November 2010 Commit Log

Number of Commits:
100
Number of Active Developers:
10
guilkey 2010-11-30 16:31 Rev.: 46525

Fix warning in a conditional.

0 lines of code changed in 2 files:

harman 2010-11-30 15:58 Rev.: 46524

Global:
removed the testing version of getFineLevelRangeNodes(), using the original
version that has been cleaned up.

getCoarseLevelRange()
-now passing in the IntVector boundaryLayer and number of ghost cells.
Previously, boundaryLayer was hardcoded to (0,0,0)

AMRMPM
interpolateToParticlesAnUpdate_CFI()
- now interpolating gvelocityStar_fine & gAcceleration_fine to the CFI
particles on the coarse level.

interpolateParticlestoGrid()
- removed debugging code.

interpolateParticlestoGrid_CFI()
- using cell based getCoarseLevelRange instead of nodal based getCoarseLevelRangeNodes()

interpolateToParticlesAndUpdate()
- removed hardwired vel & acc

109 lines of code changed in 6 files:

jas 2010-11-30 14:51 Rev.: 46523

Remove the #define for setting the hash_map. Now 1.42 version of boost will compile cleanly.


0 lines of code changed in 2 files:

jsutherland 2010-11-29 18:17 Rev.: 46521

This should fix the build errors on Wasatch3P.

0 lines of code changed in 2 files:

jsutherland 2010-11-29 18:09 Rev.: 46520

Fix up an input file and activate another regression test for wasatch. This one tests staggered transport equations (diffusion only)

4 lines of code changed in 1 file:

guilkey 2010-11-24 22:48 Rev.: 46518

Band-Aid the computes and requires so this will run multi-proc without doing
the grid reset.

6 lines of code changed in 1 file:

guilkey 2010-11-24 16:40 Rev.: 46517

Arrange the calculations in a way that won't make the buildbot mad.

7 lines of code changed in 1 file:

guilkey 2010-11-24 15:15 Rev.: 46516

Fix the deformation which was getting reset to the unadjusted value even though
pressure stabilization was applied.

2 lines of code changed in 1 file:

ffjhl 2010-11-24 14:15 Rev.: 46515

added PressureStabilization to avoid spurious fluid oscillations.

0 lines of code changed in 2 files:

ffjhl 2010-11-24 14:14 Rev.: 46514

Replaced eigen method with getEigenValues; the former's known to
be problematic.

0 lines of code changed in 2 files:

jsutherland 2010-11-24 10:41 Rev.: 46513

turn off debug version of a test that was very slow in debug.

2 lines of code changed in 1 file:

harman 2010-11-23 09:26 Rev.: 46512

if the command mpirun is not in the user's path then warn them to either put
it in their path or set the environmental variable MPIRUN.

0 lines of code changed in 2 files:

harman 2010-11-23 09:10 Rev.: 46511

reduced output.

1 lines of code changed in 1 file:

jthornoc 2010-11-23 07:02 Rev.: 46510

Added check for null pointer from TabProps. Removed a debugging print
statement.

7 lines of code changed in 1 file:

harman 2010-11-22 18:21 Rev.: 46509

If the command mpirun does not exist in the users path then warn the user
to set the environmental variable MPIRUN and exit.


16 lines of code changed in 1 file:

dav 2010-11-22 18:06 Rev.: 46508

1) OSX does not support --static flag. 2) Check for/require serialization lib for boost.

13 lines of code changed in 2 files:

dav 2010-11-22 17:35 Rev.: 46507

M configure.ac
M configure

- Allow --without-hypre and --without-petsc (to turn off hypre/petsc).
Note, for historical reasons (Ie, I think we just didn't do it),
none of the other configure options (I belive) support --without.
It probably would be nice to add them in.

M CCA/Components/Solvers/AMR/sub.mk
M CCA/Components/Solvers/sub.mk

- Indentation

270 lines of code changed in 6 files:

harman 2010-11-22 16:56 Rev.: 46506

Do not make a library out of this subdir.

0 lines of code changed in 1 file:

harman 2010-11-22 16:39 Rev.: 46505

added a few tabs.
Still won't compile.

6 lines of code changed in 2 files:

harman 2010-11-22 16:04 Rev.: 46504

added sub.mk

76 lines of code changed in 1 file:

harman 2010-11-22 16:02 Rev.: 46503

moved Oren's multi-level solver into a separate directory.

5772 lines of code changed in 111 files:

jsutherland 2010-11-22 12:59 Rev.: 46502

fix previous commit to turn of boost::serialization output in TabProps.

1 lines of code changed in 1 file:

guilkey 2010-11-22 12:06 Rev.: 46501

1. Add the "C" variable to the copy constructor and outputProblemSpec.
2. Add the ability to ignite from a plane, rather than just at a point.

0 lines of code changed in 2 files:

harman 2010-11-22 11:38 Rev.: 46500

more descriptive task names that are passed into the taskgraph

1 lines of code changed in 7 files:

jsutherland 2010-11-22 10:14 Rev.: 46499

Default TabProps to HDF5 output for now

1 lines of code changed in 1 file:

guilkey 2010-11-19 18:46 Rev.: 46497

Fix uninitialized variable.

1 lines of code changed in 1 file:

jthornoc 2010-11-19 18:37 Rev.: 46496

Yet another missing ghost cell.

0 lines of code changed in 2 files:

harman 2010-11-19 18:18 Rev.: 46495

added the previous version of getFineLevelRangeNodes_old() so the RT will pass.

49 lines of code changed in 3 files:

jthornoc 2010-11-19 17:49 Rev.: 46494

Ghost number requirement for area fraction needed fixin'

2 lines of code changed in 1 file:

jthornoc 2010-11-19 16:28 Rev.: 46493

TabPropsInterface.cc/.h
-> Allow for enthalpy to be turned on but radiation be off for adiabatic
cases
-> Use BSpline pointers rather than strings for table lookup
-> Check bounds on derived mixture fraction for coal tables
-> Complete rehash of how property BCs are being computed
-> Removed d_hl_outlet and d_hl_pressure in favor of the previous
comment

Discretization_new.h
-> Fixed a bug in super_bee for internal boundary conditions
-> Changed the diffusion calculation for now at boundaries so that they
behave exactly like the old Arches boundary conditions. We should
revisit this.

tableProbe.m
-> Read in a string.

arches_spec.xml
-> updated spec to match above changes

298 lines of code changed in 4 files:

harman 2010-11-19 12:01 Rev.: 46492

added ./ to doTestScript command.

0 lines of code changed in 2 files:

luitjens 2010-11-19 11:52 Rev.: 46491

Added an assert

1 lines of code changed in 1 file:

jsutherland 2010-11-19 10:55 Rev.: 46490

- add convection regression tests in wasatch. These also test parallelization (though not completely).

7 lines of code changed in 1 file:

jsutherland 2010-11-18 20:56 Rev.: 46489

- push a bit more information into the FieldManager for debugging purposes.
- add some documentation to Array3Window.h. There was a land-mine there that I walked into. Hopefully this will save the next poor soul some pain and suffering.
- add some code to set the process rank as a field variable. This isn't currently active in Wasatch, but can be easily plugged in if needed.

110 lines of code changed in 5 files:

harman 2010-11-18 17:56 Rev.: 46487

added TaskDBG

0 lines of code changed in 2 files:

harman 2010-11-18 17:18 Rev.: 46486

Compiler warnings

interpolateParticlesToGrid_CFI()
get the zoi variable for the entire fine patch instead of getRegion.

