qaas_compilers.csv
| app_name | compiler | option # | flags | #MPI | #OMP | time(s) | GFlops/s | Spd w.r.t orig | Spd w.r.t gcc |
|---|
| amg | orig | 0 | default | 1 | 64 | 45.77 | 0 | 1 | 1 |
| amg | gcc | 0 | default | 1 | 64 | 45.57 | 0 | 1.01 | 1 |
| amg | armclang | 1 | -O2 -mcpu=native -armpl | 1 | 64 | 46.14 | 0 | 1 | 0.99 |
| amg | armclang | 2 | -O2 -mcpu=native -fno-vectorize -armpl | 1 | 64 | 45.8 | 0 | 1 | 1 |
| amg | armclang | 3 | -O3 -mcpu=native -armpl | 1 | 64 | 45.74 | 0 | 1.01 | 1 |
| amg | armclang | 4 | -O3 -mcpu=native -fno-vectorize -armpl | 1 | 64 | 45.72 | 0 | 1.01 | 1 |
| amg | armclang | 5 | -Ofast -mcpu=native -armpl | 1 | 64 | 45.82 | 0 | 1 | 1 |
| amg | armclang | 6 | -Ofast -mcpu=native -fno-vectorize -armpl | 1 | 64 | 45.92 | 0 | 1 | 1 |
| amg | gcc | 1 | -O2 -mcpu=native -funroll-loops -larmpl | 1 | 64 | 46.39 | 0 | 0.99 | 0.99 |
| amg | gcc | 2 | -O2 -mcpu=native -fno-tree-vectorize -fno-openmp-simd -funroll-loops -larmpl | 1 | 64 | 46.64 | 0 | 0.99 | 0.98 |
| amg | gcc | 3 | -O3 -mcpu=native -funroll-loops -larmpl | 1 | 64 | 45.47 | 0 | 1.01 | 1.01 |
| amg | gcc | 4 | -O3 -mcpu=native -fno-tree-vectorize -fno-openmp-simd -funroll-loops -larmpl | 1 | 64 | 45.38 | 0 | 1.01 | 1.01 |
| amg | gcc | 5 | -Ofast -mcpu=native -funroll-loops -larmpl | 1 | 64 | 45.38 | 0 | 1.01 | 1.01 |
| amg | gcc | 6 | -Ofast -mcpu=native -fno-tree-vectorize -fno-openmp-simd -funroll-loops -larmpl | 1 | 64 | 45.48 | 0 | 1.01 | 1.01 |
