Niklas Neher

Validation: dynamical pressure boundaries

1

based on #900 # Poiseuille Flow 2D ### RMSEP error (%) # Hagen-Poiseuille Flow 3D ### RMSEP error (%) # Pulsatile Channel Flow 3D ### velocity profile in x-direction

based on #935 Note that the reference paper uses ISPH # Reference Yu et al. (2024) | displacement | pressure | |----------|---------------| | | | # This PR

The variable name `volume` is misleading, since it stores the kernel values, which are the inverse of volume.

Roadmap: robust open boundaries for non-trivial flow scenarios

1

This PR serves as a roadmap to track the necessary changes and implementations required to enable robust open boundary simulations for non-trivial flow scenarios. The primary goal is to systematically...

Add shifting zone

1

In this PR, I aimed to further improve the robustness of the open boundary implementation. To test this, I simulated a pulse wave propagation scenario with `prescribed_pressure_in = 1000` and...

Open boundary functionality fails on AMDGPU because of `findall` bug

1

https://github.com/trixi-framework/TrixiParticles.jl/blob/114a57b1338992ea56ce36aa429bcc11e62e7278/src/general/buffer.jl#L53 See https://github.com/JuliaGPU/AMDGPU.jl/pull/771 https://github.com/JuliaGPU/AcceleratedKernels.jl/issues/41

Cray MPI autodetection fails on AMD MI300A node (missing `mpi_gtl_hsa` preload)

1

On a node with 4× AMD MI300A GPUs, the automatic detection of the setup via `MPIPreferences.use_system_binary(vendor="cray")` fails. ```julia julia> MPIPreferences.use_system_binary(vendor="cray") ERROR: ArgumentError: Collection has multiple elements, must contain exactly 1...