Caused by 6d832ee5b3
The bone names were flipped correctly, however, we were missing the
actual FCurve BezTriple flipping.
This was caused by passing the wrong FCurve into
`do_curve_mirror_flippping` -- this function is checking for existence
of the FCurve in the `KeyframeCopyBuffer`, so we gotta pass
`fcurve_in_copy_buffer` (instead of the new fcu).
Pull Request: https://projects.blender.org/blender/blender/pulls/135381
Caused by 40ac21e5a5 [does not remember/resore the previous ScrArea &
ARegion anymore].
Without this, the operator reporting might get confused by using the
wrong area/region, so added back
NOTE: I tried to just notify `NC_SPACE | ND_SPACE_INFO_REPORT` in
`screen_opengl_render_end` (same as in `wm_operator_finished` >
`wm_operator_register`), but to no avail...
Think this is quite good pratice to leave us with the original area/
region anyways though.
Thx @brecht for improvements (restoring in `screen_opengl_render_init`
already, also taking care of restoring in some early out cases)
Co-authored-by: Brecht Van Lommel <brecht@blender.org>
Pull Request: https://projects.blender.org/blender/blender/pulls/135394
For this to work, we have to take into account handles in TransVert as
well.
Behavior is the same as in legacy curves (meaning that if the control
point itself is selected, the handles are ignores).
Part of #133448 (same thing for grease pencil, which I plan to also make
part of TransVert, but better solve for Curves first, so we can share
code here).
Pull Request: https://projects.blender.org/blender/blender/pulls/134945
The compositor backdrop is still visible even after a new completely
different file is loaded. This is because compositor contexts cache
things like the backdrop, and contexts are stored on interactive
compositor renders, which are not freed until Blender is closed. To fix
this, we cleanup the interactive compositor renders that were created
upon file load.
Pull Request: https://projects.blender.org/blender/blender/pulls/135390
`OpenSubdiv_Buffer` is a wrapper that was introduced at the time
that Blender couldn't use CPP directly. It contains a pointer to
a VertBuf and callbacks to use GPU module on that buffer.
This PR replaces OpenSubdiv_Buffer with `blender::gpu::VertBuf` and
removes the wrapper.
NOTE: OpenSubdiv tests are added to blender_test executable to make the
library dependencies not to complicated.
Pull Request: https://projects.blender.org/blender/blender/pulls/135389
Enable VK loader debug messages when running blender_debug_gpu_*
scripts.
Allows easier debugging for users not familiar with using terminal.
It won't log anything when using OpenGL.
Vulkan loading logging can be big (1000 lines) but gives a lot of insight
how vulkan and related applications are configured on the user system
Sadly it does contains actual paths but think it is OK as the user is still
responsible for uploading it in the tracker.
An example of a log has been attached.
See #135327.
Pull Request: https://projects.blender.org/blender/blender/pulls/135399
Add this operator to the dropdown menu to the right of the attribute list
for curves and point cloud geometries. For these geometry types the
operator can run in edit mode too, unlike meshes.
This commits fixes a type error issue when reporting that a addon was
designed for a version of Blender newer than the current one.
Thanks to @Rakete for the fix.
Ref: !135392
This was caused by the normals not being available (default to
`vec3(0)`) which produced `NaN`s down the line after the
unsafe normalize. Changing to `safe_normalize` fixes the issue.
Now ccl_device sets inlining and ccl_device_inline forces inlining.
This matches more closely with what is currently done for cuda and metal
backends.
I've measured from 1% to 6% overall performance improvement in rendering
benchmark scenes on Arc B580, as well as a small decrease in compile
time.
- Avoid redundant computaiton of curve type index masks
- Parallelize index buffer build (always add space for a cyclic segment)
- Avoid overhead of GPU index buffer "add vert" utilities
- Add utility for consistent order of point and handle data
- Make some variable names more consistent
- Avoid mixing multiple abstraction levels in the same function
Added entries for operators missing in the Curves edit mode context and
topbar menus as mentioned in #105253.
Ordered menu entries based on legacy and mesh menus but it's not
identical as the operator set is reduced. Prioritized keeping group
classification consistent between the menus when deriving the layout.
Pull Request: https://projects.blender.org/blender/blender/pulls/134935
threading::parallel_for implements its own check to avoid
threading when the data length is below the grain size, and
its overhead is lower anyway, since it doesn't use a function
call for every element.
Pull Request: https://projects.blender.org/blender/blender/pulls/135316