[Ffmpeg-devel-irc] ffmpeg-devel.log.20120404

[burek]

[00:05] <Compn> hyc : btw, libav has started its GSOC rtmp task, and they are talking about api wrt rtmpdump / librtmp on libav-devel , if you are interested in joining that discussion
[00:09] <Compn> maybe they arent talking about librtmp api
[00:10] <Compn> the discussion is hard to follow
[00:11] <kierank> they are writing their own rtmp code
[00:12] <Compn> yeah but it looks like they dont want options inside url
[00:12] <Compn> but want -app etc
[00:12] <Compn> which would need some changes to librtmp to keep compatability
[00:13] <Compn> http://lists.libav.org/pipermail/libav-devel/2012-April/025298.html
[00:22] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * rb5d433d306 10ffmpeg/libavformat/utils.c: 
[00:22] <CIA-17> ffmpeg: lavf: increase RELATIVE_TS_BASE
[00:22] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[00:22] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * rae3ee72e02 10ffmpeg/libavformat/utils.c: 
[00:22] <CIA-17> ffmpeg: lavf: print some more things in TS debug statements
[00:22] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[01:36] <CIA-17> ffmpeg: 03Hans-Kristian Arntzen 07master * rb991b300f2 10ffmpeg/libswscale/ (swscale_internal.h swscale_unscaled.c): 
[01:36] <CIA-17> ffmpeg: Unscaled Planar RGB -> RGB support in swscale.
[01:36] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[01:36] <CIA-17> ffmpeg: Signed-off-by: Derek Buitenhuis <derek.buitenhuis at gmail.com>
[01:37] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * r41a097aada 10ffmpeg/: 
[01:37] <CIA-17> ffmpeg: Merge remote-tracking branch 'qatar/master'
[01:37] <CIA-17> ffmpeg: * qatar/master:
[01:37] <CIA-17> ffmpeg:  Unscaled Planar RGB -> RGB support in swscale.
[01:37] <CIA-17> ffmpeg: Conflicts:
[01:37] <CIA-17> ffmpeg:  libswscale/swscale_unscaled.c
[01:37] <CIA-17> ffmpeg: Merged-by: Michael Niedermayer <michaelni at gmx.at>
[02:07] <media> hi all
[02:07] <media> michaelni you here?
[02:08] <media> any here?
[02:11] <media> machaelni if you read this, just to update you and devs something for sure is broke in ffmpeg I've tried all I can think of and live tv with vdr and sound problems still persist.
[02:12] <media> I still get the errors we talked about the other day as follows
[02:13] <media> [h264 @ 0x8ac6540] mmco: unref short failure
[02:13] Last message repeated 1 time(s).
[02:13] <media> [h264 @ 0x8ac6540] illegal short term buffer state detected
[02:13] <media> [h264 @ 0x8ac6540] Cannot combine reference and non-reference fields in the same frame
[02:13] <media> [h264 @ 0x8ac6540] If you want to help, upload a sample of this file to ftp://upload.ffmpeg.org/MPlayer/incoming/ and contact the ffmpeg-devel mailing list.
[02:13] <media> [h264 @ 0x8ac6540] decode_slice_header error
[02:13] <media> [h264 @ 0x8ac6540] reference picture missing during reorder
[02:13] <media> [h264 @ 0x8ac6540] Missing reference picture
[02:14] <media> thanks bye
[04:07] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * r3360b8517a 10ffmpeg/libavcodec/h264_refs.c: 
[04:07] <CIA-17> ffmpeg: h264: fix seeking in low delay streams without IDR
[04:07] <CIA-17> ffmpeg: Fixes Ticket1165
[04:07] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[04:55] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * ra4e359a3f9 10ffmpeg/libavcodec/h263dec.c: 
[04:55] <CIA-17> ffmpeg: mpeg4: dont reset picture_num for xvid
[04:55] <CIA-17> ffmpeg: Fixes Ticket1162
[04:55] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[06:09] <ramiro> hi
[06:10] <ramiro> 'make test' is gone, what's the current way to test changes?
[06:13] <Daemon404> are you referring to running the regressions tests?
[06:13] <ramiro> yes
[06:14] <Daemon404> make fate, though this wont use the fate sampels suite
[06:14] <Daemon404> just vsynth1/2 etc
[06:14] <Daemon404> otherwise make fate-rsync, make fate (configured with --samples or SAMPLES env var)
[06:16] <ramiro> hm, that'll take a while to download
[06:17] <Daemon404> ~500 mb iirc
[09:31] <CIA-17> ffmpeg: 03ami_stuff 07release/0.10 * ra854d00acd 10ffmpeg/libavcodec/x86/dsputil_yasm.asm: 
[09:31] <CIA-17> ffmpeg: Replace SSE2 instruction in scalarproduct_float_sse() by SSE equivalent.
[09:31] <CIA-17> ffmpeg: Fixes an AAC decoding issue with the sample from ticket #213 on machines
[09:31] <CIA-17> ffmpeg: with SSE but without SSE2.
[09:31] <CIA-17> ffmpeg: Based on 89411a by Reimar.
[09:31] <CIA-17> ffmpeg: (cherry picked from commit f6b78638086beae9bcab672d4c9de1790be5a928)
[09:32] <CIA-17> ffmpeg: 03Michael Niedermayer 07release/0.10 * rfdc6f6507c 10ffmpeg/libavcodec/h264_refs.c: 
[09:32] <CIA-17> ffmpeg: h264: fix seeking in low delay streams without IDR
[09:32] <CIA-17> ffmpeg: Fixes Ticket1165
[09:32] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[09:32] <CIA-17> ffmpeg: (cherry picked from commit 3360b8517a1f478c4102072e5eadd8ba78be0538)
[09:32] <CIA-17> ffmpeg: 03ami_stuff 07release/0.6 * rb56606e6bc 10ffmpeg/libavcodec/x86/dsputil_yasm.asm: 
[09:32] <CIA-17> ffmpeg: Replace SSE2 instruction in scalarproduct_float_sse() by SSE equivalent.
[09:32] <CIA-17> ffmpeg: Fixes an AAC decoding issue with the sample from ticket #213 on machines
[09:32] <CIA-17> ffmpeg: with SSE but without SSE2.
[09:32] <CIA-17> ffmpeg: Based on 89411a by Reimar.
[09:32] <CIA-17> ffmpeg: (cherry picked from commit f6b78638086beae9bcab672d4c9de1790be5a928)
[09:32] <CIA-17> ffmpeg: 03Michael Niedermayer 07release/0.10 * ra56eaa024f 10ffmpeg/libavcodec/h263dec.c: 
[09:32] <CIA-17> ffmpeg: mpeg4: dont reset picture_num for xvid
[09:32] <CIA-17> ffmpeg: Fixes Ticket1162
[09:32] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[09:32] <CIA-17> ffmpeg: (cherry picked from commit a4e359a3f98650dab3d2e93f067658e20fa9c0d7)
[09:32] <CIA-17> ffmpeg: 03Stefano Sabatini 07release/0.10 * rd076d0febd 10ffmpeg/libavfilter/vf_fade.c: 
[09:32] <CIA-17> ffmpeg: lavfi/fade: fix black level for non studio-level pixel formats
[09:33] <CIA-17> ffmpeg: Replace SSE2 instruction in scalarproduct_float_sse() by SSE equivalent.
[09:33] <CIA-17> ffmpeg: Fixes an AAC decoding issue with the sample from ticket #213 on machines
[09:33] <CIA-17> ffmpeg: with SSE but without SSE2.
[09:33] <CIA-17> ffmpeg: Based on 89411a by Reimar.
[09:33] <CIA-17> ffmpeg: (cherry picked from commit f6b78638086beae9bcab672d4c9de1790be5a928)
[09:33] <CIA-17> ffmpeg: 03Stefano Sabatini 07release/0.9 * r5667eaf010 10ffmpeg/libavfilter/vf_fade.c: 
[09:33] <CIA-17> ffmpeg: lavfi/fade: fix black level for non studio-level pixel formats
[09:33] <CIA-17> ffmpeg: Fix trac ticket #1139, regression introduced in 8c1fb50d077d5f954.
[09:33] <CIA-17> ffmpeg: (cherry picked from commit 95ce0ddcfe99182365e0e57f5f41d7f1a01c57eb)
[09:33] <CIA-17> ffmpeg: 03ami_stuff 07release/0.8 * r10848d0862 10ffmpeg/libavcodec/x86/dsputil_yasm.asm: 
[09:33] <CIA-17> ffmpeg: Replace SSE2 instruction in scalarproduct_float_sse() by SSE equivalent.
[09:34] <CIA-17> (15 lines omitted)
[11:50] <CIA-17> ffmpeg: 03Clément BSsch 07master * re29f66d64a 10ffmpeg/ffmpeg.c: ffmpeg: fix a few typo.
[13:23] <ubitux> ffmpeg.org down?
[13:30] <Compn> server was upgraded, i think
[13:31] <Compn> seems like it
[13:31] <ubitux> seems back :)
[13:31] <Compn> Broadcast message from root (pts/1) (Wed Apr  4 11:18:44 2012):
[13:31] <Compn> The system is going down for system halt NOW!
[15:01] <ubitux> i can't believe how apple isn't following at all its own specifications
[15:01] <av500> what is not to believe?
[15:01] <ubitux> :(
[15:02] <ubitux> they're violating it so much i wonder if i'm even reading the good spec
[15:02] <Compn> lol
[15:02] <Compn> what spec are you reading ?
[15:02] <ubitux> quicktime, 2001
[15:02] <Compn> sounds like an old spec
[15:02] <av500> 2001?
[15:02] <ubitux> the only one i found
[15:02] <av500> they made that spec into a movie
[15:02] <Compn> haha
[15:03] <ubitux> :D
[15:03] <ubitux> https://developer.apple.com/standards/classicquicktime.html
[15:03] <ubitux> (the "current" web one isn't "up to date" either)
[15:05] <av500> ubitux: what quick feature are you wasting your time with?
[15:05] <ubitux> timecode again
[15:05] <ubitux> tmcd track etc
[15:06] <ubitux> at the moment i counted 3 things not matching the specs
[15:06] <ubitux> (for that single feature)
[15:09] <michaelni> only 3 ?
[15:09] <ubitux> i'm not done yet
[15:09] <ubitux> timecode is just a frame id ffs T_T
[15:10] <ubitux> fortunately the iso wasn't stupid enough to put it in the iso base media standard or the mp4 one
[15:10] <av500> and it's used a lot?
[15:11] <av500> other than in classic files from 2001?
[15:17] <ubitux> it seems to be muxed all the time with the apple softwares
[15:17] <ubitux> take the big buck bunny mov for instance
[15:17] <ubitux> there is a tmcd track
[15:17] <av500> ignoring that works fine here :)
[15:17] <Compn> ubitux : mail mike and see if hes got an updated one
[15:17] <ubitux> it's used for editing software
[15:18] <ubitux> Compn: an updated one of what?
[15:18] <Compn> Apple QuickTime container specification, dated 2007-09-04: 
[15:18] <Compn> mirror: http://multimedia.cx/mirror/qtff-2007-09-04.pdf
[15:18] <Compn> http://wiki.multimedia.cx/index.php?title=Mirror
[15:18] <ubitux> huh.
[15:18] <Compn> original: http://developer.apple.com/mac/library/documentation/QuickTime/QTFF
[15:18] <ubitux> wait, where does this come from?
[15:19] <Compn> no idea if thats what you need.
[15:19] <mateo`> av500: the tmcd track is also used by the broadcasters
[15:20] <ubitux> the 2007 seems as bad as the 2001 one
[15:20] <Compn> it should be worse
[15:20] <ubitux> :D
[15:20] <Compn> mp4 never gets better .
[15:21] <ubitux> mp4 is actually "better"
[15:21] <ubitux> at least the specs is nice :p
[15:21] <ubitux> and there is less insanity imop
[15:21] <ubitux> -p
[15:21] <Compn> always trying to stuff new ideas into an old idea
[15:23] <ubitux> anyway, thanks for that specs
[15:23] <ubitux> still, i don't know where it was grabbed
[15:23] <CIA-17> ffmpeg: 03Nicolas George 07master * r8a9ae37fee 10ffmpeg/libavdevice/ (timefilter.c timefilter.h): timefilter: allow to evaluate at other times.
[15:23] <Compn> like i said, ask mike :)
[15:23] <CIA-17> ffmpeg: 03Nicolas George 07master * r12292f35a3 10ffmpeg/ (4 files in 3 dirs): v4l2: allow to convert monotonic timestamps.
[15:24] <MP4_maniac> 22:21 (ubitux) mp4 is actually "better"
[15:25] <ubitux> yes
[15:33] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * re3e89b6d8b 10ffmpeg/ (4 files in 2 dirs): 
[15:33] <CIA-17> ffmpeg: libavfilter: Add interlace detect filter.
[15:33] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[15:44] <Compn> ooo interlace detection?
[15:44] <Compn> thats newsworthy 
[15:44] <Compn> i've seen lots and lots of requests for such a feature over the years
[15:45] <Compn> nice one michaelni :)
[15:46] <Compn> i now expect we'll see a ton of interlaced materials that get incorrect results 
[15:46] <Compn> non-interlaced too
[15:49] <iive> now ffmpeg needs good inverse telecine filter.
[15:49] <j-b> take the one from VLC
[15:50] <j-b> http://git.videolan.org/?p=vlc.git;a=blob;f=modules/video_filter/deinterlace/algo_ivtc.c;hb=HEAD
[15:51] <ubitux> there is just more comment ;)
[16:24] <CIA-17> ffmpeg: 03Nicolas George 07master * re8497a338e 10ffmpeg/libavfilter/ (drawutils.c drawutils.h): drawutils: implement uniform and mask blending.
[16:24] <CIA-17> ffmpeg: 03Nicolas George 07master * ra63712d37c 10ffmpeg/libavfilter/vf_drawtext.c: vf_drawtext: use drawutils.
[16:24] <CIA-17> ffmpeg: 03Nicolas George 07master * r6c7b5b7b33 10ffmpeg/libavfilter/vf_ass.c: vf_ass: use drawutils.
[16:26] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * r398491ae01 10ffmpeg/libavfilter/vf_idet.c: 
[16:26] <CIA-17> ffmpeg: vf_idet: pass context to av_log()
[16:26] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[16:26] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * r02567bd73c 10ffmpeg/libavfilter/vf_idet.c: 
[16:26] <CIA-17> ffmpeg: vf_idet: remove unused function
[16:26] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[16:26] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * r5d9cfd8726 10ffmpeg/libavfilter/vf_idet.c: 
[16:26] <CIA-17> ffmpeg: vf_idet: move statistic variables into the context
[16:26] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[16:26] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * r2941a937f2 10ffmpeg/libavfilter/vf_idet.c: 
[16:26] <CIA-17> ffmpeg: vf_idet: fix indention
[16:26] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[16:36] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * rbaf0c79a4d 10ffmpeg/libavfilter/vf_idet.c: 
[16:36] <CIA-17> ffmpeg: vf_idet: set interlaced flags
[16:36] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[16:36] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * r6fb35dbad1 10ffmpeg/libavfilter/vf_idet.c: 
[16:36] <CIA-17> ffmpeg: vf_idet: print accumulated statistics at end
[16:36] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[16:58] <ramiro> hi
[17:00] <Compn> yo
[17:10] <kriegerod> michaelni, is the fix for this pushed already? http://ffmpeg.org/trac/ffmpeg/ticket/387
[17:10] <kriegerod> checkout from today afternoon seem to be broken
[17:16] <michaelni> kriegerod, just tried, works for me
[17:16] <michaelni> whats your command line and what happens ?
[17:28] <kriegerod> ah, my fail - i tried to set profile baseline to flv video encoder, and it apparently does not have it
[18:16] <CIA-17> ffmpeg: 03Tomas Härdin 07master * ra61fada380 10ffmpeg/libavformat/mxfdec.c: (log message trimmed)
[18:16] <CIA-17> ffmpeg: mxfdec: Fix regression on files from Pinnacle Thunder
[18:16] <CIA-17> ffmpeg: The reason for this is that such files have IndexTableSegments which when parsed
[18:16] <CIA-17> ffmpeg: cover EditUnit ranges like this:
[18:16] <CIA-17> ffmpeg:  [0,1)
[18:16] <CIA-17> ffmpeg:  [249,250)
[18:16] <CIA-17> ffmpeg:  [249,377)
[18:16] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * rc22fd9bd9c 10ffmpeg/libavcodec/allcodecs.c: (log message trimmed)
[18:16] <CIA-17> ffmpeg: allcodecs: move libaacplus down.
[18:16] <CIA-17> ffmpeg: We choose the first encoder by default and libaccplus has a
[18:16] <CIA-17> ffmpeg: quite limited set of supported bitrates/sample rates.
[18:16] <CIA-17> ffmpeg: Thus leading to failure by default in many cases when it is
[18:16] <CIA-17> ffmpeg: enabled at compile time.
[18:17] <CIA-17> ffmpeg: Moving it down means that the other aac encoders are favored
[19:07] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * rca26c2c30f 10ffmpeg/libavformat/utils.c: 
[19:07] <CIA-17> ffmpeg: lavf: add mp1/2 to determinable_frame_size()
[19:07] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[19:19] <CIA-17> ffmpeg: 03Derek Buitenhuis 07master * r6da08c9986 10ffmpeg/libavcodec/w32pthreads.h: 
[19:19] <CIA-17> ffmpeg: w32threads: Make pthread_cond_wait follow POSIX
[19:19] <CIA-17> ffmpeg: pthread_cond_wait is supposed to return an integer,
[19:19] <CIA-17> ffmpeg: and indeed does sometimes. Fix its function declaration
[19:19] <CIA-17> ffmpeg: to match its behavior and POSIX.
