flw

The `flw` instruction loads a single-precision floating-point value from memory at address `xs1` + `imm` into floating-point register `fd`. It does not modify the bits being transferred; in particular, the payloads of non-canonical NaNs are preserved.

flw fd, imm(xs1)