In some cases, the directory structure we want in the filesystem is not
exactly what we have in target/
For example, when systemd is used on a read-only rootfs, /var must be a
tmpfs. However, we may have packages that install stuff in there, and
set important rights (via the permission-table). So, at build time, we
need /var to be a symlink to the remanent location (/usr/share/factory)
while at runtime we need /var to be a directory.
One option would have been to have /var as a real directory even during
build time, and in a target-finalize hook, move everything out of there
and into the "factory" location. However, that's not possible because
it's too early: some packages may want to set ownership and/or acces
rights on directories or files in /var, and this is only done in the
fakeroot script, which is called only later during the assembling of the
filesystem images.
Also, there would have been no way to undo the tweak (i.e. we need to
restore the /var symlink so that subsequent builds continue to work) if
it were done as a target-finalize hook.
The only solution is to allow packages to register pre- and post-hooks
that are called right before and right after the rootfs commands are
executed, and inside in the fakeroot script.
We can however not re-use the BR2_ROOTFS_POST_FAKEROOT_SCRIPT feature
either because it is done before the filesystem command, but there is
nothing that is done after. Also, we don't want to add to, and modify a
user-supplied variable.
So, we introduce two new variables that packages can set to add the
commands they need to run to tweak the filesystem right at the last
moment.
Those hooks are not documented on-purpose; they are probably going to
only ever be used by systemd.
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Reviewed-by: Romain Naour <romain.naour@gmail.com>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
For systemd, we create a simple /etc/fstab with only an entry for /, as
systemd otherwise automatically mounts what it needs where it needs it.
systemd does not like that the content of /var be symlinks to /tmp,
especially journald that starts before /tmp is mounted, and thus the
journal files are hidden from view, which causes quite a bit of fuss...
Instead, move the current /var to a sysv-only skeleton.
systemd at install time will create the /var content it needs, so we
just create an empty /var for systemd.
systemd would create /home and /srv at runtime if they are missing, but
it is better to create them right now, to simplify supporting systemd on
a RO filesystem in the (near) future.
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Cc: Maxime Hadjinlian <maxime.hadjinlian@gmail.com>
Reviewed-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Currently, we use the same skeleton for sysv-like init systems and
systemd, even though systemd has some peculiarities that makes our
default skeleton unfit.
So, we'll need to provide different skeletons (really, only part of
it) for sysv-like and systemd. In addition, in order to support the
"no init system" (BR2_INIT_NONE) use case, we introduce a "none"
skeleton.
Introduce three new skeleton packages, aptly named skeleton-sysv,
skeleton-systemd and skeleton-none. All three are providers of the
skeleton virtual package, in lieu of the skeleton-common package,
which is now a simple dependency of all three new skeletons.
Those packages are empty for now. In followup changes:
- sysv-specific stuff will be moved out of skeleton-common and into
skeleton-sysv;
- systemd-specific stuff will be added to skeleton-systemd.
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
[Arnout:
- merge with the patch that enables the BR2_INIT_NONE case
- simplify the BR2_PACKAGE_SKELETON_COMMON_ONLY select logic]
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
[Thomas:
- remove the BR2_PACKAGE_SKELETON_COMMON_ONLY logic, and instead
introduce a separate skeleton-none package for the BR2_INIT_NONE]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
We don't really care what's going in staging, as long as it is properly
setup for merged/non-merged /usr, especially for the lib/ directory.
So we can just copy the skeleton as-is.
This simplifies maintenance, should we ever need to tweak the layout:
we'd just have to do it once in the skeleton directory to have it
propagated to both target and staging.
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Reviewed-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
We now have two packages that can act as a skeleton, skeleton-common,
also known as our default skeleton, and skeleton-custom.
This means that the skeleton package can be a standard virtual package
now.
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Package variables should always be prefixed with the package name. For
SKELETON_COMMON_ROOT_PASSWORD, this means we set the variable a second
time based on the value it previously had. Fortunately, this is fine
for make, even for recursively expanded variables.
This also allows to simplify the condition of the empty password - it's
not needed to set it again to empty if it already was empty.
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Move all the handling of the default skeleton into a new package,
skeleton-common.
We don't name it skeleton-default, because it will be further split
later, into a skeleton for sysv and another for systemd, with some parts
still common between the two. So just name it skeleton-common right now;
this will save us a rename later.
While we're at it, also assign to SKELETON_COMMON_TARGET_FINALIZE_HOOKS
instead of directly to the global FINALIZE_HOOKS. Therefore, we don't
need to do all of that in a condition BR2_PACKAGE_SKELETON_COMMON==y.
Note: it would be technically sound to move the skeleton files together
within a sub-directory of the skeleton-common package. However, we refer
the user to those files, from various locations (manual, packages). It
will indeed be easier for the user to find those files in
system/skeleton/ rather than in package/skeleton-common/skeleton/
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Reviewed-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
[Arnout: remove the mkdir $(STAGING_DIR)/usr/include which was removed
in skeleton.mk in master.]
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
For target install, we don't need to check the merged-usr symlinks
any more, it was already checked before.
For staging, instead of creating directories, just copy the same
skeleton, which was already checked to be correct.
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
[Arnout: split off in a separate patch + wrote commit message]
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
In addition to checking the symlinks in the merged usr case, also check
whether /bin, /lib, /sbin and their /usr/... counterparts exist in the
non-merged case.
Missing directories are collected in the variable
SKELETON_CUSTOM_MISSING_DIRS. For symmetry and because it's more
logical, rename SKELETON_CUSTOM_NOT_MERGED_USR to
SKELETON_CUSTOM_NOT_MERGED_USR_DIRS.
When a directory is indeed missing, "stat" will print an error.
Buildroot *also* prints an error, which is clearer. So remove the error
from stat by redirecting it to /dev/null.
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
[Arnout: split off in a separate patch + wrote commit message]
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
skeleton-custom.mk contains a huge condition for BR_BUILDING and for
BR2_ROOTFS_MERGED_USR. This makes the code difficult to follow, so
rework the condition a bit so that only the parts that really need to
be conditional are indeed conditional.
Note that there is no overhead in needlessly evaluation the "stat"
commands. Indeed, the assignments use late evaluation so the "stat"
is only executed when evaluating the condition - when skeleton-custom
is not selected, stat is never called.
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
[Arnout: split off in a separate patch + wrote commit message]
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
For the custom skeleton, we practicaly do nothing, except ensure it
contains the basic, required directories, and that those are properly
setup wrt. merged /usr.
Furthermore, our current skeleton is not fit for systemd, and we'll
have to split things out into various skeletons.
So, off-load the custom skeleton into its own package.
Thus, the existing skeleton package is now limited to:
- when using our default skeleton, install and tweak it properly;
- when using a custom skeleton, do nothing except for depending on
the skeleton-custom package.
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
[Arnout: split off in a separate patch doing only this]
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
The "builtin" kernel does not boot a systemd-based system, so
we resort to building the same one as currently used by our
qemu_arm_vexpress_defconfig.
We test the 8 following combinations:
- busybox, read-only, without network
- busybox, read-only, with network
- busybox, read-write, without network
- busybox, read-write, with network
- basic systemd, read-write, network w/ ifupdown
- basic systemd, read-write, network w/ networkd
- full systemd, read-write, network w/ networkd
- no init system, read-only, without network
The tests just verify what the /sbin/init binary is, and that we were
able to grab an IP address. More tests can be added later, for example
to check each systemd features (journal, tmpfiles...)
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Cc: Arnout Vandecappelle <arnout@mind.be>
[Arnout: update .gitlab-ci.yml]
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
samba4 picks up dbus as dependency if it was built before:
Checking for dbus : yes
Checking for header dbus/dbus.h : yes
Checking for library dbus-1 : yes
There is no configure option to control dbus support so we just make
sure dbus is built before samba4 to have reproducible builds.
