LUTI4

Lookup table read with 4-bit indices

This instruction copies indexed 8-bit or 16-bit elements from the one or two table vectors to the destination vector using packed 4-bit indices from a segment of the source vector. A segment corresponds to a portion of the source vector that is consumed in order to fill the destination vector. The segment is selected by the vector segment index.

Encoding: Advanced SIMD

Variants: FEAT_AdvSIMD && FEAT_LUT (FEAT_AdvSIMD && FEAT_LUT)

313029282726252423222120191817161514131211109876543210
01001110010000
Qop2RmlenopRnRd

Byte (len == x1 && op == 0)

LUTI4 <Vd>.16B, { <Vn>.16B }, <Vm>[<index>]

Halfword (op == 1)

LUTI4 <Vd>.8H, { <Vn1>.8H, <Vn2>.8H }, <Vm>[<index>]

Decoding algorithm

if !IsFeatureImplemented(FEAT_AdvSIMD) || !IsFeatureImplemented(FEAT_LUT) then
    EndOfDecode(Decode_UNDEF);
if len<0> == '0' && op == '0' then EndOfDecode(Decode_UNDEF);

constant integer d = UInt(Rd);
constant integer n = UInt(Rn);
constant integer m = UInt(Rm);
constant integer isize = 4;
constant integer esize = 8 << UInt(op);
constant integer ntblr = 1 << UInt(op);
constant integer part = if op == '0' then UInt(len<1>) else UInt(len);

Operation

CheckFPAdvSIMDEnabled64();
constant integer elements = 128 DIV esize;
constant integer ibase = elements * part;
constant bits(128) indices = V[m, 128];
constant bits(128) table1  = V[n+0, 128];
constant bits(128) table2  = if ntblr == 2 then V[(n+1) MOD 32, 128] else Zeros(128);
bits(128) result;
bits(esize) res;

for e = 0 to elements-1
    constant integer index = UInt(Elem[indices, ibase + e, isize]);
    if index < elements then
        res = Elem[table1, index, esize];
    else
        assert ntblr == 2;
        res = Elem[table2, index - elements, esize];
    Elem[result, e, esize] = res;

V[d, 128] = result;

Explanations

<Vd>: Is the name of the SIMD&FP destination register, encoded in the "Rd" field.
<Vn>: Is the name of the SIMD&FP table register, encoded in the "Rn" field.
<Vm>: Is the name of the SIMD&FP source register, encoded in the "Rm" field.
<index>: For the "Byte" variant: is the vector segment index, in the range 0 to 1, encoded in the "len<1>" field.
<index>: For the "Halfword" variant: is the vector segment index, in the range 0 to 3, encoded in the "len" field.
<Vn1>: Is the name of the first SIMD&FP table register, encoded in the "Rn" field.
<Vn2>: Is the name of the second SIMD&FP table register, encoded as "Rn" plus 1 modulo 32.

Operational Notes

If PSTATE.DIT is 1: