**** BEGIN LOGGING AT Fri Mar 01 02:59:57 2019
Mar 01 06:10:48 <Herrie|Laptop> Morning!
Mar 01 06:10:57 <Herrie|Laptop> JaMa: I don't have anything pending for Halium 5.1
Mar 01 06:11:39 <Herrie|Laptop> Seems that Halium guys use the UBPorts android_system_core in their manifest and not their own :S https://github.com/Halium/android/blob/halium-5.1/default.xml#L205
Mar 01 06:12:35 <Herrie|Laptop> bshah: Any reason for this?
Mar 01 06:13:26 <Herrie|Laptop> bshah: I can PR the same changes to UBPorts, or we could cherry pick the one commit from UBPorts that we're missing from UBPorts and switch to Halium android_system_core repo
Mar 01 06:13:26 <bshah> umm.
Mar 01 06:13:43 <bshah> I'd say cherry-pick + switch
Mar 01 06:14:13 <Herrie|Laptop> bshah: OK
Mar 01 06:14:19 <Herrie|Laptop> Let me do that in a minute
Mar 01 06:25:00 <Herrie|Laptop> bshah: https://github.com/Halium/android_system_core/pull/7
Mar 01 06:25:04 <Herrie|Laptop> Only a single commit ;)
Mar 01 06:28:59 <Herrie|Laptop> bshah: And this one for the switch: https://github.com/Halium/android/pull/31
Mar 01 08:44:27 <JaMa> Herrie|Laptop: this should fix the 5.1 build?
Mar 01 08:44:49 <Herrie|Laptop> JaMa: Yes once bshah merges it
Mar 01 08:50:01 <JaMa> ok
Mar 01 08:50:04 <JaMa> thanks
Mar 01 08:50:57 <bshah> Herrie|Laptop: JaMa: done
Mar 01 08:51:06 <JaMa> cool
Mar 01 08:52:14 <Herrie|Laptop> bshah: Thnx!
Mar 01 10:09:25 <Tofe> elvispre: thanks for testing it; as soon as I saw that it didn't fix anything on rosy, I began to seriously doubt my fix... But now it's confirmed, it's not the right lead
Mar 01 11:16:56 <JaMa> Herrie|Laptop: http://jenkins.nas-admin.org/view/halium/job/halium-luneos-5.1-build/23/console
Mar 01 11:17:17 <JaMa> fatal: Couldn't find remote ref refs/heads/halium-5.1
Mar 01 11:17:17 <JaMa> error: Cannot fetch Halium/android_system_core from https://github.com/Halium/android_system_core
Mar 01 11:17:20 <JaMa> error: Exited sync due to fetch errors
Mar 01 11:26:20 <Herrie|Laptop> JaMa: Aarrgh... 5.1 is a bit of a pain compared to 7.1.
Mar 01 11:26:33 <Herrie|Laptop> Need to specify the revision in manifest.xml
Mar 01 11:27:14 <Herrie|Laptop> In https://github.com/Halium/android/commit/f8e985c17b2821ed827bd24205111bd9a312d5d4 add: revision="ubp-5.1"
Mar 01 11:27:22 <Herrie|Laptop> Can you manually test that and I'll PR to bshah
Mar 01 11:33:33 <JaMa> Herrie|Laptop: it's still running even after failed fetch, will try after it's done
Mar 01 11:33:45 <Herrie|Laptop> JaMa: Thnx, I'll try here too
Mar 01 11:53:59 <JaMa> Herrie|Laptop: should it be revision="refs/heads/ubp-5.1" ?
Mar 01 11:58:01 <JaMa> repo sync didn't fail, testing the build
Mar 01 11:59:10 <Herrie|Laptop> JaMa: Well it seems they were using the long names previously for some reason
Mar 01 11:59:50 <Herrie|Laptop> On 7.1 we mainly use the short variants
Mar 01 11:59:54 <Herrie|Laptop> On 5.1 seems a mixed bag
Mar 01 12:36:25 <JaMa> Herrie|Laptop: still fails to build: /usr/include/c++/7/bits/stl_deque.h:582:43: error: call of overloaded ‘swap(std::size_t&, std::size_t&)’ is ambiguous
Mar 01 12:41:23 <JaMa> ttps://github.com/Halium/android_system_core/commit/f2d250969041d6b728c0bc071d5a1b5b92daf19e#diff-6b4ce3fa12a69bed69df6b9bfca91a2d might be a fix
Mar 01 13:47:53 <JaMa> nope, not the right fix
Mar 01 13:59:43 <Herrie|Laptop> JaMa: Weird let me check logs
Mar 01 14:01:29 <Herrie|Laptop> JaMa: You hae more log?
Mar 01 14:01:35 <Herrie|Laptop> This was locally right?
Mar 01 14:24:54 <JaMa> on jenkins but by hand
Mar 01 14:25:50 <JaMa> btw: is there a reason why workspace halium-luneos-5.1-build
Mar 01 14:26:07 <JaMa> is using ~/halium-luneos-5.1 directory instead of its own workspace?
