**** BEGIN LOGGING AT Thu Feb 02 03:00:02 2017 Feb 02 08:42:57 good morning Feb 02 10:09:29 florian_kc: parcel sent ? Feb 02 10:14:51 florian_kc: update - just arrived Feb 02 10:16:46 +1 Feb 02 10:17:11 and I owe florian_kc a beer Feb 02 10:33:36 * florian is back from the meeting Feb 02 10:33:44 dl9pf: very good Feb 02 11:44:06 what is the recipe specific sysroot built from? Feb 02 11:44:29 what the recipe says its depends are Feb 02 11:45:03 there's a sysroot-components/ in tmp which has each recipe's sysroot contribution, those are glued together to make the recipe sysroot Feb 02 11:45:08 is it built from packages Feb 02 11:45:51 I ahve depend just not seems to make it into sysroot Feb 02 11:46:18 the files from each recipe that go into the sysroot is unchanged Feb 02 11:46:23 its just the sysroot is smaller than it used to be Feb 02 11:46:46 *unless* you have a nasty recipe that was writing directly to the sysroot instead of into the staging area, but that would have failed if built from sstate Feb 02 11:46:58 numpy Feb 02 11:55:26 urg found it in sysroot :) Feb 02 11:55:29 that works :) Feb 02 16:45:21 WARNING: npuhd-0.0.0+gitAUTOINC+f80e16a1e6-r0 do_package: QA Issue: npuhd: Files/directories were installed but not shipped in any package: Feb 02 16:45:21 /home/balister/src/oe-core/build/tmp-glibc/work/armv7ahf-neon-oe-linux-gnueabi/npuhd/0.0.0+gitAUTOINC+f80e16a1e6-r0/recipe-sysroot-native/usr/lib/python2.7/site-packages/npuhd.py Feb 02 16:45:40 I'm guessing installing in native systroot is really bad? Feb 02 16:46:03 what the hell Feb 02 16:46:06 (yes) Feb 02 16:46:17 lol Feb 02 16:46:18 yes Feb 02 16:46:21 ... Feb 02 16:46:29 I was surprised it caught that even Feb 02 16:46:39 fscking cmake Feb 02 16:46:46 and the idiots that write cmake files Feb 02 16:46:55 I'm rather tempted to play with process namespacing to make parts of the build directory / tmpdir read only to tasks.. Feb 02 16:46:57 liekly the latter in this case Feb 02 16:47:09 make it impossible to write to sysroot paths directly in do_install, etc Feb 02 16:47:14 hmm Feb 02 16:48:22 that would be handy Feb 02 16:51:32 I think I'll be sorting some mcake in the morning **** ENDING LOGGING AT Fri Feb 03 03:00:00 2017