**** BEGIN LOGGING AT Sat Feb 15 02:59:57 2020 Feb 15 14:30:37 @Saur: This seems to fix the problem: https://github.com/RobertBerger/meta-java/commit/a97d27ea0d6701df222c79fb1ce983fc39308c0c Feb 15 14:32:27 @Saur: if you think there is a cleaner solution please let me know. Feb 15 14:54:49 Good morning. I remember seeing a warning with a list of sw components under LGPL during the boot. I can't seem to enable that featuer again. Can someone point me to the doc? Feb 15 14:57:00 New news from stackoverflow: Showing a list of software with LGPL during boot Feb 15 15:06:13 Found my answer - it's from arago-gpl-notice from meta-arago. It only displays GPLv3 also. Feb 15 15:44:29 RP: http://errors.yoctoproject.org/Errors/Details/391145/ could it be related to sysrooting problem you are working on ? Feb 15 15:44:45 I have also sent a minor upgrade to webkitgtk which fixes the problem Feb 15 15:51:54 zeddii: hey, any plans to update linux-yocto 4.19.y to latest 4.19 ? Feb 15 15:53:28 zeddii: also, kernel/sched/fair.c doesn't even build, probably due to outdated RT patch Feb 15 15:54:01 khem: don't think so Feb 15 15:55:16 RP: this minor upgrade helped this was clang compiler btw Feb 15 15:55:36 I saw this error on few builds Feb 15 16:20:48 khem: thanks, I've pulled it in to test Feb 15 16:22:44 hm does VOLATILE_BINDS create any kind of systemd bindmount service files so I can do After= in a systemd service? Feb 15 16:38:11 Ad0: yes it does Feb 15 16:38:35 cool thanks Feb 15 16:38:45 I need to figure out the naming convention Feb 15 16:38:46 look at meta/recipes-core/volatile-binds/volatile-binds.bb Feb 15 16:38:53 yeah reading it now thanks :D Feb 15 16:39:03 services.append("%s.service" % what[1:].replace("/", "-")) Feb 15 16:39:12 usually its name of binds / is replaced wit - Feb 15 16:39:14 right Feb 15 16:39:19 that's not enough I'm afraid Feb 15 16:39:32 system-connection has a dash in it's name, it needs to be escaped Feb 15 16:40:21 https://www.freedesktop.org/software/systemd/man/systemd-escape.html Feb 15 16:42:19 so the correct name is for instance etc-NetworkManager-system\\x2dconnections.mount Feb 15 16:42:19 if the path is /etc/NetworkManager/system-connections Feb 15 16:59:11 should that recipe have a) "home made" systemd-escape stuff or call the systemd-escape binary? Feb 15 17:00:28 the escape string is just a hex ascii representation tho Feb 15 17:03:32 I will improve on it and maybe submit a patch to have better compat Feb 15 17:03:41 the unit name needs to match the path 100% Feb 15 17:32:50 sure Feb 15 17:59:46 can a bbappend overwrite a python function to "patch" it ? Feb 15 18:00:02 perhaps not Feb 15 18:00:22 yeah I want to test it in practice without modifying the original volatile-binds.bb Feb 15 18:10:52 hm just saw that it ends with .service so it may be other ruels Feb 15 18:30:36 maybe denix knows if this is dangerous or not to not have the systemd compatible names Feb 15 19:01:52 what's the best strategy to install some files on one machine but not for another?, having it in a custom dir that does NOT match the machine name under the recipe and do_install_append_MACHINE ? Feb 15 19:13:43 SRC_URI_machine += ? Feb 15 19:15:49 of course, it was even there from before *smack* I better take some time off lol Feb 15 20:20:53 Ad0: "SRC_URI_machine +=" is almost always wrong, SRC_URI_append_machine = " foo" Feb 15 20:30:39 ok Feb 15 20:30:46 I see either/or all over the place Feb 15 20:31:49 is there a clear-cut += vs _append rule? :) Feb 15 21:03:56 Ad0: SRC_URI_machine will override whatever was set in SRC_URI before Feb 15 21:04:36 Ad0: so it won't _add_ anything extra to existing SRC_URI even when combined with += Feb 15 21:04:55 Ad0: _append_machine will keep what was set there and just append extra items to it Feb 15 21:13:45 really Feb 15 21:29:37 RP: it seems meson bbclass changes are not doing good. my builds are breaking in parsing Feb 15 22:39:31 khem: ok, I'll put those on hold Feb 15 22:46:00 khem: we should probably start to think about making at least meta-oe parsing part of our incoming patch tests Feb 15 23:13:59 this time it is meta-gnome, but yes I agree wider testing would be good as long as it is quick **** ENDING LOGGING AT Sun Feb 16 02:59:57 2020