adds diamond tiling test

[lukastruemper]

No description provided.

[daisytuner]

Daisytuner Report - mlir_torch (chamomile)

@@                                   Benchmarks                                   @@
=====================================================================================
  Benchmark              Time        ΔTime       Thr         Energy      ΔEnergy     
=====================================================================================
# linear_torch           6.17 s      N/A         N/A         1492.30 J   N/A         
# linear_none            15.80 s     N/A         N/A         3942.12 J   N/A         
# linear_sequential      15.80 s     N/A         N/A         3863.74 J   N/A         
# linear_openmp          15.76 s     N/A         N/A         3856.71 J   N/A         
# linear_cuda            13.42 s     N/A         N/A         2578.72 J   N/A         
# matmul_torch           5.98 s      N/A         N/A         1442.44 J   N/A         
# matmul_none            10.65 s     N/A         N/A         2826.05 J   N/A         
# matmul_sequential      10.44 s     N/A         N/A         2809.04 J   N/A         
# matmul_openmp          10.56 s     N/A         N/A         2840.21 J   N/A         
# matmul_cuda            10.23 s     N/A         N/A         1963.24 J   N/A         

[daisytuner]

Daisytuner Report - python_npbench (zinnia)

@@                                   Benchmarks                                   @@
=====================================================================================
  Benchmark              Time        ΔTime       Thr         Energy      ΔEnergy     
=====================================================================================
# adi_numpy              1.32 s      -0.31%      N/A         131.88 J    -0.34%      
# adi_omp                15.05 s     -0.72%      N/A         1496.32 J   +0.39%      
# adi_cuda               4.73 s      +0.21%      N/A         458.46 J    +0.19%      
# adi_seq_tuning         15.98 s     -0.49%      N/A         1508.49 J   -0.49%      
# atax_numpy             2.16 s      -0.19%      N/A         223.56 J    -0.13%      
# atax_omp               2.97 s      -2.30%      N/A         374.23 J    -3.48%      
# atax_cuda              4.12 s      +0.20%      N/A         424.86 J    +0.23%      
# atax_seq_tuning        3.72 s      -0.14%      N/A         376.84 J    +0.03%      
# gemm_numpy             1.21 s      -0.62%      N/A         193.06 J    -0.60%      
# gemm_omp               1.11 s      +0.05%      N/A         162.52 J    +0.06%      
# gemm_cuda              10.58 s     -0.45%      N/A         1005.79 J   -0.49%      
# gemm_seq_tuning        1.11 s      -0.34%      N/A         161.67 J    -0.49%      
# gesummv_numpy          1.76 s      +0.38%      N/A         251.34 J    +0.31%      
# gesummv_omp            1.98 s      +1.31%      N/A         313.02 J    +0.24%      
# gesummv_cuda           8.31 s      -0.39%      N/A         1000.57 J   -0.18%      
# gesummv_seq_tuning     6.65 s      -0.16%      N/A         812.91 J    -0.12%      
# gemver_numpy           1.10 s      +0.71%      N/A         168.37 J    +0.92%      
# gemver_omp             848.23 ms   -1.56%      N/A         109.38 J    -2.50%      
# gemver_cuda            3.88 s      -0.79%      N/A         388.06 J    -0.79%      
# gemver_seq_tuning      4.51 s      +0.52%      N/A         436.81 J    +0.60%      
# k2mm_numpy             1.20 s      +0.22%      N/A         196.51 J    +0.33%      
# k2mm_omp               3.54 s      +0.10%      N/A         660.46 J    -1.27%      
# k2mm_cuda              13.59 s     +0.02%      N/A         1294.20 J   +0.40%      
# k2mm_seq_tuning        3.62 s      +0.14%      N/A         466.40 J    +0.03%      
# k3mm_numpy             1.03 s      -0.20%      N/A         181.63 J    -0.23%      
# k3mm_omp               5.55 s      -0.21%      N/A         949.26 J    -1.46%      
# k3mm_cuda              19.83 s     -0.41%      N/A         1870.93 J   -0.45%      
# k3mm_seq_tuning        5.73 s      -0.09%      N/A         792.18 J    -0.11%      
# mvt_numpy              2.42 s      -0.43%      N/A         248.54 J    -0.32%      
# mvt_omp                2.75 s      +0.18%      N/A         285.28 J    +0.13%      
# mvt_cuda               3.36 s      +0.39%      N/A         343.04 J    +0.34%      
# mvt_seq_tuning         2.75 s      -0.38%      N/A         285.40 J    -0.27%      
# symm_numpy             785.66 ms   -1.03%      N/A         81.05 J     -0.73%      
# symm_omp               6.01 s      -1.63%      N/A         589.82 J    -1.18%      
# symm_seq_tuning        8.50 s      +0.31%      N/A         810.07 J    +0.30%      
# syr2k_numpy            886.06 ms   -1.46%      N/A         90.00 J     -1.46%      
# syr2k_omp              9.89 s      -0.29%      N/A         940.62 J    -0.21%      
# syr2k_cuda             1.62 s      -1.06%      N/A         168.74 J    -1.06%      
# syr2k_seq_tuning       9.85 s      -0.17%      N/A         936.89 J    -0.15%      
# syrk_numpy             782.94 ms   +1.02%      N/A         80.44 J     +0.80%      
# syrk_omp               6.01 s      +0.30%      N/A         578.23 J    +0.29%      
# syrk_cuda              1.50 s      -1.65%      N/A         157.61 J    -1.44%      
# syrk_seq_tuning        5.98 s      +0.55%      N/A         574.92 J    +0.43%      
# trmm_numpy             876.78 ms   +0.01%      N/A         89.33 J     -0.30%      
# trmm_omp               728.67 ms   +3.51%      N/A         92.80 J     +3.97%      
# trmm_seq_tuning        3.42 s      +1.02%      N/A         324.73 J    -0.01%      

[daisytuner]

Daisytuner Report - mlir_torch_models (chamomile)

@@                                   Benchmarks                                   @@
=====================================================================================
  Benchmark              Time        ΔTime       Thr         Energy      ΔEnergy     
=====================================================================================
# bn_conv_bn_relu_maxpool_torch18.59 s     -0.97%      N/A         3580.48 J   -2.46%      
# bn_conv_bn_relu_maxpool_run_none3.27 s      -0.06%      N/A         657.08 J    -1.12%      
# bn_conv_bn_relu_maxpool_run_sequential3.28 s      -0.52%      N/A         659.17 J    -2.19%      
# bn_conv_bn_relu_maxpool_run_openmp3.36 s      +1.50%      N/A         693.47 J    +2.09%      
# bn_conv_bn_relu_maxpool_run_cuda3.71 s      -0.25%      N/A         724.40 J    -1.64%      

[daisytuner]

Daisytuner Report - mlir_torch_layers (chamomile)

@@                                   Benchmarks                                   @@
=====================================================================================
  Benchmark              Time        ΔTime       Thr         Energy      ΔEnergy     
=====================================================================================
# batchnorm_torch        19.01 s     -1.56%      N/A         3724.81 J   +2.94%      
# batchnorm_run_none     6.22 s      -0.00%      N/A         1210.91 J   +4.77%      
# batchnorm_run_sequential6.65 s      +1.76%      N/A         1292.29 J   +6.43%      
# batchnorm_run_openmp   5.74 s      -1.09%      N/A         1353.36 J   +3.36%      
# batchnorm_run_cuda     8.05 s      -0.42%      N/A         1577.89 J   +4.76%      
# conv2d_torch           18.58 s     -0.05%      N/A         3643.22 J   +4.52%      
# conv2d_run_openmp      5.02 s      +0.56%      N/A         1216.13 J   +4.82%      
# conv2d_run_cuda        6.84 s      +0.28%      N/A         1342.97 J   +5.56%      
# linear_torch           6.13 s      +0.67%      N/A         1475.96 J   +3.84%      
# linear_run_none        11.53 s     -0.90%      N/A         3135.96 J   +1.24%      
# linear_run_sequential  9.99 s      +1.16%      N/A         2778.56 J   +4.71%      
# linear_run_openmp      9.73 s      -0.36%      N/A         2868.10 J   +2.66%      
# linear_run_cuda        9.20 s      +0.12%      N/A         1791.89 J   +5.03%      
# matmul_torch           6.12 s      +1.01%      N/A         1477.66 J   +4.05%      
# matmul_run_none        11.56 s     -0.38%      N/A         3147.59 J   +1.87%      
# matmul_run_sequential  9.87 s      -0.73%      N/A         2748.98 J   +2.79%      
# matmul_run_openmp      9.69 s      +0.16%      N/A         2857.66 J   +4.04%      
# matmul_run_cuda        9.21 s      +1.58%      N/A         1792.57 J   +6.27%      
# pooling_torch          26.14 s     +1.14%      N/A         5211.23 J   +5.78%      
# pooling_run_none       25.27 s     -0.25%      N/A         4849.27 J   +4.48%      
# pooling_run_sequential 25.68 s     -0.82%      N/A         4918.95 J   +3.71%      
# pooling_run_openmp     17.12 s     -1.17%      N/A         3657.86 J   +3.28%      
# pooling_run_cuda       31.15 s     -0.25%      N/A         6061.07 J   +4.67%      
# relu_torch             18.93 s     -0.07%      N/A         3716.56 J   +4.65%      
# relu_run_none          5.25 s      +0.22%      N/A         1021.44 J   +4.67%      
# relu_run_sequential    6.35 s      +0.31%      N/A         1232.58 J   +4.72%      
# relu_run_openmp        5.70 s      +0.31%      N/A         1344.55 J   +4.45%      
# relu_run_cuda          8.28 s      +0.10%      N/A         1625.79 J   +5.31%