Commit Graph

331 Commits

Author SHA1 Message Date
manuelk
9158a430ac Fix Far::PatchTables control vertex indexing to use 4 CVs to represent the 0-ring
- we were allocating 20 CVs for the Gregory basis, but only need 4
 (the 20 CVs are stored in the stencil table)
2014-11-12 13:20:55 -08:00
manuelk
e8210886a8 Extend stencil-driven Gregory basis code support to Osd::EvalLimit (cpu) 2014-11-11 17:35:52 -08:00
manuelk
82a53c4278 Fix bug in cpuEvalLimitKernel.cpp
apparently an offset to a basis vertex was incorrectly set
2014-10-31 17:06:34 -07:00
David G Yu
1e00056d8a Improved stability on OSX drivers by not declaring unused shader in/out vars. 2014-10-23 15:59:05 -07:00
David G. Yu
b11d5e22f0 Fixed d3d11 and gl mesh wrappers to be
consistent wrt the MeshInterface API.
2014-10-23 13:45:57 -07:00
David G. Yu
6cc24d32fd Fixed missing namespace closing brace. 2014-10-23 13:43:45 -07:00
David G. Yu
fce3dcca00 Fixed vec array initializer. 2014-10-23 13:41:27 -07:00
manuelk
f4c2f8d9e1 Fix Windows build of Osd OMP CPU kernel
note: we #include <vector> instead malloc.h for portability
2014-10-23 12:07:12 -07:00
manuelk
f145f70fd2 Refactor Far::StencilTablesFactory
- re-implement the pool allocator
    - use templates to remove code redundancy between regular & limit stencils
    - leverage [] operator overloading to simplify stencil factorization
    - add the ability to treat subdivision levels independently (see below)

- refactor Far::TopologyRefiner::Interpolate<>() methods to pass buffers by reference
  (allows overloading of [] operator)

- rename some of the stencil factory options
- propagate changes to Osd / examples / tutorials...
2014-10-21 16:36:26 -07:00
manuelk
a16d558a3e Fix GLSL compute code version defintion
- remove #version declaration from the kernel code
- move it in front of shader sources before compiling to prevent some drivers from throwing errors

fixes #360
2014-10-20 18:09:30 -07:00
Takahito Tejima
c0907c7bc1 Introduces 'single-crease patch' : implements "Efficient Evaluation of Semi-Smooth Creases in
Catmull-Clark Subdivision Surfaces", Niessner et al, Eurographics 2012.

This change includes;
-topology identification for single-crease patch during adaptive refinement.
-patch array population (similar to boundary)
-sharpness buffer generation
-glsl shader

Eval stuffs will be coming.
2014-10-13 08:52:09 -07:00
manuelk
cea1dd1880 Change FarStencilTables indices from 'unsigned int' to Far::Index type 2014-10-09 16:36:23 -07:00
manuelk
a79832b3a1 Far::PatchTables re-factor & clean-up
- remove Descriptor iterators
- switch 'unsigned int' indices to typed Far::Index

