**** BEGIN LOGGING AT Wed Mar 13 02:59:57 2019
Mar 13 04:13:44 <learningc> Where is the cross-compiler located when I first install the Yocto package?
Mar 13 07:46:30 <jeanba> Hey, I'm trying to install a configuration file from my package into a configuration directory of another package (I'm adding a monit config file for my package into /etc/monit.d/), but bitbake complains: file /etc/monit.d conflicts between attempted installs of monit-5.25.2-r0.0.cortexa7hf_neon and xdaqbled-1.0+git1+e8cc47dff4-r0.1.cortexa7hf_neon
Mar 13 10:01:21 <yocti> New news from stackoverflow: How to place executable in to specied directory (path) under yocto buils <https://stackoverflow.com/questions/55138725/how-to-place-executable-in-to-specied-directory-path-under-yocto-buils>
Mar 13 12:23:19 <klemen> Hi, I'm having problems with the RTC clock on RK3288
Mar 13 12:23:30 <klemen> the RTC is PCF8563
Mar 13 12:23:57 <klemen> the kernel is this: https://github.com/rockchip-linux/kernel
Mar 13 12:24:04 <klemen> version 4.4.167
Mar 13 12:24:15 <klemen> release-4.4 branch
Mar 13 12:24:33 <klemen> here's what dmesg says:
Mar 13 12:25:05 <klemen> https://pastebin.com/grP5V8jT
Mar 13 12:25:34 <klemen> does anyone know what I can do?
Mar 13 14:02:08 <yocti> New news from stackoverflow: Why should one delete sstate-cache when deleting tmp in Yocto build? <https://stackoverflow.com/questions/55143217/why-should-one-delete-sstate-cache-when-deleting-tmp-in-yocto-build>
Mar 13 14:32:13 <yocti> New news from stackoverflow: Best current practice on versioning Yocto layer set <https://stackoverflow.com/questions/55144283/best-current-practice-on-versioning-yocto-layer-set>
Mar 13 17:32:47 <yocti> New news from stackoverflow: Why is a Yocto recipe being skipped? <https://stackoverflow.com/questions/55147646/why-is-a-yocto-recipe-being-skipped>
Mar 13 18:02:53 <yocti> New news from stackoverflow: Is someone able to build Yocto for Xilinx ZCU102? <https://stackoverflow.com/questions/55148394/is-someone-able-to-build-yocto-for-xilinx-zcu102>
Mar 13 18:05:39 <mrk377> QUESTION: I tried to create a symbolic link in IMAGE_POSTPROCESS_COMMAND to a non-existent file in /usr/local/testing (ln -s ${IMAGE_ROOTFS}/home/root/testing  /usr/local).  It fails.  Is this possible or does the scripting prevent this?
Mar 13 18:33:32 <rburton> mrk377: arguments the wrong way around?
Mar 13 18:33:47 <rburton> target then link name
Mar 13 18:38:54 <mrk377> haha.  Roger.
Mar 13 19:03:32 <Quazil1> What is the bitbake command to drop into the dev environment? bitbake -c <something> ?
Mar 13 19:08:50 <kergoth> devshell
Mar 13 20:35:36 <Striking7> Hey all. I'm having some strange issues with a fresh clone of yocto, using devtool
Mar 13 20:35:42 <Striking7> relevant output: https://pastebin.com/ujGTiV7a
Mar 13 20:36:00 <Striking7> You'll see around line 9 it's feeding the user agent stuff into wget
Mar 13 20:36:32 <Striking7> Again, fresh clone straight from the repository, I only ran oe-init-build-env followed by devtool latest-version xmodmap
Mar 13 20:36:59 <Striking7> It does it for busybox too
Mar 13 20:37:36 <kergoth> looks like maybe it ran an external command to get some value and it wrote something to stdout it wasn't expecting, and then blindly went to use that
Mar 13 20:37:36 <Striking7> https://pastebin.com/bnc4AtrS <-- same, but for busybox
Mar 13 20:37:39 <kergoth> odd
Mar 13 20:37:54 <Striking7> Yeah, it's been puzzling me a bit too.
