MythLogBot@irc.freenode.net :: #mythtv

Daily chat history

Current users (70):

aloril, Anssi, benklop, betheynyx, blafu, blahdodo, brfransen, Captain_Murdoch, ChanServ, Chutt_, clever, CyberJacob, davic, dblain, dekarl, dmfrey, eharris, ElmerFudd, enyc, gary_buhrmaster, ghoti, gigem, gregl, GreyFoxx, Guest63620, Hydr0p0nX, Hydroponx, ikevin, jab416171, jams, jheizer_, joki, jpabq, jpharvey__, jst, jya, knowledgejunkie, KooeeeeennN, kukks, lapion, libsci, logan-, mad_enz, markspieth, membiblio, MitchCapper, MythBuild, MythLogBot, nephyrin`, peper03, poptix, pppingme, rhpot1991, rich0, sheedy-away, sphery, stuarta, stuartm, sunbird, taylorr, tgm4883, Tobbe5178, tris, vincent42, wagner__, Warped, XDS2010, xris, _charly_, _iwc
Sunday, November 20th, 2016, 00:00 UTC
[00:00:25] jya: Could be. Your fix doesn't make much sense to me. I'll look into later today hopefully.
[00:00:40] jya: If that's what Qt is doing, a security bug must be opened with Qt
[00:00:54] jya: But I'd be surprised if they did.
[00:01:00] clever: a lot of objects in QT store the data in the d property of the class, which is shared between several instances, so the copy constructor copies the d pointer, rather then the data inside d
[00:01:17] clever: and then if you try to modify one of them, it will dup the d class, and detach from the origina;
[00:01:24] jya: clever: that would be true if the object wasn't made const
[00:01:41] jya: As it's const, you can't even increase its reference counter.
[00:01:46] clever: ah
[00:02:31] peterbennett: clever: but in this case it seems if you delete one of the objects will it delete the original data, thereby messing up the others.
[00:02:56] jya: Now if they copy the pointer d and keep that. It's still UAF and is very bad.
[00:04:42] peterbennett: Ok I will download the QT source and see if I can find out what they are doing.
[00:05:01] clever: its also on github, so you can just read it
[00:06:29] peterbennett: oh ok
[00:06:49] clever: https://github.com/qt/qtbase
[00:07:39] peterbennett: gotta go, bye
[00:30:36] peterbennett (peterbennett!~Peter@mythtv/developer/peterbennett) has quit (Quit: Leaving.)
[01:52:40] mad_enz (mad_enz!~mad_enz@2607:f090:488d:2ab:c56c:b071:5a13:eb9a) has quit (Remote host closed the connection)
[03:11:12] peper03 (peper03!~peper03@mythtv/developer/peper03) has quit (Ping timeout: 244 seconds)
[03:27:50] SteveGoodey (SteveGoodey!~steve@host86-154-46-109.range86-154.btcentralplus.com) has joined #mythtv
[04:10:33] peterbennett (peterbennett!~Peter@c-75-67-202-199.hsd1.ma.comcast.net) has joined #mythtv
[04:10:42] peterbennett (peterbennett!~Peter@c-75-67-202-199.hsd1.ma.comcast.net) has quit (Changing host)
[04:10:43] peterbennett (peterbennett!~Peter@mythtv/developer/peterbennett) has joined #mythtv
[04:13:36] peterbennett: jya: I looked at the QT source and what you said is correct, and QT looks like it is doing the correct thing, so I am going to revisit that bug and work out a better fix.
[04:14:20] peterbennett: jya: My original theory for the seg fault does not work
[04:30:45] dekarl1 (dekarl1!~dekarl@mythtv/developer/dekarl) has joined #mythtv
[04:32:07] dekarl (dekarl!~dekarl@mythtv/developer/dekarl) has quit (Ping timeout: 268 seconds)
[04:36:54] peterbennett (peterbennett!~Peter@mythtv/developer/peterbennett) has quit (Quit: Leaving.)
[04:50:47] SteveGoodey (SteveGoodey!~steve@host86-154-46-109.range86-154.btcentralplus.com) has quit (Quit: Konversation terminated!)
[05:05:13] peper03 (peper03!~peper03@mythtv/developer/peper03) has joined #mythtv
[05:12:40] amessina (amessina!~amessina@unaffiliated/amessina) has quit (Quit: Konversation terminated!)
[07:13:04] Roklobster (Roklobster!~Dirkka@ppp118-209-57-1.lns20.mel4.internode.on.net) has joined #mythtv
[07:34:36] dekarl1 is now known as dekarl
[07:41:41] dekarl: a helpful helper helped me discover the greatness of mythweb.com
[07:53:09] dekarl (dekarl!~dekarl@mythtv/developer/dekarl) has quit (Ping timeout: 268 seconds)
[08:02:24] ** enyc thinks... what.... **
[08:53:44] SteveGoodey (SteveGoodey!~steve@host86-151-54-68.range86-151.btcentralplus.com) has joined #mythtv
[09:13:42] dekarl (dekarl!~dekarl@mythtv/developer/dekarl) has joined #mythtv
[11:18:36] jya: how can I call mythtranscode without having a matching recording ? I mean just a filename to convert
[11:20:36] Roklobster (Roklobster!~Dirkka@ppp118-209-57-1.lns20.mel4.internode.on.net) has quit (Remote host closed the connection)
[11:24:40] peper03 (peper03!~peper03@mythtv/developer/peper03) has quit (Ping timeout: 268 seconds)
[11:29:27] peper03 (peper03!~peper03@mythtv/developer/peper03) has joined #mythtv
[11:34:42] jya: --video it was
[11:59:25] jya: is deleting the original recording before transcoding something expected??
[12:05:20] jya: stuartm: ping
[12:06:49] stuartm: no, definitely not
[12:36:04] dekarl: is this about the deletion in mythtranscode issue? #12845
[12:36:04] ** MythLogBot http://code.mythtv.org/trac/ticket/12845 **
[13:15:46] stuartm: fwiw, although I've not tested that patch it looks harmless
[13:17:11] stuartm: I'm not really sure how it fixes the issue of the original recording being deleted before the transcode occurs though. That seems more like we're hitting the cleanup function even in the case of an error occurring during the transcode
[13:20:32] joki (joki!~joki@p54861ADD.dip0.t-ipconnect.de) has joined #mythtv
[13:48:58] rich0_ (rich0_!~quassel@gentoo/developer/rich0) has quit (Read error: Connection reset by peer)
[13:52:43] rich0_ (rich0_!~quassel@gentoo/developer/rich0) has joined #mythtv
[14:11:59] warpme (warpme!~warpme@89-67-153-39.dynamic.chello.pl) has joined #mythtv
[14:13:20] warpme: jya: great work with ffmpef resync. All works OK for me except quite frequeng segfaults in LiveTV. Here are 2 examplary bt: http://pastebin.com/KjB2Rtxv http://pastebin.com/f1FcL0kZ
[14:14:21] warpme: jya: also I see noticable slower channel change (1..2sec->5..7sec). Is this intentional or rather side effect?
[14:14:39] jya: warpme: please open a bug, attach the backtrace, and if possible use a non optimised build, your backtrace are colose to unusable
[14:14:53] warpme: sure. np.
[14:15:03] jya: can't see how channel change could be affected at all.
[14:16:16] warpme: jya: ok – maybe this is kind coincidence with chnges on CA in my sat provider. no problem.
[14:17:59] jya: looking a fifo mode in transcode.cpp, how did this ever work... when frames are dropped, logged as dropped, they are still written regardless
[14:18:27] stuarta: afternoon all
[14:23:26] jya: good afternoon
[14:24:21] warpme: jya: FYI: #12929
[14:24:21] ** MythLogBot http://code.mythtv.org/trac/ticket/12929 **
[14:25:00] jya: warpme: you must compile with --compile-type=debug
[14:25:28] warpme: jya: yes it is compiled with debug.
[14:25:52] jya: must have some extran -Ox in your configure command
[14:26:04] jya: no way everything would be optimised out otherwise
[14:26:25] jya: how often do you get those segfault?
