Mesh: Remove redundant custom data pointers

For copy-on-write, we want to share attribute arrays between meshes
where possible. Mutable pointers like `Mesh.mvert` make that difficult
by making ownership vague. They also make code more complex by adding
redundancy.

The simplest solution is just removing them and retrieving layers from
`CustomData` as needed. Similar changes have already been applied to
curves and point clouds (e9f82d3dc7ee, 410a6efb747f). Removing use of
the pointers generally makes code more obvious and more reusable.

Mesh data is now accessed with a C++ API (`Mesh::edges()` or
`Mesh::edges_for_write()`), and a C API (`BKE_mesh_edges(mesh)`).

The CoW changes this commit makes possible are described in T95845
and T95842, and started in D14139 and D14140. The change also simplifies
the ongoing mesh struct-of-array refactors from T95965.

**RNA/Python Access Performance**
Theoretically, accessing mesh elements with the RNA API may become
slower, since the layer needs to be found on every random access.
However, overhead is already high enough that this doesn't make a
noticible differenc, and performance is actually improved in some
cases. Random access can be up to 10% faster, but other situations
might be a bit slower. Generally using `foreach_get/set` are the best
way to improve performance. See the differential revision for more
discussion about Python performance.

Cycles has been updated to use raw pointers and the internal Blender
mesh types, mostly because there is no sense in having this overhead
when it's already compiled with Blender. In my tests this roughly
halves the Cycles mesh creation time (0.19s to 0.10s for a 1 million
face grid).

Differential Revision: https://developer.blender.org/D15488

1 files changed, 8 insertions, 6 deletions

diff --git a/source/blender/blenkernel/intern/mesh_calc_edges.cc b/source/blender/blenkernel/intern/mesh_calc_edges.cc
index 31e20750cf2..8a8960fb8a7 100644
--- a/source/blender/blenkernel/intern/mesh_calc_edges.cc
+++ b/source/blender/blenkernel/intern/mesh_calc_edges.cc
@@ -80,9 +80,10 @@ static void add_existing_edges_to_hash_maps(Mesh *mesh,
                                             uint32_t parallel_mask)
 {
   /* Assume existing edges are valid. */
+  const Span<MEdge> edges = mesh->edges();
   threading::parallel_for_each(edge_maps, [&](EdgeMap &edge_map) {
     const int task_index = &edge_map - edge_maps.data();
-    for (const MEdge &edge : Span(mesh->medge, mesh->totedge)) {
+    for (const MEdge &edge : edges) {
       OrderedEdge ordered_edge{edge.v1, edge.v2};
       /* Only add the edge when it belongs into this map. */
       if (task_index == (parallel_mask & ordered_edge.hash2())) {
@@ -96,10 +97,11 @@ static void add_polygon_edges_to_hash_maps(Mesh *mesh,
                                            MutableSpan<EdgeMap> edge_maps,
                                            uint32_t parallel_mask)
 {
-  const Span<MLoop> loops{mesh->mloop, mesh->totloop};
+  const Span<MPoly> polys = mesh->polygons();
+  const Span<MLoop> loops = mesh->loops();
   threading::parallel_for_each(edge_maps, [&](EdgeMap &edge_map) {
     const int task_index = &edge_map - edge_maps.data();
-    for (const MPoly &poly : Span(mesh->mpoly, mesh->totpoly)) {
+    for (const MPoly &poly : polys) {
       Span<MLoop> poly_loops = loops.slice(poly.loopstart, poly.totloop);
       const MLoop *prev_loop = &poly_loops.last();
       for (const MLoop &next_loop : poly_loops) {
@@ -157,10 +159,11 @@ static void update_edge_indices_in_poly_loops(Mesh *mesh,
                                               Span<EdgeMap> edge_maps,
                                               uint32_t parallel_mask)
 {
-  const MutableSpan<MLoop> loops{mesh->mloop, mesh->totloop};
+  const Span<MPoly> polys = mesh->polygons();
+  MutableSpan<MLoop> loops = mesh->loops_for_write();
   threading::parallel_for(IndexRange(mesh->totpoly), 100, [&](IndexRange range) {
     for (const int poly_index : range) {
-      MPoly &poly = mesh->mpoly[poly_index];
+      const MPoly &poly = polys[poly_index];
       MutableSpan<MLoop> poly_loops = loops.slice(poly.loopstart, poly.totloop);
 
       MLoop *prev_loop = &poly_loops.last();
@@ -242,7 +245,6 @@ void BKE_mesh_calc_edges(Mesh *mesh, bool keep_existing_edges, const bool select
   CustomData_reset(&mesh->edata);
   CustomData_add_layer(&mesh->edata, CD_MEDGE, CD_ASSIGN, new_edges.data(), new_totedge);
   mesh->totedge = new_totedge;
-  mesh->medge = new_edges.data();
 
   /* Explicitly clear edge maps, because that way it can be parallelized. */
   clear_hash_tables(edge_maps);