[Ffmpeg-devel-irc] ffmpeg-devel.log.20120807

[burek]

[00:27] <CIA-41> ffmpeg: 03Michael Niedermayer 07master * rcde4741177 10ffmpeg/libavcodec/h264.c: 
[00:27] <CIA-41> ffmpeg: h264: disable assert on current_picture_ptr being null.
[00:27] <CIA-41> ffmpeg: It is possible in various error pathes as well as gap handling
[00:27] <CIA-41> ffmpeg: that this has already been allocated. Its not clear why that
[00:27] <CIA-41> ffmpeg: would be a problem with the current code, thus disable the
[00:27] <CIA-41> ffmpeg: assert to avoid common assert failure when asserts are enabled.
[00:27] <CIA-41> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[01:26] <CIA-41> ffmpeg: 03Michael Niedermayer 07master * r3865ec2ace 10ffmpeg/libavcodec/proresenc_kostya.c: 
[01:26] <CIA-41> ffmpeg: proresenc_kostya: do not attempt to free random things
[01:26] <CIA-41> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[01:29] <iive> LoL
[01:55] <durandal_1707> copy-paste fail
[02:24] <durandal_1707> http://transcoding.wordpress.com/2011/11/16/careful-with-audio-resampling-using-ffmpeg/
[04:46] <CIA-41> ffmpeg: 03Michael Niedermayer 07master * r29d1df66ad 10ffmpeg/libavcodec/mpegaudiodec.c: 
[04:46] <CIA-41> ffmpeg: mpegaudiodec: replace assert() by check under #ifdef DEBUG
[04:46] <CIA-41> ffmpeg: The assert can be false with some invalid inputs, the check is
[04:46] <CIA-41> ffmpeg: too expensive to always do though for just a warning message.
[04:46] <CIA-41> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[05:12] <durandal_1707> hmm if that warning is not really relevant just make is av_dlog
[06:04] <CIA-41> ffmpeg: 03Paul B Mahol 07master * r011ce89dba 10ffmpeg/ (5 files in 2 dirs): 
[06:04] <CIA-41> ffmpeg: avcodec: add bmp parser
[06:04] <CIA-41> ffmpeg: Signed-off-by: Paul B Mahol <onemda at gmail.com>
[06:11] <michaelni> could be done, but while the av_dlog() can be remvoed by the preprocessor the check would not
[06:11] <michaelni> and that may have negative speed effects, the ifdef avoids this possibility
[06:12] Action: michaelni doesnt trust gcc in removing if(a > b) ;
[06:12] <michaelni> without sideeffects that is
[06:12] <durandal_1707> check is not needed ....
[06:15] <michaelni> please elaborate
[06:16] <durandal_1707> av_dlog is DEBUG and not WARNING....
[06:16] <michaelni> the code checks if e is too small and then prints a warning
[06:17] <durandal_1707> i see no point in  that...
[06:18] <durandal_1707> but whatever, there should be some macro to make such stuff takes less code
[06:18] <michaelni> yes
[06:20] <michaelni> some av_cond_dlog(condition, context, level, message, ...) maybe that gets converted to if(condition) av_log ...
[06:21] Action: michaelni is tired falls asleep on his keyboard
[07:33] <CIA-41> ffmpeg: 03Clément BSsch 07master * rcf753d5b59 10ffmpeg/doc/filters.texi: 
[07:33] <CIA-41> ffmpeg: doc: fix misspelled "convergence".
[07:33] <CIA-41> ffmpeg: Found-by: David Rice
[10:41] <ubitux> ok, edgedetect filter complete
[10:41] <ubitux> (will submit tonight)
[10:41] <ubitux> do we have a "difference" filter of some kind?
[10:41] <mateo`> ubitux: great :)
[10:41] <ubitux> :D
[10:42] <ubitux> http://ubitux.fr/pub/pics/edgedetect.png
[10:42] <ubitux> http://ubitux.fr/pub/pics/edgedetect2.png
[10:42] <ubitux> it seems the results are consistent with the examples i see
[13:32] <CIA-41> ffmpeg: 03Michael Niedermayer 07master * r7febc5aa93 10ffmpeg/libavcodec/arm/ (h264dsp_init_arm.c h264pred_init_arm.c): 
[13:32] <CIA-41> ffmpeg: arm: fix compile with disable-optimizations on android
[13:32] <CIA-41> ffmpeg: Fixes Ticket1241
[13:32] <CIA-41> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[13:45] <CIA-41> ffmpeg: 03Michael Niedermayer 07release/0.11 * ra19010f7a6 10ffmpeg/libavcodec/wmaenc.c: 
[13:45] <CIA-41> ffmpeg: wmaenc: dont mess with the bitrate.
