Gald you solve this! Previously I have also linked a library ( GitHub - stevengj/nlopt: library for nonlinear optimization, wrapping many algorithms for global and local, constrained or unconstrained, optimization · GitHub ) dynamically once … all binaries broke and took me quite a while to solve it 
By the way, if you have not paid FFTW3 already, KFR GitHub - kfrlib/kfr: Fast, modern C++ DSP framework, FFT, Sample Rate Conversion, FIR/IIR/Biquad Filters (SSE, AVX, AVX-512, ARM NEON, RISC-V RVV) · GitHub might be a better choice, or even Apple vDSP and Intel IPP. I have done some reserach recently (in case you are interested, Share some FFT benchmark results ) and it seems that FFTW3 performs kind of bad on M chips 