:: #mythtv

Daily chat history

Current users (72):

aca20031, aloril, Anssi, Beirdo_, bobweaver, brfransen, Chutt_, clever, coling, Cougar, danielk221, dblain, dekarl, ElmerFudd, ephemer0l, fetzerch, foobum, foxbuntu`, gary_buhrmaster, ghoti, Gibby, gregL, GreyFoxx, J-e-f-f-A, jams_, jarle, jarryd, jheizer, joe_____, joki, jpabq, jpabq_, jpharvey, jst, jwhite, jya, kormoc, KungFuJesus, kurre2, kwmonroe, laga, moparisthebest, mrand, MythBuild, MythLogBot, neufeld, nyloc, poptix, purserj, pvr4me, rhpot1991, rsiebert, seld, Sharky112065, SmallR2002, sphery, sraue, stuartm, superm1, tgm4883, Tobbe5178, tonsofpcs, tris, unforgiven512, wagnerrp, wahrhaft_, whoDat_, wilmoore-misc, wolfgang2, XDS2010_, xris, _charly_
Saturday, June 15th, 2013, 00:07 UTC
[00:07:49] Chutt (Chutt! has quit (Read error: Connection reset by peer)
[00:18:16] Chutt (Chutt! has joined #mythtv
[00:19:08] Chutt (Chutt! has quit (Read error: Connection reset by peer)
[00:20:17] Chutt (Chutt! has joined #mythtv
[00:25:28] Chutt_ (Chutt_! has joined #mythtv
[00:25:41] Chutt (Chutt! has quit (Ping timeout: 248 seconds)
[01:11:20] NightMonkey (NightMonkey!~NightrMon@pdpc/supporter/professional/nightmonkey) has quit (Quit: Body blow! Body blow!)
[01:13:37] skd5aner (skd5aner! has quit (Ping timeout: 248 seconds)
[01:22:27] stichnot (stichnot!~stichnot@mythtv/developer/stichnot) has joined #mythtv
[02:07:53] stichnot (stichnot!~stichnot@mythtv/developer/stichnot) has quit (Ping timeout: 256 seconds)
[02:09:40] _nyloc_ (_nyloc_! has quit (Read error: Operation timed out)
[02:11:00] nyloc (nyloc! has joined #mythtv
[03:01:24] peper03 (peper03! has quit (Ping timeout: 260 seconds)
[03:02:41] peper03 (peper03! has joined #mythtv
[03:24:04] stichnot (stichnot! has joined #mythtv
[03:24:04] stichnot (stichnot!~stichnot@mythtv/developer/stichnot) has joined #mythtv
[03:24:04] stichnot (stichnot! has quit (Changing host)
[03:29:46] fetzerch (fetzerch!~quassel@unaffiliated/fetzerch) has quit (Ping timeout: 256 seconds)
[03:31:21] fetzerch (fetzerch!~quassel@unaffiliated/fetzerch) has joined #mythtv
[03:57:23] joki (joki! has quit (Ping timeout: 260 seconds)
[04:02:02] joki (joki! has joined #mythtv
[05:41:36] stoffel (stoffel! has joined #mythtv
[06:01:14] jya_ (jya_!~jyavenard@mythtv/developer/jya) has quit (Quit: jya_)
[06:13:06] jya_ (jya_!~jyavenard@mythtv/developer/jya) has joined #mythtv
[06:44:46] wilmoore-misc (wilmoore-misc! has joined #mythtv
[06:45:08] stoffel (stoffel! has quit (Ping timeout: 240 seconds)
[07:04:21] SteveGoodey (SteveGoodey! has joined #mythtv
[07:15:27] jya_ (jya_!~jyavenard@mythtv/developer/jya) has quit (Quit: jya_)
[07:29:13] jya_ (jya_!~jyavenard@mythtv/developer/jya) has joined #mythtv
[09:00:41] SteveGoodey (SteveGoodey! has quit (Quit: Konversation terminated!)
