Correct method to specify -mfpu compile flag?

edit

Hi, I raised an issue in github for this because I thought it might be a bug (well not a bug, but something that could be fixed/improved): https://github.com/opencv/opencv/issu... But it got flagged as a question, so thought I'd ask it here as well/instead.

Basically trying to understand exactly what the ENABLE_NEON ENABLE_VFPV3 cmake flags do, and how I can specify -mfpu=neon-fp-armv8 in the compile? If I turn both these flags off then I can specify -mfpu, but I don't know if this will turn off NEON/VFP optimisations within the opencv code itself?