frieder has quit [Read error: Connection reset by peer]
goliath has joined #yocto
leon-anavi has joined #yocto
mckoan|away is now known as mckoan
troth has quit [Ping timeout: 250 seconds]
rfuentess has joined #yocto
troth has joined #yocto
gsalazar has joined #yocto
selff has joined #yocto
<selff>
hi everyone, im trying to activate otg mode in cm4 yocto image, but i couldnt see any information about it. in cm4 with raspbian, it was enough to write otg_mode = "1" in config.txt.
<selff>
i was able to enable i2c and dwc2 usb host mode support from the documentation in rpi-config_git.bb.
<selff>
but i couldnt find otg mode
michalkotyla has joined #yocto
troth has quit [Ping timeout: 248 seconds]
<amahnui[m]>
Please I wish to ask if there is documentation on how to install yocto on freebsd as the commands given for Ubuntu are not working for it.
<amahnui[m]>
Also I wish to ask why my cursor is not showing on the yocto window
<qschulz>
amahnui[m]: there does not seem to be a list of packages to install on FreeBSD, so you'll have to figure it out yourself I guess?
<qschulz>
Maybe that's part of the outreachy internship goal?
<qschulz>
What **I** would do is try to find the name of the packages in FreeBSD that corresponds to the name of the package that is said to be needed on Ubuntu
<amahnui[m]>
I wish to ask if a mouse can be used on the qemu-system window
troth has joined #yocto
<amahnui[m]>
My disappears when I bring it to the window
manuel1985 has quit [Ping timeout: 240 seconds]
<kranzo>
Using git subpath, patches will apply either for devtool or for bitbake because bitbake is patching on the subpath and devtool on the base repo... is there a way to fix that? I'm on dunfell
jmiehe has joined #yocto
zen_coder has quit [Quit: Konversation terminated!]
<kranzo>
Or maybe a better question: what is ${S} in the context of devtool ?
Guest10 has joined #yocto
<Guest10>
Hey. What are the downsides of using pip on Yocto?
<LetoThe2nd>
Guest10: are there any upsides?
<Guest10>
Being able to install python packages on runtime easily?
<kranzo>
Guest10: well thats not using pip on yocto but on a target right?
<Guest10>
Oh yeah, I meant on a target, mb
<LetoThe2nd>
Guest10: breaking reproducibility, requiring network access, requiring a writeable filesystem, breaking license tracking, depending on intended module pulling in a lot of hidden dependencies... do i need to go on?
<kranzo>
LetoThe2nd: you are just faster in typing right now ...
<qschulz>
LetoThe2nd: not even sure it works fine with wheels packages either, I don't know exactly how they identify the tag to be used (manylinux2014, etc...)
<qschulz>
Guest10: ^
Guest10 has quit [Quit: Connection closed]
<qschulz>
also, if there are source-only packages you download from registries, you'll need to compile them on target, so you'll need a toolchain and -dev packages for headers
<LetoThe2nd>
a.k.a. hidden dependencies
troth has quit [Ping timeout: 240 seconds]
<amahnui[m]>
I also wish to ask if the bugs on https://wiki.yoctoproject.org/wiki/Newcomers are still up to date now and how to know if has been assigned to someone or if it has been solved already.
frieder_ has quit [Ping timeout: 240 seconds]
Schiller has joined #yocto
beneth has joined #yocto
Schiller has quit [Client Quit]
OutBackDingo has quit [Ping timeout: 260 seconds]
troth has joined #yocto
<tomzy_0>
Hi, I have some questions regarding the Yocto release cycle. Assuming that the versions are described as MAJOR.MINOR.PATCH, we can assume that the LTS release will always raise the MAJOR version and the Stable release will always raise the MINOR version? Assuming the next LTS will be released in ~ 2 years, should it be version 5.0? The second
<amahnui[m]>
rburton_: It was mentioned on the outreachy portal that there will be a co-mentor for freebsd, Please I wish to
<qschulz>
amahnui[m]: just click on the ID link and check the status in Bugzilla.
<LetoThe2nd>
tomzy_0: it is likely, but not set in stone. don't base anything on the assumption, plwase.
<qschulz>
amahnui[m]: if no-one has commented on the bugzilla entry and it is not assigned, assign it to you and start working on it if you want
<tomzy_0>
LetoThe2nd: ACK, thanks for answer.
<amahnui[m]>
qschulz: that was what I was worried about
<amahnui[m]>
I thought it would not reflect if someone took it.
<amahnui[m]>
Thanks I will check some unassigned bugs and begin working on it
frieder_ has joined #yocto
frieder_ has quit [Ping timeout: 250 seconds]
<RP>
tomzy_0: we have no version policy which says LTS == MAJOR. It happens we did it that way this time but who knows in future
<tomzy_0>
Ok, and as for the patches for given version? They are added periodically after every month/week or so?
Schiller has joined #yocto
frieder_ has joined #yocto
<RP>
tomzy_0: patches get merged 2/3 times a week, releases have been every ~5 weeks but we may try and speed that up a little maybe once every 3-4 weeks for a PATCH level release
<Schiller>
i need to add a step in my build between the source ./oe-init-build-env and the bitbake build command. we got our own layer where i need to edit the .templateconf file. Where should i store my sed command?
<tomzy_0>
RP: Thanks!
frieder_ has quit [Ping timeout: 248 seconds]
<LetoThe2nd>
Schiller: automate the build with kas :-)
<Schiller>
LetoThe2nd: Can you explain it further. I dont know what <kas> is
<Schiller>
LetoThe2nd: Thanks. Maybe i forgot to mention that i run the YPAutobuilder Project already with the yocto-autobuilder-helper. I think an integration of another software is a little bit over the top (correct me if im mistaken). For my own setup i really just need to add an buildstep where i run an sed command.
<LetoThe2nd>
qschulz: coworkjer of mine offered to do a kas introduction at YPS :-)
<LetoThe2nd>
Schiller: indeed, you forgot to mention that, because then kas clearly doesn't apply.
florian has joined #yocto
<Schiller>
Im pretty sure i have to edit the run-config from the yocto-auto-helper/scripts folder to add an additional buildstep before the bitbake command. Are there any dependencies with variables etc which i have to take care of or can i just add a step the sed command with no worries?
Maxxed has joined #yocto
frieder_ has joined #yocto
<amahnui[m]>
qschulz: I just noticed that the package `build-essential` probably does have a freebsd alternative based of the pkgs.org link.
<RP>
Schiller: that should work
manuel__ has joined #yocto
<hmw[m]>
Hi when i run
<hmw[m]>
QSqlDatabase db = QSqlDatabase::addDatabase("QMYSQL", name);
<hmw[m]>
when i run in compiled debug mode without gdb connected the application runs fine
<hmw[m]>
in debug mode ( gdb connected ) the application that i run crashes
frieder has joined #yocto
manuel_ has quit [Ping timeout: 240 seconds]
frieder_ has quit [Ping timeout: 256 seconds]
<hmw[m]>
and i get a siggill
<amahnui[m]>
<amahnui[m]> "qschulz: I just noticed that the..." <- After doing some research, `build-essential` is already part of the freebsd base system
<qschulz>
amahnui[m]: please takes lots of notes so that you/we can document this later on
<Schiller>
when i try to build my image i get the following ERROR: ParseError at /home/pokybuild3/yocto-worker/meta-plusoptix/build/build/conf/auto.conf:17: unparsed line: 'BB_TASK_NICE_LEVEL:task-testimage = '0'' can someone explain how the auto.conf is generated and how the Error occurs? Like i said before i also eddited the .templateconf in the builddir
<Schiller>
/meta-poky/meta-plusoptix/ to be able to use our own layer.
manuel_ has quit [Ping timeout: 240 seconds]
florian_kc has joined #yocto
florian has quit [Read error: Connection reset by peer]
<qschulz>
Schiller: double quotes instead of single quotes for 0 ?
<qschulz>
I don't know if single quotes are supported as I've never used them
florian__ has joined #yocto
florian_kc has quit [Ping timeout: 256 seconds]
<td4>
You have used single quote then double quote '0"
<td4>
ah, no, it's from the error log
<Schiller>
td4 yes its from error log. i think i also found the problem
<Schiller>
td4 i need to edit my .templateconf before the source but the source already happens in step 5 (Unpack shared repositories) and i eddited the file aftewards.
