When making a `headless` macOS build, linking would fail complaining
about missing `GHOST_ContextCGL` symbols, which are required by the Metal
backend. This commit fixes this issue by compiling the macOS CGL GHOST
Context if the Metal backend is enabled, instead of only depending on
`WITH_HEADLESS`, similarly to what's done on Linux/other *nix systems.
Pull Request: https://projects.blender.org/blender/blender/pulls/130847
Joining and Docking tag the source and target areas for redrawing. But
there are times when areas are split and recombined as part of the
processes. In some cases areas that are not either source or target
must also be tagged for redraw. Especially for Outliner and 3DView as
they try to avoid full redraws with RGN_DRAW_NO_REBUILD, but needed
here.
Pull Request: https://projects.blender.org/blender/blender/pulls/130801
#126806 stopped the drawing of horizontal scrollbars when the areas
were very short. However this still left their hit areas still
active. This PR properly removes the scrollbar in this situation by
also changing the v2d.scroll flags.
Pull Request: https://projects.blender.org/blender/blender/pulls/130764
These are useless now that PointerRNA has explicit default values, and
become a problem when real constructors are added to this struct. Simply
use the default empty value initialization instead.
Pull Request: https://projects.blender.org/blender/blender/pulls/130927
When the recent files list is shown on the Splash screen, the
horizontal size is constrained and so longer items will have their
names truncated. This PR adds the file name as the first item in the
popup tooltip.
Pull Request: https://projects.blender.org/blender/blender/pulls/130875
This replaces the existing C API of `BLO_Write_IDBuffer` with a C++ API. This
helps because:
* No need for explicit freeing.
* Can use more generic `DynamicStackBuffer` utility instead of having a separate
implementation of that.
Additionally, the API is changed so that a new `BLO_Write_IDBuffer` is created
for each `ID` instead of reusing the same for multiple IDs. This simplifies the
code quite a bit and allows for better use of the RAII pattern.
I expect the performance to be the same as before. In theory, there could be a
small speedup, because the `BLO_Write_IDBuffer` is not allocated separately
anymore, but that should be negligible.
No functional changes are expected.
Pull Request: https://projects.blender.org/blender/blender/pulls/130452
The converter is an abstraction that takes a base mesh to be subdivided
and provides its topology information to the OpenSubdiv library. It does
this with one level of indirection: first extracting the base mesh
topology with a virtual function call per element to local arrays,
then giving the information in those arrays to OpenSubdiv.
That level of indirection also handles cache invalidation for the
intermediate data structures which optimize repeated subdivisions of a
changing base mesh with constant topology. However, these days the mesh
data is stored with simpler to compare data arrays, and we also have
implicit sharing which provides another way to detect unchanged
shared data.
As a very first step to a design where we use OpenSubdiv more directly
and don't store duplicate topology arrays for the base mesh, this PR
provides the converter with the mesh's face offsets array directly
rather than using function calls. For multires reshape the temporary
format is changed to match.
Next steps will do the same thing for face vertices ("corner verts" in
Blender lingo), edges, and creases. Then we can remove the "converter"
indirection completely, then we can work on a better cache invalidation
strategy using implicit sharing. That's a ways off though. On its own,
this PR should just reduce function call overhead a bit.
Reference #130917.
Pull Request: https://projects.blender.org/blender/blender/pulls/130241
The HIP compiler version was not being detected on first run in a fresh
build folder since HIP requires perl, and perl wasn't installed yet due
to the compiler detection running before the installs into the build
folder.
this change re-orders the compiler detection and environment preparation
so perl is available when we detect the hip compiler version.
Implements #130836:
interpolate_*_wrapmode_fl take InterpWrapMode wrap_u and wrap_v arguments.
U and V coordinate axes can have different wrap modes: clamp/extend,
border/zero, wrap/repeat.
Note that this removes inconsistency where cubic interpolation was
returning zero for samples completely outside the image, but all other
functions were not, and the behavior was not matching the function
documentation either.
Use the new functions in the new compositor CPU backend.
Possible performance impact for other places (e.g. VSE): measured on
4K resolution, transformed (scaled and rotated) 4K EXR image:
- Nearest filter: no change,
- Bilinear filter: no change,
- Cubic BSpline filter: slight performance decrease, IMB_transform
19.5 -> 20.7 ms (Ryzen 5950X, VS2022). Feels acceptable.
Pull Request: https://projects.blender.org/blender/blender/pulls/130893
When resizing window some platforms could freeze. The cause was that
depending on a platform the swap chain could get out of date when
acquiring the next swapchain image or when presenting the swap chain
image to the queue.
We only covered one scenario. This PR implements the scenario to
recreating the swap chain when the image could not be acquired.
Pull Request: https://projects.blender.org/blender/blender/pulls/130909
Ensure no bezier curves for offset modifier, this makes it consistent
with the rest of the grease pencil modifiers for now and we will make
bezier support for the rest of the modifiers later.
Pull Request: https://projects.blender.org/blender/blender/pulls/130901
Prefetching happens on a background thread by design, and so it was not loading
any fonts and using the default monospace font.
Address this by making all font usage within VSE use "unique" BLF font objects,
and protecting concurrent access to them or their state within VSE itself:
- SeqFontMap structure to hold a mutex, file path -> font ID map (for file
based fonts), name -> font ID map (for datablock fonts).
- seq_load_font_file, seq_load_font_mem, seq_unload_font that use the above
instead of calling into BLF directly.
- Text effect rendering part guards with a mutex, so that font render state and
glyph cache does not get trashed.
- SeqFontMap is global instead of some Scene-specific runtime data (e.g.
scene->ed), because right now there's a hard max limit of total number of
fonts (64), and if each copy of the VSE data would start to load their own
unique fonts, that limit could get easily reached.
- BLF font loading/unloading code is now thread safe. The rest of BLF drawing
is still not!
If at some point in the future BLF font usage becomes thread safe, and font
drawing becomes "stateless", this whole machinery can be removed.
Pull Request: https://projects.blender.org/blender/blender/pulls/130542
In Blender 4.3 all the EEVEE Legacy compatibility Python API calls for
materials in were removed. All Python code that makes use of that API
need to be updated to make use of the new API.
This commit updates two built in Python scripts to use the new API
to avoid errors like the one reported in #130822
Candidate for 4.3.1 corrective release
Pull Request: https://projects.blender.org/blender/blender/pulls/130873
This workaround the issue of the sync requiring a view.
This simply stores the rv3d members that are needed for
calling `view_dist_get`.
This is a workaround to avoid any difference in behavior
with legacy overlay while not violating the design principles.
A better fix would be to compute the offset differently
solely based on the `View` itself. But that would change
the display and should be done in another task.
The "is edit mode" checks in the draw module have been simplified
to just check the object mode. We need to be a bit more careful to
avoid retrieving the original mesh from a non-mesh object.
This commit improves the "get pre-modified mesh" function with
better comments and checks that reflect the relatively recent cases
where the original and evaluated geometry data types don't match.
It's always called with a mesh object too.
Pull Request: https://projects.blender.org/blender/blender/pulls/130850
