shoragan has quit [Read error: Connection reset by peer]
hnez has quit [Read error: Connection reset by peer]
OnkelUlla has quit [Read error: Connection reset by peer]
OnkelUlla has joined #yocto
shoragan has joined #yocto
beartama has quit [Remote host closed the connection]
beartama has joined #yocto
shoragan has quit [Quit: quit]
hnez has joined #yocto
shoragan has joined #yocto
davidinux1 has quit [Ping timeout: 268 seconds]
davidinux1 has joined #yocto
Abp has joined #yocto
Abp has quit [Ping timeout: 260 seconds]
Daanct12 has joined #yocto
Abp has joined #yocto
Abp has quit [Ping timeout: 255 seconds]
jclsn has quit [Ping timeout: 268 seconds]
jclsn has joined #yocto
lexano has quit [Ping timeout: 272 seconds]
khem has quit [Quit: WeeChat 4.2.2]
khem has joined #yocto
tgamblin has quit [Ping timeout: 252 seconds]
Xagen has joined #yocto
xmn has joined #yocto
<Hammdist>
so I demonstrated sufficient patience for riscv32 yocto to build .... but when I launch it with the qemu launcher thing it takes a very long time to boot, and the boot messages appear character by character really slowly .. is this expected?
<Hammdist>
also the system clock seems to run much slower than realtime because the dmesg-like lines are being printed with the timestamp around 2s but it has been running for many minutes now
jmd has joined #yocto
beartama has quit [Remote host closed the connection]
beartama has joined #yocto
tlammi has joined #yocto
lthadeus has joined #yocto
PaowZ has joined #yocto
<Hammdist>
however, because I have qemu-user installed on the host, I am able to mount the image and just chroot into it (no system-level emulation). I don't have gcc (or gccgo, which I am most interested in). how can I get gccgo added to the image?
alperak has joined #yocto
Abp has joined #yocto
Abp has quit [Read error: Connection reset by peer]
Abp has joined #yocto
Lihis has quit [Quit: Quitting]
Lihis has joined #yocto
<JaMa>
khem: https://git.openembedded.org/meta-openembedded-contrib/commit/?h=yoe/mut&id=42d64e3dfb40e3af847c673040d27207f8d832b8 has TODO for nodejs, instead of the work around from jansa/master-gcc14 branch and some of these are no longer needed as you added fixes before that (librelp, pcapplusplus)
lthadeus has quit [Ping timeout: 246 seconds]
Abp has quit [Ping timeout: 256 seconds]
Abp has joined #yocto
<khem>
JaMa: this is not going in as it is, I have already removed code parts for librelp and pcapplusplus
<khem>
ideally each warning will be fixed in its own way hopefully
<khem>
my first attempt was to make it such that it can keep working with gcc-13 as well.
lthadeus has joined #yocto
Abp has quit [Read error: Connection reset by peer]
Abp has joined #yocto
davidinux1 has quit [Ping timeout: 260 seconds]
davidinux1 has joined #yocto
c-thaler has joined #yocto
PaowZ has quit [Ping timeout: 240 seconds]
goliath has joined #yocto
<JaMa>
khem: all these -Wno-errors are backwards compatible with gcc-13 except return-mismatch and declaration-missing-parameter-type calloc-transposed-args template-id-cdtor (which I didn't use in meta-oe/oe-core), right?
<JaMa>
khem: I agree the issues should be fixed, I just added these as work around to see how badly world is broken (especially our own components) and for that I had to successfuly build the dependencies first
Abp has quit [Ping timeout: 255 seconds]
Abp has joined #yocto
enok has joined #yocto
Daanct12 has quit [Quit: WeeChat 4.2.2]
leon-anavi has joined #yocto
Jookia has joined #yocto
<Jookia>
o/ Is it possible to override a .inc file in another layer?
