Selector

Repository: - CTuning archive repository CTuning experiments CTuning experiments - remote repository at cmind CTuning experiments - remote repository at unidapt CTuning experiments - work remote repository at cmind CTuning repository CTuning setup CTuning shared CTuning tmp CTuning with unexpected behavior CTuning with validated behavior Cm-experiments-201408xx-fgg-motivation-compiler-flag-tuning-mess-1 Cm-experiments-201408xx-fgg-motivation-compiler-flag-tuning-mess-2 Cm-experiments-201408xx-fgg-motivation-compiler-flag-tuning-mess-3 Default (main) repository Demo Remote at c-mind.org Remote at c-mind.org - shared Repository for tests Repository for tmp data Shared repository Shared repository - extended User repository - Grigori Fursin Module: - Collective Tuning exploration space Collective Tuning scenario - compiler optimizations Collective Tuning scenario - extract program static MILEPOST features Collective Tuning scenario - universal benchmarking Collective Tuning scenario - universal exploration Use multiple graphs: Points in space: ludcmp-icc-12.1-intel-core2-centrino-0002-2100-pad0 ludcmp-icc-12.1-intel-core2-centrino-2046-2048-0.8GHz-pad0 ludcmp-icc-12.1-intel-core2-centrino-2046-2048-2.2GHz-pad0 ludcmp-icc-12.1-intel-core2x-centrino-0002-2100-pad0 ludcmp-icc-12.1-intel-i5-dataset-0002-4096-pad0 ludcmp-icc-12.1-intel-i5-dataset-0508-0514-pad0 ludcmp-icc-12.1-intel-i5-dataset-0508-0514-pad0-low-freq ludcmp-icc-12.1-intel-i5-dataset-0508-0514-pad1 susan-c-icc-12.1-intel-dist-mix-30 susan-c-icc-12.1-intel-dist-perf-30 susan-c-icc-12.1-intel-dist-perf-30-in-one susan-c-icc-12.1-intel-dist-power-30 Graph engine: Google Web Services: 2D Column chart Google Web Services: 2D Scatter chart Python MatPlotLib: 1D Density graph Python MatPlotLib: 2D Column chart Python MatPlotLib: 2D Scatter chart Use raw dimensions: Convert text axes to float: X axis flattened key: Accelerator compiler Additional compiler flags Alchemist plugin Alchemist sub-plugin Benchmark internal execution ID Benchmark score Build exit code Build script name Build script name Build script type Build script type Calibrate Calibrate to time (sec.) Calibrate until max iterations cJSON library Cleaning script name collective tuning setup Command line Compilation type Compilation type Compilation type of linked libraries Compiler compiler flag: /O3 compiler flag: /Qunroll compiler flag: openmp cTuning accelerator compiler description cTuning CC Compiler cTuning CC Plugins cTuning compiler description CVM P value Dataset name Dataset UID Dataset UOA Date/time finished Date/time started Enter (existing) cTuning space UOA Extract after this pass in MILEPOST GCC Fine-grain transformations: force unroll factor Individual factors can be selected through a json file. Force input file Force output file Frequency (Hz) Frequency string (Hz) host OS If OpenMP is used, number of threads Installed perf supports output to file? Intel VTune Keep all produced files Kernel execution time in seconds Likwid installation Lillie P value Main function/kernel repetition Measure performance group (DATA, ENERGY, FLOPS_DP, L2, L2CACHE, MEM, TLB, BRANCH, etc) Measure using event set (INSTR_RETIRED_ANY:FIXC0,CPU_CLK_UNHALTED_CORE:FIXC1) Measured CPU frequency MILEPOST GCC Compiler Mode Mode Mode Notes Number of kernel repetitions (if supported) OpenME cTuning plugin OpenME library OS description Output correct? Parallelize (OpenMP)? Pearson P value pipeline failed? PLUTO compiler Prefix command line Prepared compiler flags Processor description Processor pinning (0; 0-1; 1,2-3,4; etc) Program Program Program binary size (bytes) Program compilation time (with module overhead) (sec.) Program execution CPI Program execution instructions retired Program execution time (with module overhead) (sec.) Program execution time normalized (with module overhead) (sec.) Program name Program UID Prune program by class Refresh from templates if exist Run exit code Save input json to file (for reproducibility/off-line tuning) - pipeline will not run! Save output json to file Select existing cTuning space Show input json (without running pipeline to prepare for off-line tuning) Show output json Skip target file check Statistical characteristic: Program compilation time (with module overhead) (sec.) (ALL) Statistical characteristic: Program compilation time (with module overhead) (sec.) (MAX) Statistical characteristic: Program compilation time (with module overhead) (sec.) (MIN) Statistical characteristic: Program compilation time (with module overhead) variation (DELTA PERCENT / 100) Statistical characteristic: Program compilation time (with module overhead) variation (DELTA) Statistical characteristic: Program execution time (with module overhead) (NUM) Statistical characteristic: Program execution time (with module overhead) (sec.) (ALL) Statistical characteristic: Program execution time (with module overhead) (sec.) (MAX) Statistical characteristic: Program execution time (with module overhead) (sec.) (MIN) Statistical characteristic: Program execution time normalized (with module overhead) (sec.) (ALL) Statistical characteristic: Program execution time normalized (with module overhead) (sec.) (MAX) Statistical characteristic: Program execution time normalized (with module overhead) (sec.) (MIN) Statistical characteristic: Program execution time normalized (with module overhead) variation (DELTA PERCENT / 100) Statistical characteristic: Program execution time normalized (with module overhead) variation (DELTA) Statistical characteristic: Program execution time variation (DELTA PERCENT / 100) Statistical characteristic: Program execution time variation (DELTA) System description target accelerator processor target OS target processor Tile factor Tile factors topology (dim1;dim2;dim3;;...) Tile? Tuning objective Turn on this pipeline section Turn on this pipeline section Turn on this pipeline section Turn on this pipeline section Turn on this pipeline section Turn on this pipeline section Turn on this pipeline section Turn on this pipeline section Turn on this pipeline section Turn on this pipeline section Turn on this pipeline section Turn on this pipeline section Turn on this pipeline section Turn on this pipeline section Turn on this pipeline section Turn on this pipeline section Turn on this pipeline section Turn on this pipeline section Turn on this pipeline section Universal Alchemist plugin Unroll factor Unroll? Use kernel/codelet instrumentation Use only base flag Use OpenMP parallelization flag Use sudo Use sudo User User-friendly system name Working repository X axis type: float integer uoa text X axis name: Figure size X (inch):   Figure size Y (inch):   DPI:   Detach graph: Fix additional dimensions:

Graph

Save as png eps pdf