Use safer copying method
1 lines of code changed in 1 file:
Undid an optimization to the load balancer I added earlier. This appears to be causing some problems.
1 lines of code changed in 1 file:
Compiler Error
1 lines of code changed in 1 file:
Fixed a bug in the load balancer when using the SFC code that could lead to an incorrect SFC curve.
14 lines of code changed in 1 file:
Fixed a bug that shows up when used in AMR runs.
0 lines of code changed in 2 files:
Update location of glui.h to GL directory.
11 lines of code changed in 3 files:
A few more changes to mpi_test. Mostly just changes to make the logic more clear.
0 lines of code changed in 2 files:
Fixed a bug which could lead to a less than optimal load balance.
Reduced the amount of data that is broadcast.
24 lines of code changed in 1 file:
Tidying up a bit.
28 lines of code changed in 1 file:
more info
0 lines of code changed in 2 files:
trying to turn on ability to search doxygen
0 lines of code changed in 2 files:
changed the timeout on async tests to timeout when progress is not being made for 50 seconds instead of 50 seconds total time.
10 lines of code changed in 1 file:
save correct number of particles
1 lines of code changed in 1 file:
Yet another change to the boundary conditions.
Going back to setting symmetry conditions on the
extra nodes, but now we also set symmetry conditions
on the plane of regular nodes ALL THE WAY TO THE EDGE
OF THE EXTRA NODES IN THE OTHER TWO DIMENSIONS.
45 lines of code changed in 1 file:
dataVerify_Pull_driver: changed rsync option -rl -> -a
dataVerify_driver: added rsync option -t (preserve times)
0 lines of code changed in 4 files:
Add the configuration file for the master controller for the buildbot
running on retriever.sci.utah.edu.
197 lines of code changed in 1 file:
-Minor simplification of the conditional statements
- "-m" option now works
0 lines of code changed in 2 files:
Use old artificial viscosity code but in the newer framework.
13 lines of code changed in 1 file:
Get out some additional data.
0 lines of code changed in 2 files:
removed scripts that compile scirun, these are points of confusion.
-updated documentation
0 lines of code changed in 4 files:
-removed unused script
-updated documentation
0 lines of code changed in 2 files:
removed unused script highwater_check.pl
updated documentation
0 lines of code changed in 2 files:
This script is no longer used and a point of confusion.
0 lines of code changed in 1 file:
changed logic for comparing the input.xml(org) files. This should eliminate warning when comparing checkpoint directories
0 lines of code changed in 2 files:
Fixes to allow use on a Mac. Make 'find' commands cleaner.
0 lines of code changed in 2 files:
It turns out that the code I was pretty sure was never used
IS used in MPMICE. Uncommented the code for pressure BCs.
2 lines of code changed in 1 file:
Fixed a compile error related to accidentally reusing a variable name.
0 lines of code changed in 2 files:
added spaces around conditional statement. shell is retarded
0 lines of code changed in 2 files:
syntax error
1 lines of code changed in 1 file:
indentation
detect if using make or gmake
95 lines of code changed in 1 file:
In MPMBoundCond.cc, delete the code for acceleration boundary
conditions, this has not been used for a long time. Comment out
the code for pressure boundary conditions. This has probably never
been used, but I'm keeping it around for a little while longer.
More importantly, change the way in which symmetry BCs are set
when using gimp. Previously, the extra node values were set in
the same way that normal boundary nodes were set, by zeroing out
the normal component, and leaving the other two alone. Scott B.
has convinced me through empirical evidence that just zeroing
all values out on the extra nodes is more effective. This will
change the answers for some problems that use gimp and symmetry BCs.
15 lines of code changed in 2 files:
removed extraneous "." from the file path
Removed --size-only option. If a file/directory doesn't exist on the receiving
machine rsync throws an error. This may slow down the verification process
significantly.
2 lines of code changed in 1 file:
If the user is replacing the goldstand with the -f option they can now add the log file comment via the command line
19 lines of code changed in 2 files:
Turn back on all the safety checks...
25 lines of code changed in 1 file:
MVAPICH2/1.0 requires -lrt
16 lines of code changed in 2 files:
Fixes to pscp2. Now records a (temporary) list of files that have been transferred. Records (permanently in the UDA) information about the file transfer. Allows for 'picking up' in the middle of transfer that stopped. Allows for running when you don't have permissions to write to the UDA. See http://www.csafe.utah.edu/wiki/index.php/Documentation/UsersGuide/DataTransfer
186 lines of code changed in 3 files:
Fix for OSX
8 lines of code changed in 1 file:
output CC, CXX, F77 in the configure command script
0 lines of code changed in 2 files:
Delete some commented out code.
