June 2007 Commit Log

Number of Commits:
108
Number of Active Developers:
12
harman 2007-06-29 17:56 Rev.: 37929

-gridstats
dump out a few more details about the grid.

10 lines of code changed in 1 file:

borodai 2007-06-29 13:42 Rev.: 37926

Removed Kumar's arrass.

0 lines of code changed in 1 file:

luitjens 2007-06-29 13:02 Rev.: 37925

Removed spew

0 lines of code changed in 1 file:

luitjens 2007-06-29 12:40 Rev.: 37924

Add the percent time in overhead to the ComponentTimes:+ spew

9 lines of code changed in 1 file:

luitjens 2007-06-29 10:50 Rev.: 37920

Calculate the percent time spent doing AMR grid related things. This
will be used to control dilation eventually.

61 lines of code changed in 2 files:

luitjens 2007-06-28 00:18 Rev.: 37912

MatZeroRows, MatZeroRowsIS, and ISCreateGeneral are collective
operations. In addition d_DOF may be empty on some processors while not
empty on other processors. This causes a bug where collective
operations are not called on all processors. The solution is to move
these calls outside of the if(d_DOF.size()>0).


7 lines of code changed in 1 file:

jas 2007-06-27 19:50 Rev.: 37911

Fix temperature_solve so it works for multi-patch/multi-processor.

21 lines of code changed in 1 file:

jas 2007-06-27 14:31 Rev.: 37906

Removed #if 0 stuff related to old version of temperature solve.

0 lines of code changed in 1 file:

harman 2007-06-26 12:03 Rev.: 37899

Don't execute PFE when the DataArchive is not present --bryan

2 lines of code changed in 1 file:

ramanuja 2007-06-25 22:24 Rev.: 37898

Removed some printf() to make it faster.
Added some time-checks to ensure desired results.
cleaned some comments.

9 lines of code changed in 1 file:

jas 2007-06-25 16:19 Rev.: 37897

Reimplement the function of interpolation particle temperatures to the grid
nodes in a more robust fashion. This is turned on when <temperature_solve>
is set to true in <MPM>. Note that you must have a particle resolution of
at least 2,2,2.

250 lines of code changed in 2 files:

ramanuja 2007-06-25 13:12 Rev.: 37896

Fixed the author name and the description.

7 lines of code changed in 1 file:

worthen 2007-06-25 12:59 Rev.: 37895

Make bulletproofing be called on the finest level. Errors normally caught for level 0 had been masked when more than one level had been manually defined.

3 lines of code changed in 2 files:

dav 2007-06-25 12:28 Rev.: 37894

M compare_scalar.cc

- Use fabs() (instead of abs()) so it will compile.
- Removed my name as I didn't write it... ;)
- Removed the description of what it does as it looks like it is just
a direct copy of the info from compare_mms.
- Cleaned up commented out includes.
- Coding standard return lines.

9 lines of code changed in 1 file:

worthen 2007-06-25 11:26

Fix seg-fault when printing once-per-proc tasks.
In Switcher::problemSetup, check to see if the regridder needs to be turned on (like on a restart)

26 lines of code changed in 3 files:

sbrown 2007-06-22 16:24 Rev.: 37889

AIX/IntelCC linker locks up trying to find log(int). This fixes building on AIX. Please check this.

1 lines of code changed in 1 file:

luitjens 2007-06-22 13:14 Rev.: 37887

Minor tweak to the load balancer logic

1 lines of code changed in 1 file:

ramanuja 2007-06-22 10:49 Rev.: 37886

The make file modifications for compiling compare_scalar command line utility.

0 lines of code changed in 2 files:

ramanuja 2007-06-22 10:45 Rev.: 37885

A command-line utility for error analysis of advection of passive scalar using ice.

388 lines of code changed in 2 files:

worthen 2007-06-21 14:32 Rev.: 37881

Allow the Regridder to be turned on and off when switching

37 lines of code changed in 5 files:

harman 2007-06-21 10:23 Rev.: 37867

moved fine level conditional above scinew

6 lines of code changed in 1 file:

harman 2007-06-21 09:46 Rev.: 37866

dump out diagnostics when switching criteria has been met.

0 lines of code changed in 2 files:

harman 2007-06-21 09:33 Rev.: 37865

Only compute mass burned on the finest level.

5 lines of code changed in 1 file:

jas 2007-06-20 18:41 Rev.: 37864

Add the endcap surface area only when we have end caps in input file.

2 lines of code changed in 1 file:

harman 2007-06-20 18:13 Rev.: 37863

excess debugging output

0 lines of code changed in 1 file:

harman 2007-06-20 15:29 Rev.: 37861

initialize temp_CC_mpm to 0

dump out diagnostics when a cell "burning."

6 lines of code changed in 1 file:

jas 2007-06-20 13:20 Rev.: 37857

Add end cap surface area computation.

8 lines of code changed in 1 file:

jas 2007-06-20 13:19 Rev.: 37856

Add endcap average data to the computation of heat fluxes.

68 lines of code changed in 4 files:

harman 2007-06-20 12:48 Rev.: 37855

added level index and reformatted DebugStream TaskDBG

6 lines of code changed in 1 file:

harman 2007-06-20 11:58 Rev.: 37853

Get pXLabel from the oldDW instead of the newDW.

3 lines of code changed in 1 file:

harman 2007-06-20 10:37 Rev.: 37851

progress towards it working on multiple processors

Compute temp_CC_mpm in all cells where cmass > 1e100

31 lines of code changed in 1 file:

luitjens 2007-06-20 10:26 Rev.: 37850

Include the time in MPI_Wait{some/all} in the SCI_DEBUG ComponentTimes:+

6 lines of code changed in 4 files:

worthen 2007-06-19 15:33 Rev.: 37848

uncommit fake arches

2 lines of code changed in 1 file:

worthen 2007-06-19 15:31 Rev.: 37847

Only use NC_CCweight on MPM levels

62 lines of code changed in 5 files:

harman 2007-06-19 14:51 Rev.: 37846

switchInitialize()
-added warning to user

--------------------------------------------------------------
ImpMPM: the heat flux BC cannot be applied on the timestep
immediately after a component switch. The computes/requires
cannot be met and one pseudo timestep must take place
---------------------------------------------------------------

7 lines of code changed in 1 file:

harman 2007-06-19 14:48 Rev.: 37845

Tidied up the "carry over" output.

6 lines of code changed in 1 file:

harman 2007-06-19 14:14 Rev.: 37843

Clarification of what "0" is.

14 lines of code changed in 8 files:

worthen 2007-06-19 12:51 Rev.: 37842

Call switchInitialize after switch (fix conditional)

2 lines of code changed in 1 file:

worthen 2007-06-18 23:29 Rev.: 37839

Fix Switcher bugs - see message 37838

48 lines of code changed in 7 files:

kuzimmer 2007-06-18 18:41 Rev.: 37834

for the time being, get rid of multi-threading while reading particles

15 lines of code changed in 1 file:

harman 2007-06-18 16:33 Rev.: 37833

accidentally committed debugging statement

0 lines of code changed in 1 file:

harman 2007-06-18 16:30 Rev.: 37832

Fixed ImpMPM + AMR restart bug

MPMFlags.cc:
- moved section of code related to AMR to ImpMPM.cc and AMRMPM.cc
<AMR>
<MPM>
min_grid_level & max_grid_level
</MPM>
<AMR>

ImpMPM:
- Cleaned up printSchedule() and added a few printTask()

39 lines of code changed in 3 files:

luitjens 2007-06-18 14:46 Rev.: 37831

Free the memory for sub-cycle timesteps when using W-Cycle. Previously this would not be deleted until the end of the coarsest timestep leading to excess memory usage.

Theoritically this could provide up to a factor of (refinement ratio)^(num levels-1) improvement in memory usage.

Bryan & Justin

30 lines of code changed in 6 files:

harman 2007-06-18 10:24 Rev.: 37829

missing line from previous commit

1 lines of code changed in 1 file:

harman 2007-06-18 10:16 Rev.: 37828

Compute the "switch" on every level.