Signed-off-by: Bernd Kuhls <bernd.kuhls@t-online.de>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
This patch allows building samba4 with uClibc again after rpc support
was removed from uClibc. Building with musl is still broken:
The first error
../nsswitch/wins.c: In function ‘_nss_wins_gethostbyname_r’:
../nsswitch/wins.c:272:15: error: ‘NETDB_INTERNAL’ undeclared (first use in this function)
*h_errnop = NETDB_INTERNAL;
^~~~~~~~~~~~~~
../nsswitch/wins.c:272:15: note: each undeclared identifier is reported only once for each function it appears in
../nsswitch/wins.c:349:14: error: ‘NETDB_SUCCESS’ undeclared (first use in this function)
*h_errnop = NETDB_SUCCESS;
^~~~~~~~~~~~~
could be solved by applying
https://git.alpinelinux.org/cgit/aports/tree/main/samba/netdb-defines.patch
but then the build fails at
../ctdb/tests/src/test_mutex_raw.c: In function ‘main’:
../ctdb/tests/src/test_mutex_raw.c:170:28: error: ‘pthread_mutex_t {aka struct <anonymous>}’ has no member named ‘__data’
printf("pid=%u\n", mutex->__data.__owner);
^~
Therefore we disable musl support until a fix can be found.
Signed-off-by: Bernd Kuhls <bernd.kuhls@t-online.de>
[Thomas:
- instead of setting SAMBA4_CONF_ENV within the libtirpc condition,
introduce SAMBA4_CFLAGS and SAMBA4_LDFLAGS variables and use them
- add missing dependency on host-pkgconf, needed because we call
pkg-config to get the flags from libtirpc]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Now that binutils 2.29 has been introduced, and we have moved to
2.28.1 as the default version, it is time to drop support for binutils
2.26.x.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Now that binutils 2.29 has been released, it is time to move to
binutils 2.28.x as the default binutils version, instead of 2.27.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Both our target and host gdb depend on ncurses (host-ncurses for
host-gdb, of course). However, while for the target we passs
--with-curses, we are not doing this for the host variant. Due to
this, host-gdb default to using the termcap library: if such a library
is available on the build system, it will be used instead of the
host-ncurses we have built. This causes the host gdb binary to depend
on a library that we do not provide in $(HOST_DIR), breaking the
principle of a standalone SDK (which should only depend on the C
library).
To solve this, we simply pass --with-curses in HOST_GDB_CONF_OPTS,
which forces host-gdb to use the host-ncurses library.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Reviewed-by: Romain Naour <romain.naour@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
UTF-8 formatted file breaks installations on some systems:
Traceback (most recent call last):
File "setup.py", line 34, in <module>
long_description=open('README.rst').read() + '\n\n' + open('HISTORY.rst').read(),
File "/home/peko/autobuild/instance-1/output/target/usr/lib/python3.6/encodings/ascii.py", line 26, in decode
return codecs.ascii_decode(input, self.errors)[0]
UnicodeDecodeError: 'ascii' codec can't decode byte 0xc2 in position 1125: ordinal not in range(128)
This patch replaces '§' with 'S' meaning 'Section' and hence converts file to plain ASCII.
Fixes:
http://autobuild.buildroot.net/results/4189b2ddcd3d59c629af78e4d455a29c77f8c683/
Signed-off-by: Yegor Yefremov <yegorslists@googlemail.com>
[Thomas: add autobuilder reference.]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Our current skeleton is tailored to sysv-like init systems; it is not
fit for systemd-based systems. So, in upcoming changes, we'll add
another skeleton for systemd.
This means we can no longer have the current skeleton default to 'y', or
it would be enabled also for systemd, which would be incorrect.
So, we remove the default to 'y' but have it selected by the default
skeleton choice.
However, we do not yet have a way to directly build (really, install)
the custom skeleton, it is built (really, installed) as a dependency of
the default skeleton. So we must also forcibly select the default
skeleton when using a custom one.
Until we have the means to do only one or the other; i.e. when we have a
virtual skeleton.
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Reviewed-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Setting the root pasword is done in a target-finalize hook, so we do not
need to enforce a dependency from the skeleton onto host-mkpasswd.
Dropping that dependency will simplify making skeleton a virtual
package (in up-coming changes).
Instead, it is now selected as any other package. As such, it is
guaranteed to be built before target-finalize.
