Skip to content
Snippets Groups Projects

Repository graph

You can move around the graph by using the arrow keys.
Select Git revision
  • BdG
  • BdG-Akai
  • BdG-merge-arrays-in-Bauer-solver
  • BdG_adaptive_integration
  • BdG_constraining
  • KKRIMP-BdG
  • KKRIMP-BdG-GPU
  • KKRIMP-BdG-GPU-BatchedLU
  • KKRIMP-BdG-GPU-GDYSON
  • KKRIMP-BdG-GPU-merge-sll-arrays
  • KKRIMP-BdG-aixcelerate
  • KKRIMP-BdG-test-Philipp
  • develop
  • develop-kkrsusc
  • kkrnano-chebyshev-tfQMRgpu
  • kkrnano-chebyshev-tfQMRgpu-jedi
  • kkrnano-constraint-bfields
  • kkrnano-dev2017
  • kkrnano-dev2017_gpu
  • kkrnano-dev2017_juqueenworkshop
  • v3.6
  • v3.5
  • pre-v3.5
  • v3.4
  • v3.3
  • v3.2
  • v3.1
  • pkkprime-v1.1
  • pkkprime-v1.0
  • rhoq-v1.1
  • kkrimp-v1.3
  • kkrhost-v3.0
  • kkrhost-v2.4
  • kkrhost-v2.3
  • kkrsusc-v1.0.1
  • kkrsusc-v1.0
  • kkrhost-v2.2
  • rhoq-v1.0
  • voronoi-v1.0
  • kkrhost-v2.1
40 results
Created with Raphaël 2.2.010May98543227Apr26252416151410Mar1Feb31Jan2524232019181716Dec1587529Nov23211716151484227Oct26251022Sep1217Aug228Jul27262521854130Jun21712May9526Apr85329Mar23221148Feb331Jan252110Dec62128Nov171615108543128Oct27262522211918141187654330Sep2823109818Aug10954321Jul121128May251811765428Apr26211914128129Mar261715121110943226Feb2523221931Jan27262521201918151318Dec1711843130Nov27261817161312109654330Oct28272322212098762128Sep22211124Aug18764331Jul30292422212017161413876230Jun292523221815430May2726252098Added output command line option to gdyson miniappMove initialization to GPUSome cleanupWorking batched versionBatched routine with batch size of 1 gives correct resultRefactor problematic loop in rll_local_solutionsreference manuscript for stable irregular solutionsKKRIMP-BdG-GPU-…KKRIMP-BdG-GPU-merge-sll-arraysuntested: merge arrays also for sllAdd openACC pragmas to move data to the GPU and avoid data transfer between host and the devicerefactored to make more modular (still the batched LU only uses single calls)linearsolve gpu version compileMerge branch 'KKRIMP-BdG-GPU' of https://iffgit.fz-juelich.de/kkr/jukkr into KKRIMP-BdG-GPUlinearsolve gpu versionMove pointers to cudalibs module and do a test running with larger batch sizesforgot to add a file in previous commitCreating linearsolve_dc_gpu subroutineTest different number of batchesRefactor rll_local_solutions to prepare for using batched LU on GPUsRemove serial option of rll_global_solutions and fix rll_local_solutions from previous mergeMerge branch 'KKRIMP-BdG-GPU' into KKRIMP-BdG-GPU-BatchedLUFix batched version of LU callsAdding missing #ifdef statementsPut outline for GPU version of LU in Dyson equationremoving unused variableMinor cleanupremove timing_pause statementKKRIMP-BdG-GPU-…KKRIMP-BdG-GPU-GDYSONUse GPUlibraries in sll_local and refactor gdysonFirst version using batched LU with batch size 1Merge branch 'KKRIMP-BdG-GPU-GDYSON' into 'KKRIMP-BdG-GPU'Fixing problems related to GMAT2Removing typoRestructuring the dyson standalone test code and adding timersSmall fixesFix compilation with intel compilerAdd icolor2 argument for timing calls in sll_local_solutionsComplete source list and make gdyson mini app compile on jureca with NVIDIA compilerFirst versionMerge branch 'BdG-merge-arrays-in-Bauer-solver' into KKRIMP-BdG-GPUUndo changes in kkrflex main routine (we work on rllsll mini-app for now)Initial version that uses cusolver GPU library for LU factorization in rll_local_solutions