**** BEGIN LOGGING AT Tue May 14 02:59:57 2019
May 14 07:03:00 <Herrie> Morning!
May 14 07:03:33 <Tofe> Morning!
May 14 07:06:14 <Tofe> Herrie: I didn't test it yet, but we need to add "com.palm.service.accounts" to the outbounds of "com.palm.tempdb" in db8 api-permissions
May 14 07:06:54 <Tofe> mmmh wait, that's weird
May 14 07:07:08 <Tofe> why would tempdb need to know how it contacts
May 14 07:07:51 <Tofe> damn, I'll need to investigate a bit more
May 14 07:12:27 <Herrie> Tofe: Yeah some of these are a bit weird
May 14 07:13:37 <Herrie> ka6sox: ping
May 14 07:20:40 <Herrie> Tofe: So some SRCREV bumps
May 14 07:20:41 <Herrie> https://github.com/webOS-ports/meta-webos-ports/pull/350
May 14 07:20:58 <Herrie> I think we need to check sleepd at some point. Morphis did quite some commits in the past which we don't have in the OSE code now
May 14 07:22:23 <Herrie> Especially to do with display state & suspend handling it seems
May 14 07:23:04 <Tofe> ok yes
May 14 07:24:44 <Herrie> I noticed it was pointing to my fork in GitHub so added a WebOS-Ports/webOS-OSE branch to sleepd in webos-ports and pulled in everything from OSE and added my systemd commit
May 14 07:33:46 <Tofe> https://github.com/webOS-ports/luna-next/pull/142 little "cleanup"
May 14 07:42:26 <Herrie|Laptop> Tofe: Merged & bumped
May 14 07:42:32 <Herrie|Laptop> Rebooted my builder and it doesn't come up :S
May 14 07:42:57 <Herrie|Laptop> There goes my ZNC :P
May 14 07:45:27 <Tofe> thanks
May 14 08:15:58 <Herrie|Laptop> Well let's put Jenkins to work then :P
May 14 08:21:53 <Tofe> exactly!
May 14 08:24:23 <JaMa> hehe poor jenkins just finished :)
May 14 08:24:38 <Herrie|Laptop> JaMa: Yes LOL
May 14 08:25:02 <Herrie|Laptop> JaMa: The Qt host contamination was there for qemux86 BTW
May 14 08:25:11 <JaMa> yes, I'm removin the sstate for it now
May 14 08:25:28 <JaMa> also the sstate for old arch
May 14 08:25:44 <JaMa> already updated jenkins-scripts to cleanup -halium archs
May 14 08:29:01 <Herrie|Laptop> Seems we have a power outage at home :P
May 14 08:29:09 <Herrie|Laptop> That might explain why my builder doesn't come up :P
May 14 08:29:26 <Tofe> it doesn't have its own pocket nuclear battery? damn
May 14 08:29:54 <Herrie|Laptop> Tofe: Well I would need to make the whole house run on a UPS or something... Nah ;)
May 14 08:31:08 <JaMa> find sstate-cache/ -name \*:aarch64:\* -delete; find sstate-cache/ -name \*:cortexa8t2hf-neon:\* -delete; find sstate-cache/ -name \*:cortexa8hf-neon:\* -delete; find sstate-cache/ -name \*:qtquickcontrols:i586-webos-linux:\* -delete
May 14 08:31:15 <JaMa> done on jenkins and milla
May 14 08:31:50 <JaMa> remaining archs:
May 14 08:31:53 <JaMa> find sstate-cache/ ! -type d  | sed 's/.*sstate:[^:]*://g' | sed 's/-webos-linux.*$//g' | grep -v ^: | grep -v ^x86_64-linux: | sort -u  | xargs  | sed 's/ /,/g'
May 14 08:31:57 <JaMa> aarch64-cortexa53,aarch64-halium,all,core2-64,cortexa7hf-neon-vfpv4,cortexa7t2hf-neon-vfpv4,cortexa8hf-neon-halium,cortexa8t2hf-neon-halium,hammerhead,i586,mako,mido,onyx,qemux86,qemux86_64,raspberrypi2,raspberrypi3,raspberrypi3_64,rosy,tenderloin,tissot,x86_64
May 14 08:35:43 <Tofe>  JaMa: do you really gain substantial time in selecting only these sstate ? looks like 95% will be rebuilt anyway
May 14 08:42:52 <JaMa> why should it rebuilt 95%?
