Fix crash bug when interpolating node centered data in GatherFields.
0 lines of code changed in 2 files:
Add an IntersectTri module for self intersecting all of the triangles in a TriSurfMesh.
333 lines of code changed in 5 files:
Add a find_closest_elems call that returns all of the elements that are closest, not just one of them.
0 lines of code changed in 2 files:
Add a TriTriIntersection test.
0 lines of code changed in 4 files:
more updates and files
1195 lines of code changed in 3 files:
updated figures to reflect the addition of scheduleFinalizeTimestep.
0 lines of code changed in 4 files:
Backup current version of startTester
0 lines of code changed in 2 files:
Disable shaders if there's no GLEW
0 lines of code changed in 2 files:
Added scheduleSetBCsInterpolated( sched, patches, matls);
to the scheduleTimeAdvance for MPMICE
2 lines of code changed in 2 files:
Make the g.velocityInterp(olated) feel the effect of contact. Previously,
it was not, and that was causing single velocity contact to not give the
same answers as normal single material simulations, and thus all the contact
results were fouled up to some degree.
71 lines of code changed in 2 files:
Remove debugging print.
0 lines of code changed in 2 files:
Added a colliding bars option, problem_type=5.
0 lines of code changed in 2 files:
Added a colliding bars problem, problem_type 5.
0 lines of code changed in 2 files:
Eliminate the use of reserve when specifying the vectors need for the
LinearInterpolator.
2 lines of code changed in 1 file:
moving particle shader into Kurt package, as it seems easier to work with than the branch and I can keep it working with the rest of the trunk.
912 lines of code changed in 8 files:
Eliminate the use of reserve when specifying the vectors need for the
LinearInterpolator.
0 lines of code changed in 2 files:
Update f_theta at the same time that vol_frac is updated (after models).
8 lines of code changed in 1 file:
Check box variables more strictly.
9 lines of code changed in 1 file:
Tolerance given now will be used as absolute tolerance,
not as relative one. Name of the tag to set tolerance in input file
is res_tol (ugly, but to match Hypre solver).
3 lines of code changed in 1 file:
Add volume rendering tests for all of the shader paths.
1590 lines of code changed in 12 files:
Set gui variables when the box changes so that they cache properly. Fixes crash bug in BioTensor.
4 lines of code changed in 1 file:
Add Neumann BCs.
0 lines of code changed in 6 files:
Added documention and removed debuging output.
118 lines of code changed in 8 files:
A fairly lame 1D implicit MPM code.
330 lines of code changed in 1 file:
Fix volume rendering path for all shaded colormap2 cases.
60 lines of code changed in 1 file:
Fix volume rendering path for all unshaded ColorMap2 cases.
82 lines of code changed in 6 files:
Fix the 1D colormap path for all cases (in particular normal generation).
35 lines of code changed in 1 file:
Minor change, indenting
73 lines of code changed in 1 file:
Add guards for old version of petsc and implemented work around for
missing VecGetValues() for old version of petsc.
12 lines of code changed in 1 file:
Add guards around VecScale for argument order differences between 2.2.1
and 2.3.x.
5 lines of code changed in 1 file:
Eliminate the use of reserve when specifying the vectors need for the
LinearInterpolator.
Use single temperature field.
Add input file option for doing a linear solve of the grid temperatures
based on the particle temperatures in the interpolateParticlesToGrid().
This option is currently turned off.
Modify heat conduction algorithm so it more closely follows the FEM approach
in setting up the stiffness matrix and Q vector.
RT tests will change as a result of the algorithm modification.
483 lines of code changed in 4 files:
Remove unused variable.
0 lines of code changed in 1 file:
Determine the number of DOF surrounding a node that is on the boundary. This
information is used to zero out the rows/columns in the removeDOFHeat().
40 lines of code changed in 2 files:
Eliminate the use of reserve when specifying the vectors need for the
LinearInterpolator.
4 lines of code changed in 2 files:
Eliminate the use of reserve when specifying the vectors need for the
LinearInterpolator.
0 lines of code changed in 2 files:
Eliminate the use of reserve when specifying the vectors need for the
LinearInterpolator.
8 lines of code changed in 5 files:
Save ppms in movie mode when png support not available.
13 lines of code changed in 1 file:
Clean up compiler warnings.
2 lines of code changed in 1 file:
Move first contact step and compute stress tensor ahead of NCToCC_0,
as I think it should be. This should only change answers for MPMICE
problems that include at least two MPM matls.
3 lines of code changed in 1 file:
Clip bricks against the view so that bricks which are not in the view are not drawn at all.
7 lines of code changed in 1 file:
Added relative standard deviation to BNRStats and modified stdv formula
9 lines of code changed in 1 file:
Fix saving of all-in-one materials
2 lines of code changed in 1 file:
Corrected debug stream BNRStats calculation of stats and modified output format
Added flags to the debug stream BNRPatches.
