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 2.2.03Aug30Jul28262522201817121086130Jun242018171614131110985432129May282726211916109121Apr1611931Mar30161065432129Feb1242129Jan2717161587Added different versions of CSR Light to script and benchmarkAdded different versions of CSR LightCode cleaningAdded original CSR LightFixed compilation errorFixed blocks fillingAdded setBlocks method, commented getColumns in for cyclesBug fix for CSR MultiVector, optimizations for CSR LightWithoutAtomicDivided CSR LightWithoutAtomic by 4 kernelsOptimizations for all kernelsUncommented kernelsOptimizations for CSR AdaptiveUncommented kernelsCompilation error fixMemory optimizations, changes to CSR Adaptive and unit testsChanged CSR AdaptiveIncreased block sizes, optimizations for CSR Light, new logic for CSR DynamicAdded exceptions handling to tnl-benchmark-spmv.Fixed linking of DistributedMatrixTest.One more fix of linking -lcudadevrt.Fixed script for processing results of SpMV benchmark.Fixed linking of libcudadevrt.Finishing rebase.Fixing includes of std headers in ConfigEntryType.Temporary deleted using of shared memoryCSR Adaptive optimizationsBug fixesFixed script and benchmarkAdded CSRLightWithoutAtomic, small optimizationsRefactoring, added CSR MultiVector(a lot of warps for one row)Fix of SpMV benchmark and update of Python script for SpMV benchmark results processing.Possible fix for compilationFixed bugsAdded CSR Vector with dynamic parallelismAdded libcudadevrt to fix problem with linking.Changes to CSR SpMV functionsProblem with functionMerge branch 'JK/VTK' into 'develop'Generalized DistributedMeshSynchronizer to allow synchronization of plain arrays with multiple values per elementFixed distributeSubentities