Blender Git Commits

USD import: orientation bug.

Uninitialized USDMeshReader::m_isLeftHanded flag was
causing the vertex winding orientation to randomly flip,
causing incorrect shading and other issues.

Sculpt Expand: Force expand to work only on active connected components

Fix crash in versioning code

Sculpt Expand: Fix keymap entry for snapping

Scupt: Fix failed scene projection with ortographic views

USD import: handle primvars:normals if specified.

Per Pixar UsdGeomPointBased documentation: If 'normals' and
'primvars:normals' are both specified, the latter has precedence.

Sculpt: Fix wrong vertex normals in Scene Project

USD import: Handle empty vertex-varying normals.

If point normals are empty, now invoking BKE_mesh_calc_normals()
to avoid black renders. Since BKE_mesh_calc_normals() requires
edges to be defined, moving point normal calculation to follow
reading polys.

EEVEE: Cleanup: Remove use of macros for the loops

EEVEE: Fix occlusion applied on SSR

Merge branch 'master' into xr-actions-D9124

XR: Use common functions for modal_3d operators

Reduces the amount of similar code for XR invoke_3d/modal_3d
operators that manipulate view3d params.

Add TODO for merging into master

Fix Cycles CUDA kernels for Ampere not building with CUDA 11

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.

Cycles: Add CUDA 11 build support

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

EEVEE: Add back support for user occlusion on eevee_specular BSDF

Merge branch 'master' into temp-gpencil-fill

Fix Cycles CUDA kernels for Ampere not building with CUDA 11

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.

Cycles: hack make update to build this temporary branch with 2.83 libraries

Cycles: Add CUDA 11 build support

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