Statistics
| Branch: | Revision:

ffmpeg / libavcodec / h264.c @ 1968e438

History | View | Annotate | Download (311 KB)

# Date Author Comment
1968e438 01/10/2009 10:57 PM Carl Eugen Hoyos

Use only MpegEncContext in ff_vdpau_h264_add_data_chunk.

Originally committed as revision 16520 to svn://svn.ffmpeg.org/ffmpeg/trunk

7fd3c09e 01/10/2009 02:45 AM Carl Eugen Hoyos

Move vdpau_h264_set_reference_frames() call from vdpauvideo.c into h264.c.

Originally committed as revision 16513 to svn://svn.ffmpeg.org/ffmpeg/trunk

7eb6ed78 01/09/2009 10:22 PM Carl Eugen Hoyos

Move VDPAU check, make future VDPAU patches smaller.

Originally committed as revision 16508 to svn://svn.ffmpeg.org/ffmpeg/trunk

090846f3 01/07/2009 11:48 PM Carl Eugen Hoyos

Rename ff_VDPAU as ff_vdpau.

Originally committed as revision 16492 to svn://svn.ffmpeg.org/ffmpeg/trunk

218a6022 01/07/2009 11:03 PM Diego Biurrun

Only #include svq3.c if the SVQ3 decoder is enabled.

Originally committed as revision 16487 to svn://svn.ffmpeg.org/ffmpeg/trunk

8b6871ed 01/07/2009 01:58 AM Diego Biurrun

Disable SVQ3 checks if SVQ3 support is not compiled in.

Originally committed as revision 16462 to svn://svn.ffmpeg.org/ffmpeg/trunk

26ef65da 01/05/2009 08:46 AM Jason Garrett-Glaser

50l: Reenable error resilience for H264 (Patch by Dark Shikari).

Originally committed as revision 16434 to svn://svn.ffmpeg.org/ffmpeg/trunk

f2c214a1 01/05/2009 12:48 AM Carl Eugen Hoyos

Cosmetics: Fix indentation.

Originally committed as revision 16433 to svn://svn.ffmpeg.org/ffmpeg/trunk

369122dd 01/04/2009 11:55 PM NVIDIA Corporation

Add VDPAU hardware accelerated decoding for H264 which can be used by
video players.

Original patch by NVIDIA corporation.

Originally committed as revision 16431 to svn://svn.ffmpeg.org/ffmpeg/trunk

802e9146 12/24/2008 06:39 PM Michael Niedermayer

export refs

Originally committed as revision 16306 to svn://svn.ffmpeg.org/ffmpeg/trunk

62ea19c0 12/24/2008 03:23 AM Michael Niedermayer

Optimize pred_pskip_motion()
7 cpu cycles faster on pentium dual.

Originally committed as revision 16304 to svn://svn.ffmpeg.org/ffmpeg/trunk

ebfb0449 12/23/2008 10:06 PM Michael Niedermayer

Indent

Originally committed as revision 16302 to svn://svn.ffmpeg.org/ffmpeg/trunk

05d3fd44 12/23/2008 10:04 PM Michael Niedermayer

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

9963b332 12/23/2008 09:12 PM Michael Niedermayer

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

2f2893ce 12/23/2008 07:12 PM Michael Niedermayer

indent

Originally committed as revision 16296 to svn://svn.ffmpeg.org/ffmpeg/trunk

d42fc4a8 12/23/2008 07:10 PM Michael Niedermayer

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

593af7cd 12/23/2008 06:31 PM Michael Niedermayer

Optimize esc removal code.

Originally committed as revision 16294 to svn://svn.ffmpeg.org/ffmpeg/trunk

2d76bf39 12/23/2008 05:50 PM Michael Niedermayer

Indent

Originally committed as revision 16292 to svn://svn.ffmpeg.org/ffmpeg/trunk

8140955d 12/23/2008 05:49 PM Michael Niedermayer

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

abb27cfb 12/23/2008 01:11 AM Michael Niedermayer

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

e08715d3 12/23/2008 12:38 AM Michael Niedermayer

Optimize 0 0 0-3 search, 45% faster on pentium dual.

