Floatingpoint complex add with rotate (predicated)
Add the real and imaginary components of the active floatingpoint complex numbers from the first source vector to the complex numbers from the second source vector which have first been rotated by 90 or 270 degrees in the direction from the positive real axis towards the positive imaginary axis, when considered in polar representation, equivalent to multiplying the complex numbers in the second source vector by ±j beforehand. Destructively place the results in the corresponding elements of the first source vector. Inactive elements in the destination vector register remain unmodified.
Each complex number is represented in a vector register as an even/odd pair of elements with the real part in the evennumbered element and the imaginary part in the oddnumbered element.
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  1  1  0  0  1  0  0  size  0  0  0  0  0  rot  1  0  0  Pg  Zm  Zdn 
if !IsFeatureImplemented(FEAT_SVE) && !IsFeatureImplemented(FEAT_SME) then UNDEFINED; if size == '00' then UNDEFINED; constant integer esize = 8 << UInt(size); constant integer g = UInt(Pg); constant integer dn = UInt(Zdn); constant integer m = UInt(Zm); constant boolean sub_i = (rot == '0'); constant boolean sub_r = (rot == '1');
<Zdn> 
Is the name of the first source and destination scalable vector register, encoded in the "Zdn" field. 
<T> 
Is the size specifier,
encoded in

<Pg> 
Is the name of the governing scalable predicate register P0P7, encoded in the "Pg" field. 
<Zm> 
Is the name of the second source scalable vector register, encoded in the "Zm" field. 
<const> 
Is the const specifier,
encoded in

CheckSVEEnabled(); constant integer VL = CurrentVL; constant integer PL = VL DIV 8; constant integer pairs = VL DIV (2 * esize); constant bits(PL) mask = P[g, PL]; constant bits(VL) operand1 = Z[dn, VL]; constant bits(VL) operand2 = if AnyActiveElement(mask, esize) then Z[m, VL] else Zeros(VL); bits(VL) result; for p = 0 to pairs1 acc_r = Elem[operand1, 2 * p + 0, esize]; acc_i = Elem[operand1, 2 * p + 1, esize]; if ActivePredicateElement(mask, 2 * p + 0, esize) then elt2_i = Elem[operand2, 2 * p + 1, esize]; if sub_i then elt2_i = FPNeg(elt2_i, FPCR); acc_r = FPAdd(acc_r, elt2_i, FPCR); if ActivePredicateElement(mask, 2 * p + 1, esize) then elt2_r = Elem[operand2, 2 * p + 0, esize]; if sub_r then elt2_r = FPNeg(elt2_r, FPCR); acc_i = FPAdd(acc_i, elt2_r, FPCR); Elem[result, 2 * p + 0, esize] = acc_r; Elem[result, 2 * p + 1, esize] = acc_i; Z[dn, VL] = result;
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:
Internal version only: aarchmrs v202403_relA, pseudocode v202403_rel, sve v202403_rel ; Build timestamp: 20240326T09:45
Copyright © 20102024 Arm Limited or its affiliates. All rights reserved. This document is NonConfidential.