IRC log of #maemo-ssu for Wednesday, 2013-07-03

sixwheeledbeastcan anybody tell me or point me to the changes to boot time in CSSU-T since 7.2. I have had a number of reports of stripping swap on start up. Has rcS-late changed?00:03
*** discopig has joined #maemo-ssu00:11
*** discopig has quit IRC00:11
*** discopig has joined #maemo-ssu00:11
*** arcean_ has joined #maemo-ssu00:24
*** tg has quit IRC00:24
*** arcean has quit IRC00:26
*** tg has joined #maemo-ssu00:32
RaimuI bet it has redone rcS-late and altered the fstabbing habits.00:57
sixwheeledbeastit seems this way. Updating to T8.2 on test device to try an reproduce.01:02
sixwheeledbeastIt's uninstalled lock-applet... grr01:06
DocScrutinizer05yes, but you can simply install it again01:38
DocScrutinizer05and yes, T8.x massively changed stuff in fstab and iirc changed rcS* for that. And it seems this change hasn't been sufficiently QA tested against all relevant possible system configs we're facing out there, heck not even against removing uSD card originally01:41
DocScrutinizer05I guess a lot of the reswap and whatnot-swap packages will conflict with this fstab patch, as well as most if not all of customized filesystems and/or partitioning01:43
*** kolp has quit IRC01:48
RaimuFor me it was a matter of rewriting "ext4" into a bunch of places again and rewriting in some of the swap-specific things I'd added on my own.01:55
RaimuFrom what I gather it should be easier from now on to keep those customized parts intact as long as those packages get an update.01:56
RaimuNot that I'm assuming they necessarily will.01:57
DocScrutinizer05yes, it sounds like a solution to the problem it creates02:01
DocScrutinizer05as in "next time you'll not have this problem again, you got this time caused by the solution getting implemented"02:02
DocScrutinizer05like a surgery that ensues 4 weeks in hospital, with the purpose to allow future operations on the very same issue will only need 1..2 weeks02:07
Raimu*laugh*02:07
RaimuYeah...02:07
* DocScrutinizer05 ponders cancellation insurance cancellation insurance02:12
RaimuLike, if you're insured for the situation that your insurance for the cancellation fees is itself canceled?02:12
DocScrutinizer05yeah, for the situation when you want to cancel your cancellation fee insurance02:14
RaimuIt's sad how insurance companies like making insurance getting a ridiculous game.02:14
DocScrutinizer05when you buy it, it will ensure you against costs to get rid of it02:15
RaimuHaha.02:15
DocScrutinizer05but of course only when you paid it02:15
*** aap__ has quit IRC02:18
DocScrutinizer05solution to the generic immanent problem, self referential02:18
Raimu:)02:19
*** aap has quit IRC02:19
RaimuNight o/02:20
*** futpib has quit IRC02:22
*** discopig is now known as bromide02:24
*** unclouded has joined #maemo-ssu02:33
*** arcean_ has quit IRC02:49
*** LauRoman has quit IRC02:54
*** M4rtinK has quit IRC03:09
*** LaoLang_cool has joined #maemo-ssu03:19
*** LaoLang_cool has quit IRC03:25
*** povbot_ has joined #maemo-ssu04:04
-barjavel.freenode.net- [freenode-info] channel flooding and no channel staff around to help? Please check with freenode support: http://freenode.net/faq.shtml#gettinghelp04:04
