PCMPESTRM (XMM, XMM, I8) - Latency

Experiment 1

• Instruction: PCMPESTRM XMM1, XMM2, 2
• Chain instructions: SHUFPD XMM1, XMM0, 0;SHUFPD XMM1, XMM1, 0;SHUFPD XMM1, XMM1, 0;SHUFPD XMM1, XMM1, 0;SHUFPD XMM1, XMM1, 0;SHUFPD XMM1, XMM1, 0;SHUFPD XMM1, XMM1, 0;SHUFPD XMM1, XMM1, 0;SHUFPD XMM1, XMM1, 0;SHUFPD XMM1, XMM1, 0;SHUFPD XMM1, XMM1, 0;SHUFPD XMM1, XMM1, 0;SHUFPD XMM1, XMM1, 0;SHUFPD XMM1, XMM1, 0;SHUFPD XMM1, XMM1, 0
• Chain latency: 15
• Code:

     0:	66 0f 3a 60 ca 02    	pcmpestrm xmm1,xmm2,0x2
     6:	66 0f c6 c8 00       	shufpd xmm1,xmm0,0x0
     b:	66 0f c6 c9 00       	shufpd xmm1,xmm1,0x0
    10:	66 0f c6 c9 00       	shufpd xmm1,xmm1,0x0
    15:	66 0f c6 c9 00       	shufpd xmm1,xmm1,0x0
    1a:	66 0f c6 c9 00       	shufpd xmm1,xmm1,0x0
    1f:	66 0f c6 c9 00       	shufpd xmm1,xmm1,0x0
    24:	66 0f c6 c9 00       	shufpd xmm1,xmm1,0x0
    29:	66 0f c6 c9 00       	shufpd xmm1,xmm1,0x0
    2e:	66 0f c6 c9 00       	shufpd xmm1,xmm1,0x0
    33:	66 0f c6 c9 00       	shufpd xmm1,xmm1,0x0
    38:	66 0f c6 c9 00       	shufpd xmm1,xmm1,0x0
    3d:	66 0f c6 c9 00       	shufpd xmm1,xmm1,0x0
    42:	66 0f c6 c9 00       	shufpd xmm1,xmm1,0x0
    47:	66 0f c6 c9 00       	shufpd xmm1,xmm1,0x0
    4c:	66 0f c6 c9 00       	shufpd xmm1,xmm1,0x0
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM2, XMM2

• Show nanoBench command
• Results:
  • Instructions retired: 16.0
  • Core cycles: 21.46
  • Reference cycles: 18.01
  • UOPS_RETIRED.ALL: 26.0

Experiment 2

• Instruction: PCMPESTRM XMM1, XMM2, 2
• Chain instructions: PSHUFD XMM1, XMM0, 0;PAND XMM1, XMM1;PAND XMM1, XMM1;PAND XMM1, XMM1;PAND XMM1, XMM1;PAND XMM1, XMM1;PAND XMM1, XMM1;PAND XMM1, XMM1;PAND XMM1, XMM1;PAND XMM1, XMM1;PAND XMM1, XMM1;PAND XMM1, XMM1;PAND XMM1, XMM1;PAND XMM1, XMM1;PAND XMM1, XMM1
• Chain latency: 15
• Code:

     0:	66 0f 3a 60 ca 02    	pcmpestrm xmm1,xmm2,0x2
     6:	66 0f 70 c8 00       	pshufd xmm1,xmm0,0x0
     b:	66 0f db c9          	pand   xmm1,xmm1
     f:	66 0f db c9          	pand   xmm1,xmm1
    13:	66 0f db c9          	pand   xmm1,xmm1
    17:	66 0f db c9          	pand   xmm1,xmm1
    1b:	66 0f db c9          	pand   xmm1,xmm1
    1f:	66 0f db c9          	pand   xmm1,xmm1
    23:	66 0f db c9          	pand   xmm1,xmm1
    27:	66 0f db c9          	pand   xmm1,xmm1
    2b:	66 0f db c9          	pand   xmm1,xmm1
    2f:	66 0f db c9          	pand   xmm1,xmm1
    33:	66 0f db c9          	pand   xmm1,xmm1
    37:	66 0f db c9          	pand   xmm1,xmm1
    3b:	66 0f db c9          	pand   xmm1,xmm1
    3f:	66 0f db c9          	pand   xmm1,xmm1
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM2, XMM2

• Show nanoBench command
• Results:
  • Instructions retired: 16.0
  • Core cycles: 21.49
  • Reference cycles: 17.84
  • UOPS_RETIRED.ALL: 26.0

Experiment 3 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM1, XMM2, 2
• Chain instructions: SHUFPD XMM1, XMM0, 0;SHUFPD XMM1, XMM1, 0;SHUFPD XMM1, XMM1, 0;SHUFPD XMM1, XMM1, 0;SHUFPD XMM1, XMM1, 0;SHUFPD XMM1, XMM1, 0;SHUFPD XMM1, XMM1, 0;SHUFPD XMM1, XMM1, 0;SHUFPD XMM1, XMM1, 0;SHUFPD XMM1, XMM1, 0;SHUFPD XMM1, XMM1, 0;SHUFPD XMM1, XMM1, 0;SHUFPD XMM1, XMM1, 0;SHUFPD XMM1, XMM1, 0;SHUFPD XMM1, XMM1, 0
• Chain latency: 15
• Code:

     0:	66 0f 3a 60 ca 02    	pcmpestrm xmm1,xmm2,0x2
     6:	66 0f c6 c8 00       	shufpd xmm1,xmm0,0x0
     b:	66 0f c6 c9 00       	shufpd xmm1,xmm1,0x0
    10:	66 0f c6 c9 00       	shufpd xmm1,xmm1,0x0
    15:	66 0f c6 c9 00       	shufpd xmm1,xmm1,0x0
    1a:	66 0f c6 c9 00       	shufpd xmm1,xmm1,0x0
    1f:	66 0f c6 c9 00       	shufpd xmm1,xmm1,0x0
    24:	66 0f c6 c9 00       	shufpd xmm1,xmm1,0x0
    29:	66 0f c6 c9 00       	shufpd xmm1,xmm1,0x0
    2e:	66 0f c6 c9 00       	shufpd xmm1,xmm1,0x0
    33:	66 0f c6 c9 00       	shufpd xmm1,xmm1,0x0
    38:	66 0f c6 c9 00       	shufpd xmm1,xmm1,0x0
    3d:	66 0f c6 c9 00       	shufpd xmm1,xmm1,0x0
    42:	66 0f c6 c9 00       	shufpd xmm1,xmm1,0x0
    47:	66 0f c6 c9 00       	shufpd xmm1,xmm1,0x0
    4c:	66 0f c6 c9 00       	shufpd xmm1,xmm1,0x0
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM2, XMM2;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 16.0
  • Core cycles: 21.45
  • Reference cycles: 17.84
  • UOPS_RETIRED.ALL: 26.0

Experiment 4 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM1, XMM2, 2
• Chain instructions: PSHUFD XMM1, XMM0, 0;PAND XMM1, XMM1;PAND XMM1, XMM1;PAND XMM1, XMM1;PAND XMM1, XMM1;PAND XMM1, XMM1;PAND XMM1, XMM1;PAND XMM1, XMM1;PAND XMM1, XMM1;PAND XMM1, XMM1;PAND XMM1, XMM1;PAND XMM1, XMM1;PAND XMM1, XMM1;PAND XMM1, XMM1;PAND XMM1, XMM1
• Chain latency: 15
• Code:

     0:	66 0f 3a 60 ca 02    	pcmpestrm xmm1,xmm2,0x2
     6:	66 0f 70 c8 00       	pshufd xmm1,xmm0,0x0
     b:	66 0f db c9          	pand   xmm1,xmm1
     f:	66 0f db c9          	pand   xmm1,xmm1
    13:	66 0f db c9          	pand   xmm1,xmm1
    17:	66 0f db c9          	pand   xmm1,xmm1
    1b:	66 0f db c9          	pand   xmm1,xmm1
    1f:	66 0f db c9          	pand   xmm1,xmm1
    23:	66 0f db c9          	pand   xmm1,xmm1
    27:	66 0f db c9          	pand   xmm1,xmm1
    2b:	66 0f db c9          	pand   xmm1,xmm1
    2f:	66 0f db c9          	pand   xmm1,xmm1
    33:	66 0f db c9          	pand   xmm1,xmm1
    37:	66 0f db c9          	pand   xmm1,xmm1
    3b:	66 0f db c9          	pand   xmm1,xmm1
    3f:	66 0f db c9          	pand   xmm1,xmm1
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM2, XMM2;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 16.0
  • Core cycles: 21.45
  • Reference cycles: 18.35
  • UOPS_RETIRED.ALL: 26.0

Experiment 1

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVC R12, R15; CVTSI2SD XMM0, R12D
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 42 e7          	cmovb  r12,r15
     a:	f2 41 0f 2a c4       	cvtsi2sd xmm0,r12d
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 25.07
  • Reference cycles: 21.04
  • UOPS_RETIRED.ALL: 13.0

Experiment 2

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVC R12, R15; CVTSI2SD XMM0, R12
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 42 e7          	cmovb  r12,r15
     a:	f2 49 0f 2a c4       	cvtsi2sd xmm0,r12
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 25.06
  • Reference cycles: 21.21
  • UOPS_RETIRED.ALL: 13.0

Experiment 3

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVC R12, R15; CVTSI2SS XMM0, R12D
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 42 e7          	cmovb  r12,r15
     a:	f3 41 0f 2a c4       	cvtsi2ss xmm0,r12d
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 25.0
  • Reference cycles: 21.21
  • UOPS_RETIRED.ALL: 13.0

Experiment 4

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVC R12, R15; CVTSI2SS XMM0, R12
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 42 e7          	cmovb  r12,r15
     a:	f3 49 0f 2a c4       	cvtsi2ss xmm0,r12
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 25.07
  • Reference cycles: 21.38
  • UOPS_RETIRED.ALL: 13.0

Experiment 5

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVC R12, R15; MOVD XMM0, R12D
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 42 e7          	cmovb  r12,r15
     a:	66 41 0f 6e c4       	movd   xmm0,r12d
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 21.09
  • Reference cycles: 17.84
  • UOPS_RETIRED.ALL: 13.0

Experiment 6

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVC R12, R15; MOVQ XMM0, R12
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 42 e7          	cmovb  r12,r15
     a:	66 49 0f 6e c4       	movq   xmm0,r12
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 21.09
  • Reference cycles: 17.51
  • UOPS_RETIRED.ALL: 13.0

Experiment 7

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVC R12, R15; PINSRB XMM0, R12D, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 42 e7          	cmovb  r12,r15
     a:	66 41 0f 3a 20 c4 02 	pinsrb xmm0,r12d,0x2
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 21.0
  • Reference cycles: 18.18
  • UOPS_RETIRED.ALL: 13.0

Experiment 8

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVC R12, R15; PINSRD XMM0, R12D, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 42 e7          	cmovb  r12,r15
     a:	66 41 0f 3a 22 c4 02 	pinsrd xmm0,r12d,0x2
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 21.09
  • Reference cycles: 17.51
  • UOPS_RETIRED.ALL: 13.0

Experiment 9

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVC R12, R15; PINSRQ XMM0, R12, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 42 e7          	cmovb  r12,r15
     a:	66 49 0f 3a 22 c4 02 	pinsrq xmm0,r12,0x2
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 21.09
  • Reference cycles: 17.84
  • UOPS_RETIRED.ALL: 13.0

Experiment 10

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVC R12, R15; PINSRW XMM0, R12D, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 42 e7          	cmovb  r12,r15
     a:	66 41 0f c4 c4 02    	pinsrw xmm0,r12d,0x2
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 21.09
  • Reference cycles: 17.51
  • UOPS_RETIRED.ALL: 13.0

Experiment 11

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVP R12, R15; CVTSI2SD XMM0, R12D
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 4a e7          	cmovp  r12,r15
     a:	f2 41 0f 2a c4       	cvtsi2sd xmm0,r12d
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 25.0
  • Reference cycles: 21.21
  • UOPS_RETIRED.ALL: 13.0

Experiment 12

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVP R12, R15; CVTSI2SD XMM0, R12
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 4a e7          	cmovp  r12,r15
     a:	f2 49 0f 2a c4       	cvtsi2sd xmm0,r12
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 25.0
  • Reference cycles: 20.71
  • UOPS_RETIRED.ALL: 13.0

Experiment 13

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVP R12, R15; CVTSI2SS XMM0, R12D
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 4a e7          	cmovp  r12,r15
     a:	f3 41 0f 2a c4       	cvtsi2ss xmm0,r12d
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 24.91
  • Reference cycles: 20.71
  • UOPS_RETIRED.ALL: 13.0

Experiment 14

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVP R12, R15; CVTSI2SS XMM0, R12
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 4a e7          	cmovp  r12,r15
     a:	f3 49 0f 2a c4       	cvtsi2ss xmm0,r12
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 24.98
  • Reference cycles: 21.21
  • UOPS_RETIRED.ALL: 13.0

Experiment 15

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVP R12, R15; MOVD XMM0, R12D
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 4a e7          	cmovp  r12,r15
     a:	66 41 0f 6e c4       	movd   xmm0,r12d
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 20.92
  • Reference cycles: 17.68
  • UOPS_RETIRED.ALL: 13.0

Experiment 16

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVP R12, R15; MOVQ XMM0, R12
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 4a e7          	cmovp  r12,r15
     a:	66 49 0f 6e c4       	movq   xmm0,r12
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 21.0
  • Reference cycles: 17.34
  • UOPS_RETIRED.ALL: 13.0

Experiment 17

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVP R12, R15; PINSRB XMM0, R12D, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 4a e7          	cmovp  r12,r15
     a:	66 41 0f 3a 20 c4 02 	pinsrb xmm0,r12d,0x2
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 21.0
  • Reference cycles: 17.34
  • UOPS_RETIRED.ALL: 13.0

Experiment 18

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVP R12, R15; PINSRD XMM0, R12D, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 4a e7          	cmovp  r12,r15
     a:	66 41 0f 3a 22 c4 02 	pinsrd xmm0,r12d,0x2
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 21.08
  • Reference cycles: 17.51
  • UOPS_RETIRED.ALL: 13.0

Experiment 19

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVP R12, R15; PINSRQ XMM0, R12, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 4a e7          	cmovp  r12,r15
     a:	66 49 0f 3a 22 c4 02 	pinsrq xmm0,r12,0x2
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 20.92
  • Reference cycles: 17.68
  • UOPS_RETIRED.ALL: 13.0

Experiment 20

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVP R12, R15; PINSRW XMM0, R12D, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 4a e7          	cmovp  r12,r15
     a:	66 41 0f c4 c4 02    	pinsrw xmm0,r12d,0x2
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 21.0
  • Reference cycles: 17.51
  • UOPS_RETIRED.ALL: 13.0

Experiment 21

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVS R12, R15; CVTSI2SD XMM0, R12D
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 48 e7          	cmovs  r12,r15
     a:	f2 41 0f 2a c4       	cvtsi2sd xmm0,r12d
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 25.07
  • Reference cycles: 21.21
  • UOPS_RETIRED.ALL: 13.0

Experiment 22

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVS R12, R15; CVTSI2SD XMM0, R12
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 48 e7          	cmovs  r12,r15
     a:	f2 49 0f 2a c4       	cvtsi2sd xmm0,r12
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 24.91
  • Reference cycles: 20.71
  • UOPS_RETIRED.ALL: 13.0

Experiment 23

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVS R12, R15; CVTSI2SS XMM0, R12D
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 48 e7          	cmovs  r12,r15
     a:	f3 41 0f 2a c4       	cvtsi2ss xmm0,r12d
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 25.0
  • Reference cycles: 21.21
  • UOPS_RETIRED.ALL: 13.0

Experiment 24

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVS R12, R15; CVTSI2SS XMM0, R12
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 48 e7          	cmovs  r12,r15
     a:	f3 49 0f 2a c4       	cvtsi2ss xmm0,r12
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 25.0
  • Reference cycles: 20.54
  • UOPS_RETIRED.ALL: 13.0

Experiment 25

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVS R12, R15; MOVD XMM0, R12D
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 48 e7          	cmovs  r12,r15
     a:	66 41 0f 6e c4       	movd   xmm0,r12d
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 21.08
  • Reference cycles: 17.51
  • UOPS_RETIRED.ALL: 13.0

Experiment 26

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVS R12, R15; MOVQ XMM0, R12
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 48 e7          	cmovs  r12,r15
     a:	66 49 0f 6e c4       	movq   xmm0,r12
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 21.0
  • Reference cycles: 18.18
  • UOPS_RETIRED.ALL: 13.0

Experiment 27

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVS R12, R15; PINSRB XMM0, R12D, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 48 e7          	cmovs  r12,r15
     a:	66 41 0f 3a 20 c4 02 	pinsrb xmm0,r12d,0x2
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 21.08
  • Reference cycles: 17.84
  • UOPS_RETIRED.ALL: 13.0

Experiment 28

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVS R12, R15; PINSRD XMM0, R12D, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 48 e7          	cmovs  r12,r15
     a:	66 41 0f 3a 22 c4 02 	pinsrd xmm0,r12d,0x2
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 21.0
  • Reference cycles: 17.51
  • UOPS_RETIRED.ALL: 13.0

Experiment 29

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVS R12, R15; PINSRQ XMM0, R12, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 48 e7          	cmovs  r12,r15
     a:	66 49 0f 3a 22 c4 02 	pinsrq xmm0,r12,0x2
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 21.08
  • Reference cycles: 17.51
  • UOPS_RETIRED.ALL: 13.0

Experiment 30

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVS R12, R15; PINSRW XMM0, R12D, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 48 e7          	cmovs  r12,r15
     a:	66 41 0f c4 c4 02    	pinsrw xmm0,r12d,0x2
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 20.91
  • Reference cycles: 17.84
  • UOPS_RETIRED.ALL: 13.0

Experiment 31

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVO R12, R15; CVTSI2SD XMM0, R12D
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 40 e7          	cmovo  r12,r15
     a:	f2 41 0f 2a c4       	cvtsi2sd xmm0,r12d
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 25.07
  • Reference cycles: 21.04
  • UOPS_RETIRED.ALL: 13.0

Experiment 32

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVO R12, R15; CVTSI2SD XMM0, R12
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 40 e7          	cmovo  r12,r15
     a:	f2 49 0f 2a c4       	cvtsi2sd xmm0,r12
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 24.92
  • Reference cycles: 20.54
  • UOPS_RETIRED.ALL: 13.0

Experiment 33

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVO R12, R15; CVTSI2SS XMM0, R12D
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 40 e7          	cmovo  r12,r15
     a:	f3 41 0f 2a c4       	cvtsi2ss xmm0,r12d
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 24.99
  • Reference cycles: 21.04
  • UOPS_RETIRED.ALL: 13.0

Experiment 34

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVO R12, R15; CVTSI2SS XMM0, R12
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 40 e7          	cmovo  r12,r15
     a:	f3 49 0f 2a c4       	cvtsi2ss xmm0,r12
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 25.07
  • Reference cycles: 21.21
  • UOPS_RETIRED.ALL: 13.0

Experiment 35

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVO R12, R15; MOVD XMM0, R12D
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 40 e7          	cmovo  r12,r15
     a:	66 41 0f 6e c4       	movd   xmm0,r12d
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 21.09
  • Reference cycles: 17.68
  • UOPS_RETIRED.ALL: 13.0

Experiment 36

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVO R12, R15; MOVQ XMM0, R12
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 40 e7          	cmovo  r12,r15
     a:	66 49 0f 6e c4       	movq   xmm0,r12
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 21.0
  • Reference cycles: 17.34
  • UOPS_RETIRED.ALL: 13.0

Experiment 37

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVO R12, R15; PINSRB XMM0, R12D, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 40 e7          	cmovo  r12,r15
     a:	66 41 0f 3a 20 c4 02 	pinsrb xmm0,r12d,0x2
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 21.0
  • Reference cycles: 17.84
  • UOPS_RETIRED.ALL: 13.0

Experiment 38

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVO R12, R15; PINSRD XMM0, R12D, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 40 e7          	cmovo  r12,r15
     a:	66 41 0f 3a 22 c4 02 	pinsrd xmm0,r12d,0x2
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 21.0
  • Reference cycles: 17.68
  • UOPS_RETIRED.ALL: 13.0

Experiment 39

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVO R12, R15; PINSRQ XMM0, R12, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 40 e7          	cmovo  r12,r15
     a:	66 49 0f 3a 22 c4 02 	pinsrq xmm0,r12,0x2
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 21.08
  • Reference cycles: 17.34
  • UOPS_RETIRED.ALL: 13.0

Experiment 40

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVO R12, R15; PINSRW XMM0, R12D, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 40 e7          	cmovo  r12,r15
     a:	66 41 0f c4 c4 02    	pinsrw xmm0,r12d,0x2
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 21.08
  • Reference cycles: 17.51
  • UOPS_RETIRED.ALL: 13.0

Experiment 41

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVZ R12, R15; CVTSI2SD XMM0, R12D
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 44 e7          	cmove  r12,r15
     a:	f2 41 0f 2a c4       	cvtsi2sd xmm0,r12d
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 25.07
  • Reference cycles: 21.04
  • UOPS_RETIRED.ALL: 13.0

Experiment 42

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVZ R12, R15; CVTSI2SD XMM0, R12
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 44 e7          	cmove  r12,r15
     a:	f2 49 0f 2a c4       	cvtsi2sd xmm0,r12
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 25.0
  • Reference cycles: 21.21
  • UOPS_RETIRED.ALL: 13.0

Experiment 43

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVZ R12, R15; CVTSI2SS XMM0, R12D
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 44 e7          	cmove  r12,r15
     a:	f3 41 0f 2a c4       	cvtsi2ss xmm0,r12d
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 25.0
  • Reference cycles: 20.71
  • UOPS_RETIRED.ALL: 13.0

Experiment 44

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVZ R12, R15; CVTSI2SS XMM0, R12
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 44 e7          	cmove  r12,r15
     a:	f3 49 0f 2a c4       	cvtsi2ss xmm0,r12
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 25.0
  • Reference cycles: 21.04
  • UOPS_RETIRED.ALL: 13.0

Experiment 45

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVZ R12, R15; MOVD XMM0, R12D
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 44 e7          	cmove  r12,r15
     a:	66 41 0f 6e c4       	movd   xmm0,r12d
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 21.09
  • Reference cycles: 17.51
  • UOPS_RETIRED.ALL: 13.0

Experiment 46

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVZ R12, R15; MOVQ XMM0, R12
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 44 e7          	cmove  r12,r15
     a:	66 49 0f 6e c4       	movq   xmm0,r12
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 20.91
  • Reference cycles: 17.34
  • UOPS_RETIRED.ALL: 13.0

Experiment 47

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVZ R12, R15; PINSRB XMM0, R12D, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 44 e7          	cmove  r12,r15
     a:	66 41 0f 3a 20 c4 02 	pinsrb xmm0,r12d,0x2
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 21.07
  • Reference cycles: 17.68
  • UOPS_RETIRED.ALL: 13.0

Experiment 48

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVZ R12, R15; PINSRD XMM0, R12D, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 44 e7          	cmove  r12,r15
     a:	66 41 0f 3a 22 c4 02 	pinsrd xmm0,r12d,0x2
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 21.08
  • Reference cycles: 17.68
  • UOPS_RETIRED.ALL: 13.0

Experiment 49

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVZ R12, R15; PINSRQ XMM0, R12, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 44 e7          	cmove  r12,r15
     a:	66 49 0f 3a 22 c4 02 	pinsrq xmm0,r12,0x2
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 20.92
  • Reference cycles: 17.17
  • UOPS_RETIRED.ALL: 13.0

Experiment 50

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVZ R12, R15; PINSRW XMM0, R12D, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 44 e7          	cmove  r12,r15
     a:	66 41 0f c4 c4 02    	pinsrw xmm0,r12d,0x2
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 21.08
  • Reference cycles: 17.68
  • UOPS_RETIRED.ALL: 13.0

Experiment 51 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVC R12, R15; CVTSI2SD XMM0, R12D
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 42 e7          	cmovb  r12,r15
     a:	f2 41 0f 2a c4       	cvtsi2sd xmm0,r12d
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 25.07
  • Reference cycles: 20.87
  • UOPS_RETIRED.ALL: 13.0

Experiment 52 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVC R12, R15; CVTSI2SD XMM0, R12
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 42 e7          	cmovb  r12,r15
     a:	f2 49 0f 2a c4       	cvtsi2sd xmm0,r12
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 25.05
  • Reference cycles: 21.21
  • UOPS_RETIRED.ALL: 13.0

Experiment 53 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVC R12, R15; CVTSI2SS XMM0, R12D
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 42 e7          	cmovb  r12,r15
     a:	f3 41 0f 2a c4       	cvtsi2ss xmm0,r12d
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 24.99
  • Reference cycles: 21.38
  • UOPS_RETIRED.ALL: 13.0

Experiment 54 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVC R12, R15; CVTSI2SS XMM0, R12
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 42 e7          	cmovb  r12,r15
     a:	f3 49 0f 2a c4       	cvtsi2ss xmm0,r12
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 25.06
  • Reference cycles: 21.38
  • UOPS_RETIRED.ALL: 13.0

Experiment 55 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVC R12, R15; MOVD XMM0, R12D
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 42 e7          	cmovb  r12,r15
     a:	66 41 0f 6e c4       	movd   xmm0,r12d
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 20.99
  • Reference cycles: 18.18
  • UOPS_RETIRED.ALL: 13.0

Experiment 56 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVC R12, R15; MOVQ XMM0, R12
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 42 e7          	cmovb  r12,r15
     a:	66 49 0f 6e c4       	movq   xmm0,r12
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 21.0
  • Reference cycles: 18.18
  • UOPS_RETIRED.ALL: 13.0

Experiment 57 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVC R12, R15; PINSRB XMM0, R12D, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 42 e7          	cmovb  r12,r15
     a:	66 41 0f 3a 20 c4 02 	pinsrb xmm0,r12d,0x2
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 20.97
  • Reference cycles: 17.51
  • UOPS_RETIRED.ALL: 13.0

Experiment 58 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVC R12, R15; PINSRD XMM0, R12D, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 42 e7          	cmovb  r12,r15
     a:	66 41 0f 3a 22 c4 02 	pinsrd xmm0,r12d,0x2
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 21.1
  • Reference cycles: 17.68
  • UOPS_RETIRED.ALL: 13.0

Experiment 59 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVC R12, R15; PINSRQ XMM0, R12, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 42 e7          	cmovb  r12,r15
     a:	66 49 0f 3a 22 c4 02 	pinsrq xmm0,r12,0x2
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 20.98
  • Reference cycles: 17.17
  • UOPS_RETIRED.ALL: 13.0

Experiment 60 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVC R12, R15; PINSRW XMM0, R12D, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 42 e7          	cmovb  r12,r15
     a:	66 41 0f c4 c4 02    	pinsrw xmm0,r12d,0x2
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 20.99
  • Reference cycles: 17.17
  • UOPS_RETIRED.ALL: 13.0

Experiment 61 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVP R12, R15; CVTSI2SD XMM0, R12D
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 4a e7          	cmovp  r12,r15
     a:	f2 41 0f 2a c4       	cvtsi2sd xmm0,r12d
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 25.0
  • Reference cycles: 21.04
  • UOPS_RETIRED.ALL: 13.0

Experiment 62 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVP R12, R15; CVTSI2SD XMM0, R12
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 4a e7          	cmovp  r12,r15
     a:	f2 49 0f 2a c4       	cvtsi2sd xmm0,r12
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 25.01
  • Reference cycles: 20.87
  • UOPS_RETIRED.ALL: 13.0

Experiment 63 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVP R12, R15; CVTSI2SS XMM0, R12D
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 4a e7          	cmovp  r12,r15
     a:	f3 41 0f 2a c4       	cvtsi2ss xmm0,r12d
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 25.07
  • Reference cycles: 21.04
  • UOPS_RETIRED.ALL: 13.0

Experiment 64 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVP R12, R15; CVTSI2SS XMM0, R12
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 4a e7          	cmovp  r12,r15
     a:	f3 49 0f 2a c4       	cvtsi2ss xmm0,r12
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 24.99
  • Reference cycles: 20.71
  • UOPS_RETIRED.ALL: 13.0

Experiment 65 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVP R12, R15; MOVD XMM0, R12D
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 4a e7          	cmovp  r12,r15
     a:	66 41 0f 6e c4       	movd   xmm0,r12d
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 20.98
  • Reference cycles: 17.17
  • UOPS_RETIRED.ALL: 13.0

Experiment 66 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVP R12, R15; MOVQ XMM0, R12
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 4a e7          	cmovp  r12,r15
     a:	66 49 0f 6e c4       	movq   xmm0,r12
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 20.98
  • Reference cycles: 17.68
  • UOPS_RETIRED.ALL: 13.0

Experiment 67 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVP R12, R15; PINSRB XMM0, R12D, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 4a e7          	cmovp  r12,r15
     a:	66 41 0f 3a 20 c4 02 	pinsrb xmm0,r12d,0x2
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 21.08
  • Reference cycles: 17.84
  • UOPS_RETIRED.ALL: 13.0

Experiment 68 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVP R12, R15; PINSRD XMM0, R12D, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 4a e7          	cmovp  r12,r15
     a:	66 41 0f 3a 22 c4 02 	pinsrd xmm0,r12d,0x2
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 21.09
  • Reference cycles: 17.68
  • UOPS_RETIRED.ALL: 13.0

Experiment 69 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVP R12, R15; PINSRQ XMM0, R12, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 4a e7          	cmovp  r12,r15
     a:	66 49 0f 3a 22 c4 02 	pinsrq xmm0,r12,0x2
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 20.99
  • Reference cycles: 17.68
  • UOPS_RETIRED.ALL: 13.0

Experiment 70 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVP R12, R15; PINSRW XMM0, R12D, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 4a e7          	cmovp  r12,r15
     a:	66 41 0f c4 c4 02    	pinsrw xmm0,r12d,0x2
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 21.07
  • Reference cycles: 18.01
  • UOPS_RETIRED.ALL: 13.0

Experiment 71 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVS R12, R15; CVTSI2SD XMM0, R12D
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 48 e7          	cmovs  r12,r15
     a:	f2 41 0f 2a c4       	cvtsi2sd xmm0,r12d
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 24.99
  • Reference cycles: 20.87
  • UOPS_RETIRED.ALL: 13.0

Experiment 72 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVS R12, R15; CVTSI2SD XMM0, R12
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 48 e7          	cmovs  r12,r15
     a:	f2 49 0f 2a c4       	cvtsi2sd xmm0,r12
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 24.99
  • Reference cycles: 20.87
  • UOPS_RETIRED.ALL: 13.0

Experiment 73 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVS R12, R15; CVTSI2SS XMM0, R12D
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 48 e7          	cmovs  r12,r15
     a:	f3 41 0f 2a c4       	cvtsi2ss xmm0,r12d
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 25.09
  • Reference cycles: 21.38
  • UOPS_RETIRED.ALL: 13.0

Experiment 74 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVS R12, R15; CVTSI2SS XMM0, R12
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 48 e7          	cmovs  r12,r15
     a:	f3 49 0f 2a c4       	cvtsi2ss xmm0,r12
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 24.99
  • Reference cycles: 20.87
  • UOPS_RETIRED.ALL: 13.0

Experiment 75 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVS R12, R15; MOVD XMM0, R12D
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 48 e7          	cmovs  r12,r15
     a:	66 41 0f 6e c4       	movd   xmm0,r12d
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 20.98
  • Reference cycles: 17.0
  • UOPS_RETIRED.ALL: 13.0

Experiment 76 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVS R12, R15; MOVQ XMM0, R12
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 48 e7          	cmovs  r12,r15
     a:	66 49 0f 6e c4       	movq   xmm0,r12
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 21.08
  • Reference cycles: 17.68
  • UOPS_RETIRED.ALL: 13.0

Experiment 77 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVS R12, R15; PINSRB XMM0, R12D, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 48 e7          	cmovs  r12,r15
     a:	66 41 0f 3a 20 c4 02 	pinsrb xmm0,r12d,0x2
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 20.98
  • Reference cycles: 18.18
  • UOPS_RETIRED.ALL: 13.0

Experiment 78 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVS R12, R15; PINSRD XMM0, R12D, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 48 e7          	cmovs  r12,r15
     a:	66 41 0f 3a 22 c4 02 	pinsrd xmm0,r12d,0x2
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 21.08
  • Reference cycles: 17.68
  • UOPS_RETIRED.ALL: 13.0

Experiment 79 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVS R12, R15; PINSRQ XMM0, R12, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 48 e7          	cmovs  r12,r15
     a:	66 49 0f 3a 22 c4 02 	pinsrq xmm0,r12,0x2
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 20.99
  • Reference cycles: 18.18
  • UOPS_RETIRED.ALL: 13.0

