On start the usbgx service configures the usb gadget. However if
one stops the service nothing is done, a subsequent start then
tries to configure the gadget again and fails. `systemctl restart`
is implemented by doing a stop followed by a start and thus is
also affected.
Doing a start/stop/start results with the following reported in
status:
| systemd[1]: Starting Load USB gadget schemas...
| gadget-start[801]: usbg_create_gadget() duplicate gadget name
| gadget-start[801]:
| gadget-start[801]: Error on import gadget
| gadget-start[801]: Error: USBG_ERROR_EXIST : Already exist
| gadget-start[812]: ls: write error: Device or resource busy
The run-postinsts service does execute a restart for each package
which uses the systemd class which creates a race condition that the
usbgx service is already started when run-postinsts restarts usbgx
and creates the above error output.
Fix that by unconfiguring the usb gadget on stop.
Signed-off-by: Max Krummenacher <max.krummenacher@toradex.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
The changes include:
- Change to use meson buildsystem.
- Fix build issues in OE.
- Move patches from files to plymouth subdirectory.
- Make themes build/installation optional, since a user might not want
all themes to be installed to rootfs, now he can set PLYMOUTH_THEMES
to chose the themes.
Signed-off-by: Ming Liu <liu.ming50@gmail.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Resolves an issue where python3 package generation was failing to configure.
-- Cannot probe for Python/Numpy support (because we are cross-compiling OpenCV)
-- If you want to enable Python/Numpy support, set the following variables:
-- PYTHON2_INCLUDE_PATH
-- PYTHON2_LIBRARIES (optional on Unix-like systems)
-- PYTHON2_NUMPY_INCLUDE_DIRS
-- PYTHON3_INCLUDE_PATH
-- PYTHON3_LIBRARIES (optional on Unix-like systems)
-- PYTHON3_NUMPY_INCLUDE_DIRS
Signed-off-by: Mark Hatle <mark.hatle@amd.com>
Signed-off-by: Mark Hatle <mark.hatle@kernel.crashing.org>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
"openssl rsa" works with RSA keys only. Use "openssl pkey" instead which
is a frontend that picks the right operation automatically and works
with RSA keys, eliptic curve keys and also DSA keys.
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Changes 1.26.0:
Brad House (14):
update release notes format
update format slightly
RELEASE-NOTES -> RELEASE-NOTES.md
test: fix outdated license headers
no reason to include sys/random.h all the time (#684)
Do not sanity check RR Name vs Question (#685)
autotools: fix building for 32bit windows due to stdcall symbol mangling (#689)
man ares_fds(3): mark as deprecated and add explanation (#691)
fix doxygen typo
Autotools allow make to override CFLAGS/CPPFLAGS/CXXFLAGS (#695)
Event Subsystem: No longer require integrators to have their own (#696)
adig: Differentiate between internal and server error
Release prep for c-ares 1.26.0 (#698)
clusterfuzz: enforce maximum DNS packet size due to long parser time
Erik Lax (1):
Added flags to are_dns_parse to force RAW packet parsing (#693)
Gregor Jasny (1):
cmake: improve some include related code (#680)
Changes 1.25.0:
Brad House (22):
fix test building with symbol hiding
ci: add test case for building with hidden symbol visibility
ci: disable static for symbol hiding tests
getrandom() may require sys/random.h on some systems
fix support with older google test versions
CI: Add Alpine Linux and old Ubuntu (#667)
Fix bad stub for ares__iface_ips_enumerate()
ahost should use ares_getaddrinfo() these days (#669)
Connection failure should increment server failure count first
sonarcloud: const
clang-format
ares_strsplit() rewrite as wrapper around ares__buf_split()
clang-format
Autotools warning fixes (#671)
Old MacOS SDKs require you include sys/socket.h before net/if.h (#673)
docs: host -> ip
Autotools: rework to simplify and fix recent issues (#674)
set winver consistently across build systems
autotools: update logic for building tests to provide more feedback
OSSFuzz: it assumes autotools builds a static library by default, which means the old autotools must have done that even though there were comments saying it wasn't. Disable static by default on Windows however since it can't build both simultaneously.
tests: replace google DNS with CloudFlare for reverse lookups as google's servers stopped responding properly
1.25.0 release prep (#676)
Gregor Jasny (1):
Fix minor warnings and documentation typos (#666)
Martin Chang (1):
Use SOCK_DNS extension on socket on OpenBSD (#659)
Signed-off-by: Alex Kiernan <alex.kiernan@gmail.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Some users would like to be able to share library code between multiple
applications without embedding a static copy.
Signed-off-by: William A. Kennington III <wak@google.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
This makes it possible to avoid building components which are not used
in the native or runtime tooling. Most users of nanopb just want to
build `nanopb-generator-native` and `nanopb-runtime` and not the other
two combinations as they are building software for their target
leveraging nanopb.
Signed-off-by: William A. Kennington III <wak@google.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Uses the new nanopb_PYTHON_INSTDIR_OVERRIDE to drop the previous patch.
Maintains compatability with 0.4.7 and older by adding symlinks to the
old header paths.
Signed-off-by: William A. Kennington III <wak@google.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Having useradd class inherited for the -native recipe makes no sense and
emit this kind of warning during reproducibility test:
stdio: WARNING: QA Issue: mariadb-native: native/nativesdk class is not inherited last, this can result in unexpected behaviour. Classes inherited after native/nativesdk: useradd-staticids.bbclass [native-last]
So, move the useradd inherit and associated variables to the target side
of the recipe.
Signed-off-by: Yoann Congal <yoann.congal@smile.fr>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
* nana 2.5 release is from 2005 and since then the development was
moved from savannah to github, see:
https://savannah.gnu.org/projects/nanahttps://savannah.gnu.org/news/?id=8130
* the latest commit on https://github.com/pjmaker/nana
was also 10 years ago now, so don't expect nana-3.0 release any
time soon, but the commits after the 2.5 help with some autotools
issues, so it's worth taking them
* 0001-Include-stdlib.h-for-exit-and-abort-prototypes.patch was resolved in:
bed0de4e64 (diff-44df4b12f1a47b20b5e0e107fad4f7692a17a2861337fb6e984db755feb912b6)
* sent as RFC, because I've only build tested this,
but only nana-src has fewer files (possibly due to B != S)
and nana.1 man page is still installed as before (hopefully without random
build failures now)
buildhistory-diff -p buildhistory 1b5542b7fe608fc952c94bea65103883be1b9f32 master
packages/core2-64-oe-linux/nana/nana-dbg: PKGV changed from 2.5 [default] to 2.5+git0+6d70617db8
packages/core2-64-oe-linux/nana/nana-dbg: PKGR changed from r0.0 to r0.11
packages/core2-64-oe-linux/nana/nana-dev: PKGV changed from 2.5 [default] to 2.5+git0+6d70617db8
packages/core2-64-oe-linux/nana/nana-dev: PKGR changed from r0.0 to r0.11
packages/core2-64-oe-linux/nana/nana-doc: PKGV changed from 2.5 [default] to 2.5+git0+6d70617db8
packages/core2-64-oe-linux/nana/nana-doc: PKGR changed from r0.0 to r0.11
packages/core2-64-oe-linux/nana/nana-locale: PKGV changed from 2.5 [default] to 2.5+git0+6d70617db8
packages/core2-64-oe-linux/nana/nana-locale: PKGR changed from r0.0 to r0.11
packages/core2-64-oe-linux/nana/nana-src: FILELIST: removed all items "/usr/src/debug/nana/2.5/src/L_buffer.h /usr/src/debug/nana/2.5/src/nana_error.h /usr/src/debug/nana/2.5/src/calls.h /usr/src/debug/nana/2.5/src/L_times.h"
packages/core2-64-oe-linux/nana/nana-src: PKGV changed from 2.5 [default] to 2.5+git0+6d70617db8
packages/core2-64-oe-linux/nana/nana-src: PKGR changed from r0.0 to r0.11
packages/core2-64-oe-linux/nana/nana-src: PKGSIZE changed from 9569 to 0 (-100%)
packages/core2-64-oe-linux/nana/nana-staticdev: PKGV changed from 2.5 [default] to 2.5+git0+6d70617db8
packages/core2-64-oe-linux/nana/nana-staticdev: PKGR changed from r0.0 to r0.11
packages/core2-64-oe-linux/nana/nana: PKGV changed from 2.5 [default] to 2.5+git0+6d70617db8
packages/core2-64-oe-linux/nana/nana: PKGR changed from r0.0 to r0.11
Signed-off-by: Martin Jansa <martin.jansa@gmail.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Clang 18 is fussy about it
Fixes
unittests/map.hpp:65:25: error: non-constant-expression cannot be narrowed from type 'typename std::enable_if<std::is_integral<char>::value && sizeof(char) == sizeof(char), char>::type' (aka 'char') to 'const signed char' in initializer list [-Wc++11-narrowing-const-reference]
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Fix the following error when building the recipe:
Files/directories were installed but not shipped in any package:
/lib
/lib/systemd
/lib/systemd/system
/lib/systemd/system/rtkit-daemon.service
Signed-off-by: Jean-Marc BOUCHE <jean-marc.bouche@foss.st.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Changelog:
=============
- Added the --deprecations-are-errors command line option to all build tools.
- All generated files are UTF-8 encoded.
