Functions: make copying virtual arrays to span more efficient

Sometimes functions expect a span instead of a virtual array.
If the virtual array is a span internally already, great. But if it is
not (e.g. the position attribute on a mesh), the elements have
to be copied over to a span.

This patch makes the copying process more efficient by giving
the compiler more opportunity for optimization.

1 files changed, 11 insertions, 0 deletions

diff --git a/source/blender/functions/intern/generic_virtual_array.cc b/source/blender/functions/intern/generic_virtual_array.cc
index e11501828f8..754a2156a65 100644
--- a/source/blender/functions/intern/generic_virtual_array.cc
+++ b/source/blender/functions/intern/generic_virtual_array.cc
@@ -22,8 +22,19 @@ namespace blender::fn {
  * GVArray.
  */
 
+void GVArray::materialize_to_uninitialized(void *dst) const
+{
+  this->materialize_to_uninitialized(IndexMask(size_), dst);
+}
+
 void GVArray::materialize_to_uninitialized(const IndexMask mask, void *dst) const
 {
+  BLI_assert(mask.min_array_size() <= size_);
+  this->materialize_to_uninitialized_impl(mask, dst);
+}
+
+void GVArray::materialize_to_uninitialized_impl(const IndexMask mask, void *dst) const
+{
   for (const int64_t i : mask) {
     void *elem_dst = POINTER_OFFSET(dst, type_->size() * i);
     this->get_to_uninitialized(i, elem_dst);