:: #mythtv

Daily chat history

Current users (83):

aloril, Anssi, anykey_, autopatch, brfransen, CaCtus491, cattelan, Chutt, clever, coling, Cougar, damaltor, danielk22, David_Mi1ler, dblain, dekarl1, dinamic|screen, dlblog, eharris_, ElmerFudd, foxbuntu, ghoti, GreyFoxx, highzeth, idl0r, J-e-f-f-A, jams, jarle, jcarlos, joe____, joki, jpabq, jstenback, jwh, jwhite, kc, knightr__, kurre2, kwmonroe, laga, mag0o, markcerv, MaverickTech, Mousey, mrand, mrec, MythBuild, MythLogBot, mzanetti, NightMonkey, peitolm, Peps, petefunk, pheld, poptix, purserj, rsiebert, seld, Sharky112065, skd5aner, Slasher`, SmallR2002, sphery, sraue, stichnot_, stuarta, sutula, tgm4883, ThisNewGuy, toeb, tomimo, tris, Unhelpful, vallor, Vernon_at_work, VManiac16, wagnerrp, whoDat, xavierh, XDS2010, yb0t, _charly_, _Techie_-_AFK_
Friday, June 22nd, 2012, 00:03 UTC
[00:03:04] andreax (andreax! has quit (Read error: Connection reset by peer)
[00:04:22] stichnot (stichnot!chatzilla@mythtv/developer/stichnot) has quit (Ping timeout: 265 seconds)
[00:27:50] mrand (mrand!~mrand@ubuntu/member/mrand) has joined #mythtv
[00:40:19] pheld (pheld! has quit (Quit: Leaving.)
[00:49:56] gregL (gregL! has quit (Quit: Leaving)
[00:52:22] dblain: danielk22: I tried to keep libmythservicecontracts as abstract as possible (pure interfaces) with little to no implementation (other than pure data classes). Not sure if anything has been added that would cause a dependancy on libmythbase.
[00:54:55] danielk22: dblain: Ah, I used the MythDate functions. Could easily just implement it without them. It's only two or three calls.
[00:56:07] danielk22: I wonder how it was able to include headers from libmythbase if that wasn't a dependency.. there may be other things in there..
[00:57:00] danielk22: Ah, both libmyth and libmythbase are in the include paths.
[01:04:57] dekarl1 (dekarl1! has joined #mythtv
[01:05:45] dekarl (dekarl! has quit (Ping timeout: 265 seconds)
[01:12:45] Mousey (Mousey! has quit (Remote host closed the connection)
[01:21:15] stichnot (stichnot!~chatzilla@mythtv/developer/stichnot) has joined #mythtv
[01:22:56] jya_ (jya_!~jyavenard@mythtv/developer/jya) has quit (Quit: jya_)
[01:26:16] amejia (amejia!~andres@ has joined #mythtv
[01:26:16] amejia (amejia!~andres@ has quit (Changing host)
[01:26:16] amejia (amejia!~andres@xbmc/staff/amejia) has joined #mythtv
[01:52:26] jya_ (jya_!~jyavenard@mythtv/developer/jya) has joined #mythtv
[02:07:19] amejia (amejia!~andres@xbmc/staff/amejia) has quit (Quit: Konversation terminated!)
[03:08:26] jya_ (jya_!~jyavenard@mythtv/developer/jya) has quit (Quit: jya_)
[03:16:34] NightMonkey (NightMonkey!~NightrMon@pdpc/supporter/professional/nightmonkey) has joined #mythtv
[04:58:12] Vernon_at_work (Vernon_at_work! has quit (Remote host closed the connection)
[04:58:21] Vernon_at_work (Vernon_at_work! has joined #mythtv
[05:07:12] natanojl (natanojl! has joined #mythtv
[05:17:21] NightMonkey (NightMonkey!~NightrMon@pdpc/supporter/professional/nightmonkey) has quit (Quit: Body blow! Body blow!)
[05:18:10] rsiebert (rsiebert! has joined #mythtv
[05:21:43] rsiebert_ (rsiebert_! has quit (Ping timeout: 276 seconds)
[05:53:00] cattelan is now known as cattelan_away
[06:17:19] kormoc (kormoc!~kormoc@mythtv/developer/kormoc) has quit (Quit: kormoc)
[06:27:55] natanojl (natanojl! has quit (Ping timeout: 260 seconds)
[07:13:33] sraue (sraue!~stephan@xbmc/staff/sraue) has joined #mythtv
[08:20:30] kc (kc!~Casper@unaffiliated/kc) has quit (Ping timeout: 260 seconds)
[08:20:44] jcarlos (jcarlos! has quit (Read error: Connection reset by peer)
[08:22:17] jcarlos (jcarlos! has joined #mythtv
[08:33:49] kc (kc!~Casper@unaffiliated/kc) has joined #mythtv
[08:46:02] jya_ (jya_!~jyavenard@mythtv/developer/jya) has joined #mythtv
[09:16:53] MythBuild: build #1239 of master-vista-mingw-32bit is complete: Success [3build successful] Build details are at . . . /builds/1239
[10:48:59] jya_ (jya_!~jyavenard@mythtv/developer/jya) has quit (Quit: jya_)
[11:55:57] amessina (amessina! has quit (Quit: Konversation terminated!)
[12:24:20] Goga777 (Goga777!~Goga777@ has joined #mythtv
[12:38:28] zombor (zombor!~zombor_@ has joined #mythtv
[12:38:29] zombor (zombor!~zombor_@ has quit (Changing host)
[12:38:30] zombor (zombor!~zombor_@kohana/developer/zombor) has joined #mythtv
[13:12:13] pheld (pheld! has joined #mythtv
[13:38:21] yoyolala (yoyolala!~yoyolala@unaffiliated/yoyolala) has joined #mythtv
[13:45:08] danielk22: dblain: It looks like recording.h in the contracts was including but not using mythmiscutil.h.
[13:47:29] danielk22: dblain: oh, but content.cpp is using mythmiscutil..
[13:49:32] Jordack (Jordack! has joined #mythtv
[13:51:09] danielk22: ok, so content.cpp and video.cpp are using the datacontracts/recording.h include of mythmiscutil.h, but those are linked with libmythbase; they should just include that directly (which is cleaner anyway).
[13:53:09] Goga777 (Goga777!~Goga777@ has quit (Quit: Leaving)
[13:53:42] gregL (gregL! has joined #mythtv
[14:01:09] gregL (gregL! has quit (Quit: Leaving)
[14:10:36] danielk22: dblain: I've removed the libmythbase dependencies and linking requirement.
[14:14:22] yoyolala (yoyolala!~yoyolala@unaffiliated/yoyolala) has quit (Quit: leaving)
[14:46:55] wagnerrp: is markspieth on irc?
[14:49:38] danielk22: dblain: I also noticed the LGPL applied to some code inside mythbackend. LGPL code can only be used in libraries as per section 0 and 2, but according to section 3 we can "upgrade" that code to GPLv2 or later to comply with the license. No worries, I'll take care of it.
[15:12:28] stichnot: danielk22: Are you sure about LGPL only being used in libraries? My understanding is that if you use LGPL code in libraries, it doesn't necessarily contaminate the rest of the code base with the GPL license. But aren't we already GPL?
[15:15:02] stichnot: Oh, I get it, the backend itself can't be given the LGPL license.
[15:22:15] stichnot (stichnot!~chatzilla@mythtv/developer/stichnot) has quit (Ping timeout: 260 seconds)
[15:41:29] brfransen (brfransen!~brfransen@ has quit (Read error: Operation timed out)
[15:44:51] brfransen (brfransen!~brfransen@ has joined #mythtv
[15:51:42] NightMonkey (NightMonkey!~NightrMon@pdpc/supporter/professional/nightmonkey) has joined #mythtv
[16:01:39] danielk22: Yep, the LGPL is for libraries. The intent of the license is to prevent you from copying the code into a program since because then you wouldn't be able to create a better version of the LGPL code to use with some proprietary program. But there is an exception allowing you to copy the code into a GPL program, but it requires you to update the license header to do so.
[16:02:19] danielk22: I'm presuming the justification for the exception is that you can modify the program itself if it is GPL.
[16:08:51] danielk22: stichnot: An LGPL library doesn't require the program to use GNU license, but it does require some things of the program. Such as requiring the program license to allow reverse engineering.
[16:13:55] aloril (aloril! has quit (Read error: Operation timed out)
[16:28:35] aloril (aloril! has joined #mythtv
[16:30:38] joki (joki! has quit (Ping timeout: 240 seconds)
[16:30:51] joki- (joki-! has joined #mythtv
[16:31:01] joki- is now known as joki
[16:34:58] NightMonkey (NightMonkey!~NightrMon@pdpc/supporter/professional/nightmonkey) has quit (Remote host closed the connection)
[16:40:16] stichnot (stichnot!~chatzilla@mythtv/developer/stichnot) has joined #mythtv
[16:43:52] stichnot: danielk22: does LGPL really require the program license to allow reverse engineering, or just reverse engineering of the LGPL portion?
[17:04:05] andreax (andreax! has joined #mythtv
[17:05:21] NightMonkey (NightMonkey!~NightrMon@pdpc/supporter/professional/nightmonkey) has joined #mythtv
[17:14:08] kormoc (kormoc!~kormoc@mythtv/developer/kormoc) has joined #mythtv
[17:17:32] danielk22: stichnot: It requires the program to allow reverse engineering if it was designed to use the LGPL library. But if say you wrote your program to use any libc and you then linked a LGPL libc, then it wouldn't apply. It's in sections 5&6. This part of the LGPL might be less defensible than others because it relies on the program being a derivative work of the library for it's teeth.
[17:18:54] gregL (gregL! has joined #mythtv
[17:21:25] gregL (gregL! has quit (Read error: Connection reset by peer)
[17:22:02] stichnot: danielk22: that's very interesting because at work, the generally paranoid legal team is comfortable with the use of dynamically linked LGPL libraries, but I can't imagine them being comfortable with allowing reverse engineering
[17:24:31] danielk22: I think you would really have to be trying to subvert the LGPL to really run into an issue.
[17:25:20] danielk22: Likely the stuff you guys do can't be considered a derivative works of the LGPL libraries.
[17:25:27] stichnot: Beirdo: I just noticed claiming a number of memory leaks in logging.h
[17:26:58] Mousey (Mousey! has joined #mythtv
[17:35:39] stuartm: stichnot: those look like false positives to me
[17:36:03] stichnot: ok. I didn't look into it at all.
[17:37:53] stuartm: all those pointers are free'd in the destructor
[18:09:54] stichnot: I started seeing something interesting in the frontend. In base.xml, <window name="MythBusyDialog">, <imagetype name="animation">, <delay>XXX</delay> — if I set XXX too low and try to enter the Watch Recordings screen, the animation goes on endlessly and the Watch Recordings screen never loads.
[18:13:25] Beirdo: stichnot: pretty sure those are false positives as well
[18:13:50] stichnot: ok, sorry for the false alarm :)
[18:18:13] Beirdo: no problem, shows you're paying attention :)
[18:22:10] cattelan_away is now known as cattelan
[18:44:01] yoyolala (yoyolala!~yoyolala@unaffiliated/yoyolala) has joined #mythtv
[18:49:45] yoyolala (yoyolala!~yoyolala@unaffiliated/yoyolala) has quit (Quit: leaving)
[18:54:39] danielk22: FYI Most of the strace activity in mythfrontend appears to be a side-effect of Pulse(), whenever the Qt UI thread wakes up it does a read to clear the wakeup fd, then checks the timezone (probably just a QDateTime::currentDateTime() call), then checks UDP sockets and then it does a poll() to go back to sleep.
[18:55:40] danielk22: Heh, 13 mythlogserver instances on my development machine...
[18:56:07] danielk22: Beirdo: Maybe mythlogserver should exit if it can't bind the socket?
[18:57:35] Beirdo: it does
[18:57:40] Beirdo: or it did
[18:57:46] Beirdo: did I bugger that up?
[18:57:55] Beirdo: Heh, I'll take a look :)
[18:58:06] Beirdo: and definitely it should
[18:58:27] danielk22: Anything you want me to do before killall ?
[18:58:43] Beirdo: nah, just nuke em
[18:58:48] danielk22: k
[18:59:19] Beirdo: I'm sure it will be dead easy to reproduce :)
[18:59:39] Beirdo: oh, so fun being on my last day here. :)
[18:59:54] danielk22: What is the new job?
