Files
test2/source/blender/blenkernel/BKE_mesh_sample.hh
Jacques Lucke 2cfcb8b0b8 BLI: refactor IndexMask for better performance and memory usage
Goals of this refactor:
* Reduce memory consumption of `IndexMask`. The old `IndexMask` uses an
  `int64_t` for each index which is more than necessary in pretty much all
  practical cases currently. Using `int32_t` might still become limiting
  in the future in case we use this to index e.g. byte buffers larger than
  a few gigabytes. We also don't want to template `IndexMask`, because
  that would cause a split in the "ecosystem", or everything would have to
  be implemented twice or templated.
* Allow for more multi-threading. The old `IndexMask` contains a single
  array. This is generally good but has the problem that it is hard to fill
  from multiple-threads when the final size is not known from the beginning.
  This is commonly the case when e.g. converting an array of bool to an
  index mask. Currently, this kind of code only runs on a single thread.
* Allow for efficient set operations like join, intersect and difference.
  It should be possible to multi-thread those operations.
* It should be possible to iterate over an `IndexMask` very efficiently.
  The most important part of that is to avoid all memory access when iterating
  over continuous ranges. For some core nodes (e.g. math nodes), we generate
  optimized code for the cases of irregular index masks and simple index ranges.

To achieve these goals, a few compromises had to made:
* Slicing of the mask (at specific indices) and random element access is
  `O(log #indices)` now, but with a low constant factor. It should be possible
  to split a mask into n approximately equally sized parts in `O(n)` though,
  making the time per split `O(1)`.
* Using range-based for loops does not work well when iterating over a nested
  data structure like the new `IndexMask`. Therefor, `foreach_*` functions with
  callbacks have to be used. To avoid extra code complexity at the call site,
  the `foreach_*` methods support multi-threading out of the box.

The new data structure splits an `IndexMask` into an arbitrary number of ordered
`IndexMaskSegment`. Each segment can contain at most `2^14 = 16384` indices. The
indices within a segment are stored as `int16_t`. Each segment has an additional
`int64_t` offset which allows storing arbitrary `int64_t` indices. This approach
has the main benefits that segments can be processed/constructed individually on
multiple threads without a serial bottleneck. Also it reduces the memory
requirements significantly.

For more details see comments in `BLI_index_mask.hh`.

I did a few tests to verify that the data structure generally improves
performance and does not cause regressions:
* Our field evaluation benchmarks take about as much as before. This is to be
  expected because we already made sure that e.g. add node evaluation is
  vectorized. The important thing here is to check that changes to the way we
  iterate over the indices still allows for auto-vectorization.
* Memory usage by a mask is about 1/4 of what it was before in the average case.
  That's mainly caused by the switch from `int64_t` to `int16_t` for indices.
  In the worst case, the memory requirements can be larger when there are many
  indices that are very far away. However, when they are far away from each other,
  that indicates that there aren't many indices in total. In common cases, memory
  usage can be way lower than 1/4 of before, because sub-ranges use static memory.
* For some more specific numbers I benchmarked `IndexMask::from_bools` in
  `index_mask_from_selection` on 10.000.000 elements at various probabilities for
  `true` at every index:
  ```
  Probability      Old        New
  0              4.6 ms     0.8 ms
  0.001          5.1 ms     1.3 ms
  0.2            8.4 ms     1.8 ms
  0.5           15.3 ms     3.0 ms
  0.8           20.1 ms     3.0 ms
  0.999         25.1 ms     1.7 ms
  1             13.5 ms     1.1 ms
  ```

Pull Request: https://projects.blender.org/blender/blender/pulls/104629
2023-05-24 18:11:41 +02:00

193 lines
7.0 KiB
C++

/* SPDX-License-Identifier: GPL-2.0-or-later */
#pragma once
/** \file
* \ingroup bke
*/
#include "BLI_function_ref.hh"
#include "BLI_generic_virtual_array.hh"
#include "BLI_math_vector_types.hh"
#include "FN_field.hh"
#include "FN_multi_function.hh"
#include "DNA_meshdata_types.h"
#include "BKE_attribute.h"
#include "BKE_geometry_fields.hh"
struct Mesh;
struct BVHTreeFromMesh;
namespace blender {
class RandomNumberGenerator;
}
namespace blender::bke::mesh_surface_sample {
void sample_point_attribute(Span<int> corner_verts,
Span<MLoopTri> looptris,
Span<int> looptri_indices,
Span<float3> bary_coords,
const GVArray &src,
const IndexMask &mask,
GMutableSpan dst);
void sample_point_normals(Span<int> corner_verts,
Span<MLoopTri> looptris,
Span<int> looptri_indices,
Span<float3> bary_coords,
Span<float3> src,
IndexMask mask,
MutableSpan<float3> dst);
void sample_corner_attribute(Span<MLoopTri> looptris,
Span<int> looptri_indices,
Span<float3> bary_coords,
const GVArray &src,
const IndexMask &mask,
GMutableSpan dst);
void sample_corner_normals(Span<MLoopTri> looptris,
Span<int> looptri_indices,
Span<float3> bary_coords,
Span<float3> src,
const IndexMask &mask,
MutableSpan<float3> dst);
void sample_face_attribute(Span<int> looptri_polys,
Span<int> looptri_indices,
const GVArray &src,
const IndexMask &mask,
GMutableSpan dst);
/**
* Find randomly distributed points on the surface of a mesh within a 3D sphere. This does not
* sample an exact number of points because it comes with extra overhead to avoid bias that is only
* required in some cases. If an exact number of points is required, that has to be implemented at
* a higher level.
*
* \param approximate_density: Roughly the number of points per unit of area.
* \return The number of added points.
*/
int sample_surface_points_spherical(RandomNumberGenerator &rng,
const Mesh &mesh,
Span<int> looptri_indices_to_sample,
const float3 &sample_pos,
float sample_radius,
float approximate_density,
Vector<float3> &r_bary_coords,
Vector<int> &r_looptri_indices,
Vector<float3> &r_positions);
/**
* Find randomly distributed points on the surface of a mesh within a circle that is projected on
* the mesh. This does not result in an exact number of points because that would come with extra
* overhead and is not always possible. If an exact number of points is required, that has to be
* implemented at a higher level.
*
* \param region_position_to_ray: Function that converts a 2D position into a 3D ray that is used
* to find positions on the mesh.
* \param mesh_bvhtree: BVH tree of the triangles in the mesh. Passed in so that it does not have
* to be retrieved again.
* \param tries_num: Number of 2d positions that are sampled. The maximum
* number of new samples.
* \return The number of added points.
*/
int sample_surface_points_projected(
RandomNumberGenerator &rng,
const Mesh &mesh,
BVHTreeFromMesh &mesh_bvhtree,
const float2 &sample_pos_re,
float sample_radius_re,
FunctionRef<void(const float2 &pos_re, float3 &r_start, float3 &r_end)> region_position_to_ray,
bool front_face_only,
int tries_num,
int max_points,
Vector<float3> &r_bary_coords,
Vector<int> &r_looptri_indices,
Vector<float3> &r_positions);
float3 compute_bary_coord_in_triangle(Span<float3> vert_positions,
Span<int> corner_verts,
const MLoopTri &looptri,
const float3 &position);
template<typename T>
inline T sample_corner_attribute_with_bary_coords(const float3 &bary_weights,
const MLoopTri &looptri,
const Span<T> corner_attribute)
{
return attribute_math::mix3(bary_weights,
corner_attribute[looptri.tri[0]],
corner_attribute[looptri.tri[1]],
corner_attribute[looptri.tri[2]]);
}
template<typename T>
inline T sample_corner_attribute_with_bary_coords(const float3 &bary_weights,
const MLoopTri &looptri,
const VArray<T> &corner_attribute)
{
return attribute_math::mix3(bary_weights,
corner_attribute[looptri.tri[0]],
corner_attribute[looptri.tri[1]],
corner_attribute[looptri.tri[2]]);
}
/**
* Calculate barycentric weights from triangle indices and positions within the triangles.
*/
class BaryWeightFromPositionFn : public mf::MultiFunction {
GeometrySet source_;
Span<float3> vert_positions_;
Span<int> corner_verts_;
Span<MLoopTri> looptris_;
public:
BaryWeightFromPositionFn(GeometrySet geometry);
void call(const IndexMask &mask, mf::Params params, mf::Context context) const;
};
/**
* Calculate face corner weights from triangle indices and positions within the triangles.
* The weights are 1 for the nearest corner and 0 for the two others.
*/
class CornerBaryWeightFromPositionFn : public mf::MultiFunction {
GeometrySet source_;
Span<float3> vert_positions_;
Span<int> corner_verts_;
Span<MLoopTri> looptris_;
public:
CornerBaryWeightFromPositionFn(GeometrySet geometry);
void call(const IndexMask &mask, mf::Params params, mf::Context context) const;
};
/**
* Evaluate an attribute on the input geometry and sample it with input barycentric weights and
* triangle indices.
*/
class BaryWeightSampleFn : public mf::MultiFunction {
mf::Signature signature_;
GeometrySet source_;
Span<MLoopTri> looptris_;
std::optional<bke::MeshFieldContext> source_context_;
std::unique_ptr<fn::FieldEvaluator> source_evaluator_;
const GVArray *source_data_;
eAttrDomain domain_;
public:
BaryWeightSampleFn(GeometrySet geometry, fn::GField src_field);
void call(const IndexMask &mask, mf::Params params, mf::Context context) const;
private:
void evaluate_source(fn::GField src_field);
};
} // namespace blender::bke::mesh_surface_sample