This however introduces a slight change in behaviour: previously,
host-mkpasswd would only be built if we needed to hash the root password
from its plain-text value. Now, host-mkpasswd is always built as soon as
the root password is non-empty, even if already pre-hashed.
Since host-mkpasswd is a really tiny weeny package bundled in Buildroot,
with only two C files, built as a single unit with a single gcc call,
the overhead is really minimal. Compared to the simplifications this
will allow in the skeleton packages (plural: common, sysv, systemd,
custom) to come, this overhead is acceptable.
Yet another simplification, even if small, to ease providing multiple
skeletons.
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Reviewed-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
This resolves a QML issues reported in
https://bugreports.qt.io/browse/QTBUG-61522 that will be fixed in the
5.9.2.
Signed-off-by: Petar Koretic <petar.koretic@qaap.io>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
This commmit adds an option to remove admin panel to save
speace if not needed.
Signed-off-by: Angelo Compagnucci <angelo.compagnucci@gmail.com>
[Thomas: fix indentation.]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
This commit moves the password generation as a post build step.
This prepares the option to remove the admin panel from
installation.
Signed-off-by: Angelo Compagnucci <angelo.compagnucci@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
This commit exclude some useless files from standard installation to
save space on embedded systems.
Signed-off-by: Angelo Compagnucci <angelo.compagnucci@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Since version 0.6, libcodec2 switched buildsystem from autotools to
cmake.
The patches in the libcodec2 directory were removed because they were
only relevant to the autotools build, but another patch was added to
fix the codec2.pc file being installed.
For the cmake package unittests are disabled by default. The download
site is changed to http://files.freedv.org/codec2. An option is added
to be able to install examples to target.
Signed-off-by: Olivier Schonken <olivier.schonken@gmail.com>
[Thomas: switch to 0.7, add patch to fix .pc file.]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
All patches from binutils 2.28 are kept, except patch
0906-microblaze-pr21180.patch, because it has been merged upstream as
of commit bd757ca7bf5886a4025ca02093fca1b8c5ce11a2. Other patches are
slightly adapted to differences that appeared in binutils 2.29.
The patches are now all Git formatted, so instead of having weird
sequence numbers, they have normal sequence numbers starting from 0001
and incrementing.
Since binutils 2.29 is now available as a .tar.xz tarball, we use this
format instead of .bz2 used for previous versions.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
binutils 2.28.1 is a bug-fix release of the 2.28.x branch, so let's
switch to it.
Since a xz-compressed tarball is available, we use it instead of the
bz2-compressed one.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
The default audio backend for qemu is configured at compile time. It
generates annoying warning messages to qemu's stderr when running our
tests, like these:
pulseaudio: set_sink_input_volume() failed
pulseaudio: Reason: Invalid argument
pulseaudio: set_sink_input_mute() failed
pulseaudio: Reason: Invalid argument
Explicitly set the audio backend to "none" at runtime to remove those
messages from our logs. There is no command line argument for this, so
use an environment variable when starting qemu.
Signed-off-by: Ricardo Martincoski <ricardo.martincoski@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Sometimes when a test fails in a gitlab pipeline the reason of the
failure cannot be determined using only the logfile.
Add the modified rootfs as an artifact of the job to improve
troubleshooting. To accomplish this, always use -k option from the test
infra, unconditionally add the resulting images to the artifacts, and
let the runner do the cleanup for us.
These artifacts can also be useful when a test fails locally but pass at
gitlab runners.
When the test does not generate a image, this message is displayed in
the runner log:
WARNING: test-output/*/images/*: no matching files
Cc: Arnout Vandecappelle <arnout@mind.be>
Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Ricardo Martincoski <ricardo.martincoski@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
lapack does not support in-source builds anymore:
ea081254b4
Signed-off-by: Bernd Kuhls <bernd.kuhls@t-online.de>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Upstream switched to CMake, host-pkgconf is not needed anymore.
Instead of using the latest release 1.1.6 we use current git HEAD
because it contains important patches:
"Support both static and shared library CMake builds."
c756f7f5d9
"Redo CMake configuration", this commit allows to disable building
Snappy's own tests.
be6dc3db83
Signed-off-by: Bernd Kuhls <bernd.kuhls@t-online.de>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
