8.5.2.5 : The compilation
Let's compile :
|
$ make -- Configuring done -- Generating done -- Build files have been written to: ExampleOptimisation/build [ 2%] Built target hadamard_product_O2 [ 4%] Built target hadamard_product_O1 [ 6%] Built target hadamard_product_vectorize [ 11%] Built target hadamard_product_O0 [ 15%] Built target hadamard_product_O3 [ 18%] Built target hadamard_product_Ofast [ 20%] Built target hadamard_product_intrinsics [ 22%] Built target asterics_hpc [ 25%] Built target saxpy_O2 [ 27%] Built target saxpy_O0 [ 31%] Built target saxpy_O3 [ 34%] Built target saxpy_O1 [ 36%] Built target saxpy_Ofast [ 40%] Built target saxpy_vectorize [ 43%] Built target saxpy_intrinsics [ 47%] Built target reduction_real_intrinsics_O3 [ 52%] Built target reduction_real_intrinsics_interleave2_O3 [ 56%] Built target reduction_real_vectorize_Ofast Scanning dependencies of target reduction_real_intrinsics_interleave4_O3 [ 56%] Building CXX object 4-Reduction/CMakeFiles/reduction_real_intrinsics_interleave4_O3.dir/reduction_intrinsics_interleave4.cpp.o [ 59%] Building CXX object 4-Reduction/CMakeFiles/reduction_real_intrinsics_interleave4_O3.dir/main_intrinsics_interleave4.cpp.o [ 61%] Linking CXX executable reduction_real_intrinsics_interleave4_O3 [ 61%] Built target reduction_real_intrinsics_interleave4_O3 [ 65%] Built target reduction_real_vectorize_O3 [ 70%] Built target reduction_real_O0 [ 72%] Built target reduction_O1 [ 77%] Built target reduction_O0 [ 81%] Built target reduction_real_Ofast [ 84%] Built target reduction_O3 [ 88%] Built target reduction_real_O3 [ 93%] Built target reduction_real_O1 [ 97%] Built target reduction_real_O2 [100%] Built target reduction_O2 |
Let's get the performances :
|
$ make plot_all [ 1%] Built target asterics_hpc [ 3%] Built target reduction_real_O3 [ 6%] Built target reduction_real_vectorize_Ofast [ 8%] Built target reduction_real_vectorize_O3 [ 10%] Built target reduction_real_Ofast [ 14%] Built target plot_reductionVectorize [ 15%] Built target hadamard_product_intrinsics [ 17%] Built target hadamard_product_vectorize [ 19%] Built target hadamard_product_O3 [ 23%] Built target plot_hadamardIntrinsics [ 24%] Built target hadamard_product_Ofast [ 25%] Built target hadamard_product_O2 [ 26%] Built target hadamard_product_O1 [ 29%] Built target hadamard_product_O0 [ 34%] Built target plot_hadamardBase [ 36%] Built target plot_hadamardVectorize [ 37%] Built target saxpy_intrinsics [ 40%] Built target saxpy_O3 [ 42%] Built target saxpy_vectorize [ 45%] Built target plot_saxpyIntrinsics [ 47%] Built target plot_saxpyVectorize [ 48%] Built target saxpy_Ofast [ 50%] Built target saxpy_O2 [ 51%] Built target saxpy_O0 [ 52%] Built target saxpy_O1 [ 57%] Built target plot_saxpyBase [ 59%] Built target reduction_real_intrinsics_O3 [ 64%] Built target plot_reductionIntrinsics [ 67%] Built target reduction_real_intrinsics_interleave2_O3 [ 69%] Built target reduction_real_intrinsics_interleave4_O3 Scanning dependencies of target plot_reductionIntrinsicsInterleave4 [ 70%] Run reduction_real_intrinsics_interleave4_O3 program Reduction intrinsics interleave 4 evaluateReduction : nbElement = 1024, cyclePerElement = 0.108398 cy/el, elapsedTime = 111 cy evaluateReduction : nbElement = 2048, cyclePerElement = 0.100586 cy/el, elapsedTime = 206 cy evaluateReduction : nbElement = 3072, cyclePerElement = 0.100586 cy/el, elapsedTime = 309 cy evaluateReduction : nbElement = 5024, cyclePerElement = 0.0971338 cy/el, elapsedTime = 488 cy evaluateReduction : nbElement = 10048, cyclePerElement = 0.108579 cy/el, elapsedTime = 1091 cy [ 71%] Call gnuplot reductionIntrinsicsInterleave4 [ 74%] Built target plot_reductionIntrinsicsInterleave4 [ 78%] Built target plot_reductionIntrinsicsInterleave2 [ 80%] Built target reduction_real_O2 [ 82%] Built target reduction_real_O0 [ 85%] Built target reduction_real_O1 [ 90%] Built target plot_reductionReal [ 91%] Built target reduction_O2 [ 92%] Built target reduction_O1 [ 95%] Built target reduction_O0 [ 96%] Built target reduction_O3 [100%] Built target plot_reductionBase [100%] Built target plot_all |