| Age | Commit message (Collapse) | Author |
|---|
|
Change a comment to reference our custom changes to the
sysconfig module, as dnf no longer uses distutils.
[YOCTO #14610]
Signed-off-by: Tim Orling <timothy.t.orling@intel.com>
|
|
In the comment it says we patch distutils modules, but we now are
patching sysconfig and not using disutils.
[YOCTO #14610]
Signed-off-by: Tim Orling <timothy.t.orling@intel.com>
|
|
Refactor to use newly imported bb.version.LooseVersion
[YOCTO #14610]
Signed-off-by: Tim Orling <timothy.t.orling@intel.com>
|
|
Refactor to use newly imported bb.version.LooseVersion
and replace `import distutils.sysconfig` with 'import sysconfig`.
https://www.python.org/dev/peps/pep-0632/#id23
[YOCTO #14610]
Signed-off-by: Tim Orling <timothy.t.orling@intel.com>
|
|
Signed-off-by: Tim Orling <timothy.t.orling@intel.com>
|
|
distutils has been deprecated in Python 3.10 and will be removed in
Python 3.12 (predicted release date October 2023).
Add a warning now that recommends using the roughly equivalent
setuptools-* classes.
[YOCTO #14610]
Signed-off-by: Tim Orling <timothy.t.orling@intel.com>
|
|
Give this thread pool a name for clarity.
Signed-off-by: Ross Burton <ross.burton@arm.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
base-image boots in degraded mode when using systems without display
system since there is no fb device detected and pslash service would
fail to start. Removing this image feature means that core-image-base is
complete for headless devices
[RP: tweak the false case so it builds]
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
The buildhistory_list_pkg_files function uses data from do_package, not
do_packagedata. Usally the two are restored together but it may see
a half complete directory or other races issues depending on timing.
Rework the function so that it uses the correct task dependencies. This
should avoid races but means the data is only restored to buildhistory
if the do_package or do_package_setscene tasks are restored.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
The code was assuming that the a recipe with only one srcrev wouldn't "name"
it. This isn't the case as the glibc or bzip2 recipes show, you can have
a single srcrev which is named.
We can pull the data from the fetcher and in fact we already have it, we just
need to handle the "default" case and make that code the default for all srcrev
regardless of length.
[YOCTO #14017]
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
We've bumped the minimum bitbake version past the point this fallback code
was needed, drop it.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Reworks the crate fetcher class to have it install the fetcher at recipe
finalization so that it is always available before SRC_URI is expanded.
In addition, override the value of SRCPV to also install the fetcher
when SRCPV is expanded so that AUTOREV works.
[YOCTO #10867]
Signed-off-by: Joshua Watt <JPEWhacker@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Process all systemd-sysusers configuration files in order to create
users/groups at build time. systemd-sysusers would try to create them at
run-time, but for read-only rootfs that's not possible and results in
warnings from different services:
systemd-udevd[166]: /lib/udev/rules.d/50-udev-default.rules:39 Unknown group 'render', ignoring
systemd-udevd[166]: /lib/udev/rules.d/50-udev-default.rules:40 Unknown group 'render', ignoring
systemd-udevd[166]: /lib/udev/rules.d/50-udev-default.rules:42 Unknown group 'sgx', ignoring
Signed-off-by: Vyacheslav Yurkov <uvv.mail@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Adds the rust tools to the cross and native files if present so that
projects that use both rust and meson can build
Signed-off-by: Joshua Watt <JPEWhacker@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Currently, the do_create_runtime_spdx task fails with a Key Error if a
dependency is not contained in the package providers dictionary. Add a
check before using "dep" as a key in "providers".
Signed-off-by: Andres Beltran <abeltran@linux.microsoft.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
If LICENSE_PATH is not set, then the split() will fail on a NoneType.
Signed-off-by: Saul Wold <saul.wold@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
There is a function with the same name is_work_shared() in the archiver class
this causes a conflict when both classes are included. Use work-shared as the
check in WORKDIR to allow for other packages beyond the kernel and gcc that
use a common shared-work source directory.
Signed-off-by: Saul Wold <saul.wold@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Signed-off-by: Jose Quaresma <quaresma.jose@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
The existence of missed files on the local sstate cache is the
condition to search for it on the sstate mirror.
The missed collection are populated when we search the files
on the local sstate cache.
- When we have sstate mirrors we don't need to iterate all files
again because we already know what's missing.
- When we found a file on the sstate mirror is because of they
are missing on the local sstate mirror.
So avoid checking if the hash exists on the missed collection.
- When we can't find the file on the sstate mirror we don't need
to add the hash again to the missed collection as it already exists.
Signed-off-by: Jose Quaresma <quaresma.jose@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
We only have the progress bar when there are more than 100 objects.
So we can check for this and store the result to show the progress bar.
Signed-off-by: Jose Quaresma <quaresma.jose@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
We exclude allarch packagegroups from rebuilding when their dependencies change.
The reasoning is that we are just depending on a name so having these rebuild
lots is just pointless and inefficient. We also don't want them duplicated for
multiple machines for efficiency.
In general this works fine, as long as the package names don't change. That
is also rare but there is one corner case which does catch users out - debian
package renaming. When this does break, users question sstate and so on and
lose faith in the system even if this is a known choice we made.
This commit adds an error message if an allarch packagegroup depends on any
package which shows package renaming in action (through the PKG variable
being set).
If you run into this issue you either need to remove the dependency from the
packagegroup or mark the packagegroup as tune specific, i.e. set:
PACKAGE_ARCH = "${TUNE_PKGARCH}"
before the packagegroup inherit.
[YOCTO #7298]
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Optionally allow to set QB_KERNEL_ROOT to e.g. "/dev/vda ro" to mount the
rootfs reay-only in Qemu.
Signed-off-by: Adrian Freihofer <adrian.freihofer@siemens.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
kernel.org now has a mirror of the downloads.yoctoproject.org sources
archive so include this in our mirrors list.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
empty
meson now does not allow unknown options which means we have to be
mindful of adding -Dintrospection unconditionally and provide an option
to recipes to set GIR_MESON_OPTION = "" and when recipe does that then
do not add -D<Dintrospection> option to meson. This will help recipes to
use meson 60.x+ just by adding
GIR_MESON_OPTION = ""
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Cc: Alexander Kanavin <alex.kanavin@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Meson 0.60 onwards no longer simply warns when an unrecognised option is
passed, it errors out instead. This means we can remove the logic in
insane.bbclass to check for the warnings.
Signed-off-by: Ross Burton <ross.burton@arm.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
uninative works via hashes and doesn't need the version in the tarball name but
it does make things easier to inspect in DL_DIR. There were reasons such as
ease of publication of the build tarballs but we can handle those differently
now and the signature issues from the early code aren't an issue now. From 3.4
onwards we can use a version'd name.
[YOCTO #12970]
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
* Since commit ce0579dc in bitbake (fetch2: Revert the regexp removal
for the type field and instead anchor regexp) the type regexp has
been automatically anchored at the end.
* Since commit 044fb04d in bitbake (fetch2: Allow whitespace only mirror
entries) there is no need to separate the entries in PREMIRRORS with
"\n".
Signed-off-by: Peter Kjellerstedt <peter.kjellerstedt@axis.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
* Since commit ce0579dc in bitbake (fetch2: Revert the regexp removal
for the type field and instead anchor regexp) the type regexp has
been automatically anchored at the end.
* Since commit 044fb04d in bitbake (fetch2: Allow whitespace only mirror
entries) there is no need to separate the entries in MIRRORS with "\n".
Signed-off-by: Peter Kjellerstedt <peter.kjellerstedt@axis.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Since we allow the kernel package name to be varied, we should
allow those packages to provide virtua/kernel, which allows them
to be used as primary kernels.
This change drops the use of regex, since any regex would enforce
naming (and limit what can be a provider) and add complexity./
There are currently no recipes that I found, that inherit kernel
that are not kernel recipes (kernel-base, etc, provide other
facilities), so making this provides simpler shouldn't cause
problems.
Multiple kernel providers can be dealt with in the same way as
any multiple provider.
[YOCTO: #13172]
Signed-off-by: Bruce Ashfield <bruce.ashfield@gmail.com>
squash with provider
Signed-off-by: Bruce Ashfield <bruce.ashfield@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
lnr is a script in oe-core that creates relative symlinks, with the same
behaviour as `ln --relative --symlink`. It was added back in 2014[1] as
not all of the supported host distributions at the time shipped
coreutils 8.16, the first release with --relative.
However the oldest coreutils release in the supported distributions is
now 8.22 in CentOS 7, so lnr can be deprecated and users switched to ln.
[1] 6ae3b85eaffd1b0b6914422e8de7c1230723157d
Signed-off-by: Ross Burton <ross.burton@arm.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
At the last nas outage, we realised that we don't have good mirrors of the
uninative tarball if our main system can't be accessed. kernel.org mirrors
some Yocto Project data so we've ensured uninative is there. Add the appropriate
mirror url to make use of that.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
SCons has a slightly unusual implementation of out-of-tree builds where
there is no standard way of doing it. However, if a recipe sets B to
the idiomatic ${WORKDIR}/build and passes the right arguments so that this
is used then scons will fail as it is trying to find the SConstruct file
in ${B} not ${S}.
Always pass --directory=${S} to scons so that the SConstruct file can be
found, and don't call --clean if out-of-tree builds are being used as
we would have just deleted the entire build tree already.
Signed-off-by: Ross Burton <ross.burton@arm.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Signed-off-by: Ross Burton <ross.burton@arm.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
This patch is a fixup for 676757f "sstate: fix touching files inside pseudo"
running the 'id' command inside the sstate_unpack_package
function shows that this funcion run inside the pseudo:
uid=0(root) gid=0(root) groups=0(root)
The check for [ -w ${SSTATE_PKG} ] and [ -O ${SSTATE_PKG}.siginfo ]
will always return true and the touch can fail when the real user
don't have permission or in readonly filesystem.
As the documentation refers:
- the file test operator "-w" check if the file has write permission
(for the user running the test).
- the file test operator "-O" check if you are owner of file
We can avoid this test running the touch and mask any return errors
that we have.
Signed-off-by: Jose Quaresma <quaresma.jose@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Currently, the "Organization" field for SBOMs is hard-coded in
create-spdx. Create a new variable SPDX_ORG to make this field more
generic.
Signed-off-by: Andres Beltran <abeltran@linux.microsoft.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
The empty-dirs QA check verifies that all directories specified in
QA_EMPTY_DIRS are empty. It is possible to specify why a directory is
expected to be empty by defining QA_EMPTY_DIRS_RECOMMENDATION:<path>,
which will then be included in the error message if the directory is
not empty. If it is not specified for a directory, then "but it is
expected to be empty" will be used.
Signed-off-by: Peter Kjellerstedt <peter.kjellerstedt@axis.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Previously, when shortening sstate filenames, the reserved
characters for .siginfo were not considered, when siginfo=False,
resulting in differently shortend filenames for the sstate and siginfo
files. With this change, the filenames of the truncated sstate and
siginfo files have the same basename, just as is already the case for
untruncated filenames.
Making sure that the .siginfo files always have the filename of the
corresponding sstate file plus its .siginfo suffix, also when being
truncated, makes it easier to manage the sstate cache and an sstate
mirror outside of Bitbake/Yocto.
Signed-off-by: Manuel Leonhardt <mleonhardt@arri.de>
Cc: Richard Purdie <richard.purdie@linuxfoundation.org>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
qemu-native 6.1.0 requires at least gcc v7.5:
ERROR: You need at least GCC v7.5 or Clang v6.0 (or XCode Clang v10.0)
Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Signed-off-by: Vyacheslav Yurkov <uvv.mail@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Application can depend on several overlayfs mount points. Provide a
systemd unit application can depend on to make sure all overlays are
mounted before it is started to avoid any race conditions
Signed-off-by: Bruno Knittel <Bruno.Knittel@bruker.com>
Signed-off-by: Vyacheslav Yurkov <uvv.mail@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
We should default to xz or zstd instead of gz, pick xz for now as it is more tested.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
There is uncertainty about the default branch name in git going forward.
To try and cover the different possible outcomes, add branch names to all
git:// and gitsm:// SRC_URI entries.
This update was made with the script added to contrib in this patch which
aims to help others convert other layers.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Recipes that inherit cross should also be categorized as isNative
Signed-off-by: Saul Wold <saul.wold@windriver.com>
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
|
|
This allows code reuse and future usage with relationship annotations
Signed-off-by: Saul Wold <saul.wold@windriver.com>
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
|
|
Currently, there is no clean way to pass extra parameters to the go tool link,
which is passed by the go build ldflags flag, the append needs to happen inside
the quotes of the ldflags parameter
See [YOCTO #14554].
Add a variable to allow adding extra parameters to -ldflags in the GO_LDFLAGS
variable, one of the main usecases is setting the application version.
For example, adding to the recipe something like
GO_EXTRA_LDFLAGS="-X main.Version=v1.0.0"
or
GO_EXTRA_LDFLAGS="-X main.Version=${PV}"
Signed-off-by: Ahmed Hossam <Ahmed.Hossam@opensynergy.com>
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
|
|
In order to be more efficient, we use xz as compression method
to create GPL sources archives.
Signed-off-by: Fabien Lahoudere <fabien.lahoudere@collabora.com>
[V1 was https://patchwork.openembedded.org/patch/155985/]
[Rebased]
Signed-off-by: Ian Ray <ian.ray@ge.com>
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
|
|
sstate.bbclass no longer removes empty directories to avoid a race (see
commit 4f94d929 "sstate/staging: Handle directory creation race issue").
Unfortunately Python apparently treats an empty egg-info directory as if
the version it previously contained still exists and fails if a newer
version is required, which Meson does. To avoid this, make sure there
are no empty egg-info directories from previous versions left behind.
Signed-off-by: Peter Kjellerstedt <peter.kjellerstedt@axis.com>
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
The exceptions generated with repre are more detailed but escaped the
newlines making them unreadable. Fix this.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
When rebuilds are triggered, autoconf-native can fail with:
| DEBUG: Executing shell function update_gnu_config
| install: cannot stat '[BUILDPATH]tmp/work/x86_64-linux/autoconf-native/2.71-r0/recipe-sysroot-native/usr/share/gnu-config/config.guess': No such file or directory
which is due to update_gnu_config running before extend_recipe_sysroot.
This only happens rarely since usually the prepare_recipe_sysroot
function would already have set things up and only in the invalidated
task hash cases does this rebuild in this way from configure only.
Fix the code to prepend this function instead of appending which
resolves the ordering issue.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
do_populate_sdk was added to SSTATETASKS, but had no _setscene task
created to allow it to actually run from sstate. Add it so that SDKs can
be restored from sstate.
Note that like do_image_complete, do_populate_sdk is marked with
SSTATE_SKIP_CREATION by default so sstate is not used for them; adding
this task will allow it to work if the user overrides this default
though.
Signed-off-by: Joshua Watt <JPEWhacker@gmail.com>
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
Tim Orling
Ross Burton
Khem Raj
Richard Purdie
Joshua Watt
Vyacheslav Yurkov
Andres Beltran
Saul Wold
Jose Quaresma
Adrian Freihofer
Peter Kjellerstedt
Bruce Ashfield
Manuel Leonhardt
Robert Yang
Ahmed Hossam
Ian Ray