Update locations for Wasatch3P git repos
7 lines of code changed in 4 files:
Build CUDA code for the TaylorGreen vortex
55 lines of code changed in 3 files:
reduce the amount of explicit instantiations in Wasatch to further reduce the size of the Wasatch library.
(1). Postprocessing expressions are SVOL only
(2). Parsing said expressions doesn't need to explicitly instantiate all types
132 lines of code changed in 5 files:
Remove some unnecessary explicit template instantiations to speed up compile time and reduce the size of the Wasatch library a little bit.
6 lines of code changed in 6 files:
display time after code has compiled
2 lines of code changed in 1 file:
remove a few unused variables
0 lines of code changed in 1 file:
Revert "Templated wallHT model to interface with a generic type for directional radiative fluxes. "
This reverts commit fad7a4aa53a6c03168d3c25e8d2094e6554c70f8.
170 lines of code changed in 4 files:
fix failing RT tests for Wasatch.
2 lines of code changed in 2 files:
Updated configure to match previous commit of configure.ac
4 lines of code changed in 1 file:
M configure.ac
- Updated test for checking for mpi compiler on BGQ machine.
- I don't have a new enough autoconf on Vulcan so only committing configure.ac... will commit updated configure soon.
M CCA/Components/SimulationController/AMRSimulationController.cc
- Minor syntactic cleanups.
M CCA/Components/SimulationController/SimulationController.cc
- The 'mean' is associated with the 'elap T', so place in (), as opposed to be separated with a comma.
M Core/Grid/Level.cc
- Fix ordering of #includes.
- Line up variables / initialization / etc for easier reading.
- Add some whitespace for easier reading.
- Remove tabs, standarize intention.
- Renamed a few class variables with d_.
- Add {}s.
M Core/Grid/Level.h
- Line up functions so they are easier to find.
- White space, line up variables and functions to make them eaiser to read.
- One line implementations in .h files should be on the same line as function name.
This makes parsing a .h file for member functions much easier.
- grid is now d_grid.
M Core/DataArchive/DataArchive.cc
- Remove debug print for 1000th processor.
528 lines of code changed in 6 files:
Switch convection in cqmom to use static vector instead of cc-wrapper
104 lines of code changed in 4 files:
This is the promised second commit to add the remaining files and updates needed to run the Alex Brown Soot Model using Arches.
692 lines of code changed in 9 files:
This is the first of two commits to add the Alex Brown Soot Model into the Arches code. Second commit will contain remaining source term files as well as changes in a few other files.
1251 lines of code changed in 6 files:
- removed DOS carriage returns.
- light formating.
0 lines of code changed in 2 files:
Variable name change:
t -> tstep
patch->patch1
Added bulletproofing. Does the variable exist in both DAs on this timestep?
This problem mainly occurs if <outputInitTimestep> has been specified.
* there are still issues.
0 lines of code changed in 2 files:
switch from using the cc-wrapper to static array in cqmom
47 lines of code changed in 6 files:
cleanreally
Find any *.o or *.d file in the Components directory and delete it.
If you first compile with all components enabled and then turn them off
the *.o and *.d files were left behind.
2 lines of code changed in 1 file:
turn off methane_rmcrt(). This test is causing compute_Lnorm to crash, freezing the nightly RT.
1 lines of code changed in 1 file:
added calls for float variables.
30 lines of code changed in 1 file:
Switch pows to for loops for minor speed up. This will break the last CQMOM RT
6 lines of code changed in 1 file:
Refactor the CQMOM source terms so that only one task exists to calculate the source term for each equation at once. This provides a significant speed up in CQMOM.
220 lines of code changed in 9 files:
Fix warnings.
16 lines of code changed in 20 files:
changed sign on divQ
2 lines of code changed in 1 file:
Change sign notation on heat flux divergence for RMCRT to be consistent with discrete-ordinates. This will break all RMCRT regression tests.
3 lines of code changed in 2 files:
Set OncePerProc type for Hypre's scheduleSolve().
1 lines of code changed in 1 file:
Fixing a bug in the emitted flux. The area was artificially too large.
24 lines of code changed in 1 file:
volFraction was not retrieved properly from the DW.
7 lines of code changed in 1 file:
Modifications to the wall_heatflux model.
86 lines of code changed in 2 files:
Fix a restart problem for squared sum variables when adding a new
average.
4 lines of code changed in 1 file:
Some error checking on boundaries.
29 lines of code changed in 2 files:
More cleanup on GPU logic in TaskInterface.
81 lines of code changed in 1 file:
Fix failing gpu tests for Wasatch.
4 lines of code changed in 1 file:
Fix a bug that was triggering GPU execution when Uintah said "no"
5 lines of code changed in 1 file:
Clean up some compiler warnings about extra ";"
36 lines of code changed in 35 files:
Full beta version of the diffusion code with all of the CFI corrections
implemented. Much testing still to be done.
376 lines of code changed in 14 files:
* Added output to exectimes.x.x files to indicate which timesteps that TaskExecTimes have been accumulated over.
* Updated execsummary output to indicate which timesteps that TaskExecTimes have been accumulated over.
33 lines of code changed in 6 files:
Revert commit 53533: For once per proc task, turn off all but one thread.
1 lines of code changed in 1 file:
Put the initialization of the max_concentration inside of a conditional on
doScalarDiffusion.
3 lines of code changed in 1 file:
Eliminate linking with cudart_static with static builds. This is not working on Titan due to the current CUDA Toolkit version, 5.5. Once Titan is upgraded to CUDA 6.5, we can then use cudart_static in static builds.
* generated new configure script
24 lines of code changed in 2 files:
Accurately report specs on all available on-node GPUs.
5 lines of code changed in 1 file:
Comment out multi-GPU code that's not working on local SCI cluster. Will fix this soon.
7 lines of code changed in 1 file:
Remove ProblemSetupException thrown when trying to use > 1 available GPU per node. We correctly support multiple GPUs.
5 lines of code changed in 1 file:
Remove unused type CQMOM tag
0 lines of code changed in 2 files:
Change pow to for loop multiplication for speed up. This will change RT answers slightly.
12 lines of code changed in 1 file:
Use temporary doubles for CQMOM convection for reused CC variables
114 lines of code changed in 1 file:
Progress on implementing multi-level scalar diffusion. Hopefully simplified
the interface, using the SDInterfaceModel only when necessary. Also, no longer
having a separate task for interpolating scalar data to/from the grid. Those
jobs are now done inside of the AMRMPM tasks, with conditionals on scalar
diffusion. This reduces the code duplication dramatically, and will make for
more easy to maintain code in the long run. Still need to add some CFI tasks
for the scalar diffusion calculuations.
766 lines of code changed in 10 files:
Templated wallHT model to interface with a generic type for directional radiative fluxes.
RMCRT can now be used in combination with the year 1 coal physics code.
351 lines of code changed in 4 files:
Fix a seg-fault that occurred when debugging the TaskGraph. (ie, SCI_DEBUG => TaskGraph:+)
This occurred if a task did not have a matl set or a patch set.
M Grid/Task.cc
- Add white space for easier (human) parsing.
- If matl_set or patch_set are null, then print out No-Matl-Set/No-Patch-Set.
- Added {} to a few for statements.
M Grid/Variables/ComputeSet.cc
- Add white space for easier (human) parsing.
- Added {} to some if statements.
0 lines of code changed in 4 files:
no spaces after commas for debug flags
5 lines of code changed in 3 files:
update description on how to set multiple sci_debug flags
9 lines of code changed in 2 files:
sched_rayTrace_dataOnion()
-added warning about high communication costs on coarse levels
-removed sanity debugging code.
2 lines of code changed in 2 files:
Tested on Ash, BigMac, and Aurora.
Fixes for Tony S. so that new routines can be placed in SolverInterface.cc.
Here is the fix for placing code from SolverInterface.h into SolverInterface.cc.
The two issues fixed with this commit are 1) Added SolverInterface.cc to the
list of files to compile, and 2) added Core/Math to the list of
libs that must be linked (as it was missing), both in sub.mk
M Ports/SolverInterface.cc
M Ports/SolverInterface.h
- Moved some new functions from the .h file into the .cc file.
- Align, remove tabs, clean up.
M Ports/sub.mk
- Add SolverInterface.cc to list of files to compile.
- Align, remove extra spaces, alphabetize, make eaiser to follow, remove duplicate lib listings.
- Added Core/Math to the list of libs to link against.
335 lines of code changed in 5 files:
Fix the initializatino of this term
0 lines of code changed in 2 files:
Eliminate compiler warning about comparison between signed and unsigned integer expressions.
1 lines of code changed in 1 file:
* Don't pass c++11 flag to NVCC. NVCC from CUDA <=6.0 doesn't support it. We don't use C++11 features in any NVCC compiled code anyway (yet).
* this allows us to run using C++11 features in Uintah code and CUDA 5.5 on Titan
* generate new configure script
* Fix C++11 compiler error about inability to find string literal in Allocator.cc (SCIRun::PrintTag(void*))
46 lines of code changed in 5 files:
If you need lapack, you need blas.
0 lines of code changed in 2 files:
Fix failing cqmom tests
6 lines of code changed in 2 files:
Don't exectute this task unless it is turned on.
6 lines of code changed in 1 file:
Now outputing #MPI procs, #threads per MPI proc, and total Threads, example below:
Now computing the averages for the SCI_DEBUG: ComponenTimings variables, if they exist.
#file MPIprocs threadsPerMPI totalThreads aveTaskExec aveTaskLocalComm aveTaskWaitComm
small.16.8.128.out 8 16 128 116.084 0.847692 0.0267732
small.16.16.256.out 16 16 256 103.672 1.18166 0.0549489
#file MPIprocs threadsPerMPI totalThreads startTime endTime aveMean
small.16.8.128.out 8 16 128 276.84 921.17 92.0471
small.16.16.256.out 16 16 256 247.96 825.04 82.44
22 lines of code changed in 1 file:
Don't source users .cshrc file.
1 lines of code changed in 1 file:
* Adding a nebo-ized version of the density predictor in Arches.
This new version can also compute the exact rho-f/rho relationship for
cold-flow mixing.
* Also added a get-task-by-type mechanism in the task factory base
class.
298 lines of code changed in 6 files:
Limit the warning for increasing the cuda print buffer size to proc 0.
1 lines of code changed in 1 file:
Added <SRC> to rsync command which moves results to the webpage.
2 lines of code changed in 1 file:
getItem() & getLevelItem()
- You must find the patchID or levelIndex before searching the variable DB for the label. In simulations
with a large number of patches the previous method would find the label for the wrong patch and
crash.
This bug was tricky!
0 lines of code changed in 4 files:
The recent commit of the scheduler for the velocity initialization was
occuring after CC velocities were being computed. This caused diffs in
the RT. This has been corrected.
Also, update input file to latest spec.
3 lines of code changed in 1 file:
fixed command to copy results to web page.
2 lines of code changed in 1 file:
Revert ""
This reverts commit 730400d0fb991f08d4dd64212d7909ac9d89e88b.
219 lines of code changed in 3 files:
Fixes for Tony S. so that new routines can be placed in SolverInterface.cc.
M SolverInterface.cc
M SolverInterface.h
- Moved some new functions from the .h file into the .cc file.
- Align, remove tabs, clean up.
M sub.mk
- Add SolverInterface.cc to list of files to compile.
- Align, remove extra spaces, alphabetize, make eaiser to follow.
329 lines of code changed in 3 files:
Add in the 7D internal coordinate and a source term case to CQMOM tests.
8 lines of code changed in 1 file:
Refactor the convective term for CQMOM transport equations. This provides a significant performance increase. This commit will break most of the CQMOM regression tests.
1603 lines of code changed in 12 files:
Minor fix for compliers that don't have a uint typedef
1 lines of code changed in 1 file:
Added RT for full coal physics problem. This commit includes
a 6MB chemistry table for the OFC. This 3-D table has properties
tabulated in terms of composition (coal-gas, pure O2, pure CO2),
and heat-loss. This commit will cause an arches RT failure on the
new input file OFC4.ups, which was shown to be the most physical
parameter set for the OFC.
2 lines of code changed in 1 file:
Removing a "using namespace" in the CQMOMInversion header.
Punishment will be dealt with internally.
8 lines of code changed in 1 file:
Fixing a bug in the almgren mms initial condition.
20 lines of code changed in 1 file:
* Create a separate task for initializing the velocity fields
* add the 2D, variable density MMS (Shunn/Moin) with corresponding source terms for f and continuity
* simplify the cold flow input interface
* consolidate the scalar initial condition interface into the base class
* use the solvability condition rather than the reference pressure
* fix the continuity residual check to include the effects of the source terms and the mass source from the solvability condition
* fix the continuity redidual check to include other mass source terms
1200 lines of code changed in 22 files:
Use Tony's solvability condition for the pressure.
3 lines of code changed in 1 file:
Turn on one-d pulse test.
2 lines of code changed in 1 file:
handle a few corner cases for the generalized script that computes
temporal order. Specifically, handle the following cases:
* user did not specify outputInitTimestep
* user specified outputInterval instead of outputTimestepInterval
* user did not specify an output frequency
* user did not specify max_Timesteps
Please send me feedback, bug reports, or suggestions
if you are using this script.
43 lines of code changed in 1 file:
cleaner way to copy results to html directory.
2 lines of code changed in 1 file:
May 2015 »