| Age | Commit message (Collapse) | Author |
|---|
|
Some recipes are installing libraries into ${base_libdir} (typically /lib) and
also use a foo-config binary to identify compile paths, for example
libusb-compat. Without mangling ${base_libdir} the ${base_libdir} path is
passed to the compiler, where it looks like a host path and results in
compile-host-path QA errors.
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
[YOCTO #5436]
Automatic selection of static uid/gid is needed for a dynamically generated
passwd and group file to have a deterministic outcome.
When a package is installed and instructs the system to add a new user or
group, unless it selects a static uid/gid value, the next available uid/gid
will be used. The order in which packages are installed is dynamically
computed, and may change from one installation to the next. This results
in a non-deterministic set of uid/gid values.
Enabling this code by adding USERADDEXTENSION = "useradd-staticids", and
adding a preconfigured passwd/group file will allow the continued dynamic
generation of the rootfs passwd/group files, but will ensure a deterministic
outcome. (Dynamic generation is desired so that users and groups that have
no corresponding functionality are not present within the final system image.)
The rewrite params function will override each of the fields in the
useradd and groupadd calls with the values specified. Note, the password
field is ignored as is the member groups field in the group file. If the
field is empty, the value will not be overridden. (Note, there is no way
to 'blank' a field, as this would only generally affect the 'comment' field
and there really is no reason to blank it.)
Enabling USERADD_ERROR_DYNAMIC will cause packages without static uid/gid
to generate an error and be skipped for the purpose of building. This is
used to prevent non-deterministic behavior.
USERADD_UID_TABLES and USERADD_GID_TABLES may be used to specify the name
of the passwd and group files. By default they are assumed to be
'files/passwd' and 'files/group'. Layers are searched in BBPATH order.
Signed-off-by: Mark Hatle <mark.hatle@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
When the system attempts to populate the sysroot's passwd/group files, it
does so in a single block. However, with the way it was previously
implemented, the system would always run through the code necessary to
populate the sysroot, even in the case of target packages. This had
the side effect that a cross-installed filesystem may not match a
target installed filesystem.
The code was slightly reorganized to ensure that the cross/target installed
pre-install script behavior is the same. It also moves the block that
configures the sysroot parameters to the sysroot specific section of
the code.
Also some minor validation was occuring even on nativesdk packages.
Nativesdk packages should be skipped when processing useradd ops.
Signed-off-by: Mark Hatle <mark.hatle@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Fixing the following warning:
WARNING: Variable rootfs_deb_bad_recommendations contains tabs, please remove these (/srv/ssd/sgw/poky/meta/recipes-extended/images/core-image-lsb-dev.bb)
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
Add the ability to export the tests so that they can run independently of
the build system, as is required if you want to be able to hand the test
execution off to a scheduler.
Booting/deployment of the target is still handled by the build system,
as before, only the execution of the tests happens outside of the build system.
Tests exported are the ones defined in TEST_SUITES.
No tests have been changed as interesting parts of the data store have been
exported and tests can continue to query them as before. Small adjustments were made
for a couple of oeqa modules though.
[YOCTO #5613]
Signed-off-by: Stefan Stanacar <stefanx.stanacar@intel.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
We've removed nostamp from do_rootfs in image.bbclass in OE-Core commit
8505008c115efb54d18e5f25441c7a938a32ffaf, and we should do the same for
do_populate_sdk here for the same reason - we can now rely on task
signatures so if nothing has changed, we don't need to re-run it.
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
If the package information contained unicode characters (for example, in
the package file listing) then writing them out as ASCII would fail, so
write them out using codecs.open() instead of open() using UTF-8. This
fixes ca-certificates failing in do_packagedata when buildhistory is
enabled.
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Now that bitbake is recursing into pre/postfuncs, we need to ensure
the dependencies of these functions is correct. We don't want dependencies
on MACHINE or other related variables. This patch adds in appropriate
variable exclusions to achieve this.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
OE-Core commit 259b8718a31b886f8a158aeb5de164840c9a28b2 fixed UTF-8
errors but broke decoding of escape sequences in strings (e.g.
pkg_postinst scripts had \n \t in them instead of newlines and tabs.)
We need a second call to decode() here as specifying 'string_escape' as
the second parameter won't do anything.
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Commit a5362de60c0051f16b88a40bd9cb41915bee0b0f restricted us to writing
only a single entry to the status file for any given package. This
avoids writing garbage entries (with no Status) when multiple versions
of the same package exist in the feed. However, in the case where the
same version exists for multiple architectures (for example, has been
built for both cortexa15 and generic armv7a), we do need to write out
status file entries for all of them since we can't trivially determine
which one opkg will decide to install.
Fix this by undoing a5362de60c0051f16b88a40bd9cb41915bee0b0f and writing
out an entry for everything that matches, but ensuring that we force the
Status to the correct value for each one.
Signed-off-by: Phil Blundell <pb@pbcl.net>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
read-only-rootfs
If the rootfs is read-only then we aren't going to be updating any
alternatives or modifying the password file and these binaries will
be redundant. In an ideal world we would be able to stop them from
being installed in the first place but this is non-trivial to arrange.
As a workaround in the meantime, let's just uninstall them once image
construction is finished.
Signed-off-by: Phil Blundell <pb@pbcl.net>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
In python3 this has to be taken care of, it defines a variable to
denote ABI currently its at 'm' and it uses this to construct
the directory names for installing the python headers and library
names in sysroot. e.g. it will be something like ../python3.3m/...
We need this information when we are using distutils to cross build
python extentions and want to know the locations of python headers
and libraries install locations
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
In line with python2 add distutils and setuptools
classes for handing python3
Use python-distribute instead of python-setuptools in setuptools bbclass
Remove --single-version-externally-managed since its setuptools
specific and we dont use it anymore
Do build_ext as separate step during compile
Add DISTUTILS_BUILD_EXT_ARGS for modules to pass flags to build_ext step
in setup.py
Add build_ext as sepate step during compile and add
the cross sysrooted library and headers since we are cross
compiling
Use ${PYTHON_PN} in place of hardcoding python name
Remove the /etc/share if its empty
Since OE-Core times we now have machine specific sysroots
for targets unlike before when we used arch specific sysroots
so reflect that here
Use MACHINE for sysroot when not building for build host
Python's machinery replaces directories in sysroot path to match OE's
staging area sysroots. Earlier we use to have HOST_SYS represent sysroot
always but now we use MACHINE to represent target sysroots but HOST_SYS
to represent host sysroot. This patch caters to that difference
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
This is needed to accomodate python3 alongside
python2
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
This causes issues when postinstalls have ERROR keywords
its interpreted as error and image build is cancelled
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
There would be some unexpected errors when the whole TMPDIR is located
on nfs, so add a test for it in sanity.bbclass.
Note:
The better way to get the filesystem id should be get f_fsid from struct
statvfs, but there is no f_fsid in os.stat() or os.statvfs(), so we use
'stat -f -c "%t"' here.
BTW., s/tmpdir/TMPDIR/ in the previous comment message to make them have
a uniform.
[YOCTO #5442]
Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
Automatically add 'initscripts-functions' to packages when needed.
This extra dependency is added only when the package's init script
would be used at system start-up. In other words, in a systemd based
image, if the package has already got its service files, it will not
have this extra dependency.
Signed-off-by: Chen Qi <Qi.Chen@windriver.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
If a kernel is built without any external modules (aka no CONFIG_*=m),
then during a modules_install of the kernel an empty directory is
created at /lib/modules/${KERNEL_VERIONS}/kernel. This is behaviour of
the kernel infrastructure, the directory would normally be populated
with the modules that were built.
However because of the expectations of kernel-modules-split, no packages
are created when there are no modules and an empty directory lingers.
This raises QA issues as warning or errors (depending on the distro).
The following patch changes the kernel_do_install task to check if the
directory is empty and if so removes it.
Signed-off-by: Nathan Rossi <nathan.rossi@xilinx.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
If we can get the original package name when reading
package infos in toaster.bbclass, we save it for proper
referincing in the toater ui.
Signed-off-by: Alexandru DAMIAN <alexandru.damian@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
* when log.do_package shows some unexpected dependency, people usually
need to grep package directory to find which binary was creating that
dependency, show it directly in the debug output
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
* split PRIVATE_LIBS and don't use find(), so that libfoo cannot be
found in PRIVATE_LIBS = "libfoobar"
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
On my system, doing "bitbake -c menuconfig busybox" pops up
a gnome-terminal, I can see that the compilation starts but then the
terminal automatically exits.
It turns out I don't have ncurses development package installed on my host,
and OE's host sysroot isn't being passed to gcc, so compilation fails.
The window automatically closes before I can read the error message
since the '$' in the return code check needs to be escaped.
Make sure the user can read the error message by properly checking the
return code.
Signed-off-by: Jacob Kroon <jacob.kroon@mikrodidakt.se>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Multilib QA warning was observed, as follows:
------
WARNING: Multilib QA Issue: lib32-oprofile package lib32-oprofile -
suspicious values 'kernel-vmlinux' in RRECOMMENDS
------
The package starting with 'kernel-vmlinux' should be ok with multilib QA
checking.
Signed-off-by: Ming Liu <ming.liu@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
The filenames sometimes may have strange names. With the 'awk' script
it handled a limited number of spaces in the filename and a package
installing a file named "test file with spaces" would have its name
truncated.
This patch uses the find's printf formating to simplify the code and
properly handle this case. From a testing image, the only diff produced
is:
,----[ files-in-image.txt diff ]
| --rwxr-xr-x root root 0 ./usr/bin/test\ file\ with\
| +-rwxr-xr-x root root 0 ./usr/bin/test file with spaces
`----
The options used are available since findutils 4.2.5, released in 19
Nov 2004, making it available in all supported host distributions.
Signed-off-by: Otavio Salvador <otavio@ossystems.com.br>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Scons supports -jX parallel make, so let's use that. A small scale test of a few recipes shows no failures!
Signed-off-by: Koen Kooi <koen.kooi@linaro.org>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
The run-postinsts script has been moved to ${sbindir}.
[YOCTO #5719]
Signed-off-by: Chen Qi <Qi.Chen@windriver.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
Several required perl modules may missing on the host, for example the
Text::ParseWords, Thread::Queue and Data::Dumper are not installed by
default on recent Fedora releases (19 and 20 AFAIK). There would be wild
errors if they don't exist, so check them in sanity.bbclass.
And add perl to SANITY_REQUIRED_UTILITIES.
[YOCTO #5744]
Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
Additional metadata from user-defined variable is written into
control/spec file of binary package.
Three variables are searched for adiitional package metadata:
* PACKAGE_ADD_METADATA_<PKGTYPE>_<PN>
* PACKAGE_ADD_METADATA_<PKGTYPE>
* PACKAGE_ADD_METADATA
First found variable with defined value wins.
<PN> is a package name. <PKGTYPE> is a distinct name of specific
package type:
* IPK for .ipk packages
* DEB for .deb packages
* RPM for .rpm packages
Variable can contain multiple [one-line] metadata fields separated by
literal sequence '\n'. Separator can be redefined through variable flag
'separator'. In package control/spec file separator is replaced by
newline character.
Signed-off-by: Leonid Borisenko <ive.found@gmail.com>
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
Two variables are searched for value of additional package metadata:
* PACKAGE_ADD_METADATA_<PKGTYPE>
* PACKAGE_ADD_METADATA
First found variable with defined value wins.
<PKGTYPE> is a parameter of getter and expected to be a distinct name
of specific package type. For example: 'DEB' or 'RPM'.
Variable can contain multiple [one-line] metadata fields, separated by
literal sequence '\n'. Separator can be redefined through variable flag
'separator'. Getter returns found value with separator replaced with
newline character.
As side-effect, searched variables acquired flags 'type' (equals to
'list') and 'separator'.
Signed-off-by: Leonid Borisenko <ive.found@gmail.com>
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
Previous change (086ce22b88f5ef5f75a83119a32c8b3fdcfa296d) broke
the creating of vmdk images. This protects shell expansion variables
and let dd generate the image to be transformed to vmdk by image-vmdk.class.
Signed-off-by: Joao Henrique Ferreira de Freitas <joaohf@gmail.com>
[edit to change the usage of IMAGE_FSTYPE to IS_VMDK]
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
The note issues when OECMAKE_BUILDPATH and OECMAKE_SOURCEPATH were being used
stated that an in-tree build would be done, but the default is in fact an
out-of-tree build.
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
There is a potential problem if we don't remove the previous version's
stamp, for example:
The depend chain is:
libtool-native -> autoconf-native -> m4-native
We have two m4-native: 1.4.9 and 1.4.7
1) Clean all of them to make a fresh build so that we can reproduce the
problem
$ bitbake m4-native autoconf-native libtool-native -ccleansstate
2) Build libtool-native so that the m4-native_1.4.17 will be built
$ bitbake libtool-native
3) Set PREFERRED_VERSION_m4-native = "1.4.9" and build again
$ bitbake libtool-native
4) Set PREFERRED_VERSION_m4-native = "1.4.17" and build again
$ bitbake libtool-native -ccleansstate && bitbake libtool-native
Then the build will fail:
[snip]
| m4: unrecognized option '--gnu'
| Try `m4 --help' for more information.
| autom4te: m4 failed with exit status: 1
[snip]
The is because when we change m4-native to 1.4.17 and build
libtool-native again:
5) libtool-native depends on autoconf-native, and autoconf-native's
version isn't change, so it can remove the current stamp and mirror
the sstate (the one depends on m4-native_1.4.9) from the SSTATE_DIR
correctly.
6) The mirrored autoconf-native depends on m4-native_1.4.17's
do_populate_sysroot, and the stamp is already there (which is made
by step 2), so it would do nothing, but this is incorrect, since
the one that really in the sysroot is m4-native_1.4.9, then the
error happens.
Remove previous version's stamp in sstate_clean() will fix the problem.
[YOCTO #5422]
Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
* move read_shlib_providers before registering package as provider
and show warning when different package tries to provide something
already provided.
[YOCTO #4628]
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
* prepare for reading shlibs providers only from dependency tree of
current recipe
[YOCTO #4628]
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
system_package_blacklist
* unify debug messages a bit
* old implementation allowed partial match in blacklist, it's safer
to explicitly list exact matches
* I was able to build all entries from system_package_blacklist with
icecc enabled, lets assume that they were already resolved by newer
versions (we've fixed a lot of parallel issues in recipes which were
detected even without icecc and this list is very old).
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
* for different MACHINES
* is there more elegant way to have "overridable" function so that
signature handler properly uses only the branch without
STAGING_BINDIR_TOOLCHAIN?
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
* it was introduced in
commit 3a842ec52e7d010767b13bdcb5629ac07b3ee9e7
Author: Dmitry Eremin-Solenikov <dbaryshkov@gmail.com>
Date: Fri Sep 16 10:55:16 2011 +0400
Subject: icecc.bbclass: replace with updated version
without any explanation in which case
${ICECC_CC} -print-prog-name=as
is returning as in current working directory, but will keep old
behavior just in case
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
* it's needed for use-case like this:
# Inherit icecc here, so that all builders have the same sstate signatures
INHERIT_DISTRO += "icecc"
# and then disable its function by default (so that people still need to explicity
# enable it in local.conf if they have configured icecc and want to use it.
# You need to set _empty_ value in local.conf to enable icecc function:
# ICECC_DISABLED = ""
ICECC_DISABLED ??= "1"
* so default ICECC_PARALLEL_MAKE is still empty, but we want build
to respect our PARALLEL_MAKE, unfortunately we cannot do something
like ICECC_PARALLEL_MAKE ??= "${PARALLEL_MAKE}", because that would
cause PARALLEL_MAKE to reference itself.
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
* set empty TARGET_PREFIX
This has a bit weird reason caused by unsupported setup where
external-toolchain is used in some DISTRO only for some MACHINEs
and internal is used for other MACHINEs.
Because external-toolchain usually comes with different TARGET_PREFIX
it was causing allarch recipes to have different signatures even
when they don't use toolchain at all.
Empty TARGET_PREFIX also helps to find allarch recipes which still
have default dependency on e.g. virtual/${TARGET_PREFIX}gcc.
* add TARGET_FPU just for completeness (it was used in icecc.bbclass
but now it's vardepexcluded there as well)
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
is present
When update-alternatives was part of opkg which got built in most nativesdk
scenarios, this missing dependency wasn't an issue. We now need nativesdk-opkg-utils
so we need to ensure the dependency is present in nativesdk cases.
This avoids build failures with the recent u-a move to opkg-utils.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
We have the odd situation where the CONTROL/DEBIAN directory can be removed
in the middle of the walk, the isdir() test would then fail and the walk code
would assume its a file hence we check for the names in files too.
This resolves the autobuilder failure:
error: File not found: /home/pokybuild/yocto-autobuilder/yocto-slave/nightly-fsl-arm/build/build/tmp/work/armv7a-vfp-neon-poky-linux-gnueabi/xinit/1_1.3.3-r0/package/DEBIAN
RPM build errors:
File not found: /home/pokybuild/yocto-autobuilder/yocto-slave/nightly-fsl-arm/build/build/tmp/work/armv7a-vfp-neon-poky-linux-gnueabi/xinit/1_1.3.3-r0/package/DEBIAN
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
allarch packages shouldn't use these variables.
[YOCTO #5396]
Signed-off-by: Laurentiu Palcu <laurentiu.palcu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Specify the full set of install paths (bindir, libdir, etc) for packages that
use the GNUInstallDirs module, instead of just the prefix and leaving the rest
as default (which breaks with multilib).
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Unfortunately d.keys is extremely slow. Using a list in this case should be
fine since the addtask lines are immediately above the code and aren't
going to change often.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Signed-off-by: Koen Kooi <koen.kooi@linaro.org>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Signed-off-by: Koen Kooi <koen.kooi@linaro.org>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Signed-off-by: Koen Kooi <koen.kooi@linaro.org>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Signed-off-by: Koen Kooi <koen.kooi@linaro.org>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
Ross Burton
Mark Hatle
Saul Wold
Stefan Stanacar
Paul Eggleton
Richard Purdie
Phil Blundell
Khem Raj
Robert Yang
Chen Qi
Nathan Rossi
Alexandru DAMIAN
Martin Jansa
Jacob Kroon
Ming Liu
Otavio Salvador
Koen Kooi
Leonid Borisenko
Joao Henrique Ferreira de Freitas
Laurentiu Palcu