Implement the next phases of bounds improvement design #96968.
Mainly the following changes:
Don't use `Object.runtime.bb` for performance caching volume bounds.
This is redundant with the cache in most geometry data-block types.
Instead, this becomes `Object.runtime.bounds_eval`, and is only used
where it's actually needed: syncing the bounds from the evaluated
geometry in the active depsgraph to the original object.
Remove all redundant functions to access geometry bounds with an
Object argument. These make the whole design confusing, since they
access geometry bounds at an object level.
Use `std::optional<Bounds<float3>>` to pass and store bounds instead
of an allocated `BoundBox` struct. This uses less space, avoids
small heap allocations, and generally simplifies code, since we
usually only want the min and max anyway.
After this, to avoid performance regressions, we should also cache
bounds in volumes, and maybe the legacy curve and GP data types
(though it might not be worth the effort for those legacy types).
Pull Request: https://projects.blender.org/blender/blender/pulls/114933
Move object runtime data to a separate header and allocate it separately
as `blender::bke::ObjectRuntime`. This is how node, mesh, curves, and
point cloud runtime data is stored.
Benefits:
- Allow using C++ types in object runtime data
- Reduce space required for Object struct in files
- Increase conceptual separation between DNA and runtime data
- Remove the need to add manual padding in runtime data
- Include runtime struct definition only in files that require it
Pull Request: https://projects.blender.org/blender/blender/pulls/113957
Translation markers `N_()` were added to IDs' plural names in !113912.
I also renamed some of those because I thought there were only display
names.
This commit reverts the renaming, leaving only the addition of the
markers.
Pull Request: https://projects.blender.org/blender/blender/pulls/114030
Extract:
- Sculpt filter types from the Sculpt menu. Some of these types use a
custom label, different from those defined in the operator RNA,
which was never extracted.
- "Today" and "Yesterday" from the file browser modification date.
- All name_plural from IDs, as these are used in the UI to list which
data block is to be removed, when calling outliner.orphans_purge.
Disambiguate:
- "Area", meaning the measurement of a surface as opposed to a place.
Some messages reported by Satoshi Yamasaki in #43295.
Pull Request: https://projects.blender.org/blender/blender/pulls/113912
Currently object bounds (`object.runtime.bb`) are lazily initialized
when accessed. This access happens from arbitrary threads, and
is unprotected by a mutex. This can cause access to stale data at
best, and crashes at worst. Eager calculation is meant to keep this
working, but it's fragile.
Since e8f4010611, geometry bounds are cached in the geometry
itself, which makes this object-level cache redundant. So, it's clearer
to build the `BoundBox` from those cached bounds and return it by
value, without interacting with the object's cached bounding box.
The code change is is mostly a move from `const BoundBox *` to
`std::optional<BoundBox>`. This is only one step of a larger change
described in #96968. Followup steps would include switching to
a simpler and smaller `Bounds` type, removing redundant object-
level access, and eventually removing `object.runtime.bb`.
Access of bounds from the object for mesh, curves, and point cloud
objects should now be thread-safe. Other object types still lazily
initialize the object `BoundBox` cache since they don't have
a data-level cache.
Pull Request: https://projects.blender.org/blender/blender/pulls/113465
The `lib_link` callback cannot always be fully replaced/removed, as in
some case it is also doing some validation checks, or data editing based
on the result of lib_linking internal ID pointers.
The callback has been renamed for that purpose, from `read_lib` to
`read_after_liblink`. It is now called after all ID pointers have been
fully lib-linked for the current ID, but still before the call to
`do_versions_after_linking`.
This change should not have any behavioral effect. Although in theory
the side-effect of this commit (to split lib linking itself, and the
validation/further processing code) into two completely separated steps
could have some effects, in practice none are expected, and tests did
not show any changes in behavior either..
Part of implementing #105134: Removal of readfile's lib_link & expand code.
The `expand` callback is 'trivial' to replace, since it is only iterating
over ID pointers and calling a callback.
The only change in behavior here is that some pointers that were not
processed previously will now be.
In practice this is not expected to have any real effect (usually
the IDs used by these pointers would have been expanded through other
usages anyway). But it may solve a few corner cases, undocumented issues
though.
Part of implementing #105134: Removal of readfile's lib_link & expand code.
Listing the "Blender Foundation" as copyright holder implied the Blender
Foundation holds copyright to files which may include work from many
developers.
While keeping copyright on headers makes sense for isolated libraries,
Blender's own code may be refactored or moved between files in a way
that makes the per file copyright holders less meaningful.
Copyright references to the "Blender Foundation" have been replaced with
"Blender Authors", with the exception of `./extern/` since these this
contains libraries which are more isolated, any changed to license
headers there can be handled on a case-by-case basis.
Some directories in `./intern/` have also been excluded:
- `./intern/cycles/` it's own `AUTHORS` file is planned.
- `./intern/opensubdiv/`.
An "AUTHORS" file has been added, using the chromium projects authors
file as a template.
Design task: #110784
Ref !110783.
There is no reason at all for each ID read/write callbacks to have to
deal with this generic, common data explicitely. Generic ID read/write
code is the place to handle such data (just like asset, liboverride,
etc. data are handled already).
Note behavioral change expected here.
Remove the "_for_read" suffix from methods to get geometry and geometry
components. That should be considered the default, so the suffix just
adds unnecessary text. This is consistent with the attribute API and
various implicit sharing data access methods.
Use "from_mesh" instead of "create_with_mesh". This is consistent with
the recently used naming for the `IndexMask` API.
Pull Request: https://projects.blender.org/blender/blender/pulls/110738
More consistently return geometry bounds with the `Bounds` type that
holds the min and max in one variable. This simplifies some code and
reduces the need to initialize separate min and max variables first.
Meshes now use the same `bounds_min_max()` function as curves and
point clouds, though the wrapper mesh isn't affected yet.
The motivation is to make some of the changes for #96968 simpler.
Move `GeometrySet` and `GeometryComponent` and subclasses
to the `blender::bke` namespace. This wasn't done earlier since
these were one of the first C++ classes used throughout Blender,
but now it is common.
Also remove the now-unnecessary C-header, since all users of
the geometry set header are now in C++.
Pull Request: https://projects.blender.org/blender/blender/pulls/109020
A lot of files were missing copyright field in the header and
the Blender Foundation contributed to them in a sense of bug
fixing and general maintenance.
This change makes it explicit that those files are at least
partially copyrighted by the Blender Foundation.
Note that this does not make it so the Blender Foundation is
the only holder of the copyright in those files, and developers
who do not have a signed contract with the foundation still
hold the copyright as well.
Another aspect of this change is using SPDX format for the
header. We already used it for the license specification,
and now we state it for the copyright as well, following the
FAQ:
https://reuse.software/faq/
The usage of the lib pointer was cryptic to say the least, it was
essentialy used to allow or not local IDs results in the mapping
old_id_pointer -> new_id_pointer lookup result.
Now:
- Explicitely pass a boolean to allow or not local ID as result in
lower-level code.
- Pass the 'self ID' pointer instead of its library to the whole
liblinking code (that was already the case in a few places).
Note that naming of the 'self id' pointer is currently very inconsistent
throughout the readfile liblink code, this will have to be cleaned up in
a separate step later. For now, `self_id` has been chosen for new code
as it matches the terminology in lib_query code.
The later change can also allow finer handling of lookup on undo, based
on how it was re-read (or not), should the need for this arise.
Implicit sharing means attribute ownership is shared between geometry
data-blocks, and the sharing happens automatically. So it's unnecessary
to choose whether to enable it when copying a mesh.
Add the ability to retrieve implicit sharing info directly from the
C++ attribute API, which simplifies memory usage and performance
optimizations making use of it. This commit uses the additions to
the API to avoid copies in a few places:
- The "rest_position" attribute in the mesh modifier stack
- Instance on Points node
- Instances to points node
- Mesh to points node
- Points to vertices node
Many files are affected because in order to include the new information
in the API's returned data, I had to switch a bunch of types from
`VArray` to `AttributeReader`. This generally makes sense anyway, since
it allows retrieving the domain, which wasn't possible before in some
cases. I overloaded the `*` deference operator for some syntactic sugar
to avoid the (very ugly) `.varray` that would be necessary otherwise.
