iwd and ell need to be updated in sync. This is regularly neglected.
Also the fact that they reside in different layers compicates the update process.
Beside iwd, there are not a lot of consumers for ell.
Building with internal ell makes iwd updates easier
Signed-off-by: Markus Volk <f_l_k@t-online.de>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
ver 2.20:
Fix issue with PKEX timeout and number of frequencies used.
Fix issue with handling logic for handshake failures.
Fix issue with handling ConnectedAccessPoint signal.
Signed-off-by: Markus Volk <f_l_k@t-online.de>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Make colord work out of the box
- We already create a colord user -> make use of it by passing the username
to the 'daemon_user' meson option. The conf files and systemd service file
will then be created accordingly.
- Add a backport patch to fix 'only root can write into /var' issue.
This prevents colord from starting
[https://github.com/hughsie/colord/issues/166]
- Set the runtime path for hwdata pnp.ids and
add hwdata to RDEPENDS:${PN} for non systemd images
- inherit gtk-doc to make api-documentation available
- Dont create a home-dir for colord user. It shouldn't need one.
NOTE:
colord will fail on first run with missing databases.
After second boot it should run as expected.
Signed-off-by: Markus Volk <f_l_k@t-online.de>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
The "status" function called by this script calls "pidof" to get the process id. "pidof" does not expect or operate with a full path.
Signed-off-by: Khem Raj <raj.khem@gmail.com>
They endup in undefined intrinsics demanded by compiler
error: use of undeclared identifier '__riscv_vsetvlmax_e8mf8'
Perhaps an upgrade of vendored highway module ( when it happens)
will fix it
Signed-off-by: Khem Raj <raj.khem@gmail.com>
RISCV support in highway is assuming RVV and RISCV64
however, there is RISCV32 which perhaps does not
have RVV yet
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Drop the patch to disable strip instead set the env variable.
set UJSON_BUILD_NO_STRIP=1 and get rid of one pending patch
Signed-off-by: Khem Raj <raj.khem@gmail.com>
* Drop 0001-libgcrypt.c-Fix-prototype-of-des3_encrypt-des3_decry.patch
as the issue has been fixed upstream.
* Add a patch to fix build with clang.
Signed-off-by: Yi Zhao <yi.zhao@windriver.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
ChangeLog:
https://github.com/canonical/netplan/releases/tag/1.1
* Drop the following patches as the issues have been fixed upstream:
0001-meson.build-drop-unnecessary-build-dependencies.patch
0001-networkd.c-define-scope-specific-to-case-statement.patch
* Set build option -Dtesting=false to skip testing.
Signed-off-by: Yi Zhao <yi.zhao@windriver.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Github pull requests are not recommended. When a Github contributor
wants to send patches from its repo, hint at "git request-pull"
generated emails to guide them on the right way.
Cc: Dmitry Baryshkov <dbaryshkov@gmail.com>
Signed-off-by: Yoann Congal <yoann.congal@smile.fr>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
WirePlumber 0.5.6
~~~~~~~~~~~~~~~~~
Additions:
- Implemented before/after dependencies for components, to ensure correct
load order in custom configurations (#600)
- Implemented profile inheritance in the configuration file. This allows
profiles to inherit all the feature specifications of other profiles, which
is useful to avoid copying long lists of features just to make small changes
- Added multi-instance configuration profiles, tested and documented them
- Added a ``main-systemwide`` profile, which is now the default for instances
started via the system-wide systemd service and disables features that
depend on the user session (#608)
- Added a ``wp_core_connect_fd`` method, which allows making a connection to
PipeWire via an existing open socket (useful for portal-based connections)
Fixes:
- The Bluetooth auto-switch script now uses the common event source object
managers, which should improve its stability (!663)
- Fix an issue where switching between Bluetooth profiles would temporarily
link active audio streams to the internal speakers (!655)
Signed-off-by: Markus Volk <f_l_k@t-online.de>
Signed-off-by: Khem Raj <raj.khem@gmail.com>