IRC log for #maemo-ssu on 20120720

00:38.13*** join/#maemo-ssu taziff1 (~Taziff@cyr108.internetdsl.tpnet.pl)
00:52.17*** join/#maemo-ssu wmarone__ (~wmarone@c-67-174-151-253.hsd1.ca.comcast.net)
00:56.30*** join/#maemo-ssu jon_y (~enforcer@pbz-162-141.tm.net.my)
00:59.06*** join/#maemo-ssu wmarone (~wmarone@c-67-174-151-253.hsd1.ca.comcast.net)
01:52.15*** join/#maemo-ssu taziff (~Taziff@cyr108.internetdsl.tpnet.pl)
02:18.27*** join/#maemo-ssu taziff1 (~Taziff@cyr108.internetdsl.tpnet.pl)
02:51.33*** join/#maemo-ssu amiconn_ (amiconn@rockbox/developer/amiconn)
03:35.36*** join/#maemo-ssu jonwil (~jonwil@27-33-137-199.static.tpgi.com.au)
03:43.16*** join/#maemo-ssu jonwil (~jonwil@27-33-137-199.static.tpgi.com.au)
05:23.42DocScrutinizer05~pbap
06:45.12*** join/#maemo-ssu Pali (~pali@unaffiliated/pali)
06:49.31*** join/#maemo-ssu chem|st (~chemist@svn.unix-ag.uni-kl.de)
07:08.36*** join/#maemo-ssu Guest92502 (~rd@p57B49343.dip0.t-ipconnect.de)
07:08.36*** join/#maemo-ssu _rd (~rd@p57B49343.dip0.t-ipconnect.de)
07:46.01*** join/#maemo-ssu FireFly (~firefly@firefly.xen.prgmr.com)
07:51.02*** join/#maemo-ssu Pali (~pali@unaffiliated/pali)
07:52.34*** join/#maemo-ssu FireFly (~firefly@oftn/member/FireFly)
07:54.43*** join/#maemo-ssu wmarone (~wmarone@c-67-174-151-253.hsd1.ca.comcast.net)
08:38.37*** join/#maemo-ssu luf (~luf@ip-89-103-223-40.net.upcbroadband.cz)
08:41.46PaliI created merge requests on cssu gitorious for all packages pushed to cssu-devel
08:56.54*** join/#maemo-ssu LaoLang_cool (~LaoLang_c@221.226.175.139)
09:45.11*** join/#maemo-ssu arcean (~Arcean@aaep212.neoplus.adsl.tpnet.pl)
10:12.01*** join/#maemo-ssu lardman (~simon@host-92-30-71-136.as13285.net)
10:12.01*** join/#maemo-ssu lardman (~simon@Maemo/community/contributor/lardman)
10:35.36*** join/#maemo-ssu _shadowx (~shadowx@195.24.90.254)
10:47.39*** join/#maemo-ssu lardman (~lardman@Maemo/community/contributor/lardman)
10:50.02_shadowxWhen making a call (gsm call) does the sound stream go throu pulseaudio?
10:52.04_shadowxnever mind :) pulse seems to start using much more CPU when there's active call, so I guess its involved :)
10:57.59*** join/#maemo-ssu M4rtinK (~M4rtinK@mail.melf.eu)
11:03.07tadzikhmm
11:03.23tadzikmaybe there's a way to give microphone a boost through PA then
11:03.56*** join/#maemo-ssu jd (~jade@bas2-montreal33-2925042957.dsl.bell.ca)
11:03.56*** join/#maemo-ssu jd (~jade@unaffiliated/jade)
11:12.04_shadowxmm, yep , shoud be. I'll look into later today.... thou i'm more intrested if i'll be able to make it able to do encrypted calls.... but just PoC , I doubt i'll make it usable for normal users (if a
11:12.39tadzikmy microphone sucks, more and more people complain about it
11:12.50tadzikmaybe it could be fixed somehow on the software side
11:14.08_shadowxit seems so
11:14.08_shadowxhttp://ubuntuforums.org/showthread.php?t=1454078
11:14.21_shadowxcheck out this. ... post #3
11:15.43tadzikoh, looks cool
11:15.48_shadowxso .... it shoud be possible and easy to do ... I'll check it out later today , when I get home, and tell you if it works. and maybe start topic on the forum so other ppl can do it it they need to (if there isnt already one )
11:18.12tadzikalthough there's no pacmd on maemo it seems
11:18.30tadzik_shadowx: there are a few forum topics on that, mine included
11:19.07tadzikhttp://talk.maemo.org/showthread.php?t=83613
11:19.32tadzikit describes sort of a different thing though, but I'd be fine with just boosting it
11:19.42*** join/#maemo-ssu povbot (~supybot@office.pov.lt)
11:19.42*** mode/#maemo-ssu [+v povbot] by ChanServ
11:20.22tadzikif you can find a nice way to do it I'll be most grateful
11:21.04_shadowxthere is pacmd for maemo
11:21.11_shadowxapt-get install pulseaudio-utils
11:21.12_shadowx^^^
11:21.43*** join/#maemo-ssu jonwil (~jonwil@27-33-137-199.static.tpgi.com.au)
11:22.22*** join/#maemo-ssu Pali (~pali@unaffiliated/pali)
11:22.45tadzikhehe
11:23.01tadzikokay, I'll take a closer look when I get back home today. Thanks!
11:23.12*** join/#maemo-ssu M4rtinK (~M4rtinK@mail.melf.eu)
11:24.30_shadowxnp :)
11:25.17_shadowxhey Pali, I understand i've writen an CSSU uninstaller ;) can I try it ?
11:25.45Pali_shadowx, if you want you can try it
11:32.24_shadowxwhere can I get it ?
11:35.53amiconnFor what reason someone would want to uninstall cssu?
11:38.05_shadowxFor testing purpose only.
11:50.00Paliit is part of cssu-enabler
11:50.05Paliand is in cssu-devel repo
11:53.35_shadowxOkay. Thanks
12:33.35*** join/#maemo-ssu Jade (~jade@unaffiliated/jade)
13:47.53*** join/#maemo-ssu LaoLang_cool (~LaoLang_c@221.226.175.139)
13:55.31*** join/#maemo-ssu DocScrutinizer05 (~HaleBopp@openmoko/engineers/joerg)
13:55.31*** join/#maemo-ssu DocScrutinizer (~halley@openmoko/engineers/joerg)
13:56.28*** join/#maemo-ssu luf (~luf@ip-89-103-223-40.net.upcbroadband.cz)
13:57.05luf_shadowx: How to use pacmd? It returns to me: No PulseAudio daemon running, or not running as session daemon.
13:58.27merlin1991Pali: I've fixed the false positives yesterday, uninstalling works all the way now
13:58.42Palimerlin1991, yes I saw on gitorious
13:58.55PaliI created all merge requests
13:58.59Palifor all packagesä
13:59.26merlin1991I'll go over them as soon as possible
13:59.49Paliok
13:59.54Paliall are alredy in cssu-devel
15:05.52luf:) I get pacmd working. For record: apt-get install pulseaudio-utils pulseaudio-module-extra; pactl load-module module-cli-protocol-unix; PULSE_RUNTIME_PATH=/var/run/pulse pacmd
15:39.52*** join/#maemo-ssu freemangordon (~freemango@130-204-50-168.2074221835.ddns.cablebg.net)
15:48.07ShadowXbtw, does the nokia stock kernel support 'noop' IO scheduler ?
15:51.17ShadowXand if yes, has it been discussed to make 'noop' the default IO scheduler in CSSU ?
15:52.15amiconnWhy would someone use any other scheduler on flash based storage?
15:52.21ShadowJKI wouldn't make it default
15:53.14ShadowJKCFQ in maemo5 performs vastly better than noop scheduler that was default in maemo4
15:53.22ShadowXnoop is faster on non  spinning devices , it is simples - so less operations - less power.
15:53.32ShadowJKNeither are optimal, of course
15:53.45ShadowXShadowJK, mm.... maybe i'm wrong , but i've been switching between noop and cfq for the last few days and noop seems better.
15:54.54ShadowJKWe don't have SSDs though. The flash storage behaves more like a tape drive :)
15:55.15*** join/#maemo-ssu NIN101 (~NIN@p5DD28AF3.dip0.t-ipconnect.de)
15:55.40ShadowJKIt really, really, really, really hates random access writes
15:56.22ShadowXmm..can't agree that the flash storage is closer to tape drive then to ssd , specialy when it comes to IO scheduling
15:56.36ShadowJKrandom 4k writes yield less than 50kbytes/sec, whereas pure sequential write flies along at close to 8-10 megabytes/sec :)
15:56.51freemangordonI've tried noop when I was playing with different scheduler settings and ca confirm that noop is much worse than cfq
15:57.08freemangordons/ca /can /
15:57.50ShadowXmaybe some other changes i've made are making the difference in my case....
15:57.54ShadowJKSo in that sense, the most optimal would be to always write sequentially with no seeking, like a tape drive
15:58.37freemangordoninitially it looks faster, but after a while device starts to stutters very badly
15:58.54freemangordons/stutters/stutter/
15:59.37ShadowXfreemangordon, after how much time  ~ ? can the swap fragmentation be the reason ?
16:00.44freemangordonNFC, but the end effect is bad
16:00.53ShadowX'cose i'm re-swaping few times a day and maybe that's why its behaving better in my case.
16:01.03ShadowXOk, thanks for the info :)
16:01.10ShadowJKAlready if it's not delaying/batching writes, and constantly "switching" between writing to emmc, ext, swap..
16:01.28ShadowJKthat'd cause severe slowdown
16:02.23ShadowXbtw , with what kernel was maemo4 ?
16:04.10ShadowJKbtw, bought a new sandisk 32M card, the block size is 4M. 4!! Megabytes!! This means that if you write 512 bytes at start of card, and for example 512 bytes in middle of card, the card will internally read 4M, erase 4M, write 4M, twice!
16:04.58ShadowJKSequential write about 10M/s on that thing, but due to 4M block size, once you hit swap fragmentation (or mix swap / filesystem accss on it), it' be like hitting a brick wall :)
16:07.48ShadowX32M card...4M blocksize , so 8 blocks only?
16:08.09amiconns/M/G/ I guess...
16:08.28ShadowXthat makes more sense :)
16:08.44jon_y4G blocks?
16:08.54amiconnhahaha
16:09.18amiconndidn't say s/M/G/g
16:11.26ShadowJK32G
16:11.48ShadowJK:)
16:12.05ShadowJKGetting LogFS or similar instead of ext3 would be interesting :)
16:12.13jon_ywell, 4G cells might make sense if you have 32TB :)
16:13.01ShadowXShadowJK, btw it looks like maemo5 is using the defualt IO scheduler. as was maemo4. If my search was right, maemo 4 was shipped with 2.6.16 kernel.
16:13.37ShadowXcfq was included in version 2.6.5 if i remember correctly and became default IO scheduelr around 2.6.20
16:14.08ShadowXwitch dosnt mean that 'noop' is better :) just saying , maybe that's the reason why maemo4 uses it , and maemo5 uses cfq.
16:20.03ShadowXbtw guys, sorry if I ask stupid question from time to time ....i'm still new to maemo and want to learn :D ( I got my first n900 only ~10 days ago...)
16:20.54ShadowXp.s. fsck nokia! now i'm thinking of getting second n900 :S
16:21.38ShadowJKI'm pretty sure AS was default before CFQ became the default?
16:21.49ShadowJKand before that, surely deadline was default
16:22.15ShadowXhttp://www.linuxinsight.com/cfq_to_become_the_default_i_o_scheduler.html
16:22.20ShadowXhttp://wiki.maemo.org/Compiling_on_the_N810
16:23.14ShadowJKbtw, I set /sys/block/..../queue/iosched/slice_idle to 0, to see if I can get more than month or two uptime
16:23.58ShadowXooo 1m uptime on the phone will be nice :)
16:24.02ShadowXwhat's your current best?
16:25.14ShadowJKmore than month, less than two
16:25.29ShadowXbtw you'll probably like this (if your an uptime fan)
16:25.29ShadowXroot@monitor:~# uptime
16:25.30ShadowX<PROTECTED>
16:25.52ShadowJKNot counting spare N900 that just sits idling :P
16:27.40amiconnrarely gets even close to such uptimes (which are overrated anyway)
16:28.14amiconnKernel updates kill uptimes ;)
16:28.43ShadowXamiconn, true , but how can I reboot the box :? will you have the hear to reboot box with 1k+ days uptime ? :)
16:29.04amiconnpffft
16:29.56amiconn$ uptime
16:29.56amiconn<PROTECTED>
16:30.09amiconnLast reboot was due to sata problem
16:30.18ShadowXthis is homebox ?
16:30.41amiconnThis is my homeserver/workstation combo
16:31.06amiconn(and yes, the load average is not a mistake. dnetc running on a Core i7)
16:31.34ShadowXi thought the load was 'cose youre compiling something ;D
16:32.12ShadowXbtw load 8 is not much ..can't even notice load lower then 10-15 on most servers (form the user point of view)
16:33.31amiconnWell it depends on the type of load. This one isn't noticeable because dnetc runs at nice 19
16:34.02ShadowXoh ...in that case - sure :)
16:34.02ShadowJKIt's not that I'm an uptime fan, I just hate unexpected/forced reboots
16:34.08ShadowXbtw why did you get i7 ?
16:34.50amiconnWhy not? Good performance at quite low power
16:35.04amiconnThis is an i7-2600S, so 65W TDP
16:35.56amiconnMy N900 usually gets to about 14 days uptime before it forces me to reboot (due to apps misbehaving)
16:36.13ShadowXi'm just asking :) curius why ppl give money for i3,i5,i7 ..
16:36.38ShadowXnot that its bad cpu .... its great ...
16:36.45ShadowXbut cpu power / price .... :/
16:36.54amiconnWell, what else would you choose? AMD is way slower...
16:37.08amiconndoesn't deem 275€ expensive
16:37.16amiconnThat was a year ago btw
16:38.20ShadowXwell ... its not expensive , but my point was, AMD for the same money shoud give you at least a bit more power ..and probably will drain more power too :)
16:39.19amiconnused to recommend AMD back in P4 times, but that changed when Core 2 arrived
16:39.20ShadowXdon't get me wrong i like intel :) i'm surronded only by Xeon's (with few minior expections) but AMD not that bad and most of the times you get more power for the same amount of moeny.
16:39.51amiconnAnd invest the saved money into electricity bill instead ;)
16:40.10amiconnMy machine is running 24/7
16:43.44ShadowXMine 2. Both my boxes are running non 24/7 and have always been.
16:45.53*** join/#maemo-ssu luf (~luf@ip-89-103-223-40.net.upcbroadband.cz)
17:32.50*** join/#maemo-ssu wmarone (~wmarone@c-67-174-151-253.hsd1.ca.comcast.net)
17:34.22*** join/#maemo-ssu wmarone (~wmarone@c-67-174-151-253.hsd1.ca.comcast.net)
17:42.36*** join/#maemo-ssu arcean_ (~Arcean@aaer176.neoplus.adsl.tpnet.pl)
18:03.24*** join/#maemo-ssu andre__ (~andre@ip-89-176-24-140.net.upcbroadband.cz)
18:03.24*** join/#maemo-ssu andre__ (~andre@Maemo/community/bugmaster/andre)
18:03.25tadzikwow, since when the settings work in portrait mode?
18:03.30tadzik++ to whoever did this
18:17.16*** join/#maemo-ssu nox- (noident@freebsd/developer/nox)
18:29.39lufI feel like a winner :)
18:41.42freemangordonluf: it's good to be a king :P
18:41.53lufSure :D
18:42.00freemangordonyou know that movie? history of the world?
18:42.04lufPBAP working with 3 exceptions.
18:42.17lufI'm not sure.
18:42.26lufMaybe I don't know.
18:42.59lufIt ignore first VCARD. It seems has a problem no empty line after photo and 1 ghost :D
18:43.24lufIt add "A" at the end of 1 contact's phone number.
18:44.05lufNo mess in numbers no randomely missing contacts.
18:44.16lufIt's more than I expect.
18:44.47lufBut it doesn't try to get call history from phone :( Maybe it doesn't present such feature ...
19:09.14*** join/#maemo-ssu wmarone (~wmarone@c-67-174-151-253.hsd1.ca.comcast.net)
19:26.25*** join/#maemo-ssu Guest92502 (~rd@p57B49343.dip0.t-ipconnect.de)
19:26.26*** join/#maemo-ssu _rd (~rd@p57B49343.dip0.t-ipconnect.de)
19:49.26*** join/#maemo-ssu M4rtinK (~M4rtinK@mail.melf.eu)
21:20.19lufShit. I need consultation.
21:20.26luffreemangordon, merlin1991 ?
21:20.30lufSomeone else?
21:21.14lufI imagine all 3 mistakes. I'm not sure if it's only problem of my HF so I don't know how to fix it.
21:37.30merlin1991luf: pong
21:39.33lufmerlin1991: The A suffix for number is when the contact has X-SKYPE defined.
21:39.56lufIt seems to me like annoying bug in my HF :(
21:40.49luf1) Suppress X-SKYPE on eds side isn't good idea
21:41.18luf2) Suppress X-SKYPE in obex is to scan whole VCARD string
21:41.42luf3) Ignore it means I can't use skype contacts :(
21:42.09lufOf course I'll try to contact HF manufacturer but I don't expect they fix it :(
21:42.45lufI see no configuration possibility to enable/disable SKYPE supress on demand.
21:46.42lufIt seems only 2 is option for me.
21:53.21merlin1991luf: atm I'm quite pissed so I'll think over it tomorrow? :D
22:05.14lufmerlin1991: ok. BTW what about curl?
22:06.38merlin1991didn't have the chance today, but tomorrow I should be able todo everything
22:12.17lufmerlin1991: I hope you mean today (Saturday) not tomorron (Sunday) :D
22:14.19merlin1991yeah :D
22:14.23merlin1991time is relative ;)
22:16.17luf:D
23:25.31*** join/#maemo-ssu taziff (~Taziff@cyr108.internetdsl.tpnet.pl)
23:28.08*** join/#maemo-ssu taziff (~Taziff@cyr108.internetdsl.tpnet.pl)
23:54.57*** join/#maemo-ssu ZogG (~zoggrules@109.65.230.179)

Generated by irclog2html.pl Modified by Tim Riker to work with infobot.