GitLab

Description:

VERA-dev Issue # - ####

Update README to reflect migration to code.ornl.gov

See merge request futility/Futility!339

Resolve invalid math

See merge request futility/Futility!334

Add is_between

See merge request futility/Futility!337

Initial gitlab ci test

See merge request futility/Futility!338

* Adds gatherv for rank-1 SNK and SLK integers

* Fix INTENT(IN)/INTENT(INOUT) issues

Description:
The MPI_IN_PLACE keyword needs to be swapped from the send to recv buffer in the allreduce call.
Current implementation fails to correctly sum data silently on some problems (including 3-mini in the Heavies), though this was not previously observed.

* Create addExceptionHandlerSurrogates in FutilityComputingEnvironment

This allows a client code to easily add its own exception handler as
a surrogate to all Futility exception handlers in a single call

* Fix a couple mistakes

* Fix reference to eXDMF when HDF5 is not enabled

* If at first you don't succeed...

* auto-open HDF5 and use REQUIRE more

* Fix some formatting issues in HDF5

* Add vs code to gitignore

* Update HDF5 test

* Add gatherv generic to MPI environment

* Fix typo when MPI is undefined

* Fix serial version of testParallelEnv

* Add CheckMPIF2008StandardCompiles and switch Futility_Configurations to use it

* Specify full path to fortran standard check function

* Native matrices now determine indices for data transfer at assemble-time

* Changed to Waitany for nonblocking

* Moved local computation to end

* Updated NNZ to be compatibly with PETSc o/dnnz parameters

* Resolved CI python version mismatch

* Fixed a bug where too much memory was being allocated by Native matrix init

* Add handling for arrays of strings of zero length.

* Move array allocations to the beginning to ensure the are allocated to the correct length for all cases.
Co-authored-by: Luke Cornejo <3lc@ornl.gov>

Something was being cleared twice in Trilinos due to how we
were calling various functions.  This fixes that issue and
fixes 2 other unrelated valgrind defects.  The test now passes
in release-debug with no memory leaks or other valgrind defects.

* Add a DBC_FAIL_VERBOSE logical

* Use DBC_FAIL_VERBOSE in a missed spot

* Fix uninitialized value in IAPWS module

* Make ireg intent out in regsopt

* Foundations of the XDMF mesh and file types.

* Setup of tests

* Working on reading XDMF

* Geometry processing complete

* Fixed crazy pointer issue with function by changing to subroutine

* Working up through vertices.

* Working up through vertices.

* Material import working

* Everything but mixed topo working for two pins test

* Prior to recursive assignment mods

* Working for multilevel grid after change to assignment

* Added mixed topology, just need to comment everything.

* Commented all import code and tests

* Whitespace

* Working on export. Also fixed CMake for HDF5 dependance

* XDMF export working up through leaves

* Working on writing leaves

* Working on geometry. Allocatable issue resolved

* Export working for single topology. Changed xdmf to UTF8 encoding.

* Mixed topology working

* Material ID export working

* All import export working. Need to implement tests.

* Added newline handling to xdmf

* Testing finished.

* Added distanceToLeaf function

* Added bounding box to XDMFMeshType

* Addressed thfolk suggestions, changed DBC to error handling

* Clear cell sets name with mesh%clear

* Clear boundingBox

* Addressed yl5 changes other than xml privacy

* Addressed yl5's comments

* Made children private

* Rewrote unit tests, renamed module, and removed file type

* Removed whitespace and a few blank lines.

* Add comparison of mixed precisions into IntrType.

* Make the mixed precision as an option through compiler preprocessing

* Add scaling factors for unit testing with mixed precisions.

* Add assertions for mixed precison operators.

* Address Ben and Aaron's comments.

* Fix failing test due to restructuring the operators.

* Remove unused variables.

* Adds IAPWS water properties from CTF (SHA1=eb08b78aa7) and reformats most of the routine

* Completed unit test

* Updated formatting, removed goto's, added exception handler

* Added SAVE attribute to a few variables

* Address PR comments, mostly formatting

* Removes real inequality using input tolerance

* Add missing SRK values

Description:

Gitlab Ticket # - XXXX

* Add documentation

Description:
Also correct some variable names

Gitlab Ticket # - XXXX
Co-authored-by: Benjamin Collins <bn7@ornl.gov>
Co-authored-by: Bob Salko <salkork@ornl.gov>

* disables the FMU unit test if BUILD_SHARED_LIBS is FALSE since FMU test requires shared libs (#4647)

* adds TESTONLY argument to TRIBITS_ADD_LIBRARY in the FMU test

* fix for shared library loading of testFMUBouncingBall.so (#4647)

* remove target_link_libraries from testFMU_BouncingBall CMakeLists

* Add's 5D, 6D, and 7D support for SSK, SLK, SNK, and SDK parameters

* Add support for higher order int/real arrays when reading/writing HDF5 parameter lists

* wip: adds boilerplate fmi and fmu c interface

* adds working example FMU model import

* adds wrappers for doStep getReal and setReal

* fix issue with getting c++ obj pointer

* wip: debugging illegal fmu state after init

* wip: debugging illegal fmu state after init

* adds example FMU import

* adds calls to enter and exitInitilizationMode to setupExperiment and adds output to fmu step example

* adds verbosity flag for c/c++ interface fns

* remove extraneous module imports in fmu2_f_api

* example fmu step test now runs to endTime of 0.1s

* wraps clear get/setInteger and get/setBoolean

* parse the FMU XML modelDescription on init

* enables xml parser to handle FMU born XML files and adds getValueRef and getCausality methods to FMU wrapper

* wip: get and set FMU state for FMU setRestart and rewindToRestart functionality

* adds convinience method to get and set FMU variables

* wip: adds capability to setRestart and rewindToRestart an...

* Adding overloaded assignments for geometries, and 90 degree rotation.

Description:
These functions are needed to help speed up the geometry processing in
MPACT.

VERA-dev Issue # - ####

* Removing overloaded assignment.

Description:
Should be unnecessary for types with only intrinsics and no pointers.

VERA-dev Issue # - ####

* Addressing Pull Request comments.

Description:
Adding subroutine headers and explanations.

VERA-dev Issue # - ####

Description:
For this newer version of PETSc, a parallel run using SuperLU_dist needs to
specifically call that solver (not SuperLU, which only works for serial).  Older
versions of PETSc allowed the solver to be set to SuperLU whether it was serial
or parallel.

CASL Ticket # - 4661

* Mass rename and interface improvements to Geom_* files

* WIP Add failing BWR test to unit test

* WIP first pass at refactoring combine_GraphType

Everything compiles

* WIP All futility unit tests working

* WIP add generic for Geom_Plane

* WIP fix graphs: add theta filtering for circle-circle intersections

* WIP add comments, fix small bugs, delete old combine_graphtype routine

* Fix bugs in adding circular arcs

* Remove unused variable and debug statements

* Change == to .APPROXEQA. in unit test

* Add RECURSIVE attribute to subroutines

* Add more enumerations

* Initial base class implementation

* Adds picard and modified picard types

* Adds unit test for Relaxed and Modified Picard

* Updates to comments and fixes test failures

* Switch Anderson to use base Accelerator type

* Fixing unit test

* Removes unneeded DBC statement

* Addresses review comments

* Adds more DBC checking, updates comments