Originally committed as revision 16284 to svn://svn.ffmpeg.org/ffmpeg/trunk

ec3686e8 12/22/2008 05:14 PM Michael Niedermayer

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

26695973 12/22/2008 04:56 PM Michael Niedermayer

Indent

Originally committed as revision 16278 to svn://svn.ffmpeg.org/ffmpeg/trunk

b68a4553 12/22/2008 04:46 PM Michael Niedermayer

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

1952ac37 12/22/2008 04:20 PM Michael Niedermayer

Negate 2 more variables, 1 cpu cycle faster on pentium dual.

Originally committed as revision 16276 to svn://svn.ffmpeg.org/ffmpeg/trunk

03a035e0 12/22/2008 04:14 PM Michael Niedermayer

Simplify if/else, no speed change

Originally committed as revision 16275 to svn://svn.ffmpeg.org/ffmpeg/trunk

6f3c50f2 12/22/2008 04:10 PM Michael Niedermayer

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

60c6ba7a 12/22/2008 03:54 PM Michael Niedermayer

Simplify ifs(), 8 cpu cycles faster on pentium dual

Originally committed as revision 16273 to svn://svn.ffmpeg.org/ffmpeg/trunk

127a20e3 12/22/2008 03:53 PM Michael Niedermayer

Simplify if(), 3 cpu cycles faster in pentium dual.

Originally committed as revision 16272 to svn://svn.ffmpeg.org/ffmpeg/trunk

a6493a8f 12/22/2008 09:12 AM Diego Biurrun

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

bef05f05 12/22/2008 12:10 AM Diego Biurrun

Remove a bunch of unused variables.

Originally committed as revision 16263 to svn://svn.ffmpeg.org/ffmpeg/trunk

befc8fe0 12/21/2008 05:49 PM Michael Niedermayer

Remove useless code.

Originally committed as revision 16253 to svn://svn.ffmpeg.org/ffmpeg/trunk

c212fb0c 12/21/2008 03:58 PM Michael Niedermayer

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

66c07ca9 12/20/2008 11:11 PM Michael Niedermayer

Optimize get_dct8x8_allowed().
30 cpu cycles faster on pentium dual.

Originally committed as revision 16248 to svn://svn.ffmpeg.org/ffmpeg/trunk

aac8b769 12/19/2008 02:49 PM Jason Garrett-Glaser

H.264 loopfilter speed tweaks

Originally committed as revision 16240 to svn://svn.ffmpeg.org/ffmpeg/trunk

a5805aa9 12/19/2008 09:21 AM Michael Niedermayer

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

a5b807a6 12/19/2008 02:28 AM Michael Niedermayer

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

1aea5d35 12/19/2008 02:00 AM Michael Niedermayer

Simplify ctx update in decode_cabac_mb_dqp().
no speed change

Originally committed as revision 16232 to svn://svn.ffmpeg.org/ffmpeg/trunk

7cfca0df 12/19/2008 01:50 AM Michael Niedermayer

Simplify ctx calculation in decode_cabac_mb_dqp()
no speed change

Originally committed as revision 16231 to svn://svn.ffmpeg.org/ffmpeg/trunk

d4369630 12/19/2008 01:11 AM Alexander Strange

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

712ca84c 12/19/2008 12:44 AM Jason Garrett-Glaser

Move filter_luma_intra into dsputil for later addition of asm.

Originally committed as revision 16228 to svn://svn.ffmpeg.org/ffmpeg/trunk

b9fe7063 12/19/2008 12:43 AM Jason Garrett-Glaser

Simplify chroma AC in CABAC residual decoding.

Originally committed as revision 16227 to svn://svn.ffmpeg.org/ffmpeg/trunk

8955b669 12/19/2008 12:05 AM Michael Niedermayer

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

17779f39 12/18/2008 11:52 PM Michael Niedermayer

Remove unacceptable NULL pointer hack from mc code.