[13:45] <CIA-41> ffmpeg: The bitrate is not writeable by an encoder.
[13:45] <CIA-41> ffmpeg: Fixes generation of invalid wma
[13:45] <CIA-41> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[13:45] <CIA-41> ffmpeg: 03Michael Niedermayer 07release/0.11 * r606538df6c 10ffmpeg/libavformat/oggdec.c: (log message trimmed)
[13:45] <CIA-41> ffmpeg: oggdec: check stream index before using it in ogg_get_length()
[13:45] <CIA-41> ffmpeg: Fixes crash based on a uninitialized array index read.
[13:45] <CIA-41> ffmpeg: If the read does not crash then out of array writes based
[13:45] <CIA-41> ffmpeg: on the same index might have been triggered afterwards.
[13:45] <CIA-41> ffmpeg: Found-by: inferno at chromium.org
[13:45] <CIA-41> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[13:45] <CIA-41> ffmpeg: 03Reimar Döffinger 07release/0.11 * r5231a35559 10ffmpeg/libavformat/oggdec.c: 
[13:45] <CIA-41> ffmpeg: oggdec: simplify start time calculation code.
[13:45] <CIA-41> ffmpeg: Also slightly more correct behaviour in case streams_left for
[13:45] <CIA-41> ffmpeg: some reason is 0 from the start.
[13:45] <CIA-41> ffmpeg: Signed-off-by: Reimar Döffinger <Reimar.Doeffinger at gmx.de>
[14:08] <ubitux> is there any socis student here?
[14:09] <ubitux> https://ffmpeg.org/trac/ffmpeg/ticket/1615  looks like a relatively simple task to do
[16:40] <yao_> hi, any idea why e500mc cpu is not supported in the libav's configure?
[18:07] <Daemon404> yao_, hi
[18:07] <Daemon404> im working on it
[18:07] <Daemon404> <--- Derek
[18:07] <yao_> hi Daemon404
[18:08] <Daemon404> im just doing a test build
[18:08] <yao_> thanks Daemon404, I really have no idea the cpuflags should be
[18:08] <Daemon404> configure currently doesn't support it
[18:08] <Daemon404> im adding support
[18:08] <yao_> exactly
[18:09] <Daemon404> gimme about 10 mins
[18:10] <yao_> np
[18:10] <Daemon404> yao_, you work for wind river, right?
[18:10] <yao_> yes
[18:10] <Daemon404> i interned there last year
[18:11] <yao_> really?
[18:11] <Daemon404> yup
[18:11] <yao_> usa department?
[18:11] <Daemon404> ottawa office
[18:11] <yao_> I work for Ottawa too, but I am new here
[18:12] <Daemon404> ah... i was there from jan2011->sept2011
[18:12] <Daemon404> randy knows who i am.
[18:13] <yao_> oh, Randy is on vacation
[18:13] <Daemon404> i know
[18:16] <Daemon404> yao_, brb after lunch...
[18:17] <yao_> sure
[18:17] <CIA-41> ffmpeg: 03Michael Niedermayer 07master * r2f6f2f4f73 10ffmpeg/libavcodec/mpeg12.c: 
[18:17] <CIA-41> ffmpeg: mpeg12dec: move checks out of decode picture coding extension.
[18:17] <CIA-41> ffmpeg: Fixes out of array reads
[18:17] <CIA-41> ffmpeg: Fixes Ticket1330
[18:17] <CIA-41> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[19:03] <Daemon404> yao_, i have a patch on the mailing list, but I need someone to test it on actual hardware.
[19:04] <yao_> what hardware u need?
[19:05] <Daemon404> i just need to make sure it compiles correctly and runs on an e500mc
[19:05] <Daemon404> i only have a toolchain here... nothing to test it on
[19:06] <yao_> I have an e500mc target but never test gstreamer
[19:07] <Daemon404> well... patch is sitting on the mailing list in any case :)
[19:08] <yao_> let me see whether the patch is simple enough no need testing:)
[19:12] <yao_> Daemon404: the patch is simple enough, -mcpu=e500mc is correct, our build is based on -mcpu=e500mc, I thought there is other optimization can be done with the e500mc
[19:15] <Daemon404> if youre talkign about -mhard-float, i think gcc should pass that on its own when you call -mcpu=e500mc
[19:17] <yao_> it is default when cpu is e500mc?