101 lines of code changed in 3 files:
Clip bricks against the view so that bricks which are not in the view are not drawn at all.
44 lines of code changed in 5 files:
Add a number of IsoRefine regression/example networks.
886 lines of code changed in 6 files:
Compute new hex interior nodes by pushing them through the mesh basis instead of interpolating.
0 lines of code changed in 2 files:
Compute new hex surface nodes by pushing them through the mesh basis instead of interpolating.
362 lines of code changed in 3 files:
Remove unused variable
3 lines of code changed in 1 file:
Use mesh basis to compute all interior quad points.
13 lines of code changed in 1 file:
Use mesh basis to compute some interior quad points.
6 lines of code changed in 1 file:
Add support for structured quad refinement.
51 lines of code changed in 2 files:
Add patch stats to printouts
34 lines of code changed in 1 file:
Simplified input mesh type checking, fix crash bug.
0 lines of code changed in 2 files:
RHS doesn't need to be a strict ColumnMatrix anymore (but should still be Nx1)
0 lines of code changed in 2 files:
Fix SolveMatrix memory leaks.
29 lines of code changed in 1 file:
Separate out initialization part of parallel agorithm for sci SolveMatrix methods.
195 lines of code changed in 1 file:
Use get_input_handle for input ports.
3 lines of code changed in 1 file:
Number of materials is now only printed once on startup.
2 lines of code changed in 1 file:
Quite a few DynamicLoadBalancer-AMR fixesReduce DataArchiver tasks to 1 per type per proc (instead of 1 per type per var per patch)Have MPM initialization happen only on finest level
590 lines of code changed in 13 files:
Preserve widget scale when saving ChangeFieldBounds module to network and across new fields. Perhaps reset should also reset the size? Unclear as to what reset should do.
15 lines of code changed in 1 file:
Keep track of widget mode between saves. Don't reset it either (reset only covers position).
7 lines of code changed in 1 file:
Use get_input_handle and send_output_handle for ports.
0 lines of code changed in 48 files:
Use get_input_handle and send_output_handle for ports.
0 lines of code changed in 30 files:
Added a criteria to the postfixup which splits patches if they are to
big to effectivly load balance.
65 lines of code changed in 2 files:
fix sub varinclude/sci_defs/environment_testdefs.h.in
574 lines of code changed in 16 files:
fixed exception message
1 lines of code changed in 1 file:
Include changes for LIB_DIR_HYPRE
5 lines of code changed in 2 files:
scheduleTimeAdvance(()
- adjust the tasks to reflect that scheduleConservedtoPrimitiveVars()
needs to be called twice in a single timestep.
coarsen_delP
- simplified the code, added suggested change
33 lines of code changed in 1 file:
Use get_input_handle for input ports.
140 lines of code changed in 100 files:
errorEstimate()
-fixed gradient calculation. R & L cell indices are now set to c before we
offset them..ooops
13 lines of code changed in 5 files:
Forgot to rename this one for previous commit.
1 lines of code changed in 1 file:
scheduleConservedtoPrimitive_Vars()
fixed computes & requires for model variable
1 lines of code changed in 1 file:
Use get_input_handle and send_output_handle for ports.
69 lines of code changed in 15 files:
Changed function name from "explicit" to "explicit_scalar"
since IBM fortran has "explicit" name reserved. Thanks David
for pointing this out.
103 lines of code changed in 7 files:
Use get_input_handle and send_output_handle for ports.
0 lines of code changed in 48 files:
Debug Commit for Bryan
Changes: Formatting, timers, saftey layer fix
81 lines of code changed in 3 files:
fixed a bug in the seed point, made the var name understandable, and switched to a widget for the threading gui
0 lines of code changed in 10 files:
cleaned up the lib dir to be consisted with the other lib flags i.e. SOMELIB_LIB_DIR instead of LIB_DIR_SOMELIB
438 lines of code changed in 7 files:
Use get_input_handle and send_output_handle for ports.
494 lines of code changed in 76 files:
Add removeFixedDOFHeat().
5 lines of code changed in 1 file:
removed weed from fineToCoarseOperator() argument list
6 lines of code changed in 2 files:
Remove the default argument for the fillVector().
1 lines of code changed in 1 file:
Fix prototype for fillVector().
1 lines of code changed in 1 file:
Use send_output_handle to send results downstream.
133 lines of code changed in 90 files:
ICE.cc
advectAndAdvanceInTime()
- initialize mass_adv, mom_adv....etc to 00.0
AMRICE.h
fineToCoarseOperator()
- if(conserved_var) inv_RR = 1.0
if(non-conserved_var) inv_RR = 1.0/(RR.x() * RR.y() * RR.z())
AMRICE.cc
coarsen()
- small changes
Refluxing may still have some issues.
