Back to Home / #mythtv / 2008 / 06 / Prev Day | Next Day
#mythtv IRC Logs for 2008-06-27

---Logopened Fri Jun 27 00:00:01 2008
00:03-!-reynaldo [n=rverdejo@39-194-20-190.adsl.terra.cl] has quit [Read error: 110 (Connection timed out)]
00:08-!-superm1 [n=superm1@ubuntu/member/superm1] has quit [Remote closed the connection]
00:09-!-reynaldo_ [n=rverdejo@103-252-20-190.adsl.terra.cl] has quit [Read error: 110 (Connection timed out)]
00:18-!-eharris_ is now known as eharris
00:21-!-chainsawf [n=chainsaw@121-73-114-254.cable.telstraclear.net] has joined #mythtv
00:21-!-reynaldo1 [n=rverdejo@129-253-20-190.adsl.terra.cl] has quit [Read error: 110 (Connection timed out)]
00:34-!-superm1 [n=superm1@ubuntu/member/superm1] has joined #mythtv
00:35-!-chainsawbike [n=chainsaw@121-73-114-254.cable.telstraclear.net] has quit [Read error: 111 (Connection refused)]
00:35-!-chainsawf is now known as chainsawbike
01:05-!-Lud7773 [n=nadia@c-98-202-105-66.hsd1.ut.comcast.net] has joined #mythtv
01:06-!-Lud7773 [n=nadia@c-98-202-105-66.hsd1.ut.comcast.net] has left #mythtv []
01:09-!-joobie [n=joobie@joobie.org] has joined #mythtv
01:26-!-leprechau [i=charlie@c-68-52-83-189.hsd1.tn.comcast.net] has quit [Remote closed the connection]
01:33-!-leprechau [i=charlie@c-68-52-83-189.hsd1.tn.comcast.net] has joined #mythtv
01:53-!-purserj [n=purserj@k-sit.com] has joined #mythtv
02:12-!-gnome42 [n=gnome42@206-248-129-136.dsl.teksavvy.com] has quit [Remote closed the connection]
02:23-!-noisymime [n=josh@98.65.96.58.exetel.com.au] has quit [Read error: 104 (Connection reset by peer)]
02:33-!-edannenbe [n=edannenb@mail.blooparkstudios.de] has joined #mythtv
02:51-!-bronson [n=bronson@adsl-68-122-117-135.dsl.pltn13.pacbell.net] has joined #mythtv
02:51-!-bronson [n=bronson@adsl-68-122-117-135.dsl.pltn13.pacbell.net] has left #mythtv ["Ex-Chat"]
02:56-!-grokky [n=grokky@batter.csse.unimelb.edu.au] has quit [Read error: 110 (Connection timed out)]
03:06-!-MGisbers [n=mgisbers@dslb-088-076-161-046.pools.arcor-ip.net] has quit [Read error: 110 (Connection timed out)]
03:17-!-MGisbers [n=mgisbers@192.54.45.76] has joined #mythtv
03:18-!-Ace2016_ [n=ace@79-68-204-7.dynamic.dsl.as9105.com] has joined #mythtv
03:29-!-_gunni_ [n=Gunni@xdsl-81-173-252-41.netcologne.de] has joined #mythtv
03:30-!-cskowronnek [n=cskowron@mail.blooparkstudios.de] has joined #mythtv
03:34-!-Theroc172 [i=therock@connected.through.eushells.com] has quit [Remote closed the connection]
03:46-!-|gunni| [n=Gunni@xdsl-84-44-159-184.netcologne.de] has quit [Read error: 110 (Connection timed out)]
03:47-!-danielk22 [n=danielk@p78-22.acedsl.com] has quit [Read error: 110 (Connection timed out)]
03:47-!-ahbritto [n=guest@adsl-68-124-184-167.dsl.pltn13.pacbell.net] has joined #mythtv
03:48-!-Captain_Murdoch [n=cpinkham@ip72-218-64-158.hr.hr.cox.net] has quit [Read error: 110 (Connection timed out)]
03:48-!-Tanthrix [n=tanthrix@c-71-56-144-229.hsd1.or.comcast.net] has quit [Read error: 110 (Connection timed out)]
03:49-!-Captain_Murdoch [n=cpinkham@ip72-218-64-158.hr.hr.cox.net] has joined #mythtv
03:51-!-danielk22 [n=danielk@p78-22.acedsl.com] has joined #mythtv
03:55-!-MGisbers [n=mgisbers@192.54.45.76] has quit [Remote closed the connection]
03:56-!-MGisbers [n=mgisbers@192.54.45.76] has joined #mythtv
04:03-!-purserj [n=purserj@k-sit.com] has quit ["leaving"]
04:17-!-famicom [i=famicom@c51447b09.cable.wanadoo.nl] has joined #mythtv
04:20-!-jams [n=jams@CPE-72-131-6-174.wi.res.rr.com] has quit [Read error: 110 (Connection timed out)]
04:24-!-jams [n=jams@CPE-72-131-6-174.wi.res.rr.com] has joined #mythtv
04:25<The_Ball>top
