Add a `.data<T>()` method that retrieves a mutable span. This is useful
more and more as we change to filling in vertex buffer data arrays
directly, and compared to raw pointers it's safer too because of asserts
in debug builds.
Pull Request: https://projects.blender.org/blender/blender/pulls/123338
This implement the holdout flag by switching to
the holdout case in the shader. This has a few benefits:
- Doesn't recompile the shaders.
- Makes the object infos mandatory (already the case in
practice)
- Handle transparent materials properly, keeping the
transparency working.
Fix#123284
Pull Request: https://projects.blender.org/blender/blender/pulls/123315
This implement the holdout flag by switching to
the holdout case in the shader. This has a few benefits:
- Doesn't recompile the shaders.
- Makes the object infos mandatory (already the case in
practice)
- Handle transparent materials properly, keeping the
transparency working.
Fix#123284
Pull Request: https://projects.blender.org/blender/blender/pulls/123315
When using particle system the generated coordinates where not set.
The reason is that a loose resource handle was created for particle
system that didn't contain enough information to calculate orcos.
This PR will create the object info from its parent object.
Pull Request: https://projects.blender.org/blender/blender/pulls/123307
I didn't think the BMesh extraction mode could arrive at the case where
only loose edges are requested, but turns out it can because of the
mesh wrapper system where the evaluated mesh is actually a BMesh with
deformed positions.
The mapping was read multiple times from the `map` structure in that time the `map` value could have changed (set to -1). To stop this, the mapping is read once and set to -1 earlier to reduce the possibility to other threads performing the operation and changing the mapping value.
Pull Request: https://projects.blender.org/blender/blender/pulls/123056
Not a real fix, just preventing crashes due to broken particle data.
Particle draw code can get into an invalid state, which can crash
drawing code. This just prevents the crash.
Pull Request: https://projects.blender.org/blender/blender/pulls/123075
When navigating the samples were always reset. This was introduced
by 7ec59b05ff where samples needed
to be reset when painting.
This PR solves it by separating the navigation and the painting more
clearly in the API. Also cleans up some calls that are also encapsulated
via the EEVEE Instance class.
Validated that painting and navigating still worked with these changes
applied.
Pull Request: https://projects.blender.org/blender/blender/pulls/123064
This should not happen and any failure here should be considered a bug.
But for end users better not to hang Blender, and to have a better
diagnostic for developers in bug reports.
Ref #82483
Pull Request: https://projects.blender.org/blender/blender/pulls/123023
Everything here is drawn as triangles in the end, and currently we use
index buffers either way, so "needs_tri_index" isn't a meaningful name.
What this really controls is whether the layout is flattened to duplicate
each subdiv vertex 4 times, for drawing flat shading.
Retrieve the index buffer data as a span and set it directly rather than
using the API functions to add vertices which add overhead. Also pass
all the necessary data as arguments instead of retrieving them from
the "args" struct on every use.
In a test with a 16 million vertex grid, this reduces the runtime of
generating all the wireframe index buffers by about 25%, from 95 to 76 ms
for that step of the drawing process.
Most functions in `draw_pbvh.cc` are inside of the `PBVHBatch` or
`PBVHBatches` classes, besides the more recently added attribute
copying code. This means they have access to "all the information"
and it's hard to get a sense for the proper order of calculation and
what each function is responsible for. This commit is an attempt to
clarify things a bit by changing to regular static functions.
With two viewports, and one viewport with overlays turned off, one viewport
first requests positions without calculating loose geometry, then the second
viewport can request the loose geometry index buffer. In that case the
previously-calculated positions VBO has the wrong size.
To fix this, always calculate loose geometry when the positions are requested.
That's required because we no simple way of checking whether the previously
uploaded positions included loose geometry. That was the state before recent
refactors in this area anyway.
This makes it clear that the optimization to not calculate loose geometry is
meaningless, because the positions buffer will always be requested at some
point when drawing a mesh.
Part of #116901.
This file contains the creation of four different UV edit index buffers:
triangles, lines, points, and face dots. The main changes are about making
it clearer when faces are skipped. Other than that, changes are faily
extensive but straightfoward. In a few places some parts of existing
loops have been split out and parallelized, but the bulk of the work for
each IBO remains largely single threaded.
Part of #116901.
Includes a general standardization of the code, removing constant
checks from hot loops, standardizing naming, tweaking BMesh face
iteration. And includes parallelization too.
Part of #116901.
To avoid resizing the buffer, the data is added to a temporary vector
first. That should be fine because the number of roots is expected to
be quite small compared to the size of the mesh.
