revision 16310 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-25 15:49:50 +00:00
Diego Biurrun
ddaf298cee Fix sample URL. 
Originally committed as revision 16309 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-25 15:49:02 +00:00
avcoder
bb0bfcd5ac The VC-1/WMV3 decoders depend on ENABLE_MSMPEG4_DECODER. 
patch by avcoder, ffmpeg gmail com

Originally committed as revision 16308 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-25 13:26:18 +00:00
Loren Merritt
1cca8d2423 flatten an array, since gcc fails at optimizing multidimensional arrays 
h264_idct8_add_c: 780 -> 735 cycles on conroe

Originally committed as revision 16307 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-25 01:20:37 +00:00
Michael Niedermayer
802e91466d export refs 
Originally committed as revision 16306 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-24 18:39:23 +00:00
Michael Niedermayer
53d55e9ea3 Allow decoders to set AVCodecContext.refs. 
Originally committed as revision 16305 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-24 17:38:27 +00:00
Michael Niedermayer
62ea19c061 Optimize pred_pskip_motion() 
7 cpu cycles faster on pentium dual.

Originally committed as revision 16304 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-24 03:23:38 +00:00
Michael Niedermayer
ebfb0449d3 Indent 
Originally committed as revision 16302 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-23 22:06:16 +00:00
Michael Niedermayer
05d3fd44d7 Integrate get_te0_golomb() calls into the code, this allows some checks 
to be avoided and the function is pretty small.
3% speedup, though this is probably due to changed inlining and not directly
this change.

Originally committed as revision 16301 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-23 22:04:34 +00:00
Michael Niedermayer
9963b332d1 Use get_ue_golomb_31() where possible, almost all are just in headers 
though.

Originally committed as revision 16300 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-23 21:12:19 +00:00
Michael Niedermayer
0be514625a Update golomb decoding tables so that get_*_golomb_31() return some 
larger values for too large values.

Originally committed as revision 16299 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-23 21:07:36 +00:00
Michael Niedermayer
0e921e97c1 Add get_ue_golomb_31() 
Originally committed as revision 16298 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-23 21:05:47 +00:00
Michael Niedermayer
2f2893ce9b indent 
Originally committed as revision 16296 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-23 19:12:50 +00:00
Michael Niedermayer
d42fc4a8ca Use the new VLC table for the first non trailing coeff too. 
Sadly only 5 cycles faster here on pentium dual. So maybe the
complexity is not worth it and this should be reverted ...

Originally committed as revision 16295 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-23 19:10:46 +00:00
Michael Niedermayer
593af7cdda Optimize esc removal code. 
Originally committed as revision 16294 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-23 18:31:44 +00:00
Kostya Shishkov
ffedb5c19a 352l: correct calculating number of bits for storing macroblock offset in RV3/4 
Originally committed as revision 16293 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-23 18:27:06 +00:00
Michael Niedermayer
2d76bf391a Indent 
Originally committed as revision 16292 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-23 17:50:36 +00:00
Michael Niedermayer
8140955d39 unified CAVLC level decoding LUT. 
Quite a bit faster (HPCVMOLQ_BRCM_B.264 was 3% faster here)

Originally committed as revision 16291 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-23 17:49:38 +00:00
Michael Niedermayer
5cb5023c78 Make h264 parser 50% faster. 
Originally committed as revision 16286 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-23 02:34:58 +00:00
Michael Niedermayer
abb27cfb24 100l, I broke H.264 again, forgot one hunk. 
Thanks to FATE for finding it.

Originally committed as revision 16285 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-23 01:11:56 +00:00
Michael Niedermayer
e08715d391 Optimize 0 0 0-3 search, 45% faster on pentium dual. 
Originally committed as revision 16284 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-23 00:38:45 +00:00
Andreas Öman
11c23b64ac Revert r16257: 
Include "libavutil/common.h" where we use llrint() in case ffmpeg's
own llrint() is to be used.

Originally committed as revision 16282 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-22 22:39:55 +00:00
Michael Niedermayer
ec3686e889 Simplify decode_cabac_mb_ref() a little bit, 2 cpu cycles faster on 
pentium dual.

Originally committed as revision 16279 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-22 17:14:13 +00:00
Michael Niedermayer
26695973c7 Indent 
Originally committed as revision 16278 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-22 16:56:05 +00:00
Michael Niedermayer
b68a455313 inline decode_cabac_mb_type for I & P frames, 9 cycles faster on pentium dual. 
Originally committed as revision 16277 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-22 16:46:17 +00:00
Michael Niedermayer
1952ac3713 Negate 2 more variables, 1 cpu cycle faster on pentium dual. 
Originally committed as revision 16276 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-22 16:20:13 +00:00
Michael Niedermayer
03a035e059 Simplify if/else, no speed change 
Originally committed as revision 16275 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-22 16:14:06 +00:00
Michael Niedermayer
6f3c50f2f9 Negate a few variables, this simplifies the code and makes it 5 cycles faster 
on pentium dual.

Originally committed as revision 16274 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-22 16:10:35 +00:00
Michael Niedermayer
60c6ba7aea Simplify ifs(), 8 cpu cycles faster on pentium dual 
Originally committed as revision 16273 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-22 15:54:27 +00:00
Michael Niedermayer
127a20e3b8 Simplify if(), 3 cpu cycles faster in pentium dual. 
Originally committed as revision 16272 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-22 15:53:00 +00:00
Diego Biurrun
a6493a8fbd Rename libavcodec/i386/ --> libavcodec/x86/. 
It contains optimizations that are not specific to i386 and
libavutil uses this naming scheme already.

Originally committed as revision 16270 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-22 09:12:42 +00:00
Kostya Shishkov
108c96d5e7 Decode TIFF image only after all tags have been decoded 
Originally committed as revision 16268 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-22 06:50:18 +00:00
Kostya Shishkov
8eedffde31 Some TIFFs declare -1 for the single strip height 
Originally committed as revision 16267 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-22 06:47:51 +00:00
Kostya Shishkov
89b4ec5c33 Add monochrome TIFF support 
Originally committed as revision 16266 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-22 06:43:14 +00:00
Kostya Shishkov
4386ed6922 Calculate line size variable correctly for lower bitdepths and use it for raw data copying 
Originally committed as revision 16265 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-22 06:39:31 +00:00
Kostya Shishkov
4a2f51de34 Register TIFF tags for CCITT Group 3 and 4 compression options 
Originally committed as revision 16264 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-22 06:36:10 +00:00
Diego Biurrun
bef05f05e4 Remove a bunch of unused variables. 
Originally committed as revision 16263 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-22 00:10:36 +00:00
Diego Biurrun
ebb160a0b0 Use ARCH_X86_32 instead of !ARCH_X86_64, it is more straightforward. 
Originally committed as revision 16262 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-22 00:03:30 +00:00
Anuradha Suraparaju
64bfc5845b Fix incorrectly constructed Dirac parse units that caused A/V sync loss. 
Fixes issue 694.
patch by Anuradha Suraparaju, anuradha rd.bbc.co uk

Originally committed as revision 16261 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-22 00:01:39 +00:00
Diego Biurrun
c4ff7c53e0 Do not use full include path for headers in the same directory. 
Originally committed as revision 16260 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-21 23:56:55 +00:00
Michael Niedermayer
c96bd21227 Change mpeg audio parser so it only sets frame_size, channels and bit_rate 
after it has a few valid frames.
Fixes issue762

Originally committed as revision 16259 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-21 23:50:16 +00:00
Andreas Öman
1af34f094e Include "libavutil/common.h" where we use llrint() in case ffmpeg's 
own llrint() is to be used.

