[root]/Core/Grid
BoundaryConditions
(23 files, 4039 lines)
Variables
(46 files, 7493 lines)
Author | Changes | Lines of Code | Lines per Change |
---|---|---|---|
Totals | 114 (100.0%) | 2175 (100.0%) | 19.0 |
ahumphrey | 7 (6.1%) | 1125 (51.7%) | 160.7 |
dav | 4 (3.5%) | 235 (10.8%) | 58.7 |
qymeng | 10 (8.8%) | 232 (10.7%) | 23.2 |
guilkey | 9 (7.9%) | 221 (10.2%) | 24.5 |
jsutherland | 61 (53.5%) | 186 (8.6%) | 3.0 |
tsaad | 14 (12.3%) | 125 (5.7%) | 8.9 |
harman | 5 (4.4%) | 51 (2.3%) | 10.2 |
jas | 4 (3.5%) | 0 (0.0%) | 0.0 |
Reverting the use of Boost to perform file and directory copies for restarting.
0 lines of code changed in 4 files:
This fixes the problem Todd was seeing with very large domains.
M Core/Grid/Grid.cc
- Fix the problem with precision. Before, for large numbers, adding epsilon
did not change the number (due to not enough precision). See in-code
comments explaining this fix.
- Alphabetize the #includes.
M R_Tester/toplevel/startLocalTest
- Clean up (add white space) some of the error messages.
- Add the ability to point at a 'gold standard' TestData directory that is
in a different location then the current directory. To use this feature,
set the TEST_DATA environment variable with an absolute path to the
gold standard directory of choice.
M R_Tester/toplevel/generateGoldStandards.py
- Clean up output message.
M R_Tester/helpers/selectComponents.sh
- Print out a list of possible test components to make it easier for the user
to set the TEST_COMPONENTS environment variable.
53 lines of code changed in 2 files:
Commit made on behalf of Abhishek.
This is a step toward Wasatch GPU execution.
0 lines of code changed in 2 files:
Cleaned up the .cc file, added better comments to the .h file.
22 lines of code changed in 2 files:
Copy the correct number of bytes back from the device in postD2HCopies
Cleanup some comments in Task.h
3 lines of code changed in 1 file:
Adding Andy Tonge's implementation of run time specification of domain freezing.
Access this feature by adding, e.g.,:
<cpdi_lcrit> 1.25 </cpdi_lcrit>
to the MPM (flags) section of an input file.
199 lines of code changed in 5 files:
I am backing out Dav's recent changes entirely on Grid.cc. These broke Wasatch even after my first attempt at repairing them.
14 lines of code changed in 1 file:
Here is my shot at fixing what Dav was trying to do.
Really, we should be using something like numeric_limits here.
6 lines of code changed in 1 file:
M CCA/Components/DataArchiver/DataArchiver.h
M CCA/Components/DataArchiver/DataArchiver.cc
M Core/Grid/Task.h
- Fix some indentation - add white space to make reading easier.
M CCA/Components/Arches/BoundaryCondition.cc
- Validate that all faces have been specified.
M Core/Grid/Grid.cc
- Fix for adding 'epsilon' to large numbers.
M Core/Grid/Variables/VarLabel.h
M Core/Grid/Variables/VarLabel.cc
- Remove 'using' syntax from .h file.
- Indent and use white space for easier reading.
- Use 'd_' for class variable: d_defaultCompressionMode
- Swat some evil 'endl's. :)
M Core/Math/Short27.h
M Core/Math/Matrix3.h
- Added some white space.
182 lines of code changed in 2 files:
This one has been a long time coming.
Rip out the standard namespace opening that has been polluting header files throughout the framework. This crept into all sorts of places.
DO NOT open namespaces in header files. Ever.
141 lines of code changed in 51 files:
More namespace cleanup. Grrr!
14 lines of code changed in 2 files:
to_string is in both the SCIRun and std namespaces. Due to some abuse in opening up namespaces in Uintah, some clashes were occurring on the new mac platforms. This should eliminate those clashes.
0 lines of code changed in 2 files:
Add support for post GPU kernel processing callback.
A GPU task will be called multipule times and each time an event argument will be passed.
The task function can then use the event to decide what to run when
1) Task::CPU task is scheduled to run on CPU
2) Task::GPU task is scheduled to run on GPU
3) Task::postGPU task is finished execution on GPU, DW has been copied back.
