nerdboy has quit [Remote host closed the connection]
nerdboy has joined #yocto
nerdboy has quit [Changing host]
nerdboy has joined #yocto
davidinux1 has quit [Ping timeout: 256 seconds]
davidinux1 has joined #yocto
ablu has quit [Ping timeout: 256 seconds]
jclsn has quit [Ping timeout: 256 seconds]
jclsn has joined #yocto
<vvn>
how do you guys feed the DL_DIR and SSTATE_DIR from the CI that is going to be shared with developers? do you bind mount locations in the pipeline build, or do you copy/sync after the build?
<fray>
we have a download location (artifactory or web server). WE pass that URL into our MIRRORS and SSTATE_MIRRORS.. then when the build is completely rsync/upload any new files back to the mirror..
<fray>
downside, parallel build across multiple builds could cause the same file to be downloaded more then once.. but we workaround that by attempting a 'download' pass where we run through the various targets with a --runall=do_fetch ... then run the real builds..
<fray>
(to clarify we end up with a global mirror/sstate cache.. and then a build specific one.. but everything is handled via the mirroring system. We don't share the directories, we had problems with NFS in the past.)
goliath has quit [Quit: SIGSEGV]
wooosaiiii1 has joined #yocto
wooosaiiii has quit [Ping timeout: 272 seconds]
wooosaiiii1 is now known as wooosaiiii
Abp has joined #yocto
Abp has quit [Ping timeout: 268 seconds]
amitk_ has quit [Read error: Connection reset by peer]
amitk has joined #yocto
Abp has joined #yocto
goliath has joined #yocto
davidinux1 has quit [Quit: WeeChat 3.5]
jmd has joined #yocto
amitk has quit [Ping timeout: 260 seconds]
florian has joined #yocto
jmd has quit [Remote host closed the connection]
Abp has quit [Ping timeout: 272 seconds]
Abp has joined #yocto
Abp has quit [Read error: Connection reset by peer]
Abp has joined #yocto
florian has quit [Ping timeout: 255 seconds]
rob_w has joined #yocto
Saur_Home31 has quit [Quit: Client closed]
Saur_Home31 has joined #yocto
amitk has joined #yocto
Abp has quit [Ping timeout: 246 seconds]
Abp has joined #yocto
linfax has joined #yocto
florian_kc has joined #yocto
mbulut__ has joined #yocto
florian_kc is now known as florian
GNUmoon2 has quit [Remote host closed the connection]
GNUmoon2 has joined #yocto
rfuentess has joined #yocto
legraps has joined #yocto
mvlad has joined #yocto
LocutusOfBorg has joined #yocto
mckoan|away is now known as mckoan
frieder has joined #yocto
enok has joined #yocto
Kubu_work has joined #yocto
thomas_34 has joined #yocto
enok has quit [Ping timeout: 260 seconds]
florian_kc has joined #yocto
luc4 has joined #yocto
luc4 has quit [Client Quit]
Piraty has quit [Quit: -]
legraps has quit [Remote host closed the connection]
legraps has joined #yocto
Piraty has joined #yocto
leon-anavi has joined #yocto
tnovotny has joined #yocto
Bardon has joined #yocto
Bardon_ has quit [Ping timeout: 256 seconds]
wooosaiiii has quit [Remote host closed the connection]
<MrTatillon>
ERROR: libaio-0.3.110-r0 do_compile: Function failed: do_compile (log file is located at /home/aquassay/yocto/rpi-lite/build/tmp/work/cortexa7hf-neon-vfpv4-poky-linux-gnueabi/libaio/0.3.110-r0/temp/log.do_compile.24327)
<MrTatillon>
ERROR: Logfile of failure stored in: /home/aquassay/yocto/rpi-lite/build/tmp/work/cortexa7hf-neon-vfpv4-poky-linux-gnueabi/libaio/0.3.110-r0/temp/log.do_compile.24327
<MrTatillon>
Log data follows:
<MrTatillon>
| DEBUG: Executing shell function do_compile
<MrTatillon>
| Makefile:15: recipe for target 'all' failed
<MrTatillon>
| make: *** [all] Error 2
<MrTatillon>
| WARNING: /home/aquassay/yocto/rpi-lite/build/tmp/work/cortexa7hf-neon-vfpv4-poky-linux-gnueabi/libaio/0.3.110-r0/temp/run.do_compile.24327:1 exit 1 from 'exit 1'
<MrTatillon>
| ERROR: oe_runmake failed
<MrTatillon>
| ERROR: Function failed: do_compile (log file is located at /home/aquassay/yocto/rpi-lite/build/tmp/work/cortexa7hf-neon-vfpv4-poky-linux-gnueabi/libaio/0.3.110-r0/temp/log.do_compile.24327)
<MrTatillon>
I work on VirtualBox running OpenSuse
<MrTatillon>
I try to install `libaio-devel` but anything better
<MrTatillon>
The file `libaio.h` well exist in the folder, I do not know to do ...
<MrTatillon>
Is anyone can help me ?
<MrTatillon>
Thanks
<rburton>
you cut the log off before the actual error message. also krogoth has been unmaintained for *seven years* so _please_ move to a newer release
<mckoan>
MrTatillon: please DO NOT flood the IRC channel, use pastebin !
<MrTatillon>
ok, sorry
goliath has joined #yocto
LocutusOfBorg has joined #yocto
<rburton>
MrTatillon: that log says that gcc crashed without saying why, so i'd be checking dmesg for memory failures/ooms/etc. and upgrading to a recent release and not one that was abandoned seven years ago.
tnovotny has quit [Quit: Leaving]
<MrTatillon>
rburton I thought that the error was ` io_queue_init.c:19:20: fatal error: libaio.h: No such file or directory` , I will rebuild to see full log
<MrTatillon>
Ok I will looking dmesg
<MrTatillon>
For the unmaintained release I know it, but It is due to hardware and I try to maintain this one
<MrTatillon>
Furthemore, I well built this image in an other computer, I juste try to rebuild my working project to another computer (memory trouble one the first PC)
<rburton>
looks like libaio is broken. you're on your own for that, as its seven years old now.
MrCryo has joined #yocto
MrCryo has quit [Remote host closed the connection]
MrCryo has joined #yocto
<qschulz>
trying to merge two layers from different git repos into one big repo with two subdirs instead
MrCryo has quit [Quit: MrCryo]
<qschulz>
seems like we have scritps/combo-layer for that and I think I managed to get it working (needed to apply the root commit of each repo manually though)
<rburton>
my recommendation is to not use combo-layer
<rburton>
git can do that itself now with subtree merges
<qschulz>
however, I don't see how I can enable the "From XXX rev: c0ffee" in the commit log :/
MrCryo has quit [Read error: Connection reset by peer]
MrCryo has joined #yocto
<qschulz>
ah, it's an actual merge then :) ok, good enough I guess, thanks for the pointers :)
<MrTatillon>
rburton Ok, thanks anyway
rob_w has quit [Remote host closed the connection]
frieder has quit [Ping timeout: 260 seconds]
florian__ has joined #yocto
florian has quit [Killed (NickServ (GHOST command used by florian__!~florian@dynamic-078-048-170-174.78.48.pool.telefonica.de))]
florian__ is now known as florian
florian__ has joined #yocto
florian_kc has quit [Ping timeout: 240 seconds]
legraps has quit [Ping timeout: 256 seconds]
legraps has joined #yocto
frieder has joined #yocto
legraps has quit [Ping timeout: 256 seconds]
<kilobyte_ch>
Hello, I'm trying to build an ancient Kernel within Yocto (3.4.104). Is this something, that is still doable in scarthgap? Or is this nearly impossible? I'm getting errors like these, I guess they are due to a way to modern toolchain for such an old kernel: https://github.com/linux-sunxi/meta-sunxi/issues/407
legraps has joined #yocto
<rburton>
kilobyte_ch: yeah you'll need to pass flags to stop it warning/erroring from new compiler having better diagnostics
<kilobyte_ch>
rburton: ok, thats also a way. I thought about patching the sourcecode, but disabling the errors might be the easier and cleaner way.
frieder has quit [Ping timeout: 252 seconds]
legraps has quit [Ping timeout: 255 seconds]
<qschulz>
kilobyte_ch: upgrading to latest kernel may actually end up being easier than patching the old kernel so that the compiler doesn't complain anymore :)
<qschulz>
kilobyte_ch: being curious here (wobn't be able to help), what's the use of a NAND chip to do with not being able to run mainline kernel?
<qschulz>
kilobyte_ch: I remember allwinner had a 4.9 kernel about 6 years ago already, is 3.10 really the only solution? (not that the 4.9 is going to return less warnings :) )
<qschulz>
kilobyte_ch: ah yes, I see.. I heard too many stories about those MLC already :)
Marmottus11 has joined #yocto
<qschulz>
Vendors moved to eMMC now, I think it's more cost-efficient that it was half a decade ago, and so people have stopped caring about mainlining MLC support for NAND?
<kilobyte_ch>
qschulz: I just know this 3.4 kernel. Wouldn't ever use this combination of HW/SW for a new product. This is unfortunately an already existing device.
<qschulz>
(especially since I don't think there was a big push for it at the time already (I used to work with the previous and current NAND maintainers :) )
<kilobyte_ch>
Fully agree, eMMC is the way to go nowadays.
<kilobyte_ch>
(Way easier to implement and also way more reliable than many NAND based solutions)
<qschulz>
UFS for bigger capacities I think also
<qschulz>
but not too many embedded SoCs seem to support it right now
<fray>
I'm seeing NAND for small boot roms... eMMC for boot roms and/or file systems.. and now UFS for some new designs
<qschulz>
fray: NAND? usually small boot ROMs are going for SPI-NOR no?
<fray>
issue I've seen with UFS, not a lot of direct hardware support for it (yet).. so it seems like some designs are putting it behind a USB chip.. making it difficult/impossible to directly boot from..
<fray>
qschulz: qspi and ospi is far more popular then NAND.. but I have seen it used still for boot roms
<qschulz>
fray: have only read the RK3576 from Rockchip supports it
<fray>
but ya, at this point I'm not sure why you would use raw nand in a new design.. (ones I've seen it on are pretty much updated old designs)
<fray>
(UFS though requires fewer lines then eMMC.. so more future designs going in that direction)
MrCryo has quit [Quit: MrCryo]
MrCryo has joined #yocto
<qschulz>
fray: depends on the price of the chip too :)
<fray>
ya, sure that is part of it, but I have no real insight there anymore
legraps has joined #yocto
legraps has quit [Ping timeout: 260 seconds]
Abp has quit [Read error: Connection reset by peer]
<yocton>
Xogium: UFS = Universal Flash Storage. I learn that today. Looks like the future of eMMC, faster, bigger chips and serial link instead of parallel lines.
<Xogium>
huh. Interesting
<Xogium>
for me UFS meant universal filesystem
florian has quit [Ping timeout: 252 seconds]
<Xogium>
granted, I had only one experience with spi nand, and it wasn't exactly fun. I mean, I read up on it a lot, and I was lucky that the driver for it was already made... But yeah. 128 Mbyte spi nand on mangopi r3c board. Ugh. Damn thing took literally 2 minutes to boot -- no idea if that's normal because spi ? Yeah spi, not even qspi at minimum. Still I'm glad I learned that, either way
mckoan is now known as mckoan|away
florian has joined #yocto
jmd has joined #yocto
Guest86 has joined #yocto
Abp has quit [Ping timeout: 268 seconds]
Abp has joined #yocto
MrTatillon has quit [Ping timeout: 250 seconds]
mbulut__ has quit [Ping timeout: 255 seconds]
mjm has joined #yocto
rfuentess has quit [Remote host closed the connection]
florian has quit [Ping timeout: 255 seconds]
mvlad has quit [Remote host closed the connection]
pivi has quit [Ping timeout: 260 seconds]
LocutusOfBorg has quit [Read error: Connection reset by peer]
ptsneves has joined #yocto
pivi has joined #yocto
leon-anavi has quit [Quit: Leaving]
ptsneves has quit [Read error: Connection reset by peer]
MrCryo has quit [Remote host closed the connection]
Vonter has quit [Ping timeout: 252 seconds]
dankm has quit [Remote host closed the connection]
dankm has joined #yocto
Vonter has joined #yocto
gabriele_00 has joined #yocto
Kubu_work has quit [Quit: Leaving.]
enok has joined #yocto
Haxxa has quit [Quit: Haxxa flies away.]
Haxxa has joined #yocto
<yudjinn>
hello, is it still not possible to use non-std python modules in a task? i.e. in a python task using `import libconf` and having the recipe DEPEND on `python3-libconf-native` and `inherit python3native` does correctly have the module in the sysroot, but the task cant access it (says "libconf" not found)
dgriego has quit [Ping timeout: 260 seconds]
yudjinn has quit [Quit: leaving]
yudjinn has joined #yocto
ptsneves has quit [Read error: Connection reset by peer]
ptsneves has joined #yocto
gabriele_00 has quit [Quit: gabriele_00]
<rburton>
yudjinn: tasks run in the host python
<rburton>
literally, bitbake is python code, that forks into a worker processes, and that worker runs the python. python tasks don't run in a fresh instance of the interpretter
<rburton>
you can write a little script and run that using nativepython
<yudjinn>
rburton: so the only way to ensure that is to make sure the build host is also defined for truly reproducable? I assumed the workers also chroot for the sysroots to work out
<rburton>
sorry can't understand what you're trying to ask
<rburton>
python tasks run inside bitbake itself as otherwise you'd need to build python3-native3 to run python tasks, and the bulk of the core logic is written in python tasks
kilobyte_ch has quit [Read error: Connection reset by peer]
schtobia has quit [Quit: Bye!]
schtobia has joined #yocto
ptsneves has quit [Ping timeout: 252 seconds]
<RP>
khem: thanks! I've merged those in. I'm slowly trying to widen testing now the basics work
<RP>
yudjinn: you can "bitbake buildtools-tarball" then use that set of tools to run the build?
Abp has quit [Ping timeout: 256 seconds]
florian_kc has quit [Ping timeout: 272 seconds]
enok has quit [Ping timeout: 272 seconds]
Abp has joined #yocto
Abp has quit [Ping timeout: 264 seconds]
dgriego has joined #yocto
jmd has quit [Remote host closed the connection]
Guest86 has quit [Ping timeout: 250 seconds]
toric has quit [Remote host closed the connection]