[19:19] <Daemon404> daemon404 at bbvm:~/dev/f/ffmpeg$ powerpc-poky-linux-gcc -c -mcpu=e500mc test.c -o test1.o
[19:19] <Daemon404> daemon404 at bbvm:~/dev/f/ffmpeg$ powerpc-poky-linux-gcc -c -mcpu=e500mc -mhard-float test.c -o test2.o
[19:19] <Daemon404> daemon404 at bbvm:~/dev/f/ffmpeg$ md5sum *.o
[19:19] <Daemon404> f52d25cc08954fe9d721140472580d0c  test1.o
[19:19] <Daemon404> f52d25cc08954fe9d721140472580d0c  test2.o
[19:20] <Daemon404> oh... wait i guess it would make sense to actually test a file with floating poitn arithmetic..
[19:20] <Daemon404> -_-
[19:22] <Daemon404> still identical
[19:52] <CIA-41> ffmpeg: 03Michael Niedermayer 07master * r8993c25695 10ffmpeg/libavformat/avidec.c: 
[19:52] <CIA-41> ffmpeg: avidec: fix AVStream.info memleak with dv.
[19:52] <CIA-41> ffmpeg: Fixes Ticket1334
[19:52] <CIA-41> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[20:08] <beastd> hi
[20:12] <CIA-41> ffmpeg: 03Stefano Sabatini 07master * r13bbc464ed 10ffmpeg/doc/examples/muxing.c: (log message trimmed)
[20:12] <CIA-41> ffmpeg: examples/muxing: set codec in the codec context
[20:12] <CIA-41> ffmpeg: Make add_{audio,video}_stream return the found codec, and pass it to
[20:12] <CIA-41> ffmpeg: avcodec_open2() through open_{audio,video} functions.
[20:12] <CIA-41> ffmpeg: Fix failure:
[20:12] <CIA-41> ffmpeg: No codec provided to avcodec_open2().
[20:12] <CIA-41> ffmpeg: could not open codec
[20:12] <CIA-41> ffmpeg: 03Paul B Mahol 07master * r13c5069187 10ffmpeg/ (6 files in 3 dirs): (log message trimmed)
[20:12] <CIA-41> ffmpeg: lavfi: add smptebars source
[20:12] <CIA-41> ffmpeg: Patch readapted by Stefano Sabatini, color values proposed by Tim
[20:12] <CIA-41> ffmpeg: Nicholson <nichot20 at yahoo.com>.
[20:12] <CIA-41> ffmpeg: Address trac ticket #1462.
[20:12] <CIA-41> ffmpeg: See thread:
[20:12] <CIA-41> ffmpeg: Subject: [FFmpeg-devel] [PATCH] smptebars filter
[20:23] <michaelni> hi beastd 
[20:25] Action: beastd is looking at the make uninstall fix from jamal
[21:11] <CIA-41> ffmpeg: 03Derek Buitenhuis 07master * r9b34ae4a65 10ffmpeg/configure: 
[21:11] <CIA-41> ffmpeg: configure: Add support for e500mc CPU
[21:11] <CIA-41> ffmpeg: Signed-off-by: Derek Buitenhuis <derek.buitenhuis at gmail.com>
[21:11] <CIA-41> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[21:11] <CIA-41> ffmpeg: 03Michael Niedermayer 07master * rbb7744a45b 10ffmpeg/libavcodec/pnm.c: 
[21:11] <CIA-41> ffmpeg: pnmdec: make pnm_get more robust
[21:11] <CIA-41> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[21:11] <CIA-41> ffmpeg: 03Michael Niedermayer 07master * r0049af2628 10ffmpeg/libavcodec/pnm.c: 
[21:11] <CIA-41> ffmpeg: pnmdec: make ff_pnm_decode_header() more robust
[21:11] <CIA-41> ffmpeg: Fixes ticket1321
[21:11] <CIA-41> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[22:03] <CIA-41> ffmpeg: 03Justin Ruggles 07master * r2fa57c9daf 10ffmpeg/libavformat/mpc8.c: 
[22:03] <CIA-41> ffmpeg: mpc8: set packet duration and stream start time instead of tracking frames
[22:03] <CIA-41> ffmpeg: Fixes mpc8 timestamps.
