Add EvalStencils and EvalPatches API for most of CPU and GPU evaluators.
with this change, Eval API in the osd layer consists of following parts:
- Evaluators (Cpu, Omp, Tbb, Cuda, CL, GLXFB, GLCompute, D3D11Compute)
implements EvalStencils and EvalPatches(*). Both supports derivatives
(not fully implemented though)
- Interop vertex buffer classes (optional, same as before)
Note that these classes are not necessary to use Evaluators.
All evaluators have EvalStencils/Patches which take device-specific
buffer objects. For example, GLXFBEvaluator can take GLuint directly
for both stencil tables and input primvars. Although using these
interop classes makes it easy to integrate osd into relatively
simple applications.
- device-dependent StencilTable and PatchTable (optional)
These are also optional, but can be used simply a substitute of
Far::StencilTable and Far::PatchTable for osd evaluators.
- PatchArray, PatchCoord, PatchParam
They are tiny structs used for GPU based patch evaluation.
(*) TODO and known issues:
- CLEvaluator and D3D11Evaluator's EvalPatches() have not been implemented.
- GPU Gregory patch evaluation has not been implemented in EvalPatches().
- CudaEvaluator::EvalPatches() is very unstable.
- All patch evaluation kernels have not been well optimized.
- Currently GLXFB kernel doesn't support derivative evaluation.
There's a technical difficulty for the multi-stream output.
* Added stencilTable.cpp
* Fixed the "off" variable shadow warning
* Moved constructors to cpp file
* Made LimitStencilTable constructor private
* other minor clean up.
- created new class Far::PrimvarRefiner with interpolation methods
- removed interpolation and limit methods from Far::TopologyRefiner
- replaced internal usage in Far::StencilTableFactory
- replaced usage in regressions, tutorials and examples
This is a new implementation of the stencil table construction algorithm found
in protoStencil.h. In local tests with production assets, the new algorithm is
~25% faster and significantly more stable, in terms of average performance
In one asset test, generating stencils for level 10 adaptive refinement of
BuzzLightyear was reduced from 18s to 13s.
- removed all friend declarations (no more Far declarations with Vtr)
- all protected methods made public
- intent is to move these into namespace internal
- access between Vtr classes is under review
- it takes number and pointer for the input PatchCoords.
- add derivative evaluations.
- enhance glEvalLimit example to see the derivative evaluation works.
Cleaned up the Legacy Gregory shader source by accessing buffer
data through helper functions.
Switched to performing tessellation in untransformed (object) space.
this change fixes a crash when selecting catmark_hole_test1
in glStencilViewer. If there's a hole limit stencils may not
be found for given limit location, resulting invalid stencil
entries.
- changed Vtr::LocalIndex to 16-bit integer from 8-bit
- added test shapes including valence 360 vertices
- disabled new shapes in far/regression until improved accuracy accepted
PatchTables no longer needs to friend EndCapLegacyGregoryPatchFactory.
Instead we now make the patch table factory pass in the data that needs
to be updated directly to EndCapLegacyGregoryPatchFactory.
In osd layer, we use GLPatchTable (D3D11PatchTable) as a
device-specific representation of FarPatchTables instead of
DrawContext. GLPatchTable may be used not only for drawing
but also for GPU eval APIs (not yet supported though.
We may add CudaPatchTable etc as needed).
The legacy gregory patch drawing buffers are carved out to
the separate class, named GLLegacyGregoryPatchTable.
Also face-varying data are split into client side for now, until
we add new and more robust face-varying drawing structure
(scheduled at 3.1 release)
Tentatively replicate PatchArray structure in GLPatchTables. It will
be revised in the upcoming change.
Shifting hard-coded SRV locations of legacy gregory buffers in HLSL shaders.
- changes completely deprecate AddWithWeight(T, float, float, float)
- added new EvaluateBasis() method to PatchTables
- replaced usage of old Evaluate...<T,U>() methods with EvaluateBasis()
- removed old Evaluate...<T,U>() methods
- removed now unused Interpolate...<T,U>() functions in far/interpolate.h
- moved low-level basis code from far/interpolate.* to patchBasis.*
hlslPatchGregoryBasis.hlsl is an equivalent to glslPatchGregoryBasis.
Update dxViewer to be able to switch among bspline, gregorybasis, legacy
end capping.
also fixes a bug of GLSL legacy gregory shader which had an inconsistent
resource naming with example codes.
It looks like there's still an issue of D3D11 patchParam data fetching.
we'll come back to that bug.
While this may be worth revisiting, we should first quantify the benefits and
identify the compilers that support it. Ultimately, we may never use pragma
once in favor of strictly using standard C++.
