aboutsummaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAge
* Timers: disable timer trees by defaultcartoonAnton Khirnov2024-04-14
| | | | They are very slow on Xeon Phi.
* bbox: reduce optimization levelAnton Khirnov2020-09-08
| | | | Fix build with ICC
* Stop using API removed in MPI 2.0Anton Khirnov2020-06-17
|
* Interp: allow the user to specify maximum reflevel for interpolationAnton Khirnov2018-08-07
|
* Evolve: fix tapered gridsAnton Khirnov2017-08-15
|
* Update for cartoon.Anton Khirnov2017-06-30
|
* CarpetIOHDF5: avoid variable shadowing.Anton Khirnov2013-11-05
|
* CarpetIOHDF5: remove a superfluous H5Pclose call.Anton Khirnov2013-11-05
|
* Carpet: use gdata::fence_is_energized to query electric_fenceRoland Haas2013-10-31
|
* CarpetLib: introduce accessor in gdata to query electric_fenceRoland Haas2013-10-31
| | | | | this function effectively exposes the CarpetLib parameter to Carpet without explicit parameter sharing
* Revert "CarpetLib: make electric_fence parameter restricted"Roland Haas2013-10-31
| | | | | | | This reverts commit be2bd96c56aa6dd5b1fb6654b130f3e44e049e8e. Sharing parameters between CarpetLib and Carpet can apparantly cause problems.
* CarpetLib: code layout changes when using map for allgfsRoland Haas2013-10-31
|
* LoopControl: Avoid nans in statistics outputErik Schnetter2013-10-28
|
* CarpetLib: Remove unnecessary parenthesesErik Schnetter2013-10-26
|
* CarpetLib: Remove commented-out codeErik Schnetter2013-10-26
|
* Carpet: Remove accidentally-committed extra arguments to Timer constructorIan Hinder2013-10-26
|
* Carpet: Add a barrier between prolongation and sync controlled by ↵Ian Hinder2013-10-25
| | | | sync_barriers parameter
* CarpetInterp2: Time only interpolation using Interpolate timer, and enable ↵Ian Hinder2013-10-25
| | | | more timers for tracing
* Timers: Output all clocks in XML timer tree filesIan Hinder2013-10-25
|
* CarpetInterp2: Add a barrier after interpolatingIan Hinder2013-10-25
|
* CarpetInterp2: Add parameter "interp_barrier" to add a barrier before ↵Ian Hinder2013-10-25
| | | | | | interpolating Interpolation is a global operation, so adding a barrier should not affect performance, and makes timers much more useful.
* CarpetInterp2: Add timers for fasterp interpolationIan Hinder2013-10-25
|
* LoopControl: provide omp_get_max_threads fallbackRoland Haas2013-10-24
|
* CarpetInter2: Avoid const warningsErik Schnetter2013-10-22
|
* CarpetInterp2: Make code compile when CARPETINTERP2_CHECK is not setErik Schnetter2013-10-22
|
* CarpetLib: use map instead of set for dh::gfsRoland Haas2013-10-15
| | | | | some routines assume that elements in gfs are sorted by variable index when traversing the container (eg recompose_allocate does)
* Rename new max_time_levels parameter as max_timelevels to be consistent with ↵Ian Hinder2013-10-15
| | | | other uses of "timelevels" in parameter names in Carpet
* Carpet: skip CheckFence routine if electric_fence is offRoland Haas2013-10-11
| | | | | enumerating the variables is expensive enough to show up in the timer output
* CarpetLib: make electric_fence parameter restrictedRoland Haas2013-10-11
|
* CycleClock: define CycleClock_Setup() without HAVE_TICK_COUNTERFrank Löffler2013-10-11
| | | | | | | Define a dummy CycleClock_Setup() routine for Cactus to call from the scheduler. Without this, the linker will rightfully complain about the missing symbol if the architecture doesn't have counters (HAVE_TICK_COUNTER isn't defined).
* CarpetLib: Define elapsed() when CycleClock provides no timerErik Schnetter2013-10-11
|
* CycleClock: Do not abort when no good clock is availableErik Schnetter2013-10-10
|
* Carpet: Make dependency on CycleClock optionalErik Schnetter2013-10-10
|
* Carpet: remove duplicate myproc variable in SetupGHRoland Haas2013-09-27
|
* CarpetTest: add no_split_in_directions tagRoland Haas2013-09-27
| | | | regenerate data for 2 processes, disable slabtest by default
* CarpetTest: recreate test data in new CarpetIOASCII formatRoland Haas2013-09-27
| | | | | | data is still the same, for grid-arrays I modified the files by hand, for the grid function I regenerated after checking that data is identical
* CarpetTest: make compile againRoland Haas2013-09-27
|
* CarpetSlab: pass size of allocated chunk to memory objectRoland Haas2013-09-27
|
* Carpet: fix no_split_directions implementationRoland Haas2013-09-27
| | | | had forgotten a level of indirection
* Carpet: add new group tag no_split_directionsRoland Haas2013-09-27
|
* CarpetEFenceTest: test electric fence in CarpetLibRoland Haas2013-09-27
|
* Carpet: use electric fence provided by CarpetLibRoland Haas2013-09-27
| | | | only active when CarpetLib's electric_fence parameter is set
* CarpetLib: add some code for electric fence like functionalityRoland Haas2013-09-27
|
* Carpet: correct wording of error message about number of time levelsRoland Haas2013-09-25
|
* CarpetIOHDF5: correct output method name in INFO messageRoland Haas2013-09-25
|
* CarpetLib: Add sanity check for number of time levelsErik Schnetter2013-09-25
|
* Carpet: Initialise number of timelevels at the right timeErik Schnetter2013-09-25
| | | | Also clean up code structure for setting number of time levels.
* Carpet: Comment out unused routineErik Schnetter2013-09-25
|
* Carpet: Modernize error reportingErik Schnetter2013-09-25
|
* Carpet: Introduce parameter max_time_levelsErik Schnetter2013-09-25
| | | | This parameter defines the maximum number of time levels that a variable can have when being used for time prolongation. This allows using more time levels than prolongation_order_time+1.
generated by cgit v1.2.3 (git 2.39.1) at 2024-05-09 00:21:51 +0000