Back to Home / #mythtv / 2008 / 03 / Prev Day | Next Day
#mythtv IRC Logs for 2008-03-25

---Logopened Tue Mar 25 00:00:36 2008
00:07-!-jaek [n=jaek@] has joined #mythtv
00:07-!-jaek [n=jaek@] has left #mythtv ["Leaving"]
00:22-!-Cardoe [n=Cardoe@gentoo/developer/Cardoe] has quit ["Leaving"]
00:38-!-gnome42 [] has quit [Remote closed the connection]
01:55-!-clever [] has quit [Connection timed out]
02:01-!-clever [] has joined #mythtv
02:02-!-BathoryQuorthon_ [] has joined #mythtv
02:03-!-BathoryQuorthon_ [] has quit [SendQ exceeded]
02:06-!-BathoryQuorthon [] has quit [Nick collision from services.]
02:08-!-BathoryQuorthon [] has joined #mythtv
02:42-!-dekar1 [] has quit [Read error: 110 (Connection timed out)]
02:59-!-xris [] has quit []
03:04-!-nomopofomo [] has joined #mythtv
03:04-!-nomopofomo [] has quit [Client Quit]
03:08-!-nomopofomo [] has joined #mythtv
03:08-!-nomopofomo [] has left #mythtv ["Ex-Chat"]
03:37-!-lsobral_ [n=sobral@] has joined #mythtv
03:38-!-lsobral [n=sobral@] has quit [Read error: 110 (Connection timed out)]
03:58-!-NightMonkey [n=NightMon@pdpc/supporter/sustaining/NightMonkey] has quit [Client Quit]
04:27-!-otwin_ [n=otwin@] has joined #mythtv
04:36-!-Raspberry [] has joined #mythtv
04:37-!-Raspberry [] has left #mythtv ["Konversation terminated!"]
04:41-!-otwin [n=otwin@] has quit [Read error: 110 (Connection timed out)]
04:43-!-justinh [] has joined #mythtv
04:43-!-joobie [] has joined #mythtv
04:45-!-otwin_ is now known as otwin
05:11-!-kurre2__2 [] has quit [Remote closed the connection]
05:22-!-tomimo [] has quit [Read error: 110 (Connection timed out)]
05:35-!-tomimo [] has joined #mythtv
05:38-!-kurre2 [] has joined #mythtv
05:48-!-melunko [n=hmelo@] has quit [Read error: 110 (Connection timed out)]
05:55-!-melunko [n=hmelo@] has joined #mythtv
06:11-!-joobie [] has quit ["This computer has gone to sleep"]
06:17-!-joobie [] has joined #mythtv
06:19-!-DaveMorris [] has joined #mythtv
06:27-!-CatCow [] has left #mythtv ["[IRSSI] If you bought it, a truck brought it."]
06:32-!-gardz [] has joined #mythtv
06:52-!-jhulst [n=jhulst@unaffiliated/jhulst] has quit ["Konversation terminated!"]
06:57-!-MrGandalf [] has quit ["work"]
07:00-!-dekarl [] has joined #mythtv
07:41-!-zorglu_ [] has joined #mythtv
07:48<janneg> qt4 branch running (well sort of) on a nokia n810
07:50-!-jk1joel [] has quit [Read error: 113 (No route to host)]
08:00<janneg>gbee: Were your two failed recordings scheduled to start at the same time?
08:01<gbee>janneg: think so, let me just check
08:02<gbee>yes, both started at 22:00
08:02-!-reynaldo [] has quit [Read error: 113 (No route to host)]
08:03<gbee>different cards though
08:04<janneg>my failed recordings started both at 20:00
08:04<gbee>sorry mis-read, both recordings were on the same physical card
08:04<janneg>hmm, there goes my theory of multirec being racy
08:04<janneg>and I type to slow
08:05<gbee>there was a third recording on the second card and I got the log entries mixed up
08:07<janneg>no problem, And I'm relieved, recordings blocking each other on different cards would be more severe and is harder to debug and
08:07-!-jk1joel [] has joined #mythtv
08:18-!-joobie [] has quit ["Leaving"]
08:34<janneg>gbee: testing now for qt4 qmake in mythplugins/configure
08:36<gbee>janneg: thanks
08:38<janneg>that was the easy to fix issue. now for multirec debugging
08:45-!-PointyPumper [n=pintlezz@] has quit [Read error: 104 (Connection reset by peer)]
08:56-!-splAt1 is now known as splat1
09:02-!-beavis [] has joined #mythtv
09:22-!-Cardoe [n=Cardoe@gentoo/developer/Cardoe] has joined #mythtv
09:28-!-gardz [] has quit []
09:29-!-gardz [] has joined #mythtv
09:41-!-Cardoe [n=Cardoe@gentoo/developer/Cardoe] has quit ["Leaving"]
09:48-!-Cardoe [n=Cardoe@gentoo/developer/Cardoe] has joined #mythtv
09:55-!-jmk_ [n=jmk@] has joined #mythtv
10:03<janneg>gah, failed recordings not reproduceable under gdb with appropiate logging
10:03<janneg>7x2 successfull recordings in the last hour
10:03-!-cattelan [] has quit ["This computer has gone to sleep"]
10:05-!-zorglu_ [] has left #mythtv ["Konversation terminated!"]
