aboutsummaryrefslogtreecommitdiff
path: root/Carpet
Commit message (Collapse)AuthorAge
* build the carpet2sdf utility only if $(HAVE_SDF) is set (by a thorn ↵tradke2004-12-21
| | | | | | providing the capability SDF) darcs-hash:20041221142124-3fd61-4f57ce8c775252a94b4b5b6ccf780714906e549f.gz
* CarpetWeb: Add pointer to open letter to EU ministers regarding software patentsErik Schnetter2004-12-19
| | | | darcs-hash:20041219124810-891bb-91b0c6d9e8571de3a33b0432e2d58f74da71bbba.gz
* Carpet: Update documentation of schedule binsErik Schnetter2004-12-16
| | | | darcs-hash:20041216133421-891bb-154945b77de8770636357df1ec382ce6f0763c1a.gz
* CarpetWeb/talk-2004-12-13: added missing schedule bin CCTK_RECOVER_PARAMETERStradke2004-12-16
| | | | darcs-hash:20041216125335-3fd61-4be7ff85b9a496dd1e0174618fb32d43906b145d.gz
* CarpetWeb: Move old news awayErik Schnetter2004-12-15
| | | | darcs-hash:20041215123430-891bb-0496fdf7b2f3bf69df33efbbe968ea1d74508094.gz
* CarpetWeb: Update darcs executable to version 1.0.1Erik Schnetter2004-12-14
| | | | darcs-hash:20041214205851-891bb-29bcc55594589c6b3e1b8c77a635a70c528557f3.gz
* Carpet: Update documenation of Carpet's schedule binsErik Schnetter2004-12-14
| | | | darcs-hash:20041214210303-891bb-800bdbb84b9c8a4e6e9aac3d1dd846c601feec29.gz
* CarpetWeb: Update Carpet schedule descriptionErik Schnetter2004-12-14
| | | | darcs-hash:20041214192052-891bb-1d041fb29da23c78bb7b6f254a0fe3b2bb4fbd01.gz
* CarpetWeb: Add explanation of Carpet schedule to course materialErik Schnetter2004-12-14
| | | | darcs-hash:20041214132133-891bb-a87b9ac2384372bfc79218431dd1f2195f17f754.gz
* CarpetWeb: Add example files for design walkthroughErik Schnetter2004-12-13
| | | | darcs-hash:20041213140305-891bb-502557cd15727908977eca7a385ab2b4741e0932.gz
* CarpetWeb: Update darcs executable to version 1.0.1rc2Erik Schnetter2004-12-10
| | | | darcs-hash:20041210224231-891bb-7d406894f31b628d7d5a18c4cb6f71bd2396a7e8.gz
* CarpetWeb: Announce Carpet design walkthroughErik Schnetter2004-12-07
| | | | darcs-hash:20041207160617-891bb-285436c02558b68231e3e3cb2e845c64c5b96626.gz
* CarpetWeb: Correct errors in page footersErik Schnetter2004-12-07
| | | | darcs-hash:20041207160552-891bb-a5c793d849d10d64aa655b7786c582930ecd0c6c.gz
* CarpetWeb: Add email sending scriptErik Schnetter2004-12-10
| | | | darcs-hash:20041210194930-891bb-c8db7687b63bc824e051ffcabd4268421a265905.gz
* CarpetLib:gh exposed data clamp-downswhite2004-12-10
| | | | | | | | | | | | | | | | | | | | | | | | | | | Made all variables in CarpetLib:gh to be const or non-public, to avoid confusion and insanity. Toward this end, created four member accessors for gh: const rexts & extents() const const rbnds & outer_boundaries() const const rprocs & processors() const const vector<vector<ibbox> > & bases() const This involved a few changes in several files throughout the repository: M ./Carpet/Carpet/src/Initialise.cc -2 +2 M ./Carpet/Carpet/src/Recompose.cc -8 +8 M ./Carpet/Carpet/src/SetupGH.cc -2 +2 M ./Carpet/Carpet/src/modes.cc -3 +3 M ./Carpet/CarpetIOHDF5/src/Recover.cc -4 +4 M ./Carpet/CarpetInterp/src/interp.cc -5 +5 M ./Carpet/CarpetLib/src/dh.cc -3 +3 M ./Carpet/CarpetLib/src/gh.cc -33 +33 M ./Carpet/CarpetLib/src/gh.hh -10 +28 M ./Carpet/CarpetReduce/src/mask_carpet.cc -4 +4 M ./Carpet/CarpetRegrid/src/automatic.cc -2 +2 darcs-hash:20041210131459-32473-ff0835ff0e57f6693fbe1ec23b350d1b1d18e3bc.gz
* CarpetLib/src/gh.hh member tighteningswhite2004-12-10
| | | | | | | | | | | | Tightened access to a few public members of gh class. Made a few const (they are never modified after construction), Made a couple private. There are several more with scary access: full public non-const (extents, e.g.) which is used all over and ... try not to think about it. darcs-hash:20041210004122-32473-3ed8ee7f07ea9f13400f51f580dc1d29a4258a1a.gz
* CarpetLib/src/*gf.*,*data.* const public members, canonicaliseswhite2004-12-09
| | | | | | | | | | | | | | | | | | The g* classes had exposed public member variables which were never meant to be altered from outside (I imagine altering most of them would have resulted in very bad behaviour); I made most of these const. This resulted in some complaints by the compiler about the canonical constructor and assignment ops, so I provided private dummies. This patch contains the following changes: M ./Carpet/CarpetLib/src/gdata.cc +6 M ./Carpet/CarpetLib/src/gdata.hh -2 +2 M ./Carpet/CarpetLib/src/gf.hh +5 M ./Carpet/CarpetLib/src/ggf.hh -9 +15 darcs-hash:20041209172710-32473-46d2a93e11f21f305467b375fe15910d47b8e4de.gz
* Fix checkpointing at arbitrary iterationstradke2004-12-09
| | | | | | | | | | | | | | | | | Checkpointing/recovery is now possible at arbitrary iterations. Carpet/CarpetIOHDF5/src/Recover.cc: synchronize all groups are recovery and filereader. This fixes differences in the initial data on finer levels where no interpolation from the coarser levels has been done yet. Carpet/CarpetIOHDF5/src/Checkpoint.cc: removed ugly code to self-modify the IO::checkpoint_every parameter in order to force checkpointing only on the coarsest refinement level Carpet/Carpet/src/Evolve.cc: check the termination conditions on each active refinement level (not only the coarsest one) darcs-hash:20041209135748-3fd61-92910a41aa8547473dc7e7a947fb98e8d5617517.gz
* call initial data checkpoint, analysis and output routines only for ↵tradke2004-12-09
| | | | | | | | | | refinement levels which are to be updated at the current iteration This patch effects only initial data after recovery: it prevents the analysis and output of refinement levels which weren't updated at the iteration when the checkpoint was created. darcs-hash:20041209135038-3fd61-96f33595a27f7705734ae1822427bbf02e7fe389.gz
* CarpetLib/src/data.* interpolate_from_innerloop breakdownswhite2004-12-09
| | | | | | | | | | | | | | | | | Broke down monster function data::interpolate_from_innerloop to make its overall logic visible. Added three functions void data::Check_that_the_times_are_consistent bool data::interpolate_in_time void data::interpolate_prolongate void data::interpolate_restrict Made constant 'eps' to be static file scope. Got rid of another nasty surprise return. darcs-hash:20041209121013-32473-e3164f151ce8a67871459c847e1f779a1f6d40fc.gz
* CarpetLib/src/*data* CarpetIOASCII/src/ioascii.cc MPI_Comm_rank wrappersswhite2004-12-08
| | | | | | | | | | | | | | | | | | | | | | | Removed evil assert from otherwise harmless public accessor gdata::proc(). Very systematically checked that the assert is still done everywhere this function was previously called. New functions bool this_processor_is (int procno); bool lives_on_this_processor (); wrapper messy repeated MPI_Comm_rank code. Had to modify ioascii.cc due to missing assert as explained above, but found no other instances of calls to proc(). This patch contains the following changes: M ./Carpet/CarpetIOASCII/src/ioascii.cc +1 M ./Carpet/CarpetLib/src/data.cc -41 +21 M ./Carpet/CarpetLib/src/gdata.cc -6 +19 M ./Carpet/CarpetLib/src/gdata.hh -1 +3 darcs-hash:20041208171657-32473-11da4a7ec810f97b2c82c322119fa69a9ffb4640.gz
* data.cc abs min max to fabs fmin fmaxswhite2004-12-07
| | | | | | | | | Repeated comments like Catch broken compilers that only know min(int), when the arguments of the functions could only be floating-point. So...changed to fabs fmin fmax and got rid of the Catch. darcs-hash:20041207224657-32473-0f54e7b139bd275e8906f6169400f214e968b1c5.gz
* SetupGH.cc decomposed long SetupGH functionswhite2004-12-07
| | | | | | | | | Broke up SetupGH function into smaller subroutines. Got it down to a single page, but its offspring 'allocate_map' needs further work. darcs-hash:20041207214705-32473-16294512e93ed1ad45e85b57285e2cf9f0b7da3c.gz
* CarpetIOHDF5/src/Output.cc: changed warning level for deprecated parameters ↵tradke2004-12-07
| | | | | | from 1 to 2; don't sleep(3) after such warnings darcs-hash:20041207163828-3fd61-dac602f5c3a0285ed05e79046d3fde3c57e22c33.gz
* Initialise.cc decomposed big functionswhite2004-12-06
| | | | | | | | | | | | | | | Restructured Initialise function from a being a big fat mama alien to more manageable baby aliens. Now the main function is under a page long, and some more complicated logic is dealt with by other levels of functions. Previous attempt, patch 'Initialise_decompose', was marred by misunderstanding of an evil global variable 'delta_time'. Also, it contained an unrelated file by accident, and wounded Erik's delicate formatting sensibilites. darcs-hash:20041206234531-32473-02e5b23237703ae54dcffac430244ba64d8c83fa.gz
* CarpetIOHDF5/src/Output.cc: forgot to include <unistd.h> to get sleep(3)tradke2004-12-06
| | | | darcs-hash:20041206152516-3fd61-1dec08523b8094b4decabed463b37188d85914a6.gz
* CarpetIOHDF5: fix output of DISTRIB=CONSTANT arraystradke2004-12-06
| | | | | | | | | | | | DISTRIB=CONSTANT grid variables (including scalars) are assumed to have the same values across all processors. Therefore only the portion on processor 0 will be output. The portions from other processors are compared against this processor's portion, and a level-1 warning will be issued if they don't match. During recovery, the DISTRIB=CONSTANT array portion from the checkpoint file is distributed across all processors (independently of how many there are). darcs-hash:20041206151054-3fd61-5a47be7936480027c1e6f4419a4051d223e10c7d.gz
* UNDO: Initialise_decomposeswhite2004-12-01
| | | | | | | Broke this mama of a function into a couple of levels of functional parts. darcs-hash:20041201155737-32473-d59b9f090a4d5bac0bce2d668c7ae674d545e660.gz
* CarpetIOHDF5/src/Output.cc: if deprecated parameters were set in the ↵tradke2004-12-03
| | | | | | parfile, wait 5 seconds after the warning(s) before scrolling on darcs-hash:20041203155935-3fd61-8bf992a86bf39e6cbc1abe21351c571815730f7e.gz
* CarpetIOHDF5/src/Output.cc: check for already existing datasets in an output ↵tradke2004-12-03
| | | | | | | | | | | file After recovery, it may happen that HDF5 output is requested at the same iteration when it had been done before. Therefore the code must check for this case, and remove an already existing dataset from the output file before it can be created anew. darcs-hash:20041203152858-3fd61-7703c468af4f84b6a737f9fa9a5272f4d2313724.gz
* CarpetIOHDF5: deprecate all parameters with '3D' in their namestradke2004-12-03
| | | | | | | | | | | | | | | | | | | | | CarpetIOHDF5 does output for grid variables of any dimensions, not only 3D. Therefore parameters with '3D' in their names have been marked deprecated now and should not be used anymore. They are still valid though but you will get a level-1 warning if you still use them. At some point in the future those deprecated parameters will be removed. So you should eventually fix your parameter files to substitute their occurances by their newly introduced counterparts (parameters of the same name but without the '3D'). CarpetIOHDF5/src/util/ contains a small perl script which can be applied to parfiles to automatically substitute old parameter names: ~/cactus/arrangements/Carpet/CarpetIOHDF5/src/util> ./SubstituteDeprecatedParameters.pl This perl script automatically substitutes deprecated parameter names in a parfile. Usage: ./SubstituteDeprecatedParameters.pl <parameter file> darcs-hash:20041203134032-3fd61-5d49fdff6c13f19772c6b441d5d558708dd88c71.gz
* CarpetIOHDF5/src/Checkpoint.cc: test 'IOHDF5::checkpoint_next' only at ↵tradke2004-12-03
| | | | | | coarse grid timesteps darcs-hash:20041203114454-3fd61-bc55e4ba30e567ddcee56f8dfb2f17253b7f1a1c.gz
* Comm.cc_restructureswhite2004-12-02
| | | | | | Made logic more transparent, using no surprise returns and no goto. darcs-hash:20041202153328-32473-cdea0f10bfc6d392cba04c712b7cb5360be1b573.gz
* Comm_SyncGFARRAY_renameswhite2004-12-02
| | | | | | At Eric's request, SyncGFARRAY -> SyncArrayGroup darcs-hash:20041202110106-32473-852dd965b09fab9bd1bb56985f7ecb4b75a16d63.gz
* typecase_INCLUDE_HEADER_INswhite2004-12-01
| | | | | | | | Made all Carpet thorns to include the "typecase" file from the Carpet thorn using the "include header: .. in .." mechanism, getting rid of explicit file paths. darcs-hash:20041201161810-32473-75f4594a4d9c19a11dbe9d07741678a84bd6f694.gz
* Initialise_decomposeswhite2004-12-01
| | | | | | | Broke this mama of a function into a couple of levels of functional parts. darcs-hash:20041201155737-32473-ce364db1ab14403a823d0bf0937ff4520ed04639.gz
* CarpetIOHDF5 testsuite: use IO::verbose instead of IOHDF5::verbosetradke2004-12-01
| | | | darcs-hash:20041201135259-3fd61-f3dd4087fd35e004b6a103f4dc604cd4a4a191b3.gz
* CarpetIOHDF5: allow individual output frequencies in option strings, give ↵tradke2004-12-01
| | | | | | some better info output for IO::verbose = "full" darcs-hash:20041201113424-3fd61-188206cd3e0ad315a9219fbc1b123af8ab5bff62.gz
* CarpetIOHDF5: use parameter IO::verbose instead of defining a separate ↵tradke2004-12-01
| | | | | | CarpetIOHDF5::verbose parameter darcs-hash:20041201113125-3fd61-81d942b3c8f1434ba56fee4f9160bc0f52d3cbae.gz
* CarpetIOHDF5: added example parfile with manual regriddingtradke2004-12-01
| | | | darcs-hash:20041201113013-3fd61-65a0913250d5c7d2c223f8579a1109d2063bc293.gz
* Carpet: Allow termination only every coarse grid time stepErik Schnetter2004-12-02
| | | | | | | | | | | | | | Check whether the termination condition holds only every coarse grid time step. This reduces the number of MPI_Allreduce calls dramatically; previously, we would call MPI_Allreduce after each iteration, even if there is no grid fine enough that anything would happen at this iteration. This also prevents possible trouble with restarting, which currently works only when the checkpointing happens after a full coarse grid time step. darcs-hash:20041202222326-891bb-5e247813a2a863ca1f7e9933dae7a547e0535aff.gz
* CarpetIOHDF5: some code cleanuptradke2004-11-30
| | | | darcs-hash:20041130150933-3fd61-b07a8e91c055082ff3ddebccf11a07d368c7b47c.gz
* data_fill_box_arraysswhite2004-11-29
| | | | | | | | | | | | | Pulled out repeated code into single routine data::fill_box_arrays. While this is an improvement, I don't like the fact that I've had to expose the ugly funciton in the data.hh interface, even as private. It was necessary because the inappropriate declaration of ibbox as an internal type (of inexplicit access) of the data class. Down the road, make a separate ibbox declaration. darcs-hash:20041129175753-32473-a0f983bbaf22ff404cd89a0db814faca0a8f0a7b.gz
* CarpetIOHDF5/src/iohdf5.cc: re-add this source file which seems to have been ↵tradke2004-11-29
| | | | | | removed during my last change darcs-hash:20041129172424-3fd61-8b02badb1fff6195996aeeb9c6bf486961789ffd.gz
* gh_recompose_decomposeswhite2004-11-29
| | | | | | Decomposed a long function gh::recompose into smaller functional parts. darcs-hash:20041129170456-32473-738fb31f621adce8d0437c01c7d250740763c54a.gz
* Commswhite2004-11-29
| | | | | | | | | Pulled several large functional ranges out of SyncGroup, leaving this function to hold the controlling logic. Got rid of evil returns in the midst of complicated code. darcs-hash:20041129153732-32473-640947f787526223897aeb3c79b2e523d0d02471.gz
* CarpetIOHDF5: recombine all processor components into a single HDF5 datasettradke2004-11-29
| | | | darcs-hash:20041129145133-3fd61-4899754da6016b6cf5ffcd483f55740f117cccc4.gz
* devolveswhite2004-11-27
| | | | | | Decomposed Evolve function into functional parts darcs-hash:20041127141445-32473-93edf246d7bfee7b53b4f9a055dcb587be9778f9.gz
* dh_tidyswhite2004-11-23
| | | | | | | | Fixed indentation Made order of function declarations more like the order they are called in. darcs-hash:20041123182640-32473-a83a83ad132b847bf8b219686e9c88ad277f411c.gz
* dh_decompose_fixswhite2004-11-23
| | | | | | Fixed a crashing bug. Je suis d[_\e9_]sol[_\e9_]. darcs-hash:20041123164102-32473-7746f9871f6e2f58f84825b8eb38416c622f5cb9.gz
generated by cgit v1.2.3 (git 2.39.1) at 2024-07-25 03:11:09 +0000