v1.6.0

Changelog

[1.6.0] - 2025-02-03

Added

• Add preview of Tile Cholesky factorization and solve APIs through wp.tile_cholesky(), tile_cholesky_solve()
  and tile_diag_add() (preview APIs are subject to change).
• Support for loading tiles from arrays whose shapes are not multiples of the tile dimensions.
  Out-of-bounds reads will be zero-filled and out-of-bounds writes will be skipped.
• Support for higher-dimensional (up to 4D) tile shapes and memory operations.
• Add intersection-free self-contact support in wp.sim.VDBIntegrator by passing handle_self_contact=True.
  See warp/examples/sim/example_cloth_self_contact.py for a usage example.
• Add functions wp.norm_l1(), wp.norm_l2(), wp.norm_huber(), wp.norm_pseudo_huber(), and wp.smooth_normalize()
  for vector types to a new wp.math module.
• wp.sim.SemiImplicitIntegrator and wp.sim.FeatherstoneIntegrator now have an optional friction_smoothing
  constructor argument (defaults to 1.0) that controls softness of the friction norm computation.
• Support assert statements in kernels (docs).
  Assertions can only be triggered in "debug" mode (GH-366).
• Support CUDA IPC on Linux. Call the ipc_handle() method to get an IPC handle for a wp.Event or a wp.array,
  and call wp.from_ipc_handle() or wp.event_from_ipc_handle() in another process to open the handle
  (docs).
• Add per-module option to disable fused floating point operations, use wp.set_module_options({"fuse_fp": False})
  (GH-379).
• Add per-module option to add CUDA-C line information for profiling, use wp.set_module_options({"lineinfo": True}).
• Support operator overloading for wp.struct objects by defining wp.func functions
  (GH-392).
• Add built-in function wp.len() to retrieve the number of elements for vectors, quaternions, matrices, and arrays
  (GH-389).
• Add warp/examples/optim/example_softbody_properties.py as an optimization example for soft-body properties
  (GH-419).
• Add warp/examples/tile/example_tile_walker.py, which reworks the existing example_walker.py
  to use Warp's tile API for matrix multiplication.
• Add warp/examples/tile/example_tile_nbody.py as an example of an N-body simulation using Warp tile primitives.

Changed

• Breaking: Change wp.tile_load() and wp.tile_store() indexing behavior so that indices are now specified in
  terms of array elements instead of tile multiples.
• Breaking: Tile operations now take shape and offset parameters as tuples,
  e.g.: wp.tile_load(array, shape=(m,n), offset=(i,j)).
• Breaking: Change exception types and error messages thrown by tile functions for improved consistency.
• Add an implicit tile synchronization whenever a shared memory tile's data is reinitialized (e.g. in dynamic loops).
  This could result in lower performance.
• wp.Bvh constructor now supports various construction algorithms via the constructor argument, including
  "sah" (Surface Area Heuristics), "median", and "lbvh" (docs)
• Improve the query efficiency of wp.Bvh and wp.Mesh.
• Improve memory consumption, compilation and runtime performance when using in-place vector/matrix assignments in
  kernels that have enable_backward set to False (GH-332).
• Vector/matrix/quaternion component += and -= operations compile and run faster in the backward pass
  (GH-332).
• Name files in the kernel cache according to their directory. Previously, all files began with
  module_codegen (GH-431).
• Avoid recompilation of modules when changing block_dim.
• wp.autograd.gradcheck_tape() now has additional optional arguments reverse_launches and skip_to_launch_index.
• wp.autograd.gradcheck(), wp.autograd.jacobian(), and wp.autograd.jacobian_fd() now also accept
  arbitrary Python functions that have Warp arrays as inputs and outputs.
• update_vbo_transforms kernel launches in the OpenGL renderer are no longer recorded onto the tape.
• Skip emitting backward functions/kernels in the generated C++/CUDA code when enable_backward is set to False.
• Emit deprecation warnings for the use of the owner and length keywords in the wp.array initializer.
• Emit deprecation warnings for the use of wp.mlp(), wp.matmul(), and wp.batched_matmul().
  Use tile primitives instead.

Fixed

• Fix unintended modification of non-Warp arrays during the backward pass (GH-394).
• Fix so that wp.Tape.zero() zeroes gradients passed via the grads parameter in wp.Tape.backward()
  (GH-407).
• Fix errors during graph capture caused by module unloading (GH-401).
• Fix potential memory corruption errors when allocating arrays with strides (GH-404).
• Fix wp.array() not respecting the target dtype and shape when the given data is an another array with a CUDA interface
  (GH-363).
• Negative constants evaluate to compile-time constants (GH-403)
• Fix ImportError exception being thrown during interpreter shutdown on Windows when using the OpenGL renderer
  (GH-412).
• Fix the OpenGL renderer not working when multiple instances exist at the same time (GH-385).
• Fix AttributeError crash in the OpenGL renderer when moving the camera (GH-426).
• Fix the OpenGL renderer not correctly displaying duplicate capsule, cone, and cylinder shapes
  (GH-388).
• Fix the overriding of wp.sim.ModelBuilder default parameters (GH-429).
• Fix indexing of wp.tile_extract() when the block dimension is smaller than the tile size.
• Fix scale and rotation issues with the rock geometry used in the granular collision SDF example
  (GH-409).
• Fix autodiff Jacobian computation in wp.autograd.jacobian() where in some cases gradients were not zeroed-out properly.
• Fix plotting issues in wp.autograd.jacobian_plot().
• Fix the len() operator returning the total size of a matrix instead of its first dimension.
• Fix gradient instability in rigid-body contact handling for wp.sim.SemiImplicitIntegrator and
  wp.sim.FeatherstoneIntegrator (GH-349).
• Fix overload resolution of generic Warp functions with default arguments.
• Fix rendering of arrows with different up_axis, color in OpenGLRenderer (GH-448).

v1.5.1

Changelog

[1.5.1] - 2025-01-02

Added

• Add PyTorch basics and custom operators notebooks to the notebooks directory.
• Update PyTorch interop docs to include section on custom operators
  (docs).

Fixed

• warp.sim: Fix a bug in which the color-balancing algorithm was not updating the colorings.
• Fix custom colors being not being updated when rendering meshes with static topology in OpenGL
  (GH-343).
• Fix wp.launch_tiled() not returning a Launch object when passed record_cmd=True.
• Fix default arguments not being resolved for wp.func when called from Python's runtime
  (GH-386).
• Array overwrite tracking: Fix issue with not marking arrays passed to wp.atomic_add(), wp.atomic_sub(),
  wp.atomic_max(), or wp.atomic_min() as being written to (GH-378).
• Fix for occasional failure to update .meta files into Warp kernel cache on Windows.
• Fix the OpenGL renderer not being able to run without a CUDA device available
  (GH-344).
• Fix incorrect CUDA driver function versions (GH-402).

v1.5.0

Changelog

[1.5.0] - 2024-12-02

Added

• Support for cooperative tile-based primitives using cuBLASDx and cuFFTDx, please see the tile
  documentation for details.
• Expose a reversed() built-in for iterators (GH-311).
• Support for saving Volumes into .nvdb files with the save_to_nvdb method.
• warp.fem: Add wp.fem.Trimesh3D and wp.fem.Quadmesh3D geometry types for 3D surfaces with new example_distortion_energy example.
• warp.fem: Add "add" option to wp.fem.integrate() for accumulating integration result to existing output.
• warp.fem: Add "assembly" option to wp.fem.integrate() for selecting between more memory-efficient or more
  computationally efficient integration algorithms.
• warp.fem: Add Nédélec (first kind) and Raviart-Thomas vector-valued function spaces
  providing conforming discretization of curl and div operators, respectively.
• warp.sim: Add a graph coloring module that supports converting trimesh into a vertex graph and applying coloring.
  The wp.sim.ModelBuilder now includes methods to color particles for use with wp.sim.VBDIntegrator(),
  users should call builder.color() before finalizing assets.
• warp.sim: Add support for a per-particle radius for soft-body triangle contact using the wp.sim.Model.particle_radius
  array (docs), replacing the previous
  hard-coded value of 0.01 (GH-329).
• Add a particle_radius parameter to wp.sim.ModelBuilder.add_cloth_mesh() and wp.sim.ModelBuilder.add_cloth_grid()
  to set a uniform radius for the added particles.
• Document wp.array attributes (GH-364).
• Document time-to-compile tradeoffs when using vector component assignment statements in kernels.
• Add introductory Jupyter notebooks to the notebooks directory.