20 lines of code changed in 3 files:
Moved some tasks from scheduleFinalizeTimestep to scheduleTimeAdvance.
-scheduleFinalizetimestep: scheduled on every level
-scheduleTimeAdvance: scheduled once
34 lines of code changed in 1 file:
Added Safety Layers to BNRRegridder.
Fixed PostFixup phase to correctly split patches to minimum patch size
as needed.
68 lines of code changed in 5 files:
Check in basic for BEMTransferSrcToInf module.
319 lines of code changed in 5 files:
Fix compilation error associated with CleanupPoints. Make it public and static.
4 lines of code changed in 1 file:
Minor cleanup
6 lines of code changed in 1 file:
fix bogus data string query
8 lines of code changed in 2 files:
Re-require rho from fine level (at least for the moment)
3 lines of code changed in 1 file:
fix dyn comp on 32 bit linux
28 lines of code changed in 1 file:
fillVector() now has a flag for either inserting or adding the value to the current contents.
Add a removeDOFHeat() for the removal of DOF for the implicit heat algorithm.
228 lines of code changed in 4 files:
Get Glew to work with Windows
Have Environment query values for installed SCIRun rather than compiled-in values.
Fix windows compile
619 lines of code changed in 23 files:
fillVector() now has a flag for either inserting or adding the value to the current contents.
Add a removeDOFHeat() for the removal of DOF for the implicit heat algorithm.
4 lines of code changed in 1 file:
Don't use the vector.reserve() for allocating space, just specify the size
in the constructor. This ensures that the vector.size() will be set.
111 lines of code changed in 22 files:
Fix compilation error with size_type to int casts.
0 lines of code changed in 2 files:
Fix rows update bug.
0 lines of code changed in 2 files:
Minor cleanups.
0 lines of code changed in 2 files:
Fix rows/cols numbers in the range widget to not be fixed at 100 (use input matrix size)
2 lines of code changed in 1 file:
*debug commit for Bryan
Changes:
min patch size is now constant on all levels
the main regrid function has been overhauled to potentially prepare
for safety layers (this part is may be reverted depending on how
saftey layers are handled).
97 lines of code changed in 2 files:
Add some more includes to LinearAlgebra.h, including math.h.
0 lines of code changed in 2 files:
Fix compiler warning.
0 lines of code changed in 2 files:
fix for compiler bug with dynamic_cast
4 lines of code changed in 1 file:
outputProblemSpec()
-mimic flow control logic in problemSetup.
10 lines of code changed in 1 file:
compile fix
0 lines of code changed in 2 files:
Updated configure so that the default thirdparty path is now 1.25.4 which is required.
287 lines of code changed in 2 files:
Fix FieldToNrrd conversion of tensor data. It was converting tensors to doubles and labeling them as floats. Convert to floats again.
0 lines of code changed in 4 files:
Fix crash bug when trying to draw non tensor field with tensor display turned on in showfield.
6 lines of code changed in 2 files:
Better exception message when the number of exchange coefficients is incorrect.
New message:
0 Caught exception: A ProblemSetup exception was thrown.
../src/Packages/Uintah/CCA/Components/ICE/ICE.cc:5592
The number of exchange coefficients (momentum energy) is incorrect.
Here is the correct specification:
0->1, 0->2, 0->3,
1->2, 1->3,
2->3,
41 lines of code changed in 1 file:
Fix two connections to the same ChooseNrrd module's input port.
0 lines of code changed in 2 files:
Detect attempting multiple connections to the same input port.
0 lines of code changed in 4 files:
- gcc on the MacBooks requires pointers to GL variables to have explicit GL types
38 lines of code changed in 16 files:
Make copmare_uda not depend on Core/Volume
1 lines of code changed in 1 file:
removed doICEOnLevel()
ICE is performed on all levels all the time.
53 lines of code changed in 5 files:
BioTensor fixes
11 lines of code changed in 4 files:
Make hex refinement convex before doing the refinement.
225 lines of code changed in 2 files:
Added an additional call to
scheduleConservedtoPrimitive_Vars()
right after AdvectAndAdvanceInTime(). You need primitive variables
for refineCFI() and you need to setBC on the coarse level. This task
is now called twice, once after advectAndAdvance() and also inside of
scheduleFinalize(). Answers are still not right.
75 lines of code changed in 4 files:
Refine elements of hex inputs with constant basis.
100 lines of code changed in 2 files:
Extend IsoRefine to accept lattice and structured hex inputs.
404 lines of code changed in 2 files:
Cleanup includes
0 lines of code changed in 1 file:
Added patches_per_level_per_proc input to the BNR regridder
3 lines of code changed in 1 file:
Fix compilation errors.
