ADDSD (XMM, M64) - Port Usage (IACA 2.3)
With blocking instructions for ports {'2', '3'}:
Throughput Analysis Report
--------------------------
Block Throughput: 6.50 Cycles Throughput Bottleneck: Backend. PORT2_AGU, Port2_DATA, PORT3_AGU, Port3_DATA
Port Binding In Cycles Per Iteration:
---------------------------------------------------------------------------------------
| Port | 0 - DV | 1 | 2 - D | 3 - D | 4 | 5 | 6 | 7 |
---------------------------------------------------------------------------------------
| Cycles | 0.3 0.0 | 0.3 | 6.5 6.5 | 6.5 6.5 | 0.0 | 0.3 | 0.0 | 0.0 |
---------------------------------------------------------------------------------------
| Num Of | Ports pressure in cycles | |
| Uops | 0 - DV | 1 | 2 - D | 3 - D | 4 | 5 | 6 | 7 | |
---------------------------------------------------------------------------------
| 1 | | | 0.5 0.5 | 0.5 0.5 | | | | | CP | lddqu xmm1, xmmword ptr [r14+0x40]
| 1 | | | 0.5 0.5 | 0.5 0.5 | | | | | CP | lddqu xmm2, xmmword ptr [r14+0x50]
| 1 | | | 0.5 0.5 | 0.5 0.5 | | | | | CP | lddqu xmm3, xmmword ptr [r14+0x60]
| 1 | | | 0.5 0.5 | 0.5 0.5 | | | | | CP | lddqu xmm4, xmmword ptr [r14+0x70]
| 1 | | | 0.5 0.5 | 0.5 0.5 | | | | | CP | lddqu xmm5, xmmword ptr [r14+0x80]
| 1 | | | 0.5 0.5 | 0.5 0.5 | | | | | CP | lddqu xmm6, xmmword ptr [r14+0x90]
| 1 | | | 0.5 0.5 | 0.5 0.5 | | | | | CP | lddqu xmm7, xmmword ptr [r14+0xa0]
| 1 | | | 0.5 0.5 | 0.5 0.5 | | | | | CP | lddqu xmm8, xmmword ptr [r14+0xb0]
| 1 | | | 0.5 0.5 | 0.5 0.5 | | | | | CP | lddqu xmm9, xmmword ptr [r14+0xc0]
| 1 | | | 0.5 0.5 | 0.5 0.5 | | | | | CP | lddqu xmm10, xmmword ptr [r14+0xd0]
| 1 | | | 0.5 0.5 | 0.5 0.5 | | | | | CP | lddqu xmm11, xmmword ptr [r14+0xe0]
| 1 | | | 0.5 0.5 | 0.5 0.5 | | | | | CP | lddqu xmm12, xmmword ptr [r14+0xf0]
| 2^ | 0.3 | 0.3 | 0.5 0.5 | 0.5 0.5 | | 0.3 | | | CP | addsd xmm0, qword ptr [r14]
Total Num Of Uops: 14
⇨ One μop that can only use ports {'2', '3'}
With blocking instructions for ports {'0', '1', '5'}:
Throughput Analysis Report
--------------------------
Block Throughput: 6.90 Cycles Throughput Bottleneck: Dependency chains (possibly between iterations)
Port Binding In Cycles Per Iteration:
---------------------------------------------------------------------------------------
| Port | 0 - DV | 1 | 2 - D | 3 - D | 4 | 5 | 6 | 7 |
---------------------------------------------------------------------------------------
| Cycles | 6.2 0.0 | 6.4 | 0.5 0.5 | 0.5 0.5 | 0.0 | 6.4 | 0.0 | 0.0 |
---------------------------------------------------------------------------------------
| Num Of | Ports pressure in cycles | |
| Uops | 0 - DV | 1 | 2 - D | 3 - D | 4 | 5 | 6 | 7 | |
---------------------------------------------------------------------------------
| 1 | 0.2 | 0.4 | | | | 0.3 | | | | addpd xmm1, xmm2
| 1 | 0.4 | 0.1 | | | | 0.6 | | | | addpd xmm3, xmm2
| 1 | 0.3 | 0.5 | | | | 0.2 | | | | addpd xmm4, xmm2
| 1 | 0.2 | 0.4 | | | | 0.3 | | | | addpd xmm5, xmm2
| 1 | 0.4 | 0.1 | | | | 0.4 | | | | addpd xmm6, xmm2
| 1 | 0.3 | 0.4 | | | | 0.2 | | | CP | addpd xmm7, xmm2
| 1 | 0.3 | 0.4 | | | | 0.3 | | | CP | addpd xmm8, xmm2
| 1 | 0.4 | 0.2 | | | | 0.3 | | | | addpd xmm9, xmm2
| 1 | 0.4 | 0.2 | | | | 0.3 | | | | addpd xmm10, xmm2
| 1 | 0.2 | 0.5 | | | | 0.3 | | | | addpd xmm11, xmm2
| 1 | 0.3 | 0.3 | | | | 0.3 | | | | addpd xmm12, xmm2
| 1 | 0.4 | 0.2 | | | | 0.4 | | | | addpd xmm1, xmm2
| 1 | 0.2 | 0.5 | | | | 0.2 | | | | addpd xmm3, xmm2
| 1 | 0.3 | 0.3 | | | | 0.3 | | | | addpd xmm4, xmm2
| 1 | 0.3 | 0.2 | | | | 0.4 | | | | addpd xmm5, xmm2
| 1 | 0.2 | 0.6 | | | | 0.2 | | | | addpd xmm6, xmm2
| 1 | 0.3 | 0.2 | | | | 0.4 | | | CP | addpd xmm7, xmm2
| 1 | 0.4 | 0.2 | | | | 0.3 | | | CP | addpd xmm8, xmm2
| 2^ | 0.3 | 0.5 | 0.5 0.5 | 0.5 0.5 | | 0.2 | | | | addsd xmm0, qword ptr [r14]
Total Num Of Uops: 20
⇨ One μop that can only use ports {'0', '1', '5'}
With an indexed addressing mode
With blocking instructions for ports {'2', '3'}:
Throughput Analysis Report
--------------------------
Block Throughput: 6.50 Cycles Throughput Bottleneck: Backend. PORT2_AGU, Port2_DATA, PORT3_AGU, Port3_DATA
Port Binding In Cycles Per Iteration:
---------------------------------------------------------------------------------------
| Port | 0 - DV | 1 | 2 - D | 3 - D | 4 | 5 | 6 | 7 |
---------------------------------------------------------------------------------------
| Cycles | 0.3 0.0 | 0.3 | 6.5 6.5 | 6.5 6.5 | 0.0 | 0.3 | 0.0 | 0.0 |
---------------------------------------------------------------------------------------
| Num Of | Ports pressure in cycles | |
| Uops | 0 - DV | 1 | 2 - D | 3 - D | 4 | 5 | 6 | 7 | |
---------------------------------------------------------------------------------
| 1 | | | 0.5 0.5 | 0.5 0.5 | | | | | CP | lddqu xmm1, xmmword ptr [r14+0x40]
| 1 | | | 0.5 0.5 | 0.5 0.5 | | | | | CP | lddqu xmm2, xmmword ptr [r14+0x50]
| 1 | | | 0.5 0.5 | 0.5 0.5 | | | | | CP | lddqu xmm3, xmmword ptr [r14+0x60]
| 1 | | | 0.5 0.5 | 0.5 0.5 | | | | | CP | lddqu xmm4, xmmword ptr [r14+0x70]
| 1 | | | 0.5 0.5 | 0.5 0.5 | | | | | CP | lddqu xmm5, xmmword ptr [r14+0x80]
| 1 | | | 0.5 0.5 | 0.5 0.5 | | | | | CP | lddqu xmm6, xmmword ptr [r14+0x90]
| 1 | | | 0.5 0.5 | 0.5 0.5 | | | | | CP | lddqu xmm7, xmmword ptr [r14+0xa0]
| 1 | | | 0.5 0.5 | 0.5 0.5 | | | | | CP | lddqu xmm8, xmmword ptr [r14+0xb0]
| 1 | | | 0.5 0.5 | 0.5 0.5 | | | | | CP | lddqu xmm9, xmmword ptr [r14+0xc0]
| 1 | | | 0.5 0.5 | 0.5 0.5 | | | | | CP | lddqu xmm10, xmmword ptr [r14+0xd0]
| 1 | | | 0.5 0.5 | 0.5 0.5 | | | | | CP | lddqu xmm11, xmmword ptr [r14+0xe0]
| 1 | | | 0.5 0.5 | 0.5 0.5 | | | | | CP | lddqu xmm12, xmmword ptr [r14+0xf0]
| 2 | 0.3 | 0.3 | 0.5 0.5 | 0.5 0.5 | | 0.3 | | | CP | addsd xmm0, qword ptr [r14+r13*1]
Total Num Of Uops: 14
⇨ One μop that can only use ports {'2', '3'}
With blocking instructions for ports {'0', '1', '5'}:
Throughput Analysis Report
--------------------------
Block Throughput: 6.90 Cycles Throughput Bottleneck: Dependency chains (possibly between iterations)
Port Binding In Cycles Per Iteration:
---------------------------------------------------------------------------------------
| Port | 0 - DV | 1 | 2 - D | 3 - D | 4 | 5 | 6 | 7 |
---------------------------------------------------------------------------------------
| Cycles | 6.2 0.0 | 6.3 | 0.5 0.5 | 0.5 0.5 | 0.0 | 6.4 | 0.0 | 0.0 |
---------------------------------------------------------------------------------------
| Num Of | Ports pressure in cycles | |
| Uops | 0 - DV | 1 | 2 - D | 3 - D | 4 | 5 | 6 | 7 | |
---------------------------------------------------------------------------------
| 1 | 0.4 | 0.3 | | | | 0.3 | | | | addpd xmm1, xmm2
| 1 | 0.3 | 0.3 | | | | 0.3 | | | | addpd xmm3, xmm2
| 1 | 0.2 | 0.4 | | | | 0.3 | | | | addpd xmm4, xmm2
| 1 | 0.4 | 0.3 | | | | 0.3 | | | | addpd xmm5, xmm2
| 1 | 0.3 | 0.3 | | | | 0.3 | | | | addpd xmm6, xmm2
| 1 | 0.2 | 0.3 | | | | 0.4 | | | | addpd xmm7, xmm2
| 1 | 0.4 | 0.3 | | | | 0.2 | | | CP | addpd xmm8, xmm2
| 1 | 0.3 | 0.3 | | | | 0.3 | | | | addpd xmm9, xmm2
| 1 | 0.3 | 0.2 | | | | 0.4 | | | | addpd xmm10, xmm2
| 1 | 0.4 | 0.3 | | | | 0.2 | | | | addpd xmm11, xmm2
| 1 | 0.2 | 0.4 | | | | 0.4 | | | | addpd xmm12, xmm2
| 1 | 0.3 | 0.2 | | | | 0.4 | | | | addpd xmm1, xmm2
| 1 | 0.3 | 0.3 | | | | 0.3 | | | | addpd xmm3, xmm2
| 1 | 0.2 | 0.3 | | | | 0.4 | | | | addpd xmm4, xmm2
| 1 | 0.4 | 0.3 | | | | 0.2 | | | | addpd xmm5, xmm2
| 1 | 0.2 | 0.3 | | | | 0.4 | | | | addpd xmm6, xmm2
| 1 | 0.3 | 0.4 | | | | 0.3 | | | | addpd xmm7, xmm2
| 1 | 0.4 | 0.3 | | | | 0.2 | | | CP | addpd xmm8, xmm2
| 2 | 0.2 | 0.4 | 0.5 0.5 | 0.5 0.5 | | 0.4 | | | | addsd xmm0, qword ptr [r14+r13*1]
Total Num Of Uops: 20
⇨ One μop that can only use ports {'0', '1', '5'}