This commit updates the CC6 and CC6QP platform configurations to use the latest
NXP BSP based on U-Boot v2024.04, unifying support across all NXP-based
platforms.
https://onedigi.atlassian.net/browse/DEL-9758
Signed-off-by: Arturo Buzarra <arturo.buzarra@digi.com>
This commit updates the CC6 and CC6QP platform configurations to use the latest
NXP BSP based on Linux kernel v6.6, unifying support across all NXP-based
platforms.
https://onedigi.atlassian.net/browse/DEL-9758
Signed-off-by: Arturo Buzarra <arturo.buzarra@digi.com>
With the updated firmware-ele-imx recipe, the installation logic now
requires SECOEXT_FIRMWARE_NAME to be empty when no extra firmware is
to be installed.
https://onedigi.atlassian.net/browse/DEL-9748
Signed-off-by: Javier Viguera <javier.viguera@digi.com>
This commit updates the Digi custom .bbappend recipes for FIP and TF-A to align
with the latest ST BSP release, based on the openstlinux-6.6-yocto-scarthgap-mpu-v25.06.11
tag for Yocto 5.0 (scarthgap).
https://onedigi.atlassian.net/browse/DEL-9734
Signed-off-by: Arturo Buzarra <arturo.buzarra@digi.com>
Move the address where the fitImage is loaded after the addresses
where the binaries in it are decompressed. This way, the fitImage
can grow without size restrictions.
The memory map now looks like this:
0xC0000000 Start of memory
|
| (32 MiB)
v
0xC2000000 Kernel loadaddr ($loadaddr)
|
| (32 MiB)
v
0xC4000000 DTB/DTBO load address ($fdt_addr)
| (4 MiB)
v
0xC4400000 Init ram disk ($initrd_addr)
|
|
| (64 MiB)
|
v
0xC8400000 ZIP/fitImage address ($fit_addr_r)
|
~
|
v
Signed-off-by: Hector Palacios <hector.palacios@digi.com>
IMAGE_FSTYPES was declared on each platform config file, but it has the
same values for all platforms depending on the storage media (mmc or mtd)
and whether read-only is enabled.
Move the conditional weak assignment to digi-default.inc and remove it from
each platform config.
In the case of STM platforms, since IMAGE_FSTYPES is weak-assigned by STM
layer, we still need to append/remove from it inside the platform config,
but move it to the family includes, rather than declaring it on each
specific platform.
Signed-off-by: Hector Palacios <hector.palacios@digi.com>
The generation of the sdcard image takes time and resources, and
it's not involved in the Get Started.
This can be easily re-enabled by appending the variable in the
project local.conf.
Append the variable in the build scripts, to facilitate its usage
on release builds.
Signed-off-by: Hector Palacios <hector.palacios@digi.com>
Fixes commit b143804dbb, since in nativesdk
context MACHINE_FEATURES is reset to SDK_MACHINE_FEATURES, causing OP-TEE
building tools to be missing from the generated SDK.
https://onedigi.atlassian.net/browse/DEL-9663
Signed-off-by: Javier Viguera <javier.viguera@digi.com>
This commit updates the memory layout to properly allocate space for the
different FIT image components, while ensuring total memory usage stays below
128 MiB. This avoids overlaps and ensures correct loading on memory-constrained
variants.
Final memory map:
Start of memory: # 0xC0000000
# |
# | 32 MiB reserved
# v
FIT image load address: # 0xC2000000
# |
# | 32 MiB for FIT image
# v
Kernel load address: # 0xC4000000
# |
# | 32 MiB for Kernel
# v
DTB/DTBO load address: # 0xC6000000
# |
# | Size for DTB/DTBO
# v
Total memory mapped: 96 MiB
https://onedigi.atlassian.net/browse/DEL-9634
Signed-off-by: Arturo Buzarra <arturo.buzarra@digi.com>
In theory, we already had the necessary changes to remove these images, but
two things needed tweaking:
* MULTIUBI_BUILD values use underscores instead of hyphens, so use
underscores to properly remove the default value inherited from
BOOTDEVICE_LABELS.
* STM used to incorporate a custom "stmultiubi" image type in the stm32mp
builds, but they've replaced this with the upstream "multiubi" type.
Reflect this change to avoid generating additional UBI/UBIFS images in
our builds.
Signed-off-by: Gabriel Valcazar <gabriel.valcazar@digi.com>
Commit 9c3916da94 added INSANE_SKIP
"32bit-time" to certain recipes that use 32bit APIs on the ccmp1
SOMs, but forgot to include the `pn-` prefix to really apply to
those recipes.
While on it, add two additional ones on recipes used by NXP 32-bit
platforms.
Signed-off-by: Hector Palacios <hector.palacios@digi.com>
That includes several CVE patches not present of meta-freescale's
23.2.5.imx recipe (which is based in exactly the same revision).
Similar change was done in NXP's meta-imx (see commit
99ceb057fcfdc8151c1488089d5f22363dfdb6d7).
Signed-off-by: Javier Viguera <javier.viguera@digi.com>
The binaries of certain recipes use 32-bit APIs (ioctl, stat) that produce
build warnings. Add INSANE_SKIP to prevent the warnings.
Signed-off-by: Hector Palacios <hector.palacios@digi.com>
ALSA_LIST variable is weakly set through an override (stm32mp1common or
stm32mp2common). When the override is in place, the hard assignment of
the plain variable is not enough, and the value weakly set in
meta-st-stm32mp layer takes precedence over the plain variable hard
assignment.
Clear the variable with the override and move it from machine config files
to the include file.
Signed-off-by: Hector Palacios <hector.palacios@digi.com>
Move BOOTSCHEME_LABELS from common include to platform config
Default to:
- 'opteemin' for ccmp15
- 'optee' for ccmp13
This doesn't change the behavior present in DEY-4.0, where OPTEE in ccmp15
was minimal and running on non-secure DDR.
Signed-off-by: Hector Palacios <hector.palacios@digi.com>
This commit removes kernel headers from the SDK package list to
prevent overwriting the default ones, as there are currently no
ST-specific headers required.
Reference: ebadb27d60
Signed-off-by: Hector Palacios <hector.palacios@digi.com>
This commit integrates optee-client support from the meta-st-openstlinux layer,
based on the openstlinux-6.6-yocto-scarthgap-mpu-v25.03.19 tag.
https://onedigi.atlassian.net/browse/DEL-9442
Signed-off-by: Arturo Buzarra <arturo.buzarra@digi.com>
This commit introduces the necessary settings for the kernel-fitimage class to
enable FIT image generation.
https://onedigi.atlassian.net/browse/DEL-9442
Signed-off-by: Arturo Buzarra <arturo.buzarra@digi.com>
The FIP flavor for OP-TEE + USB is managed in the meta-st-stm32mp layer through
the "optee-programmer-usb". However, since we do not require the additional
overhead introduced by the STM32CubeProgrammer tool, this commit introduces a
new FIP configuration based on OP-TEE for booting from USB.
https://onedigi.atlassian.net/browse/DEL-9442
Signed-off-by: Arturo Buzarra <arturo.buzarra@digi.com>
The sign script has been updated to support AHAB-based modules like the CCIMX8X.
As a result, there is no longer a need to maintain two separate recipes for the
signing scripts. This commit unifies them into a single recipe.
Signed-off-by: Arturo Buzarra <arturo.buzarra@digi.com>
Due to a change in systemd the default home directory is now "/root".
Modify our recipes to match with this change.
Signed-off-by: Francisco Gil <francisco.gilmartinez@digi.com>
Update to 4.4.0 version following 'lf-6.6.52_2.2.0' NXP release.
This has not been released in meta-freescale yet, so reuse 4.2.0
recipes and apply the changes from meta-imx commit
900356ea1bf71854053266eec4b92adf4552624c.
https://onedigi.atlassian.net/browse/DEL-9417
Signed-off-by: Gonzalo Ruiz <Gonzalo.Ruiz@digi.com>
This doesn't change the behavior of the linux-dey recipe, all it does is make
the LINUX_GIT_URI variable accessible to any meta-digi recipe.
Signed-off-by: Gabriel Valcazar <gabriel.valcazar@digi.com>
Now all platforms support setting the GPIO name instead of an index
as the configuration for the GPIO-enabled secure console.
Repurpose the TRUSTFENCE_GPIO_ENABLE macro to directly set the GPIO
name.
Also take the opportunity to remove TRUSTFENCE_GPIO_ENABLE and
TRUSTFENCE_CONSOLE_PASSPHRASE_ENABLE commented definitions from bbclass
and conf files, and remove a duplicated TRUSTFENCE_CONSOLE_DISABLE
definition from ccmp1.inc.
Signed-off-by: Gonzalo Ruiz <Gonzalo.Ruiz@digi.com>
When ST_OPTEE_DEBUG_TRACE=0, the source code default to a LOG_LEVEL=2
which prints a lot of messages on the boot log.
Reduce it to LOG_LEVEL=1.
Signed-off-by: Hector Palacios <hector.palacios@digi.com>
This override was recently added to meta-freescale's graphical package recipes
to easily distinguish between i.MX platforms that use the Vivante driver and
those that use the Mali one. Without this change, a lot of these recipes are
broken, so sync with the latest changes in meta-freescale's imx-base.inc file.
Signed-off-by: Gabriel Valcazar <gabriel.valcazar@digi.com>
We tested all tarball compression formats supported by poky, and although bz2
has faster (de)compression time, xz is better in terms of compression ratio:
for all of the rootfs tarballs generated for our currently supported
images/platforms (15 at the moment), the xz format saves an average of 30 MiB
per tarball compared to bz2, totalling up to 450 MiB.
