STNT1H (scalar plus immediate, strided registers)

Contiguous store non-temporal of halfwords from multiple strided vectors (immediate index)

This instruction performs a contiguous non-temporal store of halfwords 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.

A non-temporal store is a hint to the system that this data is unlikely to be referenced again soon.

Encoding: Two registers

Variants: FEAT_SME2 (ARMv9.3)

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
1	0	1	0	0	0	0	1	0	1	1	0					0	0	1										1
												imm4					msz		PNg			Rn					T	N	Zt

STNT1H { <Zt1>.H, <Zt2>.H }, <PNg>, [<Xn|SP>{, #<imm>, MUL VL}]

Decoding algorithm

if !IsFeatureImplemented(FEAT_SME2) then EndOfDecode(Decode_UNDEF);
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 = 16;
constant integer offset = SInt(imm4);

Encoding: Four registers

Variants: FEAT_SME2 (ARMv9.3)

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
1	0	1	0	0	0	0	1	0	1	1	0					1	0	1										1	0
												imm4					msz		PNg			Rn					T	N		Zt

STNT1H { <Zt1>.H, <Zt2>.H, <Zt3>.H, <Zt4>.H }, <PNg>, [<Xn|SP>{, #<imm>, MUL VL}]

Decoding algorithm

if !IsFeatureImplemented(FEAT_SME2) then EndOfDecode(Decode_UNDEF);
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 = 16;
constant integer offset = SInt(imm4);

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 = TRUE;
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[64] 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;

Explanations

<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".
<Zt1>: 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".
<Zt2>: 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".
<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.
<imm>: 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.
<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".

Operational Notes

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.