[09:16:00] danielk221 (danielk221!~danielk22@ has quit (Ping timeout: 268 seconds)
[09:28:20] SteveGoodey (SteveGoodey! has joined #mythtv
[09:38:23] jya_ (jya_!~jyavenard@mythtv/developer/jya) has quit (Ping timeout: 256 seconds)
[10:54:46] SteveGoodey (SteveGoodey! has quit (Quit: Konversation terminated!)
[11:05:32] IReboot (IReboot! has joined #mythtv
[11:06:18] cocoa117 (cocoa117!~cocoa117@ has quit (Ping timeout: 252 seconds)
[11:09:31] jya_ (jya_!~jyavenard@mythtv/developer/jya) has joined #mythtv
[11:11:34] cocoa117 (cocoa117!~cocoa117@ has joined #mythtv
[12:10:31] paul-h (paul-h!~Paul@ has joined #mythtv
[12:32:31] stuartm: paul-h: you beat me to it :)
[12:34:39] stuartm: we're now down to 92 issues with coverity,
[12:39:14] paul-h: yeah making slow progress :)
[12:42:53] SteveGoodey (SteveGoodey! has joined #mythtv
[12:49:50] paul-h: There's some crazy formatting in services api stuff
[13:18:05] SteveGoodey (SteveGoodey! has quit (Quit: Konversation terminated!)
[13:30:52] pvr4me (pvr4me! has joined #mythtv
[13:53:17] SteveGoodey (SteveGoodey! has joined #mythtv
[14:01:12] stichnot (stichnot! has joined #mythtv
[14:01:12] stichnot (stichnot!~stichnot@mythtv/developer/stichnot) has joined #mythtv
[14:01:12] stichnot (stichnot! has quit (Changing host)
[14:27:25] SteveGoodey (SteveGoodey! has quit (Quit: Konversation terminated!)
[14:35:54] pvr4me (pvr4me! has quit (Quit: pvr4me)
[14:39:39] pvr4me (pvr4me! has joined #mythtv
[15:29:18] stuartm: wagnerrp: #11601, as long as the external player facility exists, we should at least make it case-insensitive
[15:29:18] ** MythLogBot **
[15:33:48] stichnot: Even better, an empty string to indicate Internal.
[15:35:36] jpabq: stichnot: It is good that you picked where you did to insert the REC_STARTED_WRITING event. After thinking about it, I remembered that the signal monitor is not invoked for back-to-back recordings and thus the trigger would not have been sent for subsequent back-to-back on the same input.
[15:36:10] Captain_Murdoch: stuartm, I agree on case and falling back to 'empty means internal' as stichnot mentioned makes sense as well
[15:37:18] stuartm: yeah, that sounds good
[15:37:38] Captain_Murdoch: jya, when you get a chance, can you try to reproduce the double-install Theme Chooser issue since you opened the ticket for it? I wasn't able to reproduce it last night, but did put in something that I think will fix the race I was seeing when I could reproduce it on another machine once a while back.
[15:37:40] stuartm: maybe even falling back to Internal if the executable defined cannot be found
[15:37:54] Captain_Murdoch: so many "makes perfect senses" :)
[15:38:09] stuartm: heh
[15:40:00] ** Captain_Murdoch finally slapped a DVD drive into his main mythfrontend box last weekend since he never unboxed the standalone DVD players after moving cross-country a year ago. **
[15:40:04] stuartm: I'll fix it
[15:40:42] danielk221 (danielk221!~danielk22@ has joined #mythtv
[15:41:01] stuartm: Captain_Murdoch: good, or bad? It's come a very long way in the last few months, although some of the fixes and improvements haven't made it back to 0.26
[15:42:38] Captain_Murdoch: haven't tried much with it. we've ripped every DVD we have to disk already. I was rebuilding my dev machine though and had a new SATA DVD drive for that, so I threw my old PATA DVD into the main frontend. the only DVD I've played with it so far is a custom DVD that a parent made of my daughter's 1st grade class. mainly a picture DVD, but I'm not sure if it uses still frames or is actually video of still frames. have to chec
[15:42:38] Captain_Murdoch: k it out sometime to see.
[15:43:32] Captain_Murdoch: this was 0.25 as well, I needed to get my dev system up to snuff before upgrading production.
[15:44:22] cocoa117 (cocoa117!~cocoa117@ has quit (Read error: Connection reset by peer)
[15:45:20] Captain_Murdoch: and will need to upgrade master backend before I can run 0.27 on that due to Qt requirements. so I'm working on getting non-myth functionality off that machine. when I moved, I consolidated everything back onto my fileserver since I can use HDHomeRun's for everything now (cablecard for non-locals and non-cablecard for locals)
[15:46:10] Captain_Murdoch: finally getting a home office built so I can unbox the last sets of boxes sitting in the garage. :)
[15:46:50] ** Captain_Murdoch is off to breakfast and garage sales with his daughter. **
[15:46:51] stuartm: I really wanted to get LVR's encrypted dvd patch tested this weekend, but that's looking less and less likely atm
[15:47:20] Captain_Murdoch: yeah, I'd take some time to review. is that in a ticket?
[15:48:05] stuartm: might be, I have it in a dump of all his patches that he sent me once
[15:50:06] stuartm: can't find a ticket atm, I'll create one and attach the patch
[15:53:24] Captain_Murdoch: ok, thanks. I was looking for a saved copy but couldn't find it. may try to read it over if I get time tonight (and remember to). :)
[15:53:37] ** Captain_Murdoch is afk again. **
[15:54:23] stuartm: . . . from-s.patch
[15:54:42] stuartm: #11602
[15:54:42] ** MythLogBot **
[16:04:22] stichnot: jpabq: that was more your choice than mine :)
[16:04:35] stichnot: that said, I did try to test for back-to-back recordings as well
[16:07:04] jpabq: I am running a couple different versions of mythbackend (0.26 and master with different ports) on the same machine, and having a hell-of-a time getting useful logging. Is there some trick to getting mythlogserver to deal with a situation like that? How do you tell mythlogserver what port to use, for example?
[16:08:55] pvr4me (pvr4me! has quit (Quit: pvr4me)
[16:09:23] rsiebert (rsiebert! has joined #mythtv
[16:12:23] rsiebert_ (rsiebert_! has quit (Ping timeout: 240 seconds)
[16:15:41] SteveGoodey (SteveGoodey! has joined #mythtv
[16:32:39] paul-h: stuartm: you can get an up to date version of that patch from here . . .
[16:34:19] paul-h: It's in mythtv-master and is patch 0029
[16:38:08] paul-h: Captain_Murdoch: I can reproduce the double install problem – I've only tried to install one theme and while it didn't require the theme to be installed twice it didn't quite work properly either
[16:40:30] paul-h: The download completed properly, the progress dialog closed as it should but then nothing happened until the music mini-player popped up I them closed it with ESCAPE and only then did the theme change
[16:41:47] gigem: Woohoo. I finally created a torture test that got my average place time up to 8.6s. After applying my current test patch, the average place time dropped to 6.0s. I then applied anmplexid patch I'd previously set aside on top of the new patch and the average time dropped to 3.3s! I'd still like to see what affect these have on skd5aner's 42s place times before committing, but who knows when he'll be back.
[16:41:48] gigem: My test relies on creating a lot of conflicts which might not be a valid real world scenario.
[16:42:04] gigem: danielk221: Here's an update on my experimentation with disabling fdatasync. I've been running with fdatasync disable for 4 (maybe even 5) weeks now with no ill effects. One big, external-to-MythTV change I made was to lower /proc/sys/vm/dirty_expire_centisecs from the default 30s to 10s. That combination keeps nearly all of TFW writes under 1s, with an occasional (1 or 2 per day) 1s-2s write getting
[16:42:06] gigem: logged and lets the kernel efficiently smooth out the writes to disk when several recordings are in progress. The audible difference in disk activity when near the backend is very noticable.
[16:42:38] pvr4me (pvr4me! has joined #mythtv
[16:56:22] stoffel (stoffel! has joined #mythtv
[17:00:47] stuartm: paul-h: thanks
[17:04:39] pvr4me (pvr4me! has quit (Quit: pvr4me)
[17:07:36] stuartm: think everyone should at least look over that list of patches from LVR, some of them sound interesting, a few bug fixes too
[17:10:12] stichnot: gigem: that's very cool
[17:18:29] stichnot: gigem: Do you think it will ever be feasible to treat Live TV recording as incremental scheduler updates?
