Wednesday, 2012-12-19

*** FSCV has joined #mer00:01
*** FSCV has quit IRC00:06
*** shmerl has joined #mer00:07
*** plfiorini has joined #mer00:26
*** gabriel9 has quit IRC00:36
*** sonach has joined #mer00:42
*** yunta has quit IRC00:45
*** himamura_ has joined #mer00:49
*** sonach has quit IRC00:53
*** sonach has joined #mer00:53
*** M4rtinK has quit IRC00:55
*** terietor has quit IRC01:10
*** reels__ has joined #mer01:12
*** himamura_ has quit IRC01:13
*** himamura has joined #mer01:13
*** shmerl has left #mer01:15
*** reels_ has quit IRC01:16
*** icota has quit IRC01:18
*** terietor has joined #mer01:31
*** blauzahl has quit IRC01:34
*** blauzahl has joined #mer01:34
*** nizou has joined #mer01:54
*** nizou has quit IRC02:00
*** npm has quit IRC02:11
*** npm has joined #mer02:12
*** Morpog has quit IRC02:13
*** MrBlueSky125 has quit IRC02:14
*** arcean has quit IRC02:18
*** Morpog has joined #mer02:29
*** Shaan7 has quit IRC02:40
*** Shaan7 has joined #mer02:41
*** KaIRC has quit IRC02:44
*** imunsie has joined #mer02:58
*** imunsie has quit IRC03:05
*** imunsie has joined #mer03:07
*** imunsie has joined #mer03:09
*** Jucato has quit IRC03:22
*** blaroche has quit IRC03:28
*** himamura_ has joined #mer03:36
*** himamura has quit IRC03:36
*** furikku has joined #mer03:59
*** DocScrutinizer05 has quit IRC04:03
*** DocScrutinizer05 has joined #mer04:03
*** pvilja has left #mer04:08
iekkumorning04:09
*** dakovaci has joined #mer04:29
lpotterhi04:47
*** dakovaci has quit IRC04:50
*** yashshah has joined #mer04:54
*** Jucato has joined #mer05:07
*** yashshah has quit IRC05:17
*** yashshah has joined #mer05:18
*** yashshah_ has joined #mer05:22
*** yashshah has quit IRC05:22
*** disco_stu has quit IRC05:27
situMorning all05:30
*** e8johan has joined #mer05:30
*** dakovaci has joined #mer05:45
*** springz has joined #mer05:56
*** user__ has joined #mer06:04
*** dakovaci has quit IRC06:06
*** user__ has quit IRC06:07
*** dakovaci has joined #mer06:08
*** dakovaci has quit IRC06:12
*** Superpelican_ has joined #mer06:13
*** zalan has joined #mer06:16
*** Superpelican_ has quit IRC06:18
*** Superpelican_ has joined #mer06:20
*** Superpelican_ has joined #mer06:22
Superpelican_Stskeeps: How's Nemo Mobile batterylife when compared to Maemo 5?06:24
*** rdqfdx has joined #mer06:27
*** fk_lx has joined #mer06:31
*** yashshah- has joined #mer06:34
*** yashshah_ has quit IRC06:35
*** edgar2 has joined #mer06:42
StskeepsSuperpelican_: maemo5's a product, nemo is a community driven project06:50
Stskeepsmaemo5 obviously had more testing into it, so it's probably better06:50
kulvenowdays one big thing influencing the battery life is activity. If your apps are completely idle, the cpu sleeps and the overall consumption is really low. But if there's multiple applications doing something even once per minute, the overall consumption is significantly higher06:53
*** phinaliumz has joined #mer06:55
*** Superpelican_ has quit IRC06:56
*** skortela has joined #mer07:01
*** lamikr has joined #mer07:01
*** plundstr has left #mer07:04
*** plundstr has joined #mer07:04
*** zalan has quit IRC07:05
*** marquiz has quit IRC07:07
*** trip0 has quit IRC07:07
*** marquiz has joined #mer07:07
*** trip0 has joined #mer07:07
*** Shaan7 has quit IRC07:13
*** Shaan7 has joined #mer07:14
*** melonipoika has joined #mer07:21
*** calvaris has joined #mer07:25
*** calvaris has joined #mer07:25
ka6soxkulve, why significantly higher? sleepd should put it back in a low power mode quickly after that 1x/minute event07:28
*** Sfiet_Konstantin has joined #mer07:29
*** plfiorini has quit IRC07:29
*** ka6sox is now known as ka6sox-away07:30
kulvenot quickly enough07:31
*** dakovaci has joined #mer07:31
*** e8johan has quit IRC07:31
*** e8johan has joined #mer07:32
*** Morpog has quit IRC07:32
*** pirut has joined #mer07:34
*** pirut has quit IRC07:35
*** tilgovi has quit IRC07:36
*** pirut has joined #mer07:37
*** arturo182_ has quit IRC07:43
*** imunsie has quit IRC07:46
*** Sleepy_Coder has quit IRC07:48
*** veskuh has joined #mer07:49
*** Sleepy_Coder has joined #mer07:49
*** otep has quit IRC07:53
*** Morpog has joined #mer07:54
*** otep has joined #mer07:57
*** niqt has joined #mer08:00
*** yashshah__ has joined #mer08:01
*** yashshah- has quit IRC08:01
*** yunta has joined #mer08:05
*** pvanhoof has joined #mer08:05
*** gabriel9|work has joined #mer08:08
*** mvogt has quit IRC08:10
*** fk_lx has quit IRC08:10
*** Eismann_ has quit IRC08:20
*** jpetersen has joined #mer08:25
*** stefan_schmidt_w has joined #mer08:26
*** dakovaci has quit IRC08:27
niqtmorning08:32
