**** BEGIN LOGGING AT Thu Apr 25 03:00:02 2019
Apr 25 05:40:27 <Tofe> Morning!
Apr 25 06:18:22 <Herrie|Laptop> Morning!
Apr 25 06:18:31 <Herrie|Laptop> Tofe: Flashing Tenderloin now
Apr 25 06:18:40 <Herrie|Laptop> Hammerhead afterwards
Apr 25 06:26:46 <Tofe> ok, so it built ! that was my main concern :p
Apr 25 06:27:25 <Herrie|Laptop> Well it failed at first run, then 2nd run it worked
Apr 25 06:27:41 <Herrie|Laptop> Without any changes
Apr 25 06:31:59 <JaMa> morning
Apr 25 06:32:44 <Herrie|Laptop> JaMa: Morning
Apr 25 06:32:50 <Herrie|Laptop> Seems we're almost good to merge 5.12.3 :)
Apr 25 06:33:12 <Herrie|Laptop> Doing some testing on Tenderloin & Hammerhead now
Apr 25 06:35:32 <Herrie|Laptop> Tofe: Sound works on my TP (4G) in FirstUse and after FirstUse
Apr 25 06:35:34 <Herrie|Laptop> VKB not
Apr 25 06:38:02 <Tofe> vkb not working ? that's weird... could it be there is a bug which doesn't always happen
Apr 25 06:38:20 <Tofe> Your sound status is a relief :)
Apr 25 06:39:01 <Herrie|Laptop> Tofe: Still the Chrome Sandbox issue in Testr as well
Apr 25 06:39:07 <Herrie|Laptop> For the rest seems pretty OK
Apr 25 06:39:13 <Tofe> JaMa: I'll propose the PRs soon
Apr 25 06:39:15 <Herrie|Laptop> Let me check logs
Apr 25 06:41:48 <Herrie|Laptop> Tofe: maliit looks OK actually, no crashes or anything
Apr 25 06:41:55 <Herrie|Laptop> Just this warning: "Apr 25 08:39:31 tenderloin maliit-server[1109]: shared memfd open() failed: Function not implemented"
Apr 25 06:41:58 <Tofe> https://github.com/webOS-ports/meta-webos-ports/pull/338 https://github.com/shr-distribution/meta-smartphone/pull/102
Apr 25 06:42:04 <Herrie|Laptop> Which I believe was always there
Apr 25 06:42:16 <Tofe> yes, it's not new
Apr 25 06:42:29 <Tofe> what if you just restart maliit ?
Apr 25 06:43:12 <Tofe> It could be I still have an issue if the client window is created too soon or something like this
Apr 25 06:45:08 <Herrie|Laptop> Yup "systemctl restart maliit-server" seems to bring it up
Apr 25 06:45:16 <Herrie|Laptop> Also power menu works properly again
Apr 25 06:45:20 <Herrie|Laptop> Dashboards too
Apr 25 06:45:25 <Herrie|Laptop> So your fix there worked :)
Apr 25 06:46:50 <Tofe> there's still something fishy with the vkb I think
Apr 25 06:46:59 <Tofe> but most of the things are back yes
Apr 25 06:47:44 <Herrie|Laptop> The SDL2 OpenGLES test apps don't work as well I noticed
Apr 25 06:48:17 <Herrie|Laptop> Probably something small: https://bpaste.net/show/e0d38a3f0c38
Apr 25 06:48:46 <Herrie|Laptop> And also saw this one: Apr 25 08:51:49 tenderloin luna-next[1049]: WARNING: 08:51:49.146: file:///usr/palm/luna-next/shells/card/qml/Notifications/AlertWindowsArea.qml:70: TypeError: Type error
Apr 25 06:49:41 <Herrie|Laptop> Last one seems to point to https://github.com/webOS-ports/luna-next-cardshell/blob/master/qml/Notifications/AlertWindowsArea.qml#L70
Apr 25 06:56:23 <Herrie|Laptop> Tofe: I guess it might be good to exclude the /media/internal/.cache from the mediaindexer since it gives a lot of /internal/.cache/luna-next/qmlcache/f76957efe767a18d2f72b8e35544c2fd9d364842.qmlc.AIfAMk failed: Error when getting information for file “/media/internal/.cache/luna-next/qmlcache/f76957efe767a18d2f72b8e35544c2fd9d364842.qmlc.AIfAMk”: No such file or directory
Apr 25 06:56:28 <Herrie|Laptop> Which is not very useful
Apr 25 06:56:47 <Herrie|Laptop> Seems we could simply add it here: https://github.com/webOS-ports/mediaindexer/blob/6a03e931dfced4a02f96e82f3b134035b0ee9f78/src/MediaScannerServiceApp.cc#L36
Apr 25 06:56:57 <JaMa> Tofe: Herrie|Laptop: great :)
Apr 25 06:57:39 <JaMa> Tofe: PR for the remaining component change? I can take care of the merge from master to warrior
Apr 25 07:04:10 <Tofe> JaMa: well it's in one bundle currently, so if you merge from master first, I can rebase and redo the PR
Apr 25 07:07:55 <JaMa> I'll wait for it to be merged to master, then move all to warrior (together with small meta-qt5 update in webos-ports-setup)
Apr 25 07:17:17 <Tofe> JaMa: ah ok, my PR goes to warrior currently, I understand your sentence now
Apr 25 07:20:45 <Tofe> JaMa: https://github.com/webOS-ports/meta-webos-ports/pull/339 here your go
Apr 25 07:37:28 <Herrie|Laptop> Tofe: Hammerhead seems similar to Tenderloin
Apr 25 07:38:05 <Tofe> ok, so we have a lunatic vkb, some warnings, but the rest seems ok
Apr 25 07:41:37 <Herrie|Laptop> Tofe: Yup
Apr 25 07:57:55 <JaMa> Tofe: "initramfs-boot-android: split halium and generic version" isn't needed for Qt, right? But I guess it doesn't harm (one less change in mainline branch :))
Apr 25 08:01:24 <Tofe> yes, it's separate, but I included it because we'll need it soon enough -- and it's a bit cleaner anyway
Apr 25 08:02:22 <Tofe> I'll need to do a good rebase of sumo/mainline onto latest warrior, at some point
Apr 25 08:06:11 <JaMa> ok
Apr 25 08:15:01 <JaMa> all merged now
Apr 25 08:15:16 <JaMa> unstable builds now track Zeus release
Apr 25 08:16:12 <JaMa> not that many commits which weren't in warrior: https://github.com/webOS-ports/webos-ports-setup/commit/3f6ba8a25836f540886dec56db17bced7e040600
Apr 25 08:17:49 <JaMa> and https://github.com/webOS-ports/webos-ports-setup/commit/9229f7fc4f70656973eb3d9683a6caa39a68f291 which will simplify our connman bbappend a bit (which will currently be the only difference between meta-webos-ports warrior and master
Apr 25 08:18:42 <Tofe> Great, thanks, we're back to a simple situation for the near future :)
Apr 25 08:21:35 <JaMa> my version of rebased mainline changes is in jansa/warrior, but I haven't tested it recently
Apr 25 09:37:41 <Tofe> JaMa: I've also updated https://github.com/shr-distribution/meta-smartphone/pull/103 to target master
Apr 25 09:41:33 <JaMa> let me compare with my version
Apr 25 09:41:44 <JaMa> do you want it in warrior as well? I assume
Apr 25 09:41:59 <Tofe> yes, to match meta-webos-ports
Apr 25 09:42:13 <JaMa> I see the comment now :)
Apr 25 09:43:14 <Tofe> I've notified #asteroid that this might break their build, as they have bbappend on the old initrd scripts recipe
Apr 25 09:43:38 <Tofe> but it's just a rename of the bbappend to do for them
Apr 25 09:44:51 <JaMa> should I update the PR with the fixes from jansa/warrior or do you want them as review comments?
Apr 25 09:46:36 <Tofe> I didn't look at it yet
Apr 25 09:47:54 <Tofe> ah but it's the complete mainline -- no let's wait for these
Apr 25 09:52:33 <JaMa> but there were some small fixes for whitespaces and VIRTUAL-RUNTIME variables, e.g. "android-kernel-bootimg: append DTB when specified" also seems to be applicable even without mainline
Apr 25 09:59:16 <JaMa> Tofe: PR updated with some of the changes
Apr 25 11:51:35 <Tofe> thanks!
Apr 25 12:04:34 <Herrie|Laptop> JaMa: Will you kick off new testing builds once everything merged?
Apr 25 12:30:08 <JaMa> yes, on it now
Apr 25 12:30:34 <JaMa> local build finished already so it should be all in
Apr 25 12:44:42 <Herrie|Laptop> JaMa: Great!
Apr 25 13:18:14 <Herrie|Laptop> JaMa: SRCREV bump for mediaindexer you want me to PR or do directly in Warrior and you'll cherry pick to master?
Apr 25 14:07:28 <JaMa> Herrie: yes please
Apr 25 14:07:47 <JaMa> Tofe: Fetcher failure for URL: 'file://Fix-userdata-mount-options.patch looks like we missed something (or my rebase went wrong)
Apr 25 14:13:41 <Tofe> I moved back that patch from meta-webos-ports to meta-smartphones
Apr 25 14:15:47 <JaMa> hmm maybe I just didn't have the latest branch in the build, let me check
Apr 25 14:16:05 <Tofe> mmmh "meta-hp/recipes-core/initrdscripts/initramfs-scripts-halium/initramfs-boot-halium/Fix-userdata-mount-options.patch" the path doesn't look right...
Apr 25 14:16:58 <Tofe> the right path should be meta-hp/recipes-core/initrdscripts/initramfs-scripts-halium/Fix-userdata-mount-options.patch I think
Apr 25 14:17:28 <Tofe> JaMa: I do the fix, or you do ?
Apr 25 14:17:44 <JaMa> I'll
Apr 25 14:18:06 <Tofe> It's strange Herrie's build did succeed
Apr 25 14:18:20 <Tofe> anyway.
Apr 25 14:22:45 <JaMa> his builder is so fast that maybe it traveled to future and too my fix as well
Apr 25 14:26:14 <JaMa> Applying patch Fix-userdata-mount-options.patch
Apr 25 14:26:14 <JaMa> patching file scripts/halium
Apr 25 14:26:14 <JaMa> Hunk #1 succeeded at 447 with fuzz 2 (offset 42 lines).
Apr 25 14:26:22 <JaMa> will update it as well while moving it
Apr 25 14:28:05 <Herrie|Phone> I actually had it as well
Apr 25 14:28:15 <Herrie|Phone> Thought it was my Franken setup
Apr 25 14:28:31 <Herrie|Phone> So yeah should be fixed
Apr 25 14:32:38 <Tofe> Herrie|Phone: I likes JaMa explanation better :)
Apr 25 14:33:10 <Tofe> s/s/d/
Apr 25 14:33:18 <Herrie|Phone> Yeah that would be nice. AI and self-learning bitbake :P
Apr 25 14:52:19 <JaMa> https://github.com/shr-distribution/meta-smartphone/commit/527dfc4c3e284acb3eecad2233aa866e357ab134
Apr 25 15:33:17 <Herrie|Laptop> Tofe: I suspect that the VKB issue might have to do with Apr 25 08:39:33 tenderloin maliit-server[1109]: xkbcommon: ERROR: couldn't find a Compose file for locale "C.UTF-8"
Apr 25 15:33:31 <Herrie|Laptop> Which I don't recall seeing before
Apr 25 15:35:09 <Tofe> JaMa: thanks, looks better
Apr 25 15:35:25 <Tofe> Herrie|Laptop: I think I saw that one before
Apr 25 15:42:38 <Herrie|Laptop> Tofe: I've seen it in recent builds as well, but not in older I think
Apr 25 15:44:10 <Herrie|Laptop> Let me try last release to make sure
Apr 25 15:44:27 <Tofe> I'll also refresh my tissot image
Apr 25 15:58:22 <Tofe> mmh I think I introduced a mistake when I did https://github.com/webOS-ports/meta-webos-ports/commit/6cc8575742dee7ad79ba566f636934d7c9d26469
Apr 25 15:58:40 <Tofe> because now, when we reflash, we don't erase this folder... or do we?
Apr 25 16:01:44 <Herrie|Laptop> It was also there but slightly different: "May 08 23:59:35 hammerhead LunaAppManager[2265]: xkbcommon: ERROR: couldn't find a Compose file for locale "C""
Apr 25 16:03:18 <Tofe> Herrie|Laptop: do you agree we should remove luneos-data when deploying a luneos image on device? or am I missing something here?
Apr 25 16:09:12 <Tofe> proposal: https://github.com/webOS-ports/android-update-package/pull/6
Apr 25 16:42:09 <Herrie|Phone> Tofe: Not the whole luneos-data, but the single file/folder like in your PR is fine
Apr 25 16:42:38 <Tofe> yes, I revised my first implementation, it was too brutal :)
Apr 25 16:45:44 <Herrie|Phone> Yes because I think luneos-data on tenderloin is shared with legacy webOS /media/internal
Apr 25 16:46:11 <Herrie|Phone> Which would then delete all stuff on there
Apr 25 16:46:23 <Herrie|Phone> Strange that this never was an issue before though
Apr 25 16:46:50 <Herrie|Phone> Or is it due to all the recent Halium related changes it popped up?
Apr 25 16:47:00 <Tofe> no, it the commit I pointed to
Apr 25 16:47:05 <Tofe> it's*
Apr 25 16:47:40 <Tofe> we put the .firstboot_done flag in the rootfs folder, which is supposed to be as read-only as possible
Apr 25 16:48:03 <Tofe> but, it does get erased when we reflash, so the flag disappears.
Apr 25 16:56:13 <Herrie|Phone> OK well isn't that also necessary so we get first use on 1st boot?
Apr 25 16:56:22 <Herrie|Phone> Or am I not understanding the issue?
Apr 25 16:58:06 <Tofe> firstuse and firstboot have their own flag
Apr 25 16:58:36 <Tofe> but the firstuse flag is stored under /var, which is cleaned up when we see that we are doing a first boot
Apr 25 16:59:08 <Tofe> so in the end, we just need to get rid of .firstboot_done upon every reflash, one way or another
Apr 25 17:05:11 <Herrie|Phone> Tofe: Ah clear now
Apr 25 17:05:14 <Herrie|Phone> LGTM then
Apr 25 17:24:42 <Herrie|Phone> Merged
Apr 25 17:24:47 <Tofe> thx
Apr 25 17:24:55 <Herrie|Phone> What does this solve?
Apr 25 17:25:12 <Herrie|Phone> I mean any specific issue we saw?
Apr 25 17:25:21 <Herrie|Phone> Or just housekeeping?
Apr 25 17:29:48 <Tofe> It solves a future issue :)
Apr 25 17:30:10 <Tofe> when reflashing, firstboot flag wouldn't have been cleaned up, so no firstuse
Apr 25 17:30:59 <Herrie|Phone> I wonder why we didn't see this before
Apr 25 17:31:13 <Tofe> because I introduced the issue only very recently
Apr 25 17:34:54 <Herrie|Phone> Ah OK I didn't see link to previous commit
Apr 25 17:34:59 <Herrie|Phone> Now I understand
Apr 25 17:48:16 <Herrie|Phone> Tofe: I can confirm the sandbox issue in Testr was there in Doppio release too already
Apr 25 17:48:34 <Herrie|Phone> So it's not a recent qtwebengine upgrade thing
Apr 25 17:48:38 <Herrie|Phone> At least not 5.12
Apr 25 17:49:42 <Herrie|Phone> So probably 5.9.x to 5.11 upgrade regression
Apr 25 18:10:40 <JaMa> first 7 MACHINEs built on jenkins already, now will do emulator
Apr 25 18:19:01 <Herrie|Phone> JaMa: Nice!
**** ENDING LOGGING AT Fri Apr 26 02:59:56 2019