| Age | Commit message (Collapse) | Author |
|---|
|
|
|
File is replaced by pipeline_config.yaml.
|
|
Failure to return a list of the expected size & type wasn't
decrementing the value, leaking a reference.
Caused by 127b5423d6203d521acb2b96b7de5534e8dbe79a a workaround for the
real error that was fixed f5e020a7a6ad6451fcaf075ae14f7014b8a4faea.
|
|
This fixes a performance regression on Ampere cards, on specific scenes like
classroom. For cycles-x there is little difference, but this is still helpful
for LTS releases, and we need to upgrade at some point anyway.
|
|
|
|
|
|
|
|
|
|
|
|
The old code was added in {rBbaa4a9c7d4dd}
While the hotfix committed in {rB2ddecfffc3d3a3a1db4ae45e8665ca} fixed
the id_management test, the correct reason was found at
https://bugs.llvm.org/show_bug.cgi?id=47984
This matches the behavior on Linux.
|
|
Regression in d25747ee751096de2e417a7da1316bf5bf81c25a
|
|
Also resolve a crash when when displaying thumbnails, see T89868.
|
|
Scaling down images could create images with a width or height of zero.
Clamp at 1 to prevent a crash, also add an assert to scaling functions.
Ref D11956
|
|
AMD Drivers didn't report an additional space in the renderer. This made
testing for the HQ workaround fail and the issue appeared back on
certain cards.
This fix will test with surrounding spaces or if the renderer name
endswith the given string. If any of these are the case the hq normals
workaround will be enabled.
Original patch {2262d6c45adf}.
|
|
`bm_face_len_cmp` incorrectly interpreted its arguments as `BMFace *`
instead of `BMFace **`, causing an out-of-bounds read.
Ref D11637
|
|
Pre computed normals index wasn't properly aligned.
Regression from 2ec00ea0c1be1ace7cd0c7b68e43cc8e87dd07c7.
|
|
|
|
|
|
Apparently `textureSize` doesn't work with
`sampler1DArray` on this OS.
Thanks to @dave1853 for finding the source of the
problem.
|
|
Due to the way we ship the CRT on windows TBB's
malloc proxy was unable to attach it self to
the memory management functions on windows 10.
This change moves ucrtbase.dll out of the blender.crt
folder and back into the main blender folder to side
step some undesirable behaviour on win10 making TBB
once more able to attach it self.
Having this work again, should give a speed
boost in memory allocation heavy workloads
such as mantaflow.
For details on how this only failed on Win10
see T88813
|
|
Error in 87cafe92ce2f99d8da620b80e1c26f8078554f93
|
|
|
|
Enabled HQ normals workaround for this specific configuration.
|
|
The id-property iterator referenced a PyObject pointer without
increasing it's user count - allowing for errors if the value
goes out of scope during iteration.
|
|
Looping over IDProperty keys or calling iteritems leaked memory.
Error in original Python idproperty support from
8768707610fbc1cea2bde069cdfd6d3f3e2fc522
|
|
This needs to be backported to 2.83.
|
|
expected
Mistake in {rBe48c4d73d378}.
Was using the vertex index as a lookup for the loop color (instead of
the loop index).
(Issue was not present in original D1429 btw).
Maniphest Tasks: T88145
Differential Revision: https://developer.blender.org/D11212
|
|
|
|
Equivalent of D11323 for 2.83. Make `CMAKE_OSX_DEPLOYMENT_TARGET`
independent of buildbot settings and always set to 10.11.
That fixes the launch error on OS older than buildbot's.
Maniphest Tasks: T88419
Differential Revision: https://developer.blender.org/D11328
|
|
2.83.14 missed cuda/optix kernels in distribution. Better to bump the
version.
|
|
|
|
|
|
|
|
Differential Revision: https://developer.blender.org/D11115
|
|
Logic for bisect handled edges in the face crossing the plane,
but not concave N-gons containing multiple edges that lie on the plane.
|
|
Bisect-plane calculation needed to take non-uniform scale into account.
|
|
Remove custom properties from `__dir__` method result since these
can't be accessed using `__getattr__`.
Introduced in the 2.5x Python API update.
|
|
|
|
As Python can access COW ID's, ensure it's instance is kept on update.
This could happen when the "Use Self" argument was enabled for a driver.
|
|
|
|
|
|
|
|
Something in this update broke the floor() function in CUDA, instead use
floorf() like we do everywhere else in the kernel code. Thanks to Ray
Molenkamp for identifying the solution.
|
|
Action constraint without an action pointer set was crashing.
|
|
In 2.81 there was a change to increase the performance of viewport
animation rendering. This change would perform the color management on the
GPU if the only 8bit was needed. This saved CPU cycles and data
transfer.
The issue is that in the image editor or when saving the image the CM
will be reapplied. Although the speed is desired, exporting the actual
colors has more priority.
In the ticket there is an analysis that shows that shows that this fix
is the correct short term step to take. It would be better that the
render result is aware of the color space of its buffers so the applying
color management could be skipped when saving to disk or drawing in the
image editor.
The issue with this change is the performance penalty it has.
Reviewed By: Brecht van Lommel
Maniphest Tasks: T79999
Differential Revision: https://developer.blender.org/D10371
|
|
|
|
|
|
With this patch the build system checks whether the "CUDA10_NVCC_EXECUTABLE" CMake
variable is set and if so will use that to build sm_30 kernels. Similarily for sm_8x kernels it
checks "CUDA11_NVCC_EXECUTABLE". All other kernels are built using the default CUDA
toolkit. This makes it possible to use either the CUDA 10 or CUDA 11 toolkit by default and
only selectively use the other for the kernels where its a hard requirement.
Reviewed By: brecht
Differential Revision: https://developer.blender.org/D9179
|
|
Running Blender on Ampere cards was already possible with ptx, this fix is
needed to support building CUDA binaries.
Note the CUDA version used for official Blender builds is still 10, this is
merely the change to make it possible for those using CUDA 11 and specifying
the sm_8x kernels to be compiled.
Found by Milan Jaros.
|
|
|