[19:19] <CIA-17> ffmpeg: Signed-off-by: Derek Buitenhuis <derek.buitenhuis at gmail.com>
[19:19] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * rfde63146d2 10ffmpeg/: 
[19:19] <CIA-17> ffmpeg: Merge remote-tracking branch 'dwbuiten/master'
[19:19] <CIA-17> ffmpeg: * dwbuiten/master:
[19:19] <CIA-17> ffmpeg:  w32threads: Make pthread_cond_wait follow POSIX
[19:19] <CIA-17> ffmpeg: Merged-by: Michael Niedermayer <michaelni at gmx.at>
[19:25] <ramiro> how is the output from seek-test supposed to be interpreted?
[20:01] <michaelni> ramiro, seektest seeks to random spots and then prints the timestamps and other stuff from the packets that get demuxed after seeking
[20:01] <michaelni> if the seek target and the packet time stamps lay far appart then thats bad
[20:02] <michaelni> the forward vs backward seek flag and the stream index used for sseking are also randomly set
[20:32] <ramiro> michaelni: asfenc was so bad at generating a search index that the regression tests never actually tested it
[20:32] <ramiro> it always went by ff_seek_frame_binary()
[20:33] <ramiro> with my patches wmp is extra happy to seek anywhere and resume proper playback quickly enough (no audio delay)
[20:38] <bcoudurier> nice
[20:39] <CIA-17> ffmpeg: 03Reimar Döffinger 07master * ra45605a1ba 10ffmpeg/ (6 files in 2 dirs): 
[20:39] <CIA-17> ffmpeg: Add AVIO_FLAG_DIRECT.
[20:39] <CIA-17> ffmpeg: Allows avoiding the buffer when using avio read, write and seek functions.
[20:39] <CIA-17> ffmpeg: When using the ffmpeg executable -avioflags direct can be used to enable
[20:39] <CIA-17> ffmpeg: this mode for input files, but has no effect on output files.
[20:39] <CIA-17> ffmpeg: Signed-off-by: Reimar Döffinger <Reimar.Doeffinger at gmx.de>
[20:41] <michaelni> ramiro, ok but why does the seek test worsen, i dont understand it ?
[20:41] <ramiro> me neither, I'm still looking into that
[20:41] <michaelni> ok
[21:15] <CIA-17> ffmpeg: 03Reimar Döffinger 07master * r6cecb8915a 10ffmpeg/tests/ref/fate/avio-direct: 
[21:15] <CIA-17> ffmpeg: Add forgotten avio-direct reference file.
[21:15] <CIA-17> ffmpeg: Signed-off-by: Reimar Döffinger <Reimar.Doeffinger at gmx.de>
[21:33] <CIA-17> ffmpeg: 03Samuel Pitoiset 07master * rfaba4a9b88 10ffmpeg/libavformat/rtmpproto.c: 
[21:33] <CIA-17> ffmpeg: rtmp: update supported audio codecs value
[21:33] <CIA-17> ffmpeg: The audio codecs property is composed by all values except
[21:33] <CIA-17> ffmpeg: SUPPORT_SND_INTEL (0x0008) and SUPPORT_SND_UNUSED (0x0010) which are
[21:33] <CIA-17> ffmpeg: unused.
[21:33] <CIA-17> ffmpeg: Signed-off-by: Luca Barbato <lu_zero at gentoo.org>
[21:33] <CIA-17> ffmpeg: 03Raffaele Sena 07master * r34d908c083 10ffmpeg/libavformat/rtmpproto.c: 
[21:33] <CIA-17> ffmpeg: rtmp: implement bandwidth notification
[21:33] <CIA-17> ffmpeg: Improve compatibility with some servers.
[21:33] <CIA-17> ffmpeg: Signed-off-by: Luca Barbato <lu_zero at gentoo.org>
[21:33] <CIA-17> ffmpeg: 03Aneesh Dogra 07master * r6d21f49879 10ffmpeg/libavcodec/pcm.c: 
[21:33] <CIA-17> ffmpeg: pcm: K&R formatting cosmetics
[21:33] <CIA-17> ffmpeg: Signed-off-by: Diego Biurrun <diego at biurrun.de>
[21:34] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * r4480edcf31 10ffmpeg/: (log message trimmed)
[21:34] <CIA-17> ffmpeg: Merge remote-tracking branch 'qatar/master'
[21:34] <CIA-17> ffmpeg: * qatar/master:
[21:34] <CIA-17> ffmpeg:  w32threads: Make pthread_cond_wait follow POSIX
[21:34] <CIA-17> ffmpeg:  cosmetics: Consistently place static, inline and av_cold attributes/keywords.
[21:34] <CIA-17> ffmpeg:  sbrdsp: Use standard multiple inclusion guards.
[21:34] <CIA-17> ffmpeg:  pcm: K&R formatting cosmetics
[00:00] --- Thu Apr  5 2012