BCS 746-000945

Felix Gerking authored 2 years ago and Jonas Höppner committed 2 years ago

BCS 746-000943

Felix Gerking authored 2 years ago and Jonas Höppner committed 2 years ago

Now the correct phy reg is set explicitly. Use the driver specific bindings
for the phy reset and set the correct reset timings.

BCS 746-000932

BCS 746-000936

Tobias Kahlki authored 2 years ago and Jonas Höppner committed 2 years ago

First draft of the integration of the LCDIF and the PWM backlight.
The changes haven't been tested yet.

BCS 746-000944
BCS 746-000951

Tobias Kahlki authored 2 years ago and Jonas Höppner committed 2 years ago

BCS 746-000953

Tobias Kahlki authored 2 years ago and Jonas Höppner committed 2 years ago

BCS 746-000952

Tobias Kahlki authored 2 years ago and Jonas Höppner committed 2 years ago

The MC3 uses an AC97 codec for audio/touch input.
Added a first version of the AC97 codec to the devicetree.

Note: The communication couldn't be tested yet and might not work.

Tobias Kahlki authored 2 years ago and Jonas Höppner committed 2 years ago

The eMMC is connected to usdhc3, but we need it on mmc0.
Re-ordered the mmcs to reflect this change.

--

Commit: seco-ne/yocto/infrastructure/gitlab-ci@d25cdebf

Really fix potential error message in retrigger job

TypeError: 'NoneType' object is not subscriptable

Example:
https://git.seco.com/seco-ne/yocto/manifest/-/jobs/129201

Add the device tree for the touch display set SCX1001255GGU06 including
the LCD panel FG1001L0DSSWMG01 and an SIS touch.
The panel is connected via LVDS.

The current state does not work as expected. It seems the MIPI signals
are not clocked out correctly.

--

Commit: seco-ne/yocto/infrastructure/gitlab-ci@3eef782e

deploy_gitlab_ci: integrate only correct projects into the manifest

Previously we integrated all projects into the manifest, even those that
belonged to a different manifest branch. That led to all manifests
having the same project revisions, even if they should be different due
to different intgrated project branches.

Save a list of integration sources for each manifest branch separately
now, so that we can check which project belongs to which manifest branch
later.

--

Commit: seco-ne/yocto/infrastructure/gitlab-ci@2b851ac6

Fix potential error message in retrigger job

TypeError: 'NoneType' object is not subscriptable
Example:
https://git.seco.com/seco-ne/yocto/infrastructure/ci-test/minimal-manifest/-/jobs/127610

--

Commit: seco-ne/yocto/infrastructure/gitlab-ci@c3fd0ecc

Check if variable is defined before checking it value

--

Commit: seco-ne/yocto/infrastructure/gitlab-ci@c1be0f4b

Fix jinja2 "or" operator

--

Commit: seco-ne/yocto/infrastructure/gitlab-ci@8da534ec

Run integration pipeline for target branch only

When we're integrating a single branch into a single manifest branch,
there is no need to run the intgeration pipeline for all configured
integrations, but only for those that match the target branch of the
current merge request.

--

Commit: seco-ne/yocto/infrastructure/gitlab-ci@8c94b725

ci-test: add job to check for correct foo branch

This should catch an error where during a gitlab-ci merge process the
wrong foo branch is integrated into the manifest, which happened in
Yocto.

--

Commit: seco-ne/yocto/infrastructure/gitlab-ci@163c7e1e

Add yml extension to jinja2 templates

--

Commit: seco-ne/yocto/infrastructure/gitlab-ci@59523aa0

check job: disable verbose output

--

Commit: seco-ne/yocto/infrastructure/gitlab-ci@28bf0065

retrigger: add job states to log message

--

Commit: seco-ne/yocto/infrastructure/gitlab-ci@fd742017

deploy_gitlab_ci: fix target branch in integration merge request

We need to pass the correct target branch to the merge request creation
instead of using the project's default branch.

--

Commit: seco-ne/yocto/infrastructure/gitlab-ci@38379173

Run yamllint job on gitlab-ci integrations

Otherwise no pipeline is created at all in these merge requests, which
prevents the MR from being merged.

--

Commit: seco-ne/yocto/infrastructure/gitlab-ci@dce4f8a3

changelog_generator: Fix machine project ID

We have deleted the original project due to git corruption. Replace with
the new project's ID.

--

Commit: seco-ne/yocto/infrastructure/gitlab-ci@a7aa7c26

deploy_gitlab_ci: fix project argument passing

--

Commit: seco-ne/yocto/infrastructure/gitlab-ci@8a38f194

Put all manifest intgeration jobs into a single stage

--

Commit: seco-ne/yocto/infrastructure/gitlab-ci@a09564c1

deploy_gitlab_ci: explicitly pass gitlab-ci project and branch

--

Commit: seco-ne/yocto/infrastructure/gitlab-ci@172e7625

Improve error messages

--

Commit: seco-ne/yocto/infrastructure/gitlab-ci@f0f5a370

Improve naming of functions and variables

--

Commit: seco-ne/yocto/infrastructure/gitlab-ci@c8b484ca

Retrigger MRs from all integrating projects

Instead of getting the projects from the manifest file, search them via
their INTEGRATION variable.

Improve log messages on this occasion: print URL instead of just IDs.

--

Commit: seco-ne/yocto/infrastructure/gitlab-ci@5e75bfc1

Deploy gitlab-ci to multiple branches

Add different integration jobs for different manifest branches. The
merge stage only has one job for each manifest, though, because
otherwise we cannot guarantee a consistent state for all branches.

Extend the deploy script for this purpose, so that it can deploy to
multiple manifest branches at once.

--

Commit: seco-ne/yocto/infrastructure/gitlab-ci@771dcff1

Fix temporary dir path

We were passing the complete TemporaryDirectory object to the repo clone
function instead of just the path string, resulting in the repo being
cloned into a local dir "<TemporaryDirectory '/tmp/tmphwakypf8'>".

Fix this to actually use the generated temp dir.

This change makes it necessary to keep the TemporaryDirectory object
reference until we don't need the directory anymore, otherwise it will
be removed immediately.

--

Commit: seco-ne/yocto/infrastructure/gitlab-ci@dce219c8

Print less important output in grey color

--

Commit: seco-ne/yocto/infrastructure/gitlab-ci@5e36261d

deploy_gitlab_ci: remove hard-coded integration targets

Get targets dynamically using the INTEGRATION variable instead.

--

Commit: seco-ne/yocto/infrastructure/gitlab-ci@94cd6781

Recurse to child pipelines on retrigger

This makes it possible to retrigger jobs that are not part of the
given pipeline but in any nested child pipeline of it.

--

Commit: seco-ne/yocto/infrastructure/gitlab-ci@fb1219a7

Ignore __pycache__ in scripts directory

--

Commit: seco-ne/yocto/infrastructure/gitlab-ci@0c29c995

Refactoring: rename arguments for consistency

--

Commit: seco-ne/yocto/infrastructure/gitlab-ci@d461090b

Remove hard coded project and branch variables

MANIFEST_PROJECT, MASTER_BRANCH_MANIFEST, MASTER_BRANCH_PROJECT

--

Commit: seco-ne/yocto/infrastructure/gitlab-ci@a4aae414

Add rules for integration pipeline

--

Commit: seco-ne/yocto/infrastructure/gitlab-ci@bfa38b24

Improve log messages

--

Commit: seco-ne/yocto/infrastructure/gitlab-ci@1c9670d9

commit_and_push: do not pass branch

Branch can be implicitly determined via the repo.

BCS 746-000931

The pull-up of the CMD line of the eMMC isn't placed on the hardware.
Therefore, the internal pull-up of the pad is required, for the CMD line
to function properly.
Also increased the clock rate to 400 MHz (see NXP EVK) and assigned the
correct regulators (the eMMC uses fixed power rails for 1.8 V and 3.3 V).

Note: The eMMC isn't working yet.

Added nodes for the generic voltages and three nodes for the eMMC and
SD-Card (the SD-Card requires a 1V8 and 3V3 rail for the high-speed mode).

What works:
* fngsystem (kirkstone) ramfs boot via uuu
* for UART2 an alias to serial0 is set
* ethernet, at the moment no phy-reg is set explictily
* RTC, we need to check if specific backup source settings are required:
root@FLASH-N-GO:~# dmesg | grep rtc
[    1.519835] snvs_rtc 30370000.snvs:snvs-rtc-lp: registered as rtc1
[    1.960681] rtc-rv8803 0-0032: Voltage low, temperature compensation stopped.
[    1.967833] rtc-rv8803 0-0032: Voltage low, data loss detected.
[    1.977571] rtc-rv8803 0-0032: Voltage low, data is invalid.

BCS 746-000818

The Trizeps 8 Plus v1r3 revision has a new audio codec that is not
yet supported. Disable the old codec anyways to prevent from
unexpected behaviour.

A duplicate of the seco logo was present in the logos dir.

The image file for the fngsystem boot logo was missing.
Re-added the file to fix the kernel build.

Both logos need to be enabled by default. The logo that is added to the
kernel is picked through the defconfig and the override in the kernel
include file.

Cherry-pick from af765e15

Felix Gerking authored 2 years ago and Tobias Kahlki committed 2 years ago

Added wakeupsource and interrupt for RTC rv3028.
Now wakealarm the is available at: /sys/class/rtc/rtc0/wakealarm

BCS 746-000767

(cherry picked from commit a72cb6a2)

Jonas Höppner authored 3 years ago and Tobias Kahlki committed 2 years ago

Measured the pixelclock on the lvds interface the frequencies generated
by media_dips2_pix clock do not match the calculated values by the
driver. The driver selects a devider between 1 and 64 ( according to the
    ref manual) but the measured frequencies only end up with
frequencies matching a devider from 1,7,14,21,35.
This patch changes the selected of deviders to that list, so the
frequency is rounded to a actually working frequency.

BCS 746-000495

(cherry picked from commit 70382bea)

Jonas Höppner authored 3 years ago and Tobias Kahlki committed 2 years ago

The imx8mp ldb driver had an hardcoded overwrite for the pixelclock as
the video_pll is not able to generate all needed frequencies.
But the hardcoded value only fits to one mode, probably used on the
EVK.
Removing this override leads to a pixelclock that may be generated with
a devider from the PLL, which ends up with a frquency much closer to what
is expected.

Also it is not the ldb clock but the lcdif clock used for this, so this
seems to be the wrong place for modifying the clock anyway.

BCS 746-000502

(cherry picked from commit 86df943d)

Felix Gerking authored 2 years ago and Tobias Kahlki committed 2 years ago

The pwm-legacy-backlight driver is removed in the 5.15 kernel. Use
pwm-backlight instead.

BCS 746-001009

For some reason, without the hdmiphy the device stalls during the boot
process. Further investigation into why this happens is required.

BCS 746-001009

The ethernet phy and audio codec have changed in the new revision of the
Trizeps 8 Plus. Removed the unwanted properties from the ethphy0 node
and set the address to 0.
The audio codec hasn't been added yet.

Cherry-pick from a40f7a72

BCS 746-000835

Changed the driver to probe without interrupt gpio, as the irq is not
routed on the trizeps8plus board. It seems to be used for debug output
on pmic errors only, so it should work without.

Cherry-pick from 7739e8c9


(cherry picked from commit 3ec08c4d)

The first approach of selecting a RTC didn't work as expected.
The active RTC must reside on the rtc0 alias because otherwise
the hwclock utilities can't access the RTC.
For this to work, we now have a placeholder in the main device tree
that is filled with the actual values in one of the two overlays.
This breaks with the naming conventions for I2C bus devices, but
right now, we don't have a better solution.

BCS 746-000877

(cherry picked from commit 11963a85)

The TANARO will be produced with one of two RTC variants.
One of them will be activated during the installation by
applying one of the overlays.

BCS 746-000877

(cherry picked from commit 9b1710b2)