Experiment 80 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVS R12, R15; PINSRW XMM0, R12D, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 48 e7          	cmovs  r12,r15
     a:	66 41 0f c4 c4 02    	pinsrw xmm0,r12d,0x2
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 20.99
  • Reference cycles: 17.34
  • UOPS_RETIRED.ALL: 13.0

Experiment 81 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVO R12, R15; CVTSI2SD XMM0, R12D
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 40 e7          	cmovo  r12,r15
     a:	f2 41 0f 2a c4       	cvtsi2sd xmm0,r12d
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 24.97
  • Reference cycles: 20.87
  • UOPS_RETIRED.ALL: 13.0

Experiment 82 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVO R12, R15; CVTSI2SD XMM0, R12
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 40 e7          	cmovo  r12,r15
     a:	f2 49 0f 2a c4       	cvtsi2sd xmm0,r12
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 24.97
  • Reference cycles: 21.04
  • UOPS_RETIRED.ALL: 13.0

Experiment 83 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVO R12, R15; CVTSI2SS XMM0, R12D
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 40 e7          	cmovo  r12,r15
     a:	f3 41 0f 2a c4       	cvtsi2ss xmm0,r12d
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 25.07
  • Reference cycles: 20.87
  • UOPS_RETIRED.ALL: 13.0

Experiment 84 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVO R12, R15; CVTSI2SS XMM0, R12
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 40 e7          	cmovo  r12,r15
     a:	f3 49 0f 2a c4       	cvtsi2ss xmm0,r12
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 25.08
  • Reference cycles: 21.38
  • UOPS_RETIRED.ALL: 13.0

Experiment 85 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVO R12, R15; MOVD XMM0, R12D
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 40 e7          	cmovo  r12,r15
     a:	66 41 0f 6e c4       	movd   xmm0,r12d
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 21.1
  • Reference cycles: 17.51
  • UOPS_RETIRED.ALL: 13.0

Experiment 86 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVO R12, R15; MOVQ XMM0, R12
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 40 e7          	cmovo  r12,r15
     a:	66 49 0f 6e c4       	movq   xmm0,r12
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 21.06
  • Reference cycles: 17.51
  • UOPS_RETIRED.ALL: 13.0

Experiment 87 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVO R12, R15; PINSRB XMM0, R12D, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 40 e7          	cmovo  r12,r15
     a:	66 41 0f 3a 20 c4 02 	pinsrb xmm0,r12d,0x2
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 21.08
  • Reference cycles: 18.01
  • UOPS_RETIRED.ALL: 13.0

Experiment 88 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVO R12, R15; PINSRD XMM0, R12D, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 40 e7          	cmovo  r12,r15
     a:	66 41 0f 3a 22 c4 02 	pinsrd xmm0,r12d,0x2
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 21.06
  • Reference cycles: 17.84
  • UOPS_RETIRED.ALL: 13.0

Experiment 89 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVO R12, R15; PINSRQ XMM0, R12, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 40 e7          	cmovo  r12,r15
     a:	66 49 0f 3a 22 c4 02 	pinsrq xmm0,r12,0x2
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 20.98
  • Reference cycles: 18.18
  • UOPS_RETIRED.ALL: 13.0

Experiment 90 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVO R12, R15; PINSRW XMM0, R12D, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 40 e7          	cmovo  r12,r15
     a:	66 41 0f c4 c4 02    	pinsrw xmm0,r12d,0x2
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 21.08
  • Reference cycles: 17.34
  • UOPS_RETIRED.ALL: 13.0

Experiment 91 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVZ R12, R15; CVTSI2SD XMM0, R12D
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 44 e7          	cmove  r12,r15
     a:	f2 41 0f 2a c4       	cvtsi2sd xmm0,r12d
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 25.07
  • Reference cycles: 21.38
  • UOPS_RETIRED.ALL: 13.0

Experiment 92 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVZ R12, R15; CVTSI2SD XMM0, R12
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 44 e7          	cmove  r12,r15
     a:	f2 49 0f 2a c4       	cvtsi2sd xmm0,r12
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 25.06
  • Reference cycles: 21.21
  • UOPS_RETIRED.ALL: 13.0

Experiment 93 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVZ R12, R15; CVTSI2SS XMM0, R12D
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 44 e7          	cmove  r12,r15
     a:	f3 41 0f 2a c4       	cvtsi2ss xmm0,r12d
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 25.07
  • Reference cycles: 21.21
  • UOPS_RETIRED.ALL: 13.0

Experiment 94 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVZ R12, R15; CVTSI2SS XMM0, R12
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 44 e7          	cmove  r12,r15
     a:	f3 49 0f 2a c4       	cvtsi2ss xmm0,r12
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 24.98
  • Reference cycles: 21.21
  • UOPS_RETIRED.ALL: 13.0

Experiment 95 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVZ R12, R15; MOVD XMM0, R12D
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 44 e7          	cmove  r12,r15
     a:	66 41 0f 6e c4       	movd   xmm0,r12d
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 21.0
  • Reference cycles: 17.84
  • UOPS_RETIRED.ALL: 13.0

Experiment 96 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVZ R12, R15; MOVQ XMM0, R12
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 44 e7          	cmove  r12,r15
     a:	66 49 0f 6e c4       	movq   xmm0,r12
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 20.99
  • Reference cycles: 18.18
  • UOPS_RETIRED.ALL: 13.0

Experiment 97 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVZ R12, R15; PINSRB XMM0, R12D, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 44 e7          	cmove  r12,r15
     a:	66 41 0f 3a 20 c4 02 	pinsrb xmm0,r12d,0x2
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 20.99
  • Reference cycles: 17.68
  • UOPS_RETIRED.ALL: 13.0

Experiment 98 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVZ R12, R15; PINSRD XMM0, R12D, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 44 e7          	cmove  r12,r15
     a:	66 41 0f 3a 22 c4 02 	pinsrd xmm0,r12d,0x2
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 21.08
  • Reference cycles: 17.51
  • UOPS_RETIRED.ALL: 13.0

Experiment 99 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVZ R12, R15; PINSRQ XMM0, R12, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 44 e7          	cmove  r12,r15
     a:	66 49 0f 3a 22 c4 02 	pinsrq xmm0,r12,0x2
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 21.09
  • Reference cycles: 17.34
  • UOPS_RETIRED.ALL: 13.0

Experiment 100 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM0, XMM1, 2
• Chain instructions: CMOVZ R12, R15; PINSRW XMM0, R12D, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c1 02    	pcmpestrm xmm0,xmm1,0x2
     6:	4d 0f 44 e7          	cmove  r12,r15
     a:	66 41 0f c4 c4 02    	pinsrw xmm0,r12d,0x2
  

• Init:

  VZEROALL;
  PXOR XMM0, XMM0;
  PXOR XMM1, XMM1;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 21.08
  • Reference cycles: 17.51
  • UOPS_RETIRED.ALL: 13.0

Experiment 1

• Instruction: PCMPESTRM XMM2, XMM1, 2
• Chain instructions: SHUFPD XMM1, XMM0, 0;SHUFPD XMM1, XMM1, 0;SHUFPD XMM1, XMM1, 0;SHUFPD XMM1, XMM1, 0;SHUFPD XMM1, XMM1, 0;SHUFPD XMM1, XMM1, 0;SHUFPD XMM1, XMM1, 0;SHUFPD XMM1, XMM1, 0;SHUFPD XMM1, XMM1, 0;SHUFPD XMM1, XMM1, 0;SHUFPD XMM1, XMM1, 0;SHUFPD XMM1, XMM1, 0;SHUFPD XMM1, XMM1, 0;SHUFPD XMM1, XMM1, 0;SHUFPD XMM1, XMM1, 0
• Chain latency: 15
• Code:

     0:	66 0f 3a 60 d1 02    	pcmpestrm xmm2,xmm1,0x2
     6:	66 0f c6 c8 00       	shufpd xmm1,xmm0,0x0
     b:	66 0f c6 c9 00       	shufpd xmm1,xmm1,0x0
    10:	66 0f c6 c9 00       	shufpd xmm1,xmm1,0x0
    15:	66 0f c6 c9 00       	shufpd xmm1,xmm1,0x0
    1a:	66 0f c6 c9 00       	shufpd xmm1,xmm1,0x0
    1f:	66 0f c6 c9 00       	shufpd xmm1,xmm1,0x0
    24:	66 0f c6 c9 00       	shufpd xmm1,xmm1,0x0
    29:	66 0f c6 c9 00       	shufpd xmm1,xmm1,0x0
    2e:	66 0f c6 c9 00       	shufpd xmm1,xmm1,0x0
    33:	66 0f c6 c9 00       	shufpd xmm1,xmm1,0x0
    38:	66 0f c6 c9 00       	shufpd xmm1,xmm1,0x0
    3d:	66 0f c6 c9 00       	shufpd xmm1,xmm1,0x0
    42:	66 0f c6 c9 00       	shufpd xmm1,xmm1,0x0
    47:	66 0f c6 c9 00       	shufpd xmm1,xmm1,0x0
    4c:	66 0f c6 c9 00       	shufpd xmm1,xmm1,0x0
  

• Init:

  VZEROALL;
  PXOR XMM2, XMM2;
  PXOR XMM1, XMM1

• Show nanoBench command
• Results:
  • Instructions retired: 16.0
  • Core cycles: 23.49
  • Reference cycles: 19.36
  • UOPS_RETIRED.ALL: 26.0

Experiment 2

• Instruction: PCMPESTRM XMM2, XMM1, 2
• Chain instructions: PSHUFD XMM1, XMM0, 0;PAND XMM1, XMM1;PAND XMM1, XMM1;PAND XMM1, XMM1;PAND XMM1, XMM1;PAND XMM1, XMM1;PAND XMM1, XMM1;PAND XMM1, XMM1;PAND XMM1, XMM1;PAND XMM1, XMM1;PAND XMM1, XMM1;PAND XMM1, XMM1;PAND XMM1, XMM1;PAND XMM1, XMM1;PAND XMM1, XMM1
• Chain latency: 15
• Code:

     0:	66 0f 3a 60 d1 02    	pcmpestrm xmm2,xmm1,0x2
     6:	66 0f 70 c8 00       	pshufd xmm1,xmm0,0x0
     b:	66 0f db c9          	pand   xmm1,xmm1
     f:	66 0f db c9          	pand   xmm1,xmm1
    13:	66 0f db c9          	pand   xmm1,xmm1
    17:	66 0f db c9          	pand   xmm1,xmm1
    1b:	66 0f db c9          	pand   xmm1,xmm1
    1f:	66 0f db c9          	pand   xmm1,xmm1
    23:	66 0f db c9          	pand   xmm1,xmm1
    27:	66 0f db c9          	pand   xmm1,xmm1
    2b:	66 0f db c9          	pand   xmm1,xmm1
    2f:	66 0f db c9          	pand   xmm1,xmm1
    33:	66 0f db c9          	pand   xmm1,xmm1
    37:	66 0f db c9          	pand   xmm1,xmm1
    3b:	66 0f db c9          	pand   xmm1,xmm1
    3f:	66 0f db c9          	pand   xmm1,xmm1
  

• Init:

  VZEROALL;
  PXOR XMM2, XMM2;
  PXOR XMM1, XMM1

• Show nanoBench command
• Results:
  • Instructions retired: 16.0
  • Core cycles: 23.44
  • Reference cycles: 19.86
  • UOPS_RETIRED.ALL: 26.0

Experiment 3 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM2, XMM1, 2
• Chain instructions: SHUFPD XMM1, XMM0, 0;SHUFPD XMM1, XMM1, 0;SHUFPD XMM1, XMM1, 0;SHUFPD XMM1, XMM1, 0;SHUFPD XMM1, XMM1, 0;SHUFPD XMM1, XMM1, 0;SHUFPD XMM1, XMM1, 0;SHUFPD XMM1, XMM1, 0;SHUFPD XMM1, XMM1, 0;SHUFPD XMM1, XMM1, 0;SHUFPD XMM1, XMM1, 0;SHUFPD XMM1, XMM1, 0;SHUFPD XMM1, XMM1, 0;SHUFPD XMM1, XMM1, 0;SHUFPD XMM1, XMM1, 0
• Chain latency: 15
• Code:

     0:	66 0f 3a 60 d1 02    	pcmpestrm xmm2,xmm1,0x2
     6:	66 0f c6 c8 00       	shufpd xmm1,xmm0,0x0
     b:	66 0f c6 c9 00       	shufpd xmm1,xmm1,0x0
    10:	66 0f c6 c9 00       	shufpd xmm1,xmm1,0x0
    15:	66 0f c6 c9 00       	shufpd xmm1,xmm1,0x0
    1a:	66 0f c6 c9 00       	shufpd xmm1,xmm1,0x0
    1f:	66 0f c6 c9 00       	shufpd xmm1,xmm1,0x0
    24:	66 0f c6 c9 00       	shufpd xmm1,xmm1,0x0
    29:	66 0f c6 c9 00       	shufpd xmm1,xmm1,0x0
    2e:	66 0f c6 c9 00       	shufpd xmm1,xmm1,0x0
    33:	66 0f c6 c9 00       	shufpd xmm1,xmm1,0x0
    38:	66 0f c6 c9 00       	shufpd xmm1,xmm1,0x0
    3d:	66 0f c6 c9 00       	shufpd xmm1,xmm1,0x0
    42:	66 0f c6 c9 00       	shufpd xmm1,xmm1,0x0
    47:	66 0f c6 c9 00       	shufpd xmm1,xmm1,0x0
    4c:	66 0f c6 c9 00       	shufpd xmm1,xmm1,0x0
  

• Init:

  VZEROALL;
  PXOR XMM2, XMM2;
  PXOR XMM1, XMM1;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 16.0
  • Core cycles: 23.52
  • Reference cycles: 19.53
  • UOPS_RETIRED.ALL: 26.0

Experiment 4 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM2, XMM1, 2
• Chain instructions: PSHUFD XMM1, XMM0, 0;PAND XMM1, XMM1;PAND XMM1, XMM1;PAND XMM1, XMM1;PAND XMM1, XMM1;PAND XMM1, XMM1;PAND XMM1, XMM1;PAND XMM1, XMM1;PAND XMM1, XMM1;PAND XMM1, XMM1;PAND XMM1, XMM1;PAND XMM1, XMM1;PAND XMM1, XMM1;PAND XMM1, XMM1;PAND XMM1, XMM1
• Chain latency: 15
• Code:

     0:	66 0f 3a 60 d1 02    	pcmpestrm xmm2,xmm1,0x2
     6:	66 0f 70 c8 00       	pshufd xmm1,xmm0,0x0
     b:	66 0f db c9          	pand   xmm1,xmm1
     f:	66 0f db c9          	pand   xmm1,xmm1
    13:	66 0f db c9          	pand   xmm1,xmm1
    17:	66 0f db c9          	pand   xmm1,xmm1
    1b:	66 0f db c9          	pand   xmm1,xmm1
    1f:	66 0f db c9          	pand   xmm1,xmm1
    23:	66 0f db c9          	pand   xmm1,xmm1
    27:	66 0f db c9          	pand   xmm1,xmm1
    2b:	66 0f db c9          	pand   xmm1,xmm1
    2f:	66 0f db c9          	pand   xmm1,xmm1
    33:	66 0f db c9          	pand   xmm1,xmm1
    37:	66 0f db c9          	pand   xmm1,xmm1
    3b:	66 0f db c9          	pand   xmm1,xmm1
    3f:	66 0f db c9          	pand   xmm1,xmm1
  

• Init:

  VZEROALL;
  PXOR XMM2, XMM2;
  PXOR XMM1, XMM1;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 16.0
  • Core cycles: 23.51
  • Reference cycles: 19.86
  • UOPS_RETIRED.ALL: 26.0

Experiment 1

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVC R12, R15; CVTSI2SD XMM0, R12D
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 42 e7          	cmovb  r12,r15
     a:	f2 41 0f 2a c4       	cvtsi2sd xmm0,r12d
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 27.06
  • Reference cycles: 22.73
  • UOPS_RETIRED.ALL: 13.0

Experiment 2

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVC R12, R15; CVTSI2SD XMM0, R12
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 42 e7          	cmovb  r12,r15
     a:	f2 49 0f 2a c4       	cvtsi2sd xmm0,r12
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 26.98
  • Reference cycles: 22.73
  • UOPS_RETIRED.ALL: 13.0

Experiment 3

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVC R12, R15; CVTSI2SS XMM0, R12D
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 42 e7          	cmovb  r12,r15
     a:	f3 41 0f 2a c4       	cvtsi2ss xmm0,r12d
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 26.91
  • Reference cycles: 22.39
  • UOPS_RETIRED.ALL: 13.0

Experiment 4

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVC R12, R15; CVTSI2SS XMM0, R12
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 42 e7          	cmovb  r12,r15
     a:	f3 49 0f 2a c4       	cvtsi2ss xmm0,r12
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 27.06
  • Reference cycles: 22.56
  • UOPS_RETIRED.ALL: 13.0

Experiment 5

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVC R12, R15; MOVD XMM0, R12D
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 42 e7          	cmovb  r12,r15
     a:	66 41 0f 6e c4       	movd   xmm0,r12d
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 23.06
  • Reference cycles: 19.7
  • UOPS_RETIRED.ALL: 13.0

Experiment 6

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVC R12, R15; MOVQ XMM0, R12
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 42 e7          	cmovb  r12,r15
     a:	66 49 0f 6e c4       	movq   xmm0,r12
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 23.0
  • Reference cycles: 19.19
  • UOPS_RETIRED.ALL: 13.0

Experiment 7

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVC R12, R15; PINSRB XMM0, R12D, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 42 e7          	cmovb  r12,r15
     a:	66 41 0f 3a 20 c4 02 	pinsrb xmm0,r12d,0x2
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 23.06
  • Reference cycles: 19.53
  • UOPS_RETIRED.ALL: 13.0

Experiment 8

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVC R12, R15; PINSRD XMM0, R12D, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 42 e7          	cmovb  r12,r15
     a:	66 41 0f 3a 22 c4 02 	pinsrd xmm0,r12d,0x2
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 23.06
  • Reference cycles: 19.53
  • UOPS_RETIRED.ALL: 13.0

Experiment 9

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVC R12, R15; PINSRQ XMM0, R12, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 42 e7          	cmovb  r12,r15
     a:	66 49 0f 3a 22 c4 02 	pinsrq xmm0,r12,0x2
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 23.07
  • Reference cycles: 19.7
  • UOPS_RETIRED.ALL: 13.0

Experiment 10

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVC R12, R15; PINSRW XMM0, R12D, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 42 e7          	cmovb  r12,r15
     a:	66 41 0f c4 c4 02    	pinsrw xmm0,r12d,0x2
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 22.92
  • Reference cycles: 19.19
  • UOPS_RETIRED.ALL: 13.0

Experiment 11

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVP R12, R15; CVTSI2SD XMM0, R12D
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 4a e7          	cmovp  r12,r15
     a:	f2 41 0f 2a c4       	cvtsi2sd xmm0,r12d
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 26.97
  • Reference cycles: 22.39
  • UOPS_RETIRED.ALL: 13.0

Experiment 12

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVP R12, R15; CVTSI2SD XMM0, R12
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 4a e7          	cmovp  r12,r15
     a:	f2 49 0f 2a c4       	cvtsi2sd xmm0,r12
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 26.98
  • Reference cycles: 22.73
  • UOPS_RETIRED.ALL: 13.0

Experiment 13

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVP R12, R15; CVTSI2SS XMM0, R12D
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 4a e7          	cmovp  r12,r15
     a:	f3 41 0f 2a c4       	cvtsi2ss xmm0,r12d
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 26.96
  • Reference cycles: 22.39
  • UOPS_RETIRED.ALL: 13.0

Experiment 14

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVP R12, R15; CVTSI2SS XMM0, R12
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 4a e7          	cmovp  r12,r15
     a:	f3 49 0f 2a c4       	cvtsi2ss xmm0,r12
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 27.06
  • Reference cycles: 22.56
  • UOPS_RETIRED.ALL: 13.0

Experiment 15

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVP R12, R15; MOVD XMM0, R12D
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 4a e7          	cmovp  r12,r15
     a:	66 41 0f 6e c4       	movd   xmm0,r12d
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 22.98
  • Reference cycles: 19.36
  • UOPS_RETIRED.ALL: 13.0

Experiment 16

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVP R12, R15; MOVQ XMM0, R12
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 4a e7          	cmovp  r12,r15
     a:	66 49 0f 6e c4       	movq   xmm0,r12
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 23.06
  • Reference cycles: 19.53
  • UOPS_RETIRED.ALL: 13.0

Experiment 17

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVP R12, R15; PINSRB XMM0, R12D, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 4a e7          	cmovp  r12,r15
     a:	66 41 0f 3a 20 c4 02 	pinsrb xmm0,r12d,0x2
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 22.98
  • Reference cycles: 19.36
  • UOPS_RETIRED.ALL: 13.0

Experiment 18

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVP R12, R15; PINSRD XMM0, R12D, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 4a e7          	cmovp  r12,r15
     a:	66 41 0f 3a 22 c4 02 	pinsrd xmm0,r12d,0x2
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 23.06
  • Reference cycles: 19.7
  • UOPS_RETIRED.ALL: 13.0

Experiment 19

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVP R12, R15; PINSRQ XMM0, R12, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 4a e7          	cmovp  r12,r15
     a:	66 49 0f 3a 22 c4 02 	pinsrq xmm0,r12,0x2
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 23.06
  • Reference cycles: 19.7
  • UOPS_RETIRED.ALL: 13.0

Experiment 20

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVP R12, R15; PINSRW XMM0, R12D, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 4a e7          	cmovp  r12,r15
     a:	66 41 0f c4 c4 02    	pinsrw xmm0,r12d,0x2
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 23.06
  • Reference cycles: 19.86
  • UOPS_RETIRED.ALL: 13.0

Experiment 21

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVS R12, R15; CVTSI2SD XMM0, R12D
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 48 e7          	cmovs  r12,r15
     a:	f2 41 0f 2a c4       	cvtsi2sd xmm0,r12d
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 26.97
  • Reference cycles: 22.39
  • UOPS_RETIRED.ALL: 13.0

Experiment 22

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVS R12, R15; CVTSI2SD XMM0, R12
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 48 e7          	cmovs  r12,r15
     a:	f2 49 0f 2a c4       	cvtsi2sd xmm0,r12
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 26.98
  • Reference cycles: 22.39
  • UOPS_RETIRED.ALL: 13.0

Experiment 23

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVS R12, R15; CVTSI2SS XMM0, R12D
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 48 e7          	cmovs  r12,r15
     a:	f3 41 0f 2a c4       	cvtsi2ss xmm0,r12d
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 27.05
  • Reference cycles: 22.56
  • UOPS_RETIRED.ALL: 13.0

Experiment 24

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVS R12, R15; CVTSI2SS XMM0, R12
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 48 e7          	cmovs  r12,r15
     a:	f3 49 0f 2a c4       	cvtsi2ss xmm0,r12
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 26.91
  • Reference cycles: 22.89
  • UOPS_RETIRED.ALL: 13.0

Experiment 25

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVS R12, R15; MOVD XMM0, R12D
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 48 e7          	cmovs  r12,r15
     a:	66 41 0f 6e c4       	movd   xmm0,r12d
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 23.0
  • Reference cycles: 19.19
  • UOPS_RETIRED.ALL: 13.0

Experiment 26

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVS R12, R15; MOVQ XMM0, R12
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 48 e7          	cmovs  r12,r15
     a:	66 49 0f 6e c4       	movq   xmm0,r12
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 23.06
  • Reference cycles: 19.86
  • UOPS_RETIRED.ALL: 13.0

Experiment 27

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVS R12, R15; PINSRB XMM0, R12D, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 48 e7          	cmovs  r12,r15
     a:	66 41 0f 3a 20 c4 02 	pinsrb xmm0,r12d,0x2
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 23.06
  • Reference cycles: 19.7
  • UOPS_RETIRED.ALL: 13.0

Experiment 28

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVS R12, R15; PINSRD XMM0, R12D, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 48 e7          	cmovs  r12,r15
     a:	66 41 0f 3a 22 c4 02 	pinsrd xmm0,r12d,0x2
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 23.06
  • Reference cycles: 19.53
  • UOPS_RETIRED.ALL: 13.0

Experiment 29

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVS R12, R15; PINSRQ XMM0, R12, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 48 e7          	cmovs  r12,r15
     a:	66 49 0f 3a 22 c4 02 	pinsrq xmm0,r12,0x2
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 23.0
  • Reference cycles: 19.19
  • UOPS_RETIRED.ALL: 13.0

Experiment 30

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVS R12, R15; PINSRW XMM0, R12D, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 48 e7          	cmovs  r12,r15
     a:	66 41 0f c4 c4 02    	pinsrw xmm0,r12d,0x2
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 22.91
  • Reference cycles: 19.19
  • UOPS_RETIRED.ALL: 13.0

Experiment 31

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVO R12, R15; CVTSI2SD XMM0, R12D
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 40 e7          	cmovo  r12,r15
     a:	f2 41 0f 2a c4       	cvtsi2sd xmm0,r12d
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 27.0
  • Reference cycles: 23.06
  • UOPS_RETIRED.ALL: 13.0

Experiment 32

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVO R12, R15; CVTSI2SD XMM0, R12
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 40 e7          	cmovo  r12,r15
     a:	f2 49 0f 2a c4       	cvtsi2sd xmm0,r12
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 27.05
  • Reference cycles: 22.56
  • UOPS_RETIRED.ALL: 13.0

Experiment 33

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVO R12, R15; CVTSI2SS XMM0, R12D
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 40 e7          	cmovo  r12,r15
     a:	f3 41 0f 2a c4       	cvtsi2ss xmm0,r12d
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 27.0
  • Reference cycles: 23.23
  • UOPS_RETIRED.ALL: 13.0

Experiment 34

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVO R12, R15; CVTSI2SS XMM0, R12
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 40 e7          	cmovo  r12,r15
     a:	f3 49 0f 2a c4       	cvtsi2ss xmm0,r12
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 27.05
  • Reference cycles: 22.56
  • UOPS_RETIRED.ALL: 13.0

Experiment 35

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVO R12, R15; MOVD XMM0, R12D
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 40 e7          	cmovo  r12,r15
     a:	66 41 0f 6e c4       	movd   xmm0,r12d
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 23.0
  • Reference cycles: 19.36
  • UOPS_RETIRED.ALL: 13.0

Experiment 36

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVO R12, R15; MOVQ XMM0, R12
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 40 e7          	cmovo  r12,r15
     a:	66 49 0f 6e c4       	movq   xmm0,r12
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 23.0
  • Reference cycles: 19.19
  • UOPS_RETIRED.ALL: 13.0

Experiment 37

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVO R12, R15; PINSRB XMM0, R12D, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 40 e7          	cmovo  r12,r15
     a:	66 41 0f 3a 20 c4 02 	pinsrb xmm0,r12d,0x2
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 22.92
  • Reference cycles: 19.19
  • UOPS_RETIRED.ALL: 13.0

Experiment 38

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVO R12, R15; PINSRD XMM0, R12D, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 40 e7          	cmovo  r12,r15
     a:	66 41 0f 3a 22 c4 02 	pinsrd xmm0,r12d,0x2
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 22.97
  • Reference cycles: 19.36
  • UOPS_RETIRED.ALL: 13.0

Experiment 39

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVO R12, R15; PINSRQ XMM0, R12, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 40 e7          	cmovo  r12,r15
     a:	66 49 0f 3a 22 c4 02 	pinsrq xmm0,r12,0x2
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 23.07
  • Reference cycles: 19.53
  • UOPS_RETIRED.ALL: 13.0

Experiment 40

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVO R12, R15; PINSRW XMM0, R12D, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 40 e7          	cmovo  r12,r15
     a:	66 41 0f c4 c4 02    	pinsrw xmm0,r12d,0x2
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 23.07
  • Reference cycles: 19.36
  • UOPS_RETIRED.ALL: 13.0

Experiment 41

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVZ R12, R15; CVTSI2SD XMM0, R12D
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 44 e7          	cmove  r12,r15
     a:	f2 41 0f 2a c4       	cvtsi2sd xmm0,r12d
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 27.05
  • Reference cycles: 22.56
  • UOPS_RETIRED.ALL: 13.0

Experiment 42

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVZ R12, R15; CVTSI2SD XMM0, R12
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 44 e7          	cmove  r12,r15
     a:	f2 49 0f 2a c4       	cvtsi2sd xmm0,r12
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 27.05
  • Reference cycles: 22.56
  • UOPS_RETIRED.ALL: 13.0

Experiment 43

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVZ R12, R15; CVTSI2SS XMM0, R12D
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 44 e7          	cmove  r12,r15
     a:	f3 41 0f 2a c4       	cvtsi2ss xmm0,r12d
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 27.07
  • Reference cycles: 22.56
  • UOPS_RETIRED.ALL: 13.0

Experiment 44

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVZ R12, R15; CVTSI2SS XMM0, R12
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 44 e7          	cmove  r12,r15
     a:	f3 49 0f 2a c4       	cvtsi2ss xmm0,r12
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 27.06
  • Reference cycles: 22.89
  • UOPS_RETIRED.ALL: 13.0

Experiment 45

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVZ R12, R15; MOVD XMM0, R12D
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 44 e7          	cmove  r12,r15
     a:	66 41 0f 6e c4       	movd   xmm0,r12d
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 22.99
  • Reference cycles: 19.7
  • UOPS_RETIRED.ALL: 13.0

Experiment 46

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVZ R12, R15; MOVQ XMM0, R12
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 44 e7          	cmove  r12,r15
     a:	66 49 0f 6e c4       	movq   xmm0,r12
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 23.01
  • Reference cycles: 19.19
  • UOPS_RETIRED.ALL: 13.0

Experiment 47

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVZ R12, R15; PINSRB XMM0, R12D, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 44 e7          	cmove  r12,r15
     a:	66 41 0f 3a 20 c4 02 	pinsrb xmm0,r12d,0x2
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 23.07
  • Reference cycles: 19.86
  • UOPS_RETIRED.ALL: 13.0

Experiment 48

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVZ R12, R15; PINSRD XMM0, R12D, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 44 e7          	cmove  r12,r15
     a:	66 41 0f 3a 22 c4 02 	pinsrd xmm0,r12d,0x2
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 22.92
  • Reference cycles: 19.19
  • UOPS_RETIRED.ALL: 13.0

Experiment 49

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVZ R12, R15; PINSRQ XMM0, R12, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 44 e7          	cmove  r12,r15
     a:	66 49 0f 3a 22 c4 02 	pinsrq xmm0,r12,0x2
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 23.0
  • Reference cycles: 19.19
  • UOPS_RETIRED.ALL: 13.0

Experiment 50

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVZ R12, R15; PINSRW XMM0, R12D, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 44 e7          	cmove  r12,r15
     a:	66 41 0f c4 c4 02    	pinsrw xmm0,r12d,0x2
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 23.0
  • Reference cycles: 19.19
  • UOPS_RETIRED.ALL: 13.0

Experiment 51 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVC R12, R15; CVTSI2SD XMM0, R12D
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 42 e7          	cmovb  r12,r15
     a:	f2 41 0f 2a c4       	cvtsi2sd xmm0,r12d
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 26.99
  • Reference cycles: 22.89
  • UOPS_RETIRED.ALL: 13.0

Experiment 52 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVC R12, R15; CVTSI2SD XMM0, R12
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 42 e7          	cmovb  r12,r15
     a:	f2 49 0f 2a c4       	cvtsi2sd xmm0,r12
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 27.0
  • Reference cycles: 23.06
  • UOPS_RETIRED.ALL: 13.0

Experiment 53 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVC R12, R15; CVTSI2SS XMM0, R12D
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 42 e7          	cmovb  r12,r15
     a:	f3 41 0f 2a c4       	cvtsi2ss xmm0,r12d
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 27.06
  • Reference cycles: 22.39
  • UOPS_RETIRED.ALL: 13.0

Experiment 54 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVC R12, R15; CVTSI2SS XMM0, R12
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 42 e7          	cmovb  r12,r15
     a:	f3 49 0f 2a c4       	cvtsi2ss xmm0,r12
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 26.98
  • Reference cycles: 22.89
  • UOPS_RETIRED.ALL: 13.0

Experiment 55 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVC R12, R15; MOVD XMM0, R12D
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 42 e7          	cmovb  r12,r15
     a:	66 41 0f 6e c4       	movd   xmm0,r12d
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 22.99
  • Reference cycles: 19.19
  • UOPS_RETIRED.ALL: 13.0

Experiment 56 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVC R12, R15; MOVQ XMM0, R12
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 42 e7          	cmovb  r12,r15
     a:	66 49 0f 6e c4       	movq   xmm0,r12
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 22.99
  • Reference cycles: 19.36
  • UOPS_RETIRED.ALL: 13.0

Experiment 57 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVC R12, R15; PINSRB XMM0, R12D, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 42 e7          	cmovb  r12,r15
     a:	66 41 0f 3a 20 c4 02 	pinsrb xmm0,r12d,0x2
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 22.99
  • Reference cycles: 19.19
  • UOPS_RETIRED.ALL: 13.0

Experiment 58 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVC R12, R15; PINSRD XMM0, R12D, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 42 e7          	cmovb  r12,r15
     a:	66 41 0f 3a 22 c4 02 	pinsrd xmm0,r12d,0x2
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 22.99
  • Reference cycles: 19.19
  • UOPS_RETIRED.ALL: 13.0

Experiment 59 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVC R12, R15; PINSRQ XMM0, R12, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 42 e7          	cmovb  r12,r15
     a:	66 49 0f 3a 22 c4 02 	pinsrq xmm0,r12,0x2
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 23.06
  • Reference cycles: 19.19
  • UOPS_RETIRED.ALL: 13.0

Experiment 60 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVC R12, R15; PINSRW XMM0, R12D, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 42 e7          	cmovb  r12,r15
     a:	66 41 0f c4 c4 02    	pinsrw xmm0,r12d,0x2
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 22.98
  • Reference cycles: 19.19
  • UOPS_RETIRED.ALL: 13.0

Experiment 61 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVP R12, R15; CVTSI2SD XMM0, R12D
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 4a e7          	cmovp  r12,r15
     a:	f2 41 0f 2a c4       	cvtsi2sd xmm0,r12d
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 26.99
  • Reference cycles: 23.23
  • UOPS_RETIRED.ALL: 13.0

Experiment 62 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVP R12, R15; CVTSI2SD XMM0, R12
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 4a e7          	cmovp  r12,r15
     a:	f2 49 0f 2a c4       	cvtsi2sd xmm0,r12
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 26.99
  • Reference cycles: 22.56
  • UOPS_RETIRED.ALL: 13.0

Experiment 63 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVP R12, R15; CVTSI2SS XMM0, R12D
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 4a e7          	cmovp  r12,r15
     a:	f3 41 0f 2a c4       	cvtsi2ss xmm0,r12d
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 26.99
  • Reference cycles: 23.23
  • UOPS_RETIRED.ALL: 13.0

Experiment 64 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVP R12, R15; CVTSI2SS XMM0, R12
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 4a e7          	cmovp  r12,r15
     a:	f3 49 0f 2a c4       	cvtsi2ss xmm0,r12
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 26.99
  • Reference cycles: 23.23
  • UOPS_RETIRED.ALL: 13.0

Experiment 65 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVP R12, R15; MOVD XMM0, R12D
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 4a e7          	cmovp  r12,r15
     a:	66 41 0f 6e c4       	movd   xmm0,r12d
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 23.0
  • Reference cycles: 19.19
  • UOPS_RETIRED.ALL: 13.0

Experiment 66 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVP R12, R15; MOVQ XMM0, R12
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 4a e7          	cmovp  r12,r15
     a:	66 49 0f 6e c4       	movq   xmm0,r12
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 23.06
  • Reference cycles: 19.7
  • UOPS_RETIRED.ALL: 13.0

Experiment 67 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVP R12, R15; PINSRB XMM0, R12D, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 4a e7          	cmovp  r12,r15
     a:	66 41 0f 3a 20 c4 02 	pinsrb xmm0,r12d,0x2
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 23.05
  • Reference cycles: 19.53
  • UOPS_RETIRED.ALL: 13.0

Experiment 68 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVP R12, R15; PINSRD XMM0, R12D, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 4a e7          	cmovp  r12,r15
     a:	66 41 0f 3a 22 c4 02 	pinsrd xmm0,r12d,0x2
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 23.06
  • Reference cycles: 19.53
  • UOPS_RETIRED.ALL: 13.0

Experiment 69 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVP R12, R15; PINSRQ XMM0, R12, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 4a e7          	cmovp  r12,r15
     a:	66 49 0f 3a 22 c4 02 	pinsrq xmm0,r12,0x2
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 22.99
  • Reference cycles: 19.19
  • UOPS_RETIRED.ALL: 13.0

Experiment 70 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVP R12, R15; PINSRW XMM0, R12D, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 4a e7          	cmovp  r12,r15
     a:	66 41 0f c4 c4 02    	pinsrw xmm0,r12d,0x2
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 22.99
  • Reference cycles: 19.19
  • UOPS_RETIRED.ALL: 13.0

Experiment 71 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVS R12, R15; CVTSI2SD XMM0, R12D
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 48 e7          	cmovs  r12,r15
     a:	f2 41 0f 2a c4       	cvtsi2sd xmm0,r12d
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 27.05
  • Reference cycles: 22.39
  • UOPS_RETIRED.ALL: 13.0

Experiment 72 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVS R12, R15; CVTSI2SD XMM0, R12
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 48 e7          	cmovs  r12,r15
     a:	f2 49 0f 2a c4       	cvtsi2sd xmm0,r12
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 26.99
  • Reference cycles: 23.06
  • UOPS_RETIRED.ALL: 13.0

Experiment 73 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVS R12, R15; CVTSI2SS XMM0, R12D
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 48 e7          	cmovs  r12,r15
     a:	f3 41 0f 2a c4       	cvtsi2ss xmm0,r12d
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 26.99
  • Reference cycles: 23.23
  • UOPS_RETIRED.ALL: 13.0

Experiment 74 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVS R12, R15; CVTSI2SS XMM0, R12
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 48 e7          	cmovs  r12,r15
     a:	f3 49 0f 2a c4       	cvtsi2ss xmm0,r12
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 26.99
  • Reference cycles: 23.23
  • UOPS_RETIRED.ALL: 13.0

Experiment 75 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVS R12, R15; MOVD XMM0, R12D
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 48 e7          	cmovs  r12,r15
     a:	66 41 0f 6e c4       	movd   xmm0,r12d
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 22.99
  • Reference cycles: 19.19
  • UOPS_RETIRED.ALL: 13.0

Experiment 76 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVS R12, R15; MOVQ XMM0, R12
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 48 e7          	cmovs  r12,r15
     a:	66 49 0f 6e c4       	movq   xmm0,r12
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 23.0
  • Reference cycles: 19.19
  • UOPS_RETIRED.ALL: 13.0

Experiment 77 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVS R12, R15; PINSRB XMM0, R12D, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 48 e7          	cmovs  r12,r15
     a:	66 41 0f 3a 20 c4 02 	pinsrb xmm0,r12d,0x2
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 23.07
  • Reference cycles: 19.53
  • UOPS_RETIRED.ALL: 13.0

Experiment 78 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVS R12, R15; PINSRD XMM0, R12D, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 48 e7          	cmovs  r12,r15
     a:	66 41 0f 3a 22 c4 02 	pinsrd xmm0,r12d,0x2
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 22.99
  • Reference cycles: 19.19
  • UOPS_RETIRED.ALL: 13.0

Experiment 79 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVS R12, R15; PINSRQ XMM0, R12, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 48 e7          	cmovs  r12,r15
     a:	66 49 0f 3a 22 c4 02 	pinsrq xmm0,r12,0x2
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 22.99
  • Reference cycles: 19.19
  • UOPS_RETIRED.ALL: 13.0

Experiment 80 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVS R12, R15; PINSRW XMM0, R12D, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 48 e7          	cmovs  r12,r15
     a:	66 41 0f c4 c4 02    	pinsrw xmm0,r12d,0x2
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 22.98
  • Reference cycles: 19.19
  • UOPS_RETIRED.ALL: 13.0

Experiment 81 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVO R12, R15; CVTSI2SD XMM0, R12D
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 40 e7          	cmovo  r12,r15
     a:	f2 41 0f 2a c4       	cvtsi2sd xmm0,r12d
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 26.97
  • Reference cycles: 23.06
  • UOPS_RETIRED.ALL: 13.0

Experiment 82 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVO R12, R15; CVTSI2SD XMM0, R12
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 40 e7          	cmovo  r12,r15
     a:	f2 49 0f 2a c4       	cvtsi2sd xmm0,r12
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 26.99
  • Reference cycles: 23.23
  • UOPS_RETIRED.ALL: 13.0

Experiment 83 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVO R12, R15; CVTSI2SS XMM0, R12D
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 40 e7          	cmovo  r12,r15
     a:	f3 41 0f 2a c4       	cvtsi2ss xmm0,r12d
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 27.05
  • Reference cycles: 22.39
  • UOPS_RETIRED.ALL: 13.0

Experiment 84 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVO R12, R15; CVTSI2SS XMM0, R12
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 40 e7          	cmovo  r12,r15
     a:	f3 49 0f 2a c4       	cvtsi2ss xmm0,r12
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 26.99
  • Reference cycles: 22.39
  • UOPS_RETIRED.ALL: 13.0

Experiment 85 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVO R12, R15; MOVD XMM0, R12D
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 40 e7          	cmovo  r12,r15
     a:	66 41 0f 6e c4       	movd   xmm0,r12d
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 23.05
  • Reference cycles: 19.19
  • UOPS_RETIRED.ALL: 13.0

Experiment 86 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVO R12, R15; MOVQ XMM0, R12
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 40 e7          	cmovo  r12,r15
     a:	66 49 0f 6e c4       	movq   xmm0,r12
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 22.99
  • Reference cycles: 19.19
  • UOPS_RETIRED.ALL: 13.0

Experiment 87 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVO R12, R15; PINSRB XMM0, R12D, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 40 e7          	cmovo  r12,r15
     a:	66 41 0f 3a 20 c4 02 	pinsrb xmm0,r12d,0x2
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 22.98
  • Reference cycles: 19.19
  • UOPS_RETIRED.ALL: 13.0

Experiment 88 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVO R12, R15; PINSRD XMM0, R12D, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 40 e7          	cmovo  r12,r15
     a:	66 41 0f 3a 22 c4 02 	pinsrd xmm0,r12d,0x2
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 23.06
  • Reference cycles: 19.7
  • UOPS_RETIRED.ALL: 13.0

Experiment 89 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVO R12, R15; PINSRQ XMM0, R12, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 40 e7          	cmovo  r12,r15
     a:	66 49 0f 3a 22 c4 02 	pinsrq xmm0,r12,0x2
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 23.09
  • Reference cycles: 19.7
  • UOPS_RETIRED.ALL: 13.0

Experiment 90 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVO R12, R15; PINSRW XMM0, R12D, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 40 e7          	cmovo  r12,r15
     a:	66 41 0f c4 c4 02    	pinsrw xmm0,r12d,0x2
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 23.07
  • Reference cycles: 19.53
  • UOPS_RETIRED.ALL: 13.0

Experiment 91 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVZ R12, R15; CVTSI2SD XMM0, R12D
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 44 e7          	cmove  r12,r15
     a:	f2 41 0f 2a c4       	cvtsi2sd xmm0,r12d
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 26.99
  • Reference cycles: 22.05
  • UOPS_RETIRED.ALL: 13.0

Experiment 92 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVZ R12, R15; CVTSI2SD XMM0, R12
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 44 e7          	cmove  r12,r15
     a:	f2 49 0f 2a c4       	cvtsi2sd xmm0,r12
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 26.99
  • Reference cycles: 22.56
  • UOPS_RETIRED.ALL: 13.0

Experiment 93 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVZ R12, R15; CVTSI2SS XMM0, R12D
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 44 e7          	cmove  r12,r15
     a:	f3 41 0f 2a c4       	cvtsi2ss xmm0,r12d
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 26.99
  • Reference cycles: 22.73
  • UOPS_RETIRED.ALL: 13.0

Experiment 94 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVZ R12, R15; CVTSI2SS XMM0, R12
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 44 e7          	cmove  r12,r15
     a:	f3 49 0f 2a c4       	cvtsi2ss xmm0,r12
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 26.99
  • Reference cycles: 22.73
  • UOPS_RETIRED.ALL: 13.0

Experiment 95 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVZ R12, R15; MOVD XMM0, R12D
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 44 e7          	cmove  r12,r15
     a:	66 41 0f 6e c4       	movd   xmm0,r12d
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 22.98
  • Reference cycles: 19.19
  • UOPS_RETIRED.ALL: 13.0

Experiment 96 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVZ R12, R15; MOVQ XMM0, R12
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 44 e7          	cmove  r12,r15
     a:	66 49 0f 6e c4       	movq   xmm0,r12
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 22.98
  • Reference cycles: 19.19
  • UOPS_RETIRED.ALL: 13.0

Experiment 97 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVZ R12, R15; PINSRB XMM0, R12D, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 44 e7          	cmove  r12,r15
     a:	66 41 0f 3a 20 c4 02 	pinsrb xmm0,r12d,0x2
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 23.06
  • Reference cycles: 19.7
  • UOPS_RETIRED.ALL: 13.0

Experiment 98 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVZ R12, R15; PINSRD XMM0, R12D, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 44 e7          	cmove  r12,r15
     a:	66 41 0f 3a 22 c4 02 	pinsrd xmm0,r12d,0x2
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 22.99
  • Reference cycles: 19.36
  • UOPS_RETIRED.ALL: 13.0

Experiment 99 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVZ R12, R15; PINSRQ XMM0, R12, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 44 e7          	cmove  r12,r15
     a:	66 49 0f 3a 22 c4 02 	pinsrq xmm0,r12,0x2
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 23.06
  • Reference cycles: 19.7
  • UOPS_RETIRED.ALL: 13.0

Experiment 100 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM1, XMM0, 2
• Chain instructions: CMOVZ R12, R15; PINSRW XMM0, R12D, 2
• Chain latency: ≥3
• Code:

     0:	66 0f 3a 60 c8 02    	pcmpestrm xmm1,xmm0,0x2
     6:	4d 0f 44 e7          	cmove  r12,r15
     a:	66 41 0f c4 c4 02    	pinsrw xmm0,r12d,0x2
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM0, XMM0;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 3.0
  • Core cycles: 23.06
  • Reference cycles: 19.36
  • UOPS_RETIRED.ALL: 13.0

Experiment 1

• Instruction: PCMPESTRM XMM1, XMM2, 2
• Chain instruction: CVTSD2SI EAX, XMM0
• Chain latency: ≥1
• Code:

     0:	66 0f 3a 60 ca 02    	pcmpestrm xmm1,xmm2,0x2
     6:	f2 0f 2d c0          	cvtsd2si eax,xmm0
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM2, XMM2

• Show nanoBench command
• Results:
  • Instructions retired: 2.0
  • Core cycles: 18.53
  • Reference cycles: 15.49
  • UOPS_RETIRED.ALL: 12.0

Experiment 2

• Instruction: PCMPESTRM XMM1, XMM2, 2
• Chain instruction: CVTSD2SI RAX, XMM0
• Chain latency: ≥1
• Code:

     0:	66 0f 3a 60 ca 02    	pcmpestrm xmm1,xmm2,0x2
     6:	f2 48 0f 2d c0       	cvtsd2si rax,xmm0
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM2, XMM2

• Show nanoBench command
• Results:
  • Instructions retired: 2.0
  • Core cycles: 18.65
  • Reference cycles: 15.66
  • UOPS_RETIRED.ALL: 12.0

Experiment 3

• Instruction: PCMPESTRM XMM1, XMM2, 2
• Chain instruction: CVTSS2SI EAX, XMM0
• Chain latency: ≥1
• Code:

     0:	66 0f 3a 60 ca 02    	pcmpestrm xmm1,xmm2,0x2
     6:	f3 0f 2d c0          	cvtss2si eax,xmm0
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM2, XMM2

• Show nanoBench command
• Results:
  • Instructions retired: 2.0
  • Core cycles: 18.58
  • Reference cycles: 15.49
  • UOPS_RETIRED.ALL: 12.0

Experiment 4

• Instruction: PCMPESTRM XMM1, XMM2, 2
• Chain instruction: CVTSS2SI RAX, XMM0
• Chain latency: ≥1
• Code:

     0:	66 0f 3a 60 ca 02    	pcmpestrm xmm1,xmm2,0x2
     6:	f3 48 0f 2d c0       	cvtss2si rax,xmm0
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM2, XMM2

• Show nanoBench command
• Results:
  • Instructions retired: 2.0
  • Core cycles: 18.63
  • Reference cycles: 15.49
  • UOPS_RETIRED.ALL: 12.0

Experiment 5

• Instruction: PCMPESTRM XMM1, XMM2, 2
• Chain instruction: CVTTSD2SI EAX, XMM0
• Chain latency: ≥1
• Code:

     0:	66 0f 3a 60 ca 02    	pcmpestrm xmm1,xmm2,0x2
     6:	f2 0f 2c c0          	cvttsd2si eax,xmm0
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM2, XMM2

• Show nanoBench command
• Results:
  • Instructions retired: 2.0
  • Core cycles: 18.53
  • Reference cycles: 15.49
  • UOPS_RETIRED.ALL: 12.0

Experiment 6

• Instruction: PCMPESTRM XMM1, XMM2, 2
• Chain instruction: CVTTSD2SI RAX, XMM0
• Chain latency: ≥1
• Code:

     0:	66 0f 3a 60 ca 02    	pcmpestrm xmm1,xmm2,0x2
     6:	f2 48 0f 2c c0       	cvttsd2si rax,xmm0
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM2, XMM2

• Show nanoBench command
• Results:
  • Instructions retired: 2.0
  • Core cycles: 18.64
  • Reference cycles: 15.49
  • UOPS_RETIRED.ALL: 12.0

Experiment 7

• Instruction: PCMPESTRM XMM1, XMM2, 2
• Chain instruction: CVTTSS2SI EAX, XMM0
• Chain latency: ≥1
• Code:

     0:	66 0f 3a 60 ca 02    	pcmpestrm xmm1,xmm2,0x2
     6:	f3 0f 2c c0          	cvttss2si eax,xmm0
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM2, XMM2

• Show nanoBench command
• Results:
  • Instructions retired: 2.0
  • Core cycles: 18.68
  • Reference cycles: 15.66
  • UOPS_RETIRED.ALL: 12.0

Experiment 8

• Instruction: PCMPESTRM XMM1, XMM2, 2
• Chain instruction: CVTTSS2SI RAX, XMM0
• Chain latency: ≥1
• Code:

     0:	66 0f 3a 60 ca 02    	pcmpestrm xmm1,xmm2,0x2
     6:	f3 48 0f 2c c0       	cvttss2si rax,xmm0
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM2, XMM2

• Show nanoBench command
• Results:
  • Instructions retired: 2.0
  • Core cycles: 18.55
  • Reference cycles: 15.66
  • UOPS_RETIRED.ALL: 12.0

Experiment 9

• Instruction: PCMPESTRM XMM1, XMM2, 2
• Chain instruction: EXTRACTPS EAX, XMM0, 2
• Chain latency: ≥1
• Code:

     0:	66 0f 3a 60 ca 02    	pcmpestrm xmm1,xmm2,0x2
     6:	66 0f 3a 17 c0 02    	extractps eax,xmm0,0x2
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM2, XMM2

• Show nanoBench command
• Results:
  • Instructions retired: 2.0
  • Core cycles: 18.55
  • Reference cycles: 15.49
  • UOPS_RETIRED.ALL: 12.0

Experiment 10

• Instruction: PCMPESTRM XMM1, XMM2, 2
• Chain instruction: MOVD EAX, XMM0
• Chain latency: ≥1
• Code:

     0:	66 0f 3a 60 ca 02    	pcmpestrm xmm1,xmm2,0x2
     6:	66 0f 7e c0          	movd   eax,xmm0
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM2, XMM2

• Show nanoBench command
• Results:
  • Instructions retired: 2.0
  • Core cycles: 18.58
  • Reference cycles: 15.49
  • UOPS_RETIRED.ALL: 12.0

Experiment 11

• Instruction: PCMPESTRM XMM1, XMM2, 2
• Chain instruction: MOVMSKPD EAX, XMM0
• Chain latency: ≥1
• Code:

     0:	66 0f 3a 60 ca 02    	pcmpestrm xmm1,xmm2,0x2
     6:	66 0f 50 c0          	movmskpd eax,xmm0
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM2, XMM2

• Show nanoBench command
• Results:
  • Instructions retired: 2.0
  • Core cycles: 18.65
  • Reference cycles: 15.49
  • UOPS_RETIRED.ALL: 12.0

Experiment 12

• Instruction: PCMPESTRM XMM1, XMM2, 2
• Chain instruction: MOVMSKPS EAX, XMM0
• Chain latency: ≥1
• Code:

     0:	66 0f 3a 60 ca 02    	pcmpestrm xmm1,xmm2,0x2
     6:	0f 50 c0             	movmskps eax,xmm0
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM2, XMM2

• Show nanoBench command
• Results:
  • Instructions retired: 2.0
  • Core cycles: 18.68
  • Reference cycles: 15.49
  • UOPS_RETIRED.ALL: 12.0

Experiment 13

• Instruction: PCMPESTRM XMM1, XMM2, 2
• Chain instruction: MOVQ RAX, XMM0
• Chain latency: ≥1
• Code:

     0:	66 0f 3a 60 ca 02    	pcmpestrm xmm1,xmm2,0x2
     6:	66 48 0f 7e c0       	movq   rax,xmm0
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM2, XMM2

• Show nanoBench command
• Results:
  • Instructions retired: 2.0
  • Core cycles: 18.56
  • Reference cycles: 15.32
  • UOPS_RETIRED.ALL: 12.0

Experiment 14

• Instruction: PCMPESTRM XMM1, XMM2, 2
• Chain instruction: PEXTRB EAX, XMM0, 2
• Chain latency: ≥1
• Code:

     0:	66 0f 3a 60 ca 02    	pcmpestrm xmm1,xmm2,0x2
     6:	66 0f 3a 14 c0 02    	pextrb eax,xmm0,0x2
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM2, XMM2

• Show nanoBench command
• Results:
  • Instructions retired: 2.0
  • Core cycles: 18.65
  • Reference cycles: 15.66
  • UOPS_RETIRED.ALL: 12.0

Experiment 15

• Instruction: PCMPESTRM XMM1, XMM2, 2
• Chain instruction: PEXTRD EAX, XMM0, 2
• Chain latency: ≥1
• Code:

     0:	66 0f 3a 60 ca 02    	pcmpestrm xmm1,xmm2,0x2
     6:	66 0f 3a 16 c0 02    	pextrd eax,xmm0,0x2
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM2, XMM2

• Show nanoBench command
• Results:
  • Instructions retired: 2.0
  • Core cycles: 18.6
  • Reference cycles: 15.49
  • UOPS_RETIRED.ALL: 12.0

Experiment 16

• Instruction: PCMPESTRM XMM1, XMM2, 2
• Chain instruction: PEXTRQ RAX, XMM0, 2
• Chain latency: ≥1
• Code:

     0:	66 0f 3a 60 ca 02    	pcmpestrm xmm1,xmm2,0x2
     6:	66 48 0f 3a 16 c0 02 	pextrq rax,xmm0,0x2
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM2, XMM2

• Show nanoBench command
• Results:
  • Instructions retired: 2.0
  • Core cycles: 18.67
  • Reference cycles: 15.49
  • UOPS_RETIRED.ALL: 12.0

Experiment 17

• Instruction: PCMPESTRM XMM1, XMM2, 2
• Chain instruction: PEXTRW EAX, XMM0, 2
• Chain latency: ≥1
• Code:

     0:	66 0f 3a 60 ca 02    	pcmpestrm xmm1,xmm2,0x2
     6:	66 0f c5 c0 02       	pextrw eax,xmm0,0x2
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM2, XMM2

• Show nanoBench command
• Results:
  • Instructions retired: 2.0
  • Core cycles: 18.51
  • Reference cycles: 15.49
  • UOPS_RETIRED.ALL: 12.0

Experiment 18

• Instruction: PCMPESTRM XMM1, XMM2, 2
• Chain instruction: PMOVMSKB EAX, XMM0
• Chain latency: ≥1
• Code:

     0:	66 0f 3a 60 ca 02    	pcmpestrm xmm1,xmm2,0x2
     6:	66 0f d7 c0          	pmovmskb eax,xmm0
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM2, XMM2

• Show nanoBench command
• Results:
  • Instructions retired: 2.0
  • Core cycles: 18.67
  • Reference cycles: 15.32
  • UOPS_RETIRED.ALL: 12.0

Experiment 19 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM1, XMM2, 2
• Chain instruction: CVTSD2SI EAX, XMM0
• Chain latency: ≥1
• Code:

     0:	66 0f 3a 60 ca 02    	pcmpestrm xmm1,xmm2,0x2
     6:	f2 0f 2d c0          	cvtsd2si eax,xmm0
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM2, XMM2;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 2.0
  • Core cycles: 18.67
  • Reference cycles: 15.66
  • UOPS_RETIRED.ALL: 12.0

Experiment 20 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM1, XMM2, 2
• Chain instruction: CVTSD2SI RAX, XMM0
• Chain latency: ≥1
• Code:

     0:	66 0f 3a 60 ca 02    	pcmpestrm xmm1,xmm2,0x2
     6:	f2 48 0f 2d c0       	cvtsd2si rax,xmm0
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM2, XMM2;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 2.0
  • Core cycles: 18.65
  • Reference cycles: 15.66
  • UOPS_RETIRED.ALL: 12.0

Experiment 21 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM1, XMM2, 2
• Chain instruction: CVTSS2SI EAX, XMM0
• Chain latency: ≥1
• Code:

     0:	66 0f 3a 60 ca 02    	pcmpestrm xmm1,xmm2,0x2
     6:	f3 0f 2d c0          	cvtss2si eax,xmm0
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM2, XMM2;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 2.0
  • Core cycles: 18.74
  • Reference cycles: 15.99
  • UOPS_RETIRED.ALL: 12.0

Experiment 22 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM1, XMM2, 2
• Chain instruction: CVTSS2SI RAX, XMM0
• Chain latency: ≥1
• Code:

     0:	66 0f 3a 60 ca 02    	pcmpestrm xmm1,xmm2,0x2
     6:	f3 48 0f 2d c0       	cvtss2si rax,xmm0
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM2, XMM2;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 2.0
  • Core cycles: 18.64
  • Reference cycles: 15.15
  • UOPS_RETIRED.ALL: 12.0

Experiment 23 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM1, XMM2, 2
• Chain instruction: CVTTSD2SI EAX, XMM0
• Chain latency: ≥1
• Code:

     0:	66 0f 3a 60 ca 02    	pcmpestrm xmm1,xmm2,0x2
     6:	f2 0f 2c c0          	cvttsd2si eax,xmm0
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM2, XMM2;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 2.0
  • Core cycles: 18.61
  • Reference cycles: 15.32
  • UOPS_RETIRED.ALL: 12.0

Experiment 24 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM1, XMM2, 2
• Chain instruction: CVTTSD2SI RAX, XMM0
• Chain latency: ≥1
• Code:

     0:	66 0f 3a 60 ca 02    	pcmpestrm xmm1,xmm2,0x2
     6:	f2 48 0f 2c c0       	cvttsd2si rax,xmm0
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM2, XMM2;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 2.0
  • Core cycles: 18.63
  • Reference cycles: 15.32
  • UOPS_RETIRED.ALL: 12.0

Experiment 25 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM1, XMM2, 2
• Chain instruction: CVTTSS2SI EAX, XMM0
• Chain latency: ≥1
• Code:

     0:	66 0f 3a 60 ca 02    	pcmpestrm xmm1,xmm2,0x2
     6:	f3 0f 2c c0          	cvttss2si eax,xmm0
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM2, XMM2;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 2.0
  • Core cycles: 18.75
  • Reference cycles: 15.66
  • UOPS_RETIRED.ALL: 12.0

Experiment 26 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM1, XMM2, 2
• Chain instruction: CVTTSS2SI RAX, XMM0
• Chain latency: ≥1
• Code:

     0:	66 0f 3a 60 ca 02    	pcmpestrm xmm1,xmm2,0x2
     6:	f3 48 0f 2c c0       	cvttss2si rax,xmm0
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM2, XMM2;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 2.0
  • Core cycles: 18.66
  • Reference cycles: 15.66
  • UOPS_RETIRED.ALL: 12.0

Experiment 27 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM1, XMM2, 2
• Chain instruction: EXTRACTPS EAX, XMM0, 2
• Chain latency: ≥1
• Code:

     0:	66 0f 3a 60 ca 02    	pcmpestrm xmm1,xmm2,0x2
     6:	66 0f 3a 17 c0 02    	extractps eax,xmm0,0x2
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM2, XMM2;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 2.0
  • Core cycles: 18.6
  • Reference cycles: 15.49
  • UOPS_RETIRED.ALL: 12.0

Experiment 28 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM1, XMM2, 2
• Chain instruction: MOVD EAX, XMM0
• Chain latency: ≥1
• Code:

     0:	66 0f 3a 60 ca 02    	pcmpestrm xmm1,xmm2,0x2
     6:	66 0f 7e c0          	movd   eax,xmm0
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM2, XMM2;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 2.0
  • Core cycles: 18.63
  • Reference cycles: 15.49
  • UOPS_RETIRED.ALL: 12.0

Experiment 29 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM1, XMM2, 2
• Chain instruction: MOVMSKPD EAX, XMM0
• Chain latency: ≥1
• Code:

     0:	66 0f 3a 60 ca 02    	pcmpestrm xmm1,xmm2,0x2
     6:	66 0f 50 c0          	movmskpd eax,xmm0
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM2, XMM2;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 2.0
  • Core cycles: 18.62
  • Reference cycles: 15.66
  • UOPS_RETIRED.ALL: 12.0

Experiment 30 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM1, XMM2, 2
• Chain instruction: MOVMSKPS EAX, XMM0
• Chain latency: ≥1
• Code:

     0:	66 0f 3a 60 ca 02    	pcmpestrm xmm1,xmm2,0x2
     6:	0f 50 c0             	movmskps eax,xmm0
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM2, XMM2;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 2.0
  • Core cycles: 18.66
  • Reference cycles: 15.32
  • UOPS_RETIRED.ALL: 12.0

Experiment 31 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM1, XMM2, 2
• Chain instruction: MOVQ RAX, XMM0
• Chain latency: ≥1
• Code:

     0:	66 0f 3a 60 ca 02    	pcmpestrm xmm1,xmm2,0x2
     6:	66 48 0f 7e c0       	movq   rax,xmm0
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM2, XMM2;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 2.0
  • Core cycles: 18.6
  • Reference cycles: 15.66
  • UOPS_RETIRED.ALL: 12.0

Experiment 32 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM1, XMM2, 2
• Chain instruction: PEXTRB EAX, XMM0, 2
• Chain latency: ≥1
• Code:

     0:	66 0f 3a 60 ca 02    	pcmpestrm xmm1,xmm2,0x2
     6:	66 0f 3a 14 c0 02    	pextrb eax,xmm0,0x2
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM2, XMM2;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 2.0
  • Core cycles: 18.67
  • Reference cycles: 16.16
  • UOPS_RETIRED.ALL: 12.0

Experiment 33 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM1, XMM2, 2
• Chain instruction: PEXTRD EAX, XMM0, 2
• Chain latency: ≥1
• Code:

     0:	66 0f 3a 60 ca 02    	pcmpestrm xmm1,xmm2,0x2
     6:	66 0f 3a 16 c0 02    	pextrd eax,xmm0,0x2
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM2, XMM2;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 2.0
  • Core cycles: 18.72
  • Reference cycles: 16.16
  • UOPS_RETIRED.ALL: 12.0

Experiment 34 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM1, XMM2, 2
• Chain instruction: PEXTRQ RAX, XMM0, 2
• Chain latency: ≥1
• Code:

     0:	66 0f 3a 60 ca 02    	pcmpestrm xmm1,xmm2,0x2
     6:	66 48 0f 3a 16 c0 02 	pextrq rax,xmm0,0x2
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM2, XMM2;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 2.0
  • Core cycles: 18.6
  • Reference cycles: 15.32
  • UOPS_RETIRED.ALL: 12.0

Experiment 35 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM1, XMM2, 2
• Chain instruction: PEXTRW EAX, XMM0, 2
• Chain latency: ≥1
• Code:

     0:	66 0f 3a 60 ca 02    	pcmpestrm xmm1,xmm2,0x2
     6:	66 0f c5 c0 02       	pextrw eax,xmm0,0x2
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM2, XMM2;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 2.0
  • Core cycles: 18.61
  • Reference cycles: 15.49
  • UOPS_RETIRED.ALL: 12.0

Experiment 36 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM1, XMM2, 2
• Chain instruction: PMOVMSKB EAX, XMM0
• Chain latency: ≥1
• Code:

     0:	66 0f 3a 60 ca 02    	pcmpestrm xmm1,xmm2,0x2
     6:	66 0f d7 c0          	pmovmskb eax,xmm0
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM2, XMM2;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 2.0
  • Core cycles: 18.65
  • Reference cycles: 15.15
  • UOPS_RETIRED.ALL: 12.0

Experiment 1

• Instruction: PCMPESTRM XMM1, XMM2, 2
• Chain instructions: CMOVC R15D, R15D;MOVSX RAX, R15D;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX
• Chain latency: 17
• Code:

     0:	66 0f 3a 60 ca 02    	pcmpestrm xmm1,xmm2,0x2
     6:	45 0f 42 ff          	cmovb  r15d,r15d
     a:	49 63 c7             	movsxd rax,r15d
     d:	48 63 c0             	movsxd rax,eax
    10:	48 63 c0             	movsxd rax,eax
    13:	48 63 c0             	movsxd rax,eax
    16:	48 63 c0             	movsxd rax,eax
    19:	48 63 c0             	movsxd rax,eax
    1c:	48 63 c0             	movsxd rax,eax
    1f:	48 63 c0             	movsxd rax,eax
    22:	48 63 c0             	movsxd rax,eax
    25:	48 63 c0             	movsxd rax,eax
    28:	48 63 c0             	movsxd rax,eax
    2b:	48 63 c0             	movsxd rax,eax
    2e:	48 63 c0             	movsxd rax,eax
    31:	48 63 c0             	movsxd rax,eax
    34:	48 63 c0             	movsxd rax,eax
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM2, XMM2

• Show nanoBench command
• Results:
  • Instructions retired: 17.0
  • Core cycles: 36.91
  • Reference cycles: 31.14
  • UOPS_RETIRED.ALL: 27.0

Experiment 2

• Instruction: PCMPESTRM XMM1, XMM2, 2
• Chain instructions: CMOVP R15D, R15D;MOVSX RAX, R15D;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX
• Chain latency: 17
• Code:

     0:	66 0f 3a 60 ca 02    	pcmpestrm xmm1,xmm2,0x2
     6:	45 0f 4a ff          	cmovp  r15d,r15d
     a:	49 63 c7             	movsxd rax,r15d
     d:	48 63 c0             	movsxd rax,eax
    10:	48 63 c0             	movsxd rax,eax
    13:	48 63 c0             	movsxd rax,eax
    16:	48 63 c0             	movsxd rax,eax
    19:	48 63 c0             	movsxd rax,eax
    1c:	48 63 c0             	movsxd rax,eax
    1f:	48 63 c0             	movsxd rax,eax
    22:	48 63 c0             	movsxd rax,eax
    25:	48 63 c0             	movsxd rax,eax
    28:	48 63 c0             	movsxd rax,eax
    2b:	48 63 c0             	movsxd rax,eax
    2e:	48 63 c0             	movsxd rax,eax
    31:	48 63 c0             	movsxd rax,eax
    34:	48 63 c0             	movsxd rax,eax
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM2, XMM2

• Show nanoBench command
• Results:
  • Instructions retired: 17.0
  • Core cycles: 37.01
  • Reference cycles: 31.48
  • UOPS_RETIRED.ALL: 27.0

Experiment 3

• Instruction: PCMPESTRM XMM1, XMM2, 2
• Chain instructions: CMOVS R15D, R15D;MOVSX RAX, R15D;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX
• Chain latency: 17
• Code:

     0:	66 0f 3a 60 ca 02    	pcmpestrm xmm1,xmm2,0x2
     6:	45 0f 48 ff          	cmovs  r15d,r15d
     a:	49 63 c7             	movsxd rax,r15d
     d:	48 63 c0             	movsxd rax,eax
    10:	48 63 c0             	movsxd rax,eax
    13:	48 63 c0             	movsxd rax,eax
    16:	48 63 c0             	movsxd rax,eax
    19:	48 63 c0             	movsxd rax,eax
    1c:	48 63 c0             	movsxd rax,eax
    1f:	48 63 c0             	movsxd rax,eax
    22:	48 63 c0             	movsxd rax,eax
    25:	48 63 c0             	movsxd rax,eax
    28:	48 63 c0             	movsxd rax,eax
    2b:	48 63 c0             	movsxd rax,eax
    2e:	48 63 c0             	movsxd rax,eax
    31:	48 63 c0             	movsxd rax,eax
    34:	48 63 c0             	movsxd rax,eax
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM2, XMM2

• Show nanoBench command
• Results:
  • Instructions retired: 17.0
  • Core cycles: 36.92
  • Reference cycles: 31.14
  • UOPS_RETIRED.ALL: 27.0

Experiment 4

• Instruction: PCMPESTRM XMM1, XMM2, 2
• Chain instructions: CMOVO R15D, R15D;MOVSX RAX, R15D;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX
• Chain latency: 17
• Code:

     0:	66 0f 3a 60 ca 02    	pcmpestrm xmm1,xmm2,0x2
     6:	45 0f 40 ff          	cmovo  r15d,r15d
     a:	49 63 c7             	movsxd rax,r15d
     d:	48 63 c0             	movsxd rax,eax
    10:	48 63 c0             	movsxd rax,eax
    13:	48 63 c0             	movsxd rax,eax
    16:	48 63 c0             	movsxd rax,eax
    19:	48 63 c0             	movsxd rax,eax
    1c:	48 63 c0             	movsxd rax,eax
    1f:	48 63 c0             	movsxd rax,eax
    22:	48 63 c0             	movsxd rax,eax
    25:	48 63 c0             	movsxd rax,eax
    28:	48 63 c0             	movsxd rax,eax
    2b:	48 63 c0             	movsxd rax,eax
    2e:	48 63 c0             	movsxd rax,eax
    31:	48 63 c0             	movsxd rax,eax
    34:	48 63 c0             	movsxd rax,eax
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM2, XMM2

• Show nanoBench command
• Results:
  • Instructions retired: 17.0
  • Core cycles: 36.92
  • Reference cycles: 30.81
  • UOPS_RETIRED.ALL: 27.0

Experiment 5

• Instruction: PCMPESTRM XMM1, XMM2, 2
• Chain instructions: CMOVZ R15D, R15D;MOVSX RAX, R15D;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX
• Chain latency: 17
• Code:

     0:	66 0f 3a 60 ca 02    	pcmpestrm xmm1,xmm2,0x2
     6:	45 0f 44 ff          	cmove  r15d,r15d
     a:	49 63 c7             	movsxd rax,r15d
     d:	48 63 c0             	movsxd rax,eax
    10:	48 63 c0             	movsxd rax,eax
    13:	48 63 c0             	movsxd rax,eax
    16:	48 63 c0             	movsxd rax,eax
    19:	48 63 c0             	movsxd rax,eax
    1c:	48 63 c0             	movsxd rax,eax
    1f:	48 63 c0             	movsxd rax,eax
    22:	48 63 c0             	movsxd rax,eax
    25:	48 63 c0             	movsxd rax,eax
    28:	48 63 c0             	movsxd rax,eax
    2b:	48 63 c0             	movsxd rax,eax
    2e:	48 63 c0             	movsxd rax,eax
    31:	48 63 c0             	movsxd rax,eax
    34:	48 63 c0             	movsxd rax,eax
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM2, XMM2

• Show nanoBench command
• Results:
  • Instructions retired: 17.0
  • Core cycles: 36.91
  • Reference cycles: 31.14
  • UOPS_RETIRED.ALL: 27.0

Experiment 6 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM1, XMM2, 2
• Chain instructions: CMOVC R15D, R15D;MOVSX RAX, R15D;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX
• Chain latency: 17
• Code:

     0:	66 0f 3a 60 ca 02    	pcmpestrm xmm1,xmm2,0x2
     6:	45 0f 42 ff          	cmovb  r15d,r15d
     a:	49 63 c7             	movsxd rax,r15d
     d:	48 63 c0             	movsxd rax,eax
    10:	48 63 c0             	movsxd rax,eax
    13:	48 63 c0             	movsxd rax,eax
    16:	48 63 c0             	movsxd rax,eax
    19:	48 63 c0             	movsxd rax,eax
    1c:	48 63 c0             	movsxd rax,eax
    1f:	48 63 c0             	movsxd rax,eax
    22:	48 63 c0             	movsxd rax,eax
    25:	48 63 c0             	movsxd rax,eax
    28:	48 63 c0             	movsxd rax,eax
    2b:	48 63 c0             	movsxd rax,eax
    2e:	48 63 c0             	movsxd rax,eax
    31:	48 63 c0             	movsxd rax,eax
    34:	48 63 c0             	movsxd rax,eax
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM2, XMM2;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 17.0
  • Core cycles: 37.0
  • Reference cycles: 31.48
  • UOPS_RETIRED.ALL: 27.0

Experiment 7 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM1, XMM2, 2
• Chain instructions: CMOVP R15D, R15D;MOVSX RAX, R15D;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX
• Chain latency: 17
• Code:

     0:	66 0f 3a 60 ca 02    	pcmpestrm xmm1,xmm2,0x2
     6:	45 0f 4a ff          	cmovp  r15d,r15d
     a:	49 63 c7             	movsxd rax,r15d
     d:	48 63 c0             	movsxd rax,eax
    10:	48 63 c0             	movsxd rax,eax
    13:	48 63 c0             	movsxd rax,eax
    16:	48 63 c0             	movsxd rax,eax
    19:	48 63 c0             	movsxd rax,eax
    1c:	48 63 c0             	movsxd rax,eax
    1f:	48 63 c0             	movsxd rax,eax
    22:	48 63 c0             	movsxd rax,eax
    25:	48 63 c0             	movsxd rax,eax
    28:	48 63 c0             	movsxd rax,eax
    2b:	48 63 c0             	movsxd rax,eax
    2e:	48 63 c0             	movsxd rax,eax
    31:	48 63 c0             	movsxd rax,eax
    34:	48 63 c0             	movsxd rax,eax
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM2, XMM2;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 17.0
  • Core cycles: 36.99
  • Reference cycles: 31.31
  • UOPS_RETIRED.ALL: 27.0

Experiment 8 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM1, XMM2, 2
• Chain instructions: CMOVS R15D, R15D;MOVSX RAX, R15D;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX
• Chain latency: 17
• Code:

     0:	66 0f 3a 60 ca 02    	pcmpestrm xmm1,xmm2,0x2
     6:	45 0f 48 ff          	cmovs  r15d,r15d
     a:	49 63 c7             	movsxd rax,r15d
     d:	48 63 c0             	movsxd rax,eax
    10:	48 63 c0             	movsxd rax,eax
    13:	48 63 c0             	movsxd rax,eax
    16:	48 63 c0             	movsxd rax,eax
    19:	48 63 c0             	movsxd rax,eax
    1c:	48 63 c0             	movsxd rax,eax
    1f:	48 63 c0             	movsxd rax,eax
    22:	48 63 c0             	movsxd rax,eax
    25:	48 63 c0             	movsxd rax,eax
    28:	48 63 c0             	movsxd rax,eax
    2b:	48 63 c0             	movsxd rax,eax
    2e:	48 63 c0             	movsxd rax,eax
    31:	48 63 c0             	movsxd rax,eax
    34:	48 63 c0             	movsxd rax,eax
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM2, XMM2;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 17.0
  • Core cycles: 37.0
  • Reference cycles: 30.97
  • UOPS_RETIRED.ALL: 27.0

Experiment 9 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM1, XMM2, 2
• Chain instructions: CMOVO R15D, R15D;MOVSX RAX, R15D;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX
• Chain latency: 17
• Code:

     0:	66 0f 3a 60 ca 02    	pcmpestrm xmm1,xmm2,0x2
     6:	45 0f 40 ff          	cmovo  r15d,r15d
     a:	49 63 c7             	movsxd rax,r15d
     d:	48 63 c0             	movsxd rax,eax
    10:	48 63 c0             	movsxd rax,eax
    13:	48 63 c0             	movsxd rax,eax
    16:	48 63 c0             	movsxd rax,eax
    19:	48 63 c0             	movsxd rax,eax
    1c:	48 63 c0             	movsxd rax,eax
    1f:	48 63 c0             	movsxd rax,eax
    22:	48 63 c0             	movsxd rax,eax
    25:	48 63 c0             	movsxd rax,eax
    28:	48 63 c0             	movsxd rax,eax
    2b:	48 63 c0             	movsxd rax,eax
    2e:	48 63 c0             	movsxd rax,eax
    31:	48 63 c0             	movsxd rax,eax
    34:	48 63 c0             	movsxd rax,eax
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM2, XMM2;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 17.0
  • Core cycles: 37.0
  • Reference cycles: 31.48
  • UOPS_RETIRED.ALL: 27.0

Experiment 10 (with clean upper 32 bits)

• Instruction: PCMPESTRM XMM1, XMM2, 2
• Chain instructions: CMOVZ R15D, R15D;MOVSX RAX, R15D;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX;MOVSX RAX, EAX
• Chain latency: 17
• Code:

     0:	66 0f 3a 60 ca 02    	pcmpestrm xmm1,xmm2,0x2
     6:	45 0f 44 ff          	cmove  r15d,r15d
     a:	49 63 c7             	movsxd rax,r15d
     d:	48 63 c0             	movsxd rax,eax
    10:	48 63 c0             	movsxd rax,eax
    13:	48 63 c0             	movsxd rax,eax
    16:	48 63 c0             	movsxd rax,eax
    19:	48 63 c0             	movsxd rax,eax
    1c:	48 63 c0             	movsxd rax,eax
    1f:	48 63 c0             	movsxd rax,eax
    22:	48 63 c0             	movsxd rax,eax
    25:	48 63 c0             	movsxd rax,eax
    28:	48 63 c0             	movsxd rax,eax
    2b:	48 63 c0             	movsxd rax,eax
    2e:	48 63 c0             	movsxd rax,eax
    31:	48 63 c0             	movsxd rax,eax
    34:	48 63 c0             	movsxd rax,eax
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM2, XMM2;
  MOV EDX, EDX;

• Show nanoBench command
• Results:
  • Instructions retired: 17.0
  • Core cycles: 36.98
  • Reference cycles: 31.14
  • UOPS_RETIRED.ALL: 27.0

Experiment 1

• Instruction: PCMPESTRM XMM1, XMM2, 2
• Chain instruction: CVTSD2SI EDX, XMM0
• Chain latency: ≥1
• Code:

     0:	66 0f 3a 60 ca 02    	pcmpestrm xmm1,xmm2,0x2
     6:	f2 0f 2d d0          	cvtsd2si edx,xmm0
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM2, XMM2

• Show nanoBench command
• Results:
  • Instructions retired: 2.0
  • Core cycles: 19.67
  • Reference cycles: 16.67
  • UOPS_RETIRED.ALL: 12.0

Experiment 2

• Instruction: PCMPESTRM XMM1, XMM2, 2
• Chain instruction: CVTSD2SI RDX, XMM0
• Chain latency: ≥1
• Code:

     0:	66 0f 3a 60 ca 02    	pcmpestrm xmm1,xmm2,0x2
     6:	f2 48 0f 2d d0       	cvtsd2si rdx,xmm0
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM2, XMM2

• Show nanoBench command
• Results:
  • Instructions retired: 2.0
  • Core cycles: 19.69
  • Reference cycles: 16.67
  • UOPS_RETIRED.ALL: 12.0

Experiment 3

• Instruction: PCMPESTRM XMM1, XMM2, 2
• Chain instruction: CVTSS2SI EDX, XMM0
• Chain latency: ≥1
• Code:

     0:	66 0f 3a 60 ca 02    	pcmpestrm xmm1,xmm2,0x2
     6:	f3 0f 2d d0          	cvtss2si edx,xmm0
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM2, XMM2

• Show nanoBench command
• Results:
  • Instructions retired: 2.0
  • Core cycles: 19.64
  • Reference cycles: 16.5
  • UOPS_RETIRED.ALL: 12.0

Experiment 4

• Instruction: PCMPESTRM XMM1, XMM2, 2
• Chain instruction: CVTSS2SI RDX, XMM0
• Chain latency: ≥1
• Code:

     0:	66 0f 3a 60 ca 02    	pcmpestrm xmm1,xmm2,0x2
     6:	f3 48 0f 2d d0       	cvtss2si rdx,xmm0
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM2, XMM2

• Show nanoBench command
• Results:
  • Instructions retired: 2.0
  • Core cycles: 19.69
  • Reference cycles: 16.5
  • UOPS_RETIRED.ALL: 12.0

Experiment 5

• Instruction: PCMPESTRM XMM1, XMM2, 2
• Chain instruction: CVTTSD2SI EDX, XMM0
• Chain latency: ≥1
• Code:

     0:	66 0f 3a 60 ca 02    	pcmpestrm xmm1,xmm2,0x2
     6:	f2 0f 2c d0          	cvttsd2si edx,xmm0
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM2, XMM2

• Show nanoBench command
• Results:
  • Instructions retired: 2.0
  • Core cycles: 19.69
  • Reference cycles: 16.5
  • UOPS_RETIRED.ALL: 12.0

Experiment 6

• Instruction: PCMPESTRM XMM1, XMM2, 2
• Chain instruction: CVTTSD2SI RDX, XMM0
• Chain latency: ≥1
• Code:

     0:	66 0f 3a 60 ca 02    	pcmpestrm xmm1,xmm2,0x2
     6:	f2 48 0f 2c d0       	cvttsd2si rdx,xmm0
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM2, XMM2