3 lines of code changed in 1 file:
Use GLEW to determine graphics capabilities. This makes GLEW a requirement.
326 lines of code changed in 4 files:
Add test networks for IsoRefine.
292 lines of code changed in 2 files:
Fix port name
1 lines of code changed in 1 file:
Fix crash bug for constant basis rendering of 3D fields.
4 lines of code changed in 1 file:
Reworked communication structure to avoid searching queues for
completed communication. Now waitsome is called instead to complete
communication.
958 lines of code changed in 4 files:
Move data transmitter sources and code cleanup.
0 lines of code changed in 14 files:
Added a post fixup phase which splits the largest patches until the
desired number of patches are reached. This should help the load
balancer.
107 lines of code changed in 5 files:
Fixed a bug that caused serial runs to produce one big patchset
19 lines of code changed in 2 files:
Serial runs should no longer use MPI
238 lines of code changed in 4 files:
Fix flaot compile error
1 lines of code changed in 1 file:
BNR when flags disperse
40 lines of code changed in 1 file:
added dependency for the Core/Algorithms/Math lib
0 lines of code changed in 2 files:
Round two to make this more streamlined
210 lines of code changed in 3 files:
Realized that the streamlineIntegrators should be basic integrators and moved the streamline colinear point clean up back to the module
449 lines of code changed in 10 files:
Fixed a bug that occurred when rank 0 had 0 flags
Fixed a bug that occured when any processor had 0 flags
26 lines of code changed in 2 files:
moved the alogrithm guts out of the module and placed it into the algorithm dir
530 lines of code changed in 5 files:
Minimum patch size now accounts for cell refinement ratio
4 lines of code changed in 1 file:
Removed spew.
0 lines of code changed in 1 file:
Fixed a bug in the patch fixer that would produce incorrect patchsets.
7 lines of code changed in 1 file:
Fixed a bug that would cause two tasks to potentially be assigned the
same tag.
15 lines of code changed in 1 file:
Change variable names to match nameing convention
559 lines of code changed in 8 files:
Big Change: Single level results shouldn't change
ICE.cc, MPMICE.cc, ICE.h & MPMICE.h
scheduleTimeAdvance()
- moved all tasks after advectAndAdvance() to scheduleFinalizeTimestep()
AMRICE.cc & AMRICE.h
coarsen(),
fineToCoarseOperator(),
reflux_applyCorrectionFluxes()
reflux_computeCorrectionFluxes()
- operate on the conserved quantities mass_adv, mom_adv, eng_adv....etc.
impAMRICE.cc
moved scheduleConservedtoPrimitive_Vars() to the very bottom of the timestep.
185 lines of code changed in 7 files:
More progress on BNR
391 lines of code changed in 4 files:
Configure will now respect the value of PETSC_ARCH set by the user. The default behavior is the same as before, but now if you install petsc in a new location, you can set
PETSC_ARCH before running configure, and it will be properly detected.
4 lines of code changed in 2 files:
Allow for quiet .d file creation
0 lines of code changed in 2 files:
M CCA/Components/Regridder/BNRTask.h
Indented, untabified, use c++ comments, pulled into 80 columns (this
only effected the spacing on comments), other general readability
updates.
Killed the namespace pollution of "using namespace std/Uintah" in the
.h file.
Still needs to have class member variables updated to use naming
convention (ie: _ on names).
M testprograms/sub.mk
M testprograms/BNRRegridder/sub.mk
Add in full list of libraries to all compilation on SGI/Mac.
M testprograms/BNRRegridder/bnrtest.cc
Use "using namespace Uintah" in the .cc file.
Also:
> const int X=700, Y=700;
Please don't use global variables that have a single letter for their
name... this makes it impossible to search quickly for them and in
general is just a bad idea. Please fix this when you get a chance.
158 lines of code changed in 6 files:
Compiles under gcc4
2 lines of code changed in 1 file:
Fix locking/hanging bug
3 lines of code changed in 1 file:
CullIntersection which *should* work for BNR Regridder
1 lines of code changed in 1 file:
Fix potential problem with incorrect UPS files, allow user to run AMR without a regridder
34 lines of code changed in 4 files:
Removed lines that clipped variances about 0.9 by 2/3.
3 lines of code changed in 1 file:
Test program for the Berger-Rigoutos regridder.
176 lines of code changed in 3 files:
Initial Berger-Rigoutos regridder (Not fully functional or tested).
1932 lines of code changed in 9 files:
Handle bad particle case correctly
9 lines of code changed in 3 files:
Try #2 at new subcycle
59 lines of code changed in 2 files:
Comment out an (99%-likely-benign) error/warning
2 lines of code changed in 1 file:
added tvar->var_adv to transported variables destructor.
1 lines of code changed in 1 file:
fixed bug in allocating >4G data on 64-bit machines.
0 lines of code changed in 2 files: