FTZ status on bela?

What is the status of subnormal IEEE 754 floating point numbers support on ARM ? Is the FTZ bit set to on by the OS ? Are then any slowdown issues like on Intel CPU when subnormals samples are produced en FTZ is off? (see for iOS: https://stackoverflow.com/questions/7346521/subnormal-ieee-754-floating-point-numbers-support-on-ios-arm-devices-iphone-4.)

Denormals are flushed to zero in hardware. This happens by default on the NEON fpu, and it's enabled manually in the VFP fpu (
https://github.com/BelaPlatform/Bela/blob/master/core/math_runfast.c ).
So there are no slowdowns associated with denormals on Bela.