Merge branch 'main' into parallel-hdf5

trixi-framework · Sep 6, 2023 · d6ab704 · d6ab704
2 parents 712cb79 + f098ea2
commit d6ab704
Show file tree

Hide file tree

Showing 14 changed files with 53 additions and 41 deletions.
diff --git a/.github/workflows/SpellCheck.yml b/.github/workflows/SpellCheck.yml
@@ -10,4 +10,4 @@ jobs:
       - name: Checkout Actions Repository
         uses: actions/checkout@v3
       - name: Check spelling
-        uses: crate-ci/[email protected].5
+        uses: crate-ci/[email protected].9
diff --git a/.zenodo.json b/.zenodo.json
@@ -15,7 +15,7 @@
             "orcid": "0000-0002-1752-1158"
         },
         {
-            "affiliation": "Applied Mathematics, University of Hamburg, Germany",
+            "affiliation": "Numerical Mathematics, Johannes Gutenberg University Mainz, Germany",
             "name": "Ranocha, Hendrik",
             "orcid": "0000-0002-3456-2277"
         },

diff --git a/AUTHORS.md b/AUTHORS.md
@@ -12,7 +12,7 @@ provided substantial additions or modifications. Together, these two groups form
 * [Gregor Gassner](https://www.mi.uni-koeln.de/NumSim/gregor-gassner),
   University of Cologne, Germany
 * [Hendrik Ranocha](https://ranocha.de),
-  University of Hamburg, Germany
+  Johannes Gutenberg University Mainz, Germany
 * [Andrew Winters](https://liu.se/en/employee/andwi94),
   Linköping University, Sweden
 * [Jesse Chan](https://jlchan.github.io),

diff --git a/Project.toml b/Project.toml
@@ -1,7 +1,7 @@
 name = "Trixi"
 uuid = "a7f1ee26-1774-49b1-8366-f1abc58fbfcb"
 authors = ["Michael Schlottke-Lakemper <[email protected]>", "Gregor Gassner <[email protected]>", "Hendrik Ranocha <[email protected]>", "Andrew R. Winters <[email protected]>", "Jesse Chan <[email protected]>"]
-version = "0.5.40-pre"
+version = "0.5.41-pre"
 
 [deps]
 CodeTracking = "da1fd8a2-8d9e-5ec2-8556-3022fb5608a2"

diff --git a/README.md b/README.md
@@ -247,7 +247,7 @@ Schlottke-Lakemper](https://lakemper.eu)
 (RWTH Aachen University/High-Performance Computing Center Stuttgart (HLRS), Germany) and
 [Gregor Gassner](https://www.mi.uni-koeln.de/NumSim/gregor-gassner)
 (University of Cologne, Germany). Together with [Hendrik Ranocha](https://ranocha.de)
-(University of Hamburg, Germany), [Andrew Winters](https://liu.se/en/employee/andwi94)
+(Johannes Gutenberg University Mainz, Germany), [Andrew Winters](https://liu.se/en/employee/andwi94)
 (Linköping University, Sweden), and [Jesse Chan](https://jlchan.github.io) (Rice University, US),
 they are the principal developers of Trixi.jl.
 The full list of contributors can be found in [AUTHORS.md](AUTHORS.md).

diff --git a/docs/literate/src/files/differentiable_programming.jl b/docs/literate/src/files/differentiable_programming.jl
@@ -128,7 +128,7 @@ condition_number = cond(V)
 # you can compute the gradient of an entropy-dissipative semidiscretization with respect to the
 # ideal gas constant of the compressible Euler equations as described in the following. This example
 # is also available as the elixir
-# [examples/special\_elixirs/elixir\_euler\_ad.jl](https://github.com/trixi-framework/Trixi.jl/blob/main/examples/special_elixirs/elixir_euler_ad.jl)
+# [`examples/special_elixirs/elixir_euler_ad.jl`](https://github.com/trixi-framework/Trixi.jl/blob/main/examples/special_elixirs/elixir_euler_ad.jl)
 
