aboutsummaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAge
...
* CarpetLib: store OpenMP thread number in dist objectRoland Haas2012-04-09
|
* CarpetRegrid2: Ensure that the refinement hierarchy is symmetry about the ↵Erik Schnetter2012-04-04
| | | | | | origin, if desired Introduce a new parameter that tests whether the refinement hierarchy is symmetric about the origin.
* CarpetLib: Improve comment of dh data structuresErik Schnetter2012-04-04
|
* Carpet: Time OutputGH operationsErik Schnetter2012-04-04
|
* Carpet: Do not count ghost zones when calculating speedErik Schnetter2012-04-04
| | | | When calculating the speed of a simulation, ignore ghost zones. Use two counters, one including the outer boundaries, the other excluding it.
* CarpetIOASCII: Declare function prototypeErik Schnetter2012-04-04
|
* Make timer tree output precision and threshold configurable with parametersIan Hinder2012-04-04
|
* CarpetLib: output regions when not-contained error is detected inRoland Haas2012-03-08
| | | | restrict_3d_cc_rf2 and restrict_3d_vc_rf2
* Carpet: make OutputGridCoordinates fully multipatch awareRoland Haas2012-03-26
| | | | | | | This is achieved by using domain information in domainspecs the way OutputGrids does. The key functional change was to use Multipatch_ConvertFromPhysicalBoundary (used to fill domainspecs) rather than CoordBase::ConvertFromPhysicalBoundary.
* Backed out changeset 74ddd545dbb1Roland Haas2012-03-20
| | | | | I give up. I cannot get this right today. The previous patch would have allowed a thread to switch sockets which is undesirable.
* Carpet: let thread run on any core [n0,n0+num_threads) instead of binding toRoland Haas2012-03-20
| | | | | | | single core This is the behaviour of eg. Kraken's aprun --cc numa_node. Avoids using omp_get_thread_num().
* Backed out changeset 27a049a40a81Roland Haas2012-03-20
| | | | | sigh, there _is_ a difference between omp_get_num_threads() and omp_get_thread_num() after all.
* Carpet: replace omp_get_num_threads() by mynthreads to compile without OpenMPRoland Haas2012-03-20
|
* CarpetIOF5: Continue to implement file readingErik Schnetter2012-03-18
|
* Carpet: Check CPU affinity per thread, not per processErik Schnetter2012-03-17
|
* Carpet: Don't warn when there are more cores then threadsErik Schnetter2012-03-16
|
* Carpet: Don't use CPU_COUNTErik Schnetter2012-03-16
|
* Carpet: Add example parameter files for CPU affinity handlingErik Schnetter2012-03-16
|
* MergeErik Schnetter2012-03-16
|\
| * Carpet: Add parameter set_cpuaffinityErik Schnetter2012-03-16
| |
* | CarpetIOScalar: fix handling of io_truncate_filesRoland Haas2012-03-13
|/
* CarpetIOF5: Make generic reading workErik Schnetter2012-03-07
|
* CarpetIOScalar: add option all_reduction_in_one file to collect all reductionRoland Haas2012-03-03
| | | | results in a single file (with filename *.scalars.asc)
* Carpet: Avoid superfluous mode changes when synchronisingErik Schnetter2012-03-03
|
* Carpet: Check same_number_of_components_on_each_processErik Schnetter2012-03-03
|
* CarpetReduce: Warn about reductions in local modeErik Schnetter2012-03-03
|
* CarpetLib: Cache end() iterator in bboxsetErik Schnetter2012-03-03
|
* CarpetIOF5: Make outputting scalars and CCTK_INT variables workErik Schnetter2012-03-01
|
* CarpetInterp2: Make compile with CCTK_REAL=floatErik Schnetter2012-03-01
|
* MergeErik Schnetter2012-03-01
|\
| * Carpet: Make compile with CCTK_REAL=floatErik Schnetter2012-03-01
| |
* | Carpet: Clean up codeErik Schnetter2012-02-29
| | | | | | | | | | Move timer handling into SyncGroupsInScheduleBlock. Set up list of groups to synchronize only once.
* | CarpetLib: Ensure that time and memory statistics are outputErik Schnetter2012-02-29
|/
* Carpet: Correct representation of host and model namesErik Schnetter2012-02-27
|
* CarpetReduce: Remove unused variablesErik Schnetter2012-02-25
|
* CarpetLib: Add Fortran wrappers for CarpetLib timersErik Schnetter2012-02-25
|
* CarpetLib: Avoid warning about sign changeErik Schnetter2012-02-25
|
* CarpetLib: Remove support OMIT_F90Erik Schnetter2012-02-25
| | | | OMIT_F90 was already non-functional.
* CarpetIOHDF5: Simplify makefileErik Schnetter2012-02-25
|
* CarpetIOHDF5: More internal error checkingErik Schnetter2012-02-25
|
* CarpetIOASCII: Add prototype for read_next_set to avoid warningErik Schnetter2012-02-25
|
* CarpetIOASCII: White space changeErik Schnetter2012-02-25
|
* Carpet: Change some #warnings into commentsErik Schnetter2012-02-25
|
* Carpet: LoadBalanceReal: UpdateErik Schnetter2012-02-25
| | | | | Set global parameters before decomposing. Declare C++ prototype.
* Carpet: Reformat some code in Timers.ccErik Schnetter2012-02-25
|
* Carpet: Restore output precision after TimerNode outputErik Schnetter2012-02-25
|
* Carpet: Initialise Carpet::in_analysis_binErik Schnetter2012-02-25
|
* CarpetLib: Change API to obtain pointer to grid function dataErik Schnetter2012-02-25
| | | | | | | | Change the API to obtain a pointer to grid function data: - Use a function "typed_data_pointer" instead of overloading the () operator (because this looks nicer) - Don't use a virtual function (because this isn't needed) - Update all uses
* CarpetLib: Rewrite some loops to avoid Intel compiler segfaultErik Schnetter2012-02-25
|
* CarpetLib: Re-enable functional API for vect classErik Schnetter2012-02-24
|