rob_w has quit [Remote host closed the connection]
janvermaete[m] has joined #yocto
otavio_ has quit [Remote host closed the connection]
otavio has joined #yocto
Austriker has joined #yocto
rfried is now known as simond
simond is now known as rfried
simond47 has joined #yocto
<rfried>
Hi simond47
Austriker has quit [Quit: Client closed]
<amahnui[m]>
The package debianutils is not having an alternative for freebsd
<qschulz>
amahnui[m]: IIRC, it's for the `which` binary
<qschulz>
so just which should be there I assume
<qschulz>
amahnui[m]: there are other distributions supported which don't have the exact same packages
<qschulz>
so you can also take inspiration from those. e.g. fedora requires which package
<Schiller>
im curious. I'm working with the YPAutobuilder/yocto-autobuilder-helper. In which script is the oe-init-build-env getting sourced or where can i edit the .templateconf? I need it eddited before the source is happening.
tomzy_0 has quit [Quit: Client closed]
<RP>
Schiller: you can see it in run-config in bitbakecmd(), the cmd = ". ./oe-init-build-env; %s" % cmd
kranzo has quit [Quit: Client closed]
selff has quit [Quit: Client closed]
OutBackDingo has joined #yocto
Minvera has joined #yocto
xmn has joined #yocto
akiCA has joined #yocto
<amahnui[m]>
<qschulz> "so you can also take inspiration..." <- Alright I will check for it
<amahnui[m]>
Thanks
<Schiller>
RP: i'm sorry for asking but i dont have any knowledge in python. Where would i put the sed ShellCommand when i want to edit the .templateconf before the sourcing.
pbergin has quit [Quit: Leaving]
codavi has joined #yocto
akiCA has quit [Ping timeout: 246 seconds]
<manuel__>
Hi all. I'd like to enable license tracking in my image. I added LICENSE_CREATE_PACKAGE = "1" and NO_RECOMMENDATIONS = "1" (to not install the license package for *every* package, as they are recommended by default) to my local.conf. The build completes fine. Then I add os-release-lic to my image, and bitbake aborts with "bitbake aborts with Nothing RPROVIDES 'os-release-lic'". Why is that? Has anyone any advice?
<manuel__>
'os-release-lic' does get listed by `oe-pkgdata-util list-pkgs`.
<zeddii>
rburton: RP: we are having a packaging discussion on the meta-virt mailing list. Is it true that procps really hardcodes /usr/lib ? That seems like it should be causing issues throughout a bunch of packages when multlib is enabled, etc. Have either of you run into that ?
<manuel__>
qschulz: Pinging you, since you didn't hear from me for so long ;)
paulg has quit [Remote host closed the connection]
<qschulz>
manuel__: the docs seem to indicate that you need COPY_LIC_DIRS and COPY_LIC_MANIFEST too
<manuel__>
qschulz: That's not how I read them. COPY_LIC_DIRS is merely another solution to the same problem.
<manuel__>
I mean, that's how I read it
<qschulz>
manuel__: tried to grep into poky but nothing gets returned for os-release-lic
sakoman has joined #yocto
<qschulz>
manuel__: IMAGE_FEATURES would probably need an addditional lic-pkgs in your image recipe?
<RP>
zeddii: that seems strange to me. We run multilib builds on the AB
<manuel__>
qschulz: Same here. LICENSE_CREATE_PACKAGE gets read in pokys license.bbclass, but I couldn't see os-release including that bbclass.
GLumen has joined #yocto
<qschulz>
manuel__: license.bbclass is probably inherited by every package
<qschulz>
the same way base.bbclass is, but cannot guarantee
<manuel__>
qschulz: Makes sense
<RP>
zeddii: I guess it is internally coded and nobody has hit it until now
rsalveti has joined #yocto
<zeddii>
indeed. I find it strange (but what isn't strange about systemd ?) .. but I'm rolling with it. Will keep an eye out for it in other places.
<RP>
zeddii: I try not to talk about systemd ;-)
<LetoThe2nd>
RP: the first rule of systemd is...
<RP>
Schiller: since it is getting sourced for any bitbake command, just create a dummy bitbake command like "bitbake -p" which runs it once to get your template setup, then sed it in the next command, then you're good?
<RP>
LetoThe2nd: not for that reason though ;-)
selff has quit [Quit: Client closed]
alex__ has joined #yocto
<alex__>
hi everyone, I've created an npm package recipe and I get this error:
<alex__>
I've tried to add INSANE_SKIP_${PN} += "already-stripped" but I still get the error, why?
ahs3 has joined #yocto
ahs3 has quit [Client Quit]
<LetoThe2nd>
alex__: your npm package is bundling random binaries for random platforms. only sane solution is to get rid of those and force a rebuild from source that matches your build.
manuel__ is now known as manuel1985
florian__ is now known as florian
<Saur[m]>
manuel__: I have a faint memory that there is a problem with using `BAD_RECOMMENDATIONS` for a package and then trying to add required dependency on it. I assume it is the same if you ignore all default recommendations using `NO_RECOMMENDATIONS`. If I remember correctly this is a problem with rpm.
florian has quit [Read error: Connection reset by peer]
florian has joined #yocto
frieder has quit [Ping timeout: 250 seconds]
abelloni has quit [Ping timeout: 250 seconds]
beneth has quit [Ping timeout: 248 seconds]
<alex__>
LetoThe2nd: thank you, I'll look into what's the right way to do that
vmeson has quit [Remote host closed the connection]
vmeson has joined #yocto
<manuel1985>
When exactly is yocto reparsing the recipes? I though it's whenever local.conf changes, but now I observed a few times that I changed local.conf and yocto completed the "parsing recipes" stage within just seconds
<fray>
reparsing happens when certain hash values change. So if you add a comment, it may not reparse.
<fray>
If you changed a variable you believe should have changed behavior, I'd suggest asking on the mailing list as it might be a bug.
frieder has joined #yocto
td4 has quit [Quit: Client closed]
beneth has joined #yocto
frieder has quit [Ping timeout: 240 seconds]
<Schiller>
RP: i fixed it. quick question. is the line 179 in run-config the first source in the YPAutobuilder project?
<Schiller>
RP: when i run "echo '# Template settings' > .templateconf; echo 'TEMPLATECONF=${TEMPLATECONF:-meta-plusoptix/conf}' >> .templateconf; rm -r build/conf; . ./oe-init-build-env; %s" % cmd
<Schiller>
without rm it doesn't build as expected. it just edits the .templateconf.
<alex__>
LetoThe2ndjust to understand, those files have been stripped by? because I think I don't understand this and I don't have time to do it properly
<RP>
Schiller: No, there are other calls such as layer-config
frieder has joined #yocto
<hushmoney>
how do i know when to use += or _append or :append =?
<LetoThe2nd>
alex__: no. this has nothing to do with stripping. its the npm package including random binaries. its not about stripping, its about not shipping an arm binary to your x86 target, and vice versa.
<qschulz>
hushmoney: :append vs _append => honister and later releases should use :append
<qschulz>
hushmoney: if you have an older release, if it is recent enough, it should be compatible with both syntaxes
<hushmoney>
ok reading that pdf i think has given me some clue to ask another question that i hadn't been able to define before. FOO_bar_append <-- is bar always the MACHINE? can there be context-specific vars tracking other variables?
<qschulz>
hushmoney: bar can be anything in OVERRIDES
<qschulz>
this includes the distro name for example
<qschulz>
and if it's not part of OVERRIDES, then it's actually a variable called FOO_bar to which _append appends stuff
<hushmoney>
ah-ha, i understand. thanks!
BhsTalel has joined #yocto
<BhsTalel>
Hello, How to disable qemu-native-debian from being built (cuz I am building for other machine)?
<alex__>
LetoThe2nd: got it, thanks
<rfs613>
vmeson: I didn't realize it was legal to switch MACHINE like that. I always keep multiple build directories, one per MACHINE.
<qschulz>
rfs613: machine-specific recipes are built in machine-specific directories so it should be just fine
nk058[m] has joined #yocto
<rfs613>
qschulz: okay, but evidently license files overlap... and I'd expect the same for tmp/deploy/cve too
goliath has quit [Quit: SIGSEGV]
<qschulz>
rfs613: nothing is bug-free :)
<rfs613>
we can dream, can't we? :P
<nk058[m]>
Hi all. I am building an application which depends on a library. I have added that library in DEPENDS of that application's recipe. I am building an esdk, so I added library-dev into TOOLCHAIN_TARGET_TASK:append = " library-dev" in the image recipe. However, I do not see any header file and library in the Sysroot of esdk for target. I have read many discussions regarding the same problem, but I still cannot make it work.
<nk058[m]>
Also, is Sysroot for target and host on esdk is inside the `sysroot` in top directory of extracted esdk?
Guest5 has joined #yocto
<kergoth>
RP: not sure if you saw my comment/question about f-strings from friday. do you know if there's a thread about the contention on this anywhere?
