<kanavin_>
RP: right, I didn't know there's a new ticket
<kanavin_>
yeah, HTTP Error 504: Gateway Timeout
<RP>
kanavin_: we created a new one since we'd changed infra and wanted to draw a line
<kanavin_>
for go-runtime
<RP>
kanavin_: if you do go through and see which artefacts they are, could you add a summary?
<RP>
I've been meaning to have a look at that
<kanavin_>
RP: Yes, I'll do that.
florian has joined #yocto
mckoan|away is now known as mckoan
<RP>
kanavin_: I did go through our sstate objects and tried to reduce the size of them where we could too
kanavin_ has quit [Ping timeout: 265 seconds]
kanavin has joined #yocto
<mcfrisk>
those CDN issues reminds me of a virtualization provider and their virtual block device infra which broke under virtual servers doing yocto builds. block device timeouts are not well handled by linux kernel..
kanavin has quit [Read error: Connection reset by peer]
kanavin has joined #yocto
jmiehe has joined #yocto
kanavin has quit [Remote host closed the connection]
kanavin has joined #yocto
rfuentess has quit [Remote host closed the connection]
rob_w has quit [Remote host closed the connection]
<RP>
adrianalin: thanks, it helps to know the issue is there in master
<RP>
adrianalin: no promising on if/when someone can look at it but it is at least tracked now
<adrianalin>
thank you too!
<RP>
adrianalin: we did make some recent tweaks to rust so I was hoping one of them might have helped but clearly now
<RP>
not
<adrianalin>
do you know what may cause this issue? some hints maybe? I looked on this for a few days now but without success (checked env vars in build.rs, cmake, all seem good); maybe is a cargo issue?
<RP>
adrianalin: The segfault probably means it is mixing two things it shouldn't such as host and target or sdk and target or host and sdk
<RP>
adrianalin: that is what I'd guess at
<RP>
adrianalin: strace the failing command and see what it accesses?
<adrianalin>
hmm would require a strace to follow all processes but there might be lots of output
<adrianalin>
*child
<RP>
adrianalin: lots of output but you save to a file and filter
<adrianalin>
yes, I will try this way
florian has quit [Ping timeout: 252 seconds]
Mayur has joined #yocto
mkazantsev has joined #yocto
<Mayur>
hello guys,
<Mayur>
i want to chnage compiler for a single recipe. GCCVERSION in local.conf acc to docs will chnage it for all. so what can be done here?
<rburton>
Mayur: have you already written the recipes for that different compiler?
<rburton>
(you can't just switch compiler)
leon-anavi has quit [Ping timeout: 268 seconds]
<Mayur>
no i have recipe for that app for now.
<rburton>
if you _need_ to then you'll need to write your own recipes for the entire toolchain that can coexist with the standard compiler, and change the depends in your recipe. much easier to fix the code that you need a different compiler for.
<Mayur>
ok thaks for the info. and how can i change it then for a single recipe / which variable needs to be set. can not find it in docs.
<rburton>
you turn off the default dependencies and then add your own for the recipes you wrote for the entire toolchain
<rburton>
i'll say again: don't do this
<rburton>
you either have a codegen bug in the compiler (fix it) or you have old code which the new compiler doesn't like (fix the code)
Xogium has quit [Ping timeout: 260 seconds]
<rburton>
or you have something actually special like "i want to build some firmware for a different cpu" in which case there are examples for that in meta-arm or meta-ti. the meta-arm approach works but is horrible, the meta-ti approach is more complex but better.
leon-anavi has joined #yocto
<rburton>
usual 'how to ask a question on the internet' point applies: explain what you're actually trying to solve, don't ask how to do what you think you need to do. often there's a better way but if nobody knows the context then nobody can help.
florian has joined #yocto
Xogium has joined #yocto
<Mayur>
Actual Problem:
<Mayur>
We have a C++ lib which is working fine at the moment for diff targets / Products. g++12.0 works fine. Our target with yocto 4.2 with gcc12 was working fine till now.
<Mayur>
Now we are updating yocto to 5.0 which brings gcc13.3.. So now our recipe for that code is failing. Fixing that app code is too much effort and will also cause problems for other owners which are using / compuling the same code. So trying to make it backward compatible with old gcc which was before.
ThomasRoos has joined #yocto
<JaMa>
Mayur: having 2 different toolchains is more effort than fixing the code
enok has quit [Quit: enok]
enok has joined #yocto
rfuentess has quit [Ping timeout: 252 seconds]
ThomasRoos has quit [Client Quit]
<Mayur>
yes came to know it now. Was thinking that it is just a variable in a recipe where you can define explicit version same as you can define PREFERED_VERSIOn for a package. but now i got it that its not simple like that one.