Changed

• Drop support for Python 3.7; Python 3.8 is now the minimum-supported version.
• Promote the wp.Int, wp.Float, and wp.Scalar generic annotation types to the public API.
• warp.fem: Simplify querying neighboring cell quantities when integrating on sides using new
  wp.fem.cells(), wp.fem.to_inner_cell(), wp.fem.to_outer_cell() operators.
• Show an error message when the type returned by a function differs from its annotation, which would have led to the compilation stage failing.
• Clarify that wp.randn() samples a normal distribution of mean 0 and variance 1.
• Raise error when passing more than 32 variadic argument to the wp.printf() built-in.

Fixed

• Fix place setting of paddle backend.
• warp.fem: Fix tri-cubic shape functions on quadrilateral meshes.
• warp.fem: Fix caching of integrand kernels when changing code-generation options.
• Fix wp.expect_neq() overloads missing for scalar types.
• Fix an error when a wp.kernel or a wp.func object is annotated to return a None value.
• Fix error when reading multi-volume, BLOSC-compressed .nvdb files.
• Fix wp.printf() erroring out when no variadic arguments are passed (GH-333).
• Fix memory access issues in soft-rigid contact collisions (GH-362).
• Fix gradient propagation for in-place addition/subtraction operations on custom vector-type arrays.
• Fix the OpenGL renderer's window not closing when clicking the X button.
• Fix the OpenGL renderer's camera snapping to a different direction from the initial camera's orientation when first looking around.
• Fix custom colors being ignored when rendering meshes in OpenGL (GH-343).
• Fix topology updates not being supported by the the OpenGL renderer.

v1.4.2

Changelog

[1.4.2] - 2024-11-13

Changed

• Make the output of wp.print() in backward kernels consistent for all supported data types.

Fixed

• Fix to relax the integer types expected when indexing arrays (regression in 1.3.0).
• Fix printing vector and matrix adjoints in backward kernels.
• Fix kernel compile error when printing structs.
• Fix an incorrect user function being sometimes resolved when multiple overloads are available with array parameters with different dtype values.
• Fix error being raised when static and dynamic for-loops are written in sequence with the same iteration variable names (GH-331).
• Fix an issue with the Texture Write node, used in the Mandelbrot Omniverse sample, sometimes erroring out in multi-GPU environments.
• Code generation of in-place multiplication and division operations (regression introduced in a69d061)(GH-342).

v1.4.1

Changelog

[1.4.1] - 2024-10-15

Fixed

• Fix iter_reverse() not working as expected for ranges with steps other than 1 (GH-311).
• Fix potential out-of-bounds memory access when a wp.sparse.BsrMatrix object is reused for storing matrices of different shapes.
• Fix robustness to very low desired tolerance in wp.fem.utils.symmetric_eigenvalues_qr.
• Fix invalid code generation error messages when nesting dynamic and static for-loops.
• Fix caching of kernels with static expressions.
• Fix ModelBuilder.add_builder(builder) to correctly update articulation_start and thereby articulation_count when builder contains more than one articulation.
• Re-introduced the wp.rand*(), wp.sample*(), and wp.poisson() onto the Python scope to revert a breaking change.

