The SDK2.0 assembly code for the RP2350 uses:As I understand it, it has 32 single precision floating point registers, on which single precision operations, including division and square root can be performed. Pairs of these registers may be combined into double precision registers, however only double precision load and store operations - no arithmetic - are supported, so I'm not sure what use that is. However, separately, there is a double precision floating point coprocessor, which accelerates double precision floating point arithmetic. So the ARM processor itself does single precision and the coprocessor does double precision.
Code:
.fpu fpv5-sp-d16Statistics: Posted by lurk101 — Sun Sep 08, 2024 9:06 pm