Hi, Actually the Yocto SDK sets LD variable correctly : here is its value : aarch64-poky-linux-ld --sysroot=/media/talel/data/sdk-multigate/sysroots/aarch64-poky-linux Now I'm stuck with linking,

Update Zephyr to the latest version. Tested via zephyr-philosophers on qemu-cortex-m3. Signed-off-by: Jon Mason <jon.mason@...> --- classes/zephyr-kernel-src.bbclass | 10

A quick review of this looks fine. I'm currently swamped with some work stuff and ELC-E this week. I intend to get to this (and a few other contributions next week. If you don't see anything from

If you use volatile-binds recipe, you can override folders and files and it will select overlayfs or bind-mount

Thanks for the suggestion! If I should overlay just part of /etc I'm down to single files - and that does not seem to be possible. I'm trying to figure out if I can solve my problem using bind mounts

I meant follow the lead from one of the emails you pointed and in addition add a dependency on grub-native so you get the tool you are looking for during build. > > > >

Thanks Khem: Not entirely certain what that means. Apologies, I'm learning about 10 different things at the same time here. I probably should have also asked, should I expect grubenv to be

perhaps a post edit function and add grub-native to DEPENDS so you can find needed binaries during cross build.

Hi: I am using grub as my bootloader, and enabling it via EFI_PROVIDER = "grub-efi" That all seems to work fine, except that grubenv is not being created anywhere. I can run grub-editenv to create

Pass LDFLAGS as CLFAGS as well, you need to use --syroot option during link as well > >

Hello Talel <bhstalel@...> wrote: many options to handle such situations. usually, workdirs is strictly specific to one recipe, no other recipe gets to peak into it. however if you have

You can add a function and use nostamp on it so it gets executed every time, something

Hi, I'm creating a recipe for the dlib library. Source also contains python bindings so my recipe looks like: SUMMARY = "A toolkit for making real world machine learning and data analysis

Sorry this is rest of the mail: I need to use the macchina WORKDIR in a different recipe , so that I can compile only my sample. For example, in my recipe I export those global variable used in the

Hi All, I have a macchina.io recipe that fetches the source from Github and compiles it correctly. In the WORKDIR , there is a samples directory with many examples, and each example has a Makefile,

Over in the OpenBMC project, we rebased on yocto master (https://gerrit.openbmc-project.xyz/c/openbmc/openbmc/+/37488). After this we started seeing intermittent issues with our os-release.bb recipe.

Hi, Thanks for the reply. Setting LD to $CC didn't work as well.

Thanks, no problem :)

Hi, In my project, I ended up by overlaying not the entire /etc but only subparts I needed to modify. (/etc/machine-id is maybe already a bind mount from /run/machine-id) Hope it helps, Best

I see that your linker is not using proper sysroot perhaps using LD=${CC} might help.