IRC log for #maemo-ssu on 20120613

00:01.09merlin1991osso-wlan or osso-wlancond?
00:02.34okiaswhat I written exactly
00:03.18merlin1991ah yeah it actually contains wlancond
00:03.37merlin1991might interrupt wlan conenctivity when you replace the package
00:04.53okiashaha, and what about "apt"? :-D
00:05.05okiasthat could interrupt further apt-get. right? :-D
00:05.21okiasI guess manual removing package is not acceptable :-D
00:05.40merlin1991yea
00:05.46merlin1991just do it as the last one
00:06.02okiasmerlin1991: do you use lastest cssu?
00:06.09merlin1991ie do it together with installing mp-fremantle-generic-pr
00:06.15merlin1991okias: ofc I do :D
00:06.25okiasmerlin1991: maybe, if you could bump all installed packages and I'll install which missing
00:07.29merlin1991I suppose you mean dump and not bump? :D
00:07.51okiasyeah
00:08.07okiassorry. it's pretty late here in czech republic (02.08)
00:08.22merlin1991no worries, I'm from vienna :)
00:08.29okiasLOL
00:08.31okiasOMG
00:08.35okiasproblem FIXED
00:08.54merlin1991huh?
00:08.54okiashttp://paste.debian.net/174199/ check it! :-D
00:09.33okiasI guess deinstall says I removed it... so it include even packages accidentally removed
00:09.34okias:-)
00:10.48merlin1991hm my list seems to be too big
00:11.14okiasmerlin1991: no problem, I got it! :-)
00:11.35merlin1991you sure you don't want to check against a valid list?
00:12.10okiasno problem, I have small clue what is what ;-) but if you have it, upload it and I'll check it for sure!:-)
00:12.28merlin1991btw how did you generate that list?
00:13.59okiasdpkg --get-selections AND deinstall mean, that I removed package, but not PURGEd ;-) so that's all package which I had installed
00:15.15merlin1991http://paste.debian.net/174201
00:15.52merlin1991that's from a n900 on cssu-thumb, so just ignore the kernel-cssu* packages if you're on testing
00:19.20okiasso at least I didn't removed so much packages ;-)
00:20.37okiasmerlin1991: thanks for list, I'll do diff ;-)
00:24.31okiasoh still some packages missing! again - thanks
00:25.03merlin1991bear in mind some of those packages are simply things from extras ;)
00:29.39okiasmerlin1991: yeah, I browser it carefully, but for example cifs modul I must have! :-D
00:31.45okiasmerlin1991: btw. does have nokia some dependency tracking? is any way to figure if I need every "libraries and support packages"?
00:32.10merlin1991phew
00:32.42merlin1991I guess the've put some thought into the metapackge, so that stuff should be neccesary
00:52.56*** join/#maemo-ssu ekze (nyan@2002:5046:e474:1::3)
01:03.15*** join/#maemo-ssu amiconn_ (amiconn@rockbox/developer/amiconn)
01:26.48*** part/#maemo-ssu okias (~okias@2001:1ab0:7e1e:e005:dead::feed)
01:40.59*** join/#maemo-ssu okias (~okias@2001:1ab0:7e1e:e005:dead::feed)
01:41.36okiasis possible to remove hildon-welcome ?
01:42.13okiaswithout breaking boot. I just dont need start video sequence
02:43.53*** join/#maemo-ssu amiconn_ (quassel@rockbox/developer/amiconn)
02:51.39*** part/#maemo-ssu okias (~okias@2001:1ab0:7e1e:e005:dead::feed)
06:47.07*** join/#maemo-ssu freemangordon (freemangor@130-204-50-168.2074221835.ddns.cablebg.net)
07:38.39*** join/#maemo-ssu arcean (~arcean@aadd136.neoplus.adsl.tpnet.pl)
07:43.02*** join/#maemo-ssu M4rtinK2 (~M4rtinK@ip-89-102-207-166.net.upcbroadband.cz)
08:02.35*** join/#maemo-ssu MrPingu (~chatzilla@86.92.226.97)
08:07.10*** join/#maemo-ssu arcean_ (~arcean@aacz183.neoplus.adsl.tpnet.pl)
08:09.10*** join/#maemo-ssu Guest8322 (~dafox@ip51cc571d.speed.planet.nl)
08:52.28*** join/#maemo-ssu dafox (~dafox@ip51cc571d.speed.planet.nl)
10:14.45*** join/#maemo-ssu dafox (~dafox@ip51cc571d.speed.planet.nl)
10:21.51chem|stfreemangordon: merlin1991 how far are we from getting libcurl3 to cssu-stable?
