LetoThe2nd changed the topic of #yocto to: Welcome to the Yocto Project | Learn more: https://www.yoctoproject.org | Community: https://www.yoctoproject.org/community | IRC logs: http://irc.yoctoproject.org/irc/ | Having difficulty on the list, with someone on the list or on IRC, contact Yocto Project Community Manager Letothe2nd | CoC: https://www.yoctoproject.org/community/code-of-conduct
simonew has joined #yocto
enok71 has quit [Ping timeout: 260 seconds]
lexano has quit [Ping timeout: 255 seconds]
<merit> bleh
<merit> still not really sure on how to _use_ those patches once they made it into the sysroot, but they're there
florian_kc has quit [Ping timeout: 240 seconds]
qschulz has quit [Remote host closed the connection]
qschulz has joined #yocto
davidinux has quit [Ping timeout: 264 seconds]
davidinux has joined #yocto
starblue has quit [Ping timeout: 252 seconds]
starblue has joined #yocto
jclsn has quit [Ping timeout: 255 seconds]
jclsn has joined #yocto
Vonter has joined #yocto
Net147 has quit [Ping timeout: 264 seconds]
<merit> https://ctxt.io/2/AACItSZfFg < this feels really close
<merit> got a few typos, and I think it's actually right
<merit> well.. I think it works
germ has joined #yocto
germ has quit [Changing host]
germ has joined #yocto
germ has quit [Client Quit]
ablu has quit [Read error: Connection reset by peer]
germ has joined #yocto
germ has quit [Changing host]
germ has joined #yocto
ablu has joined #yocto
pedrowiski has quit [Remote host closed the connection]
<merit> wait.. no it's not
Net147 has joined #yocto
Net147 has quit [Changing host]
Net147 has joined #yocto
wmills_ has quit [Ping timeout: 256 seconds]
wmills_ has joined #yocto
<merit> the "recipe-sysroot" is empty unless I build without the class applied, first
sev99 has quit [Ping timeout: 250 seconds]
Xagen has quit [Quit: Textual IRC Client: www.textualapp.com]
<merit> https://ctxt.io/2/AACIoQUKFg < okay... gotta do it in an afters for the recipe sysroot populate task
simonew has quit [Remote host closed the connection]
simonew has joined #yocto
xmn has quit [Ping timeout: 264 seconds]
simonew has quit [Ping timeout: 260 seconds]
Chaser has joined #yocto
Chaser has quit [Remote host closed the connection]
thomas_34 has joined #yocto
roussinm has quit [Quit: WeeChat 3.3-dev]
wmills__ has joined #yocto
TundraMan has joined #yocto
rfs613- has joined #yocto
shoragan_ has joined #yocto
wmills_ has quit [Remote host closed the connection]
marka has quit [Quit: ZNC 1.8.2 - https://znc.in]
rfs613 has quit [Quit: restart]
ssweeny has quit [Ping timeout: 268 seconds]
ssweeny has joined #yocto
shoragan has quit [Ping timeout: 268 seconds]
xantoz has quit [Ping timeout: 268 seconds]
xantoz has joined #yocto
jmd has joined #yocto
<thomas_34> Hi guys, I have a hard time to understand how certain terms are defined. For example, what is a target? What is a machine and is it different from a "machine configuration"?
tlwoerner has quit [Ping timeout: 256 seconds]
<thomas_34> https://docs.yoctoproject.org/dev-manual/building.html?highlight=multiconfig#building-a-simple-image Chapter 4: "The target is the name of the recipe you want to build."
<thomas_34> https://docs.yoctoproject.org/dev-manual/building.html?highlight=multiconfig#building-images-for-multiple-targets-using-multiple-configurations : "You can use a single bitbake command to build multiple images or packages for different targets where each image or package requires a different configuration (multiple configuration builds)."
<thomas_34> I'm confused here. "Target" in the first definition means cleary something different compared to the second sentence.
<ablu> thomas_34: In the first example "target" references to the "bitbake <target>" above.
<thomas_34> ablu, ahh okay. So when I want to communicate with "yocto-people", what do they understand under the term "target"?
<ablu> thomas_34: Typically it is your build result. What that is probably depends on your context.
<thomas_34> Okay, so its not the "piece of hardware", which I build software with yocto/bitbake?
<ablu> I would call it "piece of software built for a particular configuration (hardware, tunable, ...)". But I am not sure whether one will be able to find a definition that fits everything.
<thomas_34> Ok, thank you ablu for your definition
kpo has quit [Ping timeout: 246 seconds]
dlan has quit [Ping timeout: 256 seconds]
dlan has joined #yocto
zpfvo has joined #yocto
linfax has joined #yocto
ray-san has joined #yocto
<thomas_34> Is there a difference in terms of definition between "machine" and "machine configuration"? Or is machine just the abbreviation for machine configuration?
<michaelo> Hi rburton ! This sounds like very cool. Title + SHA is fine. You can send this to the docs@ mailing list, or directly to me if you prefer. Did you figure out "manually" whether each commit impacts the documentation or not?
mckoan|away is now known as mckoan
frieder has joined #yocto
jmd has quit [Remote host closed the connection]
jmd has joined #yocto
enok71 has joined #yocto
jmd has quit [Remote host closed the connection]
rob_w has joined #yocto
ak77 has quit [Read error: Connection reset by peer]
<mckoan> thomas_34: MACHINE is a variable defining the file containing machine configuration
leon-anavi has joined #yocto
ak77 has joined #yocto
Kubu_work has joined #yocto
enok71 has quit [Ping timeout: 268 seconds]
prabhakalad has quit [Ping timeout: 252 seconds]
prabhakalad has joined #yocto
luc4 has joined #yocto
luc4 has quit [Client Quit]
enok71 has joined #yocto
florian has joined #yocto
mbulut has joined #yocto
mbulut has quit [Client Quit]
Guest40 has joined #yocto
<Ad0> where is the location of the config file used in kernel configure? it is copied somewhere right?
<Ad0> I have a working defconfig, but I want to change it into fragments
<Guest40> Good all, I am trying to create a yocto, which has two recipes, one "mount-points" which does the mount of the disks, and in the other one I want to create a directory and add a file to that directory "utilities.bb", in "utilities.bb" I have specified with RDEPEND that "mount-points" must be executed first so there is no conflict with
<Guest40> "utilities.bb" but I am not able to create the directory after the mount.
<Guest40> I have tried to create the directory in / instead of in the disk I am mounting and it has worked without problem so I understand that the recipe works fine and the problem is that it will try to mount later and it will step on the directory that is created with "utilities.bb".
<Guest40> thanks for the help.
<Guest40> Translated with DeepL.com (free version)
speeder has joined #yocto
<Ad0> that's in the layer, not the working dir
<Ad0> my idea is to remove the defconfig from my layer and let it use the vanilla config, then replace the config in the working directory so it will be created as a fragment when done
<Ad0> easier to see what's different so I can remove the shit I don't need
<landgraf> Ad0: The document I linked covered this task iirc
<Ad0> I don't think it covers the internals like where the config ends up in the working tree
<Ad0> I need to find that path and replace the file
<landgraf> Ad0: iiuc you want to have vanila defconfig + fragments in your layer, right?
<Ad0> yes, but I already have a "complete" modified defconfig in my layer already. when I start the task it will use that file . but I will temporarily remove the file and rather replace it in the working tree so a fragment is created
<Ad0> with several devices with different defconfigs + kernel upgrades I think fragments would be easier
<Ad0> instead of full defconfigs per device
<Ad0> also it will remove some bbappend scripting
<landgraf> Ad0: just run find tmp* -name .config in your builddir and you'll find it
<Ad0> yeah from memory last time I checked there were several
Vonter has quit [Ping timeout: 260 seconds]
<landgraf> after -c menuconfig of course
Vonter has joined #yocto
<landgraf> Ad0: I think you're looking for work-shared/*/kernel-build-artifacts/.config
<Ad0> could be
<Ad0> I will just try a -c menuconfig and diff the files till I find one that matches
<landgraf> there're several but not all are related to kernel.
<Ad0> thanks! I think the path you have there is correct
<enok71> Anyone who knows how kernel-fitimage.class works and how to make it create config sections that u-boot can actually use to add a bunch of .dtbo overlay images included in the fitimage file to the included .dtb image at boot time, as described in the u-boot documentation? It looks to me like those config sections emitted by kernel-fitimage are not entirely useful for u-boot. I'm using the latest master HEAD from the poky git repo.
<Ad0> landgraf, does it generate fragments for me based on the diff from menuconfig?
<Ad0> the article you gave me seems like i have to manually make my own
<Ad0> maybe I didn't read well enough
<Ad0> bitbake -c diffconfig virtual/kernel
<enok71> Or perhaps nobody uses dtbo in this way and instead creates multiple complete .dtb files and no .dtbos? A waste of space though, as 99% of such dtbs will be identical.
<rburton> merit: do the apply_sysroot_patches as a postfunc on do_patch. because patch runs before configure and DEPENDS is short-hand for do_configure[depends] you'll need to add the dependency on the other recipe explicitly.
<Ad0> "Could not locate BSP definition for my-machine/standard and no defconfig was provided" but it is based on raspberry pi tho.
<Ad0> probably because I set MACHINEOVERRIDES
<rburton> by default the yocto-kernel framework uses MACHINE as KMACHINE
<rburton> if you're not providing a bsp definition then make sure you set KMACHINE to the right one
<Ad0> thanks
<Ad0> that var is nowhere in meta-raspberrypi tho
<Ad0> something else sets that maybe
<rburton> as i said, default is MACHINE
<Ad0> thanks a lot
<Ad0> so you can set a different machine for kernel
<Ad0> it worked setting it to it but I guess I have to find the right name ... Could not locate BSP definition for raspberrypi3/standard and no defconfig was provided
<rburton> have you created a new machine that pulls in the rpi machine?
<Ad0> no, just require conf/machine/include/arm/armv7a/tune-cortexa7.inc and include conf/machine/include/rpi-base.inc
<rburton> meta-raspberrypi doesn't even use linux-yocto, so you either need to use the meta-rpi kernel, or provide your own config
enok71 has quit [Remote host closed the connection]
<Ad0> ok
<Ad0> I will just copy the defconfig from the linux-raspberry pi and use that as a baseline. what sucks is that I have to think about updating that too
<rburton> the rpi kernel pulls the rpi fork of linux which has a defconfig in
<Ad0> yeah
<rburton> use that :)
<Ad0> linux-raspberrypi/1_5.15.92+gitAUTOINC+509f4b9d68_14b35093ca-r0/defconfig
<Ad0> so if I use git pull on that layer I might have to update it
<rburton> recipes-kernel/linux/linux-raspberrypi.inc:KBUILD_DEFCONFIG:raspberrypi4 ?= "bcm2711_defconfig"
<rburton> if you use their kernel tree then you can just use the defconfig it provides
<Ad0> maybe I can set the KBUILD_DEFCONFIG in my bbappend for my machines then?
<rburton> if you have rpi hardware then i don't know why you don't just use the rpi kernel in meta-rpi, but yes you could recreate the recipe
<Ad0> I seen that name but not gotten into it (meta-rpi) , not sure what the diff is between that and meta-raspberrypi
<rburton> i was being lazy, meta-raspberrypi is what i meant
<Ad0> yeah I have seen some alternate layer for meta-rasperrypi
<Ad0> I use the kernel there but
<Ad0> setting KBUILD_DEFCONFIG in my bbappend worked
speeder_ has joined #yocto
speeder has quit [Ping timeout: 268 seconds]
enok71 has joined #yocto
vladest1 has joined #yocto
wooosaiiii1 has joined #yocto
speeder__ has joined #yocto
ablu_ has joined #yocto
<Guest40> Good all, I am trying to create a yocto, which has two recipes, one "mount-points" which does the mount of the disks, and in the other one I want to create a directory and add a file to that directory "utilities.bb", in "utilities.bb" I have specified with RDEPEND that "mount-points" must be executed first so there is no conflict with
<Guest40> "utilities.bb" but I am not able to create the directory after the mount.
<Guest40> I have tried to create the directory in / instead of in the disk I am mounting and it has worked without problem so I understand that the recipe works fine and the problem is that it will try to mount later and it will step on the directory that is created with "utilities.bb".
Xywzel_ has joined #yocto
ablu has quit [Ping timeout: 252 seconds]
vladest has quit [Ping timeout: 252 seconds]
wooosaiiii has quit [Ping timeout: 252 seconds]
Xywzel has quit [Ping timeout: 252 seconds]
vladest1 is now known as vladest
svuorela_ has quit [Ping timeout: 252 seconds]
derRichard has quit [Ping timeout: 252 seconds]
<rburton> Guest40: mounts happen on boot, but recipes write files at build time. what are you _actually_ trying to do?
wooosaiiii1 is now known as wooosaiiii
derRicha1d has joined #yocto
speeder_ has quit [Ping timeout: 268 seconds]
svuorela has joined #yocto
<Ad0> thanks rburton
speeder_ has joined #yocto
ablu_ is now known as ablu
speeder__ has quit [Ping timeout: 264 seconds]
<Guest40> rburton i'm trying to mount sda2 on /disk2 and then create a directory on that disk, /disk2/directory1, I mount with one recipe and make the directory with another, and the one that makes de directory has the RDEPENDS of mount
starblue has quit [Ping timeout: 264 seconds]
starblue has joined #yocto
<rburton> Guest40: think about the operation order. mount points happen on the target during boot. if you want to split a filesystem up into partitions then wic can do that, look up the --exclude-path option
<Guest40> rburton: ok , i will check it, thankyou
thomas_34 has quit [Quit: Client closed]
Vonter has quit [Ping timeout: 264 seconds]
Vonter has joined #yocto
plat51 has joined #yocto
lexano has joined #yocto
alessioigor has joined #yocto
plat51 has quit [Quit: Client closed]
Guest40 has quit [Quit: Client closed]
alessioigor has quit [Remote host closed the connection]
thomas_34 has joined #yocto
<thomas_34> If I define OE_CMAKE = "foobar" in my recipe.bb and inherite the "base" class, should then show up the string "foobar" somewhere in run.do_compile file of that recipe?
rfs613- is now known as rfs613
rfuentess has joined #yocto
Michael_Guest has joined #yocto
<Ad0> how do I verify that a config fragmet made it into the kernel config ? bitbake virtual/kernel -c kernel_configcheck -f returns nothing
<Ad0> also the fragment made from pasting my kernel config over the original have much of the same variables with the same settings as the original Oo
joekale has joined #yocto
speeder__ has joined #yocto
speeder_ has quit [Ping timeout: 268 seconds]
thomas_34 has quit [Quit: Client closed]
speeder_ has joined #yocto
speeder__ has quit [Ping timeout: 240 seconds]
alessioigor has joined #yocto
paw has joined #yocto
speeder__ has joined #yocto
thomas_34 has joined #yocto
speeder_ has quit [Ping timeout: 255 seconds]
pedrowiski has joined #yocto
pedrowiski has quit [Ping timeout: 252 seconds]
<rburton> Ad0: bitbake virtual/kernel -c showconfig will tell you the path of the .config file that was actually used
<Ad0> thanks
<rburton> (or just look in the build tree)
tlwoerner has joined #yocto
Guest84 has joined #yocto
<Guest84> Hello
<Guest84> Is it possible to downgrade python3 to 3.8 in kirkstone?
<rburton> not easily
enok71 has quit [Quit: enok71]
enok71 has joined #yocto
<Guest84> I was thinking checking out gatesharth and stealing their python3 recipe, add a PREFERRED_VERSION in local.conf
<Guest84> Now problem is recipes syntax
rob_w has quit [Remote host closed the connection]
<Guest84> There is a script which can try and take care of that but python3 is a massive set of recipes
<rburton> why do you need to downgrade?
<Guest84> would that work?
<rburton> python3 is one recipe
<Guest84> The python3 directory is massive
<rburton> sure, but the recipe is just python3_*.bb
<paulbarker> Hey folks, here's something I've not thought about before but came up in discussions today - should we be providing an older linux-libc-headers version if our BSP provides a kernel version older than the one in oe-core?
<Guest84> Nothing else to change but the python3_blah.bb file?
<paulbarker> E.g. Scarthgap will have linux-libc-headers 6.6, but our BSP will use the 6.1 CIP kernel by default
enok71 has quit [Ping timeout: 268 seconds]
<Guest84> Let me try that and I’ll come back to you
<Guest84> rburton: Thanks
<paulbarker> https://docs.kernel.org/kbuild/headers_install.html explicitly says that "Kernel headers are backwards compatible, but not forwards compatible."
<rburton> zeddii: ^^^
<paulbarker> My reading of that is that we should be providing the older linux-libc-headers version in our BSP and using it when we build an older kernel
<zeddii> it's been stated that way for years. in practice, we've hit a grand total of zero issues.
<zeddii> I'd be strongly against any changes in core to the headers, but people can do whatever they want in their own layers.
<paulbarker> zeddii: We definitely don't need any changes in core.
<zeddii> my opinion stands on any layer, but I just obviously have no input on them.
<paulbarker> I expect issues would occur if someone used flashy new kernel features that weren't present in our older BSP kernel, but adoption of new APIs/etc is usually slow
<zeddii> if you change the libc-headers, you are then repsponsible for the complete testing of all your applications and there's not much OEcore or any maintenance can do if you run into an issues.
<zeddii> paulbarker: and that's a pathalogically evil program that is going directly at kernel headers and structures. if something goes through libc, they are well protected.
<zeddii> this is very well trodden ground.
<zeddii> (the libc-headers question)
<zeddii> I answered it about 13 years ago at Wind River, and the answer still hasn't changed.
<paulbarker> zeddii: From that I imply that Go is pathologically evil, and I don't disagree haha
<zeddii> indeed!
sev99 has joined #yocto
<RP> paulbarker: keep in mind that linux-libc-headers works in conjunction with glibc and the toolchain to decide what to do. glibc has the notion of "OLDEST_KERNEL" and the "guarantees" are also from components like glibc.
<RP> paulbarker: you really don't want or need multiple toolchains and libc-headers versions around
<RP> paulbarker: as zeddii says, whilst you may not have considered it, it has been given a lot of thought and real world testing over the last decade+
<paulbarker> So, the conclusion would be - libc provides enough of a buffer to prevent issues
<zeddii> the only class of applications that even have a chance of running into issues are exchanging structures with the kernel AND if the kernel for some reason decided to re-order a structure. The userspace ABI promise is in play, regardless of what libc-headers says.
<zeddii> if something is that tied to the kernel, you'd hope that same application is checking the version, etc.
<jdiez> when modifying (with `devtool modify`) a build tool that is required by another recipe, is it enough to do `devtool build <buildtool>`? the changes I made don't seem to be visible when I build the recipe that depends on this build tool
<paulbarker> Thanks zeddii and RP, I appreciate the input. Will discuss with the rest of our team
<Ad0> lol got tired of network manager's BS and disabled ipv6
<rburton> RP: so i ran all of the ptests against poky-altcfg. all passed... except valgrind which explodes for god knows what reasons
<vmeson> rburton: you called my son...
jmd has joined #yocto
<rburton> the gdb tests specifically fail dramatically
<vmeson> rburton: is this just master or are you up to something interesting?
thomas_34 has quit [Ping timeout: 250 seconds]
Michael_Guest has quit [Quit: Client closed]
Guest84 has quit [Quit: Client closed]
<jdiez> doesn't look like `devtool build <devtool>` is sufficient to use the new build tool in my dependent recipe. Also, using `devtool modify` and `devtool finish` on an already-patched recipe seems to only write out the most recent patch, overwriting previous ones :(
<RP> rburton: are you saying we should expand the test matrix? :)
<RP> rburton: and is that more or less useful than musl ptest?
Guest84 has joined #yocto
Guest84 has quit [Client Quit]
<rburton> vmeson: master with poky-altcfg
<rburton> RP: i think its fair that it actually works before expanding the matrix
<vmeson> rburton: I'll take a look when I get some time.
paw has quit [Ping timeout: 252 seconds]
Saur75 has quit [Quit: Client closed]
<RP> rburton: we also don't really have the arm builder capacity to do this on arm
<rburton> i don't mind what host its on
Saur75 has joined #yocto
<RP> x86 is obviously easier
<RP> rburton: might be an idea to put the AB config in place so we can enable when ready? We should do this for musl too?
<rburton> yeah
mckoan is now known as mckoan|away
linfax has quit [Ping timeout: 268 seconds]
<gmorell> /w 2
florian has quit [Quit: Ex-Chat]
enok71 has joined #yocto
goliath has quit [Quit: SIGSEGV]
ray-san has quit [Ping timeout: 268 seconds]
<paulbarker> zeddii: We see no warnings building Linux 5.10 with the latest master, despite OLDEST_KERNEL being set to 5.15
<paulbarker> https://git.yoctoproject.org/poky/tree/meta/classes-recipe/kernel.bbclass#n647 is: check_oldest_kernel[vardepsexclude] += "OLDEST_KERNEL KERNEL_VERSION"
<paulbarker> Doesn't that stop the check function from re-running if KERNEL_VERSION is changed?
Guest40 has joined #yocto
<Guest40> hi , i want to add file to a data partition, not to root. How can i achive this? i've got the partition created with wic and .wks, but don't know how to add the files and directories i need
xmn has joined #yocto
speeder_ has joined #yocto
leon-anavi has quit [Remote host closed the connection]
speeder__ has quit [Ping timeout: 252 seconds]
Saur75 has quit [Quit: Client closed]
enok71 has quit [Ping timeout: 260 seconds]
Saur75 has joined #yocto
starblue has quit [Ping timeout: 252 seconds]
speeder__ has joined #yocto
Kubu_work has quit [Quit: Leaving.]
<olani> jdiez: Did you add the build tool to DEPENDS of your modified recipe?
speeder_ has quit [Ping timeout: 252 seconds]
<jdiez> olani: I was modifying the recipe/source code of a build tool that was in the DEPENDS of another recipe. Due to it being a build tool, I needed to build the -native version of the package, which I could not do with `devtool`
wmills__ has quit [Remote host closed the connection]
wmills__ has joined #yocto
<rburton> that should work
clever has quit [Ping timeout: 260 seconds]
speeder_ has joined #yocto
florian has joined #yocto
clever has joined #yocto
speeder__ has quit [Ping timeout: 272 seconds]
enok71 has joined #yocto
Saur75 has quit [Quit: Client closed]
Xagen has joined #yocto
Saur75 has joined #yocto
zpfvo has quit [Remote host closed the connection]
speeder_ has quit [Ping timeout: 240 seconds]
rfuentess has quit [Remote host closed the connection]
|Xagen has joined #yocto
Xagen has quit [Ping timeout: 255 seconds]
Guest40 has quit [Quit: Client closed]
florian has quit [Ping timeout: 240 seconds]
pbiel has joined #yocto
<pbiel> I'm a bit confused abut how should I approach the following problem in Yocto.
<pbiel> For example I have a machine and I would like to achieve the following:
<pbiel> on the other hand for the same machine for a production build I would like to add some patches to u-boot that remove all the unnecessary u-boot commands and leave only bootm command, the same goes for the image type I would like to use for the production build - fitImage and what's more signed
<pbiel> for development builds I would like to use extlinux configuration and store let's say 10 different device tree blobs on /boot partition to allow switching between them on boot for the purpose of some tests
<pbiel> Basically this can be applied to more use cases, not only u-boot.
<pbiel> Whenever I searched for the answer to this question, most answers concerned the image generation where the solution is simple - just create two image variants: *dev, *production, and install appropriate packages accordingly. But this cannot be applied as things like kernel device tree, uboot config and so on are configured in machine file.
<pbiel> What is the canonical solution for such problems?
pedrowiski has joined #yocto
pedrowiski has quit [Changing host]
pedrowiski has joined #yocto
bielpa_ has joined #yocto
enok71 has quit [Ping timeout: 255 seconds]
florian has joined #yocto
pedrowiski has quit [Ping timeout: 252 seconds]
<khem> Linux 6.8 ships with Ubuntu 24.04 when it is released next month, so I guess its also going to be a lts like ?
<rburton> who knows :)
<khem> yeah
<khem> pbiel: I would say, define new distro_features dev and prod and make the needed changes under those since the changes are not only in recipe space but also global namespace
kpo has joined #yocto
<pbiel> Well that makes sens actually. I'm looking at this moment at the ATF recpipe for one of NXP's boards, and in fact, in that recipe they change if DISTRO_FEATURES contains secure. If so they use UBOOT_BINARY that is preconfigured to support secure boot, otherwise they use standard u-boot config
<pbiel> they check*
mattsm has quit [Ping timeout: 255 seconds]
mattsm has joined #yocto
mattsm6 has joined #yocto
mattsm has quit [Ping timeout: 272 seconds]
mattsm has joined #yocto
mattsm6 has quit [Ping timeout: 272 seconds]
frieder has quit [Remote host closed the connection]
mattsm2 has joined #yocto
mattsm4 has joined #yocto
mattsm has quit [Ping timeout: 268 seconds]
mattsm4 is now known as mattsm
mattsm2 has quit [Ping timeout: 264 seconds]
simonew has joined #yocto
kpo has quit [Quit: Konversation terminated!]
enok71 has joined #yocto
sev99 has quit [Quit: Client closed]
joekale has quit [Remote host closed the connection]
joekale has joined #yocto
Haxxa has quit [Quit: Haxxa flies away.]
Haxxa has joined #yocto
johnmcnuggets has joined #yocto
Saur75 has quit [Quit: Client closed]
Saur75 has joined #yocto
<jonmason> pidge: is it you who is making a mess with screens and qemuarmv5? ;-)
Xagen has joined #yocto
|Xagen has quit [Ping timeout: 252 seconds]
Kubu_work has joined #yocto
starblue has joined #yocto
<RP> jonmason: it was. rburton said you were just the person to help sort it all out! :)
<jonmason> RP: just like rburton to throw me under the bus while out on vacation ;-)
<jonmason> Updated the bug, I _think_ I know the issue, just not sure if they fix is what we want to do
Zapados has joined #yocto
<RP> jonmason: he said you were due back and be eager to help :)
<jonmason> lmao
<jonmason> I must remember when he's out on vacation
<Zapados> How do I layer kas YMLs?  For example, parent.yml has some BBLAYERS ?= in it.  Then I want child.yml, header: includes: -parent.yml, then append some more BBLAYERS in there.  Any way to do that?
<jonmason> Zapados: we're doing it in meta-arm
<jonmason> there is an include method
<jonmason> one sec and i'll link you
<Zapados> jonmason I don't see that extending BBLAYERS though.
<vvn> hi there -- I have an ERROR: No recipes in default available for: <paths-to-dynamic-multimedia-layer>/libcamera_%.bbappend, but I do have meta-multimedia in my BBLAYERS. Any idea what could be the issue? Layers order or priority maybe?
<jonmason> Zapados: I thought you were generally asking how to layer them
<jonmason> you can see in base.yml how to change variables
<jonmason> that is what is included in that qemuarm.yml file
<jonmason> and we tweak it further in testimage.yml
<jonmason> and in xen.yml
vladest has quit [Quit: vladest]
<jonmason> then you just do something like "kas build ci/qemuarm.yml:ci/xen.yml:ci/testimage.yml"
<Zapados> jonmason looking at testimage.yml, what additional layers are added there?   It's likely I'm thinking about this wrong...?