We aren't done yet... but its a step in the right direction
2014-10-09 14:48:50 -07:00
manuelk
639788abbe Fix vec array initialization in GLSL & HLSL (should fix some DX compiler problems)
Fixes #354
2014-10-04 16:22:43 -07:00
manuelk
d3fa91deaa Fix incorrect assert in Osd::DrawContext 2014-10-03 10:35:00 -07:00
manuelk
82e7a1bcc5 Minor EvalLimit naming refactor
- also added some additional animation to the glEvalLimit code example
2014-09-22 16:01:43 -07:00
manuelk
ad402d3623 Fix inconsistent file access permissions 2014-09-17 13:45:50 -07:00
Nathan Litke
78ae830794 Fixed Osd::Mesh constructors that initialize member variables from user variables. 2014-09-17 13:41:45 -07:00
Manuel Kraemer
8fc5371ec9 Merge pull request #350 from nathan-at-digitalfish/build_android
Fix build for Android
2014-09-12 17:27:57 -07:00
manuelk
c66ee724a8 WIP limit stencils factory
todo:
- fix rotated tangents on patches
- add "end-caps" (bilinear or Gregory ?)
- add pure bilinear stencil mode
2014-09-12 16:59:16 -07:00
manuelk
e538261f7d Fix Cuda compile warning in osd/cudaKernel.cu 2014-09-12 10:43:18 -07:00
Julien Demouth
29b2d033e3 Introduces a kernel for NUM_ELEMENTS=4. It's faster and not buggy,
anymore.
2014-09-12 10:22:30 -07:00
Julien Demouth
cf0fe60ab4 Add a first pass of optimization in CUDA kernels
Conflicts:
	opensubdiv/osd/cudaKernel.cu
2014-09-12 10:21:46 -07:00
Nathan Litke
d7311801c1 Fix compiler errors for OpenGL-ES. 2014-09-11 16:30:50 -07:00
Takahito Tejima
b3f6f52bb9 Fixes the bug of glShareTopology. This example mutates patchDescriptor in drawcontext, so it needs a write access.
Adding another GetPatchArray() returning non-const reference. We should think about better API for this.
2014-09-10 15:24:16 -07:00
manuelk
c399655dcc Landing 3.0.0.alpha
Sync'ing the 'dev' branch with the 'feature_3.0dev' branch at commit 68c6d11fc36761ae1a5e6cdc3457be16f2e9704a

The branch 'feature_3.0dev' is now locked and preserved for historical purposes.
2014-09-05 15:07:46 -07:00
Nathan Litke
0128417447 Allow NEON to be disabled on Android. Works similar to GCD on Mac OS X. 2014-08-07 13:10:54 -07:00
Rachid El Guerrab
035e58f44a Added subdivision kernels for ARM NEON
* assembler kernels are based on the C implementation in neonKernel.cpp
* enable assembler kernel functions in neonComputeController.cpp with #define USE_ASM_KERNELS 1
2014-08-05 22:09:12 -07:00
Manuel Kraemer
df719eb9ed Revert "added memcpyasync and streams to cuda backend"
This reverts commit 84212cd725.
2014-08-02 12:40:08 -07:00
Frost
84212cd725 added memcpyasync and streams to cuda backend 2014-07-24 10:18:09 -04:00
Sergey Sharybin
1ebf2449dd Fix mismatch in declaration and usage of OsdCudaComputeRestrictedVertexA
Unused argument `pass` was defined in the CUDA kernel and it was never
passed to this function from the C++ code. This argument is also wasn't
used by the function itself.
2014-07-03 16:54:15 +06:00
Sergey Sharybin
7ac06db165 Fix crashes when using rather low-end cards like Intel ones
Solved by checking on run-time whether texture buffer objects
are supported.

When building with GLEW library doing compile-time check is
not enough, because actual information about existing features
is only known on runtime.