 # First, we create a semidiscretization of the compressible Euler equations.
 

diff --git a/docs/literate/src/files/index.jl b/docs/literate/src/files/index.jl
@@ -116,7 +116,7 @@
 
 # ## Examples in Trixi.jl
 # Trixi.jl already contains several more coding examples, the so-called `elixirs`. You can find them
-# in the folder [`examples`](https://github.com/trixi-framework/Trixi.jl/blob/main/examples/).
+# in the folder [`examples/`](https://github.com/trixi-framework/Trixi.jl/blob/main/examples/).
 # They are structured by the underlying mesh type and the respective number of spatial dimensions.
 # The name of an elixir is composed of the underlying system of conservation equations (for instance
 # `advection` or `euler`) and other special characteristics like the initial condition

diff --git a/docs/src/callbacks.md b/docs/src/callbacks.md
@@ -15,7 +15,7 @@ control, adaptive mesh refinement, I/O, and more.
 
 ### CFL-based time step control
 Time step control can be performed with a [`StepsizeCallback`](@ref). An example making use
-of this can be found at [examples/tree_2d_dgsem/elixir\_advection\_basic.jl](https://github.com/trixi-framework/Trixi.jl/blob/main/examples/tree_2d_dgsem/elixir_advection_basic.jl)
+of this can be found at [`examples/tree_2d_dgsem/elixir_advection_basic.jl`](https://github.com/trixi-framework/Trixi.jl/blob/main/examples/tree_2d_dgsem/elixir_advection_basic.jl)
 
 ### Adaptive mesh refinement
 Trixi.jl uses a hierarchical Cartesian mesh which can be locally refined in a solution-adaptive way.
@@ -24,12 +24,12 @@ passing an [`AMRCallback`](@ref) to the ODE solver. The `AMRCallback` requires a
 [`ControllerThreeLevel`](@ref) or [`ControllerThreeLevelCombined`](@ref) to tell the AMR
 algorithm which cells to refine/coarsen.
 
-An example elixir using AMR can be found at [examples/tree_2d_dgsem/elixir\_advection\_amr.jl](https://github.com/trixi-framework/Trixi.jl/blob/main/examples/tree_2d_dgsem/elixir_advection_amr.jl).
+An example elixir using AMR can be found at [`examples/tree_2d_dgsem/elixir_advection_amr.jl`](https://github.com/trixi-framework/Trixi.jl/blob/main/examples/tree_2d_dgsem/elixir_advection_amr.jl).
 
 ### Analyzing the numerical solution
 The [`AnalysisCallback`](@ref) can be used to analyze the numerical solution, e.g. calculate
 errors or user-specified integrals, and print the results to the screen. The results can also be
-saved in a file. An example can be found at [examples/tree_2d_dgsem/elixir\_euler\_vortex.jl](https://github.com/trixi-framework/Trixi.jl/blob/main/examples/tree_2d_dgsem/elixir_euler_vortex.jl).
+saved in a file. An example can be found at [`examples/tree_2d_dgsem/elixir_euler_vortex.jl`](https://github.com/trixi-framework/Trixi.jl/blob/main/examples/tree_2d_dgsem/elixir_euler_vortex.jl).
 In [Performance metrics of the `AnalysisCallback`](@ref) you can find a detailed
 description of the different performance metrics the `AnalysisCallback` computes.
 
@@ -38,15 +38,15 @@ description of the different performance metrics the `AnalysisCallback` computes
 #### Solution and restart files
 To save the solution in regular intervals you can use a [`SaveSolutionCallback`](@ref). It is also
 possible to create restart files using the [`SaveRestartCallback`](@ref). An example making use
-of these can be found at [examples/tree_2d_dgsem/elixir\_advection\_extended.jl](https://github.com/trixi-framework/Trixi.jl/blob/main/examples/tree_2d_dgsem/elixir_advection_extended.jl).
+of these can be found at [`examples/tree_2d_dgsem/elixir_advection_extended.jl`](https://github.com/trixi-framework/Trixi.jl/blob/main/examples/tree_2d_dgsem/elixir_advection_extended.jl).
 An example showing how to restart a simulation from a restart file can be found at
-[examples/tree_2d_dgsem/elixir\_advection\_restart.jl](https://github.com/trixi-framework/Trixi.jl/blob/main/examples/tree_2d_dgsem/elixir_advection_restart.jl).
+[`examples/tree_2d_dgsem/elixir_advection_restart.jl`](https://github.com/trixi-framework/Trixi.jl/blob/main/examples/tree_2d_dgsem/elixir_advection_restart.jl).
 
