Fixes a warning when building btree with cmake - #19644

Thanks for the PR!

While you're at it, you may also want to add -Wno-deprecated-non-prototype to also make life easier for some versions of Clang as well, see

BTree's codebase is still clinging onto old standards and newer compilers tend to like that kind of stuff less and less as time goes on (there's also a PR for enabling compilation with MUSL, but that's still in the review queue).

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 98.55%. Comparing base (1c3c201) to head (7193279).
⚠️ Report is 2 commits behind head on master.

@@            Coverage Diff             @@
##           master   #19644      +/-   ##
==========================================
- Coverage   98.58%   98.55%   -0.04%     
==========================================
  Files         182      182              
  Lines       23322    23322              
  Branches        5        5              
==========================================
- Hits        22993    22986       -7     
- Misses        328      335       +7     
  Partials        1        1              

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

Code size report:

Reference:  extmod: Add USB Network (NCM) driver implementation. [1c3c201]
Comparison: extmod/extmod.cmake: Fixes a warning when building btree with cmake. [merge of 7193279]
  mpy-cross:    +0 +0.000% 
   bare-arm:    +0 +0.000% 
minimal x86:    +0 +0.000% 
   unix x64:    +0 +0.000% standard
      stm32:    +0 +0.000% PYBV10
      esp32:    +0 +0.000% ESP32_GENERIC
     mimxrt:    +0 +0.000% TEENSY40
        rp2:    +0 +0.000% RPI_PICO_W
       samd:    +0 +0.000% ADAFRUIT_ITSYBITSY_M4_EXPRESS
  qemu rv32:    +0 +0.000% VIRT_RV32

Then it would be probably better to use the exact same flags as we do in extmod/extmod.mk:
https://github.com/micropython/micropython/blob/master/extmod/extmod.mk#L449

Then both .cmake and .mk would be consistent, and the suppressions are consistent with the ones from micropython/examples/natmod/btree/Makefile.

Agree?

That makes sense, feel free to update the PR (keeping it a single commit if possible).

Done. Please double-check with your clang setup if possible, I only verified the gcc build.

We got a failed test for qemu_arm (thread/stress_heap.py), currently trying to reproduce this locally.
We also lost 0.04% of coverage somewhere along the way?!

I am a bit time constrained right now, but as far as I can see the qemu_arm build does not even use cmake. Is this maybe a flaky test @agatti ?

I'd say the changes look good.

For the rest, coverage check is obviously mistaken here (that happens sometimes) and that specific test for qemu_arm has been flaky for some time. There's a PR to skip the AES stress test on QEMU-based jobs, hopefully will be up for merging in v1.30.