Originally committed as revision 16225 to svn://svn.ffmpeg.org/ffmpeg/trunk

04618b98 12/18/2008 11:31 PM Michael Niedermayer

Check ref values in CABAC H.264 for validity.

Originally committed as revision 16224 to svn://svn.ffmpeg.org/ffmpeg/trunk

c25ac15a 12/18/2008 06:25 PM Michael Niedermayer

Move idct_(dc)add closer to where it is needed.

Originally committed as revision 16223 to svn://svn.ffmpeg.org/ffmpeg/trunk

aebb5d6d 12/18/2008 06:14 PM Michael Niedermayer

indent

Originally committed as revision 16222 to svn://svn.ffmpeg.org/ffmpeg/trunk

96465b90 12/18/2008 06:12 PM Michael Niedermayer

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

6456d6d8 12/18/2008 05:42 PM Michael Niedermayer

s/h->cbp_table[mb_xy]/h->cbp/

Originally committed as revision 16220 to svn://svn.ffmpeg.org/ffmpeg/trunk

04824298 12/18/2008 04:11 PM Michael Niedermayer

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

93445d16 12/18/2008 03:51 PM Jason Garrett-Glaser

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

c375d87f 12/18/2008 03:32 PM Michael Niedermayer

Remove if() surrounding decode_cabac_mb_type() that can never be true.

Originally committed as revision 16217 to svn://svn.ffmpeg.org/ffmpeg/trunk

c325b505 12/18/2008 03:21 PM Michael Niedermayer

Remove unreachable else clause, found by dark shikari.

Originally committed as revision 16216 to svn://svn.ffmpeg.org/ffmpeg/trunk

dae006d7 12/18/2008 11:17 AM Michael Niedermayer

Remove useless IS_8x8DCT check i forgot, spotted by dark shikari.

Originally committed as revision 16215 to svn://svn.ffmpeg.org/ffmpeg/trunk

1eb96035 12/18/2008 04:13 AM Michael Niedermayer

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

62bc966f 12/18/2008 03:04 AM Michael Niedermayer

Remove redundant nnz variable.

Originally committed as revision 16209 to svn://svn.ffmpeg.org/ffmpeg/trunk

0a8ca22f 12/18/2008 02:56 AM Michael Niedermayer

indent

Originally committed as revision 16208 to svn://svn.ffmpeg.org/ffmpeg/trunk

2fd1f0e0 12/18/2008 02:53 AM Michael Niedermayer

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...

49c084a7 12/18/2008 01:10 AM Michael Niedermayer

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

621561cd 12/18/2008 12:46 AM Michael Niedermayer

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

4080e67c 12/17/2008 02:53 AM Michael Niedermayer

Replace != 0 || check by |
3 cpu cycles faster

Originally committed as revision 16183 to svn://svn.ffmpeg.org/ffmpeg/trunk

ad9ca7e7 12/17/2008 02:35 AM Michael Niedermayer

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

ac0623b2 12/16/2008 10:16 PM Michael Niedermayer

Fix indention, also do a little vertical alignment of changed lines.

Originally committed as revision 16176 to svn://svn.ffmpeg.org/ffmpeg/trunk

6120a343 12/16/2008 09:08 PM Michael Niedermayer

Factorize 3 multiplications out, code becomes 3 cpu cycles faster.
(not significant as thats just per MB)

Originally committed as revision 16174 to svn://svn.ffmpeg.org/ffmpeg/trunk

1dd488e9 12/16/2008 08:43 PM Michael Niedermayer

Move ENABLE_SMALL back to the per MB check, as otherwise gcc wont remove
the code.

Originally committed as revision 16173 to svn://svn.ffmpeg.org/ffmpeg/trunk

89db0bae 12/16/2008 08:26 PM Michael Niedermayer

Do not recalculate constant part of is_complex for every MB.

