blender/intern/cycles/bvh/bvh4.cpp

/*
 * Adapted from code copyright 2009-2010 NVIDIA Corporation
 * Modifications Copyright 2011, Blender Foundation.
 *
 * Licensed under the Apache License, Version 2.0 (the "License");
 * you may not use this file except in compliance with the License.
 * You may obtain a copy of the License at
 *
 * http://www.apache.org/licenses/LICENSE-2.0
 *
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 */

#include "bvh/bvh4.h"

#include "render/mesh.h"
#include "render/object.h"

#include "bvh/bvh_node.h"
#include "bvh/bvh_unaligned.h"

CCL_NAMESPACE_BEGIN

/* Can we avoid this somehow or make more generic?
 *
 * Perhaps we can merge nodes in actual tree and make our
 * life easier all over the place.
 */

BVH4::BVH4(const BVHParams &params_,
           const vector<Mesh *> &meshes_,
           const vector<Object *> &objects_)
    : BVH(params_, meshes_, objects_)
{
  params.bvh_layout = BVH_LAYOUT_BVH4;
}

namespace {

BVHNode *bvh_node_merge_children_recursively(const BVHNode *node)
{
  if (node->is_leaf()) {
    return new LeafNode(*reinterpret_cast<const LeafNode *>(node));
  }
  /* Collect nodes of one layer deeper, allowing us to have more children in an inner layer. */
  assert(node->num_children() <= 2);
  const BVHNode *children[4];
  const BVHNode *child0 = node->get_child(0);
  const BVHNode *child1 = node->get_child(1);
  int num_children = 0;
  if (child0->is_leaf()) {
    children[num_children++] = child0;
  }
  else {
    children[num_children++] = child0->get_child(0);
    children[num_children++] = child0->get_child(1);
  }
  if (child1->is_leaf()) {
    children[num_children++] = child1;
  }
  else {
    children[num_children++] = child1->get_child(0);
    children[num_children++] = child1->get_child(1);
  }
  /* Merge children in subtrees. */
  BVHNode *children4[4];
  for (int i = 0; i < num_children; ++i) {
    children4[i] = bvh_node_merge_children_recursively(children[i]);
  }
  /* Allocate new node. */
  BVHNode *node4 = new InnerNode(node->bounds, children4, num_children);
  /* TODO(sergey): Consider doing this from the InnerNode() constructor.
   * But in order to do this nicely need to think of how to pass all the
   * parameters there. */
  if (node->is_unaligned) {
    node4->is_unaligned = true;
    node4->aligned_space = new Transform();
    *node4->aligned_space = *node->aligned_space;
  }
  return node4;
}

}  // namespace

BVHNode *BVH4::widen_children_nodes(const BVHNode *root)
{
  if (root == NULL) {
    return NULL;
  }
  if (root->is_leaf()) {
    return const_cast<BVHNode *>(root);
  }
  BVHNode *root4 = bvh_node_merge_children_recursively(root);
  /* TODO(sergey): Pack children nodes to parents which has less that 4
   * children. */
  return root4;
}

void BVH4::pack_leaf(const BVHStackEntry &e, const LeafNode *leaf)
{
  float4 data[BVH_QNODE_LEAF_SIZE];
  memset(data, 0, sizeof(data));
  if (leaf->num_triangles() == 1 && pack.prim_index[leaf->lo] == -1) {
    /* object */
    data[0].x = __int_as_float(~(leaf->lo));
    data[0].y = __int_as_float(0);
  }
  else {
    /* triangle */
    data[0].x = __int_as_float(leaf->lo);
    data[0].y = __int_as_float(leaf->hi);
  }
  data[0].z = __uint_as_float(leaf->visibility);
  if (leaf->num_triangles() != 0) {
    data[0].w = __uint_as_float(pack.prim_type[leaf->lo]);
  }

  memcpy(&pack.leaf_nodes[e.idx], data, sizeof(float4) * BVH_QNODE_LEAF_SIZE);
}

void BVH4::pack_inner(const BVHStackEntry &e, const BVHStackEntry *en, int num)
{
  bool has_unaligned = false;
  /* Check whether we have to create unaligned node or all nodes are aligned
   * and we can cut some corner here.
   */
  if (params.use_unaligned_nodes) {
    for (int i = 0; i < num; i++) {
      if (en[i].node->is_unaligned) {
        has_unaligned = true;
        break;
      }
    }
  }
  if (has_unaligned) {
    /* There's no unaligned children, pack into AABB node. */
    pack_unaligned_inner(e, en, num);
  }
  else {
    /* Create unaligned node with orientation transform for each of the
     * children.
     */
    pack_aligned_inner(e, en, num);
  }
}

void BVH4::pack_aligned_inner(const BVHStackEntry &e, const BVHStackEntry *en, int num)
{
  BoundBox bounds[4];
  int child[4];
  for (int i = 0; i < num; ++i) {
    bounds[i] = en[i].node->bounds;
    child[i] = en[i].encodeIdx();
  }
  pack_aligned_node(
      e.idx, bounds, child, e.node->visibility, e.node->time_from, e.node->time_to, num);
}

void BVH4::pack_aligned_node(int idx,
                             const BoundBox *bounds,
                             const int *child,
                             const uint visibility,
                             const float time_from,
                             const float time_to,
                             const int num)
{
  float4 data[BVH_QNODE_SIZE];
  memset(data, 0, sizeof(data));

  data[0].x = __uint_as_float(visibility & ~PATH_RAY_NODE_UNALIGNED);
  data[0].y = time_from;
  data[0].z = time_to;

  for (int i = 0; i < num; i++) {
    float3 bb_min = bounds[i].min;
    float3 bb_max = bounds[i].max;

    data[1][i] = bb_min.x;
    data[2][i] = bb_max.x;
    data[3][i] = bb_min.y;
    data[4][i] = bb_max.y;
    data[5][i] = bb_min.z;
    data[6][i] = bb_max.z;

    data[7][i] = __int_as_float(child[i]);
  }

  for (int i = num; i < 4; i++) {
    /* We store BB which would never be recorded as intersection
     * so kernel might safely assume there are always 4 child nodes.
     */
    data[1][i] = FLT_MAX;
    data[2][i] = -FLT_MAX;

    data[3][i] = FLT_MAX;
    data[4][i] = -FLT_MAX;

    data[5][i] = FLT_MAX;
    data[6][i] = -FLT_MAX;

    data[7][i] = __int_as_float(0);
  }

  memcpy(&pack.nodes[idx], data, sizeof(float4) * BVH_QNODE_SIZE);
}

void BVH4::pack_unaligned_inner(const BVHStackEntry &e, const BVHStackEntry *en, int num)
{
  Transform aligned_space[4];
  BoundBox bounds[4];
  int child[4];
  for (int i = 0; i < num; ++i) {
    aligned_space[i] = en[i].node->get_aligned_space();
    bounds[i] = en[i].node->bounds;
    child[i] = en[i].encodeIdx();
  }
  pack_unaligned_node(e.idx,
                      aligned_space,
                      bounds,
                      child,
                      e.node->visibility,
                      e.node->time_from,
                      e.node->time_to,
                      num);
}

void BVH4::pack_unaligned_node(int idx,
                               const Transform *aligned_space,
                               const BoundBox *bounds,
                               const int *child,
                               const uint visibility,
                               const float time_from,
                               const float time_to,
                               const int num)
{
  float4 data[BVH_UNALIGNED_QNODE_SIZE];
  memset(data, 0, sizeof(data));

  data[0].x = __uint_as_float(visibility | PATH_RAY_NODE_UNALIGNED);
  data[0].y = time_from;
  data[0].z = time_to;

  for (int i = 0; i < num; i++) {
    Transform space = BVHUnaligned::compute_node_transform(bounds[i], aligned_space[i]);

    data[1][i] = space.x.x;
    data[2][i] = space.x.y;
    data[3][i] = space.x.z;

    data[4][i] = space.y.x;
    data[5][i] = space.y.y;
    data[6][i] = space.y.z;

    data[7][i] = space.z.x;
    data[8][i] = space.z.y;
    data[9][i] = space.z.z;

    data[10][i] = space.x.w;
    data[11][i] = space.y.w;
    data[12][i] = space.z.w;

    data[13][i] = __int_as_float(child[i]);
  }

  for (int i = num; i < 4; i++) {
    /* We store BB which would never be recorded as intersection
     * so kernel might safely assume there are always 4 child nodes.
     */

    data[1][i] = NAN;
    data[2][i] = NAN;
    data[3][i] = NAN;

    data[4][i] = NAN;
    data[5][i] = NAN;
    data[6][i] = NAN;

    data[7][i] = NAN;
    data[8][i] = NAN;
    data[9][i] = NAN;

    data[10][i] = NAN;
    data[11][i] = NAN;
    data[12][i] = NAN;

    data[13][i] = __int_as_float(0);
  }

  memcpy(&pack.nodes[idx], data, sizeof(float4) * BVH_UNALIGNED_QNODE_SIZE);
}

/* Quad SIMD Nodes */

void BVH4::pack_nodes(const BVHNode *root)
{
  /* Calculate size of the arrays required. */
  const size_t num_nodes = root->getSubtreeSize(BVH_STAT_NODE_COUNT);
  const size_t num_leaf_nodes = root->getSubtreeSize(BVH_STAT_LEAF_COUNT);
  assert(num_leaf_nodes <= num_nodes);
  const size_t num_inner_nodes = num_nodes - num_leaf_nodes;
  size_t node_size;
  if (params.use_unaligned_nodes) {
    const size_t num_unaligned_nodes = root->getSubtreeSize(BVH_STAT_UNALIGNED_INNER_COUNT);
    node_size = (num_unaligned_nodes * BVH_UNALIGNED_QNODE_SIZE) +
                (num_inner_nodes - num_unaligned_nodes) * BVH_QNODE_SIZE;
  }
  else {
    node_size = num_inner_nodes * BVH_QNODE_SIZE;
  }
  /* Resize arrays. */
  pack.nodes.clear();
  pack.leaf_nodes.clear();
  /* For top level BVH, first merge existing BVH's so we know the offsets. */
  if (params.top_level) {
    pack_instances(node_size, num_leaf_nodes * BVH_QNODE_LEAF_SIZE);
  }
  else {
    pack.nodes.resize(node_size);
    pack.leaf_nodes.resize(num_leaf_nodes * BVH_QNODE_LEAF_SIZE);
  }

  int nextNodeIdx = 0, nextLeafNodeIdx = 0;

  vector<BVHStackEntry> stack;
  stack.reserve(BVHParams::MAX_DEPTH * 2);
  if (root->is_leaf()) {
    stack.push_back(BVHStackEntry(root, nextLeafNodeIdx++));
  }
  else {
    stack.push_back(BVHStackEntry(root, nextNodeIdx));
    nextNodeIdx += root->has_unaligned() ? BVH_UNALIGNED_QNODE_SIZE : BVH_QNODE_SIZE;
  }

  while (stack.size()) {
    BVHStackEntry e = stack.back();
    stack.pop_back();

    if (e.node->is_leaf()) {
      /* leaf node */
      const LeafNode *leaf = reinterpret_cast<const LeafNode *>(e.node);
      pack_leaf(e, leaf);
    }
    else {
      /* Inner node. */
      /* Collect nodes. */
      const BVHNode *children[4];
      const int num_children = e.node->num_children();
      /* Push entries on the stack. */
      for (int i = 0; i < num_children; ++i) {
        int idx;
        children[i] = e.node->get_child(i);
        assert(children[i] != NULL);
        if (children[i]->is_leaf()) {
          idx = nextLeafNodeIdx++;
        }
        else {
          idx = nextNodeIdx;
          nextNodeIdx += children[i]->has_unaligned() ? BVH_UNALIGNED_QNODE_SIZE : BVH_QNODE_SIZE;
        }
        stack.push_back(BVHStackEntry(children[i], idx));
      }
      /* Set node. */
      pack_inner(e, &stack[stack.size() - num_children], num_children);
    }
  }

  assert(node_size == nextNodeIdx);
  /* Root index to start traversal at, to handle case of single leaf node. */
  pack.root_index = (root->is_leaf()) ? -1 : 0;
}

void BVH4::refit_nodes()
{
  assert(!params.top_level);

  BoundBox bbox = BoundBox::empty;
  uint visibility = 0;
  refit_node(0, (pack.root_index == -1) ? true : false, bbox, visibility);
}

void BVH4::refit_node(int idx, bool leaf, BoundBox &bbox, uint &visibility)
{
  if (leaf) {
    /* Refit leaf node. */
    int4 *data = &pack.leaf_nodes[idx];
    int4 c = data[0];

    BVH::refit_primitives(c.x, c.y, bbox, visibility);

    /* TODO(sergey): This is actually a copy of pack_leaf(),
     * but this chunk of code only knows actual data and has
     * no idea about BVHNode.
     *
     * Would be nice to de-duplicate code, but trying to make
     * making code more general ends up in much nastier code
     * in my opinion so far.
     *
     * Same applies to the inner nodes case below.
     */
    float4 leaf_data[BVH_QNODE_LEAF_SIZE];
    leaf_data[0].x = __int_as_float(c.x);
    leaf_data[0].y = __int_as_float(c.y);
    leaf_data[0].z = __uint_as_float(visibility);
    leaf_data[0].w = __uint_as_float(c.w);
    memcpy(&pack.leaf_nodes[idx], leaf_data, sizeof(float4) * BVH_QNODE_LEAF_SIZE);
  }
  else {
    int4 *data = &pack.nodes[idx];
    bool is_unaligned = (data[0].x & PATH_RAY_NODE_UNALIGNED) != 0;
    int4 c;
    if (is_unaligned) {
      c = data[13];
    }
    else {
      c = data[7];
    }
    /* Refit inner node, set bbox from children. */
    BoundBox child_bbox[4] = {BoundBox::empty, BoundBox::empty, BoundBox::empty, BoundBox::empty};
    uint child_visibility[4] = {0};
    int num_nodes = 0;

    for (int i = 0; i < 4; ++i) {
      if (c[i] != 0) {
        refit_node((c[i] < 0) ? -c[i] - 1 : c[i], (c[i] < 0), child_bbox[i], child_visibility[i]);
        ++num_nodes;
        bbox.grow(child_bbox[i]);
        visibility |= child_visibility[i];
      }
    }

    if (is_unaligned) {
      Transform aligned_space[4] = {
          transform_identity(), transform_identity(), transform_identity(), transform_identity()};
      pack_unaligned_node(
          idx, aligned_space, child_bbox, &c[0], visibility, 0.0f, 1.0f, num_nodes);
    }
    else {
      pack_aligned_node(idx, child_bbox, &c[0], visibility, 0.0f, 1.0f, num_nodes);
    }
  }
}

CCL_NAMESPACE_END