Commit message (Collapse) | Author | Age | ||
---|---|---|---|---|
... | ||||
* | CarpetLib: Ensure that time and memory statistics are output | Erik Schnetter | 2012-09-11 | |
| | ||||
* | CarpetLib: Add Fortran wrappers for CarpetLib timers | Erik Schnetter | 2012-09-11 | |
| | ||||
* | CarpetLib: Avoid warning about sign change | Erik Schnetter | 2012-09-11 | |
| | ||||
* | CarpetLib: Remove support OMIT_F90 | Erik Schnetter | 2012-09-11 | |
| | | | | OMIT_F90 was already non-functional. | |||
* | Carpet: Change some #warnings into comments | Erik Schnetter | 2012-09-11 | |
| | ||||
* | CarpetLib: Change API to obtain pointer to grid function data | Erik Schnetter | 2012-09-11 | |
| | | | | | | | | 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 segfault | Erik Schnetter | 2012-09-11 | |
| | ||||
* | CarpetLib: Re-enable functional API for vect class | Erik Schnetter | 2012-09-11 | |
| | ||||
* | CarpetLib: Add vect::first and vect::last | Erik Schnetter | 2012-09-11 | |
| | ||||
* | CarpetLib: Remove/correct ATTRIBUTE_CONST and ATTRIBUTE_PURE declarations | Erik Schnetter | 2012-09-11 | |
| | ||||
* | CarpetLib: Declare gh::level_did_change as const | Erik Schnetter | 2012-09-11 | |
| | ||||
* | CarpetLib: Provide idiv/imod functions that always round down | Erik Schnetter | 2012-09-11 | |
| | | | | | | | Provide idiv/imod functions for signed integer division that always round towards negative infinity. Also provide vector version of these. Use these in the bbox output routines to ensure proper rounding. | |||
* | Carpet/CarpetLib: Move stack backtrace code to its own file | Erik Schnetter | 2012-09-11 | |
| | | | | | Also add implementation of backtrace_symbols to output line number information. | |||
* | CarpetLib: Move some OpenMP parallelisation into the operators | Erik Schnetter | 2012-09-11 | |
| | ||||
* | CarpetLib: Add preliminary support for DGFE operators | Erik Schnetter | 2012-09-11 | |
| | ||||
* | CarpetLib: Only produce refluxing debug output with CarpetLib::verbose | Erik Schnetter | 2012-09-11 | |
| | ||||
* | Merge | Erik Schnetter | 2012-09-11 | |
| | ||||
* | CarpetLib: wrap some debugging-only assert statement in an if block | roland haas | 2012-09-11 | |
| | ||||
* | CarpetLib: Improve prolongation with single precision. | Barry Wardell | 2012-09-11 | |
| | | | | | * Allow larger errors * Disable 11th order as single precision isn't good enough for it to be accurate. | |||
* | CarpetLib: Allow larger errors in single precision | Erik Schnetter | 2012-09-11 | |
| | | | | Allow larger prolongation errors in single precision. | |||
* | Merge | Erik Schnetter | 2012-09-11 | |
|\ | ||||
| * | Re-organise some refluxing internals | Erik Schnetter | 2012-09-11 | |
| | | | | | | | | | | | | | | | | | | | | | | Re-organise some of the internal details of refluxing. Refluxing requires restricting fluxes from fine to coarse grids. Previously, Carpet would internally store bounding boxes that were offset by 1/2 grid point, and adjust (correct) these boxes in various places in a rather ad-hoc manner. This is now cleaned up. Remove the (unused) "accumulate" operator, and the (unused) "accumulate" prolongation type. | |||
* | | CarpetLib: Correct white space | Erik Schnetter | 2012-09-11 | |
| | | ||||
* | | Cast several things to CCTK_REAL to support | Ian Hinder | 2012-09-11 | |
|/ | | | | | | | | single-precision --- Carpet/Carpet/src/Timing.cc | 2 +- Carpet/CarpetLib/src/gdata.cc | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) | |||
* | CarpetLib: add empty constructor to struct slab to make gcc happy | Roland Haas | 2012-09-11 | |
| | ||||
* | CarpetLib: Don't declare dummy variable const | Erik Schnetter | 2012-09-11 | |
| | ||||
* | CarpetLib: Correct argument name | Erik Schnetter | 2012-09-11 | |
| | ||||
* | CarpetLib: Allocate space for dstbox before restricting/prolongating | Erik Schnetter | 2012-09-11 | |
