--

Commit: seco-ne/yocto/infrastructure/gitlab-ci@c23b4e5d

Some debug output for comparing build in cleaos and yocto-hamburg

--

Commit: seco-ne/tools/gf-emc-test-suite@705ea75c

helper: Add script to enable/disable the standard output on the terminal

This is required by some tests that exchange data on the default tty.

MODV-98

--

Commit: seco-ne/kernel/linux-guf@ffc9ae13

dts:imx6ull-nallino: set default CMA to 128 MiB

Set the default amount of allocated CMA in device tree to 128 MiB for Nallino

--

Commit: seco-ne/kernel/linux-guf@29a46793

dts:imx6qdl-santvend: Set default CMA to 256 MiB

Double the default CMA allocation amount to 256 MiB only for Santvend devices.

--

Commit: seco-ne/kernel/linux-guf@5e42d5cd

dts:imxqdl-san Set default CMA to 128 MiB

Set the default CMA allocation amount of SAN* device to 128 MiB in the device tree.

--

Commit: seco-ne/yocto/infrastructure/gitlab-ci@8a7e9c94

YT-291: Remove unnecessary dependencies from deploy stage.

  During the implementation of YT-93, some unrelated dependencies were
  mistakenly added to the deploy stage.
  This caused issues in the release build for certain custom projects.
  This commit removes those unnecessary dependencies.

--

Commit: seco-ne/tools/gf-emc-test-suite@51241f32

udp_echo: Change output path for binaries to ../bin

The binaries have not been installed into the image, because they are
expected to be put into ../bin.

--

Commit: seco-ne/yocto/infrastructure/gitlab-ci@081272c1

YT-260: Inherit restrictions from the parent Confluence page

  In each release, a Confluence page is created with expected restrictions
  specifying who can edit or update the page.
  Confluence automatically inherits "read" restrictions from the parent page,
  but "edit/update" restrictions are not inherited by design in Confluence.
  This ensures that restrictions are manually inherited from the parent page.

  Since the Confluence Python API module lacks functionality for
  setting restrictions, direct HTTP requests are used with the
  `requests` library instead of the Confluence module.

--

Commit: seco-ne/kernel/linux-imx-kuk@3462ce77

dts:arm64: Add GPIO names for MV expander and clean up existing names

Add GPIO names for the GPIO expander on the Modular Vision carrier.
Also reformat the existing GPIO names on the expander of the
SOM-SMARC-MX8M-Plus.

--

Commit: seco-ne/tools/gf-emc-test-suite@c6b94949

runtest: Add GPIOs to emission/immunity tests for MV

--

Commit: seco-ne/tools/gf-emc-test-suite@85f38e02

scripts: Add simple GPIO loopback script

This uses gpioset/gpioget to set an output and read back the value from
an input.

--

Commit: seco-ne/yocto/infrastructure/gitlab-ci@1740c4a8

YT-262: Hide bootloader artifacts on the Release page

  Customers often misinterpret the presence of bootloader artifacts
  on the Release page as a sanctioned method for updating the bootloader.
  Currently, these artifacts are intended for internal use only.

--

Commit: seco-ne/yocto/infrastructure/gitlab-ci@a2951867

YT-187: Update Confluence release page to feature UUU tools archive

      Currently, we distribute a variety of .sh and .bat scripts
      to aid in flashing devices using the UUU tool.
      To simplify the release page and enhance user-friendliness,
      all scripts have been bundled into a single archive.
      This commit updates the release page to represent the new archive,
      eliminating the display of individual UUU script files.

--

Commit: seco-ne/yocto/infrastructure/gitlab-ci@145ffd9f

YT-229: Initiate LAVA tests for imx8mp device type

  New, Modular Vision, device were added to LAVA rack
  Start running tasts on this device.
  Device type imx8mp

--

Commit: seco-ne/tools/gf-emc-test-suite@e1d9212c

scripts: Add gpu_load.sh script

The script converts the times in /sys/kernel/debug/gc/idle
into percent values over some time.
Tested on mx8mp.

PBB-46

Change-Id: I7f3b01b7ad8f3774533dff8c502033306f353cee

--

Commit: seco-ne/yocto/infrastructure/gitlab-ci@ff906931

YT-160: Workaround to prevent CI integration failure in 'merge-yocto' step

  The GitLab CI integration job frequently fails on the 'merge-yocto' step,
  specifically at the "git fetch" stage.
  The issue occurs because Git fails to fetch the .gitlab-ci submodule
  due to an incorrect reference, likely caused by the shallow clone
  (--depth=1) configuration, which limits the commit history.

  Extending the clone depth to --depth=2 helps prevent this error
  by providing a slightly deeper commit history,
  which resolves the reference issue.

--

Commit: seco-ne/tools/gf-emc-test-suite@f40be4f1

pbb:env: Use network test service and increase range for ADC

--

Commit: seco-ne/tools/gf-emc-test-suite@9cea6335

pbb:env: Add original test-script

--

Commit: seco-ne/yocto/infrastructure/gitlab-ci@37b396e8

YT-247: Automate Initiation of Build Stages in Scheduled Pipelines

  Enable automatic initiation of the build stage for the SDK
  and the test stage for pipelines within scheduled pipelines.

--

Commit: seco-ne/yocto/infrastructure/gitlab-ci@e3a76466

scripts: Add simple python script to convert license.manifest to md table

--

Commit: seco-ne/kernel/linux-imx-kuk@25b3628e

arm64:dts: Reduce drive-strength for USDHC2 to x2

Note: 0x2 equals a drive-strength of x4 while 0x4 equals x2.

--

Commit: seco-ne/kernel/linux-imx-kuk@10f71c8f

arm64:dts: Reduce drive-strength for USDHC2

The signal over-/undershoots to much with the drive-strength of x6.

--

Commit: seco-ne/3rd-party/kuk/uboot-imx-kuk@7db73164



board:nallino: add initial nallino board

It is based on Rocko U-Boot for Nallino.

Signed-off-by: Mikhail Vanyulin <mikhail.vanyulin@rtsoft.de>

--

Commit: seco-ne/3rd-party/kuk/uboot-imx-kuk@66a7daa6



seco_mx6_base_env.h: make loadaddresses configurable

Signed-off-by: Mikhail Vanyulin <mikhail.vanyulin@rtsoft.de>

--

Commit: seco-ne/tools/yocto-install@90c7cc26



fng-install-mx6-boot.sh: check for boards with custom u-boot

Signed-off-by: Mikhail Vanyulin <mikhail.vanyulin@rtsoft.de>

--

Commit: seco-ne/3rd-party/kuk/uboot-imx-kuk@235c56ee



seco-mx6: add defconfigs for santino and santino-lt boards

Signed-off-by: Mikhail Vanyulin <mikhail.vanyulin@rtsoft.de>

--

Commit: seco-ne/3rd-party/kuk/uboot-imx-kuk@4495e2d8



seco-mx6: initialize correct uart for console on santino-lt

Signed-off-by: Mikhail Vanyulin <mikhail.vanyulin@rtsoft.de>

--

Commit: seco-ne/3rd-party/kuk/uboot-imx-kuk@d65422b1



seco-mx6: add config options to select santino and santino-lt

CONFIG_BOARD_IS_SANTINO and CONFIG_BOARD_IS_SANTINO_LT were added and
will be used to determine for which board u-boot is being built.

Signed-off-by: Mikhail Vanyulin <mikhail.vanyulin@rtsoft.de>

--

Commit: seco-ne/3rd-party/kuk/uboot-imx-kuk@7a2a3758



dts: add dts for santino and santino-lt boards

Santino and Santino-LT are based on iMX6S and iMX6DL CPUs and do not
work properly with standard dts which is for iMX6Q CPU.

Signed-off-by: Mikhail Vanyulin <mikhail.vanyulin@rtsoft.de>

--

Commit: seco-ne/3rd-party/kuk/uboot-imx-kuk@3a7bb782



imx6: board-detection: add mx6dl to board ids

Signed-off-by: Mikhail Vanyulin <mikhail.vanyulin@rtsoft.de>

--

Commit: seco-ne/tools/gf-emc-test-suite@4e400ee5

mv:mx8mp: Fix issues with running MV emission and immunity tests

--

Commit: seco-ne/tools/gf-emc-test-suite@e58319d0

mv:mx8mp: fix emission and immunity tests for modular vision

Fixed the tests so they will run on MODV.
Removed unecessary tests.