 #### Time series
 Sometimes it is useful to record the evaluations of state variables over time at
 a given set of points. This can be achieved by the [`TimeSeriesCallback`](@ref), which is used,
 e.g., in
-[examples/tree_2d_dgsem/elixir\_acoustics\_gaussian\_source.jl](https://github.com/trixi-framework/Trixi.jl/blob/main/examples/tree_2d_dgsem/elixir_acoustics_gaussian_source.jl).
+[`examples/tree_2d_dgsem/elixir_acoustics_gaussian_source.jl`](https://github.com/trixi-framework/Trixi.jl/blob/main/examples/tree_2d_dgsem/elixir_acoustics_gaussian_source.jl).
 The `TimeSeriesCallback` constructor expects a semidiscretization and a list of points at
 which the solution should be recorded in regular time step intervals. After the
 last time step, the entire record is stored in an HDF5 file.
@@ -113,12 +113,12 @@ will yield the following plot:
 Some callbacks provided by Trixi.jl implement specific features for certain equations:
 * The [`LBMCollisionCallback`](@ref) implements the Lattice-Boltzmann method (LBM) collision
   operator and should only be used when solving the Lattice-Boltzmann equations. See e.g.
-  [examples/tree_2d_dgsem/elixir\_lbm\_constant.jl](https://github.com/trixi-framework/Trixi.jl/blob/main/examples/tree_2d_dgsem/elixir_lbm_constant.jl)
+  [`examples/tree_2d_dgsem/elixir_lbm_constant.jl`](https://github.com/trixi-framework/Trixi.jl/blob/main/examples/tree_2d_dgsem/elixir_lbm_constant.jl)
 * The [`SteadyStateCallback`](@ref) terminates the time integration when the residual steady state
   falls below a certain threshold. This checks the convergence of the potential ``\phi`` for
-  hyperbolic diffusion. See e.g. [examples/tree_2d_dgsem/elixir\_hypdiff\_nonperiodic.jl](https://github.com/trixi-framework/Trixi.jl/blob/main/examples/tree_2d_dgsem/elixir_hypdiff_nonperiodic.jl).
+  hyperbolic diffusion. See e.g. [`examples/tree_2d_dgsem/elixir_hypdiff_nonperiodic.jl`](https://github.com/trixi-framework/Trixi.jl/blob/main/examples/tree_2d_dgsem/elixir_hypdiff_nonperiodic.jl).
 * The [`GlmSpeedCallback`](@ref) updates the divergence cleaning wave speed `c_h` for the ideal
-  GLM-MHD equations. See e.g. [examples/tree_2d_dgsem/elixir\_mhd\_alfven\_wave.jl](https://github.com/trixi-framework/Trixi.jl/blob/main/examples/tree_2d_dgsem/elixir_mhd_alfven_wave.jl).
+  GLM-MHD equations. See e.g. [`examples/tree_2d_dgsem/elixir_mhd_alfven_wave.jl`](https://github.com/trixi-framework/Trixi.jl/blob/main/examples/tree_2d_dgsem/elixir_mhd_alfven_wave.jl).
 
 ## Usage of step callbacks
 Step callbacks are passed to the `solve` method from the ODE solver via the keyword argument
@@ -152,7 +152,7 @@ more callbacks, you need to turn them into a `CallbackSet` first by calling
 ## Stage callbacks
 [`PositivityPreservingLimiterZhangShu`](@ref) is a positivity-preserving limiter, used to enforce
 physical constraints. An example elixir using this feature can be found at
-[examples/tree_2d_dgsem/elixir\_euler\_positivity.jl](https://github.com/trixi-framework/Trixi.jl/blob/main/examples/tree_2d_dgsem/elixir_euler_positivity.jl).
+[`examples/tree_2d_dgsem/elixir_euler_positivity.jl`](https://github.com/trixi-framework/Trixi.jl/blob/main/examples/tree_2d_dgsem/elixir_euler_positivity.jl).
 
 ## Implementing new callbacks
 Since Trixi.jl is compatible with [OrdinaryDiffEq.jl](https://github.com/SciML/OrdinaryDiffEq.jl),
@@ -162,4 +162,4 @@ Step callbacks are just called [callbacks](https://diffeq.sciml.ai/latest/featur
 Stage callbacks are called [`stage_limiter!`](https://diffeq.sciml.ai/latest/solvers/ode_solve/#Explicit-Strong-Stability-Preserving-Runge-Kutta-Methods-for-Hyperbolic-PDEs-(Conservation-Laws)).
 
 An example elixir showing how to implement a new simple stage callback and a new simple step
-callback can be found at [examples/tree_2d_dgsem/elixir\_advection\_callbacks.jl](https://github.com/trixi-framework/Trixi.jl/blob/main/examples/tree_2d_dgsem/elixir_advection_callbacks.jl).
+callback can be found at [`examples/tree_2d_dgsem/elixir_advection_callbacks.jl`](https://github.com/trixi-framework/Trixi.jl/blob/main/examples/tree_2d_dgsem/elixir_advection_callbacks.jl).
diff --git a/docs/src/index.md b/docs/src/index.md
@@ -324,7 +324,7 @@ Schlottke-Lakemper](https://lakemper.eu)
 (RWTH Aachen University/High-Performance Computing Center Stuttgart (HLRS), Germany) and
 [Gregor Gassner](https://www.mi.uni-koeln.de/NumSim/gregor-gassner)
 (University of Cologne, Germany). Together with [Hendrik Ranocha](https://ranocha.de)
-(University of Hamburg, Germany) and [Andrew Winters](https://liu.se/en/employee/andwi94)
+(Johannes Gutenberg University Mainz, Germany) and [Andrew Winters](https://liu.se/en/employee/andwi94)
 (Linköping University, Sweden), and [Jesse Chan](https://jlchan.github.io) (Rice University, US),
 they are the principal developers of Trixi.jl.
 The full list of contributors can be found under [Authors](@ref).

diff --git a/docs/src/meshes/dgmulti_mesh.md b/docs/src/meshes/dgmulti_mesh.md
@@ -81,16 +81,20 @@ type, but will be more efficient at high orders of approximation.
 ## Trixi.jl elixirs on simplicial and tensor product element meshes
 
 Example elixirs with triangular, quadrilateral, and tetrahedral meshes can be found in
-the `examples/dgmulti_2d` and `examples/dgmulti_3d` folders. Some key elixirs to look at:
+the [`examples/dgmulti_2d/`](https://github.com/trixi-framework/Trixi.jl/blob/main/examples/dgmulti_2d/)
+and [`examples/dgmulti_3d/`](https://github.com/trixi-framework/Trixi.jl/blob/main/examples/dgmulti_3d/)
+folders. Some key elixirs to look at:
 
-* `examples/dgmulti_2d/elixir_euler_weakform.jl`: basic weak form DG discretization on a uniform triangular mesh.
+* [`examples/dgmulti_2d/elixir_euler_weakform.jl`](https://github.com/trixi-framework/Trixi.jl/blob/main/examples/dgmulti_2d/elixir_euler_weakform.jl):
+  basic weak form DG discretization on a uniform triangular mesh.
   Changing `element_type = Quad()` or `approximation_type = SBP()` will switch to a quadrilateral mesh
   or an SBP-type discretization. Changing `surface_integral = SurfaceIntegralWeakForm(flux_ec)` and
   `volume_integral = VolumeIntegralFluxDifferencing(flux_ec)` for some entropy conservative flux
   (e.g., [`flux_chandrashekar`](@ref) or [`flux_ranocha`](@ref)) will switch to an entropy conservative formulation.
-* `examples/dgmulti_2d/elixir_euler_triangulate_pkg_mesh.jl`: uses an unstructured mesh generated by
-  [Triangulate.jl](https://github.com/JuliaGeometry/Triangulate.jl).
-* `examples/dgmulti_3d/elixir_euler_weakform.jl`: basic weak form DG discretization on a uniform tetrahedral mesh.
+* [`examples/dgmulti_2d/elixir_euler_triangulate_pkg_mesh.jl`](https://github.com/trixi-framework/Trixi.jl/blob/main/examples/dgmulti_2d/elixir_euler_triangulate_pkg_mesh.jl):
+  uses an unstructured mesh generated by [Triangulate.jl](https://github.com/JuliaGeometry/Triangulate.jl).
+* [`examples/dgmulti_3d/elixir_euler_weakform.jl`](https://github.com/trixi-framework/Trixi.jl/blob/main/examples/dgmulti_3d/elixir_euler_weakform.jl):
+  ´basic weak form DG discretization on a uniform tetrahedral mesh.
   Changing `element_type = Hex()` will switch to a hexahedral mesh. Changing
   `surface_integral = SurfaceIntegralWeakForm(flux_ec)` and
   `volume_integral = VolumeIntegralFluxDifferencing(flux_ec)` for some entropy conservative flux

diff --git a/docs/src/overview.md b/docs/src/overview.md
@@ -5,7 +5,7 @@ conservation laws. Thus, it is not a monolithic PDE solver that is configured at
 via parameter files, as it is often found in classical numerical simulation codes.
 Instead, each simulation is configured by pure Julia code. Many examples of such
 simulation setups, called *elixirs* in Trixi.jl, are provided in the
-[examples](https://github.com/trixi-framework/Trixi.jl/blob/main/examples)
+[`examples/`](https://github.com/trixi-framework/Trixi.jl/blob/main/examples)
 folder.
 
 Trixi.jl uses the method of lines, i.e., the full space-time discretization is separated into two steps;
@@ -77,7 +77,7 @@ Further information can be found in the
 ## Next steps
 
 We explicitly encourage people interested in Trixi.jl to have a look at the
-[examples](https://github.com/trixi-framework/Trixi.jl/blob/main/examples)
+[`examples/`](https://github.com/trixi-framework/Trixi.jl/blob/main/examples)
 bundled with Trixi.jl to get an impression of what is possible and the general
 look and feel of Trixi.jl.
 Before doing that, it is usually good to get an idea of

diff --git a/docs/src/parallelization.md b/docs/src/parallelization.md
@@ -53,16 +53,24 @@ a system-provided MPI installation with Trixi.jl can be found in the following s
 
 ### [Using a system-provided MPI installation](@id parallel_system_MPI)
 
-When using Trixi.jl with a system-provided MPI backend the underlying [`p4est`](https://github.com/cburstedde/p4est)
-library needs to be compiled with the same MPI installation. Therefore, you also need to use
-a system-provided `p4est` installation (for notes on how to install `p4est` see e.g.
-[here](https://github.com/cburstedde/p4est/blob/master/README), use the configure option
-`--enable-mpi`). In addition, [P4est.jl](https://github.com/trixi-framework/P4est.jl) needs to
-be configured to use the custom `p4est` installation. Follow the steps described
-[here](https://github.com/trixi-framework/P4est.jl/blob/main/README.md) for the configuration.
+When using Trixi.jl with a system-provided MPI backend the underlying
+[`p4est`](https://github.com/cburstedde/p4est) and [`t8code`](https://github.com/DLR-AMR/t8code)
+libraries need to be compiled with the same MPI installation. Therefore, you also need to
+use system-provided `p4est` and `t8code` installations (for notes on how to install `p4est`
+and `t8code` see e.g. [here](https://github.com/cburstedde/p4est/blob/master/README) and
+[here](https://github.com/DLR-AMR/t8code/wiki/Installation), use the configure option
+`--enable-mpi`). Note that `t8code` already comes with a `p4est` installation, so it suffices
+to install `t8code`. In addition, [P4est.jl](https://github.com/trixi-framework/P4est.jl) and
+[T8code.jl](https://github.com/DLR-AMR/T8code.jl) need to be configured to use the custom
+installations. Follow the steps described
+[here](https://github.com/DLR-AMR/T8code.jl/blob/main/README.md#installation) and
+[here](https://github.com/trixi-framework/P4est.jl/blob/main/README.md#installation) for the
+configuration. The paths that point to `libp4est.so` (and potentially to `libsc.so`) need to be
+the same for P4est.jl and T8code.jl. This could e.g. be `libp4est.so` that usually can be found
+in `lib/` or `local/lib/` in the installation directory of `t8code`.
 In total, in your active Julia project you should have a LocalPreferences.toml file with sections
-`[MPIPreferences]` and `[P4est]` as well as an entry `MPIPreferences` in your Project.toml to
-use a custom MPI installation.
+`[MPIPreferences]`, `[T8code]` and `[P4est]` as well as an entry `MPIPreferences` in your
+Project.toml to use a custom MPI installation.
 
 
 ### [Usage](@id parallel_usage)

diff --git a/docs/src/restart.md b/docs/src/restart.md
@@ -18,15 +18,15 @@ save_restart = SaveRestartCallback(interval=100,
 Make this part of your `CallbackSet`.
 
 An example is
-[```examples/examples/structured_2d_dgsem/elixir_advection_extended.jl```](https://github.com/trixi-framework/Trixi.jl/blob/main/examples/structured_2d_dgsem/elixir_advection_extended.jl).
+[`examples/examples/structured_2d_dgsem/elixir_advection_extended.jl`](https://github.com/trixi-framework/Trixi.jl/blob/main/examples/structured_2d_dgsem/elixir_advection_extended.jl).
 
 
 ## [Perform the simulation restart](@id restart_perform)
 Since all of the information about the simulation can be obtained from the
 last snapshot, the restart can be done with relatively few lines
 in an extra elixir file.
 However, some might prefer to keep everything in one elixir and
-conditionals like ```if restart``` with a boolean variable ```restart``` that is user defined.
+conditionals like `if restart` with a boolean variable `restart` that is user defined.
 
 First we need to define from which file we want to restart, e.g.
 ```julia
@@ -50,7 +50,7 @@ time the one form the snapshot:
 tspan = (load_time(restart_filename), 2.0)
 ```
 
-We now also take the last ```dt```, so that our solver does not need to first find
+We now also take the last `dt`, so that our solver does not need to first find
 one to fulfill the CFL condition:
 ```julia
 dt = load_dt(restart_filename)
@@ -63,7 +63,7 @@ ode = semidiscretize(semi, tspan, restart_filename)
 
 You should now define a [`SaveSolutionCallback`](@ref) similar to the
 [original simulation](https://github.com/trixi-framework/Trixi.jl/blob/main/examples/structured_2d_dgsem/elixir_advection_extended.jl),
-but with ```save_initial_solution=false```, otherwise our initial snapshot will be overwritten.
+but with `save_initial_solution=false`, otherwise our initial snapshot will be overwritten.
 If you are using one file for the original simulation and the restart
 you can reuse your [`SaveSolutionCallback`](@ref), but need to set
 ```julia
@@ -86,4 +86,4 @@ Now we can compute the solution:
 sol = solve!(integrator)
 ```
 
-An example is in `[``examples/structured_2d_dgsem/elixir_advection_restart.jl```](https://github.com/trixi-framework/Trixi.jl/blob/main/examples/structured_2d_dgsem/elixir_advection_restart.jl).
+An example is in [`examples/structured_2d_dgsem/elixir_advection_restart.jl`](https://github.com/trixi-framework/Trixi.jl/blob/main/examples/structured_2d_dgsem/elixir_advection_restart.jl).