Originally committed as revision 16257 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-21 21:53:42 +00:00
Michael Niedermayer
ed4c638a07 Fix PTS returned by the demuxers for mpeg4 without a decoder. 
More precissely fix has_b_frames.
Fixes issue757

Originally committed as revision 16256 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-21 21:47:29 +00:00
Michael Niedermayer
c62da7972b Move CODEC_FLAG_LOW_DELAY into ff_mpeg4_decode_picture_header(). 
Originally committed as revision 16255 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-21 21:30:36 +00:00
Michael Niedermayer
befc8fe086 Remove useless code. 
Originally committed as revision 16253 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-21 17:49:40 +00:00
Michael Niedermayer
c212fb0cb1 Only execute clear_blocks() when needed. 
+0.3% speedup for both aladin & cathedral.

Originally committed as revision 16252 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-21 15:58:42 +00:00
Robert Swain
897c1703d0 AAC: Reindent after last commit 
Originally committed as revision 16251 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-21 05:32:55 +00:00
Alex Converse
aab5413323 AAC: Fix AAC prediction when used in conjunction with the CPE 
Patch by Alex Converse (alex converse gmail com)

Originally committed as revision 16250 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-21 05:28:14 +00:00
Alex Converse
99665a21f4 AAC: Don't write an illegal sampling_index in a PCE to the MPEG4AudioConfig 
member

Patch by Alex Converse (alex converse gmail com)

Originally committed as revision 16249 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-21 05:08:32 +00:00
Michael Niedermayer
66c07ca96f Optimize get_dct8x8_allowed(). 
30 cpu cycles faster on pentium dual.

Originally committed as revision 16248 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-20 23:11:30 +00:00
Kostya Shishkov
6beb8b2612 RV30 decoder 
Originally committed as revision 16247 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-20 18:06:01 +00:00
Kostya Shishkov
541e8f3bec RV30 loop filter 
Originally committed as revision 16245 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-20 06:09:54 +00:00
Reynaldo H. Verdejo Pinochet
3f16ed15df Cosmetics by Kenan Gillet. Part 1 of 3 of his 'qcelp: silence handling' 
changeset.

Originally committed as revision 16244 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-20 02:38:34 +00:00
Stefano Sabatini
64e74b68a4 Put under #if LIBAVCODEC_VERSION_MAJOR < 53 the deprecated functions 
av_set_string() and av_set_string2(), they will be dropped at the next
major bump.

Originally committed as revision 16243 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-19 22:37:56 +00:00
Stefano Sabatini
f16dd7e65a Remove calls to deprecated av_set_string2() with calls to 
av_set_string3().

Originally committed as revision 16241 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-19 22:29:30 +00:00
Jason Garrett-Glaser
aac8b76983 H.264 loopfilter speed tweaks 
Originally committed as revision 16240 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-19 14:49:17 +00:00
Jason Garrett-Glaser
a22eff36f9 Port x264 deblocking code to libavcodec. 
This includes SSE2 luma deblocking code and both MMXEXT and SSE2 luma
intra deblocking code for H.264 decoding. This assembly is available
under --enable-gpl and speeds decoding of Cathedral by 7%.

Originally committed as revision 16239 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-19 13:45:13 +00:00
Michael Niedermayer
6a1846eb4a Document new idct requirements. 
Originally committed as revision 16237 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-19 09:34:18 +00:00
Michael Niedermayer
a5805aa9d1 Fix decoding with the plain C idcts of 
FRExt/HPCAMOLQ_BRCM_B
FRExt/HPCAQ2LQ_BRCM_B
FRExt/HPCVMOLQ_BRCM_B

Originally committed as revision 16236 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-19 09:21:54 +00:00
Kostya Shishkov
bbc888d0c5 a table for upcoming RV30 loop filtering 
Originally committed as revision 16235 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-19 06:29:35 +00:00
Jason Garrett-Glaser
40c7d0ae3c Add automatic prefix handling to yasm functions. Does nothing now, but will 
be useful for porting x264 asm in the future.

Originally committed as revision 16234 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-19 03:01:08 +00:00
Michael Niedermayer
a5b807a6c1 Replace /2 by >>1 in decode_cabac_mb_dqp() 
3 cpu cycles speed up on pentium dual.

Originally committed as revision 16233 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-19 02:28:51 +00:00
Michael Niedermayer
1aea5d35e5 Simplify ctx update in decode_cabac_mb_dqp(). 
no speed change

Originally committed as revision 16232 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-19 02:00:33 +00:00
Michael Niedermayer
7cfca0dfd8 Simplify ctx calculation in decode_cabac_mb_dqp() 
no speed change

Originally committed as revision 16231 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-19 01:50:20 +00:00
Alexander Strange
d43696309a Clear FF_INPUT_BUFFER_PADDING_SIZE bytes at the end of NALs in rbsp_buffer. 
Fixes valgrind uninitialized value warnings at the end of decoding H.264
frames.

Originally committed as revision 16230 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-19 01:11:52 +00:00
Jason Garrett-Glaser
712ca84c21 Move filter_luma_intra into dsputil for later addition of asm. 
Originally committed as revision 16228 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-19 00:44:51 +00:00
Jason Garrett-Glaser
b9fe706305 Simplify chroma AC in CABAC residual decoding. 
Originally committed as revision 16227 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-19 00:43:30 +00:00
Michael Niedermayer
8955b66950 Optimize ctx calculation in decode_cabac_mb_mvd(), code by dark shikari. 
The case for 16x16 blocks becomes 10 cpu cycles faster on pentium dual,
i could not find a speed difference in the case of subblocks though.

Originally committed as revision 16226 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-19 00:05:39 +00:00
Michael Niedermayer
17779f39b6 Remove unacceptable NULL pointer hack from mc code. 
Originally committed as revision 16225 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-18 23:52:32 +00:00
Michael Niedermayer
04618b98e3 Check ref values in CABAC H.264 for validity. 
Originally committed as revision 16224 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-18 23:31:10 +00:00
Michael Niedermayer
c25ac15a07 Move idct_(dc)add closer to where it is needed. 
Originally committed as revision 16223 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-18 18:25:11 +00:00
Michael Niedermayer
aebb5d6d96 indent 
Originally committed as revision 16222 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-18 18:14:38 +00:00
Michael Niedermayer
96465b90a1 Reorder ifs in chroma hl_decode_mb to avoid a duplicate transform_bypass 
check.
14 cpu cycles speedup on Pentium Dual

Originally committed as revision 16221 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-18 18:12:59 +00:00
Michael Niedermayer
6456d6d87c s/h->cbp_table[mb_xy]/h->cbp/ 
Originally committed as revision 16220 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-18 17:42:44 +00:00
Michael Niedermayer
04824298a9 Faster CAVLC decoding of trailing_ones. Based on a patch by dark shikari. 
decode_residual is about 3.3% faster.

Originally committed as revision 16219 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-18 16:11:19 +00:00
Jason Garrett-Glaser
93445d1617 Replace i by trailing_ones, part of a patch by dark shikari. 
No speed change meassureable with START/STOP_TIMER, but this is needed
for future optimizations.

