**** BEGIN LOGGING AT Thu Dec 29 03:00:00 2016 Dec 29 03:56:48 Herrie, you should probably post a comment on http://forums.enyojs.com/discussion/2564/installing-2-7-with-npm-gives-me-errors as the error messages are similar. Dec 29 05:56:27 DougReeder: Got npm 2.15.11 Dec 29 06:02:24 DougReeder: With cmder it seems to work now Dec 29 09:43:20 morning Dec 29 10:36:13 Morning Dec 29 10:43:24 Herrie|Pre3: I managed to reuse our current cm10.1 image on maguro, using the new lxc method; so there is an intermediate path which lets us work gently on cm12.1 migration for maguro and TP while having the mako already on cm12.1 Dec 29 10:43:54 Tofe: Nice :) Dec 29 10:44:21 Does the new lxc method also open up the multi-rom possibility? Dec 29 10:45:19 I... don't know how multi-rom works :p Dec 29 10:46:38 ygm :P Dec 29 10:47:02 It's over my head since it's low level linux stuff but I think you understand this :P Dec 29 10:47:28 Do you know if mounting /dev/mapper/store-cm--system on /system for TP would be a stable enough partition path ? Dec 29 10:49:03 maybe I should just try and see :) Dec 29 10:50:14 ok, should work Dec 29 10:52:26 Herrie: I can also migrate grouper along the way, but it will all be theoretical Dec 29 10:52:37 Tofe: OK Dec 29 10:52:55 I'm doing a nightly build on Jenkins now to confirm all is behaving Dec 29 10:53:10 Afterwards we can start migrating things across I guess Dec 29 10:53:46 Ah issue already :P Dec 29 10:54:06 PRservice jenkins.nas-admin.org:8585 not available Dec 29 10:54:26 novaldex|away : ^ Dec 29 10:54:44 uh, there is no fstab for grouper ? that's quite weird Dec 29 10:54:46 We've seen that one before :P Dec 29 11:52:03 hey Herrie, no problem, 2 secs Dec 29 11:56:08 Herrie: should be fixed now Dec 29 13:44:38 novaldex|away: Thnx seems to behave better now :) I.e. build is running Dec 29 14:13:19 novaldex|away: All seems OK now. Dec 29 14:13:27 For the build Dec 29 14:13:41 Need to see the rsync later Dec 29 14:13:55 Did you complete all the maintenance including the lxc stuff etc? Dec 29 14:14:18 I noticed a new Jenkins version is available. Is that something you will tackle or scoutcamper? Dec 29 16:06:20 Herrie|Pre3: I'll begin PR'ing stuff, but it's better if we discuss it a little bit before merging :) Dec 29 16:08:06 Tofe: Sure :) Dec 29 16:08:30 Anyway I already said we're not having a release in January so we got time to fix loose ends :P Dec 29 16:17:57 Hopefully the current november release was quite a good one Dec 29 16:24:04 The first PR is for adapting init.rc for our needs for CM-12.1. As no build uses this branch yet, it's pretty safe :) https://github.com/webOS-ports/android_system_core/pull/2 Dec 29 16:29:20 Herrie, JaMa: for meta-webos-ports and meta-smartphone, against which branch should I PR ? master ? krogoth ? Dec 29 16:30:45 I'll begin with a PR against krogoth, and if it's wrong I'll redo it correctly Dec 29 16:34:27 thanks Herrie, glad the builds are better & working Dec 29 16:34:55 we weren't ready for lxc/lxd just yet, so we've upgraded xen (as was our original plan) Dec 29 16:35:18 lxd will come later once we've had more chance to test it out Dec 29 16:35:19 novaldex: OK. Dec 29 16:35:30 thanks for the Jenkins update note, we'll go take a look Dec 29 16:35:43 Jenkins is what scout does normally I think Dec 29 16:36:19 i'm all for learning & taking some load off the others, i'll ask him & see what he feels like Dec 29 16:36:41 Tofe: krogoth normally. JaMa will cherry pick for others if needed. Dec 29 16:36:48 ok Dec 29 16:37:30 https://github.com/webOS-ports/meta-webos-ports/pull/205 and https://github.com/shr-distribution/meta-smartphone/pull/15 go together Dec 29 16:38:35 I'll also have to commit https://github.com/Tofee/android_kernel_google_msm/tree/mako/3.4/cm-12.1 into a new branch of https://github.com/shr-distribution/linux/tree/mako/3.4/master somehow Dec 29 16:40:03 Tofe: JaMa can make a branch for you in shr-distribution Dec 29 16:40:51 that would be nice Dec 29 16:41:21 if he can initialize it from my branch, that'd be even perfect :p Dec 29 16:45:40 Tofe: you got LuneOS working on TP with 12.1? Dec 29 16:46:28 nizovn: TP wasn't tried yet. Dec 29 16:47:19 didn't try yet; but that shouldn't be a big problem now Dec 29 16:48:02 Tofe: what you mean by new lxc method? Dec 29 16:48:15 i think /dev/mapper/store-cm--system might not present on TP Dec 29 16:49:38 nizovn: it's not *that* new, actually. It's just that I moved from using initramfs image + our cm10.1 build to initramfs image + minimal hybris android layer + original /system partition Dec 29 16:50:42 in my PR, I didn't migrate TP and maguro yet; I just use our current CM10.1 as the hybris android layer. It's not minimal at all, but it works as well. Dec 29 16:50:57 i don't like that we would depend on existing /system partition Dec 29 16:51:20 e.g. TP can be LuneOS only device Dec 29 16:51:41 well, we do already depend on having the correct firmwares **** BEGIN LOGGING AT Thu Dec 29 16:57:07 2016 Dec 29 16:57:33 ok Dec 29 16:58:33 For our CM10.1 wop build for tenderloin, a really simple fix would be to bind-mount the intermediate android layer to /system Dec 29 16:58:35 just thinking that having full android would require much more space than just /system Dec 29 16:59:05 Tofe: We use CM11 for Tenderloin not 10.1 afaik Dec 29 16:59:36 Herrie|Pre3: well, my sentence works too if we replace 10.1 with 11 :p Dec 29 17:00:31 I just mean we don't *have* to upgrade the CM version we're using today. The new approach is flexible enough to let us use only the current binaries if we want to. Dec 29 17:01:38 ok Dec 29 17:01:57 nizovn: anyway I haven't tested yet my PR for tenderloin (only for maguro), so I might as well do the bind-mount approach for CM11 as a start Dec 29 17:02:57 The big advantage of all this is that building a new hybris layer for CM12.1 for a given target device is quite easy now Dec 29 17:03:32 and you end up with a <20MB tgz Dec 29 17:05:16 sounds good :) Dec 29 17:28:53 Herrie, it’s good to hear you can build Messaging now. Can you send SMSs? **** BEGIN LOGGING AT Thu Dec 29 18:01:37 2016 Dec 29 18:11:42 DougReeder: Yes and no... Sending SMS works but chatthreader doesn't like it Dec 29 18:11:49 That could be Enyo 2.7 or my changes recently... Dec 29 18:12:07 We need to update webos-telephonyd anyway for the timestamp to be 13 digits Dec 29 18:12:16 For incoming SMS timestamp is wrong Dec 29 18:18:26 I'll test that shortly Dec 29 18:19:27 To see if it's your changes or it was already there Dec 29 19:31:41 nizovn/Tofe: You have any proper solution for a 13 digit EPOCH timestamp with miliseconds in C++? Dec 29 19:32:55 We currently have: https://github.com/webOS-ports/webos-telephonyd/blob/master/src/telephonyservice_sms.c#L103 which only does 10 digits Dec 29 19:33:51 And various bits that use: https://github.com/webOS-ports/webos-telephonyd/blob/9c9ead1f564646e698a62498238ac937cccf645c/drivers/ofono/ofonomessage.c#L117 Dec 29 19:35:00 you mean *1000? Dec 29 19:35:05 I came across the following in DB8 code: https://github.com/openwebos/db8/blob/7204e597a1774a72bddd3393a56b73eb1f2b174c/test/db/MojDbPerfTest.cpp#L352 which uses a custom header it seems: https://github.com/openwebos/db8/blob/7204e597a1774a72bddd3393a56b73eb1f2b174c/inc/core/MojTime.h Dec 29 19:35:23 But I also saw that C++ 11 has something with chrono that could take care of it Dec 29 19:35:30 Not sure what would be the preferred way Dec 29 19:37:25 Chrono way is discussed in https://stackoverflow.com/questions/16177295/get-time-since-epoch-in-milliseconds-preferably-using-c11-chrono Dec 29 19:37:52 And also came across another solution on this gist: https://gist.github.com/sevko/d23646ba07c77c15fde9 Dec 29 19:42:47 Herrie: yes, seems we can use ^ example, but using clock_gettime instead of gettimeofday Dec 29 19:48:44 nizovn: OK I know all 3 would eventually work, just wonder which would have preference Dec 29 19:48:53 I guess the third one is the cleanest one Dec 29 21:56:03 Herrie, what are the symptoms when the chatthreader doesn’t like it? **** ENDING LOGGING AT Fri Dec 30 03:00:00 2016