Activity by Clock Time
| Directory | Changes | Lines of Code | Lines per Change |
|---|---|---|---|
| Totals | 331 (100.0%) | 20353 (100.0%) | 61.4 |
| Packages/Uintah/CCA/Components/Models/Radiation/ | 94 (28.4%) | 9421 (46.3%) | 100.2 |
| Packages/Uintah/CCA/Components/Models/Radiation/fortran/ | 197 (59.5%) | 8458 (41.6%) | 42.9 |
| Packages/Uintah/CCA/Components/Models/test/ | 5 (1.5%) | 1078 (5.3%) | 215.6 |
| Packages/Uintah/CCA/Components/MPMArches/ | 6 (1.8%) | 544 (2.7%) | 90.6 |
| Packages/Uintah/CCA/Components/Arches/ | 9 (2.7%) | 520 (2.6%) | 57.7 |
| Packages/Uintah/CCA/Components/MPMArches/fortran/ | 6 (1.8%) | 192 (0.9%) | 32.0 |
| Packages/Uintah/CCA/Components/Arches/Mixing/ | 12 (3.6%) | 124 (0.6%) | 10.3 |
| Packages/Uintah/CCA/Components/Models/ | 2 (0.6%) | 16 (0.1%) | 8.0 |
Extended to calculation without table (gets temperature from ICE, and
hardwires other inputs)
Now can do radiation every N timesteps
Added lots of comments
409 lines of code changed in 8 files:
Indentation
189 lines of code changed in 6 files:
1.
Fixed bug in nongray calculations with Spherical Harmonics.
Previously this would not run in parallel. For this, needed to
change all requires to requires with one ghost cell (in schedule
intensitysolve). Also, in same function, changed abskg, esrcg,
and shgamma to requires from modifies. These are now getCopied
with one ghost cell each.
2.
Fixed bug in m_radwsgg, so that now corners and edges of patches are not
accessed...was causing a floating point exception.
3.
Some spacing and indentation, and garbage removal.
4.
Still need to parallelize fluxes in spherical harmonics.
61 lines of code changed in 7 files:
1. Removed hardwiring of radiation solver switches in linear solvers;
now done dynamically according to radiation solver choice and linear
solver choice and specified in input file
2. Automated selection of linear solver and preconditioner choices
3. Print warnings and throw exceptions for certain solver choices
100 lines of code changed in 2 files:
"Make it so" that PETSc/HYPRE knows about the logical for spherical harmonics,
and sets a numerical parameter correctly. HYPRE parameter is set correctly;
PETSc is still hardwired. Minor printout change in rdombmcalc.
Minor indentation change in RadiationModel.
50 lines of code changed in 9 files:
Changed names of subroutines (changing filenames is not enough! :-) in
fortran directory of Models/Radiation from x to m_x to avoid further
conflicts with Arches/MPMArches
23 lines of code changed in 21 files:
Made mmWallCellTypeInit require cellType from old DataWarehouse,
even when it is not needed on restart. This is necessary because
otherwise cellType is never required on restart, causing it to have
material none, which is why the restart was failing.
Now MPMArches restart tests should pass on the checkpoints (once
the GS have been updated). (Thanks, Bryan!)
6 lines of code changed in 1 file:
Changed name of subroutine from fixradval to m_fixradval. The use of
fixradval was causing a conflict with the Arches fixradval and therefore
causing MPMArches tests to fail to complete. Now they should complete.
I will update the GS when the tests run next.
0 lines of code changed in 6 files:
Fixed bug in previous commit...we need cellType for boundaries,
so created a cellType variable in DW. Initialized and used
as a constCCVariable afterwards. BCs are now correctly computed.
Temperature is now a constCCVariable in scheduleIntensitySolve.
rdombmcalc and rdombc now use cellType. pcell is a const in rdomsolve
and rshsolve, rshresults, and radcal.
radcal and radwsgg now use temperature as a constCCVariable.
Replaced xx, yy, and zz in rsh* by dxep, dxpw, etc.
Activated rdombmcalc for P1.
Added RadiationConstVars to boundaryCondition.
Added label for intensity. This is needed for parallelization of
P1. More work needs to be done here, including splitting of the
P1 calculation into a solve step and a flux/source calculation step.
Removed some unnecessary variables in the fortran routines, such
as volq and src in rdomsolve.F; these are done in separate routines
now.
Split rdombc into two parts: one for fire calculations and another,
as a modification, for test problems. Will later separate test
problems altogether from fire stuff (insofar as problem specification
is concerned).
Added absolute norms to rdombmcalc, so that we can compare
parallel results with single processor results.
Made arean, areat double instead of OffsetArray1.
Some indentation.
Code works for both DO and P1, for both test problem 1 and
actual fire. Results for DO compare well with Arches results;
still have to check P1 results. Works with both PETSc and HYPRE.
777 lines of code changed in 30 files:
Fixed uninitialized variables.
Fixed bug in RadiationDriver constructor.
Added logic for test problems.
Commented out wall stuff in rdomsolve. Keeping said stuff for
future use. Also commented out some pcell ne ffield stuff in
rdomsolve. Need to examine this in a little bit more detail.
Added metrics in rdombmcalc to check for test problem results in
multiprocessor case.
Code now runs in parallel, and produces correct results for
test problem. Need to check results in fire case.
235 lines of code changed in 12 files:
Some quick fixes for case to run in parallel. Still tentative...
need to understand what is going on.
26 lines of code changed in 3 files:
Fixed bugs. Radiation now appears to be running in serial mode.
31 lines of code changed in 5 files:
Renamed classes and fortran routines to avoid conflicts with
Arches. Arches now runs.
7482 lines of code changed in 118 files:
All Arches Radiation modules incorporated into Models infrastructure, in a form
that can be used by ICE. Everything compiles.
8512 lines of code changed in 66 files:
NonAdiabaticTable: at present just a copy of AdiabaticTable, to be used
for calculations with heat loss (radiation)
1077 lines of code changed in 4 files:
Remove tablename from names of dynamically created variables using input
file specification
1 lines of code changed in 1 file:
stuff for sulfur chemistry
283 lines of code changed in 16 files:
Calculate mmCellType_MPM, mmCellType_CutCell in addition to mmCellType,
based on void_frac_MPM and void_frac_CutCell, respectively (BoundaryCondition.cc)
Added access function and set variable for d_cutCells (BoundaryCondition.h)
Moved setInitialGuess to be near schedule_setInitialGuess (ExplicitSolver.cc)
Minor stuff in Properties.cc
355 lines of code changed in 4 files:
Read in walls from Jeremy's geometry
192 lines of code changed in 6 files:
Read in walls from Jeremy's geometry
Streamlined logic:
void_frac calculation is done now after calculating void_frac_MPM and void_
frac_CutCell
Stairstepping is removed from MPMArches; will be done exclusively in
Arches->BoundaryCondition::mmWallCellTypeInit.
Created void_frac_MPM, void_frac_CutCell, mmCellType_MPM, and mmCellType_CutCell
Copying void and solid fractions from one step to another is done separately
(independent of void fraction selection)
Changed bc on nbars and epsg from zero Dirichlet to zero Neumann
544 lines of code changed in 6 files: