SQRTSD (XMM, XMM)

Experiment 1

Instruction: SQRTSD XMM2, XMM0
Chain instructions: ORPD XMM2, XMM0; ORPD XMM2, XMM1; ANDPD XMM2, XMM1; ORPD XMM2, XMM2; ORPD XMM2, XMM2; ORPD XMM2, XMM2; ORPD XMM2, XMM2; ORPD XMM2, XMM2; ORPD XMM2, XMM2; ORPD XMM2, XMM2; ORPD XMM2, XMM2; ORPD XMM2, XMM2; ORPD XMM2, XMM2; ORPD XMM2, XMM2; ORPD XMM2, XMM2; ORPD XMM2, XMM2; ORPD XMM2, XMM2; ORPD XMM2, XMM2; ORPD XMM2, XMM2; ORPD XMM2, XMM2; ORPD XMM2, XMM2; ORPD XMM2, XMM2; ORPD XMM2, XMM2;
Chain latency: ≥23

Code:

   0:	f2 0f 51 d0          	sqrtsd xmm2,xmm0
   4:	66 0f 56 d0          	orpd   xmm2,xmm0
   8:	66 0f 56 d1          	orpd   xmm2,xmm1
   c:	66 0f 54 d1          	andpd  xmm2,xmm1
  10:	66 0f 56 d2          	orpd   xmm2,xmm2
  14:	66 0f 56 d2          	orpd   xmm2,xmm2
  18:	66 0f 56 d2          	orpd   xmm2,xmm2
  1c:	66 0f 56 d2          	orpd   xmm2,xmm2
  20:	66 0f 56 d2          	orpd   xmm2,xmm2
  24:	66 0f 56 d2          	orpd   xmm2,xmm2
  28:	66 0f 56 d2          	orpd   xmm2,xmm2
  2c:	66 0f 56 d2          	orpd   xmm2,xmm2
  30:	66 0f 56 d2          	orpd   xmm2,xmm2
  34:	66 0f 56 d2          	orpd   xmm2,xmm2
  38:	66 0f 56 d2          	orpd   xmm2,xmm2
  3c:	66 0f 56 d2          	orpd   xmm2,xmm2
  40:	66 0f 56 d2          	orpd   xmm2,xmm2
  44:	66 0f 56 d2          	orpd   xmm2,xmm2
  48:	66 0f 56 d2          	orpd   xmm2,xmm2
  4c:	66 0f 56 d2          	orpd   xmm2,xmm2
  50:	66 0f 56 d2          	orpd   xmm2,xmm2
  54:	66 0f 56 d2          	orpd   xmm2,xmm2
  58:	66 0f 56 d2          	orpd   xmm2,xmm2
  5c:	66 0f 56 d2          	orpd   xmm2,xmm2

Init:

MOV RAX, 0x465a61fe1acdc21c;
MOV [R14+0], RAX;
MOV [R14+8], RAX;
MOVUPD XMM2, [R14];
MOVUPD XMM1, [R14]

Show nanoBench command
Results:
- Instructions retired: 24.0
- Core cycles: 37.0
- Reference cycles: 28.77
- UOPS_RETIRED.ANY: 24.0

Experiment 2

Instruction: SQRTSD XMM2, XMM0

Code:

   0:	f2 0f 51 d0          	sqrtsd xmm2,xmm0

Init:

MOV RAX, 0x3ff0000000000000;
MOV [R14+0], RAX;
MOV [R14+8], RAX;
MOVUPD XMM2, [R14];
MOVUPD XMM1, [R14]

Show nanoBench command
Results:
- Instructions retired: 1.0
- Core cycles: 12.0
- Reference cycles: 9.01
- UOPS_RETIRED.ANY: 1.0

Experiment 3

Instruction: SQRTSD XMM2, XMM0
Chain instructions: ORPD XMM2, XMM0; ORPD XMM2, XMM1; ANDPD XMM2, XMM1; ORPD XMM2, XMM2; ORPD XMM2, XMM2; ORPD XMM2, XMM2; ORPD XMM2, XMM2; ORPD XMM2, XMM2; ORPD XMM2, XMM2; ORPD XMM2, XMM2; ORPD XMM2, XMM2; ORPD XMM2, XMM2; ORPD XMM2, XMM2; ORPD XMM2, XMM2; ORPD XMM2, XMM2; ORPD XMM2, XMM2; ORPD XMM2, XMM2; ORPD XMM2, XMM2; ORPD XMM2, XMM2; ORPD XMM2, XMM2; ORPD XMM2, XMM2; ORPD XMM2, XMM2; ORPD XMM2, XMM2;
Chain latency: ≥23

Code:

   0:	f2 0f 51 d0          	sqrtsd xmm2,xmm0
   4:	66 0f 56 d0          	orpd   xmm2,xmm0
   8:	66 0f 56 d1          	orpd   xmm2,xmm1
   c:	66 0f 54 d1          	andpd  xmm2,xmm1
  10:	66 0f 56 d2          	orpd   xmm2,xmm2
  14:	66 0f 56 d2          	orpd   xmm2,xmm2
  18:	66 0f 56 d2          	orpd   xmm2,xmm2
  1c:	66 0f 56 d2          	orpd   xmm2,xmm2
  20:	66 0f 56 d2          	orpd   xmm2,xmm2
  24:	66 0f 56 d2          	orpd   xmm2,xmm2
  28:	66 0f 56 d2          	orpd   xmm2,xmm2
  2c:	66 0f 56 d2          	orpd   xmm2,xmm2
  30:	66 0f 56 d2          	orpd   xmm2,xmm2
  34:	66 0f 56 d2          	orpd   xmm2,xmm2
  38:	66 0f 56 d2          	orpd   xmm2,xmm2
  3c:	66 0f 56 d2          	orpd   xmm2,xmm2
  40:	66 0f 56 d2          	orpd   xmm2,xmm2
  44:	66 0f 56 d2          	orpd   xmm2,xmm2
  48:	66 0f 56 d2          	orpd   xmm2,xmm2
  4c:	66 0f 56 d2          	orpd   xmm2,xmm2
  50:	66 0f 56 d2          	orpd   xmm2,xmm2
  54:	66 0f 56 d2          	orpd   xmm2,xmm2
  58:	66 0f 56 d2          	orpd   xmm2,xmm2
  5c:	66 0f 56 d2          	orpd   xmm2,xmm2

Init:

MOV RAX, 0x3ff0000000000000;
MOV [R14+0], RAX;
MOV [R14+8], RAX;
MOVUPD XMM2, [R14];
MOVUPD XMM1, [R14]

Show nanoBench command
Results:
- Instructions retired: 24.0
- Core cycles: 37.0
- Reference cycles: 28.6
- UOPS_RETIRED.ANY: 24.0

Experiment 1

Instruction: SQRTSD XMM0, XMM2
Chain instructions: ORPD XMM2, XMM0; ORPD XMM2, XMM1; ANDPD XMM2, XMM1; ORPD XMM2, XMM2; ORPD XMM2, XMM2; ORPD XMM2, XMM2; ORPD XMM2, XMM2; ORPD XMM2, XMM2; ORPD XMM2, XMM2; ORPD XMM2, XMM2; ORPD XMM2, XMM2; ORPD XMM2, XMM2; ORPD XMM2, XMM2; ORPD XMM2, XMM2; ORPD XMM2, XMM2; ORPD XMM2, XMM2; ORPD XMM2, XMM2; ORPD XMM2, XMM2; ORPD XMM2, XMM2; ORPD XMM2, XMM2; ORPD XMM2, XMM2; ORPD XMM2, XMM2; ORPD XMM2, XMM2;
Chain latency: ≥23

Code:

   0:	f2 0f 51 c2          	sqrtsd xmm0,xmm2
   4:	66 0f 56 d0          	orpd   xmm2,xmm0
   8:	66 0f 56 d1          	orpd   xmm2,xmm1
   c:	66 0f 54 d1          	andpd  xmm2,xmm1
  10:	66 0f 56 d2          	orpd   xmm2,xmm2
  14:	66 0f 56 d2          	orpd   xmm2,xmm2
  18:	66 0f 56 d2          	orpd   xmm2,xmm2
  1c:	66 0f 56 d2          	orpd   xmm2,xmm2
  20:	66 0f 56 d2          	orpd   xmm2,xmm2
  24:	66 0f 56 d2          	orpd   xmm2,xmm2
  28:	66 0f 56 d2          	orpd   xmm2,xmm2
  2c:	66 0f 56 d2          	orpd   xmm2,xmm2
  30:	66 0f 56 d2          	orpd   xmm2,xmm2
  34:	66 0f 56 d2          	orpd   xmm2,xmm2
  38:	66 0f 56 d2          	orpd   xmm2,xmm2
  3c:	66 0f 56 d2          	orpd   xmm2,xmm2
  40:	66 0f 56 d2          	orpd   xmm2,xmm2
  44:	66 0f 56 d2          	orpd   xmm2,xmm2
  48:	66 0f 56 d2          	orpd   xmm2,xmm2
  4c:	66 0f 56 d2          	orpd   xmm2,xmm2
  50:	66 0f 56 d2          	orpd   xmm2,xmm2
  54:	66 0f 56 d2          	orpd   xmm2,xmm2
  58:	66 0f 56 d2          	orpd   xmm2,xmm2
  5c:	66 0f 56 d2          	orpd   xmm2,xmm2