10:20-!-nordenm_ [] has joined #mythtv
10:26<laga>danielk22: i have seen people reporting 'odd' colours on ati radeon cards during video playback, eg people looked like smurfs. the issue went away when they adjusted huet to about 50%.. i've verified that their Xv adapter name still matches the one in calc_hue_base() (libs/libmythtv/videoout_xv.cpp), maybe it was fixed in the driver?
10:29<danielk22>laga: calc_hue_base() doesn't fix a bug; there are many valid 'no change' hue values that are valid. Are we sure the result of calc_hue_base() is still being applied?
10:30-!-nordenm [] has quit [Read error: 110 (Connection timed out)]
10:30<GreyFoxx>Do we do anything differently when first launching the backend when it comes to accessing the firewire devices than we do for later recordings?
10:31<GreyFoxx>Lately I've gotten a lot of instances where the busreset seems to fail, but stop/start mythbackend and it works and picks up recording
10:33<laga>danielk22: no, i'll take a look
10:33<GreyFoxx>After the first recording, we do a bus reset which returns chan 63, get no data and call busreset 3 more times and each gets chan -1.
10:33<GreyFoxx>If I stop and start the backend it just "works"
10:36-!-Nikas [] has joined #mythtv
10:37<laga>danielk22: i'm not quite sure how calc_hue_base() is supposed to work.. will it set the initial hue value to 50%? because for the guy who had that problem, the hue slider was at 0
10:39-!-grokky [] has quit []
10:41<danielk22>laga: It should set it to 0% on nvidia cards and 50% on most others.
10:44<danielk22>greyfoxx: it reads we are failing to recover from a bus reset. These happen whenever a device on the bus is power cycled, a device is plugged in or unplugged, or when the recorder isn't getting any data. Possibly the firewire drivers have changed and we need to adjust...
10:45<GreyFoxx>Ok. In my case there is no other firewire device, the STB hasn't been power cycled, (but 4 days ago the backend was)
10:46<GreyFoxx>but over the last several days to a week I've had more and more failed firewire recordings
10:47<GreyFoxx>if I restart the backend in the middle of a failed recording it records the rest of the show happily
10:47-!-PointyPumper [i=Pintlezz@] has joined #mythtv
10:49<GreyFoxx>usually failures coincide with a need to change the channel
10:49<GreyFoxx>so if I record 3 shows in a row on channel 600 they are fine
10:49<GreyFoxx>if a change is required then it seems to be more likely to fail
10:50<GreyFoxx>but this is really just annectotal observations at this point
10:52<laga>danielk22: i can actually reproduce the problem with my radeon card, so i'll try to debug it..
10:52<janneg>anything left before merging qt4 back to trunk?
10:53-!-cattelan [] has joined #mythtv
10:54<janneg>I wasn't able to reproduce the failed simultaneous multirec recordings in 10 tries
10:57-!-th1 [] has joined #mythtv
10:57-!-jgarvey [] has joined #mythtv
11:06<danielk22>thx laga
11:07<danielk22>greyfoxx: Have you tried disabling the auto resets? (I think the setting is in mythtv-setup).
11:08<GreyFoxx>I didn't know that was an option
11:08<GreyFoxx>I'll try that
11:08<GreyFoxx>I'm testing several recordings in arow right now
11:09<GreyFoxx>so far 4 in arow on the same channel have worked, next I'll try one that causes a channel change first
11:12<janneg>bah, svn merge is plain stupid. merging back to trunk causes tons of conflicts
11:14<janneg>svn status | grep "^C" | wc -l
11:15-!-th1 [] has quit [Read error: 110 (Connection timed out)]
11:16<janneg>does anything speaks against applying svn diff ...trunk/mythtv ...branches/mythtv-qt4 instead of a proper merge?
11:16<janneg>Chutt: ^^^
11:17<janneg>danielk22: how did you merged multirec back to trunk?
11:18<danielk22>janne don't recall.. usually svn merge, unless there were problems, then I might have used a diff
11:18<danielk22>using a diff means you have to watch out for adds and removes.
11:19<danielk22>removes will just blank the file after a diff, not actually remove it. adds need to be manually added.
11:19<Cardoe>you could also use
11:19<danielk22>Probably not applicable here, but svn properties are not carried over by a diff
11:19<Cardoe>which is a user-side implementation of some of the merge features that SVN 1.5 will offer
11:19<Cardoe>if things don't merge right
11:20<Cardoe>basically you'll get 3-way merge help
11:20<gbee>danielk22: re adds? Really? I've been using svn diff > foo.diff and patch -p0 < foo.diff to add new files to another checkout for a while
11:21<danielk22>gbee: you need to do an "svn add" on the new files.. they are there, svn just doesn't know about them.
11:21<janneg>gbee: you have to remember to do svn rm and svn add on the files
11:21<gbee>doh, yeah, sorry
11:21<danielk22>the diff approach also loses any svn move history, but again probably not applicable here.
11:22<janneg>I think I go the diff patch approach. I don't think I've motivated enough to resolve conflicts in 140 files
11:23-!-skamithi [] has joined #mythtv
11:24<gbee>janneg: are you going to try and fix the multirec bug before merging back to trunk?
11:26<janneg>gbee: no, I couldn't reproduce it with -v channel,record under gdb
11:27<janneg>100% success in 10 attempts
11:28<gbee>ok, maybe we should post a list of known bugs/issues to the -dev list so we hopefully don't get multiple reports of things we already know about? Just a thought, but I suspect the 3 or more warnings already posted to the dev list might not be enough ;)
11:29<gbee>there are already people posting to the -user list about themes no longer working with trunk, despite my warnings about mythui
11:30<janneg>I had planned to mention them in the commit message but I could also write another announcement
11:31<janneg>grr, I want another vcs. svn diff trunk mythtv-qt4 takes ages
11:32<gbee>don't make me learn another vcs ;)
11:33<janneg>428 files changed, 5197 insertions(+), 3615 deletions(-)
11:34<stuarta>i sense pain in the future
11:57-!-gnome42 [] has joined #mythtv
12:12<Chutt>janneg, as long as you're sure nothing will get removed by doing a diff..
12:13-!-Nikas [] has quit ["( :: NoNameScript 4.2 :: )"]
12:13<janneg>Chutt: sure. I'm verifying the diff atm
12:15<laga>danielk22: right, xv_hue_base isn't applied anymore.. i'll supply a patch later
12:16<laga>danielk22: if my assumption is correct, it was broken 5 months ago. i wonder why nobody has complained since then
12:17*janneg guesses nobody uses ati cards with mythtv
12:17*stuarta shouts me!
12:18<stuarta>however i don't use their drivers
12:20<laga>stuarta: it happens with the free drivers
12:21<laga>janneg: you only see it when you enable "xv picture controls"
12:23*stuarta suspects he hasn't enabled "xv picture controls"
12:25<gbee>odd, recording made last night on my parents box isn't there today, no log messages suggesting it auto-expired and there is a transcode log entry showing it was definately recorded and not zero-byte (3.9Gb) but no sign of it
12:26<gbee>think I need to enable the deleted group stuff for them, because I suspect they accidently deleted it and just don't remember doing it :)
12:27-!-xris [] has joined #mythtv
12:32-!-onixian [n=xian@] has joined #mythtv
12:32-!-onixian [n=xian@] has quit [Remote closed the connection]
12:35-!-lcase [] has joined #mythtv
12:36<janneg>yeah, aac decoder patch posted for review on ffmpeg-dev
12:48<GreyFoxx>danielk22: It looks like my lockups are related to changing channels, maybe not waiting long enough for the channel change to occur on the STB?
12:49<GreyFoxx>I have no failures recording if I don't change channels, but do get failues if a channel change is involved
12:49<GreyFoxx>It's as if we aren't waiting long enough for data to arrive after issuing the channel change before assuming we need to reset the bus
12:49<GreyFoxx>I'll try it with turning off the bus reset option
12:54<danielk22>gf: k, maybe we need to up the timeout, or make it configurable..
13:02-!-JoeBorn [] has quit [Read error: 110 (Connection timed out)]
13:10-!-jmk_ [n=jmk@] has quit ["Leaving"]
13:10-!-nwahmaet [] has joined #mythtv
13:11-!-DaveMorris [] has quit ["Leaving."]
13:20<sphery>What it says: If you choose to submit a bug report, please make sure to include a brief description of what you were doing, along with the following backtrace as an attachment (please don't paste the whole thing into the ticket).
13:20<sphery>What the user sees (ref #5035, #5026, #5018...): ... submit a bug report ... paste the whole thing
13:28-!-xri1 [] has joined #mythtv
13:28-!-xris [] has quit [Read error: 104 (Connection reset by peer)]
13:36-!-xris [] has joined #mythtv
13:37-!-bio_ [] has quit [Read error: 110 (Connection timed out)]
13:38-!-xri1 [] has quit [Read error: 104 (Connection reset by peer)]
13:39<justinh>the folks who are complaining about themes no longer working can eat cake :)
13:39<justinh>or they can just wait for a while til I get my head around what needs to change & I can help them
13:39<justinh>as in... give enough hints for fixes to emerge
13:43-!-melunko [n=hmelo@] has quit [Read error: 110 (Connection timed out)]
13:43-!-renatofilho^_ [n=renato@] has quit [Read error: 110 (Connection timed out)]
13:44-!-feiner [] has quit [Read error: 104 (Connection reset by peer)]
13:44-!-renatofilho^_ [n=renato@] has joined #mythtv
13:45-!-melunko [n=hmelo@] has joined #mythtv
13:45-!-lsobral [n=sobral@] has joined #mythtv
13:45-!-lsobral_ [n=sobral@] has quit [Read error: 110 (Connection timed out)]
13:59-!-xri1 [] has joined #mythtv
13:59-!-xris [] has quit [Read error: 104 (Connection reset by peer)]
14:01-!-feiner [] has joined #mythtv
14:04-!-jmk_ [n=jmk@] has joined #mythtv
14:05<GreyFoxx>turning off the auto bus reset seems to have made it better.
14:05-!-cattelan [] has quit ["Leaving"]
14:15-!-lcase [] has quit [Remote closed the connection]
14:18-!-xri1 [] has quit []
14:23-!-xris [] has joined #mythtv
14:23-!-keithclark [] has joined #mythtv
14:24-!-keithclark [] has left #mythtv ["Leaving"]
14:33-!-phrag [n=phrag@unaffiliated/phrag] has joined #mythtv
14:34-!-phrag [n=phrag@unaffiliated/phrag] has left #mythtv ["Leaving"]
14:34-!-renatofilho^_ [n=renato@] has quit [Remote closed the connection]
14:34<Cardoe>justinh: you might hear some murmur from Gentoo users... I'm going to try to have mythtv-themes pull in both repos
14:37-!-renatofilho^ [n=renato@] has joined #mythtv
14:39-!-hti_pro [] has joined #mythtv
14:43-!-TelnetManta [] has quit ["Ex-Chat"]
14:44<hti_pro>Firstly i feel that this issue is developer related, and I hope I am not crossing any lines. If this issue is not relevant to the channel please direct me to the correct place. I have an idea for a startup company that could involve mythtv. This company would primarily be interested in offering a whole house distributed pvr application. I am interested in using mythtv to do this. I would not be selling the software of course, in a
14:46<Cardoe>Someone needs to tell nigel
14:46<Cardoe>Because he's ignoring me
14:46<Cardoe>64 bit Intel machines should not have -march=k8
14:46<Cardoe>it's -march=nocona
14:47<Cardoe>-march=k8 will potentially lead to bad instructions in future GCC versions when they actually make use of the full x86_64 instruction set
14:47<Cardoe>since k8 specifies the AMD instruction set
14:53<gbee>danielk22: have you got the message about your email address bouncing email from Robert?
14:53<janneg>Cardoe: I'll fix it after the next ffmpeg merge
14:54<gbee>hti_pro: we tend to reserve this channel for code related discussion, if you are looking for permission etc then best contact the lead developer, Isaac, in person
14:55*gbee ducks a swipe from Chutt
14:55<hti_pro>thank you very much
14:55<xris>hti_pro: from personal experience, I can pretty safely say that Isaac doesn't really care, as long as you follow the terms of the GPL.
14:56<xris>just be aware that the schedules direct agreement won't allow you to sell/provide SD memberships for people for US/Canada listings.
14:56<xris>so you'll probably have to work something out with TMS.
14:59<gbee>hti_pro: as xris said, myth is GPL'd so you are free to do whatever you want, but I think we'd ask that you be prepared to handle your customers support requests - a couple of guys selling mythtv boxes commercially come to us whenever they are asked a question, or a feature is requested by their customers - that's not really on, because they are paid to look after their customers, we aren't paid a penny
15:00<gbee>if you require changes or features in the mythtv codebase then either make them yourself or hire someone to do it
15:01<hti_pro>I totally agree, I am interested in taking feature requests at some point, at which time I plan to have my own hand in the development, and I would like to do it in the interest of the entire mythtv community not just my customers
15:04<hti_pro>alright well I thank all of you for your time and input and look forward to the continued development in mythtv and I would also like to say that I am very happy with the current status of mythtv and that you guys have done a great job thus far.
15:06-!-hti_pro [] has left #mythtv []
15:12<gbee>anyone know of a linux tool for finding unused headers?
15:12<gbee>or a gcc switch ...
15:16<janneg>no, unfortunately not. I've searched already last week
15:17<janneg>last compile test before the merge is running
15:17*janneg can't look at diffs anymore
15:21*Cardoe cheers janneg
15:28<gbee>where are global keybindings defined? I need to add a backspace binding for the textedit widget
15:29<gbee>okay - grepping for different strings finally paid off
15:45<janneg>final diffstat (mythtv and mythplugins) before commit: 716 files changed, 8257 insertions(+), 10368 deletions(-)
15:48<janneg>and committed
15:51<gbee>lets see how many conflicts there are in my tree ...
15:52<gbee>six, not too bad
15:53<Chutt>the change is too big for trac
15:55<Chutt>janneg, sending email to the -dev list?
15:57<janneg>Chutt: huh, it shouldn't be larger than a ffmpeg sync
15:57<Chutt>apparently it is =)
15:57<janneg>I'm composing it
15:57<Chutt>it's not showing the diff in trac, just gives you the option to download it
15:57<gbee>touches more files
15:57-!-JoeBorn [] has joined #mythtv
16:05<Cardoe>ebuild added to Gentoo just to punish the people that annoy me
16:05<gbee>what, us? :p
16:06<Cardoe>heh. nah
16:06<Cardoe>well maybe danielk22...
16:06<Cardoe>but he knows why..
16:06*Cardoe pokes danielk22 with a ./configure line
16:07<Cardoe>he closed some users ticket saying the Gentoo maintainer must be using the wrong ./configure line..
16:07<Cardoe>except I had danielk22 review the ./configure line I was building 0.21 with before I added it to Gentoo
16:08<gbee>not looking forward to the noise this will generate on the -dev list and trac, but lets hope that I'm just being overly pessimistic and that people did actually read the warnings we gave
16:10<Chutt>xris, so we didn't go with a small logo, huh
16:12<xris>Chutt: that's just the sample. could switch, but I think you were the only vote for the small logo
16:12<xris>tshirt with small pocket-area logo looks a little funny, imho
16:12<Chutt>well, i won't be wearing that :p
16:13<xris>could always get some polos for the board. :)
16:14-!-ahbritto [] has joined #mythtv
16:16<xris>but... do weigh in on the logo size.. maybe the rest of the board will actually say something.
16:18<Chutt>maybe if it wasn't so tall
16:18<Chutt>but then it'd be squished
16:20<janneg>the diff was 28k lines
16:20-!-czth [n=dbrobins@nat/microsoft/x-c676430452b971be] has joined #mythtv
16:20-!-czth_ [n=dbrobins@nat/microsoft/x-3ba4dae479634a1b] has joined #mythtv
16:21<xris>Chutt: yeah, I thought about a 2/3 size one, but then it just looks weird/small
16:24*Chutt updates to TOT
16:24<Chutt>production box, even
16:25-!-czth_ [n=dbrobins@nat/microsoft/x-3ba4dae479634a1b] has quit [Client Quit]
16:27<xris>how long before qt4 stuff gets merged into head?
16:27<Chutt>it's in now
16:27<justinh>Cardoe: only in 0.21-fixes I hope.... I can still hardly even bear to look at the 4:3 ones
16:27<janneg>-40 minutes
16:28<Cardoe>justinh: yeah that's what I'm working with
16:28<xris>Chutt: ahh, cool. guess I should recompile
16:28<xris>of course, that would require that I could GET to my home network... stupid comcast...
16:28<Cardoe>xris: use a different port then 22 ;)
16:29<Cardoe>and use something like and ddclient
16:29<Chutt>no qt4 on that box
16:30<justinh>Cardoe: AFAIK they're all good with 0.21 still. the mythui port would involve a lot of work I think, so I'm looking into doing something else. I even considered kicking glass-wide into shape for 0.21 but to me it's lost its shine. need to break new ground
16:31<xris>Cardoe: no.. my connection is down... I have a business line -- no blocked ports.
16:32<Cardoe>xris: oh
16:32<Cardoe>xris: well then.. I agree.. stupid comcast. :-D
16:33-!-mo0dbo0m [] has quit [Client Quit]
16:44<Chutt><command-line>: warning: missing terminating " character
16:44<Chutt><command-line>: warning: missing terminating " character
16:44<Chutt>version.cpp:3: error: expected primary-expression before ';' token
16:44<Chutt>grep: "/root/mythtv/mythtv"/libs/libmyth/mythcontext.h: No such file or directory
16:44<Chutt>(grep from before version.cpp compilation)
16:49<janneg>Chutt: are you sure you're using qt4 qmake?
16:49<Chutt>no, actually
16:49<Chutt>i thought you changed it to use qmake-qt4 if it existed?
16:49<janneg>I did
16:50<Chutt>it's using qmake on the commandline for everything
16:50<Chutt>not qmake-qt4
16:51<janneg>I had to delete the Makefiles in my trunk checkout before it actually used the configured qmake
16:51<gbee>it's using QTDIR here which points to the qt3 qmake, I have to manually override with --qmake
16:51<Chutt>it's still using qmake here
16:51<Chutt>did a distclean
16:52<gbee>even when qmake in /usr/bin is symlinked to the qt4 version it uses the qt3 one, very annoying
16:52<janneg>Chutt: if the standard qmake is qt4 it will use it
16:52<Chutt>'qmake' is qt3
16:52<Chutt>'qmake-qt4' is qt4
16:53<Chutt>there we go
16:53<Chutt>'make distclean' didn't delete _all_ the Makefiles
16:53<gbee>janneg: any chance we can parse the output of qmake --version in configure and bail before trying to compile?
16:53<janneg>gbee: that's already there
16:53<Chutt>the makefile in libs was still qmake, not qmake-qt4
16:55-!-Tanthrix [] has quit [Read error: 110 (Connection timed out)]
16:55<gbee>after a distclean
16:56<gbee>libs and programs definately should have been removed
16:57<gbee>looks like I was just hitting the same issue as Chutt, though that doesn't explain why forcing --qmake worked
16:57*gbee shrugs
17:03-!-sveinung [] has joined #mythtv
17:10<Chutt>i wonder why dbcheck.cpp takes so long to compile
17:10<Chutt>it's basiacally just a huge list of strings
17:15-!-|gunni| [] has joined #mythtv
17:15-!-|gunni| [] has quit [Read error: 104 (Connection reset by peer)]
17:15<gbee>completely refusing to tune on my second tuner atm, backend restart didn't help
17:16-!-melunko [n=hmelo@] has quit [Read error: 113 (No route to host)]
17:25-!-jgarvey [] has quit [Read error: 110 (Connection timed out)]
17:26-!-jgarvey [] has joined #mythtv
17:28-!-joobie [n=joobie@] has joined #mythtv
17:29-!-kormoc [n=kormoc@unaffiliated/kormoc] has joined #mythtv
17:29-!-___sveinung [] has quit [Read error: 110 (Connection timed out)]
17:29-!-JoeBorn [] has quit [Read error: 110 (Connection timed out)]
17:29-!-skamithi [] has quit ["WeeChat 0.2.3"]
17:30<gbee>any benefit to using "-exec rm -f {};" instead of -delete
17:30-!-nwahmaet [] has left #mythtv []
17:33<kormoc>gbee, supported by older versions of find, but -delete is the 'better way(tm)(r)'
17:34<janneg>-exec rm is traditional
17:34-!-|gunni| [] has joined #mythtv
17:34<janneg>Chutt, gbee: make sure to revert the 4 not generated makefiles in the tree
17:34<Chutt>what 4?
17:34<gbee>contrib ones for a start
17:35<Chutt>i don't build anything in contrib, so =)
17:35<Chutt>and i checked that i had no local changes
17:36<janneg>Makefiles not generated by qmake but handwritten and under version control
17:42*kormoc shoots Colin Guthrie. Just let the topic die already...
17:44<Chutt>janneg, none of those are used in a normal make =)
17:48<janneg>Chutt: sure, but the changes might get comitted by the next svn ci
17:50-!-JoeBorn [] has joined #mythtv
17:51-!-Dave123 [] has quit [Connection timed out]
17:52<gnome42>gbee: there are other real advantages to -exec rm ... although they are alluding me atm :) (ie. stay away from -delete, it'll eventually burn ya IMHO :)
17:52<kormoc>gnome42, I seem to recall it being the other way around, -delete is the 'better' way
17:53<gnome42>gbee: oh, I remember one advantage now ...
17:53<gbee>been using it since it was added to find, but I thought I'd better ask whether there was a good reason people were still using -exec rm
17:53<gbee>since I'm usually the last to know these things
17:54<gnome42>-delete does not fork so if your 'found' list is very long you will run out of chars on the cmdline in a hurry
17:54<kormoc>you sure?
17:54<gnome42>AIX is the worst for that
17:54<gnome42>kormoc: i think so :) But I'd have to recheck to be sure
17:55<gnome42>-exec rm forks and rm for each I'm pretty sure
17:55<kormoc>it actually queues them up iirc
17:55<janneg>I would consider not forking for every file an advantage
17:56<gnome42>the cmdline limit in linux is huge though
17:56-!-clever [] has quit [Connection timed out]
17:56<gnome42>janneg: I would too but, in practice -exec rm is always safe
17:57<janneg>yes, and forking is pretty cheap on linux
17:57<gnome42>especially compared to the IO ;)
18:01-!-|gunni| [] has quit ["KVIrc 3.2.4 Anomalies"]
18:01-!-Cardoe [n=Cardoe@gentoo/developer/Cardoe] has quit ["Leaving"]
18:01-!-clever [] has joined #mythtv
18:02-!-xris [] has quit []
18:03<kormoc>gnome42, 100,000 8 char filenames -delete worked fine, rm * did not "Argument list too long"
18:03<kormoc>gnome42, I do believe that -delete doesn't fork *but* it runs the unlink syscall once per file (strace agrees with this)
18:04-!-jgarvey [] has quit ["Leaving"]
18:05<kormoc>gnome42, and given my ARG_MAX is 131,072 bytes, 900,000 bytes is a good bit over that
18:05<gnome42>kormoc: I believe ya. It was that superstar Zijlstra's fault! :))
18:09<gnome42>limitless on linux, so it's just a portability concern I guess.
18:12<janneg>gnome42: the -2k lines for qt4 port are coming from killing mythbrowser. the port added 1500-2000 lines. big part of it being db updates though
18:12<gnome42>I wonder if there's issues if there's spaces in the filenames, dunno
18:13<gnome42>janneg: oh, I see. How much without the DB updates considered? (if you happen to know)
18:13-!-jmk_ [n=jmk@] has quit ["Leaving"]
18:16<kormoc>gnome42, also, if it matters, 100,000 files, -delete takes 2.6s real time, -exec rm takes 127.6s
18:16<janneg>gnome42: libmythtv/dbcheck.cpp adds 550 lines
18:17<gnome42>kormoc: thx, I was curious about the times.
18:18<kormoc>gnome42, and they both handle files with spaces correctly
18:20<gnome42>k, good to know (for linux aways). Gotta run
18:20*kormoc waves
18:23-!-MrGandalf [] has joined #mythtv
18:24-!-Cardoe [n=Cardoe@gentoo/developer/Cardoe] has joined #mythtv
18:31<janneg>gnome42: most of the added lines are the db updates. ~1200 lines
18:36-!-Tanthrix [] has joined #mythtv
18:37-!-mattwire [] has joined #mythtv
18:47-!-xris [] has joined #mythtv
18:49<xris>grumble. can't blame comcast for lost internet at home.. firewall kernel panic.
18:57-!-beavis [] has quit ["Verlassend"]
19:03-!-Dibblah [] has quit [No route to host]
19:12-!-Dibblah [] has joined #mythtv
19:16-!-grokky [] has joined #mythtv
19:21-!-purserj [] has joined #mythtv
19:24<Chutt>janneg, mythphone crashes mythfrontend on startup for me
19:24<Chutt>assert due to a qwidget being created outside of the gui thread
19:26<Chutt>my fonts are all wrong with mepo-wide :(
19:26<Chutt>probably just need to get the newer version of the theme
19:31<Chutt>it's like it's using two different fonts on screen
19:32<Chutt>the 'a's and 't's are lower than every other character
19:33<janneg>Chutt: do you use mythphone? i.e. has it a valid config?
19:33<Chutt>never used it
19:33<Chutt>other than compiling/installing it
19:34<janneg>I can't reproduce it. I can even enter the mythphone screen
19:34<Chutt>i'll check it out later
19:34<Chutt>might be that it repros on my non-production system =)
19:34<Chutt>everything (except the fonts in mepo-wide) looks ok on my main box
19:34<Chutt>not that i've done more than live-tv and playing back a recording yert
19:37<janneg>text is for some reason smaller than with qt3. but all glyphs looks the same
19:38*janneg is installing debian etch in a virtual machine to test compilation with qt4.2
19:43-!-MrGandalf [] has quit [Read error: 110 (Connection timed out)]
19:47-!-TelnetManta [] has joined #mythtv
19:48-!-Dave123 [] has joined #mythtv
19:49<Chutt>did anyone ever see that flickering in playback?
