This fixes a threading issue (T73593) between drivers that write to the
same memory address. Driver nodes in the depsgraph now get relations to
each other in order to ensure serialisation.
These relations are only added between drivers that target the same
struct in RNA, which is determined by removing everything after the last
period. For example, a driver with data path
`pose.bones["Arm_L"].rotation_euler[2]` will be grouped with all other
drivers on that datablock with a data path that starts with
`pose.bones["Arm_L"]` to form a 'driver group'.
To find a suitable relation within such a driver group, say the relation
(from → to), a depth-first search is performed (turned out to be
marginally faster than a breadth-first in my test case) to see whether
this will create a cycle, and to see whether there already is such a
connection (direct or transitive). This is done by recursively
inspecting the incoming connections of the 'to' node and thereby walking
from it towards the 'from' node. This is an order of magnitde faster
than inspecting the outgoing connections of the 'from' node.
This approach generalises the special case for array properties, so the
code to support that special case has been removed from
`DepsgraphRelationBuilder::build_animdata_drivers()`.
A test on the Spring rig [1] shows that this process adds approximately
8% to the build time of the dependency graph. In my test case, it takes
28 ms for this process on a total 329 ms construction time. However,
since it also made some code obsolete, it only adds 24 ms (=8%) to the
construction time. I have experimented with a simple cache to keep track
of known-connected (from, to) node pairs, but this did not significantly
improve the timing.
Note that animation data and drivers are already connected by a
relation, which means that animating a field and also changing it with a
driver will not cause conflicts.
[1] https://cloud.blender.org/p/spring/5d30a1076249366fa1939cf1
Differential Revision: https://developer.blender.org/D6905
Reviewed By: sergey, mont29
This partially fixes T73593.
The `add_relation(driver_key, property_entry_key, ...);` call can fail
in the following situation:
- A collection is linked, and instanced into the scene by an Empty.
- The collection contains an object with a driver on its `hide_render` or
`hide_viewport` property.
As the object doesn't exist as a real object in the scene, it's added with
`base_index=-1` to the depsgraph (see `DepsgraphNodeBuilder::build_collection()`).
As a result the node for syncing the restrictflags back to the base
isn't present in the depsgraph, and the `add_relation()` call failed.
This commit fixes the warning, simply by not attempting to add the
offending relation.
Using a struct here allows to change given parameters to the callbacks
without having to edit all callbacks functions, which is always noisy
and time consuming.
This patch fixes {T73051}. The cause of the issue was the absence of
relations in the depsgraph between IK solvers of overlapping IK chains.
Reviewed By: sergey, brecht
Differential Revision: https://developer.blender.org/D6700
This dependency was removed in fd0bc7e002,
as there already were dependencies ANIMATION -> PARAMETERS and
PARAMETERS -> CACHE, making ANIMATION -> CACHE unnecessary.
Upon subsequent inspection, the ANIMATION -> PARAMETERS dependency was
there due to the fallback behaviour in
`RNANodeQuery::construct_node_identifier()`. Now this is no longer
relied upon, and the required relation is made explicit again.
The dependency graph has to know whether a driver must be re-evaluated
every frame due to a dependency on the current frame number. For python
drivers it was using a heuristic based on searching for certain sub-
strings in the expression, notably including '('.
When the expression is actually evaluated using Python, this can't be
easily improved; however if the Simple Expression evaluator is used,
this check can be done precisely by accessing the parsed data.
Differential Revision: https://developer.blender.org/D6624
It was printing "frame" time instead of FPS.
Other interesting thing to get solved is to solve "drop" in FPS
when there is an idle time in user input.
Current usecase is limited to observing FPS when there is a continuous
stream of events: for example, keep moving vertex in edit mode while
watching debug prints.
The FPS here is measured based on a timestamp from when depsgraph
was previously evaluated.
Allows to ease investigating performance improvements/regressions
which are not related on animation system but on modifications on
a single frame (such as transforming vertex in edit mode).
When there are ID properties on an object, and these are animated and
used by a driver, the depsgraph has proper connections between ACTION →
ID PROPERTY → DRIVER.
When these properties are defined on a mesh, however, the depsgraph
relations are incorrectly created between GEOMETRY → PROPERTIES_EXIT →
DRIVER (because it's assumed that 'source = ENTRY' implies 'geometry').