04:25<The_Ball>sorry, wrong window
04:35-!-Therock_ [i=therock@connected.through.eushells.com] has joined #mythtv
05:24-!-grokky [n=grokky@ppp59-167-139-21.lns3.mel6.internode.on.net] has joined #mythtv
05:30-!-Tanthrix [n=tanthrix@c-71-56-144-229.hsd1.or.comcast.net] has joined #mythtv
06:16-!-Ace2016__ [n=ace@79-68-204-7.dynamic.dsl.as9105.com] has joined #mythtv
06:28-!-Ace2016_ [n=ace@79-68-204-7.dynamic.dsl.as9105.com] has quit [Read error: 110 (Connection timed out)]
06:36-!-cskowronnek [n=cskowron@mail.blooparkstudios.de] has quit ["Ex-Chat"]
06:39-!-cskowronnek [n=cskowron@mail.blooparkstudios.de] has joined #mythtv
06:41-!-Ace2016_ [n=ace@79-68-204-7.dynamic.dsl.as9105.com] has joined #mythtv
06:47-!-Ace2016__ [n=ace@79-68-204-7.dynamic.dsl.as9105.com] has quit [Read error: 104 (Connection reset by peer)]
06:47-!-joobie [n=joobie@joobie.org] has quit ["Leaving"]
06:49-!-famicom_ [i=famicom@c51447b09.cable.wanadoo.nl] has joined #mythtv
06:52-!-joobie [n=joobie@joobie.org] has joined #mythtv
06:54-!-MrGandalf [n=mgandalf@cpe-72-225-44-161.rochester.res.rr.com] has quit ["blah"]
07:01-!-sphery_ [n=mdean@user-0c6sj46.cable.mindspring.com] has joined #mythtv
07:01-!-famicom [i=famicom@c51447b09.cable.wanadoo.nl] has quit [Read error: 110 (Connection timed out)]
07:06-!-MGisbers [n=mgisbers@192.54.45.76] has quit []
07:11-!-sphery__ [n=mdean@user-0c6sj46.cable.mindspring.com] has joined #mythtv
07:17-!-sphery [n=mdean@user-0c6sj46.cable.mindspring.com] has quit [Read error: 110 (Connection timed out)]
07:18-!-sphery__ [n=mdean@user-0c6sj46.cable.mindspring.com] has quit [Read error: 113 (No route to host)]
07:19-!-sphery [n=mdean@user-0c6sj46.cable.mindspring.com] has joined #mythtv
07:26-!-sphery_ [n=mdean@user-0c6sj46.cable.mindspring.com] has quit [Read error: 110 (Connection timed out)]
07:54-!-sphery [n=mdean@user-0c6sj46.cable.mindspring.com] has quit [Read error: 113 (No route to host)]
08:13-!-edannenbe [n=edannenb@mail.blooparkstudios.de] has quit [Remote closed the connection]
08:14-!-edannenbe [n=edannenb@mail.blooparkstudios.de] has joined #mythtv
08:18-!-edannenbe [n=edannenb@mail.blooparkstudios.de] has quit [Read error: 104 (Connection reset by peer)]
08:19-!-edannenbe [n=edannenb@mail.blooparkstudios.de] has joined #mythtv
08:21-!-edannenbe [n=edannenb@mail.blooparkstudios.de] has quit [Remote closed the connection]
08:21-!-edannenbe [n=edannenb@mail.blooparkstudios.de] has joined #mythtv
08:39-!-edannenbe [n=edannenb@mail.blooparkstudios.de] has quit [Remote closed the connection]
08:43-!-sphery [n=mdean@user-0c6sj46.cable.mindspring.com] has joined #mythtv
08:47-!-edannenbe [n=edannenb@mail.blooparkstudios.de] has joined #mythtv
09:03-!-famicom__ [i=famicom@c51447b09.cable.wanadoo.nl] has joined #mythtv
09:09-!-gustave_ [n=chatzill@S0106001cf0b8088c.cg.shawcable.net] has joined #mythtv
09:21-!-Cardoe [n=Cardoe@gentoo/developer/Cardoe] has joined #mythtv
09:29-!-famicom_ [i=famicom@c51447b09.cable.wanadoo.nl] has quit [Read error: 110 (Connection timed out)]
09:42-!-Amun [n=Jonsi@c-71-205-199-128.hsd1.mi.comcast.net] has quit []
09:42-!-Amun [n=Jonsi@c-71-205-199-128.hsd1.mi.comcast.net] has joined #mythtv
09:49-!-jmk [n=jmk@64.73.34.172] has joined #mythtv
09:56-!-grokky [n=grokky@ppp59-167-139-21.lns3.mel6.internode.on.net] has quit []
10:00-!-foxbuntu [n=foxxxxbu@12-219-21-84.client.mchsi.com] has quit ["Leaving"]
10:31-!-edannenbe [n=edannenb@mail.blooparkstudios.de] has quit ["Ex-Chat"]
10:33-!-MrGandalf [i=mgandalf@cpe-72-225-41-14.rochester.res.rr.com] has joined #mythtv
10:45-!-squidly [n=craig@HoodLUG/member/squidly] has joined #mythtv
10:45-!-squidly [n=craig@HoodLUG/member/squidly] has left #mythtv ["Leaving"]
11:01-!-reynaldo [n=rverdejo@91-200-20-190.adsl.terra.cl] has joined #mythtv
11:10-!-beandog [n=steve@gentoo/developer/beandog] has joined #mythtv
11:12-!-mattwire [n=mattwire@host86-146-213-137.range86-146.btcentralplus.com] has joined #mythtv
11:14-!-iamlindoro_ [n=iamlindo@140.239.95.222] has joined #mythtv
11:19-!-gnome42 [n=gnome42@206-248-129-136.dsl.teksavvy.com] has joined #mythtv
11:26-!-MrGandalf [i=mgandalf@cpe-72-225-41-14.rochester.res.rr.com] has quit ["home"]
11:28-!-reynaldo [n=rverdejo@91-200-20-190.adsl.terra.cl] has quit [Read error: 113 (No route to host)]
11:32-!-edannenbe [n=bird@BAEc6ff.bae.pppool.de] has joined #mythtv
11:47-!-famicom [i=famicom@c51447b09.cable.wanadoo.nl] has joined #mythtv
11:53-!-famicom_ [i=famicom@c51447b09.cable.wanadoo.nl] has joined #mythtv
11:58-!-joobie [n=joobie@joobie.org] has quit ["This computer has gone to sleep"]
12:02-!-famicom__ [i=famicom@c51447b09.cable.wanadoo.nl] has quit [Read error: 110 (Connection timed out)]
12:04-!-famicom [i=famicom@c51447b09.cable.wanadoo.nl] has quit [Read error: 110 (Connection timed out)]
12:35-!-ahbritto [n=guest@adsl-68-124-184-167.dsl.pltn13.pacbell.net] has quit [Client Quit]
12:40-!-cskowronnek [n=cskowron@mail.blooparkstudios.de] has quit ["Ex-Chat"]
12:48*Cardoe kicks DVD stuff
13:06-!-stoffel_ [n=sfr@p57B4E7C0.dip.t-dialin.net] has joined #mythtv
13:29<danielk22>cardoe: 'profile' is actually a good way to generate a mythtv for release, ubuntu also uses this.
13:30-!-czth_ [n=dbrobins@nat/microsoft/x-e8d8204a24257fd8] has joined #mythtv
13:33-!-czth__ [n=dbrobins@nat/microsoft/x-c75da33a2a62ec7f] has joined #mythtv
13:41-!-MrGandalf [n=mgandalf@cpe-72-225-44-161.rochester.res.rr.com] has joined #mythtv
13:47-!-czth [n=dbrobins@nat/microsoft/x-f335b36842a3f851] has quit [Connection timed out]
13:49-!-czth_ [n=dbrobins@nat/microsoft/x-e8d8204a24257fd8] has quit [Read error: 110 (Connection timed out)]
13:57<Cardoe>danielk22: yeah. it's decent for debugging as well
13:58<Cardoe>but there have been some situations where it's not
13:58<Cardoe>danielk22: now if you can fix the dvd player part.. so the WAF ratio goes up.. cause she's pissed right now
13:59<danielk22>what
14:00<danielk22>s the problem?
14:00<danielk22>the dvd player is really stanley's area of expertise.
14:02<Cardoe>http://svn.mythtv.org/trac/ticket/5478
14:02<Cardoe>I managed to reproduce the issue listed at the end without pausing the dvd ever
14:03<Cardoe>which almost makes me think it's hardware
14:03<Cardoe>but no errors in dmesg
14:04<Cardoe>I'm only logging "important,general" currently..
14:05<danielk22>This might actually be in an overlap of stuff I know about and stanley knows about.
14:05<Chutt>i just use a ps3 for dvd playback :p
14:05<danielk22>"-v playback,audio" might help with debugging...
14:06<Chutt>danielk22, we should someday split up GetFrame
14:06<Chutt>so that it doesn't process audio when video's behind/etc
14:07<danielk22>yep
14:07<danielk22>it has troubles with audio only files too.
14:07<danielk22>it works, but it's ugly
14:07<Chutt>that's the cause of the Audio buffer overflow stuff
14:08<Chutt>in cardoe's log, at least
14:09<Cardoe>danielk22: Alright. I'll change my logging and re-run the test
14:10<Cardoe>danielk22: thanks
14:15<Cardoe>any of you guys try gcc 4.3.x?
14:15<Cardoe>and trunk?
14:21<clever>im using 4.1.3
14:23-!-beata [n=beata@c-69-142-6-232.hsd1.nj.comcast.net] has joined #mythtv
14:49<_Joemama>are most computer remote controls compatible with mythtv?
14:59<danielk22>Joemama wrong channel, you want #mythtv-users
15:15-!-czth__ is now known as czth
15:25<stoffel_>creatix
15:34-!-moodboom [n=moodboom@cpe-075-177-134-090.nc.res.rr.com] has quit [Client Quit]
15:37-!-iamlindoro_ [n=iamlindo@140.239.95.222] has quit []
15:52-!-Anduin [n=awithers@adsl-69-110-19-95.dsl.pltn13.pacbell.net] has quit [Read error: 113 (No route to host)]
15:52-!-Anduin [n=awithers@adsl-69-110-19-95.dsl.pltn13.pacbell.net] has joined #mythtv
15:53-!-iamlindoro_ [n=iamlindo@140.239.95.222] has joined #mythtv
16:11-!-tomimo [n=kurre@xdsl-83-150-88-111.nebulazone.fi] has quit ["Gotta get going ..."]
16:11-!-|gunni| [n=Gunni@xdsl-84-44-233-23.netcologne.de] has joined #mythtv
16:26-!-_gunni_ [n=Gunni@xdsl-81-173-252-41.netcologne.de] has quit [Read error: 110 (Connection timed out)]
16:31-!-mace_ [n=mace@cpc1-john1-0-0-cust917.renf.cable.ntl.com] has quit [Remote closed the connection]
16:31-!-mace [n=mace@debian/developer/mace] has joined #mythtv
16:33-!-tomimo [n=kurre@xdsl-83-150-88-111.nebulazone.fi] has joined #mythtv
16:38-!-streamtrade [n=jsass@24.32.27.162] has joined #MythTV
16:45-!-gustave_ [n=chatzill@S0106001cf0b8088c.cg.shawcable.net] has quit ["ChatZilla 0.9.83 [Firefox 3.0/2008052912]"]
16:47-!-Anduin1 [n=awithers@adsl-69-110-6-123.dsl.pltn13.pacbell.net] has joined #mythtv
16:48-!-stoffel_ [n=sfr@p57B4E7C0.dip.t-dialin.net] has quit ["leaving"]
16:51-!-Anduin1 [n=awithers@adsl-69-110-6-123.dsl.pltn13.pacbell.net] has quit [Connection reset by peer]
16:51-!-Anduin [n=awithers@adsl-69-110-19-95.dsl.pltn13.pacbell.net] has quit [Read error: 60 (Operation timed out)]
16:55-!-jmk [n=jmk@64.73.34.172] has quit ["Leaving"]
16:58<Chutt>hrm
16:58<Chutt>i wonder if qt4 allows painting of qwidgets into qimages without too much effort
17:09-!-famicom__ [i=famicom@c51447b09.cable.wanadoo.nl] has joined #mythtv
17:10-!-Fenix-Dark [n=scott@ool-4352729d.dyn.optonline.net] has joined #mythtv
17:10-!-Fenix-Dark [n=scott@ool-4352729d.dyn.optonline.net] has left #mythtv ["Leaving"]
17:22-!-stoth [n=stoth@ool-18bfe594.dyn.optonline.net] has joined #mythtv
17:24-!-beandog [n=steve@gentoo/developer/beandog] has quit ["Leaving"]
17:26-!-famicom_ [i=famicom@c51447b09.cable.wanadoo.nl] has quit [Read error: 110 (Connection timed out)]
17:42-!-Ace2016_ [n=ace@79-68-204-7.dynamic.dsl.as9105.com] has quit [Connection timed out]
17:48-!-Ace2016_ [n=ace@79-68-204-7.dynamic.dsl.as9105.com] has joined #mythtv
17:51-!-iamlindoro_ [n=iamlindo@140.239.95.222] has quit []
18:02-!-Serphil [i=54ca2cb8@gateway/web/ajax/mibbit.com/x-8533862b3fa61e0d] has joined #mythtv
18:02-!-Cardoe [n=Cardoe@gentoo/developer/Cardoe] has quit ["Leaving"]
18:13-!-mattwire [n=mattwire@host86-146-213-137.range86-146.btcentralplus.com] has quit ["Leaving"]
18:13-!-Serphil [i=54ca2cb8@gateway/web/ajax/mibbit.com/x-8533862b3fa61e0d] has quit ["http://www.mibbit.com ajax IRC Client"]
18:15-!-|gunni| [n=Gunni@xdsl-84-44-233-23.netcologne.de] has quit ["KVIrc 3.2.4 Anomalies http://www.kvirc.net/"]
18:19-!-stoth [n=stoth@ool-18bfe594.dyn.optonline.net] has quit [Remote closed the connection]
18:44-!-beoba [n=fsoh@unaffiliated/beoba] has joined #mythtv
18:50-!-danielk22 [n=danielk@p78-22.acedsl.com] has quit [Read error: 104 (Connection reset by peer)]
18:53-!-danielk22 [n=danielk@p78-22.acedsl.com] has joined #mythtv
19:11-!-foxbuntu [n=foxxxxbu@12-219-21-84.client.mchsi.com] has joined #mythtv
19:12-!-lsobral [n=sobral@200.184.118.132] has quit ["Ex-Chat"]
19:18-!-robthebob [n=rn114@78-32-214-33.no-dns-yet.enta.net] has joined #mythtv
19:33-!-Ace2016_ [n=ace@79-68-204-7.dynamic.dsl.as9105.com] has quit [Connection timed out]