*** Martix has joined #mer08:33
*** chriadam is now known as chriadam|away08:34
*** gabriel9 has joined #mer08:34
*** gabriel9|work has quit IRC08:35
*** lamikr has quit IRC08:36
*** blam has quit IRC08:37
*** yunta has quit IRC08:40
*** zenvoid has joined #mer08:42
*** melonipoika has quit IRC08:44
*** Martix has quit IRC08:47
*** slaine has joined #mer08:54
*** yashshah has joined #mer09:00
*** mdfe has joined #mer09:00
*** blam has joined #mer09:01
*** yashshah__ has quit IRC09:02
*** lamikr has joined #mer09:06
*** Nicd- has quit IRC09:08
*** himamura_ has quit IRC09:11
*** ravirdv has joined #mer09:11
*** ortylp has joined #mer09:14
*** sonach has left #mer09:15
*** himamura has joined #mer09:16
Bostikmorning guys09:23
Stskeepsmorn bostik09:23
Bostiktry to make one short trip to town and it still takes 2h...09:27
*** springz has quit IRC09:29
*** popey has quit IRC09:32
*** himamura_ has joined #mer09:35
*** dakovaci has joined #mer09:36
*** himamura_ has quit IRC09:37
ShadowJKIt's all about the total number of wakeups :)09:37
*** notmart has joined #mer09:37
*** himamura has quit IRC09:37
*** springz has joined #mer09:37
ShadowJKOn N900+Maemo5: "while true; do sleep 5 ; done" (sleep 5s, do nothing, repeat) uses about 3 times more power than idle :)09:39
*** popey has joined #mer09:40
ShadowJKOne thing waking up once a minute isn't a big impact, but if there are several things waking up, it starts adding up :)09:41
kulveI would guess that it's same on all platforms. Just on some platforms with limited multitasking it's easier to limit the wakeups because there's not so many 3rd party apps running in the background09:41
ShadowJKyeah09:41
*** niqt_ has joined #mer09:41
ShadowJKI suspect maemo5 is better than others at this, because it was designed to not limit multitasking09:42
*** hazchemix has joined #mer09:43
*** niqt has quit IRC09:44
*** mdfe has quit IRC09:46
*** mdfe has joined #mer09:48
*** whi5key has joined #mer09:49
*** yunta has joined #mer09:49
*** ortylp has quit IRC09:51
*** rcg-work has joined #mer09:52
*** ortylp has joined #mer09:53
rcg-workmorning all09:53
*** Bryanstein has quit IRC09:54
*** springz has quit IRC09:55
*** skortela has quit IRC09:55
*** dakovaci has quit IRC09:56
rcg-workSage, could you also have a look at: https://build.pub.meego.com/request/show/755710:01
rcg-worki need this as dependency for the upcoming qtodo version10:01
*** Bryanstein has joined #mer10:01
*** yashshah has quit IRC10:03
Sagercg-work: sure10:07
Sageseems ok now accepted10:08
*** Grimmer has quit IRC10:08
rcg-workSage, thanks :)10:08
*** Grimmer has joined #mer10:09
*** reels__ has quit IRC10:12
*** reels has joined #mer10:12
*** auri__ has joined #mer10:22
*** arcean has joined #mer10:24
*** lizardo has joined #mer10:25
*** M4rtinK has joined #mer10:25
sledgesmoorning10:25
*** dakovaci has joined #mer10:27
yuntalbt: ping10:29
*** tanty has joined #mer10:29
*** Morpog has quit IRC10:33
lbtyunta: pong10:34
*** yunta has quit IRC10:34
*** jstaniek has joined #mer10:41
niqt_hi rcg10:42
rcg-workhey niqt_10:44
rcg-workmorning sledges10:44
niqt_rcg are you developing with mer-qtcreator?10:49
*** zhxt has joined #mer10:54
*** varikonniemi has quit IRC10:55
*** CosmoHill has joined #mer10:55
*** yunta has joined #mer10:56
*** varikonniemi has joined #mer10:57
*** blam has quit IRC10:57
*** thetet has joined #mer10:57
rcg-workniqt_, nope11:00
rcg-workam currently messing with nexus7 moslo and pa on n711:00
*** dijenerate has quit IRC11:01
kysseMESS WITH IT.11:03
*** Morpog has joined #mer11:03
BostikStskeeps: one thing picked up from qt-releasing which might be interesting for you - qtquick1 is not part of -rcX; hence, no guarantees when it'll get updated to work with essential modules11:10
Stskeepsyeah, figured as much11:10
kulveSage: I created the packages to :devel and not to :stable. I guess that was a mistake?11:11