• Show nanoBench command
• Results:
  • Instructions retired: 2.0
  • Core cycles: 19.57
  • Reference cycles: 16.67
  • UOPS_RETIRED.ALL: 12.0

Experiment 7

• Instruction: PCMPESTRM XMM1, XMM2, 2
• Chain instruction: CVTTSS2SI EDX, XMM0
• Chain latency: ≥1
• Code:

     0:	66 0f 3a 60 ca 02    	pcmpestrm xmm1,xmm2,0x2
     6:	f3 0f 2c d0          	cvttss2si edx,xmm0
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM2, XMM2

• Show nanoBench command
• Results:
  • Instructions retired: 2.0
  • Core cycles: 19.59
  • Reference cycles: 16.5
  • UOPS_RETIRED.ALL: 12.0

Experiment 8

• Instruction: PCMPESTRM XMM1, XMM2, 2
• Chain instruction: CVTTSS2SI RDX, XMM0
• Chain latency: ≥1
• Code:

     0:	66 0f 3a 60 ca 02    	pcmpestrm xmm1,xmm2,0x2
     6:	f3 48 0f 2c d0       	cvttss2si rdx,xmm0
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM2, XMM2

• Show nanoBench command
• Results:
  • Instructions retired: 2.0
  • Core cycles: 19.66
  • Reference cycles: 16.5
  • UOPS_RETIRED.ALL: 12.0

Experiment 9

• Instruction: PCMPESTRM XMM1, XMM2, 2
• Chain instruction: EXTRACTPS EDX, XMM0, 2
• Chain latency: ≥1
• Code:

     0:	66 0f 3a 60 ca 02    	pcmpestrm xmm1,xmm2,0x2
     6:	66 0f 3a 17 c2 02    	extractps edx,xmm0,0x2
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM2, XMM2

• Show nanoBench command
• Results:
  • Instructions retired: 2.0
  • Core cycles: 19.57
  • Reference cycles: 16.33
  • UOPS_RETIRED.ALL: 12.0

Experiment 10

• Instruction: PCMPESTRM XMM1, XMM2, 2
• Chain instruction: MOVD EDX, XMM0
• Chain latency: ≥1
• Code:

     0:	66 0f 3a 60 ca 02    	pcmpestrm xmm1,xmm2,0x2
     6:	66 0f 7e c2          	movd   edx,xmm0
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM2, XMM2

• Show nanoBench command
• Results:
  • Instructions retired: 2.0
  • Core cycles: 19.63
  • Reference cycles: 16.67
  • UOPS_RETIRED.ALL: 12.0

Experiment 11

• Instruction: PCMPESTRM XMM1, XMM2, 2
• Chain instruction: MOVMSKPD EDX, XMM0
• Chain latency: ≥1
• Code:

     0:	66 0f 3a 60 ca 02    	pcmpestrm xmm1,xmm2,0x2
     6:	66 0f 50 d0          	movmskpd edx,xmm0
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM2, XMM2

• Show nanoBench command
• Results:
  • Instructions retired: 2.0
  • Core cycles: 19.65
  • Reference cycles: 16.5
  • UOPS_RETIRED.ALL: 12.0

Experiment 12

• Instruction: PCMPESTRM XMM1, XMM2, 2
• Chain instruction: MOVMSKPS EDX, XMM0
• Chain latency: ≥1
• Code:

     0:	66 0f 3a 60 ca 02    	pcmpestrm xmm1,xmm2,0x2
     6:	0f 50 d0             	movmskps edx,xmm0
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM2, XMM2

• Show nanoBench command
• Results:
  • Instructions retired: 2.0
  • Core cycles: 19.68
  • Reference cycles: 16.67
  • UOPS_RETIRED.ALL: 12.0

Experiment 13

• Instruction: PCMPESTRM XMM1, XMM2, 2
• Chain instruction: MOVQ RDX, XMM0
• Chain latency: ≥1
• Code:

     0:	66 0f 3a 60 ca 02    	pcmpestrm xmm1,xmm2,0x2
     6:	66 48 0f 7e c2       	movq   rdx,xmm0
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM2, XMM2

• Show nanoBench command
• Results:
  • Instructions retired: 2.0
  • Core cycles: 19.58
  • Reference cycles: 16.5
  • UOPS_RETIRED.ALL: 12.0

Experiment 14

• Instruction: PCMPESTRM XMM1, XMM2, 2
• Chain instruction: PEXTRB EDX, XMM0, 2
• Chain latency: ≥1
• Code:

     0:	66 0f 3a 60 ca 02    	pcmpestrm xmm1,xmm2,0x2
     6:	66 0f 3a 14 c2 02    	pextrb edx,xmm0,0x2
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM2, XMM2

• Show nanoBench command
• Results:
  • Instructions retired: 2.0
  • Core cycles: 19.62
  • Reference cycles: 16.67
  • UOPS_RETIRED.ALL: 12.0

Experiment 15

• Instruction: PCMPESTRM XMM1, XMM2, 2
• Chain instruction: PEXTRD EDX, XMM0, 2
• Chain latency: ≥1
• Code:

     0:	66 0f 3a 60 ca 02    	pcmpestrm xmm1,xmm2,0x2
     6:	66 0f 3a 16 c2 02    	pextrd edx,xmm0,0x2
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM2, XMM2

• Show nanoBench command
• Results:
  • Instructions retired: 2.0
  • Core cycles: 19.54
  • Reference cycles: 16.33
  • UOPS_RETIRED.ALL: 12.0

Experiment 16

• Instruction: PCMPESTRM XMM1, XMM2, 2
• Chain instruction: PEXTRQ RDX, XMM0, 2
• Chain latency: ≥1
• Code:

     0:	66 0f 3a 60 ca 02    	pcmpestrm xmm1,xmm2,0x2
     6:	66 48 0f 3a 16 c2 02 	pextrq rdx,xmm0,0x2
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM2, XMM2

• Show nanoBench command
• Results:
  • Instructions retired: 2.0
  • Core cycles: 19.65
  • Reference cycles: 16.5
  • UOPS_RETIRED.ALL: 12.0

Experiment 17

• Instruction: PCMPESTRM XMM1, XMM2, 2
• Chain instruction: PEXTRW EDX, XMM0, 2
• Chain latency: ≥1
• Code:

     0:	66 0f 3a 60 ca 02    	pcmpestrm xmm1,xmm2,0x2
     6:	66 0f c5 d0 02       	pextrw edx,xmm0,0x2
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM2, XMM2

• Show nanoBench command
• Results:
  • Instructions retired: 2.0
  • Core cycles: 19.58
  • Reference cycles: 16.5
  • UOPS_RETIRED.ALL: 12.0

Experiment 18

• Instruction: PCMPESTRM XMM1, XMM2, 2
• Chain instruction: PMOVMSKB EDX, XMM0
• Chain latency: ≥1
• Code:

     0:	66 0f 3a 60 ca 02    	pcmpestrm xmm1,xmm2,0x2
     6:	66 0f d7 d0          	pmovmskb edx,xmm0
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM2, XMM2

• Show nanoBench command
• Results:
  • Instructions retired: 2.0
  • Core cycles: 19.65
  • Reference cycles: 16.67
  • UOPS_RETIRED.ALL: 12.0

Experiment 1

• Instruction: PCMPESTRM XMM1, XMM2, 2
• Chain instructions: CMOVC R15D, R15D;MOVSX RDX, R15D;MOVSX RDX, EDX;MOVSX RDX, EDX;MOVSX RDX, EDX;MOVSX RDX, EDX;MOVSX RDX, EDX;MOVSX RDX, EDX;MOVSX RDX, EDX;MOVSX RDX, EDX;MOVSX RDX, EDX;MOVSX RDX, EDX;MOVSX RDX, EDX;MOVSX RDX, EDX;MOVSX RDX, EDX;MOVSX RDX, EDX
• Chain latency: 17
• Code:

     0:	66 0f 3a 60 ca 02    	pcmpestrm xmm1,xmm2,0x2
     6:	45 0f 42 ff          	cmovb  r15d,r15d
     a:	49 63 d7             	movsxd rdx,r15d
     d:	48 63 d2             	movsxd rdx,edx
    10:	48 63 d2             	movsxd rdx,edx
    13:	48 63 d2             	movsxd rdx,edx
    16:	48 63 d2             	movsxd rdx,edx
    19:	48 63 d2             	movsxd rdx,edx
    1c:	48 63 d2             	movsxd rdx,edx
    1f:	48 63 d2             	movsxd rdx,edx
    22:	48 63 d2             	movsxd rdx,edx
    25:	48 63 d2             	movsxd rdx,edx
    28:	48 63 d2             	movsxd rdx,edx
    2b:	48 63 d2             	movsxd rdx,edx
    2e:	48 63 d2             	movsxd rdx,edx
    31:	48 63 d2             	movsxd rdx,edx
    34:	48 63 d2             	movsxd rdx,edx
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM2, XMM2

• Show nanoBench command
• Results:
  • Instructions retired: 17.0
  • Core cycles: 38.0
  • Reference cycles: 31.82
  • UOPS_RETIRED.ALL: 27.0

Experiment 2

• Instruction: PCMPESTRM XMM1, XMM2, 2
• Chain instructions: CMOVP R15D, R15D;MOVSX RDX, R15D;MOVSX RDX, EDX;MOVSX RDX, EDX;MOVSX RDX, EDX;MOVSX RDX, EDX;MOVSX RDX, EDX;MOVSX RDX, EDX;MOVSX RDX, EDX;MOVSX RDX, EDX;MOVSX RDX, EDX;MOVSX RDX, EDX;MOVSX RDX, EDX;MOVSX RDX, EDX;MOVSX RDX, EDX;MOVSX RDX, EDX
• Chain latency: 17
• Code:

     0:	66 0f 3a 60 ca 02    	pcmpestrm xmm1,xmm2,0x2
     6:	45 0f 4a ff          	cmovp  r15d,r15d
     a:	49 63 d7             	movsxd rdx,r15d
     d:	48 63 d2             	movsxd rdx,edx
    10:	48 63 d2             	movsxd rdx,edx
    13:	48 63 d2             	movsxd rdx,edx
    16:	48 63 d2             	movsxd rdx,edx
    19:	48 63 d2             	movsxd rdx,edx
    1c:	48 63 d2             	movsxd rdx,edx
    1f:	48 63 d2             	movsxd rdx,edx
    22:	48 63 d2             	movsxd rdx,edx
    25:	48 63 d2             	movsxd rdx,edx
    28:	48 63 d2             	movsxd rdx,edx
    2b:	48 63 d2             	movsxd rdx,edx
    2e:	48 63 d2             	movsxd rdx,edx
    31:	48 63 d2             	movsxd rdx,edx
    34:	48 63 d2             	movsxd rdx,edx
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM2, XMM2

• Show nanoBench command
• Results:
  • Instructions retired: 17.0
  • Core cycles: 37.92
  • Reference cycles: 31.98
  • UOPS_RETIRED.ALL: 27.0

Experiment 3

• Instruction: PCMPESTRM XMM1, XMM2, 2
• Chain instructions: CMOVS R15D, R15D;MOVSX RDX, R15D;MOVSX RDX, EDX;MOVSX RDX, EDX;MOVSX RDX, EDX;MOVSX RDX, EDX;MOVSX RDX, EDX;MOVSX RDX, EDX;MOVSX RDX, EDX;MOVSX RDX, EDX;MOVSX RDX, EDX;MOVSX RDX, EDX;MOVSX RDX, EDX;MOVSX RDX, EDX;MOVSX RDX, EDX;MOVSX RDX, EDX
• Chain latency: 17
• Code:

     0:	66 0f 3a 60 ca 02    	pcmpestrm xmm1,xmm2,0x2
     6:	45 0f 48 ff          	cmovs  r15d,r15d
     a:	49 63 d7             	movsxd rdx,r15d
     d:	48 63 d2             	movsxd rdx,edx
    10:	48 63 d2             	movsxd rdx,edx
    13:	48 63 d2             	movsxd rdx,edx
    16:	48 63 d2             	movsxd rdx,edx
    19:	48 63 d2             	movsxd rdx,edx
    1c:	48 63 d2             	movsxd rdx,edx
    1f:	48 63 d2             	movsxd rdx,edx
    22:	48 63 d2             	movsxd rdx,edx
    25:	48 63 d2             	movsxd rdx,edx
    28:	48 63 d2             	movsxd rdx,edx
    2b:	48 63 d2             	movsxd rdx,edx
    2e:	48 63 d2             	movsxd rdx,edx
    31:	48 63 d2             	movsxd rdx,edx
    34:	48 63 d2             	movsxd rdx,edx
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM2, XMM2

• Show nanoBench command
• Results:
  • Instructions retired: 17.0
  • Core cycles: 37.99
  • Reference cycles: 31.98
  • UOPS_RETIRED.ALL: 27.0

Experiment 4

• Instruction: PCMPESTRM XMM1, XMM2, 2
• Chain instructions: CMOVO R15D, R15D;MOVSX RDX, R15D;MOVSX RDX, EDX;MOVSX RDX, EDX;MOVSX RDX, EDX;MOVSX RDX, EDX;MOVSX RDX, EDX;MOVSX RDX, EDX;MOVSX RDX, EDX;MOVSX RDX, EDX;MOVSX RDX, EDX;MOVSX RDX, EDX;MOVSX RDX, EDX;MOVSX RDX, EDX;MOVSX RDX, EDX;MOVSX RDX, EDX
• Chain latency: 17
• Code:

     0:	66 0f 3a 60 ca 02    	pcmpestrm xmm1,xmm2,0x2
     6:	45 0f 40 ff          	cmovo  r15d,r15d
     a:	49 63 d7             	movsxd rdx,r15d
     d:	48 63 d2             	movsxd rdx,edx
    10:	48 63 d2             	movsxd rdx,edx
    13:	48 63 d2             	movsxd rdx,edx
    16:	48 63 d2             	movsxd rdx,edx
    19:	48 63 d2             	movsxd rdx,edx
    1c:	48 63 d2             	movsxd rdx,edx
    1f:	48 63 d2             	movsxd rdx,edx
    22:	48 63 d2             	movsxd rdx,edx
    25:	48 63 d2             	movsxd rdx,edx
    28:	48 63 d2             	movsxd rdx,edx
    2b:	48 63 d2             	movsxd rdx,edx
    2e:	48 63 d2             	movsxd rdx,edx
    31:	48 63 d2             	movsxd rdx,edx
    34:	48 63 d2             	movsxd rdx,edx
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM2, XMM2

• Show nanoBench command
• Results:
  • Instructions retired: 17.0
  • Core cycles: 38.0
  • Reference cycles: 31.98
  • UOPS_RETIRED.ALL: 27.0

Experiment 5

• Instruction: PCMPESTRM XMM1, XMM2, 2
• Chain instructions: CMOVZ R15D, R15D;MOVSX RDX, R15D;MOVSX RDX, EDX;MOVSX RDX, EDX;MOVSX RDX, EDX;MOVSX RDX, EDX;MOVSX RDX, EDX;MOVSX RDX, EDX;MOVSX RDX, EDX;MOVSX RDX, EDX;MOVSX RDX, EDX;MOVSX RDX, EDX;MOVSX RDX, EDX;MOVSX RDX, EDX;MOVSX RDX, EDX;MOVSX RDX, EDX
• Chain latency: 17
• Code:

     0:	66 0f 3a 60 ca 02    	pcmpestrm xmm1,xmm2,0x2
     6:	45 0f 44 ff          	cmove  r15d,r15d
     a:	49 63 d7             	movsxd rdx,r15d
     d:	48 63 d2             	movsxd rdx,edx
    10:	48 63 d2             	movsxd rdx,edx
    13:	48 63 d2             	movsxd rdx,edx
    16:	48 63 d2             	movsxd rdx,edx
    19:	48 63 d2             	movsxd rdx,edx
    1c:	48 63 d2             	movsxd rdx,edx
    1f:	48 63 d2             	movsxd rdx,edx
    22:	48 63 d2             	movsxd rdx,edx
    25:	48 63 d2             	movsxd rdx,edx
    28:	48 63 d2             	movsxd rdx,edx
    2b:	48 63 d2             	movsxd rdx,edx
    2e:	48 63 d2             	movsxd rdx,edx
    31:	48 63 d2             	movsxd rdx,edx
    34:	48 63 d2             	movsxd rdx,edx
  

• Init:

  VZEROALL;
  PXOR XMM1, XMM1;
  PXOR XMM2, XMM2

• Show nanoBench command
• Results:
  • Instructions retired: 17.0
  • Core cycles: 37.99
  • Reference cycles: 31.65
  • UOPS_RETIRED.ALL: 27.0