Signed-off-by: Wang Mingyu <wangmy@fujitsu.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
SRCREV = "86a5627dbeac8d9d9bc34326a758d6a477e876e4"
is 15 commits after v2.7.0 tag and configure.ac says 2.7.0 as well:
86a5627dbe/configure.ac (L4)
Signed-off-by: Martin Jansa <martin.jansa@gmail.com>
: drop 'r' from gitr and ${SRCPV}
* +git is preferred over +gitr since 2012:
https://lists.openembedded.org/g/openembedded-core/topic/72154716#16837
some recipes still use +gitr, because updating it would cause PV going backwards
but without share PRservice the version-going-backwards QA issues are quite
common and most people probably cannot depend on always-increasing version number
already
Signed-off-by: Martin Jansa <martin.jansa@gmail.com>
* avoid installed-vs-shipped with multilib when baselib is lib32 or lib64 instead of just lib
ERROR: QA Issue: lib32-flatcc: Files/directories were installed but not shipped in any package:
/usr/lib
/usr/lib/libflatcc.a
/usr/lib/libflatccrt.a
Please set FILES such that these items are packaged. Alternatively if they are unneeded, avoid installing them or delete them within do_install.
lib32-flatcc: 3 installed and not shipped files. [installed-vs-shipped]
Signed-off-by: Martin Jansa <martin.jansa@gmail.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Changelog:
=============
- test suite now works correctly on IPv6-only hosts
- fixes for building against LibreSSL
- ne_uri_parse() fixes for handling URI with no path and catch
some invalid URIs which were allowed
- retry requests after a 408 response on a persisted connection
- 207 error strings are cleaned and compressed to a single line
- fixed thread-safety in ne_rfc1123_date where gmtime_r is available
- ne_lock_refresh() fixed to use a non-idempotent request
- TLS name verification updated to match RFC 9110/6125, added strict
handling of IP literals vs DNS names
Signed-off-by: Wang Mingyu <wangmy@fujitsu.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Changelog:
==============
- tests: Default to /tmp for create_sparse_tempfile()
- tests: Avoid setting up intermediary loop device for the nvme target
- tests: Remove unreliable nvme attribute checks
- lvm-dbus: Fix leaking error
- lvm-dbus: Avoid using already-freed memory
- utils: Add expected printf string annotation
- fs: Report reason for open() and ioctl() failures
- ci: Add an action to compile libblockdev with different compilers
- Sync spec with downstream
- Add BDPluginSpec constructor and use it in plugin_specs_from_names
- overrides: Remove unused 'sys' import
- ci: Manually prepare spec file for Packit
- ci: Remove the custom version command for Packit
- swap: Add support for checking label and UUID format
- fs: Add a function to check label format for F2FS
- fs: Add a generic function to check for fs info availability
- fs: Fix allowed UUID for generic mkfs with VFAT
- fs: Add support for getting filesystem min size for NTFS and Ext
- tests: Remove some obsolete rules to skip tests
- Mark NVDIMM plugin as deprecated since 3.1
- part: Fix potential double free when getting parttype
- tests: Use BDPluginSpec constructor in LVM DBus plugin tests
- python: Add a deepcopy function to our structs
- Fix missing progress initialization in bd_crypto_luks_add_key
- tests: Skip some checks for btrfs errors with btrfs-progs 6.6.3
Signed-off-by: Wang Mingyu <wangmy@fujitsu.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Changelog:
Add clock in upper right corner
License-Update: Copyright year updated to 2024.
Signed-off-by: Wang Mingyu <wangmy@fujitsu.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Changelog:
=============
-Clients are now allowed to request more than 32 encodings (#108)
-Zlib streams are now preserved when a client switches between encodings
Signed-off-by: Wang Mingyu <wangmy@fujitsu.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Changes:
===========
- The specification expands set of Unicode characters seen by Markdown
parser as a punctuation. Namely all Unicode general categories P
(punctuation) and S (symbols) are now seen as such.
- The definition of HTML comment has been changed so that '<!-->' and
'<!--->' are also recognized as HTML comments.
- HTML tags recognized as HTML block starting condition of type 4 has been
updated, namely a tag '<source>' has been removed, whereas '<search>'
added.
Signed-off-by: Wang Mingyu <wangmy@fujitsu.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Changelog:
============
- Do not dedupe sig and sig-of-checksum when loading
- Fix the installed tests
- Show the sig-of-checksum results clearer on the CLI
Signed-off-by: Wang Mingyu <wangmy@fujitsu.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Changelog:
============
-[oss-fuzz] fuzzing support
-[Fuzzing] fix harness
-[Fuzzing] using cifuzz for PR
-Specify C++11 std in meson build
-Add ini_ prefix even to static names so inih can be used as an #include
Signed-off-by: Wang Mingyu <wangmy@fujitsu.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Changelog:
==========
- Updated documentation to refer to RFC 8949
- Improvements to 'cbor_describe'
- Fix 'cbor_string_set_handle' not setting the codepoint count
- 'cbor_load' will no longer fail on input strings that are well-formed but not valid UTF-8
- All decoders like 'cbor_load' and 'cbor_stream_decode' will accept all well-formed tag values
Signed-off-by: Wang Mingyu <wangmy@fujitsu.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
