reatmon_ has quit [Remote host closed the connection]
reatmon_ has joined #yocto
paulg has quit [Ping timeout: 252 seconds]
dankm has quit [Remote host closed the connection]
dankm has joined #yocto
sotaoverride has joined #yocto
rjones2 has quit [Quit: Client closed]
dankm has quit [Remote host closed the connection]
dankm has joined #yocto
jclsn has quit [Ping timeout: 252 seconds]
jclsn has joined #yocto
MattWeb__ has quit [Quit: Client closed]
lexano has quit [Ping timeout: 276 seconds]
pbsds3 has quit [Ping timeout: 244 seconds]
Jones42__ has joined #yocto
paulg has joined #yocto
jkale has joined #yocto
joekale has quit [Ping timeout: 244 seconds]
Jones42_ has quit [Ping timeout: 260 seconds]
jmiehe has joined #yocto
jmiehe has quit [Client Quit]
ablu has quit [Ping timeout: 245 seconds]
ablu has joined #yocto
Articulus has joined #yocto
Guest88 has joined #yocto
<Guest88>
hi all
<Guest88>
I have just cloned poky and checkout kirkstone, but during the build get stuck at fetching this repo: git://anongit.freedesktop.org/git/virglrenderer
<Guest88>
effectively it seems not available. How it can be possible that a branch updated 7 days ago refer to a repo not reachable?
heidaren has quit [Remote host closed the connection]
heidaren has joined #yocto
ehussain has joined #yocto
pbsds3 has joined #yocto
<zeddii>
09:03 zeddii: that particular document won't really help you. I've been through it about a hundred times, you just run into the guid/uid mapping that requires setup and permissions, I've been working on modifying the docker source code to not require it in an install configuration.
<zeddii>
I'm doing an engine/vfs agnostic uid/guid workaround, but I'm hitting a segfault, more to do on that front yet.
joekale has joined #yocto
jkale has quit [Ping timeout: 248 seconds]
enok has joined #yocto
goliath has joined #yocto
enok has quit [Ping timeout: 276 seconds]
alessioigor has joined #yocto
CrazyGecko has joined #yocto
xmn has quit [Ping timeout: 276 seconds]
enok has joined #yocto
enok has quit [Ping timeout: 244 seconds]
rob_w has joined #yocto
enok has joined #yocto
enok has quit [Ping timeout: 265 seconds]
enok has joined #yocto
filipm has joined #yocto
mckoan|away is now known as mckoan
zpfvo has joined #yocto
prabhakalad has quit [Quit: Konversation terminated!]
prabhakalad has joined #yocto
<filipm>
filipm What is the proper way to create an image that depends on another image on scarthgap?
<filipm>
I've tried to use DEPENDS += "image2" in image1.bb, but I get ERROR: image1-1-1.0-r0 do_populate_lic_deploy: Couldn't find license information for dependency image2.
<filipm>
My goal is to create have clean partitions for different applications. I.e. one rootfs for system deps, another for apps that are mounted as separate partitions.
Guest0118 has joined #yocto
<mckoan>
filipm: you can extend an existing image using a .bbappend or include the image you depend on using: require theotherimage.bb
<mckoan>
filipm: about your error, declare LICENSE = "CLOSED"
enok has quit [Ping timeout: 272 seconds]
rfuentess has joined #yocto
enok has joined #yocto
Daanct12 has joined #yocto
<filipm>
mckoan, I've tried this before, but the result was that one rootfs has /appfs populated; I want the mount points to be empty. The only way I could think of to be able to have clean mount points to use two rootfs recipes. The license is already set to CLOSED.
