MythLogBot@irc.freenode.net :: #mythtv

Daily chat history

Current users (94):

alan`, aloril, Anduin, anykey_, beata, Beirdo, boshhead, brfransen, caelor, Captain_Murdoch, carmony, castlec, cattelan, cdev, clever, coling, Cougar, dagar, danielk22, Dave123, Dave123-road, davide, dlblog, eharris, elmojo, elvum_, f33dMB, foobum, foxbuntu, ghoti, Gibby, gigem, gregL, GreyFoxx, hads, high-rez, iamlindoro, ikonia, J-e-f-f-A, j-rod|afk, jafa, jams, jannau, joe___, jpabq, jpabq-, jpharvey_, jstenback, justinh, justpaul, jwhite, jya, kc, kenni, Kevin`, kha, knightr, kormoc, kurre, laga, lofidellity, m4xx, mag0o, markk_, MavT, mrand, MythLogBot, natanojl, neolive, NightMonkey, nutron, okolsi, ozatomic, paul-h, pheld, pmcenery, PointyPumper, poptix, purserj, reynaldo, rhpot1991, rooaus, skd5aner, sphery, Splat1, stuarta, stuartm, superm1, tgm4883, ThisNewGuy, tris, xris, ybot, _charly_
Friday, December 10th, 2010, 00:01 UTC
[00:01:14] gkeen316 (gkeen316!~gkeen316@d75-156-147-11.abhsia.telus.net) has quit (Ping timeout: 260 seconds)
[00:07:25] jya: jannau: re: cpuid, I can't believe it was just that... ahaha
[00:11:24] sphery: danielk22: so the CPU+ instead of Normal thing was much simpler than I last thought--I had the right fix in my patch, before, but we had a different bug I've since fixed that prevented the fix from working properly before.
[00:12:00] danielk22: sphery: heh, glad it's fixed now.
[00:12:28] danielk22: markk_: Cool, I'll update and try it out.
[00:33:35] xris: wagnerrp: if I have time tonight, I think I'm going to create a code.mythtv repo in git (not github) and we can use that for your work and a couple other things.
[00:33:53] wagnerrp: ok
[00:45:15] Dave123 (Dave123!~dave@cpe-74-74-222-96.rochester.res.rr.com) has quit (Ping timeout: 245 seconds)
[00:46:56] Dave123 (Dave123!~dave@cpe-74-74-222-96.rochester.res.rr.com) has joined #mythtv
[00:51:09] stuartm_ (stuartm_!~gbee@cpc4-derb9-0-0-cust534.8-3.cable.virginmedia.com) has joined #mythtv
[00:51:30] stuartm (stuartm!~gbee@cpc4-derb9-0-0-cust534.8-3.cable.virginmedia.com) has quit (Ping timeout: 245 seconds)
[01:09:25] jpabq (jpabq!~jpabq@174-28-182-216.albq.qwest.net) has quit (Ping timeout: 240 seconds)
[01:09:27] jpabq- (jpabq-!~jpabq@174-28-182-216.albq.qwest.net) has quit (Read error: Operation timed out)
[01:11:02] jpabq (jpabq!~jpabq@174-28-182-216.albq.qwest.net) has joined #mythtv
[01:11:32] jpabq- (jpabq-!~jpabq@174-28-182-216.albq.qwest.net) has joined #mythtv
[01:39:57] gigem (gigem!~david@host137.12.intrusion.com) has quit (Remote host closed the connection)
[01:40:25] gigem (gigem!~david@host137.12.intrusion.com) has joined #mythtv
[01:52:53] jya: does the git repo contains old branches like the hdaudio-branch?
[01:53:42] jya: trying to go back in time find where something broke
[01:55:40] jya: [25035]
[01:55:40] MythLogBot: SVN 25035: (branch master) https://github.com/MythTV/mythtv/commit/67bcd82a
[02:13:07] danielk22: jya: I'm pretty sure only the active branches were ported. In order to port most of the old branches we would have had to rebuild svn with those branches as branches from root instead of root/mythtv. Not impossible but very labor intensive.
[02:13:29] danielk22: svn is still there for the history...
[02:13:32] jya: danielk22: the hdaudio branch was a branch from root
[02:13:42] jya: yeah, I wanted to use git bisect
[02:13:54] jya: to easily find out where it broke
[02:14:47] jya: talked to foobum regarding the SSE3 instruction in the SSE2 code, interestingly, it was never picked up because his AMD cpu reports as SSE2 only, yet SSE3 code works
[02:15:02] danielk22: jya: In that case I don't know if it's there, but with svn it is fairly easy to bisect using revision numbers.
[02:15:11] jya: wonder if AMD silently failed to report SSE3 just to avoid paying royalties
[02:15:32] jya: yeah, I can do similar to bisect, but it's more involving
[02:16:03] danielk22: jya: Some of SSE3 is present in contemporary AMD processors that don't report SSE3 since they don't support the full spec.
[02:16:56] jya: i see.
[02:17:24] jya: well, back to May 1st, and it is still broken... wierd, it works on foobum machine just fine (he's still running version as of may 30th)
[02:17:53] danielk22: Did foobum really see a speed up with that SSE3 code. I'm doubtful of it just looking at it. It seems like memory access should dominate timings.
[02:18:36] jya: I never looked that part of the code (rarely use timestretch) but for the floating point conversion, yes, the improvement was massive
[02:18:44] jya: 11–12 times faster
[02:18:56] jya: that's in audiooutpututil.cpp
[02:19:05] jya: that's where I concentrated my testing
[02:19:32] jya: as it's the most significant slow down that could have occurred
[02:19:38] danielk22: jya: 11–12 than the SSE2 code, or than C++ code without SSE math enabled?
[02:19:54] jya: SSE vs C++
[02:20:09] jya: that's compiling as standard
[02:20:16] jya: not sure what the compiler does
[02:20:31] jya: the SSE code process 16 bytes in one go
[02:20:32] danielk22: That I can see, gcc will use the old FPU unless you tell it explicitly to use SSE.
