UMOPA (2-way)

Unsigned integer sum of outer products and accumulate

This instruction works with a 32-bit element ZA tile.

The unsigned integer sum of outer products and accumulate instructions multiply the sub-matrix in the first source vector by the sub-matrix in the second source vector. The first source holds SVLS×2 sub-matrix of unsigned 16-bit integer values, and the second source holds 2×SVLS sub-matrix of unsigned 16-bit integer values.

Each source vector is independently predicated by a corresponding governing predicate. When a 16-bit source element is inactive, it is treated as having the value 0.

The resulting SVLS×SVLS widened 32-bit integer sum of outer products is then destructively added to the 32-bit integer destination tile. This is equivalent to performing a 2-way dot product and accumulate to each of the destination tile elements.

Each 32-bit container of the first source vector holds 2 consecutive column elements of each row of a SVLS×2 sub-matrix, and each 32-bit container of the second source vector holds 2 consecutive row elements of each column of a 2×SVLS sub-matrix.

SME2
(FEAT_SME2)

313029282726252423222120191817161514131211109876543210
10100001100ZmPmPnZn010ZAda
u0S

UMOPA <ZAda>.S, <Pn>/M, <Pm>/M, <Zn>.H, <Zm>.H

if !HaveSME2() then UNDEFINED; constant integer esize = 32; integer a = UInt(Pn); integer b = UInt(Pm); integer n = UInt(Zn); integer m = UInt(Zm); integer da = UInt(ZAda); boolean sub_op = FALSE; boolean unsigned = TRUE;

Assembler Symbols

<ZAda>

Is the name of the ZA tile ZA0-ZA3, encoded in the "ZAda" field.

<Pn>

Is the name of the first governing scalable predicate register P0-P7, encoded in the "Pn" field.

<Pm>

Is the name of the second governing scalable predicate register P0-P7, encoded in the "Pm" field.

<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.

Operation

CheckStreamingSVEAndZAEnabled(); constant integer VL = CurrentVL; constant integer PL = VL DIV 8; constant integer dim = VL DIV esize; bits(PL) mask1 = P[a, PL]; bits(PL) mask2 = P[b, PL]; bits(VL) operand1 = Z[n, VL]; bits(VL) operand2 = Z[m, VL]; bits(dim*dim*esize) operand3 = ZAtile[da, esize, dim*dim*esize]; bits(dim*dim*esize) result; integer prod; for row = 0 to dim-1 for col = 0 to dim-1 bits(esize) sum = Elem[operand3, row*dim+col, esize]; for k = 0 to 1 if ActivePredicateElement(mask1, 2*row + k, esize DIV 2) && ActivePredicateElement(mask2, 2*col + k, esize DIV 2) then prod = (Int(Elem[operand1, 2*row + k, esize DIV 2], unsigned) * Int(Elem[operand2, 2*col + k, esize DIV 2], unsigned)); if sub_op then prod = -prod; sum = sum + prod; Elem[result, row*dim+col, esize] = sum; ZAtile[da, esize, dim*dim*esize] = result;

Operational information

If PSTATE.DIT is 1:


Internal version only: aarchmrs v2023-12_rel, pseudocode v2023-12_rel, sve v2023-12_rel ; Build timestamp: 2023-12-15T16:46

Copyright © 2010-2023 Arm Limited or its affiliates. All rights reserved. This document is Non-Confidential.