This only makes ti so CPU backend works, GLSL backends still
requires some work if we want them to make working. Not sure
it worth doing this now.
2014-07-02 22:17:58 +06:00
Nathan Litke
e4d7f3d495 Fixed omissions in osd/mesh.h.
* added `OsdMeshInterface::GetFarMesh` and `OsdMesh::GetFarMesh` to match `OsdGLMesh` and `OsdD3D11Mesh`
* added `interleaved` argument to `OsdMesh::Refine` to match `OsdMeshInterface::Refine`
2014-06-27 02:15:27 -07:00
Nathan Litke
299966fd5c Fixed an error in createCLBuffer that occurs when the buffer size is zero. 2014-06-24 12:14:41 -07:00
Nathan Litke
b7a763853c Added the CATMARK_RESTRICTED_VERT_VERTEX_A, CATMARK_RESTRICTED_VERT_VERTEX_B1, and CATMARK_RESTRICTED_VERT_VERTEX_B2 kernels which compute vertices resulting from the refinement of a smooth or (fully) sharp vertex.
* CATMARK_RESTRICTED_VERT_VERTEX_A handles k_Crease and k_Corner rules
* CATMARK_RESTRICTED_VERT_VERTEX_B1 handles regular k_Smooth and k_Dart rules
* CATMARK_RESTRICTED_VERT_VERTEX_B2 handles irregular k_Smooth and k_Dart rules
2014-06-23 15:59:43 -07:00
Nathan Litke
634187f6c5 Fixed a bug in the CUDA kernel. 2014-06-23 14:49:10 -07:00
Nathan Litke
a6e179367b Fixed an error in the DirectCompute shader. This resolves https://github.com/PixarAnimationStudios/OpenSubdiv/issues/313. 2014-06-09 19:43:02 -07:00
Nathan Litke
94c2a4293c Fixed a compiler error in tbbKernel.cpp. Addressed comments from https://github.com/PixarAnimationStudios/OpenSubdiv/pull/310. 2014-06-03 14:49:30 -07:00
Nathan Litke
0af14f8ac6 Added the CATMARK_RESTRICTED_EDGE_VERTEX kernel which computes vertices resulting from the refinement of a smooth or (fully) sharp edge. 2014-05-30 17:44:25 -07:00
Manuel Kraemer
626921d82e Merge pull request #308 from nathan-at-digitalfish/new_face_vertex_kernels
New face vertex kernels
2014-05-30 10:56:12 -07:00
Nathan Litke
24b4aea686 Fixed a compiler error in the GLSL Transform Feedback kernels on OS X/Mavericks. 2014-05-30 00:55:25 -07:00
Nathan Litke
29b51d80af Implemented CATMARK_QUAD_FACE_VERTEX and CATMARK_TRI_QUAD_FACE_VERTEX kernels for DirectX 11, OpenMP, and TBB platforms. Fixed a bug in the GLSL Transform Feedback kernels. Minor changes for consistent formatting. 2014-05-29 23:20:04 -07:00
Nathan Litke
b769f4f60d Implemented CATMARK_QUAD_FACE_VERTEX and CATMARK_TRI_QUAD_FACE_VERTEX kernels for CUDA, GLSL Transform Feedback, and GLSL Compute platforms. 2014-05-28 17:21:37 -07:00
manuelk
de57f43db1 Fix doxygen generation errors 2014-05-28 16:19:49 -07:00
Nathan Litke
064115bbe8 Added the CATMARK_QUAD_FACE_VERTEX and CATMARK_TRI_QUAD_FACE_VERTEX kernels.
* The CATMARK_QUAD_FACE_VERTEX kernel calculates the face-vertex for a quadrilateral face.  It applies to every face after the first subdivision step, and may be applied for the first subdivision step of a quadrilateral coarse mesh.

* The CATMARK_TRI_QUAD_FACE_VERTEX kernel calculates the face-vertex for a triangle or quadrilateral face.  It may be applied for the first subdivision step of a coarse mesh composed of triangles and/or quadrilaterals.

* Both kernels calculate each face-vertex using four vertex indices (triangles are specified by repeating the third index).  Therefore neither kernel uses the F_ITa codex table, and instead the first vertex offset in the F_IT index table is stored in the FarKernelBatch's table offset.
2014-05-28 13:52:12 -07:00
Nathan Litke
95aca6ed21 Minor changes to fix typographical errors, and for consistency across class method prototypes. 2014-05-28 13:40:51 -07:00
Takahito Tejima
7f2b65ba46 Reordering includes to address a compile error on mac/glew environment 2014-05-23 13:23:24 -07:00
Sergey Sharybin
6ef232c95a Integrate CLEW into osd library and examples
If the system has CLEW installed (which is detected by recently
added FindCLEW routines) then OpenSubduv would be compiled against
this library.

It makes binaries and libraries more portable across the systems,
so it's possible to run the same binary on systems with and without
OpenCL SDK installed.

The most annoying part of the change is updating examples to load
OpenCL libraries, but ideally code around controllers and interface
creation is to be de-duplicated anyway.

Based on the pull request #303 from Martijn Berger
2014-05-22 09:10:59 +02:00
manuelk
d7afc82133 Fix compiler warnings in OsdClVertexBuffer 2014-05-16 14:52:43 -07:00
manuelk
b74f45f68d Decrease compiler warning thresholds and fix outstanding warnings (continued)
- turn off some of icc's remarks (mostly because of tbb)
- fix many of icc -w3 remarks (more to fix once i can work around icc 14.0 linker barfing)
2014-05-15 18:03:44 -07:00