v.1.4.0

CHANGELOG

[1.4.0] - 2024-10-01

Added

• Support for a new wp.static(expr) function that allows arbitrary Python expressions to be evaluated at the time of
  function/kernel definition (docs).
• Support for stream priorities to hint to the device that it should process pending work
  in high-priority streams over pending work in low-priority streams when possible
  (docs).
• Adaptive sparse grid geometry to warp.fem (docs).
• Support for defining wp.kernel and wp.func objects from within closures.
• Support for defining multiple versions of kernels, functions, and structs without manually assigning unique keys.
• Support for default argument values for user functions decorated with wp.func.
• Allow passing custom launch dimensions to jax_kernel() (GH-310).
• JAX interoperability examples for sharding and matrix multiplication (docs).
• Interoperability support for the PaddlePaddle ML framework (GH-318).
• Support wp.mod() for vector types (GH-282).
• Expose the modulo operator % to Python's runtime scalar and vector types.
• Support for fp64 atomic_add, atomic_max, and atomic_min (GH-284).
• Support for quaternion indexing (e.g. q.w).
• Support shadowing builtin functions (GH-308).
• Support for redefining function overloads.
• Add an ocean sample to the omni.warp extension.
• warp.sim.VBDIntegrator now supports body-particle collision.
• Add a contributing guide to the Sphinx docs .
• Add documentation for dynamic code generation (docs).

Changed

• wp.sim.Model.edge_indices now includes boundary edges.
• Unexposed wp.rand*(), wp.sample*(), and wp.poisson() from the Python scope.
• Skip unused functions in module code generation, improving performance.
• Avoid reloading modules if their content does not change, improving performance.
• wp.Mesh.points is now a property instead of a raw data member, its reference can be changed after the mesh is initialized.
• Improve error message when invalid objects are referenced in a Warp kernel.
• if/else/elif statements with constant conditions are resolved at compile time with no branches being inserted in the generated code.
• Include all non-hidden builtins in the stub file.
• Improve accuracy of symmetric eigenvalues routine in warp.fem.

Fixed

• Fix for wp.func erroring out when defining a Tuple as a return type hint (GH-302).
• Fix array in-place op (+=, -=) adjoints to compute gradients correctly in the backwards pass
• Fix vector, matrix in-place assignment adjoints to compute gradients correctly in the backwards pass, e.g.: v[1] = x
• Fix a bug in which Python docstrings would be created as local function variables in generated code.
• Fix a bug with autograd array access validation in functions from different modules.
• Fix a rare crash during error reporting on some systems due to glibc mismatches.
• Handle --num_tiles 1 in example_render_opengl.py (GH-306).
• Fix the computation of body contact forces in FeatherstoneIntegrator when bodies and particles collide.
• Fix bug in FeatherstoneIntegrator where eval_rigid_jacobian could give incorrect results or reach an infinite
  loop when the body and joint indices were not in the same order. Added Model.joint_ancestor to fix the indexing
  from a joint to its parent joint in the articulation.
• Fix wrong vertex index passed to add_edges() called from ModelBuilder.add_cloth_mesh() (GH-319).
• Add a workaround for uninitialized memory read warning in the compute-sanitizer initcheck tool when using wp.Mesh.
• Fix name clashes when Warp functions and structs are returned from Python functions multiple times.
• Fix name clashes between Warp functions and structs defined in different modules.
• Fix code generation errors when overloading generic kernels defined in a Python function.
• Fix issues with unrelated functions being treated as overloads (e.g., closures).
• Fix handling of stream argument in array.__dlpack__().
• Fix a bug related to reloading CPU modules.
• Fix a crash when kernel functions are not found in CPU modules.
• Fix conditions not being evaluated as expected in while statements.
• Fix printing Boolean and 8-bit integer values.
• Fix array interface type strings used for Boolean and 8-bit integer values.
• Fix initialization error when setting struct members.
• Fix Warp not being initialized upon entering a wp.Tape context.
• Use kDLBool instead of kDLUInt for DLPack interop of Booleans.

v1.3.3

[1.3.3] - 2024-09-04

• Bug fixes
  • Fix an aliasing issue with zero-copy array initialization from NumPy introduced in Warp 1.3.0.
  • Fix wp.Volume.load_from_numpy() behavior when bg_value is a sequence of values.