<Guest88>
Hi, I have just cloned poky and checkout kirkstone, but during the build get stuck at fetching a repo: git://anongit.freedesktop.org/git/virglrenderer. Effectively it seems not available. How it can be possible that a branch updated 7 days ago refer to a repo not reachable?
ahussain has joined #yocto
simonl has quit [Ping timeout: 260 seconds]
ehussain has quit [Ping timeout: 248 seconds]
ahussain is now known as ehussain
simonl has joined #yocto
<Guest0118>
Perhaps it is kept this way to avoid invalidating existing sstate-caches. Or maybe it's a low priority fix since it is easily fixable by adding a MIRROR. But I'm just guessing.
vvn has quit [Ping timeout: 265 seconds]
vvn has joined #yocto
leon-anavi has joined #yocto
simonl has quit [Ping timeout: 252 seconds]
<rburton>
Guest88: because the AB has a download mirror, and that site being down appears to be a recent thing.
<rburton>
Guest88: the proper fix would be to change the url to gitlab.freedesktop.org, trivially fixed
alperak has joined #yocto
mbulut has joined #yocto
Guest9 has joined #yocto
<Guest9>
Hello!! am trying to add a dtbo to yocto kirkstone build but have been unable to do so so far despite everything i tried from online suggestions, any advice welcome
<Guest9>
the device am working on is a nvidia xavier nx som on top of a custom board, need the dtbo to describe the 40m pins header and most notably the SPI interface to reach a IIO sensor (ICM 20689)
<rburton>
Guest88: confirmed "cgit has been down intermittently for a couple of weeks. It needs manual fixing each time." sending a patch to switch to gitlab would be great
<Guest88>
rburton thanks! yes I have fix it in that way, but the question is more in general, if it is common that a officila release, not so old like kirkstone, could be broken.
<rburton>
no, the repo should be in the source mirror, so if you're using poky then it should fetch that instead (after the git server times out)
<Guest88>
ah! no it tell me search on MIRRORS but than fails. By the way I can submit the patch if is useful.
<rburton>
so share your fetch log where it fails the mirror fetch if you want, because it should have worked
<Guest88>
OK! so strange that the build hang for 3h without check on mirrors
<Guest88>
ok I check the log, thanks
<RP>
Guest88: the hang might be the clue, it perhaps didn't get as far as timing out and hitting the mirror
<rburton>
Guest88: backporting the gitlab change would be the best thing to do, as it looks like the old infra is falling apart at fdo
<RP>
rburton: I wonder if we're missing a timeout somewhere though
<Guest9>
Hello!! am trying to add a dtbo to yocto kirkstone build but have been unable to do so so far despite everything i tried from online suggestions, any advice welcome
<Guest9>
the device am working on is a nvidia xavier nx som on top of a custom board, need the dtbo to describe the 40m pins header and most notably the SPI interface to reach a IIO sensor (ICM 20689)
viric has joined #yocto
<LetoThe2nd>
Guest9: then probably a good start would be to show what you did, respectively describe *exactly* what you want to do. Do you need it on the boot partition? Just on the root filesystem? In which ways does your approach fail?
<Guest9>
LetoThe2nd: The dtbo file is a fragment that I need to add to the dts, so far I have created the following recipe to replace the existing dtbo file in the tegra-binaries layer:
<Guest9>
the aforementioned hdr.tgz contains the following:
<LetoThe2nd>
Guest9: first, please no big pastes in here.
<LetoThe2nd>
Guest9: and second, a dtbo technically is a separate device tree blob that is loaded at some later point in time. but "apply to the dts" sounds like you want to patch it. that's already two very different things.
<Guest9>
LetoThe2nd: maybe I didn't use the proper terminology, apologies
<Guest9>
the purpose of this is to create a dtbo file that describes the 40,pins header
<LetoThe2nd>
Guest9: it's not about apologising, it's about understanding what you actually want to do. because if you ask the wrong questions, then everybody including google will also give you the wrong answers, e.g. those that do not work for you.
<Guest9>
ok, so with the proper question: how to load a modified dtbo file ?
