← Home

CLASTA (scalar)

Conditionally extract element after last to general-purpose register

This instruction extracts and zero-extends the element after the Last active element, or if the Last active element is the final element it extracts element zero, from the source vector register, and destructively places the zero-extended value in the destination and first source general-purpose register. If there are no Active elements, the least significant element-size bits of the destination and first source general-purpose register are destructively zero-extended.

SVE class

(FEAT_SVE || FEAT_SME)

313029282726252423222120191817161514131211109876543210
00000101size110000101PgZmRdn
B

Encoding

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

Decode

if !IsFeatureImplemented(FEAT_SVE) && !IsFeatureImplemented(FEAT_SME) then EndOfDecode(Decode_UNDEF); end; let esize : integer{} = 8 << UInt(size); let g : integer = UInt(Pg); let dn : integer = UInt(Rdn); let m : integer = UInt(Zm); let csize : integer{} = if esize < 64 then 32 else 64; let isBefore : boolean = FALSE;

Assembler Symbols

<R>

Is a width specifier, encoded in size:

size <R>
00 W
01 W
10 W
11 X
<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>

Is the size specifier, encoded in size:

size <T>
00 B
01 H
10 S
11 D

Operation

CheckSVEEnabled(); let VL : integer{} = CurrentVL(); let PL : integer{} = VL DIV 8; let elements : integer = VL DIV esize; let mask : bits(PL) = P{}(g); let operand1 : bits(esize) = X{}(dn); let operand2 : bits(VL) = Z{}(m); var result : bits(csize); var last : integer = LastActiveElement{PL}(mask, esize); if last < 0 then result = ZeroExtend{csize}(operand1); else if !isBefore then last = last + 1; if last >= elements then last = 0; end; end; result = ZeroExtend{csize}(operand2[last*:esize]); end; X{csize}(dn) = result;

Operational information

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


Version 2026.06 — Copyright © 2010-2026 Arm Limited or its affiliates.

This site is provided as a community resource and is NOT affiliated with nor endorsed by Arm Limited.