aboutsummaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAge
* fix typo in name of include fileHEADmasterrhaas2013-08-16
| | | | git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@526 b61c5cb5-eaca-4651-9a7a-d64986f99364
* add missing file Vectors.ccrhaas2013-08-04
| | | | git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@525 b61c5cb5-eaca-4651-9a7a-d64986f99364
* add code to align allocated memory for vectorizationrhaas2013-08-02
| | | | git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@524 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Catch internal inconsistency with too few grid points per processeschnett2013-01-08
| | | | | | | Check that there are enough grid points to send while synchronizing. git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@519 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Move MPI support from flesh to a thorneschnett2012-07-21
| | | | | | | | | | Move MPI support from flesh to thorn ExternalLibraries/MPI. This also requires thorns that call MPI directly to declare this in their configuration.ccl. Existing configurations using MPI need to include ExternalLibraries/MPI into their thorn list. git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@516 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Introduce cctk_ash, retire cctk_lssheschnett2012-07-04
| | | | | | | | | | | | Introduce cctk_ash, describing the process-local array shape that has been allocated. This may be larger than cctk_lsh, the process-local shape that should be used. Retire cctk_lssh and related infrastructure to handle staggered grid functions. git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@515 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Initialise cGroupDynamicData.lssh, if presenteschnett2012-06-11
| | | | git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@514 b61c5cb5-eaca-4651-9a7a-d64986f99364
* reduce warn level for fallback in topology method. The 'new' method is ↵knarf2011-09-21
| | | | | | expected to fail in some cases, so this should not be of level COMPLAIN. This level was meant for development, but this change is now old enough git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@509 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Correct synchronisation erroreschnett2011-07-06
| | | | | | | | Correct error checking logic that would otherwise prevent group 0 from being synchronised. git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@508 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Add missing #include "cctk_Arguments.h"eschnett2011-01-21
| | | | git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@506 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Correct function signatureseschnett2011-01-21
| | | | git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@505 b61c5cb5-eaca-4651-9a7a-d64986f99364
* silence compiler warningknarf2010-10-12
| | | | git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@502 b61c5cb5-eaca-4651-9a7a-d64986f99364
* check return value of malloc and free space for all casesknarf2009-09-19
| | | | git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@499 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Make sure we do not call malloc(0). This should not be harmful, but better ↵knarf2009-09-18
| | | | | | be safe. git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@498 b61c5cb5-eaca-4651-9a7a-d64986f99364
* bugfix for the "automatic" topologyy generator:knarf2009-09-18
| | | | | | | | | | | | | | | | | | | | | | | | | | It used to fall back to the old method, when some of the spcifications for nprocs have already been set, because it should not overwrite something e.g. specified by parameters. If this is not the case, nprocs is usually initialized with 0es. This was good. PUGH also tries to be smart and for very small extends in one direction (smaller than 2*ghostsize) it presets nprocs in that direction with 1. This is also good. Now, PUGH presets this before the "automatic" topology generator kicks in, which means this sees the "1" and falls back to the old method. This old method however has problems with exactly this type of grid and gets it wrong (overwriting that "1"). This patch now lets the automatic method only fall back to the old method if it arrives at a different answer than preset values (if set). (This fallback is neccessary because it cannot handle all cases which the old method does.) This fixes the problem because the automatic method also arrives at the preset value of "1" (and can set the other dimentions which have not been preset). git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@497 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Do not set up topologies for lower-dimensional grid functions if theseschnetter2009-01-28
| | | | | | | | | are not allowed, as indicated by the flesh parameter Cactus::allow_mixeddim_gfs. This prevents problems if such topologies do not exist, e.g. on small grids such as 3x3x200. git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@493 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Include Licence in READMEknarf2009-01-27
| | | | git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@492 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Unify README file formatknarf2009-01-26
| | | | git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@491 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Fixing error messages for PUGH_QueryGroupStorage(), when it is passed anrideout2009-01-18
| | | | | | | invalid group. git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@490 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Correct error in error checking: malloc can return NULL if 0 bytes areschnetter2008-03-20
| | | | | | | allocated. git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@489 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Use MPI_COMM_WORLD as communicator if a NULL pointer was passed for the cGHtradke2007-12-06
| | | | | | | | | argument into PUGH's overloadable for CCTK_Barrier(). This allows CCTK_Barrier() to be called even before a grid hierarchy exists (eg. during PARAM_CHECK). git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@488 b61c5cb5-eaca-4651-9a7a-d64986f99364
* * Cast the expressions for the sizes of malloc calls to size_t. This fixes ↵rideout2007-04-10
| | | | | | | | | PR 2088. * Check if malloc returns a null pointer, and if so die with a level 0 warning. git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@487 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Fixed bug in topology setup for dimensions with zero grid points.tradke2006-11-29
| | | | git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@486 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Check that the user-specified the number of processors sums up to the totaltradke2006-08-03
| | | | | | | number of processors available only for topology dimensions > 1. git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@485 b61c5cb5-eaca-4651-9a7a-d64986f99364
* If the user specified the number of processors for each dimensiontradke2006-07-31
| | | | | | | check that they sum up to the total number of processors available. git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@484 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Fix description of PUGH::processor_topology_[123]d_* parameters.tradke2006-07-31
| | | | git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@483 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Provide new overload function for CCTK_SyncGroupsByDirI().tradke2006-07-06
| | | | | | | | | | | Do not overload CCTK_SyncGroup() anymore by default. This patch requires the most recent changes in the flesh API to provide the new overloadable function CCTK_SyncGroupsByDirI(), as discussed in thread http://www.cactuscode.org/old/pipermail/developers/2006-June/004933.html. git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@482 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Applied patch from Erik Schnetter: rename PUGHi_GenerateTopology totradke2006-06-23
| | | | | | | | PUGH_GenerateTopology an export it to other thorns. git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@481 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Erik's patch to "Make PUGH's memory statistics work again".tradke2006-06-15
| | | | | | | | | See http://www.cactuscode.org/old/pipermail/patches/2005-November/000118.html for the start of the thread. The final patch is not available on the patches mailing list but can eventually be refered to via http://www.aei.mpg.de/~jthorn/spool/pugh-memstats-2.diff. git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@480 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Registering the TraditionalTopology routine as "manual" so you can set ↵dstark2006-03-07
| | | | | | processor_topology="manual" and set processor_topology_* parameters. git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@479 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Added missing include of <stdlib.h> (for calloc).tradke2006-01-12
| | | | git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@478 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Don't make PUGH::physical2logical and PUGH::processor_topology restrictedtradke2005-12-19
| | | | | | | | parameters (but private instead) so that PUGH and Carpet can still be compiled together. git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@477 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Minor change for Pelican (AIX): included stdlib.h for qsort.dstark2005-12-15
| | | | git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@476 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Changes from Dylan Stark to allow registration of topologygoodale2005-12-14
| | | | | | | | | | | | | | | | | | | | routines and introducing a physical to logical processor mapping. Now it is possible to register custom topology routines if you don't like the two ones we have at the moment (the traditional one and the new one from Frank Loeffler). You can also provide routines to optimise the placement of neighbouring processes on the processors by giving a custom mapping from the numbers PUGH uses for the processors to the actual MPI processor. Please treat this registration as semi-volatile for the moment and coordinate with Dylan if you fancy using it. git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@475 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Correct email address in copyright statement.schnetter2005-10-22
| | | | git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@474 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Moving parameters into their correct scope sections (based upon therideout2005-10-04
| | | | | | | | comments), and merging two private parameter sections. This commit is purely cosmetic (and fixes incorrect comments). git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@472 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Apply Frank Loeffler's patch to improve the processor distribution ofschnetter2005-09-26
| | | | | | | | | | | | | | | | | | | | | | | | PUGH: Up to now, PUGH does not take the actual grid size into account while setting the number of processors/domains in each direction. This gives problems, e.g. using boxes like 200x7x7 (with 3 ghostzones). It also does not really do a good job in situations with non near-cubic domains (like in bitant mode it devides first in the z-direction, which is usually the smallest). I inserted another decomposition algorithm which is looking for the longest direction and devides this if possible. Because it was done to be easy to understand, it can fail in certain situations, in which it is gracefully falling back to the old decomposition (e.g. a cubic box and 9 processors). To include this algorithm I added a keyword "automatic_old" to PUGH::processor_topology. The new behaviour is the default; this keyword re-enables the old behaviour. git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@471 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Improve periodic boundaries on small domains: allow the number ofschnetter2005-09-26
| | | | | | | | | | | | ghost zones to be larger than the number of interior points if there is only one processor in the corresponding direction. This is necessary to allow two ghost zones when there is only one interior grid point, which is often used to set up lower-dimensional domains. Second part of this patch. git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@470 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Improve periodic boundaries on small domains: allow the number ofschnetter2005-09-26
| | | | | | | | | | ghost zones to be larger than the number of interior points if there is only one processor in the corresponding direction. This is necessary to allow two ghost zones when there is only one interior grid point, which is often used to set up lower-dimensional domains. git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@469 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Insert several checks whether variable indices passed into routinesschnetter2005-09-26
| | | | | | | are legal, and whether malloc returned a null pointer. git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@468 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Using neighbours to correctly setup bbox and ownership arrays.dstark2005-06-20
| | | | git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@466 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Set cctk_bbox to zero for periodic boundaries.schnetter2005-04-18
| | | | | | | This closes PR 483. git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@465 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Take group vector length into account when deciding how to allocateschnetter2005-03-24
| | | | | | | storage git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@464 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Allow realloc to return NULL for a zero-sized requestschnetter2005-03-24
| | | | git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@463 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Report an error for GroupDynamicData for variables with zeroschnetter2005-03-24
| | | | | | | variables, because PUGH cannot handle this case git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@462 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Correct termination criterion when dt<0schnetter2005-03-24
| | | | git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@461 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Implement the option "PUGH::local_size_includes_ghosts", which allowsschnetter2005-03-20
| | | | | | | | specifying the processor-local unghosted grid size -- this is the grid size that should stay constant for a parallel performance test. git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@460 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Take the vector length into account when allocating storage.schnetter2005-02-19
| | | | git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@459 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Changed the default for boolean parameter PUGH::padding_active from "yes" totradke2005-02-04
| | | | | | | | "no" because it doesn't seem to be doing any good on Intel processors and actually decreases performance on Opterons. git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@458 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Use CCTK_BYTE instead of CCTK_CHARschnetter2005-01-29
| | | | git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@457 b61c5cb5-eaca-4651-9a7a-d64986f99364
generated by cgit v1.2.3 (git 2.39.1) at 2024-05-04 00:36:27 +0000