20:11-!-lyricnz [n=simonrob@d220-238-85-206.dsl.vic.optusnet.com.au] has joined #mythtv
20:21<knowledgejunkie>should mythbackend resident memory usage be fairly static over time? Since the memory-leak ticket war I've been measuing usage every 10m - and in 4 days RSS has grown 4x
20:22<clever>mine is at 51mb atm
20:22<clever>but the rss can vary alot over time depending on how much is swaped out
20:22<clever>virtual shouldnt change much(assuming its not accessing files)
20:32-!-Jimbo_ [i=hidden-u@relay.optinc.com] has left #mythtv []
20:36<knowledgejunkie>I'm monitoring RSS and VSZ - the sum of these has grown from 350MB to 550MB over the 4 day period
20:37<clever>virtual is the entire memory space the program has
20:37<knowledgejunkie>RSS from 31MB to 123MB and VSZ from 321MB to 423MB
20:37<clever>rss is the chunk thats resident in ram
20:38<clever>my rss is at 69mb so far
20:38<clever>wait thats the fe
20:38<clever>backend rss 52mb
20:38<clever>virtual 275mb
20:38<clever>mythtv 412 1.3 10.3 282124 53676 pts/1 Sl+ Jun24 60:26 mythbackend -v network,upnp
20:38<clever>and its currently recording
20:40<knowledgejunkie>root 25385 1.4 11.8 423360 123172 ? Ssl Jun23 102:09 mythbackend --daemon --logfile /var/log/mythtv/mythbackend.log -v important,general --pidfile /var/run/mythbackend.pid --noupnp
20:43<knowledgejunkie>When I last shutdown mythbackend after ~2months uptime, VSZ was at 1.6GB - surely that isn't normal
20:48-!-beoba [n=fsoh@unaffiliated/beoba] has left #mythtv ["One short sleep past, we wake eternally"]
20:55<clever>yeah that isnt normal
20:55<clever>copy the /proc/PID/maps file
20:55<clever>that will tell you exactly where in the virtual memory the space is
20:56<clever>when i run it on my mythbackend
20:56<clever>i see these as the top 3(smallest first)
20:56<clever>b4cf1000-b553e000 r-xp 00000000 08:13 5652850 /usr/lib/libGLcore.so.1.0.9639
20:56<clever>b74bd000-b7f80000 r-xp 00000000 fe:00 40095239 /media/mainlv/root/7.10/lib/libmythtv-0.22.so.0.22.0
20:56<clever>0816e000-0a2f9000 rw-p 0816e000 00:00 0 [heap]
21:00-!-joobie [n=joobie@joobie.org] has joined #mythtv
21:02-!-joobie [n=joobie@joobie.org] has quit [Client Quit]
21:02<knowledgejunkie>don't have /proc/PID...
21:02<clever>i can see most of the virtual ram for mythbackend is to my heap
21:02<clever>replace PID with the pid of mythbackend
21:03<knowledgejunkie>oh... dear... it's... late... :)
21:03<clever>cat /proc/412/maps |/home/clever/bin/mapsort |sort -n|tail -n3
21:03<clever>(mapsort is a php script i made to aid)
21:03<clever>all it does is print the diff between the 2 numbers b74bd000-b7f80000
21:03<clever>(the size of the maping)
21:04<clever>http://pastebin.com/m1a38e974
21:04<knowledgejunkie>thanks
21:04<clever>(thats a tab in the quotes)
21:05<clever>(the last set of quotes)
21:05<knowledgejunkie>ok
21:05<clever>cat /proc/412/maps |/home/clever/bin/mapsort |sort -n|tail -n3
21:05<clever>or a bit simpler
21:06<clever>cat /proc/`pidof mythbackend`/maps |/home/clever/bin/mapsort |sort -n|tail -n3
21:06-!-qfour20 [i=qfour20@s15286609.onlinehome-server.com] has joined #mythtv
21:07<knowledgejunkie>got 'em
21:07<knowledgejunkie>10485760 b7346000-b7d46000 rwxp b7346000 00:00 0
21:07<knowledgejunkie>11472896 00667000-01158000 r-xp 00000000 03:02 4653409 /usr/local/lib/libmythtv-0.21.so.0.21.0
21:07<knowledgejunkie>44802048 08d0c000-0b7c6000 rwxp 08d0c000 00:00 0 [heap]
21:07<clever>looks pretty much the same as mine(except the glcore)
21:08<knowledgejunkie>yep - no GL or UPNP on that old backend
21:08<clever>the ones with the rwxp flags and no filename i think are extra chunks from new/malloc
21:08<knowledgejunkie>it does use EIT scanning though, and does have an ivtv card...
21:08<clever>ivtv here too
21:08<clever>a problem with alocing large chunks on the heap
21:09<clever>if you aloc a big chunk then a small one
21:09<clever>you get BIGBIGBIGBIGsmall
21:09<clever>then when you delete the big one you get ____________small
21:09<clever>giant hole in the heap that you cant release to the system:P
21:09<knowledgejunkie>nasty
21:09<clever>malloc/new will not use the heap if the block is 'too big'
21:10<clever>instead it makes a seperate maping
21:10<clever>like a 2nd heap
21:10<clever>but it cant be extended like the heap(doesnt work with brk())
21:12<knowledgejunkie>Is this a general Linux memory management issue rather than specific to MythTV? (I'm not yet a C programmer)
21:12<clever>its general memory management in c++ i think
21:12<clever>http://www.linuxdevcenter.com/pub/a/linux/2006/11/30/linux-out-of-memory.html
21:12<clever>thats where i learned a good deal of it
21:15<knowledgejunkie>thanks for the link - skimming now
21:16<clever>linux(and windows) have virtual memory
21:16<clever>in linux you can map files to certain addresses in the memory with mmap(which shows up in /proc/PID/maps)
21:17<clever>but theres a few special maps allready inplace
21:17<clever>like the heap&stack
21:19<knowledgejunkie>right
21:19<knowledgejunkie>does this in anyway rule out memory leaks in an app where this memory behaviour is observed?
21:19<knowledgejunkie>should I next think about running valgrind on mythbackend?
21:20-!-Cardoe [n=Cardoe@gentoo/developer/Cardoe] has joined #mythtv
21:21<Cardoe>well.. I just attempted to play Apollo 13 while I was out of the house and logged it
21:21<clever>knowledgejunkie: valgrind will help find out if its realy leaking
21:21<clever>and exactly where
21:23<clever>memory fragmentation(within the heap) could let it grow over time, but it should level off and not keep growing
21:23<clever>playing with the first 2 tiny c programs will show how the system will run out of ram after using 3gig(on a 32bit box)
21:23<clever>or less if you actualy use that ram(enless you got 3gig of ram+swap)
21:24<knowledgejunkie>1GB RAM and 1GB swap on the current machine
21:25-!-joobie [n=joobie@joobie.org] has joined #mythtv
21:25<clever>the linux kernel will by default let you alocate more ram then you have
21:25<clever>so i could alocate 3gig of ram on a 21mb ram box
21:25<clever>aslong as i just dont try to write to much of it
21:26<knowledgejunkie>I won't ask why that's a good idea
21:26<clever>some programs may alocate a ton of ram and not use it
21:26<clever>but you can turn it off if you dont like the idea:P
21:26<jamesd__>clever, it doesn't matter how much you write to it.. since it will swap it out as needed no matter how long it takes.
21:27<clever>jamesd__: enless i dont have the swap:P
21:27<clever>for example knowledgejunkie has 2gig of ram+swap
21:27<jamesd__>true
21:27<clever>so he can alocate a 3gig chunk of ram(roughtly)
21:27<clever>but only write to 2gig of it
21:27<clever>assuming NOTHING else was running
21:27-!-joobie [n=joobie@joobie.org] has quit [Client Quit]
21:28<clever>and because libs&heap&stack are randomly laid out in the virtualmem, you cant get 3gig in 1 peice
21:28<clever>just 98% of it in chunks
21:29-!-grokky [n=grokky@ppp59-167-139-21.lns3.mel6.internode.on.net] has joined #mythtv
21:30<clever>64bit solves a number of these problems
21:31<clever>you could alocate a 1tb chunk of ram and not even use 50% of the address space
21:31<knowledgejunkie>even if I have less than 4GB?
21:31<clever>yes
21:31<clever>because the kernel doesnt actualy use the ram you alocate
21:31<clever>until you write to it
21:31<jamesd__>knowledgejunkie, its called virtual memory for a reason.
21:31<clever>that too
21:32<clever>'There is no guarantee that allocated memory area is physically contiguous; it's only virtually contiguous.'
21:33<clever>(the data could easily get fragmented within psysical memory)
21:34<Chutt>my backend's been up for 11 days (with trunk), and has a resident size of 40MB
21:34<clever>but the resident can be anything, depending on how much other programs shoved out
21:34<Chutt>no swap in use :p
21:34<clever>lol
21:35<clever>for example my backend is 71mb resident
21:35<Chutt>the _frontend_ is using 215MB, however
21:35<clever>i just used a prog from http://www.linuxdevcenter.com/pub/a/linux/2006/11/30/linux-out-of-memory.html?page=1 to eat 150mb of ram
21:35<clever>pushing other crap out of the way
21:36<clever>now the backend has 39mb resident!
21:36<knowledgejunkie>my SDTV frontend is using 303MB virtual (85MB resident)
21:36<clever>mythtv 13502 14.3 7.9 240696 40864 pts/2 Sl+ 21:24 10:19 mythfrontend
21:36<clever>that varries wildly depending on theme
21:36<knowledgejunkie>mythlcdserver is somehow using nearly 60MB...
