ST1W (scalar plus immediate, strided registers)

Contiguous store of words from multiple strided vectors (immediate index)

Contiguous store of words from elements of two or four strided vector registers to the memory address generated by a 64-bit scalar base and immediate index which is multiplied by the vector's in-memory size, irrespective of predication, and added to the base address.

Inactive elements are not written to memory.

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

Two registers
(FEAT_SME2)

313029282726252423222120191817161514131211109876543210
101000010110imm4010PNgRnT0Zt
msz<1>msz<0>N

ST1W { <Zt1>.S, <Zt2>.S }, <PNg>, [<Xn|SP>{, #<imm>, MUL VL}]

if !IsFeatureImplemented(FEAT_SME2) then UNDEFINED; constant integer n = UInt(Rn); constant integer g = UInt('1':PNg); constant integer nreg = 2; constant integer tstride = 8; constant integer t = UInt(T:'0':Zt); constant integer esize = 32; constant integer offset = SInt(imm4);

Four registers
(FEAT_SME2)

313029282726252423222120191817161514131211109876543210
101000010110imm4110PNgRnT00Zt
msz<1>msz<0>N

ST1W { <Zt1>.S, <Zt2>.S, <Zt3>.S, <Zt4>.S }, <PNg>, [<Xn|SP>{, #<imm>, MUL VL}]

if !IsFeatureImplemented(FEAT_SME2) then UNDEFINED; constant integer n = UInt(Rn); constant integer g = UInt('1':PNg); constant integer nreg = 4; constant integer tstride = 4; constant integer t = UInt(T:'00':Zt); constant integer esize = 32; constant integer offset = SInt(imm4);

Assembler Symbols

<Zt1>

For the two registers variant: is the name of the first scalable vector register Z0-Z7 or Z16-Z23 to be transferred, encoded as "T:'0':Zt".

For the four registers variant: is the name of the first scalable vector register Z0-Z3 or Z16-Z19 to be transferred, encoded as "T:'00':Zt".

<Zt2>

For the two registers variant: is the name of the second scalable vector register Z8-Z15 or Z24-Z31 to be transferred, encoded as "T:'1':Zt".

For the four registers variant: is the name of the second scalable vector register Z4-Z7 or Z20-Z23 to be transferred, encoded as "T:'01':Zt".

<Zt3>

Is the name of the third scalable vector register Z8-Z11 or Z24-Z27 to be transferred, encoded as "T:'10':Zt".

<Zt4>

Is the name of the fourth scalable vector register Z12-Z15 or Z28-Z31 to be transferred, encoded as "T:'11':Zt".

<PNg>

Is the name of the governing scalable predicate register PN8-PN15, with predicate-as-counter encoding, encoded in the "PNg" field.

<Xn|SP>

Is the 64-bit name of the general-purpose base register or stack pointer, encoded in the "Rn" field.

<imm>

For the two registers variant: is the optional signed immediate vector offset, a multiple of 2 in the range -16 to 14, defaulting to 0, encoded in the "imm4" field.

For the four registers variant: is the optional signed immediate vector offset, a multiple of 4 in the range -32 to 28, defaulting to 0, encoded in the "imm4" field.

Operation

CheckStreamingSVEEnabled(); constant integer VL = CurrentVL; constant integer PL = VL DIV 8; constant integer elements = VL DIV esize; constant integer mbytes = esize DIV 8; bits(64) base; bits(64) addr; bits(VL) src; constant bits(PL) pred = P[g, PL]; constant bits(PL * nreg) mask = CounterToPredicate(pred<15:0>, PL * nreg); constant boolean contiguous = TRUE; constant boolean nontemporal = FALSE; integer transfer = t; constant boolean tagchecked = n != 31; constant AccessDescriptor accdesc = CreateAccDescSVE(MemOp_STORE, nontemporal, contiguous, tagchecked); if !AnyActiveElement(mask, esize) then if n == 31 && ConstrainUnpredictableBool(Unpredictable_CHECKSPNONEACTIVE) then CheckSPAlignment(); else if n == 31 then CheckSPAlignment(); base = if n == 31 then SP[] else X[n, 64]; addr = AddressAdd(base, offset * nreg * elements * mbytes, accdesc); for r = 0 to nreg-1 src = Z[transfer, VL]; for e = 0 to elements-1 if ActivePredicateElement(mask, r * elements + e, esize) then Mem[addr, mbytes, accdesc] = Elem[src, e, esize]; addr = AddressIncrement(addr, mbytes, accdesc); transfer = transfer + tstride;

Operational information

If PSTATE.DIT is 1, the timing of this instruction is insensitive to the value of the data being loaded or stored when its governing predicate register contains the same value for each execution.


Internal version only: aarchmrs v2024-03_relA, pseudocode v2024-03_rel, sve v2024-03_rel ; Build timestamp: 2024-03-26T09:45

Copyright © 2010-2024 Arm Limited or its affiliates. All rights reserved. This document is Non-Confidential.