[17:44:18] dekarl: stuartm: the patches are a bottomless pit ;) (In a positive sense)
[17:45:00] stuartm: yeah, just pulling the rest of them now
[17:45:59] stuartm: A process that's slowed by the need to edit every message with a trac ticket number :)
[17:46:38] stuartm: we need a coverity hook added, which will close the coverity defect automatically
[17:46:59] stuartm: let me re-phrase, I need to write a hook
[17:47:27] dekarl: btw, i've been thinking if its possible to have a report with all the compiler warnings from the buildbots, similar to cppcheck and coverty... and a script to update the fixes-release notes from git logs
[17:47:35] dekarl: I think I can do the latter one
[17:48:26] stuartm: 78ddf3835 is a beauty
[17:49:18] stuartm: dekarl: yeah I was thinking the same about the compiler warnings, something easier to browse with links to the code would be good, and probably not too hard
[17:51:23] pvr4me (pvr4me! has joined #mythtv
[17:58:01] stuartm: an overview page would be good too, showing the counts for each tool/compiler
[18:00:19] dekarl: btw, who can add me on github, so I can close pull requests?
[18:02:01] dekarl: or close them for me ;) mythtv #46,#47
[18:02:01] ** MythLogBot **
[18:02:01] ** MythLogBot **
[18:06:01] danielk221: gigem: this is a backend only machine?
[18:06:49] danielk221: were you playing video during any of those 1–2 second writes, any hiccups?
[18:11:06] SteveGoodey (SteveGoodey! has quit (Quit: Konversation terminated!)
[18:11:55] gary_buhrmaster (gary_buhrmaster!~gtb@2001:470:80e4:3::36) has joined #mythtv
[18:13:45] gary_buhrmaster: stuartm: re patches: NP. Better too many people fix than none (I seem to recall we had this discussion before). I just wanted to (try to) keep the tickets outstanding under control (if I did not notice it now, it likely wold have been months, and then at least I would not remember the reference).
[18:14:13] SteveGoodey (SteveGoodey! has joined #mythtv
[18:16:39] gary_buhrmaster: stuartm: And thanks for looking at the others (and sorry for the terrible workflow in regards to closing tickets.)
[18:39:37] gigem: stichnot: It kind of already does incremental scheduling for live TV, but there are plenty of holes where it can still break down. What it currently does is perform a place-only reschedule approximately 2 minutes before a recording is to start. If the user changes channels or inputs or another user enters live TV after that reschedule, the scheduler won't change what inputs it intends to use. Ultimately, I
[18:39:39] gigem: think live TV needs to be more tightly coupled to the scheduler where the scheduler tells the frontend what input to use when entering live TV or changing channels.
[18:42:31] gigem: danielk221: Yes, this is on my dedicated, backend-only production system. I likely was watching some recordings when the long writes took place, but I haven't noticed any more playback glitches than normal. My development system is a backend/frontend combo, so I can run some tests on it if you'd like.
[18:50:27] stuartm: gary_buhrmaster: the workflow issue isn't your fault and we may be able to address that through a git alias + script
[18:52:23] paul-h: Is the double !! intentional here? – . . . er.cpp#n1514
[18:54:20] dekarl: c65a45682845 added a bunch of them
[18:54:49] pvr4me (pvr4me! has quit (Quit: pvr4me)
[18:57:53] danielk221: paul-h: yep, it just converts the value from an int to bool.
[18:59:25] paul-h: OK thanks never seen that before :)
[19:00:08] dekarl (dekarl! has quit (Read error: Operation timed out)
[19:01:18] danielk221: It is rarely necessary, but this code was generated from a search & replace where I wanted the result to have the same type & precedence as the original to avoid regressions.
[19:01:40] danielk221: On this line it isn't necessary.
[19:04:23] dekarl (dekarl! has joined #mythtv
[19:04:27] paul-h: The result of (cacheMode & kCacheIgnoreDisk) should be a bool anyway?