May 14 08:44:26 <Tofe> didn't you remove all state for arm32 and aarch64 ?
May 14 08:44:51 <JaMa> not all, just the one without -halium
May 14 08:45:03 <Tofe> ah ok, yes, I missed that part
May 14 08:45:42 <JaMa> looks like something was changed in ls2, so it will rebuild quite a lot, but not because of me :)
May 14 08:46:12 <JaMa> and the duplicated sstate will be cleaned by normal sstate-cleanup job, this manual cleanup was needed only for the unused arch (until we have hammerhead mainline)
May 14 08:46:52 <JaMa> in theory we should also drop the .ipk files with old arch
May 14 08:47:32 <Tofe> it *might* be me :)
May 14 08:48:23 <JaMa> Herrie bumped the SRCREV, we can blame him :)
May 14 08:49:04 <JaMa> but progress is good, better than jenkins sitting idle :)
May 14 08:49:34 <Tofe> yep
May 14 08:50:23 <Herrie|Laptop> JaMa: Well it was needed, Tofe added some more useful logging to LS2 ;)
May 14 09:21:28 <novaldex> Herrie|Laptop: ping, saw your ping to ka6sox, i'll give him a reminder later when I hear from him, but anything I can answer?
May 14 09:22:05 <Herrie|Laptop> novaldex: Well issues, but that's with him. Just we'd need to update the channel topic here as well at some point, and I think he's the only one who can do this
May 14 09:22:53 <novaldex> yes I think you're right, he is the only one that can
May 14 09:23:09 <novaldex> I think if you send him the line as you'd like it to read he'll take care of it
May 14 09:29:09 <Herrie|Laptop> OK
May 14 12:39:16 <Herrie|Laptop> Tofe: This should bring us up to date with OSE in general. https://github.com/webOS-ports/meta-webos-ports/pull/351/files
May 14 12:39:44 <Herrie|Laptop> And this one might actually be useful: https://github.com/webosose/nodejs-module-webos-sysbus/pull/1
May 14 12:42:20 <Tofe> It's ready for merge?
May 14 12:42:21 <JaMa> Herrie|Laptop: btw: instead of setting ;branch=something after ${WEBOS_PORTS_GIT_REPO_COMPLETE} you should just set WEBOS_GIT_PARAM_BRANCH = "something"
May 14 12:43:00 <Herrie|Laptop> Tofe: Cannot build test, but yes
May 14 12:43:05 <Herrie|Laptop> JaMa: Will do that for future ;)
May 14 12:43:09 <Herrie|Laptop> Learning every day :)
May 14 12:43:25 <JaMa> the PR adds few more cases
May 14 12:44:48 <Herrie|Laptop> I can have a look see how many we have and clean them all up :P ?
May 14 12:44:54 <Herrie|Laptop> Anyway on a roll today LOL
May 14 12:44:59 <JaMa> sure
May 14 12:51:03 <Herrie|Laptop> JaMa: this only works for our webos-ports repos and OSE repos right? Not for gitlab etc
May 14 13:01:47 <JaMa> correct, it's part of WEBOS_PORTS_GIT_REPO_COMPLETE (and other _COMPLETE) variables
May 14 13:02:08 <JaMa> if you append ;branch=something then there are 2 branch parameters
May 14 13:09:44 <JaMa> e.g. bitbake -e nodejs-module-webos-sysbus | grep ^SRC_URI=
May 14 13:09:45 <JaMa> SRC_URI="git://github.com/webOS-ports/nodejs-module-webos-sysbus;branch=webOS-ports/master;branch=webosose"
May 14 13:22:42 <Herrie|Laptop> JaMa: all done
May 14 13:23:37 <Herrie|Laptop> 30 recipes updated ;)
May 14 13:25:12 <Tofe> nice !
May 14 13:26:40 <Tofe> you renamed all these branches already?
May 14 13:31:52 <Herrie|Laptop> Tofe: Well I basically cloned the webosose ones into webOS-ports/webOS-OSE and updated recipes
May 14 13:32:12 <Herrie|Laptop> And deleted the webosose ones (since anyway these were only used in recent builds, not in any releases)
May 14 13:37:22 <Tofe> ok, let's merge then
May 14 13:38:12 <Herrie|Laptop> Not build tested, but Jenkins will do that soon enough for us :P
May 14 13:38:17 <Herrie|Laptop> Mostly should be OK though
May 14 13:38:26 <Herrie|Laptop> Otherwise will fix it later today ;)
May 14 13:38:34 <Tofe> let's see
May 14 13:49:45 <Herrie|Laptop> Hammerhead is almost done, so we'll see soon enough with the next build ;)
May 14 13:59:06 <Herrie|Laptop> Tofe: you seen https://twitter.com/kdecommunity/status/1126822683059027969 ?
May 14 13:59:17 <Herrie|Laptop> Seems that they don't have much graphics glitches there?
May 14 14:00:52 <Tofe> Less than us, yes; but it seems related to transparency somehow
May 14 14:02:03 <Tofe> and we use more complex graphics also; kde uses a lot of flat design nowadays
May 14 14:02:30 <Tofe> but maybe there's something else
May 14 14:03:54 <Herrie|Laptop> Ah ok
May 14 14:04:17 <Herrie|Laptop> The build failed for Mako, pushed a quick fix
May 14 14:05:00 <Tofe> bshah: is plasma mobile based on QtWayland ? It's still kwin, right?
May 14 14:06:36 <PureTryOut[m]> Plasma Mobile uses KWin as the compositor yes
May 14 14:07:12 <Tofe> I don't think it's using QtWayland, but I'm not really sure
May 14 14:11:12 <Tofe> seems to be pure Wayland implementation yes, looking at kwin's code
May 14 14:14:59 <Herrie|Laptop> Tofe: Seems that qemux86 is behaving better, so quick fix worked :P
May 14 14:17:04 <Tofe> good!
May 14 14:18:01 <Herrie|Laptop> It will need to do a pretty much full rebuild due to ls2 anyway but well ;)
May 14 14:19:21 <JaMa> I have good reproducer for pseudo issue, so I was bisecting the qtbase changes to find out where it got broken and now discovered that last 10+ iterations were silently failing with /bin/sh: 1: /jenkins/mjansa/build-nodistro-master/BUILD/work/i586-oe-linux/pseudo-test-qmake/1.0-r0/recipe-sysroot-native/usr/bin/qmake: not found :/
May 14 14:20:05 <Herrie|Laptop> Ah not finding qmake is not good ;)
May 14 14:22:03 <JaMa> not good, but when qmake doesn't install anything, nothing gets broken in pseudo db
May 14 14:22:12 <JaMa> so the package_qa wasn't failing
May 14 14:25:20 <JaMa> this is also a bit annoying :)
May 14 14:25:23 <JaMa> docker-shr qemux86-64@ ~/build/oe-core $ dash -c 'echo -e "R22:\n\tfoo"'
May 14 14:25:23 <JaMa> -e R22: foo
May 14 14:25:23 <JaMa> docker-shr qemux86-64@ ~/build/oe-core $ bash -c 'echo -e "R22:\n\tfoo"'
May 14 14:25:23 <JaMa> R22: foo
May 14 14:26:36 <JaMa> luckily there is /bin/echo
May 14 15:27:08 <Herrie|Laptop> JaMa: I pushed 2 fixes to Warrior that I ran into while building on Jenkins so far
May 14 15:27:27 <Herrie|Laptop> Might be more to come, but looks like that might be it
May 14 15:27:45 <Herrie|Laptop> Couldn't build test locally due to builder being down after power outage and will only get to it in about 2 hrs time ;)
May 14 16:23:21 <Herrie|Laptop> Tofe: https://github.com/Herrie82/qt5-qpa-hwcomposer-plugin/commits/herrie/upstream-LuneOS-changes
May 14 16:23:30 <Herrie|Laptop> Need to build test still, then plan to PR upstream
May 14 16:43:09 <Tofe> Herrie|Laptop: looks good
May 14 16:43:30 <Herrie|Laptop> Tofe: Yeah we kept accumulating patches, so decided to spend some time to rework it so we could upstream it ;)
May 14 16:43:37 <Herrie|Laptop> Just need to build test when home
May 14 16:43:44 <Herrie|Laptop> But should be pretty straight forward in general
May 14 17:58:10 <Herrie> Tofe: We use the QPA only on actual devices right, not on Qemux86?
May 14 17:58:50 <Tofe> right, it's a hwcomposer qpa here
May 14 17:59:17 <Herrie> Would build test be enough, or I should test it on device as well?
May 14 18:01:00 <Tofe> for that commit, a build test is enough I thnk
May 14 18:01:35 <Tofe> https://photos.app.goo.gl/17M6rTEbqjUXd9HVA some little bit nicer video that doesn't show too much how many artefacts we get :p
May 14 18:02:26 <Herrie> Well I can charge my Hammerhead and test there.. .It's an all or nothing usually right?
May 14 18:02:27 <Tofe> (I'm rebuilding the whole pinephone image, so it'll take a bit to debug that -- and I'm not even sure where to begin with)
May 14 18:02:49 <Tofe> Herrie: right
May 14 18:03:29 <Herrie> That looks a lot better no? Or just showing differnt things?
May 14 18:03:48 <Tofe> Herrie: just showing what works, instead of focusing on what doesn't :p
May 14 18:04:24 <Tofe> some guy wanted to do an enthousiast blog post about pinephone progress, so that'll fit better
May 14 18:05:26 <Tofe> I'll try to put that on youtube, somehow there's no obvious sharing option between google photos and youtube... go figure
May 14 18:08:11 <Tofe> ah yes ok, I've got to do it from youtube. ok.
May 14 18:11:11 <Tofe> https://www.youtube.com/watch?v=hIok2W5fyk0 hop
May 14 18:19:26 <Herrie> :)
May 14 18:19:53 <Herrie> Tofe: What am I doing wrong here? qeglfscontext.h:45:35: error: missing binary operator before token "("
May 14 18:19:54 <Herrie>  #if QT_VERSION >= QT_VERSION_CHECK(5, 8, 0)
May 14 18:21:10 <Tofe> Missing #include ? let me check
May 14 18:21:59 <Tofe> try #include <QtGlobal>
May 14 18:48:48 <Herrie> That improves it a little, have another mystery now
May 14 18:49:43 <Herrie> qeglfscontext.cpp: In member function 'virtual void* QEglFSContext::eglSurfaceForPlatformSurface(QPlatformSurface*)':  qeglfscontext.cpp:103:28: error: 'QEGLPbuffer' does not name a type; did you mean 'wl_buffer'?
May 14 18:49:51 <Herrie> Will put kids to bed, can focus better then ;P
May 14 18:56:22 <Tofe> Probably also a missing include for the QEGLPbuffer type, but not sure here, without the source code
May 14 19:41:03 <Herrie> Wrong copy & paste it seems :P
May 14 19:42:56 <Herrie> But you were right
May 14 19:50:26 <Herrie|Laptop> Copied the wrong include :P
May 14 19:50:31 <Herrie|Laptop> So that's why it didn't work
May 14 19:50:33 <Herrie|Laptop> Now compiling
May 14 19:50:37 <Herrie|Laptop> Building Hammerhead now
May 14 19:52:12 <Herrie|Laptop> Tofe: Found the cause of ADB?
May 14 21:11:43 <Herrie> Tofe: Can confirm that ADB also doens't work on Hammerhead
May 14 21:21:34 <Tofe> the adb issue is caused by a missing "/var/usb-debugging-enabled" file
May 14 21:22:04 <Tofe> probably the consequence of my tinkering with the firstuse flashing...
**** BEGIN LOGGING AT Wed May 15 01:32:39 2019
**** BEGIN LOGGING AT Wed May 15 01:42:14 2019
**** BEGIN LOGGING AT Wed May 15 01:52:50 2019
**** BEGIN LOGGING AT Wed May 15 02:02:22 2019
**** ENDING LOGGING AT Wed May 15 02:59:57 2019