- added new class to far/topologyLevel.h
- updated TopologyRefiner to manage set of TopologyLevels internally
- added TopologyRefiner method to retrieve TopologyLevel
- redefined obsolete TopologyRefiner methods in terms of TopologyLevel
As a preparation for retiring DrawContext, move SupportsAdaptiveTessellation
method to examples/common/glUtils, which is renamed and namespaced
from gl_common.{cpp,h} to be consistent to other files.
Same renamings applied to other example files.
Remove DrawRegistry from osd layer and put a simple shader caching
utility into examples/common. osd layer only provides patch shader
snippet and let client configure and compile the code. Clients also
maintain the lifetime of shader object, which is preferable for the
actual application integration.
update all examples to use the new scheme.
These are now redundant since all bspline patches are encoded in
the patch tables consistently using 16 point indices with boundary
and corner edges indicated in the boundary mask of the patch params.
My earlier change which simplified the categorization of
patch types broke evaluation for boundary and corner patches.
Previously, boundary and corner patches were always rotated
into a canoncial orientation by permuting the point indices
of the patch. This was convenient in some cases, but generally
made things unecessarily complicated, since the parameterization
of the patch had to be counter-rotated to compensate.
Now patches always remain correctly oriented with respect
to the underlying surface topology and evaluation of boundary
and corner patches is accommodated by simply adjusting the
spline weights to account for the missing/invalid patch
points along boundary and corner edges.
There is more to clean up and optimize, but this restores
correct behavior.
Since unified shading work already removed subPatch info from
Osd::PatchDescriptor, the difference between Far::PatchDescriptor and
Osd::PatchDescriptor is just maxValence and numElements. They are used
for legacy gregory patch drawing.
Both maxValence and numElements are actually constant within a topology
(drawContext). This change move maxValence to DrawContext and let client
manage numElements, then we can eliminate Osd::PatchDescriptor and simply
use Far::PatchDescritor instead.
This is still an intermediate step toward further DrawRegistry refactoring.
For the time being, adding EffectDesc struct to include maxValence and
numValence to be maintained by the clients. They will be cleaned up later.
The side benefit of this change is we no longer need to recompile regular b-spline
shaders for the different max-valences.
- added Far::GetGregoryWeights() to work directly with 20 weights
- simplified tensor-product evaluation of Bezier and BSpline (will more
readily support higher order derivatives in near future)
- fixed Bezier derivative scaling issue (off by factor of 3.0)
* noted incorrectness of Gregory derivatives (correction will accompany
support for higher order derivatives in near future)
- Remove MeshPtexData bit from Osd::MeshBits. It's not used any more
- Rename ptexIndexBuffer in D3D11DrawContext to paramParamBuffer
- Remove Is/SetPtexEnabled from D3D11DrawRegistry
In OpenSubdiv 2.x, we encapsulated subdivision tables into
compute context in osd layer since those tables are order-dependent
and have to be applied in a certain manner. In 3.0, we adopted stencil
table based refinement. It's more simple and such an encapsulation is
no longer needed. Also 2.0 API has several ownership issues of GPU
kernel caching, and forces unnecessary instantiation of controllers
even though the cpu kernels typically don't need instances unlike GPU ones.
This change completely revisit osd client facing APIs. All contexts and
controllers were replaced with device-specific tables and evaluators.
While we can still use consistent API across various device backends,
unnecessary complexities have been removed. For example, cpu evaluator
is just a set of static functions and also there's no need to replicate
FarStencilTables to ComputeContext.
Also the new API delegates the ownership of compiled GPU kernels
to clients, for the better management of resources especially in multiple
GPU environment.
In addition to integrating ComputeController and EvalStencilController into
a single function Evaluator::EvalStencils(), EvalLimit API is also added
into Evaluator. This is working but still in progress, and we'll make a followup
change for the complete implementation.
-some naming convention changes:
GLSLTransformFeedback to GLXFBEvaluator
GLSLCompute to GLComputeEvaluator
-move LimitLocation struct into examples/glEvalLimit.
We're still discussing patch evaluation interface. Basically we'd like
to tease all ptex-specific parametrization out of far/osd layer.
TODO:
-implments EvalPatches() in the right way
-derivative evaluation API is still interim.
-VertexBufferDescriptor needs a better API to advance its location
-synchronization mechanism is not ideal (too global).
-OsdMesh class is hacky. need to fix it.
Changing all device kernels to take two buffer identifiers for
source and destination separately.
This change is an intermediate step toward upcoming context/controller
refactoring.
Previously we have a limitation that the source and destination
vertex buffer has to be a single buffer, since the subdivision
kernels are iteratively applied by level.
With stencil tables, we don't have such a limitation any more,
so we may want to apply stencils from seprate source buffer to
another.
To specifiy the output location within the destination buffer,
we can use VertexBufferDescriptor.offset. This allows us not only
configuring arbitrary batching scheme, but also relaxing the
limitation that source and destination buffers are in same
interleaved layout. For examples, we could include derivatives only
in the destination buffer, which doesn't need to be allocated in
the source buffer.
we're teasing out ptex specific data from core osd entities,
so there's no reason to keep ptex texturing utilities in core osd.
move them into example libs and let clients assemble shader snippets
as needed.
Also removing older ptex texturing code (without mipmap)
Each patch has a corresponding patchParam. This is a set of three values
specifying additional information about the patch:
faceId -- topological face identifier (e.g. Ptex FaceId)
bitfield -- refinement-level, non-quad, boundary, transition, uv-offset
sharpness -- crease sharpness for a single-crease patch
These are stored in OsdPatchParamBuffer indexed by the value returned
from OsdGetPatchIndex() which is a function of the current PrimitiveID
along with an optional client provided offset.
Accessors are provided to extract values from a patchParam. These are
all named OsdGetPatch*().
While drawing patches, the patchParam is condensed into a patchCoord which
has four values (u, v, faceLevel, faceId). These patchCoords are treated
as int values during per-prim processing but are converted to float values
during per-vertex processing where the values are interpolated.
Also, cleaned up more of the shader namespace by giving an Osd prefix
to public functions, and consolidated boundary and transition handling
code into the PatchCommon shader files. The functions determining
tessellation levels are now all named OsdGetTessLevel*().
- resolves DX-CL interop functions in Osd::ClD3D11VertexBuffer.
- enable CL kernels in DX build.
- more cleanup in test harnesses, adding D3D11 initializations into DeviceContext.
- add new defines OPENSUBDIV_HAS_OPENGL and OPENSUBDIV_HAS_DX for convenience.
- removed default value for its <SIZE> parameter
- updated all usage to specify a value for <SIZE>
- added explicit element destruction missing from destructor
- corrected comment regarding VLA's being non-standard
refactor CL/CUDA specific initialization stuffs into
examples/common/clDeviceContext and cudaDeviceContext, and
update examples to use those structs.
also
- remove CL/CUDA tests from osd_regression. The tests for those kernels will be covered by glImaging.
- update cuda initialization to use the GL-interoperable device if available.
- remove CL specialization from glShareTopology, following the same pattern as we took in the previous OsdGLMesh refactoring. (still something strange with XFB kernels though)
- fix file permissions.
The previous change to the gathering of patch points went
a bit too far. Near non-manifold features it is important
to be careful when traversing the faces in a level to avoid
assumptions that are valid only for manifold topology.
Also, removed Vtr::Level::gatherQuadRegularPatchPoints().
This method was added in my previous change, but it is
unsafe to use in the presence of non-manifold topology.
Removed OpenCL/D3D11 specialization and add DEVICE_CONTEXT as a template
parameter. For the kernels which don't need a context object (e.g.
CPU, OpenGL, cuda) just ignore the context, and for the kernels which
use a context (e.g. OpenCL, DirectX) takes a context or a user-defined
class as which encapsulates device contexts. Note that OpenCL requires
two objects, cl_context and cl_command_queue. The user-defined
class must provide GetContext() and GetCommandQueue() for strongly typed
binding to osd VertexBuffers and ComputeContexts.
Osd::Mesh and MeshInterface have been used as a handy harness to host
multiple GPU kernels and graphics APIs. However it has CL/DirectX
specializations and duplicates large amount of plubming code. With this
change, glMesh.h and d3d11Mesh.h become just typedefs and all logic is
put into mesh.h without specializations.
Also cleaned up unused header files and code formatting.
Now a boundary and corner patch remains
aligned with its underlying parametric
orientation. This simplifies both the
gathering of patch vertices and downstream
evaluation of patches.
Added a method to Vtr::Level which gathers
the 16 patch points for a B-spline patch
even if the patch has boundary or corner
edges. The undefined patch vertex index
values along boundary and corner edges are
assigned Vtr::INDEX_INVALID.
In order to simplifiy the process of drawing
B-spline patches with boundary or corner
edges, the Far::PatchTablesFactory will
replace any invalid vertex indices with
a known good value, i.e. the index of the
first patch face vertex.
Single-crease patches are still a slightly
special case, which will be resolved later.
- rename "Regular end cap" to "BSplineBasis end cap"
- revert templating and add EndCapType into PatchTablesFactory::Options.
- make EndCapFactories internal in PatchTablesFactory.
- move end cap stencils into PatchTables, keep them relative to the max level.
- add a utility StencilTablesFactory::AppendEndCapStencilTables to splice and factorize endcap stencil tables.
Remove the ptex-specific code from the Far::TopologyRefiner and instead provide it in a separate class Far::PtexIndices. Clients who need to use the Ptex API need to first build a Far::PtexIndices object by providing it with a refiner.