<JaMa>
correct and spending the time to fix the code is better for long term as you cannot be stuck on gcc-12 forever (there is already gcc-14 in styhead which brings many more build failures in not-well-maintained code)
<kanavin>
Mayur, if the c++ is written correctly, then it should work with both gcc 12 and 13. What are the actual fails?
<mcfrisk>
select the correct C++ standard version for the build, and fix the errors which are likely major bugs anyway
<mcfrisk>
disabling warnings as errors may be needed for the short term, but eventually those should be fixed too. there are likely lost of real bugs..
enok has quit [Ping timeout: 252 seconds]
rfuentess has joined #yocto
<Mayur>
Yes. Thanks for the info all. setting the same C++14 stand in cmake and its still failing at many positions like
<Mayur>
error: 'uint32_t' in namespace 'std' does not name a type; did you mean 'wint_t'?
<JaMa>
well that's just missing include which is backwards compatible
<bjdooks>
hmm, no stdint.h equivalent
<JaMa>
cstdint
<Mayur>
yes i know but why its working under gcc12 then
<JaMa>
the gcc even shows you hint which #include is missing
<JaMa>
Mayur: because some other header file included it for you transitively and now it doesn't
<Mayur>
ah ok.
<Mayur>
will tryy to solve these errors if possible
<Mayur>
thank you all for the info
<mcfrisk>
the include changes are really simple to solve in the end. work but manageable.
<mcfrisk>
other distros have done gcc updates and have possibly noted down common bugs and their fixes in large code bases
<JaMa>
I've added around 100 .patch files for our own components for gcc-13 and about 100 -Wno-errors for gcc-14 :/
<rburton>
i think gcc 13 was the one where they stopped the headers including other headers which was great for breaking code that didn't explicitly include everything it needed
* RP
wonders which patch in master-next is breaking go-runtime
* RP
worries it is the gold one
prabhakalad has quit [Ping timeout: 244 seconds]
prabhakalad has joined #yocto
uartz has joined #yocto
enok has joined #yocto
enok has quit [Client Quit]
enok71 has joined #yocto
enok71 has quit [Client Quit]
enok has joined #yocto
<rburton>
RP: huh
tec4 has quit [Quit: bye!]
tec4 has joined #yocto
enok71 has joined #yocto
enok has quit [Ping timeout: 276 seconds]
enok71 is now known as enok
Vonter has quit [Quit: WeeChat 4.5.1]
CrazyGecko has quit [Quit: Konversation terminated!]
BoergeSt has quit [Quit: WeeChat 4.1.1]
adrianalin has quit [Quit: Client closed]
dgriego has quit [Quit: Bye]
dgriego has joined #yocto
nerdboy has quit [Ping timeout: 252 seconds]
florian has quit [Quit: Ex-Chat]
mckoan is now known as mckoan|away
Mayur has quit [Quit: Client closed]
adrianalin has joined #yocto
kienan has joined #yocto
<RP>
rburton: I can't merge it until we resolve that one :/
<kienan>
hi, I'm building a distro with INITRAMFS_IMAGE="core-image-minimal-initramfs". In my case, I'm also installing kernel-modules into this rootfs. However, the kernel-modules for the core-image-minimal-initramfs and my other image are different and I don't catch why that might be. The two images use the same preferred virtual/kernel. Any ideas on what I should check?
Xagen has joined #yocto
leon-anavi has quit [Quit: Leaving]
nerdboy has joined #yocto
nerdboy has quit [Changing host]
nerdboy has joined #yocto
amitk has joined #yocto
enok has quit [Ping timeout: 260 seconds]
amitk_ has quit [Ping timeout: 260 seconds]
adrianalin has quit [Quit: Client closed]
enok has joined #yocto
florian has joined #yocto
mjm_ has quit [Quit: mjm_]
enok has quit [Quit: enok]
enok71 has joined #yocto
enok71 is now known as enok
enok has quit [Ping timeout: 252 seconds]
enok has joined #yocto
mjm has joined #yocto
mjm has quit [Client Quit]
mkazantsev has quit [Remote host closed the connection]
rfuentess has quit [Remote host closed the connection]
hnez has quit [Quit: No Ping reply in 180 seconds.]
zeddii has quit [Remote host closed the connection]
astlep5504018066 has quit [Read error: Connection reset by peer]