*** niqt has joined #mer11:12
Sagekulve: use :devel only for now11:12
kulveok, then I did the right thing :)11:12
Sageyes :)11:12
*** ALoGeNo has joined #mer11:13
*** yunta has quit IRC11:15
*** niqt_ has quit IRC11:15
*** dijenerate has joined #mer11:20
*** whi5key has quit IRC11:21
*** Martix has joined #mer11:28
*** arcean_ has joined #mer11:48
*** arcean has quit IRC11:50
*** zalan has joined #mer11:50
*** e8johan has quit IRC11:51
*** e8johan has joined #mer11:56
*** niqt has quit IRC11:58
*** blam has joined #mer11:58
*** icota has joined #mer12:01
*** arcean_ is now known as arcean12:03
kulveis there some automatic way of adding those package groups to repos from patterns.xml? I.e. some specifically named components or do I need to run manually "osc meta pattern" for each repo?12:05
*** dakovaci has quit IRC12:06
Sagekulve: there is automatic way to do that yes.12:20
Sagekulve: https://github.com/nemomobile/nemo-patterns/tree/patterns-n950 for example is for n95012:21
Sageit is packaged and then the package is put to the repository and boss automatically does the patterns12:21
*** aportale has joined #mer12:23
*** sivang has joined #mer12:24
sivangdiscussion welcome, http://www.mail-archive.com/mer-general@lists.merproject.org/msg01012.html12:24
*** rcg-work has quit IRC12:24
kulveSage: thanks12:25
situStskeeps: Congrats, so you're now a chief engineer.12:25
Stskeepssitu: thanks12:26
CosmoHillStskeeps: congrats12:26
Bostikputting the 'pro' back in 'propellerhead'?12:26
Bostikand that was NOT an insult12:27
sivangcongrets, but you were already cheif architect right?12:27
Stskeepsno12:27
*** DocScrutinizer05 has quit IRC12:27
Stskeepsi'm project architect in mer, but at jolla i was just an engineer12:27
Stskeepsnow i'm chief research engineer, or something12:28
Stskeeps:P12:28
* sivang bows and causes a Jolla tide for Stskeeps 12:28
*** DocScrutinizer05 has joined #mer12:28
Sagekulve: that git tree is using tag_git script for packaging12:29
*** yunta has joined #mer12:33
*** auri__ has quit IRC12:34
*** auri__ has joined #mer12:34
*** wanggjghost has joined #mer12:34
*** yunta has quit IRC12:35
*** rcg-work has joined #mer12:40
*** KaIRC has joined #mer12:40
*** ortylp has quit IRC12:44
*** whi5key has joined #mer12:52
Bostikthe things we do... /tmp/gold/ld: fatal error: libQt5WebKit.so.5.0.0: mmap: failed to allocate 585291404 bytes for output file: Cannot allocate memory12:55
Stskeepshow about we just ditch the debug symbols?12:55
Stskeeps:P12:55
* Bostik finds a bigger pen12:56
BostikStskeeps: we already do :)12:56
Stskeeps..12:56
Bostikthat is without debug symbols, with symbol filter, and with gold12:56
*** CosmoHill has quit IRC12:59
Bostiklet's see if I can drop all the fancy stuff out (again)13:00
*** whi5key has quit IRC13:01
*** CosmoHill has joined #mer13:03
*** phinaliumz has quit IRC13:05
sivanghas anybody done work already on supporting the HTML5/JS developer story in Qt Creator btw?13:09
*** thetet has quit IRC13:14
*** arcean has quit IRC13:17
*** reels_ has joined #mer13:18
*** reels has quit IRC13:20
rozhkov_Sage: is it possible to get debugsymbols for connman packaged?13:20
Stskeeps /etc/zypp/repos.d13:21
Stskeepsenable the debug repo in mer-core.repo13:21
Stskeepszypper ref13:21
rozhkov_Stskeeps: thanks!13:21
*** FSCV has joined #mer13:22
aportaleAnyone who had this issue with the SDK Control Center? http://pastie.org/555266513:23
*** Sfiet_Konstantin has quit IRC13:26
*** arcean has joined #mer13:28
*** reels_ has quit IRC13:28
*** jayrulez has quit IRC13:38
bfederau_Sorry for asking again. I need some help setting up a local Mer OBS. I don't find any error message in the OBS log files, but the spinner of the build result box in the project is permantently running and I get no results. I have no idea where to look to solve this issue.13:47
*** kallela has quit IRC13:48
*** kallela has joined #mer13:48
Stskeepsbfederau_: check backend logs?13:48
*** kallela has quit IRC13:48
*** vgrade_ has joined #mer13:49
bfederau_Stskeeps: yeah. one othe last entris look like "POST localhost:5352/search/request?match=%  ....... 200"13:50
*** spoofy has quit IRC13:50
Stskeepsbfederau_: is anything listneing on localhost 5352?13:50
bfederau_Stskeeps: netstat says: tcp        0      0 0.0.0.0:5352            0.0.0.0:*               LISTEN13:51