[1.3.2] - 2024-08-30

• Bug fixes
  • Fix accuracy of 3x3 SVD wp.svd3 with fp64 numbers (GH-281).
  • Fix module hashing when a kernel argument contained a struct array (GH-287).
  • Fix a bug in wp.bvh_query_ray() where the direction instead of the reciprocal direction was used
    (GH-288).
  • Fix errors when launching a CUDA graph after a module is reloaded. Modules that were used during graph capture
    will no longer be unloaded before the graph is released.
  • Fix a bug in wp.sim.collide.triangle_closest_point_barycentric() where the returned barycentric coordinates may be
    incorrect when the closest point lies on an edge.
  • Fix 32-bit overflow when array shape is specified using np.int32.
  • Fix handling of integer indices in the input_output_mask argument to autograd.jacobian and
    autograd.jacobian_fd (GH-289).
  • Fix ModelBuilder.collapse_fixed_joints() to correctly update the body centers of mass and the
    ModelBuilder.articulation_start array.
  • Fix precedence of closure constants over global constants.
  • Fix quadrature point indexing in wp.fem.ExplicitQuadrature (regression from 1.3.0).
• Documentation improvements
  • Add missing return types for built-in functions.
  • Clarify that atomic operations also return the previous value.
  • Clarify that wp.bvh_query_aabb() returns parts that overlap the bounding volume.

[1.3.1] - 2024-07-27

• Remove wp.synchronize() from PyTorch autograd function example
• Tape.check_kernel_array_access() and Tape.reset_array_read_flags() are now private methods.
• Fix reporting unmatched argument types

[1.3.0] - 2024-07-25

• Warp Core improvements

  • Update to CUDA 12.x by default (requires NVIDIA driver 525 or newer), please see README.md for commands to install CUDA 11.x binaries for older drivers
  • Add information to the module load print outs to indicate whether a module was
    compiled (compiled), loaded from the cache (cached), or was unable to be
    loaded (error).
  • wp.config.verbose = True now also prints out a message upon the entry to a wp.ScopedTimer.
  • Add wp.clear_kernel_cache() to the public API. This is equivalent to wp.build.clear_kernel_cache().
  • Add code-completion support for wp.config variables.
  • Remove usage of a static task (thread) index for CPU kernels to address multithreading concerns (GH-224)
  • Improve error messages for unsupported Python operations such as sequence construction in kernels
  • Update wp.matmul() CPU fallback to use dtype explicitly in np.matmul() call
  • Add support for PEP 563's from __future__ import annotations (GH-256).
  • Allow passing external arrays/tensors to wp.launch() directly via __cuda_array_interface__ and __array_interface__, up to 2.5x faster conversion from PyTorch
  • Add faster Torch interop path using return_ctype argument to wp.from_torch()
  • Handle incompatible CUDA driver versions gracefully
  • Add wp.abs() and wp.sign() for vector types
  • Expose scalar arithmetic operators to Python's runtime (e.g.: wp.float16(1.23) * wp.float16(2.34))
  • Add support for creating volumes with anisotropic transforms
  • Allow users to pass function arguments by keyword in a kernel using standard Python calling semantics
  • Add additional documentation and examples demonstrating wp.copy(), wp.clone(), and array.assign() differentiability
  • Add __new__() methods for all class __del__() methods to handle when a class instance is created but not instantiated before garbage collection
  • Implement the assignment operator for wp.quat
  • Make the geometry-related built-ins available only from within kernels
  • Rename the API-facing query types to remove their _t suffix: wp.BVHQuery, wp.HashGridQuery, wp.MeshQueryAABB, wp.MeshQueryPoint, and wp.MeshQueryRay
  • Add wp.array(ptr=...) to allow initializing arrays from pointer addresses inside of kernels (GH-206)

• warp.autograd improvements:

