| Commit message (Collapse) | Author | Age |
| |
|
|
|
|
| |
Changes to other coeffs as well due to a different Maple version used.
|
|
|
|
| |
git-svn-id: http://svn.aei.mpg.de/numrel/AEIThorns/AEILocalInterp/trunk@60 0f49ee68-0e4f-0410-9b9c-b2c123ded7ef
|
|
|
|
| |
git-svn-id: http://svn.aei.mpg.de/numrel/AEIThorns/AEILocalInterp/trunk@59 0f49ee68-0e4f-0410-9b9c-b2c123ded7ef
|
|
|
|
| |
git-svn-id: http://svn.aei.mpg.de/numrel/AEIThorns/AEILocalInterp/trunk@56 0f49ee68-0e4f-0410-9b9c-b2c123ded7ef
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This leads to a slight change in behaviour. Currently, AEILocalInterp
traverses the list of points sequentially, and aborts when the first
error is encountered. After parallelisation, there is no fixed order
in which the points are traversed, and if several errors are
encountered, any one of the errors may be returned, not necessarily
the first. I am not aware of any thorn that would or should rely on
such an ordering.
This patch also adds "restrict" and "const" statements that may
improve performance as it gives the compiler more information about
dependencies between pointers.
git-svn-id: http://svn.aei.mpg.de/numrel/AEIThorns/AEILocalInterp/trunk@55 0f49ee68-0e4f-0410-9b9c-b2c123ded7ef
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
In AEIThorns/AEILocalInterp/src/InterpLocalUniform.c
1370 if (value_not_set)
1371 then {
1372 /* default operand will use each input exactly once, */
1373 /* but this only makes since if N_input_arrays == N_output_arrays
*/
1374 if (N_input_arrays != N_input_arrays)
1375 then {
1376 free(input_array_offsets);
1377 free(operand_indices);
1378 CCTK_VWarn(ERROR_MSG_SEVERITY_LEVEL,
The test is always going to fail and data is not going to be freed up.
Based on the comment, I think that the test should compare N_input_arrays
to N_output_arrays.
git-svn-id: http://svn.aei.mpg.de/numrel/AEIThorns/AEILocalInterp/trunk@52 0f49ee68-0e4f-0410-9b9c-b2c123ded7ef
|
|
|
|
|
|
|
|
| |
caught by Ian Hawke (with help from the Portland Group compiler :)
... the fix is to just add Yet Another { } pair ...
git-svn-id: http://svn.aei.mpg.de/numrel/AEIThorns/AEILocalInterp/trunk@47 0f49ee68-0e4f-0410-9b9c-b2c123ded7ef
|
|
|
|
| |
git-svn-id: http://svn.aei.mpg.de/numrel/AEIThorns/AEILocalInterp/trunk@46 0f49ee68-0e4f-0410-9b9c-b2c123ded7ef
|
|
|
|
|
|
|
|
|
| |
AEILocalInterp::log_interp_coords = true
causes this thorn to write a log file on each processor
giving the grid and (some of) the interpolation coordinates
git-svn-id: http://svn.aei.mpg.de/numrel/AEIThorns/AEILocalInterp/trunk@45 0f49ee68-0e4f-0410-9b9c-b2c123ded7ef
|
|
|
|
|
|
|
| |
instead of void.
git-svn-id: http://svn.aei.mpg.de/numrel/AEIThorns/AEILocalInterp/trunk@44 0f49ee68-0e4f-0410-9b9c-b2c123ded7ef
|
|
|
|
| |
git-svn-id: http://svn.aei.mpg.de/numrel/AEIThorns/AEILocalInterp/trunk@43 0f49ee68-0e4f-0410-9b9c-b2c123ded7ef
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
CCTK_ERROR_INTERP_DELTA_X_ZERO if the grid spacing is zero along one or
more axes; previously it would return the generic "I got a NaN/infinity
or other non-finite value" error code CCTK_ERROR_INTERP_COORD_NAN for
this case.
*** IMPORTANT ***
This change requires a cvs-update of the Cactus flesh (more accurately,
the file src/include/cctk_Interp.h in the flesh ) in order to define
the new error code. If you don't do this cvs update, AEILocalInterp
won't compile.
git-svn-id: http://svn.aei.mpg.de/numrel/AEIThorns/AEILocalInterp/trunk@42 0f49ee68-0e4f-0410-9b9c-b2c123ded7ef
|
|
|
|
|
|
|
|
|
|
|
| |
this would have been triggered (causing a spurious
> CCTK_InterpLocalUniform(): internal error!
> unexpected result -4 from AEILocalInterp_molecule_posn()
> at pt=0 axis=0!
if an interpolation coordinate is a NaN or other non-finite quantity
git-svn-id: http://svn.aei.mpg.de/numrel/AEIThorns/AEILocalInterp/trunk@41 0f49ee68-0e4f-0410-9b9c-b2c123ded7ef
|
|
|
|
| |
git-svn-id: http://svn.aei.mpg.de/numrel/AEIThorns/AEILocalInterp/trunk@40 0f49ee68-0e4f-0410-9b9c-b2c123ded7ef
|
|
|
|
| |
git-svn-id: http://svn.aei.mpg.de/numrel/AEIThorns/AEILocalInterp/trunk@39 0f49ee68-0e4f-0410-9b9c-b2c123ded7ef
|
|
|
|
| |
git-svn-id: http://svn.aei.mpg.de/numrel/AEIThorns/AEILocalInterp/trunk@38 0f49ee68-0e4f-0410-9b9c-b2c123ded7ef
|
|
|
|
| |
git-svn-id: http://svn.aei.mpg.de/numrel/AEIThorns/AEILocalInterp/trunk@37 0f49ee68-0e4f-0410-9b9c-b2c123ded7ef
|
|
|
|
| |
git-svn-id: http://svn.aei.mpg.de/numrel/AEIThorns/AEILocalInterp/trunk@36 0f49ee68-0e4f-0410-9b9c-b2c123ded7ef
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
a grid boundary, the actual interpolation was sometimes done as if
the interpolation coordinate was displaced by one grid spacing :( :(
also add more debugging code (introduced to help track down this bug,
but there's no harm in leaving it in for possible future use)
also add explicit checks for a couple of error cases:
* interpolation coordinate is not a finite number (eg NaN or infinity)
* grid origin/delta not a finite number (eg NaN or infinity)
* grid spacing is 0 (we need to divide by this!)
also add explicit note about the test driver in this directory
git-svn-id: http://svn.aei.mpg.de/numrel/AEIThorns/AEILocalInterp/trunk@35 0f49ee68-0e4f-0410-9b9c-b2c123ded7ef
|
|
|
|
|
|
|
|
| |
in the parameter table (default is still no debugging)
* add some additional debugging code
git-svn-id: http://svn.aei.mpg.de/numrel/AEIThorns/AEILocalInterp/trunk@34 0f49ee68-0e4f-0410-9b9c-b2c123ded7ef
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
grid (so the interpolation fails), then (if the user didn't request
us to be silent) we print a warning message.
This commit adds the text
"(this may be caused by a global interpolation with\n"
" driver::ghost_size too small)\n"
to this warning message, to cue the user about a common cause of
this problem.
This "fixes" bug #2022, where the quotation marks are because that
bug was actually reported against CactusBase/LocalInterp (but the
same problem applies to this thorn).
git-svn-id: http://svn.aei.mpg.de/numrel/AEIThorns/AEILocalInterp/trunk@33 0f49ee68-0e4f-0410-9b9c-b2c123ded7ef
|
|
|
|
|
|
|
|
|
|
|
| |
* fix the title and one other place: this is thorn AEILocalInterp,
not thorn LocalInterp
* switch to new-style latex cactus style
* a few small wording clarifications & typo fixes
* add a reference on the non-smoothness of interpolation errors
git-svn-id: http://svn.aei.mpg.de/numrel/AEIThorns/AEILocalInterp/trunk@32 0f49ee68-0e4f-0410-9b9c-b2c123ded7ef
|
|
|
|
|
|
|
|
|
|
|
| |
> AEILocalInterp/src/InterpLocalUniform.c:2058 where the return statement is
> marked as an "/*** ERROR RETURN ***/" but is actually also reached in the
> non-error case. I believe there are some curly brackets missing, this return
> statement needs to be enclosed in the preceding if-block along with the
> CCTK_VWarn() call.
git-svn-id: http://svn.aei.mpg.de/numrel/AEIThorns/AEILocalInterp/trunk@29 0f49ee68-0e4f-0410-9b9c-b2c123ded7ef
|
|
|
|
| |
git-svn-id: http://svn.aei.mpg.de/numrel/AEIThorns/AEILocalInterp/trunk@28 0f49ee68-0e4f-0410-9b9c-b2c123ded7ef
|
|
|
|
| |
git-svn-id: http://svn.aei.mpg.de/numrel/AEIThorns/AEILocalInterp/trunk@27 0f49ee68-0e4f-0410-9b9c-b2c123ded7ef
|
|
|
|
|
|
|
| |
(all inside #ifdef AEILOCALINTERP_DEBUG or AEILOCALINTERP_DEBUG2)
git-svn-id: http://svn.aei.mpg.de/numrel/AEIThorns/AEILocalInterp/trunk@26 0f49ee68-0e4f-0410-9b9c-b2c123ded7ef
|
|
|
|
| |
git-svn-id: http://svn.aei.mpg.de/numrel/AEIThorns/AEILocalInterp/trunk@25 0f49ee68-0e4f-0410-9b9c-b2c123ded7ef
|
|
|
|
|
|
|
| |
-- no code change for the normal case (#undef AEILOCALINTERP_DEBUG)
git-svn-id: http://svn.aei.mpg.de/numrel/AEIThorns/AEILocalInterp/trunk@24 0f49ee68-0e4f-0410-9b9c-b2c123ded7ef
|
|
|
|
| |
git-svn-id: http://svn.aei.mpg.de/numrel/AEIThorns/AEILocalInterp/trunk@23 0f49ee68-0e4f-0410-9b9c-b2c123ded7ef
|
|
|
|
|
|
|
| |
now that Cactus defines these
git-svn-id: http://svn.aei.mpg.de/numrel/AEIThorns/AEILocalInterp/trunk@22 0f49ee68-0e4f-0410-9b9c-b2c123ded7ef
|
|
|
|
|
|
|
| |
interpolation operator "uniform cartesian"
git-svn-id: http://svn.aei.mpg.de/numrel/AEIThorns/AEILocalInterp/trunk@21 0f49ee68-0e4f-0410-9b9c-b2c123ded7ef
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
either outside the grid, or inside but too close to the grid boundary,
to also include the grid spacings. A sample of the new msg is:
WARNING[L1,P0] (AEILocalInterp):
CCTK_InterpLocalUniform():
interpolation point is either outside the grid,
or inside but too close to the grid boundary!
0-origin interpolation point number pt=691
interpolation point (x,y)=(2.26354,-0.591776)
grid x_min(delta_x)x_max = -0.1683(0.1122)3.30989
grid y_min(delta_y)y_max = -0.448799(0.224399)6.50758
git-svn-id: http://svn.aei.mpg.de/numrel/AEIThorns/AEILocalInterp/trunk@20 0f49ee68-0e4f-0410-9b9c-b2c123ded7ef
|
|
|
|
| |
git-svn-id: http://svn.aei.mpg.de/numrel/AEIThorns/AEILocalInterp/trunk@19 0f49ee68-0e4f-0410-9b9c-b2c123ded7ef
|
|
|
|
|
|
|
|
|
|
|
| |
to be only first accurate :( :(
(it was a typo in the maple script generating the coefficients)
(I'm still investigating why my interpolator tests didn't spot this
before)
git-svn-id: http://svn.aei.mpg.de/numrel/AEIThorns/AEILocalInterp/trunk@18 0f49ee68-0e4f-0410-9b9c-b2c123ded7ef
|
|
|
|
| |
git-svn-id: http://svn.aei.mpg.de/numrel/AEIThorns/AEILocalInterp/trunk@16 0f49ee68-0e4f-0410-9b9c-b2c123ded7ef
|
|
|
|
| |
git-svn-id: http://svn.aei.mpg.de/numrel/AEIThorns/AEILocalInterp/trunk@15 0f49ee68-0e4f-0410-9b9c-b2c123ded7ef
|
|
|
|
| |
git-svn-id: http://svn.aei.mpg.de/numrel/AEIThorns/AEILocalInterp/trunk@14 0f49ee68-0e4f-0410-9b9c-b2c123ded7ef
|
|
|
|
| |
git-svn-id: http://svn.aei.mpg.de/numrel/AEIThorns/AEILocalInterp/trunk@13 0f49ee68-0e4f-0410-9b9c-b2c123ded7ef
|
|
|
|
|
|
|
|
|
|
|
| |
print a level 1 warning if we return an error code due to a point
being outside (or too close to the boundary of) the grid
if the key "suppress_warnings" is in the parameter table,
don't print these warnings
git-svn-id: http://svn.aei.mpg.de/numrel/AEIThorns/AEILocalInterp/trunk@12 0f49ee68-0e4f-0410-9b9c-b2c123ded7ef
