FCLASS.H
Float Classify (Half)
FCLASS.H rd, rs1
Examines a 16-bit floating-point number and generates a classification bitmask.
Details
Classifies the half-precision (16-bit) floating-point value in rs1, writing a 10-bit one-hot result to integer rd. Bits represent: negative infinity, negative normal, negative subnormal, negative zero, positive zero, positive subnormal, positive normal, positive infinity, signalling NaN, quiet NaN.
Pseudocode Operation
R[rd] = classify_half(F[rs1]);
Example
FCLASS.H t0, a0
Encoding
Binary Layout
1110000
31:25
00000
24:20
rs1
19:15
001
14:12
rd
11:7
1010011
6:0
Operands
-
rd
Dest (Int) -
rs1
Src (Half)