This repository has been archived on 2023-10-09. You can view files and clone it, but cannot push or open issues or pull requests.
Files
blender-archive/source/blender/blenkernel/depsgraph_private.h

170 lines
5.5 KiB
C++
Raw Normal View History

/*
* ***** BEGIN GPL LICENSE BLOCK *****
*
* This program is free software; you can redistribute it and/or
* modify it under the terms of the GNU General Public License
* as published by the Free Software Foundation; either version 2
* of the License, or (at your option) any later version.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program; if not, write to the Free Software Foundation,
2010-02-12 13:34:04 +00:00
* Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
*
* The Original Code is Copyright (C) 2004 Blender Foundation.
* All rights reserved.
*
* Contributor(s): none yet.
*
* ***** END GPL LICENSE BLOCK *****
*/
2011-02-27 20:40:57 +00:00
/** \file blender/blenkernel/depsgraph_private.h
* \ingroup bke
*/
#ifndef __DEPSGRAPH_PRIVATE_H__
#define __DEPSGRAPH_PRIVATE_H__
#include "BKE_depsgraph.h"
#include "DNA_constraint_types.h"
#include "BKE_constraint.h"
Fix depsgraph to compute more accurate links for collision & force. Current implementation more or less indiscriminately links physics objects to colliders and forces, ignoring precise details of layer checks and collider groups. The new depsgraph seemed to lack some such links at all. The relevant code in modifiers suffers from a lot of duplication. Different physics simulations use independent implementations of collision and similar things, which results in a lot of variance: * Cloth collides with objects on same or visible layer with dupli. * Softbody collides with objects on same layer without dupli. * Non-hair particles collide on same layer with dupli. * Smoke uses same code as cloth, but needs different modifier. * Dynamic paint "collides" with brushes on any layer without dupli. Force fields with absorption also imply dependency on colliders: * For most systems, colliders are selected from same layer as field. * For non-hair particles, it uses the same exact set as the particles. As a special quirk, smoke ignores smoke flow force fields; on the other hand dependency on such field implies dependency on the smoke domain. This introduces two utility functions each for old and new depsgraph that are flexible enough to handle all these variations, and uses them to handle particles, cloth, smoke, softbody and dynpaint. One thing to watch out for is that depsgraph code shouldn't rely on any properties that don't cause a graph rebuild when changed. This was violated in the original code that was building force field links, while taking zero field weights into account. This change may cause new dependency cycles in cases where necessary dependencies were missing, but may also remove cycles in situations where unnecessary links were previously created. It's also now possible to solve some cycles by switching to explicit groups, since they are now properly taken into account for dependencies. Differential Revision: https://developer.blender.org/D2141
2016-08-12 18:23:29 +03:00
struct Scene;
struct Group;
struct EffectorWeights;
struct ModifierData;
/* **** DAG relation types *** */
/* scene link to object */
#define DAG_RL_SCENE (1 << 0)
/* object link to data */
#define DAG_RL_DATA (1 << 1)
/* object changes object (parent, track, constraints) */
#define DAG_RL_OB_OB (1 << 2)
/* object changes obdata (hooks, constraints) */
#define DAG_RL_OB_DATA (1 << 3)
/* data changes object (vertex parent) */
#define DAG_RL_DATA_OB (1 << 4)
/* data changes data (deformers) */
#define DAG_RL_DATA_DATA (1 << 5)
#define DAG_NO_RELATION (1 << 6)
#define DAG_RL_ALL_BUT_DATA (DAG_RL_SCENE | DAG_RL_OB_OB | DAG_RL_OB_DATA | DAG_RL_DATA_OB | DAG_RL_DATA_DATA)
#define DAG_RL_ALL (DAG_RL_ALL_BUT_DATA | DAG_RL_DATA)
#define DAGQUEUEALLOC 50
enum {
DAG_WHITE = 0,
DAG_GRAY = 1,
DAG_BLACK = 2
};
typedef struct DagAdjList {
struct DagNode *node;
Result of 2 weeks of quiet coding work in Greece :) Aim was to get a total refresh of the animation system. This is needed because; - we need to upgrade it with 21st century features - current code is spaghetti/hack combo, and hides good design - it should become lag-free with using dependency graphs A full log, with complete code API/structure/design explanation will follow, that's a load of work... so here below the list with hot changes; - The entire object update system (matrices, geometry) is now centralized. Calls to where_is_object and makeDispList are forbidden, instead we tag objects 'changed' and let the depgraph code sort it out - Removed all old "Ika" code - Depgraph is aware of all relationships, including meta balls, constraints, bevelcurve, and so on. - Made depgraph aware of relation types and layers, to do smart flushing of 'changed' events. Nothing gets calculated too often! - Transform uses depgraph to detect changes - On frame-advance, depgraph flushes animated changes Armatures; Almost all armature related code has been fully built from scratch. It now reveils the original design much better, with a very clean implementation, lag free without even calculating each Bone more than once. Result is quite a speedup yes! Important to note is; 1) Armature is data containing the 'rest position' 2) Pose is the changes of rest position, and always on object level. That way more Objects can use same Pose. Also constraints are in Pose 3) Actions only contain the Ipos to change values in Poses. - Bones draw unrotated now - Drawing bones speedup enormously (10-20 times) - Bone selecting in EditMode, selection state is saved for PoseMode, and vice-versa - Undo in editmode - Bone renaming does vertexgroups, constraints, posechannels, actions, for all users of Armature in entire file - Added Bone renaming in NKey panel - Nkey PoseMode shows eulers now - EditMode and PoseMode now have 'active' bone too (last clicked) - Parenting in EditMode' CTRL+P, ALT+P, with nice options! - Pose is added in Outliner now, with showing that constraints are in the Pose, not Armature - Disconnected IK solving from constraints. It's a separate phase now, on top of the full Pose calculations - Pose itself has a dependency graph too, so evaluation order is lag free. TODO NOW; - Rotating in Posemode has incorrect inverse transform (Martin will fix) - Python Bone/Armature/Pose API disabled... needs full recode too (wait for my doc!) - Game engine will need upgrade too - Depgraph code needs revision, cleanup, can be much faster! (But, compliments for Jean-Luc, it works like a charm!) - IK changed, it now doesnt use previous position to advance to next position anymore. That system looks nice (no flips) but is not well suited for NLA and background render. TODO LATER; We now can do loadsa new nifty features as well; like: - Kill PoseMode (can be option for armatures itself) - Make B-Bones (Bezier, Bspline, like for spines) - Move all silly button level edit to 3d window (like CTRL+I = add IK) - Much better & informative drawing - Fix action/nla editors - Put all ipos in Actions (object, mesh key, lamp color) - Add hooks - Null bones - Much more advanced constraints... Bugfixes; - OGL render (view3d header) had wrong first frame on anim render - Ipo 'recording' mode had wrong playback speed - Vertex-key mode now sticks to show 'active key', until frame change -Ton-
2005-07-03 17:35:38 +00:00
short type;
int count; /* number of identical arcs */
Result of 2 weeks of quiet coding work in Greece :) Aim was to get a total refresh of the animation system. This is needed because; - we need to upgrade it with 21st century features - current code is spaghetti/hack combo, and hides good design - it should become lag-free with using dependency graphs A full log, with complete code API/structure/design explanation will follow, that's a load of work... so here below the list with hot changes; - The entire object update system (matrices, geometry) is now centralized. Calls to where_is_object and makeDispList are forbidden, instead we tag objects 'changed' and let the depgraph code sort it out - Removed all old "Ika" code - Depgraph is aware of all relationships, including meta balls, constraints, bevelcurve, and so on. - Made depgraph aware of relation types and layers, to do smart flushing of 'changed' events. Nothing gets calculated too often! - Transform uses depgraph to detect changes - On frame-advance, depgraph flushes animated changes Armatures; Almost all armature related code has been fully built from scratch. It now reveils the original design much better, with a very clean implementation, lag free without even calculating each Bone more than once. Result is quite a speedup yes! Important to note is; 1) Armature is data containing the 'rest position' 2) Pose is the changes of rest position, and always on object level. That way more Objects can use same Pose. Also constraints are in Pose 3) Actions only contain the Ipos to change values in Poses. - Bones draw unrotated now - Drawing bones speedup enormously (10-20 times) - Bone selecting in EditMode, selection state is saved for PoseMode, and vice-versa - Undo in editmode - Bone renaming does vertexgroups, constraints, posechannels, actions, for all users of Armature in entire file - Added Bone renaming in NKey panel - Nkey PoseMode shows eulers now - EditMode and PoseMode now have 'active' bone too (last clicked) - Parenting in EditMode' CTRL+P, ALT+P, with nice options! - Pose is added in Outliner now, with showing that constraints are in the Pose, not Armature - Disconnected IK solving from constraints. It's a separate phase now, on top of the full Pose calculations - Pose itself has a dependency graph too, so evaluation order is lag free. TODO NOW; - Rotating in Posemode has incorrect inverse transform (Martin will fix) - Python Bone/Armature/Pose API disabled... needs full recode too (wait for my doc!) - Game engine will need upgrade too - Depgraph code needs revision, cleanup, can be much faster! (But, compliments for Jean-Luc, it works like a charm!) - IK changed, it now doesnt use previous position to advance to next position anymore. That system looks nice (no flips) but is not well suited for NLA and background render. TODO LATER; We now can do loadsa new nifty features as well; like: - Kill PoseMode (can be option for armatures itself) - Make B-Bones (Bezier, Bspline, like for spines) - Move all silly button level edit to 3d window (like CTRL+I = add IK) - Much better & informative drawing - Fix action/nla editors - Put all ipos in Actions (object, mesh key, lamp color) - Add hooks - Null bones - Much more advanced constraints... Bugfixes; - OGL render (view3d header) had wrong first frame on anim render - Ipo 'recording' mode had wrong playback speed - Vertex-key mode now sticks to show 'active key', until frame change -Ton-
2005-07-03 17:35:38 +00:00
unsigned int lay; // for flushing redraw/rebuild events
const char *name;
struct DagAdjList *next;
} DagAdjList;
typedef struct DagNode {
int color;
short type;
float x, y, k;
void *ob;
void *first_ancestor;
Result of 2 weeks of quiet coding work in Greece :) Aim was to get a total refresh of the animation system. This is needed because; - we need to upgrade it with 21st century features - current code is spaghetti/hack combo, and hides good design - it should become lag-free with using dependency graphs A full log, with complete code API/structure/design explanation will follow, that's a load of work... so here below the list with hot changes; - The entire object update system (matrices, geometry) is now centralized. Calls to where_is_object and makeDispList are forbidden, instead we tag objects 'changed' and let the depgraph code sort it out - Removed all old "Ika" code - Depgraph is aware of all relationships, including meta balls, constraints, bevelcurve, and so on. - Made depgraph aware of relation types and layers, to do smart flushing of 'changed' events. Nothing gets calculated too often! - Transform uses depgraph to detect changes - On frame-advance, depgraph flushes animated changes Armatures; Almost all armature related code has been fully built from scratch. It now reveils the original design much better, with a very clean implementation, lag free without even calculating each Bone more than once. Result is quite a speedup yes! Important to note is; 1) Armature is data containing the 'rest position' 2) Pose is the changes of rest position, and always on object level. That way more Objects can use same Pose. Also constraints are in Pose 3) Actions only contain the Ipos to change values in Poses. - Bones draw unrotated now - Drawing bones speedup enormously (10-20 times) - Bone selecting in EditMode, selection state is saved for PoseMode, and vice-versa - Undo in editmode - Bone renaming does vertexgroups, constraints, posechannels, actions, for all users of Armature in entire file - Added Bone renaming in NKey panel - Nkey PoseMode shows eulers now - EditMode and PoseMode now have 'active' bone too (last clicked) - Parenting in EditMode' CTRL+P, ALT+P, with nice options! - Pose is added in Outliner now, with showing that constraints are in the Pose, not Armature - Disconnected IK solving from constraints. It's a separate phase now, on top of the full Pose calculations - Pose itself has a dependency graph too, so evaluation order is lag free. TODO NOW; - Rotating in Posemode has incorrect inverse transform (Martin will fix) - Python Bone/Armature/Pose API disabled... needs