GCC and FPU

Looking at the gcc codebase it seems far from trivial to start fiddling with it.

I wouldn't dare (for now) to think about async ops. Llvm looks much better

in this regard. There is a unfinished nios2 port at ellcc.org, did anyone check

it out?

Following is a sample code generated by gcc for nios2 and powerpc:

double x3(double x)

{

return x*x*x;

}

-- nios2-elf-gcc -O2 (v3.4.6)

mov r6,r4

mov r7,r5

custom 0,zero,r6,r7 (fwrx)

custom 15,r5,r6,r7 (fmuld)

custom 4,r4,zero,zero (frdy)

custom 0,zero,r4,r5 (fwrx)

custom 15,r5,r6,r7 (fmuld)

custom 4,r4,zero,zero (frdy)

mov r2,r4

mov r3,r5

ret

-- powerpc-eabi-gcc -O2 (v3.3.6)

fmul f0,f1,f1

fmul f0,f0,f1

fmr f1,f0

blr