[19:00:15] Beirdo: Software Development on the BlackBerry 10 (middleware) at RIM
[19:00:47] Beirdo: rather than system admin :)
[19:00:58] Beirdo: I'd so far rather be doing dev work
[19:01:32] Beirdo: so Monday, I start something completely different :)
[19:02:24] danielk22: :)
[19:04:15] Beirdo: decided not to even tae a week between
[19:15:11] Beirdo: how was the initialTimer segfaulting? It shouldn't have been able to, just curious
[19:15:41] Beirdo: the only way it should have gotten there is with m_initialTimer set
[19:16:20] Beirdo: musta missed something
[19:24:02] Beirdo: ah yes
[19:24:16] Beirdo: as you were saying... qApp->quit isn't always going to exit.
[19:24:19] Beirdo: frick
[19:24:29] Beirdo: Exception during logging socket setup: Connection refused
[19:24:36] Beirdo: but then doesn't quit
[19:25:03] brfransen (brfransen!~brfransen@ has quit (Read error: Operation timed out)
[19:25:24] Beirdo: 2012-06–22 12:23:44.666228 E Exception during socket setup: Address already in use
[19:25:31] Beirdo: actually that's the salient one
[19:26:07] Captain_Murdoch: are we still supposed to be able to use the TZ env var setting to run a FE on a client in a different time zone than the MBE in 0.25 or was that funcationality removed somehow?
[19:26:50] brfransen (brfransen!~brfransen@ has joined #mythtv
[19:34:45] stuartm: Beirdo: yeah, qApp->quit() requires the event loop, i.e. qApp->exec() has been called
[19:35:08] Beirdo: well, it is in the logserver
[19:35:27] stuartm: which was causing me no end of difficulties when I was wanting the frontend to abort if no translations were available
[19:35:30] Beirdo: so I'm sure there's something else I'm missing at the moment, like it never GETS to the exec()
[19:35:45] Beirdo: nothing like some debugging ;)
[19:46:22] danielk22: Beirdo: The m_initialTimer==NULL happened very shortly after mythfrontend started. I'd guess some kind of race.
[19:46:50] Beirdo: interesting
[19:47:09] Beirdo: can't hurt to null-check though
[19:50:05] Beirdo: so... what IS the right way to tell this hosebeast to quit?
[19:53:24] stuartm: forceful language, with a hint of violence
[19:53:31] Beirdo: hehe
[19:53:44] Beirdo: I know abort() will do it, but that hardly seems the right way
[20:10:57] Beirdo: hehe, exit(0) causes a mutex lock failure
[20:15:23] danielk22: ::exit(0) or qApp::exit(0) ?
[20:15:59] Beirdo: qApp->exit(0) is the same as qApp->quit()
[20:16:19] danielk22: Do those cause the mutex lock failure?
[20:16:27] Beirdo: no, that was bad code
[20:16:32] Beirdo: I'm fixing it
[20:17:42] Beirdo: haha, now it segfaults
[20:17:50] Beirdo: sigh. I'll get it right
[20:18:07] Beirdo: the problem is the interworking of the logging thread and the logserver thread
[20:18:19] Beirdo: the edge case needed a bit of work, it seems
[20:49:06] Captain_Murdoch (Captain_Murdoch!~Captain_M@mythtv/developer/CaptainMurdoch) has quit (Ping timeout: 272 seconds)
[21:19:51] dblain (dblain!~dblain@mythtv/developer/dblain) has quit (Ping timeout: 252 seconds)
[21:26:08] amejia (amejia!~andres@xbmc/staff/amejia) has joined #mythtv
[21:32:51] Jordack (Jordack! has quit (Quit: Two canibals sit down to eat a clown. One looks at the other and asks "Does this taste funny to you?")
[21:38:29] zombor (zombor!~zombor_@kohana/developer/zombor) has quit (Remote host closed the connection)
[21:45:25] danielk22: stichnot: I'm not getting the high CPU use and strace shows almost no activity on the backend, but a just started mythfrontend instance is using nearly 4 GB.. The virtual memory use may be key. I tend to over-provision memory and swap on my systems which may explain why CPU use remains low for me.
[21:46:33] dblain (dblain! has joined #mythtv
[21:46:33] dblain (dblain!~dblain@mythtv/developer/dblain) has joined #mythtv
[21:46:33] dblain (dblain! has quit (Changing host)
[21:49:02] stichnot: And it probably depends on how aggressive the malloc implementation is about dealing with fragmentation, which may in turn depend on the particulars of the system. My backend machine has 4GB and modest swap. (And it's running a 32-bit OS.)
[21:50:07] danielk22: 32 bit.. that may be important. My mythfrontend is already taking up more memory after a 60 seconds of runtime than is allowed in a 32 bit program.
[21:51:34] zombor (zombor!~zombor_@ has joined #mythtv
[21:51:34] zombor (zombor!~zombor_@ has quit (Changing host)
[21:51:34] zombor (zombor!~zombor_@kohana/developer/zombor) has joined #mythtv
[21:51:46] stichnot: You don't happen to have your no-delete reference counter debugging config defined, do you?
[21:51:59] jya_ (jya_!~jyavenard@mythtv/developer/jya) has joined #mythtv
[21:52:18] danielk22: Nope, but I am recompiling right now with leak detection..
[21:55:18] danielk22: I have a feeling the memory being allocated isn't even used. Otherwise I'd expect the resident portion to be larger.
[21:58:57] danielk22: Err the backend leaked 1 reference counted object.. a MythSocket.. hardly a large leak.
[22:05:31] zombor (zombor!~zombor_@kohana/developer/zombor) has quit (Remote host closed the connection)
[22:15:22] danielk22: — Umm, this does not inspire a lot of confidence in zeromq.
[22:17:44] danielk22: At least he has now realized he shouldn't use C++ exceptions.
[22:23:51] jya_ (jya_!~jyavenard@mythtv/developer/jya) has quit (Quit: jya_)
[22:44:29] aloril (aloril! has quit (Ping timeout: 244 seconds)
[22:57:20] stichnot: Would anyone mind trying something simple? In your theme's base.xml, find <window name="MythBusyDialog">, and in the <imagetype name="animation"> element change delay to something ridiculously small like 10. Then restart the frontend and try entering the Watch Recordings screen. In my case the busy dialog goes forever and the Watch Recordings screen never comes up.
[23:00:32] aloril (aloril! has joined #mythtv
[23:05:50] danielk22: stichnot: no lockup here
[23:08:32] stichnot: danielk22: is it a powerful frontend? (mine is an ION)
[23:08:53] danielk22: stichnot: i5 laptop
[23:10:11] stichnot: I guess pretty much anything is powerful compared to an ION...
[23:10:45] danielk22: Even with a build going on it doesn't lockup
[23:11:14] danielk22: stichnot: if it is locking up a backtrace would probably help find it..
[23:11:44] danielk22: I'm not so sure a SIGABRT would work right now, but running in gdb should work.
[23:12:10] stichnot: It doesn't actually lock up. The busy animation goes on forever like the screen is still being prepared. The CPU runs at 100%.
[23:12:34] danielk22: Still should tell us what it is waiting for..
[23:13:00] danielk22: I recall things like that happening when the background loading was first added.
[23:13:09] stichnot: Running in the debugger shows it's continuously trying to draw rounded rectangles, fwiw
[23:14:01] stichnot: this is backed up by tons of reference counting output when running "-v all"
[23:15:04] stichnot: Extra logging shows that MythScreenType::Close() and CloseBusyPopup() are not being called, even after WaitForLoadToComplete() completes.
[23:19:58] stichnot: There are probably particulars of the theme (Blue Abstract in this case) that contribute to whatever triggers the problem. For example, the dialog screen darkens everything with an alpha=110 black background, so the animation may somehow be fighting with the objects behind it
[23:20:50] stichnot: anyway, more important things to debug if I'm the only one with the problem...
[23:25:30] amejia_ (amejia_!~andres@xbmc/staff/amejia) has joined #mythtv
[23:26:24] amejia (amejia!~andres@xbmc/staff/amejia) has quit (Ping timeout: 245 seconds)
[23:38:41] andreax (andreax! has quit (Read error: Connection reset by peer)
[23:52:49] kormoc (kormoc!~kormoc@mythtv/developer/kormoc) has quit (Quit: kormoc)
[23:59:51] stichnot (stichnot!~chatzilla@mythtv/developer/stichnot) has quit (Ping timeout: 265 seconds)

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