VBLENDVPS (YMM, YMM, YMM, YMM) - Port Usage (IACA 2.3)


With blocking instructions for ports {'0', '1'}:

Throughput Analysis Report
--------------------------
Block Throughput: 5.00 Cycles       Throughput Bottleneck: Backend. Port0, Port1

Port Binding In Cycles Per Iteration:
---------------------------------------------------------------------------------------
|  Port  |  0   -  DV  |  1   |  2   -  D   |  3   -  D   |  4   |  5   |  6   |  7   |
---------------------------------------------------------------------------------------
| Cycles | 5.0    0.0  | 5.0  | 0.0    0.0  | 0.0    0.0  | 0.0  | 2.0  | 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       |     |           |           |     |     |     |     | CP | vaddpd xmm4, xmm5, xmm6
|   1    |           | 1.0 |           |           |     |     |     |     | CP | vaddpd xmm7, xmm5, xmm6
|   1    | 1.0       |     |           |           |     |     |     |     | CP | vaddpd xmm8, xmm5, xmm6
|   1    |           | 1.0 |           |           |     |     |     |     | CP | vaddpd xmm9, xmm5, xmm6
|   1    | 1.0       |     |           |           |     |     |     |     | CP | vaddpd xmm10, xmm5, xmm6
|   1    |           | 1.0 |           |           |     |     |     |     | CP | vaddpd xmm11, xmm5, xmm6
|   1    | 1.0       |     |           |           |     |     |     |     | CP | vaddpd xmm12, xmm5, xmm6
|   1    |           | 1.0 |           |           |     |     |     |     | CP | vaddpd xmm4, xmm5, xmm6
|   1    | 1.0       |     |           |           |     |     |     |     | CP | vaddpd xmm7, xmm5, xmm6
|   1    |           | 1.0 |           |           |     |     |     |     | CP | vaddpd xmm8, xmm5, xmm6
|   2    |           |     |           |           |     | 2.0 |     |     |    | vblendvps ymm0, ymm1, ymm2, ymm3
Total Num Of Uops: 12

With blocking instructions for ports {'0', '5'}:

Throughput Analysis Report
--------------------------
Block Throughput: 5.00 Cycles       Throughput Bottleneck: Backend. Port0, Port5

Port Binding In Cycles Per Iteration:
---------------------------------------------------------------------------------------
|  Port  |  0   -  DV  |  1   |  2   -  D   |  3   -  D   |  4   |  5   |  6   |  7   |
---------------------------------------------------------------------------------------
| Cycles | 5.0    0.0  | 2.0  | 0.0    0.0  | 0.0    0.0  | 0.0  | 5.0  | 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       |     |           |           |     |     |     |     | CP | movq mmx0, mmx1
|   1    |           |     |           |           |     | 1.0 |     |     | CP | movq mmx2, mmx1
|   1    | 1.0       |     |           |           |     |     |     |     | CP | movq mmx3, mmx1
|   1    |           |     |           |           |     | 1.0 |     |     | CP | movq mmx4, mmx1
|   1    | 1.0       |     |           |           |     |     |     |     | CP | movq mmx5, mmx1
|   1    |           |     |           |           |     | 1.0 |     |     | CP | movq mmx6, mmx1
|   1    | 1.0       |     |           |           |     |     |     |     | CP | movq mmx7, mmx1
|   1    |           |     |           |           |     | 1.0 |     |     | CP | movq mmx0, mmx1
|   1    | 1.0       |     |           |           |     |     |     |     | CP | movq mmx2, mmx1
|   1    |           |     |           |           |     | 1.0 |     |     | CP | movq mmx3, mmx1
|   2    |           | 2.0 |           |           |     |     |     |     |    | vblendvps ymm0, ymm1, ymm2, ymm3
Total Num Of Uops: 12

With blocking instructions for ports {'1', '5'}:

Throughput Analysis Report
--------------------------
Block Throughput: 5.00 Cycles       Throughput Bottleneck: Backend. Port1, Port5

Port Binding In Cycles Per Iteration:
---------------------------------------------------------------------------------------
|  Port  |  0   -  DV  |  1   |  2   -  D   |  3   -  D   |  4   |  5   |  6   |  7   |
---------------------------------------------------------------------------------------
| Cycles | 2.0    0.0  | 5.0  | 0.0    0.0  | 0.0    0.0  | 0.0  | 5.0  | 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 |           |           |     |     |     |     | CP | lea r8w, ptr [r14]
|   1    |           |     |           |           |     | 1.0 |     |     | CP | lea r9w, ptr [r14]
|   1    |           | 1.0 |           |           |     |     |     |     | CP | lea r10w, ptr [r14]
|   1    |           |     |           |           |     | 1.0 |     |     | CP | lea r11w, ptr [r14]
|   1    |           | 1.0 |           |           |     |     |     |     | CP | lea r12w, ptr [r14]
|   1    |           |     |           |           |     | 1.0 |     |     | CP | lea bx, ptr [r14]
|   1    |           | 1.0 |           |           |     |     |     |     | CP | lea cx, ptr [r14]
|   1    |           |     |           |           |     | 1.0 |     |     | CP | lea dx, ptr [r14]
|   1    |           | 1.0 |           |           |     |     |     |     | CP | lea r8w, ptr [r14]
|   1    |           |     |           |           |     | 1.0 |     |     | CP | lea r9w, ptr [r14]
|   2    | 2.0       |     |           |           |     |     |     |     |    | vblendvps ymm0, ymm1, ymm2, ymm3
Total Num Of Uops: 12

With blocking instructions for ports {'0', '1', '5'}:

Throughput Analysis Report
--------------------------
Block Throughput: 4.00 Cycles       Throughput Bottleneck: Backend. Port0, Port1, Port5

Port Binding In Cycles Per Iteration:
---------------------------------------------------------------------------------------
|  Port  |  0   -  DV  |  1   |  2   -  D   |  3   -  D   |  4   |  5   |  6   |  7   |
---------------------------------------------------------------------------------------
| Cycles | 4.0    0.0  | 4.0  | 0.0    0.0  | 0.0    0.0  | 0.0  | 4.0  | 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       |     |           |           |     |     |     |     | CP | vcvtdq2ps xmm4, xmm5
|   1    |           | 1.0 |           |           |     |     |     |     | CP | vcvtdq2ps xmm6, xmm5
|   1    |           |     |           |           |     | 1.0 |     |     | CP | vcvtdq2ps xmm7, xmm5
|   1    | 1.0       |     |           |           |     |     |     |     | CP | vcvtdq2ps xmm8, xmm5
|   1    |           | 1.0 |           |           |     |     |     |     | CP | vcvtdq2ps xmm9, xmm5
|   1    |           |     |           |           |     | 1.0 |     |     | CP | vcvtdq2ps xmm10, xmm5
|   1    | 1.0       |     |           |           |     |     |     |     | CP | vcvtdq2ps xmm11, xmm5
|   1    |           | 1.0 |           |           |     |     |     |     | CP | vcvtdq2ps xmm12, xmm5
|   1    |           |     |           |           |     | 1.0 |     |     | CP | vcvtdq2ps xmm4, xmm5
|   1    | 1.0       |     |           |           |     |     |     |     | CP | vcvtdq2ps xmm6, xmm5
|   2    |           | 1.0 |           |           |     | 1.0 |     |     | CP | vblendvps ymm0, ymm1, ymm2, ymm3
Total Num Of Uops: 12
⇨ 2 μops that can only use ports {'0', '1', '5'}