in your fortran app recipe you have to add DEPENDS = "libgfortran"

Post log.do_complile perhaps that will have the needed link line

#honister #make #yocto

It’s linker cmdline which is interesting here please post it and that will help to see further

#honister #make #yocto

Yes There already is beagleV machine and it will be derived from that I have plans to add it if Someone is working on it please let me know

its possible to build baremetal toolchains, e.g. look at meta-riscv baremetal machines e.g. https://github.com/riscv/meta-riscv/blob/master/conf/machine/baremetal-riscv32.conf you can do MACHINE=barem

you might have to figure out how the clone should work firstly. then you can either encode it into SRC_URI, but I would say be careful since you might leak password and url info into packages, other o

#yocto

which release branch are you using? <tagliapietra.alessandro@...> wrote:

Ok then next thing you want to check is where the reported stripped file is located and even if it does not work then perhaps look into removing these files if they are not needed

<tagliapietra.alessandro@...> wrote: what package folders do you see under packages-split/ folder in your build area for this recipe ?

The prebuilds are for other architectures. so perhaps you can rm them in a do_install_append() or if they are really needed you perhaps would also need INSANE_SKIP_<name of package> = "arch" eventuall

This will be valiant effort but my advice is don't go alone if you want to undertake this SOC sdks have lot of nitty gritty issues that will pop up along the way and unless you have someone from SOC s

Oh I see. we do support aarch64 as build host, so if you are trying 64bit OS on rpi then it should work although, rpi4 may not have enough muscle to build, perhaps using SSD with rpi4 might help a bit

are you on a x86_64 host? Secondly try to uncomment SDKMACHINE setting in local.conf and see if it helps. Also check if your default shell is bash or dash.

OK it should be ok then. Can you describe your steps to setup the system? maybe there is some caveat

It’s complaining about SDKMACHINE setting which is a separate variable, can you check your local.conf if it’s set to something

<richard.purdie@...> wrote: right, I think shallow clone should be default IMO for all git fetcher tarballs

if you are using linux-yocto for kernel then it is surely supported , so as long as it is added to kernel recipe it should work ok. On Tue, Mar 15, 2022 at 3:58 AM Monsees, Steven C (US) via lists.yoc

On Mon, Mar 14, 2022 at 12:57 PM Monsees, Steven C (US) via lists.yoctoproject.org <steven.monsees=baesystems.com@...> wrote: these are some config files for doing what ? from what

Thanks Matthias There is a patch staged in master-next to fix this. Can you try master-next meanwhile <matthias.klein@...> wrote:

can you check /boot/config.txt on target and see if this change is persisting in your image or not