**** BEGIN LOGGING AT Sat Jul 14 03:00:00 2018 Jul 14 09:07:17 Does yocto verify downloaded packages? Jul 14 09:57:25 Hi Jul 14 09:58:27 I'm having a problem when using shell commands in recipe functions Jul 14 09:58:29 New news from stackoverflow: How to increase boot partition of hdd.img yocto Jul 14 09:58:59 for example if I use ls ${WORKDRI} in a do_unpack_append function I get a syntax error Jul 14 10:15:44 if I use the same command in the configure function then it works Jul 14 10:16:55 moham96: do_unpack is a python function Jul 14 10:18:00 I tried do_fetch also and it failed, so I can't use shell commands in these functions? Jul 14 10:18:54 moham96: since they're both python functions, appending shell to them isn't going to work Jul 14 10:19:09 moham96: you could add a postfunc which could be in shell Jul 14 10:21:27 bitbake was messing my kernel source after unpacking and creating corrupt symlink(symlink to a long string instead of a path), I was trying to debug this Jul 14 10:51:56 I'm running into http://lists.openembedded.org/pipermail/openembedded-core/2017-November/144205.html similar issues to this on rocko with multilib image Jul 14 10:52:06 is there some solution ? Jul 14 10:52:17 seems the patches were ignored Jul 14 12:48:32 Marex: there were some fixes in master. Some of the patch series was rather ugly :( Jul 14 12:50:12 RP: ha Jul 14 12:50:22 RP: is multilib ext sdk known to be broken-ish ? Jul 14 12:50:30 *ext sdk generation Jul 14 12:53:17 Marex: I'd say master stands a much better chance of working Jul 14 12:53:48 RP: I'm stuck on rocko because of the xilinx metalayer trash Jul 14 12:54:06 RP: I managed to avoid using most of it, but apparently not all just yet Jul 14 12:56:27 Marex: with the people we have its hard enough to fix the issues in master without fixing all the stable releases as well :( Jul 14 12:56:48 Marex: I know there are still some multilib issues left too :( Jul 14 12:57:51 RP: I'll keep digging :) Jul 14 17:45:28 The script meta/recipes-core/initscripts/initscripts-1.0/mountall.sh does a kill -USR1 1 Jul 14 17:45:49 USR1 is considered by busybox init has a 'halt' Jul 14 17:47:09 So if I use the original mountall.sh, it starts, and as soon as it goes past rcS, it shuts down Jul 14 17:49:55 sysvinit: src/init.c: * SIGUSR1 means close and reopen /dev/initctl **** ENDING LOGGING AT Sun Jul 15 03:00:01 2018