[Ffmpeg-devel-irc] ffmpeg-devel.log.20150730

[burek]

[00:01:40 CEST] <cone-183> ffmpeg 03wm4 07master:bf51fcd304d5: rawdec: fix mjpeg probing
[00:01:54 CEST] <jnoring> Hi--what's the status of hosting for ffmpeg?  I see there's a long thread on the mailing list, but not sure if there was a conclusion.
[00:18:22 CEST] <cone-183> ffmpeg 03wm4 07master:4c6beaed9210: rawdec: fix mjpeg probing buffer size check
[01:26:09 CEST] <cone-183> ffmpeg 03James Almer 07master:7e38340b831a: hmac: add support for SHA-2
[01:26:10 CEST] <cone-183> ffmpeg 03Michael Niedermayer 07master:865564c466c3: Merge commit '7e38340b831af8e3271f22bd66b9da14953c02af'
[01:26:11 CEST] <cone-183> ffmpeg 03Michael Niedermayer 07master:8c5cc7b7fba0: avutil/hmac: support another ABI difference to the fork
[01:35:55 CEST] <cone-183> ffmpeg 03James Almer 07master:e59f7cd89e7f: fate: add test vectors for HMAC SHA and SHA-2
[01:35:56 CEST] <cone-183> ffmpeg 03Michael Niedermayer 07master:277a5005b78d: Merge commit 'e59f7cd89e7f4f096f8456d63191adec2f8b1083'
[01:38:20 CEST] <jamrial> michaelni: libav is about to force a major bump, so i think it's better to just add the new enums and keep ours renamed as deprecated until then, when the two projects can become synced without ABI problems
[01:38:42 CEST] <jamrial> i'll send a patch
[01:48:51 CEST] <cone-183> ffmpeg 03James Almer 07master:65dd6a1f84cc: rtmpproto: use AVHMAC instead of a custom implementation
[01:48:52 CEST] <cone-183> ffmpeg 03Michael Niedermayer 07master:15a7a3cf0aaa: Merge commit '65dd6a1f84cc5c9a91d279e8e8b72ff7adf58d31'
[02:03:47 CEST] <jamrial> michaelni: patch in the ml
[02:27:12 CEST] <BtbN> "The compiler is GCC 4.2.1."
[02:27:16 CEST] <BtbN> oO
[02:28:39 CEST] <jamrial> lol
[02:32:37 CEST] <cone-183> ffmpeg 03James Almer 07master:bc397246aa08: avutil/hmac: fix API/ABI compatibility with the fork
[02:37:24 CEST] <BtbN> "and what we've used to build FFmpeg for quite a few releases now"
[02:37:37 CEST] <BtbN> well, must have been realy quite a few ones
[02:38:32 CEST] <jamrial> maybe this is some new mips code added in 2.7 that uses some preprocessor features that weren't available in 4.2
[02:38:53 CEST] <jamrial> fate has a handful of mips clients, but the oldest compiler used there is 4.3, and it builds fine
[02:40:04 CEST] <jamrial> and really, they should update. i recall gcc 4.2 would miscompile shit all the time, at least on x86
[03:22:05 CEST] <cone-183> ffmpeg 03Michael Niedermayer 07master:363995296f65: doc/APIchanges: fill in some missing things
[03:29:20 CEST] <cone-183> ffmpeg 03James Almer 07master:222d7619cea3: avfilter/vf_pnsr: fix author name
[04:57:33 CEST] <cone-183> ffmpeg 03James Almer 07master:6ce3d9fc4a70: avformat/rtmpproto: free hmac context properly
[05:00:16 CEST] <cone-183> ffmpeg 03James Zern 07master:88df16fe6dd1: libvpxenc: quiet coded_frame deprecation warnings
[05:00:17 CEST] <cone-183> ffmpeg 03James Zern 07master:9723d439fdeb: libvpxenc: report pict_type/error via side-data
[05:11:06 CEST] <cone-183> ffmpeg 03Michael Niedermayer 07master:bf2474c74f2c: avcodec/mpeg12enc: extend QP range to 28 for non linear quantizers
[05:11:07 CEST] <cone-183> ffmpeg 03Michael Niedermayer 07master:f977e69dacf0: avcodec/mpegvideo_enc: Ignore QMAX if VBV constraints are exceeded
[06:04:55 CEST] <rcombs> ubitux: mind reviewing this, if you've got a moment? https://ffmpeg.org/pipermail/ffmpeg-devel/2015-July/175655.html
[12:24:01 CEST] <cone-638> ffmpeg 03James Almer 07master:9487ffd4c02b: rtmpproto: free hmac context properly
[12:24:02 CEST] <cone-638> ffmpeg 03Michael Niedermayer 07master:dd0dbaccb74e: Merge commit '9487ffd4c02b9e261562d43735490068c0df0d4b'
[12:36:38 CEST] <cone-638> ffmpeg 03Martin Storsjö 07master:bf0cef5c3a11: checkasm: Include io.h for isatty, if available
[12:36:39 CEST] <cone-638> ffmpeg 03Michael Niedermayer 07master:1919827f2c2b: Merge commit 'bf0cef5c3a114df452e5476167634dd8f51eb448'
[14:10:14 CEST] <BBB> do we test abi compat with fork in any systematic way? e.g. do we compile fate against libav-libraries and then run with binary libraries from ffmpeg and see if it passes?
[14:11:11 CEST] <nevcairiel> it probably wouldnt work anyway =p
[14:11:30 CEST] <nevcairiel> maintaining ABI compat is futile anyway, its already quite fragile as it is
[14:28:43 CEST] <wm4> he's gone, but this is how we "test" it: http://git.videolan.org/?p=ffmpeg.git;a=blob;f=libavutil/utils.c;h=2e3b1e03b4d30e01897b2020433859ec0ff716ae;hb=HEAD#l44
[14:43:33 CEST] <cone-638> ffmpeg 03Ganesh Ajjanagadde 07master:92e62f49cf74: ffmpeg: modify tty state when stderr is redirected
[14:52:55 CEST] <wm4> BBB: <wm4> he's gone, but this is how we "test" it: http://git.videolan.org/?p=ffmpeg.git;a=blob;f=libavutil/utils.c;h=2e3b1e03b4d30e01897b2020433859ec0ff716ae;hb=HEAD#l44
[14:53:15 CEST] <wm4> or does this count as trolling?P
[14:53:22 CEST] <BBB> thats trolling, yes
[14:53:25 CEST] <BBB> do we have functional tests?
[14:53:27 CEST] <BBB> :-p
[14:53:34 CEST] <wm4> not that I know of
[14:54:00 CEST] <wm4> and this whole thing was just a bad hack at the start of the fork business anyway
[14:54:12 CEST] <wm4> even if there was a need once (which I doubt), it's long gone
[14:58:52 CEST] <ubitux> we're probably going to be able to clean up some "compatibility" stuff in the future
[14:59:19 CEST] <BBB> that would be nice
[14:59:39 CEST] <BBB> I mean, unless we have functional tests, I dont believe its useful
[14:59:45 CEST] <BBB> it most likely doesnt work very reliably anyway
[14:59:50 CEST] <BBB> which is worse than not working at all
[16:03:09 CEST] <cone-638> ffmpeg 03Michael Niedermayer 07master:2d80de0be2b8: avcodec/v408enc: remove coded_frame usage
[16:03:10 CEST] <cone-638> ffmpeg 03Michael Niedermayer 07master:a0123b602364: avcodec/v308enc: Remove coded_frame usage
[16:22:27 CEST] <cone-638> ffmpeg 03Michael Niedermayer 07master:9ebffc5bba36: avcodec/yuv4enc: drop coded_frame usage
[16:22:28 CEST] <cone-638> ffmpeg 03Michael Niedermayer 07master:77d8fd5c9c1d: avcodec/y41penc: Drop coded_frame usage
[16:34:26 CEST] <cone-638> ffmpeg 03Luca Barbato 07master:5ae178539b91: http: Add the trailing endlines if they are missing
[16:34:27 CEST] <cone-638> ffmpeg 03Michael Niedermayer 07master:8e67599505e3: Merge commit '5ae178539b91d25710b7bb322d156c31aea9f8bf'
[16:42:19 CEST] <ubitux> rcombs: ah yeah sorry i forgot to answer
[16:42:32 CEST] <ubitux> going to have a look now
[16:43:33 CEST] <wm4> is it correct that incoming/system.php renders something?
[16:45:02 CEST] <wm4> or did it get h4x0red
[16:47:14 CEST] <ubitux> rcombs: you can have tag_name local to the sscanf if scope, right?
[16:48:06 CEST] <wm4> rcombs: where's your mp4 sidx parsing patch? didn't you write one?
[17:02:29 CEST] <ubitux> rcombs: looks pretty fine otherwise
[17:49:46 CEST] <durandal_170> mpv uses audioconvert... in debian
[17:51:15 CEST] <jamrial> durandal_170: the lavu header, not lavc's functions
[17:53:57 CEST] <wm4> durandal_170: huh? it never did
[17:56:40 CEST] <nevcairiel> Why is it supposed to be our job to fix downstreams
[17:57:14 CEST] <nevcairiel> If they are unmaintained, stop using them
[17:57:23 CEST] <nevcairiel> If they have maintainers, notify then
[18:01:38 CEST] <wm4> the was also a deprecation period of 3 years or longer
[18:04:09 CEST] <nevcairiel> Yeah these are really old things
[18:05:31 CEST] <cone-638> ffmpeg 03Luca Barbato 07master:0c73a5a53cc9: hls: Save and forward avio options
[18:05:32 CEST] <cone-638> ffmpeg 03Michael Niedermayer 07master:ba12ba859aab: Merge commit '0c73a5a53cc97f4291bbe9e1e68226edf6161744'
[18:37:48 CEST] <cone-638> ffmpeg 03James Almer 07master:6877537ff321: hmac: add missing version bump and APIChanges entry
[18:37:49 CEST] <cone-638> ffmpeg 03Michael Niedermayer 07master:d560c61125b3: Merge commit '6877537ff321136cf78ab03b8b91e6b6f218db5b'
[18:45:48 CEST] <cone-638> ffmpeg 03Martin Storsjö 07master:764ec7014972: rtsp: Only interpret $ as interleaved packet indicator at the start of replies
[18:45:49 CEST] <cone-638> ffmpeg 03Michael Niedermayer 07master:bc4bdc6d6152: Merge commit '764ec70149728be82304c163ccc4e280f1629201'
[19:24:56 CEST] <cone-638> ffmpeg 03Michael Niedermayer 07release/2.5:ef905b1b27d5: avcodec/diracdec: Check slices malloc and propagate error code
[19:24:57 CEST] <cone-638> ffmpeg 03Paul B Mahol 07release/2.5:ff68bf7a1bf3: avformat/nut: support WavPack
[19:24:58 CEST] <cone-638> ffmpeg 03Andreas Cadhalpun 07release/2.5:74a88a509c6e: avidec: check for valid bit_rate range
[19:24:59 CEST] <cone-638> ffmpeg 03Andreas Cadhalpun 07release/2.5:18189ebab3cf: libshine: fix support for shine 3.0
[19:25:00 CEST] <cone-638> ffmpeg 03Andreas Cadhalpun 07release/2.5:98aedbf9fe38: mpegaudiodec: copy AVFloatDSPContext from first context to all contexts
[19:25:01 CEST] <cone-638> ffmpeg 03Andreas Cadhalpun 07release/2.5:983dc64772a0: vc1dec: use get_bits_long and limit the read bits to 32
[19:25:02 CEST] <cone-638> ffmpeg 03Andreas Cadhalpun 07release/2.5:d8c82ed321da: huffyuvdec: validate image size
[19:25:03 CEST] <cone-638> ffmpeg 03Andreas Cadhalpun 07release/2.5:ffcfab882b66: snow: remove an obsolete av_assert2
[19:25:04 CEST] <cone-638> ffmpeg 03Andreas Cadhalpun 07release/2.5:5e833d958c09: Changelog update
[19:38:00 CEST] <cone-638> ffmpeg 03wm4 07release/2.5:c809cf1c5b1b: rawdec: fix mjpeg probing
[19:38:01 CEST] <cone-638> ffmpeg 03wm4 07release/2.5:1eb646ec9f87: rawdec: fix mjpeg probing buffer size check
[20:13:46 CEST] <cone-638> ffmpeg 03Michael Niedermayer 07master:dfe8e3b3819d: avcodec/avuienc: Drop coded_frame usage
[20:13:47 CEST] <cone-638> ffmpeg 03Michael Niedermayer 07master:802e51299e6d: avcodec/r210enc: Drop coded_frame usage
[20:48:01 CEST] <slacker> Hi, is it possible to use harware accleration to accelrate video encoding using ffmpeg?  I haven't installed proprietary drivers.  VAAPI with intel intregrated graphics a good idea?
[20:54:25 CEST] <llogan> that's a probably question for #ffmpeg, unless you're intending to do some development work on FFmpeg
[20:55:13 CEST] <slacker> umm... I'll post it there, Thank you!
[23:57:13 CEST] <cone-228> ffmpeg 03Michael Niedermayer 07master:d2077c860e56: Revert "ffmpeg: modify tty state when stderr is redirected"
[00:00:00 CEST] --- Fri Jul 31 2015