Jones42_ has joined #yocto
alessioigor has quit [Ping timeout: 256 seconds]
Jones42__ has quit [Read error: Connection reset by peer]
<Guest9>
the first recipe I mentioned has the purpose to replace the existing dtbo file in the kernel with one of the same name but a different content
<Guest9>
then added the following lines to local.conf
<viric>
Is it normal that changing DISTRO_FEATURES then rebuilds fail?
<Guest9>
LetoThe2nd all of the above fails
<LetoThe2nd>
viric: if you change something to a breaking state, then it fails...
<viric>
For example, if I do: DISTRO_FEATURES:append=" debuginfo", then I have problems rebuilding things like ca-certificates, because there are /usr/share/licenses/ files not in FILES
<viric>
But if I do -c cleanall ca-certificates, it seems then it works
<viric>
It's like some missing dependency thing failing?
<viric>
So I want to append "debuginfo" and then rebuild the image again.
<viric>
and many recipes fail with that /usr/share/licenses/ leftover
<LetoThe2nd>
Guest9: yeah, but fails "how"? so does that modified file of your tgz end up in the image?
<LetoThe2nd>
viric: ah ok that's what you mean. Well I would just delete tmp, and rebuild.
<viric>
but that's a bug, isn't it?
<viric>
I should never need to delete t mp
<LetoThe2nd>
Guest9: after looking at the reference manual, I can't find any trace of KERNEL_DEVICETREE_APPLY_OVERLAYS nor OVERLAY_DTB_FILE, hence those either somehow BSP specific probably
<LetoThe2nd>
viric: hmmm not so sure here. I would actually word it the other way round: if you have concerns about deleting tmp then that's a bug, because everything in tmp should be transient and completely reproducible.
<viric>
But it takes ages to build all
<viric>
I only want to save me the time of rebuilding
<LetoThe2nd>
viric: nope, a handful of minutes, if even.
<Guest9>
LetoThe2nd: Understood, so how would you go about loading a dtbo to the device in kirkstone?
<viric>
when you say "delete tmp", then what tmp are you referring to?
<LetoThe2nd>
viric: what you're probably thinking of is sstate, and yes, THAT ONE you should never delete.
<LetoThe2nd>
viric: well I'm referring to the tmp directory in your build directory.
<viric>
that holds all build outputs, isn't it?
<LetoThe2nd>
Guest9: first I would make sure that it actually ends up where I want it, second I would find out what the mechanism for loading is in my specific boards case. if I'm using something u-boot for example, it might be the one loading it. on some board configurations, the kernel allows run-time loading.
<viric>
If I delete that, I have to recompile everything
<LetoThe2nd>
Guest9: I have no experience on the Nvidia stuff, so I can't answer it for you, I can just help you find out where you might be going/thinking wrong.
<LetoThe2nd>
viric: sstate holds the build output. tmp is completely transient.
enok has quit [Ping timeout: 255 seconds]
<viric>
ah. I'll try
Guest9 has quit [Quit: Client closed]
Guest9 has joined #yocto
<LetoThe2nd>
viric: so I repeat: deleting tmp is common. deleting sstate is not. actually, it is very usual to not even have sstate in the build, but to put it into a separate place and share it across many builds.
<Guest9>
LetoThe2nd: in older versions of yoocto (dunfell) the dtbo was added to /boot/ directory, in kirkstone, the /boot/ directory only contains the efi directory, nothing else.
<viric>
ah
Guest88 has quit [Ping timeout: 256 seconds]
<viric>
is sstate always growing in size? How does drop data from there not reqwuired anymore?
<LetoThe2nd>
viric: it always grows until you manually remove things.
<viric>
LetoThe2nd: like by date with 'find ... -delete' for example?
<LetoThe2nd>
Guest9: ok that sounds like they have changed the boot strategy somehow.