--

Commit: seco-ne/tools/gf-emc-test-suite@79b163a4

mv:mx8mp: Add keypad_init to emission test

--

Commit: seco-ne/tools/gf-emc-test-suite@0ade1106

mv:mx8mp: Adjust emission and immunity to MV interfaces

--

Commit: seco-ne/tools/gf-emc-test-suite@c32469d8

mv:mx8mp: Copy pbb tests for MV + SOM-SMARC-MX8M-Plus

--

Commit: seco-ne/kernel/linux-imx-kuk@2a225181

[MYON2][CONXM] add support for ATM 7" LVDS

- TOUCH_INT and TOUCH_RESET removed from myon2-pinfunc.h because
  they may change
- legacy-drivers for backlight and lvds-bridge are no longer available
- pinctrl for touch added
- reserved@board section removed
- hog-gpios for pwm and backlight removed, this is done by the drivers
- more generic name for the hdmi bridge

--

Commit: seco-ne/kernel/linux-imx-kuk@36332200

|MYON2] new device tree for Myon2+ConXM+ATM0700L61CT added

--

Commit: seco-ne/kernel/linux-imx-kuk@a88d8cb6

arm64:dts: Add touch to MV + G156HAN02

--

Commit: seco-ne/yocto/infrastructure/gitlab-ci@f6b301dc

YT-227 Fix submit LAVA task for smoke tests

  Smoke tests use a single type of test (boot.jinja2).
  However, some parameters need to be dynamic, such as device type.
  To address this, the `test-plan-args-var` introduced
  what allows environment variables with extra parameters to be used
  in the LAVA job generation script.
  The pipeline has been adjusted accordingly.

--

Commit: seco-ne/yocto/infrastructure/gitlab-ci@be2a04ab

Enable verbose mode for submit_test.py

--

Commit: seco-ne/yocto/infrastructure/gitlab-ci@cf0071f2

YT-234: Fix ci-test pipelines (was broken with YT-65 MR)

 The YT-65 MR refactored the repo_checkout job,
 which is also used in CI-test pipelines.
 Added the necessary changes to the CI-test code.

--

Commit: seco-ne/tools/gf-emc-test-suite@981be2fa

rs485echo: Support all baudrates using termios2 ioctl directly from kernel header

The kernel has support for setting the baudrate as integer using BOTHER
as id with the termios2 struct. It seems that this is not yet supported
in the libc, so we are using kernel headers directly.

--

Commit: seco-ne/tools/gf-emc-test-suite@ccb24247

rs485echo: Support more baudrates using 'custom_devisor' way

--

Commit: seco-ne/yocto/infrastructure/gitlab-ci@667964a6

Fix typos

--

Commit: seco-ne/yocto/infrastructure/gitlab-ci@bd7e9366

YT-65 We perform a repository checkout in the manifest project for each build job.
  Each pipeline builds code for multiple MACHINES.
  This results in multiple downloads from each repository in the manifest.
  Occasionally, certain repositories may fail to fetch data.
  To avoid fetching the same data repeatedly for each build job in the pipeline,
  the `repo_checkout` stage is added.
  In this stage, the repo tool prepares the repository structure once
  and stores it in the cache.
  Subsequent build jobs use the pre-fetched data from the cache.

--

Commit: seco-ne/tools/gf-emc-test-suite@1248f778

scripts:network: Clean-up UDP echo network scripts

--

Commit: seco-ne/tools/gf-emc-test-suite@3bf94dc0

tests:network: Add UDP echo server and client

The server echos incoming packets on port 7 back to the sender.
The echo services used to be available in initd, but are not present in systemd based
systems anymore.
The client sends test packets with a sequence number every 100 ms.

--

Commit: seco-ne/tools/gf-emc-test-suite@e4619854

scripts:network: Add UDP network test service

--

Commit: seco-ne/kernel/linux-imx-kuk@9ad2bc3d

panel-dt/panel-lvds: switch sel6-8-gpio to low in unprepare

There is an issue with a display showing artifacts after leaving the display
blank for some time. If sel6_8 line stays high during the power-off of the
display, the display gets side-powered by the sel6_8 line. And the power
sequence during the next power up does not reset the display in a good state.