Init:

MOV RAX, 0x465a61fe1acdc21c;
MOV [R14+0], RAX;
MOV [R14+8], RAX;
MOVUPD XMM2, [R14];
MOVUPD XMM1, [R14]

Show nanoBench command
Results:
- Instructions retired: 24.0
- Core cycles: 43.0
- Reference cycles: 33.45
- UOPS_RETIRED.ANY: 24.0

Experiment 2

Instruction: SQRTSD XMM2, XMM2

Code:

   0:	f2 0f 51 d2          	sqrtsd xmm2,xmm2

Init:

MOV RAX, 0x3ff0000000000000;
MOV [R14+0], RAX;
MOV [R14+8], RAX;
MOVUPD XMM2, [R14];
MOVUPD XMM1, [R14]

Show nanoBench command
Results:
- Instructions retired: 1.0
- Core cycles: 18.0
- Reference cycles: 14.21
- UOPS_RETIRED.ANY: 1.0

Experiment 3

Instruction: SQRTSD XMM0, XMM2
Chain instructions: ORPD XMM2, XMM0; ORPD XMM2, XMM1; ANDPD XMM2, XMM1; ORPD XMM2, XMM2; ORPD XMM2, XMM2; ORPD XMM2, XMM2; ORPD XMM2, XMM2; ORPD XMM2, XMM2; ORPD XMM2, XMM2; ORPD XMM2, XMM2; ORPD XMM2, XMM2; ORPD XMM2, XMM2; ORPD XMM2, XMM2; ORPD XMM2, XMM2; ORPD XMM2, XMM2; ORPD XMM2, XMM2; ORPD XMM2, XMM2; ORPD XMM2, XMM2; ORPD XMM2, XMM2; ORPD XMM2, XMM2; ORPD XMM2, XMM2; ORPD XMM2, XMM2; ORPD XMM2, XMM2;
Chain latency: ≥23

Code:

   0:	f2 0f 51 c2          	sqrtsd xmm0,xmm2
   4:	66 0f 56 d0          	orpd   xmm2,xmm0
   8:	66 0f 56 d1          	orpd   xmm2,xmm1
   c:	66 0f 54 d1          	andpd  xmm2,xmm1
  10:	66 0f 56 d2          	orpd   xmm2,xmm2
  14:	66 0f 56 d2          	orpd   xmm2,xmm2
  18:	66 0f 56 d2          	orpd   xmm2,xmm2
  1c:	66 0f 56 d2          	orpd   xmm2,xmm2
  20:	66 0f 56 d2          	orpd   xmm2,xmm2
  24:	66 0f 56 d2          	orpd   xmm2,xmm2
  28:	66 0f 56 d2          	orpd   xmm2,xmm2
  2c:	66 0f 56 d2          	orpd   xmm2,xmm2
  30:	66 0f 56 d2          	orpd   xmm2,xmm2
  34:	66 0f 56 d2          	orpd   xmm2,xmm2
  38:	66 0f 56 d2          	orpd   xmm2,xmm2
  3c:	66 0f 56 d2          	orpd   xmm2,xmm2
  40:	66 0f 56 d2          	orpd   xmm2,xmm2
  44:	66 0f 56 d2          	orpd   xmm2,xmm2
  48:	66 0f 56 d2          	orpd   xmm2,xmm2
  4c:	66 0f 56 d2          	orpd   xmm2,xmm2
  50:	66 0f 56 d2          	orpd   xmm2,xmm2
  54:	66 0f 56 d2          	orpd   xmm2,xmm2
  58:	66 0f 56 d2          	orpd   xmm2,xmm2
  5c:	66 0f 56 d2          	orpd   xmm2,xmm2

Init:

MOV RAX, 0x3ff0000000000000;
MOV [R14+0], RAX;
MOV [R14+8], RAX;
MOVUPD XMM2, [R14];
MOVUPD XMM1, [R14]

Show nanoBench command
Results:
- Instructions retired: 24.0
- Core cycles: 43.0
- Reference cycles: 32.93
- UOPS_RETIRED.ANY: 24.0

SQRTSD (XMM, XMM) - Latency

Operands

Latency operand 1 → 1: ≤12

Latency operand 2 → 1: ≤18

Latency operand 1 → 1: ≤12

Experiment 1

Experiment 2

Experiment 3

Latency operand 2 → 1: ≤18

Experiment 1

Experiment 2

Experiment 3