[22:03] <CIA-41> ffmpeg: 03Justin Ruggles 07master * rfdbeae4490 10ffmpeg/libavcodec/mpc8.c: 
[22:03] <CIA-41> ffmpeg: mpc8: add a flush function
[22:03] <CIA-41> ffmpeg: Ensures that the next frame decoded after seeking will be decoded as a
[22:03] <CIA-41> ffmpeg: keyframe.
[22:03] <CIA-41> ffmpeg: 03Michael Niedermayer 07master * rbc773d0d42 10ffmpeg/: (log message trimmed)
[22:03] <CIA-41> ffmpeg: Merge commit 'f5d2c597e99af218b0d4d1cf9737c7e68ee934e4'
[22:03] <CIA-41> ffmpeg: * commit 'f5d2c597e99af218b0d4d1cf9737c7e68ee934e4':
[22:03] <CIA-41> ffmpeg:  build: fix library installation on cygwin
[22:04] <CIA-41> ffmpeg:  mpc8: add a flush function
[22:04] <CIA-41> ffmpeg:  mpc8: set packet duration and stream start time instead of tracking frames
[22:04] <CIA-41> ffmpeg: Conflicts:
[22:04] <CIA-41> ffmpeg: 03Yaakov Selkowitz 07master * rf5d2c597e9 10ffmpeg/configure: 
[22:04] <CIA-41> ffmpeg: build: fix library installation on cygwin
[22:04] <CIA-41> ffmpeg: This installs libraries using the proper names and locations,
[22:04] <CIA-41> ffmpeg: generates an import lib for the DLL, and drops no longer needed
[22:04] <CIA-41> ffmpeg: linker flags.
[22:04] <CIA-41> ffmpeg: Signed-off-by: Mans Rullgard <mans at mansr.com>
[22:39] <CIA-41> ffmpeg: 03jamal 07master * r7c5466d4fc 10ffmpeg/common.mak: 
[22:39] <CIA-41> ffmpeg: checkheaders: Fix directory creation in out-of-tree builds.
[22:39] <CIA-41> ffmpeg: checkheaders wasn't creating folders as needed, so if it was run
[22:39] <CIA-41> ffmpeg: immediately after the configure script it would fail as soon as
[22:39] <CIA-41> ffmpeg: it tried to compile headers inside the ARCH folders.
[22:39] <CIA-41> ffmpeg: Signed-off-by: Alexander Strasser <eclipse7 at gmx.net>
[22:39] <CIA-41> ffmpeg: 03jamal 07master * r49440853d0 10ffmpeg/library.mak: 
[22:39] <CIA-41> ffmpeg: build: Fix some paths in uninstall-libs
[22:39] <CIA-41> ffmpeg: Folder and file names weren't being separated with a slash.
[22:39] <CIA-41> ffmpeg: This resulted in .dll.a, .lib and .def files not being removed on uninstall.
[22:39] <CIA-41> ffmpeg: Signed-off-by: Alexander Strasser <eclipse7 at gmx.net>
[22:51] <CIA-41> ffmpeg: 03Anton Khirnov 07master * r104e10fb42 10ffmpeg/ (5 files in 2 dirs): lavc: add AV prefix to codec ids.
[22:51] <CIA-41> ffmpeg: 03Michael Niedermayer 07master * r7a72695c05 10ffmpeg/: (log message trimmed)
[22:51] <CIA-41> ffmpeg: Merge commit '36ef5369ee9b336febc2c270f8718cec4476cb85'
[22:51] <CIA-41> ffmpeg: * commit '36ef5369ee9b336febc2c270f8718cec4476cb85':
[22:51] <CIA-41> ffmpeg:  Replace all CODEC_ID_* with AV_CODEC_ID_*
[22:51] <CIA-41> ffmpeg:  lavc: add AV prefix to codec ids.
[22:51] <CIA-41> ffmpeg: Conflicts:
[22:51] <CIA-41> ffmpeg:  doc/APIchanges
[22:51] <CIA-41> ffmpeg: 03Anton Khirnov 07master * r36ef5369ee 10ffmpeg/ (582 files in 8 dirs): Replace all CODEC_ID_* with AV_CODEC_ID_*
[22:52] <durandal_1707> what a waste
[22:53] <j-b> waste?
