October 2009 Commit Log

Number of Commits:
218
Number of Active Developers:
11
charlesreid 2009-10-31 20:33 Rev.: 45003

Adding some functionality to the LU solver (ability to calculate determinant as a check for singular matrix)

Cleaning up DQMOM solver, adding ability to calculate and save out the moments
(This required a call in ExplicitSolver and a map to store the labels in ArchesLabel)

Also adding a compiler directive that can be used to dump matrices A, B, X, and Resid (this causes a big slowdown but will be useful for small/periodic/verification cases)

462 lines of code changed in 6 files:

charlesreid 2009-10-31 14:53 Rev.: 45002

Fixing a typo that's causing a "no compute for NewCCVelocity" crash

1 lines of code changed in 1 file:

luitjens 2009-10-30 23:19 Rev.: 45001

Fixed a bug I accidentally added to the tiled regridder.

1 lines of code changed in 1 file:

jthornoc 2009-10-30 20:05 Rev.: 45000

Arches.cc:
1) Now copying phi into old_phi after initialization at the beginning of the simulation
2) Properly initializing ccVelocity (gas) at the beginning of the simulation
3) Initializing the particle velocity to the gas velocity at the start (we can debate if this is the right way to handle this)

DqMomEqn.cc:
1) Subtracting the Ax=b source instead of zeroing out the RHS (duh!) for the verification procedure

PartVel.cc:
1) Properly initializing the particle velocity at the start of a new timestep

21 lines of code changed in 3 files:

jpedel 2009-10-30 17:49 Rev.: 44999

Removing debugging statements

0 lines of code changed in 1 file:

jpedel 2009-10-30 17:37 Rev.: 44998

Using isnan function to test if dqmom source terms are nan

2 lines of code changed in 1 file:

jpedel 2009-10-30 17:35

Initializating particle velocities; this was causing nan on the boundaries

11 lines of code changed in 2 files:

luitjens 2009-10-30 14:24 Rev.: 44995

Change int->unsigned int.

Fixed bullet proofing

4 lines of code changed in 1 file:

luitjens 2009-10-30 14:17 Rev.: 44994

Forgot the header.

0 lines of code changed in 2 files:

luitjens 2009-10-30 14:03 Rev.: 44993

Encode the patch set into less bits prior to performing the all gather. This should reduce the amount of data sent by at least a factor of 3 and possibly a factor of 6 on some machines.

98 lines of code changed in 1 file:

luitjens 2009-10-30 13:25 Rev.: 44992

Turn off parallel space filling curve generation.

1 lines of code changed in 1 file:

jpedel 2009-10-29 21:20 Rev.: 44991

Adding statements to prevent source terms to be NaN

4 lines of code changed in 1 file:

jthornoc 2009-10-29 18:01 Rev.: 44990

Forgot to actually add the new files (MMS1).

Changed = back to += (we need += because of multiple quadrature nodes...) in source terms but now we are initializing the source terms to zero before computing.

181 lines of code changed in 5 files:

jthornoc 2009-10-29 17:36 Rev.: 44989

1) added a "mms1" source term
2) added extra sources to dqmom equations for verification/debugging
3) added underscores for RHS, Fconv, Fdiff for naming consistency
4) added initialization for "mms1"
5) changed some += to = in a few coal models including coalgasmomentum.cc, coalgasdevol, and constantsrcterm
6) added compiler directive (VERIFY_DQMOM_TRANSPORT) for DQMOMEqn.cc for verification purposes. Note that ScalarEqn.cc doesn't need this because scalars are in general open to an arbitrary number of source terms.

138 lines of code changed in 9 files:

harman 2009-10-29 16:27 Rev.: 44988

Simulation state is plotted at the bottom of the timestep.
Now plotting stress on the particles
Minor cleanup of debugging code.
Configured for translation of a bar in the x+ dir.

63 lines of code changed in 1 file:

harman 2009-10-29 13:28 Rev.: 44987

-int8() -> int32()
-added bulletproofing to catch if a region is not divisble by R.dx.
-positionToNode()
add offset of 1 to the node number only in the first region (I hate indices starting at 1)
added print statement (commented out)

This has lots of error checking and plotting code just for this problem.

Now have 3 levels with a refinement ratio of 4 on each of the finer levels working
pure translation of a particle through the domain.

188 lines of code changed in 1 file:

jas 2009-10-29 12:33 Rev.: 44986

Add timeout of one hour to each test.

1 lines of code changed in 1 file:

jas 2009-10-28 20:56 Rev.: 44985

Use the 4.4 compilers for the build.

1 lines of code changed in 1 file:

jas 2009-10-28 20:11 Rev.: 44984

The squeeze_32 is now using gcc 4.4 family of compilers and is doing a static
library build.

13 lines of code changed in 1 file:

jas 2009-10-28 16:27 Rev.: 44983

remove advect_2L_MI from the mpmice_test cases since it was duplicated in
the mpmice_amr_test_cases.

1 lines of code changed in 1 file:

jas 2009-10-28 16:04 Rev.: 44982

Add the explode2D_amr test to the mpmice_amr set of test cases.

25 lines of code changed in 1 file:

dav 2009-10-28 15:19 Rev.: 44981

Static link fix.

0 lines of code changed in 2 files:

dav 2009-10-28 15:11 Rev.: 44980

Static link fix.

0 lines of code changed in 2 files:

dav 2009-10-28 15:05 Rev.: 44979

Static link fix.

0 lines of code changed in 2 files:

dav 2009-10-28 14:52 Rev.: 44978

static link fix

0 lines of code changed in 2 files:

luitjens 2009-10-28 14:37 Rev.: 44977

Added the DL library to the static linking line

1 lines of code changed in 1 file:

harman 2009-10-28 13:33 Rev.: 44976

The bar is now specified using bar_min and bar_max. This allows the user to put
1 particle anywhere in the domain.

15 lines of code changed in 1 file:

dav 2009-10-28 13:23 Rev.: 44975

Add lib necessary for (dynamic) mac linking.

1 lines of code changed in 1 file:

dav 2009-10-28 12:57 Rev.: 44974

Move thread lib toward end of link line for static build.

5 lines of code changed in 3 files:

dav 2009-10-28 12:43 Rev.: 44973

update_uda not ready for prime time yet.

1 lines of code changed in 1 file:

dav 2009-10-28 12:41 Rev.: 44972

Cosmetics

0 lines of code changed in 2 files:

dav 2009-10-28 12:39 Rev.: 44971


Allow for building 'sus' (et al) statically. Use the configure flag: --enable-static. For the most
part, you will not want a static build. However, in two cases you might:

1) To check for circular dependencies in the code. (Dynamic libs are forgiving in this regard.)
2) For machines that don't support shared libs (eg: AIX, some new micro-kernel Linux clusters, etc.)

Unfortunately when static linking, library order is important. There is no good way (that I know of)
to tell our build system to place the libs in a specific order, so I have hardcoded the libraries
that are necessary to static link each executable. (See CORE_STATIC_PSELIBS and CORE_STATIC_LIBS in
configVars.mk.)

Note, SUS previously build statically for AIX and Redstorm. A number of hard coded #defines for these
architectures have been replaced with more general code.

If a source code file needs to know that it is building statically (most won't, but some do), then it
will need to #include <sci_defs/compile_defs.h> and use the "STATIC_BUILD" #define'd var.

Some specific notes:

M testprograms/TestMatrix3/testmatrix3.cc

Since it is used in a library, it can't have a "main()" (as this causes two mains to show up when linking).

M configure.ac

- Go back to using 'autoconf' version 2.61 as the standard as it is available many places, and I can't
find 2.63 anywhere.
- Fixed a few uses of a $var that would cause shell script errors when not defined. (Placed it withing ""
so that it becomes an empty string (as opposed to a missing string).


5455 lines of code changed in 44 files:

dav 2009-10-28 12:27 Rev.: 44970

Better comments.

0 lines of code changed in 2 files:

dav 2009-10-28 12:26 Rev.: 44969

Cosmetics

0 lines of code changed in 2 files:

dav 2009-10-28 12:23 Rev.: 44968

Place holder functions for future use (inserted to avoid future confusion).

8 lines of code changed in 3 files:

dav 2009-10-28 12:15 Rev.: 44967

Remove circular dependency by moving WAIT_FOR_DEBUGGER stuff to Core/Exceptions

55 lines of code changed in 8 files:

jas 2009-10-28 12:08 Rev.: 44966

Fix compiler warnings.

4 lines of code changed in 1 file:

jas 2009-10-28 10:13 Rev.: 44965

Fix for some compiler warnings.

4 lines of code changed in 3 files:

luitjens 2009-10-28 09:41 Rev.: 44964

removed some spew.

0 lines of code changed in 1 file:

luitjens 2009-10-28 09:19 Rev.: 44963

Reduced the restrictions on the tiled regridder. Patch boundaries no longer need to line up (though the tile size must still divide evenly into the domain).

Removed the dynamic tile size code and other general cleanup.

93 lines of code changed in 2 files:

luitjens 2009-10-28 08:43 Rev.: 44962

Possible fix for circular dependency.

18 lines of code changed in 6 files:

jas 2009-10-27 20:48 Rev.: 44961

Fix compiler warning.

7 lines of code changed in 1 file:

jpeterson 2009-10-27 20:05 Rev.: 44960

added note on hard limit

2 lines of code changed in 1 file:

luitjens 2009-10-27 14:55 Rev.: 44959

Renable scrubbing. Disabled an error criteria that doesn't seem to be valid.

3 lines of code changed in 2 files:

luitjens 2009-10-27 14:51 Rev.: 44958

Added a new SCI_DEBUG called WaitForDebugger.

This will cause sus to wait for a debugger if it crashes or throws an exception.

To use add WaitForDebugger:+ to your SCI_DEBUG line. When a crash occurs you will see a message that contains
hostname and the pid. Then open a terminal on the host and attach gdb using `gdb --pid <pid>`.

25 lines of code changed in 10 files:

guilkey 2009-10-27 14:08 Rev.: 44957

Put a conditional for dealing with a particular erosion algorithm back in. This
should get the RT passing again without the need for replacing gold standards.

5 lines of code changed in 1 file:

luitjens 2009-10-27 12:51 Rev.: 44956

Removed unnecessary error checking. The computes/requires framework should handle these checks.

202 lines of code changed in 22 files:

luitjens 2009-10-27 12:49 Rev.: 44955

Allocate cellInfo

5 lines of code changed in 1 file:

jthornoc 2009-10-27 12:20 Rev.: 44954

1) Turned off scrubbing per Justin's recommendation
2) For mpmArches, cellInformation is computed in doMomExchange and then required from the new_dw thereafter.
3) Removed the empty emitNormal because it is no longer needed

10 lines of code changed in 5 files:

luitjens 2009-10-27 10:45 Rev.: 44952

Fixed compiler warning.

0 lines of code changed in 2 files:

harman 2009-10-26 15:12 Rev.: 44951

- added option to eject from main time integration loop after max_timesteps is reached.
- positionToNode()
computation of node now takes into account different dx for each region

14 lines of code changed in 1 file:

jthornoc 2009-10-26 14:49 Rev.: 44950

Fixing the missing computes for cellInfo.

19 lines of code changed in 3 files:

jthornoc 2009-10-26 14:42 Rev.: 44949

Implementing an empty emitNormal for PerPatch type variables so that the code can get past the checkpointing.

5 lines of code changed in 1 file:

luitjens 2009-10-26 11:47 Rev.: 44948

Removed the level index queries they alread exist under different names.

6 lines of code changed in 3 files:

luitjens 2009-10-26 10:39 Rev.: 44947

Added queries to get the maximum and minimum cell index for the level.

Compute and store the total number of cells on the level.

Changed totalCells to return the stored number of cells instead of looping through all patches and counting the cells.

30 lines of code changed in 2 files:

harman 2009-10-26 09:29 Rev.: 44946

dt is now dynamically computed.

findNodesAndWeights()
- turned off comparison of old vs new method of computing shape functions

20 lines of code changed in 1 file:

harman 2009-10-23 14:53 Rev.: 44945

-added a new problem called "advectBlock" which is just translation of particles
-BCValue -> velG_BCValue
-Each problem is responsible for setting velG_BCValue

72 lines of code changed in 1 file:

harman 2009-10-23 10:24 Rev.: 44944

- print out node position and region of influcence variables Lx during intialization
- change pre allocation (1,NP) -> (NP,1) (1,NN) -> (NN,1), (1,Bignum)->(BigNum,1)
so it's easier to read when printing data to the screen.
- computing total momentum every timestep
- Writing conserved quantities (totalMom, totalEng, KE, SE) to a file.
- Writing errors( totalEng_err, totalMom_err, and tipDeflect_err) to a file

84 lines of code changed in 1 file:

luitjens 2009-10-23 10:15 Rev.: 44943

use delete[] instead of delete.

Added some more debugging spew.


28 lines of code changed in 1 file:

jas 2009-10-22 20:57 Rev.: 44941

Compiler warnings.

Throw exceptions if call Patch::numFaces and Patch::invalidFace.

9 lines of code changed in 4 files:

luitjens 2009-10-22 18:04 Rev.: 44940

Added some debugging output

6 lines of code changed in 1 file:

harman 2009-10-22 16:24 Rev.: 44939

- include dx in the derivative of the shape function calculation (Gs)
- added findNodesAndWeights_gimp() which is equation 15. (not used)
- added version of findNodesAndWeights() that computes equation 14.
- added section to compute the zone of influence Lx_minus Lx_plus for a
composite grid with several regions with different dx.
- added code to compute the node position for a composite grid with several
regions with different dx.

All machinery is in place to run problems with mesh refinement.
Currently, setup with 3 regions, all at the same resolution.

128 lines of code changed in 1 file:

jas 2009-10-22 15:03 Rev.: 44938

Compiler warning fixes.

56 lines of code changed in 24 files:

luitjens 2009-10-22 13:49 Rev.: 44937

Added more profiling to RGTimes

38 lines of code changed in 1 file:

jthornoc 2009-10-22 13:42 Rev.: 44936

Fixing sub.mk to remove warning. Thanks Dav!

0 lines of code changed in 2 files:

luitjens 2009-10-22 13:27 Rev.: 44935

removed spew.

0 lines of code changed in 1 file:

luitjens 2009-10-22 13:08 Rev.: 44934

Reuse the grid and modified some of the reuse logic for materials.

10 lines of code changed in 1 file:

sshankar 2009-10-22 12:28 Rev.: 44933

Optimizations for speeding up the uda reading

37 lines of code changed in 1 file:

jthornoc 2009-10-22 11:11 Rev.: 44932

Gave SpatialOps a lobotomy. Now linking in all factories, models, sources, and equations from the Arches code base rather than maintaining two copies. This should resolve any naming conflicts.

27 lines of code changed in 46 files:

jpeterson 2009-10-22 11:07 Rev.: 44931

Changed default directory to inputs

0 lines of code changed in 2 files:

jpeterson 2009-10-22 11:05 Rev.: 44930

Moved script to scripts directory, added ability to specify base directory.

100 lines of code changed in 1 file:

luitjens 2009-10-22 10:37 Rev.: 44929

Fixed a bug in the tiled regridder that would cause regridding to occur more often than necessary.

Added the debug stream RGTimes which profiles the tiled regridder.