Mar 01 14:27:56 <JaMa> http://jenkins.nas-admin.org/view/halium/job/halium-luneos-5.1-build/ws/log.halium-5.1.txt
Mar 01 14:58:55 <Herrie|Laptop> JaMa: This one doesn't ring a bell to me
Mar 01 15:25:48 <Herrie|Laptop> JaMa: Hmmz getting it here now too
Mar 01 15:25:55 <Herrie|Laptop> While I had it working before
Mar 01 15:26:01 <Herrie|Laptop> Let me try to debug a bit
Mar 01 15:28:09 <Herrie|Laptop> I had 5.1 building for various targets clean last week here :S
Mar 01 15:33:16 <Herrie|Laptop> JaMa: I guess this might help
Mar 01 15:33:18 <Herrie|Laptop> https://github.com/LineageOS/android_libnativehelper/commit/c250bc97079a841c3f2c3aa6679784dab7f7d16a
Mar 01 15:37:26 <Herrie|Laptop> It's weird because I tried it clean earlier and it worked at my end :S
Mar 01 15:37:33 <Herrie|Laptop> Yup that solves it
Mar 01 15:37:58 <Herrie|Laptop> Just need to check if it also works on 16.04
Mar 01 15:38:06 <Herrie|Laptop> Starting VirtualBox
Mar 01 15:38:34 <JaMa> Herrie|Laptop: yes I'm looking at the same diff when comparing libnativehelper in 7.1 and 5.1
Mar 01 15:39:12 <Herrie|Laptop> JaMa: It succeeded to build here
Mar 01 15:39:21 <Herrie|Laptop> I took it from CM 13.1 but that's fine
Mar 01 15:39:42 <Herrie|Laptop> Just has less diffs v.s. CM14.1/Android 7.1 ;)
Mar 01 15:41:04 <Herrie|Laptop> JaMa: Seems we're using CM repo there directly. If you can confirm it works at your end we can ask bshah to fork it into Halium and I can PR there
Mar 01 15:42:03 <Herrie|Laptop> And LineageOS PR process is a pain :P
Mar 01 15:43:17 <JaMa> running on jenkins http://jenkins.nas-admin.org/view/halium/job/halium-luneos-5.1-build/ws/log.halium-5.1.fix.txt
Mar 01 15:43:29 <JaMa> with http://jenkins.nas-admin.org/view/halium/job/halium-luneos-5.1-build/ws/libnativehelper.swap.diff
Mar 01 15:45:48 <JaMa> Tofe: I've switched unstable back to thud and changed the gcc version to use 7, that should fix the build for devices with old kernels, then we can upgrade gcc separately on the rest of Yocto upgrade to unblock future upgrades (like to Warrior)
Mar 01 15:46:26 <JaMa> https://github.com/webOS-ports/meta-webos-ports/commit/ee36f5eae417565be6a076fde0886b5b8f33f0a0
Mar 01 15:56:11 <Tofe> JaMa: ok, thanks
Mar 01 15:56:33 <JaMa> still we should probably start with rocko and go from there
Mar 01 15:57:21 <JaMa> I'm trying to fix the qemux86 in pyro, but still getting distracted by some other issues :)
Mar 01 16:00:58 <Tofe> :)
**** BEGIN LOGGING AT Fri Mar 01 17:27:35 2019
Mar 01 18:32:08 <JaMa> Herrie: tenderloin built ok with that libnativehelper patch with some non-fatal errors like this one: cp: cannot stat '/home/jenkins/halium-luneos-5.1/out/target/product/tenderloin/root/init.recovery.*.rc': No such file or directory
Mar 01 18:32:51 <JaMa> haven't tried mako and others, because for this build I've disabled the device setup and repository cleanup (to prevent removing my local change), so mako failed shortly
Mar 01 18:45:13 <JaMa> Tofe: do we still need to use https://github.com/Tofee/halium-devices/commits/tofe/tissot in the override? Looks like it was merged to Halium few months ago https://github.com/Halium/halium-devices/commit/27c10c0fca0083505aac72d7915de7180f08d197
Mar 01 18:46:29 <JaMa> Tofe: also there is ~/halium-luneos-all directory on jenkins which I don't see where it would be referenced, but it was updated today by something
Mar 01 18:46:33 <JaMa> /home/jenkins/halium-luneos-all/.repo/local_manifests:
Mar 01 18:46:35 <JaMa> override_halium_device.xml
Mar 01 18:46:40 <JaMa> there is just .repo directory, nothing else in there
Mar 01 18:47:17 <JaMa> 6.9M    .repo/
Mar 01 18:51:30 <Tofe> JaMa: then the override isn't needed anymore -- for now. We can just comment it out.
Mar 01 18:51:57 <Tofe> as for halium-luneos-all directory... could it be that the script which gets the override xml puts it in the wrong directory ?
Mar 01 18:52:31 <JaMa> I've grepped halium-luneos-all in jenkins-job.sh and whole halium-build-5.1 and nothing was shown
Mar 01 18:54:13 <Tofe> weird
Mar 01 19:05:41 <JaMa> we're also using our own android repo for 7.1 https://github.com/webos-ports/android.git so this might be missing there, but that's more a reason to just merge it there I guess
Mar 01 19:06:28 <JaMa> or do we need different manifest (only for 7.1)?
Mar 01 19:08:13 <Tofe> we use subtree in that repo, to help keeping in pace with upstream's repo
Mar 01 19:09:14 <Tofe> (I included a shell script to help people update it when they want to)
**** ENDING LOGGING AT Sat Mar 02 02:59:57 2019