rp2: Build with -fno-math-errno to use the hardware sqrt instruction.

Gadgetoid · dpgeorge · commit 772df1ae574a · 2026-07-17T23:33:03.000+10:00
Without -fno-math-errno the compiler must keep sqrt()/sqrtf() as library
calls so they can set errno on a domain error, even though the Cortex-M33
FPU has a single VSQRT.F32 instruction. MicroPython's math module
detects domain errors via isnan/isinf checks on the result rather than
errno, so disabling errno here is safe and lets sqrt leverage hardware.

Benchmarked on a Pico 2 (RP2350, perfbench N=150 M=100, avg of 3):
misc_mandel (complex abs() in its inner loop) improves by ~12%, with no
measurable change to non-sqrt benchmarks and a slightly smaller binary.

Signed-off-by: Phil Howard &lt;github@gadgetoid.com&gt;
diff --git a/ports/rp2/CMakeLists.txt b/ports/rp2/CMakeLists.txt
@@ -532,6 +532,11 @@ target_compile_options(${MICROPY_TARGET} PRIVATE
     -Wall
     -Werror
     -g  # always include debug information in the ELF
+
+    # Let the compiler lower sqrt()/sqrtf() to the hardware VSQRT instruction
+    # instead of a library call.  MicroPython's math module detects domain
+    # errors via isnan/isinf on the result, not errno, so this is safe.
+    -fno-math-errno
 )
 
 target_link_options(${MICROPY_TARGET} PRIVATE