[19:08:59] paul-h: stuartm: LVR's 0101 patch is interesting it does fix the flicker in MythGallery but not for remote images sadly, for some reason they don't appear to be caching properly so they always download
[19:10:04] danielk221: paul-h: no the result is an int, but the || that follows it will convert it to a bool.
[19:12:20] danielk221: shortcut operators like && and || force the conversions.. where you need to do the conversion to bool explicitly is if you say want to use ^ as a logical xor.
[19:29:53] exoon (exoon! has joined #mythtv
[19:42:22] pvr4me (pvr4me! has joined #mythtv
[19:52:07] wolfgang2 (wolfgang2! has quit (Ping timeout: 264 seconds)
[19:56:30] wolfgang2 (wolfgang2! has joined #mythtv
[20:13:56] exoon (exoon! has quit (Remote host closed the connection)
[20:19:04] exoon (exoon! has joined #mythtv
[20:26:41] pvr4me (pvr4me! has quit (Quit: pvr4me)
[20:35:26] stuartm: paul-h: when I last looked at the remote image caching I discovered that there was a problem with the way we check whether the cached file needs updating, basically we do a stat on the remote file to see when it was last modified and that introduces a delay to the process
[20:39:31] stuartm: there might be a simple solution to all flashing issues; we simply don't Clear() the existing MythUIImage until the new image has been successfully loaded
[20:40:59] stuartm: in effect double buffering
[20:44:49] cesman (cesman!~cesman@pdpc/supporter/professional/cesman) has quit (Quit: Konversation terminated!)
[20:56:03] stuartm: working on a proof-of-concept patch
[21:15:07] peper03 (peper03! has quit (Quit: Konversation terminated!)
[21:18:26] stuartm: paul-h: does the following patch improve the flickering at all?
[21:30:05] amessina (amessina!~amessina@2001:470:c1dc:7779:d6be:d9ff:fe8d:7c1e) has quit (Quit: Konversation terminated!)
[21:30:41] SteveGoodey (SteveGoodey! has quit (Quit: Konversation terminated!)
[21:47:13] Seeker` (Seeker`!~cjo20@unaffiliated/seeker) has quit (Ping timeout: 248 seconds)
[21:48:55] Seeker` (Seeker`! has joined #mythtv
[21:48:55] Seeker` (Seeker`! has quit (Changing host)
[21:48:56] Seeker` (Seeker`!~cjo20@unaffiliated/seeker) has joined #mythtv
[21:53:04] pvr4me (pvr4me! has joined #mythtv
[21:55:28] stoffel (stoffel! has quit (Ping timeout: 246 seconds)
[22:08:39] neufeld (neufeld! has quit (Read error: Connection reset by peer)
[22:08:47] neufeld (neufeld! has joined #mythtv
[22:09:42] kormoc (kormoc!~kormoc@mythtv/developer/kormoc) has quit (Ping timeout: 264 seconds)
[22:09:51] kormoc (kormoc!~kormoc@mythtv/developer/kormoc) has joined #mythtv
[22:10:35] paul-h: stuartm: the patch wouldn't apply cleanly but it was only the first hunk that failed. It does stop the flicker though :)
[22:10:55] paul-h: stuartm: I could be wrong but it doesn't appear that remote images ever get loaded from either the memory or disk cache even though they do end up in the disk cache at least
[22:12:37] jya_ (jya_!~jyavenard@mythtv/developer/jya) has quit (Quit: jya_)
[22:16:20] paul-h: MythNews is a good test for that if you scroll up/down in the articles list you would think that after it had loaded the image once it would come from the cache but it doesn't it is always re-downloaded
[22:40:14] IReboot (IReboot! has quit (Quit: Ex-Chat)
[22:44:30] exoon (exoon! has quit (Remote host closed the connection)
[22:53:34] paul-h (paul-h!~Paul@ has quit (Quit: Konversation terminated!)
[23:19:12] pvr4me_ (pvr4me_! has joined #mythtv
[23:21:13] pvr4me (pvr4me! has quit (Ping timeout: 246 seconds)
[23:21:14] pvr4me_ is now known as pvr4me

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