| Thursday, January 31st, 2013, 00:27 UTC | ||
| [00:27:41] | devinheitmueller (devinheitmueller!~devinheit@pool-108-54-72-165.nycmny.fios.verizon.net) has quit (Quit: devinheitmueller) | |
| [00:34:59] | jheizer__ (jheizer__!~jon@c-98-226-220-178.hsd1.il.comcast.net) has joined #mythtv | |
| [00:39:35] | NightMonkey (NightMonkey!~NightrMon@pdpc/supporter/professional/nightmonkey) has quit (Ping timeout: 260 seconds) | |
| [00:43:22] | devinheitmueller (devinheitmueller!~devinheit@pool-108-54-72-165.nycmny.fios.verizon.net) has joined #mythtv | |
| [00:49:54] | jheizer__ (jheizer__!~jon@c-98-226-220-178.hsd1.il.comcast.net) has quit (Ping timeout: 264 seconds) | |
| [00:51:25] | NightMonkey (NightMonkey!~NightrMon@pdpc/supporter/professional/nightmonkey) has joined #mythtv | |
| [01:01:06] | jheizer__ (jheizer__!~jon@c-98-226-220-178.hsd1.il.comcast.net) has joined #mythtv | |
| [01:01:18] | bas-t (bas-t!~tycho@52484E89.cm-4-1b.dynamic.ziggo.nl) has quit (Quit: Ex-Chat) | |
| [01:16:20] | dmfrey (dmfrey!~dmfrey@64-121-93-243.c3-0.smt-ubr1.atw-smt.pa.cable.rcn.com) has joined #mythtv | |
| [01:38:31] | joki- (joki-!~joki@p54864A08.dip.t-dialin.net) has joined #mythtv | |
| [01:38:36] | joki (joki!~joki@p54864B84.dip.t-dialin.net) has quit (Ping timeout: 264 seconds) | |
| [01:38:45] | joki- is now known as joki | |
| [01:45:59] | neufeld_AFK is now known as neufeld | |
| [01:57:04] | neufeld: | I wonder if anybody can help me out with my patch on #11108. It's a real bug that isn't fixed in head, the patch I've proposed modifies three lines and adds two. I'm unclear on one point, why prerollseconds is initialized to zero outside of the loop at mythbackend/scheduler.cpp:1749 ? It's later updated from the gCoreContext in the loop itself, but is there a reason that lookup isn't performed before the loop |
| [01:57:04] | ** MythLogBot http://code.mythtv.org/trac/ticket/11108 ** | |
| [01:57:04] | neufeld: | starts up? The extra two lines are an underflow guard, just to make sure that I don't create any conditions that wouldn't be experienced in the former flow (the unpatched code can't return a negative number for secs_to_next, so I bound it below by 1, as that number is a valid result in the unpatched code). Finally, as I don't have write permission to the repository, can somebody, if they like the patch, please |
| [01:57:04] | neufeld: | apply it? I'm running with it locally, and it does fix the bug for me. Thank you. |
| [02:02:20] | gigem: | neufeld: it's on my list of things to look at. Ping me during the day tomorrow (US time) and I'll try to get to it. |
| [02:02:39] | neufeld: | gigem: OK, thanks. |
| [02:08:56] | lentferj_ (lentferj_!~lentferj@p579B74F2.dip.t-dialin.net) has quit (Ping timeout: 252 seconds) | |
| [02:09:20] | mrand (mrand!~mrand@ubuntu/member/mrand) has quit (Read error: Connection reset by peer) | |
| [02:10:00] | lentferj (lentferj!~lentferj@87.155.120.34) has joined #mythtv | |
| [02:12:40] | jheizer__ (jheizer__!~jon@c-98-226-220-178.hsd1.il.comcast.net) has quit (Ping timeout: 246 seconds) | |
| [02:19:42] | neufeld is now known as neufeld_AFK | |
| [02:37:33] | danielk22: | devinheitmueller: The IVTV VBI parsing code is used exclusively for EIA-608 CC. We can't really dump it since it is used to display the captions from existing recordings unless there is some sort of transition plan, that plan could be run mythccextractor on all recordings to generate SRT files, but there needs to be some plan. |
| [02:39:17] | danielk22: | devinheitmueller: I e-mailed whoever was working on ivtv vbi insertion a long time ago (Klaus) and he felt it wouldn't be possible to insert the CC in a standards compliant way in the driver because there wasn't enough space. |
| [02:41:21] | danielk22: | We also decode XDS, but that is in the CC stream. I don't think it is a great loss if we lose that for existing recordings, but it would be very good to have in whatever you are working on. |
| [02:48:43] | danielk22: | gigem: The only thing I like about the Delete Recordings screen is being able to see the file size easily. But with how little I use it these days I can't really argue for it. It's just easier to add hard drives these days than it is to actively delete the stuff. |
| [02:50:48] | Sharky112065 is now known as Sharky-AFK | |
| [02:51:08] | jheizer__ (jheizer__!~jon@c-98-226-220-178.hsd1.il.comcast.net) has joined #mythtv | |
| [02:55:08] | devinheitmueller: | danielk22: Sure, understood. Figured if we've got EIA-608 automatically embedded in the user data section of the stream, would avoid the need for IVTV parsing, but I obviously wasn't thinking of existing recordings. |
| [02:55:34] | devinheitmueller: | I'm not 100% sure it's possible for IVTV, but I'm in the process of doing it for cx18. |
| [02:55:45] | devinheitmueller: | (and the data sheets for the cx23416 suggest it can be done) |
| [02:57:44] | danielk22: | devinheitmueller: We should be able to decode that automatically if you are doing EIA-608 embedding in EIA-708 packets. |
| [03:01:27] | skd5aner: | gigem: oh, and I said "broken glass" earlier – but meant to say "Tinted Glass" – I actually love this theme, it's great! I might go as far as nominating it for consideration for default theme (I know stuartm was looking for potential new default theme candidates) |
| [03:01:37] | skd5aner: | gigem: also, it too shows total disk usage in the watch recordings screen |
| [03:01:39] | devinheitmueller: | Well in this context I'm letting the hardware do the embedding, but yeah it's the same mechanism used for EIA-708 in MPEG2. |
| [03:05:01] | wagnerrp: | skd5aner: isn't everyone's default theme 'mythbuntu'? |
| [03:05:35] | wagnerrp: | hmm... i guess only for 88.3% of users :P |
| [03:05:51] | stichnot (stichnot!~stichnot@mythtv/developer/stichnot) has quit (Ping timeout: 245 seconds) | |
| [03:06:06] | skd5aner: | danielk22: any of the themes could be updated to include file size in the watch recordigns screen – several already do :) |
| [03:06:36] | skd5aner: | wagnerrp: heh... yea, I haven't looked at the smolt data in a while, but sounds about right |
| [03:10:28] | jheizer__ (jheizer__!~jon@c-98-226-220-178.hsd1.il.comcast.net) has quit (Ping timeout: 248 seconds) | |
| [03:27:13] | jheizer__ (jheizer__!~jon@c-98-226-220-178.hsd1.il.comcast.net) has joined #mythtv | |
| [03:32:22] | NightMonkey (NightMonkey!~NightrMon@pdpc/supporter/professional/nightmonkey) has quit (Quit: Body blow! Body blow!) | |
| [03:52:55] | stichnot (stichnot!~stichnot@adsl-69-110-235-166.dsl.pltn13.pacbell.net) has joined #mythtv | |
| [03:52:55] | stichnot (stichnot!~stichnot@mythtv/developer/stichnot) has joined #mythtv | |
| [03:52:55] | stichnot (stichnot!~stichnot@adsl-69-110-235-166.dsl.pltn13.pacbell.net) has quit (Changing host) | |
| [04:05:08] | jheizer__ (jheizer__!~jon@c-98-226-220-178.hsd1.il.comcast.net) has quit (Ping timeout: 248 seconds) | |
| [04:13:32] | devinheitmueller (devinheitmueller!~devinheit@pool-108-54-72-165.nycmny.fios.verizon.net) has quit (Quit: devinheitmueller) | |
| [04:40:27] | Oleg_ (Oleg_!~mine@pool-71-183-180-16.nycmny.east.verizon.net) has joined #mythtv | |
| [04:40:37] | Oleg_ (Oleg_!~mine@pool-71-183-180-16.nycmny.east.verizon.net) has left #mythtv () | |
| [04:42:09] | dmfrey (dmfrey!~dmfrey@64-121-93-243.c3-0.smt-ubr1.atw-smt.pa.cable.rcn.com) has quit (Quit: Ex-Chat) | |
| [04:58:09] | fetzerch (fetzerch!~quassel@unaffiliated/fetzerch) has quit (Ping timeout: 248 seconds) | |
| [04:59:55] | fetzerch (fetzerch!~quassel@unaffiliated/fetzerch) has joined #mythtv | |
| [05:32:03] | Goga777 (Goga777!~Goga777@95-30-20-60.broadband.corbina.ru) has joined #mythtv | |
| [05:33:08] | kodaws (kodaws!~koda@host42-8-static.81-94-b.business.telecomitalia.it) has joined #mythtv | |
| [05:57:54] | stichnot: | danielk22: I found a serious problem that affects playback of live TV but not completed recordings (and affects in-progress recordings to a small extent). When playback starts, avformat_find_stream_info() scans the start of the file for various info. Whenever it reads a packet, av_get_packet() calls ffio_limit() which I believe is supposed to ensure you don't try to read past eof. The... |
| [05:57:56] | stichnot: | ...first time ffio_limit() is called, it calls avio_size() to determine the limit, which ultimately calls RemoteFile::GetFileSize(). This unfortunately returns the size of the file when the RemoteFile object was created, which may be very small since little time has elapsed between starting the live TV recording and starting playback. That in turn makes av_get_packet() truncate packets and... |
| [05:57:58] | stichnot: | ...it all goes downhill from there. |
| [06:00:52] | stichnot: | I don't know how to make RemoteFile::GetFileSize() return something reasonable for live TV or in-progress recordings. Since avio_size() caches the result, it has to be "right" the first time. |
| [06:11:53] | Goga777 (Goga777!~Goga777@95-30-20-60.broadband.corbina.ru) has quit (Remote host closed the connection) | |
| [06:13:01] | stichnot: | If anyone is having problems with live TV startup or channel changes, you can try commenting out the call to ffio_limit() in av_get_packet() in external/FFmpeg/libavformat/utils.c . |
| [06:14:42] | stichnot: | This gets my PVR-150 working pretty well with live TV. |
| [06:21:00] | stichnot (stichnot!~stichnot@mythtv/developer/stichnot) has quit (Ping timeout: 252 seconds) | |
| [06:28:10] | wahrhaft (wahrhaft!~quassel@cpe-24-210-71-0.columbus.res.rr.com) has quit (Read error: Operation timed out) | |
| [06:30:34] | wahrhaft (wahrhaft!~quassel@cpe-24-210-71-0.columbus.res.rr.com) has joined #mythtv | |
| [06:34:07] | stichnot (stichnot!~stichnot@adsl-68-127-209-31.dsl.pltn13.pacbell.net) has joined #mythtv | |
| [06:34:07] | stichnot (stichnot!~stichnot@adsl-68-127-209-31.dsl.pltn13.pacbell.net) has quit (Changing host) | |
| [06:34:07] | stichnot (stichnot!~stichnot@mythtv/developer/stichnot) has joined #mythtv | |
| [06:36:12] | stichnot: | This use of ffio_limit() was introduced in the ffmpeg used in 0.26, so it doesn't necessarily explain the regressions people have reported in 0.25, though there could be something similar happening there under a different name. |
| [06:40:00] | Sharky-AFK is now known as Sharky112065 | |
| [06:40:04] | rsiebert_ (rsiebert_!~quassel@g231186186.adsl.alicedsl.de) has joined #mythtv | |
| [06:43:16] | rsiebert (rsiebert!~quassel@g224250034.adsl.alicedsl.de) has quit (Ping timeout: 248 seconds) | |
| [06:45:51] | stichnot (stichnot!~stichnot@mythtv/developer/stichnot) has quit (Ping timeout: 245 seconds) | |
| [06:56:46] | Tobbe5178 (Tobbe5178!~asdf@h186n5-sv-a13.ias.bredband.telia.com) has joined #mythtv | |