bfederau_Stskeeps: and I thought that the number 200 in the log means HTTP code 200 for everything ok13:53
*** lizardo has quit IRC13:55
*** jukkaeklund has joined #mer13:56
*** vgrade_ has quit IRC13:59
*** Hoolxi has joined #mer14:04
[ol]Stskeeps: Something's strange is happening with MAKEDEV package. I'm unable to git clote its repo.14:09
[ol]git clone ssh://ol@review.merproject.org:29418/mer-core/MAKEDEV14:10
[ol]Cloning into 'MAKEDEV'...14:10
[ol]fatal: '/mer-core/MAKEDEV': not a Gerrit project14:10
[ol]fatal: The remote end hung up unexpectedly14:10
[ol]I was thinking that it was removed, but no, it's here: http://gitweb.merproject.org/gitweb?p=mer-core/MAKEDEV.git;a=summary14:10
*** yashshah has joined #mer14:11
*** sivang has quit IRC14:12
*** herangr has joined #mer14:15
*** CosmoHill has quit IRC14:16
Sageisn't MAKEDEV obsoleted?14:21
[ol]Sage: May be it is, but udev requires it.14:22
[ol]Should I remove this requirement from udev's spec file?14:23
Sageumm...14:24
*** CosmoHill has joined #mer14:24
Sage[ol]: udev was merged to systemd and those shouldn't depend it anymore14:25
*** Superpelican_ has joined #mer14:25
[ol]Sage: Do you mean, we don't need udev package anymore?14:26
CosmoHillsage, have you heard about eudev?14:27
Stskeeps[ol]: /me looks14:30
*** arturo182 has joined #mer14:31
Stskeeps[ol]: uhhh..14:32
*** Superpelican_ has left #mer14:35
*** vgrade_ has joined #mer14:36
*** spre has joined #mer14:37
*** jstaniek has quit IRC14:38
*** Hoolxi has quit IRC14:39
Stskeeps[ol]: that works for me. let me look at permissions14:39
Sage[ol]: yes we have udev package but it is provided by systemd14:40
Sage[ol]: see http://gitweb.merproject.org/gitweb?p=mer-core/systemd.git;a=summary14:40
SageStskeeps: we should really start marking packages obsoleted and remove the content from the git trees and adding obsoleted info file or something14:41
Stskeepsyes14:41
Stskeeps[ol]: try now?14:42
SageI think there is alrady ~10 not used packages in that gitweb14:42
[ol]Are you sure that udev is obsolete? I see udev package in Fedora.14:42
Stskeeps[ol]: udev binary package, not source package14:42
[ol]And it was built from udev SRPM, not systemd.14:42
Sage[ol]: what version? see http://lwn.net/Articles/490413/14:43
[ol]Stskeeps: Now cloning MAKEDEV works.14:43
Stskeeps[ol]: ok, i messed up permissions - sorry about that14:43
Stskeepsand thanks for noticing14:43
vgrade_Qt5.0 FINAL14:43
*** vicmiclovich has joined #mer14:44
Stskeepsfinally14:44
Stskeeps:P14:44
[ol]Should we switch to Qt5?14:45
Stskeepswe provide qt4 and qt5 side by side14:45
[ol]Is Qt5 source compatible to Qt4?14:46
*** icota has quit IRC14:46
Bostikmostly14:46
*** jukkaeklund has quit IRC14:47
[ol]That's great! This means that there's no need to keep Qt4 for compatibility.14:47
Stskeepswell, in practice you do14:48
[ol]There's still qt3 package in Fedora. And some packages still depend on it...14:48
Stskeepssome will want to still build qt4-mer products, so14:49
*** zenvoid has quit IRC14:50
*** niqt has joined #mer14:51
*** arcean has quit IRC14:51
*** zenvoid has joined #mer14:51
*** thetet has joined #mer14:55
*** jayrulez has joined #mer14:56
[ol]Uh oh... "udev < 184 is obsoleted by (installed) systemd-187-1.x86_64"14:56
*** ravirdv has quit IRC14:57
*** lizardo has joined #mer14:58
[ol]Stskeeps: I have all packages from the list you provided built successfully.14:58
Stskeepscongratulations :)14:58
Stskeepslet's start with patches to gerrit and get those in, then i can load in them to a obs and kickstart the build14:59
vgrade_looks like qtwebkit has been split up into qtwebkit1/215:00
*** Jay_BEE has quit IRC15:00
vgrade_ignore me, its still one lump15:04
*** spoofy has joined #mer15:05
Stskeepsi think it might make sense, if possible, to build them independently..15:05
*** niqt has quit IRC15:08
*** Jay_BEE has joined #mer15:09
*** Nicd- has joined #mer15:17
*** niqt has joined #mer15:18
*** CosmoHill has quit IRC15:20
*** ortylp has joined #mer15:21
*** melo_ has joined #mer15:22
*** hazchemix has quit IRC15:24
*** CosmoHill has joined #mer15:25
*** melo_ has quit IRC15:26
*** notmart has quit IRC15:30
*** notmart has joined #mer15:31
*** notmart has joined #mer15:31
Bostikvgrade_: webkit2 is the split-process plugin system, but the Qt part has indeed been split into "just core" and "UI parts"15:33
*** thopiekar has joined #mer15:33
*** rcg-work has quit IRC15:34
BostikFFS!15:34
Stskeepsmm?15:36
*** Aurium has joined #mer15:36
Bostikdom/Document.h and webkit2<->fullscreen-api stuff again15:37
Stskeepsah15:37
BostikI should have most of tomorrow free for looking into that finally with proper thought15:38
Bostikthat jungle of multiple build-system configs and indirections needs more than a machete15:38
*** himamura has joined #mer15:38
*** edgar2 has quit IRC15:39
Stskeeps /usr/bin/ld: cannot find -lQtCLucene15:40
Stskeepshmm15:40
Stskeepsah, nm15:40
Stskeepspatch issue15:40
*** jstaniek has joined #mer15:44
*** himamura has quit IRC15:45
*** himamura has joined #mer15:45
*** thopiekar_ has joined #mer15:47
*** ortylp has quit IRC15:48
*** thopiekar has quit IRC15:51
*** knuty_ has joined #mer15:52
*** biatch0 has quit IRC15:54
*** edgar2 has joined #mer15:58
*** wanggjghost has quit IRC16:00
*** vicmiclovich has quit IRC16:01
*** ortylp has joined #mer16:02
*** herangr has quit IRC16:03
*** gabriel9 has quit IRC16:04
*** lamikr has quit IRC16:10
*** mdfe has quit IRC16:12
*** niqt has quit IRC16:17
*** jayrulez has quit IRC16:28
*** slaine has quit IRC16:28
*** NIN101 has joined #mer16:28
*** pirut has quit IRC16:31
*** pvanhoof has quit IRC16:34
*** Sfiet_Konstantin has joined #mer16:34
*** NIN101 has quit IRC16:37
*** vgrade_ has quit IRC16:38
*** Sfiet_Konstantin has quit IRC16:38
*** pvanhoof has joined #mer16:46
*** plundstr has left #mer16:49
*** NIN101 has joined #mer16:50
*** zhxt has quit IRC16:56
*** thopiekar_ has quit IRC17:02
*** thopiekar has joined #mer17:04
*** knuty_ has quit IRC17:07
*** ortylp has quit IRC17:17
*** stefan_schmidt_w has quit IRC17:27
*** arcean has joined #mer17:31
*** calvaris has quit IRC17:36
*** Sfiet_Konstantin has joined #mer17:37
*** thetet has quit IRC17:43
*** Morpog has quit IRC17:48
*** SfietKonstantin has joined #mer17:48
*** Sfiet_Konstantin has quit IRC17:49
*** chouchoune has quit IRC17:55
*** lizardo has quit IRC17:55
*** chouchoune has joined #mer17:55
*** ortylp has joined #mer17:55
*** icota has joined #mer17:58
*** blam_ has joined #mer18:04
*** blam has quit IRC18:04
*** nizou has joined #mer18:12
*** yashshah has quit IRC18:15
*** yashshah has joined #mer18:15
*** rcg has joined #mer18:23
*** ortylp has quit IRC18:26
*** aportale has left #mer18:41
*** terietor has quit IRC18:41
*** icota has quit IRC18:44
*** shmerl has joined #mer18:46
*** lpotter has quit IRC18:48
*** lpotter has joined #mer18:51
*** SfietKonstantin has quit IRC18:53
rcgSage, on b.m.o you have busybox in the nemo:devel:hw:n950-n9 project19:00
rcghowever, we also use it for the nexus7 moslo19:00
Sageyes, well it is because of moslo requirement19:00
rcgwould a nemo:devel:hw:common project make sense?19:01
Sagewell, there isn't much things like that really. busybox is something that should go to mer core19:01
rcgalright19:01
rcgso i will upload a busybox into the n7 hw adaptation for now19:01
Sagejust needs quite a lot of packaging and conditions in place.19:02
*** lpotter-home has joined #mer19:02
rcgonce there is a busybox in mer core i would check if the n7 moslo also works with this one and delete the n7 specific package then19:02
rcgdoes this make sense to you?19:02
Sageyes19:03
Sagesame thing with n95019:03
Sagealso the mtev xorg driver just went to mer core so we can drop that soon from those projects as well19:03
*** lpotter has quit IRC19:04
rcgah, nice19:04
*** lpotter-home is now known as lpotter19:04
rcgwrt kexec stuff, we use some special patches to enable kexec hardboot19:04
Sagethere has been quite a lot of temporary solutions but while the adaptation amount increases we get better in this19:04
rcgalright19:04
*** terietor has joined #mer19:05
rcgbasically i want to reduce unneeded duplication and wanted to give a heads up so that you are aware of this19:05
Sagesure19:05
*** shmerl has quit IRC19:06
*** dijenerate has quit IRC19:07
Sagedoing a :hw:common would be very hard to handle in general. hw adaptations seem to have < 20 packages in them and some of these are binary blobs that require certain versions of certain libs etc.19:07