21:36<Chutt>and resolution
21:37<clever>yeah
21:37<hads>I can't remember how long mines been up (probably a few weeks) and mythbackend is using 38MB RES (fixes).
21:37<knowledgejunkie>I'm using MythCenter-Wide @720x576
21:38<clever>project greyhem 1024x768
21:38<knowledgejunkie>As my recording schedule is empty for tomorrow - I'll give valgrind a whirl and see if it throws anything up
21:38<clever>ive lowered my res before to be closer to sdtv(less downscaling for the card)
21:39<clever>but then the aspect ratio of many windows was weird
21:39<Chutt>root 13821 3.4 22.4 332972 217936 pts/2 SLl+ Jun24 146:46 mythfrontend
21:39-!-beandog [n=steve@gentoo/developer/beandog] has joined #mythtv
21:39<Chutt>(1080p, mepo-wide)
21:39<danielk22>re: mythlcdserver, you are probably counting the 61 libs which are shared by a number of programs on your machine.
21:39<clever>knowledgejunkie: with a hardware encoder, you can probly valgrind as you record
21:39<knowledgejunkie>checking top, i'm currently using about 8% swap
21:39<clever>5% cpu*10 is still only 50%
21:40<knowledgejunkie>my capture devices are either DVB-{S,T} or ivtv
21:40<clever>danielk22: yeah trying to track down every page of ram gets complex when you add shared
21:40<Chutt>i wouldn't be surprised if there's a leak in, say, the upnp code
21:40<clever>and even worse when you add copy on write
21:40<clever>i sometimes run the oom script(2nd) from http://www.linuxdevcenter.com/pub/a/linux/2006/11/30/linux-out-of-memory.html
21:41<Chutt>random stuff on the network hits that fairly often
21:41<clever>to flush stale data into swap
21:41<knowledgejunkie>renato: mythlcdserver - 31452 mythtv 15 0 57724 13m 10m S 0.0 1.4 0:17.08 mythlcdserver
21:41<Chutt>which would make stuff grow randomly if it were leaking
21:41<clever>Chutt: my upnp code randomly gets stuck calling the same syscall in a inf loop
21:41<danielk22>for me with full debugging symbols mythlcdserver uses 3480k
21:41<clever>causing massive system cpu usage
21:41<Chutt>clever, that's been fixed in trunk, iirc
21:42<clever>Chutt: people keep telling me that:P
21:42<Chutt>ah
21:42<Chutt>heh
21:42<Chutt>actually, i thought that was fixed pre 21
21:42-!-Amun [n=Jonsi@c-71-205-199-128.hsd1.mi.comcast.net] has quit []
21:42<clever>finding --version
21:42<clever>MythTV Version : 17442M
21:42<clever>MythTV Branch : trunk
21:42-!-Amun [n=Jonsi@c-71-205-199-128.hsd1.mi.comcast.net] has joined #mythtv
21:42<clever>qt4
21:42<Chutt>funky
21:42<clever>yea
21:43<Chutt>i haven't seen it, and i use upnp all the time
21:43<clever>i never even touch upnp and it happens:P
21:43<clever>but it tends to take a few months
21:43<Chutt>vista supports playing back from it nicely =)
21:43<clever>i tracked it down perfectly last time
21:43<clever>a certain function was claiming bytes are ready to be read(select i think)
21:44<Chutt>yeah
21:44<Chutt>and they weren't
21:44<clever>then it would use an ioctl to count them(0 bytes left)
21:44<Chutt>so it was looping
21:44<clever>so it went back to select
21:44<clever>and acording to netstat the socket was closing
21:44<Chutt>tight loop, no way to exit it
21:44<Chutt>100% cpu
21:44<clever>so a simple read() in nonblocking mode, would cause it to read 0 and know its closed
21:44<clever>blocking mode i mean
21:44<clever>but the code is designed in a nonblocking way
21:44<clever>and isnt checking that its actualy closed when theres 0 left
21:45<clever>i suspect if i hangup the upnp conn without warning i could trigger it
21:45<clever>and mythtv may be doing just that with it freq timeouts(damn slow hardware)
21:47<Cardoe>this place is now jumpin
21:47<clever>wee
21:47<clever>core dump
21:48<clever>with a simple typo
21:48<clever>http://192.168.1.60:6543
21:49<knowledgejunkie>clever: thanks so much for the memory pointers (no pun intended :) )
21:50<clever>lol
21:51<clever>#0 0xb5879922 in QString::fromLatin1_helper (str=0x19f <Address 0x19f out of bounds>, size=-1) at tools/qstring.cpp:3165
21:51<clever>its trying to use a QString at 0x19f
21:51<Chutt>heh
21:51<clever>#2 0x0807956f in QString (this=0xae0f7f40, ch=0x19f <Address 0x19f out of bounds>) at /usr/include/qt4/QtCore/qstring.h:371
21:52<clever>#3 0xb6bca200 in MythSocket::readStringList (this=0xac59cb70, list=@0xae0f8190, quickTimeout=false) at mythsocket.cpp:491
21:53<clever>#4 0x080c3fb1 in MainServer::ProcessRequestWork (this=0xb190fa00, sock=0xac59cb70) at mainserver.cpp:301
21:53<Chutt>is that a VERBOSE call in your tree?