[02:21:25] jya: with SSE code for the float conversion, it's almost as fast as a plain memory copy
[02:22:44] danielk22: But only thing the SSE3 code eliminates is two a half-cycle operations over the SSE2 code. But those cycles won't cost you anything because they are hidden by memory throughput limitations.
[02:22:57] jya: ok, now I'm annoyed, I thought I would find a revision where upconversion to LPCM did work ; but I can't find one now
[02:23:26] jya: so now I have no version that do work, and they are all broken in a different fashion.. wtf..
[02:23:43] jya: I've been stuck on this for days... haven't been to work in 3 days :)
[02:24:02] jya: I ... have ..... to .... fix ..... it !
[02:24:39] jya: it's werid, the videoplayer issues a pause audio command, and never comes out of it
[02:24:51] jya: i don't get it...
[02:25:11] jya: i need fresh eyes to look over it..
[02:26:43] jya: it is okay if I add a command line switch the prevent checking if the database or mythbackend schema is correct? I test with a backend running 0.24, frontend is running trunk. what I'm testing do not involve the backend at all, so I always have to modify the dbcheck.cp pand mythversion.h before I compile... time waster
[02:29:27] danielk22: jya: there is already a switch. I wish I could remember it, but it slips my mind. Something about expert mode.
[02:29:36] jya: ahhh
[02:29:59] jya: so you're telling me I've been wasting my time for nothing :P
[02:31:54] sphery: jya: haven't read all of backlog, but no, don't add it
[02:31:59] sphery: it's already there
[02:32:04] jya: and it is ?
[02:32:12] sphery: not something I'll mention in here
[02:32:44] jya: obviously people using it can be expected to know what they are doing ...
[02:32:46] Beirdo^2 (Beirdo^2!~beirdo@unaffiliated/beirdo) has joined #mythtv
[02:53:08] Beirdo^2 (Beirdo^2!~beirdo@unaffiliated/beirdo) has quit (Remote host closed the connection)
[03:23:59] markk_: elmojo: would appreciate it if you could sanitise http://mythtv.pastebin.ca/2015291 for me. it's a fix for #9349. It's a little more invasive than absolutely necessary but I can't see any problem here.
[03:58:44] markk_: sphery: thanks for sorting the database stuff for display profiles – I always avoid touching the db :)
[04:00:03] sphery: markk_: heh, I felt much more comfortable there than doing the libmpeg stuff--since it went too deep into player code :)
[04:01:04] gkeen316 (gkeen316!~gkeen316@d75-156-147-11.abhsia.telus.net) has joined #mythtv
[04:02:06] sphery: so thanks for your reviewing that change, yesterday. and thanks to danielk22 for helping me figure out which changes to make to the remaining profiles
[04:06:52] jya: sphery : will existing profile be updated
[04:07:07] jya: or the user will have to delete it manually and only then it will be re-created
[04:08:13] sphery: yeah, if they're using one of the profile groups I delete, it will then use GetDefaultProfileName(), which will choose the most appropriate one (Normal), and if that doesn't exist, then profiles[0]
[04:35:20] rooaus (rooaus!~cameron@ppp59-167-126-229.static.internode.on.net) has quit (Ping timeout: 240 seconds)
[04:40:55] gkeen316 (gkeen316!~gkeen316@d75-156-147-11.abhsia.telus.net) has quit (Ping timeout: 240 seconds)
[04:50:33] rooaus (rooaus!~cameron@ppp59-167-126-229.static.internode.on.net) has joined #mythtv
[04:53:42] jya: anyone using 6 channels analog or hdmi out who could try something for me ?
[05:04:17] PointyPumper (PointyPumper!~pintlezz@190.244.73.13) has quit (Ping timeout: 265 seconds)
[05:22:20] NightMonkey (NightMonkey!debian-tor@pdpc/supporter/professional/nightmonkey) has quit (Ping timeout: 245 seconds)
[05:55:40] gigem (gigem!~david@host137.12.intrusion.com) has quit (*.net *.split)
[05:55:40] MaverickTech (MaverickTech!~MaverickT@dns2.arel.com.au) has quit (*.net *.split)
[05:55:40] clever (clever!~clever@142.167.220.215) has quit (*.net *.split)
[05:55:40] cesman (cesman!~cecil@pdpc/supporter/professional/cesman) has quit (*.net *.split)
[05:55:41] Kevin` (Kevin`!~kevin@router.kwzs.be) has quit (*.net *.split)
[05:59:49] gigem (gigem!~david@host137.12.intrusion.com) has joined #mythtv
[05:59:49] MaverickTech (MaverickTech!~MaverickT@dns2.arel.com.au) has joined #mythtv
[05:59:49] clever (clever!~clever@142.167.220.215) has joined #mythtv
[05:59:49] cesman (cesman!~cecil@pdpc/supporter/professional/cesman) has joined #mythtv
[05:59:49] Kevin` (Kevin`!~kevin@router.kwzs.be) has joined #mythtv
[06:04:50] kenni (kenni!~kenni@pfsense.dhcp.pop.k-net.dk) has quit (Ping timeout: 250 seconds)
[06:06:49] wagnerrp (wagnerrp!~wagnerrp_@NR-FT1-66-42-240-159.fuse.net) has quit (Ping timeout: 255 seconds)
[06:07:06] NightMonkey (NightMonkey!debian-tor@pdpc/supporter/professional/nightmonkey) has joined #mythtv
[06:09:18] wagnerrp (wagnerrp!~wagnerrp_@NR-FT1-66-42-240-159.fuse.net) has joined #mythtv
[06:36:05] stoffel (stoffel!~quassel@p57B4A854.dip.t-dialin.net) has joined #mythtv
[07:00:32] kenni (kenni!~kenni@pfsense.dhcp.pop.k-net.dk) has joined #mythtv
[07:27:09] MaverickTech (MaverickTech!~MaverickT@dns2.arel.com.au) has quit (Ping timeout: 240 seconds)
[07:30:29] cesman (cesman!~cecil@pdpc/supporter/professional/cesman) has quit (Ping timeout: 240 seconds)
[07:40:21] el_duerino (el_duerino!~afey@2001:4bd8:2:2:223:aeff:fe91:6669) has joined #mythtv
[07:40:26] el_duerino (el_duerino!~afey@2001:4bd8:2:2:223:aeff:fe91:6669) has left #mythtv ()
[07:40:59] jya: holly cow... the issue wasn't mythtv code, but alsa default hardware buffer size
[07:41:14] jya: why didn't I think of that earlier ..
[07:45:05] jya: I had installed some patched alsa driver for the hd audio passthrough, which had reset my audio buffer to 64kB
[08:08:28] cesman (cesman!~cecil@pdpc/supporter/professional/cesman) has joined #mythtv
[08:11:07] eharris (eharris!~eharris@99-179-7-82.lightspeed.austtx.sbcglobal.net) has quit (Ping timeout: 240 seconds)
[08:18:42] eharris (eharris!~eharris@99-179-7-82.lightspeed.austtx.sbcglobal.net) has joined #mythtv
[08:25:27] xris: jya: http://svn.mythtv.org/
[08:42:46] jya: xris: very fancy html :P
[08:43:37] jya: could we have it starting on trac right away, and put the link to github and other resources on the trac frontpage ?
[08:44:50] pheld (pheld!~heldal@cl-5.osl-01.no.sixxs.net) has joined #mythtv
[08:44:58] eharris (eharris!~eharris@99-179-7-82.lightspeed.austtx.sbcglobal.net) has quit (Read error: Operation timed out)
[08:54:11] eharris (eharris!~eharris@99-179-7-82.lightspeed.austtx.sbcglobal.net) has joined #mythtv
[09:23:56] stoffel (stoffel!~quassel@p57B4A854.dip.t-dialin.net) has quit (Remote host closed the connection)
[09:27:16] kormoc is now known as kormoc_afk
[09:54:03] jamesba (jamesba!~jamesba@gateb.kw.bbc.co.uk) has joined #mythtv
[10:03:10] NightMonkey (NightMonkey!debian-tor@pdpc/supporter/professional/nightmonkey) has quit (Ping timeout: 245 seconds)
[10:08:44] stuartm_ is now known as stuartm
[10:11:05] cesman (cesman!~cecil@pdpc/supporter/professional/cesman) has quit (Read error: Connection reset by peer)
[10:11:06] deaman (deaman!~deaman@nat/trolltech/x-jtsgjqqidnxzchdy) has joined #mythtv
[10:26:43] cesman (cesman!~cecil@pdpc/supporter/professional/cesman) has joined #mythtv
[10:31:57] cesman (cesman!~cecil@pdpc/supporter/professional/cesman) has quit (Read error: Connection reset by peer)
[11:27:26] okolsi (okolsi!~mythtv@a88-112-74-247.elisa-laajakaista.fi) has quit (Quit: leaving)
[11:43:35] knightr (knightr!~knightr@69-165-170-178.dsl.teksavvy.com) has quit (Ping timeout: 245 seconds)
[11:44:21] knightr (knightr!~knightr@69-165-170-178.dsl.teksavvy.com) has joined #mythtv
[11:45:04] stuartm: first properly attributed patch committed using git :)
[11:47:36] stoffel (stoffel!~quassel@p57B4A854.dip.t-dialin.net) has joined #mythtv
[11:59:34] cesman (cesman!~cecil@pdpc/supporter/professional/cesman) has joined #mythtv
[12:10:59] cesman (cesman!~cecil@pdpc/supporter/professional/cesman) has quit (Read error: Connection reset by peer)
[12:19:36] jya: stuartm: Congrats !
[12:20:01] cocoa117 (cocoa117!~cocoa117@wk-29-208.guest.rdg.ac.uk) has joined #mythtv
[12:22:00] deaman (deaman!~deaman@nat/trolltech/x-jtsgjqqidnxzchdy) has quit (Quit: deaman)
[12:41:48] gigem_ (gigem_!~david@host137.12.intrusion.com) has joined #mythtv
[12:45:29] gigem (gigem!~david@host137.12.intrusion.com) has quit (Ping timeout: 240 seconds)
[12:52:08] gkeen316 (gkeen316!~gkeen316@d75-156-147-11.abhsia.telus.net) has joined #mythtv
[13:12:35] gkeen316 (gkeen316!~gkeen316@d75-156-147-11.abhsia.telus.net) has quit (Ping timeout: 272 seconds)
[13:20:12] MaverickTech (MaverickTech!~MaverickT@111.86.233.220.static.exetel.com.au) has joined #mythtv
[14:01:44] cesman (cesman!~cecil@pool-71-254-162-41.lsanca.fios.verizon.net) has joined #mythtv
[14:01:44] cesman (cesman!~cecil@pdpc/supporter/professional/cesman) has joined #mythtv
[14:01:44] cesman (cesman!~cecil@pool-71-254-162-41.lsanca.fios.verizon.net) has quit (Changing host)
[14:03:44] j-rod|afk is now known as j-rod
[14:08:49] cocoa117 (cocoa117!~cocoa117@wk-29-208.guest.rdg.ac.uk) has quit (Quit: Leaving)
[14:41:07] jya: janneg: are you planning to perform a ffmpeg resync soon ?
[14:43:21] jannau: yes
[14:43:41] jannau: soon as in this year
[14:48:58] danielk22: xris: perhaps some redirects ? trac.mythtv.org, fisheye.mythtv.org, jira.mythtv.org, svn.mythtv.org -> old code trac ?
[14:49:32] danielk22: PS I have no idea how I can browse the old code in trac right now
[15:04:40] jya: jannau: great. I want to import some changeset from ffmpeg related to SPDIF changes. I have imported the patch file. How cna I import them with git ?
