[Ffmpeg-devel-irc] ffmpeg-devel.log.20130912

[burek]

[11:18] <cone-47> ffmpeg.git 03Diego Biurrun 07master:5045f6bdfabf: configure: Fix wrong reference to user support mailing list
[11:18] <cone-47> ffmpeg.git 03Michael Niedermayer 07master:ae77f06c691d: Merge commit '5045f6bdfabfa5d3374756d2c23f762fb042acae'
[11:26] <cone-47> ffmpeg.git 03Martin Storsjö 07master:48d6556dd46d: dsicin: Add some basic sanity checks for fields read from the file
[11:26] <cone-47> ffmpeg.git 03Michael Niedermayer 07master:7e88dbe9358f: Merge commit '48d6556dd46d4f4fac10d0f4a819e314887cd50e'
[11:39] <cone-47> ffmpeg.git 03Martin Storsjö 07master:8d928023f953: smacker: Don't return packets in unallocated streams
[11:39] <cone-47> ffmpeg.git 03Michael Niedermayer 07master:33cf70a8b88c: Merge commit '8d928023f953a28692ba27071a448259134b103b'
[11:48] <cone-47> ffmpeg.git 03Martin Storsjö 07master:710b0e270259: smacker: Avoid integer overflow when allocating packets
[11:48] <cone-47> ffmpeg.git 03Michael Niedermayer 07master:c16cd5559349: Merge commit '710b0e27025948b7511821c2f888ff2d74a59e14'
[12:01] <cone-47> ffmpeg.git 03Martin Storsjö 07master:d002fce24a72: smacker: Check malloc return values
[12:01] <cone-47> ffmpeg.git 03Michael Niedermayer 07master:d510d70df02b: Merge commit 'd002fce24a7206e95f9492031f45adc35711e183'
[12:52] <cone-47> ffmpeg.git 03Martin Storsjö 07master:0679cec6e880: smacker: Make sure we don't fill in huffman codes out of range
[12:52] <cone-47> ffmpeg.git 03Michael Niedermayer 07master:f8932d990228: Merge commit '0679cec6e8802643bbe6d5f68ca1110a7d3171da'
[13:06] <cone-47> ffmpeg.git 03Martin Storsjö 07master:f3d57dc69145: smacker: Free memory properly if the init function fails
[13:06] <cone-47> ffmpeg.git 03Michael Niedermayer 07master:af539319af3a: Merge commit 'f3d57dc69145f1b7acb4870da9ce60378190a1fd'
[13:38] <cone-47> ffmpeg.git 03Martin Storsjö 07master:49568851bf17: shorten: Use a checked bytestream reader for the wave header
[13:38] <cone-47> ffmpeg.git 03Michael Niedermayer 07master:a44d35d90ffa: Merge commit '49568851bf1700e3d9ea9cda29208d0df3c2c38b'
[13:45] <cone-47> ffmpeg.git 03Martin Storsjö 07master:b26742cc3085: shorten: Break out of loop looking for fmt chunk if none is found
[13:45] <cone-47> ffmpeg.git 03Michael Niedermayer 07master:4caf8c51de93: Merge commit 'b26742cc308552f242ee2bf93b07a3ff509f4edc'
[13:57] <cone-47> ffmpeg.git 03Martin Storsjö 07master:183b9d843a95: ape: Don't allow the seektable to be omitted
[13:57] <cone-47> ffmpeg.git 03Michael Niedermayer 07master:fcfbc58a1cd7: Merge commit '183b9d843a9533774fabd3984a52f3987001acbc'
[14:07] <cone-47> ffmpeg.git 03Michael Niedermayer 07master:0d61f2600107: zmbvdec: Check the buffer size for uncompressed data
[14:07] <cone-47> ffmpeg.git 03Michael Niedermayer 07master:d6e650d88594: Merge commit '0d61f260010707f3028b818e8b24598e1a83d696'
[14:20] <cone-47> ffmpeg.git 03Martin Storsjö 07master:459f2b393a3f: mpc8: Check the seek table size parsed from the bitstream
[14:20] <cone-47> ffmpeg.git 03Michael Niedermayer 07master:ad9a877a6d77: Merge commit '459f2b393a3f89ed08d10fbceb4738d1429f268e'
[14:32] <cone-47> ffmpeg.git 03Martin Storsjö 07master:17d57848fc14: mpc8: Make sure the first stream exists before parsing the seek table
[14:32] <cone-47> ffmpeg.git 03Michael Niedermayer 07master:83d2b7e9a25e: Merge commit '17d57848fc14e82f76a65ffb25c90f2f011dc4a0'
[14:39] <cone-47> ffmpeg.git 03Martin Storsjö 07master:0f678c0214dc: aic: Validate values read from the bitstream
[14:39] <cone-47> ffmpeg.git 03Michael Niedermayer 07master:72abfba88458: Merge commit '0f678c0214dccb355ed8955077a2bea46984fbc8'
[14:43] <durandal_1707> libavcodec/aic.c:226:43: warning: comparison of unsigned expression < 0 is always false [-Wtautological-compare]