zpfvo has joined #yocto
Kubu_work has joined #yocto
jclsn has quit [Quit: WeeChat 4.2.2]
jclsn has joined #yocto
frieder has joined #yocto
mvlad has joined #yocto
mbulut has joined #yocto
gsalazar has joined #yocto
c-thaler has quit [Quit: Client closed]
jmd has quit [Remote host closed the connection]
<LetoThe2nd>
yo dudX
<JaMa>
Jookia: yes, see BBPATH variable
mcfrisk_ is now known as mcfrisk
tnovotny has joined #yocto
Guest22 has joined #yocto
<Guest22>
hello. building an yocto image for colibri imx6dl. shall i use kernel from meta-toradex-nxp or meta-freescale
jmd has joined #yocto
mcfrisk has quit [Quit: leaving]
enok has quit [Ping timeout: 260 seconds]
mcfrisk has joined #yocto
mckoan|away is now known as mckoan
<mckoan>
Guest22: the MACHINE steers the kernel selection
<mckoan>
Guest22: therefore it will be meta-toradex-nxp
<Guest22>
for some reason, llinux-toradex_5.4-2.3.x.bb is getting compiled (from meta-freescale-3rdpary)
<Guest22>
and i did
<Guest22>
IMX_DEFAULT_BSP = "nxp"
<Guest22>
include conf/machine/include/colibri-imx6.inc
florian_kc has joined #yocto
<mckoan>
Guest22: yes, my fault. the machine colibri-imx6 is into meta-freescale-3rdpary and the kernel is in its recipes-kernel
<Xogium>
sometimes I wish vendors had actual irc channels...
<JaMa>
some would probably ignore irc messages as they do with github issue tickets or e-mails to ML :)
<Xogium>
yeah
<Xogium>
:/
<Xogium>
but irc is really one of the best ways for screen reader users to communicate. All modern ui suck
<JaMa>
I always wish there would be more vendors which don't suck (more than 0 would be nice)
<Xogium>
the less worse I've found so far for my use, both on hardware design and on software stack is st. Their yocto distro aside, they are really not too bad
<Xogium>
but it always depend on what your use case is of course
florian__ has joined #yocto
<JaMa>
yes, it depends on what HW you're looking for, once it has some HW features to accelerate AI or media, then the BSP gets much worse in my experience, because they will try to at least demo some of the parts in variously weird ways
<Xogium>
yeah among other things
amurray_tgp has joined #yocto
<Xogium>
I try to stick to the software components like bootloader/kernel but use them on my own i.e: outside of their own distro, so far that has worked not too badly
lthadeus has joined #yocto
enok has joined #yocto
tlammi has quit [Quit: Client closed]
pvogelaar has joined #yocto
enok has quit [Ping timeout: 268 seconds]
<Jookia>
JaMa: I'll try it out. I'm trying to patch over a bug :(
<Jookia>
is it best to just use a patched git branch openembedded to patch bugs in it instead of trying to fix them with layers?
<Jookia>
setting BBPATH doesn't seem to really help me override qemu.inc
Guest22 has quit [Quit: Client closed]
<Jookia>
ok yeah i'll just fork it
<RP>
Jookia: layers usually work out much easier in the long run. If you set BBPATH correctly, you should be able to override an inc file
<Jookia>
wouldn't it just mean if i update yocto then it'll be mixing my old qemu.inc with new qemu recipes?
<Jookia>
that seems kind of troubling
<RP>
Jookia: that is what you asked to do :)
<RP>
Jookia: you could create a copy of the recipe in your layer if you prefer
<Jookia>
my overall goal is to patch a bug in openembedded-core so i was wondering what the workflow is for that
florian_kc is now known as florian
locutusofborg has quit [Read error: Connection reset by peer]
Saur_Home has quit [Quit: Client closed]
Saur_Home has joined #yocto
locutusofborg has joined #yocto
tlammi has joined #yocto
locutusofborg has quit [Changing host]
locutusofborg has joined #yocto
<JaMa>
Jookia: ideally fork for your local test, send the fix to openembedded-core first and only if it's not applied soon enough then apply it in your local layer with a link to upstream submitted fix
<Jookia>
JaMa: how do you apply it in a local layer? i don't think you can just apply patches to bitbake recipes?
<Jookia>
i'm not bothering to send them upstream as they don't affect supported distros
<Jookia>
it'll take a while for the bug to appear in supported distros
<JaMa>
Jookia: and if it's applied in newer upstream release than what you're using, I'm happy with "backports" layers like https://github.com/webosose/meta-webosose/tree/master/meta-webos-backports which you can just drop when upgrading to newer release (to make sure you won't carry your own qemu.inc copy forever)
<JaMa>
Jookia: if it takes a while for OE to reproduce the bug you already know about, then it will take even longer for someone else to fix it once it starts getting reproduced
<Jookia>
it's more that the bug happens when you use a newer python version or newer glib on host
<Jookia>
so maybe it'll break in the next ubuntu. it breaks on arch
<JaMa>
which I usually do with gentoo on host, so I for one I'm interested in what you're trying to fix in qemu
<JaMa>
and for some other builds I used 24.04 since before it reached alpha, so yes, I might trigger this qemu issue relatively soon
<Jookia>
JaMa: oh. qemu.inc leaks /usr/lib in to PKGCONFIG and on arch this tries to mix in host glib and qemu-native glib which errors. second bug is new python complains about an uninitialzed variable in the esdk
<Jookia>
i don't want to really spend time on a mailing list fighting over whether its okay to leak PKGCONF paths in to qemu's build
<JaMa>
the first one is surely upstreamable, right?
michal has joined #yocto
<Jookia>
i don't know, it might break qemu on very old systems that rely on leaking in host SDL
<JaMa>
ok, then I don't really want to spend time teaching you how to use BBPATH to provide your own qemu.inc
<Jookia>
fair enough
<Jookia>
i ended up just using git patching anyway
Saur_Home has quit [Quit: Client closed]
Saur_Home has joined #yocto
<Jookia>
thanks for the help anyway
<michal>
Hello, I am trying to separate /etc and /opt folders to new partitions. ETC is mounted in initramfs, before automatically changing root to main rootfs, and OPT is added to /etc/fstab. It works in general, but I occur a weird situation: my custom C++ + Java application has problem with reading /proc/meminfo file - that file is not visible for my application. From shell, when I login as a root I see that file, everyting is OK. I dont know, maybe
<michal>
it is important - my custom app is executed by systemd service. Separated partition have also automatically generated services: opt.mount i etc.mount.
<michal>
Maybe someone had a similar issue, or know where to look to find solution? If I put it back to one partition, then everythig is OK.
<mckoan>
michal: what's the user of your C++ + Java application ?
<ldywicki>
@michal why you would write /proc/meminfo?
<ldywicki>
aaa, sorry, you want to read it
<mckoan>
michal: what's the output of mount
<ldywicki>
proc fs does not come from fstab afaik
xmn has quit [Ping timeout: 268 seconds]
<michal>
mckoan: proc line from mount command: "proc on /proc type proc (rw,relatime)" - this same output I got where /etc i /opt is in one, main partition
Jookia has left #yocto [#yocto]
<ldywicki>
@michal do you have any indication of application error/bug? Do you read procfs from Java or C or combine both?
Jookia has joined #yocto
<michal>
ldywicki: its just a java call for C++ function where file is open and fgets return null for first line
<mckoan>
michal: can you run 'cat /proc/meminfo' from CLI ?
<Jookia>
does your systemd unit have security features enabled
<michal>
mckoan: I can, and it works. Problem is only with my application
tlammi has quit [Quit: Client closed]
<michal>
Jookia: no, service file contains only environment variables, ExecStartPre, ExecStart and baic after/wantedby configuration
<ldywicki>
@michal is there any raeson why you fetch contents of /proc via cpp? You can get these through JVM. I been doing some sysfs scans from java and these are working with no issues.
<Jookia>
michal: you may need to strace the service to see what's happening
tlammi has joined #yocto
<michal>
ldywicki: yes, but it worked before my changes (moving /etc and /opt to different partition). Of course I can rework that solution, but I wish to uderstand whats happen - maybe my approach with mounting /etc in initramfs and /opt in fstab is not correct
<michal>
Jookia: I will try to debug it more form application perspective
<Jookia>
moving a partition shouldn't affect partitions like /proc or friends. with systemd /proc should always be available for applications
<Jookia>
the only thing that i could think of is that you have somehow launched your application with the wrong root directory
<Jookia>
or there's some security feature preventing you reading /proc/meminfo
<ldywicki>
One thing which comes to my mind is missing sysctl.conf or so. I wouldn't worry much about /opt for now. Its /etc/ which can get you into trouble as multiple things are looking for information there. Since systemd kicks in your boot process itself is perfectly fine (I had a bit of struggle few weeks ago with that).
lthadeus has quit [Remote host closed the connection]
lthadeus has joined #yocto
lthadeus has quit [Quit: Leaving]
tlammi has quit [Ping timeout: 250 seconds]
lexano has joined #yocto
Piraty has quit [Ping timeout: 256 seconds]
locutusofborg has quit [Read error: Connection reset by peer]
florian__ has quit [Ping timeout: 264 seconds]
locutusofborg has joined #yocto
Hubi has joined #yocto
enok has joined #yocto
locutusofborg has quit [Ping timeout: 252 seconds]
locutusofborg has joined #yocto
<Jookia>
JaMa: BTW I apologize for my tone and if I upset you
Herdinger has joined #yocto
amurray_tgp has quit [Quit: Client closed]
ablu has quit [Ping timeout: 260 seconds]
ablu has joined #yocto
Guest21 has joined #yocto
tgamblin has joined #yocto
Hubi has quit [Quit: Client closed]
enok has quit [Ping timeout: 268 seconds]
enok has joined #yocto
|Xagen has joined #yocto
|Xagen has quit [Client Quit]
Xagen has quit [Ping timeout: 252 seconds]
Herdinger has quit [Quit: Client closed]
enok has quit [Ping timeout: 256 seconds]
xmn has joined #yocto
lthadeus has joined #yocto
Piraty has joined #yocto
enok has joined #yocto
sotaoverride is now known as Guest2723
Guest2723 has quit [Killed (tungsten.libera.chat (Nickname regained by services))]
ctraven_ is now known as sotaoverride
sotaover1ide has joined #yocto
lthadeus has quit [Remote host closed the connection]
Xagen has joined #yocto
lthadeus has joined #yocto
goliath has quit [Quit: SIGSEGV]
lthadeus has quit [Quit: Leaving]
Guest21 has quit [Quit: Client closed]
enok has quit [Ping timeout: 268 seconds]
tnovotny has quit [Quit: Leaving]
jmd has quit [Remote host closed the connection]
Abp has quit [Read error: Connection reset by peer]
<khem>
RP: few of gcc-14 patches did not make into master-next look at top 5 patches on contrib/kraj/gcc-14 branch
<khem>
JaMa: kids get you annoyed not upset, but age I agree
<khem>
:)
<khem>
rburton:quite oe-eeky :)
Xagen has joined #yocto
Abp has joined #yocto
Abp has quit [Ping timeout: 252 seconds]
Abp has joined #yocto
<JaMa>
khem: I don't know what's the root cause, but I'm grumpy most of the time of last couple years ;/
<JaMa>
I should drink more beers and build less images I guess
<JaMa>
and with that I've triggered 100+ image builds to keep jenkins servers warm while I sleep, gnite
Abp has quit [Ping timeout: 252 seconds]
<ldywicki>
@JaMa its jenkins, I would blame it. ;) To be fair, its one of most advanced ci/cd tools which gets nasty when it comes to maintenance and configuration management.