griefith/test2
1
0
Fork 0
Code Issues Pull Requests Actions 1 Packages Projects Releases Wiki Activity

Mesh: Limit threading for copying positions to GPU data

Browse Source 
See b99c1abc3a for more information about how using fewer
threads for just copying data can improve performance. In my simple
test file with mesh data re-uploaded every frame, this improved
performance from 23.5 FPS to 25.5 FPS (almost 9%).
This commit is contained in:
Hans Goudey
2024-03-28 18:37:23 -04:00
parent 5ed99e6361
commit 510874f7b9
1 changed files with 14 additions and 8 deletions
Download Patch File Download Diff File Expand all files Collapse all files

22
source/blender/draw/intern/mesh_extractors/extract_mesh_vbo_pos.cc
View File

@@ -48,14 +48,20 @@ static void extract_pos_init(const MeshRenderData &mr,
MutableSpan vbo_data(static_cast<float3 *>(GPU_vertbuf_get_data(vbo)),
GPU_vertbuf_get_vertex_len(vbo));
if (mr.extract_type == MR_EXTRACT_MESH) {
array_utils::gather(
mr.vert_positions, mr.corner_verts, vbo_data.take_front(mr.corner_verts.size()));
extract_mesh_loose_edge_positions(mr.vert_positions,
mr.edges,
mr.loose_edges,
vbo_data.slice(mr.corners_num, mr.loose_edges.size() * 2));
array_utils::gather(
mr.vert_positions, mr.loose_verts, vbo_data.take_back(mr.loose_verts.size()));
threading::memory_bandwidth_bound_task(
mr.vert_positions.size_in_bytes() + mr.corner_verts.size_in_bytes() +
vbo_data.size_in_bytes() + mr.loose_edges.size(),
[&]() {
array_utils::gather(
mr.vert_positions, mr.corner_verts, vbo_data.take_front(mr.corner_verts.size()));
extract_mesh_loose_edge_positions(
mr.vert_positions,
mr.edges,
mr.loose_edges,
vbo_data.slice(mr.corners_num, mr.loose_edges.size() * 2));
array_utils::gather(
mr.vert_positions, mr.loose_verts, vbo_data.take_back(mr.loose_verts.size()));
});
}
else {
*static_cast<float3 **>(tls_data) = vbo_data.data();