Conversation
Mainly this PR adds a laplacian for face-centered B to `CleanupDivergence`, which is automatically used if the B_CT package is loaded. It doesn't actually seem faster than the cell-centered version despite having a smaller stencil. It also adds a "PtoU" for face-centered fields, which takes a cell-centered B and interpolates it to faces. Yes, this means interpolating interpolated values, which is not ideal but doesn't seem egregious. Finally, it adds a version of the `resize` test with face-centered fields. I also rephrased the face divB in terms of derivatives for clarity, spit out Flux-CT device functions into a header like Face-CT has, and modified some uncommon device-side B field functions a bit. Finally, this adds an unrelated feature allowing re-initialization of electron temps at user option, designed for restarting ideal simulations w/electrons enabled.
Face-centered B cleanup
…. Conserves total energy at least?
Fix face-centered B cleaning with outflow boundary conditions
Polar averaging for angular momentum
Fix restarting ideal runs w/EMHD
Pull in restart+new input file
CI fixes and version bumps
These must have been reintroduced in a merge; Linking was always done with the versions in kharma/prob/elec, so these were just dead code.
Update AddGeoSource to the form listed in docs
…to fix/update-parameters
implicit.hpp was importing the entire EMHD namespace, avoid that. Other cleanup.
Respect frame-switch parameters for mixed fluid frames. Rename fluid/drift to normal/drift to reflect our usage (note original inspiration *is* fluid/drift)
Fix compiling floors under Nvidia, maybe a touch slower Fix MHDModes test: upwinded CT converges @2.15-2.3, allow this
WIP: Split the EMHD kernel
gs05_c CT scheme no longer needs velocities preserved. UCT1/2 will, but I'm not sure we'll implement those, revive this code if we do. Other removals/reshuffles as I looked up parameters
All of KHARMA's parameter files were drifting out of date. This: 1. Removes all ghost_zones = true, which is always unnecessary unless debugging something nasty 2. Moves `reconstruction` param into `<flux>`, generally also adds `type=llf` for clarity 3. Moves all boundary declarations to `<boundaries>`, where they are even needed anymore. 4. When emulating iharm3d, set `flux_ct` and `onedw` explicitly, and name the file to reflect iharm3d "mode." 5. Thoroughly modern mad.par and sane.par reflecting best practices 6. Remove refinement pars from unrefined meshes
* Remove some tests from the GPU version of all_pars which are expected to fail -- too much memory or non-GPU dependent. * restart & re-init tests are broken with new mad.par/sane.par. They always work on dedicated machines but fail in CI for some reason. Maybe it's MPI? Disable that to see.
* Fix 2D restart test for new mad.par/sane.par, add EMHD * Remove EMHD tests from NVHPC CI, EMHD is buggy on specifically the NVHPC stack, so we will avoid it until this is fixed. * Accommodate CUDA+GCC on Darwin, see above * Move ApplyEMHDLimits, conventionally it should be in .cpp file
* 2D restarts needed another option * actually remove conducting_atmo from NVHPC tests * Point people to the wiki now that it's """good"""
Parameter updates and code/comment cleanup
Contributor
Author
|
Finally a release with net negative line count! The tests passed if you click through, GitHub limits the number of status updates per commit and our mirror with GitLab hits the limit if the tests take too long. |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
More transmissive boundaries, more stringent tests, more documentation, more cleaner, more faster, more better.
Highlights:
boundaries/excise_polar_flux=truedevelopand new B Cleanup w/GMG #109)./run.sh -r restart_file.rhdf -i new_input_parameters.par