[15:05:14] jya: like this one
[15:05:14] jya: http://git.ffmpeg.org/?p=ffmpeg;a=commitdiff_ . . . ea30c3738461
[15:05:33] jya: I tried git am -p1 file.patch
[15:05:45] jya: I get error: libavformat/Makefile: does not exist in index
[15:05:59] jya: never done a git am before. I'm a bit confused on how it works
[15:06:44] gigem_ is now known as gigem
[15:10:00] jannau: jya: you have to rewrte the patch, git am just accepts git format-patch output
[15:10:22] jannau: and that won't be compatible between ffmpeg and mythtv
[15:10:22] jya: that patch is a git format-patch I believe
[15:10:46] jya: i wanted to make things cleaner and easier to track by using identical commit as what they did in ffmpeg
[15:11:13] jannau: jya: please wait for the sync
[15:11:28] jya: so I have to do it by hand and apply the author and committer ?
[15:12:02] jya: I'm not doing this on the master don't worry.. I also have changes that aren't in ffmpeg yet
[15:26:12] danielk22: jya: The audio code should use VERBOSE(VB_AUDIO, LOC + "blah" [+ ENO]); style debugging. The debugging statements there and in the signal monitoring code dated from before we had settled on the LOC + X + ENO format.
[15:26:33] Kunalagon (Kunalagon!~Kunalagon@212.200.241.54) has joined #mythtv
[15:27:02] jya: ok... will be easy, all verbose comments in the audio code is a macro VBAUDIO, VBERROR etc..
[15:29:53] danielk22: yeah, I realize the existing debugging were mostly not of that format and it wasn't obvious what the format should be.. only excuse is that the audio code was old and very fragile so few dared touch it.
[15:32:18] danielk22: Beirdo: ^^^ I noticed you were putting strerror(errno) directly in verbose macros instead of using the ENO macro.
[15:33:40] cocoa117 (cocoa117!~cocoa117@wk-29-208.guest.rdg.ac.uk) has joined #mythtv
[15:33:44] jya: what's ENO ?
[15:34:40] danielk22: It's just a macro that calls strerror ... it prints both the human readable error and the errno, so it makes sense to the user and we can look up the errno if the human message is in russian.
[15:35:00] jya: ok.
[15:36:10] jya: i use almost no errno, most framework have their own error code, and a way to retrieve the value (like snd_strerror for alsa0)
[15:36:52] jya: danielk22: what's the format for LOC ?
[15:37:17] jya: currently, I have "ALSA: " for alsa, PulseAudio for pulse etc..
[15:38:32] danielk22: Generally the name of the class, optionally containing something that will tell you which instance it was. If the class name is long then something like "ALSA: " "Pulse: " etc is fine.
[15:39:39] Captain_Murdoch: stuartm, now you get to document how you did that attribution on the UsingGit page. :)
[15:39:41] danielk22: The idea is for any developer to be able to find the file where the error is being reported from, and in the case of multiple instances follow the trail of warnings before and after the error to figure out what happened.
[15:39:46] jya: currently, VBAUDIO macro is VERBOSE(VB_AUDIO, LOC + "message")
[15:39:56] jya: where LOC is "ALSA: " , etc...
[15:39:56] stuartm: danielk22: browsing trunk at particular point in time? If you go to the commit list you can follow the 'tree' link at the right to see the state of the tree at the point the commit was made – https://github.com/MythTV/mythtv/commits/ – the same link should be on the individual commit pages as well
[15:40:00] jya: isn't it good ?
[15:40:38] stuartm: to translate an SVN commit to it's Git equivalent just type the [{number}] in here, MythLogBot will post a link to the Git commit
[15:41:16] stuartm: e.g. \[12345\] becomes [12345]
[15:41:16] MythLogBot: SVN 12345: (branch master) https://github.com/MythTV/mythweb/commit/18d0b946
[15:41:49] stuartm: hmm, guess I didn't really need to escape those brackets in the demonstration :)
[15:42:17] danielk22: jya: VERBOSE(VB_AUDIO, LOC + "message") is fine. We just want the code itself to be standard and not contain extra macros for things we already have macros for.. It's so that any one of us can read the code easily without having to look up VBAUDIO to make sure it doesn't have a side effect such as setting a class instance error string.
[15:43:01] jya: So you want me to change all the VBAUDIO(blah) with VERBOSE(VB_AUDIO, LOC + ... ) ?
[15:43:39] jya: [45678]
[15:43:39] MythLogBot: No match for SVN revision 45678
[15:43:43] jya: cool :0
[15:43:44] danielk22: jya: yup, it's obviously a minor thing but these things add up.
[15:44:10] jya: danielk22: sigh... took me hours to change all the VERBOSE(VB_AUDIO) into VBAUDIO ..
[15:44:20] danielk22: stuartm: I am more concerned about browsing the branches that didn't make it to git.
[15:45:32] ** jya looking so much forward to reformat all the strings so they fit in 80 columns ... **
[15:46:09] danielk22: jya: I understand, I did the exact same thing when I started with the project. It took a while just to get everyone to agree on adding LOC to the standard invocation was acceptible.
[15:46:45] jya: but I am invoking LOC :)
[15:47:26] stuartm: danielk22: ah, I didn't realise we'd dropped branches in the transition
[15:49:23] danielk22: jya: I also had a strong preference for 90 columns, but I totally understand the need for some standard.
[15:49:42] jya: so can we make it 130 then ? :0
[15:49:52] jya: let's set a standard
[15:49:57] danielk22: hehe
[15:50:23] jya: or 110, that way I can fit exactly 2 emacs window on my laptp side by side
[15:50:24] danielk22: Isn't that the ffmpeg standard ? ;]
[15:50:41] jya: formatting to 80 columns has been such a pain
[15:51:00] danielk22: We wouldn't want to be accused of copyinhg, so 133 makes more sense ;]
[15:51:25] jya: yeah, but have just realised that 133 x 2 doesn't fit on my laptop unless I drop the font size
[15:52:57] jannau: ffmpeg uses 80 but not strictly
[15:53:02] danielk22: I have my font size adjusted so I can fit exactly three 80 column windows on each screen.
[15:53:57] danielk22: jannau: I always end up using one 240 column window per screen with ffmpeg.
[15:54:02] jya: the ffmpeg source I'm looking at right now, it's about 100
[15:54:28] jya: 163 characters this line
[15:55:19] j-rod is now known as j-rod|afk
[15:59:37] jya: right... time for bed... good night everyone
[16:01:22] gkeen316 (gkeen316!~gkeen316@207.228.113.179) has joined #mythtv
[16:02:49] j-rod|afk is now known as j-rod
[16:06:05] danielk22: stuarta: do you have any tasks for me wrt to the mythtv-setup stuff?
[16:06:52] stuarta: i want to do a design first. i should put out an email to -developers to solicit input
[16:08:05] stuarta: my feeling is that every action should be available through an interface, which our in backend setup leverages, but it also allows things like mythweb to be used to update configuration
[16:09:55] stuartm: we shouldn't get ahead of ourselves, the aim for now is just to replace mythtv-setup with the in-built webserver, the long term goals should be kept in mind but our focus should remain on the immediate tasks
[16:11:33] natanojl (natanojl!~jonatan@c83-252-237-63.bredband.comhem.se) has joined #mythtv
[16:11:42] stuarta: the first thing that does come to mind is that every configuration item that currently requires the backend to restart needs to be modified so that it only internally restarts
[16:11:44] stuartm: so getting the web-server improvements and framework into trunk should be a priority, it then needs to be hooked into danielk22's settings stuff (patch in trunk) and from there we can focus on smaller goals e.g. hardware setup/auto-detection, scanning, xmltv config
[16:12:05] stuartm: stuarta: aye, that as well
[16:13:31] stuarta: since, effectively, the backend will have to start with zero configuration
[16:14:08] [1]gkeen316 ([1]gkeen316!~gkeen316@199.16.181.210) has joined #mythtv
[16:15:51] [1]gkeen316 ([1]gkeen316!~gkeen316@199.16.181.210) has quit (Read error: Connection reset by peer)
[16:16:11] [1]gkeen316 ([1]gkeen316!~gkeen316@207.228.113.178) has joined #mythtv
[16:16:19] stuarta: stuartm: with the right design, making the configuation actions be done via a webservice, is no more difficult that doing what you suggest
[16:16:27] stuartm: someone needs to map out the overall layout/flow of setup but without getting bogged down in the minutiae e.g. on a brand new install first screen shown needs to prompt for language/locale, then where do we go? Does it follow a wizard format where the user is lead through the process? How should we break down the steps?
[16:16:32] danielk22: stuarta: The patch does address some of the bootstrap stuff. But I hadn't fully figured it out obviously. I anticipate there will be some bugs revealed when we do things like restart recorders which has never been done in the past even if the code theoretically supports it.
[16:17:03] stuarta: which patch are you talking about?
[16:17:12] gkeen316 (gkeen316!~gkeen316@207.228.113.179) has quit (Ping timeout: 265 seconds)
[16:17:23] stuarta: i wouldn't bother to restart recorders.
[16:17:58] danielk22: The one I sent you.. if it is not in there, I either already added it to trunk or it got munged out in some merge.
[16:17:59] stuarta: let them continue what they are doing, only newly started up recorders would start with the updated config. far simper
[16:18:15] stuarta: oh, i haven't looked at email today. been in bed ill
[16:19:30] stuarta: i'll get to it tonight once little man has gone to bed
[16:19:31] danielk22: stuarta: sent it several weeks ago.. hmm, I can dig it up, but it's on the hard drive of the laptop that went up in flames. (Lenovo still hasn't even mailed the replacement and it's been 8 days already..grr)
[16:19:52] stuarta: oh right, i'll have a look
[16:24:25] stuarta: hmmm, no in my inbox, will have a look in mythtv inbox
[16:32:30] gkeen316 (gkeen316!~gkeen316@199.16.181.210) has joined #mythtv
[16:34:02] [2]gkeen316 ([2]gkeen316!~gkeen316@199.16.181.210) has joined #mythtv
[16:34:52] [1]gkeen316 ([1]gkeen316!~gkeen316@207.228.113.178) has quit (Ping timeout: 276 seconds)
[16:36:54] stuarta: danielk22: no idea where it is
[16:36:55] gkeen316 (gkeen316!~gkeen316@199.16.181.210) has quit (Ping timeout: 245 seconds)
[16:49:27] stuartm: it's not the one attached to this ticket? http://code.mythtv.org/trac/ticket/8262
[16:50:09] danielk22: stuartm: hehe, it is.. I had forgotten about my "insurance policy" ;)
[16:50:23] [2]gkeen316 ([2]gkeen316!~gkeen316@199.16.181.210) has quit (Read error: Connection reset by peer)
[17:12:12] stuartm: huh, how come my repo has lots of local changes that I didn't make?
[17:12:47] stuartm: schemawizard.cpp, audiooutputalsa.cpp, tv_play.cpp, Terra/menu-ui.xml
[17:13:48] stuartm: hmm, think the g-n-w has messed up somehow
[17:21:17] okolsi (okolsi!~mythtv@unaffiliated/okolsi) has joined #mythtv
[17:25:10] danielk22: stuartm: g-n-w will not update the files if you have updated them in another working directory. You need to do a "git reset --hard" to get the updates (but this will also wipe out any local changes).
[17:26:16] danielk22: stuartm: I believe this is what caused my merge + merge revert commit yesterday. I did a git merge in one working dir and a git commit in a different working dir.
[17:27:16] danielk22: I still have no idea why it didn't auto-commit after the merge.
[17:40:17] gigem (gigem!~david@host137.12.intrusion.com) has quit (Remote host closed the connection)
[17:40:43] gigem (gigem!~david@host137.12.intrusion.com) has joined #mythtv
[17:42:53] jamesba (jamesba!~jamesba@gateb.kw.bbc.co.uk) has quit (Remote host closed the connection)
[17:50:02] cocoa117 (cocoa117!~cocoa117@wk-29-208.guest.rdg.ac.uk) has quit (Quit: Leaving)
[17:50:53] sphery: danielk22: I still think Captain_Murdoch's idea of splitting up mythbackend with/before the backend settings stuff would make things easier. Make a mythtv binary that a) provides the webserver interface, and b) runs the appropriate binaries, as configured for the current host; then it can stop/start mythbackend (which does capture), or mythjobqueue as necessary. Also avoids the whole problem of making current mythbackend start up ...
[17:50:59] sphery: ... without any capture cards defined/configured. And I don't think it would be too hard to get done fast.
[17:53:32] danielk22: sphery: stuarta: One of the things I had hoped to get out of this was to be able to reconfigure one recorder and have it autorestarted without affecting ongoing recordings.. obviously not a feature needed by many, but very useful to a developer.
[17:54:19] danielk22: Also it would mean a faster restart for regular users :)
[17:55:02] sphery: ah, yeah, that would either require having multiple mythrecorder processes (one per recording) or modifying the mythbackend I proposed to allow it to reload part of its configuration
[17:55:30] sphery: the one process per recording would have the side effect of "fixing" Udo's "mythbackend still eats memory" issue :)
[17:56:22] stuartm: danielk22: I think it's useful to the average user, I can't begin to count the number of times when because of the DVB nodes moving around on a power cycle I'd find myself hours later with in-progress recordings on one tuner and others failing on another – the only fix was to re-run mythtv-setup to fix the device ordering and restart the backend, but doing that meant interrupting/stopping the unaffected recordings
[17:57:26] danielk22: I hope you mean one per 'recorder' not 'recording' I'm adding back to back recordings in the recording branch soon for non-livetv recordings (already there for livetv obviously). Which wouldn't help Udo at all.
[17:57:27] stuartm: sphery: "fixes" like that are dubious at best :)
[17:58:13] danielk22: I don't think anything would satisfy Udo anyway...
[17:58:41] stuartm: and what effect would one process per-recorder plus supporting processes such as mythbackend have on our overall memory/processor requirements?
[17:59:16] sphery: recorder or recording... either would work :)
[17:59:46] danielk22: stuartm: well yeah, that happens to me too.. I have quick release plugs on all the hw so I can just switch the cables when the hardware devices switch around on reboot. So I have a HW solution for that problem :P
[18:00:01] sphery: I don't think it would change memory requirements much at all since we use a lot of (primarily?) shared libs
[18:00:26] duerF (duerF!~tommi@pat.ok.is) has joined #mythtv
[18:00:59] danielk22: stuartm: It would increase it a bit, but so long as virtual memory is enabled it would be pretty minimal.
[18:04:50] sphery: stuartm: oh, and FWIW, splitting our current mythbackend process would also allow us to run /only/ the required portion on a given host. For example, running mythjobqueue is about 1/10 the memory usage of mythbackend--see http://www.gossamer-threads.com/lists/mythtv/users/347829#347829
[18:05:19] danielk22: Under linux each thread is a process anyway.. It just happens to share some memory with other processes without needing explicit calls to do so.
[18:06:19] danielk22: sphery: I think that is really the biggest arguement for it, not that it simplifies the mythtv-setup code.
[18:07:01] sphery: yeah, definitely a big advantage
[18:11:15] kormoc_afk is now known as kormoc
[18:13:05] stuarta: sphery: i like that idea of splitting out to a simple master backend, and all the recorders are slaves etc...
[18:13:18] Beirdo: danielk22: I have them in particular in myth_system in the forking area. They are needed there, but in other places, probably should be changed to ENO
[18:13:55] wagnerrp: would that be each recorder gets its own process, spawned as needed?
[18:14:58] sphery: sounds like that's the approach that would work best for danielk if we decide to go down that route... even if not, splitting mythbackend into mythtv/mythbackend(or mythrecorder)/mythjobqueue/... makes sense
[18:15:10] danielk22: Beirdo: yeah, obviously in the critical area of the forking code we want to stay away from QString.. do we have some comment to that effect there ("Danger Will Robinson!") ?
[18:15:56] Beirdo: I believe so
[18:16:10] Beirdo: if not, I'll be sure we add even fatter warnings in the new code :)
[18:16:11] stuarta: i'd lean toward 1 recorder
[18:16:18] stuarta: per hardware device
[18:16:26] stuarta: simplifies multirec handling etc
[18:16:35] Beirdo: the other places would have me forgetting about ENO ;)
[18:19:00] wagnerrp (wagnerrp!~wagnerrp_@NR-FT1-66-42-240-159.fuse.net) has quit (Quit: Leaving)
[18:19:53] Beirdo: hey. Did the gihub/trac integration actually WORK on #9173?
[18:20:05] wagnerrp (wagnerrp!~wagnerrp_@NR-FT1-66-42-240-159.fuse.net) has joined #mythtv
[18:20:07] Beirdo: or did stuartm add that by hand?
[18:21:26] wagnerrp: in the near term, i would like to get mythjobqueue cut out of the backend and into a separate process
[18:22:05] kormoc: wagnerrp, it is, no?
[18:22:16] wagnerrp: i mean remove the jobqueue thread from mythbackend
[18:22:26] wagnerrp: require the jobqueue to be run as a separate process
[18:22:44] wagnerrp: and put a proper job scheduler in the master backend
[18:22:57] wagnerrp: instead of the opportunistic setup its currently using
[18:23:09] kormoc: I thought the backend just shelled out and launched mythjobqueue by hand?
[18:23:28] Beirdo: for each job, I think
[18:23:37] wagnerrp: no, mythjobqueue is just an alternate queue
[18:23:38] Beirdo: rather than have the queue managed by an external process
[18:23:46] wagnerrp: if you dont want to run the backend which includes a queue
[18:30:03] Captain_Murdoch: and by 'queue', you mean "virtual free-for-all by which whichever backend happens to check next gets to run the job"  :)
[18:30:19] Captain_Murdoch: and that's my code, so it's ok for me to say. :)
[18:32:39] Beirdo: hehe
[18:32:59] wagnerrp: right, thats what i meant by put a proper scheduler in the master backend
[18:33:49] wagnerrp: it will be a lot easier to track which clients can run a jobqueue if all they do is run a jobqueue
[18:41:13] duerF (duerF!~tommi@pat.ok.is) has quit (Quit: Leaving)
[18:41:20] stuartm: Beirdo: by hand, just followed the expected format because it's easy to parse
[18:44:23] stuartm: Beirdo: are we clear whether it's the github hook which fails or the trac plugin?
[18:50:45] Beirdo: it is the trac plugin
[18:51:18] Beirdo: the logs show the hook hits it (the trac logs), but it doesn't seem to parse the messages correctly or soemthing
[18:53:11] Beirdo: I can look at it over the weekend, I guess. My python knowledge ain't the strongest, but I should be able to find my way through it to see what its issue is
[18:55:36] stuartm: we're running the latest version of the plugin?
[18:55:59] Beirdo: the latest that was available at the time, yes
[18:59:50] stuartm: hmm, the git repo for the plugin last saw activity in Feb 09
[19:00:46] Beirdo: yeah
[19:00:54] Beirdo: we can fork their repo and fix it
[19:01:04] Beirdo: anyways, I got a meeting, be back in a bit
[19:02:58] stuartm: are we using FastCGI?
[19:03:09] stuartm: there is a patch in their pull requests to fix that
[19:09:42] Steve_Goodey (Steve_Goodey!~steve@host86-145-237-137.range86-145.btcentralplus.com) has joined #mythtv
[19:29:06] danielk22: markk_: Just FYI Your latest changes in video playback have made the OSD UI _much_ more responsive here.
[19:47:19] stoffel (stoffel!~quassel@p57B4A854.dip.t-dialin.net) has quit (Remote host closed the connection)
[19:52:40] Beirdo: nice
[19:52:54] Beirdo: stuartm: no, we are using WSGI for trac
[19:53:06] Beirdo: I think that's the acronym :)
[19:54:32] stuartm: Beirdo: ah, these patches might be worth a look anyway, they are pretty recent – https://github.com/davglass/github-trac/pull/4
[19:55:15] Beirdo: yeah, likely good to look at
[19:57:44] ** Beirdo makes a list for the weekend... github plugin... buildbot (hopefully) **
[19:57:57] gigem (gigem!~david@host137.12.intrusion.com) has quit (Read error: Connection reset by peer)
[19:58:29] Beirdo: ... splitting -commits into 3 MLs.
[19:58:36] Beirdo: sounds like a fun weekend
[19:59:40] Beirdo: if any of you guys use the WOL and mythshutdown/mythwelcome stuff, I wouldn't mind if you could give the functionality a quick test on the mythsystem-rewrite branch
[20:00:04] Beirdo: those are the only remaining new changes I couldn't test yet, (other than Windows)
[20:00:18] Beirdo: anyways, lunch. Wow, today's flying
[20:00:18] gigem (gigem!~david@host137.12.intrusion.com) has joined #mythtv
[20:00:46] wagnerrp: Beirdo: ive been running it for about a week now, and the only problem so far was that mythtv-setup issue
[20:04:00] stuartm: is there a simple way to get the diff for a particular commit?
[20:04:23] wagnerrp: on github or the command line?
[20:05:14] stuartm: command line, local commit which hasn't been pushed
[20:05:24] wagnerrp: git show <hash>
[20:05:30] stuartm: thanks
[20:06:47] purserj (purserj!~purserj@hosting.collaborynth.com.au) has quit (Ping timeout: 276 seconds)
[20:12:27] stuartm: markk_: I'd just like to second what danielk22 said, it's much more responsive
[20:14:54] stuartm: and I might be imagining it because I don't use it very often, but livetv seems faster and more stable
[20:19:36] iamlindoro: There's circumstantial evidence that LiveTV stability has taken a serious nosedive in .24.... hope that it did help but suspect there are a lot of nagging problems left
[20:20:02] iamlindoro: The "Error Opening Jump Program file Buffer" stuff that every third user seems to see is really a black mark, it'd be nice to get to the bottom of why so many people see it
[20:24:17] danielk22: iamlindoro: I started looking at it. It looks like the new threading model for the video player blocks some updates so if there isn't at least 2048KB of data in the recording file when we probe the file we get stuck in an infinitine loop.
[20:26:11] stuarta: i get the audio starting before the video these days
[20:26:14] iamlindoro: danielk22: Is that for the issue where the frontend needs killing, or the timeout/failure back to the menu?
[20:26:32] iamlindoro: I see a lot more complaints of the latter, but I've heard of the former too
[20:26:36] danielk22: It should eventually fail back to the menu, but it can take a minute.
[20:26:43] iamlindoro: Ah, yeah, sounds about right then
[20:26:54] danielk22: I've seen both. But the latter is much more common.
[20:40:14] NightMonkey (NightMonkey!debian-tor@pdpc/supporter/professional/nightmonkey) has joined #mythtv
[20:49:05] PointyPumper (PointyPumper!~pintlezz@190.244.73.13) has joined #mythtv
[20:58:20] Steve_Goodey (Steve_Goodey!~steve@host86-145-237-137.range86-145.btcentralplus.com) has quit (Remote host closed the connection)
[20:59:05] jarle (jarle!~jarle@70.84-234-133.customer.lyse.net) has quit (Remote host closed the connection)
[20:59:58] purserj (purserj!~purserj@hosting.collaborynth.com.au) has joined #mythtv
[21:07:49] purserj (purserj!~purserj@hosting.collaborynth.com.au) has quit (Read error: Operation timed out)
[21:15:34] purserj (purserj!~purserj@hosting.collaborynth.com.au) has joined #mythtv
[21:16:11] Steve_Goodey (Steve_Goodey!~steve@host86-145-237-137.range86-145.btcentralplus.com) has joined #mythtv
[21:19:24] tomimo (tomimo!~kurre@xdsl-83-150-88-111.nebulazone.fi) has quit (Quit: Ex-Chat)
[21:23:05] wagnerrp (wagnerrp!~wagnerrp_@NR-FT1-66-42-240-159.fuse.net) has quit (Quit: Leaving)
[21:26:04] stuarta: anyone else getting a link error "undefined reference to `bd_mouse_select(bluray*, long, unsigned short, unsigned short)'" ???
