**** BEGIN LOGGING AT Tue Feb 13 03:00:02 2018 Feb 13 05:35:48 Morning! Feb 13 06:42:47 thejsa: These might be of help? https://github.com/ATLFlight/meta-sdflight-bsp/blob/master/recipes-core/dtbtool/dtbtool_1.1.bb and http://git.yoctoproject.org/cgit/cgit.cgi/meta-qcom/tree/recipes-devtools/skales/skales_git.bb?h=master Feb 13 06:43:23 And this one: http://git.yoctoproject.org/cgit/cgit.cgi/meta-qcom/tree/recipes-kernel/linux/linux-qcom-bootimg.inc?h=pyro Feb 13 06:44:42 And this one as well: https://github.com/ATLFlight/meta-sdflight-bsp/blob/master/recipes-kernel/linux/linux-sdflight.bb Feb 13 06:44:48 That should give some pointers I guess Feb 13 06:46:22 Tofe: Hmmz interesting error for Angler: https://bpaste.net/show/cff1a7f482bc Feb 13 06:46:25 Haven't seen this before Feb 13 08:17:04 Herrie: it's a bit weird for the dtb... how does it build on Android ? Feb 13 08:18:12 Herrie: for angler, could it be "vendor" is a symlink to itself or to its parent folder ? Feb 13 08:19:51 Tofe: Yeah it seems that might be the case. Feb 13 08:25:03 It has 0 bytes in the archive, so hard to say Feb 13 08:26:22 For DTB at Android side seems they use a new tool for QCom specifically Feb 13 08:26:51 abootimg hasn't been updated since 2011 it seems and there are some forks with modifications by the looks of it, that added the dt.img support Feb 13 08:52:00 mmh Feb 13 08:53:33 Seems PMOS guys also ran into it already and documented it a bit: https://wiki.postmarketos.org/wiki/QCDT Feb 13 08:54:58 Such nice fragmentation LOL... https://wiki.postmarketos.org/wiki/Dtbtool Feb 13 10:15:43 Tofe: On Angler seems that /vendor symlinks to /system/vendor. And /system/vendor symlinks to /vendor Feb 13 10:15:46 That won't work :P Feb 13 10:28:05 Herrie|Laptop: nope ! :) Feb 13 10:44:32 OK I guess I just add a "rm ${WORKDIR}/system/vendor" in the do_install_prepend() for now to "fix" it Feb 13 10:44:39 But should be sorted properly sometime Feb 13 11:48:20 Tofe: Good thing is that I now have an image ;) Feb 13 11:48:26 Probably not working, but it's building Feb 13 11:56:53 Ah, good :) Feb 13 11:57:59 So when linux_unix-10 is around he can test :P Feb 13 12:01:00 Tofe: I noticed that the mer/halium guys name their device specific rules 70-devicename.rules instead of devicename.rules which we use. This seems to have to do with the order of priority of the rules. I guess it would make sense for us to do similar? Feb 13 12:02:43 Herrie: Those links seem to be what we need but the trick is implementing them into our build flow, I don't think I have the skills to do that Feb 13 12:02:54 thejsa: I can give it a stab I guess Feb 13 12:03:28 Yocto/Bitbake has a bit of a steep learning curve but it's very flexible and scalable for builds :) Feb 13 12:03:35 might need to inherit from meta-qcom perhaps? Feb 13 12:04:01 I'm primarily a web developer, haha, quite new to embedded development like this Feb 13 12:05:17 thejsa: Same here ;) Feb 13 12:05:26 I can mock together a few things and follow instructions LOL Feb 13 12:24:25 linux_unix-10: Got something to test for you ;) Feb 13 12:25:46 http://www.vve-goudenleeuw.nl/luneos-dev-package-angler-20180213104209-0-0.zip Feb 13 12:26:00 I expect dragons, but worth to try and pull some logs Feb 13 13:19:49 Thanks! I'll have a go at it probably this weekend since that's when I can test this without people worrying why I'm not answering their messages. Feb 13 13:20:18 Y'know because the cell radio might not work ;) Feb 13 13:24:16 linux_unix-10: I doubt it'll boot at a first go, we'd need to add udev rules etc, but it builds already Feb 13 13:24:32 Just on Halium side 1 issue that I found while creating the image.... Feb 13 13:24:54 The /vendor is symlinked to /system/vendor and /system/vendor is symlinked to /vendor, therefore looping Feb 13 13:24:58 Yocto didn't like that Feb 13 13:27:09 thejsa: It seems we might be able to take https://github.com/ATLFlight/meta-sdflight-bsp/blob/master/recipes-core/dtbtool/dtbtool_1.1.bb and adjust the SRC_URI to point to https://github.com/LineageOS/android_device_motorola_athene/blob/16854cc7ca3589a17a7be66910aed8fe4e1e57fe/dtbtool/dtbtool.c instead Feb 13 13:27:44 And then try to do something similar as in https://github.com/ATLFlight/meta-sdflight-bsp/blob/master/recipes-kernel/linux/linux-sdflight.bb Feb 13 13:35:06 I'll play around a bit Feb 13 14:23:15 Herrie|Laptop: Yeah, adding udev rules is to be done on the first boot manually. Feb 13 14:31:56 linux_unix-10: I have no idea how far it'll get in terms of booting, this is just what I can do without an actual device for now ;) Feb 13 14:32:06 But Nexus devices are straightforward usually at least Feb 13 20:40:17 Tofe: sfos guys pointed me to init for the android_property_service and I found a PR from bshah there which seems to describe my issue: https://github.com/Halium/android_system_core/pull/1 Feb 13 20:40:24 So I'm going to rebuild using that branch Feb 13 20:40:32 That might sort the issue with hwcomposer eventually Feb 13 20:44:06 Herrie: yes, that's a very good pointer indeed Feb 13 21:17:24 Just we'd might need to rebuild the Halium bits for Athene, Onyx & Angler as well :P Feb 13 21:21:17 yes, quite for sure Feb 13 21:21:54 Herrie: what is your preferred way ? a patch, or a fork ? Feb 13 21:22:41 we already apply a set of patches, so that might be quite easy to add one corresponding to this PR Feb 13 21:26:06 Hello there. Is recommended to install latest stable version of LuneOs to Nexus 5? What should do be done? And, what app run in this Os? Excuse me for ignorance but I am very novice with this subject **** ENDING LOGGING AT Wed Feb 14 03:00:01 2018