Commit to allow for building minimum set of Wasatch files when
Wasatch is not required, but Arches is.
M CCA/Components/Wasatch/Operators/sub.mk
M CCA/Components/Wasatch/sub.mk
M CCA/Components/sub.mk
M Makefile.in
M build_scripts/program.mk
- Indent properly, remove badly placed tabs.
M configVars.mk.in
M configure
M configure.ac
78 lines of code changed in 10 files:
Another tweak to the tortuous logic required to keep this house of cards
standing.
6 lines of code changed in 1 file:
Add pTemperatureGradient to the scheduleRefine.
4 lines of code changed in 1 file:
One of several commits to convert the rayLocation variable into physical coordinates. Currently,
it's defined as the cell index + some fraction of a cell. This is confusing on a multi-level grid.
Rename rayLocation() -> ray_orgin()
37 lines of code changed in 7 files:
Don't compute relocate pTemperatureGradient for AMR (for now).
4 lines of code changed in 1 file:
Put a carry forward for pTemperatureGradientLabel in ImpMPM.cc so that the
switcher will keep working.
20 lines of code changed in 4 files:
fix Wasatch RT.
13 lines of code changed in 1 file:
Add support for computing partial derivatives of properties from TabProps through the TabPropsEvaluator.
Note that this does not yet have parser support (I am not sure if we want to allow the user that much rope to hang themselves with).
176 lines of code changed in 1 file:
1. Reformulate the heat conduction code a bit so that it is consistent with
the momentum update in using the analog of a USL (update stress last) method.
That is, the temperature gradient on the particles is computed using the time
integrated temperature on the grid at the end of a timestep.
2. Implement gradient enhanced projection of the particle temperatures to
the grid. Currently this is turned on by setting
UseGradientEnhancedVelocityProjection. These should probably have their own
flags.
3. Commented out the code for the so-called momentum formulation in SerialMPM.
It was never shown to be superior, and it is currently out of date with the
update of the velocity gradient, etc. Will likely remove, eventually.
This commit will change answers and hence break the RT.
71 lines of code changed in 9 files:
Simplify particle drag adn gravity models. For constant properties utilize type constant model. This will break CQMOM RT.
33 lines of code changed in 3 files:
Removing fracture code.
0 lines of code changed in 1 file:
Removed empty space next to All so methane_RCCE runs in local tests, removed redundant methane_RCCE test.
1 lines of code changed in 1 file:
We can now use the static version of the CUDA runtime lib on Titan.
12 lines of code changed in 2 files:
Fix error message to specify correct CLA.
6 lines of code changed in 1 file:
Make a dummy index for wVel in 2D cases to pass through StaticVector. This should fix failing dbg build tests.
12 lines of code changed in 1 file:
Set a default value for UseGradientEnhancedVelocityProjection.
1 lines of code changed in 1 file:
Make use of gradient enhanced velocity projection a run time option in
SerialMPM and AMRMPM.
31 lines of code changed in 4 files:
Destroy the new varlabel. This should fix failing memory tests.
3 lines of code changed in 1 file:
Following r53735, varLabel name change boundFlux -> RMCRTboundFlux for GPU code.
9 lines of code changed in 1 file:
Add new regression tests for CQMOM interactino with stair step walls
6 lines of code changed in 1 file:
Revert "Change location of new wasatch_3p git archive."
This reverts commit 539240c65086abc81359cc1635263baea11ac42e.
4 lines of code changed in 1 file:
minor formatting cleanup
8 lines of code changed in 1 file:
Change the wall interaction model in CQMOM to approximate stair-step cartesian geometry walls as diagonal angles in the flow cells next to walls
1521 lines of code changed in 3 files:
Change location of new wasatch_3p git archive.
4 lines of code changed in 1 file:
added PIDX code to DataArchiver
425 lines of code changed in 9 files:
repair failing RMCRT Wasatch test
2 lines of code changed in 1 file:
varLabel name change boundFlux -> RMCRTboundFlux
Added a task stencilToDBLs that:
copies CCVariabble<stencil 7> values into 6 CCVariable< double>
mimics output of DO radiation model.
179 lines of code changed in 4 files:
remove unnecessary Wasatch dependencies from the ParticlesHelper.
9 lines of code changed in 2 files:
Exploratory commit, fixing problem where geometry objects are creating having no cells.
In the past, this criteria was more aggressive, and was changed to be more relaxed.
The criteria in this commit is between these two extremes.
1 lines of code changed in 1 file:
updated git lines.
added titan machine.
21 lines of code changed in 3 files:
added 2 test problems.
<< "__________________________________\n"
<< " RMCRT:Initialization (6)\n"
<< " abskg = 5\n"
<< " T = ( ( 1 + sin(4.5 * 2Pi * x) ) * 1000^4 )^ 0.25 \n"
<< "__________________________________\n";
<< "__________________________________\n"
<< " RMCRT:Initialization (7)\n"
<< " abskg = 5.0 * exp ( -2 * x - 0.1)\n"
<< " T = ( 2*1000^4 * x + 1000^4)^0.25 \n"
<< "__________________________________\n"
To use
<benchmark dir= {0,1,2} > {6,7} </benchmark>
Removed redundant benchmark tags.
74 lines of code changed in 3 files:
cosmetic changes
21 lines of code changed in 6 files:
Fix error that occurred with stand-alone tools (compute_lnorm,
lineextract). The 'bug' does not show up when runnign sus... I
believe the error occurs if queryGrid() is called more than once on a
single DataArchive.
M DataArchive.cc
In queryGrid(), if the grid has already been queried, it is possible
to short circuit out of the routine... moved these two lines:
timedata.d_patchInfo.clear();
timedata.d_matlInfo.clear();
to after the short circuit. This avoids the d_patchInfo and
d_matlInfo from being reset in this case.
0 lines of code changed in 2 files:
Added
<?xml version="1.0" encoding="ISO-8859-1"?>
to header so editors recognize these files as xml formatted
48 lines of code changed in 94 files:
Show start/end time for SpatialOps build. Update Baja compiler version number.
9 lines of code changed in 2 files:
Delete the inputs/<component>/tmp directory, which stores the modified ups files.
24 lines of code changed in 5 files:
turned an exception into a warning.
2 lines of code changed in 1 file:
level: added operator<<
DWDatabase: print()
- outputing patch or level information depending on domainType.
17 lines of code changed in 3 files:
get rid of compiler warnings in Wasatch
5 lines of code changed in 5 files:
print()
- more descriptive output.
0 Name: abskg domain ptr: 0x1b664e0 ID: 2 matl:0
0 Name: sigmaT4 domain ptr: 0x1b665a0 ID: 3 matl:0
0 Name: abskg domain ptr: 0x1b665a0 ID: 3 matl:0
0 Name: cellType domain ptr: 0x1b664e0 ID: 2 matl:0
0 Name: sigmaT4 domain ptr: 0x1b66480 ID: 1 matl:0
0 Name: cellType domain ptr: 0x1b665a0 ID: 3 matl:0
2 lines of code changed in 1 file:
convert the OneSidedDirichletBC to GPU
5 lines of code changed in 1 file:
Ben Isaac, Derek Harris, and Tony Saad: provide a slightly more robust solution to the fix that Ben and Derek proposed for the exploding memory usage on restarts. The fact that there is no legitimate load balancer when (previously) assigning BCs on restarts causes every MPI rank to see all other patches in the domain and assign BCs to all those patches. With the solution committed here, do not assign BCs when the load balancer is NULL. instead, postpone the BC assignment to after the load balancer has been properly set. This commit is 100% backwards compatible and passes the Wasatch RT.
17 lines of code changed in 6 files:
Eliminated a few steps getting started.
cd opt/StandAlone
make link_orderAccuracy
orderAccuracy/runTests
244 lines of code changed in 5 files:
Adding back in the Collide_AMR test.
1 lines of code changed in 1 file:
Fixed a one character bug. I can barely believe this code even ran before,
setting all particleIDs on a patch to the same value.
1 lines of code changed in 1 file:
Fix missing separator complaint from make
2 lines of code changed in 1 file:
masterScript: The user's path is now at the beginning of the path.
RMCRT.m RMCRT_wrapper
- pass in the path to the postProcessing tools so dat files can be used.
*.tst
- Filled in. Number of rays is varied in all instances.
whatToRun.xml
- added other benchmark tests.
62 lines of code changed in 3 files:
fix and enable Upwind and Extrapolant operatorson GPUs
16 lines of code changed in 5 files:
Moved ups file to input/Arches/RMCRT. Fixed hardcoded path to tables. Minor formatting.
Created symbolic links to the ups files. Now all ups files are in one spot.
Moved tst files to test_config_files/Arches/RMCRT
5 lines of code changed in 11 files:
Added 3 benchmarks for verification of radiation models to orderAccuracy suite.
Benchmarks were generated using a semi-analytical solution to heat flux divergence solution
for three cases.
1. sinusoidal temperature field and constant absorption coefficient with walls
at 1500K.
2. linear temperature field with exponential decay of absorption coefficient.
with walls at 0K.
3. linear temperature field with exponential decay of absorption coefficient
with walls at 0K and a small with a hot square at 1000K.
This commit added a new arches radproperties inheritance class called "special" for the purposes of these tests.
3017 lines of code changed in 22 files:
simplify the evaluation of the Extrapolant operator
4 lines of code changed in 1 file:
Switch wall checking in CQMOM from cellType to volfraction
101 lines of code changed in 2 files:
Missed adjusting this file for the introduction of the Alex Brown Soot Model in Arches.
20 lines of code changed in 1 file:
The new location of the Wasatch3P repos is still causing problems on systems with git versions that do not support an inline config option. This should fix this.
4 lines of code changed in 1 file:
This will hopefully fix the ssl cert issues that arose on a few of the build/test platforms after my changes yesterday.
My best guess at this point is that the issue was due to outdated ssl libraries on those platforms.
4 lines of code changed in 1 file: