PobodysNerfect has quit [Ping timeout: 256 seconds]
d-s-e has joined #yocto
olani- has joined #yocto
Saur has joined #yocto
bps2 has quit [Ping timeout: 240 seconds]
bps has joined #yocto
bps has quit [Changing host]
bps has joined #yocto
amitk__ has joined #yocto
<qschulz>
o/
amitk has quit [Ping timeout: 264 seconds]
florian has joined #yocto
florian_kc has joined #yocto
amitk__ has quit [Ping timeout: 256 seconds]
jclsn has joined #yocto
simonsilie[m] has joined #yocto
starblue has quit [Ping timeout: 256 seconds]
starblue has joined #yocto
d-s-e has quit [Ping timeout: 240 seconds]
silbe has joined #yocto
pidge has joined #yocto
seninha has joined #yocto
prabhakarlad has quit [Quit: Client closed]
<dacav>
Hi. I've got a question about -native builds. We have a recipe X.bb that depends on Y-native, so that the tool provided by Y can be used while building X. We had problems building Y-native, and to our surprise we learned that it is built with the host compiler, while we would have expected it to be built with the yocto-provided native compiler.
<dacav>
Is this normal/intended or are we doing something wrong?
<dacav>
We are building under kirkstone
<dacav>
The thing we find weird is that `./tmp/hosttools/gcc
<dacav>
...is a softlink to /usr/bin/gcc`
sgw has quit [Quit: Leaving.]
sgw has joined #yocto
alejandr1 has quit [Ping timeout: 246 seconds]
alejandr1 has joined #yocto
otavio has joined #yocto
<JaMa>
dacav: that's expected, there is no recipe for gcc-native, the one from host is used, you can use buildtools tarball for some things
d-s-e has joined #yocto
amitk_ has quit [Ping timeout: 256 seconds]
amitk has joined #yocto
amitk has quit [Ping timeout: 256 seconds]
amitk has joined #yocto
<LetoThe2nd>
I guess langdale will be EOL after 4.1.4?
ptsneves has joined #yocto
goliath has quit [Remote host closed the connection]
<jclsn>
Does anyone know a comprehensive how-to for signing ATF binaries?
<jclsn>
I need to implement the root of trust for our Layerscape processor and the NXP docs I could find are scattered and hard to understand
<jclsn>
I am actually wishing back HAB at this point xD
rob_w has quit [Remote host closed the connection]
<qschulz>
jclsn: if tf-a is part of the u-boot fit image, and you sign the u-boot fit image and make the spl check the signature of it before loading it, then that should be enough
<jclsn>
qschulz: I think on Layerscape processors it is a FIP image and not FI(T) image. I am also confused
<jclsn>
It would be nice if there were a unified standard like UEFI for x86 one day ^^
<dacav>
jclsn: I'm doing something similar. Our current solution is to include the firmware that we want to have verified in the fit image. Our uboot has the offsets in flash determined at runtime, to point at the offset where the firmware is located according to the fit image.
<jclsn>
dacav: I think it is a bit different in our case. The TF-A is loading U-Boot. Boot chain is BootROM -> TF-A -> U-Boot -> Kernel
<jclsn>
I am just trying to make sense of this, so a may be wrong
<dacav>
JaMa: I see. But then it seems weird that the SDK (at least *our* sdk) is shipping a x86_64 compiler, if it is not used inside yocto. Is that just to enable SDK users to compile in case gcc is not on their system?
Thorn has quit [Quit: For Sale: Parachute. Only used once, never opened, small stain.]
<dacav>
jclsn: yes, it looks like this is a different approach :)
<jclsn>
dacav: What is your approach? Is it a Layerscape processor as well? They all should require the same boot stages, so I don't see how it could be different
<dacav>
We have a layerscape processor too, but we have also a secure enclave running on a different device
<dacav>
We verify uboot and a fit image containing kernel, devtree, rootfs and fman
<jclsn>
But u-boot is still verfied by the TF-A
<jclsn>
should be
<jclsn>
and the TF-A verified by the Boot ROM code
<dacav>
u-boot is (going to be) verified by the secure verification device.
<dacav>
The application within the secure verification device is verified by the device's firmware
<dacav>
So I _think_ our solution is different. But I'm not very knowledgeable of the model you linked above
<jclsn>
Ok thanks. I am going to keep on reading and work myself through this jungle. At some point I will make sense of this
<jclsn>
At least it seems that the TF-A or TBB is some standardized procedure and not proprietary sh!% like HAB
<dacav>
This world can be a little complicated. Good luck
<dacav>
Yes, I believe the stuff we are making is a little on the proprietary side. I'm not familiar with HAB though
<jclsn>
HAB is the same for i.MX processors
<jclsn>
It is also horrible to configure, but at least there is a comprehensive guide to walk you through the steps
<mckoan>
jclsn: HAB works like a charm on NXP iMX6 (and AHAB on iMX8)
<jclsn>
mckoan: Yeah, I know. i.MX8 uses hab as well. I think only the i.MX8 Quad uses AHAB or something.
<jbo>
Hey guys. I connected a Quectel EC25 module via USB to my system. I added lsusb shows the device. However, doing an mmcli -L always yields: error: couldn't get bus: Could not connect: No such file or directory
<jbo>
is there anything obvious I'm missing? I'm using the core-image-minimal with some packages added (such as usbutils, modemmanager etc)
zpfvo has quit [Ping timeout: 265 seconds]
jquaresma[m] has joined #yocto
zpfvo has joined #yocto
<dacav>
Hi. I'd need to have xxd(1) available in the SDK, since it is needed to build u-boot. I've added nativesdk-xxd to TOOLCHAIN_HOST_TASK, after extending the vim recipe with vim_%.bbappend so that it adds BBCLASSEXTEND += "nativesdk". (xxd is provided by vim)
<dacav>
I can successfully populate_sdk.
<dacav>
Yet xxd is *not* available after I source the sh file in /opt
<dacav>
Any idea of why it doesn't end up theree? :)
sakoman has joined #yocto
kscherer has joined #yocto
frieder has quit [Ping timeout: 240 seconds]
olani- has quit [Ping timeout: 250 seconds]
olani- has joined #yocto
goliath has quit [Quit: SIGSEGV]
<JPEW>
dacav: Does it work if you add nativesdk-vim instead?
<JPEW>
dacav: Ya, I think you might need that instead. xxd isn't packaged independently from vim AFAICT
bps has quit [Ping timeout: 250 seconds]
frieder has joined #yocto
Minvera has joined #yocto
alessioigor has quit [Quit: alessioigor]
alessioigor has joined #yocto
frieder has quit [Remote host closed the connection]
<jbo>
I have a USB device requiring the driver called/named "option". On a desktop, I can do this: 'modprobe option' and then everything works as expected.
<jbo>
I am desperately trying to figure out how to make that work on my Yocto
<jbo>
this is presumably provided by a kernel-module-{modname} but I can't figure out which one it is.
<jbo>
any help?
<JPEW>
jbo: I don't know a whole lot about autoloading, but I think the kernel module is supposed to have some metadata that describes when it should be loaded?
<JPEW>
Perhaps your module is missing that information
<jbo>
JPEW, well, I am working off the core-image-minimal so I think it's simply not in there. I have an overaly adding a bunch of stuff like alsa-utils, usbutils etc. Seems like I am just missing that driver/module.
zpfvo has quit [Quit: Leaving.]
<khem>
JaMa: no, it does not anymore since I do not use OE servers to run builds anymore
mckoan is now known as mckoan|away
camus has quit [Read error: Connection reset by peer]
pidge has quit [Remote host closed the connection]
florian_kc has joined #yocto
florian_kc has quit [Ping timeout: 256 seconds]
Haxxa has quit [Quit: Haxxa flies away.]
dmoseley has quit [Ping timeout: 265 seconds]
dmoseley has joined #yocto
Haxxa has joined #yocto
sakoman has quit [Quit: Leaving.]
sakoman has joined #yocto
Marian54 has joined #yocto
<Marian54>
Hi all,
<Marian54>
I'm building and itb image with kernel + rootfs.
<Marian54>
In the end archive I'm seeing a file that I don't know from what package/reciepe was installed.
<Marian54>
How can I find what reciepe is installing that file in the sysroot.
<Marian54>
The file is "gcc" :) ...
<Marian54>
Regards,
<Marian54>
Marian.
schtobia has quit [Quit: Bye!]
schtobia has joined #yocto
alimon has quit [Remote host closed the connection]
florian_kc has joined #yocto
Minvera has quit [Ping timeout: 268 seconds]
kscherer has quit [Quit: Konversation terminated!]
<Xagen>
when creating a .sh file after populating the sdk, and extracting it using `-y -d <path>`, is there a way to preserve the original file metadata for the extracted files so they look the same as when they were packaged?
jclsn has quit [Quit: WeeChat 3.8]
jclsn has joined #yocto
<fray>
ok
jclsn has quit [Client Quit]
jclsn has joined #yocto
alessioigor has quit [Remote host closed the connection]
bps has joined #yocto
bps has quit [Changing host]
bps has joined #yocto
akiCA has joined #yocto
akiCA has quit [Ping timeout: 268 seconds]
florian_kc has quit [Ping timeout: 256 seconds]
bps has quit [Ping timeout: 265 seconds]
paulg has quit [Ping timeout: 265 seconds]
chep has quit [Ping timeout: 240 seconds]
chep has joined #yocto
paulg has joined #yocto
GNUmoon2 has quit [Remote host closed the connection]
GNUmoon2 has joined #yocto
<marex>
khem: looking at LVGL now
Marian54 has quit [Quit: Client closed]
<marex>
khem: hmmmm ... https://git.ostc-eu.org/rzr/dialog-lvgl repo is "archived" , CI on that repo is not passing anymore , shall we just drop that recipe ?
<marex>
hm, maybe this can be fixed ...
<marex>
uhhhh, that's another recipe which pulls in the whole damn lvgl , except this time , the version is hard-coded in Makefile and uh, that Makefile runs git clone
<marex>
words escape me
<rfs613>
marex: what, no docker container ? no calld to AWS APIs? :P
<marex>
as long as it doesn't start the other editor to do the build, that's fine
<rfs613>
heopfully they at least use nodejs to read a yaml file ...