Whilst rngd has to run as root, we can significantly constrain its
permissions (network is only required if nistbeacon is enabled).
(From OE-Core rev: 3ed70b755d0b60e61e0871f8b0cca2e2ab5e13f9)
Signed-off-by: Alex Kiernan <alex.kiernan@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
rngd.service has a constraint of Before=sysinit.target, installing as
part of multi-user.target is an odd choice as unless it's pulled in by a
different dependency, an other target would not start it.
(From OE-Core rev: eeb1e236dab087b7565dbbf6979e2b4c03e56e91)
Signed-off-by: Alex Kiernan <alex.kiernan@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
sshd.socket launches the templated sshd@.service, so by moving the
sshdgenkeys.service to sshd.socket, key generation can start in advance
of a connection.
(From OE-Core rev: af38a39b4bed9e43c5075008be47ca72191a489e)
Signed-off-by: Alex Kiernan <alex.kiernan@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
part.get("uuid") return the Partition UUID from the partition
table instead of Filesystem UUID. This lead to swap partition
UUID not match/change when wic write expand swap partition.
change it to read the filesystem UUID using blkid. The output
from blkid should looks like this:
wic-partvzhiwq3s: LABEL="swap1" UUID="04e55c19-3f3f-4491-9e32-44eea8daa827" VERSION="1" TYPE="swap" USAGE="other"
[YOCTO #13313]
(From OE-Core rev: 683297eefebe83f848daad9927871242ab28ef91)
Signed-off-by: Lee Chee Yang <chee.yang.lee@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
This is useful when tests misbehave and get stuck, or when
a significant increase in testing time is undesirable and
needs to be caught automatically.
(From OE-Core rev: d77546e910ad9048f0057f4465716d417b810065)
Signed-off-by: Alexander Kanavin <alex.kanavin@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Python's unittest will not propagate exceptions upside
of itself, but rather will just catch and print them.
The working way to make it stop is to send a SIGINT
(e.g. simulate a ctrl-c press), which will make it exit
with a KeyboardInterrupt exception.
This also makes pressing ctrl-c twice from bitbake work
again (previously hanging instances of bitbake and qemu were
left around, and bitbake would no longer start until they
were killed manually).
(From OE-Core rev: 06568a06ca22ee279d1829d26d4c38738233c06a)
Signed-off-by: Alexander Kanavin <alex.kanavin@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
legacy.patch
removed since it is included in 2020.1
(From OE-Core rev: 5402e91bdca4d4068d93d9374974006d034fbdbf)
Signed-off-by: Wang Mingyu <wangmy@cn.fujitsu.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Existing weston test available make sure that a process
for weston-desktop-shell exist when image boot up.
Enhance weston tests by:
- execute weston-info to make sure weston interface(s)
are initialized
- execute weston and make sure it can initialize a
new wayland compositor (retry checking for
wayland processes up to 5 times)
- enable weston logging for debugging when fail
to initialize wayland compositor
[YOCTO# 10690]
(From OE-Core rev: d06552c588bde8fc15c23d8fc2eb00a4243cbc3b)
Signed-off-by: Yeoh Ee Peng <ee.peng.yeoh@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
https://www.boost.org/users/history/version_1_72_0.html documents a
"Known Issue" and has a revert patch for an issue that causes code to
fail to compile that includes the coroutine function. Without this
patch, code which includes the asymmetric_coroutine.hpp will fail to
compile.
(From OE-Core rev: e25e3f151c324272990ad6370c5bc71fd5d05c8c)
Signed-off-by: Andrew Geissler <geissonator@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
If there is a datastore within a datastore (e.g. BB_ORIGENV) then
get-hash() doesn;t correclty handle the contents using the memory
address instead of the contents.
This is a patch from dominik.jaeger@nokia.com which addresses
this problem. Its been low priority since we don't include
BB_ORIGENV anywhere this would cause an issue as standard.
[YOCTO #12473]
(Bitbake rev: 1a8bcfc1eb89ccff834ba68fb514330b510976a2)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Some companies are using large numbers of patch files in SRC_URI.
Rightly or wrongly that exposes a performance problem where the code
does not handle the large string manipulations in a way which works
efficienty in python.
This is a modified version of a patch from z00539568
<zhangyifan46@huawei.com153340508@qq.com which addresses the performance
problem. I modified it to use a more advanced regex, retain the "*" check
and cache the regex.
[YOCTO #13824]
(Bitbake rev: c07f374998903359ed55f263c86466d05aa39b68)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
When a website is behind a CDN like Cloudflare there may be a "Browser
Integrity Check" or other test applied to requests before they are
allowed through to the server. Downloading via wget passes these tests
as headers are set appropriately, however the Python urllib module may
fail these tests unless additional headers are set. This causes
Wget.checkstatus() to fail where Wget.download() would actually succeed.
For Cloudflare in particular a valid User-Agent is needed, it's easy to
add this to the headers in Wget.checkstatus(). The user agent string is
copied from Wget._fetch_index().
(Bitbake rev: 4679d3cdb9cdf23f3962aa61c599ad7474591f9f)
Signed-off-by: Paul Barker <pbarker@konsulko.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Since commit [tinfoil: Simplify remote datastore connections][1] and
[tinfoil: Add back ability to parse on top of a datastore][2] applied,
bitbake run command parseRecipeFile with param config_data.dsindex rather
than config_data.
While calling tinfoil.parse_recipe_file() with one config_data (with the same
config_data.dsindex) to parse multiple recipes, it will mess up insane check.
It broke update_layer.py on layerindex, here are the simplified steps:
[snip]
t= bb.tinfoil.Tinfoil()
t.prepare()
data = bb.data.createCopy(t.config_data)
fn = "path_to/oe-core/meta/recipes-graphics/images/core-image-clutter.bb"
t.parse_recipe_file(fn, appends=False, config_data=data)
fn = "path_to/oe-core/meta/recipes-graphics/packagegroups/packagegroup-core-x11-base.bb"
t.parse_recipe_file(fn, appends=False, config_data=data)
| File "path_to/oe-core/meta/classes/insane.bbclass", line 1303,
in __anon_1304__path_to_oe_core_meta_classes_insane_bbclass
| bb.fatal("Fatal QA errors found, failing task.")
[snip]
In above failure, RDEPENDS is assigned `${PACKAGE_INSTALL} ${LINGUAS_INSTALL}
${IMAGE_INSTALL_DEBUGFS}' in core-image-clutter.bb, but it broke insane check
on packagegroup-core-x11-base.bb
>From commit [remotedata: enable transporting datastore from the client to
the server][3], it create a new DataSmart to save receive_datastore's remote_data
Similarly, make a copy of config_data(with different config_data.dsindex) could
fix the issue.
[1] http://git.openembedded.org/bitbake/commit/?id=85e03a64dd0a4ebe71009ec4bdf4192c04a9786e
[2] http://git.openembedded.org/bitbake/commit/?id=4618da2094189e4d814b7d65672cb65c86c0626a
[3] http://git.openembedded.org/bitbake/commit/?id=784d2f1a024efe632fc9049ce5b78692d419d938
(Bitbake rev: a3074807974536e370289c25fddcb9ad93cbc137)
Signed-off-by: Hongxu Jia <hongxu.jia@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
0001-xdr_float-do-not-include-bits-endian.h.patch
removed since it is included in 1.2.6
(From OE-Core rev: c940792e1492923036cae97010f9945e7ad43a32)
Signed-off-by: Zheng Ruoqin <zhengrq.fnst@cn.fujitsu.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Add more i915 pciids to make graphic work well on some new boards
like TigerLake.
(From OE-Core rev: 64e93cbb6660f1cb60fa2599f2fd87e44e6c7565)
Signed-off-by: Liwei Song <liwei.song@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
buildall-qemu simplifies the process of build testing an upgraded or
patched recipe by cycling through the build steps for each available qemu
target, with desired LIBC specified by the user as an option (defaulting
to both glibc and musl if no option is provided). While building, a log
file with the name "<recipe>-buildall.log" is written, containing a PASS
or FAIL line upon completion of the build for each architecture. For now,
qemu targets are not selectable (i.e. the recipe is built for all qemu
targets found in meta/conf/machine), but this functionality can be added
in the future along with other useful options.
The log file created by buildall-qemu also includes some basic system
info (e.g. build start time, hostname, host OS, host kernel). Since it is
not guaranteed that tools such as lsb_release will be available on the
host (it isn't by default on my Fedora machine), this information is
collected manually. Additionally, the previous run's log is retained for
comparison by renaming it in the format <recipe>-buildall.log.old once a
new set of builds is triggered for the same recipe.
We've seen multiple variations of this concept in use as one-liners and
as bash functions in dotfiles, so it seemed appropriate to submit it in
script form to oe-core for everyone to use.
Sample log output:
BUILDALL-QEMU LOG FOR aspell
START TIME: 2020-02-11_19:50:02
HOSTNAME: yow-tgamblin-fedora2
HOST OS: Fedora 31 (Server Edition)
HOST KERNEL: 5.4.10-200.fc31.x86_64
===============
BUILD RESULTS:
[glibc]
PASS: qemuarmv5
PASS: qemux86
PASS: qemuppc
PASS: qemumips64
PASS: qemux86-64
PASS: qemumips
PASS: qemuarm
PASS: qemuarm64
PASS: qemuriscv64
[musl]
PASS: qemuarmv5
PASS: qemux86
PASS: qemuppc
PASS: qemumips64
PASS: qemux86-64
FAIL: qemumips
FAIL: qemuarm
FAIL: qemuarm64
FAIL: qemuriscv64
===============
PASSED: 14
FAILED: 4
(From OE-Core rev: ee64b7db6651df4ed419e3d2649b4fb2f2faf830)
Signed-off-by: Trevor Gamblin <trevor.gamblin@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Add support for running wic images with EFI as testimage.
Introduces a variable called QEMU_USE_OVMF for configuration.
(From OE-Core rev: 3af8aaff68ed332d812ea7dc184d392700ad7882)
Signed-off-by: Konrad Weihmann <kweihmann@outlook.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
-License-Update: Copyright year updated to 2020.
(From OE-Core rev: 76e142a370c66dd7929c9df4401aae8346f2dd68)
Signed-off-by: Wang Mingyu <wangmy@cn.fujitsu.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Add volatiles files to create /var/log/cups directory rather than create
it directly when do_install.
(From OE-Core rev: 315689f58536dec4042ef9880c227a69e71e749d)
Signed-off-by: Yi Zhao <yi.zhao@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
When cross-compiling, do not change scripts to use host
versions of perl and gawk.
Also, use INSANE_SKIP to suppress QA complaints if perl
or gawk are not on the target.
(From OE-Core rev: 8972fe5581b9fe8ef14d539001758bb13bca6737)
Signed-off-by: Joe Slater <joe.slater@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Shebang lines longer than 128 characters can give an error
depending on the operating system.
This implements a test that signals an error when locating a
faulty shebang.
YOCTO: #11053
(From OE-Core rev: e694b09cacf8eb0b7e189f449816d3d6a337cb37)
Signed-off-by: Domarys Correa <domarys.correa@ossystems.com.br>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Content of the optional parameter will be appended to the rootfs-device
in the qemu configuration, in case QB_ROOTFS_OPT is not specified.
By default this is empty.
Example use cases are:
Defining 'readonly' when using squashfs, so multiple instances of qemu
can share the same base image, something that cannot be done by just
specifying 'snapshot'.
Defining 'bootindex=0' which helps to get past the EFI shell in
ovmf-binary. This also enables the use case of running WIC images with
EFI bootloader through the testimage.bbclass.
(From OE-Core rev: e9b8c194636cb5505774a2a71bf54450580dd5b8)
Signed-off-by: Konrad Weihmann <kweihmann@outlook.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
0001-atk_test_util.h-add-missing-sys-time.h-include.patch was removed since it was included in 2.34.2
(From OE-Core rev: a2d1a3349fe92276ba28028209858c4c34ef527a)
Signed-off-by: Zheng Ruoqin <zhengrq.fnst@cn.fujitsu.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
no-dll.patch
removed since it is no available for 1.0.23
(From OE-Core rev: 5d55b5e3c5b41929e348a2c9ab69ce0ba774f0cb)
Signed-off-by: Wang Mingyu <wangmy@cn.fujitsu.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
musl-ioctl.patch
removed since it is included in 2.4.101
(From OE-Core rev: 015e631cd18d982df03d4c829c0f8229a2face38)
Signed-off-by: Wang Mingyu <wangmy@cn.fujitsu.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
This variable is already defined in image.bbclass and there's not need
to redefine it here.
(From OE-Core rev: 3ab3b10d14798d27c8e5a096daab36d1bad64e80)
Signed-off-by: Bartosz Golaszewski <bgolaszewski@baylibre.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
On ubuntu 18.04.1, it does not provides `mdir' by default
which caused `wic ls **.wic' failed on fat partition
...
$ wic ls build/tmp-glibc/deploy/images/xilinx-zynqmp/wrlinux-image-std-xilinx-zynqmp.wic
ERROR: Can't find executable 'mdir'
...
Add nativesdk-mtools to buildtools-tarball and use buildtools
to provide mdir
(From OE-Core rev: f35a6586d62a98589ef6454f36b8fbeb30029c1b)
Signed-off-by: Hongxu Jia <hongxu.jia@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
If there is no soundcards on the target (e.g. qemu), the pkp_postinst
function will report an error:
alsactl: load_state:1735: No soundcards found...
pkg_run_script: package "alsa-state" postinst script returned status 19.
opkg_configure: alsa-state.postinst returned 19.
Pass '-g' option to alsactl to ignore this error.
(From OE-Core rev: 100fce2a65582b9032cb9412fbe8ca572a0549b1)
Signed-off-by: Yi Zhao <yi.zhao@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The pathlib module is for Object-oriented filesystem paths
It also provides a lot of handy utilities for checking on
paths. This seems to justify adding it to the core package
along side os, sys, and the other *path libraries.
[YOCTO #13670]
(From OE-Core rev: e5d3a6b2fb1fad62dec6389b16fec258df0b573d)
Signed-off-by: Tim Orling <timothy.t.orling@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
This is a bugfix release:
ver 0.31:
Fix issue with verification of the second certificate in chain.
Fix issue with handling trusted CA matching in verification.
(From OE-Core rev: ea92fd8ec153146d613516c4d32476f27ad1f815)
Signed-off-by: Pierre-Jean Texier <pjtexier@koncepto.io>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Documentation says that if you exit 0 in a pkg_postinst it will marked as
installed.
If you exit 0, before running postinst-intercepts defer_to_first_boot, the
pkg_postinst_ontarget script will not be present on target.
The "exit 0" in tzdata makes it difficult to have a bbappend with a
pkg_postinst_target step when you have `INSTALL_TIMEZONE_FILE = 0`
(From OE-Core rev: 338a67951c375d907b36b87d4115ee5489df722e)
Signed-off-by: Maxime Roussin-Bélanger <maxime.roussinbelanger@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
coroutines in ruby 2.7+ needs ucontext APIs which are not available in
musl but an external library is available to provide them so use it
Use cached values for ac_cv_func_isnan and ac_cv_func_isinf this is not
detected correctly by configure on musl
on ARM drop using old arm32 implementation of coroutine which is slow and
inefficient
(From OE-Core rev: 406e7a085b0650bdba9f19633dadb7e6ff0149e5)
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
License-Update: Updated copyright years [1]
Latest master 0.10.x+ has added support for mips/mips64, which should
help compile ruby on musl for these architectures
Switch SRC_URI to github upstream URI
Check for common arches before checking others in map_kernel_arch
Drop already upstreamed patches
[1] d31eaabbaf
(From OE-Core rev: aa3b25b649cfe5d30cc0d8a539fbbcc9efdb4fbd)
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Builds like native-openjdk, really wants a to link
some tools against the static version. Since when
using the extended tarball, its the only place to
get it, add the library.
(From OE-Core rev: 59c4a3fdbbfd5a6aaba7e0a1675dcd5866a7f3a4)
Signed-off-by: Jeremy Puhlman <jpuhlman@mvista.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