*** ChanServ sets mode: +v povbot_04:04
*** tom____ has joined #maemo-ssu04:06
*** DocScrutinizer51 has quit IRC04:10
*** RST38h has quit IRC04:10
*** Macer has quit IRC04:10
*** povbot has quit IRC04:10
*** tom has quit IRC04:10
*** DocScrutinizer51 has joined #maemo-ssu04:12
*** Macer has joined #maemo-ssu04:15
*** amiconn_ has joined #maemo-ssu05:01
*** amiconn has quit IRC05:01
*** amiconn_ is now known as amiconn05:01
*** nox- has quit IRC05:11
*** RST38h has joined #maemo-ssu05:50
*** amiconn has quit IRC05:59
*** amiconn_ has joined #maemo-ssu05:59
*** amiconn_ is now known as amiconn05:59
*** psycho_oreos has joined #maemo-ssu06:26
*** tg has quit IRC06:43
*** sixwheeledbeast has quit IRC06:53
*** sixwheeledbeast has joined #maemo-ssu06:57
*** psycho_oreos has quit IRC07:03
*** psycho_oreos has joined #maemo-ssu07:05
*** psycho_oreos has quit IRC07:09
*** sixwheeledbeast has quit IRC07:13
*** sixwheeledbeast has joined #maemo-ssu07:34
*** sixwheeledbeast has quit IRC07:54
*** sixwheeledbeast has joined #maemo-ssu07:59
*** sixwheeledbeast has quit IRC08:10
*** M13 has joined #maemo-ssu08:13
*** sixwheeledbeast has joined #maemo-ssu08:31
*** bromide has quit IRC08:37
*** discopig has joined #maemo-ssu08:40
*** sixwheeledbeast has quit IRC08:43
*** freemangordon has joined #maemo-ssu08:46
*** sixwheeledbeast has joined #maemo-ssu08:50
*** futpib has joined #maemo-ssu08:53
*** tg has joined #maemo-ssu09:02
*** sixwheeledbeast has quit IRC09:06
*** Pali has joined #maemo-ssu09:07
*** sixwheeledbeast has joined #maemo-ssu09:12
*** oldtopman has quit IRC09:12
*** LauRoman has joined #maemo-ssu09:18
*** discopig has quit IRC09:43
*** tg has quit IRC09:43
*** sixwheeledbeast has quit IRC09:46
*** sixwheeledbeast has joined #maemo-ssu09:57
*** infobot has quit IRC09:58
*** XDS2010_ has quit IRC10:00
*** freemangordon_ has joined #maemo-ssu10:00
*** freemangordon has quit IRC10:00
*** sixwheeledbeast has quit IRC10:02
*** XDS2010_ has joined #maemo-ssu10:02
*** sixwheeledbeast has joined #maemo-ssu10:11
*** discopig has joined #maemo-ssu10:26
*** s1gk1ll_ has joined #maemo-ssu11:07
*** s1gk1ll has quit IRC11:09
*** discopig has quit IRC11:15
*** discopig has joined #maemo-ssu11:15
*** Martix_ has quit IRC11:23
*** futpib_ has joined #maemo-ssu11:27
*** futpib has quit IRC11:27
*** g3kk3r has quit IRC11:30
*** g3kk3r has joined #maemo-ssu11:30
*** s1gk1ll_ is now known as s1gk1ll11:35
*** tg has joined #maemo-ssu11:47
*** M4rtinK has joined #maemo-ssu11:47
*** M4rtinK has quit IRC12:26
*** freemangordon_ has quit IRC12:30
*** freemangordon has joined #maemo-ssu12:38
*** tg has quit IRC12:42
*** jon_y has quit IRC12:46
*** discopig is now known as bromide12:50
*** kolp has joined #maemo-ssu13:20
*** LauRoman has quit IRC13:23
*** tg has joined #maemo-ssu13:23
*** lizardo has joined #maemo-ssu13:25
merlin1991Pali: freemangordon ping14:17
freemangordonmerlin1991: pong14:17
Palimerlin1991: pong14:17
merlin1991since we don't want to stop freemangordon from celebrating his birthday I guess we can move the meeting to tomorrow?14:18
freemangordonmerlin1991: well, I am busy tomorrow14:19
freemangordonso lets make it today, I 'll participate as much as I can14:19
merlin1991I'd rather pick a day you can fully participate since a lot of the points were brought in by you14:20
freemangordonmerlin1991: not exactly, kcssu, glibc and bug triaging are not exactly mine14:21
freemangordonbut ok14:21
merlin1991hrmpfh, my sisters n900 went down with a broken usb port14:22
sixwheeledbeastfreemangordon: Happy Birthday :D14:23
merlin1991freemangordon, Pali so whatabout the rest of the week (starting with friday) when can you be around?14:23
freemangordonsaturday?14:24
PaliI will be without internet connection from friday to monday14:24
merlin1991we do have a great sense of timing, don't we?14:25
freemangordon:D14:32
merlin1991I guess the best option is today then14:32
freemangordonyeah14:32
merlin1991ah well :/14:32
*** M13 has quit IRC14:45
Drathirfreemangordon: happy birthday from me too...15:41
Drathirmerlin1991: any chance to fix usb?15:42
merlin1991I'm going to solder in a new usb port15:42
Drathirmerlin1991: can i ask about price that new one ? or only avaible take off fron other n900?15:43
merlin19912,8$ for  a set of 20 :D15:44
merlin1991http://dx.com/p/diy-micro-usb-5-pin-female-smt-socket-connector-silver-20-piece-pack-13413215:44
merlin1991hm, I'm going mental here, the friggin microsd / lightsensor pcb won't go off15:52
*** M13 has joined #maemo-ssu15:54
merlin1991ah ffs it's just wayyyy too strong flue15:55
merlin1991*glue*15:55
*** M13 has quit IRC16:23
*** oldtopman has joined #maemo-ssu16:32
*** M13 has joined #maemo-ssu16:41
Drathirmerlin1991: go inside n900 si veru hard?16:42
Drathiris*16:42
*** jon_y has joined #maemo-ssu16:45
*** M13 has joined #maemo-ssu16:59
*** M13 has quit IRC17:01
*** M13 has joined #maemo-ssu17:02
*** M13 has quit IRC17:04
*** M13 has joined #maemo-ssu17:04
merlin1991not really17:04
*** futpib_ has quit IRC17:11
*** futpib_ has joined #maemo-ssu17:17
*** freemangordon has quit IRC17:30
Drathirmerlin1991: /me need too little improve my usb port to but n900 is my main and not yet before try go inside...17:30
*** M13 has quit IRC17:31
*** M13 has joined #maemo-ssu17:31
Drathirimprove mean try to better fit port size because i think for all this time using of become little bigger than before...17:32
*** freemangordon has joined #maemo-ssu17:41
*** sixwheeledbeast has quit IRC18:32
*** sixwheeledbeast has joined #maemo-ssu18:42
*** sixwheeledbeast has quit IRC18:50
*** sixwheeledbeast has joined #maemo-ssu18:51
*** sixwheeledbeast has left #maemo-ssu18:52
*** sixwheeledbeast has joined #maemo-ssu18:54
*** sixwheeledbeast has quit IRC18:59
*** sixwheeledbeast has joined #maemo-ssu19:20
*** sixwheeledbeast has quit IRC19:27
*** NIN101 has joined #maemo-ssu19:39
*** Vlad_on_the_road has joined #maemo-ssu20:11
*** Sc0rpius has quit IRC20:44
*** Sc0rpius has joined #maemo-ssu20:46
*** Sc0rpius has quit IRC20:59
*** arcean has joined #maemo-ssu21:02
*** Sc0rpius has joined #maemo-ssu21:12
*** futpib_ has quit IRC21:29
*** Martix_ has joined #maemo-ssu21:30
*** M4rtinK has joined #maemo-ssu21:35
*** M13 has quit IRC21:37
Palimerlin1991: is meeting today?21:40
*** Sc0rpius has quit IRC21:48
*** Martix_ has quit IRC21:49
*** xes has joined #maemo-ssu21:55
*** Sc0rpius has joined #maemo-ssu21:56
freemangordonPali: yes22:00
Palifreemangordon: ok, when?22:00
freemangordonin 10 seconds from now :D22:00
Paliok :-)22:00
* freemangordon wonders where is merlin199122:01
*** Sc0rpius has quit IRC22:02
*** futpib_ has joined #maemo-ssu22:05
* merlin1991 is here now22:08
merlin1991sorry for the delay, I just got home22:08
merlin1991freemangordon: arcean Pali ping22:09
Palimerlin1991: pong22:09
freemangordonmerlin1991: pong22:09
*** Sc0rpius has joined #maemo-ssu22:10
freemangordonmerlin1991: did you mail luf?22:11
arceanhi22:11
freemangordonSc0rpius: hi :)22:11
freemangordonarcean: hi22:11
arceanfreemangordon: happy birthday :)22:12
merlin1991freemangordon: luf said he has about no time this week and will read the backlog22:12
freemangordonmerlin1991: ok22:12
merlin1991also yep happy birthday :)22:12
freemangordonarcean: thanks22:12
freemangordonthanks22:12
freemangordonmerlin1991: shall we start?22:12
freemangordonI guess we're all here22:12
merlin1991yep looks like, so step 1 gathering of topics22:12
merlin1991I've got "new" packages and the maintainer situation22:13
freemangordonfrom here: KCSSU, glibc and... umm...22:13
* freemangordon tries to rmrmber the 3rd thing22:13
freemangordon*remember22:13
freemangordon<freemangordon> merlin1991: kernel-cssu, glibc and h-d bug. also a little of bug triaging will be good to be done22:14
freemangordonok, h-d bug seems to be fixed22:14
freemangordonso from me it is KCSSU and glibc22:14
freemangordonPali: what is the situation with PA and the missing sounds?22:15
freemangordonmerlin1991: BTW is chem|st aware there is a meeting?22:16
merlin1991uh I just thought of him myself22:16
freemangordon:(22:16
freemangordonmerlin1991: seems Pali is quit today, so lets add e2fsprogs and upstart to the agenda22:17
freemangordon*quiet22:17
Paliwhat is with PA? hm I do not know22:17
PaliI tried to remove both nokia and our patch for decreasing memory allocation size22:17
Paliand for some users it helped and for some not...22:18
freemangordonPali: what is the status, did you find whic is the broken versiona nd what breaks it?22:18
Palineed to look at it22:18
freemangordonPali: could it be my backported patch the problem?22:18
Paliproblem is that I have no problems with ringing or any other sound22:18
freemangordonPali: me neither22:18
Paliso in next days, I will compile all cssu versions and upload it to server22:18
freemangordonand PA with all that patches is in CSSU-thumb since forever22:18
Paliso TMO users can check which version working and which not22:19
Paliand then we can look at patches between working --> non working versions22:19
freemangordonPali: :nod:22:19
merlin1991Pali: they are still in the repo:22:19
merlin1991http://repository.maemo.org/community-testing/pool/fremantle/free/p/pulseaudio/22:19
Paliok, better then22:19
merlin1991hm actually there are only 2 we ever released22:20
Palibut not all versions22:20
freemangordonmerlin1991: I see only cssu1 and cssu422:20
Palionly cssu1 and cssu422:20
Palicssu3 is already on my website22:20
Paliand I will compile also cssu222:20
merlin1991I just checked we only ever updated pulse twice22:20
Palion my website is also version with removed memory patches22:20
merlin1991in T6 and in T822:20
Paliyes, but there are more commits in git22:21
Paliand need to know which commit/patch broke PA22:21
merlin1991ah I thought you meant all released versions, not all versions in the git history22:21
Palicssu1 working cssu3 not22:21
merlin1991now it makes sense22:21
Palithis wrote TMO people22:21
freemangordonPali: merlin1991: I still don;t get it why there is no single report for PA in cssu-thumb22:21
Palimaybe because thumb version using less memory?22:22
PaliI bet that memory alocation is problem22:22
Palilook at PA in network mode22:22
freemangordonPali: come on, the difference in the binary size could not be the problem22:22
Paliif memory block is too smal (defaul in PR1.3) it show lot of errors and sound is bad22:23
freemangordonPali: 0.9.15-1maemo44+0m5+0cssu4+thumb0 is in cssu-thumb for ages22:23
merlin1991freemangordon: we could possibly be looking at a bug in gcc22:23
freemangordonyep22:23
merlin1991but lets first root out after which patch it goes bad22:23
Palimaybe PA using some other strategy and looking for RAM usage?22:23
freemangordonok22:23
merlin1991because even if it is gcc that is going bad we still have to find which bit makes it do that :D22:24
Palior compiler behaviour? (not bug, but gcc 4.7 can do some optimalization and some code can be compiled differently)22:24
freemangordonmerlin1991: by default PA is compiled with auto vectorization enabled :(22:24
freemangordon-ffast-math too22:24
freemangordonok, agreed22:24
freemangordonooh, and -O322:24
Palifast math is option which tell GCC that floting point sum is comutative and asociative?22:24
freemangordonwe hit that -O3 with Qt already22:25
Paliif yes, this could be funny with sound files :-)22:25
freemangordonPali: could you build -O0 versions too?22:25
PaliI can try22:25
freemangordonIIRC DEB_BUILD_OPTIONS=noopt22:25
Palibut do not know if it will be easy or not...22:26
freemangordonnot sure if PA looks for that22:26
freemangordonok22:26
merlin1991yep it has ifneq (,$(findstring noopt,$(DEB_BUILD_OPTIONS))) \ CFLAGS += -O0 -rdynamic -fno-omit-frame-pointer -mapcs-frame22:26
freemangordon:)22:26
freemangordontoldya22:26
freemangordon:P22:26
freemangordonmerlin1991: ooh, DocScrutinizer05 thinks sending PA logs to syslog is a bad idea, lets append that to the queue. If he is here to motivate his point22:28
freemangordonmerlin1991: shall we continue with KCSSU22:29
freemangordon?22:29
freemangordonand glibc, as these are related somehow22:29
merlin1991I've got a bigger point first22:29
freemangordonok22:29
merlin1991maintaininer(s)22:29
freemangordonof what?22:29
merlin1991cssu, and also give me asecond to write .)22:29
freemangordon:)22:29
* freemangordon hands merlin1991 a second22:30
merlin1991I'm currently (err rather the past months) swamped with stuff todo and have been mostly inactive regarding cssu activity, up to the point where I've got valid complaints, so I'm throwing the position out there if anyone thinks he/she can do a better job22:30
merlin1991I'm happy to keep going, but it will be a rather slow pace22:31
merlin1991thoughts, slaps on the back, outrage?22:33
freemangordonmerlin1991: I'd rather say keep it going, as (as I see it) it is only me and Pali who qualify fot the job (besides you). You know what will happen if I become maintainer of CSSU, so I prefer to not be ;)22:33
freemangordon*for22:33
freemangordonmerlin1991: besides the lack of time, are there any other reasons?22:34
merlin1991lack of time simply22:35
freemangordonmerlin1991: also, do you expect you'll have time someday?22:35
merlin1991spring next year things will normalize for me :/22:36
freemangordonooh :(22:36
freemangordonPali: here?22:36
freemangordonarcean: here?22:36
Palihere22:36
arceanhere22:36
freemangordongood :). what do you think?22:36
arceanslow pace should be enough, of course if it's not a problem for merlin199122:38
freemangordonmerlin1991: arcean: Pali: lets decide that we'll pester merlin only when we think there is enough material for new -T, along with the list of what have to be included. is that sound ok?22:40
freemangordondoes even :)22:41
Paliok22:41
merlin1991nah pester me whenever you feel like it22:41
PaliI will prepare new enabler22:41
merlin1991the earlier you pester the earlier I can react :D22:41
Paliand when it will be ready I will ping22:42
freemangordonmerlin1991: but please, if you feel MAG way someday, make it clear :)22:43
freemangordonmerlin1991: what about repos?22:44
freemangordon*the repos22:44
merlin1991apart from getting autobuilder to work again repos are low maintainance22:44
merlin1991and I hope I can finally tackle that this weekend22:44
freemangordonyeah, the autobuild22:44
freemangordongood22:44
freemangordonnext point?22:45
PaliPG key was updated :-) so I belive all current CSSU will update to that version til october22:45
Pali*GPG22:45
merlin1991last saturday turned against all initial plans into a day of dead sleeping :/22:45
*** oldtopman has quit IRC22:45
merlin1991next point is kcssu I guess22:45
freemangordonPali: is it in CSSU-S too?22:45
Paliyes22:46
freemangordonfine22:46
merlin1991yep chem|st pushed the modest and the enabler update out22:46
Paliproblem with CSSU GPG key already fixed22:46
freemangordonok22:46
Paliproblem with Nokia GPG key is WONTFIX22:46
freemangordonwell, we stll can try fmtx-enabler hack22:46
freemangordon*still22:46
freemangordonok, kcssu22:47
Paliif maemo infra has space + has permission for mirror downloads.maemo.nokia.com/fremantle, I can create new GPG key, resign repo and add this key to CSSU22:47
freemangordonPali: hmm, the solition would be to replace that key22:48
Paliyes22:48
Palibut automatically will work only for CSSU users22:48
freemangordonPali: we can at least try fmtx trick, to see if it will work22:48
DocScrutinizer05autobuilder: http://privatepaste.com/145b0770c322:48
*** aap has joined #maemo-ssu22:49
Paliwhat is it?22:49
*** aap__ has joined #maemo-ssu22:49
DocScrutinizer05we need somebody to look into all this mess, or it will go down eventually22:49
freemangordonDocScrutinizer05: <merlin1991> and I hope I can finally tackle that this weekend22:49
freemangordonok, KCSSU22:49
freemangordonPali: you merged KP52, ain't?22:49
Palido not remember22:50
Paligoing to look22:50
Paliyes, kp52 is merged into kernel repo in cssu gitorious22:51
freemangordonPali: do we need anything else here http://wiki.maemo.org/CSSU_kernel_assess assessed?22:52
*** aap__ has quit IRC22:52
*** aap has quit IRC22:52
freemangordonsorry, I just can;t remember :)22:52
Paliwhite patches22:53
freemangordonah, yes22:53
Palionly red & greed are decided to not include or include22:53
freemangordonPali: is that table updated with stuff from KP52?22:53
*** aap has joined #maemo-ssu22:53
Paliyes: 18:26, 24. jan 2013 pali (updated patches list to match kp52 version)22:54
Paliat least wiki history saying that22:54
freemangordonmerlin1991: I guess fixing glibc repo problem will happen (hopefully) during the weekend, correct?22:55
merlin1991hopefully22:55
merlin1991but it's going to be fun to find22:55
freemangordonPali: ok. how are you with your free time tomorrow?22:55
Palimaybe, probably not22:56
freemangordonFriday?22:56
freemangordonI mean during the day, not in the evening22:56
merlin1991Pali said he has no internet friday<>monday22:56
freemangordonoh, yes22:56
freemangordonPali: ok, I'll try to fill whatever I can tomorrow, will pester you for the rest22:57
Palilot of patches has already Y or N (but do not have colored lines)22:57
freemangordonyes22:58
Palipatches like nilfs only adding new drivers (in ko modules)22:58
Palithis only increasing size nothing more (no autoload unless user using nilfs on SD card)22:58
*** aap__ has joined #maemo-ssu22:59
*** oldtopman has joined #maemo-ssu22:59
freemangordoniirc we already decided that .kos are Y by default22:59
Paliand need to decide if we include 1) battery drivers, 2) overlock support 3) hostmode support22:59
PaliI think 1) should be NO22:59
freemangordon1 is no :)22:59
freemangordon2 should be fine22:59
freemangordonas it is disabled by default23:00
freemangordonnot sure about 323:00
Palithere is comment for 2)  OC patches break powertop23:00
freemangordonbut If 3 depends on 1, it should be No too23:00
Palialso when OC is disabled?23:00
freemangordonPali: it is disabled by default if I got your question right23:01
freemangordonPali: you need to explicitly enable it23:01
PaliI'm thinking about 3)...23:01
freemangordonPali: oh, 3 does not depend on 123:01
PaliI think that hostmode will work also without kernel charger drivers23:01
freemangordonyep23:01
Palibut then needs userspace charger driver :D23:02
Palieg booston23:02
Paliso my hostmode package with menu will not work23:02
freemangordonPali: does not matter, as it (hostmode) is disabled by default too23:02
freemangordonPali: there is KP for that :)23:03
Paliyes it is disabled when you boot kernel :-)23:03
freemangordonhmm...23:03
freemangordonmerlin1991: arcean: what about 3?23:03
PaliDocScrutinizer05: what do you think? :D23:04
freemangordonDocScrutinizer05: yeah23:04
merlin1991tbh, if I go hostmode, I use palis menu which seems to depend on the battery drivers? so then no23:04
freemangordonmerlin1991: no23:04
freemangordonafaik Pali's menu depends on KP23:04
Palino23:05
freemangordonhmm23:05
Palimenu depends on some kernel which provides hostmode and charger kernel drivers23:05
merlin1991..23:05
freemangordonPali: well, ok :D23:05
Paliwe can provide charger/battery drivers in separate DEB pacakge23:05
freemangordonmerlin1991: the original hostmode UI does not depend on charging drivers23:05
Palihm, right23:06
merlin1991freemangordon: yea, but the original ui is far from a reason to include it in cssu23:06
Palioriginal MAG's hostmode UI working fine23:06
freemangordonPali: lets disable it for now23:06
freemangordon(3 that is)23:06
Paliso that UI can work with possible included hostmode patches in CSSU23:06
Paliok23:06
Paliso disable charger/battery drivers and hostmode23:07
freemangordonPali: those who have that UI are already in KP23:07
freemangordons/in/on/23:07
Paliand enable overclock patches? or not?23:07
merlin1991btw Pali how far did you get with upstreaming those patches?23:07
Palibattery patches?23:07
merlin1991yep23:07
Palior charger?23:07
merlin1991both23:07
merlin1991I think battery you got upstream iirc23:08
Palibattery patches was all upstream (expect that one which exporting register file in sysfs)23:08
Palialso rx51_battery patch is upstreamed (that for temperature and design capacity)23:08
DocScrutinizer05if you want to disable hostmode, you have to disable SR and OC by same rationale23:08
freemangordonDocScrutinizer05: :nod:23:08
Palibq2415x_charger is upstream, but platform data not23:08
Paliplatform data depends on some changes in isp1704 and musb driver23:09
Palibut today I finished that patches and I will send it to ML for upstreaming23:09
freemangordonoooh, nice :D23:09
Palialso I refactored arm errata patch for thumb2 isa23:10
freemangordonPali: I guess we can include them in CSSU , along with hostmode and OC once they are upstreamed23:10
Pali(moved all secure code to one file as maintainers wanted)23:10
freemangordonPali: you're my hero :D23:10
Palifreemangordon: please look at this patch: https://gitorious.org/linux-n900/linux-n900/commit/10d633415878cb5bb8e5b11b8cf9349e5f15937b23:12
freemangordonPali: so, 1,2 and 3 disabled. for now23:12
freemangordonagree?23:12
Paliand this: https://gitorious.org/linux-n900/linux-n900/commit/f58f12560ae21397493e6e102ce4a7248b81c25523:12
Paliok, disabled 1 2 and 323:12
Palifor now23:12
freemangordonPali: I still have birthday, will look tomorrow :P23:12
Paliok you have also weekend :-ň23:13
freemangordonyeah :)23:13
PaliI wrote to #maemo-alternatives I pushed new patches on top of 3.10 kernel23:13
Palitoday23:13
freemangordonok, based on ^^^ I'll try to fill that table23:13
Paliok23:13
freemangordononce we have eveything (KCSSU and glibc) in place, we'll pester merlin for a new update, including kcssu, glibc, e2fsprogs, h-d and upstart23:14
freemangordonand maybe PA23:15
Paliand new enabler?23:15
freemangordonoh, yes23:15
Paliok23:15
freemangordonPali: hmm23:15
freemangordonwe need some kernel detection, so we won;t overwrite KP/kcssu-thumb23:15
Pali"We need info from Luke-Jr why ans if this patch is really needed"23:15
freemangordondoes yours work?23:16
Paliyes, it working23:16
Palialso working for u-boot23:16
freemangordonPali: sure? as when I installed u-boot it was unable to detect my kernel :P23:16
PaliI think that u-boot-flasher already depends on package where is detector23:16
freemangordonI was on kernel-cssu323:17
Palithere is one problem with memory allocation23:17
merlin1991btw for a new release we have an old pull request for calendar-backend23:17
Palifor some reason sometimes kernel refuse malloc(2MB)23:17
freemangordonPali: malloc? in kernel?23:17
Paliand then detect application cannot unpack kernel zImage23:17
freemangordonmerlin1991: which one?23:17
Palino in userspace23:17
Palimalloc returns NULL23:17
freemangordonhmm, well, just retry it :P23:17
Paliand kernel in dmesg wrote that *some nokia kernel patch* refused allocation because of some limits23:18
merlin1991yeah a while $pointer == NULL is probably a good idea ;)23:18
Palithat limits are set at last lines in rcS-late23:18
*** arcean has quit IRC23:18
PaliI will try something23:18
freemangordonPali: do we need the whole kernel extracted?23:19
freemangordonuncompressed that is23:19
merlin1991freemangordon: http://gitorious.org/community-ssu/calendar-backend/merge_requests/1 I have no idea about the other one23:19
Palibecause /dev/mtd* is character device, full mtd(kernel) must be read from mtd to memory23:19
Paliand mtd(kernel) is 2MB23:19
freemangordonaah, mmap does not work :(23:20
Paliunpacking with libz is on the fly with some buffer on stack23:20
Paliand mmap does not working -> yes23:20
Palimmap working on /dev/mtd in some upstream 3.x kernels23:20
PaliI already played with this23:20
freemangordonirrelevant :P23:20
*** tg has quit IRC23:21
freemangordonPali: well, do something about it, we need kernel detection23:21
PaliI will try to look at those nokia patches in kernel how to force kernel to allocate 2MB connected memory23:21
freemangordonwe should overwrite *ONLY* stock kernel23:21
freemangordonconnected?23:21
Palimaybe there is some ulimit/restriction when running that process dpkg/apt/HAM?23:21
freemangordonyou mean linear?23:22
Paliyeah, linear23:22
freemangordonooh, but wy do you need it like that?23:22
Paliwithout holes :D23:22
Palibecause it is hard to write it...23:23
freemangordonPali: I am facing similar problems with bc-cat, kernel sometimes refuse to alloc_pages_exact for something like 384K23:23
Paliand what to do whit that luke-jr patch? (yellow color)23:23
Palifreemangordon: alloc_pages in dmes?23:23
Pali*dmesg23:23
freemangordonyep23:23
*** arcean has joined #maemo-ssu23:23
Paliso this is same problem23:24
freemangordonPali: I don't think it is Nokia to blame, you're just under memory pressure23:24
freemangordonPali: but why do you need linear RAM?23:25
Paliand it is not possible to swap something?23:25
Palibecause of libz23:25
freemangordonPali: maybe if you try to flush fs cache and retry after that23:25
Paliwill try23:26
freemangordonPali: but the only remedy I can think of is retrying23:26
Palisleep for 1s and try again23:27
Paliand after 10 fails return no version23:27
freemangordonPali: flush fs caches, sleep for 1s and retry23:27
Palinow when malloc fail returning no version23:27
Paliok23:27
Palihow to flush fs caches? sync(); ?23:28
freemangordonPali: /proc/sys/vm...23:28
freemangordonor whatever it was23:29
freemangordonPali: sync flushes the cache, but does not free it23:29
Paliman 2 sync: sync, syncfs - commit buffer cache to disk23:29
Palisync() causes all buffered modifications to file metadata and data to be written to the underlying file systems.23:29
Palisync() is always successful.23:29
freemangordonPali: but it does not free the caches23:30
Palihm, how to do that?23:30
freemangordonecho "3" > /proc/sys/vm/drop_caches23:30
Paliok23:30
freemangordonnot sure about "3" google for it23:30
*** Vlad_on_the_road has quit IRC23:31
freemangordonok, guys, gtg23:31
*** _rd has joined #maemo-ssu23:31
Paliany other point?23:31
freemangordonhave 30 more minutes to celebrate my birthday :)23:31
merlin1991hf !23:31
freemangordonPali: none I am aware of23:31
freemangordon~hf23:31
merlin1991*have fun*23:32
Paliok, then I'm going offline23:32
freemangordonaah :). thanks, night23:32
Palibye23:32
Pali:-)23:32
merlin1991night everyone23:32
*** Pali has quit IRC23:33
*** sixwheeledbeast has joined #maemo-ssu23:42
*** bromide has quit IRC23:43
*** discopig has joined #maemo-ssu23:48
*** LauRoman has joined #maemo-ssu23:49

Generated by irclog2html.py 2.15.1 by Marius Gedminas - find it at mg.pov.lt!