This has the advantage of keeping the API on the TopologyRefiner a little cleaner. The ptex methods were const but were mutating state with const_casts. The new mechanism still achieves the same lazy initialization behavior by forcing clients to instantiate them exactly when needed.
A disadvantage of this approach is that the PatchTablesFactory creates its own PtexIndices and throws it out after the patch tables are created. This is great if you're never going to need the ptex indices again, but not so great if you will need them again.
computes edge lengths using limit surface points. Made this
the default screen-space metric so that we avoid cracks when
using Gregory Basis or Regular B-spline end caps.
The alternative method which computes edge lengths using the
distance between B-spline control points is still available.
Added a diagram and comments to explain how the control
points and limit points are organized.
This change moves all gregory patch generation from Far::PatchTablesFactory
so that we can construct patch tables without stencil tables as well as client
can chose any end patch strategies (we have 3 options for now: legacy 2.x style
gregory patch, gregory basis patch and experimental regular patch approximation).
Also Far::EndCapGregoryBasisPatchFactory provides index mapping from patch index
to vtr face index, which can be used for single gregory patch evaluation on top
of refined points, without involving heavier stencil tables generation.
- modified main LimitMask queries to include subdivision Rule
- split internal mask assignments into corner, crease and smooth cases
- adapted all three schemes to the new structure
- completed limit tangents for Loop
- update TopologyRefiner Limit methods to pass newly required Rule
This is the first step to tease off Osd compute controller/contexts
from Far API.
Currently FarStencilTable only creates a kernelbatch for the entire range,
so we can use [0, numStencils) for all cases instead of KernelBatch.
This might not be true if we apply non-factorized level-wise stencils,
then we'll add another modular utility to serve those cases.
PatchTablesFactory fills 20 indices topology into patchtable, and use it for eval and draw.
note: currently screen-space adaptive tessellation of gregory basis patches is
broken and cracks appear around them.
- new Options for Refine() methods for base face and vertex ordering
- removed ignored/unused "full topology" choice from AdaptiveOptions
- added base face and vertex ordering logic to Refinement
- addition of TopologyRefiner members for component counts and max valence
- refactoring of Level additions to update all new member totals
- addition of GetMaxValence() to TopologyRefiner
- updated PatchTablesFactory to user new GetMaxValence() method
- renaming of "Hole" methods for TopologyRefiner and Vtr::Level
- updated original FVar inspection to deal witn non-manifold vertices
- updated FVar refinement to handle > 2 faces per edge
- couple of fixes to use parent topology when child not present
level before we do any further population from the parent as that may
further modify the child's _maxEdgeFaces. When _maxEdgeFaces is incorrect
we may end up with stack corruption problems when
Far::TopologyRefiner::interpolateChildVertsFromEdges() goes to use this data
to allocate some stack space.
- regular patches now generated for faces with non-manifold vertices
- non-manifold features treated as boundary edges and vertices
- other minor cleanup/improvement in patch identification method
- commented on approx of smooth x-ordinary corner with regular sharp patch
- added edge-face local-indices to Vtr::Level members
- added internal and public access to edge-face local-indices
- updated refinement methods to use and populate edge-face local-indices
- updated refinement methods to be sensitive to degenerate edges
- added Vtr::VALENCE_LIMIT constant for internal use
- updated Vtr::Level topology completion to fail on valence overflow
- simplfied non-manifold edge handling in Vtr::Level topology completion
- updated warnings for all topology construction failures
- added detection and warning when passing non-triangles for Loop
- changed return type of Vtr array empty() to bool
- renamed and adjusted arguments to Vtr gather methods for consistency
- removed lone Vtr gather method that took an explicit offset
- updated usage of revised Vtr gather methods in Far
- fixed warnings redeclaring names of local variables already in use
- extend Far::PatchTables data structures & interfaces to store requisite
information for channels of face-varying bi-cubic patches
- implement gather function in Far::PatchTablesFactory to populate face-varying
channels with adaptive patches
- extend accessor interface in Vtr::Level
- propagate code fall-out throughout OpenSubdiv code base, examples & tutorials
- extend vtrViewer code to visualize tessellated bi-cubic face-varying patches
- added new tag for sharp corner, now used in isolation
- split single semi-sharp tag in two for vertex and edge sharpness
- updated tag propagation and all tests of previous semi-sharp tag
- added clear() methods for all tags and simplified initialization
- added Vtr::FVarLevel method to resolve correct index in level 0 case
- updated tag retrieval methods that used face-value array as input
- changed signatures to accept face-verts when needed and updated usage
- added methods to FVarLevel to combine FVar tags with vertex tags
- use composite of combined tag to find FVar features needing isolation
- added more tagging related to the "linear borders" interpolation case
