Castro

Castro copied to clipboard

switch sum_utils to use AMReX volwgtsum

1

AMReX now has a `volwgtsum` function: https://github.com/AMReX-Codes/amrex/pull/2961 we should be able to use this in place of the Castro version in `sum_utils.cpp`

zingale

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

zingale

output the location where the timestep is set

6

## PR summary addresses #328 ## PR motivation ## PR checklist - [ ] test suite needs to be run on this PR - [ ] this PR will change...

zingale

this doesn't yet work for parallel (MPI) ## PR summary ## PR motivation ## PR checklist - [ ] test suite needs to be run on this PR - [...

zingale

amrex::Error::0::Error: initial data has rho <~ small_dens !!!

3

Following the steps to test the code ,I met some faluts. ./Castro3d.gnu.MPI.ex inputs-sod-x ... amrex::Error::0::Error: initial data has rho

zams2cc

[WIP] Compute the shock flag once -- at the start of the timestep

1

## PR summary This changes the way the shock flag is computed for disabling burning in shocks. Now we compute it at the top of the timestep, before the first...

zingale

it should not matter which comes first, since reactions don't change density, but this swap will allow us to precompute the shock flag with sources before the first burn. ##...

zingale

gpu running slower than cpu for subchandra problem

3

My subchandra simulations frequently run much faster when using cpu’s compared to gpu’s, when the opposite should be true. The plot below was made to compare the cpu vs. gpu...

brady-ryan

split c-linter action in 2

2

Right now the c-linter action does clang-tidy and claims to do cppcheck. We should split it in two. For the clang-tidy, we should use the build process that is in...

zingale

see https://github.com/AMReX-Codes/amrex/pull/2954

zingale