CLASTB (scalar)

Conditionally extract last element to general-purpose register

From the source vector register extract the last active element, and then zero-extend that element to destructively place in the destination and first source general-purpose register. If there are no active elements then destructively zero-extend the least significant element-size bits of the destination and first source general-purpose register.

Encoding: SVE

Variants: FEAT_SVE || FEAT_SME (FEAT_SVE || FEAT_SME)

313029282726252423222120191817161514131211109876543210
00000101110001101
sizeBPgZmRdn

CLASTB <R><dn>, <Pg>, <R><dn>, <Zm>.<T>

Decoding algorithm

if !IsFeatureImplemented(FEAT_SVE) && !IsFeatureImplemented(FEAT_SME) then
    EndOfDecode(Decode_UNDEF);
constant integer esize = 8 << UInt(size);
constant integer g = UInt(Pg);
constant integer dn = UInt(Rdn);
constant integer m = UInt(Zm);
constant integer csize = if esize < 64 then 32 else 64;
constant boolean isBefore = TRUE;

Operation

CheckSVEEnabled();
constant integer VL = CurrentVL;
constant integer PL = VL DIV 8;
constant integer elements = VL DIV esize;
constant bits(PL) mask = P[g, PL];
constant bits(esize) operand1 = X[dn, esize];
constant bits(VL) operand2 = Z[m, VL];
bits(csize) result;
integer last = LastActiveElement(mask, esize);

if last < 0 then
    result = ZeroExtend(operand1, csize);
else
    if !isBefore then
        last = last + 1;
        if last >= elements then last = 0;
    result = ZeroExtend(Elem[operand2, last, esize], csize);

X[dn, csize] = result;

Explanations

<R>: <dn>: Is the number [0-30] of the source and destination general-purpose register or the name ZR (31), encoded in the "Rdn" field.
<Pg>: Is the name of the governing scalable predicate register P0-P7, encoded in the "Pg" field.
<Zm>: Is the name of the source scalable vector register, encoded in the "Zm" field.
<T>:

Operational Notes

If FEAT_SME is implemented and the PE is in Streaming SVE mode, then any subsequent instruction which is dependent on the general-purpose register written by this instruction might be significantly delayed.