<Guest9>
it is now efi based which it was not before
<LetoThe2nd>
viric: yup
<LetoThe2nd>
Guest9: I see. The question now is, what comes after efi, and can it load & apply the dtbo
<LetoThe2nd>
Guest9: my suggestion would be to ask the OE4T people. it sounds Nvidia specific, and that's their home turf. or alternatively, your board vendor.
<viric>
is it usual to track local.conf with git?
<viric>
or every developer should have its own local.conf?
<LetoThe2nd>
viric: local.conf is not tracked. you should manage its setup, for example with the new layer setup tooling by kanavin_, or if you're on older releases you might want to look into kas for example.
<viric>
thank you
<viric>
Is there any way I don't produce ptest packages for all recipes? Or dbg packages for all recipes? To limit that a bit.
<viric>
It says: The first is the “setscene” stage. ... If nothing exists, the second phase, which is the setscene stage, ...
<viric>
Both named setscene?
filipm has quit [Quit: Client closed]
sakoman has joined #yocto
<viric>
does debuginfod conflict with rm_work ?
sa7mfo_ has joined #yocto
<sa7mfo_>
Hello, If I build two images, one for the "main-rootfs", and one image that includes a kernel+initramfs. How do I integrate that kernel+initramfs image into /boot of my "main-rootfs" image?
alessioigor has quit [Ping timeout: 256 seconds]
alessioigor has joined #yocto
MrCryo has joined #yocto
sakoman has quit [Ping timeout: 260 seconds]
ehussain has quit [Ping timeout: 260 seconds]
ehussain has joined #yocto
Guest0118 has quit [Ping timeout: 256 seconds]
ehussain has quit [Read error: Connection reset by peer]
sakoman has joined #yocto
florian has joined #yocto
mvlad has joined #yocto
<rburton>
viric: no
<viric>
rburton: ok
florian has quit [Ping timeout: 248 seconds]
lexano has joined #yocto
Guest9 has quit [Quit: Client closed]
xmn has joined #yocto
rob_w has quit [Remote host closed the connection]
Xagen has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
Jones42_ has quit [Read error: Connection reset by peer]
alessioigor has quit [Remote host closed the connection]
alessioigor has joined #yocto
<viric>
maybe ptest and rm_work collide? It's like it's not deleting anything at all from tmp, with ptest
Daanct12 has quit [Quit: WeeChat 4.3.5]
<viric>
maybe rm_work && CACHE= are incompatible? I can't find what made rm_work not work
goliath has quit [Quit: SIGSEGV]
tleb has quit [Remote host closed the connection]
sfo has quit [Remote host closed the connection]
tostr has quit [Remote host closed the connection]
raghavgururajan has quit [Remote host closed the connection]
jonesv has quit [Remote host closed the connection]
tleb has joined #yocto
tostr has joined #yocto
raghavgururajan has joined #yocto
jonesv has joined #yocto
sfo has joined #yocto
jmiehe has joined #yocto
Jones42 has joined #yocto
goliath has joined #yocto
Xagen has joined #yocto
<rburton>
viric: can you define 'not work'?
<viric>
rburton: hm I think it was wrong perception... I saw big directories under 'work' but it's effectively because I had ctrl-c'ed their build. So they were in some intermediate state.
alessioigor has quit [Remote host closed the connection]
heidaren has quit [Quit: Bye ~]
<rburton>
yeah rm_work only runs if a recipe finishes the build task successfully
<viric>
mckoan: ah remove duplicated, for cache management. nice.
prabhakalad has quit [Ping timeout: 260 seconds]
BrianL has joined #yocto
mbulut has quit [Ping timeout: 260 seconds]
BrianL is now known as blilly
jsbronder has quit [Remote host closed the connection]
jsbronder has joined #yocto
alperak has quit [Quit: Connection closed for inactivity]
amitk_ has quit [Ping timeout: 260 seconds]
<landgraf>
what can be the reason of missing "recipe-sysroot/etc/passwd" file during do_package_write_rpm task execution? It causes "Exception: FileNotFoundError: [Errno 2] No such file or directory: ". Not 100% reproducible (I'd say not even 1%)
mjm has quit [Quit: mjm]
mckoan is now known as mckoan|away
jmd has joined #yocto
leon-anavi has quit [Quit: Leaving]
<viric>
oh I'm testing the debuginfod but it fails to provide the source to packages that are in devtool workspace
Minvera has joined #yocto
zpfvo has quit [Remote host closed the connection]
<rburton>
how are you running the debuginfod? the oe-debuginfod script only looks at deployed packages, so are you actually doing a full build of the recipes or just the compile step?
<viric>
I run "bitbake package"
<viric>
that creates the rpm
<viric>
but the debug info has file locations relative to my computer, not relative to /usr...
<viric>
I don't understand how debuginfod translates relative directories to absolute directories
<viric>
So when I attach gdb, for my binary it fails to find /home/myname/git/..../workspace/sources/package/..../main.cpp
<rburton>
looks like the path remapping needs to be extended
<viric>
what is the path remapping?
<rburton>
see DEBUG_PREFIX_MAP
<viric>
is that something I can do in local.conf?
<rburton>
append to it, sure
<viric>
I see.
<viric>
It'd be nicer if there was an upstream change that covered this for workspaces / externalsrc / etc.
<rburton>
sure, you're probab;y hitting a bad interaction between externalsrc and the remapping, a patch would be great if you root cause it
Chaser has joined #yocto
Chaser has quit [Client Quit]
rob_w has joined #yocto
rfuentess has quit [Remote host closed the connection]
amitk has joined #yocto
blilly has quit [Quit: Client closed]
prabhakalad has joined #yocto
goliath has quit [Quit: SIGSEGV]
rjones2 has joined #yocto
<rjones2>
Hello, does anyone know if SPDX 3.0 is now the default for scarthgap or only master?
Ad0 has quit [Ping timeout: 252 seconds]
<tortoise>
I'm having some trouble with building a rust project in yocto. The project has both crate.io and git dependencies. It's the git dependencies that seem to be causing the issue. Where I'm at now it appears that during do_compile the cargo command finds the git repos that I downloaded with additional SRC_URI's but it fails when it goes to build them. I don't really know 100% whats going on, but I have suspicions
<tortoise>
Although the error seems to indicate that it's unable to update because the Cargo.toml of the git dependecy is a virtual manifest, I think it might actually be due to no network access during do_compile
<tortoise>
I'm wondering if anyone has considered modifying cargo.bbclass or cargo_common.bbclass so that do_fetch calls cargo vendor to bring in all the dependencies
<tortoise>
I think this where things go off the rails, I might be able to fix this and submit a patch upstream
mulk has joined #yocto
<rburton>
rjones2: to be sure, master only as it only just merged
Vonter has quit [Ping timeout: 245 seconds]
Vonter has joined #yocto
cabazon50 has joined #yocto
cabazon50 has quit [Quit: Client closed]
<tlwoerner>
moto-timo: is "inherit pypi" not considered best practice anymore?
<tlwoerner>
is "inherit python_setuptools_build_meta" better
<tlwoerner>
are those two incompatible? complimentary?
Haxxa has quit [Quit: Haxxa flies away.]
Haxxa has joined #yocto
alessioigor has quit [Remote host closed the connection]
mvlad has quit [Remote host closed the connection]
jmd has quit [Remote host closed the connection]
rob_w has quit [Read error: Connection reset by peer]
<moto-timo>
complementary
<moto-timo>
look at the bbclasses, they do entirely different things
<moto-timo>
'inherit setuptools3' is in general no longer best practice
<moto-timo>
there might be a WORKDIR vs UNPACKDIR issue in pypi.bbclass. I have a local patch under test, but I didn't convince myself I was correct yet.
amitk has quit [Ping timeout: 252 seconds]
<rburton>
pypi is _just_ a way of sourcing the sdists
Xagen has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]