take a look at following link, it will help solve this hopefully https://wiki.gentoo.org/wiki/Project:Toolchain/Binutils_2.34_porting_notes/undefined_reference_to_bfd_get_section_* <michael.opdenacker

if you are using latest release then py2 is more or less not needed unless you use perhaps packages like chromium ( which is also now fixed in meta-browser ) but this is for oe-core and meta-oe layers

#python

Yes it’s supported although it’s not default for qemumips so the simple trick you can do is change the DEFAULTTUNE setting in the qemumips.conf away from mips32r2

We have dropped uclibc support long time ago so any effort to support uClibc based system will be a bit of work and we have also dropped stale patches from metadata so even if you are able to get exte

lgtm. Thanks

<quentin.schulz@...> wrote: sounds good.

just adding layer is not enough, you have to add it to your image as well via IMAGE_INSTALL or some other way as indirect dependency

<quentin.schulz@...> wrote: perhaps applying the sed expression via do_configure:prepend() is simple ? and maybe make it rk3399 specific with do_configure:prepend:rk3399

I think using qemux86-64 might be your best bet, its fairly automated with runqemu script all you would need to do is package up the gtests as ptest package with a runner

I agree with you that broken is not good as it was for sometime, but now with the latest supported release in 3.4 series ( which is 3.4.1) things should be good.

Do we still need this patch ? I think now that dot release is out it’s less of a problem. Version specific patching will set a different preset for the layer to carry unexcercised patches

Looks good

tf-a built with clang is bloated for rk3399 SOCs with 2.6+ it needs looking into, until then use gcc always to build it Signed-off-by: Khem Raj <raj.khem@...> Cc: Jon Mason <jon.mason@...> C

This is a known issue look at https://github.com/kraj/meta-clang/issues/119 You can workaround it by https://github.com/kraj/meta-clang/issues/119#issuecomment-518317888 On Thu, Dec 9, 2021 at 11:22 A

The changes are the same as master and it applies fine on master without any patch fuzz either, can you check if there are some other patches are applied to this recipe from other layers. I wonder if

That’s ok to remove it but more importantly does it work now without this option

Cool

meta-arm has dropped exporting virtual/trusted-firmware-a and expects direct use of trusted-firmware-a on depends sections Signed-off-by: Khem Raj <raj.khem@...> --- recipes-bsp/u-boot/u-boot%.b

This has been included upstream see [1] [1] https://review.trustedfirmware.org/plugins/gitiles/TF-A/trusted-firmware-a/+/f943b7c8e292e3aad2fcbdd0a37505f62b3b4c87%5E%21/#F0 Signed-off-by: Khem Raj <raj

<gwilson@...> wrote: I wonder if its some overlap in device tree overlays perhaps you can open a github issue here https://github.com/agherzan/meta-raspberrypi/issues