Expand description
GPU fluid simulation: Navier-Stokes, LBM kernels, SPH kernels, vortex confinement, and shallow-water equations — CPU mock backend.
All structures here simulate GPU buffer management and dispatch via
in-memory Vecf64` buffers so that the logic can be unit-tested without
a real GPU device.
Structs§
- Fluid
GpuBuffer - A double-buffered GPU field for ping-pong updates.
- Fluid
GpuPipeline - Ordered fluid GPU pipeline with memory bandwidth estimation.
- LBMBoundary
Kernel Spec - Specification of an LBM boundary kernel.
- LBMCollision
Kernel Spec - Specification of an LBM collision kernel.
- LBMGpu
Kernels - GPU kernel manager for a Lattice Boltzmann Method simulation.
- LBMStreaming
Spec - Specification of an LBM streaming kernel.
- Navier
Stokes Gpu - GPU (CPU mock) Navier-Stokes solver on a 3-D Cartesian grid.
- Pipeline
Pass - A single pass (dispatch) in the fluid GPU pipeline.
- SPHGpu
Kernels - GPU SPH particle data for density/pressure/viscosity kernels.
- Vortex
Confinement - GPU vortex confinement — amplifies small-scale vortical structures to counteract numerical dissipation.
- Water
SimGpu - GPU shallow-water equation (SWE) solver for height field and ripple simulation on a 2-D grid.
Enums§
- Barrier
Kind - Synchronization barrier kind between GPU passes.
- LBMBoundary
Kind - Variants of supported LBM boundary conditions.
Constants§
- D3Q19_Q
- D3Q19 LBM velocity directions (19 discrete velocities).
- D3Q19_
WEIGHTS - D3Q19 lattice weights.