full recode too (wait for my doc!) - Game engine will need upgrade too - Depgraph code needs revision, cleanup, can be much faster! (But, compliments for Jean-Luc, it works like a charm!) - IK changed, it now doesnt use previous position to advance to next position anymore. That system looks nice (no flips) but is not well suited for NLA and background render. TODO LATER; We now can do loadsa new nifty features as well; like: - Kill PoseMode (can be option for armatures itself) - Make B-Bones (Bezier, Bspline, like for spines) - Move all silly button level edit to 3d window (like CTRL+I = add IK) - Much better & informative drawing - Fix action/nla editors - Put all ipos in Actions (object, mesh key, lamp color) - Add hooks - Null bones - Much more advanced constraints... Bugfixes; - OGL render (view3d header) had wrong first frame on anim render - Ipo 'recording' mode had wrong playback speed - Vertex-key mode now sticks to show 'active key', until frame change -Ton-
2005-07-03 17:35:38 +00:00
int ancestor_count;
unsigned int lay; /* accumulated layers of its relations + itself */
unsigned int scelay; /* layers due to being in scene */
uint64_t customdata_mask; /* customdata mask */
int lasttime; /* if lasttime != DagForest->time, this node was not evaluated yet for flushing */
int BFS_dist; /* BFS distance */
int DFS_dist; /* DFS distance */
int DFS_dvtm; /* DFS discovery time */
int DFS_fntm; /* DFS Finishing time */
struct DagAdjList *child;
struct DagAdjList *parent;
struct DagNode *next;
Threaded object update and EvaluationContext Summary: Made objects update happening from multiple threads. It is a task-based scheduling system which uses current dependency graph for spawning new tasks. This means threading happens on object level, but the system is flexible enough for higher granularity. Technical details: - Uses task scheduler which was recently committed to trunk (that one which Brecht ported from Cycles). - Added two utility functions to dependency graph: * DAG_threaded_update_begin, which is called to initialize threaded objects update. It will also schedule root DAG node to the queue, hence starting evaluation process. Initialization will calculate how much parents are to be evaluation before current DAG node can be scheduled. This value is used by task threads for faster detecting which nodes might be scheduled. * DAG_threaded_update_handle_node_updated which is called from task thread function when node was fully handled. This function decreases num_pending_parents of node children and schedules children with zero valency. As it might have become clear, task thread receives DAG nodes and decides which callback to call for it. Currently only BKE_object_handle_update is called for object nodes. In the future it'll call node->callback() from Ali's new DAG. - This required adding some workarounds to the render pipeline. Mainly to stop using get_object_dm() from modifiers' apply callback. Such a call was only a workaround for dependency graph glitch when rendering scene with, say, boolean modifiers before displaying this scene. Such change moves workaround from one place to another, so overall hackentropy remains the same. - Added paradigm of EvaluaitonContext. Currently it's more like just a more reliable replacement for G.is_rendering which fails in some circumstances. Future idea of this context is to also store all the local data needed for objects evaluation such as local time, Copy-on-Write data and so. There're two types of EvaluationContext: * Context used for viewport updated and owned by Main. In the future this context might be easily moved to Window or Screen to allo per-window/per-screen local time. * Context used by render engines to evaluate objects for render purposes. Render engine is an owner of this context. This context is passed to all object update routines. Reviewers: brecht, campbellbarton Reviewed By: brecht CC: lukastoenne Differential Revision: https://developer.blender.org/D94
2013-12-26 17:24:42 +06:00
/* Threaded evaluation routines */
uint32_t num_pending_parents; /* number of parents which are not updated yet
* this node has got.
* Used by threaded update for faster detect whether node could be
* updated aready.
*/
bool scheduled;
/* Runtime flags mainly used to determine which extra data is to be evaluated
* during object_handle_update(). Such an extra data is what depends on the
* DAG topology, meaning this flags indicates the data evaluation of which
* depends on the node dependencies.
*/
short eval_flags;
} DagNode;
typedef struct DagNodeQueueElem {
struct DagNode *node;
struct DagNodeQueueElem *next;
} DagNodeQueueElem;
typedef struct DagNodeQueue {
DagNodeQueueElem *first;
DagNodeQueueElem *last;
int count;
int maxlevel;
struct DagNodeQueue *freenodes;
} DagNodeQueue;
2014-01-17 17:35:03 +11:00
/* forest as we may have more than one DAG unconnected */
typedef struct DagForest {
ListBase DagNode;
struct GHash *nodeHash;
int numNodes;
bool is_acyclic;
int time; /* for flushing/tagging, compare with node->lasttime */
bool ugly_hack_sorry; /* prevent type check */
bool need_update;
} DagForest;
// queue operations
DagNodeQueue *queue_create(int slots);
void queue_raz(DagNodeQueue *queue);
void push_queue(DagNodeQueue *queue, DagNode *node);
void push_stack(DagNodeQueue *queue, DagNode *node);
DagNode *pop_queue(DagNodeQueue *queue);
DagNode *get_top_node_queue(DagNodeQueue *queue);
void queue_delete(DagNodeQueue *queue);
// Dag management
DagForest *dag_init(void);
DagForest *build_dag(struct Main *bmain, struct Scene *sce, short mask);
void free_forest(struct DagForest *Dag);
DagNode *dag_find_node(DagForest *forest, void *fob);
DagNode *dag_add_node(DagForest *forest, void *fob);
DagNode *dag_get_node(DagForest *forest, void *fob);
DagNode *dag_get_sub_node(DagForest *forest, void *fob);
void dag_add_relation(DagForest *forest, DagNode *fob1, DagNode *fob2, short rel, const char *name);
Fix depsgraph to compute more accurate links for collision & force. Current implementation more or less indiscriminately links physics objects to colliders and forces, ignoring precise details of layer checks and collider groups. The new depsgraph seemed to lack some such links at all. The relevant code in modifiers suffers from a lot of duplication. Different physics simulations use independent implementations of collision and similar things, which results in a lot of variance: * Cloth collides with objects on same or visible layer with dupli. * Softbody collides with objects on same layer without dupli. * Non-hair particles collide on same layer with dupli. * Smoke uses same code as cloth, but needs different modifier. * Dynamic paint "collides" with brushes on any layer without dupli. Force fields with absorption also imply dependency on colliders: * For most systems, colliders are selected from same layer as field. * For non-hair particles, it uses the same exact set as the particles. As a special quirk, smoke ignores smoke flow force fields; on the other hand dependency on such field implies dependency on the smoke domain. This introduces two utility functions each for old and new depsgraph that are flexible enough to handle all these variations, and uses them to handle particles, cloth, smoke, softbody and dynpaint. One thing to watch out for is that depsgraph code shouldn't rely on any properties that don't cause a graph rebuild when changed. This was violated in the original code that was building force field links, while taking zero field weights into account. This change may cause new dependency cycles in cases where necessary dependencies were missing, but may also remove cycles in situations where unnecessary links were previously created. It's also now possible to solve some cycles by switching to explicit groups, since they are now properly taken into account for dependencies. Differential Revision: https://developer.blender.org/D2141
2016-08-12 18:23:29 +03:00
typedef bool (*DagCollobjFilterFunction)(struct Object *obj, struct ModifierData *md);
void dag_add_collision_relations(DagForest *dag, struct Scene *scene, Object *ob, DagNode *node, struct Group *group, int layer, unsigned int modifier_type, DagCollobjFilterFunction fn, bool dupli, const char *name);
void dag_add_forcefield_relations(DagForest *dag, struct Scene *scene, Object *ob, DagNode *node, struct EffectorWeights *eff, bool add_absorption, int skip_forcefield, const char *name);
void graph_print_queue(DagNodeQueue *nqueue);
void graph_print_queue_dist(DagNodeQueue *nqueue);
void graph_print_adj_list(DagForest *dag);
#endif /* __DEPSGRAPH_PRIVATE_H__ */