Add comments explaining why weak form is only for conserved terms (#1676

) * Add comments explaining why weak form is only for conserved terms * Update src/solvers/dg.jl Co-authored-by: Andrew Winters <[email protected]> * Update src/solvers/dgsem_structured/dg_2d.jl Co-authored-by: Andrew Winters <[email protected]> * Update src/solvers/dg.jl Co-authored-by: Andrew Winters <[email protected]> * Update src/solvers/dgsem_structured/dg_3d.jl Co-authored-by: Andrew Winters <[email protected]> * Update src/solvers/dgsem_structured/dg_2d.jl Co-authored-by: Andrew Winters <[email protected]> * Update src/solvers/dgsem_tree/dg_2d.jl Co-authored-by: Andrew Winters <[email protected]> * Update src/solvers/dgsem_tree/dg_1d.jl Co-authored-by: Andrew Winters <[email protected]> * Update src/solvers/dgsem_tree/dg_3d.jl Co-authored-by: Andrew Winters <[email protected]> * Update src/solvers/dgsem_tree/dg_2d.jl Co-authored-by: Andrew Winters <[email protected]> * Update src/solvers/dgsem_tree/dg_3d.jl Co-authored-by: Andrew Winters <[email protected]> * Update src/solvers/dgsem_structured/dg_3d.jl Co-authored-by: Andrew Winters <[email protected]> * Update src/solvers/dgsem_tree/dg_1d.jl Co-authored-by: Andrew Winters <[email protected]> * Update src/solvers/dg.jl Co-authored-by: Hendrik Ranocha <[email protected]> --------- Co-authored-by: Andrew Winters <[email protected]> Co-authored-by: Hendrik Ranocha <[email protected]>
trixi-framework · Oct 21, 2023 · fcab6a9 · fcab6a9
1 parent fc00ac1
commit fcab6a9
Show file tree

Hide file tree

Showing 6 changed files with 40 additions and 0 deletions.
diff --git a/src/solvers/dg.jl b/src/solvers/dg.jl
@@ -40,6 +40,11 @@ standard textbooks.
   Nodal Discontinuous Galerkin Methods: Algorithms, Analysis, and
   Applications
   [doi: 10.1007/978-0-387-72067-8](https://doi.org/10.1007/978-0-387-72067-8)
+
+`VolumeIntegralWeakForm()` is only implemented for conserved terms as 
+non-conservative terms should always be discretized in conjunction with a flux-splitting scheme, 
+see [`VolumeIntegralFluxDifferencing`](@ref).
+This treatment is required to achieve, e.g., entropy-stability or well-balancedness.
 """
 struct VolumeIntegralWeakForm <: AbstractVolumeIntegral end
 

diff --git a/src/solvers/dgsem_structured/dg_2d.jl b/src/solvers/dgsem_structured/dg_2d.jl
@@ -49,6 +49,13 @@ function rhs!(du, u, t,
     return nothing
 end
 
+#=
+`weak_form_kernel!` is only implemented for conserved terms as 
+non-conservative terms should always be discretized in conjunction with a flux-splitting scheme, 
+see `flux_differencing_kernel!`.
+This treatment is required to achieve, e.g., entropy-stability or well-balancedness.
+See also https://github.com/trixi-framework/Trixi.jl/issues/1671#issuecomment-1765644064
+=#
 @inline function weak_form_kernel!(du, u,
                                    element,
                                    mesh::Union{StructuredMesh{2}, UnstructuredMesh2D,

diff --git a/src/solvers/dgsem_structured/dg_3d.jl b/src/solvers/dgsem_structured/dg_3d.jl
@@ -49,6 +49,13 @@ function rhs!(du, u, t,
     return nothing
 end
 
+#=
+`weak_form_kernel!` is only implemented for conserved terms as 
+non-conservative terms should always be discretized in conjunction with a flux-splitting scheme, 
+see `flux_differencing_kernel!`.
+This treatment is required to achieve, e.g., entropy-stability or well-balancedness.
+See also https://github.com/trixi-framework/Trixi.jl/issues/1671#issuecomment-1765644064
+=#
 @inline function weak_form_kernel!(du, u,
                                    element,
                                    mesh::Union{StructuredMesh{3}, P4estMesh{3}},

diff --git a/src/solvers/dgsem_tree/dg_1d.jl b/src/solvers/dgsem_tree/dg_1d.jl
@@ -144,6 +144,13 @@ function calc_volume_integral!(du, u,
     return nothing
 end
 
+#=
+`weak_form_kernel!` is only implemented for conserved terms as 
+non-conservative terms should always be discretized in conjunction with a flux-splitting scheme, 
+see `flux_differencing_kernel!`.
+This treatment is required to achieve, e.g., entropy-stability or well-balancedness.
+See also https://github.com/trixi-framework/Trixi.jl/issues/1671#issuecomment-1765644064
+=#
 @inline function weak_form_kernel!(du, u,
                                    element, mesh::Union{TreeMesh{1}, StructuredMesh{1}},
                                    nonconservative_terms::False, equations,

diff --git a/src/solvers/dgsem_tree/dg_2d.jl b/src/solvers/dgsem_tree/dg_2d.jl
@@ -194,6 +194,13 @@ function calc_volume_integral!(du, u,
     return nothing
 end
 
+#=
+`weak_form_kernel!` is only implemented for conserved terms as 
+non-conservative terms should always be discretized in conjunction with a flux-splitting scheme, 
+see `flux_differencing_kernel!`.
+This treatment is required to achieve, e.g., entropy-stability or well-balancedness.
+See also https://github.com/trixi-framework/Trixi.jl/issues/1671#issuecomment-1765644064
+=#
 @inline function weak_form_kernel!(du, u,
                                    element, mesh::TreeMesh{2},
                                    nonconservative_terms::False, equations,

diff --git a/src/solvers/dgsem_tree/dg_3d.jl b/src/solvers/dgsem_tree/dg_3d.jl
@@ -223,6 +223,13 @@ function calc_volume_integral!(du, u,
     return nothing
 end
 
+#=
+`weak_form_kernel!` is only implemented for conserved terms as 
+non-conservative terms should always be discretized in conjunction with a flux-splitting scheme, 
+see `flux_differencing_kernel!`.
+This treatment is required to achieve, e.g., entropy-stability or well-balancedness.
+See also https://github.com/trixi-framework/Trixi.jl/issues/1671#issuecomment-1765644064
+=#
 @inline function weak_form_kernel!(du, u,
                                    element, mesh::TreeMesh{3},
                                    nonconservative_terms::False, equations,