Float literals don't fit in an immediate
There is no "load immediate float" instruction — a 32-bit constant can't be encoded inline. Instead MWCC parks the value in the small data area (SDA) and loads it with lfs (load floating single), addressed relative to r2/r13. A function that doubles its argument, for example, compiles to:
lfs f0, ... # load 2.0f from the SDA
fmuls f1, f0, f1 # single-precision multiply
blr
The ... is a relocation the linker fills in; in the disassembler you'll see a concrete SDA-relative offset off r2, e.g. lfs f0, 0x20(r2) (or a symbolic lfs f0, lit@sda21(r2)). When you spot lfs feeding straight into an fmuls/fadds, the original C almost certainly had a float literal (an f suffix constant) in the expression. The constant loaded by lfs is the literal value itself — read it from the disassembly symbol or the SDA entry.
Your task
Write quarter taking an f32 x to match the target assembly above.