Pull Request: https://projects.blender.org/blender/blender/pulls/107059
Similar to e35f971da1. We aren't meant to rely on the
zero-initiatialization on creation anyway. I observed a small (a few
percent) decrease in minimum runtime in the geometry nodes points node.
The position attribute has special meaning for point clouds, and
meshes and curves have access methods for the attribute as well.
This saves boilerplate and gives more consistency between types.
This integrates the new implicit-sharing system (from fbcddfcd68)
with `CustomData`. Now the potentially long arrays referenced by custom
data layers can be shared between different systems but most importantly
between different geometries. This makes e.g. copying a mesh much cheaper
because none of the attributes has to be copied. Only when an attribute
is modified does it have to be copied.
Also see the original design task: #95845.
This reduces memory and improves performance by avoiding unnecessary
data copies. For example, the used memory after loading a highly
subdivided mesh is reduced from 2.4GB to 1.79GB. This is about 25%
less which is the expected amount because in `main` there are 4 copies
of the data:
1. The original data which is allocated when the file is loaded.
2. The copy for the depsgraph allocated during depsgraph evaluation.
3. The copy for the undo system allocated when the first undo step is
created right after loading the file.
4. GPU buffers allocated for drawing.
This patch only gets rid of copy number 2 for the depsgraph. In theory
the other copies can be removed as part of follow up PRs as well though.
-----
The patch has three main components:
* Slightly modified `CustomData` API to make it work better with implicit
sharing:
* `CD_REFERENCE` and `CD_DUPLICATE` have been removed because they are
meaningless when implicit-sharing is used.
* `CD_ASSIGN` has been removed as well because it's not an allocation
type anyway. The functionality of using existing arrays as custom
data layers has not been removed though.
* This can still be done with `CustomData_add_layer_with_data` which
also has a new argument that allows passing in information about
whether the array is shared.
* `CD_FLAG_NOFREE` has been removed because it's no longer necessary. It
only existed because of `CD_REFERENCE`.
* `CustomData_copy` and `CustomData_merge` have been split up into a
functions that do copy the actual attribute values and those that do
not. The latter functions now have the `_layout` suffix
(e.g. `CustomData_copy_layout`).
* Changes in `customdata.cc` to make it actually use implicit-sharing.
* Changes in various other files to adapt to the changes in `BKE_customdata.h`.
Pull Request: https://projects.blender.org/blender/blender/pulls/106228
This simplifies the usage of the API and is preparation for #104478.
The `CustomData_add_layer` and `CustomData_add_layer_named` now have corresponding
`*_with_data` functions that should be used when creating the layer from existing data.
Pull Request: https://projects.blender.org/blender/blender/pulls/105708
The goal is to give technical artists the ability to optimize modifier usage
and/or geometry node groups for performance. In the long term, it
would be useful if Blender could provide its own UI to display profiling
information to users. However, right now, there are too many open
design questions making it infeasible to tackle this in the short term.
This commit uses a simpler approach: Instead of adding new ui for
profiling data, it exposes the execution-time of modifiers in the Python
API. This allows technical artists to access the information and to build
their own UI to display the relevant information. In the long term this
will hopefully also help us to integrate a native ui for this in Blender
by observing how users use this information.
Note: The execution time of a modifier highly depends on what other
things the CPU is doing at the same time. For example, in many more
complex files, many objects and therefore modifiers are evaluated at
the same time by multiple threads which makes the measurement
much less reliable. For best results, make sure that only one object
is evaluated at a time (e.g. by changing it in isolation) and that no
other process on the system keeps the CPU busy.
As shown below, the execution time has to be accessed on the
evaluated object, not the original object.
```lang=python
import bpy
depsgraph = bpy.context.view_layer.depsgraph
ob = bpy.context.active_object
ob_eval = ob.evaluated_get(depsgraph)
modifier_eval = ob_eval.modifiers[0]
print(modifier_eval.execution_time, "s")
```
Differential Revision: https://developer.blender.org/D17185
Use a consistent style for declaring the names of struct members
in their declarations. Note that this convention was already used in
many places but not everywhere.
Remove spaces around the text (matching commented arguments) with
the advantage that the the spell checking utility skips these terms.
Making it possible to extract & validate these comments automatically.
Also use struct names for `bAnimChannelType` & `bConstraintTypeInfo`
which were using brief descriptions.
Bounding box calculation can be a large in some situations, especially
instancing. This patch caches the min and max of the bounding box in
runtime data of meshes, point clouds, and curves, implementing part of
T96968.
Bounds are now calculated lazily-- only after they are tagged dirty.
Also, cached bounds are also shared when copying geometry data-blocks
that have equivalent data. When bounds are calculated on an evaluated
data-block, they are also accessible on the original, and the next
evaluated ID will also share them. A geometry will stop sharing bounds
as soon as its positions (or radii) are changed.
Just caching the bounds gave a 2-3x speedup with thousands of mesh
geometry instances in the viewport. Sharing the bounds can eliminate
recalculations entirely in cases like copying meshes in geometry nodes
or the selection paint brush in curves sculpt mode, which causes a
reevaluation but doesn't change the positions.
**Implementation**
The sharing is achieved with a `shared_ptr` that points to a cache mutex
(from D16419) and the cached bounds data. When geometries are copied,
the bounds are shared by default, and only "un-shared" when the bounds
are tagged dirty.
Point clouds have a new runtime struct to store this data. Functions
for tagging the data dirty are improved for added for point clouds
and improved for curves. A missing tag has also been fixed for mesh
sculpt mode.
**Future**
There are further improvements which can be worked on next
- Apply changes to volume objects and other types where it makes sense
- Continue cleanup changes described in T96968
- Apply shared cache design to more expensive data like triangulation
or normals
Differential Revision: https://developer.blender.org/D16204
Don't add the radius attribute to point clouds by default, since not
having a radius attribute should be a valid state. The radius is only
set when a radius attribute also exists on curves.
This is the conventional way of dealing with unused arguments in C++,
since it works on all compilers.
Regex find and replace: `UNUSED\((\w+)\)` -> `/*$1*/`
This adds a utility function to copy the data from a PointCloud
outside of the main database to one that is in the database. This
is similar to `BKE_mesh_nomain_to_mesh`.
Ref D11592
When resizing mesh and curves attribute storage, avoid initializing the
new memory for basic types. Also, avoid skipping "no free" layers; all
layers should be reallocated to the new size since they may be accessed.
The semantics introduced in 25237d2625 are essential for this
change, because otherwise we don't have a way to construct non-trivial
types in the new memory.
In a basic test of the extrude node, I observed a performance
improvement of about 30%, from 55ms to 42ms.
Differential Revision: https://developer.blender.org/D15818
Replace `mesh_attributes`, `mesh_attributes_for_write` and the point
cloud versions with methods on the `Mesh` and `PointCloud` types.
This makes them friendlier to use and improves readability.
Differential Revision: https://developer.blender.org/D15907
When allocating new `CustomData` layers, often we do redundant
initialization of arrays. For example, it's common that values are
allocated, set to their default value, and then set to some other
value. This is wasteful, and it negates the benefits of optimizations
to the allocator like D15082. There are two reasons for this. The
first is array-of-structs storage that makes it annoying to initialize
values manually, and the second is confusing options in the Custom Data
API. This patch addresses the latter.
The `CustomData` "alloc type" options are rearranged. Now, besides
the options that use existing layers, there are two remaining:
* `CD_SET_DEFAULT` sets the default value.
* Usually zeroes, but for colors this is white (how it was before).
* Should be used when you add the layer but don't set all values.
* `CD_CONSTRUCT` refers to the "default construct" C++ term.
* Only necessary or defined for non-trivial types like vertex groups.
* Doesn't do anything for trivial types like `int` or `float3`.
* Should be used every other time, when all values will be set.
The attribute API's `AttributeInit` types are updated as well.
To update code, replace `CD_CALLOC` with `CD_SET_DEFAULT` and
`CD_DEFAULT` with `CD_CONSTRUCT`. This doesn't cause any functional
changes yet. Follow-up commits will change to avoid initializing
new layers where the correctness is clear.
Differential Revision: https://developer.blender.org/D15617
This mirrors the C++ attribute API better, separates the implementation
of attributes from CustomData slightly, and makes functions simpler,
clearer, and safer.
Also fix an issue with removing an attribute caused by 97712b018d
meant the first attribute with the given type was removed instead of
the attribute with the given name.