[14:26:57] jya: all the time or sometimes, what happen if you change channels between once with AC3 audio
[14:27:08] warpme: jya: quite frequently. Avg. once per 10–20 channel changes
[14:27:17] jya: seems that it's the spdif encoder (used for passthrough)
[14:27:45] jya: my guess is that it's only because you're switching between channels that do not always use 5.1 audio
[14:28:04] jya: what if you disable AC3 and DTS passthrough in the audio settings?
[14:29:07] jya: now that I think about it, everytime we did a resync, there were issues with the digital passthrough
[14:30:09] warpme: jya: yes. currently difficult to say (can’t recall exact channels) but indded I recall that it segfaults when changing channel from 2.0 to 5.1 or vice versa.
[14:30:22] warpme: jya: let me test
[14:37:29] warpme: jya: indeed turning-off DD and DTS (effectivelly sound is now 2.0) causes no segfaults on well segfaulting 2 channels. I’m checking rest of my 67HD channels.
[14:37:47] jya: sorry don't get that..
[14:37:59] jya: no segfauls on well segfaulting?
[14:39:19] warpme: jya: sorry. I have combination of 2 chanells where ffmpeg3.2 always segfaults. Now (with sound 2.0) no segfaults on those 2 channels. now checking rest of my channels
[14:45:09] ** stuarta wanders off again **
[14:45:20] rich0_ is now known as rich0
[14:53:44] mad_enz (mad_enz!~mad_enz@2607:f090:488d:2ab:ada3:1dd:e762:d024) has joined #mythtv
[14:56:53] warpme: jya: just finished checking on 112 channels. With sound forced to 2.0 no segfaults. Enabling just DD (and changing speakers type from 2.0 to 5.1) – segfaults starts again.
[14:57:22] jya: ok, so it's the SPDIF encoder, I'll have a look tomorrow
[15:01:22] warpme: jya: indeed. FYI: I done test with sound config where 2.0 sound is upmixed to 5.1. Starting tv channel with 2.0 sound gives immediatelly segfault.
[15:02:19] tgm4883 (tgm4883!uid23806@ubuntu/member/tgm4883) has quit (Quit: Connection closed for inactivity)
[15:27:12] amessina (amessina!~amessina@unaffiliated/amessina) has joined #mythtv
[16:18:46] Hydr0p0nX (Hydr0p0nX!~Hydr0p0nX@66-191-154-200.dhcp.mtgm.al.charter.com) has quit (Ping timeout: 256 seconds)
[16:46:00] warpme (warpme!~warpme@89-67-153-39.dynamic.chello.pl) has quit (Quit: warpme)
[17:20:04] tgm4883 (tgm4883!uid23806@ubuntu/member/tgm4883) has joined #mythtv
[18:26:04] Roklobster (Roklobster!~Dirkka@ppp118-209-57-1.lns20.mel4.internode.on.net) has joined #mythtv
[18:35:01] Hydr0p0nX (Hydr0p0nX!~Hydr0p0nX@66-191-154-200.dhcp.mtgm.al.charter.com) has joined #mythtv
[20:04:16] amessina (amessina!~amessina@unaffiliated/amessina) has quit (Read error: Connection reset by peer)
[20:15:47] Roklobster (Roklobster!~Dirkka@ppp118-209-57-1.lns20.mel4.internode.on.net) has quit (Remote host closed the connection)
[21:51:11] SteveGoodey (SteveGoodey!~steve@host86-151-54-68.range86-151.btcentralplus.com) has quit (Quit: Konversation terminated!)
[23:27:36] dmfrey: I am trying to get an mpeg2 decoder to plug into ExoPlayer, which is gonna be the video player i'm using in the android app. However, I am unsure where to start looking for this. Can I use ffmpeg for this? I have a toolchain in place that builds ffmpeg for android and it is currently handling the ac3 audio playback. Any thoughts from here?
[23:51:50] dmfrey: what i'm finding is libmpeg2, but I am not really seeing any code updates since 2008. is there something else that handles mpeg2?

IRC Logs collected by BeirdoBot.
Please use the above link to report any bugs.