March 2017 Commit Log

Number of Commits:
78
Number of Active Developers:
13
tsaad 2017-03-31 13:06 Rev.: 56385

replace !=0 with != nullptr when checking the validity of ProbleSpecP. This was causing compilation issues on my mac.

2 lines of code changed in 1 file:

guilkey 2017-03-30 23:28 Rev.: 56383

Fixes to the logic of old and new localized.

4 lines of code changed in 1 file:

guilkey 2017-03-30 23:26 Rev.: 56382

Formatting.

1 lines of code changed in 1 file:

guilkey 2017-03-30 21:56 Rev.: 56381

Rearranging a few lines for efficiency, and so that failTime is computed
after time of localization is set.

11 lines of code changed in 1 file:

guilkey 2017-03-30 21:48 Rev.: 56380

Re-enable ability to delete a troublesome particle.

2 lines of code changed in 1 file:

harman 2017-03-30 15:21 Rev.: 56379

The user can set:

<exitOnCrash> false</exitOnCrash>

which allows the remaining tests to keep running if an exception is thrown. For example if you wanted to
test the max CFL you can get away.

24 lines of code changed in 1 file:

milo 2017-03-28 22:30 Rev.: 56371

added german_lignite to ash database for emissivity calculations

18 lines of code changed in 1 file:

harman 2017-03-28 08:35 Rev.: 56363

Comment out assertion that's hit when a single task has 2 modifies for the same variable.


2 lines of code changed in 1 file:

bisaac 2017-03-28 00:52 Rev.: 56362

This commit adds particle shrinkage and swelling, modifies the density function to compute the literal density with dqmom, and makes input files consistent with coal_density. Shouldn't change any answers.

904 lines of code changed in 11 files:

dav 2017-03-27 17:07 Rev.: 56361


- Some tweaks to cleanup w3p build.

M ExprLib-build/doit
M RadProps-build/doit
M TabProps-build/doit
M SpatialOps-build/doit

- Don't need continuation chars (\) in a string.
- Put all the CUDA flags into the CUDA if section.

M NSCBC-build/doit

- There apparently isn't a "make clean" in NSCBC.

0 lines of code changed in 10 files:

cgritton 2017-03-27 16:28 Rev.: 56360

Added feature to calculate values of potential in the extra cells based on specified boundary conditions

155 lines of code changed in 1 file:

jsutherland 2017-03-27 16:10 Rev.: 56359

Commit for Josh McConnell to fix builds on Titan.

1 lines of code changed in 1 file:

guilkey 2017-03-23 16:57 Rev.: 56356

Update comment regarding system requirements for this utility.

1 lines of code changed in 1 file:

dav 2017-03-23 16:40 Rev.: 56355


Adding back in the TIFF make variables. Note, when TIFF (and JPEG, etc) were ripped out of Uintah,
a lot of the infrastructure for them was left in (for example, configure kept checking for them), so
to put it back all I had to do was resurrect the TIFF vars in the Makefile (well, actually in
configVars.mk which is included into the Makefile).

M configVars.mk.in

- Add in HAVE_TIFF and TIFF_LIBRARY flags.

M StandAlone/tools/extractors/sub.mk

- Since I added back the HAVE_TIFF and TIFF_LIBRARY flags to the make system, I am reverting
the change to this sub.mk as it now (again) just works.


4 lines of code changed in 2 files:

harman 2017-03-23 16:38 Rev.: 56354

Jim's fix to the nan issue.


4 lines of code changed in 1 file:

harman 2017-03-23 15:51 Rev.: 56353

SerialMPM, AMRMPM:
purged findRogueParticles()
updateErosionParameter()

Constitutive Models:

Reducing redundant code!

Removed the local version of pLocalizedLabel that the CMs were managing
and then copying to pLocalizedMPMLabel. There's only pLocalizedMPMLabel.
The Constituive models _can_ compute it and the damage models will modify it.

purged: addRequiresDamageParameter(), getDamageParameter()


DamageModels:
purged: computeScalarDamage().

167 lines of code changed in 37 files:

dav 2017-03-23 15:32 Rev.: 56352


I think TIFF support was yanked out of Uintah a long time ago, but Todd asked me to look at compiling the
left over particle2tiff.cc file. These commits allow for that to happen (assuming libtiff is installed
in the system library location correctly). I am not adding any configure support, so you will have to
manually >


M extractors/particle2tiff.cc

- Few minor tweeks to get the file to compile with the newerer Core. I have not actually tried running
it to see if it actually works.

M extractors/sub.mk

- Manually added "-ltiff" the LIBS variable. Before (I think) configure, figured this out for you.
- Also, to make this work, you need to manually comment out (use: #) the check for HAVE_TIFF. So do the following:

#ifeq ($(HAVE_TIFF),yes)
.
.
.
#endif

5 lines of code changed in 2 files:

cgritton 2017-03-23 11:28 Rev.: 56351

Cleaning up some of the compiler warnings

6 lines of code changed in 1 file:

cgritton 2017-03-23 10:27 Rev.: 56350

Coupled finite volume solver for Gauss's equation with exisiting mpm code.

1552 lines of code changed in 27 files:

bisaac 2017-03-23 09:38 Rev.: 56348

Adding the new char oxidation model with the particle shrinkage soure-term, and additional functions within ParticleTools. This should not impact regression tests.

1364 lines of code changed in 9 files:

jthornoc 2017-03-22 13:49 Rev.: 56347

Remove some unused boost headers.

Get system pressure for brown soot model from the table.

17 lines of code changed in 4 files:

dav 2017-03-21 11:58 Rev.: 56343


M doit

If a W3P repository directory doesn't exist, git clone it... Before it assumed that if one was missing,
they all were. Now just check and update one at a time. This probably only occurs if an error happened
during the initial checkout, or if a new lib is added to the 3P.

0 lines of code changed in 2 files:

dav 2017-03-21 11:51 Rev.: 56342

_M .

M doit

- More verbose.
- Quit if error detected.

M HelperBuildScripts/checkEnv.sh

M ExprLib-build/doit
M RadProps-build/doit
M TabProps-build/doit

- Display CMAKE command.

M SpatialOps-build/doit

- Display CMAKE command.
- Quit if error in cmake.

M README

- Cleanup/update.


100 lines of code changed in 9 files:

guilkey 2017-03-21 11:12 Rev.: 56341

Add a requires for p.mass.

1 lines of code changed in 1 file:

harman 2017-03-20 17:01 Rev.: 56340

Moved the HancockMacKenzie and JohnsonCook damage models from the guts of the CM into separate tasks, consistent
with the Brittle and Threshold damage models. Updated the relevant ups and xml include files.

** Some of the nightly MPM tests will fail with differences in p.damages. The differences are small and
due to differences in pStress used in the computation.

516 lines of code changed in 17 files:

milo 2017-03-20 11:11 Rev.: 56339

Adding David Lignell soot changes. This should not affect regression tests.

106 lines of code changed in 3 files:

bisaac 2017-03-17 14:45 Rev.: 56338

Added a new regression test which uses the new emissivity model and thermal conductivity model. Doesn't effect any other regression tests.

3 lines of code changed in 1 file:

derekhar 2017-03-17 13:29 Rev.: 56337

Added pressure to oxidationSmith. Also corrected rawcoal computed for use in rate clipping.

25 lines of code changed in 2 files:

bisaac 2017-03-16 16:59 Rev.: 56336

Added Hadley thermal conductivity model. This will change the answers for 1GW_RT, however these changes are expected. Also cleanup an unused flag in 1GW_RT.ups.

257 lines of code changed in 2 files:

cgritton 2017-03-16 15:43 Rev.: 56335

Working solver for electrostatic based on Gauss's law.

287 lines of code changed in 9 files:

dav 2017-03-15 18:16 Rev.: 56334


Todd asked that the SCI_DEBUG name be treated in a case-insensitive manner...
So now you can do:

setenv SCI_DEBUG my_debug_stream:+
or
setenv SCI_DEBUG MY_DEBUG_STREAM:+
or
setenv SCI_DEBUG my_DEBUG_stream:+

and they will all work and refer to the same thing.

M Core/Util/DebugStream.h
M Core/Util/DebugStream.cc

- Fixed formatting, naming conventions, spacing, etc.

125 lines of code changed in 2 files:

derekhar 2017-03-15 15:43 Rev.: 56333

Changed convergence tolerance in charoxy_smith to use a relative tolerance, based on fastest reaction.

5 lines of code changed in 1 file:

harman 2017-03-15 14:24 Rev.: 56332

ElasticPlasticHP::computeStressTensorImplicit()
- Consistent with computeStressTensor, compute pLocalizedMPM = false.

ConstitutiveModelFactory::create()
- minor code consolidation.

5 lines of code changed in 2 files:

tsaad 2017-03-15 14:23 Rev.: 56331

Fix w3p build on Titan.

2 lines of code changed in 1 file:

jthornoc 2017-03-15 12:02 Rev.: 56330

Adding access to the table keys through the ChemHelper.
This will prevent passing the entire table around when one just needs a
small subset of information.

22 lines of code changed in 3 files:

harman 2017-03-15 11:23 Rev.: 56328

computeStressTensorImplicit
- calculation of strain energy is consistent with what is in computeStressTensor() task.

13 lines of code changed in 1 file:

derekhar 2017-03-14 15:34 Rev.: 56327

Added unit-vectors for setting non-orthogonal boundary conditions.

57 lines of code changed in 2 files:

dav 2017-03-13 17:59 Rev.: 56326


Fix code to remove compiler warnings spewing out of GCC 6.1. These are (I hope) purely cosmetic changes.

M Core/Math/TntJama/tnt_fortran_array2d_utils.h
M Core/Math/TntJama/tnt_fortran_array3d_utils.h
M Core/Math/TntJama/tnt_array2d_utils.h
M Core/Math/TntJama/tnt_array3d_utils.h

- Fixed truly atrocious (random?) formatting...
- Remove tabs.
- Coding standard fixes: Use {} for 'if' statements, function return type on its own line, etc.

M CCA/Components/MPM/SerialMPM.cc

- Fix incorrect indentation.
- Coding standard fixes: Use {} for 'if' statements, function return type on its own line, whites pace, alignmnet of variables, etc.

681 lines of code changed in 5 files:

harman 2017-03-13 16:24 Rev.: 56325

ElasticPlasticHP:
- removed old damage model that was replaced by the threshold model.

ThresholdModel:
- Now allowing p.LocalizedMPM+ to be computed upstream.

test_hypoviscoelastic_rad_ret.ups:
- updated to new damage model specs.

22 lines of code changed in 3 files:

harman 2017-03-13 15:24 Rev.: 56324


Added wiring to MPMMaterial so pLocalizedMPM can be computed upstream of the damage models.

ElasticPlasticHP.cc
- now computing p.localizedMPM+
- Added some of Jim's suggested changes.

test_hypoviscoelastic_radial_return: fails due to different number of checkpoints.


149 lines of code changed in 15 files:

harman 2017-03-13 15:16 Rev.: 56323

formatting.

139 lines of code changed in 2 files:

jthornoc 2017-03-13 14:37 Rev.: 56321

Update private member naming. Still needs some TLC.

213 lines of code changed in 2 files:

jthornoc 2017-03-13 14:37 Rev.: 56320

Allow an intrusion inlet to specify only a fraction of the total scalar
variables. Unspecified variables are assumed to have a zero flux.

Also, start fixing the variable naming in DQMOM.

292 lines of code changed in 4 files:

jas 2017-03-10 16:32 Rev.: 56317

Update for changes to petsc 3.6+

10 lines of code changed in 3 files:

ahumphrey 2017-03-10 12:28 Rev.: 56314

Some edits that didn't make it into the last committ.

Use auto specifier for C++11 compiler type deduction (especially when decl is verbose).
Use preincrement when looping over ADTs.

8 lines of code changed in 1 file:

ahumphrey 2017-03-10 12:08 Rev.: 56313

Use RAII style std::lock_guard<std::mutex>.
Easier and safer to deal with. No explicit locking or unlocking required - locked when created, unlocked when DTOR called (when out of scope).

20 lines of code changed in 1 file:

bisaac 2017-03-09 18:09 Rev.: 56312

Added the new ash-emissivity model. This will break 1GW_RT because we have added new variables. The values saved in the .uda are unchanged.

278 lines of code changed in 5 files:

derekhar 2017-03-09 17:03 Rev.: 56311

corrected warning/message for ellipse boundary object.

1 lines of code changed in 1 file:

cgritton 2017-03-09 12:03 Rev.: 56310

Removed s from comments, should not have been there.

0 lines of code changed in 1 file:

cgritton 2017-03-09 11:58 Rev.: 56309

Updated boundary conditions to include Neumann bc and added basic code for Gauss's law

898 lines of code changed in 13 files:

harman 2017-03-09 10:57 Rev.: 56308

added comment

1 lines of code changed in 1 file:

harman 2017-03-08 17:58 Rev.: 56307

Fixed the strain energy calc. -Jim.

StrainEnergy.dat
Disks_complex vs goldstandards:Disks_complex

Comparing StrainEnergy.dat... *** failed
greatest relative error: %0.946582604201865
at times: 0.057123919016210639 / 0.057123919016210639
values: 10.64879348504321 / 10.54799385835641
and first signifant relative error: %0.110636060559455
at times: 0.032087934780574333 / 0.032087934780574333
values: 0.93112880285718025 / 0.93009863863096487

greatest absolute error: 0.192542355905516
at times: 0.041295456745894979 / 0.041295456745894979
values: 86.47849557805489 / 86.285953222149374
and first signifant absolute error: %0.10301642262156
at times: 0.032087934780574333 / 0.032087934780574333
values: 0.93112880285718025 / 0.93009863863096487


11 lines of code changed in 1 file:

jthornoc 2017-03-08 17:55 Rev.: 56306

Added a non-normal, massflow rate condition.

3 lines of code changed in 1 file:

harman 2017-03-08 17:46 Rev.: 56305

-corrected output problem spec label.
-fixed initialization of d_charTime and pTimeOfLoc.

* The results for disks_complex compare exactly with the goldstandards

9 lines of code changed in 1 file:

harman 2017-03-08 11:03 Rev.: 56304

The erosion code has been consolidated and moved to a new ErosionModel. It is independent of the CM and
each MPM material can apply a different erosion algorithm. Previously, one algorithm for all matls.
The default model is none.


MPM tests pass, some MPMICE tests fail due to the number of checkpoints changing.

559 lines of code changed in 28 files:

ahumphrey 2017-03-08 01:13 Rev.: 56303

Correctly compartmentalized DetailedTask and DetailedTasks classes with race condition fixed.

1315 lines of code changed in 6 files:

ahumphrey 2017-03-07 23:27 Rev.: 56302

Cleanup DWDatabase. Fix compiler warning.

107 lines of code changed in 2 files:

ahumphrey 2017-03-07 20:27 Rev.: 56301

Backing out committ 56299 for now. The optimizations made in this commit have triggered the DWDatabase race condition Brad and I discussed at a high level recently.

1332 lines of code changed in 19 files:

allen 2017-03-07 16:40 Rev.: 56300

clean up of the PIDX checkpointing

77 lines of code changed in 1 file:

ahumphrey 2017-03-07 16:21 Rev.: 56299

Separate decl/impl of DetailedTask and DetailedTasks classes.
This drastically reduces confusion when trying to study any of this code (e.g. MPI engine and TG compile stuff).
Also removed significant amount of dead and/or redundant code.
More work to conform to what seems to be forming into a more widely adpoted coding standard.

1507 lines of code changed in 19 files:

ahumphrey 2017-03-06 22:53 Rev.: 56298

Fix GPU build errors

2 lines of code changed in 1 file:

ahumphrey 2017-03-06 22:39 Rev.: 56297

Cleanup up unused #includes, and remove remnants from original SingleProcessor Scheduler support.

26 lines of code changed in 2 files:

ahumphrey 2017-03-06 22:06 Rev.: 56296

More infrastructure cleanup.

92 lines of code changed in 5 files:

allen 2017-03-06 17:18 Rev.: 56295

minor depandic name changes to the UPS

61 lines of code changed in 5 files:

jthornoc 2017-03-06 15:36 Rev.: 56294

Pulling the coal_mass_clip model. With our rate clipping, not only is
this model unnecessary, but is actually causes cases to go unstable.

While four RT tests used this model, only coal_RadPropsPlank seems to be
affected.

1 lines of code changed in 5 files:

allen 2017-03-06 14:33 Rev.: 56291

changed the GUI and updated it for new var names

138 lines of code changed in 5 files:

allen 2017-03-03 17:00 Rev.: 56289

added new vars to the in-situ

32 lines of code changed in 3 files:

allen 2017-03-03 16:49 Rev.: 56288

added new vars to the in-situ

189 lines of code changed in 7 files:

allen 2017-03-03 15:25 Rev.: 56287

add min/max bounds to the in-situ vars

89 lines of code changed in 1 file:

harman 2017-03-03 14:50 Rev.: 56286

Purged some old code that's no longer needed.

0 lines of code changed in 6 files:

allen 2017-03-03 14:09

add min/max bounds to the in-situ vars, fixed handle checks

223 lines of code changed in 13 files:

harman 2017-03-03 11:38 Rev.: 56283

Moved scheduleUpdateStress_DamageModel() from scheduleComputeStressTensor() to
scheduleTimeAdvanced()

The IMPM regression tests now run to completion but the number of checkpoint variables has changed.

3 lines of code changed in 2 files:

harman 2017-03-02 18:23 Rev.: 56282

SerialMPM:
- moved common damage related methods to MPMCommon

AMRMPM & ImpMPM
- incorporated new damage related plumbing.
- purged compute/requires and initialization of pLocalizedMPM. It's handled via the
damage Models.

* All AMRMPM tests pass.

112 lines of code changed in 6 files:

harman 2017-03-02 16:13 Rev.: 56281


cmdLine()
- return stdout, stderr and the process return value.

Fixed bug introduced in previous commit related to the gpu test.


9 lines of code changed in 1 file:

harman 2017-03-02 11:14 Rev.: 56279

-removed instructions on how to replace the GS when a test fails.

-added function that should keep the error msg from failed tests from becoming scrambled.


18 lines of code changed in 1 file:

harman 2017-03-01 17:44 Rev.: 56277

Pulled the damage code from the constitutive models and put it in a separate class. This improves
the reusability when added to other CMs. Purged some dead wood.

Note the strain energy is not decremented when a particle is damaged. This should be fixed.

Nightly MPM and MPMICE regression tests pass except for several in which the number of checkpoints variables
differs.

1228 lines of code changed in 21 files:

harman 2017-03-01 17:12 Rev.: 56276

-removed messages about how to update the GS.
-show the user why compareUda failed if it's something besides a tolerance that's been exceeded.

6 lines of code changed in 1 file:

ahumphrey 2017-03-01 17:08 Rev.: 56275

More infrastructure cleanup, specifically DWDatabase before fixing a discovered race condition.

309 lines of code changed in 6 files:

February 2017 »

Generated by StatSVN 0.7.0