Signed saturating doubling multiply high (unpredicated)
Multiply then double the corresponding signed elements of the first and second source vectors, and place the most significant half of the results in the corresponding elements of the destination vector register. Each result element is saturated to the N-bit element's signed integer range -2(N-1) to (2(N-1))-1. This instruction is unpredicated.
Variants: FEAT_SVE2 || FEAT_SME (FEAT_SVE2 || FEAT_SME)
31 | 30 | 29 | 28 | 27 | 26 | 25 | 24 | 23 | 22 | 21 | 20 | 19 | 18 | 17 | 16 | 15 | 14 | 13 | 12 | 11 | 10 | 9 | 8 | 7 | 6 | 5 | 4 | 3 | 2 | 1 | 0 |
0 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 1 | 0 | 1 | 1 | 1 | 0 | 0 | |||||||||||||||||
size | Zm | R | Zn | Zd |
---|
SQDMULH <Zd>.<T>, <Zn>.<T>, <Zm>.<T>
if !IsFeatureImplemented(FEAT_SVE2) && !IsFeatureImplemented(FEAT_SME) then EndOfDecode(Decode_UNDEF); constant integer esize = 8 << UInt(size); constant integer n = UInt(Zn); constant integer m = UInt(Zm); constant integer d = UInt(Zd);
CheckSVEEnabled(); constant integer VL = CurrentVL; constant integer elements = VL DIV esize; constant bits(VL) operand1 = Z[n, VL]; constant bits(VL) operand2 = Z[m, VL]; bits(VL) result; for e = 0 to elements-1 constant integer element1 = SInt(Elem[operand1, e, esize]); constant integer element2 = SInt(Elem[operand2, e, esize]); constant integer res = 2 * element1 * element2; Elem[result, e, esize] = SignedSat(res >> esize, esize); Z[d, VL] = result;
If PSTATE.DIT is 1: