MOVUPS (M128, XMM) - Port Usage (IACA 2.3)


With blocking instructions for port '4':

Throughput Analysis Report
--------------------------
Block Throughput: 11.00 Cycles       Throughput Bottleneck: Backend. Port4

Port Binding In Cycles Per Iteration:
---------------------------------------------------------------------------------------
|  Port  |  0   -  DV  |  1   |  2   -  D   |  3   -  D   |  4   |  5   |  6   |  7   |
---------------------------------------------------------------------------------------
| Cycles | 0.0    0.0  | 0.0  | 3.7    0.0  | 3.7    0.0  | 11.0 | 0.0  | 0.0  | 3.7  |
---------------------------------------------------------------------------------------


| Num Of |                    Ports pressure in cycles                     |    |
|  Uops  |  0  - DV  |  1  |  2  -  D  |  3  -  D  |  4  |  5  |  6  |  7  |    |
---------------------------------------------------------------------------------
|   2^   |           |     | 0.3       | 0.3       | 1.0 |     |     | 0.3 | CP | mov qword ptr [r14+0x40], r8
|   2^   |           |     | 0.3       | 0.3       | 1.0 |     |     | 0.3 | CP | mov qword ptr [r14+0x48], r8
|   2^   |           |     | 0.3       | 0.3       | 1.0 |     |     | 0.3 | CP | mov qword ptr [r14+0x50], r8
|   2^   |           |     | 0.3       | 0.3       | 1.0 |     |     | 0.3 | CP | mov qword ptr [r14+0x58], r8
|   2^   |           |     | 0.3       | 0.3       | 1.0 |     |     | 0.3 | CP | mov qword ptr [r14+0x60], r8
|   2^   |           |     | 0.3       | 0.3       | 1.0 |     |     | 0.3 | CP | mov qword ptr [r14+0x68], r8
|   2^   |           |     | 0.3       | 0.3       | 1.0 |     |     | 0.3 | CP | mov qword ptr [r14+0x70], r8
|   2^   |           |     | 0.3       | 0.3       | 1.0 |     |     | 0.3 | CP | mov qword ptr [r14+0x78], r8
|   2^   |           |     | 0.3       | 0.3       | 1.0 |     |     | 0.3 | CP | mov qword ptr [r14+0x80], r8
|   2^   |           |     | 0.3       | 0.3       | 1.0 |     |     | 0.3 | CP | mov qword ptr [r14+0x88], r8
|   2^   |           |     | 0.3       | 0.3       | 1.0 |     |     | 0.3 | CP | movups xmmword ptr [r14], xmm0
Total Num Of Uops: 22
⇨ One μop that can only use port '4'

With blocking instructions for ports {'2', '3', '7'}:

Throughput Analysis Report
--------------------------
Block Throughput: 11.00 Cycles       Throughput Bottleneck: Backend. Port4

Port Binding In Cycles Per Iteration:
---------------------------------------------------------------------------------------
|  Port  |  0   -  DV  |  1   |  2   -  D   |  3   -  D   |  4   |  5   |  6   |  7   |
---------------------------------------------------------------------------------------
| Cycles | 0.0    0.0  | 0.0  | 3.7    0.0  | 3.7    0.0  | 11.0 | 0.0  | 0.0  | 3.7  |
---------------------------------------------------------------------------------------


| Num Of |                    Ports pressure in cycles                     |    |
|  Uops  |  0  - DV  |  1  |  2  -  D  |  3  -  D  |  4  |  5  |  6  |  7  |    |
---------------------------------------------------------------------------------
|   2^   |           |     | 0.3       | 0.3       | 1.0 |     |     | 0.3 | CP | mov qword ptr [r14+0x40], r8
|   2^   |           |     | 0.3       | 0.3       | 1.0 |     |     | 0.3 | CP | mov qword ptr [r14+0x48], r8
|   2^   |           |     | 0.3       | 0.3       | 1.0 |     |     | 0.3 | CP | mov qword ptr [r14+0x50], r8
|   2^   |           |     | 0.3       | 0.3       | 1.0 |     |     | 0.3 | CP | mov qword ptr [r14+0x58], r8
|   2^   |           |     | 0.3       | 0.3       | 1.0 |     |     | 0.3 | CP | mov qword ptr [r14+0x60], r8
|   2^   |           |     | 0.3       | 0.3       | 1.0 |     |     | 0.3 | CP | mov qword ptr [r14+0x68], r8
|   2^   |           |     | 0.3       | 0.3       | 1.0 |     |     | 0.3 | CP | mov qword ptr [r14+0x70], r8
|   2^   |           |     | 0.3       | 0.3       | 1.0 |     |     | 0.3 | CP | mov qword ptr [r14+0x78], r8
|   2^   |           |     | 0.3       | 0.3       | 1.0 |     |     | 0.3 | CP | mov qword ptr [r14+0x80], r8
|   2^   |           |     | 0.3       | 0.3       | 1.0 |     |     | 0.3 | CP | mov qword ptr [r14+0x88], r8
|   2^   |           |     | 0.3       | 0.3       | 1.0 |     |     | 0.3 | CP | movups xmmword ptr [r14], xmm0
Total Num Of Uops: 22
⇨ One μop that can only use ports {'2', '3', '7'}

With an indexed addressing mode


With blocking instructions for port '4':

Throughput Analysis Report
--------------------------
Block Throughput: 11.00 Cycles       Throughput Bottleneck: Backend. Port4

Port Binding In Cycles Per Iteration:
---------------------------------------------------------------------------------------
|  Port  |  0   -  DV  |  1   |  2   -  D   |  3   -  D   |  4   |  5   |  6   |  7   |
---------------------------------------------------------------------------------------
| Cycles | 0.0    0.0  | 0.0  | 3.7    0.0  | 3.7    0.0  | 11.0 | 0.0  | 0.0  | 3.7  |
---------------------------------------------------------------------------------------


| Num Of |                    Ports pressure in cycles                     |    |
|  Uops  |  0  - DV  |  1  |  2  -  D  |  3  -  D  |  4  |  5  |  6  |  7  |    |
---------------------------------------------------------------------------------
|   2^   |           |     |           | 0.3       | 1.0 |     |     | 0.6 | CP | mov qword ptr [r14+0x40], r8
|   2^   |           |     | 0.3       | 0.3       | 1.0 |     |     | 0.3 | CP | mov qword ptr [r14+0x48], r8
|   2^   |           |     | 0.3       | 0.3       | 1.0 |     |     | 0.3 | CP | mov qword ptr [r14+0x50], r8
|   2^   |           |     | 0.3       | 0.3       | 1.0 |     |     | 0.3 | CP | mov qword ptr [r14+0x58], r8
|   2^   |           |     | 0.3       | 0.3       | 1.0 |     |     | 0.3 | CP | mov qword ptr [r14+0x60], r8
|   2^   |           |     | 0.6       | 0.3       | 1.0 |     |     | 0.2 | CP | mov qword ptr [r14+0x68], r8
|   2^   |           |     | 0.2       | 0.3       | 1.0 |     |     | 0.5 | CP | mov qword ptr [r14+0x70], r8
|   2^   |           |     | 0.3       | 0.3       | 1.0 |     |     | 0.3 | CP | mov qword ptr [r14+0x78], r8
|   2^   |           |     | 0.3       | 0.3       | 1.0 |     |     | 0.3 | CP | mov qword ptr [r14+0x80], r8
|   2^   |           |     | 0.3       | 0.3       | 1.0 |     |     | 0.3 | CP | mov qword ptr [r14+0x88], r8
|   2    |           |     | 0.6       | 0.3       | 1.0 |     |     |     | CP | movups xmmword ptr [r14+r13*1], xmm0
Total Num Of Uops: 22
⇨ One μop that can only use port '4'

With blocking instructions for ports {'2', '3'}:

Throughput Analysis Report
--------------------------
Block Throughput: 5.50 Cycles       Throughput Bottleneck: Backend. PORT2_AGU, PORT3_AGU

Port Binding In Cycles Per Iteration:
---------------------------------------------------------------------------------------
|  Port  |  0   -  DV  |  1   |  2   -  D   |  3   -  D   |  4   |  5   |  6   |  7   |
---------------------------------------------------------------------------------------
| Cycles | 0.0    0.0  | 0.0  | 5.5    5.0  | 5.5    5.0  | 1.0  | 0.0  | 0.0  | 0.0  |
---------------------------------------------------------------------------------------


| Num Of |                    Ports pressure in cycles                     |    |
|  Uops  |  0  - DV  |  1  |  2  -  D  |  3  -  D  |  4  |  5  |  6  |  7  |    |
---------------------------------------------------------------------------------
|   1    |           |     | 1.0   1.0 |           |     |     |     |     | CP | lddqu xmm1, xmmword ptr [r14+0x40]
|   1    |           |     |           | 1.0   1.0 |     |     |     |     | CP | lddqu xmm2, xmmword ptr [r14+0x50]
|   1    |           |     | 1.0   1.0 |           |     |     |     |     | CP | lddqu xmm3, xmmword ptr [r14+0x60]
|   1    |           |     |           | 1.0   1.0 |     |     |     |     | CP | lddqu xmm4, xmmword ptr [r14+0x70]
|   1    |           |     | 1.0   1.0 |           |     |     |     |     | CP | lddqu xmm5, xmmword ptr [r14+0x80]
|   1    |           |     |           | 1.0   1.0 |     |     |     |     | CP | lddqu xmm6, xmmword ptr [r14+0x90]
|   1    |           |     | 1.0   1.0 |           |     |     |     |     | CP | lddqu xmm7, xmmword ptr [r14+0xa0]
|   1    |           |     |           | 1.0   1.0 |     |     |     |     | CP | lddqu xmm8, xmmword ptr [r14+0xb0]
|   1    |           |     | 1.0   1.0 |           |     |     |     |     | CP | lddqu xmm9, xmmword ptr [r14+0xc0]
|   1    |           |     |           | 1.0   1.0 |     |     |     |     | CP | lddqu xmm10, xmmword ptr [r14+0xd0]
|   2    |           |     | 0.5       | 0.5       | 1.0 |     |     |     | CP | movups xmmword ptr [r14+r13*1], xmm0
Total Num Of Uops: 12
⇨ One μop that can only use ports {'2', '3'}