| [06:59:07] | dekarl (dekarl!~dekarl@p4FE854A2.dip.t-dialin.net) has quit (Ping timeout: 256 seconds) | |
| [06:59:30] | stichnot (stichnot!~stichnot@adsl-68-127-102-161.dsl.pltn13.pacbell.net) has joined #mythtv | |
| [06:59:30] | stichnot (stichnot!~stichnot@adsl-68-127-102-161.dsl.pltn13.pacbell.net) has quit (Changing host) | |
| [06:59:30] | stichnot (stichnot!~stichnot@mythtv/developer/stichnot) has joined #mythtv | |
| [07:02:27] | dekarl (dekarl!~dekarl@p4FE84D9E.dip.t-dialin.net) has joined #mythtv | |
| [07:09:21] | FabriceMG (FabriceMG!~Thunderbi@217.112.59.207) has joined #mythtv | |
| [07:11:29] | dblain_ (dblain_!~dblain@mythtv/developer/dblain) has quit (Ping timeout: 248 seconds) | |
| [07:24:10] | Sharky112065 is now known as Sharky-Sleep | |
| [07:32:33] | kodabws (kodabws!~koda@host42-8-static.81-94-b.business.telecomitalia.it) has joined #mythtv | |
| [07:35:58] | kodaws (kodaws!~koda@host42-8-static.81-94-b.business.telecomitalia.it) has quit (Ping timeout: 248 seconds) | |
| [07:36:24] | kodabbws (kodabbws!~koda@host42-8-static.81-94-b.business.telecomitalia.it) has joined #mythtv | |
| [07:39:47] | kodabws (kodabws!~koda@host42-8-static.81-94-b.business.telecomitalia.it) has quit (Ping timeout: 248 seconds) | |
| [08:31:29] | skd5aner (skd5aner!~skd5aner@50-90-30-141.res.bhn.net) has quit (Read error: Connection reset by peer) | |
| [08:31:47] | skd5aner (skd5aner!~skd5aner@50-90-30-141.res.bhn.net) has joined #mythtv | |
| [08:50:52] | bas-t (bas-t!~tycho@52484E89.cm-4-1b.dynamic.ziggo.nl) has joined #mythtv | |
| [09:19:15] | Merlin83b (Merlin83b!~Daniel@2a00:1ee0:3:1337:f0c5:7ea2:c4fe:e7f8) has joined #mythtv | |
| [09:46:59] | bas-t (bas-t!~tycho@52484E89.cm-4-1b.dynamic.ziggo.nl) has quit (Quit: Ex-Chat) | |
| [11:25:01] | MythBuild (MythBuild!~MythBuild@alcor.mythtv.org) has quit (Remote host closed the connection) | |
| [11:25:11] | MythBuild (MythBuild!~MythBuild@alcor.mythtv.org) has joined #mythtv | |
| [11:25:38] | stuarta: | lets see if that fixes the doxygen build |
| [11:34:01] | MythBuild: | build #1629 of doxygen-master is complete: Failure [4failed rsync to server] Build details are at http://code.mythtv.org/buildbot/builders/doxy . . . /builds/1629 blamelist: Stuart Morgan <smorgan@mythtv.org >, Alex Vasilyev <sandybigboy@gmail.com >, Jonatan Lindblad <jlindblad@mythtv.org >, Paul Gardiner <mythtv@glidos.net >, John Poet <jpoet@mythtv.org >, Richard |
| [11:34:01] | MythBuild: | <peper03@yahoo.com >, Raymond Wagner <rwagner@mythtv.org >, Karl Dietz <dekarl@mythtv.org >, Morris Cavestro <morriscavestro@yahoo.it >, David Engel <dengel@mythtv.org >, Greg Cockburn <greg@performancemagic.com >, Nicolas Riendeau <nriendeau@mythtv.org >, John Patrick Poet <jpoet@mythtv.org >, Wayne McDougall <waynemcdougall@gmail.com >, Jim Stichnoth <jstichnoth@mythtv.org >, |
| [11:34:01] | MythBuild: | Daniel Thor Kristjansson <danielk@cuymedia.net > |
| [11:41:40] | MythBuild (MythBuild!~MythBuild@alcor.mythtv.org) has quit (Remote host closed the connection) | |
| [11:41:53] | MythBuild (MythBuild!~MythBuild@alcor.mythtv.org) has joined #mythtv | |
| [11:48:00] | MythBuild: | build #1630 of doxygen-master is complete: Success [3build successful] Build details are at http://code.mythtv.org/buildbot/builders/doxy . . . /builds/1630 |
| [11:50:09] | stuarta: | much better :) |
| [11:58:08] | bas-t (bas-t!~tycho@52484E89.cm-4-1b.dynamic.ziggo.nl) has joined #mythtv | |
| [12:59:45] | danielk22: | stichnot: I believe the issues people had with 0.25 were all fixed early in the 0.26 cycle, which is why the issues were surprising to me. |
| [13:01:26] | danielk22: | stichnot: Is ffio_limit() called when ffmpeg is streaming. i.e. if you feed it from stdio it can't seek and can't know the file size.. if so can we return that ffio_limit() sentinel value? |
| [13:05:27] | danielk22: | ffmpeg seems to have introduced many features in the last few years that assume fixed files rather than broadcast streams. The jump to the 'end' of the file to read the pts/dts values for instance. |
| [13:05:29] | danielk22: | This tells you absolutely nothing about real streams both because the end of the current file may not be the end of the file in 10 milliseconds and because pts/dts discontinuities are allowed and even somewhat common in real streams. |
| [13:06:13] | danielk22: | All it does is make playback startup slower and make ffmpeg make bad assumptions about the bitrate and program length. |
| [13:11:34] | stuartm: | if only we had better relations with ffmpeg/libav and their devs hung out here and on the mailing lists, I'm not sure they really know the problems faced by the applications that use their library and 'real world' scenarios |
| [13:12:37] | mrand (mrand!~mrand@ubuntu/member/mrand) has joined #mythtv | |
| [13:34:39] | knightr (knightr!~knightr@mythtv/developer/knightr) has quit (Read error: Connection reset by peer) | |
| [13:39:58] | devinheitmueller (devinheitmueller!~devinheit@pool-108-54-72-165.nycmny.fios.verizon.net) has joined #mythtv | |
| [13:40:51] | kodabbws (kodabbws!~koda@host42-8-static.81-94-b.business.telecomitalia.it) has quit (Ping timeout: 248 seconds) | |
| [13:43:06] | stichnot: | danielk22: I'll look into the streaming issue right now. estimate_timings() suggests two possible checks on streamed input – the AVFMT_NOFILE flag, and ic->pb->seekable which we actually set to false for live TV in AvFormatDecoder::InitByteContext(). |