0 lines of code changed in 1 file:
problemSetup()
- removed the cout_norm statements they are rarely ever used.
6 lines of code changed in 1 file:
changed label names so they are consist with ICE label
mass_source_CC -> modelMass_src
momentum_source_CC -> modelMom_src
energy_source_CC -> modelEng_src
sp_vol_source_CC -> modelVol_src
density_CC -> rho_CC
velocity_CC -> vel_CC
temperature_CC -> temp_CC
pressure_CC -> press_CC
This should not change the answers
179 lines of code changed in 27 files:
Fix formatting for "_" that is used extensively in variable names.
0 lines of code changed in 2 files:
I'm going to need a tetanus shot to cure this bite in my butt. This
should fix guni2DRT.
1 lines of code changed in 1 file:
Mac Fix: Needs all these libs to successfully link.
0 lines of code changed in 2 files:
First pass through formatting the text.
463 lines of code changed in 1 file:
Apparently the PGI compiler on REDSTORM is different from PGI everywhere else... sigh.
0 lines of code changed in 2 files:
Don't believe that we need to compile the cd_templates when not building DATAFLOW.
0 lines of code changed in 2 files:
Add a heat equation example that will show an explicit and implicit
formulation. The implicit formulation will demonstrate the use of the
petsc interface plus various boundary conditions.
256 lines of code changed in 6 files:
Additional examples.
751 lines of code changed in 1 file:
Allow for easy use of different SSH (via env var). Cleaner, clearer status messages. Start/End time output.
44 lines of code changed in 1 file:
Initial checkin of the Uintah application programing interface.
195 lines of code changed in 1 file:
Save level number in nrrd.
9 lines of code changed in 1 file:
Throw an exception if the boundary conditions have not been read in properly
and the patch data structures are empty.
14 lines of code changed in 1 file:
Fix --with-glew flag to configure. (Though you probably shouldn't use it anyway, as it defaults to the SCIRun 3P which has the 'blessed' glew in it.)
10 lines of code changed in 2 files:
Added Labels for Angio component.
63 lines of code changed in 1 file:
Add labels for new Angio component.
166 lines of code changed in 1 file:
Adding a new component. This is the beginnings of a discrete model
for angiogenesis. Still much work to be done.
3234 lines of code changed in 23 files:
Modified the output of getstacktrace
11 lines of code changed in 1 file:
Deleted some commented code (SSP) and moved some code around (HEP).
5 lines of code changed in 2 files:
Don't output error message if stringstream fails because material "all" is being used
2 lines of code changed in 1 file:
moved the calculation of adding the hydrostatic pressure to the relative pressure
from *.f to *.cc file.
This should not change the answers
11 lines of code changed in 5 files:
Minor cleanup in the conditional statements that determine when to run debug tests.
Weekly tests are only run with optimized code.
0 lines of code changed in 2 files:
moved:
calculatePressureCoeff() & mmModifyPressureCoeffs()
from Discretization.cc -> PressureSolver.cc
This should not change the answers
102 lines of code changed in 4 files:
pruned #includes
0 lines of code changed in 6 files:
pruned #includes
0 lines of code changed in 1 file:
added:
-dbg_only
-opt_only
options to the RT.
49 lines of code changed in 3 files:
Move the dissipation from artificial viscosity out of the conditional
on being plastic. Still commented out for now.
5 lines of code changed in 1 file:
More pressCoeff from 7 CCVariable<doubles> -> 1 CCVariable<Stencil7>
9 lines of code changed in 1 file:
Converted pressCoeff from 7 CCVariable<doubles> -> 1 CCVariable<Stencil7>
Discretization:
- added new templated function compute_Ap() which duplicates fortran:apcal_all
except works with stencil 7 data. After all variables are converted to stencil7 we can
use this in place of apcal_all.
BoundaryCondition
- added templated function copy_stencil7() which copies the stencil7 data
in and out of the CCVariable<double> arrays. This is used before and after
the fortran pressureBC routines. This is gross hack until the fortran BC routines
are written in c++
This should not change the answers.
199 lines of code changed in 10 files:
cleaned up the #includes
3 lines of code changed in 1 file:
Cleaned up prototypes in Discretization.cc.
No need to pass a variable into a function if it isn't used.
Shouldn't change the answers
38 lines of code changed in 8 files:
calculatePressureCoeff()
- cleaned up the prototype
computeDivergence()
- replaced 3 loops -> cellIterator
This should not change the answers
28 lines of code changed in 4 files:
moved code from fortran routine mm_modify_prescoef into C++
Descretization::mmModifyPressureCoeffs()
This should not change the answers
14 lines of code changed in 5 files:
changed solve -> sched_solve to reflect what is really going on.
14 lines of code changed in 4 files:
Added Key/Value pairs with more info to output NRRDs. Add 'data' line to particle NRRDs.
36 lines of code changed in 3 files:
ssh -> $(SSH)
0 lines of code changed in 2 files:
cleaner formatting of output
1 lines of code changed in 1 file:
Print out level extents (Helpful for RTRT).
15 lines of code changed in 1 file:
Moved -maxdepth option to be the first option to eliminate a warning.
0 lines of code changed in 2 files:
For now, use the unrotated rate of deformation tensor in
computing the artificial viscosity. This shouldn't matter,
as we are just using the trace of the tensor, which should
be invariant, but the polar decomposition code is dicey at
best.
35 lines of code changed in 3 files:
Added (and used in sus) the option to make parsing of the .scirunrc
file silent.
M Core/Util/Environment.h
M Core/Util/Environment.cc
Added beSilent option so no print statements occur when parsing (or
failing to parse) .scirunrc. This defaults to false (ie, defaults to
old behavior of being verbose).
Moved some 'global' functions to just be local as they shouldn't
be/arent called from anywhere but Environment.cc.
M Packages/Uintah/StandAlone/sus.cc
Use the beSilent option (for create_sci_environment()) so that large
runs don't spam the output with "Parsing... not found", etc.
31 lines of code changed in 3 files:
Remove the ExampleLabels and replace with an explicit declaration of the
VarLabel data types for clarity.
80 lines of code changed in 14 files:
Remove the examples label datatype and explicitly show the creation of
VarLabels for clarity.
24 lines of code changed in 2 files:
Fixed a bug that would cause mpi to crash on a restart.
5 lines of code changed in 1 file:
The Switcher component used to record the UDA name in the <input_file>
section (of the input.xml file):
<SimulationComponent>
...
<subcomponent>
...
<input_file>HTContainer_arches_mpmice.uda.006/input.xml.0</input_file>
</subcomponent>
...
</SimulationComponent>
This causes a problem when trying to restart an UDA if you run from a
directory not directly containing the UDA directory (or if the UDA was
renamed). I've updated the code to not record the UDA name (now it
just puts input.xml.#). I've also updated the code such that if it
sees an uda name (in the input.xml file), it removes it, and then it
adds in the UDA name as specified when sus was run.
M CCA/Components/Parent/Switcher.h
M CCA/Components/Parent/ComponentFactory.cc
Pass the uda name into the Switcher.
M CCA/Components/Parent/Switcher.cc
- Handle input_file name correctly (for both new and old UDAs).;
- Don't add the UDA name to the <input_file> tag in new UDAs.
- Formatting tweaks to make code more readable.
- Rename variable 'in' to 'input_file' so I could actually search through the file for it. ;)
M StandAlone/inputs/ups_spec.xml
Update the ups spec file for SimulationComponents. Note, the Switcher
component is still not correctly specified and should be updated by
someone who has worked on and knows it.
46 lines of code changed in 5 files:
Verify that MPI is in use when running arches... as arches needs petsc which needs mpi...
9 lines of code changed in 1 file:
Fix the computes for p_qLabel so that it will work in parallel.
4 lines of code changed in 2 files:
Add p_q to the carryForward code used by RigidMaterial.
0 lines of code changed in 2 files:
Only assign pvelGrads if Fracture is being used.
1 lines of code changed in 1 file:
Small optimization to the load balancing algorithm.
34 lines of code changed in 1 file:
Whoa Nelly! Move the calculation of p_qLabel_preReloc (aka the artificial
viscosity term) from SerialMPM to each of the individual constitutive models.
This should be more efficient and will allow me to add the dissipative heat
that results. That is coming next.
331 lines of code changed in 23 files:
Put pvelGrads back in for the fracture code.
3 lines of code changed in 1 file:
Change the way in which the various measures of deformation
are computed to a way in which the velocity gradient is
available right within computeStressTensor. This should NOT
change any answers, but will then be followed up by computing
the artificial viscosity within the models, not in a separate
task in SerialMPM.
120 lines of code changed in 13 files:
Multiply pdTdt by delT to get a temperature increment in particle
temperature.
(I swear I already committed this fix once...)
1 lines of code changed in 1 file:
Rewrote possiblyCreateDetailedDependency.
The old method would only combine the dependency with one other matching dependency. If the dependency matched two other dependencies the could later have problems.
The new method will now combine with all matching dependencies by combining dependencies one at a time and repeatly trying to combine the new dependecy with the old dependencies
until the dependency can no longer be combined.
Also added some comments.....
195 lines of code changed in 1 file:
Made a small optimization.
17 lines of code changed in 1 file:
Added a few operators
22 lines of code changed in 2 files: