Age | Commit message (Collapse) | Author |
|
Signed-off-by: David P <megver83@parabola.nu>
|
|
Signed-off-by: David P <megver83@parabola.nu>
|
|
This is done by enabling CONFIG_COMPAT.
Though when importing the configuration on Linux
v5.4.8 and enabling this configuration with:
make ARCH=arm64 CROSS_COMPILE=aarch64-linux-gnu- xconfig
it updated other parts of the the .config as well.
Signed-off-by: Denis 'GNUtoo' Carikli <GNUtoo@cyberdimension.org>
|
|
Without that patch, /boot/vmlinuz-linux-libre-aarch64 ends
up being an Image.gz, however some u-boot configuration
and/or versions seem to not like that and fail with the
following message:
Bad Linux ARM64 Image magic!
Making sure that vmlinuz-linux-libre-aarch64 is an Image
results in the board booting. However it then fails with:
Failed to execute /init (error -8)
This might be because the board I used for testing doesn't
support runing 32bit binaries and/or because we need to
enable more configuration in the kernel to enable that.
Signed-off-by: Denis 'GNUtoo' Carikli <GNUtoo@cyberdimension.org>
|
|
The chromebook package has also been disabled due to the following
error:
| ==> Starting package_linux-libre-aarch64-chromebook()...
| FATAL ERROR: Couldn't open "arch/arm/boot/zImage": No such file or directory
| mkimage: Can't read kernel.itb.tmp: Invalid argument
| ==> ERROR: A failure occurred in package_linux-libre-aarch64-chromebook().
| Aborting...
It could be fixed later on if there is some interest.
Signed-off-by: Denis 'GNUtoo' Carikli <GNUtoo@cyberdimension.org>
|
|
Signed-off-by: Denis 'GNUtoo' Carikli <GNUtoo@cyberdimension.org>
|
|
- Compiling a toolchain takes time.
- We also need to make sure to get the toolchain right, usually it
requires 3 steps (gcc-initial, gcc-intermediate, gcc-final) and
here we're trying to do it in two to save compilation time.
- The 2 step toolchain worked for newlib but we don't know PKGBUILD
examples that do that for glibc based toolchains.
- We don't need a libc to build Linux.
Because of all that, we depends on the aarch64 gcc-initial toolchain
for building it now.
Once we have pushed aarch64 gcc, we will need to add it again as
a dependency for building this pacakge, in order to be able to build
this package and aarch64 userspace applications as well with the same
compiler.
Signed-off-by: Denis 'GNUtoo' Carikli <GNUtoo@cyberdimension.org>
|
|
Signed-off-by: David P <megver83@parabola.nu>
|
|
Signed-off-by: Denis 'GNUtoo' Carikli <GNUtoo@cyberdimension.org>
|
|
Some System On a Chip (SOC) have 64bit ARM CPUs that are also
capable of running arm 32bit applications.
Since:
- Parabola doesn't have an aarch64 package repository yet
- The devicetree of devices using such System On a Chip
are not compiled when using the armv7 defconfig
- Drivers or other platform support code for such hardware
might also not be compiled in with the armv7 defconfig
- This approach has already been tested with the linux-libre-x86_64
kenrel
It's then a good idea to add support for such devices by adding
an arrch64 kernel as the maintenance and work is minimal.
The downside is that the external kernel modules that are either
compiled against the kernel or using DKMS will not work by default.
It's however still possible to compile them by hand by using
ARCH= and CROSS_COMPILE= as it is done for this kernel, so it
should be relatively easy to add support for the non-dkms modules.
As for DKMS, someone would need to look into it to understand how
to pass it the make flags (ARCH= and CROSS_COMPILE=) that are
required to automatically build modules.
Signed-off-by: Denis 'GNUtoo' Carikli <GNUtoo@cyberdimension.org>
|
|
Signed-off-by: Denis 'GNUtoo' Carikli <GNUtoo@cyberdimension.org>
|
|
Signed-off-by: David P <megver83@parabola.nu>
|
|
Signed-off-by: David P <megver83@parabola.nu>
|
|
reason: looks unused and without love
Signed-off-by: David P <megver83@parabola.nu>
|
|
Signed-off-by: David P <megver83@parabola.nu>
|
|
Signed-off-by: David P <megver83@parabola.nu>
|
|
Signed-off-by: David P <megver83@parabola.nu>
|
|
Signed-off-by: Denis 'GNUtoo' Carikli <GNUtoo@cyberdimension.org>
|
|
Signed-off-by: David P <megver83@parabola.nu>
|
|
Signed-off-by: Denis 'GNUtoo' Carikli <GNUtoo@cyberdimension.org>
|
|
Signed-off-by: David P <megver83@parabola.nu>
|
|
Signed-off-by: David P <megver83@parabola.nu>
|
|
required
Sometimes the defconfig and the source code are out of sync, this
can result in issues like this one:
| ==> Starting build()...
| scripts/kconfig/conf --syncconfig Kconfig
| *
| * Restart config...
| *
| *
| * General architecture-dependent options
| *
| OProfile system profiling (OPROFILE) [M/n/y/?] m
| OProfile multiplexing support (EXPERIMENTAL) (OPROFILE_EVENT_MULTIPLEX) [N/y/?] n
| Kprobes (KPROBES) [Y/n/?] y
| Optimize very unlikely/likely branches (JUMP_LABEL) [Y/n/?] y
| Static key selftest (STATIC_KEYS_SELFTEST) [N/y/?] n
| Stack Protector buffer overflow detection (STACKPROTECTOR) [Y/n/?] (NEW)
This makes sure that in such case, the default choice is used instead of
asking the user about it.
Signed-off-by: Denis 'GNUtoo' Carikli <GNUtoo@cyberdimension.org>
|
|
While updating this kenrel to 5.1.3-gnu, the 5.0.2-gnu defconfig
was accidentally commited.
This change fixes it by importing linux-libre's defconfig for
x86_64.
Signed-off-by: Denis 'GNUtoo' Carikli <GNUtoo@cyberdimension.org>
|
|
Signed-off-by: Denis 'GNUtoo' Carikli <GNUtoo@cyberdimension.org>
|
|
about rcn-libre patch
|
|
Signed-off-by: David P <megver83@parabola.nu>
|
|
Signed-off-by: Denis 'GNUtoo' Carikli <GNUtoo@cyberdimension.org>
|
|
Signed-off-by: David P <megver83@parabola.nu>
|
|
Signed-off-by: Denis 'GNUtoo' Carikli <GNUtoo@cyberdimension.org>
|
|
Signed-off-by: David P <megver83@parabola.nu>
|
|
Signed-off-by: David P <megver83@parabola.nu>
|
|
Signed-off-by: David P <megver83@parabola.nu>
|
|
although patches may still work for 4.9, the support is over
Signed-off-by: David P <megver83@parabola.nu>
|
|
Signed-off-by: Denis 'GNUtoo' Carikli <GNUtoo@cyberdimension.org>
|
|
Signed-off-by: David P <megver83@parabola.nu>
|
|
The previous PKGBUILD was build and tested.
However I removed the following files after testing
and forgott to regenrate the sha512sum:
'kernel.its', 'kernel.keyblock', 'kernel_data_key.vbprivk'
I didn't re-test after that since it already took hours to
build it my build machine and that this PKGBUILD is based
on a well tested PKGBUILD.
I didn't bump the package revision in this commit because the
package cound't build before.
Signed-off-by: Denis 'GNUtoo' Carikli <GNUtoo@cyberdimension.org>
|
|
With this kernel, on machines with an x86_64 compatible CPU, you
can run an i686 userspace with an x86_64 kernel.
This for instance enables to use the same i686 rootfs on computers
with 32bit and 64bit CPUs, while still being able to take advantage
of a 64bit kernel on the machines that supports it.
The users will then need to force i686 in pacman.conf like that:
[options]
Architecture = i686
They will also need to select the right kernel at boot.
Signed-off-by: Denis 'GNUtoo' Carikli <GNUtoo@cyberdimension.org>
|
|
Signed-off-by: David P <megver83@parabola.nu>
|
|
Signed-off-by: David P <megver83@parabola.nu>
|
|
Signed-off-by: David P <megver83@parabola.nu>
|
|
Signed-off-by: David P <megver83@parabola.nu>
|
|
Signed-off-by: David P <megver83@parabola.nu>
|
|
Because ArchLinux32 has linux-pae in [extra]
Also update config
Signed-off-by: David P <megver83@parabola.nu>
|
|
Signed-off-by: David P <megver83@parabola.nu>
|
|
Signed-off-by: David P <megver83@parabola.nu>
|
|
Signed-off-by: David P <megver83@parabola.nu>
|
|
Signed-off-by: David P <megver83@parabola.nu>
|
|
Signed-off-by: David P <megver83@parabola.nu>
|
|
Signed-off-by: David P <megver83@parabola.nu>
|