Carry upstream-submitted patch that adds SDL_TESTS_SHARED_LIB option,
enabling test executables to be built with dynamic linking against
libsdl2.
Signed-off-by: Antonios Christidis <a-christidis@ti.com>
Signed-off-by: Ryan Eatmon <reatmon@ti.com>
Fix BeagleBadge DT which is now k3-am62l3-beaglebadge in TI Linux
6.18 and TI U-Boot 26.01 build name which was updated alongside Linux
DT name change.
Signed-off-by: Judith Mendez <jm@ti.com>
Signed-off-by: Ryan Eatmon <reatmon@ti.com>
The DCE library depends on the firmware to function, the firmware has
no dependency on any one userspace library. Switch the runtime dependency
around here.
Signed-off-by: Andrew Davis <afd@ti.com>
Signed-off-by: Ryan Eatmon <reatmon@ti.com>
The conflicting firmware packages that necessitate this bbappend are
all here in meta-ti-bsp, so this will be needed by any distro using
this layer. Move this bbappend out of meta-arago to this layer.
Signed-off-by: Andrew Davis <afd@ti.com>
Signed-off-by: Ryan Eatmon <reatmon@ti.com>
Upstream the BOARD name for the EVM is "am62lx-evm". Our downstream
used just "am62lx" but was updated to optionally also work with the
same name as upstream. This allows for easier migration. Switch to
using the upstream-safe name here for the EVM. This will need to
be done anyway when we switch to upstream-only for TF-A, doing it
now just makes it easier to switch between the two.
The other AM62L based boards already override the default with their
correct BOARD names, and now that the EVM does too, drop the default.
All boards will need their own BOARD name as TF-A does DRAM config
which is board specific, so dropping the default prevents mistakenly
not updating this name and causing hard to debug DRAM/boot issues.
Signed-off-by: Andrew Davis <afd@ti.com>
Signed-off-by: Ryan Eatmon <reatmon@ti.com>
Keeping all IPC related recipes in one "ipc" directory is not very well
organized. Like most other recipes give this one its own directory.
Signed-off-by: Andrew Davis <afd@ti.com>
Signed-off-by: Ryan Eatmon <reatmon@ti.com>
Keeping all IPC related recipes in one "ipc" directory is not very well
organized. Like most other recipes give this one its own directory.
Signed-off-by: Andrew Davis <afd@ti.com>
Signed-off-by: Ryan Eatmon <reatmon@ti.com>
Migrate mesa from v24.0.1 -> v25.2.8. Edit the ti-bsp logic to
build/package mesa-pvr v25 for both Rogue and SGX cores.
Remove some duplicate variables from mesa-pvr-25.inc in order to work with
the main bitbake recipe mesa-pvr_25.2.8.bb.
Signed-off-by: Antonios Christidis <a-christidis@ti.com>
Signed-off-by: Ryan Eatmon <reatmon@ti.com>
Add local copies of mesa.inc from oe-core, renamed with version suffixes
to avoid conflicts:
- mesa-pvr-24.inc: verbatim copy for v24.0.1 (existing content, renamed)
- mesa-pvr-25.inc: verbatim copy for v25.2.8 (new)
With the above new naming scheme, update check_yocto_rules.json.
Signed-off-by: Antonios Christidis <a-christidis@ti.com>
Signed-off-by: Ryan Eatmon <reatmon@ti.com>
Non-AM62LX devices use upstream TF-A. Update this to the 2.15 tag.
Signed-off-by: Suhaas Joshi <s-joshi@ti.com>
Signed-off-by: Ryan Eatmon <reatmon@ti.com>
Beaglebadge has only 256Mb OSPI flash and except for
arago-bootstrap-base-image and arago-tiny-image other arago images
doesn't fit into OSPI flash. Hence remove ubifs and ubi from
IMAGE_FSTYPES to prevent unnecessary image generation and build
failures when adding packages to the arago-default-image.
Those who want to run arago-bootstrap-base-image and arago-tiny-image
on OSPI can enable UBIFS in local.conf .
Signed-off-by: Yogesh Hegde <y-hegde@ti.com>
Signed-off-by: Ryan Eatmon <reatmon@ti.com>
Bump to commit e3b1e426 which includes graceful degradation of fTPM TA
when RPMB storage is unavailable, preventing TA panic and memory dumps.
Signed-off-by: Shiva Tripathi <s-tripathi1@ti.com>
Signed-off-by: Ryan Eatmon <reatmon@ti.com>
Some k3-am62x-sk* overlays no longer apply to AM62A/AM62D boards,
update overlay list packaged for AM62D so that the k3-am62a7-sk*
overlays are included as well as the k3-am62x-sk* overlays
that do apply to AM62D board. This means ti/k3-am62x-sk generic
prefix must now be droped.
Signed-off-by: Judith Mendez <jm@ti.com>
Signed-off-by: Ryan Eatmon <reatmon@ti.com>
Update READMEs as follows:
* Don't point to CGit interface, as the old URL is no longer valid
* Use https:// instead of git:// protocol due to deprecation
* Add missing revisions and use the same ordering
* Correct some sub-layer dependencies
* Specify wrynose branch instead of master
Signed-off-by: Denys Dmytriyenko (TI) <denys@konsulko.com>
Signed-off-by: Ryan Eatmon <reatmon@ti.com>
Add a standalone initramfs-module-netsetup recipe that installs
83-netsetup, an initramfs-framework module that configures the network
interface via DHCP before the NFS root mount. The recipe's RDEPENDS
pulls in CPSW and its dependencies, so the kernel modules and DHCP
client are only included when NFS boot support is needed.
Include initramfs-module-netsetup in packagegroup-ti-core-initramfs
gated on :bsp-next, where AM65_CPSW_NUSS became a loadable module.
TI staging kernels (6.6/6.12/6.18) still have CPSW built-in (=y).
Signed-off-by: Moteen Shah <m-shah@ti.com>
Signed-off-by: Ryan Eatmon <reatmon@ti.com>
* Bump linux-bb.org from 6.12.43 to 6.12.57
* Update DTs for am62 platforms, such as beagleplay and pocketbeagle2
Signed-off-by: Denys Dmytriyenko (TI) <denys@konsulko.com>
Signed-off-by: Ryan Eatmon <reatmon@ti.com>
New recipes for GPU DDK 26.1. Update BSP configuration files with new
preferred recipe versions.
Signed-off-by: Antonios Christidis <a-christidis@ti.com>
Signed-off-by: Ryan Eatmon <reatmon@ti.com>
Apply upstream candidate fix for an assert in
weston_surface_apply_subsurface_order that fires when subsurfaces have
no views. Without this patch, the assert fires when running Chromium
and causes the display to crash.
Signed-off-by: Thorsten Lannynd <t-lannynd@ti.com>
Signed-off-by: Ryan Eatmon <reatmon@ti.com>
All supported TF-A handle this automatically based on the firmware, the
flag no longer does anything as it isn't checked in TF-A now. Remove this.
Signed-off-by: Andrew Davis <afd@ti.com>
Signed-off-by: Ryan Eatmon <reatmon@ti.com>
RM_WORK_EXCLUDE was used as a workaround for sourceipk in Arago, which
is long since gone.
Signed-off-by: Denys Dmytriyenko (TI) <denys@konsulko.com>
Signed-off-by: Ryan Eatmon <reatmon@ti.com>
Set additional LAYERRECOMMENDS for dynamic layers.
Signed-off-by: Denys Dmytriyenko (TI) <denys@konsulko.com>
Signed-off-by: Ryan Eatmon <reatmon@ti.com>
Set meta-ti-test layer to explicitly depend on meta-ti-bsp for ti-soc
definition.
Signed-off-by: Denys Dmytriyenko (TI) <denys@konsulko.com>
Signed-off-by: Ryan Eatmon <reatmon@ti.com>
As discussed on the mailing list, it is not always safe to have a
version-agnostic bbappend, as there could be different versions
of weston recipe provided by downstream layers, not compatible
with our patch, resulting in breakage. Make our bbappend again
version-specific.
Signed-off-by: Denys Dmytriyenko (TI) <denys@konsulko.com>
Signed-off-by: Ryan Eatmon <reatmon@ti.com>
When I rebased before moving from meta-arago to meta-ti, I accidentally
removed the wrong patch. Net result was this orphaned patch was left
and imported incorrectly. Remove it.
Signed-off-by: Ryan Eatmon <reatmon@ti.com>
We are moving the meta-arago-test layer to meta-ti-test. This commit
imports all check_yocto_rules from meta-arago-test.
Signed-off-by: Ryan Eatmon <reatmon@ti.com>
We are moving the meta-arago-test layer to meta-ti-test. This commit
imports the ti-test packagegroup that uses all of the previous commits..
