**** BEGIN LOGGING AT Wed Apr 26 03:00:01 2017 Apr 26 08:10:27 I'm trying to switch my init.d system to a systemd system but when i give uboot custom_params="init=/bin/systemd" to start systemd. but when i boot it returns: Requested init /boot/systemd failed Apr 26 09:12:41 khem: thanks Apr 26 12:22:00 khem: https://autobuilder.yocto.io/tgrid. note how the top row is going more green than the row below. the only difference is gcc6 vs gcc7: top row has all the fixes but doesn't actually switch compiler. i think its aborting on startup. Apr 26 14:51:29 khem: more buildsets are passing on the ab, gcc7 is definitely writing bad code for eg mips Apr 26 14:55:27 rburton, fix of the day Apr 26 14:55:29 http://cgit.openembedded.org/openembedded-core/tree/meta/lib/oe/utils.py#n68 Apr 26 14:55:34 http://cgit.openembedded.org/bitbake/tree/lib/bb/utils.py#n711 Apr 26 14:55:38 duplicated Apr 26 14:56:53 ant_work: welllll Apr 26 14:57:02 MLPREFIX is an OE specific thing Apr 26 14:57:08 so you can't just move the oe code to bitbake Apr 26 14:58:53 still, is called prune_suffix und checks prefix? Apr 26 17:44:05 rburton: ok, seems ppc and mips are not as good Apr 26 18:17:07 wmat, you changed how users are handeled on wiki? Apr 26 18:29:38 khem: yeah Apr 26 18:29:58 khem: arm64 failed in my respin but i think that was AB failure Apr 26 18:35:04 Crofton|work: well, i haven't re-enabled the account request extension yet Apr 26 18:35:31 ok Apr 26 18:35:44 just logged in and went to check for old requests Apr 26 18:35:49 login box was differenet Apr 26 18:36:09 yeah, that's turned off right now Apr 26 18:38:08 editing works Apr 26 18:43:17 yeah, it should for anyone with an account currently Apr 26 20:25:59 wmat: did you see JaMa's email on the mailing list re wiki permissions? Apr 26 22:04:53 bluelightning: yes, i saw it, just got home though Apr 26 22:05:33 Crofton|work: as a quick fix, just make Jama an administrator Apr 26 22:16:54 Crofton|work: nvm, i did it Apr 26 22:37:19 khem, around? Apr 26 22:38:40 I see again the images in tmp-musl are bigger than tmp-glibc...jffs2, ubi, ... all apart .tar.gz Apr 26 22:39:05 the tarball is 14 vs 16MB Apr 26 22:39:44 obvious diffs Apr 26 22:40:34 > glibc-binary-localedata-en-gb armv5e 2.25-r0 Apr 26 22:40:34 > glibc-binary-localedata-en-us armv5e 2.25-r0 Apr 26 22:40:56 > libc6 armv5e 2.25-r0 Apr 26 22:41:06 > libnss-mdns armv5e 0.10-r7 Apr 26 22:41:16 > locale-base-en-gb armv5e 2.25-r0 Apr 26 22:41:17 > locale-base-en-us armv5e 2.25-r0 Apr 26 22:41:29 < musl armv5e 1.1.16+git0+54807d47ac-r0 Apr 26 22:41:32 that's all Apr 26 22:42:08 why do the MTD images result bigger is yet unclear... Apr 26 22:46:49 khem, the uncompressed tarball confirms: Apr 26 22:47:27 33M musl vs 39M glibc (core-image-base plus mtd utils) Apr 26 23:21:12 khem: doh, on second rebuild the images shrunk...? Apr 26 23:27:19 seems happening on the first build from scratch. I'll tru to debug Apr 26 23:27:25 gn Apr 27 00:52:02 is there a way to specify multiple tasks/packages in a single invocation of bitbake? Apr 27 00:52:18 like for example, running cleanall+build on 2 separate packages Apr 27 00:55:50 i've tried a couple things but it seems both the clean and build tasks run at the same time so it gives an error similar to "temp is not empty" Apr 27 01:11:38 m4t, you have been around forever! Apr 27 01:11:51 I always think UK ham radi ocallsign Apr 27 01:20:44 i took a several year absence (since '10 or so) but yeah i've been around lol Apr 27 01:21:22 i was using a couple of those dht-walnut ppc405gp boards for a nas and ntp server waaay back. Apr 27 01:36:35 m4t: there's no way to specify that tasks specified should be run consecutively rather than in any way that dependencies allow, I have to admit I was considering whether we could add such an option just the other day though Apr 27 01:42:43 bluelightning: thx Apr 27 01:43:15 yeah i was digging through the bitbake source for a minute and nothing stood out to me w/rt that **** ENDING LOGGING AT Thu Apr 27 03:00:00 2017