No extra dependencies are pulled in, since xz-native is already being pulled in
for all of our image builds, so the only drawback to this change is the
increased compression time (+7.34s on average per tarball).
https://onedigi.atlassian.net/browse/DEL-9459
Signed-off-by: Gabriel Valcazar <gabriel.valcazar@digi.com>
This commit clears the Cortex-M33 applications list to prevent including all
the Cortex-M33 demo applications from the ST layer. In the future, if we have
custom Digi examples, they should be added here.
Signed-off-by: Arturo Buzarra <arturo.buzarra@digi.com>
.ext4 files are identical to .ext4.gz, but uncompressed. .tar.xz files are the
exact same as .tar.bz2, but compressed with a different format.
Remove these artifacts to reduce storage overhead and to match the rest of our
platforms.
Signed-off-by: Gabriel Valcazar <gabriel.valcazar@digi.com>
This replicates meta-imx commit addc08b02f9a, which was added to be able to
re-use packages in i.MX8/i.MX9 builds. Without this change, there are
compilation errors when building recent versions of the onnxruntime package
because some .S files are built with specific -march switches that conflict
with our current tunings. For example, when building for the ccimx8x-sbc-pro:
cc1: error: switch '-mcpu=cortex-a35+crc+crypto' conflicts with '-march=armv8.2-a+fp16' switch [-Werror]
[...]HalfGemmKernelNeon.S:151: Error: selected processor does not support `fmla v20.8h,v16.8h,v0.h[0]'
Using the generic tuning solves this issue, and according to the log in the
original meta-imx commit, this has minimal to no impact on binaries, so use
said tuning in all relevant platforms.
Signed-off-by: Gabriel Valcazar <gabriel.valcazar@digi.com>
The value of this variable is always the same within a platform, so it doesn't
make sense to have different variables for it in each recipe that requires it.
While at it, it seems like all currently supported platforms in DEY 5.0 use the
same value of "wayland-1", presumably because they all use similar versions of
wayland/weston. Set this as the default value for all platforms for now. As we
support more platforms, we can adjust this value if needed, but if it ends up
being the same for all platforms, we can just hardcode it.
https://onedigi.atlassian.net/browse/DEL-9404
Co-authored-by: Isaac Hermida <isaac.hermida@digi.com>
Signed-off-by: Isaac Hermida <isaac.hermida@digi.com>
Signed-off-by: Gabriel Valcazar <gabriel.valcazar@digi.com>
This commit removes kernel headers from the SDK package list to prevent
overwriting the default ones, as there are currently no ST-specific headers
required.
https://onedigi.atlassian.net/browse/DEL-9381
Signed-off-by: Arturo Buzarra <arturo.buzarra@digi.com>
This commit updates the default FIP generation variables to align with the
common stm32mp2 configuration files from the latest meta-st-stm32mp layer,
based on the openstlinux-6.6-yocto-scarthgap-mpu-v24.11.06 tag for Yocto 5.0
(Scarthgap).
https://onedigi.atlassian.net/browse/DEL-9381
Signed-off-by: Arturo Buzarra <arturo.buzarra@digi.com>
Update ccmp2 platform to use the latest U-Boot v2023.10 based on the
openstlinux-6.6-yocto-scarthgap-mpu-v24.11.06 tag for Yocto 5.0 (scarthgap).
https://onedigi.atlassian.net/browse/DEL-9381
Signed-off-by: Arturo Buzarra <arturo.buzarra@digi.com>
Update ccmp2 platform to use the latest Linux kernel v6.6.48 based on the
openstlinux-6.6-yocto-scarthgap-mpu-v24.11.06 tag for Yocto 5.0 (scarthgap).
https://onedigi.atlassian.net/browse/DEL-9381
Signed-off-by: Arturo Buzarra <arturo.buzarra@digi.com>
This commit removes the gcnano recipes integrated into meta-digi, as a new
version (v6.4.19) has been released in the latest ST BSP, based on the
openstlinux-6.6-yocto-scarthgap-mpu-v24.11.06 tag for Yocto 5.0 (Scarthgap).
https://onedigi.atlassian.net/browse/DEL-9381
Signed-off-by: Arturo Buzarra <arturo.buzarra@digi.com>
This commit removes all outdated OPTEE-OS recipes and synchronizes the Digi custom
.bbappend with the latest v4.0.0 from the ST BSP release, based on the
openstlinux-6.6-yocto-scarthgap-mpu-v24.11.06 tag for Yocto 5.0 (scarthgap).
https://onedigi.atlassian.net/browse/DEL-9381
Signed-off-by: Arturo Buzarra <arturo.buzarra@digi.com>
This commit removes all outdated TF-A recipes and synchronizes the Digi custom
.bbappend with the latest v2.10 from the ST BSP release, based on the
openstlinux-6.6-yocto-scarthgap-mpu-v24.11.06 tag for Yocto 5.0 (scarthgap).
https://onedigi.atlassian.net/browse/DEL-9381
Signed-off-by: Arturo Buzarra <arturo.buzarra@digi.com>
Arturo Buzarra
Javier Viguera
Hector Palacios
Gabriel Valcazar
Isaac Hermida
Francisco Gil
Gonzalo Ruiz