Signed-off-by: Ryan Eatmon <reatmon@ti.com>
We are moving the meta-arago-test layer to meta-ti-test. This commit
imports all of the recipes under dynamic-layers.
Signed-off-by: Ryan Eatmon <reatmon@ti.com>
We are moving the meta-arago-test layer to meta-ti-test. This commit
imports all of the recipes under recipes-support.
Signed-off-by: Ryan Eatmon <reatmon@ti.com>
We are moving the meta-arago-test layer to meta-ti-test. This commit
imports all of the recipes under recipes-kernel.
Signed-off-by: Ryan Eatmon <reatmon@ti.com>
We are moving the meta-arago-test layer to meta-ti-test. This commit
imports all of the recipes under recipes-graphics.
Signed-off-by: Ryan Eatmon <reatmon@ti.com>
We are moving the meta-arago-test layer to meta-ti-test. This commit
imports all of the recipes under recipes-extende.
Signed-off-by: Ryan Eatmon <reatmon@ti.com>
We are moving the meta-arago-test layer to meta-ti-test. This commit
imports all of the recipes under recipes-devtools.
Signed-off-by: Ryan Eatmon <reatmon@ti.com>
We are moving the meta-arago-test layer to meta-ti-test. This commit
imports all of the recipes under recipes-connectivity.
Signed-off-by: Ryan Eatmon <reatmon@ti.com>
We are moving the meta-arago-test layer to meta-ti-test. This commit
imports all of the recipes under recipes-bsp.
Signed-off-by: Ryan Eatmon <reatmon@ti.com>
We are moving the meta-arago-test layer to meta-ti-test. This commit
imports all of the recipes under recipes-benchmark.
Signed-off-by: Ryan Eatmon <reatmon@ti.com>
We are moving the meta-arago-test layer to meta-ti-test. This is the
first commit that creates the layer directory structure. Follow on
patches will import all of the various recipes.
Signed-off-by: Ryan Eatmon <reatmon@ti.com>
The board specific firmware is added in the generic AM69 SoC include
instead of the AM69-SK board configuration. This follows the changes
already made for the J784s4 SoC and EVM board in
commit 27615406 ("conf: machine: j784s4: Move ti-eth-fw-j784s4 to EVM conf")
Move the machine-essential recommend from the SoC include
to the AM69-SK configuration.
Signed-off-by: Franz Schnyder <franz.schnyder@toradex.com>
Signed-off-by: Ryan Eatmon <reatmon@ti.com>
The am62lxx SoC default UART is ttyS0 since it doesn't have the
R5 cores taking up a UART. The k3 default SERIAL_CONSOLES
inherited includes ttyS2, which causes systemd to wait indefinitely
for a console that never appears, significantly increasing boot time
and degrading user experience.
Override SERIAL_CONSOLES to specify only ttyS0, allowing the boot
process to proceed normally without waiting for unavailable consoles.
Signed-off-by: Yogesh Hegde <y-hegde@ti.com>
Signed-off-by: Ryan Eatmon <reatmon@ti.com>
It was pointed out that j722s-evm-k3r5 was incorrectly setup and would
not build correctly out of the box. This is due to the fact that we set
both UBOOT_MACHINE and UBOOT_CONFIG by default. This was a mistake.
While looking into this newish feature I also found that another
platform was using a prepend, when it really should just be setting the
value for tie-test-builds.
Signed-off-by: Ryan Eatmon <reatmon@ti.com>
The `ti-eth-fw-j784s4` firmware is added in the generic J784s4 SoC
include, which is therefore used for all the J784s4-based machines.
That firmware seems to be developed specifically for the EVM, as it
takes control of pins used for the Ethernet board setup on the EVM. On
non-EVM boards, like the Aquila-AM69, those signals are used for other
functions, so enabling the firmware in the SoC include is too broad
and breaks functionality.
Move the machine-essential recommend from the SoC include
to the EVM configuration.
Signed-off-by: Franz Schnyder <franz.schnyder@toradex.com>
Signed-off-by: Ryan Eatmon <reatmon@ti.com>
LCPD Automation Script
Antonios Christidis
Judith Mendez
Andrew Davis
Suhaas Joshi
Yogesh Hegde
Shiva Tripathi
Denys Dmytriyenko (TI)
Moteen Shah
Thorsten Lannynd
Ryan Eatmon
Franz Schnyder