So switch the sel6_8 gpio to low, when the VDD-LCD is powered of, and enable it
to the correct state, just before the VDD-LCD is powered on again.

(cherry picked from commit 4f570f563b1ea85d901b8db98af6ad629c4aa1ba)

sel68

--

Commit: seco-ne/kernel/linux-guf@4f570f56

panel-dt: switch sel6-8-gpio to low in unprepare

There is an issue with a display showing artifacts after leaving the display
blank for some time. If sel6_8 line stays high during the power-off of the
display, the display gets side-powered by the sel6_8 line. And the power
sequence during the next power up does not reset the display in a good state.

So switch the sel6_8 gpio to low, when the VDD-LCD is powered of, and enable it
to the correct state, just before the VDD-LCD is powered on again.

--

Commit: seco-ne/yocto/infrastructure/gitlab-ci@a7e07f19

YT-180 Generation of the firmware package failed due to a typo
  in the code that checks of existance of the variable.
  What caused  TypeError: unhashable type: 'list'

--

Commit: seco-ne/yocto/infrastructure/gitlab-ci@7bd78810

PBB-32 Fix suffix addition in FNGo pipeline for custom projects
 due to an issue in GitLab where escaped variables are not correctly
 passed to child pipelines, a workaround is required.
 As a result, we use raw variables.
 The DEFERRED_RELEASE_VERSION variable was introduced for this purpose.
 However, for custom builds, we have an additional place where we modify
 the DEFERRED_RELEASE_VERSION variable to add a suffix.
 Therefore, extra interim variables need to be added to handle the workaround
 and ensure we get the raw variable in the end.

--

Commit: seco-ne/3rd-party/kuk/uboot-imx-kuk@b11d5d7b

cmd:seco_ne: Export bootmode on ENV bootselect set

The bootmode environment variable wasn't set to the new bootselect value
when setting the boot order.

--

Commit: seco-ne/3rd-party/kuk/uboot-imx-kuk@70b5836b

cmd:seco-ne: Add oneshot modes to ENV bootselect

Add oneshot, oneshotA and oneshotB modes to the ENV bootselect
implementation.

--

Commit: seco-ne/yocto/infrastructure/gitlab-ci@83f4f610

Take group path from group instead of options

--

Commit: seco-ne/yocto/infrastructure/gitlab-ci@93eabeb3

YT-157 Adjust GitLab Backup Script to avoid shared projects
 Ensure backups are only processed for direct member projects of a group.
 If a project is shared with a group, backing it up without the appropriate
 permissions can cause the backup to fail.

--

Commit: seco-ne/tools/yocto-install@520c20f1

install-scripts: initial support of GPT

Added a new parameter to choose between partition table type:
GPT (GUID Partition Table) or MBR (Master Boot Record).

The default partition table type remains MBR for SECO i.MX6/8 boards,
and GPT can be activated by passing a "--partition-table-type=gpt" parameter
to fngsystem-self-init.sh script.

--

Commit: seco-ne/kernel/modules/ilitek@44c49560

driver:irq: Remove disable_irq/irq_enable from threaded_irq routine

The disable/enable in the irq routine leads to a race-condition when
using an IRQ that is connected to a "slow bus" (e.g. a GPIO expander
connected via I2C). In general, the disable/enable should not be
required when using a threaded_irq, since the Kernel disables the IRQ
while in the routine anyways.

--

Commit: seco-ne/kernel/modules/ilitek@f8c900f3

repo: Ignore Makefile

--

Commit: seco-ne/kernel/modules/ilitek@222578ee

repo: Add gitignore file

--

Commit: seco-ne/tools/gf-emc-test-suite@da00ff31

setup-network: Add guard to not modify the interface just used for ssh access

--

Commit: seco-ne/yocto/infrastructure/gitlab-ci@77a91f92

YT-93 Improve release workflow

--

Commit: seco-ne/yocto/infrastructure/gitlab-ci@edbfded2

YT-164 Sort artifacts on confluence release page for FNG system

--

Commit: seco-ne/kernel/linux-guf@a1781083



logo: remove fng system logo

It will be loaded as external logo.

Signed-off-by: Mikhail Vanyulin <mikhail.vanyulin@rtsoft.de>