<    April 2017    >
Su Mo Tu We Th Fr Sa  
                   1  
 2  3  4  5  6  7  8  
 9 10 11 12 13 14 15  
16 17 18 19 20 21 22  
23 24 25 26 27 28 29  
30
00:04 durandal_1707 joined
00:24 McGuyver joined
00:25 McGuyver joined
00:35 XSoul joined
00:52 edong23 joined
00:52 iml_ joined
00:52 azeam joined
01:53 azeam_afk joined
02:20 azeam_afk joined
02:26 chewey joined
02:44 azeam joined
03:02 oneforall2 joined
04:58 hfb joined
05:07 ggherdov-_ joined
05:08 edong23_ joined
05:09 d0ggie_ joined
05:13 ivanich joined
05:13 SirCmpwn_ joined
05:20 michaelni joined
05:53 madgoat joined
05:53 madgoat left
05:57 durandal_170 joined
05:57 durandal_170 joined
06:26 PaulePanter joined
06:34 michaelni joined
06:47 Unity joined
06:47 Unity joined
07:43 mrklintscher joined
07:45 ivanich joined
08:29 spaam joined
08:29 spaam joined
08:44 spaam joined
08:45 spaam joined
08:46 spaam joined
09:33 irclogger_com joined
09:33 Topic for
10:10 cybrNaut joined
10:28 scartozzo joined
10:29 scartozzo joined
10:54 courrier joined
10:55 iive joined
10:55 iive joined
11:13 durandal_1707 joined
11:36 durandal_1707 joined
11:43 spaam left
12:03 durandal_1707 joined
12:03 durandal_1707 joined
12:21 cesdo joined
12:35 mrklintscher joined
12:51 durandal_170 joined
12:51 durandal_170 joined
13:16 scartozzo joined
13:25 cristian_c joined
13:27 cybrNaut joined
13:59 durandal_1707 joined
13:59 durandal_1707 joined
14:18 hfb joined
14:59 iml_ joined
15:20 kelnoky joined
16:42 cybrNaut joined
16:48 iml_ joined
16:54 cybrNaut joined
16:57 cybrNaut joined
17:37 azeam_afk joined
17:37 azeam joined
18:26 Qantourisc joined
18:28 <Qantourisc> Weird problem: video is lagging ... for 500mS (with new PC (sound-card) )
18:29 <Qantourisc> is async "normal" ?
18:29 <Qantourisc> and just correct it, or shouild this normally be in sync per default
18:35 krabador joined
18:40 <Qantourisc> lavdopts=threads=16 <= that was the cause
19:22 azeam joined
19:33 balkamos joined
19:34 azeam_afk joined
19:39 ivanich joined
19:39 iml_ joined
19:54 azeam_afk joined
20:19 nahra` joined
21:00 courrier joined
21:01 nahra` joined
21:12 nahra` joined
22:38 AndrewR joined
22:44 <AndrewR> hi ... I run into interesting crash while trying to measure how fast dv amd mjpeg encoding can be (on multicore machine). https://pastebin.com/QbUFr6b9 - unfortunately not so much info, probably I need to recompile mencoder with more debug ...File pond.dv from http://libdv.sourceforge.net/
22:45 <AndrewR> *and
22:53 <gnarface> does it have to be mencoder? this might be easier with ffmpeg
23:02 <AndrewR> gnarface, probably ... but I was just playing, and run into this bug. Not even sure if my mencoder/mplayer build from ~16 feb still affected by same bug (need to check).
23:03 <AndrewR> as far as I understand mencoder itself not supported, so adding official bugs against it probably will not work (and ffmpeg git is moving target anyway)
23:08 <AndrewR> same crash with r37925
23:11 <iive> the bug seems related to ffmpeg, av_opt_free() is the one doing the double free.
23:11 <iive> AndrewR: is it a recent build, aka current mencoder and ffmpeg ?
23:12 <AndrewR> iive, it was curent few hours ago when I compiled this version...now I did git pull in ffmpeg subfolder and some updates come
23:13 <iive> yeh, git ffmpeg is doing a lot of libav merges recently
23:14 <AndrewR> iive, but those were mostly test updates and alike... I was hoping may be someone who knows mencoder can say from looking at it if something was desynced btween its use of avcodec and how it should be done now in ffmpeg ...but this hope is minimal.
23:14 <iive> michaelni: could you take a look of the backtrace? avcodec_close() calling av_opt_free() ,
23:15 <iive> AndrewR: most likely some new code frees the structure but doesn't clean the pointer.
23:16 <iive> it's a tricky because mencoder is the one crashing, so the bug might be inside ffmpeg libs OR the way mencoder calls them.
23:18 <AndrewR> iive, I assume ffmpeg itself auto-tested nowadays ....
23:30 <AndrewR> but I can't find exact same combination (x86_32/linux/gcc-4.9.x) on fate ....
23:31 <iive> it doesn't look like arch/gcc specific problem
23:34 <AndrewR> iive, thanks for looking into it...I think I will go to bed for now, leaving machine online. (most expensive quad-core music player...). If problem will not cure itself in few days ...i'll probably better to register at bugtracker anyway...
23:35 <iive> making a debug build could be of some help
23:36 <iive> also if you could run it under valgrind, it would give not only full backtrace , but where it have been freed first.
23:36 <iive> i'm sorry, i'm a little bit distracted atm