Dispatch all queued material draws. Caller owns the render pass; this method binds the pipelines + groups + meshes and issues indexed draws. `mesh_fetch` is a closure that returns `(vertex_buffer, index_buffer, index_count)` for a given (mesh_handle, mesh_idx) — lets the renderer hand over its `model_gpu_cache` without this module taking a dependency on it.
(
&'pass self,
pass: &mut wgpu::RenderPass<'pass>,
mesh_fetch: F,
)
| 1555 | /// (mesh_handle, mesh_idx) — lets the renderer hand over its |
| 1556 | /// `model_gpu_cache` without this module taking a dependency on it. |
| 1557 | pub fn dispatch<'pass, F>( |
| 1558 | &'pass self, |
| 1559 | pass: &mut wgpu::RenderPass<'pass>, |
| 1560 | mesh_fetch: F, |
| 1561 | ) |
| 1562 | where F: FnMut(u64, usize) -> Option<(&'pass wgpu::Buffer, &'pass wgpu::Buffer, u32)> |
| 1563 | { |
| 1564 | self.dispatch_with_view(pass, &self.per_view_bg, |_| true, false, mesh_fetch); |
| 1565 | } |
| 1566 | |
| 1567 | /// EN-011 — like `dispatch`, but uses a caller-supplied PerView |
| 1568 | /// bind group (so the planar-reflection pass can render the same |
no test coverage detected