Commit message (Collapse) | Author | Age | |
---|---|---|---|
* | Carpet: Add a barrier between prolongation and sync controlled by ↵ | Ian Hinder | 2013-10-25 |
| | | | | sync_barriers parameter | ||
* | CarpetInterp2: Time only interpolation using Interpolate timer, and enable ↵ | Ian Hinder | 2013-10-25 |
| | | | | more timers for tracing | ||
* | Timers: Output all clocks in XML timer tree files | Ian Hinder | 2013-10-25 |
| | |||
* | CarpetInterp2: Add a barrier after interpolating | Ian Hinder | 2013-10-25 |
| | |||
* | CarpetInterp2: Add parameter "interp_barrier" to add a barrier before ↵ | Ian Hinder | 2013-10-25 |
| | | | | | | 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 interpolation | Ian Hinder | 2013-10-25 |
| | |||
* | LoopControl: provide omp_get_max_threads fallback | Roland Haas | 2013-10-24 |
| | |||
* | CarpetInter2: Avoid const warnings | Erik Schnetter | 2013-10-22 |
| | |||
* | CarpetInterp2: Make code compile when CARPETINTERP2_CHECK is not set | Erik Schnetter | 2013-10-22 |
| | |||
* | CarpetLib: use map instead of set for dh::gfs | Roland Haas | 2013-10-15 |
| | | | | | 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 Hinder | 2013-10-15 |
| | | | | other uses of "timelevels" in parameter names in Carpet | ||
* | Carpet: skip CheckFence routine if electric_fence is off | Roland Haas | 2013-10-11 |
| | | | | | enumerating the variables is expensive enough to show up in the timer output | ||
* | CarpetLib: make electric_fence parameter restricted | Roland Haas | 2013-10-11 |
| | |||
* | CycleClock: define CycleClock_Setup() without HAVE_TICK_COUNTER | Frank Löffler | 2013-10-11 |
| | | | | | | | 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 timer | Erik Schnetter | 2013-10-11 |
| | |||
* | CycleClock: Do not abort when no good clock is available | Erik Schnetter | 2013-10-10 |
| | |||
* | Carpet: Make dependency on CycleClock optional | Erik Schnetter | 2013-10-10 |
| | |||
* | Carpet: remove duplicate myproc variable in SetupGH | Roland Haas | 2013-09-27 |
| | |||
* | CarpetTest: add no_split_in_directions tag | Roland Haas | 2013-09-27 |
| | | | | regenerate data for 2 processes, disable slabtest by default | ||
* | CarpetTest: recreate test data in new CarpetIOASCII format | Roland Haas | 2013-09-27 |
| | | | | | | 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 again | Roland Haas | 2013-09-27 |
| | |||
* | CarpetSlab: pass size of allocated chunk to memory object | Roland Haas | 2013-09-27 |
| | |||
* | Carpet: fix no_split_directions implementation | Roland Haas | 2013-09-27 |
| | | | | had forgotten a level of indirection | ||
* | Carpet: add new group tag no_split_directions | Roland Haas | 2013-09-27 |
| | |||
* | Carpet: use electric fence provided by CarpetLib | Roland Haas | 2013-09-27 |
| | | | | only active when CarpetLib's electric_fence parameter is set | ||
* | CarpetLib: add some code for electric fence like functionality | Roland Haas | 2013-09-27 |
| | |||
* | Carpet: correct wording of error message about number of time levels | Roland Haas | 2013-09-25 |
| | |||
* | CarpetIOHDF5: correct output method name in INFO message | Roland Haas | 2013-09-25 |
| | |||
* | CarpetLib: Add sanity check for number of time levels | Erik Schnetter | 2013-09-25 |
| | |||
* | Carpet: Initialise number of timelevels at the right time | Erik Schnetter | 2013-09-25 |
| | | | | Also clean up code structure for setting number of time levels. | ||
* | Carpet: Comment out unused routine | Erik Schnetter | 2013-09-25 |
| | |||
* | Carpet: Modernize error reporting | Erik Schnetter | 2013-09-25 |
| | |||
* | Carpet: Introduce parameter max_time_levels | Erik Schnetter | 2013-09-25 |
| | | | | 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. | ||
* | CarpetRegrid2: Correct wording of error message | Erik Schnetter | 2013-09-24 |
| | |||
* | CarpetInterp2: Explicitly convert values to CCTK_REAL | Erik Schnetter | 2013-08-21 |
| | |||
* | CarpetIOHDF5: check whether map exists before accessing it | Roland Haas | 2013-08-18 |
| | | | | | | | | | | this allows data files from multipatch runs to be read in with the file reader into cartesian runs if only the inner cartesian patch is required.CarpetIOHDF5: check wehter map exists before accessing it this allows data files from multipatch runs to be read in with the file reader into cartesian runs if only the inner cartesian patch is required. | ||
* | CarpetIOHDF5: remove two compiler warnings | Roland Haas | 2013-08-18 |
| | |||
* | CarpetIOHDF5: fix reading of new-style grid structures | Roland Haas | 2013-08-18 |
| | | | | | | | | last commit to CarpetIOHDF5 broke this. This commit also updates the test suite data so that it actually tests the file format. This commit adds a level 2 warning if no grid structure is found in file. | ||
* | LoopControl: Disable debug checks | Erik Schnetter | 2013-08-15 |
| | |||
* | LoopControl: Initialize fine thread communicators ahead of time | Erik Schnetter | 2013-08-15 |
| | |||
* | LoopControl: Add debug output to help track down assertion failure | Erik Schnetter | 2013-08-15 |
| | |||
* | LoopControl: Explicitly check for consistent loop arguments from the caller | Erik Schnetter | 2013-08-15 |
| | |||
* | LoopControl: Initialize num_smt_threads explicitly during loop startup | Erik Schnetter | 2013-08-15 |
| | |||
* | CarpetIOHDF5: use simple_extents rather than storage size on disk | Roland Haas | 2013-08-12 |
| | | | | | | | | | | storage size is literally the amount of space used on disk, so if eg compression is used, this is much smaller than the amount of space required to hold the data in memory. Also change type of data read in to what the memory dataspace is not what the dataspace on disk is. This way HDF5 actually converts from the on-disk representation to the in-memory one. | ||
* | CarpetLib: Correct memory allocation error in communication buffers | Erik Schnetter | 2013-08-12 |
| | |||
* | CarpetLib: Optimize memory allocation when message_count_multiplier>1 | Erik Schnetter | 2013-08-12 |
| | |||
* | CarpetIOHDF5: support IO->alias option in reader | Roland Haas | 2013-08-10 |
| | | | | | this allows the reader to read a dataset into a different variable than which one was written. Eg. GRHydro::dens in PPAnalysis::dens. | ||
* | CarpetIOHDF5: delay browsing of input files | Roland Haas | 2013-08-10 |
| | | | | this means the BrowseDataSets is only called from one location | ||
* | LoopControl: Update Fortran macros | Erik Schnetter | 2013-08-08 |
| | | | | Update Fortran macros to conform to recent changes | ||
* | LoopControl: Avoid non-standard C++ feature | Erik Schnetter | 2013-08-08 |
| |