This patch solves this by first checking whether the targeted property
is an ID property and handling it accordingly. This also made it
possible to remove some special cases from pose bone relations.
Maniphest Tasks: T73001
Reviewed By: sergey
Differential Revision: https://developer.blender.org/D6571
FreeStyle line styles were not part of the dependency graph, and
blacklisted from the Copy-on-Write system. As a result, animated
FreeStyle properties would not be updated by the animation system,
resulting in T72213. There was an explicit call to run the animation
system on the original datablocks, but that was (for good reasons)
removed in D5394.
This commit adds the FreeStyleLineStyle datablocks to the dependency
graph and allows them to be handled by the CoW system. As a result
- the UI now updates properly when properties are animated, and
- animated property values are actually used when rendering.
This commit includes @Sergey's patch P1222, which unifies two bits of
code that did the same thing: check whether datablock type is covered by
copy-on-write.
Reviewed By: sergey, brecht
Differential Revision: https://developer.blender.org/D6609
Was caused by recent animation backup added to depsgraph as a part of
another bugfix.
This commit effectively disables the animation backup, restoring bug
related on handlers, but makes it possible to interact with objects
again.
Will re-iterate over handlers+animated properties.
It was possible to have object copy-on-write happening during
action's copy-on-write, which was causing access to a freed
memory from animation backup.
Solves crash reported in T73029.
The issue was caused by special handling of animation update after
manual edits in frame_post handler: to avid loss of manual edits
done on top of animated property. This was done as a separate pass
for non-animation update after frame_post did modifications.
This caused some other side-effect: non-modified animated property
was re-setting to the value which is used in the viewport.
Idea of this solution is simple: preserve values which came from
animation update through copy-on-write process. The actual process
of this is a bit involved: need to decode RNA path and do it twice
since f-curve might point to a sub-data which pointer will change.
Since this is only done for non-active depsgraph (aka depsgraph
used for render pipeline) this is probably fine since all this
extra overhead is just a fraction of overall render process.
Differential Revision: https://developer.blender.org/D6330
This commit restores old metaball workaround which was forcing their
update from a single thread.
The root of the issue comes to the fact that metaball evaluation needs
to access metaballs from duplilists, so they are properly polygonized
with corresponding motherball which is outside of duplilist.
In a more ideal world this will be implemented in a way that will not
require iterating over all duplilists, but only through the ones which
actually contain metaballs for the given motherball. In practice this
ends up in a huge refactor in both relations builder (which meeds to
see whether there are metaballs in duplilists without actually
creating duplilist as it can not be done prior scene is evaluated)
and in metaballs area which need to use new relations information.
Additionally, metaball evaluation must become thread-safe, which is
currently not a case with dupli-object matrices. There might be issues
deeper in polygonization code which I am not aware of.
Having this forced single-thread evaluation is same as Blender 2.79
was doing.
Think it's better to have slower but simpler solution than to invest
time in refactoring area which requires deeper design changes.
Reviewed By: dfelinto
Differential Revision: https://developer.blender.org/D6539
Remove this pointer since it's linking Mesh data back to the object,
where a single edit-mesh may have multiple object users,
causing incorrect assumptions in the code.
Resolves dangling pointer part of the T72667 crash,
although there are other issues which still need to be fixed.
In EDBM_op_finish and EDBM_update_generic,
full Main lookups have been added which should be replaced with mesh
argument or the update tagging moved elsewhere.
Was only using first ID instead of all of them.
Might have been causing issues when updating motion paths of
multiple objects.
Spotted by Jack C, thanks!
Was caused by 6183688c35 (thanks ronsn for nailing it down!).
The issue is that order of copy-on-write operations is not defined, so
can not use flags set by that operation to make decision.
A collection of smaller changes that are required in the /blender/source files. A lot of them are also due to variable renaming.
Reviewed By: sergey
Maniphest Tasks: T59995
Differential Revision: https://developer.blender.org/D3855
The `DEG_depsgraph_query.h` file uses the `ITER_BEGIN` and `ITER_END`
macros defined in `BLI_iterator.h` without including that header.
No functional changes.
It is possible to have action which is not nullptr but which have no
f-curves in it (for example, animate cube's location, then delete all
f-curves).
Such situation should not add time dependency as it could slow down
scene evaluation on frame change.