Only perform burn model tests on the finest level.

161 lines of code changed in 2 files:

harman 2007-06-15 15:41 Rev.: 37817

only schedul switchCriteria tests on the finest level

8 lines of code changed in 2 files:

harman 2007-06-15 11:15 Rev.: 37816

init variables can now be initialized on a per level basis.

For multilevel simulations, particle variables must have
levels="-1"
specified. Bulletproofing will spank the uninformed.

48 lines of code changed in 2 files:

luitjens 2007-06-15 09:55 Rev.: 37815

Forgot a header

8 lines of code changed in 1 file:

worthen 2007-06-14 23:33 Rev.: 37814

Fix 'Parallel not initialized' error caused by Justin's and my parallelization of setBCTypes

22 lines of code changed in 4 files:

harman 2007-06-14 16:59 Rev.: 37813

added:
scheduleRefineInterface
scheduleRefine
scheduleCoarsen
scheduleInitialErrorEstimate
scheduleErrorEstimate

51 lines of code changed in 2 files:

harman 2007-06-14 16:45 Rev.: 37812

changed variable name:
gtemperature -> gTempAllMatls

4 lines of code changed in 1 file:

harman 2007-06-14 16:25 Rev.: 37811

for those of us who can't remember if the option is
listvariables listVariables or list_variables

30 lines of code changed in 1 file:

harman 2007-06-14 15:13 Rev.: 37807

more descriptive exception message

2 lines of code changed in 1 file:

worthen 2007-06-14 15:07 Rev.: 37806

Move finalizeLevel and assignBCS to after we are sure we are using the new grid

13 lines of code changed in 1 file:

worthen 2007-06-14 14:43 Rev.: 37805

Change switcher's carry over vars to be those not computed rather than those not required from OldDW.Add slightly better error message when a required var is not foundAdd functionality to give the computed vars

12 lines of code changed in 5 files:

harman 2007-06-14 13:27 Rev.: 37804

changed variable name gtemperature -> gTempAllMatls
removed some dead wood.

3 lines of code changed in 1 file:

luitjens 2007-06-14 13:18 Rev.: 37803

Added a new difference algorithm. This still needs to be tested and is
not active yet.

The old difference algorithm appears to be O(P^2) where the new one
should be O(P log P).

I will test and compare the two and make it active when it is ready.

169 lines of code changed in 1 file:

harman 2007-06-14 12:49 Rev.: 37801

Contents of PBXTemperature.(cc/h) has been moved to SimpleBurn.(cc/h)

SwitchingCriterialFactory.cc
- added steadyBurn switching criteria
- fixed exception

SteadyBurn.cc
- new component switching criteria based on HEChem/SteadyBurn.cc
(not working)

441 lines of code changed in 9 files:

harman 2007-06-14 12:42 Rev.: 37800

added comments.

12 lines of code changed in 2 files:

luitjens 2007-06-14 11:39 Rev.: 37798

Forgot a header

3 lines of code changed in 1 file:

luitjens 2007-06-14 10:30 Rev.: 37797

Move the costly finalizeLevel and assignBCS to below the grid equality check. This avoids performing these operations when we will not use the grid.

I will look into doing this for the hierarchical regridder also.

9 lines of code changed in 1 file:

luitjens 2007-06-14 10:28 Rev.: 37796

Modified the grid operator== to compare grids without ghostcells.

Added a function called isSimilar which returns true if two grids take up the same region but contain different patches.

42 lines of code changed in 1 file:

harman 2007-06-13 15:43 Rev.: 37793

changed how the thermalEnergy is computed

thermal_energy += pTemp[idx] * pmass[idx] * Cp

7 lines of code changed in 1 file:

borodai 2007-06-13 15:12 Rev.: 37791

Warn when heat loss exceeds actual table bounds,
not theoretical -1 to 1 bounds.

