nrf52_bsim fails on some tests after merging 29810

[nashif]

Describe the bug

west build -b nrf52_bsim tests/benchmarks/latency_measure -t run
hangs, before #29810 (by @nordic-krch ) it used to run all the way to the end.

To Reproduce
Steps to reproduce the behavior:

1. west build -b nrf52_bsim tests/benchmarks/latency_measure -t run

Expected behavior

test completes
Impact
Blocking CI in many pending PRs

Logs and console output
hangs here:

?_??: WARNING: Neither simulation id or the device number have been set. I assume you want to run without a BabbleSim phy (-nosim)
?_??: WARNING: If this is not what you wanted, check with --help how to set them
d_00: @00:00:00.000000  *** Booting Zephyr OS build zephyr-v2.4.0-2121-gde191bf3a6fb  ***
d_00: @00:00:00.000000  START - Time Measurement
d_00: @00:00:00.000000  Timing results: Clock frequency: 0 MHz

[nordic-krch]

@aescolar i think i would need a help from you here. It seems like RTC interrupt is triggered only once. I've logged COUNTER, CC[0] and EVTEN registers after setting new CC and it looks ok.

At 0 CC is set for 1sec, at one sec RTC interrupt occurs and it is set to 2sec timeout but that one never comes.

d_00: @00:00:00.000000  COUNTER:0 CC:32768
d_00: @00:00:00.000000  EVTEN:10000
d_00: @00:00:00.000000  *** Booting Zephyr OS build zephyr-v2.4.0-2096-gad3de0a76cec  ***
d_00: @00:00:00.000000  START - Time Measurement
d_00: @00:00:00.000000  Timing results: Clock frequency: 0 MHz
d_00: @00:00:01.000000  COUNTER:32768 CC:65536
d_00: @00:00:01.000000  EVTEN:10000

[aescolar]

@aescolar i think i would need a help from you here.

@nordic-krch (Today and tomorrow I'm swamped.) The default when a PR breaks master is to revert it and figure out calmly why and fix it without blocking others. If you cannot find easily what went wrong I would not delay further reverting it.