URSHL (multiple and single vector)

Multi-vector unsigned rounding shift left by vector

This instruction shifts the unsigned elements of the two or four first source vectors by the corresponding elements of the second source vector, and destructively places the rounded results in the corresponding elements of the two or four first source vectors. A positive shift amount performs a left shift, otherwise a right shift by the negated shift amount is performed.

This instruction is unpredicated.

It has encodings from 2 classes: Two registers and Four registers

Two registers
(FEAT_SME2)

313029282726252423222120191817161514131211109876543210
11000001size10Zm10100010001Zdn1
opcU

Encoding

URSHL { <Zdn1>.<T>-<Zdn2>.<T> }, { <Zdn1>.<T>-<Zdn2>.<T> }, <Zm>.<T>

Decode for this encoding

if !IsFeatureImplemented(FEAT_SME2) then EndOfDecode(Decode_UNDEF); end; let esize : integer{} = 8 << UInt(size); let dn : integer = UInt(Zdn::'0'); let m : integer = UInt('0'::Zm); let nreg : integer{} = 2;

Four registers
(FEAT_SME2)

313029282726252423222120191817161514131211109876543210
11000001size10Zm10101010001Zdn01
opcU

Encoding

URSHL { <Zdn1>.<T>-<Zdn4>.<T> }, { <Zdn1>.<T>-<Zdn4>.<T> }, <Zm>.<T>

Decode for this encoding

if !IsFeatureImplemented(FEAT_SME2) then EndOfDecode(Decode_UNDEF); end; let esize : integer{} = 8 << UInt(size); let dn : integer = UInt(Zdn::'00'); let m : integer = UInt('0'::Zm); let nreg : integer{} = 4;

Assembler Symbols

<Zdn1>

For the "Two registers" variant: is the name of the first scalable vector register of the destination and first source multi-vector group, encoded as "Zdn" times 2.

For the "Four registers" variant: is the name of the first scalable vector register of the destination and first source multi-vector group, encoded as "Zdn" times 4.

<T>

Is the size specifier, encoded in size:

size <T>
00 B
01 H
10 S
11 D
<Zdn2>

Is the name of the second scalable vector register of the destination and first source multi-vector group, encoded as "Zdn" times 2 plus 1.

<Zm>

Is the name of the second source scalable vector register Z0-Z15, encoded in the "Zm" field.

<Zdn4>

Is the name of the fourth scalable vector register of the destination and first source multi-vector group, encoded as "Zdn" times 4 plus 3.

Operation

CheckStreamingSVEEnabled(); let VL : integer{} = CurrentVL(); let elements : integer = VL DIV esize; var results : array [[4]] of bits(VL); for r = 0 to nreg-1 do let operand1 : bits(VL) = Z{}(dn+r); let operand2 : bits(VL) = Z{}(m); for e = 0 to elements-1 do let element : integer = UInt(operand1[e*:esize]); var shift : integer = ShiftSat(SInt(operand2[e*:esize]), esize); var res : integer; if shift >= 0 then res = element << shift; else shift = -shift; res = (element + (1 << (shift - 1))) >> shift; end; results[[r]][e*:esize] = res[esize-1:0]; end; end; for r = 0 to nreg-1 do Z{VL}(dn+r) = results[[r]]; end;


2026-03_rel 2026-03-26 20:48:11

Copyright © 2010-2026 Arm Limited or its affiliates. All rights reserved. This document is Non-Confidential.