71 lines of code changed in 1 file:

luitjens 2009-10-22 10:36 Rev.: 44928

Minor change in spew.

1 lines of code changed in 1 file:

jas 2009-10-21 21:48 Rev.: 44927

Compiler warnings.

2 lines of code changed in 1 file:

jas 2009-10-21 19:45 Rev.: 44926

Add SpatialOps back to the list so the code will compile.

2 lines of code changed in 3 files:

jas 2009-10-21 19:27 Rev.: 44925

Return a non-zero exit if compilation fails.

0 lines of code changed in 2 files:

jas 2009-10-21 18:27 Rev.: 44924

Do a make cleanreally first.

1 lines of code changed in 1 file:

harman 2009-10-21 17:44 Rev.: 44923

Baby steps.

Added findNodesAndWeights() function that follows the nomenclature in the Hongbing
CMES paper.

72 lines of code changed in 1 file:

jas 2009-10-21 17:12 Rev.: 44922

Add extradir to TransferTestResults class for moving data to dbg and opt
directories for the nightly builds.

8 lines of code changed in 1 file:

charlesreid 2009-10-21 16:29 Rev.: 44920

Removing SpatialOps from sub.mk files

0 lines of code changed in 2 files:

charlesreid 2009-10-21 16:06 Rev.: 44919

Forgot to do "svn move"

0 lines of code changed in 4 files:

charlesreid 2009-10-21 15:59 Rev.: 44918

Fixing naming conflict with ModelFactory by moving this model factory to SpatialOpsCoalModelFactory

111 lines of code changed in 12 files:

charlesreid 2009-10-21 12:11 Rev.: 44917

Cleaning up tabs/spacing, adding better descriptions, restoring some scalar stuff that was (accidentally?) deleted

69 lines of code changed in 2 files:

jthornoc 2009-10-21 11:29 Rev.: 44916

1) Normalizing the initial value of the dqmom variable using the scaling factor in the problemSetup of the respective equation.
2) Split initialization of weights into a separate method so that they can be used later to initialize weighted abscissas
3) Overloaded initializationFunction to deal with weighted abscissas (multiplication by a weight)

277 lines of code changed in 5 files:

luitjens 2009-10-21 11:22 Rev.: 44915

Added code that allows the space-filling curves to be generated in serial for timing comparaisons.

33 lines of code changed in 1 file:

jas 2009-10-21 11:15 Rev.: 44914

Replace ShellCommand with Configure and Compile which will keep track of
warnings.

31 lines of code changed in 1 file:

charlesreid 2009-10-21 11:02 Rev.: 44913

Swapping Y1 and Y2

4 lines of code changed in 1 file:

jas 2009-10-20 21:50 Rev.: 44911

Create TransferTestResults class that will copy the TestResults directory
from the various build slaves and store them on the buildbot webserver.

Add a link to the waterfall page so that the TestResults can be viewed
for each test.

37 lines of code changed in 1 file:

jas 2009-10-20 21:38 Rev.: 44910

Only move failed test cases to the TestResults directory.

3 lines of code changed in 1 file:

jas 2009-10-20 21:33 Rev.: 44909

Create separate TestResults directory that stores the individual
gzipped tar file for each test case.

11 lines of code changed in 1 file:

harman 2009-10-20 18:51 Rev.: 44908

Fixed so the title changes with each test. The results are crap but at least the plot title is right.

3 lines of code changed in 1 file:

harman 2009-10-20 18:11 Rev.: 44907

-turned off some of the excess output
- now saving an eps plot for every variable compared. Right now it's temp_CC, rho_CC, Press_CC, Vel_CC.
- now saving temp_CC, rho_CC, Press_CC, Vel_CC to a file for comparison plots later.


13 lines of code changed in 1 file:

harman 2009-10-20 18:08 Rev.: 44906

slight cleanup on order of accuracy plot

8 lines of code changed in 1 file:

harman 2009-10-20 18:07 Rev.: 44905

include run with 1600 grid points

0 lines of code changed in 2 files:

harman 2009-10-20 18:04 Rev.: 44904

include runs with 1600 cells

0 lines of code changed in 6 files:

harman 2009-10-20 18:02 Rev.: 44903

downshift to firstOrder advection

0 lines of code changed in 8 files:

jthornoc 2009-10-19 23:41 Rev.: 44900

Added a time verification procedure for the explicit time integrator. This is activated through a compiler directive. Some useful scripts to come to actually run the verification automagically...

142 lines of code changed in 5 files:

jas 2009-10-19 21:11 Rev.: 44899

Create a builder just for documentation.

Specify the directory where SVN checks out the source.


35 lines of code changed in 1 file:

jas 2009-10-19 19:52 Rev.: 44898

Reorganize some of the scripts for adding the TestSteps.

13 lines of code changed in 1 file:

jas 2009-10-19 19:32 Rev.: 44897

Update the nightly tests to use the new AddTestStep function.

9 lines of code changed in 1 file:

jas 2009-10-19 17:45 Rev.: 44896

Use new arguments for regression tester script.

Shorten the function to create BuildStepTests.

16 lines of code changed in 1 file:

jas 2009-10-19 17:39 Rev.: 44895

Add make link_inputs to make.py.

Use consistent names for arguments, i.e. -m --> -mpm, -i --> -ice.

