From mboxrd@z Thu Jan  1 00:00:00 1970
From: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Date: Tue, 26 Nov 2019 14:58:37 +0100
Subject: [Buildroot] [PATCH v2] package/pkg-kconfig: fix reconfigure for
 kconfig packages
In-Reply-To: <CA+_SqVYc-Rm3rw240LH37CVuZQaD6vt0iKvt=mQjeR4Pj3izug@mail.gmail.com>
References: <1574771063-29218-1-git-send-email-angelo@amarulasolutions.com>
 <20191126141838.74d804f8@windsurf>
 <CA+_SqVZMB3yr9ALL-M4Wb9cX7FNGUVxEmM0b2d8=vszT+uN67g@mail.gmail.com>
 <20191126142916.1def6761@windsurf>
 <CA+_SqVYc-Rm3rw240LH37CVuZQaD6vt0iKvt=mQjeR4Pj3izug@mail.gmail.com>
Message-ID: <20191126145837.3a28e202@windsurf>
List-Id: <buildroot.busybox.net>
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7bit
To: buildroot@busybox.net

On Tue, 26 Nov 2019 14:55:12 +0100
Angelo Compagnucci <angelo@amarulasolutions.com> wrote:

> > Instead of just reworking the commit message, I think it would instead
> > make sense to "git checkout 4b81badbcc0b25678ac6627548160702731cf393"
> > and test what was the behavior back then.  
> 
> Checked out at 4b81badbcc0b25678ac6627548160702731cf393
> 
> $ make linux-reconfigure
> umask 0022 && make -C /home/angelo/DEV/BUILDROOT/buildroot
> O=/home/angelo/DEV/BUILDROOT/br_qemu_arm/. linux-reconfigure
> rm -f /home/angelo/DEV/BUILDROOT/br_qemu_arm/build/linux-4.7/.stamp_staging_installed
> rm -f /home/angelo/DEV/BUILDROOT/br_qemu_arm/build/linux-4.7/.stamp_target_installed
> rm -f /home/angelo/DEV/BUILDROOT/br_qemu_arm/build/linux-4.7/.stamp_images_installed
> rm -f /home/angelo/DEV/BUILDROOT/br_qemu_arm/build/linux-4.7/.stamp_host_installed
> rm -f /home/angelo/DEV/BUILDROOT/br_qemu_arm/build/linux-4.7/.stamp_built
> rm -f /home/angelo/DEV/BUILDROOT/br_qemu_arm/build/linux-4.7/.stamp_kconfig_fixup_done
> rm -f /home/angelo/DEV/BUILDROOT/br_qemu_arm/build/linux-4.7/.stamp_configured
> /bin/sed -i -e "/\\<CONFIG_KERNEL_GZIP\\>/d"
> /home/angelo/DEV/BUILDROOT/br_qemu_arm/build/linux-4.7/.config
> echo 'CONFIG_KERNEL_GZIP=y' >>
> /home/angelo/DEV/BUILDROOT/br_qemu_arm/build/linux-4.7/.config
> /bin/sed -i -e "/\\<CONFIG_KERNEL_XZ\\>/d"
> /home/angelo/DEV/BUILDROOT/br_qemu_arm/build/linux-4.7/.config
> echo '# CONFIG_KERNEL_XZ is not set' >>
> /home/angelo/DEV/BUILDROOT/br_qemu_arm/build/linux-4.7/.config
> /bin/sed -i -e "/\\<CONFIG_CPU_LITTLE_ENDIAN\\>/d"
> /home/angelo/DEV/BUILDROOT/br_qemu_arm/build/linux-4.7/.config
> echo 'CONFIG_CPU_LITTLE_ENDIAN=y' >>
> /home/angelo/DEV/BUILDROOT/br_qemu_arm/build/linux-4.7/.config
> /bin/sed -i -e "/\\<CONFIG_AEABI\\>/d"
> /home/angelo/DEV/BUILDROOT/br_qemu_arm/build/linux-4.7/.config
> echo 'CONFIG_AEABI=y' >>
> /home/angelo/DEV/BUILDROOT/br_qemu_arm/build/linux-4.7/.config
> # As the kernel gets compiled before root filesystems are
> # built, we create a fake cpio file. It'll be
> # replaced later by the real cpio archive, and the kernel will be
> # rebuilt using the linux-rebuild-with-initramfs target.
> /bin/sed -i -e "/\\<CONFIG_DEVTMPFS\\>/d"
> /home/angelo/DEV/BUILDROOT/br_qemu_arm/build/linux-4.7/.config
> echo 'CONFIG_DEVTMPFS=y' >>
> /home/angelo/DEV/BUILDROOT/br_qemu_arm/build/linux-4.7/.config
> /bin/sed -i -e "/\\<CONFIG_DEVTMPFS_MOUNT\\>/d"
> /home/angelo/DEV/BUILDROOT/br_qemu_arm/build/linux-4.7/.config
> echo 'CONFIG_DEVTMPFS_MOUNT=y' >>
> /home/angelo/DEV/BUILDROOT/br_qemu_arm/build/linux-4.7/.config
> scripts/kconfig/conf  --olddefconfig Kconfig

So indeed, it only redoes the fixups, and does not regenerate the
.config from scratch. So there was no regression introduced later, it's
just that from the start, the -reconfigure behavior did not match what
the commit log said.

Thomas
-- 
Thomas Petazzoni, CTO, Bootlin
Embedded Linux and Kernel engineering
https://bootlin.com