**** BEGIN LOGGING AT Thu Aug 02 03:00:01 2018 Aug 02 05:50:42 Morning! Aug 02 05:50:45 Tofe: Seems thta Aug 02 05:51:12 that the halium-7.1 build failed. I'm not really sure why, so I kicked it off again, adding tissot as well now Aug 02 05:51:29 https://github.com/Halium/halium-devices/pull/106 should soon be merged, so that will kill the need of our android_luneos_hal I guess Aug 02 05:51:45 I guess for the Onyx we'd just fork the device repo and fix the file instead Aug 02 06:11:26 Morning! Aug 02 06:11:35 Herrie: ok thanks Aug 02 06:12:32 Also when building on my side, I remarked we still really need to remove halium-boot; at least as long as the device isn't really configured for it Aug 02 06:30:19 Tofe: what do you mean by "configured for it"? The device specific bits need to be added to halium-boot? I.e. partitions etc? Aug 02 06:30:33 yes, this kind of thing Aug 02 07:16:07 Morning Aug 02 07:27:04 Tofe: I don't see anything device specific in halium-boot? Aug 02 07:38:49 it's not in halium-boot iirc, it has to be added somehow in the device tree; that should be in halium's documation somewhere Aug 02 07:41:06 Tofe: Ah it's probably this bit for the fixup mountpoints? https://docs.halium.org/en/latest/porting/build-sources.html#include-your-device-in-fixup-mountpoints Aug 02 07:42:20 yes probably, ok it's in halium-boot not the device tree Aug 02 07:44:01 Well documentation still refers to hybris-boot Aug 02 07:44:26 But would we want to use hybris/halium boot? I thought previously we said we didn't? Aug 02 07:44:30 Or we'll need it for a/b? Aug 02 07:46:16 no, we don't need halium-boot right now Aug 02 07:46:39 I just would like the bare halium manifest to work as-is Aug 02 07:46:56 Yes that would be good Aug 02 07:47:01 we can very well build halium-boot and not use it Aug 02 07:47:03 Anyway adding the mountpoints is not rocket science ;) Aug 02 07:48:02 no, it would be better to have it a bit less centralized, but I don't really see how anyway Aug 02 07:55:59 Tofe: https://github.com/Halium/halium-devices/pull/113 Aug 02 07:56:20 That should get rid of our last bits of android_luneos_hal as soon as the other PR to halium-devices gets merged :) Aug 02 07:56:51 At least for 7.1 Aug 02 08:07:28 yup Aug 02 09:47:05 Tofe: It doesn't seem to like the Rosy build Aug 02 09:50:35 This doesn't look OK? "/tmp/jenkins4552542385952922613.sh: line 45: ./halium/devices/setup: No such file or directory" Aug 02 09:52:23 I suspect it doesn't pick up the new devices and it works for the previous ones because they were already pulled in Aug 02 09:52:28 Probably a silly typo somewhere Aug 02 10:24:32 I guess we'd need to SSH into the builder and see what's there in terms of structure Aug 02 10:56:24 Herrie|Laptop: http://jenkins.nas-admin.org/view/halium/job/halium-luneos-7.1-build/lastBuild/consoleFull the beginning of the log isn't very engaging Aug 02 11:01:33 Well we shouldn't need the old "IOError: [Errno 2] No such file or directory: '/home/jenkins/halium-luneos-7.1/.repo/local_manifests/luneos_targets.xml'' anymore Aug 02 11:01:39 So I guess some stuff just needs to get nuked :P Aug 02 11:04:49 Herrie|Laptop: I think the symlink /home/jenkins/halium-luneos-7.1/.repo/local_manifests/luneos_targets.xml is still there, pointing nowhere Aug 02 11:06:10 I've inserted a little cleanup command Aug 02 11:07:43 maybe we should cleanup luneos/luneos-hal too ? Aug 02 11:08:48 ah no, the issue is that it never really re-sync'ed because of the local_manifest issue Aug 02 11:08:52 I'll retrigger a build Aug 02 11:08:59 OK Aug 02 11:09:06 Yeah good catch on the manifest symlink Aug 02 11:10:36 now that's better Aug 02 11:11:59 I've also done the fix on 5.1 Aug 02 11:12:30 Yeha that doesn't hurt to have there either Aug 02 11:12:53 Hope the Halium guys will merge the PR soon and we can kill the android_luneos_hal for 7.1 at least Aug 02 11:13:01 For 5.1 we also don't need it anymore i think? Aug 02 11:13:10 I think I fixed it in Tenderloin manifest Aug 02 11:13:11 Let me check Aug 02 11:14:13 Seems so: https://github.com/Halium/halium-devices/blob/halium-5.1/manifests/hp_tenderloin.xml Aug 02 11:14:27 we don't apply it in the config already Aug 02 11:14:52 Ah OK Aug 02 11:15:01 Let me clean up the android_luneos_hal for this then Aug 02 11:15:04 To avoid confusion Aug 02 11:15:08 ok :) Aug 02 11:15:26 for 7.1 it seems we kind don't apply the patches at the good moment Aug 02 11:20:12 https://github.com/webOS-ports/android_luneos_hal/pull/8 Aug 02 11:20:58 thanks Aug 02 11:22:00 mmmh "prior sync failed; rebase still in progress" for luneos/luneos-hal... will it work if I just rm the luneos-hal folder before syncing ? Aug 02 11:26:52 good, it worked. Aug 02 11:27:09 now let's just wait. Aug 02 11:50:15 OK :) Aug 02 