<kergoth>
I don't see it as a big issue, was just wondering, since i ran across some interesting related projects recently
<Guest5>
Hello, I am trying to use d.setVar('X', 'value') for setting variable in custom bbclass and trying to access the value of 'X' in other per-defined class. But I am not able to get that value. Can you please me to understand whether I am missing anything or doing anything wrong?
<nk058[m]>
* Hi all. I am building an application which depends on a library. I have added that library in `DEPENDS` of that application's recipe. I am building an esdk, so I added `library-dev` into `TOOLCHAIN_TARGET_TASK:append = " library-dev"` in the image recipe. However, I do not see any header file and library in the sysroot of esdk for target. I have read many discussions regarding the same problem, but I still cannot make it work.
<nk058[m]>
Also, is Sysroot for target and host on esdk inside the `sysroot` in top directory of extracted esdk or it is in `tmp`?
_angelo_ has joined #yocto
_angelo_ has quit [Changing host]
alex__ has quit [Quit: Client closed]
rfuentess has quit [Remote host closed the connection]
frieder has quit [Remote host closed the connection]
manuel1985 has quit [Ping timeout: 256 seconds]
<kergoth>
Guest5: not en ough information. you'll only be able to access it from the same task in the same recipe
<vmeson>
rfs613: we can dream and work to fix known bugs! ;-) Want this one?
<rfs613>
vmeson: I think that one's still a bit over my head for now, and I'm still digging in the CVEs for now.
mckoan is now known as mckoan|away
leon-anavi has joined #yocto
<rfs613>
anybody know where task do_testimage, which are called in YP autobuilder, come from?
<rfs613>
ah, looks like I need to put INHERIT += "testimage" in local.conf
amitk has joined #yocto
kevinrowland has joined #yocto
<hushmoney>
can i append to a variable only when another variable exists? like SRC_URI_append = " file://${SYSTEM_USER_DTSI}" only when SYSTEM_USER_DTSI has been set by a machine override
<kergoth>
you can, but it'll be an invalid url when it's not set. 'file://' isn't likely to do well
<hushmoney>
exactly, that's why i only want to do it when the variable is set
<kergoth>
use inline python to do it conditionally. SRC_URI_apppend = "${@'file://${SYSTEM_USER_DTSI}' if d.getVar('SYSTEM_USER_DTSI') else ''}"
troth has quit [Ping timeout: 250 seconds]
<hushmoney>
lol thanks, i was jus tstarting to cobble together something that looks vaguely like that but not quite right. where is d.getVar documented?
<hushmoney>
kergoth: the URL: '${@"' is invalid and cannot be interpreted
goliath has joined #yocto
<hushmoney>
works if SYSTEM_USER_DTSI is set but if not it looks like that doesn't expand to an empty string ???
troth has joined #yocto
<kergoth>
hushmoney: two options, either SYSTEM_USER_DTSI ?= "" above where it's used, or change the python SRC_URI_apppend = "${@'file://%s' % d.getVar('SYSTEM_USER_DTSI') if d.getVar('SYSTEM_USER_DTSI') else ''}"
<kergoth>
th eoriginal should still have worked fine, though
manuel1985 has joined #yocto
leon-anavi has quit [Quit: Leaving]
<hushmoney>
thank you, assigning a default empty string works
leon-anavi has joined #yocto
<vmeson>
tgamblin: noted. I've taken the bug for now and may give it to a student in May.
kranzo has quit [Quit: Client closed]
<nk058[m]>
> Hi all. I am building an application which depends on a library. I have added that library in `DEPENDS` of that application's recipe. I am building an esdk, so I added `library-dev` into `TOOLCHAIN_TARGET_TASK:append = " library-dev"` in the image recipe. However, I do not see any header file and library in the sysroot of esdk for target. I have read many discussions regarding the same problem, but I still cannot make it work.
<nk058[m]>
> Also, is Sysroot for target and host on esdk inside the `sysroot` in top directory of extracted esdk or it is in `tmp`?
<nk058[m]>
>
<nk058[m]>
Im sorry, I don't know the policy of the group that I can repost/quote something I asked earlier. If someone could answer to that. Thank you.
<nk058[m]>
* Hi all. I am building an application which depends on a library. I have added that library in `DEPENDS` of that application's recipe. I am building an esdk, so I added `library-dev` into `TOOLCHAIN_TARGET_TASK:append = " library-dev"` in the image recipe. However, I do not see any header file and library in the sysroot of esdk for target. I have read many discussions regarding the same problem, but I still cannot make it work.
<nk058[m]>
Also, is Sysroot inside the `sysroot` in top directory of extracted esdk for target and host or it is in `tmp`?
<nk058[m]>
* asked earlier or not. If
Tokamak has joined #yocto
florian has quit [Ping timeout: 240 seconds]
Guest1410 has joined #yocto
florian has joined #yocto
<vvn>
What is the usecase for the 'from' part of mcdepends? i.e. specifying mc:foo:bar:recipe:task instead of mc::bar:recipe:task from within a recipe?
<vvn>
is the dependency not added if ${BB_CURRENT_MC} doesn't match foo?
amitk has quit [Quit: leaving]
whuang0389 has joined #yocto
<whuang0389>
Hi, I'm trying to install readonly rootfs, but Im not sure if it's actually enabled. I have IMAGE_FEATURES_append += " read-only-rootfs" in my local.conf, but I can still write to /. Is there a way to check?
<vvn>
whuang0389: the new override syntax is IMAGE_FEATURES:append = " read-only-rootfs". You can check by running 'bitbake -e <recipe> | grep ^IMAGE_FEATURES='
<whuang0389>
hm it seems like it's enabled: IMAGE_FEATURES="debug-tweaks package-management read-only-rootfs"
mvlad has quit [Remote host closed the connection]
<RP>
kergoth: I did see the comment. I'm not sure it is documented anywhere. It has come up a few times in meetings or maybe at one of the events too. Our minimum python version wasn't quite there yet and I was worried about backporting and breaking things for the LTS
<RP>
kergoth: not massively good reasons not to use them and I suspect we will have to start at some point
<RP>
kergoth: What I really don't want is "change the world" patches coming in and those are inevitable once we say we accept their use
<kergoth>
Oh yes, I've been guilty of that in the past, but committing after running conversion scripts sucks. I'd rather stuff be made to use it incrementally as we change things for other reasons
dev1990 has quit [Quit: Konversation terminated!]
paulg has quit [Quit: Leaving]
<RP>
kergoth: I'm trying to avoid this one as there is enough other stuff breaking things ;-)
<kergoth>
Huh, I didn't realize populate_sdk_ext chokes if you use it on a non-image recipe like meta-toolchain. Will have to fix that
<kergoth>
Fair enough, I was just curious about the reasoning, since options do exist. I was reading about how you can't do translations with it unless you use f-yeah or the like, but that got me thinking we should probably translate the bitbake strings someday :)
* kergoth
experimenting with 'external toolchain' support via pointing at an esdk to use its sstate artifacts and locked sigs for the toolchain components alone
<RP>
kergoth: We did once consider that but we decided most source code for software is in English so most bitbake users would have some understanding of it
<RP>
kergoth: translation resources are limited so they'd be better focused on our docs etc initally
<vvn>
should we rename TCLIBCAPPEND to TMPAPPEND?
<kergoth>
Hmm true
Guest1410 has quit [Quit: Client closed]
<vvn>
And TMPAPPEND ?= "-${BB_CURRENT_MC}" might be safer as well (catching more use cases from tweaking distro features for example)
Minvera has quit [Quit: Leaving]
linums has joined #yocto
<linums>
hi guys!
<linums>
should it be easy to create an efi loader image from an image?
<GLumen>
Does anyone actually use `--log-domains`? Is there syntax trick to getting it to work?
alimon has joined #yocto
leon-anavi has quit [Remote host closed the connection]
alimon has quit [Ping timeout: 248 seconds]
<RP>
smurray: I'm hoping you can come up with a fix for meta-agl next and the connman issue? :)
alimon has joined #yocto
<abelloni>
smurray: I'm hoping too ;)
<smurray>
RP: does it actually fail atm on the AB? I've been doing test builds with latest kirkstone today and not seen a failure
<smurray>
RP: but I've been remiss in pushing a tweak, can do so shortly
<khem>
RP: I have taken the gcc12 patch from bero and reworked a bit, it was having musl issues, which are fixes, I think Alex ran a AB build with it I will see if I can fix the failures there and send a v3 tonight
<smurray>
RP: I've pushed what I believe is a fix, fingers crossed. I'm still scratching my head as to why I wasn't seeing build failures here in my builds yesterday and today
<khem>
RP: I have pushed the patch into kraj/pokt-next on poky-contrib if you have cycles
<smurray>
RP: my apologies for forgetting about that, was off sick a couple of days, but should have remembered to look at it yesterday