| [13:54:06] | stuartm: | got a core file here with no symbol info yet I'm using a profile build? |
| [13:54:45] | kenni (kenni!~kenni@mythtv/developer/kenni) has quit (Ping timeout: 240 seconds) | |
| [13:54:45] | kenni (kenni!~kenni@port649.ds1-ly.adsl.cybercity.dk) has joined #mythtv | |
| [13:54:45] | kenni (kenni!~kenni@port649.ds1-ly.adsl.cybercity.dk) has quit (Changing host) | |
| [13:54:46] | kenni (kenni!~kenni@mythtv/developer/kenni) has joined #mythtv | |
| [14:04:35] | stuartm: | nevermind, user error |
| [14:07:36] | Seeker`: | what would the behaviour be if you ran in to this ffmpeg / streaming bug? |
| [14:07:37] | IReboot (IReboot!~doug@CPE10bf48e67915-CM00252eac6f40.cpe.net.cable.rogers.com) has joined #mythtv | |
| [14:10:07] | stichnot: | danielk22: it doesn't look like AVFMT_NOFILE is useful here. I also looked at where avformat_find_stream_info() is called from. The most interesting place is from ffplay.c, in which there is a usage message "An input file must be specified\n". So maybe streaming is really only supported for conversion, not playback. |
| [14:11:02] | bas-t: | stichnot: after switching to debian and compiling fixes 0.26 yesterday, i can not reproduce the live tv issues i had anymore. so i'm fine now |
| [14:11:35] | stichnot: | bas-t: that's too bad :) I could use more testers |
| [14:12:48] | bas-t: | stichnot: too bad indeed, but switching back to ubuntu.... what needs to be tested any way? |
| [14:14:32] | stichnot: | Seeker`: for me, starting live TV would just appear to hang for about 4 minutes, at which point it would try to stop playback but repeatedly fail due to "fail to get decoder lock" or something like that, and sometimes it would end with a segfault. |
| [14:15:22] | Seeker`: | might be what is causing the crys of "THE TV STOPPED WORKING" from my wife |
| [14:16:39] | stichnot: | Seeker`: another thing is to comment out "silence_ffmpeg_logging = true;" in AvFormatDecoder::FindStreamInfo() and then run with -v libav --loglevel debug, and you see all kinds of log messages like "Truncating packet of size xxx to 1". |
| [14:17:16] | Seeker`: | I'll give that a go this evening |
| [14:18:20] | stichnot: | bas-t: Seeker`: for now, I'd like to know how commenting out the call to ffio_limit() in av_get_packet() in external/FFmpeg/libavformat/utils.c affects live TV. This is applicable to Master and 0.26. |
| [14:18:32] | stichnot: | skd5aner: ^^^ |
| [14:19:56] | bas-t: | stichnot: where is the diff, and how exactly do you want me to test it? |
| [14:21:45] | stichnot: | let me make a quick patch |
| [14:25:33] | Seeker`: | hmm, not going to be able to test it until after the mythbuntu packages update the backend I think (still running that from their 0.27 repo, only my frontend is on directly compiled master) |
| [14:25:58] | stichnot: | http://pastebin.com/vqyqcE3X |
| [14:26:30] | stichnot: | This has the fix and also turns back on the libav logging during the initial stream scan |
| [14:33:05] | stichnot: | for testing, try entering live TV, switch channels, switch inputs, especially for the PVR-150 input, and see if it behaves properly |
| [14:36:10] | skd5aner: | stichnot: exciting stuff... for those playing along at home: #11229 and #11211 |
| [14:36:10] | ** MythLogBot http://code.mythtv.org/trac/ticket/11229 ** | |
| [14:36:10] | ** MythLogBot http://code.mythtv.org/trac/ticket/11211 ** | |
| [14:37:58] | skd5aner: | stichnot: I'll test now... this patch against 0.26-fixes? |
| [14:38:13] | skd5aner: | I guess you say it's applicable to both huh |
| [14:38:15] | skd5aner: | ? |
| [14:38:41] | stichnot: | skd5aner: yes, it should apply cleanly to Master and fixes/0.26 |
| [14:39:30] | Seeker`: | ah, mythbuntu packages were build after the last ABI bump commit, so i've compiled a new frontend. Will see if i can get some TV watched today to test |
| [14:39:32] | skd5aner: | cool, i'll report back in a few |
| [14:40:00] | skd5aner: | needless to say, live tv is always trying to tune to my analog tuner first now, so I'll know pretty much right away if it works or not |
| [14:40:17] | bas-t: | stichnot: i use satelco easywatch dvb-c pci tuners, the pvr-250 days are long gone. I'll test it anyway and make logs with and without the patch. Tomorrow i'll report back. |
| [14:40:42] | skd5aner: | stichnot: also, last night was probably the first time I attempted to watch an in progress recording of one of my PVR-500 recordings... it bombed out after several minutes every time |
| [14:40:51] | skd5aner: | played back fine once the recording was finished |
| [14:41:10] | skd5aner: | it also took ~20–30 seconds to start playback |
| [14:41:18] | skd5aner: | (during the in-progress recording) |
| [14:41:34] | stichnot: | skd5aner: I'm not up to speed on the analog live TV issues reported against 0.25, and this patch/issue doesn't apply there, so those issues could still be lurking |
| [14:42:32] | skd5aner: | yea, no clue... worked just great for me in 0.25, so not applicable in my situation |
| [14:42:46] | skd5aner: | I would say that almost has to be a different issue, if it exists at all |
| [14:43:55] | skd5aner: | stichnot: I'm assuming this patch only needs to apply to the frontend |
| [14:44:15] | bas-t (bas-t!~tycho@52484E89.cm-4-1b.dynamic.ziggo.nl) has quit (Quit: Ex-Chat) | |
| [14:46:36] | J-e-f-f-A (J-e-f-f-A!~J-e-f-f-A@pool-173-48-129-179.bstnma.fios.verizon.net) has quit (Read error: Operation timed out) | |
| [14:46:42] | Seeker`: | eugh, I can't test it because i can't get locks on any channels |
| [14:48:37] | skd5aner: | Seeker`: what tuner are you attempting to use? |
| [14:48:38] | stichnot: | skd5aner: yes, frontend only |
| [14:49:00] | Seeker`: | skd5aner: Think that was my nova-t that was failing to tune |
| [14:49:23] | Seeker`: | I've got a nova-t 500, a PCI-e DVB-S2 and a USB DVB-T2 tuner |
| [14:49:58] | Seeker` (Seeker`!~cjo20@unaffiliated/seeker) has quit (Remote host closed the connection) | |
| [14:50:09] | skd5aner: | stichnot: why would this impact the IVTV stuff, but not the digital stuff? |
| [14:52:02] | Seeker` (Seeker`!~cjo20@unaffiliated/seeker) has joined #mythtv | |
| [14:53:41] | stichnot: | skd5aner: when I added extra logging to avformat_find_stream_info(), I found that it took different paths for PVR-150 versus HDHR. Probably something about the PVR-150 format causes it to probe further into the file and run into the packet truncation problem. |
| [14:54:15] | skd5aner: | ah |
| [14:54:20] | stichnot: | In particular, I notice log messages like "Codec 0x57565455 is not in the full list." "parser not found for codec unknown_codec, packets or times may be invalid." before getting into the truncation problem |
| [14:55:53] | stichnot: | I also saw the packet truncation messages for in-progress recordings, which probably explains the 20s startup time for playback |
| [15:02:39] | laga (laga!~laga@h1626373.stratoserver.net) has quit (Ping timeout: 240 seconds) | |
| [15:02:45] | laga (laga!~laga@h1626373.stratoserver.net) has joined #mythtv | |
| [15:04:11] | Jordack (Jordack!~jordack@h69-131-44-221.plmomi.dedicated.static.tds.net) has joined #mythtv | |
| [15:12:40] | Goga777 (Goga777!~Goga777@128-71-230-128.broadband.corbina.ru) has joined #mythtv | |
| [15:14:08] | stichnot: | danielk22: the cleanest solution I can think of is to get RemoteFile::GetFileSize() to return a suitably large value for in-progress files (at least during the period ffmpeg queries it), and make sure ffmpeg doesn't do the thing on those files where it seeks to 250000 bytes from the end and probes. |
| [15:14:19] | stichnot: | better than customizing the ffmpeg code |
| [15:38:18] | J-e-f-f-A (J-e-f-f-A!~J-e-f-f-A@pool-173-48-113-214.bstnma.fios.verizon.net) has joined #mythtv | |
| [15:53:00] | stichnot (stichnot!~stichnot@mythtv/developer/stichnot) has quit (Ping timeout: 264 seconds) | |
| [15:54:36] | gregL (gregL!~greg@cpe-74-76-105-205.nycap.res.rr.com) has quit (Ping timeout: 245 seconds) | |
| [15:56:21] | gregL (gregL!~greg@cpe-74-76-105-205.nycap.res.rr.com) has joined #mythtv | |
| [15:58:51] | Sharky-Sleep is now known as Sharky112065 | |
| [15:59:18] | Merlin83b (Merlin83b!~Daniel@2a00:1ee0:3:1337:f0c5:7ea2:c4fe:e7f8) has quit (Read error: Connection reset by peer) | |
| [16:09:04] | skd5aner: | ah... bummer, stichnot's not here |
| [16:09:37] | skd5aner: | stichnot: if you read backlog, patch WORKED! I can now watch Live TV on my analog tuners (PVR-500) with no issues... glorious day! |
| [16:09:42] | gregL (gregL!~greg@cpe-74-76-105-205.nycap.res.rr.com) has quit (Ping timeout: 264 seconds) | |
| [16:11:08] | NightMonkey (NightMonkey!~NightrMon@pdpc/supporter/professional/nightmonkey) has joined #mythtv | |
| [16:14:30] | J-e-f-f-A (J-e-f-f-A!~J-e-f-f-A@pool-173-48-113-214.bstnma.fios.verizon.net) has quit (Ping timeout: 264 seconds) | |
| [16:19:21] | J-e-f-f-A (J-e-f-f-A!~J-e-f-f-A@pool-173-48-113-147.bstnma.fios.verizon.net) has joined #mythtv | |
| [16:22:46] | Goga777 (Goga777!~Goga777@128-71-230-128.broadband.corbina.ru) has quit (Remote host closed the connection) | |
| [16:25:08] | FabriceMG (FabriceMG!~Thunderbi@217.112.59.207) has quit (Quit: FabriceMG) | |
| [16:28:47] | SteveGoodey (SteveGoodey!~steve@host86-149-167-13.range86-149.btcentralplus.com) has joined #mythtv | |
| [16:36:20] | stichnot (stichnot!~stichnot@67.218.104.4) has joined #mythtv | |
| [16:36:21] | stichnot (stichnot!~stichnot@67.218.104.4) has quit (Changing host) | |
| [16:36:21] | stichnot (stichnot!~stichnot@mythtv/developer/stichnot) has joined #mythtv | |
| [16:36:38] | bas-t (bas-t!~tycho@52484E89.cm-4-1b.dynamic.ziggo.nl) has joined #mythtv | |
| [16:38:27] | stichnot: | skd5aner: I'm very happy to hear. |
| [16:39:30] | gregL (gregL!~greg@cpe-74-76-105-205.nycap.res.rr.com) has joined #mythtv | |
| [16:43:32] | jams: | stichnot, will be testing .26 shortly. Doing a checkout now |
| [16:52:33] | stichnot: | Another comment. The original issue may be exacerbated (or improved) from varying delays from tuning times, channel changing scripts, etc. In my setup, I have the PVR-150 connected to the STB's S-Video output and no channel-changing script (too lazy), so playback startup is fastest and most consistent. Other people's results may vary, which is why I'd like wider testing. |
| [16:54:00] | bas-t: | stichnot: when i test your patch, do i need to apply it to the backend as well? |
| [16:54:10] | stichnot: | no, just the frontend will do |
| [16:54:15] | bas-t: | ok |
| [16:55:06] | bas-t (bas-t!~tycho@52484E89.cm-4-1b.dynamic.ziggo.nl) has quit (Quit: Ex-Chat) | |
| [16:56:21] | skd5aner: | stichnot: my wife told me to personally thank whomever created that patch – she's the one who relies on live tv the most while she watches our daughter throughout the day and browses between game shows and cartoons |
| [16:57:28] | skd5aner: | WAF finally restored for the first time since 0.26 came out – between this patch and the one that jya created (that, unfortunatley broke playback for others and was reverted) – I'm back a mostly stable setup once again |
| [16:57:33] | stichnot: | cool, she's welcome :) |
| [16:58:34] | stichnot: | I want to get more testing and feedback, then find a better approach that won't risk regressions every time we do an ffmpeg resync. |
| [16:59:57] | stichnot: | btw, I noticed that seeking during live TV seems to have stopped working :( |
| [17:01:07] | bas-t (bas-t!~tycho@52484E89.cm-4-1b.dynamic.ziggo.nl) has joined #mythtv | |
| [17:01:10] | stichnot: | at the very least, it appears that the player isn't getting seektable updates from the backend |
| [17:01:34] | stichnot: | this is on Master |
| [17:04:56] | stichnot: | skd5aner: fwiw, your stack trace in http://code.mythtv.org/trac/attachment/ticket . . . db.mfe.2.txt is consistent with what I see (see the trace of thread 1 at the end) |
| [17:10:44] | skd5aner: | good to hear – I always worry that a dev won't be able to replicate it the same as I could |
| [17:11:28] | stichnot (stichnot!~stichnot@mythtv/developer/stichnot) has quit (Ping timeout: 246 seconds) | |
| [17:14:05] | NightMonkey (NightMonkey!~NightrMon@pdpc/supporter/professional/nightmonkey) has quit (Quit: Body blow! Body blow!) | |
| [17:14:53] | knightr (knightr!~knightr@69-165-170-178.dsl.teksavvy.com) has joined #mythtv | |
| [17:14:53] | knightr (knightr!~knightr@69-165-170-178.dsl.teksavvy.com) has quit (Changing host) | |
| [17:14:54] | knightr (knightr!~knightr@mythtv/developer/knightr) has joined #mythtv | |
| [17:21:39] | NightMonkey (NightMonkey!~NightrMon@pdpc/supporter/professional/nightmonkey) has joined #mythtv | |
| [17:21:39] | knightr: | gigem, you might want to post about the theming modifications you made on the -theming mailing list. If you are not subscribed to it if you let me know what you want me to tell them I could do it for you. |
| [17:22:20] | NightMonkey (NightMonkey!~NightrMon@pdpc/supporter/professional/nightmonkey) has quit (Read error: Connection reset by peer) | |
| [17:23:21] | skd5aner: | knightr: I thought the best idea might be to have a wiki page that indicates exactly which changes impact themeing per release, so that themers can know what's deprecated, removed, added, et |
| [17:23:43] | stichnot (stichnot!~stichnot@216.239.45.73) has joined #mythtv | |
| [17:23:43] | stichnot (stichnot!~stichnot@216.239.45.73) has quit (Changing host) | |
| [17:23:43] | stichnot (stichnot!~stichnot@mythtv/developer/stichnot) has joined #mythtv | |
| [17:27:10] | SteveGoodey (SteveGoodey!~steve@host86-149-167-13.range86-149.btcentralplus.com) has quit (Quit: Konversation terminated!) | |
| [17:28:18] | wagnerrp: | knightr: good idea... |
| [17:28:23] | SteveGoodey (SteveGoodey!~steve@host86-149-167-13.range86-149.btcentralplus.com) has joined #mythtv | |
| [17:28:30] | Steve-Goodey (Steve-Goodey!~steve@host86-149-167-13.range86-149.btcentralplus.com) has joined #mythtv | |
| [17:38:22] | Sharky112065 is now known as Sharky-AFK | |
| [17:54:49] | knightr: | skd5aner, that`s a good idea as well but I think we should mention these on the theming mailing list as well, not just document it on the wiki... |
| [17:55:27] | knightr: | wagnerrp, thanks! |
| [17:55:38] | knightr: | gotta go, ttyl... |
| [17:57:09] | NightMonkey (NightMonkey!~NightrMon@pdpc/supporter/professional/nightmonkey) has joined #mythtv | |
| [18:06:11] | Goga777 (Goga777!~Goga777@128-71-230-128.broadband.corbina.ru) has joined #mythtv | |
| [18:09:03] | jheizer__ (jheizer__!~jon@c-98-226-220-178.hsd1.il.comcast.net) has joined #mythtv | |
| [18:13:32] | gigem: | knightr: what wagnerrp said. I'll post shortly. |
| [18:17:16] | jheizer__ (jheizer__!~jon@c-98-226-220-178.hsd1.il.comcast.net) has quit (Ping timeout: 246 seconds) | |
| [18:19:28] | gigem: | neufeld_AFK: Regarding #11108, will you please add your backend logs with -v general,schedule --loglevel info to the ticket. danielk22, I think that code is trying to be too smart, especially when you consider it doesn't know whether or not preroll can be applied. I'm inclined to take a much simpler approach — if time to next recording is < 1s, then sleep exact ms, else if time is less than preroll – 60s, |
| [18:19:28] | ** MythLogBot http://code.mythtv.org/trac/ticket/11108 ** | |
| [18:19:30] | gigem: | then sleep 1s, else sleep 60s. What do you think? |
| [18:22:25] | neufeld_AFK is now known as neufeld | |
| [18:23:27] | natanojl (natanojl!~jonatan@c83-252-237-63.bredband.comhem.se) has joined #mythtv | |
| [18:25:12] | jams: | stichnot, pvr500 is now working on .26. I'm going to let it run for abit and check the live tv boundry issue |
| [18:25:47] | jams: | will also have to recheck the hvr-2250 analog |
| [18:29:11] | neufeld: | gigem: I can generate some of those, but let me just give you a quick rundown on what appears to be happening first. At scheduler.cpp:1769, it assigns secs_to_next to be the time to the next recording start time. Let's say it's 25 seconds, and prerollseconds is 15. It enters the if block at 1776. It calls "CalcTimeToNextHandleRecordingEvent()", with prerollseconds. But that function doesn't use prerollseconds, |
| [18:29:11] | neufeld: | it's an unused parameter in Scheduler::CalcTimeToNextHandleRecordingEvent() at 1916. So, it gets back 25 seconds, not 10 seconds. Then it logs the sleep for 25000 ms at 1778, hits the reschedWait() for 25000 ms, and you've missed your preroll window. |
| [18:33:19] | neufeld: | gigem: the important part of my patch simply made CalcTimeToNextHandleRecordingEvent() use that parameter. The other few lines were defensive changes to ensure that the parameters affected by that change didn't enter a previously unseen domain and so trigger unexpected behaviour. |
| [18:33:55] | ** neufeld will revert his patch and generate some logs for the ticket ** | |
| [18:36:56] | jams: | stichnot, when changing channels the live-tv chain is preserved and seeking works across boundries. The bad news is live-tv hung and failed with a videobuffer error at the 1/2 hour mark when the guide data changed. |
| [18:43:02] | peper03 (peper03!~peper03@port-92-203-68-84.dynamic.qsc.de) has joined #mythtv | |
| [18:44:19] | gigem: | neufeld: We shouldn't use prerollseconds there because we can't kknow whether or not it's going to be used without querying the recorder and we don't want to do it there. Also, unless your scheduling times are very high, you should normally be going through the else part of that if statement. That's why I want to see your logs. Either way, the scheduler used to always sleep 1s, so the easiest and simplest |
| [18:44:20] | gigem: | solution is just go back to sleeping 1s except when we know we're no where near any of the majic windows before a recording is supposed to start. |
| [18:50:22] | peper03: | Anyone know why the second call to avcodec_decode_video2 here https://github.com/MythTV/mythtv/blob/master/ . . . er.cpp#L3289 is necessary? |
| [18:50:47] | neufeld: | gigem: OK. My backend is recording for another 70 minutes, I'll be able to swap it out for the unpatched version after that. |
| [18:52:33] | peper03: | It obviously is, because it doesn't work if it's not there but 'context' doesn't appear to have been changed significantly after the first call and 'pkt' is a const pointer, so unless something sneaky is happening, I don't understand why it works. I couldn't find any significant documentation for the call either. |
| [18:52:35] | stichnot: | jams: I also want to test this with danielk22's torture test adapted from http://code.mythtv.org/trac/attachment/ticket . . . 30-sec.patch which forces a new program transition every 30 seconds. |
| [18:53:44] | jams: | cool i can apply that. Any specific logging I should be running with? |
| [18:56:16] | Goga777 (Goga777!~Goga777@128-71-230-128.broadband.corbina.ru) has quit (Ping timeout: 245 seconds) | |
| [19:01:23] | stichnot: | Off the top of my head, I would run with -v libav,playback --loglevel debug. The backend and frontend should both be running with this patch. |
| [19:01:45] | stichnot: | and if the patch doesn't apply, I have my own copy that I've updated as conflicts came in |
| [19:02:10] | jams: | failed again at program transition "waited too long for decoder to fill video buffers" |
| [19:03:19] | stichnot: | Bummer. The libav logs may be useful there. |
| [19:03:50] | jams: | will get that setup shortly |
| [19:15:14] | jams: | well so far it's made it past several forced transitions |
| [19:15:16] | bas-t (bas-t!~tycho@52484E89.cm-4-1b.dynamic.ziggo.nl) has quit (Quit: Ex-Chat) | |
| [19:31:40] | jams: | funny..with that patch applied it made it through many forced transitions as well as the data guided transition |
| [19:49:51] | Steve-Goodey (Steve-Goodey!~steve@host86-149-167-13.range86-149.btcentralplus.com) has quit (Quit: Konversation terminated!) | |
| [19:55:33] | Steve-Goodey (Steve-Goodey!~steve@host86-149-167-13.range86-149.btcentralplus.com) has joined #mythtv | |
| [20:17:06] | dijonyummy (dijonyummy!~dijonyumm@64.56.206.254) has joined #mythtv | |
| [20:17:56] | dijonyummy: | is there a better listing source than schedulesdirect, i notice recently they episode descriptions are sometimes lacking or even wrong |
| [20:18:16] | dijonyummy: | sorry i wrong channel |
| [20:18:23] | dijonyummy (dijonyummy!~dijonyumm@64.56.206.254) has left #mythtv ("Leaving") | |
| [20:25:10] | bas-t (bas-t!~tycho@52484E89.cm-4-1b.dynamic.ziggo.nl) has joined #mythtv | |
| [20:26:35] | bas-t: | stichnot: Both tests i conducted are done with fixes/0.26 : |
| [20:26:35] | bas-t: | HEAD is now at 28144cd Revert "Temporary fix for #11159: Fix playback of HD-PVR recording" |
| [20:26:35] | bas-t: | Second test inluding your patch of course. |
| [20:26:35] | bas-t: | Testing sequence: start Live TV, switch channel, switch input, switch channel, stop Live TV. |
| [20:26:35] | bas-t: | I did not observe any trouble in both tests, all was good. |
| [20:26:36] | bas-t: | I logged both tests with -v playback --loglevel debug. |
| [20:26:37] | bas-t: | Result of the first run (without the patch): http://pastebin.com/WNJfBKjh |
| [20:26:39] | bas-t: | Second run (with patch): http://pastebin.com/dGEJ1Zqh |
| [20:47:52] | NightMonkey (NightMonkey!~NightrMon@pdpc/supporter/professional/nightmonkey) has quit (Quit: Body blow! Body blow!) | |
| [21:06:31] | SteveGoodey (SteveGoodey!~steve@host86-149-167-13.range86-149.btcentralplus.com) has quit (Quit: Konversation terminated!) | |
| [21:08:45] | neufeld: | gigem: I'm running with the logging options and the old binary now, but the bug hasn't manifested on the 5 recordings since then. I'll update the bug and ping you once I get a helpful log. Thank you for looking into it. |
| [21:12:18] | Steve-Goodey (Steve-Goodey!~steve@host86-149-167-13.range86-149.btcentralplus.com) has quit (Quit: Konversation terminated!) | |
| [21:14:53] | gigem: | neufeld: Okay. |
| [21:17:45] | dblain (dblain!~dblain@mythtv/developer/dblain) has joined #mythtv | |
| [21:31:16] | jams: | stichnot, been testing for awhile now and can't seem to duplicate the failed transistions again |
| [21:35:22] | bas-t (bas-t!~tycho@52484E89.cm-4-1b.dynamic.ziggo.nl) has quit (Quit: Ex-Chat) | |
| [21:40:45] | dblain_ (dblain_!~dblain@mythtv/developer/dblain) has joined #mythtv | |
| [21:42:28] | dblain (dblain!~dblain@mythtv/developer/dblain) has quit (Ping timeout: 248 seconds) | |
| [21:51:12] | natanojl: | stichnot: I haven't tested much but your patch seems fix it. Without it I got video for a few seconds and then it froze. Haven't tried changing channel or program transitions |
| [21:53:42] | dblain (dblain!~dblain@mythtv/developer/dblain) has joined #mythtv | |
| [21:53:56] | dblain_ (dblain_!~dblain@mythtv/developer/dblain) has quit (Ping timeout: 248 seconds) | |
| [21:58:05] | dblain (dblain!~dblain@mythtv/developer/dblain) has quit (Client Quit) | |
| [22:10:43] | Sharky-AFK is now known as Sharky112065 | |
| [22:41:24] | stichnot: | bas-t (if you read this): The logs would be a lot more informative with -v playback,libav --loglevel debug because it's in the libav area where the fundamental problems are happening. Also, for proper comparison, it would be really helpful if the "unpatched" version were actually patched with the avformatdecoder.cpp change that unsuppresses the ffmpeg logging. |
| [22:41:38] | Sharky112065 is now known as Sharky-AFK | |
| [22:46:11] | stichnot: | I see that Beirdo added https://github.com/MythTV/mythtv/commit/d6413 . . . b13ac258b5ff specifically to silence the ffmpeg logging during av_find_stream_info(). Maybe I should add a mechanism to explicitly enable that logging under some command line option. |
| [22:51:56] | jams: | stichnot, anything in your patch that would cause this ? http://pastebin.com/3AmM6JT0 or is it purely mysql being mysql |
| [22:52:29] | jams: | was recording..watching live tv and brought up the programguide when the incident occured |
| [22:52:55] | Killerkiwi (Killerkiwi!~jason@60-234-255-73.bitstream.orcon.net.nz) has joined #mythtv | |
| [23:00:55] | Bray90820 (Bray90820!3251fd20@gateway/web/freenode/ip.50.81.253.32) has joined #mythtv | |
| [23:01:02] | Bray90820 (Bray90820!3251fd20@gateway/web/freenode/ip.50.81.253.32) has left #mythtv () | |
| [23:03:40] | Sharky-AFK (Sharky-AFK!~Sharky112@c-24-19-57-28.hsd1.wa.comcast.net) has quit (Ping timeout: 252 seconds) | |
| [23:03:42] | stichnot: | jams: doesn't seem to be related to the patch, but keep in mind that there are other live TV problems and race conditions lurking out there |
| [23:04:00] | jams: | sure..i'm fine with blaming mysql :) |
| [23:04:08] | stichnot: | I doubt it's mysql though... |
| [23:09:03] | petefunk (petefunk!~pfunk@unaffiliated/petefunk) has quit (Ping timeout: 264 seconds) | |
| [23:10:07] | aloril (aloril!~aloril@dsl-tkubrasgw3-54f97e-153.dhcp.inet.fi) has quit (Ping timeout: 256 seconds) | |
| [23:10:07] | aloril (aloril!~aloril@dsl-tkubrasgw3-54f97e-153.dhcp.inet.fi) has joined #mythtv | |
| [23:10:08] | petefunk (petefunk!~pfunk@198.23.147.3) has joined #mythtv | |
| [23:13:27] | Jordack (Jordack!~jordack@h69-131-44-221.plmomi.dedicated.static.tds.net) has quit () | |
| [23:26:20] | wahrhaft (wahrhaft!~quassel@cpe-24-210-71-0.columbus.res.rr.com) has quit (Remote host closed the connection) | |
| [23:28:03] | wahrhaft (wahrhaft!~quassel@cpe-24-210-71-0.columbus.res.rr.com) has joined #mythtv | |
| [23:32:14] | sl1ce (sl1ce!~johnathan@pool-100-0-73-123.bstnma.fios.verizon.net) has quit (Quit: Konversation terminated!) | |
| [23:36:14] | Sharky-AFK (Sharky-AFK!~Sharky112@c-24-19-57-28.hsd1.wa.comcast.net) has joined #mythtv | |
| [23:40:04] | Sharky-AFK is now known as Sharky112065 | |
| [23:42:31] | sl1ce (sl1ce!~johnathan@pool-100-0-73-123.bstnma.fios.verizon.net) has joined #mythtv | |
| [23:44:11] | sl1ce (sl1ce!~johnathan@pool-100-0-73-123.bstnma.fios.verizon.net) has quit (Client Quit) | |
| [23:50:00] | jheizer__ (jheizer__!~jon@c-98-226-220-178.hsd1.il.comcast.net) has joined #mythtv | |
| [23:50:48] | sl1ce (sl1ce!~johnathan@pool-100-0-73-123.bstnma.fios.verizon.net) has joined #mythtv | |
| [23:50:48] | sl1ce (sl1ce!~johnathan@pool-100-0-73-123.bstnma.fios.verizon.net) has quit (Remote host closed the connection) | |
| [23:50:49] | sl1ce (sl1ce!~johnathan@pool-100-0-73-123.bstnma.fios.verizon.net) has joined #mythtv | |
| [23:51:13] | danielk22: | stuartm: detaching the string in mythuitextedit won't fix any segfaults, just change the odds a bit. |
| [23:52:16] | danielk22: | Instead add an assert in the GetText() and SetText() that triggers when it called from outside the UI thread. Then get the backtrace.. |
| [23:58:39] | natanojl (natanojl!~jonatan@c83-252-237-63.bredband.comhem.se) has quit (Ping timeout: 260 seconds) | |
IRC Logs collected by
BeirdoBot.
Please use the above link to report any bugs.