Originally committed as revision 16172 to svn://svn.ffmpeg.org/ffmpeg/trunk

41e4055b 12/16/2008 08:09 PM Michael Niedermayer

Support the new lossless mode.

Originally committed as revision 16171 to svn://svn.ffmpeg.org/ffmpeg/trunk

40f06241 12/15/2008 01:24 AM Michael Niedermayer

Enable filter_mb_fast if CODEC_FLAG2_FAST is set.

Originally committed as revision 16141 to svn://svn.ffmpeg.org/ffmpeg/trunk

cac55c91 12/14/2008 12:24 PM Anders Grönberg

Add a few const qualifiers in appropriate places.
patch by Anders Grönberg, galileo.m2 gmail com

Originally committed as revision 16120 to svn://svn.ffmpeg.org/ffmpeg/trunk

d23b28c7 12/05/2008 09:44 PM Michael Niedermayer

Check that cpb_count is within the allowed range.

Originally committed as revision 16010 to svn://svn.ffmpeg.org/ffmpeg/trunk

c9c16cac 12/01/2008 04:24 PM Michael Niedermayer

Get rid of now unneeded pps/sps correctness code as pps/sps will not
be used at all when anything fails sanity checks.

Originally committed as revision 15972 to svn://svn.ffmpeg.org/ffmpeg/trunk

255d4e71 12/01/2008 04:04 PM Michael Niedermayer

Allocate parameter sets sanely instead of using alloc_parameter_set().
Avoids ending up with half initialized parameter sets.
Fixes issue282.

Originally committed as revision 15971 to svn://svn.ffmpeg.org/ffmpeg/trunk

ff7f75e1 11/30/2008 11:51 PM Michael Niedermayer

Fix segfault due to deallocated entries in delayed_pic after size change.
Fixes issue714.

Originally committed as revision 15965 to svn://svn.ffmpeg.org/ffmpeg/trunk

f107a02d 11/26/2008 10:22 AM Diego Biurrun

Move encoder-specific quantize_c and related tables to the H.264 encoder.

Originally committed as revision 15937 to svn://svn.ffmpeg.org/ffmpeg/trunk

ee2a957f 11/24/2008 03:53 PM Diego Biurrun

Use golomb_to_pict_type instead of its duplicate, slice_type_map.

Originally committed as revision 15926 to svn://svn.ffmpeg.org/ffmpeg/trunk

74e8b78b 11/13/2008 11:33 AM Carl Eugen Hoyos

Fix H.264 decoding.
(Problem analyzed by Uoti Urpala)

Originally committed as revision 15816 to svn://svn.ffmpeg.org/ffmpeg/trunk

3a84713a 11/12/2008 05:47 PM Roman Shaposhnik

Making it easier to send arbitrary structures as work orders to MT workers

Originally committed as revision 15804 to svn://svn.ffmpeg.org/ffmpeg/trunk

b09a7c05 11/04/2008 06:31 PM Andreas Öman

h264: Implement decoding of picture timing SEI message.

Now correct values are propagated to interlaced_frame, top_field_first
and repeat_pict in AVFrame structure.

patch by ffdshow tryouts

Originally committed as revision 15773 to svn://svn.ffmpeg.org/ffmpeg/trunk

b735aeea 10/28/2008 10:50 PM Michael Niedermayer

Replace literally hardcoded max slice number by named constant.

Originally committed as revision 15736 to svn://svn.ffmpeg.org/ffmpeg/trunk

37d3e066 10/21/2008 09:40 PM Aurelien Jacobs

uses FF_ARRAY_ELEMS() where appropriate

Originally committed as revision 15662 to svn://svn.ffmpeg.org/ffmpeg/trunk

d9ec210b 10/04/2008 09:36 AM Diego Pettenò

Mark ff_div6 and ff_rem6 static in h264.c. Patch by Diego Pettenò

Originally committed as revision 15549 to svn://svn.ffmpeg.org/ffmpeg/trunk

8d2fc163 10/02/2008 12:27 AM Carl Eugen Hoyos

Rename copy_picture to ff_copy_picture.

