Floating point arithmetic in Kernel

Question

In kernel space, I have to do matrix multiplication whose elements are floating values(say 0.8666 etc). Is it possible to do floating point arithmetic in Kernel space. I am using 3.12 Kernel on ARM based board.

Answer 1

My understanding is the kernel is always linked with FPU emulation so that it does work, but it's almost always wiser to use scaled integers.