Arm® A64 Instruction Set Architecture: Future Architecture Technologies in the A architecture profile

FRECPX

Floating-point reciprocal exponent (predicated).

Invert the exponent leaving the fractional part unchanged of each active floating-point element of the source vector, and place the results in the corresponding elements of the destination vector. Inactive elements in the destination vector register remain unmodified.

The result of this instruction can be used with FMULX to convert arbitrary elements in mathematical vector space to "unit vectors" or "direction vectors" of length 1.

if !HaveSVE() then UNDEFINED;
if size == '00' then UNDEFINED;
integer esize = 8 << UInt(size);
integer g = UInt(Pg);
integer n = UInt(Zn);
integer d = UInt(Zd);

Assembler Symbols

CheckSVEEnabled();
integer elements = VL DIV esize;
bits(PL) mask = P[g];
bits(VL) operand = Z[n];
bits(VL) result = Z[d];

for e = 0 to elements-1
    bits(esize) element = Elem[operand, e, esize];
    if ElemP[mask, e, esize] == '1' then
        Elem[result, e, esize] = FPRecpX(element, FPCR);

Z[d] = result;