[OE-core] [PATCH] kernel-yocto: run make config for KBUILD_DEFCONFIG

[Bruce Ashfield]

On Wed, Apr 26, 2017 at 9:19 AM, Stefano Babic <sbabic at denx.de> wrote:

> In case the defconfig was created with "make savedefconfig", it is not
> enough to copy it to defconfig as it is done now. In fact, running make
> ${KBUILD_DEFCONFIG}, as usually when building the kernel outside poky,
> does much more and the resulting .config file is different as the starting
> one.
> The file generated by "make savedefconfig" contains in most cases a
> subset of all options that are required for the kernel, and other
> options are inserted automatically when make is called.
>
> Signed-off-by: Stefano Babic <sbabic at denx.de>
> ---
>  meta/classes/kernel-yocto.bbclass | 3 ++-
>  1 file changed, 2 insertions(+), 1 deletion(-)
>
> diff --git a/meta/classes/kernel-yocto.bbclass
> b/meta/classes/kernel-yocto.bbclass
> index 0330270..27ff8b7 100644
> --- a/meta/classes/kernel-yocto.bbclass
> +++ b/meta/classes/kernel-yocto.bbclass
> @@ -109,7 +109,8 @@ do_kernel_metadata() {
>                                         bbwarn "defconfig detected in
> WORKDIR. ${KBUILD_DEFCONFIG} skipped"
>                                 fi
>                         else
> -                               cp -f ${S}/arch/${ARCH}/configs/${KBUILD_DEFCONFIG}
> ${WORKDIR}/defconfig
> +                                oe_runmake -C ${S} O=${B}
> ${KBUILD_DEFCONFIG}
>


This would need to be a different mode in the configuration task, not here.
This routine only gathers inputs, and doesn't process them.

I have an open bug for this, and a WIP patch that will go out early in the
2.4 release cycle.

Bruce


> +                                mv ${B}/.config ${WORKDIR}/defconfig
>                                 sccs="${WORKDIR}/defconfig"
>                         fi
>                 else
> --
> 2.7.4
>
> --
> _______________________________________________
> Openembedded-core mailing list
> Openembedded-core at lists.openembedded.org
> http://lists.openembedded.org/mailman/listinfo/openembedded-core
>



-- 
"Thou shalt not follow the NULL pointer, for chaos and madness await thee
at its end"
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openembedded.org/pipermail/openembedded-core/attachments/20170426/7c4395d2/attachment-0002.html>