Castro

Castro copied to clipboard

Can we write the advance algorithm the way non-subcycling AMR algorithms look?

3

A non-subcycling AMR version of our code (in the Strang build, not the simplified SDC build) would do the following (assumes two levels for simplicity, and ignores burning for now):...

maxpkatz

implement reference counting system for eos state

2

We should use a reference count-like system to keep track of the EOS state. Anytime an operation does something that we believe breaks thermodynamic consistency, we increment the counter. If...

zingale

Update mass fluxes at end of timestep with full timestep's worth

1

## PR summary This update makes the mass_fluxes array hold the full timestep's worth of mass fluxes at the end of finalize_advance. (This subtlety only matters if there were multiple...

maxpkatz

can we switch the fluxes MF to a fluxregister?

1

Right now, we create a MF `fluxes` where we add the fluxes during a subcycle, eventually copying them into the flux register. The hydro routines do not operate on the...

zingale

revisit the transverse update

3

Right now, we do the transverse correction to the conserved state + p. This way we don't need to use the EOS to get p on the interface when we...

zingale

## PR summary ## PR motivation ## PR checklist - [ ] test suite needs to be run on this PR - [ ] this PR will change answers in...

zingale

convert from Elixir to Async Arena

1

see this doc: https://amrex-codes.github.io/amrex/docs_html/GPU.html#async-arena

zingale

There is a Sphinx `. include:: ` directive to include the `yt_example.rst` page that is generated via `nbconvert` at build time. This works on a local machine, but it is...

zingale

Clang 12.x supports `-Wheader-guard` that ensure that the header guard matches the define immediate after. We caught one instance of a mismatch in Microphysics via ROCm. We should have the...

zingale

verify that MHD is tile safe

3

PR #2039 turned tiling back on in the MHD solver. We previously disabled it (#289), but that was back when the solver was in Fortran and not ported to the...

zingale