#mer log for Saturday, 2012-06-23

*** shrikrishna has quit IRC00:00
*** himamura has joined #mer01:11
*** swer has quit IRC01:42
*** swerden has joined #mer01:42
*** xnt14 has quit IRC01:42
*** dijenerate has quit IRC01:44
*** dionet has quit IRC01:46
*** xnt14 has joined #mer02:10
*** Alison_Chaiken has quit IRC02:11
*** tsdedst has quit IRC02:30
*** tilgovi has quit IRC02:33
*** dijenerate has joined #mer03:09
*** xnt14 has quit IRC03:20
*** furikku has joined #mer03:25
*** xnt14 has joined #mer03:25
*** [JT]- has quit IRC03:43
*** himamura has quit IRC03:52
*** himamura has joined #mer03:53
*** himamura has quit IRC04:04
*** clopez has joined #mer04:59
*** himamura has joined #mer05:06
*** rdqfdx has joined #mer05:16
Stskeepszzz05:18
*** clopez has quit IRC05:19
timophmorn05:25
*** pomm has quit IRC05:29
* Stskeeps ponders if he can do a qt-related demo today..05:33
timophI'm betting on yes05:34
timoph:)05:34
* timoph plans to try to survive zombie apocalypse today05:35
Stskeepsthey're rather time limited and i'd like to show the "change one line and deploy to device" scenario..05:35
timophyou're at QtCS?05:36
Stskeepsyeah05:36
*** gimli_ has joined #mer05:36
*** pomm has joined #mer05:36
timophwhat's the wibe among Qt devs there?05:37
timophregarding future & all that fud05:37
Stskeepsfairly good, people are very up for finishing up qt5 and getting it out the door05:37
Stskeepsit's a open source project, so05:38
timophgreat to hear05:38
timophyep. I think I said it somewhere that I believe that Qt will survive even if N totally abandons it05:38
Stskeeps:nod:05:38
timophit's just too good stuff to waste05:38
timophbtw, apparently I've done something right at work. I was told I got a raise just before leaving for vacation05:39
Stskeepscool05:39
Stskeepscongrats :)05:40
timophsomething in the lines of "don't be surprised when you check your next paycheck"05:40
timoph:)05:40
Stskeepsat least your customer seemed very happy about your work05:40
Stskeepstimoph: ah, so "where is my paycheck"? ;)05:40
timophx)05:40
* Stskeeps ponders idly05:41
Stskeepsdo we have quilt in sdk anywhere?05:41
Stskeepsoh wait, mer core has quilt05:41
timophproblem solved :)05:41
timophmer to the rescue05:41
Stskeepsi just want to try and disspell obs bad reputation05:41
timophthat reminds me, we should try to resurrect the creator-obs plugin05:42
timophand get mer target working with creator05:42
Stskeeps:nod:05:42
* timoph adds a sticky note to try the plugin after failing to survive the zombie apocalypse05:43
timophI guess it should work at least with older versions of creator. planning to check if it works with the current code05:44
timophand figure out how to start maintaining it05:45
timophcreator pluging development is a bit awkward since you need to compile the plugins with the creator code05:45
*** nj3ma has joined #mer05:46
timophthe plugins aren't independent from the creator code05:46
* timoph made testplanner/runner-ui work as qtcreator plugins back in the meego-qa times05:49
Stskeepscool05:49
Stskeepsdidnt know that05:49
timophhmmh. I wonder if that code still exists05:50
timophseems it does05:51
timophto be fair - I did the prototyping for it and based on that kyranto did the implementation that you see in the source tree05:55
timophI liked doing that stuff. building hacky proof of concept prototypes that other people used as a starting poing05:57
dm8tbrtimoph: that's the most fulfilling, isn't it. show it's possible and then let someone else take care of the menial work... ;)05:59
timophyep :)05:59
Stskeepsresearch and management? ;)06:08
*** beford has quit IRC06:12
timophhmmh. anyone remember the api to used to draw stuff on n9's standby screen06:20
* timoph wants to draw mer logo to it :)06:20
*** tilgovi has joined #mer06:22
Stskeepssomething lpm06:23
*** phaeron has joined #mer06:23
timophhmmh. seems frals has done the work already :)06:25
timophhttps://github.com/frals/lpmcustomizer06:25
Stskeepsyep06:25
phaeroncool :D06:28
*** araujo has quit IRC06:35
*** tadzik has quit IRC06:38
timophhttp://timoph.fi/kuvei/lpm.jpg \o/06:54
Stskeepsphaeron: there's really no way to ask build to just stop before build?06:54
Stskeepstimoph: \o/06:54
phaeronStskeeps: build as in obs build or the build script or local build ? :D06:56
Stskeepsas in 'build'06:56
Stskeepsie, stop after setting up chroot06:56
Stskeepsbefore running rpmbuild06:56
Stskeepsphaeron: OK, for future reference, --shell < /dev/null07:01
Stskeepsfor osc build07:01
phaeronStskeeps: osc chroot doesn't setup the chroot ?07:02
Stskeepsnot in the way i'd like it to07:02
Stskeepsi'm planning on demoing "how to make OBS a lot less painful for qt developers07:02
phaeronyeah in reference to conversation we had07:04
Stskeeps0yes07:06
*** himamura has quit IRC07:18
Stskeepsphaeron: i just made life a lot easier.. moment, will try to do a pastie demo07:20
Stskeepsphaeron: http://pastie.org/413655407:24
Stskeepsplease note no tarball step07:24
*** NIN101 has joined #mer07:30
lbtmorning ll07:31
Stskeepsmorn07:32
dm8tbrmoaning07:33
Stskeepslbt: any objections to breaking rpm strange upstream philosophies in order to make developer life easier?07:34
lbteg?07:34
Stskeepslbt: i'd like to be able to short-circuit to making binary packages07:36
Stskeepsthey dont like it as it allows people not to use pristine sources07:36
Stskeepsand well07:36
Stskeepsdevelopers are the sources of pristine sources..07:36
Stskeeps:P07:36
lbtie :   build --cd-to-this-dir-do-make-install-and-package-it src/07:37
Stskeeps.. ish07:37
Stskeepslook at my pastie07:37
lbt--rsync-src=. --rsync-dest=/home/abuild/rpmbuild/BUILD07:37
Stskeepsright07:37
lbtyeah - I was writing that at the same time as the guys who did the patches that went upstream :)07:38
phaeronit should work with a cloned git too07:39
phaeronbut different story07:39
Stskeepsyes07:40
lbtand bind-mount depending on what you want07:40
lbtwhat about using osc build to make the sb2 target?07:42
Stskeepslbt: one way of doing it, sure07:43
Stskeepsmy gripe there would be that it would be a PITA to setup rpm build directories..07:43
lbtskipprep ?07:43
Stskeepsyes07:44
Stskeepsjust added that functionality07:44
Stskeepsfor demo purpose07:44
lbtOK - rpmbuild ?07:44
Stskeepsit --short-circuit --bc instead of --ba07:45
Stskeeps.. i want a better way07:45
Stskeepsjust showing general concept07:45
lbtI'd be happy to hack rpmbuild07:45
lbtI think it forces some hoops07:45
Stskeeps:nod:07:46
lbtthe problem would be the unreliability - some spec files would ignore any hacks07:46
*** BeholdMyGlory has quit IRC07:46
*** Behold has joined #mer07:46
Stskeepssure07:47
lbtthe objective here is to synchronise/share a desktop source tree with an rpm/build/sb2 clean environment and support incremental make-driven packaging without make clean each time07:47
Stskeepsyes07:48
*** rcg has joined #mer07:48
lbtwhy didn't you say that :P07:48
Stskeepsand for the nasty packages, well, those aren't the 95% of developers anyway..07:48
Stskeeps:P07:48
lbttrue07:48
Stskeepsbut even being able to remove the tarball step is a lot less time07:48
Stskeepsso that's the angle i attack first07:48
lbtyou know - this is something I could do quite quickly; wondering if you'd rather focus on something I can't do like the llvm VMs ?07:52
Stskeepscan i tell you in 40 mins? :P07:53
lbt:)07:53
Stskeepsdemos here at 10:30 and i want to see if i can demonstrate at that point a sane development cycle07:53
lbtgo for it07:54
lbthttp://qt-project.org/groups/qt-contributors-summit-2012/wiki/Packaging-Qt ?07:55
Stskeepsi took notes07:55
* lbt goes back to the infra cleanup/security updates07:56
phaeronlbt: Stskeeps gets all the fun :D07:56
Bostik"Carsten, merproject.org, packaged qt5 since forever, etc"08:02
Bostikgood to know our timespan has been noticed08:02
Stskeepsyeah, it's meant to be understood as "in mer we've had packaged qt5 since forever"08:03
Stskeepsinstead of me doing it08:03
Stskeepsjust wrote notes too damn quickly :)08:03
Bostikhehe, good enough and personally I'm fine with that :)08:03
*** niqt has joined #mer08:14
*** niqt has quit IRC08:16
*** nj3ma has quit IRC08:16
*** tadzik has joined #mer08:25
*** alh has quit IRC08:28
*** araujo has joined #mer08:30
Stskeepslbt: OK, home:stskeeps:skipprep contains a rpm that has --skipprep functionality08:33
Stskeepsit's best combined with rpmbuild -bb08:33
Stskeepsand it skips %prep section08:33
Stskeepsthis is on ci obs fwiw08:33
*** Siosm has joined #mer08:38
lbtok08:39
Stskeepsi got it to build 'acl' this way by some osc/'build' hacker08:40
Stskeepsy08:40
* lbt idly wonders if we should permit user branches (for maintainers) in core git repos08:41
Stskeepsgerrit topic branches, maybe..08:41
lbtto ease collaboration08:42
lbtbut just an idle thought08:42
Stskeeps:nod:08:42
Stskeepsfwiw, submissions are reachable through core git repos08:43
lbtright now I'm doing the exciting job of security updates on our VMs - expect some reboots as kernels change :/08:48
lbtI won't touch anything central whilst you may be demoing08:48
Stskeepsok08:48
Stskeepsdemo was skipped, didn't manage to get to the state i wanted to be in time08:48
Stskeepsjust dont reboot CI right now if it's ok08:49
lbtsure08:49
lbtwhen you have time we'll chat about phost migrations - the 2 new ones are dual-core 8 thread, 32Gb RAM 6Tb08:50
Stskeepsok08:51
Stskeepsi have time now :P08:51
Stskeepsjust fixing some mesa-wayland stuff08:51
Stskeepsdo you have a list of VM-to-phost mappings:08:51
Stskeeps?08:51
lbtyes, run "vms" on zathras08:51
lbtgah pastie.org08:52
Stskeepsok08:53
lbthttps://gist.github.com/297760508:53
lbt.doesn't show roles like workers08:53
Stskeepsok, we can probably move crepo/csrc/cbe/cfe ?08:53
lbtI'm thinking to make phost2/4 vm hosts08:55
lbtso step 1 is 6/7 -> new workers for CI08:55
lbtclean up 2/4 to be plain phosts08:56
lbtmake VMs for CI on 2 or 408:56
lbtmove CI from monster to VMs08:56
lbtclean up and retire monster08:56
Stskeepsok08:57
lbtsay 2 for CI vms08:57
lbtwe can move the community stuff to 408:57
lbtand expand use of 5 for img08:57
lbthmm.... traffic-wise - maybe invert that08:57
lbtin amongst this there are some things that need ssl applying too08:58
lbtwe can also bring up a signer08:59
Stskeepsok09:00
*** rcg has quit IRC09:14
phaeronI read that singer09:15
lbtheh09:16
Stskeepsphaeron: sounds like a good appliance.. each time a build succeeds, it sings09:16
Stskeeps:P09:16
lbtalterego: you around ?09:16
*** rcg has joined #mer09:18
phaeronsings to the tune of your build system :D09:19
* lbt has lost the merproject DNS server...09:24
lbtI know it's around here somewhere09:24
Stskeeps..09:24
lbtit's on the internet - how hard can it be to find?09:25
Stskeepsnmap 0.0.0.0/0 ?09:25
Stskeeps:P09:25
lbtwaiting....09:26
lbtoh, I got IPv6 for phost1 aswell09:26
*** talavis has joined #mer09:26
lbtah, found it :)09:30
*** jonwil has joined #mer09:37
* Stskeeps waits for qtdeclarative to build..09:38
jonwilFinally found a repair place that can fix my N900 :)09:39
Stskeepswoo09:40
* phaeron tries to make sense of what obs is doing ..09:41
jonwilThey need to replace the whole screen assembly and its costing me $149 which is much cheaper than a new phone would be :)09:41
*** _8david` has joined #mer10:00
*** _8david has quit IRC10:02
w00tStskeeps: getting there, then10:07
Stskeepsugh10:08
Stskeepsfucking ugh10:08
w00t?10:09
Stskeepshttp://pastie.org/413702910:09
Stskeepsremember, we're on actual modern compilers..10:10
w00turrm10:11
w00tthat's intriguing10:11
Stskeepsi guess i should show it to sroedal/jlind..10:12
* Stskeeps peeks around10:12
w00twell10:14
w00tis it linking that to glib?10:14
Stskeepswell a .a file can't naturally be linked10:15
Stskeepsit should include a link to glib10:15
w00tno10:15
w00totherwise, platformsupport will be pulling in X too, and I really doubt you want that10:16
Stskeepswhat?10:16
Stskeeps:P10:16
Stskeepsglib doesn't link to x11 for me10:16
w00tno, but platformsupport contains some X stuff10:16
Stskeeps.. not sure what you're talking about10:17
*** gimli_ has quit IRC10:17
Stskeepsanyway, i solved it10:17
Stskeeps:P10:17
w00tplatformsupport links in loads of crap10:17
w00tyou do not want it to be linking directly against all the libs it uses10:17
w00totherwise you will end up with a massive dependency mess10:17
w00tthere was a thread about it a while ago, trying to find it10:18
Stskeepsyes, ok10:19
w00thttp://lists.qt-project.org/pipermail/development/2012-March/002500.html10:20
Stskeepsso, how does it make sure that it links to glib, which it uses?10:21
Stskeeps:P10:21
*** dionet has joined #mer10:22
Stskeeps@!$@ finally10:22
w00tI don't know10:22
* Stskeeps boots vm10:23
w00tthat thread never really had a proper resolution I think10:23
w00tbuilt?10:23
Stskeepsbuilt.10:23
w00tnice10:24
*** himamura has joined #mer10:24
*** Behold has quit IRC10:29
*** BeholdMyGlory has joined #mer10:30
phaeronStskeeps: in the copyprj scenario  we talked about, only project link is changed or also the repository path ?10:31
Stskeepsproject link, but in a later scenario we might also change repo path10:32
