**** BEGIN LOGGING AT Sun Nov 08 03:00:00 2020
Nov 08 03:04:28 <zeddii> thanks JaMa, I see them. And the fix makes perfect sense.
Nov 08 08:49:18 <kanavin_home> RP: thanks :) I am experimenting with switching AB matrix from sato to weston
Nov 08 09:17:20 <RP> kanavin_home: I saw the patch mentioned...
Nov 08 09:18:53 <kanavin_home> RP: the full set is here
Nov 08 09:18:53 <kanavin_home> http://git.yoctoproject.org/cgit/cgit.cgi/poky-contrib/log/?h=akanavin/package-version-updates
Nov 08 09:18:53 <kanavin_home> http://git.yoctoproject.org/cgit/cgit.cgi/yocto-autobuilder-helper/log/?h=contrib/akanavin/transition-to-weston
Nov 08 14:55:54 <RP> kanavin_home: I've messed up the controller, going to restart it :/
Nov 08 18:51:42 <khem> RP: world build doesnt look bad, only 7 apps failed
Nov 08 19:30:31 <kanavin_home> RP: thanks, seems to work
Nov 08 19:30:45 <kanavin_home> (the disabling of buildperf in regular a-full)
Nov 08 20:10:45 <RP> kanavin_home: I'll have to test a release run to double check its right but it looks hopeful
Nov 08 20:43:20 <RP> kanavin_home: I'm going to need to restart the controller again, that change wasn't quite right :(
Nov 08 20:43:41 * RP wishes there was a way to test it :/
Nov 08 21:04:50 <kanavin_home> RP: no  problem, I got enough failures to sort through :)
Nov 08 21:05:27 <kanavin_home> RP: the sato->weston patchsets will go out as RFC to gauge reaction
Nov 08 21:06:22 <kanavin_home> RP: one thing that was revealed is that weston has been broken under systemd for quite a while :-/
Nov 08 21:08:38 <RP> kanavin_home: I think we are going to need to shake up the testing matrix a bit in this cycle
Nov 08 21:08:56 <RP> kanavin_home: I want to shake up the autobuilder step handling first though...
Nov 08 21:09:28 <kanavin_home> RP: yes, I feel there's a lot of unneeded or suboptimal cruft there, but can't put my finger on it exactly
Nov 08 21:10:09 <RP> kanavin_home: there are things which need to be tested and aren't (rt comes to mind and some arm coverage)
Nov 08 21:10:55 <kanavin_home> RP: yes, for instance doing world builds on something else than x86 would be good - upstreams generally tend to only do native x86 builds
Nov 08 21:11:41 <kanavin_home> RP: on the other hand, do we still need x32? Maybe we do, but for me, it's questionable.
Nov 08 23:16:23 <RP> kanavin_home: right, we could do with extending world builds but I don't really want to add more world builds as they take a long time, hard to know what to do for the best :/
Nov 08 23:16:40 <RP> kanavin_home: x32 would regress quickly if not tested there
**** ENDING LOGGING AT Mon Nov 09 03:02:03 2020