Originally committed as revision 15508 to svn://svn.ffmpeg.org/ffmpeg/trunk

bbc78fb4 09/26/2008 06:11 PM Diego Biurrun

Remove unused variable.

Originally committed as revision 15427 to svn://svn.ffmpeg.org/ffmpeg/trunk

6c805007 09/25/2008 09:44 PM Michael Niedermayer

Avoid out of picture accesses when CODEC_FLAG_EMU_EDGE is set.
Fixes issue427.

Originally committed as revision 15418 to svn://svn.ffmpeg.org/ffmpeg/trunk

5f129a05 09/25/2008 07:01 PM Michael Niedermayer

Do not free sps & pps on width/height changes.
Fixes issue557.

Originally committed as revision 15414 to svn://svn.ffmpeg.org/ffmpeg/trunk

07dff5c7 09/25/2008 02:29 PM Michael Niedermayer

Initialize next/last_picture* as the generic code does not always
set them to sane values.

Originally committed as revision 15412 to svn://svn.ffmpeg.org/ffmpeg/trunk

49573a87 09/25/2008 12:04 PM Michael Niedermayer

Print which slices are IDR not just the slice type.

Originally committed as revision 15411 to svn://svn.ffmpeg.org/ffmpeg/trunk

cf8bb334 09/25/2008 12:41 AM Michael Niedermayer

Fix CAVLC+8x8+MBAFF loop filter.
Fixes FREXT02_JVC_C.

Originally committed as revision 15410 to svn://svn.ffmpeg.org/ffmpeg/trunk

60c9b24d 09/24/2008 11:24 PM Michael Niedermayer

Field -> MBAFF direct mode can use data from fields that are not
in the reference list. Follow the spec and no comment on the sanity of this
design ...
Fixes HPCAMAPALQ_BRCM_B

Originally committed as revision 15407 to svn://svn.ffmpeg.org/ffmpeg/trunk

aa617518 09/22/2008 08:46 PM Michael Niedermayer

A few empty lines for readability.

Originally committed as revision 15385 to svn://svn.ffmpeg.org/ffmpeg/trunk

f4d3382d 09/22/2008 08:43 PM Michael Niedermayer

Another try to fix temporal direct mode references.
Fixes at least
CAPAMA3_Sand_F.264
CVMAPAQP3_Sony_E.jsv

Originally committed as revision 15384 to svn://svn.ffmpeg.org/ffmpeg/trunk

48e025e5 09/22/2008 02:32 AM Michael Niedermayer

Fix ref_shift so that it is correct for more/all? MBAFF/PAFF mixes.
Fixes at least MR9_BT_B.

Originally committed as revision 15380 to svn://svn.ffmpeg.org/ffmpeg/trunk

471341a7 09/21/2008 11:51 PM Michael Niedermayer

indent

Originally committed as revision 15379 to svn://svn.ffmpeg.org/ffmpeg/trunk

53c193a9 09/21/2008 11:49 PM Michael Niedermayer

Correct comment in the direct mode code.
(note, yes this is unrelated to the previous simplification, the
code always behaved like it is documented now.)

Originally committed as revision 15378 to svn://svn.ffmpeg.org/ffmpeg/trunk

a0ad4083 09/21/2008 11:46 PM Michael Niedermayer

Simplify initialization of a few direct mode related variables.

Originally committed as revision 15377 to svn://svn.ffmpeg.org/ffmpeg/trunk

8b1fd554 09/21/2008 02:39 AM Michael Niedermayer

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

8f56e219 09/21/2008 02:23 AM Michael Niedermayer

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

078f42dd 09/21/2008 02:19 AM Michael Niedermayer

Set poc for field references in MBAFF.

Originally committed as revision 15373 to svn://svn.ffmpeg.org/ffmpeg/trunk

a85d0c84 09/19/2008 06:53 PM Michael Niedermayer

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

6a855aae 09/19/2008 06:49 PM Michael Niedermayer

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