Age | Commit message (Collapse) | Author |
|
There is enough upstream support for sdx55, we no longer need to use a
custom kernel recipe.
Signed-off-by: Nicolas Dechesne <nicolas.dechesne@linaro.org>
|
|
linux-linaro-qcomlt-dev: remove DT-schema-disablement patch
|
|
Respective change was merged upstream, so drop unused patch.
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
|
|
recipes-kernel: Linux 5.15 bump to rev 43aa1d1f5fcf (v5.15.6)
|
|
Changes,
43aa1d1f5fcf Merge tag 'v5.15.6' into HEAD
...
485c2c5a3cce drivers/net/wireless/ath: ath11k wmi restore mgmt_rx_event_params.chan_freq
Signed-off-by: Aníbal Limón <anibal.limon@linaro.org>
|
|
Update firmware packages
|
|
debugcc: package Bjorn's debugcc tool
|
|
The debugcc tool can be useful on the target platforms. Add a recipe to
build corresponding package.
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
|
|
recipes-graphics: mesa_git update elf-tls patch
|
|
Signed-off-by: Aníbal Limón <anibal.limon@linaro.org>
|
|
recipes-kernel: Linux 5.15 bump to rev 485c2c5a3cce (v5.15.4)
|
|
linux-linaro-qcomlt-dev: revert dtschema enforced enabling
|
|
There is no need to check dtb schema while building OE kernel images.
Not to mention that supporting dtschema properly requires significant
changes to the OE-core layer (which are still pending). With all that in
mind revert the commit making dtschema validation mandatory when
building individual dtb files.
Revert the patch here rather than in the integration tree itself, so
that users of the integration tree still can benefit from enforced dt
schema validation.
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
|
|
Changes,
485c2c5a3cce drivers/net/wireless/ath: ath11k wmi restore mgmt_rx_event_params.chan_freq
d14eb0ba93fe Merge tag 'v5.15.4' into release/qcomlt-5.15
...
d36fa951d7fe arch/arm64: defconfig enable venus clock for sm8250
...
b65f90c953f4 Merge tag 'v5.15-rc7' into release/qcomlt-5.15
Signed-off-by: Aníbal Limón <anibal.limon@linaro.org>
|
|
Rework initramfs-test-image family of images
|
|
For the initramfs-test-image family of images we have to simulate
dynamic layers/bbappends support in the form of per-layer variables.
Currently only two layers are supported this way: openembedded-layer
(meta-oe) and networking-layer (meta-networking), both are handled
manually. Allow initramfs-*-image to provide dynamic package lists for
any layer, by looping through the BBFILE_COLLECTIONS in an anonymous
Python function. This causes minor change in PACKAGE_INSTALL_foo-layer
variables syntax.
Also as we at it, move this dynamic handling to the
initramfs-tiny-image, so that any image branching from it can use this
functionality.
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
|
|
The initramfs-test-full-image is expected to contain all the tools from
initramfs-test-image. To prevent possible surprises in future, make
initramfs-test-full-image recipe include initramfs-test-image recipe
rather than duplicating it.
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
|
|
firmware-wcn6855: provide firmware for wcn6855 wifi chip
|
|
Firmware for the WCN6855 is not yet provided in the linux-firmware.
Until it is provided by the linux-firmware, package it separately.
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
|
|
Create a package with slpi firmware for the dragonboard845c.
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
|
|
db820c has apq8096 chip, not msm8996. Use correct path for firmware
files. Provide backwards compatible paths.
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
|
|
Linux kernel 5.16 is going to use more correct apq8016 instead of
msm8916 name in the firmware paths for the DB410C. Also
platform-specific wlan NV file name was changed to reflect that it
supports only the sbc board, not the whole set of prima devices.
Switch firmware-qcom-dragonboard410 to use new firmware paths by default
and provide compat symbolic links.
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
|
|
To simplify the code, use the defined shortcut instead of spelling the
whole path.
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
|
|
To simplify the code, use the defined shortcut instead of spelling the
whole path.
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
|
|
To simplify the code, use the defined shortcut instead of spelling the
whole path.
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
|
|
To simplify the code, use the defined shortcut instead of spelling the
whole path.
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
|
|
To simplify the code, use the defined shortcut instead of spelling the
whole path.
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
|
|
initramfs-test-full-image: Do not use append with += operator
|
|
this is undefined behaviour, mant times devs used them together to get
the missing space at the beginning of string which append/prepend needs
but thats not intended behaviour
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
skales: Specify branch parameter explicitly in SRC_URI
|
|
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
bootrr, diag: make sure branch and protocol are configured
|
|
recipes-graphics: mesa_git fix installation of new all-skips.txt
|
|
The default branch name for new repositories created on GitHub is now
main. While it does not affect existing repo, let's be more explicit
just in case, as there is uncertainty about how it will be managed in
general.
Also Github has announced that it will stop supporting git:// protocol
in Q1 2022, let's make sure we don't use it anymore, and use
https:// instead.
https://github.blog/2021-09-01-improving-git-protocol-security-github/#no-more-unauthenticated-git
Signed-off-by: Nicolas Dechesne <nicolas.dechesne@linaro.org>
|
|
The deqp-all-skips.txt was renamed to all-skips.txt, see:
https://gitlab.freedesktop.org/mesa/mesa/-/commit/38dff02bfb1f237ded1e2075e560ec06b52fadcf
Signed-off-by: Aníbal Limón <anibal.limon@linaro.org>
|
|
linux-linaro-qcomlt: fix overrides for the sa8155p-adp machine
|
|
Fix override syntax in the linux-linaro-qcomlt_5.15.bb used to select
machine-specific kernel for sa8155p-adp.
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
|
|
recipes-kernel: Linux 5.15 add SRCREV for 96boards platforms
|
|
The sa8155p MACHINE uses different branch to be supported, current
Linux 5.15 common release branch is in rc7 but is good enough.
Signed-off-by: Aníbal Limón <anibal.limon@linaro.org>
|
|
conf/sa8155p-adp.conf: Use 'userdata' as the intended ROOTFS partition
|
|
Since the 'system' partition is quite small (3G),
use 'userdata' partition (around 105G) as the intended
ROOTFS partition.
Signed-off-by: Bhupesh Sharma <bhupesh.sharma@linaro.org>
|
|
Split huge firmware packages
|
|
Per-board firmware packagegroups
|
|
recipes-kernel/linux: Add new .bb files for release/sa8155p-adp/qcoml…
|
|
Signed-off-by: Bhupesh Sharma <bhupesh.sharma@linaro.org>
|
|
Signed-off-by: Bhupesh Sharma <bhupesh.sharma@linaro.org>
|
|
qcom-qcommon.inc: increase INITRAMFS_MAXSIZE
|
|
Default INITRAMFS_MAXSIZE is specified as 128 MiB, which can be easily
exceeded by e.g. initramfs-kerneltest-full-image. Increase the maximum
size to 384 MiB, allowing all initramfs test images to fit.
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
|
|
Add bootrr to meta qcom
|
|
Add the 'bootrr' utility in 'initramfs-test-full-image.bb'
Signed-off-by: Bhupesh Sharma <bhupesh.sharma@linaro.org>
|