Removed the nonlinear_solver spec. Now one only needs the <ExplicitSolver> or <PicardSolver> section.
Moved <initial_dt> and <varible_dt> into the nonlinear solver section.
9 lines of code changed in 1 file:
1) Moved EKT Solver option to ExplcicitSolver or PicardSolver section depending on the solver.
2) Moved extraProjection option to ExplicitSolver or PicardSolver section
3) Moved scalarUnderflowCheck option to ExplicitSolver or PicardSolver section
10 lines of code changed in 1 file:
Enabled tau profiling when using the dynamic scheduler
42 lines of code changed in 1 file:
Removed the <mixing_model> specification. The mixing model is now set by simply adding the appropriate mixing model section in <Properties>. Also made capitalization consistent for cold flow mixing model.
Changed "model_mixture_fraction_variance" input parameter to "use_mixing_model" which is a better description of what happens when this boolean is true & moved it into the <Properties> section.
21 lines of code changed in 5 files:
Added some more tau profiling tags
6 lines of code changed in 2 files:
indentation and whitespace changes
274 lines of code changed in 1 file:
Exit out earlier if the dependency does not apply to this processor
6 lines of code changed in 1 file:
prune out some dependencies earlier.
5 lines of code changed in 2 files:
MPM's gravity is actually applied in MPMArches (for MPMArches simulations) not in SerialMPM. Moved prior change from SerialMPM.cc to MPMArches.cc.
3 lines of code changed in 2 files:
Enable ComponentTimings:+ in serial.
179 lines of code changed in 1 file:
output individual components if running in serial.
1 lines of code changed in 1 file:
Updated the order-of-accuracy input file to reflect the new inputfile format.
9 lines of code changed in 1 file:
Arches physical constants are now combined with the Uintah_specification->PhysicalConstants section rather than CFD->Arches->PhysicalConstants. The only caveat here is that (given Jim's fix) the MPM code assumes zero gravity for MPM objects (hard coded in serialMPM.cc when ArchesMPM is used). The fluid gravity is specified in the input. If non-stationary geometry is ever used in MPMARCHES this might need revisiting.
11 lines of code changed in 3 files:
Always assume that gravity is zero in MPMArches simulations.
5 lines of code changed in 1 file:
Removed doMMS. MMS is now turned on using <MMS>.
3 lines of code changed in 1 file:
Initial conditions from external files are now set by using:
<set_initial_condition inputfile="<path to file>"/>.
Also completely removed the reacting_scalar option since it was causing restart problems.
11 lines of code changed in 1 file:
Added support for picard solver given the changes in the input file.
13 lines of code changed in 1 file:
Removed the option to track a "reactingScalar". This will be replaced by the more flexible <TransportEqns> section.
7 lines of code changed in 1 file:
<transport_enthalpy> is no longer specified. Enthalpy transport is activated by simply adding the <EnthalpySolver> block to the <ExplicitSolver> block.
2 lines of code changed in 1 file:
<transport_mixture_fraction> is no longer specified. Mixture fraction transport is initiated by simply adding the <MixtureFraction> block in the <Arches> block. Note that currently you must always transport mixture fraction. This should be remedied in the future.
4 lines of code changed in 1 file:
No longer specify <transport_extra_scalars>. The transport of extra scalars is activated buy adding the <ExtraScalars> section in the <Arches> block.
3 lines of code changed in 1 file:
Made "whichTurbModel" a private member.
6 lines of code changed in 1 file:
Moved the turbulence model specification to an attribute of the <Turbulence> section. Cuts down on input clutter.
14 lines of code changed in 2 files:
Fixed a bug related to my last commit and virtual patches.
3 lines of code changed in 1 file:
readded revision 44527 and fixed a bug in that patch.
8 lines of code changed in 1 file:
Changed "grow_dt" to "init_dt" which seems much more descriptive as to what this parameter does. The code is currently reading in (now) init_dt and then setting the time step to this value. If variable_dt is true, then the init_dt value is overwritten by the CFL condition. Otherwise, the code steps @ the initial dt, init_dt.
3 lines of code changed in 2 files:
Added a input flag to turn off the heat loss warnings. By default the heat loss warnings will still print so the user must set <loud_heatloss_warning>false</loud_heatloss_warning> to turn off the warnings.
4 lines of code changed in 2 files:
Added fix for the visit lies issue
7 lines of code changed in 1 file:
Undid last commit until I get a chance to look into it more.
4 lines of code changed in 1 file:
Optimization: Short circuit out of a loop earlier.
8 lines of code changed in 1 file:
Cache the results of an expensive query and reuse it on the next iteration if the inputs are the same.
23 lines of code changed in 1 file:
Added a difference function to the compute sets. This allows us to compute a single difference instead of computing 2 differences and an intersection.
When migrating data between grids only compute the difference as the others are not needed.
85 lines of code changed in 3 files:
Added check for negative Jacobian so this will be caught before it propagates
into other sorts of badness that might cause Todd to pull out some of that
enviably thick hair of his.
19 lines of code changed in 1 file:
Added an intersection function to compute just the intersections and not also the differences. If the differences are not needed then computing them is wasteful.
Disable some expensive error checking if asserts are turned off.
64 lines of code changed in 1 file:
Cleanup some unused variable and methods
0 lines of code changed in 4 files:
Added a <NoisyNegativeDensity> to input which can be true or false to get full details if a negative density occurs. By default, the flag is set to false so the warning happens on a patch basis rather than a cell by cell basis.
Turned "WARNING" into "NOTICE" to make it sound less severe.
Added a comment in jp8_tuna_can.ups.
Added new spec in tag.
16 lines of code changed in 3 files:
Added fix for reading particle variables
14 lines of code changed in 2 files:
Added changes for reading patch data individually
512 lines of code changed in 5 files:
Do not copy from a variable outside of its range.
5 lines of code changed in 1 file:
Fixed a bug which throw a false exception when replacing empty var in DW.
23 lines of code changed in 1 file:
Fix queryRegion, changing to the extra cell/ node indices.
0 lines of code changed in 2 files:
Copy all matched foreign vars instead of first one in copyDataToNewGrid function.
40 lines of code changed in 3 files:
Move the setInvalid outside of the lock.
2 lines of code changed in 1 file:
Added a function call which makes MPI return error codes instead of aborting (this is commented out by default).
0 lines of code changed in 2 files:
Use multimap to store variables in datawarehouse.
72 lines of code changed in 3 files:
fixed an out-of-range bug when copy variables to newgrid
2 lines of code changed in 1 file:
Updated copyDataToNewGrid method to handle foreign variables correctly.
32 lines of code changed in 1 file:
Fixed a bug in the LB that could lead to different processors having different load balances.
1 lines of code changed in 1 file:
Removed a bad ASSERT
0 lines of code changed in 1 file:
Mark variables invalid when MPI is being received to them. When querying a variable do not return a variable that is invalid. This fixes many of the dynamic scheduler problems. The fixed scheduler should be unaffected.
130 lines of code changed in 5 files:
Added a flag to variables to mark if they are valid or not. This will allow the scheduler to mark variables invalid when MPI is being received to them.
9 lines of code changed in 2 files:
Renamed MixingRxnTable.cc to MixingRxnModel (for conceptual reasons) and clearly defined the mixing and reaction base class. Also picked weeds in TabPropsTable.
204 lines of code changed in 10 files:
Initialize x_, y_, and z_ to bad numbers.
0 lines of code changed in 2 files:
uninitialzed pointers
1 lines of code changed in 1 file:
compiler warnings
0 lines of code changed in 2 files:
Compiler warnings
8 lines of code changed in 1 file:
Initialize arrays to DBL_MAX to avoid compiler warnings about uninitialized values. DBL_MAX was choses to help kick up uninitialized variable problems.
0 lines of code changed in 2 files:
Initialization
1 lines of code changed in 1 file:
Created a new directory "ChemMix" that will hold the new chemistry and mixing table interface classes. All files in the "Mixing" will eventually be transfered into ChemMix, thus ridding the code of the rigidity of "NewStaticMixingTable".
4174 lines of code changed in 20 files:
Arggg! Another missed one. Svn mv tricked me.
0 lines of code changed in 2 files:
Missed another one...
0 lines of code changed in 2 files:
Mixxed some sub.mk changes
2 lines of code changed in 2 files:
Moving the furniture around a bit.
20 lines of code changed in 13 files:
Turns out I just didn't commit the new autoconf'd configure script.
0 lines of code changed in 2 files:
I thought AC_SUBST replaced non-existent variables with '', but apparently it ignores them... strange.
1075 lines of code changed in 2 files:
Added --with-electric-fence. (For Justin :)
36 lines of code changed in 4 files:
Incorporate material variability into the Kayenta model (work done by Scot Swan)
Changes to CNHDamage.cc to accomodate changes in Weibull.
317 lines of code changed in 8 files:
Compiler Warning
1 lines of code changed in 1 file:
Uninitialized variable
1 lines of code changed in 1 file:
added -mpi to command lines
0 lines of code changed in 4 files:
Detect the newer versions of openmpi.
1 lines of code changed in 1 file:
fixed a typo: should be 'use_polar_decomposition_RMB'.
Also changed the default behavior to true after more testing.
3 lines of code changed in 1 file:
Supress the printing of a stacktrace and abort messages when the signal was SIGINT. SIGINT is either generated by the user by hitting control-c or by the mpi library when an error is detected on another process. In either case the stacktrace and abort information is not useful and instead obscures the real stacktrace and abort messages.
32 lines of code changed in 4 files:
Added the pid to the beginning of each stack trace line. This should make it easier to identify which stack trace belongs to which processor.
3 lines of code changed in 1 file:
Fix parameter name typo
0 lines of code changed in 2 files:
Added preallocation commands for ILU.
9 lines of code changed in 3 files:
Added support for Petsc 3.0.0.
I still need to find the alternative for PCILUSetFill. Right now it should run with 3.0.0 but won't do preallocation when using ILU.
50 lines of code changed in 4 files:
Added a constructor to the MergeInfo struct.
5 lines of code changed in 1 file:
merged previous two versions (due to mistake in commit).
25 lines of code changed in 2 files:
75 lines of code changed in 1 file:
added d_usePolarDecompositionRMB as an option (default is false).
21 lines of code changed in 2 files:
Removed some asserts that contained non-standard mpi.
0 lines of code changed in 2 files:
Set pointers to zero after deleting them (helps eliminate memory access errors).
Added some more sanity checking.
44 lines of code changed in 12 files:
Added some sanity checking and pass a vector by reference instead of by copy.
8 lines of code changed in 3 files:
Set RHS to zero if martix is singular.
1 lines of code changed in 1 file:
Sanity Checks
6 lines of code changed in 1 file:
Added some sanity checks.
Initalized a variable.
7 lines of code changed in 1 file:
Filter out tasks that don't belong to the owning processor a bit earlier.
8 lines of code changed in 1 file:
Fix Mac build (specify required libs). Mkdir requires #include <sys/stat.h>.
10 lines of code changed in 3 files:
Add Core/Util.
2 lines of code changed in 1 file: