SSE4.1 Instruction Set Architecture (ISA) Extension

SSE4.1 instructions added with the Penryn processors:

• BLENDPD
  Blend Packed Double Precision Floating-Point Values
• BLENDPS
  Blend Packed Single Precision Floating-Point Values
• BLENDVPD
  Variable Blend Packed Double Precision Floating-Point Values
• BLENDVPS
  Variable Blend Packed Single Precision Floating-Point Values
• DPPD
  Dot Product of Packed Double Precision Floating-Point Values
• DPPS
  Dot Product of Packed Single Precision Floating-Point Values
• EXTRACTPS
  Extract Packed Single Precision Floating-Point Value
• INSERTPS
  Insert Packed Single Precision Floating-Point Value
• MOVNTDQA
  Load Double Quadword Non-Temporal Aligned Hint
• MPSADBW
  Compute Multiple Packed Sums of Absolute Difference
• PACKUSDW
  Pack with Unsigned Saturation
• PBLENDVB
  Variable Blend Packed Bytes
• PBLENDW
  Blend Packed Words
• PCMPEQQ
  Compare Packed Qword Data for Equal
• PEXTRB
  Extract Byte
• PEXTRD/PEXTRQ
  Extract Dword/Qword
• PEXTRW
  Extract Word
• PHMINPOSUW
  Packed Horizontal Word Minimum
• PINSRB
  Insert Byte
• PINSRD/PINSRQ
  Insert Dword/Qword
• PINSRW
  Insert Word
• PMAXSB
  Maximum of Packed Signed Byte Integers
• PMAXSD
  Maximum of Packed Signed Dword Integers
• PMAXUD
  Maximum of Packed Unsigned Dword Integers
• PMAXUW
  Maximum of Packed Word Integers
• PMINSB
  Minimum of Packed Signed Byte Integers
• PMINSD
  Minimum of Packed Dword Integers
• PMINUD
  Minimum of Packed Dword Integers
• PMINUW
  Minimum of Packed Word Integers
• PMOVSX (PMOVSXBW, PMOVSXBD, PMOVSXBQ, PMOVSXWD, PMOVSXWQ, PMOVSXDQ)
  Packed Move with Sign Extend
• PMOVZX (PMOVZXBW, PMOVZXBD, PMOVZXBQ, PMOVZXWD, PMOVZXWQ, PMOVZXDQ)
  Packed Move with Zero Extend
• PMULDQ
  Multiply Packed Signed Dword Integers
• PMULLD
  Multiply Packed Signed Dword Integers and Store Low Result
• PTEST
  Logical Compare
• ROUNDPD
  Round Packed Double Precision Floating-Point Values
• ROUNDPS
  Round Packed Single Precision Floating-Point Values
• ROUNDSD
  Round Scalar Double Precision Floating-Point Values
• ROUNDSS
  Round Scalar Single Precision Floating-Point Values