**** BEGIN LOGGING AT Wed Nov 20 02:59:58 2019 Nov 20 10:26:31 Crofton|work: https://fosdem.org/2020/news/2019-11-19-accepted-stands/ Nov 20 10:26:34 We have a stand Nov 20 10:27:50 paulbarker: so you will be https://www.youtube.com/watch?v=IOmMZBZGBps ? Nov 20 10:31:02 paulbarker: was there ever a year without OE stand? Nov 20 10:31:38 paulbarker: since 2007 at least Nov 20 10:31:53 hrw: Not that I know of. I've been going since 2014 and there's always been one Nov 20 10:32:18 I was manning in 2007, 2009 and maybe once or twice later Nov 20 10:34:10 LetoThe2nd: I shan't be the last! Nov 20 10:34:22 paulbarker: aw Nov 20 10:34:45 LetoThe2nd: More like https://www.youtube.com/watch?v=7woW7DmnR0E Nov 20 10:35:09 paulbarker: seriously "one of the last guys standing before the claening hordes arrive" just doesn't work as a song title. unless you're doing bossa nova, then maybe. Nov 20 10:35:58 paulbarker: heh. sounds like you have a plan! https://www.youtube.com/watch?v=bc0sJvtKrRM Nov 20 10:39:46 That's always the plan for FOSDEM Nov 20 10:41:03 :) Nov 20 14:46:06 RP:NOPACAGEDFILES patch is showing some fails in meta-oe http://errors.yoctoproject.org/Errors/Build/92679/ Nov 20 14:46:32 i wonder if they are true if so then i will look into it Nov 20 15:27:39 someone asked me if there was a way to say that a aprticular recipe can't be compile for anything but the primary machine lib (in a multilib case). I thought there was a way to flag that, but I'm not finding it Nov 20 15:28:48 I know kernel stuff generally doesn't get multilib compiled because of specific checks for kernel inherit or kernel recipe names in things.. but I still thought there was a generic way Nov 20 15:29:10 fray: NON_MULTILIB_RECIPES maybe? Nov 20 15:29:31 that would mean you'd need to know the name of each recipe.. I was looking for something that could be inserted into the recipe itself Nov 20 15:30:04 I wonder if you can do NON_MULTILIB_RECIPES = "${PN}" within the recipe tiself Nov 20 15:30:12 'er.. BPN Nov 20 16:45:41 khem: this NOPACKAGEDFILES is an experiment to see why we have recipes which don't install files and which ones they are. I've a problem with hash equivalence where I end up with empty packages so trying to debug that too Nov 20 17:16:49 RP: ALLOW_EMPTY_${PN} = "1" Nov 20 17:29:10 khem: right, but there can be other packages with files. As I said, its an experiment Nov 20 17:30:05 ok its a good one Nov 20 17:31:24 I found an interesting recipe due to this http://cgit.openembedded.org/meta-openembedded/tree/meta-oe/recipes-navigation/gpsd/gpsd-machine-conf_1.0.bb?h=master Nov 20 17:32:52 JaMa:http://cgit.openembedded.org/meta-openembedded/commit/meta-oe/recipes-navigation/gpsd/gpsd-machine-conf_1.0.bb?id=f9f3477f68fe980e72d77d5095a74acf182938d6 Nov 20 17:32:59 do we still need this recipe ? Nov 20 18:10:29 khem: I don't use gpsd (dropped whole meta-openmoko in Aug 2017), so I don't need this anymore (I don't know about others) Nov 20 18:11:51 khem: but we're still using systemd-machine-units for almost all MACHINEs and it has the same issue with empty packages Nov 20 18:22:09 JaMa: can it follow same conventions as weston-init ? Nov 20 18:23:49 I would say the convention is the same, just unlike weston-init there isn't any default content we need from systemd-machine-units nor gpsd-machine-conf Nov 20 18:25:09 we can add some bogus file to keep NOPACKAGEDFILES happy, but that doesn't make it much better imho, I guess we need some equivalent of ALLOW_EMPTY for NOPACKAGEDFILES as well Nov 20 18:29:17 my understanding is that NOPACKAGEDFILES is a hack to work out a bug? Nov 20 18:47:53 rburton: yes but it founds some stuff we can talk about in other contexts Nov 20 22:28:55 anyone need a job? https://careers.microsoft.com/us/en/job/739971/Senior-Software-Engineer Nov 20 22:40:41 just when I'm listening some podcast about Azure :) Nov 20 22:53:29 Rainy Redmond **** ENDING LOGGING AT Thu Nov 21 02:59:57 2019