10:23.06chem|stI don't want DEUs to brake their installation just because they can't wait for it any longer
10:49.43DocScrutinizer51good call
10:58.09chem|stty
11:45.44*** join/#maemo-ssu BCMM (~ben@unaffiliated/bcmm)
12:02.04freemangordonchem|st: it is not me to ask, it all depends on merlin1991, MohammadAG (as we should put it first in -testing) and luf. If you see a way I can speedup it, please tell me.
12:12.41merlin1991DEUs?
12:13.06merlin1991also, if you have cssu already installed you can upgrade with the package from extras
12:13.09Lava_Croftdumb end user?
12:13.12Lava_Croftjust a wiiiiild guess
12:13.19Lava_Croft<----
12:16.09*** join/#maemo-ssu freemangordon (freemangor@130-204-50-168.2074221835.ddns.cablebg.net)
12:29.45*** join/#maemo-ssu DocScrutinizer05 (~HaleBopp@openmoko/engineers/joerg)
12:29.58*** join/#maemo-ssu DocScrutinizer (~halley@openmoko/engineers/joerg)
13:45.56chem|stmerlin1991: ok so it is only a matter for non-cssu users?
13:46.05merlin1991yep
13:46.14chem|stok
13:52.22chem|stso it isn't in maemo-extras but in cssu repo?!
13:53.06chem|stI lack some understanding here
14:03.49freemangordonDocScrutinizer05:ping
14:27.24merlin1991chem|st: it is in maemo-extras
14:27.33merlin1991but can only be installed if you have cssu already
14:27.39merlin1991--> shouldn't actually be there
14:30.22*** join/#maemo-ssu BCMM (~ben@unaffiliated/bcmm)
14:47.18*** join/#maemo-ssu dafox (~dafox@ip51cc571d.speed.planet.nl)
15:34.57*** join/#maemo-ssu Woody14619 (~Woody@Maemo/Community/council/Woody14619)
15:55.49*** join/#maemo-ssu xmlich02 (~imlich@2001:67c:1220:80c:0:1e0:bd4e:949c)
15:58.39*** join/#maemo-ssu lizardo (lizardo@nat/indt/x-mvnitwuouikxrgmr)
16:10.52*** join/#maemo-ssu NIN101 (~NIN@p5DD289C6.dip0.t-ipconnect.de)
16:54.16chem|sthehe
16:56.06*** join/#maemo-ssu luf (~luf@ip-89-103-223-40.net.upcbroadband.cz)
16:57.54lufmerlin1991: is the format-patch for curl cssu enough or have i to use only gitorious to create merge request?
16:58.24merlin1991luf: I only dl it so far, I'll inspect it when I've finsihed my uni assignment today
16:59.19lufOk. No problem. I see no response yesterday and archive stopped on tuesday 1:30 am or so.
17:02.15lufmerlin1991: FYI just the last commit contains my changes to build on maemo. All previous is git://git.debian.org/collab-maint/curl.git
17:04.19*** join/#maemo-ssu BCMM (~ben@unaffiliated/bcmm)
17:11.04DocScrutinizer05freemangordon: pong
17:31.06merlin19917 parts down, 3 to go
17:31.12merlin1991and my head is spinning by now
17:33.08freemangordonDocScrutinizer05: it turned out that rootfs is slower than /opt according to "time cat $VERY_BIG_FILE > /dev/null"
17:33.19freemangordonon reading at least
17:33.49freemangordonany comments, as I am going to move bat Qt to /opt for CSSU-thumb
17:33.53freemangordon*back
17:33.55DocScrutinizer05quite possible
17:34.37freemangordonnot only that , but when doing  cat from rootfs, we have approximately 90% sys time
17:35.03freemangordon(and My device is OC to 805)
17:35.32RaimuIs it the file system compression that peaks the CPU?
17:35.36freemangordonso on a stock device it will be even slower
17:35.46freemangordonlooks like that is
17:35.54RaimuOr something kinkier?
17:35.57Woody14619root is ubfs, no?
17:36.03freemangordonyep
17:36.10DocScrutinizer05:nod:
17:36.11freemangordon*ubifs
17:36.28Woody14619forgot the i... am so ashamed. :)
17:36.34DocScrutinizer05while 7opt is err, prinmarily a plain blockdevice
17:36.47freemangordonyeah
17:37.19DocScrutinizer05with an own controller
17:37.37freemangordonok, I am going AFK, if you can think of any reason why to not return Qt back to /opt, please ping me
17:37.50DocScrutinizer05won't happen
17:38.09DocScrutinizer05IOW I'm sure there are no such reasons
17:38.44RaimuWhat kinda things apart from things you really need at boot should be in the rootfs, anyway?
17:38.45*** join/#maemo-ssu M4rtinK2 (~M4rtinK@89.102.207.166)
17:38.53DocScrutinizer05NAND should be used for fast logging etc
17:39.24DocScrutinizer05which it is, aiui (/var/log/*
17:39.26DocScrutinizer05)
17:40.20DocScrutinizer05after all afaik you can write sible bytes on NAND
17:40.25DocScrutinizer05single*
17:41.31DocScrutinizer05and if UBI is done by a smart hacker, they do incremental writing, that never needs 1->0 bit transitions
17:45.03DocScrutinizer05define EOF as 0x0000, escape char(0) as char(1) char(0), ecape char(1) as char(1) char(1)
17:46.17DocScrutinizer05will require reading whole logfile to get the true length, but then you can append-write single chars without any read-modify-erase-write overhead
17:47.36DocScrutinizer05define last 4 words of each block to be pointer to next block of same file
17:48.02DocScrutinizer05have a simple dos-like FAT that has a 1 for each used block and a 0 for free block
17:48.37DocScrutinizer05define 0xFFFF a deleted item
17:49.05DocScrutinizer05so you could simply delete pointers and write new addr to next place in hashtable
17:49.57DocScrutinizer05but AIUI UBI does all that already - otherwise I wouldn't know what's that magic garbage collection that happens during boot-time
18:13.50*** join/#maemo-ssu Atarii (~Atarii@77.107.156.213)
18:13.50*** join/#maemo-ssu Atarii (~Atarii@unaffiliated/atarii)
18:40.02*** join/#maemo-ssu trbs (~trbs@2001:470:d2ad:1:4a5b:39ff:fe7d:1623)
18:43.14ShadowJKonboot gc? aren't you thinking about jffs2?
18:45.32DocScrutinizer05nah, I think N900 root has more free space after boot
18:46.24DocScrutinizer05doesn't it?
18:47.26ShadowJKthat'd also happen if you upgraded apps/libs
18:47.27ShadowJKbut it's probably the free space estimate getting reset on boot
18:50.00freemangordonfun: qt-mobility 1.2 is build with no optimisations (i.e. debug)
18:55.27*** join/#maemo-ssu luf (~luf@89.103.223.40)
18:55.38RaimuReason or randomness?
18:56.39ShadowJKatleast you could build it with both opt and dbg
18:56.47ShadowJKor does that fail for c++?
18:58.07freemangordonRaimu: NFC
18:58.52freemangordondebian/rules sets DEB_CONFIGURE_EXTRA_FLAGS := -debug -silent
18:59.00RaimuHeh.
18:59.51freemangordonShadowJK: didn't get "opt and dbg"
19:00.31freemangordonMohammadAG, merlin1991: ^^^
19:01.18ShadowJKBuild it with both optimizations and debug
19:01.21ShadowJKAren't debug symbols like bundled in a separate package in maemo anyway..
19:02.13freemangordonShadowJK: I know that, just stating the fact that qt-mobility in both CSSU and extras is build in that way
19:02.33freemangordonediting configuration
19:02.37freemangordonright now
19:05.58merlin1991doing paging by hand step by step is pita
19:07.12freemangordonanyone knows who atilla77@TMO is?
19:12.27freemangordonRaimu: I agree with what you wrote at TMO, but unfortunately I have only 24 hours per day :). I will be glad to see anyone doing that measurement in a proper way
19:13.49RaimuYeah. :) I can help, but mostly when receiving orders, because I'm a noob on fire when it comes to benchmarking Linux file systems.
19:14.20freemangordonRaimu: And what exactly makes you think I am not :D
19:14.51Raimu:D
19:43.33*** join/#maemo-ssu nox- (noident@87.144.3.18)
19:43.45*** join/#maemo-ssu nox- (noident@freebsd/developer/nox)
19:44.04*** join/#maemo-ssu DocScrutinizer05 (~HaleBopp@openmoko/engineers/joerg)
19:44.55*** join/#maemo-ssu DocScrutinizer05 (~HaleBopp@openmoko/engineers/joerg)
19:56.11*** join/#maemo-ssu M4rtinK2 (~M4rtinK@46.227.13.46)
19:59.44*** join/#maemo-ssu DocScrutinizer05 (~HaleBopp@openmoko/engineers/joerg)
21:12.02*** join/#maemo-ssu smoku (~spectrum@xkh0g2.infr.xiaoka.com)
21:58.47*** join/#maemo-ssu DocScrutinizer05 (~HaleBopp@openmoko/engineers/joerg)
22:31.35*** join/#maemo-ssu M4rtinK2 (~M4rtinK@ip-89-102-207-166.net.upcbroadband.cz)
22:34.15*** part/#maemo-ssu smoku (~spectrum@xkh0g2.infr.xiaoka.com)
23:36.43*** join/#maemo-ssu int_ua (~int_ua@ip-602f.proline.net.ua)

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