<RP>
kanavin: from the backtrace, the issue is starting the fakeroot worker, which tends to mean something is wrong with pseudo
<RP>
kanavin: I'd probably dump the environment/command it is trying to run, try it manually and see what is wrong
<RP>
kanavin: it will be trying to run "pseudo bitbake-worker"
<RP>
but with more options and env
<thomas_34>
I have a recipe which produces ${S}/out/debug/A and ${S}/out/release/A. How can I control with packages that either debug/A or release/A gets installed at the same directory on the image? ${D}/somedir ?
<RP>
thomas_34: we don't support overlapping files in the same recipe, you'd need a second one
<thomas_34>
I know I can control which files belong to certain package with "FILES:{PN}-packagename += "..." ". But I cannot wrap my head around how to achieve that
<thomas_34>
Okay, RP thank you for stopping my misery to find a solution :D
<RP>
thomas_34: the reason is that in do_install, we place all the files into ${D}. They therefore can't overlap
<thomas_34>
RP, yep that was my impression. I just thought I miss some yocto feature to achieve stuff like my scenario
<kanavin>
RP: thanks, I realized I need lunch badly, and there isn't enough oxygenated blood in brain :)
florian_kc has quit [Ping timeout: 252 seconds]
<Saur>
thomas_34: You can have a pkg_postinst:${PN} function in the packages that moves the file into position.
<thomas_34>
Hmm okay. Is this approach kind of "safe"? Regarding rebuilding / sstate and so on?
zeemate has quit [Ping timeout: 272 seconds]
zeemate has joined #yocto
cyxae has joined #yocto
CrazyGecko has joined #yocto
Guest65 has joined #yocto
<rburton>
thomas_34: if you really wanted to have a single recipe instead of two (and you could have two where the second just includes the first and changes the variable it needs to) you could do the renaming during do_package. the packages would be correct and you wouldn't rely on postinsts doing the moves.
<thomas_34>
rburton thanks for pointing out a third way. I'll go with multiple recipes. It seems for me the least complicated way and lower risk to break something. Thanks for mentioned that I just could include the base-recipe and just overwrite existing variables.
<Guest65>
hi, I'm trying to write a linux kernel recipe and getting some errors. I can't see meaningful logs in log.do_compile to understand the error. Is there anything you can suggest for this?
<rburton>
don't write your own kernel recipe, use linux-yocto
<rburton>
if you must write a recipe then reuse the classes
<Guest65>
I'm reusing classes. Is there somehow a debug option so I can understand errors more?
<Guest65>
I think I got it. Although I give my own defconfig to the KBUILD_DEFCONFIG variable, another defconfig comes from somewhere.
Guest65 has quit [Quit: Client closed]
Guest65 has joined #yocto
prabhakalad has joined #yocto
<RP>
rburton: rename during do_package wouldn't work due to the way the rpm backend works
Guest65 has quit [Quit: Client closed]
Guest65 has joined #yocto
Guest65 has quit [Quit: Client closed]
goliath has quit [Quit: SIGSEGV]
thomas_34 has quit [Quit: Client closed]
<rburton>
another reason to ditch rpm :)
ptsneves has joined #yocto
chep` has joined #yocto
ptsneves has quit [Read error: Connection reset by peer]
ptsneves has joined #yocto
chep has quit [Read error: Connection reset by peer]
chep` is now known as chep
ptsneves has quit [Ping timeout: 276 seconds]
prabhakalad has quit [Quit: Konversation terminated!]
Guest65 has joined #yocto
ello- has joined #yocto
avocadosand has quit [Ping timeout: 268 seconds]
astlep5504018066 has quit [Read error: Connection reset by peer]
ello has quit [Ping timeout: 248 seconds]
ajfriesen164 has quit [Ping timeout: 248 seconds]
Fanfwe has quit [Ping timeout: 248 seconds]
NishanthMenon has quit [Ping timeout: 268 seconds]
avocadosand has joined #yocto
paulg has quit [Ping timeout: 248 seconds]
prabhakalad has joined #yocto
mithro has quit [Ping timeout: 252 seconds]
arisut has quit [Ping timeout: 248 seconds]
jkridner has quit [Ping timeout: 272 seconds]
salahaldeen has quit [Ping timeout: 252 seconds]
Crofton has quit [Ping timeout: 252 seconds]
salahaldeen has joined #yocto
jkridner has joined #yocto
dgriego has quit [Ping timeout: 248 seconds]
NishanthMenon has joined #yocto
arisut has joined #yocto
moto-timo has quit [Ping timeout: 272 seconds]
Fanfwe has joined #yocto
tortoise has quit [Ping timeout: 252 seconds]
ardo has quit [Ping timeout: 272 seconds]
dgriego has joined #yocto
<Guest65>
trying to write kernel recipe like that:
<Guest65>
but somehow it uses "olddefconfig" instead of "mymachine_defconfig". When I check the variable with "bitbake-getvar KBUILD_DEFCONFIG -r linux-foo" it returns "mymachine_defconfig" but in log.do_configure, is getting build using olddefconfig.
Tartarus has quit [Ping timeout: 272 seconds]
rynofinn____ has quit [Ping timeout: 272 seconds]
<Guest65>
does anyone have any idea how I can solve it?
frgo_ has quit [Read error: Connection reset by peer]
ardo has joined #yocto
frgo has joined #yocto
bradfa has quit [Ping timeout: 252 seconds]
ndec has quit [Ping timeout: 252 seconds]
jonmason has quit [Ping timeout: 268 seconds]
NishanthMenon has quit [Ping timeout: 252 seconds]
mansandersson86 has joined #yocto
NishanthMenon has joined #yocto
dl9pf has quit [Ping timeout: 248 seconds]
rynofinn____ has joined #yocto
Tartarus has joined #yocto
JPEW has joined #yocto
xtopher has joined #yocto
dl9pf has joined #yocto
jonmason has joined #yocto
tortoise has joined #yocto
bradfa has joined #yocto
diamondman__ has joined #yocto
astlep5504018066 has joined #yocto
<khem>
vmeson: regarding VTE g-i issue, the fundamental problem is that qemu-arm usermode dies with a illegal instruction and this illegal instruction comes from a dependent library in sysroot. So disabling clang for vte does not work. I tried to chase other deps like glib-2.0 to use clang but that did not help either
paulg has joined #yocto
<vmeson>
khem: noted... ugh.
<khem>
right approach would be to chase the illegal instruction and see if its a problem in qemu or in clang codegen
ndec has joined #yocto
vthor has quit [Excess Flood]
vthor has joined #yocto
vthor has quit [Changing host]
vthor has joined #yocto
florian has quit [Quit: Ex-Chat]
<Guest65>
Does anyone have any idea why olddefconfig is used even though I have set KBUILD_DEFCONFIG? I checked kernel.bbclass and kernel-yocto.bbclass everything seems alright.
rfuentess has quit [Remote host closed the connection]
Guest65 has quit [Quit: Client closed]
<khem>
Guest8241: is it calling KERNEL_CONFIG_COMMAND somewhere in kernel_do_configure
<khem>
RP: I send couple of remaining patches I have for clang series
thomas_34 has joined #yocto
Guest24 has joined #yocto
<Guest24>
Hi I want to maintain dev and prod version of the image in Yocto. I know that I can prepare image.bb and image-dev.bb and e.g. add some additional tools to dev img etc. What I wonder about is if there is a common way to use different flags when compiling other (application specific) recipes depending on the target image so prod vs dev. E.g. when
<Guest24>
compiling for prod I would like to compile with -O2 -DFORTIFY_SOURCE=2 but when for dev I would like to use -O0 -U_FORTIFY_SOURCE
<LetoThe2nd>
Guest24: using two separate distros
<Guest24>
LetoThe2nd so similar to images, have distro and distro-dev?
<LetoThe2nd>
Guest24: yup, basically. -dev can just include -prod and adjust the desired bits and pieces.
mckoan is now known as mckoan|away
frieder has quit [Remote host closed the connection]
cyxae_ has joined #yocto
cyxae has quit [Ping timeout: 276 seconds]
cyxae_ is now known as cyxae
<Guest24>
LetoThe2nd distro name also provides override right? So I could e.g. use EXTRA_OEMAKE:distro-dev override inside a recipe?
<LetoThe2nd>
Guest24: yup should work, if you extend DISTROOVERRIDES accordingly in the distro.conf