Originally committed as revision 16218 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-18 15:51:32 +00:00
Michael Niedermayer
c375d87f6f Remove if() surrounding decode_cabac_mb_type() that can never be true. 
Originally committed as revision 16217 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-18 15:32:07 +00:00
Michael Niedermayer
c325b5054f Remove unreachable else clause, found by dark shikari. 
Originally committed as revision 16216 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-18 15:21:15 +00:00
Michael Niedermayer
dae006d7d7 Remove useless IS_8x8DCT check i forgot, spotted by dark shikari. 
Originally committed as revision 16215 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-18 11:17:12 +00:00
Michael Niedermayer
2a5a9c2815 10l, svq3 didnt set h->cbp, this broke decoding a little. 
Originally committed as revision 16214 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-18 11:01:38 +00:00
Kostya Shishkov
efeb298ea0 cosmetics: remove one unneeded space 
Originally committed as revision 16211 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-18 06:00:12 +00:00
Michael Niedermayer
1eb960352b Do not calculate idct_dc_add/idct_add when the variables are unused. 
Originally committed as revision 16210 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-18 04:13:02 +00:00
Michael Niedermayer
62bc966f8f Remove redundant nnz variable. 
Originally committed as revision 16209 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-18 03:04:53 +00:00
Michael Niedermayer
0a8ca22f4e indent 
Originally committed as revision 16208 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-18 02:56:44 +00:00
Michael Niedermayer
2fd1f0e026 Use the new idct functions (except chroma as it was slower in benchmarks) 
cathedral +0.5% speed
aladin +0.6% speed [note aladin has been cat-ed 10 times to reduce the influence
of init time]
Speedup also verified via START/STOP_TIMER (difference was very significant
for the changed parts)

Originally committed as revision 16207 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-18 02:53:18 +00:00
Michael Niedermayer
ac22385931 H.264 idct functions that include the chroma, inter luma and intra16 luma loops 
thus avoiding the calling overhead.
New functions are not yet used.

Originally committed as revision 16206 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-18 02:36:48 +00:00
Michael Niedermayer
49c084a745 Skip non intra luma code when there is no coded luma. 
0.7% speedup for the cathedral sample.

Originally committed as revision 16203 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-18 01:10:51 +00:00
Loren Merritt
00794ed0d9 1.6x faster pred*_vertical_add_c 
Originally committed as revision 16202 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-18 01:05:06 +00:00
Michael Niedermayer
621561cdf3 Skip chroma handling when there is no coded chroma. 
0.5% overall speedup for the cathedral sample.

Originally committed as revision 16201 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-18 00:46:54 +00:00
Kostya Shishkov
17b17c53f4 add const qualifier to some pointers for input data 
Originally committed as revision 16195 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-17 20:17:07 +00:00
Kostya Shishkov
bb39171beb use precalculated mb_pos where available 
Originally committed as revision 16194 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-17 20:10:15 +00:00
Kostya Shishkov
6c3fca6479 Merge deblocking pattern with CBP for RV30/40 loop filtering 
Originally committed as revision 16192 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-17 20:00:48 +00:00
Kostya Shishkov
26bc319c1d Generate correct deblock pattern for RV30 
Originally committed as revision 16191 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-17 19:39:56 +00:00
Kostya Shishkov
f610fc81c1 Unfortunately RV30 luma thirdpel MC functions cannot be substituted with 
applying two separate filtering passes, so implement them explicitly.

Originally committed as revision 16190 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-17 19:37:45 +00:00
Kostya Shishkov
6a4583e925 4l: TIFF stores short strings inside tag, do not interpret it is as an offset 
Fixes issue 753

Originally committed as revision 16189 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-17 17:39:35 +00:00
Guillaume Poirier
5f51afd602 add AltiVec implementation of int32_to_float_fmul_scalar 
Originally committed as revision 16186 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-17 09:47:06 +00:00
Michael Niedermayer
54dab66174 Fix svq3 decoding, is_complex was not initialized. 
Originally committed as revision 16184 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-17 03:41:48 +00:00
Michael Niedermayer
4080e67c8e Replace != 0 || check by | 
3 cpu cycles faster

Originally committed as revision 16183 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-17 02:53:03 +00:00
Michael Niedermayer
ad9ca7e720 Split filter_mb_dir() out of filter_mb(). 
1% overall decoding speed up for cathedral-beta2-400extra-crop-avc.mp4
no speed change for Aladin.mpg
Benchmarks done on Pentium dual

Originally committed as revision 16182 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-17 02:35:14 +00:00