You copied the Doc URL to your clipboard.


Signed rounding shift right and accumulate (immediate).

Shift right by immediate each signed element of the source vector, preserving the sign bit, and add the rounded intermediate result destructively to the corresponding elements of the addend vector. The immediate shift amount is an unsigned value in the range 1 to number of bits per element. This instruction is unpredicated.

if !HaveSVE2() then UNDEFINED;
bits(4) tsize = tszh:tszl;
case tsize of
    when '0000' UNDEFINED;
    when '0001' esize = 8;
    when '001x' esize = 16;
    when '01xx' esize = 32;
    when '1xxx' esize = 64;
integer n = UInt(Zn);
integer da = UInt(Zda);
integer shift = (2 * esize) - UInt(tsize:imm3);

Assembler Symbols


Is the name of the third source and destination scalable vector register, encoded in the "Zda" field.

<T> Is the size specifier, encoded in tszh:tszl:
tszh tszl <T>
00 01 B
00 1x H
01 xx S
1x xx D

Is the name of the first source scalable vector register, encoded in the "Zn" field.


Is the immediate shift amount, in the range 1 to number of bits per element, encoded in "tsz:imm3".


integer elements = VL DIV esize;
bits(VL) operand1 = Z[n];
bits(VL) operand2 = Z[da];
bits(VL) result;
integer round_const = 1 << (shift - 1);

for e = 0 to elements-1
    integer element = (SInt(Elem[operand1, e, esize]) + round_const) >> shift;
    Elem[result, e, esize] = Elem[operand2, e, esize] + element<esize-1:0>;

Z[da] = result;

Operational information

This instruction might be immediately preceded in program order by a MOVPRFX instruction. The MOVPRFX instruction must conform to all of the following requirements, otherwise the behavior of the MOVPRFX and this instruction is unpredictable:

  • The MOVPRFX instruction must be unpredicated.
  • The MOVPRFX instruction must specify the same destination register as this instruction.
  • The destination register must not refer to architectural register state referenced by any other source operand register of this instruction.