- r_1 - engine_NEON1M11-0001_o52_m1 - 10 analyzed loop(s)
- r_2 - engine_NEON1M11-0001_o52_m1_ifort - 10 analyzed loop(s)
| Analysis | Count | Percentage | Weighted Count |
|---|
| ▼Loop Computation Issues– | 15 | | |
| ○Less than 10% of the FP ADD/SUB/MUL arithmetic operations are performed using FMA | 11 | 55.00 | 0.98 |
| ○Presence of a large number of scalar integer instructions | 4 | 20.00 | 0.70 |
| ▼Control Flow Issues– | 8 | | |
| ○Presence of calls | 4 | 20.00 | 0.22 |
| ○Presence of 2 to 4 paths | 2 | 10.00 | 0.06 |
| ○Presence of more than 4 paths | 2 | 10.00 | 0.04 |
| ▼Data Access Issues– | 19 | | |
| ○More than 20% of the loads are accessing the stack | 5 | 25.00 | 0.16 |
| ○Presence of indirect access | 4 | 20.00 | 0.10 |
| ○Presence of special instructions executing on a single port | 3 | 15.00 | 0.07 |
| ○Presence of constant non-unit stride data access | 3 | 15.00 | 0.07 |
| ○More than 10% of the vector loads instructions are unaligned | 3 | 15.00 | 0.07 |
| ○Presence of expensive instructions: scatter/gather | 1 | 5.00 | 0.03 |
| ▼Vectorization Roadblocks– | 17 | | |
| ○Presence of calls | 4 | 20.00 | 0.22 |
| ○Presence of indirect access | 4 | 20.00 | 0.10 |
| ○Presence of more than 4 paths | 4 | 20.00 | 0.22 |
| ○Presence of constant non-unit stride data access | 3 | 15.00 | 0.07 |
| ○Presence of 2 to 4 paths | 2 | 10.00 | 0.06 |
| ▼Inefficient Vectorization– | 4 | | |
| ○Presence of special instructions executing on a single port | 3 | 15.00 | 0.07 |
| ○Presence of expensive instructions: scatter/gather | 1 | 5.00 | 0.03 |
| Analysis | r_1 | r_2 |
|---|
| Loop Computation Issues | Less than 10% of the FP ADD/SUB/MUL arithmetic operations are performed using FMA | 6 | 5 |
|---|
| Presence of a large number of scalar integer instructions | 1 | 3 |
| Control Flow Issues | Presence of calls | 2 | 2 |
|---|
| Presence of 2 to 4 paths | 1 | 1 |
| Presence of more than 4 paths | 1 | 1 |
| Data Access Issues | Presence of constant non-unit stride data access | 1 | 2 |
|---|
| Presence of indirect access | 2 | 2 |
| More than 10% of the vector loads instructions are unaligned | 3 | 0 |
| Presence of expensive instructions: scatter/gather | 0 | 1 |
| Presence of special instructions executing on a single port | 3 | 0 |
| More than 20% of the loads are accessing the stack | 2 | 3 |
| Vectorization Roadblocks | Presence of calls | 2 | 2 |
|---|
| Presence of 2 to 4 paths | 1 | 1 |
| Presence of more than 4 paths | 2 | 2 |
| Presence of constant non-unit stride data access | 1 | 2 |
| Presence of indirect access | 2 | 2 |
| Inefficient Vectorization | Presence of expensive instructions: scatter/gather | 0 | 1 |
|---|
| Presence of special instructions executing on a single port | 3 | 0 |