| | | | | | Allocate space for dstbox instead of srcbox before restricting or prolongating. | |||
* | CarpetLib: Extend operator API to allow hyperslabbing | Erik Schnetter | 2012-09-11 | |
| | | | | | | | Extend API of grid operators and of gdata::copy_from, gdata::transfer_from to allow hyperslabbing Implement hyperslabbing in copy_3d.cc | |||
* | CarpetLib: Move template definitions to mpi_string.hh | Erik Schnetter | 2012-09-11 | |
| | | | | Define templates in .hh file, instantiate them in .cc file. | |||
* | CarpetLib: Ignore empty bboxes in bboxset::operator|=(bbox) | Erik Schnetter | 2011-12-14 | |
| | ||||
* | Improve named barriers | Erik Schnetter | 2011-12-14 | |
| | | | | | | | | | Introduce dist::barrier as low-level implementation of a named barrier. Use it in Carpet::NamedBarrier. Use the above in almost all barrier calls. | |||
* | Slight optimization of CC ENO operators. | Christian Reisswig | 2011-12-14 | |
| | ||||
* | CarpetLib: Enable more detailed output for region_t | Erik Schnetter | 2011-12-14 | |
| | ||||
* | CarpetLib: Define MPI reduction operators for complex numbers | Erik Schnetter | 2011-12-14 | |
| | ||||
* | CarpetLib: Rewrite vectorisation of prolongate_3d_rf2 | Erik Schnetter | 2011-12-14 | |
| | ||||
* | CarpetLib: Improve bboxlib algorithms | Erik Schnetter | 2011-12-14 | |
| | | | | | | | Modify the algorithm used to insert a new bbox into an existing bboxset. This reduces the computational complexity of this operation from O(n^2) to O(n), where n is the number of elements in the bboxset. This has the potential to speed up regridding significantly. | |||
* | CarpetLib: avoid simultaneous implicit and explicit instantiation of | Roland Haas | 2011-12-14 | |
| | | | | prolongate_3d_cc_eno_rf2<T,3> | |||
* | CarpetLib: Correct component and process ids while regridding | Erik Schnetter | 2011-12-14 | |
| | | | | | | | During regridding, there are two sets of component indices and process numbers, namely those of the old and of the new grid structure. Previously, Carpet would erroneously always use those of the new grid structure; this fails when the number of components changes. | |||
* | CarpetLib: Correct communication schedule for regridding prolongation | Erik Schnetter | 2011-12-14 | |
| | | | | | Correct communication schedule for regridding prolongation for cell-centred mesh refinement. | |||
* | Adapt Carpet to hierarchical timers | Ian Hinder | 2011-12-14 | |
| | | | | | | | | Specifically, remove any hierarchy information that has been added to the name of timers, as well as any code for creating timers dynamically, as these are now unnecessary. Additionally, time some previously-untimed parts of the code and make timer names in some places more consistent. | |||
* | CarpetLib: Add function allgatherv1 in mpi_string.cc | Erik Schnetter | 2011-12-14 | |
| | ||||
* | CarpetLib: Create MPI datatypes for vect<,> classes | Erik Schnetter | 2011-12-14 | |
| | ||||
* | CarpetLib: Remove some attribute(const) declarations | Erik Schnetter | 2011-12-14 | |
| | ||||
* | Select 2nd order cc-eno for 3rd order interpolation since stencil radius is ↵ | Christian Reisswig | 2011-12-14 | |
| | | | | | | | | | | larger. This is consistent with vertex-centered eno. The trouble is that Carpet derives the stencil radius from the interpolation order imposing Lagrange interpolation. I also commented out the code that checks whether we have to switch down to first order. According to Shu, this is not necessary! This way interpolation runs faster. | |||
* | CarpetLib, Carpet: Add cell-centred TVD prolongation operator | Erik Schnetter | 2011-12-14 | |
| | ||||
* | Merge. | Christian Reisswig | 2011-12-14 | |
|\ | ||||
| * | Correct index check for eno operators. | Christian Reisswig | 2011-12-14 | |
| | | ||||
* | | Merge | Erik Schnetter | 2011-12-14 | |
|\ \ | |/ |/| | ||||
| * | CarpetLib: Correct check for VECTORISE_ALIGNED_ARRAYS | Erik Schnetter | 2011-12-14 | |
| | |