SQRTSS (XMM, XMM)

With unroll_count=500 and no inner loop

Code:

   0:	f3 0f 51 d0          	sqrtss xmm2,xmm0

Init:

MOV RAX, 0x3f8000003f800000;
MOV [R14+0], RAX;
MOV [R14+8], RAX;
MOVUPS XMM0, [R14]

Show nanoBench command
Results:
- Instructions retired: 1.0
- Core cycles: 31.0
- Reference cycles: 31.0
- UOPS_RETIRED.ANY: 3.0
- UOPS_MS: 3.0

With unroll_count=500 and no inner loop

Code:

   0:	f3 0f 51 d0          	sqrtss xmm2,xmm0
   4:	0f ae e8             	lfence

Init:

MOV RAX, 0x3f8000003f800000;
MOV [R14+0], RAX;
MOV [R14+8], RAX;
MOVUPS XMM0, [R14]

Show nanoBench command
Results:
- Instructions retired: 2.0
- Core cycles: 32.0
- Reference cycles: 32.0
- UOPS_RETIRED.ANY: 4.0
- UOPS_MS: 3.0

With loop_count=1000 and unroll_count=10

Code:

   0:	f3 0f 51 d0          	sqrtss xmm2,xmm0

Init:

MOV RAX, 0x3f8000003f800000;
MOV [R14+0], RAX;
MOV [R14+8], RAX;
MOVUPS XMM0, [R14]

Show nanoBench command
Results:
- Instructions retired: 1.2
- Core cycles: 31.1
- Reference cycles: 31.1
- UOPS_RETIRED.ANY: 3.2
- UOPS_MS: 3.0

With loop_count=100 and unroll_count=100

Code:

   0:	f3 0f 51 d0          	sqrtss xmm2,xmm0

Init:

MOV RAX, 0x3f8000003f800000;
MOV [R14+0], RAX;
MOV [R14+8], RAX;
MOVUPS XMM0, [R14]

Show nanoBench command
Results:
- Instructions retired: 1.02
- Core cycles: 31.01
- Reference cycles: 31.01
- UOPS_RETIRED.ANY: 3.02
- UOPS_MS: 3.0

With 4 independent instructions

With unroll_count=200 and no inner loop

Code:

   0:	f3 0f 51 d0          	sqrtss xmm2,xmm0
   4:	f3 0f 51 d8          	sqrtss xmm3,xmm0
   8:	f3 0f 51 e0          	sqrtss xmm4,xmm0
   c:	f3 0f 51 e8          	sqrtss xmm5,xmm0

Init:

MOV RAX, 0x3f8000003f800000;
MOV [R14+0], RAX;
MOV [R14+8], RAX;
MOVUPS XMM0, [R14]

Show nanoBench command
Results:
- Instructions retired: 4.0
- Core cycles: 124.0
- Reference cycles: 124.02
- UOPS_RETIRED.ANY: 12.0
- UOPS_MS: 12.0

With loop_count=1000 and unroll_count=2

Code:

   0:	f3 0f 51 d0          	sqrtss xmm2,xmm0
   4:	f3 0f 51 d8          	sqrtss xmm3,xmm0
   8:	f3 0f 51 e0          	sqrtss xmm4,xmm0
   c:	f3 0f 51 e8          	sqrtss xmm5,xmm0

Init:

MOV RAX, 0x3f8000003f800000;
MOV [R14+0], RAX;
MOV [R14+8], RAX;
MOVUPS XMM0, [R14]

Show nanoBench command
Results:
- Instructions retired: 5.0
- Core cycles: 124.51
- Reference cycles: 124.51
- UOPS_RETIRED.ANY: 13.0
- UOPS_MS: 12.01

With loop_count=100 and unroll_count=20

Code:

   0:	f3 0f 51 d0          	sqrtss xmm2,xmm0
   4:	f3 0f 51 d8          	sqrtss xmm3,xmm0
   8:	f3 0f 51 e0          	sqrtss xmm4,xmm0
   c:	f3 0f 51 e8          	sqrtss xmm5,xmm0

Init:

MOV RAX, 0x3f8000003f800000;
MOV [R14+0], RAX;
MOV [R14+8], RAX;
MOVUPS XMM0, [R14]

Show nanoBench command
Results:
- Instructions retired: 4.1
- Core cycles: 124.06
- Reference cycles: 124.06
- UOPS_RETIRED.ANY: 12.1
- UOPS_MS: 12.01

With 8 independent instructions

With unroll_count=100 and no inner loop

Code:

   0:	f3 0f 51 d0          	sqrtss xmm2,xmm0
   4:	f3 0f 51 d8          	sqrtss xmm3,xmm0
   8:	f3 0f 51 e0          	sqrtss xmm4,xmm0
   c:	f3 0f 51 e8          	sqrtss xmm5,xmm0
  10:	f3 0f 51 f0          	sqrtss xmm6,xmm0
  14:	f3 0f 51 f8          	sqrtss xmm7,xmm0
  18:	f3 44 0f 51 c0       	sqrtss xmm8,xmm0
  1d:	f3 44 0f 51 c8       	sqrtss xmm9,xmm0

Init:

MOV RAX, 0x3f8000003f800000;
MOV [R14+0], RAX;
MOV [R14+8], RAX;
MOVUPS XMM0, [R14]

Show nanoBench command
Results:
- Instructions retired: 8.0
- Core cycles: 248.0
- Reference cycles: 248.01
- UOPS_RETIRED.ANY: 24.0
- UOPS_MS: 24.0

With loop_count=1000 and unroll_count=1

Code:

   0:	f3 0f 51 d0          	sqrtss xmm2,xmm0
   4:	f3 0f 51 d8          	sqrtss xmm3,xmm0
   8:	f3 0f 51 e0          	sqrtss xmm4,xmm0
   c:	f3 0f 51 e8          	sqrtss xmm5,xmm0
  10:	f3 0f 51 f0          	sqrtss xmm6,xmm0
  14:	f3 0f 51 f8          	sqrtss xmm7,xmm0
  18:	f3 44 0f 51 c0       	sqrtss xmm8,xmm0
  1d:	f3 44 0f 51 c8       	sqrtss xmm9,xmm0

Init:

MOV RAX, 0x3f8000003f800000;
MOV [R14+0], RAX;
MOV [R14+8], RAX;
MOVUPS XMM0, [R14]

Show nanoBench command
Results:
- Instructions retired: 10.0
- Core cycles: 249.02
- Reference cycles: 249.02
- UOPS_RETIRED.ANY: 26.0
- UOPS_MS: 24.02

With loop_count=100 and unroll_count=10

Code:

   0:	f3 0f 51 d0          	sqrtss xmm2,xmm0
   4:	f3 0f 51 d8          	sqrtss xmm3,xmm0
   8:	f3 0f 51 e0          	sqrtss xmm4,xmm0
   c:	f3 0f 51 e8          	sqrtss xmm5,xmm0
  10:	f3 0f 51 f0          	sqrtss xmm6,xmm0
  14:	f3 0f 51 f8          	sqrtss xmm7,xmm0
  18:	f3 44 0f 51 c0       	sqrtss xmm8,xmm0
  1d:	f3 44 0f 51 c8       	sqrtss xmm9,xmm0

Init:

MOV RAX, 0x3f8000003f800000;
MOV [R14+0], RAX;
MOV [R14+8], RAX;
MOVUPS XMM0, [R14]

Show nanoBench command
Results:
- Instructions retired: 8.2
- Core cycles: 248.12
- Reference cycles: 248.11
- UOPS_RETIRED.ANY: 24.2
- UOPS_MS: 24.03

Slow division

With 1 independent instruction

With unroll_count=500 and no inner loop

Code:

   0:	f3 0f 51 d0          	sqrtss xmm2,xmm0

Init:

MOV RAX, 0x72d30ff172d30ff1;
MOV [R14+0], RAX;
MOV [R14+8], RAX;
MOVUPS XMM0, [R14]

Show nanoBench command
Results:
- Instructions retired: 1.0
- Core cycles: 31.0
- Reference cycles: 31.0
- UOPS_RETIRED.ANY: 3.0
- UOPS_MS: 3.0

With unroll_count=500 and no inner loop

Code:

   0:	f3 0f 51 d0          	sqrtss xmm2,xmm0
   4:	0f ae e8             	lfence

Init:

MOV RAX, 0x72d30ff172d30ff1;
MOV [R14+0], RAX;
MOV [R14+8], RAX;
MOVUPS XMM0, [R14]

Show nanoBench command
Results:
- Instructions retired: 2.0
- Core cycles: 32.0
- Reference cycles: 32.0
- UOPS_RETIRED.ANY: 4.0
- UOPS_MS: 3.0

With loop_count=1000 and unroll_count=10

Code:

   0:	f3 0f 51 d0          	sqrtss xmm2,xmm0

Init:

MOV RAX, 0x72d30ff172d30ff1;
MOV [R14+0], RAX;
MOV [R14+8], RAX;
MOVUPS XMM0, [R14]

Show nanoBench command
Results:
- Instructions retired: 1.2
- Core cycles: 31.1
- Reference cycles: 31.1
- UOPS_RETIRED.ANY: 3.2
- UOPS_MS: 3.0

With loop_count=100 and unroll_count=100

Code:

   0:	f3 0f 51 d0          	sqrtss xmm2,xmm0

Init:

MOV RAX, 0x72d30ff172d30ff1;
MOV [R14+0], RAX;
MOV [R14+8], RAX;
MOVUPS XMM0, [R14]

Show nanoBench command
Results:
- Instructions retired: 1.02
- Core cycles: 31.01
- Reference cycles: 31.01
- UOPS_RETIRED.ANY: 3.02
- UOPS_MS: 3.0

With 4 independent instructions

With unroll_count=200 and no inner loop

Code:

   0:	f3 0f 51 d0          	sqrtss xmm2,xmm0
   4:	f3 0f 51 d8          	sqrtss xmm3,xmm0
   8:	f3 0f 51 e0          	sqrtss xmm4,xmm0
   c:	f3 0f 51 e8          	sqrtss xmm5,xmm0

Init:

MOV RAX, 0x72d30ff172d30ff1;
MOV [R14+0], RAX;
MOV [R14+8], RAX;
MOVUPS XMM0, [R14]

Show nanoBench command
Results:
- Instructions retired: 4.0
- Core cycles: 124.0
- Reference cycles: 124.01
- UOPS_RETIRED.ANY: 12.0
- UOPS_MS: 12.0

With loop_count=1000 and unroll_count=2

Code:

   0:	f3 0f 51 d0          	sqrtss xmm2,xmm0
   4:	f3 0f 51 d8          	sqrtss xmm3,xmm0
   8:	f3 0f 51 e0          	sqrtss xmm4,xmm0
   c:	f3 0f 51 e8          	sqrtss xmm5,xmm0

Init:

MOV RAX, 0x72d30ff172d30ff1;
MOV [R14+0], RAX;
MOV [R14+8], RAX;
MOVUPS XMM0, [R14]

Show nanoBench command
Results:
- Instructions retired: 5.0
- Core cycles: 124.51
- Reference cycles: 124.51
- UOPS_RETIRED.ANY: 13.0
- UOPS_MS: 12.01

With loop_count=100 and unroll_count=20

Code:

   0:	f3 0f 51 d0          	sqrtss xmm2,xmm0
   4:	f3 0f 51 d8          	sqrtss xmm3,xmm0
   8:	f3 0f 51 e0          	sqrtss xmm4,xmm0
   c:	f3 0f 51 e8          	sqrtss xmm5,xmm0

Init:

MOV RAX, 0x72d30ff172d30ff1;
MOV [R14+0], RAX;
MOV [R14+8], RAX;
MOVUPS XMM0, [R14]

Show nanoBench command
Results:
- Instructions retired: 4.1
- Core cycles: 124.06
- Reference cycles: 124.06
- UOPS_RETIRED.ANY: 12.1
- UOPS_MS: 12.01

With 8 independent instructions

With unroll_count=100 and no inner loop

Code:

   0:	f3 0f 51 d0          	sqrtss xmm2,xmm0
   4:	f3 0f 51 d8          	sqrtss xmm3,xmm0
   8:	f3 0f 51 e0          	sqrtss xmm4,xmm0
   c:	f3 0f 51 e8          	sqrtss xmm5,xmm0
  10:	f3 0f 51 f0          	sqrtss xmm6,xmm0
  14:	f3 0f 51 f8          	sqrtss xmm7,xmm0
  18:	f3 44 0f 51 c0       	sqrtss xmm8,xmm0
  1d:	f3 44 0f 51 c8       	sqrtss xmm9,xmm0

Init:

MOV RAX, 0x72d30ff172d30ff1;
MOV [R14+0], RAX;
MOV [R14+8], RAX;
MOVUPS XMM0, [R14]

Show nanoBench command
Results:
- Instructions retired: 8.0
- Core cycles: 248.0
- Reference cycles: 248.0
- UOPS_RETIRED.ANY: 24.0
- UOPS_MS: 24.0

With loop_count=1000 and unroll_count=1

Code:

   0:	f3 0f 51 d0          	sqrtss xmm2,xmm0
   4:	f3 0f 51 d8          	sqrtss xmm3,xmm0
   8:	f3 0f 51 e0          	sqrtss xmm4,xmm0
   c:	f3 0f 51 e8          	sqrtss xmm5,xmm0
  10:	f3 0f 51 f0          	sqrtss xmm6,xmm0
  14:	f3 0f 51 f8          	sqrtss xmm7,xmm0
  18:	f3 44 0f 51 c0       	sqrtss xmm8,xmm0
  1d:	f3 44 0f 51 c8       	sqrtss xmm9,xmm0

Init:

MOV RAX, 0x72d30ff172d30ff1;
MOV [R14+0], RAX;
MOV [R14+8], RAX;
MOVUPS XMM0, [R14]

Show nanoBench command
Results:
- Instructions retired: 10.0
- Core cycles: 249.02
- Reference cycles: 249.02
- UOPS_RETIRED.ANY: 26.0
- UOPS_MS: 24.02

With loop_count=100 and unroll_count=10

Code:

   0:	f3 0f 51 d0          	sqrtss xmm2,xmm0
   4:	f3 0f 51 d8          	sqrtss xmm3,xmm0
   8:	f3 0f 51 e0          	sqrtss xmm4,xmm0
   c:	f3 0f 51 e8          	sqrtss xmm5,xmm0
  10:	f3 0f 51 f0          	sqrtss xmm6,xmm0
  14:	f3 0f 51 f8          	sqrtss xmm7,xmm0
  18:	f3 44 0f 51 c0       	sqrtss xmm8,xmm0
  1d:	f3 44 0f 51 c8       	sqrtss xmm9,xmm0

Init:

MOV RAX, 0x72d30ff172d30ff1;
MOV [R14+0], RAX;
MOV [R14+8], RAX;
MOVUPS XMM0, [R14]

Show nanoBench command
Results:
- Instructions retired: 8.2
- Core cycles: 248.12
- Reference cycles: 248.12
- UOPS_RETIRED.ANY: 24.2
- UOPS_MS: 24.03

SQRTSS (XMM, XMM) - Throughput and Uops

Fast division

With 1 independent instruction

With unroll_count=500 and no inner loop

With unroll_count=500 and no inner loop

With loop_count=1000 and unroll_count=10

With loop_count=100 and unroll_count=100

With 4 independent instructions

With unroll_count=200 and no inner loop

With loop_count=1000 and unroll_count=2

With loop_count=100 and unroll_count=20

With 8 independent instructions

With unroll_count=100 and no inner loop

With loop_count=1000 and unroll_count=1

With loop_count=100 and unroll_count=10

Slow division

With 1 independent instruction

With unroll_count=500 and no inner loop

With unroll_count=500 and no inner loop

With loop_count=1000 and unroll_count=10

With loop_count=100 and unroll_count=100

With 4 independent instructions

With unroll_count=200 and no inner loop

With loop_count=1000 and unroll_count=2

With loop_count=100 and unroll_count=20

With 8 independent instructions

With unroll_count=100 and no inner loop

With loop_count=1000 and unroll_count=1

With loop_count=100 and unroll_count=10