Encoding:

Format:

CLASS.fmt 

CLASS fd, fs, fmt

CLASS.S fd,fs

CLASS.D fd,fs

Purpose:

Scalar Floating-Point Class Mask

Scalar floating-point class shown as a bit mask for Zero, Negative, Infinite, Subnormal, Quiet NaN, or Signaling

NaN.

Description:

FPR[fd] = class(FPR[fs])

Stores in fd a bit mask reflecting the floating-point class of the floating point scalar value fs.

The mask has 10 bits as follows. Bits 0 and 1 indicate NaN values: signaling NaN (bit 0) and quiet NaN (bit 1). Bits

2, 3, 4, 5 classify negative values: infinity (bit 2), normal (bit 3), subnormal (bit 4), and zero (bit 5). Bits 6, 7, 8, 9 classify positive values: infinity (bit 6), normal (bit 7), subnormal (bit 8), and zero (bit 9).

This instruction corresponds to the class operation of the IEEE Standard for Floating-Point Arithmetic 754^TM-2008.

This scalar FPU instruction also corresponds to the vector FCLASS.df instruction of MSA.

The input values and generated bit masks are not affected by the flush-subnormal-to-zero mode FCSR.FS.

The input operand is a scalar value in floating-point data format fmt. Bits beyond the width of fmt are ignored. The result is a 10-bit bitmask as described above, zero extended to fmt-width bits. Coprocessor register bits beyond fmtwidth bits are UNPREDICTABLE (e.g., for CLASS.S bits 32-63 are UNPREDICTABLE on a 64-bit FPU, while bits

32-128 bits are UNPREDICTABLE if the processor supports MSA).

Restrictions:

No data-dependent exceptions are possible.

Availability and Compatibility:

This instruction is introduced by and required as of Release 6.

CLASS.fmt is defined only for formats S and D. Other formats must produce a Reserved Instruction exception

(unless used for a different instruction).

Operation:

if not IsCoprocessorEnabled(1) 
   then SignalException(CoprocessorUnusable, 1) endif
if not IsFloatingPointImplemented(fmt)) 
   then SignalException(ReservedInstruction) endif
fin = ValueFPR(fs,fmt)
masktmp = ClassFP(fin, fmt)
StoreFPR (fd, fmt, ftmp )
/* end of instruction */
function ClassFP(tt, ts, n)
/* Implementation defined class operation. */
endfunction ClassFP

Exceptions:

Coprocessor Unusable, Reserved Instruction

Floating Point Exceptions:

Unimplemented Operation