CPU only task interface is not unchanged.
148 lines of code changed in 2 files:
1. Fix a typo in the function faceTypeToString that I added last week.
2. Allow one to access a boundary face iterator when no boundary conditions have been specified at that boundary face.
2 lines of code changed in 2 files:
Cleanup misleading typo in Task class code:
"void* device" -> "void* stream"
35 lines of code changed in 2 files:
Further interface updates to the new BC treatment in Wasatch. Today's version is based on Boundaries instead of Boundary-Conditions. As a result, a new Boundary Type was added to the Face spec in Uintah. This allows us to reason on boundary types and select appropriate boundary conditions.
39 lines of code changed in 2 files:
Cleanup old GPU interface.
0 lines of code changed in 2 files:
Implemented a lightweight version of GPUDataWarehouse, GPUGridVariable and GPUArray3D inside GPU.
This commit included a new set of interfaces for better supporting GPU task.
See example in Examples/UnifiedSchedulerTest
28 lines of code changed in 2 files:
An "in development" version of cpdi that allows the particle domains to be
frozen at some compile time chosen factor, relative to their initial size.
This variable is lcrit and it must be set in multiple locations in the code
to have the same value.
0 lines of code changed in 1 file:
rename the modifies with ghost cells to modifiesWithScratchGhost. Thanks to Guilkey and Thornock for the naming-inspiration.
24 lines of code changed in 2 files:
pay respect to the Uintah guards and create a separate interface for Task::modifies that uses ghost cells with a big warning next to it.
11 lines of code changed in 1 file:
added bulletproofing to catch situation where the number of fine cells used to compute the
coarse cell != the product of the refinement ratio.
21 lines of code changed in 1 file:
pay respect to the Uintah guards and create a separate interface for Task::modifies that uses ghost cells with a big warning next to it.
33 lines of code changed in 2 files:
allow task::modifies to specify ghost cells.
14 lines of code changed in 2 files:
Removed the last remnants of the "addMaterial" code. It hasn't been used for years.
0 lines of code changed in 2 files:
clean up some warnings coming out of various places in Uintah.
0 lines of code changed in 1 file:
Complete the implementation of findCellNodes27. I have no idea why this wasn't done previously.
If anyone is using this (MPM), please look over my changes to see if you see any problem.
11 lines of code changed in 1 file:
More descriptive variable names
outputInv -> outputInterval
checkpointInv -> checkpointInterval
updateCheckpointInterval()
using d_checkpointInterval instead of d_outputInterval in conditional
30 lines of code changed in 2 files:
Fixed a memory leak
2 lines of code changed in 1 file:
Allow user defined task to change outputInterval and checkpointInterval
Example:
schedule:
task->computes(shareState->get_outputInterval_label())
shareState->updateOutputInterval(true);
task function:
newdw->put(min_vartype(newvalue), shareState->get_outputInterval_label());
28 lines of code changed in 2 files:
Beginnings of generalized "device" tasks and work queues.
This will ultimately provide a unified approach to both GPU and MIC tasks.
1087 lines of code changed in 2 files:
Move the calculation of velocity gradient and deformation gradient out of the constitutive models,
and just have one version of the code to maintain in SerialMPM.
Other changes:
1. Add the ability to specify a number of subcycles in the calculation of F in the input file. Use the MPM Flag "minimum_subcycles_for_F"
2. Add the option to use an approximation to the exponential to compute the increment in F. Use a negative value for "minimum_subcycles_for_F", with the absolute value being the number of terms in the TS approximation.
3. Disable support for Fracture. Most fracture code persists, but won't work. It will be removed later.
This will definitely break the regression tester, but the differences have been verified to be small, or in some cases limited to the number of checkpoint variables changing (pTempPrevNew has been removed, whose brainchild was that again?).
0 lines of code changed in 1 file:
Remove reduction of delT on individual levels.
Only compute a global delT on grid.
This will save one MPI_Allreduce on each level.
3 lines of code changed in 1 file:
I think that Qingyu meant to call this setExtraCells instead of setExtraCell.
2 lines of code changed in 3 files:
Add preGridProblemSetup() for simulation component to allow problem setup before grid setup.
Add setExtraCell() to Grid class, this method can only be called when grid has not setup.
23 lines of code changed in 2 files: