Views of the floating-point extension register bank in AArch32 state
Floating-point can have different views of the extension register bank in AArch32 state.
The floating-point extension register bank can be viewed as:
- Thirty-two 64-bit
- Thirty-two 32-bit registers,
S0-S31. Only half of the register bank is accessible in this view.
- A combination of registers from these views.
64-bit floating-point registers are called double-precision registers and can contain double-precision floating-point values. 32-bit floating-point registers are called single-precision registers and can contain either a single-precision or two half-precision floating-point values.