Float literals don't fit in an immediate
PowerPC has no load-immediate for floats. A 32-bit constant just won't fit inside an instruction word. So MWCC stashes the value in the small data area (SDA) and pulls it back with lfs (load floating single), addressed off r2/r13. A function that doubles its input compiles like this:
lfs f0, ... # load 2.0f from the SDA
fmuls f1, f0, f1 # single-precision multiply
blr
Those ... are a relocation; the linker resolves them, and in the disassembler you'll instead read a real SDA-relative offset off r2, something like lfs f0, 0x20(r2) or the symbolic lfs f0, lit@sda21(r2). See an lfs flowing straight into an fmuls/fadds and you can bet the source had a float literal, an f-suffixed constant, sitting in that expression. Whatever number lfs pulls in is the literal itself. Read it off the disassembly symbol or the SDA entry.
Your task
Write quarter taking an f32 x to match the target assembly above.