changed Makefile.test, updated submodule for arduino-core-tests

[Frederikwag]

By creating this pull request you agree to the terms in CONTRIBUTING.md.
https://github.com/Infineon/.github/blob/master/CONTRIBUTING.md
--- DO NOT DELETE ANYTHING ABOVE THIS LINE ---

CONTRIBUTING.md also tells you what to expect in the PR process.

Description
The submodule "Arduino-Core-Test" in the repo "XMC-for-Arduino" linked to an older version. The version of "Arduino-Core-Test" was updated to the newest one, the different tests were executed and it was documented which tests are also working. The files where adapted accordingly.

Related Issue
It's not an issue, but tasks that still need to be done:

1. Interrupts: It exists an open ticket for the interrupt-problem. A sending interrupt is not received by an receiving pin. The existing test shows that 4/6 tests don't pass, which CAN be traced back to this. Therefore, the interrupt-functionality needs to be solved first. BUG-Ticket needs to be created!
2. Timing: the following test function is not passed: RUN_TEST_CASE(time_single_internal, testDelayMicroseconds)
   --> Expected Delay: 500000 microseconds; Actual Delay: 506483 microseconds; Tolerance: 100 microseconds (result: values not within delta 100)
3. IIC is not working for 1 board only: make FQBN=Infineon:xmc:XMC4700_Relax_Kit PORT=COM28 UNITY_PATH=\Unity test_wire_connected1_pingpong monitor --> There is an open ToDo

Context
Quiet similar to previous push (added-spi), but tried to reduce amount of changes. Therefore checked code again.

[ederjc]

Hi @Frederikwag, your commit message check fails. Please have a look here and then here.

Please let me know if you have questions.

Regarding the failing HIL checks we can see what's wrong once the commit checks are succeeding.

[jaenrig-ifx]

The fqbn vendor renaming and the ports are specific to @Frederikwag local setup.
@LinjingZhang what is the idea behind this test makefile?
I see it as a preliminary list of the tests that has been checked so far (and will be run in the HIL). Is that it?
If we want (we should if it is not temporary) enable any user to run the list of tests locally, we should provide a way to pass the hardcoded value as parameters.

[LinjingZhang]

The fqbn vendor renaming and the ports are specific to @Frederikwag local setup. @LinjingZhang what is the idea behind this test makefile? I see it as a preliminary list of the tests that has been checked so far (and will be run in the HIL). Is that it? If we want (we should if it is not temporary) enable any user to run the list of tests locally, we should provide a way to pass the hardcoded value as parameters.

Yes, it is. This is just a test command that we have validated and will be used in HIL in the future.
We can also add it to .gitignore, but for now I think it's a good reference for developers.

[LinjingZhang]

Please revise based on the review comments.
I remember we also had some test failures (invalid pins) when running gpio, not found in the description

[LinjingZhang]

for gpio, we do need a test_config.h for pin definition right? Should it be added to the PR?

[LinjingZhang]

correct added. good 👍

[LinjingZhang]

You can ask Julian to help with commit msg ;)

[LinjingZhang]

why adding echo command in .gitignore?

[Frederikwag]

Because everything what is related to SPI is a different branch.

[LinjingZhang]

Please check: https://www.geeksforgeeks.org/what-is-git-ignore-and-how-to-use-it/
and: https://www.geeksforgeeks.org/echo-command-in-linux-with-examples/

to understand real reason why you want to use echo and add the path in .gitignore and why I ask you to remove this code.

[LinjingZhang]

Unfortunately, this time the submodule points to a unknow hash tag... please double check

[LinjingZhang]

nice.

[LinjingZhang]

I forgot to mention that it would be great to add the name of the board used here. Because xmc has a lot of different targets