Monday, 2012-03-19

*** kavurt has quit IRC00:00
*** afiestas has quit IRC00:06
*** jstaniek has quit IRC00:09
*** kurtul has joined #mer00:09
*** himamura has joined #mer00:09
*** kurtul has quit IRC00:12
*** kimitake_idle has quit IRC00:27
*** brooklyn_ has joined #mer00:31
*** exman2 has joined #mer00:36
*** kimitake_idle has joined #mer00:46
*** brooklyn_ has quit IRC00:53
*** brooklyn_ has joined #mer00:58
*** brooklyn_ is now known as wala00:58
*** Alison_Chaiken has joined #mer01:39
*** Openfree` has joined #mer01:43
*** ALo has joined #mer01:57
*** ALoGeNo has quit IRC01:57
*** ALo has quit IRC01:59
*** ALoGeNo has joined #mer01:59
*** ALoGeNo has joined #mer01:59
*** himamura has quit IRC02:00
*** himamura has joined #mer02:01
*** raignarok_ has joined #mer02:04
*** raignarok has quit IRC02:07
*** Alison_Chaiken has quit IRC02:10
*** Alison_Chaiken has joined #mer02:15
*** KaIRC has quit IRC02:23
*** sandroandrade has quit IRC02:32
*** M4rtinK has quit IRC02:35
*** wala has quit IRC02:53
*** raignarok_ has quit IRC02:54
*** john__ has joined #mer03:02
*** wala has joined #mer03:03
*** john__ has quit IRC03:10
*** john___ has joined #mer03:11
*** himamura has quit IRC03:19
*** himamura has joined #mer03:19
*** himamura has quit IRC03:22
*** himamura has joined #mer03:22
*** niqt has joined #mer03:40
*** john________ has joined #mer03:47
*** john___ has quit IRC03:51
*** john________ has quit IRC04:02
*** niqt has quit IRC04:29
*** furikku has joined #mer04:50
*** wala has quit IRC04:54
*** keistom has left #mer04:59
*** disco_stu has quit IRC05:01
*** lbt has quit IRC05:36
*** lbt has joined #mer05:36
*** lbt has quit IRC05:36
*** lbt has joined #mer05:36
*** disco_stu has joined #mer05:48
*** ZiQiangHuan has joined #mer05:51
*** lbt has quit IRC05:52
*** tomeff has joined #mer05:52
*** lbt has joined #mer05:52
*** Alison_Chaiken has quit IRC06:03
*** Alison_Chaiken has joined #mer06:03
*** himamura has quit IRC06:15
*** himamura has joined #mer06:17
*** stochastic has quit IRC06:20
*** tomeff has quit IRC06:21
*** stochastic has joined #mer06:34
*** stochastic has joined #mer06:34
*** e8johan has joined #mer06:43
*** pohly has joined #mer06:43
Sage_that needs work from the UI vendors06:50
Sage_also it isn't shown in compile time as it is dbus api06:50
*** cxl000 has quit IRC06:58
*** ZiQiangHuan has quit IRC07:09
*** andre900 has joined #mer07:17
*** ZiQiangHuan has joined #mer07:19
*** JRepin has joined #mer07:23
*** niqt has joined #mer07:27
Sage_morning niqt07:27
*** eocanha has joined #mer07:28
*** u1106 has joined #mer07:28
niqtyesterday I asked mailing list where I could find the kickstart file for the latest release of Mer, but I had no answer. Is an obvious/stupid question?07:31
Sage_niqt: the only change is the release URL I think07:35
Sage_so you can modify older .ks file to fit that quite easily07:37
niqtok. i wnat build 486 image ant try it on via (CLE266=07:37
Sage_ o_007:41
Sage_what the heck happened now. i586, vm and pandaboard images built fine byt n900 and n950 images do not07:41
*** leinir has quit IRC07:43
*** andre__ has joined #mer07:44
*** andre__ has joined #mer07:44
*** cxl000 has joined #mer07:48
*** jstaniek_QFridge has joined #mer07:49
*** ZiQiangHuan has quit IRC07:50
*** ZiQiangHuan has joined #mer07:53
*** himamura has quit IRC07:59
*** himamura has joined #mer08:00
*** harbaum has joined #mer08:02
*** dcthang has joined #mer08:03
*** lbt has quit IRC08:04
*** otep has quit IRC08:10
e8johanStkeeps: repackaged the latest and greatest of xcb, but it did not do much to the framerate, at least not on my EMDG-based system08:10
e8johanStkeeps: it still sucks, even without mcompositor running, so I guess I have to do some more investigating08:11
kulveI guess I can use mic to create directly ext4 image instead of .tar.bz2 rootfs?08:14
*** dazo_afk is now known as dazo08:15
Sage_kulve: yes08:17
Sage_kulve: use raw instead of fs in cmdline08:17
Sage_then the image contains partition layout as well08:17
kulvethe "mic create help raw/fs/loop" info is a bit short..08:18
* Stskeeps looks for his coffee cup08:18
kulveSage_: thanks08:18
*** otep has joined #mer08:18
Stskeepswow, i managed to sleep from a nemo bug triage08:21
Stskeepsthat's first time :P08:21
MerBotSage_: Error: "#%/" is not a valid command.08:23
Sage_Stskeeps: vs
Sage_there is a monday welcome for you :D08:23
Stskeepsthat's the same url?08:23
Sage_armv7hl vs armv7l08:24
Stskeepsthat's it, i'm going back to bed..08:24
Sage_Release gone bad etc. :)08:24
StskeepsSage_: 0.3 works fine?08:24
Sage_yes it doesn't lack packages :)08:25
Sage_maybe rsync gone bad and something out of space?08:25
Stskeepsif you noticed our convo yesterday we were wondering if monday had come one day early08:26
Stskeepsat same time i managed to nuke my Core:armv7hl project on CI obs by accident08:26
Sage_LOL :D08:26
Stskeepsthe good news is that the resurrection procedures for Mer works08:26
Stskeepsthe bad news is that we seem to have -really- bizarre CI_CNT.B_CNT :)08:27
Stskeepslike, even before release08:27
*** Bryanstein has quit IRC08:28
*** tamisoft has joined #mer08:29
Stskeepse8johan: ok, good to know about framerate08:30
Stskeepse8johan: i presume stuff like DRI/DRI2 is enabled on your EMGD and you're using qml2 or opengl widgets?08:31
*** mdfe_ has joined #mer08:31
*** tilgovi has quit IRC08:32
* Sage_ feels like going back to bed now08:32
e8johanStskeeps: DRI and stuff are working and I'm using qml208:32
Stskeepsjust going down the checklist08:32
e8johanstill, I'll dig into it more08:32
e8johanI want to have a peek at our Arm platforms as well08:32
StskeepsSage_: should be fixable when lbt wakes up08:32
e8johanwill see what I can come up with08:33
kulveany idea what's this about: Error <creator>: Command 'extlinux' is not available.08:34
*** ericlr has joined #mer08:34
Aardkulve: install syslinux-extlinux08:35
kulveAard: yeah, just trying that as I found something from the irc logs.. Thanks08:36
*** leinir has joined #mer08:41
Sage_Stskeeps: :nod:08:45
StskeepsSage_: 0.3 is practically == 1 anyway08:47
*** JRepin has quit IRC08:47
Sage_Stskeeps: yes but well don't want to use that URL in images and don't want to hack it in just because of couple of hours of waiting08:48
*** Bryanstein has joined #mer08:50
*** afiestas has joined #mer08:54
*** andre__ has quit IRC08:54
*** smoku has joined #mer08:56
*** andre900 has quit IRC09:03
*** notmart has joined #mer09:06
Sage_Stskeeps: sent couple of package updates today that I couldn't submit during the weekend09:16
Sage_for some reason my osc connection doesn't work from my laptop anywhere else than at home :P09:17
Stskeepsthey might get bad test results because of Core:armv7hl09:18
Stskeepsjust so you know09:18
Sage_well at least libexif passed on armv7hl09:21
Sage_but thanks for reminder09:21
*** swerden has joined #mer09:26
*** swer has quit IRC09:30
*** bergie has joined #mer09:30
*** lbt has joined #mer09:39
Stskeepslbt: seems like .1 isn't complete09:39
Stskeepsnoarch in armv7hl is foobar09:40
lbtwhat's missing? is the release/ area bad?09:40
mdfe_Sage_: good morning, I get this error by using the @Intel x86 Generic Support package group.
Stskeepslbt: [09:23] <Sage_> Stskeeps: vs
mdfe_I did not define any llvmpipe packages inside my ks file09:41
lbtStskeeps: just doing that now09:42
mdfe_Stskeeps: Any idea how to build a image with mesa-x86-generic packages?09:47
lbtStskeeps: it's in obs-repos09:48
Stskeepslbt: as in, noarch packages are fine in obs-repos09:48
lbtwell, they're not noarch in there09:48
lbt-rw-r--r-- 1 prjfetcher users   1120613 Dec 16 04:06 autoconf.rpm09:48
Stskeepsthat's because there's no arch description in filenames in obs-repos09:48
lbtyeah, exactly09:48
lbtbut it is there09:49
lbtso should have synced09:49
Stskeepswell, something is foobar :P09:49
Stskeepscan we just copy 0.3 to .1 in releases/ and call it a day?09:49
lbt*nod* - I'm trying to see what09:49
lbtOK - but it may happen again09:49
lbtis this the only problem?09:50
StskeepsSage_: are OBS repos ok?09:51
Stskeepslbt: also, rename .1 to .1-broken or something09:51
lbtmaking space09:55
mdfe_lbt: could you take a look on ?09:59
mdfe_lbt: teher are two patterns.xml.gz files10:00
Stskeepsmdfe_: i think only one is referred to in the .xml for the repo i would hope10:00
lbtlooks like some kind of delete failed10:01
mdfe_Stskeeps: thynks, yes the newer one is referred10:02
lbtmdfe_: in that case I assume non-urgent so could you log a bug - I'll look at the file owner/perms at some point10:03
mdfe_Sage_: The "@Intel x86 Generic Support" package group needs the additional package "mesa-x86-generic"10:07
sledges_wkendgood morning10:13
*** sledges_wkend is now known as sledges10:13
Stskeepsmorn sledges10:14
sledges :)10:14
*** mdavey has joined #mer10:14
* sledges is after Lipstick QML, where could it be found?10:20
Stskeepslbt: looks sane10:20
Stskeepsi said .1 -> -old though :P10:20
lbtstep by step :)10:21
Stskeepssledges: CE:UX:lipstick or Lipstick10:21
lbtStskeeps: I cannot see where the script went wrong10:22
lbtunless it's in makerepo10:22
sledgesStskeeps, found it in * . a sample .ks would be handy.. will try to browse w00t around10:23
Sage_mdfe_: hmmp10:23
Sage_mdfe_: I didn't have any problems with that10:24
Sage_Stskeeps: it seems so.10:24
Sage_mdfe_: can I see your .ks file10:25
Sage_mdfe_: only the one that is referred in repomd.xmlis used from the patterns.xml.gz10:26
mdfe_I had to add the  "mesa-x86-generic" package to get it building again10:28
Sage_mdfe_: drop #mesa-dri-swrast-driver10:29
Sage_yes it is commentd but it might cause problems if the parsing isn't ok10:30
Sage_mdfe_: also you shouldn't call repos mer-shared and mer-plasma-shared but nemo-mw-shared and nemo-mw-plasmaactive for example10:31
mdfe_but there isn't any mesa-dri-swrast-driver package anymore10:31
Sage_mdfe_: yes I mean drop the line in comments just in case10:31
Sage_mdfe_: <- This has line #mesa-dri-swrast-driver. It is commented but I have seen problems with commented lines before so I would say drop that and try agains without the mesa-x86-generic10:33
mdfe_I will try10:33
Sage_mdfe_: also alsa-utils is now in mer core not in ce_tools10:34
Sage_Stskeeps: should mer minimal xorg bring dbus-x11 package?10:35
Stskeepsyes, it kinda should10:36
* sledges got lost between
Stskeepssledges: additional repo, 'lipstick' in %packages, and remove meegotouch-home from %packages10:37
sledgeswhat are MFT, XBMC, lipstick and their interrelationships10:37
timophmeegotouch fremework, xbox media center and a UI written with Qt10:38
mdfe_Sage_: no, without  "mesa-x86-generic" the image build will break10:39
*** lbt has quit IRC10:39
*** lbt has joined #mer10:39
sledgesthank you timoph, Stskeeps10:39
*** lizardo has joined #mer10:40
Sage_mdfe_: a sec10:40
Sage_mdfe_: I'll try to build it my own10:40
Stskeepscan we just make pkgmgr standard yum for the while being in mer?10:41
mdfe_Sage_: here my build log with "mesa-x86-generic"
*** DocScrutinizer has quit IRC10:42
lbtStskeeps: yes10:42
*** DocScrutinizer has joined #mer10:42
lbtStskeeps: some sb2 issues here:  these use a modified obs-rpm-install that knows about /usr/bin/zypper and accelerates it10:43
*** M4rtinK has joined #mer10:43
Stskeepslbt: yeah.. rpm --eval '%{_arch}'10:43
*** ZiQiangHuan has quit IRC10:44
*** ZiQiangHuan has joined #mer10:44
lbtsame using both -m simple and obs-rpm-install10:45
lbtreload pastie10:46
*** raignarok has joined #mer10:46
sledgesStskeeps, "...'lipstick' in %packages, and remove meegotouch-home from %packages" - are you referring to modifying an existing Nemo .ks ?10:47
Stskeepssledges: for example, but remove all @Nemo package groups10:48
*** ZiQiangHuan has quit IRC10:50
Sage_mdfe_: added that mesa-x86-generic to the x86 group. Should be fixed now10:51
*** dcthang has quit IRC10:53
lbtcoffeee then sb2 debug logs ....10:56
Stskeepslbt: i'd prefer if we concentrate on obs-rpm-install and obs-rpm-build , though that may require you to do stuff in /home10:57
Stskeepsthat is a better reference point for me10:57
lbtoh yes, I am doing11:03
lbtthe simple was to show that it's a mode problem11:03
lbtI've added mappings for my non-/home things and they're fine so I'm pretty sure you can pretend that /everything is /home :)11:04
lbtI was actually going to create sdk-build and sdk-install - that way I don't need a core-release to update sdk modes11:05
lbtclearly they should sync back eventually11:05
lbtwell s/was going to create/have created/ :)11:06
lbtjust not deployed/pushed yet11:06
*** xmlich02_ has joined #mer11:06
*** xmlich02_ has quit IRC11:08
*** xmlich02 has quit IRC11:08
*** xmlich02 has joined #mer11:08
*** tarantism has joined #mer11:10
*** KaIRC has joined #mer11:10
StskeepsREMINDER: bug triage in #mer-meeting in 48 minutes11:12
*** himamura has quit IRC11:20
Stskeepse8johan: silly question, how do you measure fps?11:20
e8johanI use swaplogger (just packaged it this morning)11:21
Stskeepsalright, is what i would have advised too11:21
e8johanthere is a spec there if you want to merge it into mer :-)11:21
*** eman has quit IRC11:21
Stskeepswe'll probably be packaging a bunch of tools from harmattan/meego to mer toosl as well11:21
*** eman has joined #mer11:22
Stskeepsif anything can be said about meego .. it certainly open source a large bunch of professional linux mobile tools11:24
Stskeepssourced, that is11:24
*** dionet has joined #mer11:25
sledgeslipstick works! but is very slow, need to wait ~5-10seconds after clicking on "phone" or "people" (running from SATA HDD plugged directly through serial link to freescale imx53)11:26
Stskeepssledges: that's probably unrelated11:27
sledgesyes, just emphasising that medium is not slow11:27
Stskeepstry something like qmlcomponentsgallery instead11:27
sledgeshmm, can't get out from "people" anymore :) mouse cursor invisible, touchscreen not yet working, any keyboard shortcuts aroudn? :)11:30
Stskeepskill from terminal?11:30
sledgesdoing it then11:30
Stskeepsif you don't have meegotouch-compositor-gestures installed11:31
sledgesmost probably not, this wasn't explicitly specified in the .ks ;)11:33
* sledges already likes lipstick bigtime, lightweight N9 UX on any embedded system, what can be better!? :)11:33
Stskeepsout of morbid curiousity, what do you see on your screen?11:34
sledgesmain menu, 9 icons, vodafone UK 00:00 pm :)11:34
Stskeepsthat's all qml btw11:36
sledgeswhat should I infer from that it's all qml ?11:38
Stskeepssledges: well, that you can customize that homescreen to your heart's extent with ease11:39
Stskeepsyou could even make it a hello kitty skin11:39
sledgesyes, along with the underlying apps base - ideal case for a lightweight GSM/UMTS-enabled industrial gadget11:40
sledges(divergence of ideas here :)))11:40
sledgesStskeeps, so if we ship a board support package with our modified lipstick for such a gadget to a industrial customer, what licensing ensues?11:41
Stskeepsw00t: wasn't lipstick bsd?11:41
Stskeepssledges: well, typical GPL applies (ie, what's in mer) but that's fairly easy to comply with11:42
timophStskeeps: seems to be lgpl 2.111:42
Stskeepsyeah, lgpl 2.111:43
sledgeslesser GPL.. I'm not into licensing at all, but if you understand it, just tell me if any additional charges imply apart from keeping the source open (hence the customer would need to open source their gadget's stuff as well?)11:43
Stskeepsso you can do commercial homescreens with it, i think11:44
* sledges would really like to understand this legislative model11:44
Stskeepsw00t: am i interpreting that right11:44
Sage_Stskeeps: k11:44
StskeepsSage_: yes, the quilt stuff looks insane..11:44
StskeepsSage_: but i want to be as close to linaro's eglibc as possible11:44
Stskeepsless support issues11:44
Stskeepssledges: so, there is no royalties involved in Mer, you may need to comply with license stuff on mer core itself and provide source for those components11:45
Stskeepssledges: you'd have to publish source of lipstick but of the homescreen itself, no, not afaik11:45
Stskeepsi'm not a lawyer, but just stating the general principles11:46
timophthat has a pretty good explanation11:46
sledgesand what should our customers do? as this is a B2B stuff11:46
timophlgpl allows linking to closed stuff for example11:47
timophbut if you modily and ship the lgpl'd stuff itself you need to share the changed11:48
Bostikaye, modifications to the *library itself* need to be published; stuff created to use library are whatever the author wants11:48
timophat least that's how I understand it11:48
Stskeepssledges: if they sell the product they may get GPL compliance requests, you can make this easy for them by providing things on a dvd11:48
Stskeepssledges: we have a mer task bug on how to make this easier11:48
Stskeepsand ideally tools to identify 'illegal' dependencies as well11:50
Stskeepssledges: there's some cases were there can be dependancies between two packages that can't legally link together11:51
*** bjarneh has joined #mer11:51
*** bjarneh has quit IRC11:52
Stskeepswe have tools that can make graphs like (beware, big svg)11:53
Stskeepsand analyze those11:54
Sage_Stskeeps: nuts :)11:54
StskeepsSage_: much cleaner than before i started grouping :P11:55
StskeepsSage_: is .1 OK now?11:55
Sage_Stskeeps: i meant the eglibc :)11:57
*** ericlr has quit IRC11:57
StskeepsSage_: oh, yeah11:57
lbtStskeeps: bug triage :)12:02
Stskeepsoh right12:02
* lbt notices his N900 time is off ...12:02
lbtand why does sb2 debugging stop at an EXEC12:03
*** ScriptRipper has joined #mer12:16
*** ScriptRipper has joined #mer12:16
*** jonnor_work has joined #mer12:25
*** sandroandrade has joined #mer12:47
lbtso Stskeeps ... can we beat SDK/sb2 into submission by c.o.p today (after pizza)12:57
Stskeepspossibly, though i'm not through my e-mail pile yet12:57
lbt:/  OK12:58
lbtSage_: we really need a RESOLVED PENDING_ACCEPT or maybe RESOLVED->ACCEPTED13:00
lbtbut anyway13:00
*** bigbluehat has joined #mer13:00
Sage_lbt: yes13:03
Sage_and BOSS to do that automatically :)13:03
Stskeepslbt: so, golden rule, -R -m obs-rpm-install for rpm and zypper13:07
Stskeeps-build for everythign else13:07
lbtStskeeps: yep - totally get that13:09
Stskeepswhere does it break in that scenario?13:09
lbtI suspect rpm13:09
lbtdamn pastie has no line#13:10
Stskeepsok, two things.. echo -n 'armv7l-meego-linux' > /etc/rpm/platform , echo -n 'arch = armv7hl' > /etc/zypp/zypp.conf13:11
Stskeepsinside the target13:11
Stskeepswhat does zypper do then?13:11
lbtshoulda checked before echo13:13
lbtbut no change13:13
*** wala has joined #mer13:13
lbtshoulda checked +file contents+ before echo13:13
Stskeeps try zypper ref13:13
lbtno change13:14
lbtzypper ref --force complains from sb213:14
lbt(ERROR)        zypper[20210/-148039968]        fwd_map_path: ResultCheck: DIFFERENT result path: C='/bin/sh', Lua='/parentroot/srv/mer/targets/mer-core-armv7l-xorg-basic-qmlviewer-0.20120315.0.0.2/bin/bash'13:14
lbtwhich seems reasonable13:14
lbtsb2 -m obs-rpm-install -R zypper in glibc13:15
lbtgives the same but with this line13:15
lbtPackage 'glibc' is not available in your repositories. Cannot reinstall, upgrade, or downgrade.13:15
lbtwhich just confirms that it doesn't see anything in the repo cache13:16
Sage_Stskeeps: eglibc review done. NOTE: Just read lines that were in the gerrit diff not the whole thing. :)13:16
StskeepsSage_: right now just running SBC/DBC on it to see if there's any fallout13:18
Stskeepsbut yeah13:18
*** raignarok has quit IRC13:18
Stskeepslbt: i need to setup a local setup to verify those which won't be right now..13:21
lbtI'm going to see what sb2 does in -d mode running sudo rpm --rebuilddb13:22
lbtzypper does the db rebuild atm13:22
lbtgah - cleaned out the sb2 logs ... and yeah, when I do sb2 -d -m simple sudo rpm --rebuilddb .... sb2 logging stops at the EXEC13:24
Stskeeps-L info might be better13:24
Stskeepsand -L info -q13:24
lbtsame problem13:25
Stskeepsdoes it stall?13:25
lbtno, it runs fine13:25
Stskeepsalso, err, you need a -R in there13:26
Stskeepsyou're doing sudo inside sb2?13:26
lbtin simple more13:26
lbtjust to see how it works ... since it does13:26
Stskeepswell that's probably why13:26
lbtthen do it in obs-rpm-install mode to compare13:26
Stskeepsuse -R instead, not sudo13:27
lbt-R isn't setup in simple mode13:27
lbterror: can't create transaction lock on /var/lib/rpm/.rpm.lock (Permission denied)13:27
Stskeepsokay, can you remove all sb2 targets, start from scratch with a image and not use simple/accel at all?13:33
Stskeepsthat makes it easier for me to replicate13:33
lbtyes - and I'll share my mode files too13:36
*** himamura has joined #mer13:46
harbaumSome sentences here sound odd without context. I e.g. just read Stskeeps' last one ...13:50
lbtdoes it help to know his wife is away for 3 weeks?13:51
*** gabrbedd has joined #mer13:54
*** tomeff has joined #mer14:03
*** bigbluehat has quit IRC14:03
*** bigbluehat has joined #mer14:07
*** tamisoft has left #mer14:09
Stskeepsone email down, 6 to go..14:11
*** bergie_ has joined #mer14:26
*** jstaniek_QFridge has quit IRC14:27
*** e8johan has left #mer14:29
*** bergie has quit IRC14:30
*** gabrbedd has quit IRC14:30
*** Termana has quit IRC14:36
*** u1106 has quit IRC14:47
Aardis it possible to tell mic to take package x from repo y, if there's the same package in repos y and z?14:50
Stskeeps--excludepkgs is always nice14:51
Aardah, the opposite way14:51
Aard'Warning: Caution, do NOT interrupt the installation, else mic cannot finish the cleanup.' -- can somebody tell the mic-guys that signal handlers exist?14:57
*** gabrbedd has joined #mer14:57
*** mdavey has quit IRC15:14
smokuAard: I'm sure they will accept a patch15:23
*** jooncheol has joined #mer15:26
*** niqt has quit IRC15:26
*** bigbluehat has quit IRC15:37
*** kthomas_vh_ has quit IRC15:41
*** pdanek has joined #mer15:43
*** harbaum has quit IRC15:46
*** Aurium has joined #mer15:47
*** Aurium has quit IRC15:53
*** kthomas_vh_ has joined #mer15:54
*** kimitake_idle is now known as kimitake15:55
*** GeorgeH has quit IRC15:57
*** jooncheol has quit IRC15:58
*** mdfe_ has quit IRC15:59
*** Aurium has joined #mer16:01
*** marquiz_ has joined #mer16:06
*** talavis has quit IRC16:08
*** leinir has quit IRC16:09
*** jonnor_work has quit IRC16:11
*** kimitake is now known as kimitake_idle16:14
*** leinir has joined #mer16:22
*** kavurt has joined #mer16:23
*** wala has quit IRC16:35
*** andre__ has joined #mer16:43
*** andre__ has joined #mer16:43
*** wala has joined #mer16:45
*** bigbluehat has joined #mer16:46
*** leinir has quit IRC16:47
*** leinir has joined #mer16:47
*** leinir has quit IRC16:47
*** leinir has joined #mer16:47
*** kavurt has quit IRC16:49
*** rolandx1_ has joined #mer16:53
*** smoku has quit IRC16:54
*** pdanek has left #mer16:54
*** tom_say has joined #mer16:57
*** raignarok has joined #mer17:01
*** NIN101 has joined #mer17:03
*** gimli has joined #mer17:10
*** gimli has joined #mer17:10
*** bigbluehat has quit IRC17:11
*** bigbluehat has joined #mer17:12
*** buckle_ is now known as buckle17:13
*** dazo is now known as dazo_afk17:15
*** talavis has joined #mer17:20
* lbt notes that the __db* files are not actually part of the db... they are files which hold lock state information and can be safely deleted in single user mode17:27
lbtso --rebuilddb is actually working fine17:27
*** NIN102 has joined #mer17:36
*** NIN101 has quit IRC17:36
Stskeepslbt: can't you activate zypper debugging?17:36
lbtit writes to var/log/zypper.log and I'm going through that now17:37
*** ScriptRipper has quit IRC17:41
*** vivijim has quit IRC17:41
*** vivijim has joined #mer17:42
*** vivijim has joined #mer17:42
*** bergie_ has quit IRC17:43
Stskeepslbt: contains @BUILD_ID@17:45
*** vivijim has quit IRC17:46
Stskeepswould it be possible to s/@BUILD_ID@/0.20120315.1/g ?17:46
lbtI thought about it - would that really be a good idea?17:47
lbtthis way I can build a 'latest'17:47
lbtor a fixed release image17:47
lbtbearing in mind "--save" puts these values into repos.d17:48
Stskeepsmy reasoning for @BUILD_ID@ being set to 0.2012031.1 is that we are sure the .ks'es will always work instead of having broken .ks'es all over the place17:48
Stskeepslesson learnt from meego17:48
lbtit may be worth pointing out in the README that the user will need --release XXX17:48
Stskeepsthe .ks is a snapshot of the kickstarter output from a certain release anyway17:49
lbtany image created that way will not be zypper up'able17:49
lbtand we like zypper up'ability17:49
lbtso this was my reasoning17:50
lbtI will say that it should have the 0.20120315.1 in the file somewhere17:51
*** ThreeM has quit IRC17:51
lbtappend "# This kickstart created as part of the 0.20120315.1 release" to all the .ks files?17:52
lbtthat way they're branded17:52
*** ThreeM has joined #mer17:52
Stskeepshow about for this release we @BUILD_ID@ -> 0.20120315.1 and set a task bug for discussing how exactly the kickstarter config -> kickstarter -> ks chain is supposed to work? i feel we need to think deeper about this17:53
lbtwe do - did you see the README17:54
*** raignarok has quit IRC17:55
Stskeepsokay, let me say this a bit clearer.. how it's supposed to work when you look at it from vendor perspective, hardware adaptations, merging, etc17:55
lbtsed -i -e's/@BUILD_ID@/0.20120315.1/' *ks17:55
Stskeepsbuild_id is a bit awkward in that aspect17:55
Stskeepsdon't you need a 'g' on that?17:56
lbtnot unless you have multiple @BUILD_ID@ per line17:56
Stskeepsalso the mind-bending thing.. what component is @BUILD_ID@ for?17:56
Stskeepscore, hw adaptation, ui ..17:57
lbtyes, I do agree17:57
Stskeepsso that's why i think we need to discuss this deeper17:57
lbtI hit that in the SDK - i hardcode M:T:Testing in my .ks17:58
lbtwhich is ugly17:58
lbtupdated README too17:58
Stskeepsso, interesting thing about the resurrection thing17:59
StskeepsCI_CNT.B_CNT > the old release's18:00
Stskeepsso far18:00
Stskeepsi wouldn't mind validating it but..18:00
lbtbut... proper versioning would be better? ?18:00
Stskeepswell, yes18:00
Stskeepsbut it amazes me it gets it right18:00
Stskeepsi had expected to loose CI_CNT.B_CT18:01
Stskeepslbt: do you have time to do a not-rsynced release of core:armv7hl alone before i start integrating things into core again?18:02
Stskeepsso we can make a proper analysis on post-nuclear-bomb-on-germany-resurrection-of-mer18:02
*** ThreeM has quit IRC18:03
lbta $TOOLS/dumpbuild "$API" "Core:armv7hl" Core:armv7hl:$RELEASE Core_armv7hl "i586 armv8el"18:04
*** ThreeM has joined #mer18:04
Stskeepsi guess for the released part18:04
Stskeepsie, createrepo'ed stuff18:04
lbtthen set RELEASE=0.20120315.1-test and let it run the rest?18:05
Stskeepsgrab_build, not dumpbuild fwiw18:05
lbtI thought you wanted a wget of the new build from obs18:06
Stskeepsi just want to do a post-mortem of the* output18:06
Stskeepsie, check if built rpm packages are newer or older18:07
lbtI understand that bit18:07
lbtwhat I'm not sure is where the rpms to be compared reside at the moment18:07 naturally18:08
lbtI thought they were still in obs18:08
Stskeepsthe core:armv7hl rebuild is done18:08
lbtyep - so grab_build does not go external - it rsyncs from one dir to another18:08
Stskeepsright, and createrepos again18:08
lbtdumpbuild goes outside using wget to pull from obs18:08
Stskeepsgrab_build rsyncs from BE repos18:09
Stskeepsie, the published repos18:09
lbtwhich line?18:09
*** andre__ has quit IRC18:10
Stskeepstools/ 0.20120315.0.3 rsync://
Stskeepsthe last one is obsrepos on be18:10
lbtah ...18:10
lbtI really want to rewrite that script :D18:10
Stskeepsyou'll get to, when we get a proper BOSS process going18:11
lbtOK - yes, I see now18:11
lbtI wonder if that's why the release-fix failed18:12
lbtI'm not going to use the script then - just try and follow it by hand18:13
*** jonnor_work has joined #mer18:13
lbtright now it'd do weird things like update -next18:13
Stskeepsfair enough18:14
Stskeepsi think it's just a good time to document the experience, hence us doing this18:14
*** eocanha has quit IRC18:27
*** bigbluehat has quit IRC18:27
*** vivijim has joined #mer18:28
*** vivijim has joined #mer18:28
Stskeepslet's see..18:30
*** bigbluehat has joined #mer18:34
Stskeepsi'd say successful resurrection19:01
Stskeepsour CI_CNT, B_CNT is still bizarre but..19:01
*** raignarok has joined #mer19:11
lbtyep - that's about what I saw19:13
lbtstill no idea about the zypper problem :)19:14
Stskeepsdidn't i make it work once? :P19:15
* Stskeeps checks logs19:15
lbtI thought I had too19:15
*** bigbluehat has quit IRC19:19
*** tommis has joined #mer19:22
*** furikku has left #mer19:27
*** JRepin has joined #mer19:29
*** harbaum has joined #mer19:36
*** Guest55802 has joined #mer19:40
*** bigbluehat has joined #mer19:42
lbtit was an interaction caused by adding /everything.... I guessed that "dummy_file" wasn't important given : set_path = session_dir.."/dummy_file"19:53
lbtoccurs several times19:53
lbtalso... accelerated_program_actions19:56
lbtit says protection = readonly_fs_always19:56
lbtwhich doesn't make sense19:56
Stskeepswhat mode are we talking about19:56
Stskeepsprotection=readonly_fs_always .. where?19:57
lbtfs_rules.lua in accelerated_program_actions19:57
lbthmm, or is that only for *exec* function calls19:59
*** harbaum has quit IRC20:02
* Stskeeps looks into upgrading 'icu'20:09
*** jstaniek has joined #mer20:15
lbthow come I hit all the roadblocks20:17
Stskeepslbt: i'm more scared that it's starting to be consistent what roadblocks people are hitting20:17
lbterror: /var/cache/zypp/packages/mer-core/armv7hl/info-4.13a-1.6.armv7hl.rpm: Header SHA1 digest: BAD20:17
Stskeepslbt: oh that one is fun20:17
Stskeepstry to cd into the target before running sb220:17
lbtin the SDK I was cd'ed to the /srv/mer/targets/blah/ root of the target20:19
lbtie :  MerSDK david:/parentroot/srv/mer/targets/mer-core-armv7l-xorg-basic-qmlviewer-0.20120315.0.0.2/root > sb2 -t mer-core -m obs-rpm-install -R zypper in glibc-devel20:20
Stskeepsand it says BAD?20:20
*** toscalix has joined #mer20:21
Stskeepstry to cd /20:21
Stskeepsand do the same20:21
Stskeepsin mer-core-blahblah/  ?20:21
Stskeepstry in that location20:22
*** talavis has quit IRC20:23
lbtI had many: error: rpmdbNextIterator: skipping h#       2 Header SHA1 digest: BAD20:23
Stskeepsget me SB2 debugging logs for that?20:23
lbtuntil I did rpm --rebuilddb20:23
Stskeepsit's NSS going crazy, fwiw20:23
lbtOK - yes, that's good to know20:24
lbtand on my side I probably ran some emulated rpm/zypper commands on the same target from a different sdk20:24
lbtie using the old fs_rules.lua20:24
*** Alison_Chaiken has quit IRC20:24
Stskeepsclean out your $HOME/.sb2 configs at times20:25
lbthmm don't have any20:25
Stskeepsor whatever it's called20:26
lbtah, you mean the target definitions in ~/.scratchbox2/20:26
*** talavis has joined #mer20:28
lbtso I have an sb2_log ...20:28
lbtscp to monter:/tmp/20:28
lbtthat's -d20:29
lbtsay if you'd prefer -L20:30
Stskeepscan i ask you to file a bug with that so we can return to it? 21:30 is beyond my mental ability to read sb2 logs20:31
lbtFWIW ... my host has ldap and nscd and nslcd20:33
Stskeepsi'm talking about netscape security services(?)20:34
Stskeepsnot nssswitch20:34
lbterr.... obviously :)20:34
lbtI'd seen rather a lot of sb2:LocatingNsswitchConf log entries so...20:35
Stskeepsi'm not sure if i should be happy or worried that all the icu patches we carry seem not to have been merged into upstream, but still apply fine, two major versions after20:37
*** andre__ has joined #mer20:37
*** andre__ has joined #mer20:37
* lbt wonders if each patch should have a bug and a link to an upstream submission20:39
*** notmart has quit IRC20:39
*** marquiz_ has quit IRC20:39
lbtlike wtf happened with that submission20:39
lbtor maybe a discrete patchzilla20:39
lbt( since bz has process hardcoded into states )20:40
*** jonnor_work has quit IRC20:40
lbtall the sb2 log entries around /lib/ seem fine20:45
Stskeepsyou normally first run the stuff later20:46
Stskeepslook for anything revolving around .chk20:47
lbtchk not mentioned in the log20:48
lbtand just FYI I'm not expecting a response.... just kinda working out loud :)20:49
*** asdfafew has joined #mer20:49
*** wubudubudubudu has quit IRC20:51
lbtthere may be a little screwup on my side here20:51
lbtsb2-init : david:/parentroot/srv/mer/targets/mer-core-armv7hl-xorg-basic-qmlviewer-latest20:52
lbtsb2      : david:/parentroot/srv/mer/targets/mer-core-armv7l-xorg-basic-qmlviewer-latest20:52
* Stskeeps sends upgrade for icu20:54
lbtso sb2 cares about cwd ....20:55
Stskeepsi got the reasoning for this at some point20:55
Stskeepsbut i can't recall it right now20:55
Stskeepsplease log that as a bug though20:57
* lbt is going for reproduction20:58
lbtwhich is not at all like replication20:59
Stskeepsdamnit, we could have used a second lbt21:00
lbtwe have a spare lpt21:01
*** lizardo has quit IRC21:02
*** bigbluehat has quit IRC21:07
lbt*sigh* package info-4.13a-1.6.armv7hl is intended for a different architecture21:08
lbtyes I've done etc/rpm/platform21:08
lbtand zypp.conf21:08
lbtI think I might take up knitting21:08
Stskeepswhat $CWD are you in?21:10
Stskeepser, PWD21:10
lbtI'm doing an rm -rf *21:10
Stskeepslet's say that if this doesn't work properly after tonight you transfer that for me for fixing, there's better things to work on21:10
Stskeepsas it obviously is having problems21:11
lbtpossibly - sometimes it just needs me to tidy up and start from a clean install that hsn't had rpm and zypper dbs butchered21:11
lbtSDK is also hacked to have new fs_rules, was zyppered up from an old version ...21:12
lbtI think that's OK thouhg21:12
*** Crnkoj has joined #mer21:14
Crnkojhey guys i need some help, i managed to break something on my n9 and now its in a boot loop, whats the best way to rescue it without flashing (i woudl actually just have to chmod one file back), or to flash it21:15
StskeepsCrnkoj: we're not really a n9 support channel, might have more luck in #harmattan , but up to step 5 might help you21:16
Stskeepsthis will void your warranty21:16
CrnkojStskeeps thanks21:16
*** GeorgeH has joined #mer21:20
*** tommis has quit IRC21:20
*** cxl000 has quit IRC21:25
lbtStskeeps: you said:   echo -n 'armv7l-meego-linux' > etc/rpm/platform21:28
lbtwhat is the meaning of armv7l?21:28
lbtwhat should it match21:29
lbtgiven echo 'arch = armv7hl' >> etc/zypp/zypp.conf21:29
Stskeepsdoesn't mean much, it basically tells rpm that the current platform is armv7l, not armv7hl, why?21:29
lbtmainly it doesn't match the value for zypp.conf21:30
Stskeepsare you doing this against an armv7l image or armv7hl21:30
lbtI coulda copied it wrong too21:30
Stskeepsif it's armv7hl it should be armv7hl21:30
lbt[19/03/2012 13:11] <Stskeeps> ok, two things.. echo -n 'armv7l-meego-linux' > /etc/rpm/platform , echo -n 'arch = armv7hl' > /etc/zypp/zypp.conf21:31
lbtcool - that'll be one issue then21:31
Stskeepsprobably, yes21:31
Stskeepsmy bad21:31
lbtI think we get to share the bads about a bit21:31
lbtsomehow I think I'm leading :)21:31
Stskeepsjust be happy you didn't get the wtfs i did when i made the mer base originally21:32
Stskeepsthat's going to cost a lot of therapy one day21:32
*** GeorgeH has quit IRC21:35
lbtI don't *think* arch= in zypp.conf is needed21:35
lbtie .. it works without it21:36
lbtzypper, not got to gcc yet21:36
Stskeepsit won't for armv7hl, i think21:36
lbtI'm doing armv7hl21:36
lbtso should I expect gcc to fail?21:37
Stskeepsi'd hope not21:37
Stskeepsglibc-devel should be there though21:37
lbtI deliberately chose not armv7l since it would be a different arch to the one you used21:37
lbtI wiped it all again and this time zypper in works fine from ~/tmp/21:44
lbtdo I need to install gcc to get ?21:49
*** andre__ has quit IRC21:49
lbtdebian has libgcc* iirc21:49
Stskeepsinstall mpc mpfr21:50
Stskeepsgmp as well21:50
Stskeepsour cross compilers don't dep on those?21:50
lbttried them - not enough21:50
lbtinto the target21:50
Stskeepswhy is it talking about libmpc ?21:51
Stskeeps /opt/cross/libexec/gcc/armv7hl-meego-linux-gnueabi/4.6.3/cc121:51
Stskeepsis naturally on SDK side21:51
lbtI'd have noticed eventually21:52
* lbt slinks away21:52
lbtbeing tired is embarrasing21:52
*** pohly has quit IRC21:53
Stskeepswell, you're just reacting rationally21:53
Stskeepsnaturally the error must come from target :)21:54
lbtand frankly, ignoring the path is habit too - there's only usually one root - not 321:55
*** Eren has joined #mer21:56
Stskeepsi really need to give some of the base packages a service check21:58
Stskeepsi don't fully understand why we regressed from perl not triggering rebuilds to that it does now21:58
lbtdid it start failing buildcompare again ?21:59
lbtso we should have /opt/cross/armv7hl-meego-linux-gnueabi/lib/ ?22:02
Stskeepslibmpc is from the sdk itself22:03
Stskeepsit's weird it doesn't have proper requirements22:03
Stskeepsthat's a bug for sure22:03
lbtmy SDK doesn't actually have gcc or anything installed22:04
Stskeepsand i would have thought installing the cross compiler would pull right deps22:04
Stskeepsprobably a bug in packaging from old cross approach22:04
lbtat least I'm finding some real bugs ...22:05
*** NIN102 has quit IRC22:05
jbosstskeeps ,you tried to get sgx woking with wayland?22:06
Stskeepsjbos: right, though there's a much more serious effort going on by rob clark from TI22:07
Stskeepsand i didn't try, i actually succeeded back then22:07
jbosok, he isn't hanging around here, isn't he?22:08
Stskeepsno, he's not, but he's in #wayland22:08
Stskeepswhy are you asking?22:08
jbosI wonder if it is possible the get working "raw" (non x) sgx 3d working on n922:09
jbosI get 2d working22:09
jbosbut not 'print3d' from examples22:10
Stskeepson n9 it's insane because of the manual update display22:10
jbosbut you should be able to turn it to automatic22:11
Stskeepsjbos: nop, doesn't work22:11
Stskeepsit did on n8x0, not on this one22:11
Stskeepsand it looks crap with automatic anyway22:12
jbosso this does not work?22:12
*** Aurium has quit IRC22:13
jbosI wonder if a while with do display-update... work22:14
jbosthere is this "manual" update script22:14
*** gimli has quit IRC22:14
* Stskeeps hands a medal to jbos for 'insane hack of the day'22:15
Stskeepsjbos: run everything fullscreen and kill compositor and you should be fairly ok22:15
Stskeepsin x1122:15
jbos(me don't like x)22:16
*** Crnkoj has left #mer22:16
Stskeepsi don't either, but on the n9 it's not a half bad idea22:16
Stskeepsrobclark's stuff is omap4 only probably though22:17
jboswell well :) guess you need to live with x on certain configurations...22:19
*** JRepin has quit IRC22:19
*** himamura has quit IRC22:19
Stskeepsout of curiousity, is this gecko on top of qt, or?22:21
*** himamura has joined #mer22:22
*** gabrbedd has quit IRC22:23
*** jooncheol has joined #mer22:24
*** jooncheol has quit IRC22:26
jbosit is22:30
jbosstill troubles me to get it compiling in mer sdk without honk22:31
jbossome udev linker error22:31
jbos-honk +qt22:31
jbosundefined reference to `udev_new' ... and so on...22:32
jboswhich is odd since udev-devel is installed and in good version22:32
Stskeepswell, i'd like to see the build log22:33
jbosjust creating a more sufficient one, it first failes with millions of /opt/cross/bin/armv7hl-meego-linux-gnueabi-ld: BFD (GNU Binutils) 2.22 assertion fail elf32-arm.c:1204922:34
jbosduring linking...22:34
Stskeepsthat one is 'normal'22:34
Stskeepsplease file a bug on that assertion22:34
*** jooncheol has joined #mer22:35
jbosso this dude is expected?22:35
Stskeepsso we remember it's there22:35
Stskeepsthe assertion happens occasionally22:35
jboshave seen this on oneiric cross compile22:36
*** jooncheol has quit IRC22:36
Stskeepswe're trying to stick very close to linaro toolchain22:37
lbtStskeeps: any problems with making a couple of sdk modes and dropping them into /usr/share/scratchbox2/modes/ as part of sdk install22:41
*** Eren has quit IRC22:42
Stskeepslbt: shouldn't be22:42
lbtI have a new rabbithole file - it seems to be working - i can pop items and clean up as I come up for air :)22:43
Stskeepsfinally, eglibc 2.15 for x86 has bootstrapped.. 180 succeeded out of 322 so far22:48
*** tomeff has quit IRC23:02
*** tilgovi has joined #mer23:03
*** parancibia has joined #mer23:04
*** parancibia has quit IRC23:06
* Stskeeps calls it a day23:06
*** kavurt has joined #mer23:09
*** tilgovi has quit IRC23:10
*** himamura has quit IRC23:30
*** himamura has joined #mer23:30
*** Guest55802 has quit IRC23:31
*** sandroandrade has quit IRC23:32
*** FireFly has quit IRC23:33
*** jstaniek has quit IRC23:33
*** Termana has joined #mer23:34
*** Termana is now known as Guest4439223:34
*** FireFly has joined #mer23:37
*** jluisn has joined #mer23:49
*** dionet has quit IRC23:49
*** himamura_ has joined #mer23:56
*** himamura has quit IRC23:59

Generated by 2.9.2 by Marius Gedminas - find it at!