[21:26:30] stuarta: i've distclean, and blown away the target dir
[21:27:36] stuarta: linking against libs/libmythtv/libmythtv-0.24.so
[21:29:09] Steve_Goodey (Steve_Goodey!~steve@host86-145-237-137.range86-145.btcentralplus.com) has quit (Remote host closed the connection)
[21:31:40] stuartm: that's the code iamlindoro just committed
[21:32:32] iamlindoro: bah
[21:32:38] iamlindoro: it ended up ourside the extern
[21:32:41] iamlindoro: I'm sorry
[21:32:45] ** stuarta sp0rks iamlindoro **
[21:33:19] ** stuarta watches tv whilst twiddling thumbs **
[21:33:37] iamlindoro: stuarta, stuartm: Fixed
[21:33:41] iamlindoro: sorry about that
[21:33:51] iamlindoro: It compiled here but I guess I didn't distclean/not sure why it worked
[21:34:10] stuarta: clearly didn't rebuild the library
[21:34:30] iamlindoro: Well the code that references it built fine, so it must have :)
[21:34:41] stuarta: building now...
[21:34:47] iamlindoro: or... well... anyway, it should be fine ;)
[21:45:20] stuarta: hmpf. plugins won't build
[21:46:37] stuarta: hmmm, libmythfreemheg, libmythlivemedia, libmythhdhomerun libraries all not found
[21:47:26] stuarta: when linking mytharchivehelper
[21:54:41] elmojo: markk_: sorry about the delay but I'll try to review that patch later tonight
[21:59:14] stuartm: stuarta: it shouldn't be linking any of those
[22:00:53] stuarta: well libmythtv seems to want them
[22:06:46] jurco (jurco!5f678264@gateway/web/freenode/ip.95.103.130.100) has joined #mythtv
[22:09:07] jurco: Hi guys. I would like to ask someone to help me to configure MythTV to run properly. I installed it day ago, finally I found some TV channels. But when I run frontend and I choose Watch TV, I get wait please... message and then nothing – back to main menu. How do I find out where the problem is?
[22:09:25] stuarta: step 1. read the topic
[22:09:52] jurco: ok, sorry. is there other mythtv channel?
[22:10:14] stuarta: step 2. read the topic again
[22:10:40] jurco: thanks ;)
[22:11:07] jurco (jurco!5f678264@gateway/web/freenode/ip.95.103.130.100) has left #mythtv ()
[22:12:02] ChanServ!ChanServ@services. changes topic to MythTV development channel. For user support, please /join #mythtv-users. Use http://mythtv.pastebin.ca/ as pastebin. This channel is logged.
[22:13:01] stuarta: eh?
[22:13:25] stuartm: stuarta: just trimming it down a bit so that it's less likely to get cutdown by some IRC clients
[22:13:46] stuarta: ah, i was wondering why chaserv was changing the topic
[22:14:19] stuartm: changed it via chanserv so that it no longer includes my hostname/ip
[22:14:48] stuartm: instead of "*** The topic was set by stuartm!~gbee@cpc1-derb9-0-0-cust743.8-3.cable.virginmedia.com on 02/11/2010 17:45." it now reads "*** The topic was set by stuartm on 10/12/2010 22:12."
[22:15:18] stuarta: make sense
[22:15:28] stuarta: oh for our project cloak
[22:16:00] ** stuarta coughs and says oww lots **
[22:36:23] gregL (gregL!~greg@cpe-74-76-125-87.nycap.res.rr.com) has quit (Ping timeout: 255 seconds)
[22:37:28] Dave123-road (Dave123-road!~dave@74.74.222.96) has quit (Quit: Leaving)
[22:39:46] gregL (gregL!~greg@cpe-74-76-125-87.nycap.res.rr.com) has joined #mythtv
[22:41:47] Dave123-road (Dave123-road!~dave@74.74.222.96) has joined #mythtv
[22:43:11] jpharvey_ (jpharvey_!~jpharvey@host86-184-21-188.range86-184.btcentralplus.com) has joined #mythtv
[22:45:43] Dave123 (Dave123!~dave@cpe-74-74-222-96.rochester.res.rr.com) has quit (Remote host closed the connection)
[22:46:30] jpharvey (jpharvey!~jpharvey@host86-185-253-248.range86-185.btcentralplus.com) has quit (Ping timeout: 245 seconds)
[22:48:40] Dave123 (Dave123!~dave@cpe-74-74-222-96.rochester.res.rr.com) has joined #mythtv
[22:50:49] Dave123 (Dave123!~dave@cpe-74-74-222-96.rochester.res.rr.com) has quit (Client Quit)
[22:50:56] Dave123 (Dave123!~dave@cpe-74-74-222-96.rochester.res.rr.com) has joined #mythtv
[22:51:08] RDV_Linux (RDV_Linux!~doug@CPE001195554bb4-CM00252eac6f40.cpe.net.cable.rogers.com) has quit (Ping timeout: 255 seconds)
[22:53:54] Dave123 (Dave123!~dave@cpe-74-74-222-96.rochester.res.rr.com) has quit (Client Quit)
[22:55:29] Dave123 (Dave123!~dave@cpe-74-74-222-96.rochester.res.rr.com) has joined #mythtv
[22:55:57] j-rod is now known as j-rod|afk
[23:05:28] justpaul (justpaul!alexdelarg@poizon.epicshells.com) has quit (Read error: Connection reset by peer)
[23:05:32] justpaul (justpaul!alexdelarg@poizon.epicshells.com) has joined #mythtv
[23:43:22] neolive_ (neolive_!~chatzilla@41.27.46.205) has joined #mythtv
[23:44:02] neolive_: hi all
[23:44:06] Kunalagon (Kunalagon!~Kunalagon@212.200.241.54) has quit (Quit: Leaving.)
[23:44:42] neolive_: is anyone here interested in broadcasting?
[23:45:24] neolive_ is now known as neolive
[23:45:26] kormoc: neolive_, not on topic for this channel
[23:56:55] MavT (MavT!~MaverickT@111.86.233.220.static.exetel.com.au) has joined #mythtv
[23:58:17] MaverickTech (MaverickTech!~MaverickT@111.86.233.220.static.exetel.com.au) has quit (Ping timeout: 264 seconds)

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