Stskeepsbut that's another story10:32
Stskeepsproject link only is first10:32
phaerontrying to word explanation10:33
Stskeepsw00t: food is ready downstairs10:37
*** arcean has joined #mer10:38
w00tooh10:39
*** kurtul has joined #mer10:40
* Stskeeps escapes from foreveralone corner10:41
* w00t is still in a network discussion10:42
lbtStskeeps: phaeron did either of you possibly update bz1 bugzilla version yesterday?10:43
Stskeepsi haven't touched anything10:44
lbtI didn't think so - just wondering10:45
lbtit may have automatically done a security update10:45
phaeronlbt: no10:46
lbt2012-06-23 06:31:27 upgrade bugzilla3 3.6.2.0-4.4 3.6.2.0-4.510:46
lbtI was asleep !10:46
Stskeepsheh10:46
lbtI am *so* glad I comitted all my changes to git10:46
dm8tbrlbt: so you're not sleep-walking but sleep-upgrading?10:46
Stskeepsthis is puppet?10:46
Stskeeps:P10:46
lbthmm .. could be10:47
lbtI'm going to shutdown bz for a bit10:53
*** BeholdMyGlory has quit IRC10:53
*** BeholdMyGlory has joined #mer10:54
phaerongoing to eat11:00
phaeronhmm I think I fixed it :D11:02
phaeronbut more testing later11:03
*** pohly has joined #mer11:11
*** tilgovi has quit IRC11:14
*** phaeron_n9 has joined #mer11:19
*** pohly has quit IRC11:21
*** wicket64 has joined #mer11:27
*** M4rtinK has joined #mer11:40
Macerwow11:45
Macerthe instructions to get mer or plasma on a tegra 2 are awful heh11:45
Stskeepswelcome to hardware adaptation, it's messy business11:46
Macerdoes anybody have any good instructions for installing on a tegra.. ie: for a tf101 with an unlocked bootloader?11:46
Maceris there something that is easily flashed that might work?11:46
MacerStskeeps: yeah :-/11:47
Macerhttp://share.basyskom.com/contour/Deployment/mer_arm_install_archive/point_of_view_nvidia_tegra2/readme.txt11:47
Macerthat is the only thing i can find that even remotely looks like soviet russia instructions about flashing11:47
Stskeepslooks less insane than other things i've seen..11:48
Stskeeps:p11:48
Maceri knwo there was a youtube video of someone running active on a tf101 but there are no good instructions on how to do it with mer/plasma11:48
Macerthat seems to be holding things back a bit.. people want to use it.. but can't install it because there are no instructions for their device ;)11:49
Stskeeps:nod;11:50
Maceri was able to get the nvflash going for my tf101 and even made a backup so my bootloader is unlocked... i just need to find something that can be flashed to it11:54
*** ALoGeNo has quit IRC11:54
Macerlike an easy tf101.img or something :)11:54
Maceris there nothing like that around?11:54
Macerif someone makes something ... i will write the instructions for you so people can actually use what you are trying to make11:55
Maceri'm sure a ton of people with older tf101s would love to get rid of android and use something a little more.... linux11:55
vgradeMacer, \o12:03
*** dionet has quit IRC12:03
Macerhey vgrade ... do you have an img for the tf101?12:04
vgradeMacer, nope, where did we get to12:06
vgradewe got you running nvflash12:07
Maceri got to the point where i was ripping the current partition table12:07
vgradei remember12:07
Macerand made the partition.txt12:07
vgradeso you have backup of the different partitions12:07
Macerbut after that i kind of gave up.. didn't really know where to go from there12:07
Maceryes12:07
Macerthat part was easy ;)12:07
vgradeon you go then12:08
Macerhttp://vetus.scientiam.org/apps/transformer/partitiontable.txt12:09
Macerthat was where i left off12:09
vgradeso you need to split your partition 612:10
vgradeinto kernel, commandline and ramdisk12:10
vgradehttp://android-dls.com/wiki/index.php?title=HOWTO:_Unpack%2C_Edit%2C_and_Re-Pack_Boot_Images12:11
Macerhttp://vetus.scientiam.org/apps/transformer/12:11
vgradethen repack them12:11
Macero.....k.12:11
vgradeand flash the boot.bin back to partition 612:11
vgradesee if it still boots anddriod12:11
Macermy transformer isn't rooted12:12
Macer(Note added by lxrose: These commands don't work if your phone is not rooted and the permissions are not set.)12:12
Macerso i guess i have to root my transformer?12:12
* Macer hopes that is easily done12:13
RaYmAnlearn nvflash :P12:13
vgradewhat commands? root in the andriod system?12:13
vgradeits not required for this hack12:13
RaYmAnroot is irrelevant when you have nvflash - you can just read and flash boot with that12:13
vgradeas long as you have an unloacked bl12:13
vgradeRaYmAn, \o12:14
vgradelooks like there is a new kernel around as well, http://forum.xda-developers.com/wiki/How_to_install_Ubuntu_on_an_Asus_EEEPad_Transformer12:14
RaYmAnwait, what? Tf101 doesn't have unlocked vs locked bls at all.12:14
MacerRaYmAn: that is what i thought12:15
MacerRaYmAn: i thought the older tf101s were unlocked12:16
RaYmAnno12:16
Macerbut later they were12:16
Maceroh... so they don't work then?12:16
RaYmAnit's more like, All TF101's are bootloader unlocked, but we managed to get the keys to the kingdom ;)12:17
Macerfrom what i understood an older B60 tf101 was able to be flashed but later models were not12:17
RaYmAnSo as far as you're concerned, all are unlocked.12:17
Maceroh... ok12:17
RaYmAnbut you can't really say some are unlocked or some aren't12:17
RaYmAnMacer: all tf101's have nvflash with our wheelie magic ;)! http://androidroot.mobi/2012/05/27/introducing-wheelie-nvflash-for-asus-transformer-tf101-b70/12:17
Macerwell.. at the time i thought the concensus was that B70+ were locked heh12:17
Macerah... ok.. well.. last i checked was before may ;)12:18
MacerRaYmAn: so is there a generic image i can flash then?12:18
RaYmAnyeah, what I'm mostly saying is that it's a misconception that some are unlocked. We just have the keys - doesn't make it unlocked :P12:18
Macerwhere it is just like.. nvflash mer.img12:18
Macerdone? :)12:18
Macerheh12:18
*** eebrah has joined #mer12:19
Maceroh wow... nice12:20
Macerok... i'm going to have to wake up and do the wheelie thing heh12:20
Macerbut i would really like to be able to flash mer/plasma to the thing12:20
Maceror mer and then install plasma onto it if that is at all possible12:20
Macerand make some instructions for the mere mortals12:21
Macerso people can actually USE mer/active12:21
Macerall of us prolatariats feel as though we have our noses pressed up against the windows watching the developer bourgeoisie wine and dine :)12:22
vgradeMacer, once you can upack, repack. Create a kernel with options a http://wiki.merproject.org/wiki/Adaptation_Guide12:24
*** eebrah has quit IRC12:25
vgradeand in your commandline add root=/dev/sdacard12:25
vgradewe can't do adaptations for all devices12:25
vgradebut we are here to help others to do bringups12:25
Macerah i see.. so it's just a matter of getting that first image going huh?12:26
vgrade?12:26
Macerwould my images be able to be used across all tf101s?12:26
Macerwhere someone can just use what i have made and skip the unpacking/repacking part?12:26
Macerand simply flash?12:26
vgradeyes12:26
vgradethe unpack repack is to make sure your toolset works12:27
Macerah ok. well... i'll read over that site in a moment and see what i can do12:27
vgradethere are lots of sites giving that information and the basyskom scripts are doing that12:27
*** jonnor has quit IRC12:28
Macerbut i haven't seen anything that is clear12:28
Macermaybe the unpacking/repacking part ... but there isn't somewhere where it is just "get this image" nvflash image.. done12:28
RaYmAnnvflash images aren't special. You just flash the boot.img created with --download :)12:29
vgradeMacer, if you want 'get this image, nvflash' then I suggest you get an Ipad12:32
Macerlol12:32
vgrade:)12:32
Stskeepsvgrade: what was the vendor of the SoC that vivaldi had, again?12:33
Macervgrade: but if it is a common img then it should b that easy ;)12:33
Macerbe12:33
vgradeStskeeps amlogic12:33
Stskeepsah thanks12:33
vgradeMacer, common in what way12:34
vgradecommon to tf's, tegra2 tabs, all andriod tabs12:35
vgradeeach needs to be different due to differences in hadrware so require different kernels12:36
Macerper model of device12:36
vgradewe did ud the same filesystem on the vega and PoV tabs, just had differnet boot.bins12:36
Stskeepsvgrade: got the mali user land libs from there?12:36
vgradeYes I have but only for a couple of versions of the API12:38
Stskeepsthat's ok, i just need one of them12:38
*** CosmoHill has joined #mer12:38
vgradeStskeeps, I can link later12:39
Stskeepsok12:39
CosmoHillwoof vgrade et Stskeeps12:41
*** dionet has joined #mer12:42
vgradeCosmoHill, |o12:42
Stskeepsvgrade: interesting, a10 egl libs are unstripped12:43
vgradeI've been working on mele a1000 and there's a guy on #arm-netbook who has been building the libs from his sdk12:45
Stskeepsok12:46
vgradehttp://rhombus-tech.net/allwinner_a10/Compile_X11_driver_for_A10/12:47
Stskeepscould you for fun ask him if there's a way to share hardware buffers in the GPU?12:47
Stskeeps.. or i could just read the page12:47
vgradeStskeeps, I'm not sure he's the guy to ask. Probably an ARM question12:48
Stskeepsok12:48
vgradeStskeeps, the sharing for wayland12:48
Stskeepsyes12:48
vgradewhat the Pi has12:49
vgradeis that a standard extension12:50
*** eebrah has joined #mer12:50
Stskeepsnot really12:50
Stskeepsbut i can do similar stuff on omap312:51
*** BeholdMyGlory has quit IRC12:55
*** BeholdMyGlory has joined #mer12:56
*** DocScrutinizer has quit IRC12:58
*** DocScrutinizer has joined #mer12:58
*** kurtul has quit IRC12:58
*** DocScrutinizer06 has joined #mer12:58
*** DocScrutinizer05 has quit IRC13:00
*** DocScrutinizer06 is now known as DocScrutinizer0513:00
*** shrikrishna has joined #mer13:03
*** tarantism has joined #mer13:21
*** dionet has quit IRC13:27
*** proyvind has quit IRC13:28
*** shrikrishna has quit IRC13:40
*** shrikrishna has joined #mer13:42
*** eebrah has quit IRC13:52
*** otep has quit IRC13:55
*** otep has joined #mer13:59
*** andre__ has joined #mer14:07
*** kurtul has joined #mer14:09
Stskeepsandre__: were you in .cz or berlin these days?14:09
andre__Stskeeps, I was in .cz until yesterday, and from tomorrow on again for a week14:10
Stskeepsalright14:10
Stskeepswe're a few people here due to qtcs so14:10
Stskeeps(berlin)14:11
andre__Ah, pity. I'm in Dresden right now :)14:11
Stskeepsah alright14:12
andre__I'm in Krakow in 9 days, but you're probably not around there :P14:12
Stskeepsyeah :P14:12
andre__maybe next year I will make it to aKademy, didn't work out this year :(14:13
Stskeepsi've been travelling 3 out of 4 weeks this month14:17
*** eebrah has joined #mer14:18
*** eebrah has quit IRC14:25
*** shrikrishna has quit IRC14:27
*** CosmoHill has quit IRC14:32
*** CosmoHill has joined #mer14:35
*** shrikrishna has joined #mer14:46
*** andre__ has quit IRC15:14
*** gabrbedd has joined #mer15:18
*** rcg has quit IRC15:19
lbtStskeeps: I've implemented rewriting in exim on our internal smarthost - this should ensure all outgoing mail appears to come from the merproject.org domain. It may affect your rules15:24
*** otep has quit IRC15:31
*** otep has joined #mer15:31
*** pomm has quit IRC15:34
*** M4rtinK has quit IRC15:37
*** CosmoHill has quit IRC15:39
*** dijenerate has quit IRC15:44
*** jonwil has quit IRC15:47
*** pomm has joined #mer15:50
*** CosmoHill has joined #mer15:53
CosmoHillhey lbt15:53
phaeronback15:54
lbthey CosmoHill15:55
lbthey phaeron15:55
*** tarantism has quit IRC15:55
*** kurtul has quit IRC15:57
CosmoHilllbt: looking at a vauxhall astra coupé turbo atm15:57
* CosmoHill looks / thinks a lot but doesn't act much15:58
*** arcean has quit IRC15:58
phaeronlbt: https://github.com/Merproject/open-build-service/commit/e679826856763fff65cd6d44541dfea49768de8e16:01
*** arcean has joined #mer16:01
lbtphaeron: makes sense16:03
phaeronI tested the obs only scenario and this fixes it. I'll test now with mds.16:05
phaeronbut it should work16:05
*** clopez has joined #mer16:09
*** gimli_ has joined #mer16:13
*** wicket64 has quit IRC16:17
*** Eismann has quit IRC16:35
*** Eismann has joined #mer16:36
*** talavis has quit IRC16:43
*** gabrbedd has quit IRC16:54
*** gimli_ has quit IRC16:58
*** tilgovi has joined #mer17:09
*** fecub has joined #mer17:11
*** pomm has quit IRC17:30
*** gimli_ has joined #mer17:37
*** clopez has quit IRC17:56
phaeronlbt: I think we can build packages of this and start testing the new mer process17:59
Stskeepsprjcopy process ?18:00
Stskeepsprogress18:00
phaeronyeah fixed the prjlink issue18:00
phaeronhttps://github.com/Merproject/open-build-service/commit/e679826856763fff65cd6d44541dfea49768de8e18:00
Stskeepscool18:01
Stskeepsso it does it right? ie, not rebuild everything18:02
phaeronyes rebuilt only one package whose srcmd5sum changed18:02
Stskeepsok, cool18:02
phaeronStskeeps: the other unrelated thing : I kludged mds to return something on  ls -b , but there is not information available about binary versions that can be returned18:04
Stskeepsok18:04
Stskeepswe have that data thoughl..18:05
phaeronhttp://pastie.org/413892718:06
phaeronso  not yet completely correct18:06
phaeronbut at least it returned  something now18:06
phaeron:d18:06
Stskeepsokay18:07
Stskeepsvery happy to see prjcopy work18:07
*** shrikrishna has quit IRC18:07
*** shrikrishna has joined #mer18:08
phaeronStskeeps: where is the data , ls obs-repos/Core\:i586\:latest/Core_i586/i586/ shows all name.rpm and _repository?view=names is only names  too18:08
Stskeepshmm, thought we generate that18:08
phaeron_repository?view=binaryversions  contains hdrmd5 and metamd518:08
Stskeepsok18:09
phaeron:)18:10
phaeronjust starting to get into mds18:10
Stskeepsi'll grab mds2 for you if you want..sec18:10
phaeronthere's a 2 ?18:10
Stskeepsyes18:11
Stskeepshttp://releases.merproject.org/~carsten/mds2.py.txt ,18:11
phaeronI also moved the serve_forever to separate thread so we can shutdown cleanly on ctrl-c18:11
Stskeeps http://releases.merproject.org/~carsten/gitmds2.py.txt18:12
Stskeepsthis uses the git stuff a lot more18:12
*** M4rtinK has joined #mer18:12
Stskeepsinstead of Core:i586 we now have Core:GITREF:SUBDIR18:12
Stskeepsnot complete, sadly, but you're welcome to take it under your wings18:13
Stskeepsthis means you can out of the blue do for example Core:v2.6.37:i58618:14
Stskeepswhich is where prjcopy+switch link comes in18:14
phaeronneed more explanation to work on it .. design document or something18:15
Stskeeps:nod:18:15
Stskeepssec18:15
*** phaeron_n9 has quit IRC18:15
Stskeepsmds2 = frontend, gitmds2.py = backend18:16
Stskeepsand design document.. really?18:16
Stskeeps:P18:16
phaeron:D18:16
phaeronat least something to explain what the hell this .py is supposed to do :)18:17
Stskeepsi've actually tried to document this one18:17
Stskeepsand add comments18:17
phaerondocstrings --> sphinx would be a good start18:17
Stskeepsheh18:18
phaeronhave you seen imager documentation ?18:18
Stskeepsyeah, i think18:18
phaeronhttp://autodoc.meego.com/mint/imager/18:19
phaeronI really like it , because I wrote it :D18:19
phaeronit even has screenshots now18:19
phaeronanyway not bragging... lets get prjcopy packaged and tested18:20
*** CosmoHill has quit IRC18:21
*** tilgovi has quit IRC18:24
*** beford has joined #mer18:30
*** CosmoHill has joined #mer18:34
*** shrikrishna has quit IRC18:35
*** dionet has joined #mer18:36
*** kurtul has joined #mer18:38
*** rcg has joined #mer18:42
*** RhymeswA has joined #mer18:51
phaeronStskeeps: implemented SIGUSR1 to list running threads for fun .. could help with the long running watcher threads19:19
*** arcean_ has joined #mer19:25
*** arcean has quit IRC19:25
Stskeepsok19:26
*** dionet has quit IRC19:33
Stskeepsi am starting to get significant suspicion that 'yum' in mer does not handle ~'s very well19:34
w00tagain?19:36
Stskeepsi went for "this can't possibly not work", something that was definately bigger19:36
Stskeepsand still boom19:36
lbt /usr/lib/python2.7/site-packages/yum/packageSack.py:from rpmUtils.miscutils import compareEVR19:45
lbtso yum uses the algorigthm there19:45
lbtuse: from rpmUtils.miscutils import stringToVersion, compareEVR19:46
lbtand you can do some verification19:46
lbtI ran into this in revs19:46
Stskeepshttp://rpm.org/ticket/5619:47
lbtI know19:47
lbthttp://wiki.meego.com/index.php?title=Packaging%2FGuidelines&action=historysubmit&diff=16825&oldid=872319:48
lbt25 Oct 2010 ....19:48
lbtI think I logged a bug that the obs ~ handling != MeeGo ~ handling19:49
Stskeepsfairly sure our ~ handling is buggy now19:50
*** Venemo_N9 has joined #mer19:50
*** nsuffys has joined #mer19:59
*** talavis has joined #mer20:01
Stskeepsman, so much technical debt..20:02
*** CosmoHill has quit IRC20:16
Venemo_N9Stskeeps, it happens to all of us20:21
lbtStskeeps: yep - that's why I'm so anal about getting the design understood early on20:21
Stskeepswhich, ~?20:22
*** fecub has quit IRC20:24
lbtmost things20:24
Stskeepsmm20:25
lbtisn't ~ defined in rpm but probably not reflected in rpmutils library20:25
Stskeepsit is there20:26
Stskeepsi traced it down and it has to be rpm being buggy..20:26
lbthow is multi ~ handled?20:27
Stskeepsnot as good as the 4.10.0 patch20:28
Stskeepsrpmutils == rpm-python20:28
lbtyes20:28
*** NIN101 has quit IRC20:28
Stskeepswhich == rpm rpmvercmp deep down20:29
lbtno20:29
Stskeepsyes20:29
lbtit doesn't call out to the C afaik20:29
Stskeepsgo look at labelCompare20:29
Stskeepsit traces nicely down to rpmvercmp in the end20:29
*** furikku has quit IRC20:29
lbthttp://lists.rpath.com/pipermail/mirrorball-commits/2010-March/000864.html20:33
lbtso I guess I was remembering the old approach20:33
Stskeepsyes, perhaps20:33
Stskeepsand that's not the rpmutils i have in sdk anyway20:34
lbtno, this was back doing REVS20:34
Macerugh. i've just ripped apart my plasma tv trying to track down which board is broken20:34
Macerit keeps losing the left half of the screen20:34
*** tilgovi has joined #mer20:35
lbtMacer: like this ?  http://imgur.com/a/Bcukp20:36
Macerno20:37
Macerit goes completely blank.. as if the left side shuts off20:37
lbtnp - some guy on a bbs I see has that "left half of screen" issue....20:37
Macerit flickers on an off as well20:37
Macersometimes it will last for hours... but sometimes it will last 5 mins and completely shut off the left side20:37
Maceri just pulled .. checked.. and reseated the ss board to see if maybe it was sending bad power to the signal board20:38
Macerbut it is almost impossible to find out which signal board is broken since there are 2 boards which control the signaling :-/20:39
Macerafter reseatingand reconnecting the ss board tho it seems to be kind of working.20:39
Macerso far so good. :) if i can fix it i was going to put this one in my son's room and i already ordered a new 46" lcd20:40
*** Venemo_N9 has quit IRC20:46
*** nsuffys has quit IRC20:57
*** kurtul has quit IRC20:58
Stskeepsw00t: qmlscene up and running on eglfs.. compositor tomorrow20:59
* Stskeeps will go crash now20:59
phaeronlbt: can you please update my email in ldap ? I tried from obs but it has no effect20:59
phaeronthanks20:59
lbtmer or meego ?21:00
lbtphaeron: and what to ?21:01
phaeronlbt: both if possible21:02
lbtgmail ?21:03
lbtdone meego21:03
phaeronyeah , in pm21:03
*** CosmoHill has joined #mer21:26
lbtadding rsync to SDK21:26
phaeronand ccache please :)21:30
phaeronlbt: ^^21:31
lbtok21:31
CosmoHillccache :)21:31
phaeronCosmoHill: you approve ? :)21:32
CosmoHillprobably21:32
CosmoHillnever used it myself21:32
CosmoHilldid give a distcc ago21:33
phaeronlbt: I tried finding the qmake cache that Stskeeps mentioned , but couldn't find much , only a blurb here : http://asoneofus.narod.ru/qt/qmake-guide.html#Details21:34
phaeronccache + distcc is quite good , used it in my LFS days21:34
lbtphaeron: I think that cache is more a config store21:36
CosmoHillLFS? :)21:36
CosmoHillyay LFS21:36
*** rdqfdx has quit IRC21:37
*** ALoGeNo has joined #mer21:37
*** ALoGeNo has joined #mer21:37
phaeronlbt: yeah  so I'll clarify from Stskeeps tomorrow21:39
CosmoHillphaeron: LFS is on this server and it's now nicer than ever21:39
phaeronthe maintenance overhead became too much21:40
phaeronanyway too tired now ..21:42
CosmoHillMrs Brown's Boys is funny21:43
CosmoHillthey leave the outtakes in the show21:43
* CosmoHill looks at Stskeeps and wonders how the hell the Mer IRC guidelines was opened on his browser21:53
alteregolbt: ah. sorry missed your messages, pm me your email and I'll send you the config file.22:04
lbtalterego: np - just seeing if you were around22:06
alteregoRight, I'll just grab it and email it now then22:09
lbtfyi I brought up the wiki with that old backup and a handcrafted one22:09
lbtthen got ldap working on it22:10
lbtso now you need to get an account via bugzilla before you can edit - should reduce spam a lot :)22:10
lbtside effect is that people may have had accounts with different usernames22:11
alteregoI have a mer bz account, have done since the beginning :P22:11
lbtoh yeah - I was meaning all users and especially wrt history22:12
alteregoYeah,22:13
alteregoCan we have both LDAP and normal?22:13
lbtwould probably defeat the point22:13
alteregoActually, yeah22:13
*** jstaniek has joined #mer22:14
alteregoHow about SSO? :)22:14
lbtwell, we have sso ... provided the first s is synchronised :)22:14
alteregoRight, sent :)22:15
alteregoHeh22:15
*** NIN101 has joined #mer22:21
*** Siosm has quit IRC22:35
*** Alison_Chaiken has joined #mer22:43
*** gimli_ has quit IRC22:48
lbt'night all22:48
CosmoHillnight lbt22:50
*** DocScrutinizer05 is now known as DocScrutinizing22:51
*** DocScrutinizing is now known as DocScrutinizer0522:53
*** DocScrutinizer05 is now known as DocScrutinizing22:58
*** DocScrutinizing is now known as DocScrutinizer0522:58
*** CosmoHill has quit IRC23:04
*** shrikrishna has joined #mer23:34
*** pomm has joined #mer23:46
*** NIN101 has quit IRC23:52

Generated by irclog2html.py 2.17.1 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!