SQRTSD (XMM, XMM) - Throughput and Uops

With unroll_count=500 and no inner loop

• Code:

     0:	f2 0f 51 d0          	sqrtsd xmm2,xmm0
  

• Init:

  MOV RAX, 0x3ff0000000000000;
  MOV [R14+0], RAX;
  MOV [R14+8], RAX;
  MOVUPD XMM0, [R14]

• Show nanoBench command
• Results:
  • Instructions retired: 1.0
  • Core cycles: 7.0
  • Reference cycles: 5.83
  • UOPS_RETIRED.ANY: 1.0
  • RETIRE_SLOTS: 1.0
  • UOPS_MS: 0.0
  • UOPS_PORT_0: 1.0
  • UOPS_PORT_1: 0.0
  • UOPS_PORT_2: 0.0
  • UOPS_PORT_3: 0.0
  • UOPS_PORT_4: 0.0
  • UOPS_PORT_5: 0.0
  • DIV_CYCLES: 7.0
  • ILD_STALL.LCP: 0.0
  • INST_DECODED.DEC0: 1.0

With unroll_count=500 and no inner loop

• Code:

     0:	f2 0f 51 d0          	sqrtsd xmm2,xmm0
     4:	0f ae e8             	lfence 
  

• Init:

  MOV RAX, 0x3ff0000000000000;
  MOV [R14+0], RAX;
  MOV [R14+8], RAX;
  MOVUPD XMM0, [R14]

• Show nanoBench command
• Results:
  • Instructions retired: 2.0
  • Core cycles: 14.25
  • Reference cycles: 11.88
  • UOPS_RETIRED.ANY: 3.0
  • RETIRE_SLOTS: 3.0
  • UOPS_MS: 3.0
  • UOPS_PORT_0: 1.0
  • UOPS_PORT_1: 1.0
  • UOPS_PORT_2: 0.0
  • UOPS_PORT_3: 0.0
  • UOPS_PORT_4: 0.0
  • UOPS_PORT_5: 0.0
  • DIV_CYCLES: 7.0
  • ILD_STALL.LCP: 0.0
  • INST_DECODED.DEC0: 2.0

With loop_count=1000 and unroll_count=10

• Code:

     0:	f2 0f 51 d0          	sqrtsd xmm2,xmm0
  

• Init:

  MOV RAX, 0x3ff0000000000000;
  MOV [R14+0], RAX;
  MOV [R14+8], RAX;
  MOVUPD XMM0, [R14]

• Show nanoBench command
• Results:
  • Instructions retired: 1.2
  • Core cycles: 7.0
  • Reference cycles: 5.83
  • UOPS_RETIRED.ANY: 1.2
  • RETIRE_SLOTS: 1.2
  • UOPS_MS: 0.0
  • UOPS_PORT_0: 1.0
  • UOPS_PORT_1: 0.0
  • UOPS_PORT_2: 0.0
  • UOPS_PORT_3: 0.0
  • UOPS_PORT_4: 0.0
  • UOPS_PORT_5: 0.0
  • DIV_CYCLES: 7.0
  • ILD_STALL.LCP: 0.0
  • INST_DECODED.DEC0: 0.01

With loop_count=100 and unroll_count=100

• Code:

     0:	f2 0f 51 d0          	sqrtsd xmm2,xmm0
  

• Init:

  MOV RAX, 0x3ff0000000000000;
  MOV [R14+0], RAX;
  MOV [R14+8], RAX;
  MOVUPD XMM0, [R14]

• Show nanoBench command
• Results:
  • Instructions retired: 1.02
  • Core cycles: 7.0
  • Reference cycles: 5.83
  • UOPS_RETIRED.ANY: 1.02
  • RETIRE_SLOTS: 1.02
  • UOPS_MS: 0.0
  • UOPS_PORT_0: 1.0
  • UOPS_PORT_1: 0.0
  • UOPS_PORT_2: 0.0
  • UOPS_PORT_3: 0.0
  • UOPS_PORT_4: 0.0
  • UOPS_PORT_5: 7.0
  • DIV_CYCLES: 7.0
  • ILD_STALL.LCP: 0.0
  • INST_DECODED.DEC0: 1.0

With unroll_count=200 and no inner loop

• Code:

     0:	f2 0f 51 d0          	sqrtsd xmm2,xmm0
     4:	f2 0f 51 d8          	sqrtsd xmm3,xmm0
     8:	f2 0f 51 e0          	sqrtsd xmm4,xmm0
     c:	f2 0f 51 e8          	sqrtsd xmm5,xmm0
  

• Init:

  MOV RAX, 0x3ff0000000000000;
  MOV [R14+0], RAX;
  MOV [R14+8], RAX;
  MOVUPD XMM0, [R14]

• Show nanoBench command
• Results:
  • Instructions retired: 4.0
  • Core cycles: 28.0
  • Reference cycles: 23.37
  • UOPS_RETIRED.ANY: 4.0
  • RETIRE_SLOTS: 4.0
  • UOPS_MS: 0.0
  • UOPS_PORT_0: 4.0
  • UOPS_PORT_1: 0.0
  • UOPS_PORT_2: 0.0
  • UOPS_PORT_3: 0.0
  • UOPS_PORT_4: 0.0
  • UOPS_PORT_5: 28.0
  • DIV_CYCLES: 28.0
  • ILD_STALL.LCP: 0.0
  • INST_DECODED.DEC0: 4.0

With loop_count=1000 and unroll_count=2

• Code:

     0:	f2 0f 51 d0          	sqrtsd xmm2,xmm0
     4:	f2 0f 51 d8          	sqrtsd xmm3,xmm0
     8:	f2 0f 51 e0          	sqrtsd xmm4,xmm0
     c:	f2 0f 51 e8          	sqrtsd xmm5,xmm0
  

• Init:

  MOV RAX, 0x3ff0000000000000;
  MOV [R14+0], RAX;
  MOV [R14+8], RAX;
  MOVUPD XMM0, [R14]

• Show nanoBench command
• Results:
  • Instructions retired: 5.0
  • Core cycles: 28.0
  • Reference cycles: 23.33
  • UOPS_RETIRED.ANY: 5.0
  • RETIRE_SLOTS: 5.0
  • UOPS_MS: 0.0
  • UOPS_PORT_0: 4.0
  • UOPS_PORT_1: 0.0
  • UOPS_PORT_2: 0.0
  • UOPS_PORT_3: 0.0
  • UOPS_PORT_4: 0.0
  • UOPS_PORT_5: 0.0
  • DIV_CYCLES: 28.0
  • ILD_STALL.LCP: 0.0
  • INST_DECODED.DEC0: 0.05

With loop_count=100 and unroll_count=20

• Code:

     0:	f2 0f 51 d0          	sqrtsd xmm2,xmm0
     4:	f2 0f 51 d8          	sqrtsd xmm3,xmm0
     8:	f2 0f 51 e0          	sqrtsd xmm4,xmm0
     c:	f2 0f 51 e8          	sqrtsd xmm5,xmm0
  

• Init:

  MOV RAX, 0x3ff0000000000000;
  MOV [R14+0], RAX;
  MOV [R14+8], RAX;
  MOVUPD XMM0, [R14]

• Show nanoBench command
• Results:
  • Instructions retired: 4.1
  • Core cycles: 28.0
  • Reference cycles: 23.33
  • UOPS_RETIRED.ANY: 4.1
  • RETIRE_SLOTS: 4.1
  • UOPS_MS: 0.0
  • UOPS_PORT_0: 4.0
  • UOPS_PORT_1: 0.0
  • UOPS_PORT_2: 0.0
  • UOPS_PORT_3: 0.0
  • UOPS_PORT_4: 0.0
  • UOPS_PORT_5: 28.0
  • DIV_CYCLES: 28.0
  • ILD_STALL.LCP: 0.0
  • INST_DECODED.DEC0: 4.02

With unroll_count=100 and no inner loop

• Code:

     0:	f2 0f 51 d0          	sqrtsd xmm2,xmm0
     4:	f2 0f 51 d8          	sqrtsd xmm3,xmm0
     8:	f2 0f 51 e0          	sqrtsd xmm4,xmm0
     c:	f2 0f 51 e8          	sqrtsd xmm5,xmm0
    10:	f2 0f 51 f0          	sqrtsd xmm6,xmm0
    14:	f2 0f 51 f8          	sqrtsd xmm7,xmm0
    18:	f2 44 0f 51 c0       	sqrtsd xmm8,xmm0
    1d:	f2 44 0f 51 c8       	sqrtsd xmm9,xmm0
  

• Init:

  MOV RAX, 0x3ff0000000000000;
  MOV [R14+0], RAX;
  MOV [R14+8], RAX;
  MOVUPD XMM0, [R14]

• Show nanoBench command
• Results:
  • Instructions retired: 8.0
  • Core cycles: 56.0
  • Reference cycles: 46.67
  • UOPS_RETIRED.ANY: 8.0
  • RETIRE_SLOTS: 8.0
  • UOPS_MS: 0.0
  • UOPS_PORT_0: 8.0
  • UOPS_PORT_1: 0.0
  • UOPS_PORT_2: 0.0
  • UOPS_PORT_3: 0.0
  • UOPS_PORT_4: 0.0
  • UOPS_PORT_5: 0.0
  • DIV_CYCLES: 56.0
  • ILD_STALL.LCP: 0.0
  • INST_DECODED.DEC0: 8.0

With loop_count=1000 and unroll_count=1

• Code:

     0:	f2 0f 51 d0          	sqrtsd xmm2,xmm0
     4:	f2 0f 51 d8          	sqrtsd xmm3,xmm0
     8:	f2 0f 51 e0          	sqrtsd xmm4,xmm0
     c:	f2 0f 51 e8          	sqrtsd xmm5,xmm0
    10:	f2 0f 51 f0          	sqrtsd xmm6,xmm0
    14:	f2 0f 51 f8          	sqrtsd xmm7,xmm0
    18:	f2 44 0f 51 c0       	sqrtsd xmm8,xmm0
    1d:	f2 44 0f 51 c8       	sqrtsd xmm9,xmm0
  

• Init:

  MOV RAX, 0x3ff0000000000000;
  MOV [R14+0], RAX;
  MOV [R14+8], RAX;
  MOVUPD XMM0, [R14]

• Show nanoBench command
• Results:
  • Instructions retired: 10.0
  • Core cycles: 56.0
  • Reference cycles: 46.66
  • UOPS_RETIRED.ANY: 10.0
  • RETIRE_SLOTS: 10.0
  • UOPS_MS: 0.0
  • UOPS_PORT_0: 8.0
  • UOPS_PORT_1: 0.0
  • UOPS_PORT_2: 0.0
  • UOPS_PORT_3: 0.0
  • UOPS_PORT_4: 0.0
  • UOPS_PORT_5: 0.0
  • DIV_CYCLES: 56.0
  • ILD_STALL.LCP: 0.0
  • INST_DECODED.DEC0: 0.1

With loop_count=100 and unroll_count=10

• Code:

     0:	f2 0f 51 d0          	sqrtsd xmm2,xmm0
     4:	f2 0f 51 d8          	sqrtsd xmm3,xmm0
     8:	f2 0f 51 e0          	sqrtsd xmm4,xmm0
     c:	f2 0f 51 e8          	sqrtsd xmm5,xmm0
    10:	f2 0f 51 f0          	sqrtsd xmm6,xmm0
    14:	f2 0f 51 f8          	sqrtsd xmm7,xmm0
    18:	f2 44 0f 51 c0       	sqrtsd xmm8,xmm0
    1d:	f2 44 0f 51 c8       	sqrtsd xmm9,xmm0
  

• Init:

  MOV RAX, 0x3ff0000000000000;
  MOV [R14+0], RAX;
  MOV [R14+8], RAX;
  MOVUPD XMM0, [R14]

• Show nanoBench command
• Results:
  • Instructions retired: 8.2
  • Core cycles: 56.0
  • Reference cycles: 46.66
  • UOPS_RETIRED.ANY: 8.2
  • RETIRE_SLOTS: 8.2
  • UOPS_MS: 0.0
  • UOPS_PORT_0: 8.0
  • UOPS_PORT_1: 0.0
  • UOPS_PORT_2: 0.0
  • UOPS_PORT_3: 0.0
  • UOPS_PORT_4: 0.0
  • UOPS_PORT_5: 0.0
  • DIV_CYCLES: 56.0
  • ILD_STALL.LCP: 0.0
  • INST_DECODED.DEC0: 8.03

With unroll_count=500 and no inner loop

• Code:

     0:	f2 0f 51 d0          	sqrtsd xmm2,xmm0
  

• Init:

  MOV RAX, 0x465a61fe1acdc21c;
  MOV [R14+0], RAX;
  MOV [R14+8], RAX;
  MOVUPD XMM0, [R14]

• Show nanoBench command
• Results:
  • Instructions retired: 1.0
  • Core cycles: 32.0
  • Reference cycles: 26.64
  • UOPS_RETIRED.ANY: 1.0
  • RETIRE_SLOTS: 1.0
  • UOPS_MS: 0.0
  • UOPS_PORT_0: 1.0
  • UOPS_PORT_1: 0.0
  • UOPS_PORT_2: 0.0
  • UOPS_PORT_3: 0.0
  • UOPS_PORT_4: 0.0
  • UOPS_PORT_5: 32.0
  • DIV_CYCLES: 32.0
  • ILD_STALL.LCP: 0.0
  • INST_DECODED.DEC0: 1.0

With unroll_count=500 and no inner loop

• Code:

     0:	f2 0f 51 d0          	sqrtsd xmm2,xmm0
     4:	0f ae e8             	lfence 
  

• Init:

  MOV RAX, 0x465a61fe1acdc21c;
  MOV [R14+0], RAX;
  MOV [R14+8], RAX;
  MOVUPD XMM0, [R14]

• Show nanoBench command
• Results:
  • Instructions retired: 2.0
  • Core cycles: 39.25
  • Reference cycles: 32.7
  • UOPS_RETIRED.ANY: 3.0
  • RETIRE_SLOTS: 3.0
  • UOPS_MS: 3.0
  • UOPS_PORT_0: 1.0
  • UOPS_PORT_1: 1.0
  • UOPS_PORT_2: 0.0
  • UOPS_PORT_3: 0.0
  • UOPS_PORT_4: 0.0
  • UOPS_PORT_5: 0.0
  • DIV_CYCLES: 32.0
  • ILD_STALL.LCP: 0.0
  • INST_DECODED.DEC0: 2.0

With loop_count=1000 and unroll_count=10

• Code:

     0:	f2 0f 51 d0          	sqrtsd xmm2,xmm0
  

• Init:

  MOV RAX, 0x465a61fe1acdc21c;
  MOV [R14+0], RAX;
  MOV [R14+8], RAX;
  MOVUPD XMM0, [R14]

• Show nanoBench command
• Results:
  • Instructions retired: 1.2
  • Core cycles: 32.0
  • Reference cycles: 26.67
  • UOPS_RETIRED.ANY: 1.2
  • RETIRE_SLOTS: 1.2
  • UOPS_MS: 0.0
  • UOPS_PORT_0: 1.0
  • UOPS_PORT_1: 0.0
  • UOPS_PORT_2: 0.0
  • UOPS_PORT_3: 0.0
  • UOPS_PORT_4: 0.0
  • UOPS_PORT_5: 0.0
  • DIV_CYCLES: 32.0
  • ILD_STALL.LCP: 0.0
  • INST_DECODED.DEC0: 0.01

With loop_count=100 and unroll_count=100

• Code:

     0:	f2 0f 51 d0          	sqrtsd xmm2,xmm0
  

• Init:

  MOV RAX, 0x465a61fe1acdc21c;
  MOV [R14+0], RAX;
  MOV [R14+8], RAX;
  MOVUPD XMM0, [R14]

• Show nanoBench command
• Results:
  • Instructions retired: 1.02
  • Core cycles: 32.0
  • Reference cycles: 26.67
  • UOPS_RETIRED.ANY: 1.02
  • RETIRE_SLOTS: 1.02
  • UOPS_MS: 0.0
  • UOPS_PORT_0: 1.0
  • UOPS_PORT_1: 0.0
  • UOPS_PORT_2: 0.0
  • UOPS_PORT_3: 0.0
  • UOPS_PORT_4: 0.0
  • UOPS_PORT_5: 0.0
  • DIV_CYCLES: 32.0
  • ILD_STALL.LCP: 0.0
  • INST_DECODED.DEC0: 1.0

With unroll_count=200 and no inner loop

• Code:

     0:	f2 0f 51 d0          	sqrtsd xmm2,xmm0
     4:	f2 0f 51 d8          	sqrtsd xmm3,xmm0
     8:	f2 0f 51 e0          	sqrtsd xmm4,xmm0
     c:	f2 0f 51 e8          	sqrtsd xmm5,xmm0
  

• Init:

  MOV RAX, 0x465a61fe1acdc21c;
  MOV [R14+0], RAX;
  MOV [R14+8], RAX;
  MOVUPD XMM0, [R14]

• Show nanoBench command
• Results:
  • Instructions retired: 4.0
  • Core cycles: 128.0
  • Reference cycles: 106.63
  • UOPS_RETIRED.ANY: 4.0
  • RETIRE_SLOTS: 4.0
  • UOPS_MS: 0.0
  • UOPS_PORT_0: 4.0
  • UOPS_PORT_1: 0.0
  • UOPS_PORT_2: 0.0
  • UOPS_PORT_3: 0.0
  • UOPS_PORT_4: 0.0
  • UOPS_PORT_5: 0.0
  • DIV_CYCLES: 128.0
  • ILD_STALL.LCP: 0.0
  • INST_DECODED.DEC0: 4.0

With loop_count=1000 and unroll_count=2

• Code:

     0:	f2 0f 51 d0          	sqrtsd xmm2,xmm0
     4:	f2 0f 51 d8          	sqrtsd xmm3,xmm0
     8:	f2 0f 51 e0          	sqrtsd xmm4,xmm0
     c:	f2 0f 51 e8          	sqrtsd xmm5,xmm0
  

• Init:

  MOV RAX, 0x465a61fe1acdc21c;
  MOV [R14+0], RAX;
  MOV [R14+8], RAX;
  MOVUPD XMM0, [R14]

• Show nanoBench command
• Results:
  • Instructions retired: 5.0
  • Core cycles: 128.0
  • Reference cycles: 106.67
  • UOPS_RETIRED.ANY: 5.0
  • RETIRE_SLOTS: 5.0
  • UOPS_MS: 0.0
  • UOPS_PORT_0: 4.0
  • UOPS_PORT_1: 0.0
  • UOPS_PORT_2: 0.0
  • UOPS_PORT_3: 0.0
  • UOPS_PORT_4: 0.0
  • UOPS_PORT_5: 0.0
  • DIV_CYCLES: 128.0
  • ILD_STALL.LCP: 0.0
  • INST_DECODED.DEC0: 0.05

With loop_count=100 and unroll_count=20

• Code:

     0:	f2 0f 51 d0          	sqrtsd xmm2,xmm0
     4:	f2 0f 51 d8          	sqrtsd xmm3,xmm0
     8:	f2 0f 51 e0          	sqrtsd xmm4,xmm0
     c:	f2 0f 51 e8          	sqrtsd xmm5,xmm0
  

• Init:

  MOV RAX, 0x465a61fe1acdc21c;
  MOV [R14+0], RAX;
  MOV [R14+8], RAX;
  MOVUPD XMM0, [R14]

• Show nanoBench command
• Results:
  • Instructions retired: 4.1
  • Core cycles: 128.0
  • Reference cycles: 106.67
  • UOPS_RETIRED.ANY: 4.1
  • RETIRE_SLOTS: 4.1
  • UOPS_MS: 0.0
  • UOPS_PORT_0: 4.0
  • UOPS_PORT_1: 0.0
  • UOPS_PORT_2: 0.0
  • UOPS_PORT_3: 0.0
  • UOPS_PORT_4: 0.0
  • UOPS_PORT_5: 0.0
  • DIV_CYCLES: 128.0
  • ILD_STALL.LCP: 0.0
  • INST_DECODED.DEC0: 4.02

With unroll_count=100 and no inner loop

• Code:

     0:	f2 0f 51 d0          	sqrtsd xmm2,xmm0
     4:	f2 0f 51 d8          	sqrtsd xmm3,xmm0
     8:	f2 0f 51 e0          	sqrtsd xmm4,xmm0
     c:	f2 0f 51 e8          	sqrtsd xmm5,xmm0
    10:	f2 0f 51 f0          	sqrtsd xmm6,xmm0
    14:	f2 0f 51 f8          	sqrtsd xmm7,xmm0
    18:	f2 44 0f 51 c0       	sqrtsd xmm8,xmm0
    1d:	f2 44 0f 51 c8       	sqrtsd xmm9,xmm0
  

• Init:

  MOV RAX, 0x465a61fe1acdc21c;
  MOV [R14+0], RAX;
  MOV [R14+8], RAX;
  MOVUPD XMM0, [R14]

• Show nanoBench command
• Results:
  • Instructions retired: 8.0
  • Core cycles: 256.0
  • Reference cycles: 213.23
  • UOPS_RETIRED.ANY: 8.0
  • RETIRE_SLOTS: 8.0
  • UOPS_MS: 0.0
  • UOPS_PORT_0: 8.0
  • UOPS_PORT_1: 0.0
  • UOPS_PORT_2: 0.0
  • UOPS_PORT_3: 0.0
  • UOPS_PORT_4: 0.0
  • UOPS_PORT_5: 0.0
  • DIV_CYCLES: 256.0
  • ILD_STALL.LCP: 0.0
  • INST_DECODED.DEC0: 8.0

With loop_count=1000 and unroll_count=1

• Code:

     0:	f2 0f 51 d0          	sqrtsd xmm2,xmm0
     4:	f2 0f 51 d8          	sqrtsd xmm3,xmm0
     8:	f2 0f 51 e0          	sqrtsd xmm4,xmm0
     c:	f2 0f 51 e8          	sqrtsd xmm5,xmm0
    10:	f2 0f 51 f0          	sqrtsd xmm6,xmm0
    14:	f2 0f 51 f8          	sqrtsd xmm7,xmm0
    18:	f2 44 0f 51 c0       	sqrtsd xmm8,xmm0
    1d:	f2 44 0f 51 c8       	sqrtsd xmm9,xmm0
  

• Init:

  MOV RAX, 0x465a61fe1acdc21c;
  MOV [R14+0], RAX;
  MOV [R14+8], RAX;
  MOVUPD XMM0, [R14]

• Show nanoBench command
• Results:
  • Instructions retired: 10.0
  • Core cycles: 256.0
  • Reference cycles: 213.33
  • UOPS_RETIRED.ANY: 10.0
  • RETIRE_SLOTS: 10.0
  • UOPS_MS: 0.0
  • UOPS_PORT_0: 8.0
  • UOPS_PORT_1: 0.0
  • UOPS_PORT_2: 0.0
  • UOPS_PORT_3: 0.0
  • UOPS_PORT_4: 0.0
  • UOPS_PORT_5: 0.0
  • DIV_CYCLES: 256.0
  • ILD_STALL.LCP: 0.0
  • INST_DECODED.DEC0: 0.1

With loop_count=100 and unroll_count=10

• Code:

     0:	f2 0f 51 d0          	sqrtsd xmm2,xmm0
     4:	f2 0f 51 d8          	sqrtsd xmm3,xmm0
     8:	f2 0f 51 e0          	sqrtsd xmm4,xmm0
     c:	f2 0f 51 e8          	sqrtsd xmm5,xmm0
    10:	f2 0f 51 f0          	sqrtsd xmm6,xmm0
    14:	f2 0f 51 f8          	sqrtsd xmm7,xmm0
    18:	f2 44 0f 51 c0       	sqrtsd xmm8,xmm0
    1d:	f2 44 0f 51 c8       	sqrtsd xmm9,xmm0
  

• Init:

  MOV RAX, 0x465a61fe1acdc21c;
  MOV [R14+0], RAX;
  MOV [R14+8], RAX;
  MOVUPD XMM0, [R14]

• Show nanoBench command
• Results:
  • Instructions retired: 8.2
  • Core cycles: 256.0
  • Reference cycles: 213.34
  • UOPS_RETIRED.ANY: 8.2
  • RETIRE_SLOTS: 8.2
  • UOPS_MS: 0.0
  • UOPS_PORT_0: 8.0
  • UOPS_PORT_1: 0.0
  • UOPS_PORT_2: 0.0
  • UOPS_PORT_3: 0.0
  • UOPS_PORT_4: 0.0
  • UOPS_PORT_5: 0.0
  • DIV_CYCLES: 256.0
  • ILD_STALL.LCP: 0.0
  • INST_DECODED.DEC0: 8.03