xsmaxqp

VSX Scalar Maximum Quad-Precision

xsmaxqp vD, vA, vB

Selects maximum of two Quad floats.

Details

Selects the larger of two quad-precision floating-point values in VSRs vA and vB according to IEEE 754 total ordering (with +0 greater than −0, and signaling NaN handling) and stores the result in VSR vD. This instruction updates FPSCR with any NaN exception flags.

Pseudocode Operation

if (vA is NaN) and (vB is NaN) then vD ← vA
else if vA is NaN then vD ← vB
else if vB is NaN then vD ← vA
else vD ← max(vA, vB)

Example

xsmaxqp vd, va, vb

Encoding

Binary Layout
63
0
vD
6
vA
11
vB
16
836
21
/
31
 
Format X-form
Opcode 0xFC000684
Extension VSX

Operands

  • vD
    Target
  • vA
    Src A
  • vB
    Src B