BranchCommit messageAuthorAge
fast_tnvRemove verification outputSuren A. Chilingaryan18 months
AgeCommit messageAuthorFilesLines
2020-03-29Remove verification outputfast_tnvSuren A. Chilingaryan2-4/+4
2020-03-29Add optimization steps for referenceSuren A. Chilingaryan10-0/+4989
2020-03-29Enable padding over Z dimmensionSuren A. Chilingaryan5-12/+14
2020-03-29Eliminate conditionals in the innermost loop to help gcc auto-vectorizerSuren A. Chilingaryan5-202/+423
2020-03-29Optimize cache usage also for the version with full back-trackingSuren A. Chilingaryan1-65/+58
2020-03-29Optimize cache usage with blocking (icc is faster and gcc is slightly slower ...Suren A. Chilingaryan1-43/+51
2020-03-27Optimized TNV routine (10x performance, 1/3 memcory consumption)Suren A. Chilingaryan9-384/+2097
2020-03-02Merge pull request #143 from vais-ral/cudaresetDaniil Kazantsev10-6/+11
2020-03-02cudaDeviceSynchronize before exitdkazanc10-6/+11
2020-02-28cudareset for fgp_tvdkazanc1-1/+1