RELEASE_NOTES

9/20/17
Release version 17.09 (September, 2017)

11/3/17
Release version 17.11 (Novermber, 2017)
Small changes to documentation and code commments.
Added a test problem with an input tracer being forced into the domain with an inflow jet.

12/1/17 
Release version 17.12 (December, 2017)
Array -> Vector because Array has been deprecated in AMReX.
Use the new MLMG Cell-Centered solver for mac projection, mac sync solves, and
  scalar diffusion.
Minor interfaces changes and comment fixes.

1/3/18
Release version 18.01 (January, 2018)
Minor changes to quiet compiler warnings.
Added tiling to hoop stress.

2/1/18
Release version 18.02 (February, 2018)
Bug fixes to projection and diffusion boundary conditions
Interfaces to the new MLMG linear solvers in amrex/Src/LinearSolvers/MLMG
Cylindrical coordinate bux fixes

4/1/18
Release version 18.04 (April, 2018)
comments and variable rennaming only

6/1/18
Release version 18.06 (June, 2018)
Small changes to some problem setups
Some improvements to the particle algorithm
Fixed some ghost cell issues in the projection

7/2/18
Release version 18.07 (July, 2018)
Fixed a 3D corner coupling bug in FPU advection for conserved quantities.
Enabled FPU advection for scalars.
Added new test problems for regression suite.
Various minor fixes to other test problems.

8/1/18
Release version 18.08 (August, 2018)
Minor changes to regression test options.
Improved work distribution in create_umac_grown

9/4/18
Minor documentation updates.

11/1/18
Fix outflow bug.
Added shear layer test problem.

12/3/18
Some outflow boundary condition fixes
OpenMP and tiling additions

1/2/19
Some GPU work

4/8/19
Various problem fixes
Various problem tweaks involving regridding
Other minor fixes

5/1/19
Minor compiler settings fixes

6/3/19
Bug fix in filling u_mac ghost cells
Fix create_umac_grown for periodic cases
Misc code cleanup; extra arguments, extraenous calls, compiler quieting
Parallel reduction cleanup
Various changes to be consistent with AMReX interfaces

7/3/19
Bug fix SUMJET, these routines were compiled even if SUMJET was not defined
Reducing the dependency of PeleLM to IAMR
Putting back HIT test case
Getting rid of GENGETFORCE, MOREGENGETFORCE, etc.. now FORT_MAKEFORCE is always compiled 
     with the most general interface that was previously called by MOREGENGETFORCE
Input parameters are now read in two separate routines: Initialize() and Initialize_specific()
     in order to reduce dependency of PeleLM to IAMR
Bug fix NavierStokesBase::velocity_advection_update()
MLMG solver is now the only one implemented and used for Diffusion and MacProj

8/1/19
Cleaned up GNUMakeFiles
Minor modification to Taylor Green test problem setup
Fixed a tiling bug in diffusion_rearrangement

9/3/19
Minor bugfixes only, some to satisfy compiler warnings, some for thread safety,
some for ghost cell filling when needed

10/1/19
Updates to ppm for GPU.
Change in logic for setting dt when the velocity is zero

11/1/19
Switch many routines to dimension-agnostic (one common code no matter the dimensionality)

12/2/19
Minor interface cleanups and problem setups

4/1/20
**This is a major update since many branches were merged into development during
the month of March
Introducing complex geometries via EB.
No longer supporting constant mu or RZ options.
Update and add regression tests.
Many updates to be compatible with changes to amrex
Various bugfixes for host and GPU calculations  
Various changes to problem setups

5/1/20
Assortment of minor bug fixes
Tiling/GPU improvements
AMReX interface updates
Minor EB improvements

6/1/20
Various EB improvements
Change default dt scaling in sync diffusion solve for PeleLM usage
Various checks on number of ghost cells and divide by zeros.

7/1/20
Small tiling additions
Small bug fixes, and minor EB and transport improvements

8/3/20
Volume weighted sum fixes
Tiling fixes to estTimeStep
Minor GPU fixes, particularly for linear solvers
amrex::Math updates
Various small bugfixes

9/1/20
Various memory, GPU, and EB-related improvements
Minor code re-org
Updates to various test problems
Add aborts for ref_ratio=4 or rz (not currently supported)
Remove dependency on C_CellMG in amrex and older solvers
Various replacements of fortran kernels to C++

10/1/20
Various GPU updates
Advection updates and bux fixes
Other bugxies to miscellaneous routines

11/2/20
Various GPU updates for BCRec and create_umac_Grown
Minor tiling fixes

12/2/20
Creation from Embiggen of a tool to extrapolate coarse checkpoint file to a finest mesh
Add a "stop_interval" parameter to allow the simulation to stop at a specfic additional time interval after the physical time of a restart checkpoint file
Remove Dpdt_Type and code related to Pressure temporal-centeredness at a Point (vs. Interval)
Test case updates

1/4/21
Various bugfixes relating to CUDA, statistics, restarting.
Several changes to support time-averaging statistics and diagnostics.

2/1/21
Change spatial discretization of multilevel grad(p)
Checkpointing fixes
Many small updates to several parts of the code, including:
  -Diffusion:diffuse_Ssync
  -InflowForce
  -GPU fixes
  -Outflow BC solvers

4/1/21
Cleanup unused options, problems, and inputs files
Fixes to ref_ratio=4 ghost cell bugs
Port some initializtion kernels to C++
Port some derived functions to C++
Port boundary filling functions to C++
Ghost cell filling bugs
New options for controlling regridding
Minor GPU and memory fixes
Remove fortran probin files
Various RZ fixes

5/3/21
Updates to EB to be compatible with AMReX changes
Updates and bugfixes to EB advection algorithm
Fix out of bounds issues related to inadequate number of ghost cells
Misc GPU fixes
New homogeneous isotropic turbulence test case

6/2/21
Changes to some MAC projectino parameters
Fixed issues with some derived quantities
Checkpoint bug fix
Fixes to hydrodynamics for GPU and EB
Updates to MOL algorithm
Fix: properly merged in changes from development

7/2/21
Documentation updates
Embedded boundary improvements and fixes
Updates to create_umac_grown for AMR hydro
Misc. AMR fixes

8/2/21
Documentation
Compiler warnings
Misc. cleanup and re-org
Various EB and advection updates and fixes

9/1/21
Diffusion changes - use new multicomponent MLMG solver
In the temperature solve, let ns.temp_cond_coef represent (lamba/c_p)
Hydrodynamic algorithm updates
Updates to create_umac_grown
Embedded boundary compile updates
Add ParmParse for max_coarsening_level in nodal and mac projections
Moved slope routines to AMReX-Hydro

10/1/21
Documentation reorganization
Misc. AMR fixes
Misc. cleanup (re-org, quiet compiler warnings, remove unused code)
Misc. compiler fixes and general updates to test and bechmark problems

11/2/21
Various GPU fixes
Cleanup to the advection and diffusion classes
Fix usage of forcing term in scalar advection

12/10/21
Update Projection and MacProj to use nodal_proj and mac_proj, respectively, for runtime options.
Misc bug fixes for special cases.
Misc. cleanup -- quiet compiler warnings, remove unused and outdated variables and comments, make use of amrex functions where possible.
Fix Docs typos.
Updates to some regression tests for better code coverage.

1/5/22
Update IAMR to reflect AMReX-Hydro changes in State Redistribution -- the old StateRedist has been removed and NewStateRedist is renamed StateRedist.
Clean up Tutorials.

2/1/22
Fix in create_umac_grown, and minor fixes in problem initialization.
Allow rz with scalar diffusivity.
Lower default tolerances for the sync.
Minor clean up and updates to regression tests.

3/1/22
Add NSB::advection_scheme to choose the advection method used. This replaces use_godunov and godunov_use_ppm.
Bug fix in Diffusion::computeExtensiveFluxes for RZ.
Fix in velocity_advection for non-conservative MOL with divU!=0.
Fix indexing error in host bcrec for EBGodunov in mac_sync. Only affects problems where the coarse-fine boundary interacts with a physical EXT_DIR or HOEXTRAP and also the coarse level .
Misc clean up.

4/1/22
Bug fix in the MAC sync for diffusive scalars.
Add option of Bell-Dawson-Shubin (BDS) advection scheme, a higher order Godunov method for scalar conservation laws.
Bring back forced homogeneous isotropic turbulence problem set up.
Misc clean up.

5/1/22
Minor documentation updates.