Sagefor example for n9xx-common there is libnl1 because of binary file requirement19:07
*** lizardo_ has joined #mer19:08
*** korhojoa has quit IRC19:09
rcgalright, makes sense19:09
*** furikku has quit IRC19:11
*** shmerl has joined #mer19:14
CosmoHillcyas19:20
*** CosmoHill has quit IRC19:20
Stskeepsevening shmerl19:21
shmerlHi Stskeeps. Was lbt around?19:21
Stskeepsunsure19:23
*** icota has joined #mer19:23
*** yashshah_ has joined #mer19:26
*** yashshah has quit IRC19:28
*** Aristide has joined #mer19:33
*** rubdos has joined #mer19:38
*** ka6sox-away is now known as ka6sox19:45
lbtshmerl: hey19:46
shmerllbt: Hi. I tested that RPM (for base SDK, i486).19:46
shmerlThe target though can't be updated.19:46
shmerlSince RPM itself doesn't work with that bug19:46
shmerlEven manually and after rebuidling RPM db.19:47
lbt553 ?19:47
shmerlI guess I can update the files purely by hand19:47
shmerlBut #65419:47
shmerl*Bug #65419:48
MerbotMer bug 654 in zypper "zypper: double free or corruption Aborted" [Major,Resolved: fixed] https://bugs.merproject.org/show_bug.cgi?id=65419:48
lbthmm - I had the same bug and rpm worked19:49
shmerlfor the target?19:49
lbtok - so maybe rm -f /var/lib/rpm/_db*19:49
lbtalso - is this on device?19:50
lbtI got the problem on my 95019:50
shmerlI.e. sb2 -m sdk-install -R rpm -Uhv ...19:50
shmerlNo, it's in the chroot SDK19:50
lbtoh19:51
lbtright - I saw the n900 in the text19:51
shmerlThat's for Nemo/N950 target under chroot.19:51
rcgwhat's the api url for b.m.o?19:52
Stskeepslbt: btw, whenever you're ready, a prerelease would b e nice19:52
lbtStskeeps: ok19:52
rcgi can't seem to find it19:52
shmerlBut the base was udpated first: libsolv-tools libsolv0 libudev systemd systemd-sysv19:52
lbtrcg: https://api.merproject.org/19:52
rcglbt, thanks19:52
lbtshmerl: OK - so the bug's confusing - it doesn't mention sb219:53
shmerlWell, the one I filed - does.19:53
shmerlbug #65519:53
MerbotMer bug 655 in SDK "zypper refresh through sb2 fails with segfault for Nemo target" [Normal,Assigned] https://bugs.merproject.org/show_bug.cgi?id=65519:53
shmerlThey seem to coinsided.19:54
lbtok - no, it wasn't a dupe of 654 :)19:54
shmerlOK :)19:54
*** melonipoika has joined #mer19:55
lbtok - so maybe rm -f /srv/mer/targets/XXXXX/var/lib/rpm/_db*19:55
lbtor sb2 -m sdk-install -R rm -f /var/lib/rpm/_db*19:56
lbtthen rebuilddb19:56
shmerlOK, I'll try that (a bit later).19:56
lbtI think we're going to have a general problem with the SDK moving from x86 __db* to native ones19:57
*** rcg has quit IRC19:58
lbtStskeeps: target date?20:02
lbtwhat's planned to go in it?20:03
lbtand pre-release or snapshot?20:03
*** rcg has joined #mer20:03
Stskeepsprerelease20:04
Stskeepswill you be around on 24/12? ;)20:04
lbtyeah20:05
Stskeepsk, well, let's make it then then20:05
*** CosmoHill has joined #mer20:15
lbtoff it goes20:15
*** Aurium has quit IRC20:21
*** yashshah_ has quit IRC20:23
Stskeepsthank you20:34
Sagelbt: we need to fix zypper libzypp libsolv chain a bit more imo20:39
Sagelbt: adding explicit requirements "Requires: x = version_of_x" to zypper -> libzypp -> libsolv20:39
lbtOK - why?20:40
Sagethat would make sure in the future that each one of those is build against each other and this would not happen20:40
Sagethere was somekind of abi break apparently that broke when people got updates for only part of that chain20:41
lbtfrom what I recall from yesterday, the chain looked well specified now20:42
Sagewell, one can update libsolv alone atm.20:44
lbtno, libsolv and -tools must match now20:44
Sageyes. well I meant those two20:44
Sagei.e., take old image and you can do rpm -Uvh libsolv0 libsolv-tools20:44
lbtand must be Requires:       libsolv-tools >= 0.1.0 for libzypp20:44
Sagethat doesn't prevent updating libsolv and breaking libzypp with that20:45
lbtwell, technically I suppose we should do Requires:       libsolv-tools >= 0.1.0, libsolv-tools < 0.2.0 for libzypp20:46
lbtbut rpm isn't as good as ruby ;)20:46
lbtI also don't see an explicit zypper depenendcy on libzypp - is that autprov ?20:47
Sageyes that is20:47
Sagebut for libsolv there isn't such thing which was the cause of the break20:48
lbtno, it goes via -tools20:48
lbtso in theory there isn't actually an abi - not if it uses -tools ?20:48
