fpu_std
fpu_std(4, &fprs[4 * offset]);
fpu_std(5, &fprs[5 * offset]);
fpu_std(6, &fprs[6 * offset]);
fpu_std(7, &fprs[7 * offset]);
fpu_std(8, &fprs[8 * offset]);
fpu_std(9, &fprs[9 * offset]);
fpu_std(10, &fprs[10 * offset]);
fpu_std(11, &fprs[11 * offset]);
fpu_std(12, &fprs[12 * offset]);
fpu_std(13, &fprs[13 * offset]);
fpu_std(14, &fprs[14 * offset]);
fpu_std(15, &fprs[15 * offset]);
fpu_std(0, &fprs[0 * offset]);
fpu_std(1, &fprs[1 * offset]);
fpu_std(2, &fprs[2 * offset]);
fpu_std(3, &fprs[3 * offset]);