21:53<clever>noe in this backtrace:P
21:53<Chutt>no, mythsocket.cpp:491
21:53<clever>ah
21:54<clever>yes
21:54<Chutt>that's rather odd.
21:55<clever>how do i show local variables?
21:55<Chutt>in gdb?
21:55<clever>yep
21:55<Chutt>um
21:55<danielk22>chutt: the cast of a QString to (const char*) is not safe
21:55<Chutt>it's been like 2 years since i've used gdb
21:55<clever>lol
21:56<Chutt>danielk22, aaah
21:56<clever>i'll just bt full
21:56<Chutt>i thought there were warnings for that?
21:56<danielk22>from the Qt4 docs: <<<Note: All the functions in this class are reentrant, except ascii(), latin1(), utf8(), and local8Bit().>>>
21:56<clever>and how often do you pay attention to the 3 compiler warnings in the 2000 lines of compiler output?
21:56<Chutt>clever, fairly often
21:56<danielk22>nah, warnings are for the weak :)
21:57<danielk22>clever: make >/dev/null
21:57<clever>lol
21:57<danielk22>clever: so only the errors & warnings show up...
21:57<clever>ah
21:57<clever>but id need to make clean first
21:57<danielk22>yup
21:57<Cardoe>danielk22: make > /dev/null ftw
21:58<danielk22>but this works with ccache :)
21:58<clever>not exactly
21:58<clever>the libupnp wants to run a svnversion on the entire tree
21:58<clever>which takes 5mins out of every single cimpile
21:58<clever>even if nothing ever changed
21:59<clever>ccache wont make it instant
21:59<danielk22>clever: I can't help that, it doesn't take that long on my machine, but it does slow things down..
21:59<clever>the system holding the recordings has never been fast at disk
21:59<clever>for some weird reason
21:59<clever>and source
21:59<Chutt>probably not a good place to record to :p
21:59<clever>source&recordings&bin&lib are all on a single nfs mount
22:00<clever>source is in /media/mainlv/sharedbuilds/7.10/mythtv/
22:00<clever>configure has --prefix= /media/mainlv/root/7.10/
22:00<clever>recordings to /media/mainlv/mythtv/
22:01<clever>and /media/mainlv/ is the mount point
22:09<Cardoe>Lemme give you a tip.. don't upgrade to gcc 4.3
22:09<Cardoe>I can't build it on 2G of RAM and 2G of swap
22:09<clever>gcc version 4.1.3 20070929 (prerelease) (Ubuntu 4.1.2-16ubuntu2)
22:09<clever>tv_play is a pain to build as it is
22:15<Chutt>need to reduce includes
22:17<clever>-rw-r--r-- 1 mythtv mythtv 260K 2008-06-25 16:15 /media/mainlv/sharedbuilds/7.10/mythtv/libs/libmythtv/tv_play.cpp
22:17<clever>its 8752 lines long
22:17<clever>by itself
22:17<clever>before includes
22:17<Chutt>that's really not much
22:17<clever>i think it needs to go on a diet:P
22:17<Chutt>it's the massive number of files that get included that's the problem
22:18<Chutt>well, files that include other files, etc
22:19<clever>-M away!
22:20<clever>or just grep it out of the existing makefile
22:20<clever>its half a screen long
22:22<danielk22>I find the files with lots of strings like dbcheck.cpp & iso639.cpp take a long time to compile..
22:29-!-danielk22 is now known as danielk_Zzzzz
22:44-!-cesman [n=cecil@pdpc/supporter/sustaining/cesman] has quit [Remote closed the connection]
23:17-!-asmussen [n=asmussen@ip68-2-153-222.ph.ph.cox.net] has joined #mythtv
23:20-!-famicom__ [i=famicom@c51447b09.cable.wanadoo.nl] has quit ["Leaving"]
23:22-!-beandog [n=steve@gentoo/developer/beandog] has quit ["Leaving"]
23:22-!-cesman [n=cecil@pdpc/supporter/sustaining/cesman] has joined #mythtv
23:31-!-rn114 [n=rn114@78-32-214-33.no-dns-yet.enta.net] has joined #mythtv
23:32<clever>knowledgejunkie: overcommit-accounting also has a ton of info i just found(in kernel docs)
23:42-!-poptix [i=poptix@poptix.net] has quit [Read error: 110 (Connection timed out)]
23:48-!-robthebob [n=rn114@78-32-214-33.no-dns-yet.enta.net] has quit [Read error: 110 (Connection timed out)]
---Logclosed Sat Jun 28 00:00:11 2008