Floating Point usage

Page under construction... Informations on this page are not guaranteed !!

• on ARM9, no choice, you have to use software (slow) floating point emulation (-msoft-float). So if you can, use fixed point arithmetic.
• on Cortex-A8, you can try to use hard Floating Point units (NEON/VFP):
  • http://pandorawiki.org/Floating_Point_Optimization
  • http://infocenter.arm.com/help/index.jsp?topic=/com.arm.doc.dui0204i/Chdehgeh.html