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
< - HIGH PRIORITY: return interpolator status for each interpolated point:
< the user (optionally) provides in the parameter table, a pointer
< to a CCTK_INT array of size N_interp_points; we store the status there
< ==> this would allow Thomas Radke to fix CactusPUGH/1563
< ==> having both of these would allow me to fix a nasty AHFinderDirect bug
document that this (still) needs to be done:
> - support excision, i.e. there should be a way to tell the interpolator
> about an excision mask, and either
> * have it return an error if the default-centering interpolator molecule
> would use data from an excised region, or preferably
> * properly implement the excision_off_centering_tolerance
> and excision_extrapolation_tolerance parameters as
> documented in the thorn guide
git-svn-id: http://svn.aei.mpg.de/numrel/AEIThorns/AEILocalInterp/trunk@11 0f49ee68-0e4f-0410-9b9c-b2c123ded7ef
|
|
|
|
|
|
|
|
|
|
| |
* add the ability to return per-point error status
* drop support for returning info on which side of the grid an
"outside the grid" point is outside on (this was tricky to implement
in combination with the per-point status)
git-svn-id: http://svn.aei.mpg.de/numrel/AEIThorns/AEILocalInterp/trunk@10 0f49ee68-0e4f-0410-9b9c-b2c123ded7ef
|
|
|
|
|
|
|
|
|
|
|
|
| |
* drop support for returning status info on which side of the grid
an outside-the-grid point is outside -- this turned out to be awkward
to implement in combinatino with per-point status, and I believe it
was only used by my apparent horizon finder to produce slightly
more informative error messages in a few cases
* convert some commented-out debugging code to #ifdef on various DEBUG symbols
git-svn-id: http://svn.aei.mpg.de/numrel/AEIThorns/AEILocalInterp/trunk@9 0f49ee68-0e4f-0410-9b9c-b2c123ded7ef
|
|
|
|
|
|
|
|
| |
Cactus build process on machines with case-insensitive filenames
(eg windoze), where 'makefile' looks a lot like 'Makefile'
git-svn-id: http://svn.aei.mpg.de/numrel/AEIThorns/AEILocalInterp/trunk@8 0f49ee68-0e4f-0410-9b9c-b2c123ded7ef
|
|
|
|
|
|
|
|
|
|
|
| |
- HIGH PRIORITY: return interpolator status for each interpolated point:
the user (optionally) provides in the parameter table, a pointer
to a CCTK_INT array of size N_interp_points; we store the status there
==> this would allow Thomas Radke to fix CactusPUGH/1563
==> having both of these would allow me to fix a nasty AHFinderDirect bug
git-svn-id: http://svn.aei.mpg.de/numrel/AEIThorns/AEILocalInterp/trunk@7 0f49ee68-0e4f-0410-9b9c-b2c123ded7ef
|
|
|
|
|
|
|
|
| |
"generalized polynomial interpolation"
is deprecated, and will probably be removed at some point
git-svn-id: http://svn.aei.mpg.de/numrel/AEIThorns/AEILocalInterp/trunk@6 0f49ee68-0e4f-0410-9b9c-b2c123ded7ef
|
|
|
|
| |
git-svn-id: http://svn.aei.mpg.de/numrel/AEIThorns/AEILocalInterp/trunk@5 0f49ee68-0e4f-0410-9b9c-b2c123ded7ef
|
|
|
|
|
|
|
| |
which included commits to RCS files with non-trunk default branches.
git-svn-id: http://svn.aei.mpg.de/numrel/AEIThorns/AEILocalInterp/trunk@3 0f49ee68-0e4f-0410-9b9c-b2c123ded7ef
|
|
git-svn-id: http://svn.aei.mpg.de/numrel/AEIThorns/AEILocalInterp/trunk@1 0f49ee68-0e4f-0410-9b9c-b2c123ded7ef
|