Mar 13 20:38:10 <kergoth> off topic, but https://lh3.googleusercontent.com/-prpPLIMykSA/XIkkk9VidaI/AAAAAAAANpk/SXutY8xLNJogQRVuNCOqcYwG2-dkJeWOgCK8BGAs/s0/2019-03-13.jpg is amusing
Mar 13 20:38:18 <Striking7> This is on a (relatively) new install of Peppermint 9, based off Ubuntu
Mar 13 20:38:44 <Striking7> Hah - so true
Mar 13 20:42:36 <paulbarker> Striking7: Check that you've not got /bin/sh linked to dash
Mar 13 20:43:26 <Striking7> paulbarker: good catch! I *did* reconfigure with bash as my default.... now I have to figure out why I did that and whether or not it's safe to undo
Mar 13 20:43:28 <paulbarker> Just a guess but I've seen a lot of weird command execution happen due to dash not interpreting expressions as expected
Mar 13 20:43:48 <paulbarker> You want to be using bash. Not dash
Mar 13 20:44:28 <kergoth> i haven't switched my /bin/sh to bash in ages. shouldn't be needed anymore, not with current master
Mar 13 20:44:33 * kergoth shrugs
Mar 13 20:44:39 <bluelightning> right, shouldn't be needed
Mar 13 20:45:05 <Striking7> I'm configured for bash, yep
Mar 13 20:45:25 <bluelightning> assuming there's no recently introduced bug, I'm wondering if this is a case of the sysroot being empty of something that devtool is assuming will be there
Mar 13 20:45:49 <Striking7> does it with dash or with bash
Mar 13 20:46:45 <Striking7> Yeah I cant remember the last time I had to mess with that, but it had to have been years ago
Mar 13 20:47:23 <paulbarker> I've definitely seen issues with dash this month but I think it was on an AGL build with rocko branch of YP
Mar 13 20:57:07 <Striking7> 'm checking out the _fetch_index right now - it looks like it's appending --useragent in there. Must be missing a quote somewhere or the like
Mar 13 20:57:21 <Striking7> I checked, and it's doing it on sumo, thud, and master for me.
Mar 13 21:03:22 <Striking7> Works fine when I do this in _fetch_index:
Mar 13 21:03:23 <Striking7>             #fetchcmd += " -O " + f.name + " --user-agent='" + agent + "' '" + uri + "'"
Mar 13 21:03:23 <Striking7>             fetchcmd += " -O " + f.name + " " + uri
Mar 13 21:03:48 <Striking7> Ditching the user agent part works fine
Mar 13 21:04:40 <kergoth> oof, that's terrible code. that is *not* how you construct a shell command in python
Mar 13 21:04:58 <Striking7> I wasn't going to judge :)
Mar 13 21:07:48 <Striking7> Whelp. Figured out my problem, and it's unrelated to yocto
Mar 13 21:08:34 <Striking7> In an attempt to log what bitbake was doing I "temporarily" shadowed wget with a shellscript that logs all urls fetched and passes through to normal wget
Mar 13 21:08:45 <Striking7> I must have been careless in how I did it
Mar 13 21:08:57 <Striking7> Thanks for helping!
Mar 13 21:09:30 <kergoth> ah, forgot to send it to stderr instead of stdout, maybe?
Mar 13 21:10:33 <Striking7> I took $* and concatenated it to a log, then passed $* to wget
Mar 13 21:10:42 <Striking7> ... within quotes
Mar 13 21:10:51 <Striking7> >< <--- my own fault, sorry to bother!
Mar 13 21:11:15 <kergoth> ah. "$@" is better :)
Mar 13 21:11:21 <kergoth> no worries, happens to us all one time or another
Mar 13 21:12:17 <Striking7> Oh neat. Well at least I learned about $@
Mar 13 21:12:18 <mrk377> QUESTION:  I want to remove the /etc/timestamp file on the image.  I tried:ROOTFS_POSTPROCESS_COMMAND_remove = "rootfs_update_timestamp" but that leave a " ;  ; " in ROOTFS_POSTPROCESS_COMMAND.  You also can use "rootfs_update_timestamp ;" but that replaces " ;" throughout variable.  Is there another way to _remove a two part string?
Mar 13 21:18:29 <bluelightning> mrk377: I don't think so, but an alternative way to do this would be to redefine rootfs_update_timestamp to do nothing
Mar 13 21:19:51 <mrk377> Oh, hmmm.  Checking how to do that.  I was just going to blow away /etc/timestamp in a IMAGE_POSTPROCESS_COMMAND
Mar 13 21:26:48 <mrk377> Thanks bluelightning.  That was easiest.
Mar 13 22:03:43 <yocti> New news from stackoverflow: bitbake rootfs large for core image minimal <https://stackoverflow.com/questions/49490009/bitbake-rootfs-large-for-core-image-minimal>
Mar 13 23:16:17 * armpit hmm, all branches failing same bitbake selftest..
**** ENDING LOGGING AT Thu Mar 14 02:59:57 2019