foundries.io CI: tests/kernel/mem_protect/stackprot fails on nrf52

[mike-scott]

Describe the bug
Due to PR: 3e255e9
sanitycheck now fails on nrf52_blenano2 -- and probably a lot of other arm HW. (log below in "Screenshots or console output")

Related CI failure:
https://ci.foundries.io/projects/zephyr/builds/5694/sanity-test-nrf52/

To Reproduce
Steps to reproduce the behavior:

1. cd tests/kernel/mem_protect/stackprot
2. mkdir build; cd build
3. cmake -DBOARD=nrf52_blenano2 ..
4. make flash
5. See error below

Expected behavior
Pass sanitycheck like previous to this commit

Impact
It's a regression for CI testing.. all of nRF52 and other arm HW will probably fail this test as a result

Screenshots or console output

 94/153 nrf52_blenano2            tests/kernel/mem_protect/stackprot/kernel.memory_protection FAILED: timeout
../tmp/outdir/nrf52_blenano2/tests/kernel/mem_protect/stackprot/kernel.memory_protection/handler.log
�***** Booting Zephyr OS v1.14.0-rc3-230-g1897896e9739 *****
Starts alternate_thread
alternate_thread: Input string is too long and stack overflowed!

***** Stack Check Fail! *****
Current thread ID = 0x20000068
Faulting instruction address = 0x8a54
Fatal fault in thread 0x20000068! Aborting.
Running test suite stackprot
===================================================================
starting test - test_stackprot
***** MPU FAULT *****
  Stacking error (context area might be not valid)
  Data Access Violation
  MMFAR Address: 0x20001c10
***** Stack Check Fail! *****
Current thread ID = 0x20000344
Faulting instruction address = 0x97e0
Fatal fault in essential thread! Spinning...

Environment (please complete the following information):

• OS: Arch Linux
• Toolchain: Zephyr SDK 0.10.0
• Commit SHA: today's master (a942fcc)

[aescolar]

@mike-scott : From your description this seems to not be related to sanitycheck, but to this particular test? Also, Is this test failing in the Zephyr master branch? (If so, the problem would not be related to your CI)
If that is the case, could you change the title to be a bit more clear that it just relates to this test and particular board?
(Something like "tests/kernel/mem_protect/stackprot fails on nrf52_blenano2" )

[mike-scott]

@aescolar sanitycheck runs this for nrf52 hardware. The "easy" way to reproduce is to run just this test. The actual regression was found via sanitycheck for real HW.

I don't see the issue with the description.

[mike-scott]

Note: this test will probably fail on a lot more hardware than just nrf52_blenano2.

I'll adjust the title to mention the test specifically.

[wentongwu]

@mike-scott Actually this change passed the CI test before check in. Anyway I will fix it ASAP.

[nashif]

@mike-scott Actually this change passed the CI test before check in. Anyway I will fix it ASAP.

CI does not run on real HW