19:50<Chutt>it's happening on my production machine
19:50-!-MrGandalf [] has joined #mythtv
19:51<laga>what flickering?
19:51<Chutt>in the qt4 port
19:51<Chutt>it's pretty obvious =)
19:52<laga>ah, i don t have the qt4 port yet :)
19:55<gbee>I've not seen any flickering
19:55<janneg>I don't have any flickering
19:57<gbee>Chutt: sure it's not related to linear deint being forced on by the playback profile changes just prior to the 0.21 release?
19:58<gbee>I do see a nasty flicker caused by linear deint being used with Nvidia TV-out and the cards flicker filter, caught me out when I upgraded relatives to 0.21
19:58-!-mattwire [] has quit ["Leaving"]
20:03<Chutt>it's blacking out the area where the preview video would be
20:03<Chutt>as well as some of the other info thingies
20:04<janneg>I hope that's not caused by my late changes. my productive system is still on r16772
20:06<Chutt>disabling the video preview fixes it
20:06<janneg>ah, that would explain it. video preview is disabled
20:06<gbee>don't have video preview enabled, but it shouldn't be running when the video is playing
20:06<Chutt>apparently it's running enough =)
20:08<gbee>playingSomething should be set false when play is called and that should shutdown the preview video
20:08<janneg>it's not visible in the watch recordings screen
20:10<gbee>playbackbox.cpp lines 2750, 2751 and PlaybackBox::drawVideo() line 1054
20:10<Chutt>gbee, i know it's _supposed_ to be off
20:11<Captain_Murdoch>the previewVideoRefreshTimer is still firing off I think and since previewVideoEnabled is still true, it still calls update(blackholebounds)
20:11<Captain_Murdoch>in PBB::timeout()
20:11<Chutt>i was just going to ask if it was still calling update(), though
20:11<Chutt>might be some wackiness with event filters
20:13<gbee>yeah, looks like CM is right, stopping the timer in play() or checking that !playingsomething before calling update() in timeout() should fix it
20:13<gbee>was that something danielk22 recently added? I don't remember seeing it before
20:13<Chutt>it's a _little_ hard to code on a 80" screen sitting 12 feet away
20:13<Chutt>at 1080p
20:17<gbee>ok, committed a fix, but I take no responsibility ;)
20:17<gbee>I'd rather have stopped the timer, but at a glance I couldn't see where I'd restart it after stopping playback (I assume it's there, I just didn't want to spend time looking for it)
20:19<Captain_Murdoch>you'd have to delve into the previewVideoState changing stuff probably.
20:19<Captain_Murdoch>actually, you might be able to do it in startPlayer() and killPlayer() (and killPlayerSafe())
20:20<gbee>what I've committed should work if we were right about the problem and it's a good enough fix until playbackbox is ported to mythui, at which point it becomes a non-issue
20:20<Chutt>excellent, thanks
20:21<Chutt>wife's in the middle of watching something now, though, so i'll have to test later =)
20:27*Captain_Murdoch fights with his frontends' shared nfsroot environment to get qt4 rpms installed
20:31-!-ToadP [] has joined #mythtv
20:40<gbee>right, I'm off to bed, changes to mythuitextedit don't quite work so I'll try again tomorrow night :/
21:11-!-Cardoe [n=Cardoe@gentoo/developer/Cardoe] has quit ["Leaving"]
21:39-!-minthome [] has joined #mythtv
22:05-!-___sveinung [] has joined #mythtv
22:23-!-frank23 [] has joined #mythtv
22:23-!-sveinung [] has quit [Read error: 110 (Connection timed out)]
22:28-!-HRearden [] has joined #mythtv
22:39-!-sveinung [] has joined #mythtv
22:45-!-ahbritto [] has quit [Client Quit]
22:56-!-___sveinung [] has quit [Read error: 110 (Connection timed out)]
22:58-!-kormoc [n=kormoc@unaffiliated/kormoc] has quit []
23:01-!-ahbritto [] has joined #mythtv
23:37-!-raidium [] has joined #mythtv
23:45-!-___sveinung [] has joined #mythtv
23:54-!-Netsplit <-> quits: HRearden, CDev, iamlindoro_, TelnetManta, raidium, sphery
23:55-!-Netsplit over, joins: raidium, TelnetManta, CDev, iamlindoro_, sphery
23:59-!-feld [] has joined #mythtv
---Logclosed Wed Mar 26 00:00:11 2008