5 lines of code changed in 1 file:

luitjens 2007-06-12 21:57 Rev.: 37785

Fixed compile when not using Petsc

0 lines of code changed in 2 files:

wight 2007-06-12 16:33 Rev.: 37783

M EquationOfStateFactory.cc
M sub.mk

Modified EquationOfStateFactory to include new EOS model: HardSphereGas
Modified sub.mk so that new EOS will make correctly

Chuck Wight - Dept of Chemistry
6/12/207

0 lines of code changed in 4 files:

wight 2007-06-12 16:29 Rev.: 37782

Added new equation of state for ICE materials: HardSphereGas

A SCIRun/src/Packages/Uintah/CCA/Components/ICE/EOS/HardSphereGas.cc
A SCIRun/src/Packages/Uintah/CCA/Components/ICE/EOS/HardSphereGas.h

useage in the UPS file (<ICE> section):

<EOS type="hard_sphere_gas">
<b> 1.4e-3 </b>
</EOS>

This EOS is a simple extension of the ideal gas EOS, but accounts for
the finite volume occupied by the molecules themselves, so causes the
computed pressure to be greater at high densities. The EOS is given by:

P(V-b) = RT (where b = specific volume of molecules)

Typical value for b is 1.4e-3 m^3/kg (e.g., for N2 or CO).

Chuck Wight - Dept of Chemistry, Univ of Utah
6/12/2007


197 lines of code changed in 2 files:

luitjens 2007-06-12 13:51 Rev.: 37774

Untabify

Parallelized part of finalizeLevel.

Bryan & Justin

182 lines of code changed in 1 file:

luitjens 2007-06-12 13:50 Rev.: 37773

Removed spew

0 lines of code changed in 1 file:

harman 2007-06-12 13:49 Rev.: 37772

formatted output.

1 lines of code changed in 1 file:

harman 2007-06-12 13:41 Rev.: 37771

Removed incorrect spew

0 lines of code changed in 2 files:

luitjens 2007-06-12 12:09 Rev.: 37768

Reduce the number of processors that participate in the regrid dyanamically at runtime

125 lines of code changed in 1 file:

luitjens 2007-06-11 18:18 Rev.: 37766

If the AMR block is defined default to turning AMR on. If doAMR is
additionally defined then set amr according to its value.

5 lines of code changed in 1 file:

guilkey 2007-06-11 18:05 Rev.: 37765

Remove flushMatrix from the constitutive models, it is now done elsewhere.

0 lines of code changed in 6 files:

luitjens 2007-06-11 16:47 Rev.: 37763

set the value of do_AMR based on the value in the specification doAMR
and not on the existance of it.

2 lines of code changed in 1 file:

luitjens 2007-06-11 15:56 Rev.: 37762

Changed AMR to doAMR.

1 lines of code changed in 1 file:

worthen 2007-06-11 15:10 Rev.: 37761

Fix string warning

1 lines of code changed in 1 file:

luitjens 2007-06-11 14:48 Rev.: 37760

on the path to implicit heat conduction with AMR.

41 lines of code changed in 2 files:

luitjens 2007-06-11 14:46 Rev.: 37759

Minor optimizations

24 lines of code changed in 1 file:

harman 2007-06-11 13:31 Rev.: 37758

added support for ParticleVariables<T>
This returns the average value over all particles in a cell.

317 lines of code changed in 1 file:

borodai 2007-06-11 12:23 Rev.: 37757

Properly create flow rate label names.

5 lines of code changed in 1 file:

sbrown 2007-06-11 12:22 Rev.: 37756

Add new 'tools' target. Use 'make -k tools' to build all utilities/extractors/etc.

3 lines of code changed in 1 file:

luitjens 2007-06-11 12:13 Rev.: 37755

Fix a few scirun things --bryan

3 lines of code changed in 2 files:

guilkey 2007-06-10 16:31 Rev.: 37752

Initialize global_offset to eliminate compiler warning

1 lines of code changed in 1 file:

luitjens 2007-06-08 15:03 Rev.: 37749

Initialize d_active

2 lines of code changed in 1 file:

luitjens 2007-06-08 10:30 Rev.: 37745

Don't pass the number of nodes into the solvers, instead use the solvers
internal state to determine the number of rows.

PetscSolver.cc
Added some debugging code.

47 lines of code changed in 11 files:

luitjens 2007-06-07 23:10 Rev.: 37743

Added a static member Parallel::getMPISize()

9 lines of code changed in 2 files:

luitjens 2007-06-07 22:27 Rev.: 37742

Only print the particle count once

1 lines of code changed in 1 file:

luitjens 2007-06-07 16:26 Rev.: 37740

Progress twoards AMR Implicit MPM.

24 lines of code changed in 3 files:

luitjens 2007-06-07 14:44 Rev.: 37738

Removed initial delt control from ice. The simulation controller does this.

0 lines of code changed in 5 files:

luitjens 2007-06-07 12:39 Rev.: 37736

When using taskDBG print out both the start and end of a task.

3 lines of code changed in 1 file:

jas 2007-06-06 17:16 Rev.: 37728

Check particles in the z direction in checkForSurface.

1 lines of code changed in 1 file:

luitjens 2007-06-06 17:07 Rev.: 37725

Backed out revision 37716 and implemented a much simpler method of
allocating the diagonal.

10 lines of code changed in 4 files:

luitjens 2007-06-06 14:23 Rev.: 37719

More progress twoard dynamically reducing the number of processors participating in the regrid at runtime.

59 lines of code changed in 1 file:

luitjens 2007-06-06 11:22 Rev.: 37716

Force allocation of the diagonal for the ImplicitHeatConduction matrix.
Petsc's preallocation information is lost once a matrix is assembled and
thus any future calls to MatSetValues can cause mallocs. The implicit
heat code adds 1's to the diagonals at a later point since the diagonals
were previously zero this means that mallocs will be called.


17 lines of code changed in 4 files:

luitjens 2007-06-06 11:14 Rev.: 37715

Begining of some code to consolodate flags onto smaller numbers of
processors. This is not active yet.

42 lines of code changed in 1 file:

luitjens 2007-06-05 23:03 Rev.: 37713

The first element and an argv array is typically reserved for the
command name. Petsc assumes this thus ignores the first option passed
in the argv array. This was causing the first option we passed petsc to
be ignored.

4 lines of code changed in 1 file:

harman 2007-06-04 17:04 Rev.: 37706

Sigh...........

1 lines of code changed in 1 file:

dav 2007-06-04 15:43 Rev.: 37705

Commented out unused vars

3 lines of code changed in 1 file:

luitjens 2007-06-04 15:22 Rev.: 37704

Initialize the useExternalQueue_ to false..

Bryan gets credit for both making and fixing this bug.... ;)


1 lines of code changed in 1 file:

luitjens 2007-06-04 14:51 Rev.: 37703

untabify

14 lines of code changed in 1 file:

worthen 2007-06-04 14:40 Rev.: 37702

Fix MultiTaskGraph problem when having an Output task with empty patch subsets

1 lines of code changed in 1 file:

luitjens 2007-06-04 14:34 Rev.: 37701

Fixed a typo....

1 lines of code changed in 1 file:

cthurston 2007-06-04 14:16 Rev.: 37700

Added sulfur balance computation.
Added ability to track elemental sulfur.

106 lines of code changed in 9 files:

luitjens 2007-06-04 12:53 Rev.: 37699

Make sure that the printParticleCount task is only called once per proc.

1 lines of code changed in 1 file:

luitjens 2007-06-04 12:34 Rev.: 37698

Fixed a bug where inconsistent initial timesteps would be taken when using more processors than patches.

1 lines of code changed in 1 file:

luitjens 2007-06-02 09:48 Rev.: 37691

Added another Petsc assert

1 lines of code changed in 1 file:

luitjens 2007-06-01 15:11 Rev.: 37687

Set the initial stash size for the matrix. Right now i'm just setting
it to a fixed number but that value should be counted.

The stash is used when a processor tries to set a matrix value that it
does not own. This happens for nodes on the plus face of a patch when
that face crosses a processor boundary.


10 lines of code changed in 1 file:

May 2007 »

Generated by StatSVN 0.4.0