-fma

(SPARC) Enables automatic generation of floating-point, fused, multiply-add instructions. 

–fns[={no|yes}]

(SPARC) Disables or enables the SPARC nonstandard floating-point mode. 

–fprecision=p

x86: Sets floating-point precision mode.

–fround=r

Sets IEEE rounding mode in effect at startup. 

–fsimple=n

Sets floating-point optimization preferences. 

–fstore

x86: Forces precision of floating-point expressions.

–ftrap=tlst

Sets IEEE trapping mode in effect at startup. 

–nofstore

x86: Disables forced precision of expression.

–xlibmieee

Causes libm to return IEEE 754 values for math routines in exceptional cases.