[14:44] <durandal_1707> nice "fix" that does not fix anything
[14:45] <Daemon404> durandal_1707, it is not unsigned in libav
[14:45] <durandal_1707> than even worse
[14:57] <cone-47> ffmpeg.git 03Martin Storsjö 07master:b1e469885362: g2meet: Allocate cursor buffers large enough to fit the aligned width
[14:57] <cone-47> ffmpeg.git 03Michael Niedermayer 07master:fa9e8bafc832: Merge commit 'b1e469885362febce3d9a4678624e44a92f77da9'
[15:02] <cone-47> ffmpeg.git 03Diego Biurrun 07master:e2cd766ca1b9: configure: Mention that icl does not build both static and shared libs
[15:02] <cone-47> ffmpeg.git 03Michael Niedermayer 07master:07f6a3d78734: Merge commit 'e2cd766ca1b9bb2ba72aca5f6ba5d8be15743984'
[15:07] <cone-47> ffmpeg.git 03Vittorio Giovara 07master:c0779a67e85d: mp3: add .mpa extension
[15:07] <cone-47> ffmpeg.git 03Michael Niedermayer 07master:44fb7de8ee6d: Merge commit 'c0779a67e85df856904ee6fab760c4233d4c2be5'
[15:17] <cone-47> ffmpeg.git 03Stefano Sabatini 07master:52ed1d6d5847: MAINTAINERS: move sections from lavc to lavu
[15:27] <cone-47> ffmpeg.git 03Alexandra Khirnova 07master:00a63bfb87af: matroskaenc: Fix stray pointers left over from av_reallocp_array refactoring
[15:27] <cone-47> ffmpeg.git 03Michael Niedermayer 07master:ba5e77814e5c: Merge remote-tracking branch 'qatar/master'
[15:35] <JEEB> also related to this channel as well, so linking what I just posted on matroska-devel: http://lists.matroska.org/pipermail/matroska-devel/2013-September/004567.html
[16:16] <cone-47> ffmpeg.git 03Paul B Mahol 07master:b211607b5c97: avfilter: various cosmetics
[16:24] <cone-47> ffmpeg.git 03Michael Niedermayer 07master:d2d96f9e47d6: vformat/smacker: use av_freep() to ensure no stale pointers remain
[16:24] <cone-47> ffmpeg.git 03Michael Niedermayer 07master:59a08d193963: avcodec/smacker: use the correct field instead of a hardcoded value for the index check in smacker_decode_tree()
[16:24] <cone-47> ffmpeg.git 03Michael Niedermayer 07master:fbab9a7c2267: avcodec/smacker: remove check that has become redundant
[16:24] <cone-47> ffmpeg.git 03Michael Niedermayer 07master:f144168d9b7c: avcodec/smacker: fix some memleaks
[16:50] <durandal_1707> michaelni: can you please comment on noise patch?
[16:56] <cone-47> ffmpeg.git 03Paul B Mahol 07master:3eba83daedaf: avfilter/vf_noise: unbreak filter when inline assembly is not present
[17:21] <cone-47> ffmpeg.git 03Michael Niedermayer 07master:51dab60c7b91: avfilter/vf_noise: dont corrupt the picture outside width x height
[17:21] <cone-47> ffmpeg.git 03Michael Niedermayer 07master:bfe144587177: avcodec/aic: fix "warning: comparison of unsigned expression < 0 is always false"
[17:26] <durandal_1707> michaelni: your change to noise filter made n->linesize redundant/unused
[17:27] <durandal_1707> michaelni: how you managed to corrupt picture?
[18:18] <durandal_1707> fflogger: ?
[18:19] <durandal_1707> actually its cone, silly me
[18:37] <cone-47> ffmpeg.git 03Paul B Mahol 07master:1b99514dfdc1: avcodec/smacker: use init_get_bits8()
[19:10] <durandal_1707> michaelni: so, apropo noise commit, i'm still waiting for answer
[19:23] <durandal_1707> but not about that damn padding area but for bytewidth change which looks to me completly unnecessary
[19:25] <ubitux> durandal_1707: what about noise + pad + overlay on the right ?
[19:25] <durandal_1707> yes, what about it?
[19:25] <ubitux> i didn't follow but noise is writing in the linesize padding?
[19:26] <durandal_1707> no, it does not right now
[19:26] <durandal_1707> but spp does
[19:28] <ubitux> ah i misunderstood the issue then
[19:46] <michaelni> spp is intended to work on decoder output (well normally) that is 8x8 DCT based so the buffer normally wont be width%8 != 0
[19:47] <durandal_1707> ...normally ....
[19:48] <durandal_1707> but i see your point, so i moved on...
[19:48] <durandal_1707> i hope to conver that inline asm to yasm
[19:56] <tg2> ahoy
[19:56] <tg2> who maintains this? http://ffmpeg.org/trac/ffmpeg/wiki/UbuntuCompilationGuide
[19:58] <durandal_1707> people that use Ubuntu
[19:59] <tg2> needs some updating as it doesn't work
[19:59] <tg2> who should I email
[20:00] <tg2> "MarkZV"
[20:04] <tg2> k i'll push the change up
[20:05] <llogan> tg2: what does not work about it?
[20:05] <tg2> you are putting yasm in ~/bin
[20:05] <tg2> but when compiling x264 it doesn't find yasm
[20:05] <tg2> you need to export PATH=~/bin:$PATH
[20:05] <tg2> then itw orks fine
[20:06] <llogan> what is your ubuntu version?
[20:06] <tg2> 12.04 LTS
[20:06] <tg2> . ~/.profile
[20:06] <tg2> doesn't do it
[20:06] <tg2> plus after all is said and done it is installed to ~/bin
[20:07] <tg2> which isn't in the path
[20:07] <tg2> after installing yasm
[20:07] <tg2> when following the guide verbatim
[20:07] <tg2> "which yasm" returns nothing
[20:08] <tg2> which is what x264 configure/build does
[20:08] <llogan> i'll take a look
[20:08] <tg2> http://pastie.org/pastes/8320613/text?key=e6qnnxnoznixtekhta5z7q
[20:08] <tg2> thats what I have
[20:08] <tg2> substituded libfaac since it install nicely from apt
[20:08] <tg2> and added zlib
[20:08] <tg2> and pic
[20:08] <tg2> to ffmpeg
[20:08] <tg2> but nothing else majorly changed
[20:09] <tg2> you can copy/paste that into terminal and install ffmpeg
[20:09] <llogan> why did you --enable-pic?
[20:09] <tg2> was in a config I used previously
[20:09] <tg2> not necessary
[20:10] <llogan> --enable-zlib is superfluous
[20:10] <tg2> k, they can be omitted then
[20:10] <tg2> but the path issue remains
[20:11] <llogan> ok. thanks for mentioning it. i'll take a look at it later today.
[20:11] <tg2> otehr than that, all good
[20:21] <durandal_1707> michaelni: what scan order is on your mind?
[21:01] <durandal_1707> if i use recursive approach so that each matrix get into cache, i'm afraid will get out of stack
[21:06] <michaelni> recursive needs just O(log n) stack
[21:08] <durandal_1707> where n is?
[21:10] <durandal_1707> hmm
[21:12] <michaelni> n as in pixels or dimension or width if its a square
[21:13] <michaelni> also in theory you can implement it without stack or any equivalnent
[22:28] <cone-47> ffmpeg.git 03Carl Eugen Hoyos 07master:52cf08b4c885: Store the video bit_rate in the context when muxing mxf.
[22:28] <cone-47> ffmpeg.git 03Carl Eugen Hoyos 07master:d73565d5ddf4: Use rc_max_rate if no video bit_rate was specified when muxing mxf_d10.
[22:28] <cone-47> ffmpeg.git 03Michael Bradshaw 07master:7412a4a48f23: libopenjpegdec.c: Correctly scale gray16 output if precision < 16
[22:28] <cone-47> ffmpeg.git 03Carl Eugen Hoyos 07master:f99f7f40257c: Set bits_per_raw_sample when reading j2k frames.
[22:28] <cone-47> ffmpeg.git 03Michael Niedermayer 07master:e2774b3e3b51: Merge remote-tracking branch 'cehoyos/master'
[23:20] <cone-47> ffmpeg.git 03Carl Eugen Hoyos 07release/1.1:b5e2ef771d89: Store the video bit_rate in the context when muxing mxf.
[23:20] <cone-47> ffmpeg.git 03Carl Eugen Hoyos 07release/1.1:9fde820d8e3c: Use rc_max_rate if no video bit_rate was specified when muxing mxf_d10.
[23:20] <cone-47> ffmpeg.git 03Carl Eugen Hoyos 07release/1.2:3a6fbb0c8e06: Store the video bit_rate in the context when muxing mxf.
[23:20] <cone-47> ffmpeg.git 03Carl Eugen Hoyos 07release/1.2:b4f90013ab4f: Use rc_max_rate if no video bit_rate was specified when muxing mxf_d10.
[23:20] <cone-47> ffmpeg.git 03Carl Eugen Hoyos 07release/2.0:aaef59d53538: Store the video bit_rate in the context when muxing mxf.
[23:20] <cone-47> ffmpeg.git 03Carl Eugen Hoyos 07release/2.0:09bc4be3dbb3: Use rc_max_rate if no video bit_rate was specified when muxing mxf_d10.
[00:00] --- Fri Sep 13 2013