Multi-vector unsigned saturating rounding shift right narrow by immediate
This instruction shifts right by an immediate value the unsigned integer value in each element of the four source vectors, and places the rounded results in the quarter-width destination elements. Each result element is saturated to the quarter-width N-bit element's unsigned integer range 0 to (2N)-1. The immediate shift amount is an unsigned value in the range 1 to number of bits per source element.
This instruction is unpredicated.
Variants: FEAT_SME2 (ARMv9.3)
| 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 |
| 1 | 1 | 0 | 0 | 0 | 0 | 0 | 1 | 1 | 1 | 1 | 0 | 1 | 1 | 0 | 0 | 1 | |||||||||||||||
| tsize | imm5 | N | Zn | op | U | Zd | |||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
UQRSHR <Zd>.<T>, { <Zn1>.<Tb>-<Zn4>.<Tb> }, #<const>
if !IsFeatureImplemented(FEAT_SME2) then EndOfDecode(Decode_UNDEF); if tsize == '00' then EndOfDecode(Decode_UNDEF); constant integer esize = 8 << HighestSetBit(tsize); constant integer n = UInt(Zn:'00'); constant integer d = UInt(Zd); constant integer shift = (8 * esize) - UInt(tsize:imm5);
CheckStreamingSVEEnabled(); constant integer VL = CurrentVL; constant integer elements = VL DIV (4 * esize); bits(VL) result; for r = 0 to 3 constant bits(VL) operand = Z[n+r, VL]; for e = 0 to elements-1 constant bits(4 * esize) element = Elem[operand, e, 4 * esize]; constant integer res = (UInt(element) + (1 << (shift-1))) >> shift; Elem[result, r*elements + e, esize] = UnsignedSat(res, esize); Z[d, VL] = result;