RCR (R32, I8) - Latency

Experiment 1

• Instruction: RCR R8D, 2
• Chain instructions: MOVSX R8, R8D;MOVSX R8, R8D;MOVSX R8, R8D;MOVSX R8, R8D;MOVSX R8, R8D;MOVSX R8, R8D;MOVSX R8, R8D;MOVSX R8, R8D;MOVSX R8, R8D
• Chain latency: 9
• Code:

     0:	41 c1 d8 02          	rcr    r8d,0x2
     4:	4d 63 c0             	movsxd r8,r8d
     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
    13:	4d 63 c0             	movsxd r8,r8d
    16:	4d 63 c0             	movsxd r8,r8d
    19:	4d 63 c0             	movsxd r8,r8d
    1c:	4d 63 c0             	movsxd r8,r8d
  

• Show nanoBench command
• Results:
  • Instructions retired: 10.0
  • Core cycles: 12.67
  • Reference cycles: 12.58
  • UOPS_RETIRED.ALL: 17.0

Experiment 2

• Instruction: RCR R8D, 2
• Code:

     0:	41 c1 d8 02          	rcr    r8d,0x2
  

• Show nanoBench command
• Results:
  • Instructions retired: 1.0
  • Core cycles: 5.0
  • Reference cycles: 4.87
  • UOPS_RETIRED.ALL: 8.0

Experiment 3 (with dependency-breaking instructions)

• Instruction: RCR R8D, 2
• Chain instruction: TEST R15, R15
• Code:

     0:	41 c1 d8 02          	rcr    r8d,0x2
     4:	4d 85 ff             	test   r15,r15
  

• Show nanoBench command
• Results:
  • Instructions retired: 2.0
  • Core cycles: 5.99
  • Reference cycles: 6.06
  • UOPS_RETIRED.ALL: 9.0

Experiment 1

• Instruction: RCR R8D, 2
• Chain instructions: CMOVC R15D, R15D;MOVSX R8, R15D;MOVSX R8, R8D;MOVSX R8, R8D;MOVSX R8, R8D;MOVSX R8, R8D;MOVSX R8, R8D;MOVSX R8, R8D;MOVSX R8, R8D;MOVSX R8, R8D
• Chain latency: 11
• Code:

     0:	41 c1 d8 02          	rcr    r8d,0x2
     4:	45 0f 42 ff          	cmovb  r15d,r15d
     8:	4d 63 c7             	movsxd r8,r15d
     b:	4d 63 c0             	movsxd r8,r8d
     e:	4d 63 c0             	movsxd r8,r8d
    11:	4d 63 c0             	movsxd r8,r8d
    14:	4d 63 c0             	movsxd r8,r8d
    17:	4d 63 c0             	movsxd r8,r8d
    1a:	4d 63 c0             	movsxd r8,r8d
    1d:	4d 63 c0             	movsxd r8,r8d
    20:	4d 63 c0             	movsxd r8,r8d
  

• Show nanoBench command
• Results:
  • Instructions retired: 11.0
  • Core cycles: 15.33
  • Reference cycles: 15.3
  • UOPS_RETIRED.ALL: 19.0

Experiment 1

• Instruction: RCR R8D, 2
• Chain instructions: MOVSX R8, R8D;MOVSX R8, R8D;MOVSX R8, R8D;MOVSX R8, R8D;MOVSX R8, R8D;MOVSX R8, R8D;MOVSX R8, R8D;MOVSX R8, R8D;TEST R8D, R8D
• Chain latency: 9
• Code:

     0:	41 c1 d8 02          	rcr    r8d,0x2
     4:	4d 63 c0             	movsxd r8,r8d
     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
    13:	4d 63 c0             	movsxd r8,r8d
    16:	4d 63 c0             	movsxd r8,r8d
    19:	4d 63 c0             	movsxd r8,r8d
    1c:	45 85 c0             	test   r8d,r8d
  

• Show nanoBench command
• Results:
  • Instructions retired: 10.0
  • Core cycles: 12.68
  • Reference cycles: 12.52
  • UOPS_RETIRED.ALL: 17.0

Experiment 2

• Instruction: RCR R8D, 2
• Chain instruction: TEST R8D, R8D
• Chain latency: 1
• Code:

     0:	41 c1 d8 02          	rcr    r8d,0x2
     4:	45 85 c0             	test   r8d,r8d
  

• Show nanoBench command
• Results:
  • Instructions retired: 2.0
  • Core cycles: 6.01
  • Reference cycles: 5.84
  • UOPS_RETIRED.ALL: 9.0

Experiment 3 (with dependency-breaking instructions)

• Instruction: RCR R8D, 2
• Chain instructions: TEST R8D, R8D;;MOV R8D, 0
• Chain latency: 1
• Code:

     0:	41 c1 d8 02          	rcr    r8d,0x2
     4:	45 85 c0             	test   r8d,r8d
     7:	41 b8 00 00 00 00    	mov    r8d,0x0
  

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 6.0
  • Reference cycles: 6.12
  • UOPS_RETIRED.ALL: 10.0

Experiment 1

• Instruction: RCR R8D, 2
• Chain instructions: CMC;CMC;CMC;CMC;CMC;CMC;CMC;CMC
• Chain latency: 8
• Code:

     0:	41 c1 d8 02          	rcr    r8d,0x2
     4:	f5                   	cmc    
     5:	f5                   	cmc    
     6:	f5                   	cmc    
     7:	f5                   	cmc    
     8:	f5                   	cmc    
     9:	f5                   	cmc    
     a:	f5                   	cmc    
     b:	f5                   	cmc    
  

• Show nanoBench command
• Results:
  • Instructions retired: 9.0
  • Core cycles: 12.64
  • Reference cycles: 12.64
  • UOPS_RETIRED.ALL: 16.0

Experiment 2

• Instruction: RCR R8D, 2
• Code:

     0:	41 c1 d8 02          	rcr    r8d,0x2
  

• Show nanoBench command
• Results:
  • Instructions retired: 1.0
  • Core cycles: 5.0
  • Reference cycles: 5.1
  • UOPS_RETIRED.ALL: 8.0

Experiment 3 (with dependency-breaking instructions)

• Instruction: RCR R8D, 2
• Chain instruction: MOV R8D, 0
• Code:

     0:	41 c1 d8 02          	rcr    r8d,0x2
     4:	41 b8 00 00 00 00    	mov    r8d,0x0
  

• Show nanoBench command
• Results:
  • Instructions retired: 2.0
  • Core cycles: 5.99
  • Reference cycles: 5.84
  • UOPS_RETIRED.ALL: 9.0