16 lines of code changed in 2 files:

guilkey 2009-10-19 16:55 Rev.: 44894

Threw some crap against the wall, these changes seem to stick.

10 lines of code changed in 3 files:

jas 2009-10-19 16:10 Rev.: 44893

Simply the addition of tests using a function.

52 lines of code changed in 1 file:

guilkey 2009-10-19 14:27 Rev.: 44891

This should get the ImpMPM RT running again. Getting computes and requires
in place.

Also, deleting ZerilliArmstrongPlastic, which was apparently replaced by
ZAPlastic.

240 lines of code changed in 73 files:

charlesreid 2009-10-19 14:12 Rev.: 44890

Fixing some problems with 'requires' statements, and making initialization function error message more clear

15 lines of code changed in 4 files:

jas 2009-10-19 12:21 Rev.: 44887

Remove builders that aren't used.

Use relative paths for commands.


35 lines of code changed in 1 file:

guilkey 2009-10-19 12:16 Rev.: 44886

This should get the MPMF regression test running again.

6 lines of code changed in 2 files:

jas 2009-10-16 22:06 Rev.: 44885

Use the helper script make.py for compiling.

11 lines of code changed in 1 file:

jas 2009-10-16 20:15 Rev.: 44884

Don't specify the python interpreter.

0 lines of code changed in 1 file:

jas 2009-10-16 20:05 Rev.: 44883

Helper script for buildbot compilation.

19 lines of code changed in 1 file:

jthornoc 2009-10-16 16:40 Rev.: 44882

Removing this check for length size. It doesn't work because the upper limit check is arbitrary.

5 lines of code changed in 1 file:

jthornoc 2009-10-16 14:24 Rev.: 44881

Changed the solver tolerance checking.

19 lines of code changed in 2 files:

jas 2009-10-16 13:22 Rev.: 44880

Remove commented out lines.

0 lines of code changed in 1 file:

jas 2009-10-16 12:03 Rev.: 44879

Use relative path for the configure line.

31 lines of code changed in 1 file:

guilkey 2009-10-16 11:10 Rev.: 44878

More fixes to cmputes and requires. Still doesn't run.

1 lines of code changed in 1 file:

guilkey 2009-10-16 11:06 Rev.: 44877

Add some computes. This gets past the errors kicked up by Justin's changes,
but now it is segfaulting.

3 lines of code changed in 1 file:

guilkey 2009-10-15 10:44 Rev.: 44874

Added compute(pFiberDir,...) to scheduleInitialize.

1 lines of code changed in 1 file:

guilkey 2009-10-14 23:06 Rev.: 44873

Trying to clean up fall out from Justin's recent commits.

28 lines of code changed in 4 files:

jas 2009-10-14 21:22 Rev.: 44872

Add the UCF tests and append the test case you are running to the TestCase
directory name. Upon any failure of a test, the TestCase directory will
be tar gzipped.

31 lines of code changed in 1 file:

jas 2009-10-14 16:41 Rev.: 44871

Only run the 32 bit debug compile and the optimize build and test if
the 64bit debug compile succeeds.

15 lines of code changed in 1 file:

luitjens 2009-10-14 11:58 Rev.: 44870

Also comment out the requires so buildbot will stop yelling at us.

2 lines of code changed in 1 file:

luitjens 2009-10-14 11:02 Rev.: 44869

Commented compute out until we can resolve the multiple computes issue.

1 lines of code changed in 1 file:

harman 2009-10-14 10:47 Rev.: 44868

Fixed computes/requires for delT

19 lines of code changed in 7 files:

luitjens 2009-10-14 10:24 Rev.: 44867

Added a missing compute.

1 lines of code changed in 1 file:

luitjens 2009-10-14 10:19 Rev.: 44866

Added some missing computes/requries.

4 lines of code changed in 1 file:

luitjens 2009-10-14 10:06 Rev.: 44865

Removed debug code.

1 lines of code changed in 1 file:

harman 2009-10-14 09:54 Rev.: 44864

Fixed the delT computes and requires

21 lines of code changed in 8 files:

luitjens 2009-10-14 09:53 Rev.: 44863

Fixed some requires issues.

2 lines of code changed in 1 file:

charlesreid 2009-10-13 23:29 Rev.: 44862

Fixing minor typo

1 lines of code changed in 1 file:

harman 2009-10-13 17:57 Rev.: 44861

added ability to perform order of accuracy tests on the density, temperature, pressure
and velocity simultaneously. This is a rough draft and needs some polish.


144 lines of code changed in 6 files:

luitjens 2009-10-13 11:15 Rev.: 44860

Turned on computes/requires checking.

2 lines of code changed in 1 file:

charlesreid 2009-10-12 20:23 Rev.: 44859

Changing some >/<'s to <=/>='s

9 lines of code changed in 1 file:

luitjens 2009-10-12 17:34 Rev.: 44858

Added an exception to the requires checking for the copyDataToNew grid task because it runs outside of the normal schedule process.

7 lines of code changed in 1 file:

guilkey 2009-10-12 17:07 Rev.: 44857

Details, details. _preReloc variables come from the NewDW, not the old.

1 lines of code changed in 1 file:

guilkey 2009-10-12 16:59 Rev.: 44856

Crossing my fingers that this will resolve Justin's problem. Namely, change
the requires of pErosionLabel to pErosionLabel_preReloc.

1 lines of code changed in 1 file:

luitjens 2009-10-12 15:45 Rev.: 44855

Readded missing requires.

1 lines of code changed in 1 file:

harman 2009-10-12 15:23 Rev.: 44854

added tests 2 -> 4 from "Riemann Solvers and Numerical Methods for Fluid Dynamics"
They're commented out.

782 lines of code changed in 12 files:

jthornoc 2009-10-12 15:02 Rev.: 44853

A little clean-up and/or slight modifications of the previous commit (from Charles).

21 lines of code changed in 8 files:

luitjens 2009-10-12 14:56 Rev.: 44852

More missing requires.

6 lines of code changed in 3 files:

luitjens 2009-10-12 14:26 Rev.: 44851

Added some missing computes.

3 lines of code changed in 1 file:

luitjens 2009-10-12 13:56 Rev.: 44850

Added missing require.

1 lines of code changed in 1 file:

luitjens 2009-10-12 13:49 Rev.: 44849

added missing compute.

1 lines of code changed in 1 file:

luitjens 2009-10-12 13:45 Rev.: 44848

getWhichDW() now takes a parameter to specify which running task is making the request. Previously it would assume the innermost task which means parent tasks would return the wrong dw.

24 lines of code changed in 2 files:

harman 2009-10-12 13:42 Rev.: 44847

added test1 of Toro's book. This is currently commented out.

8 lines of code changed in 1 file:

charlesreid 2009-10-12 12:39 Rev.: 44845

Fixing some compiler warnings

20 lines of code changed in 2 files:

charlesreid 2009-10-12 12:20 Rev.: 44844

- Fixing initialization bugs in DQMOMEqn
- Adding uniform constant/step initialization functions to DQMOMEqn class
- Adding environment-specific constant/step initialization functions to DQMOMEqn class
- Fixing print statements in equation classes to use proc0cout instead of cout
- Fixing several indentation issues in equation classes

The following files had only minor/cosmetic changes:
* PartVel.cc
* ExplicitSolver.cc
* DQMOM.cc
* ExplicitTimeInt.h
* CoalGasMomentum.cc
* CoalGasDevol.cc
* Arches.cc
* ScalarEqn.h
* DQMOMEqn.h

The following files had major changes:
* ScalarEqn.cc
* DQMOMEqn.cc
* EqnBase.cc

986 lines of code changed in 14 files:

luitjens 2009-10-12 11:20 Rev.: 44843

Fixed some more computes/requires errors.

12 lines of code changed in 4 files:

luitjens 2009-10-12 10:59 Rev.: 44842

Added missing computes/requires.

3 lines of code changed in 2 files:

guilkey 2009-10-11 22:28 Rev.: 44841

Make deletion of "rogue" particles a runtime option. There is still a hardwired
parameter that should probably be an option as well. Also, added the task
findRogueParticles so that this will work correctly in parallel. This task is
only scheduled if the run time option is set.

131 lines of code changed in 6 files:

guilkey 2009-10-11 21:09 Rev.: 44840

Oops, forgot to include the .h file.

1 lines of code changed in 1 file:

guilkey 2009-10-11 21:08 Rev.: 44839

Changed my mind, make heating due to artificial viscosity true by default.

3 lines of code changed in 1 file:

guilkey 2009-10-11 21:03 Rev.: 44838

Make heating due to artificial viscosity optional, off by default.

16 lines of code changed in 2 files:

guilkey 2009-10-11 13:02 Rev.: 44837

p_q is always computed (artifical viscosity) so it should always be scheduled.

4 lines of code changed in 1 file:

guilkey 2009-10-11 12:54 Rev.: 44836

Comment out the code I added Friday until I make it a run time option.

2 lines of code changed in 1 file:

charlesreid 2009-10-10 22:06 Rev.: 44835

Fixing some indentation issues
Adding some helpful output to Arches class

106 lines of code changed in 4 files:

guilkey 2009-10-09 18:30 Rev.: 44833

Added code to remove particles that have "gone rogue". Eventually, this will
be an input file option, with knobs.

58 lines of code changed in 1 file:

jthornoc 2009-10-09 18:18 Rev.: 44831

Added a mechanism for initializing the transported scalar variables, derived from EqnBase, to a function. Need to modify DQMOMeqn to use this as well.

111 lines of code changed in 5 files:

guilkey 2009-10-09 18:15 Rev.: 44830

Adding pLocalizedMPMLabel so that SerialMPM can know about failed particles.

5 lines of code changed in 2 files:

guilkey 2009-10-09 17:44 Rev.: 44829

Added an AllowNoShear option to ElasticPlastic. Other modifications to
attempt to make this model more robust as particles "localize". Removed
unneeded line of code from JCPlastic.cc

47 lines of code changed in 3 files:

jthornoc 2009-10-09 17:24 Rev.: 44828

Fixing some weird requires statements.

10 lines of code changed in 2 files:

luitjens 2009-10-09 15:32 Rev.: 44827

More missing computes/requires

6 lines of code changed in 2 files:

luitjens 2009-10-09 15:05 Rev.: 44826

Whoever added cellInformation to Arches forgot a couple of computes/requires......


36 lines of code changed in 10 files:

harman 2009-10-09 14:53 Rev.: 44825

pre-allocate grid variables

Globals variables are slow, now passing numRegions and Regions into function arguments

Moved plotting code into separate functions

It now works with multiple regions, with a refinement ratio of 1 on each region.

226 lines of code changed in 1 file:

luitjens 2009-10-09 13:40 Rev.: 44824

Fix some requires problems.

5 lines of code changed in 1 file:

luitjens 2009-10-09 13:40 Rev.: 44823

treat put with a replace as a compute.

Remove delT exemption from computes/modifies/requires checking.

20 lines of code changed in 1 file:

luitjens 2009-10-09 10:53 Rev.: 44822

Removed compute because it is scheduled in MPM::scheduleComputeStressTensor if artifical viscosity is set to on. Jim needs to look into the proper fix.

1 lines of code changed in 1 file:

jthornoc 2009-10-09 10:51 Rev.: 44821

Accidently was destroying a varlabel twice.

0 lines of code changed in 1 file:

luitjens 2009-10-09 10:42 Rev.: 44820

Added some missing requires.

5 lines of code changed in 1 file:

luitjens 2009-10-09 10:35 Rev.: 44819

Added some missing computes.

3 lines of code changed in 2 files:

luitjens 2009-10-09 10:18 Rev.: 44818

Added missing require

1 lines of code changed in 1 file:

harman 2009-10-09 09:14 Rev.: 44817

-added Regions Structure.

-added function positionToNode()

-Using postionToNode in findNodesAndWeights() & findNodesAndWeightGradients()

The time integration loop has variable dx incorportated, the initialization does not.

94 lines of code changed in 1 file:

luitjens 2009-10-08 20:20 Rev.: 44816

Readded delt requires for ICE

11 lines of code changed in 1 file:

luitjens 2009-10-08 20:02 Rev.: 44814

Fixed some faulty logic in the task graph that would incorrectly determine which dw was being used when using the W-cycle (this should not change the answers).

2 lines of code changed in 1 file:

charlesreid 2009-10-08 18:32 Rev.: 44813

Changing some cout's to proc0cout's to reduce amount of stuff being printed

12 lines of code changed in 1 file:

charlesreid 2009-10-08 18:04 Rev.: 44812

Resolving a naming conflict (& getting rid of a bug in DQMOMEqn initialization)

267 lines of code changed in 16 files:

jas 2009-10-08 15:08 Rev.: 44811

Archive failing test cases.

3 lines of code changed in 1 file:

luitjens 2009-10-08 14:26 Rev.: 44810

Update the delt logic in AMRSimulation controller to be consistant.

Components must compute and store delt on a per level basis. Right now the infastructure then uses the per-level delt to compute the global delt.

49 lines of code changed in 18 files:

jpedel 2009-10-08 13:48

More accurate heat transfer model

175 lines of code changed in 2 files:

harman 2009-10-08 11:11 Rev.: 44806

added bulletproofing
- if a particle leaves the domain then warn the user and exit the time
integration loop.

0 lines of code changed in 2 files:

harman 2009-10-08 11:05 Rev.: 44805

- fixed typo in problem_type strcmp
- consolidated problem dependent parameters into one area
- printing out timestep information for each timestep
- added bulletproofing
if a paricle leaves the domain then warn the user and
exit the time integration loop
- added problem_type, NN, PPC to the header in the output files

42 lines of code changed in 1 file:

jas 2009-10-08 09:18 Rev.: 44804

Fix misspelling.

1 lines of code changed in 1 file:

harman 2009-10-07 17:37 Rev.: 44803

more thrashing
- eliminated problem types 2 & 3
- problem types are now described with strings
(impulsiveBar, oscillator, colliidingBars)
- added bulletproofing

- preallocate particle and grid variables for speed
- Nuked G, K, dug
- variable name changes
mg -> massG
vg -> velG
sigp -> stressP
Fig -> intForceG
Feg -> extForceG
ag -> accl_G
vg_new -> vel_new_G

- added titles to the final plot
- now plotting tip deflection error vs time
- now writing particle data at final timestep to particleData.dat
- now writing grid data at final timestep to gridData.dat

133 lines of code changed in 1 file:

jas 2009-10-07 17:04 Rev.: 44802

Remove the username and password for svn access.

0 lines of code changed in 1 file:

harman 2009-10-07 14:02 Rev.: 44801

formatting,
- 2 space indentation
- spaces between +-/*

Variable name changes
volp -> volP
dvp -> dvelP
mp -> massP
vp -> velP

* this should not change the answers

209 lines of code changed in 1 file:

jas 2009-10-07 13:56 Rev.: 44800

Add svn username and password to svnpolling and the various builders.

20 lines of code changed in 1 file:

luitjens 2009-10-07 12:32 Rev.: 44799

added constLevelP handle

0 lines of code changed in 2 files:

guilkey 2009-10-07 11:36 Rev.: 44798

Make ApproachContact syntactically more like its big brother, FrictionContact,
so it will be easier to identify the differences between the two.

40 lines of code changed in 1 file:

harman 2009-10-07 11:15 Rev.: 44797

changed filenames. Matlab doesn't like numbers in the filenames.

330 lines of code changed in 2 files:

harman 2009-10-07 10:54 Rev.: 44796

initial commit. This is just a copy of one_d_mpm_new.m

0 lines of code changed in 1 file:

harman 2009-10-07 10:48 Rev.: 44795

Reverted change.
There is some ambiguity in the scheduling and computing of delT that needs to get hammered
out before this change can be committed.

8 lines of code changed in 1 file:

harman 2009-10-07 09:34 Rev.: 44794

BIG BUG, pointed out by Justin.

- moved put(delt) outside the patch loop.

This potentially could change all of the multi-patch problem answers. If it doesn't, then I've
been living right.

11 lines of code changed in 1 file:

harman 2009-10-07 09:01 Rev.: 44793

pulled a weed

0 lines of code changed in 4 files:

harman 2009-10-06 21:59 Rev.: 44792

More work on riemann test

58 lines of code changed in 5 files:

harman 2009-10-06 21:53 Rev.: 44791

added <otherFilesToCopy> tag
These files will be copied to the testing directory

0 lines of code changed in 2 files:

qymeng 2009-10-06 16:57 Rev.: 44790

Updated task graph validation code.
Uncomment "#if 0" in hasGetAccess() and hasPutAccess() functions to enable checking.

113 lines of code changed in 2 files:

charlesreid 2009-10-06 14:23 Rev.: 44789

Fixing a few more std::runtime's to Uintah type errors

12 lines of code changed in 2 files:

charlesreid 2009-10-06 14:09 Rev.: 44788

Making errors more descriptive (changing from std_runtime to Uintah type error)

1 lines of code changed in 1 file:

harman 2009-10-05 18:07 Rev.: 44787

initial commit for the riemann shock tube order of accuracy test

453 lines of code changed in 4 files:

harman 2009-10-05 18:04 Rev.: 44786

bulletproofing when the comparison utility can't be found

0 lines of code changed in 2 files:

harman 2009-10-05 17:22 Rev.: 44785

fixed debugging print statement

2 lines of code changed in 1 file:

charlesreid 2009-10-05 15:11 Rev.: 44784

Making changes in heat transfer model parameters

23 lines of code changed in 1 file:

guilkey 2009-10-05 13:16 Rev.: 44783

Reverting accidentally committed changes.

3 lines of code changed in 1 file:

guilkey 2009-10-05 12:27 Rev.: 44782

Removed left over allocateAndPut of NC_CCweights that should have been removed
when these computes were pushed up to SerialMPM. Thanks Justin and Qingyu for
pointing this out.

5 lines of code changed in 2 files:

charlesreid 2009-10-03 18:20 Rev.: 44781

- Fixing the constant source term model
- Changing Kobayashi Sarofim model to get weights and weighted abscissas from old_dw, not new_dw (I don't think this actually changes anything, as they should be the same value before the DQMOM linear solve)
- Various message/comment fixes (cosmetic, do not actually change anything)

75 lines of code changed in 3 files:

jas 2009-10-02 16:43 Rev.: 44780

Remove getNumVelFields from SimulationState. This function is not used.

0 lines of code changed in 8 files:

jthornoc 2009-10-02 16:37 Rev.: 44779

Compiler warnings

1 lines of code changed in 3 files:

guilkey 2009-10-02 11:03 Rev.: 44778

Add some requires for p.size based on errors Justin found. More to come.

2 lines of code changed in 1 file:

charlesreid 2009-10-01 17:13 Rev.: 44777

Fixing a few more incorrectly-named variables.

7 lines of code changed in 2 files:

charlesreid 2009-10-01 17:05 Rev.: 44776

Removing parts of Kobayashi Sarofim model that look for gas_temperature as a label. This model uses the gas temperature as the temperature, by default; if the user specifies a particle temperature internal coordinate, the model will use that instead.

17 lines of code changed in 2 files:

jpedel 2009-10-01 15:31 Rev.: 44775

Using Ubhayakar values (more realistic)

5 lines of code changed in 1 file:

jpedel 2009-10-01 15:16 Rev.: 44774

Fixing my own bug

1 lines of code changed in 1 file:

jpedel 2009-10-01 15:07 Rev.: 44773

Fixing a small bug with d_radiation

3 lines of code changed in 1 file:

luitjens 2009-10-01 14:42 Rev.: 44772

Explicitly set the values for the WhichDW enum instead of relying on the compiler to do the right thing...

0 lines of code changed in 2 files:

harman 2009-10-01 13:29 Rev.: 44771

scheduleComputeLagrangianSpecificVolume()
- added missing requires for press_CC

5 lines of code changed in 1 file:

jas 2009-10-01 12:39 Rev.: 44770

Add gravity to the output problem spec function.

4 lines of code changed in 1 file:

charlesreid 2009-10-01 12:00 Rev.: 44768

Fixing a few typos and reverting one change made in my last commit (radiation was accidentially turned off by default)

6 lines of code changed in 2 files:

luitjens 2009-10-01 10:06 Rev.: 44767

Added a bit more information to LBOut

4 lines of code changed in 1 file:

September 2009 »

Generated by StatSVN 0.4.0