MFC v4.4.2

What's Changed

• Add weno_avg back by @haochey in #172
• Syscheck: Run a system check before running MFC & More by @henryleberre in #198
• Fix doxygen for m_bubbles.fpp by @sbryngelson in #202

Full Changelog: v4.4.1...v4.4.2

MFC v4.4.1

What's Changed

• Minor bug fix with polytropic qbmm and cleaning up p_main by @anandrdbz in #183
• Model Patches: STLs & (some) OBJs by @henryleberre in #181
• Added example case for Lax shock tube problem (1D) by @RasmitDevkota in #190
• Workaround for a CMake regression (CMAKE_Fortran_PREPROCESS) by @henryleberre in #193
• Generated visualizations for 1D_laxshocktube by @RasmitDevkota in #192
• Address #188 by @wilfonba in #197
• Patches Patch by @wilfonba in #195
• Hardcoded ICs Update by @wilfonba in #201
• Intel Compiler Bug by @anandrdbz in #161

New Contributors

• @RasmitDevkota made their first contribution in #190

Full Changelog: v4.4.0...v4.4.1

MFC v4.4.0

What's Changed

• Fix #143, --wait, and mpi-autodetecting template files by @henryleberre in #146
• Update case.py by @anandrdbz in #145
• New CI using phoenix and slurm by @sbryngelson in #147
• HLL + CBC fix by @anandrdbz in #148
• Test Cases: 2D Taylor Green Vortex by @haochey in #150
• Fixed a bug on pi_inf and enabled subgrid bubble model for viscous flow by @lee-hyeoksu in #151
• Remove intel compilers for now by @sbryngelson in #159
• Documentation for qbmm by @anandrdbz in #158
• hypoelastic examples cases fix by @wilfonba in #152
• Fix the computation of vcfl by @haochey in #155
• Update index.html by @sbryngelson in #164
• [#162] Continue CI when runs fails (matrix & jobs) by @henryleberre in #165
• Add Featured Simulations to the MFC website by @henryleberre in #166
• Update to CI, embers queue test by @sbryngelson in #170
• Upstream fixes from henryleberre/[amdgpu, gcc-acc] and other things by @henryleberre in #167
• Support analytically defining patch variables in all case files by @henryleberre in #176
• Update run-phoenix-release-gpu.sh by @sbryngelson in #178
• QBMM + Preston by @anandrdbz in #180

Full Changelog: v4.3.6...v4.4.0

MFC v4.3.6

What's Changed

• Add support for counting lines of code via ./mfc.sh count, remove ./mfc.sh cloc by @sbryngelson in #137
• Example case for 3D turbulent mixing layer by @lee-hyeoksu in #133
• Single precision fix by @wilfonba in #138
• Fix a bug in s_superposition_instability_wave by @lee-hyeoksu in #140
• Add Nsight Docs by @anshgupta1234 in #141
• Add intel compilers by @anshgupta1234 in #142

Full Changelog: v4.3.5...v4.3.6

MFC v4.3.5

What's Changed

• Helper Functions by @wilfonba in #91
• Manual inlining of s_quad for performance by @wilfonba in #118
• Restarting Cases by @anshgupta1234 in #120
• Links Changed & Deleting defaults.yaml reference by @anshgupta1234 in #122
• MPI_Abort fix by @wilfonba in #125
• remove unused vars by @sbryngelson in #126
• Move startup checks to their own file by @sbryngelson in #127
• Removing more parent scope overwrites and unused variables by @sbryngelson in #128
• Update CMakeLists.txt for --debug fix on nvidia GPUs by @sbryngelson in #130
• No more global overwrites by @sbryngelson in #131
• Added --debug to ci.yml by @anshgupta1234 in #129
• Update running.md by @anshgupta1234 in #135
• Remove Post-Process Tests from --debug by @anshgupta1234 in #136

Full Changelog: v4.3.4...v4.3.5

MFC v4.3.4

What's Changed

• Viscous fix by @wilfonba in #101
• fix Bridges2 load by @sbryngelson in #106
• Post Process Tests by @anshgupta1234 in #107
• Conversions Update by @anshgupta1234 in #86
• Check if num_patches mismatches the specified patches by @anshgupta1234 in #110
• Fix Getting Started Link and Update testing.md by @anshgupta1234 in #111
• Fix h5py bug by using h5dump by @sbryngelson in #112
• Update Getting Started by @anshgupta1234 in #116
• Weak Scaling, Parameter Forwarding, Profiling, ... by @henryleberre in #117

Full Changelog: v4.3.3...v4.3.4

MFC v.4.3.3

What's Changed

• add 1D and quasi-1D hypoelastic examples by @js-spratt in #77
• Fix idx by @anshgupta1234 in #78
• Add Q_M-criterion to post_process by @lee-hyeoksu in #79
• Add iostat check when reading .inp files by @belericant in #80
• Fix #81 and #82 by @henryleberre in #84
• Remove -f from rm in s_delete_directory by @belericant in #85
• Fix #92 by @henryleberre in #93
• mfc.sh load updated for Phoenix SLURM by @anshgupta1234 in #96
• Update case.py by @anandrdbz in #99
• Variables and Patch Documentation by @wilfonba in #100

New Contributors

• @lee-hyeoksu made their first contribution in #79
• @belericant made their first contribution in #80

Full Changelog: v4.3.2...v4.3.3

MFC v4.3.2

What's Changed

• Use src//include/.fpp for Fypp includes by @henryleberre in #67
• Fixed #48 by @anshgupta1234 in #68
• Fixed error with calling conservative to primitive vars in m_data_output by @anshgupta1234 in #73
• Move easily portable and repeated MPI calls to common/ by @sbryngelson in #75
• Fix post-process bug & Small fixes by @henryleberre in #74

Full Changelog: v4.3.1...v4.4.0

MFC v4.3.1

What's Changed

• remove surface tension by @sbryngelson in #58
• modularize patch check in pre_process by @sbryngelson in #59
• new initial condition modules to modularize by @sbryngelson in #60
• CBC issue fixed by @anandrdbz in #61
• added inline riemann avg state by @sbryngelson in #62
• Small fixes by @henryleberre in #64
• Fix #65 by @anshgupta1234 in #66

Full Changelog: v4.3.0...v4.3.1

MFC v4.3.0

What's Changed

• Cray Fortran + Docs + Toolchain Refactor by @henryleberre in #52
• Update m_mpi_proxy.fpp by @sbryngelson in #55

Includes:

Cray / CCE

Introduce support for the compilation of pre_process and simulation with the Cray Fortran compiler (ftn) for CPUs with limited GPU support. The GPU version of simulation doesn't work yet because of a compiler bug. I am waiting on the vendor to fix this issue. Most GPU-related changes/fixes are included in this Pull Request. These include:

• Ensuring HDF5 and FFTW3 are not configured/built by the CMake superbuild file for dependencies when Cray is the target compiler. We create dummy CMake targets in this case.
• Adding the correct flags for compilation with Cray. Since ACC offloading is enabled by default, we supply -h noacc -x acc when targeting CPUs and/or building a target other than simulation.
• Creating a macros.fpp file to seamlessly handle the allocation of GPU arrays with a single call.
  Adding an entry to mfc.sh load for Crusher.
  mfc.sh
• We now keep track and install Python dependencies into the virtualenv in a far simpler and more efficient manner.
  Added mfc.sh cloc.
• Toolchain/

Refactor (for a final time) the toolchain code

• modes have been replaced with simpler and more versatile switches: --[no-]mpi, --[no-]gpu, and --[no-]debug. These are options that can be turned on and off independently.
• We properly handle cases where CMake configuration has failed or was incomplete. We do so by checking for the existence of a CMakeCache.txt file.
• Removed the user .yml configuration file.
• Parsing of template files has been improved and simplified. Expressions inside {} are now directly passed to eval() with special access to the dictionary of parsed command-line arguments.
• References to member classes within the global MFCState are no longer. Modules are now publicly available and export their functionality to others.

Website / Documentation

• MFC's website and documentation have been updated for a smoother user onboarding experience. Note: We now have to build and install the most recent release of Doxygen for the documentation to render properly.
• I also added bibtex syntax highlighting to the MFC paper citation.

CMake

• case.fpp files are gone from all targets except simulation.
• CMakeLists.txt handles src/common code more carefully:
• src/common code is now preprocessed into src//autogen where is the name of the current target being compiled. This fixes bugs that resulted from different targets updating the same (common) src/common/autogen/*.f90 source files and causing undesirable recompilations.
  src/common is now part of the Fypp include path.

Windows

I updated the mfc.bat file to work better!

Full Changelog: v4.2.0...v4.3.0