Other changes include:
- Replace buffer out of bounds check with an assert
(it was only done for the byte).
- Step over pixels on each X span
instead of calculating the offset for each pixel.
- Use const variables.
Paths were checked for being at least 3 characters in length
before being added to FS_CATEGORY_OTHER.
It looks like this was originally done to prevent the preferences
attempting to add "//" paths which is now handled by `BLI_path_is_rel`.
As it happens the old logic prior to [0] was incorrect and asserted
if textures was set to "//textures" for example.
Now only check the path isn't empty.
[0]: 968f156fae
Excluding "/boot" mount point would also exclude directories
beginning with boot, without being a sub-directory.
Resolve by adding a version of STRPREFIX that delimits on null
terminator or slash.
Note that moving to the templated versions of array allocation functions
in [0] caused integer overflow regressions because multiplying by the
`sizeof(type)` caused the following values to be cast to a size_t.
[0]: e85ebb24fe
For Text Objects make sure we don't render a glyph for tab or other
control characters. Return nothing for carriage return and line feed,
and treat the others as a space character.
Pull Request: https://projects.blender.org/blender/blender/pulls/137374
When Timeline (and similar editors) is reduced to a just the height of
two headers you can still change the play head. Except if you have any
markers and then marker selection operator interferes. This PR sets a
minimum area height for marker operations. This minimum differs if you
hide the header.
Pull Request: https://projects.blender.org/blender/blender/pulls/137312
Previously, the BVH bounds were updated with a top-down approach
starting at the root. As explained in #136471, algorithms that visit
every node of the tree can become a bottleneck when sculpting high-res
meshes.
This patch refactors `flush_bounds_to_parents` by using a bottom-up
approach starting from the dirty leaf nodes. This requires calculating
and storing the parents of the nodes during the BVH build.
The change reduces the overhead measured in #136471 to near zero. The
same tests give the following results:
| Radius | Node Ratio | Speedup |
|--------------|---------|------------|
| 0.10m | 297 | 1.25x |
| 0.25m | 297 | 1.23x |
| 0.50m | 551 | 1.18x |
| 1.00m | 1930 | 1.13x |
where node ratio is the ratio between total nodes and nodes affected by
the brush. For reference, typical ratios when sculpting something like
a character are very roughly:
- tens when sculpting main forms
- low hundreds when sculpting secondary forms
- high hundreds to thousands when sculpting tertiary details
Pull Request: https://projects.blender.org/blender/blender/pulls/137019
Import the initial set of rest transforms as the Armature's rest pose,
even if there's no animation for the bones. The rest transforms are
already being written out during export.
Pull Request: https://projects.blender.org/blender/blender/pulls/135079
* Switches `Sculpt` for `Brush` since the former was just used to access
the latter.
* Simplifies the code to a single ternary.
* Renames and moves the method into the ed::sculpt_paint namespace
Pull Request: https://projects.blender.org/blender/blender/pulls/137368
For various purposes we traverse the computation done by a node tree (e.g. for
gizmos and socket usage infeferencing). For that we generally have to keep track
of the compute context we're in at any given time. During the traversal, it's
common to enter and exist the same compute contexts multiple times. Currently,
we'd always build a new compute context when that happens. That happens even
though the old one is generally still around, because other data may reference
it. This patch implements a `ComputeContextHash` type that avoids rebuilding the
same compute contexts over and over again.
I'm considering to also replace the usage of `ComputeContextBuilder` with this
cache somehow, but will see how that works out.
The reason I'm working on this now is that I have to traverse the node tree a
bit again to find where closures might be evaluated. I wanted to be able to
cache the compute contexts for a while already.
Pull Request: https://projects.blender.org/blender/blender/pulls/137360
The issue is the following:
- The stabilized normal for the main pass is calculated in
`calc_brush_plane`
- The normal is (wrongly) recalculated in `update_sculpt_normal`. This
is a common problem that other brushes have. The specific issue with the
Plane brush is that the recomputed normal is not stabilized.
- For symmetry passes, this non-stabilized normal is then assigned to
the brush in `calc_brush_plane`
This PR fixes the issue by explicitly returning false in
`sculpt_brush_needs_normal` when the Plane brush is used. A more proper
refactor would be needed, but this can easily be ported to 4.4.
Pull Request: https://projects.blender.org/blender/blender/pulls/137271
Use `blender::Stack` instead of `BLI_Stack`. Theoretically this can
improve performance because it makes use of inlining to remove function
call overhead and give the compiler more information. I observed a build
time improvement of about 4%, from 1156 to 1113 ms with 30k cube objects.
Notably, now `deg_graph_flush_visibility_flags` doesn't call any
non-inlined functions.
Pull Request: https://projects.blender.org/blender/blender/pulls/137355
This commit renames the `types.hh` file to `brushes.hh` as an attempt to
make brush implementations more independent and begin to unwind and
remove much of the logic in `sculpt.cc`.
As a first step, this moves the node mask calculation exposed for the
plane brush into `plane.cc` and exposes the function header into
`types.hh`.
Pull Request: https://projects.blender.org/blender/blender/pulls/137020
They are actually already some literals with the `f` suffix
that are in our shader codebase and we never had problem in
the past 5 years (or even 8 years).
So I think it is safe to do and improves convergence of codestyles.
Pull Request: https://projects.blender.org/blender/blender/pulls/137352
The goal is to log information about which closures are evaluated where. This
information is not exposed in the UI yet, but will be needed to be able to debug
the evaluation and inspect socket values within closures.
Pull Request: https://projects.blender.org/blender/blender/pulls/137351
I observed allocation becoming a bottleneck when building the depsgraph
with scenes with many simple data-blocks. One of the main culprits was
the struct that encodes relations between nodes in the graph.
Instead of allocating each `Relation` with a separate allocation call,
combine them into a `LinearAllocator`. That is must faster because it
allocates in large chunks and just bumps an offset on each allocation.
In a test file with 30 thousand cube objects, I observe a 1.18x
improvement in depsgraph evaluation time, from 1370 to 1163 ms.
The depsgraph isn't completely re-allocated when it's rebuilt, so the
allocator added in this PR has to be cleared manually. In the future,
it can be used for other structs or potentially strings.
Pull Request: https://projects.blender.org/blender/blender/pulls/137303
A couple of memory leak fixes for the vulkan backend.
We increment the submission_id on render_graphs upon reset. This
triggers cleanup of anything tracked as a VKResourceTracker. Notably
uniform buffers created for push constant fallbacks. This fixes a memory
leak that was accumulating VKUniformBuffers every frame without cleaning
them up.
Reset resource pools when a swapchain image is presented. This ends up
calling vkResetDescriptorPool, freeing up descriptor set resources. This
fixes a memory leak that was accumulate descriptor sets and pools over
time without freeing them.
Pull Request: https://projects.blender.org/blender/blender/pulls/137305
This will be needed to determine if there are volumes in the scene, before
allocation passes to aid volume sampling.
The kernels are now also loaded in the middle of scene update, at a place
where kernel features are known but before the kernels are needed for
displacement and background light evaluation..
Updating the camera to final resolution for progressive refinement still
happens later.
Pull Request: https://projects.blender.org/blender/blender/pulls/137228
On certain platforms the 3d cursor wasn't visible. The reason was that
the shader read garbage in `gpu_attr_1_fetch_unorm8` and interpreted the
data to be packed, resulting in loading incorrect colors.
Pull Request: https://projects.blender.org/blender/blender/pulls/137336
Point count is missing in object and sculpt mode for Curves object.
Added an extra if block that handles this case. Added a new variable
`totcurvepoints` to track curve points count. Otherwise it is aggregated with
other points count (eg. mesh) which result in wrong number in status bar.
Pull Request: https://projects.blender.org/blender/blender/pulls/137199
Found this during #137299
Python error observed when calling batch rename operator on sequencer
strips. `change_effect_input` operator doesn't have any enum property.
Remove operator_enum to fix the error.
Pull Request: https://projects.blender.org/blender/blender/pulls/137331
The crash was caused by an overflow in the opgl_path_segment_storage
array. It happened because shadow catcher paths would write to the
segments but not clear them. This made it so the next render loop
iteration for the main path starts with non-empty segments in the
guiding data.
Disable training when megakernel is called for the shadow catcher
state.
To ensure this issue is not forgotten when the guiding is ported to
GPU add asserts in the `guiding.h`. While it is a no-op for default
GPU kernels sometimes we do compile debug kernels. But also it acts
as a plain-text reminder to the future-us in working on the code.
There is now also an assert before the main path megakernel to help
catching such cases in the future.
Pull Request: https://projects.blender.org/blender/blender/pulls/137291
As mentioned in #132028, search box is intentionally skipped for lib
override hierarchy view mode of outliner. But search filtering from
previous mode still builds the tree with searched elements. Added a new
function that checks the outliner view mode. If it's lib
override-hierarchy, skip search filtering.
See images attached in PR description.
Pull Request: https://projects.blender.org/blender/blender/pulls/137086
