**** BEGIN LOGGING AT Mon Jul 25 02:59:56 2022
Jul 25 07:46:02 <mckoan> good morning
Jul 25 09:14:12 <reichi> good morning
Jul 25 09:14:41 <reichi> does anyone have an Idea why this whould happen on kirkstone when building samba (ubuntu 18.04):
Jul 25 09:14:41 <reichi> "| qemu-aarch64: Could not open '/lib/ld-linux-aarch64.so.1': No such file or directory"
Jul 25 09:15:14 <reichi> the file is in the recipe sysroot
Jul 25 09:54:25 <ptsneves> reichi: this looks like host contamination.
Jul 25 09:58:42 <reichi> yeah
Jul 25 09:59:06 <reichi> it is a migration from pyro (yeah I know, that's quite the leap)
Jul 25 10:02:14 <reichi> It's guess it's time to dig...
Jul 25 10:06:58 <ptsneves> reichi: i can tell you that i built samba with the latest master about a month ago for armv5 and the build was successful.
Jul 25 10:07:08 <reichi> yeah
Jul 25 10:07:14 <ptsneves> on ubuntu 20.04
Jul 25 10:07:19 <reichi> because i am migrating a "big" full, distribution
Jul 25 10:07:33 <reichi> in can't tell for sure that there's nothing on "my" side
Jul 25 10:07:35 <reichi> causing this
Jul 25 10:49:56 <reichi> hmm
Jul 25 12:38:38 <Tartarus> Check your layer for patches and other changes in the bbappend you maybe don't need, on samba?
Jul 25 12:45:53 <reichi> I'm pretty sure theres nothing left
Jul 25 12:46:04 <reichi> everyhting looks fine
Jul 25 12:46:28 <reichi> i already found out that using ./configure in waf-samba.bbclass seems to use host python
Jul 25 12:47:07 <reichi> that's fixed by explicitly calling "python3 ./buildtools/bin/waf configure"
Jul 25 12:47:43 <reichi> I have a "custom" target and I think it's probably also related to that
Jul 25 12:49:20 <reichi> now i have to find out why qemu_binary in waf-samba.bbclass is not used when expected
Jul 25 12:49:30 <reichi> I'm pretty sure that leads to the root cause...
Jul 25 14:59:40 <mort> is there any semantic difference at all between `foo:append += "bar"` and `foo:append = " bar"`
Jul 25 15:00:04 <mort> `foo:append = " bar"` looks a bit scary because it *looks* as if you've overwriting the appended data
Jul 25 16:07:58 <JPEW> mort: You are not overwriting the append data. `append` is a special "operator" that stacks as you might expect
Jul 25 16:08:13 <mort> right
Jul 25 16:08:21 <mort> the syntax doesn't exactly make it look like that
Jul 25 16:09:36 <mort> especially with `foo_append = " bar"`, it could've very well been the case that bitbake just has some code somewhere which does "${foo}${foo_append}"
Jul 25 16:09:46 <mort> I bet that's why everyone does 'foo_append += "bar"'
Jul 25 16:11:10 * JPEW is most assuredly guilt of that also
Jul 25 16:12:45 <Tartarus> Wasn't it true that it was, at some point in history tho?  Or am I now mis-remembering too
Jul 25 16:14:48 <JPEW> Tartarus: If it was, it was a long time ago
Jul 25 16:54:41 <Tartarus> I have more literal gray hair than I like to admit these days ;)
Jul 25 17:12:35 <reichi> :append += is acutually deprecated, iirc
Jul 25 17:13:33 <mort> it is
Jul 25 17:14:11 <mort> that's why I asked actually, before mass-replacing :append += "x" with :append = " x" I wanna make sure that's not gonna subtly break anything lol
Jul 25 17:36:59 <Saur[m]> mort: Current versions of OE will warn if you use `FOO:append += "..."` so you do want to replace any such occurrences.
**** ENDING LOGGING AT Tue Jul 26 02:59:56 2022