<jonmason> if you want layers, xen.yml pulls in meta-virt
starblue has quit [Ping timeout: 268 seconds]
<jonmason> essentially, you are modifying the local.conf via the tag "local_conf_header:"
<Zapados> jonmason so to extend BBLAYERS I could do something like BBLAYERS:append = " test"?
<Zapados> Ahh, this was stacking:
<Zapados> bblayers_conf_header:
<Zapados>     meta-xxxbase: |
<Zapados>         LCONF_VERSION = "7"
<Zapados>         BBPATH = "${TOPDIR}"
<Zapados>         BBFILES ?= ""
<Zapados>     layer-order: |
<Zapados>         BBLAYERS ?= " \
<Zapados> Bad idea maybe?
rsalveti has quit [Quit: Connection closed for inactivity]
jmiehe has joined #yocto
<jonmason> you only need the secondary names if you are going to override them in another yml
<jonmason> but that looks like it might work
<Zapados> i'd like to extend them, not override.  Say there are 30 BBLAYERS entries in first YML.  Second YML needs to add 2 more layers.  I just want to extend the list by the 2 more.
jwinarsk has joined #yocto
<jwinarsk> What are the known working references for the pipewire/wireplumber recipes in master? My primary interest is video right now
vladest has joined #yocto
Zapados has quit [Quit: Client closed]
<pidge> jonmason: Yes, I was. I'll check out the bug tomorrow when I'm more awake.
jmd has quit [Remote host closed the connection]
<KanjiMonster> Zapados: you shouldn't need to manually set BBLAYERS (or bblayers.conf), kas would usually autogenerate conf/bblayers.conf based on the defined layers: in repos: in all included yml files
Zapados has joined #yocto
<Zapados> Oof my connection dropped.  :P
Saur75 has quit [Quit: Client closed]
Saur75 has joined #yocto
<khem> vvn: interesting. I have the bbappend from meta-rpi dynamic layer and it does not complain like yours, so I wonder if this recipe is somehow masked in your distro setup
<vvn> I'm wondering the same thing. In my company layer, I do have a dynamic bbappend for libcamera as well. How can I debug that?
<khem> grep for libcamera in all layers
* vvn was expecting some magic bitbake-layers trick from khem
Zapados has quit [Quit: Client closed]
Zapados has joined #yocto
johnmcnuggets has quit [Quit: Leaving]
goliath has joined #yocto
sugarbeet has quit [Ping timeout: 256 seconds]
sugarbeet has joined #yocto
xmn_ has joined #yocto
xmn has quit [Ping timeout: 252 seconds]
<rburton> Zapados: normally you let kas deal with BBLAYERS and don't assign it yourself
<Zapados> rburton ahh!
alessioigor has quit [Ping timeout: 264 seconds]
<rburton> as jon said, meta-arm has a non-trivial but not complicated setup involving lots of fragments that get overlaid
starblue has joined #yocto
joekale has quit [Ping timeout: 260 seconds]
enok71 has quit [Ping timeout: 268 seconds]
enok71 has joined #yocto
jmiehe has quit [Quit: jmiehe]
florian has quit [Ping timeout: 260 seconds]
enok71 has quit [Ping timeout: 264 seconds]
olani- has joined #yocto
<Zapados> rburton if I clone meta-arm but name it meta-arm_TEST, does the repo: meta-arm: break because it doesn't have a "path:"?
pedrowiski has joined #yocto
<jwinarsk> Has anyone built swiftshader for riscv64? I'm hitting error with this function: `llvm::jitlink::InProcessMemoryManager::Create()`. There are couple upstream bugs.
<vmeson> For the curioius: https://github.com/google/swiftshader SwiftShader is a high-performance CPU-based implementation of the Vulkan graphics API.