I've split these changes out from this PR #16057

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 98.57%. Comparing base (f4e4599) to head (a70a90c).
Report is 7 commits behind head on master.

@@           Coverage Diff           @@
##           master   #16313   +/-   ##
=======================================
  Coverage   98.57%   98.57%           
=======================================
  Files         164      164           
  Lines       21349    21349           
=======================================
  Hits        21045    21045           
  Misses        304      304           

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

Code size report:

   bare-arm:    +0 +0.000% 
minimal x86:    +0 +0.000% 
   unix x64:    +0 +0.000% standard
      stm32:    +0 +0.000% PYBV10
     mimxrt:    +0 +0.000% TEENSY40
        rp2: -1952 -0.214% RPI_PICO_W[incl +4(bss)]
       samd:    +0 +0.000% ADAFRUIT_ITSYBITSY_M4_EXPRESS
  qemu rv32:    +0 +0.000% VIRT_RV32

For anyone that wants to test but doesn't want to build the firmware, here is a zipped UF2 file for the ARM build of 2d4b02b.

I have a project with different topics (webserver, I2C, buttons, asynio, WLAN, MQTT, neo led, logging, ...) and this firmware runs stable for me. I could run the project without any modifications from the Pico W.

Requesting a look at #16180 (and related #16181), a confirmed lightsleep regression for Pico W and Pico 2 W

I did a full test of this PR on:

• Pico
• Pico W
• Pico2, in ARM and RISCV mode
• Pico2 W, in ARM and RISCV mode

Everything seems to work as it should except the following:

• There are some known failures, like using the native emitter to run tests/thread/stress_schedule.py.
• tests/extmod/machine1.py test fails due to the issue fixed by tests/run-tests.py: Set name of injected test module to '__main__'. #16433.
• tests/ports/rp2/rp2_dma.py fails on Pico2 W because the DMA channel that's allocated to Python in this test is different to the expected one, I guess because the expected DMA channel is used by the WiFi driver. To fix this, that test should probably be rewritten to use unittest.
• tests/thread/thread_stdin.py locks up the Pico2 W when in RISCV mode. I'm not sure why, but I'm happy to let this through and fix it separately.

Should we include unittest in the list of installed modules? To avoid having to remember to manually install it to run the tests

Should we include unittest in the list of installed modules? To avoid having to remember to manually install it to run the tests

I'd rather not do that. The unittest package is pretty large when frozen in to firmware and not all ports have the space for it. Also, it's not a very common package for users to use, so takes up flash space unnecessarily.

If you're using a particular board for running the MicroPython test suite, it's relatively easy to do mpremote mip install unittest once on that board and then leave it installed. Then it will work with the tests from then on.

I squashed down the last two changes you made @peterharperuk . This looks good now, it's very clean thank you!