11:56:03 Will take a while I guess :P Aug 02 11:56:09 But it's good to have them build all there Aug 02 11:58:42 yes Aug 02 12:19:53 Seems it doesn't work for mido somehow Aug 02 12:21:58 I guess due to error: Cannot remove project "device/oneplus/onyx": uncommitted changes are present Aug 02 12:22:09 But that should be sorted with the force-sync I think :S Aug 02 12:25:40 Or we should commit the changes to the repos somehow before starting a new machine Aug 02 12:25:44 Just I'm not sure how to do that :P Aug 02 12:33:15 something like repo forall -vc "git reset --hard" Aug 02 12:34:59 I'm giving it a try Aug 02 12:58:32 Tofe: I'd need to add that locally to my script as well... Aug 02 12:58:42 I now usually just delete the device and vendor folder :P Aug 02 13:28:20 Tofe: Seems to work ? :) Aug 02 13:28:25 I hope so Aug 02 13:28:31 mido starts to build it seems now :) Aug 02 14:57:13 OK mido succeeded it's now continuing to rosy so seems to work Aug 02 16:28:58 Herrie: ninja: error: 'vendor/motorola/athene/proprietary/lib/hw/camera.msm8952.so', needed by '/home/jenkins/halium-luneos-7.1/out/target/product/athene/system/lib/hw/camera.msm8952.so', missing and no known rule to make it :( Aug 02 16:41:06 Tofe: I've seen that one locally as well Aug 02 16:41:12 It's an issue with proprietary files I guess Aug 02 16:41:19 I think in general we could remove athene from build Aug 02 16:41:28 Since we never really got far with it Aug 02 16:41:34 I'll have a look to see if I can fix it later Aug 02 17:43:47 Herrie, JaMa: https://github.com/shr-distribution/meta-smartphone/pull/73 and https://github.com/webOS-ports/meta-webos-ports/pull/295 should bring tissot almost up to date with what I have, except that we need the updated Halium image build in Jenkins Aug 02 17:45:10 I'll be on vacation in August, so I'll not necessarily be able to tinker much my devices... so at least you'll be able to play a bit on your side. Though it's still uneasy to debug the issues, because adb doesn't work yet :-/ Aug 02 18:02:04 At least some of that "rebase still in progress" stuff was caused by trying to apply patches in the wrong directory. cd into folder that doesn't exist. [Start but fail to] apply patch in cwd. Aug 02 18:02:05 Tofe: the monospace font is probably provided by ttf-liberation-mono package, not narrow Aug 02 18:10:02 nizovn: you sure ? It does currently work with this rdepends Aug 02 18:12:12 i didn't try that, but the recipe seems to install only narrow fonts: https://github.com/openembedded/meta-openembedded/blob/pyro/meta-oe/recipes-graphics/ttf-fonts/ttf-liberation-sans-narrow_1.07.4.bb Aug 02 18:13:05 and there is separate ttf-liberation-mono package here: https://github.com/openembedded/meta-openembedded/blob/pyro/meta-oe/recipes-graphics/ttf-fonts/ttf-liberation_0.2.bb Aug 02 18:15:05 mmh weird, I must have been lucky Aug 02 18:22:22 nizovn: better ? :) Aug 02 18:23:00 yes, thanks :) Aug 02 19:49:18 Tofe: I've disabled Athene for now and restarted a build on Jenkins Aug 02 19:49:26 I'll have a look into Athene later Aug 02 19:49:35 I fixed this locally, thought I PR-ed it, but maybe not :S Aug 02 19:50:04 Ah see I did Aug 02 19:50:10 Just need to PR somewhere I guess Aug 02 19:50:14 https://github.com/Herrie82/android_device_motorola_athene/commit/ea949e8d5445a0118a1f834f739ab394d8ed006f Aug 02 19:54:41 Tofe: OK PR-ed: https://github.com/Lyudmila17/android_device_motorola_athene/pull/1 Aug 02 19:54:50 Lyudmila17 is usually pretty quick Aug 02 19:54:56 So we could reinstate the Athene again Aug 02 19:56:43 Tofe: OK halium-devices setup script changes wer ejust merged Aug 02 19:56:55 So I'll remove the patches from android_luneos_hal now Aug 02 19:59:09 https://github.com/webOS-ports/android_luneos_hal/pull/9 Aug 02 20:08:52 Tofe: And Athene fix merged :D https://github.com/Lyudmila17/android_device_motorola_athene/pull/1 Aug 02 20:11:45 "Lyudmila17 is usually pretty quick" :-) Aug 02 20:25:41 elvispre: Seems the LineageOS guys have their own review system, they don't really use pull requests Aug 02 20:25:45 At least that's what I recalled Aug 02 20:26:13 Fair enough. I wasn't expecting it to be top of their agenda anyway. Aug 02 20:30:29 https://wiki.lineageos.org/submitting-patch-howto.html Aug 02 20:31:21 Maybe when I get an extra 24 hours a day I might get around to that :/ Aug 02 20:34:02 (Or if I happen to want to build a LineageOS image anyway.) Aug 02 20:35:22 That's what I mean :P Aug 02 20:35:29 I never bothered either tbh Aug 02 20:35:35 Have more useful things to do :P Aug 02 20:45:12 :-) Aug 02 21:30:29 wow great Aug 02 21:30:42 merged Aug 02 21:33:28 Tofe: Your meta-wop PR looks good to me Aug 02 21:33:33 Any reason we keep TLS patch for arm? **** ENDING LOGGING AT Fri Aug 03 03:00:00 2018