RSQRTSS (XMM, XMM)

Experiment 1

Instruction: RSQRTSS XMM2, XMM0
Chain instructions: ORPS XMM2, XMM0; ORPS XMM2, XMM1; ANDPS XMM2, XMM1; ORPS XMM2, XMM2; ORPS XMM2, XMM2; ORPS XMM2, XMM2; ORPS XMM2, XMM2;
Chain latency: ≥7

Code:

   0:	f3 0f 52 d0          	rsqrtss xmm2,xmm0
   4:	0f 56 d0             	orps   xmm2,xmm0
   7:	0f 56 d1             	orps   xmm2,xmm1
   a:	0f 54 d1             	andps  xmm2,xmm1
   d:	0f 56 d2             	orps   xmm2,xmm2
  10:	0f 56 d2             	orps   xmm2,xmm2
  13:	0f 56 d2             	orps   xmm2,xmm2
  16:	0f 56 d2             	orps   xmm2,xmm2

Init:

VZEROALL;
MOV RAX, 0x72d30ff172d30ff1;
MOV [R14+0], RAX;
MOV [R14+8], RAX;
MOVUPS XMM2, [R14];
MOVUPS XMM1, [R14]

Show nanoBench command
Results:
- MPERF: 10.43
- APERF: 12.0
- UOPS: 8.0

Experiment 2

Instruction: RSQRTSS XMM2, XMM0

Code:

   0:	f3 0f 52 d0          	rsqrtss xmm2,xmm0

Init:

VZEROALL;
MOV RAX, 0x3f8000003f800000;
MOV [R14+0], RAX;
MOV [R14+8], RAX;
MOVUPS XMM2, [R14];
MOVUPS XMM1, [R14]

Show nanoBench command
Results:
- MPERF: 4.31
- APERF: 5.0
- UOPS: 1.0

Experiment 3

Instruction: RSQRTSS XMM2, XMM0
Chain instructions: ORPS XMM2, XMM0; ORPS XMM2, XMM1; ANDPS XMM2, XMM1; ORPS XMM2, XMM2; ORPS XMM2, XMM2; ORPS XMM2, XMM2; ORPS XMM2, XMM2;
Chain latency: ≥7

Code:

   0:	f3 0f 52 d0          	rsqrtss xmm2,xmm0
   4:	0f 56 d0             	orps   xmm2,xmm0
   7:	0f 56 d1             	orps   xmm2,xmm1
   a:	0f 54 d1             	andps  xmm2,xmm1
   d:	0f 56 d2             	orps   xmm2,xmm2
  10:	0f 56 d2             	orps   xmm2,xmm2
  13:	0f 56 d2             	orps   xmm2,xmm2
  16:	0f 56 d2             	orps   xmm2,xmm2

Init:

VZEROALL;
MOV RAX, 0x3f8000003f800000;
MOV [R14+0], RAX;
MOV [R14+8], RAX;
MOVUPS XMM2, [R14];
MOVUPS XMM1, [R14]

Show nanoBench command
Results:
- MPERF: 10.43
- APERF: 12.0
- UOPS: 8.0

Experiment 1

Instruction: RSQRTSS XMM0, XMM2
Chain instructions: ORPS XMM2, XMM0; ORPS XMM2, XMM1; ANDPS XMM2, XMM1; ORPS XMM2, XMM2; ORPS XMM2, XMM2; ORPS XMM2, XMM2; ORPS XMM2, XMM2;
Chain latency: ≥7

Code:

   0:	f3 0f 52 c2          	rsqrtss xmm0,xmm2
   4:	0f 56 d0             	orps   xmm2,xmm0
   7:	0f 56 d1             	orps   xmm2,xmm1
   a:	0f 54 d1             	andps  xmm2,xmm1
   d:	0f 56 d2             	orps   xmm2,xmm2
  10:	0f 56 d2             	orps   xmm2,xmm2
  13:	0f 56 d2             	orps   xmm2,xmm2
  16:	0f 56 d2             	orps   xmm2,xmm2

Init:

VZEROALL;
MOV RAX, 0x72d30ff172d30ff1;
MOV [R14+0], RAX;
MOV [R14+8], RAX;
MOVUPS XMM2, [R14];
MOVUPS XMM1, [R14]

Show nanoBench command
Results:
- MPERF: 10.48
- APERF: 12.0
- UOPS: 8.0

Experiment 2

Instruction: RSQRTSS XMM2, XMM2

Code:

   0:	f3 0f 52 d2          	rsqrtss xmm2,xmm2

Init:

VZEROALL;
MOV RAX, 0x3f8000003f800000;
MOV [R14+0], RAX;
MOV [R14+8], RAX;
MOVUPS XMM2, [R14];
MOVUPS XMM1, [R14]

Show nanoBench command
Results:
- MPERF: 4.36
- APERF: 5.0
- UOPS: 1.0

Experiment 3

Instruction: RSQRTSS XMM0, XMM2
Chain instructions: ORPS XMM2, XMM0; ORPS XMM2, XMM1; ANDPS XMM2, XMM1; ORPS XMM2, XMM2; ORPS XMM2, XMM2; ORPS XMM2, XMM2; ORPS XMM2, XMM2;
Chain latency: ≥7

Code:

   0:	f3 0f 52 c2          	rsqrtss xmm0,xmm2
   4:	0f 56 d0             	orps   xmm2,xmm0
   7:	0f 56 d1             	orps   xmm2,xmm1
   a:	0f 54 d1             	andps  xmm2,xmm1
   d:	0f 56 d2             	orps   xmm2,xmm2
  10:	0f 56 d2             	orps   xmm2,xmm2
  13:	0f 56 d2             	orps   xmm2,xmm2
  16:	0f 56 d2             	orps   xmm2,xmm2

Init:

VZEROALL;
MOV RAX, 0x3f8000003f800000;
MOV [R14+0], RAX;
MOV [R14+8], RAX;
MOVUPS XMM2, [R14];
MOVUPS XMM1, [R14]

Show nanoBench command
Results:
- MPERF: 10.31
- APERF: 12.0
- UOPS: 8.0

RSQRTSS (XMM, XMM) - Latency

Operands

Latency operand 1 → 1: ≤5

Latency operand 2 → 1: ≤5

Latency operand 1 → 1: ≤5

Experiment 1

Experiment 2

Experiment 3

Latency operand 2 → 1: ≤5

Experiment 1

Experiment 2

Experiment 3