FDOT (8-bit floating-point to half-precision, vector)
8-bit floating-point dot product to half-precision (vector)
This instruction computes the fused sum-of-products of a group of two 8-bit
floating-point values held in each 16-bit element of the first and
second source vectors. The half-precision sum-of-products are scaled by
2-UInt(FPMR.LSCALE[3:0]), before being destructively added without
intermediate rounding to the corresponding half-precision elements
of the destination vector.
The 8-bit floating-point encoding format for the elements of the first
source vector is selected by FPMR.F8S1.
The 8-bit floating-point encoding format for the elements of the second
source vector is selected by FPMR.F8S2.
0
0
0
1
1
1
0
0
1
0
1
1
1
1
1
1
FDOT <Vd>.<Ta>, <Vn>.<Tb>, <Vm>.<Tb>
if !IsFeatureImplemented(FEAT_FP8DOT2) then UNDEFINED;
constant integer d = UInt(Rd);
constant integer n = UInt(Rn);
constant integer m = UInt(Rm);
constant integer datasize = if Q == '1' then 128 else 64;
constant integer esize = 16;
constant integer elements = datasize DIV esize;
<Vd>
Is the name of the SIMD&FP destination register, encoded in the "Rd" field.
<Ta>
Is an arrangement specifier,
<Vn>
Is the name of the first SIMD&FP source register, encoded in the "Rn" field.
<Tb>
Is an arrangement specifier,
<Vm>
Is the name of the second SIMD&FP source register, encoded in the "Rm" field.
CheckFPMREnabled(); CheckFPAdvSIMDEnabled64();
constant bits(datasize) operand1 = V[n, datasize];
constant bits(datasize) operand2 = V[m, datasize];
constant bits(datasize) operand3 = V[d, datasize];
bits(datasize) result;
for e = 0 to elements-1
constant bits(esize) op1 = Elem[operand1, e, esize];
constant bits(esize) op2 = Elem[operand2, e, esize];
bits(esize) sum = Elem[operand3, e, esize];
sum = FP8DotAddFP(sum, op1, op2, FPCR, FPMR);
Elem[result, e, esize] = sum;
V[d, datasize] = result;