VPERMQ (YMM, M256, I8) - Throughput and Uops (IACA 3.0)
With a non-indexed addressing mode
With 1 independent instruction
Throughput Analysis Report
--------------------------
Block Throughput: 1.00 Cycles Throughput Bottleneck: Dependency chains
Loop Count: 48
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 | 0.5 0.5 | 0.5 0.5 | 0.0 | 1.0 | 0.0 | 0.0 |
--------------------------------------------------------------------------------------------------
| Num Of | Ports pressure in cycles | |
| Uops | 0 - DV | 1 | 2 - D | 3 - D | 4 | 5 | 6 | 7 |
-----------------------------------------------------------------------------------------
| 2 | | | 0.5 0.5 | 0.5 0.5 | | 1.0 | | | vpermq ymm0, ymmword ptr [r14], 0x2
Total Num Of Uops: 2
With 13 independent instructions
Throughput Analysis Report
--------------------------
Block Throughput: 13.00 Cycles Throughput Bottleneck: Backend
Loop Count: 22
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 | 6.5 6.5 | 6.5 6.5 | 0.0 | 13.0 | 0.0 | 0.0 |
--------------------------------------------------------------------------------------------------
| Num Of | Ports pressure in cycles | |
| Uops | 0 - DV | 1 | 2 - D | 3 - D | 4 | 5 | 6 | 7 |
-----------------------------------------------------------------------------------------
| 2 | | | 0.5 0.5 | 0.5 0.5 | | 1.0 | | | vpermq ymm0, ymmword ptr [r14], 0x2
| 2 | | | 0.5 0.5 | 0.5 0.5 | | 1.0 | | | vpermq ymm1, ymmword ptr [r14+0x20], 0x2
| 2 | | | 0.5 0.5 | 0.5 0.5 | | 1.0 | | | vpermq ymm2, ymmword ptr [r14+0x40], 0x2
| 2 | | | 0.5 0.5 | 0.5 0.5 | | 1.0 | | | vpermq ymm3, ymmword ptr [r14+0x60], 0x2
| 2 | | | 0.5 0.5 | 0.5 0.5 | | 1.0 | | | vpermq ymm4, ymmword ptr [r14+0x80], 0x2
| 2 | | | 0.5 0.5 | 0.5 0.5 | | 1.0 | | | vpermq ymm5, ymmword ptr [r14+0xa0], 0x2
| 2 | | | 0.5 0.5 | 0.5 0.5 | | 1.0 | | | vpermq ymm6, ymmword ptr [r14+0xc0], 0x2
| 2 | | | 0.5 0.5 | 0.5 0.5 | | 1.0 | | | vpermq ymm7, ymmword ptr [r14+0xe0], 0x2
| 2 | | | 0.5 0.5 | 0.5 0.5 | | 1.0 | | | vpermq ymm8, ymmword ptr [r14+0x100], 0x2
| 2 | | | 0.5 0.5 | 0.5 0.5 | | 1.0 | | | vpermq ymm9, ymmword ptr [r14+0x120], 0x2
| 2 | | | 0.5 0.5 | 0.5 0.5 | | 1.0 | | | vpermq ymm10, ymmword ptr [r14+0x140], 0x2
| 2 | | | 0.5 0.5 | 0.5 0.5 | | 1.0 | | | vpermq ymm11, ymmword ptr [r14+0x160], 0x2
| 2 | | | 0.5 0.5 | 0.5 0.5 | | 1.0 | | | vpermq ymm12, ymmword ptr [r14+0x180], 0x2
Total Num Of Uops: 26
With an indexed addressing mode
With 1 independent instruction
Throughput Analysis Report
--------------------------
Block Throughput: 1.00 Cycles Throughput Bottleneck: Dependency chains
Loop Count: 48
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 | 0.5 0.5 | 0.5 0.5 | 0.0 | 1.0 | 0.0 | 0.0 |
--------------------------------------------------------------------------------------------------
| Num Of | Ports pressure in cycles | |
| Uops | 0 - DV | 1 | 2 - D | 3 - D | 4 | 5 | 6 | 7 |
-----------------------------------------------------------------------------------------
| 2 | | | 0.5 0.5 | 0.5 0.5 | | 1.0 | | | vpermq ymm0, ymmword ptr [r14+r13*1], 0x2
Total Num Of Uops: 2
With 13 independent instructions
Throughput Analysis Report
--------------------------
Block Throughput: 13.00 Cycles Throughput Bottleneck: Backend
Loop Count: 22
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 | 6.5 6.5 | 6.5 6.5 | 0.0 | 13.0 | 0.0 | 0.0 |
--------------------------------------------------------------------------------------------------
| Num Of | Ports pressure in cycles | |
| Uops | 0 - DV | 1 | 2 - D | 3 - D | 4 | 5 | 6 | 7 |
-----------------------------------------------------------------------------------------
| 2 | | | 0.5 0.5 | 0.5 0.5 | | 1.0 | | | vpermq ymm0, ymmword ptr [r14+r13*1], 0x2
| 2 | | | 0.5 0.5 | 0.5 0.5 | | 1.0 | | | vpermq ymm1, ymmword ptr [r14+r13*1+0x20], 0x2
| 2 | | | 0.5 0.5 | 0.5 0.5 | | 1.0 | | | vpermq ymm2, ymmword ptr [r14+r13*1+0x40], 0x2
| 2 | | | 0.5 0.5 | 0.5 0.5 | | 1.0 | | | vpermq ymm3, ymmword ptr [r14+r13*1+0x60], 0x2
| 2 | | | 0.5 0.5 | 0.5 0.5 | | 1.0 | | | vpermq ymm4, ymmword ptr [r14+r13*1+0x80], 0x2
| 2 | | | 0.5 0.5 | 0.5 0.5 | | 1.0 | | | vpermq ymm5, ymmword ptr [r14+r13*1+0xa0], 0x2
| 2 | | | 0.5 0.5 | 0.5 0.5 | | 1.0 | | | vpermq ymm6, ymmword ptr [r14+r13*1+0xc0], 0x2
| 2 | | | 0.5 0.5 | 0.5 0.5 | | 1.0 | | | vpermq ymm7, ymmword ptr [r14+r13*1+0xe0], 0x2
| 2 | | | 0.5 0.5 | 0.5 0.5 | | 1.0 | | | vpermq ymm8, ymmword ptr [r14+r13*1+0x100], 0x2
| 2 | | | 0.5 0.5 | 0.5 0.5 | | 1.0 | | | vpermq ymm9, ymmword ptr [r14+r13*1+0x120], 0x2
| 2 | | | 0.5 0.5 | 0.5 0.5 | | 1.0 | | | vpermq ymm10, ymmword ptr [r14+r13*1+0x140], 0x2
| 2 | | | 0.5 0.5 | 0.5 0.5 | | 1.0 | | | vpermq ymm11, ymmword ptr [r14+r13*1+0x160], 0x2
| 2 | | | 0.5 0.5 | 0.5 0.5 | | 1.0 | | | vpermq ymm12, ymmword ptr [r14+r13*1+0x180], 0x2
Total Num Of Uops: 26