[22:54] <durandal_1707> last do ranom rename every next random day
[22:59] <CIA-41> ffmpeg: 03Paul B Mahol 07master * r3662f787e7 10ffmpeg/libavcodec/gifdec.c: 
[22:59] <CIA-41> ffmpeg: gifdec: pass avctx to av_log()
[22:59] <CIA-41> ffmpeg: Signed-off-by: Paul B Mahol <onemda at gmail.com>
[22:59] <CIA-41> ffmpeg: 03Paul B Mahol 07master * rf64288960b 10ffmpeg/libavcodec/gifdec.c: 
[22:59] <CIA-41> ffmpeg: gifdec: remove redundant "gif: " from av_dlog()
[22:59] <CIA-41> ffmpeg: Signed-off-by: Paul B Mahol <onemda at gmail.com>
[22:59] <CIA-41> ffmpeg: 03Paul B Mahol 07master * r2da5a5ce00 10ffmpeg/libavcodec/gifdec.c: 
[22:59] <CIA-41> ffmpeg: gifdec: return more meaningful error codes
[22:59] <CIA-41> ffmpeg: Signed-off-by: Paul B Mahol <onemda at gmail.com>
[23:09] <CIA-41> ffmpeg: 03Mans Rullgard 07master * ra3df4781f4 10ffmpeg/ (18 files in 3 dirs): 
[23:09] <CIA-41> ffmpeg: x86: add colons after labels
[23:09] <CIA-41> ffmpeg: nasm prints a warning if the colon is missing.
[23:09] <CIA-41> ffmpeg: Signed-off-by: Mans Rullgard <mans at mansr.com>
[23:09] <CIA-41> ffmpeg: 03Mans Rullgard 07master * r2b140a3d09 10ffmpeg/libavcodec/x86/ (h264_deblock_10bit.asm rv34dsp.asm rv40dsp.asm): 
[23:09] <CIA-41> ffmpeg: x86: use 32-bit source registers with movd instruction
[23:09] <CIA-41> ffmpeg: yasm tolerates mismatch between movd/movq and source register size,
[23:09] <CIA-41> ffmpeg: adjusting the instruction according to the register. nasm is more
[23:09] <CIA-41> ffmpeg: strict.
[23:09] <CIA-41> ffmpeg: Signed-off-by: Mans Rullgard <mans at mansr.com>
[23:09] <CIA-41> ffmpeg: 03Mans Rullgard 07master * r7238265052 10ffmpeg/libavutil/x86/x86inc.asm: 
[23:09] <CIA-41> ffmpeg: x86: fix rNmp macros with nasm
[23:09] <CIA-41> ffmpeg: For some reason, nasm requires this. No harm done to yasm.
[23:09] <CIA-41> ffmpeg: Signed-off-by: Mans Rullgard <mans at mansr.com>
[23:09] <CIA-41> ffmpeg: 03Mans Rullgard 07master * redd8226795 10ffmpeg/libavutil/x86/x86inc.asm: 
[23:09] <CIA-41> ffmpeg: x86: fix build with nasm 2.08
[23:09] <CIA-41> ffmpeg: It appears that something goes wrong in old nasm versions when the
[23:09] <CIA-41> ffmpeg: %+ operator is used in the last argument of a macro invocation and
[23:09] <CIA-41> ffmpeg: this argument is tested with %ifdef within the macro. This patch
[23:09] <CIA-41> ffmpeg: rearranges the macro arguments such that the %+ operator is never
[23:09] <CIA-41> ffmpeg: * qatar/master:
[23:09] <CIA-41> ffmpeg:  x86: fix build with nasm 2.08
[23:09] <CIA-41> ffmpeg:  x86: use nop cpu directives only if supported
[23:09] <CIA-41> ffmpeg:  x86: fix rNmp macros with nasm
[23:09] <CIA-41> ffmpeg:  build: add trailing / to yasm/nasm -I flags
[23:09] <CIA-41> ffmpeg: 03Mans Rullgard 07master * rf295fee2c9 10ffmpeg/Makefile: 
[23:10] <CIA-41> ffmpeg: build: add trailing / to yasm/nasm -I flags
[23:10] <CIA-41> ffmpeg: nasm requires a trailing / on paths specified with -I.
[23:10] <CIA-41> ffmpeg: It does no harm with yasm.
[23:10] <CIA-41> ffmpeg: Signed-off-by: Mans Rullgard <mans at mansr.com>
[23:39] <saste> ffmpeg-soc: This list is about development of the FFmpeg Google Summer of Code projects. 
[23:40] <saste> ffmpeg-soc is not used since ages, should we remove it from "contact"?
[23:41] <j-b> yes
[23:43] <saste> i agree, will send a patch soon
[00:00] --- Wed Aug  8 2012