SageI think you are missing my point :)20:49
lbtcould be20:49
Stskeepseither way, libsolv should have upped their soname if they upped api20:49
Stskeepser, abi20:49
lbtdo zypp and solv both assume a binary format and not use shared code?20:50
Sageyes, that too. But I was saying that we could prevent it anyway with explicit ver dependencies20:50
Stskeepsmight make sense to tighten it, yes20:50
Sagelibsolv, libzypp and zypper are anyway very tight to each other.20:50
lbtso tighter than   Requires:       libsolv-tools >= 0.1.020:51
SageThe problem we had now was that vendors used stuff that wasn't build against the same core. Thus there was dependnecy from package X to libzypp version x.y. Which prevented libzypp update but libsolv got upated fine as there wasn't anything preventing that20:51
Sagelbt: you still miss the point here :)20:51
Sagelbt: if it is >= you can always install newer libsolv* that might break stuff without anyone complaining20:52
lbtyou can also install a bug-fixed libsolv without having to redo zypper20:52
SageThus I would suggest using %define libsolv-version %{expand:%(rpm -q --qf '[%%{version}]' libsolv0)}20:53
Sageand then using that as requirement.20:53
lbtso you're re-inventing soname as an rpm macro?20:53
Sagewell, I'm just saying that zypper is one of the components that should never break in updates as it gets the updates.20:54
Stskeepslbt: nah, a tighter binding20:55
Stskeepslbt: libfoo.so.1 can be anything20:55
Stskeepspackage = %{libsolv-version}  is "it needs to be this binary"20:55
lbtI guess if it's a special case for a sensitive package20:55
*** lizardo_ has quit IRC20:55
Sagealso I only saying %{version} there which still allows rebuilds to be installed as it doesn't do -%{release} though not sure if that should be there as well20:56
Sagelbt: yes, this is a bit special case imo.20:56
*** notmart has quit IRC20:57
lbtand this is libzypp -> libsolv20:59
*** Eliel has quit IRC20:59
_ThomasDoes anyone here have an idea if QT5 has a rendering backend for OpenGLES2?  If all I want is one app running on my device, I don't see X11 as a good option?21:08
*** Grimmer has quit IRC21:08
Sageyes, zypper -> libzypp works as they change so on each version /usr/lib/libzypp.so.1200 is 12.2.021:08
*** melonipoika has quit IRC21:12
*** Grimmer has joined #mer21:13
*** yunta has joined #mer21:19
*** NIN101 has quit IRC21:23
*** erbo has quit IRC21:24
vgrade_Thomas: sure does, look at eglfs. match with framebuffer gles/egl drivers21:25
vgrade_Thomas: also we have qtwayland running on eglfs on Pi21:26
*** jpetersen_ has joined #mer21:28
vgrade_Thomas: http://www.youtube.com/watch?v=k4yrJiDH2LI21:29
vgrade_Thomas: also https://wiki.merproject.org/wiki/Zephyr project is designed for that usecase21:31
*** Aristide has quit IRC21:32
*** Aristide has joined #mer21:32
*** jpetersen has quit IRC21:32
*** lpotter has quit IRC21:32
shmerlvgrade: Do you think Nvidia will start supporting Wayland for Tegra too?21:33
*** lpotter has joined #mer21:35
*** CosmoHill has quit IRC21:36
*** CosmoHill has joined #mer21:41
_Thomasvgrade: Someone should do that our dongle ;)21:41
Stskeepsmali used to have a wayland compositor at st-e i thinjk21:42
_ThomasStskeeps: I heard about it too, but I don't know the state21:43
Stskeepsyeah, this was ages ago21:44
_ThomasStskeeps: I talked to the guy that wrote it at Linaro Connect21:44
*** aboyer has joined #mer21:44
_ThomasWe chatted about it briefly while talking about a different topic21:44
Stskeepsanyway.. it should be possible to do with mali sources, not exactly rocket science with hwmem/ump/etc21:45
_ThomasNo, but is it better to do qtwayland than qteglfs / directfb?21:46
Stskeepswell, for one app it doesn't matter21:48
Stskeepsfor multiple, it atters21:48
*** disco_stu has joined #mer21:48
Stskeepseglfs basically means "put this on framebuffer", ish :P21:49
Stskeepsfullscreen egl, etc21:49
_ThomasEGLFS means EGL FullScreen, I found out21:49
*** tanty has quit IRC21:49
_ThomasBut they combine EGLFS and DirectFB also?21:49
Stskeepsyou can, but i've never used that personally21:49
_ThomasSo basically, if you want more than one app, and don't care about x11, qtwayland is the better route?21:50
_ThomasIs wayland mature enough?21:50
Stskeepsyes, if you can get proper wayland support for your gfx stack, it's better to go that way21:50
Stskeepsi mean, you don't need a full KDE on there.. :P21:51
_ThomasI take it that qtwebkit and all the other bells and whistles of Qt works on any backend?21:51
Stskeepsi can't speak for qtwebkit as it's a monster, but yeah, that's the idea21:51
StskeepsQPA in qt abstracts it out nicely21:52
*** Eismann has joined #mer21:54
shmerlwayland is considered to have a stable release now, so they are making 1.0.x releases.21:56
*** icota has quit IRC21:58
shmerlIf you use wayland you can get all their potential future addons with it. (Like network transparency for example). Otherwise - you are on your own.21:58
*** zenvoid has quit IRC22:00
*** niqt has joined #mer22:00
*** Aristide has quit IRC22:00
_Thomasshmerl: Wayland is considered to have a stable protocol22:02
_Thomasshmerl: That is why they are making 1.x releases22:02
shmerlRight, they are still going to add features.22:02
Stskeepsdoesn't take much for the actual buffer sharing anyway22:03
Stskeepsone of my implementations is <300 lines of code, when using gralloc and android native handles22:03
shmerlHere is an interesting example: https://www.youtube.com/watch?v=L8zwnqKysfI&t=1h11m51s22:07
*** hasselmm_ has quit IRC22:09
[ol]Do you guys know, is there any macro in Mer rpm to check whether the package is built in Mer? (Like %{fedora} on Fedora, %{rhel) on Red Hat, %{suse_version} on SUSE.)22:15
*** mvogt has joined #mer22:15
*** rubdos has quit IRC22:17
*** Morpog has joined #mer22:17
*** blam_ has quit IRC22:18
*** cxl000 has quit IRC22:19
*** zalan has quit IRC22:20
_ThomasWayland + Weston seems really cool :)22:22
Stskeepsyou should see qml compositor then..22:23
Stskeepsdid you see the wolfenstein compositor?22:23
_Thomasno? :)22:23
Stskeepshttp://www.youtube.com/watch?v=_FjuPn7MXMs22:23
_Thomascool :)22:26
_ThomasBut I guess you need to recompile all apps to run on top of wayland, right? :)22:26
_ThomasAnd maybe rewrite?22:26
Stskeepswell, if they're qt5 already, then no22:27
Stskeepsit's abstracted in qpa22:27
_Thomasok22:27
Stskeepsyou normally attack the problem at toolkit side22:27
Stskeepsgtk, qt, etc22:27
_Thomasand I guess qt is the one that is most mature?22:27
Stskeepsit depends on definition but it's okay well along22:28
Stskeepsi think the gtk guys also did a pretty good job but ubuntu doesn't seem to want to enable it in their packages22:28
Stskeeps.. if you trust phoronix22:28
shmerlWolf compositor, lol :)22:29
*** himamura has quit IRC22:29
_ThomasSo, the ideal would be if there was a distribution for ARM-based platforms that had all this22:30
Stskeepsyeah, well, mer comes with wayland oob22:32
*** pvanhoof has quit IRC22:32
Stskeepssoon final qt5 too22:32
_Thomasis there an howto somewhere on how to build qt5 from scratch?  I tried searching on the wiki for qt, but didn't find any22:33
_Thomas(I was hoping for x-compiling)22:33
Stskeepslook at guides for rpi maybe22:34
shmerlStskeeps: Since qt5 just went stable, do you think Jolla will be able to be base Sailfish on it out the box too, or only in subsequent releases?22:36
Stskeepsdunno22:39
shmerlLooks a bit tight. But qt5 and qt4 will be able to coexist? So even if it'll be based on qt4, one could pull qt5 from Mer repo, and use it there?22:40
*** CosmoHill has quit IRC22:41
*** nizou has quit IRC22:42
Stskeepsyes22:42
shmerlThat's good.22:44
*** dakovaci has joined #mer22:45
*** blam_ has joined #mer22:48
SpeedEvilStskeeps: cool! (wolf3d)22:52
*** niqt has quit IRC23:01
*** CosmoHill has joined #mer23:04
*** jayrulez has joined #mer23:05
*** icota has joined #mer23:11
*** dakovaci has quit IRC23:15
*** Aristide has joined #mer23:15
*** spangles has joined #mer23:23
*** spangles has quit IRC23:25
*** gabriel9 has joined #mer23:25
*** thopiekar has quit IRC23:37
*** fcorrea has quit IRC23:37
*** fcorrea has joined #mer23:42
*** fcorrea has joined #mer23:43
*** blaroche has joined #mer23:44
*** Aristide has quit IRC23:45
*** NeuhNeuh has joined #mer23:45
*** rcg has quit IRC23:46
*** chriadam|away is now known as chriadam23:52
CosmoHillnight night23:56
*** CosmoHill has quit IRC23:56

Generated by irclog2html.py 2.11.0 by Marius Gedminas - find it at mg.pov.lt!