7 lines of code changed in 1 file:

harman 2010-11-18 15:53 Rev.: 46485

Added task to zero out the nodal data on the coarse level nodes at the coarse fine
interface.

scheduleCoarsenNodalData_CFI()/ coarsenNodalData_CFI()
- pass in a flag to the task that indicates to either coarsen or zero the data.


77 lines of code changed in 2 files:

harman 2010-11-18 14:26 Rev.: 46484

removed unused enum.

0 lines of code changed in 4 files:

harman 2010-11-18 12:01 Rev.: 46483

Ooops. This should have been committed.

11 lines of code changed in 1 file:

harman 2010-11-18 11:59 Rev.: 46482

scheduleInterpolateParrticlesToGrid_CFI()
- schedule gZOILabel for the fine level patch

scheduleInterpolateToParticlesAndUpdate()
- removed scheduling for the fine/coarse levels. This is handled by a different task.
- added compute for diagnostic variable gSumWeightsLabel

computeZoneOfInfluence()
- pulled out the hackery from the previous commit

interpolateToParticlesAndUpdate()
- removed all coarse level + fine level computations. This task only operates
on a single level.

Added new task interpolateToParticlesAndUpdate_CFI()
- This interpolates nodal data from the fine level to the coarse level particles
near the coarse/fine interface.

It basically performs the major operations that I want it to do for a very simple 2L grid.

176 lines of code changed in 1 file:

harman 2010-11-18 11:47 Rev.: 46481


getFineLevelRangeNodes()
using the function computeVariableExtents to determine the fine patch low/high
indices. This allows for boundary layer cells. Eliminated the improper
use of ghost cells.

54 lines of code changed in 3 files:

harman 2010-11-18 11:16 Rev.: 46480

getModifiable() & getParticleVariable() for particle subsets
more descriptive exception message

2 lines of code changed in 1 file:

guilkey 2010-11-17 09:30 Rev.: 46475

Use proc0cout to reduce some spew.

7 lines of code changed in 1 file:

guilkey 2010-11-16 22:27 Rev.: 46474

Only spew logmes info on processor 0.

0 lines of code changed in 2 files:

guilkey 2010-11-16 21:58 Rev.: 46473

Fixed outputProblemSpec to insert the correct tag of cohesive_zone instead of
material.

0 lines of code changed in 2 files:

harman 2010-11-16 16:40 Rev.: 46472

more information in the ostream operator <<

0 lines of code changed in 2 files:

harman 2010-11-15 17:06 Rev.: 46471

added boost library to spatialOps -Dav

1 lines of code changed in 1 file:

harman 2010-11-15 16:00 Rev.: 46470

removed Jim's scheduling of interpolateParticlesToGrid(). This should have been removed
in previous commit

1 lines of code changed in 1 file:

jpeterson 2010-11-15 15:25 Rev.: 46469

Moved crack burn criteria into computeNumPPC because it already loops over all particles. Should increase speed.
Should not change numbers.

46 lines of code changed in 3 files:

jpeterson 2010-11-15 12:34 Rev.: 46468

Some much needed commenting, housecleaning and reorganization. Combined two nested for loop structures together.

138 lines of code changed in 1 file:

jthornoc 2010-11-13 04:46 Rev.: 46467

cout --> proc0cout

0 lines of code changed in 2 files:

jthornoc 2010-11-13 04:19 Rev.: 46466

Fix a warning.

1 lines of code changed in 1 file:

jthornoc 2010-11-13 03:44 Rev.: 46465

Generalizing the Westbrook & Dryer model.

0 lines of code changed in 4 files:

guilkey 2010-11-12 19:20 Rev.: 46464

Minor cleanup

0 lines of code changed in 2 files:

guilkey 2010-11-12 18:34 Rev.: 46463

Fixed a bug in how the Weibull distribution was being used.

1 lines of code changed in 1 file:

jthornoc 2010-11-12 14:41 Rev.: 46462

Passing variable by reference.
Updated ups file to match spec.

8 lines of code changed in 3 files:

harman 2010-11-12 11:50 Rev.: 46461

startLocalTest
- pass in "unknown" for the argument dbg_opt into the python script

runSusTests
- added vertical spacing.
- added additional warnings if the user is trying to do something silly
- Tell the user that the gold standard is being created.
- only set the outputpath & weboutputpath if running the nightly RT AND the
the environmental variables (HTMLOG,WEBLOG) exist.

30 lines of code changed in 2 files:

harman 2010-11-12 10:54 Rev.: 46460

fixed "if dialog exists" conditional.

0 lines of code changed in 2 files:

harman 2010-11-12 10:35 Rev.: 46459

Global:
added note about performance tests and debug builds

Wasatch:
- no longer running performance tests. Performance tests are not run with dbg builds

37 lines of code changed in 18 files:

guilkey 2010-11-11 19:14 Rev.: 46458

Remove the condition on interpolator type in the Neumann BC code.

0 lines of code changed in 2 files:

guilkey 2010-11-11 18:56 Rev.: 46456

Fix timestep size calculation

0 lines of code changed in 2 files:

guilkey 2010-11-11 18:54 Rev.: 46455

Formatting

1 lines of code changed in 1 file:

harman 2010-11-11 17:35 Rev.: 46454

removed debugging code

0 lines of code changed in 1 file:

dav 2010-11-11 17:00 Rev.: 46453

partial fix for SVN checking, still is not complete.

14 lines of code changed in 1 file:

dav 2010-11-11 16:50 Rev.: 46452

Fix python version error problem with exit. Add info for OSX user wrt 'dialog'.

23 lines of code changed in 2 files:

harman 2010-11-11 16:12 Rev.: 46451

Added -svnDiff and -svnStat to sus command line.
This allows users to diff sus_log.txt files, gold standard vs localRT

2 lines of code changed in 1 file:

harman 2010-11-11 16:10 Rev.: 46450

using cout instead of cerr

0 lines of code changed in 2 files:

dav 2010-11-11 15:34 Rev.: 46448

Added ability to set TEST_COMPONENTS environment variable so that user
input is not requried to run the (local) tests/generate gold standards.

TEST_COMPONENTS should be set to something like "MPM ICE ARCHES"

M R_Tester/toplevel/startLocalTest

- Use TEST_COMPONENTS env var

M R_Tester/toplevel/generateGoldStandards.py

- Added a number of !!!FIXME!!!s that should be implemented to make
the script more foolproof.
- exit if wrong version of python

M R_Tester/helpers/selectComponents.sh

- Use TEST_COMPONENTS env var
- Added some documentation

69 lines of code changed in 3 files:

harman 2010-11-11 15:20 Rev.: 46447

Added bulletproofing, warn the user and exit if the component has not been compiled.

15 lines of code changed in 1 file:

jthornoc 2010-11-11 14:27 Rev.: 46446

Core/Grid/BoundaryConditions/*:
* Added the ability to name a <Face> tag using the attribute "name". Ie:
<Face side="x-" name="my_face"> …
and retrieve that name. By default, the name is "NotSet" if no name is specified.

BoundaryCondition.cc:
* Removed some references to reacting scalar (replaced by new scalar transport)
* Replaced some fortran with C++ implementations
* Added ability to specify new types of inlet including
- Flat Profile
- Function
- Vector
- Input file
Note that the current code should behave as before. Some of the above capabilities now needs implementation and polish.
* Removed some old extra scalar code.
* Added FlowInlet specs.

TabPropsInterface and MixingRxnModel.h:
* Added calculation of third mixture fraction from 2 conserved mixture fractions if using coal (eta and fp are now specified as attributes of the <coal> tag.
* Removed <adiabatic> spec in TabProps for now since the radiation model solely determines if the case is adiabatic or not. Now d_adiabatic is set depending on the radiation model specification.

arches_spec.xml:
* Removed a few unused tags and added spec for changes above.

ups_spec.xml:
* Added attribute spec for changes listed above for <BoundaryConditions><Face>

691 lines of code changed in 17 files:

harman 2010-11-11 14:15 Rev.: 46445

Use the verion of python in the user's path, not /usr/bin/python


Arches.py, ICE.py, Wasatch.py:
- local regression tests are run on "All" operating systems. You should
set this is you're going to run it on Linux and Darwin.

29 lines of code changed in 12 files:

harman 2010-11-11 14:11 Rev.: 46444

makefile:
- pass in SRCTOP_ABS to startLocalTester

startLocalTest
- using SRCTOP_ABS for paths to symbolic links
- changed dialog window geometry for OSX

generateGoldStandards
- throw error if version of python is < 2.4
- use python in the users path , not /usr/bin

selectComponents
- changed dialog window geometry for OSX

runSusTests
- hardening: compare OS strings in all uppercase
- New version of python crash if you try to access an environmental variable
that does not exist. Set default values of outputpath & weboutputpath, modify
if nightly test is being run.


32 lines of code changed in 5 files:

harman 2010-11-11 11:27 Rev.: 46442

added commas

2 lines of code changed in 1 file:

dav 2010-11-10 17:21 Rev.: 46441

use src abs

1 lines of code changed in 1 file:

jsutherland 2010-11-10 16:02 Rev.: 46440

Add a TabProps test in Wasatch

3 lines of code changed in 1 file:

harman 2010-11-09 16:47 Rev.: 46439

fixed the mpirun command for passing in environmental variables.
This had been broken for a long time. The memory tests were being performed correctly.
I'm not sure why.


1 lines of code changed in 1 file:

guilkey 2010-11-09 16:45 Rev.: 46438

Add more information about the particles that are being deleted.

3 lines of code changed in 1 file:

harman 2010-11-09 15:29 Rev.: 46437

- indentation
- removed dead code
- added exits inside of the conditional tree.
- default return code is 1 so it fails
- fixed a long standing syntax error.

0 lines of code changed in 2 files:

harman 2010-11-08 17:48 Rev.: 46436

sus
- use thread::exitAll() to return the error code from sus

generateGoldStandards
- bark at the user if sus returns anything besides 0.

9 lines of code changed in 2 files:

jpeterson 2010-11-08 17:15 Rev.: 46435

Some commenting and cleaning. Added default output of Cv parameter when using JWL/C EOS so that restarts can occur.

11 lines of code changed in 1 file:

guilkey 2010-11-08 16:33 Rev.: 46434

Replace the use of "eigen" with "getEigenValues". eigen seems to have a bug
which causes it to get into a loop from which it never returns.

10 lines of code changed in 1 file:

harman 2010-11-08 15:32 Rev.: 46433

changed script name to be more descriptive

72 lines of code changed in 4 files:

harman 2010-11-08 13:54 Rev.: 46432

Makefile:
- fixed path to generateGoldStandards script

<component>.py
- added comment

startLocalTest
- better error messages
- added loop over all selected component tests

generateGoldStandards
- direct run output to sus.log
- use mpirun if the # processors > 1.0
mpirun -np 1 <command> is valid.

runSusTests
- better error messages

70 lines of code changed in 14 files:

jsutherland 2010-11-06 15:36 Rev.: 46431

- Turn off some operators that we don't need for now.
- quiet a compiler warning

34 lines of code changed in 3 files:

ffjhl 2010-11-05 14:16 Rev.: 46430

Neumann b.c. for heat conduction added by merging
temperature b.c. with pressure b.c. intended for MPMICE.
Should not change existing results.

26 lines of code changed in 1 file:

dav 2010-11-04 15:22 Rev.: 46429

M Makefile.in

- Add in the 'make gold_standards' target.
- Remove some of the i/o from runLocalRT as it can be confusing if error
conditions occur during the runs. Any print statements like this need
to be moved into the script itself at the appropriate location(s).

M R_Tester/ICE.py
M R_Tester/ARCHES.py

- Clean up the modUPS stuff.
- Turn back on ARCHES methane8patch test... I think this should work...
Not sure why it was 'broken' for Todd's side of things... Since Todd
is out I haven't been able to test this fix for the nightly tester.
I'll take a look at the results tomorrow and see where we are.

M R_Tester/toplevel/startLocalTest

- Don't let the script run from dumb/emacs terminals as the 'dialog'
thing doesn't handle them correctly.
- Run python with '-u' (unbuffered output). This fixes Todd's problem
of not seeing any output (from 'tee') until the program completely
finishes.

M R_Tester/toplevel/generateGoldStandards.py

- Produce the correct (I hope) TestData directory. Was
TestData/ICE/advect.ups, now is
TestData/ICE/advect/advect.ups (etc).
- Put in user input for deleting previous gold standards.
- FIXME: still need to do some auto-testing to see if
configure/svn have changed and let the user know...

AM R_Tester/helpers/selectComponents.sh

- A helper script to bring up Todd's GUI to get a list of the
gold standards that the user wishes to generate.

M R_Tester/helpers/runSusTests.py

- There was a problem with env HTMLLOG and WEBLOG not being
set for the local tester. So check for this condition
and don't use them if so.

192 lines of code changed in 7 files:

luitjens 2010-11-03 09:23 Rev.: 46427

Disabled methane8patch until Dav can fix the testing scripts.

3 lines of code changed in 1 file:

harman 2010-11-02 15:10 Rev.: 46426

cleaned out unnecessary
#includes

Added some formatting and documentation.

0 lines of code changed in 2 files:

harman 2010-11-02 12:20 Rev.: 46425

Pass the environmental variable "LOCAL_OR_NIGHTLY_TEST" in to the runSusTests script.

Only create the webpage if running the Nightly RT.

127 lines of code changed in 3 files:

harman 2010-11-02 11:01 Rev.: 46424

removed scalability code. It's not being used.

19 lines of code changed in 1 file:

harman 2010-11-01 17:29 Rev.: 46423

if the user is running a local RT and a test's goldstandard does not exist
the exit with an error message.

8 lines of code changed in 1 file:

harman 2010-11-01 17:12 Rev.: 46422

changes to account for the ".py" extension added to the component files.

2 lines of code changed in 2 files:

harman 2010-11-01 16:38 Rev.: 46421

take into account the addition of ".py" to all the component tests

1 lines of code changed in 1 file:

harman 2010-11-01 16:31 Rev.: 46420

updated functions getNightlyTests() & getLocalTests() to return the right tests.

18 lines of code changed in 9 files:

dav 2010-11-01 16:24 Rev.: 46419


Add 'generateGoldStandard' python script. Renamed the test files (ICE, MPM, etc) to have a '.py' extension (as they are python files).

976 lines of code changed in 25 files:

harman 2010-11-01 16:15 Rev.: 46418

fixed error in conditional

1 lines of code changed in 1 file:

jeff788 2010-11-01 12:52 Rev.: 46417

added localization capability to Kayenta

360 lines of code changed in 3 files:

harman 2010-11-01 12:28 Rev.: 46416

Only email John if the GS is owned by csafe-tester. You don't want to spam
John everytime a local regression test GS is replaced.

5 lines of code changed in 1 file:

harman 2010-11-01 11:34 Rev.: 46415

Improved error messages

Redirect command output to /dev/null. I don't want to see that csafe group doesn't exist.

Tee the output from the running tests to stdout and the log file.

9 lines of code changed in 3 files:

harman 2010-11-01 10:16 Rev.: 46414

-added dialog user input for selecting the component.
-added bulletproofing

51 lines of code changed in 2 files:

October 2010 »

Generated by StatSVN 0.4.0