CMLA (vectors)

Complex integer multiply-add

This instruction multiplies the duplicated real components for rotations 0 and 180, or imaginary components for rotations 90 and 270, of the integral numbers in the first source vector by the corresponding complex number in the second source vector rotated by 0, 90, 180 or 270 degrees in the direction from the positive real axis towards the positive imaginary axis, when considered in polar representation.

This instruction then adds the products to the corresponding components of the complex numbers in the addend vector, and destructively places the results in the corresponding elements of the addend vector. This instruction is unpredicated.

These transformations permit the creation of a variety of multiply-add and multiply-subtract operations on complex numbers by combining two of these instructions with the same vector operands but with rotations that are 90 degrees apart.

Each complex number is represented in a vector register as an even/odd pair of elements with the real part in the even-numbered element and the imaginary part in the odd-numbered element.

SVE2
(FEAT_SVE2 || FEAT_SME)

313029282726252423222120191817161514131211109876543210
01000100size0Zm0010rotZnZda
op

Encoding

CMLA <Zda>.<T>, <Zn>.<T>, <Zm>.<T>, <const>

Decode for this encoding

if !IsFeatureImplemented(FEAT_SVE2) && !IsFeatureImplemented(FEAT_SME) then EndOfDecode(Decode_UNDEF); end; let esize : integer{} = 8 << UInt(size); let n : integer = UInt(Zn); let m : integer = UInt(Zm); let da : integer = UInt(Zda); let sel_a : integer = UInt(rot[0]); let sel_b : integer = UInt(NOT(rot[0])); let sub_r : boolean = (rot[0] != rot[1]); let sub_i : boolean = (rot[1] == '1');

Assembler Symbols

<Zda>

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

<T>

Is the size specifier, encoded in size:

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

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

<Zm>

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

<const>

Is the const specifier, encoded in rot:

rot <const>
00 #0
01 #90
10 #180
11 #270

Operation

CheckSVEEnabled(); let VL : integer{} = CurrentVL(); let pairs : integer = VL DIV (2 * esize); let operand1 : bits(VL) = Z{}(n); let operand2 : bits(VL) = Z{}(m); let operand3 : bits(VL) = Z{}(da); var result : bits(VL); for p = 0 to pairs-1 do let elt1_a : integer = SInt(operand1[(2 * p + sel_a)*:esize]); let elt2_a : integer = SInt(operand2[(2 * p + sel_a)*:esize]); let elt2_b : integer = SInt(operand2[(2 * p + sel_b)*:esize]); let elt3_r : bits(esize) = operand3[(2 * p + 0)*:esize]; let elt3_i : bits(esize) = operand3[(2 * p + 1)*:esize]; let product_r : integer = elt1_a * elt2_a; let product_i : integer = elt1_a * elt2_b; if sub_r then result[(2 * p + 0)*:esize] = elt3_r - product_r; else result[(2 * p + 0)*:esize] = elt3_r + product_r; end; if sub_i then result[(2 * p + 1)*:esize] = elt3_i - product_i; else result[(2 * p + 1)*:esize] = elt3_i + product_i; end; end; Z{VL}(da) = result;

Operational information

This instruction is a data-independent-time instruction as described in About PSTATE.DIT.

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


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.