672 AMD Geode™ LX Processors Data Book
Instruction Set
33234H
PFMAX Packed Floating-Point MAXimum 2
MMX Register1 with MMX Register2 0F0F [11 mm1
mm2] A4
MMX reg 1[dword] <--- MMX reg 1 [dword] --- if (MMX reg 1 [dword] >
MMX reg 2 [dword])
MMX reg 1 [dword] <--- MMX reg 2 [dword] --- if (MMX reg 1 [dword]
NOT > MMX reg 2 [dword])
MMX Register with Memory64 0F0F [mod mm r/m]
A4
MMX reg [dword] <--- MMX reg [dword] --- if (MMX reg [dword] >
Memory64 [dword])
MMX reg [dword] <--- Memory [dword --- if (MMX reg [dword] NOT >
Memory64 [dword])
PFMIN Packed Floating - Point MINimum 2
MMX Register 1 with MMX Register2 0F0F [11 mm1
mm2] 94
MMX reg 1 [dword] <--- MMX reg 1 [dword] --- if (MMX reg 1 [dword] <
MMX reg 2 [dword])
MMX reg 1 [dword] <--- MMX reg 1 [dword] --- if (MMX reg 1 [dword]
NOT < MMX reg 2 [dword])
MMX register1 with Mwnory64 0F0F [mod mm r/m]
94
MMX reg [dword] <--- MMX reg [dword] --- if (MMX reg [dword] <
Memory64 [dword])
MMX reg [dword] <--- Memory64 [dword] --- if (MMX reg [dword] NOT
< Memory64 [dword])
PFMUL Packed Floating-Point Multiplication 2
MMX Register 1 with MMX Register 2 0F0F [11 mm1
mm2] B4
MMX reg 1 [dword] <--- sat --- MMX reg 1 [dword] * MMX reg 2 [dword]
MMX Register with Memory64 0F0F [mod mm 2]
B4
MMX reg [dword] <--- sat --- MMX reg [dword] * Memory64 [dword]
PFNACC Packed Floating-Point Negative Accumulate 2
MMX Register1 with MMX Register2 0F0F [11 mm1
mm2] 8A
MMX reg 1 [low dword] <--- (MMX reg 1 [low dword] - MMX reg 1 [high
dword])
MMX reg 1 [high dword] <--- (MMX reg 2 [low dword] - MMX reg 2 [high
dword])
MMX Register with Memory64 0F0F [mod mm r/m]
8A
MMX reg [low dword] <--- (MMX reg [low dword] - MMX reg [high
dword])
MMX reg [high dword] <--- (Memory64 [low dword] - Memory64 [high
dword])
PFPNACC Packed Floating-Point Mixed Positive-Negative Accumulate 2
MMX Register1 with MMX Register2 0F0F [11 mm1
mm2] 8E
MMX reg 1 [low dword] <--- (MMX reg 1 [low dword] - MMX reg 1 [high
dword])
MMX reg 1 [high dword] <--- (MMX reg 2 [low dword] + MMX reg 2
[high dword])
MMX Register with Memory64 0F0F [mod mm r/m]
8E
MMX reg [low dword] <--- (MMX reg [low dword] - MMX reg [low
dword])
MMX reg [high dword] <--- (Memory64 [low dword] - Memory64 [high
dword])
PFRCP Floating-Point Reciprocal Approximation 21
MMX Register1 with MMX Register2 0F0F [11 mm1
mm2] 96
MMX reg 1 [low dword] <--- sat --- reciprocal --- MMX reg 2 [low dword]
MMX reg 1 [high dword] <--- sat --- reciprocal --- MMX reg 2 [low
dword]
MMX Register with Memory64 0F0F [mod mm r/m]
96
MMX reg [Low dword] <--- sat --- reciprocal --- Memory64 [low dword]
MMX reg [high dword] <--- sat --- reciprocal --- Memory64 [low dword]
PFRCPV Floating-Point Reciprocal Vector 23
MMX Register1 with MMX Register 0F0F [11 mm1
mm2] 86
MMX reg 1 [low dword] <---sat --- reciprocal --- MMX reg 2 [low dword]
MMX reg 1 [high dword] <--- sat --- reciprocal MMX reg 2 [high dword]
MMX Register with Memory64 0F0F [mod mm r/m]
86
MMX reg [low dword] <---sat --- reciprocal Value - Memory64 [low
dword]
MMX reg [high dword] <--- sat --- reciprocal value - Memory64 [high
dword]
PFRCPIT1 Packed Floating-Point Reciprocal, First Iteration Step 11, 2
MMX Register1 with MMX Register 2 0F0F [11 mm1
mm2] A6
MMX reg 1 [dword] <--- move --- MMX reg 2 [dword]
MMX Register with Memory64 0F0F [mod mm r/m]
A6
MMX reg [dword] <-- move --- Memory64 [dword]
PFRCPIT2 Packed Floating-Point Reciprocal/Reciprocal Square Root, Second Iteration Step 11, 2
MMX Register 1 with MMX Register 2 0FDF [11 mm1
mm2] B6
MMX reg 1 [dword] <--- move --- MMX reg 2 [dword]
MMX Register with Memory64 0FDF [mod mm r/m]
B6
MMX reg [dword] <--- move --- Memory64 [dword]
Table 8-30. AMD 3DNow!™ Technology Instruction Set (Continued)
AMD 3DNow!™ Instructions Opcode/imm8 Operation
Clk
Cnt Notes