Skip to content
Snippets Groups Projects

Repository graph

You can move around the graph by using the arrow keys.
Select Git revision
No matching results
An error occurred while fetching branches. Retry the search.
An error occurred while fetching tags. Retry the search.
Created with Raphaël 2.2.010Feb93230Jan292524231916826Dec2120151413121110765432130Nov2928262221201918171615141312111098765432131Oct3029272524232221191716764Added GridTraverserBenchmarkHelper.Added synchrounous/asynchronous modes for grid traversers.Fixed calculation of bandwidth in the traverser benchmarksTraverser benchmarks: added explicit cast to RealFixed order of indices in the traverser benchmarksFixes in traversers benchmark.GridTraverser_impl.h splitted into GridTraverser_1D.hpp, GridTraverser_2D.hpp and GridTraverser_3D.hpp.GridTraverser_impl.h splitted into GridTraverser_1D.hpp, GridTraverser_2D.hpp and GridTraverser_3D.hpp.Fixed tnl-benchmark-traversers.hFixing traversers benchmark kernels.Fixed CUDA travresers benchmark tests.Traversers benchmark tests can be configures as list of tests.Added method containsValue to List.Refactoring.Analyzing grid traversers.Optimized conditional OpenMP traversing in 2D and 3D grid traversers - cells only.Added configuration parameter 'reset' to Benchmark.Added traversers benchmark test with mesh function.Deleting old code.GridTraversersBenchmark.h splitted into GridTraversersBenchmark_1D.h, GridTraversersBenchmark_2D.h and GridTraversersBenchmark_3D.h.GridTraversersBenchmark.h splitted into GridTraversersBenchmark_1D.h, GridTraversersBenchmark_2D.h and GridTraversersBenchmark_3D.h.Implemented traversers benchmark test - parallel for with a grid entity.Added traversers benchmark test - parallel for with a grid entity.Fixing indexes ordering in parallel for in traversers benchmark.Additional fixes of the traversers benchmark tests.Traversers benchmark tests can be controled from the command line.Fixed cell type in traversers benchmark.Added flag -g to compilation of the traversers benchmark.Timing can be turned off in the becnhmark - for better profiling.Added traversers benchmarks with boundaries.Splitting Benchmarks.h into Benchmarks, Logging and FunctionTimer.Added traversers benchmark tests without reseting.Added benchmark function timing without reset function to measure CPUCUDA device synchronization is performed only for CUDA benchmarks.Added pure-C test to traversers benchmark.Changing verbose form bool to int to have three levels of verbosity inFixed traversers benchmark test using traverser.Fixed indexing and data set size in traversers benchmark.Changing minTime in Benchmark from int to double.Fixed memory bandwidth in traversers benchmark.