MOV (R8l, M8) - Latency

Experiment 1

• Instruction: MOV R8B, byte ptr [R14]
• Code:

     0:	45 8a 06             	mov    r8b,BYTE PTR [r14]
  

• Show nanoBench command
• Results:
  • Instructions retired: 1.0
  • Core cycles: 1.0
  • Reference cycles: 0.67
  • UOPS_EXECUTED.THREAD: 2.0

Experiment 2

• Instruction: MOV R8B, byte ptr [R14]
• Chain instructions: MOVSX R8, R8B;MOVSX R8, R8D;MOVSX R8, R8D;MOVSX R8, R8D;MOVSX R8, R8D
• Chain latency: 5
• Code:

     0:	45 8a 06             	mov    r8b,BYTE PTR [r14]
     3:	4d 0f be c0          	movsx  r8,r8b
     7:	4d 63 c0             	movsxd r8,r8d
     a:	4d 63 c0             	movsxd r8,r8d
     d:	4d 63 c0             	movsxd r8,r8d
    10:	4d 63 c0             	movsxd r8,r8d
  

• Show nanoBench command
• Results:
  • Instructions retired: 6.0
  • Core cycles: 6.0
  • Reference cycles: 5.32
  • UOPS_EXECUTED.THREAD: 7.0

Experiment 1

• Instruction: MOV R8B, byte ptr [R14]
• Chain instructions: MOVSX R8, R8B;XOR R14, R8;XOR R14, R8;XOR R14, R8;XOR R14, R8
• Chain latency: 5
• Code:

     0:	45 8a 06             	mov    r8b,BYTE PTR [r14]
     3:	4d 0f be c0          	movsx  r8,r8b
     7:	4d 31 c6             	xor    r14,r8
     a:	4d 31 c6             	xor    r14,r8
     d:	4d 31 c6             	xor    r14,r8
    10:	4d 31 c6             	xor    r14,r8
  

• Show nanoBench command
• Results:
  • Instructions retired: 6.0
  • Core cycles: 11.0
  • Reference cycles: 9.98
  • UOPS_EXECUTED.THREAD: 7.0

Experiment 1

• Instruction: MOV R8B, byte ptr [R14+R13]
• Chain instructions: MOVSX R8, R8B;XOR R13, R8;XOR R13, R8;XOR R13, R8;XOR R13, R8
• Chain latency: 5
• Code:

     0:	47 8a 04 2e          	mov    r8b,BYTE PTR [r14+r13*1]
     4:	4d 0f be c0          	movsx  r8,r8b
     8:	4d 31 c5             	xor    r13,r8
     b:	4d 31 c5             	xor    r13,r8
     e:	4d 31 c5             	xor    r13,r8
    11:	4d 31 c5             	xor    r13,r8
  

• Init:

  XOR R13, R13

• Show nanoBench command
• Results:
  • Instructions retired: 6.0
  • Core cycles: 11.01
  • Reference cycles: 10.42
  • UOPS_EXECUTED.THREAD: 7.0

Experiment 1

• Instruction: MOV R8B, byte ptr [R14]
• Chain instructions: MOVSX R12, R8B;MOVSX R12, R12d;MOVSX R12, R12d;MOVSX R12, R12d;mov [R14], R12B
• Chain latency: ≥5
• Code:

     0:	45 8a 06             	mov    r8b,BYTE PTR [r14]
     3:	4d 0f be e0          	movsx  r12,r8b
     7:	4d 63 e4             	movsxd r12,r12d
     a:	4d 63 e4             	movsxd r12,r12d
     d:	4d 63 e4             	movsxd r12,r12d
    10:	45 88 26             	mov    BYTE PTR [r14],r12b
  

• Show nanoBench command
• Results:
  • Instructions retired: 6.0
  • Core cycles: 8.0
  • Reference cycles: 7.32
  • UOPS_EXECUTED.THREAD: 8.0