aboutsummaryrefslogtreecommitdiff
path: root/src
Commit message (Collapse)AuthorAge
* Updated to use new syntax for MainLoopIndex.goodale2000-10-12
| | | | | | | | | | Fixed header files so prototype for above is included. Should now work on Hitachi machine. Tom git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@284 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Calculate the total allocated storage in MBytes.tradke2000-10-11
| | | | git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@283 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Fixed buglet in printf statementallen2000-10-09
| | | | | | | CactusPUGH-PUGH/462 git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@282 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Removed old fileallen2000-09-23
| | | | git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@281 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Correct linear index to start with when iterating over elementstradke2000-09-22
| | | | | | | of an array. If you have periodic BC this index is not 0. git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@280 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Macro name changesallen2000-09-22
| | | | git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@279 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Make the option to report on storage steerableallen2000-09-21
| | | | git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@278 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Reactivated timers.tradke2000-09-18
| | | | | | | Added routine PUGH_PrintTimingInfo() to output time spent in communication. git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@275 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Fixed type of the timer variable from double to int.tradke2000-09-18
| | | | git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@274 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Removed the first_array_in_group flag from the pGA. It's not needed anymore.tradke2000-09-18
| | | | git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@273 b61c5cb5-eaca-4651-9a7a-d64986f99364
* The connectivity and the extras structure for CCTK_ARRAY variablestradke2000-09-18
| | | | | | | are now freed in DestroyPGH(). This fixes BR 436. git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@272 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Destroy the comm buffer, connectivity, and extras structure for each group oncetradke2000-09-18
| | | | | | | before the group's variables are destroyed. git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@271 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Don't need to set first_array_in_group anymore.tradke2000-09-18
| | | | | | | Untabified source. git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@270 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Checking for a NULL cGH for MyProc and nProcs.goodale2000-09-16
| | | | | | | Tom git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@269 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Erik's changes to PUGH's communication scheme.tradke2000-09-15
| | | | | | | | | | | Now send requests aren't simply freed anymore but rather explicitely finished by calling MPI_Waitall() on all of them for a given communication. This solves the problem of MPICH and LAM on some Linux SMP machines where it needed a long time to complete a communication. git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@268 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Allocate/free array of stati for outstanding send requests.tradke2000-09-15
| | | | git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@267 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Added an array of stati to comm structure for outstanding send requests.tradke2000-09-15
| | | | git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@266 b61c5cb5-eaca-4651-9a7a-d64986f99364
* New parameter cacheline_mult which can be used to spread the cachelinesgoodale2000-09-04
| | | | | | | | | | which consecutive GFs are aligned on. The default for this parameter is better than no spreading, but still not optimal. Tom git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@265 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Added an index for arrays to the pGA and pGH which is needed to make thatallen2000-09-01
| | | | | | | | | Cache lines are associated uniquely with each timelevel of GA This unique index is used now when storage is assigned for the GA git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@264 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Fixed reductions of arrays on a per-element base which was broken aftertradke2000-08-28
| | | | | | | | my last commits. This fixes both CactusEinstein-Einstein/420 and CactusPUGH-PUGH/422. git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@262 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Forgot to free the connectivity and extras when destroying CCTK_ARRAY vars.tradke2000-08-25
| | | | | | | Thanks to Philip who found this memory leak. git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@261 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Allow ghostzone sizes of 0. This was buggy before.tradke2000-08-04
| | | | git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@260 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Added ReductionSum.c to list of sources.tradke2000-08-01
| | | | git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@258 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Changed names of registered reduction operators to betradke2000-08-01
| | | | | | | PUGH_Reduction<op>GVs() and PUGH_Reduction<op>Arrays() resp. git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@257 b61c5cb5-eaca-4651-9a7a-d64986f99364
* All reduction operators are stagger-aware now and can be applied totradke2000-08-01
| | | | | | | arrays of arbitrary dimensions. git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@256 b61c5cb5-eaca-4651-9a7a-d64986f99364
* #ifdef _cplusplus should read #ifdef __cplusplus.tradke2000-08-01
| | | | git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@255 b61c5cb5-eaca-4651-9a7a-d64986f99364
* #ifdef _cplusplus should read #ifdef __cplusplus.tradke2000-08-01
| | | | git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@254 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Backing out my change from last night, as it is now redundant.goodale2000-07-21
| | | | | | | Tom git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@252 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Fixing memory problem which can cause core dumps with timelevels. Thisgoodale2000-07-21
| | | | | | | | | is a temporary fix as some of this code needs to be redone properly. Tom git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@251 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Removing silly code.goodale2000-07-20
| | | | | | | Tom git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@250 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Stopping a coredump from DestroyPGH when thre are variables withgoodale2000-07-20
| | | | | | | | | timelevels. Tom git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@249 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Changed assert(0) to abort() in PUGH_Abort, and commented it out.allen2000-07-20
| | | | git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@248 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Compiler warning on originallen2000-07-20
| | | | git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@247 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Things picked up by purify.goodale2000-07-19
| | | | | | | Tom git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@246 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Uncommenting memory freeing on all timelevelsallen2000-07-19
| | | | git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@245 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Quick and dirty fix to prevent access to freed memory when destroying Array ↵goodale2000-07-19
| | | | | | | | | | | | groups. This is far from perfect, and seems to now not free all the memory, but is far better than a segfault. Tom git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@244 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Registered reduction routines return an error code, and split reduction.callen2000-07-19
| | | | | | | into several files git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@243 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Putting back the loop over time levels in DestroyPGHallen2000-07-18
| | | | git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@242 b61c5cb5-eaca-4651-9a7a-d64986f99364
* #ifdef _cplusplus should better be #ifdef __cplusplustradke2000-07-17
| | | | git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@241 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Changes for CCTK_Exit and CCTK_Abortallen2000-07-14
| | | | | | | | | New argument lists Added MPI_Abort to overload CCTK_Abort git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@240 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Removed old debugging statementsallen2000-07-12
| | | | git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@239 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Tidying warnings and bracketsallen2000-07-10
| | | | git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@238 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Coding style fix and code consistency change.goodale2000-07-09
| | | | | | | Tom git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@237 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Fixing assignment of too much memory.goodale2000-07-09
| | | | | | | Tom git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@236 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Immediately free MPI send requests after MPI_Isend() so that we don't havetradke2000-06-27
| | | | | | | to MPI_Wait() on them. git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@234 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Commit the derived datatype which is used to communicate CCTK_COMPLEX data.tradke2000-06-27
| | | | git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@233 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Bugfix for giving PUGH info output.tradke2000-06-27
| | | | git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@232 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Use CCTK functions to advance interation numberallen2000-06-26
| | | | git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@231 b61c5cb5-eaca-4651-9a7a-d64986f99364
* Bug fix for non-mpi variable declarationsallen2000-06-25
| | | | git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@230 b61c5cb5-eaca-4651-9a7a-d64986f99364
* PUGH_Report is now dimension independent, and reports on the decompositionallen2000-06-24
| | | | | | | of grid functions for dimensions for which there is at least 1 GF. git-svn-id: http://svn.cactuscode.org/arrangements/CactusPUGH/PUGH/trunk@229 b61c5cb5-eaca-4651-9a7a-d64986f99364
generated by cgit v1.2.3 (git 2.39.1) at 2024-07-05 14:34:10 +0000