  • New warp.autograd module with utility functions gradcheck(), jacobian(), and jacobian_fd() for debugging kernel Jacobians (docs)
  • Add array overwrite detection, if wp.config.verify_autograd_array_access is true in-place operations on arrays on the Tape that could break gradient computation will be detected (docs)
  • Fix bug where modification of @wp.func_replay functions and native snippets would not trigger module recompilation
  • Add documentation for dynamic loop autograd limitations

• warp.sim improvements:

  • Improve memory usage and performance for rigid body contact handling when self.rigid_mesh_contact_max is zero (default behavior).
  • The mask argument to wp.sim.eval_fk() now accepts both integer and boolean arrays to mask articulations.
  • Fix handling of ModelBuilder.joint_act in ModelBuilder.collapse_fixed_joints() (affected floating-base systems)
  • Fix and improve implementation of ModelBuilder.plot_articulation() to visualize the articulation tree of a rigid-body mechanism
  • Fix ShapeInstancer __new__() method (missing instance return and *args parameter)
  • Fix handling of upaxis variable in ModelBuilder and the rendering thereof in OpenGLRenderer

• warp.sparse improvements:

  • Sparse matrix allocations (from bsr_from_triplets(), bsr_axpy(), etc.) can now be captured in CUDA graphs; exact number of non-zeros can be optionally requested asynchronously.
  • bsr_assign() now supports changing block shape (including CSR/BSR conversions)
  • Add Python operator overloads for common sparse matrix operations, e.g A += 0.5 * B, y = x @ C

• warp.fem new features and fixes:

  • Support for variable number of nodes per element
  • Global wp.fem.lookup() operator now supports wp.fem.Tetmesh and wp.fem.Trimesh2D geometries
  • Simplified defining custom subdomains (wp.fem.Subdomain), free-slip boundary conditions
  • New field types: wp.fem.UniformField, wp.fem.ImplicitField and wp.fem.NonconformingField
  • New streamlines, magnetostatics and nonconforming_contact examples, updated mixed_elasticity to use a nonlinear model
  • Function spaces can now export VTK-compatible cells for visualization
  • Fixed edge cases with NanoVDB function spaces
  • Fixed differentiability of wp.fem.PicQuadrature w.r.t. positions and measures

v1.3.2

[1.3.2] - 2024-08-30

• Bug fixes
  • Fix accuracy of 3x3 SVD wp.svd3 with fp64 numbers (GH-281).
  • Fix module hashing when a kernel argument contained a struct array (GH-287).
  • Fix a bug in wp.bvh_query_ray() where the direction instead of the reciprocal direction was used
    (GH-288).
  • Fix errors when launching a CUDA graph after a module is reloaded. Modules that were used during graph capture
    will no longer be unloaded before the graph is released.
  • Fix a bug in wp.sim.collide.triangle_closest_point_barycentric() where the returned barycentric coordinates may be
    incorrect when the closest point lies on an edge.
  • Fix 32-bit overflow when array shape is specified using np.int32.
  • Fix handling of integer indices in the input_output_mask argument to autograd.jacobian and
    autograd.jacobian_fd (GH-289).
  • Fix ModelBuilder.collapse_fixed_joints() to correctly update the body centers of mass and the
    ModelBuilder.articulation_start array.
  • Fix precedence of closure constants over global constants.
  • Fix quadrature point indexing in wp.fem.ExplicitQuadrature (regression from 1.3.0).
• Documentation improvements
  • Add missing return types for built-in functions.
  • Clarify that atomic operations also return the previous value.
  • Clarify that wp.bvh_query_aabb() returns parts that overlap the bounding volume.

[1.3.1] - 2024-07-27

• Remove wp.synchronize() from PyTorch autograd function example
• Tape.check_kernel_array_access() and Tape.reset_array_read_flags() are now private methods.
• Fix reporting unmatched argument types

[1.3.0] - 2024-07-25

• Warp Core improvements
  • Update to CUDA 12.x by default (requires NVIDIA driver 525 or newer), please see README.md for commands to install CUDA 11.x binaries for older drivers
  • Add information to the module load print outs to indicate whether a module was
    compiled (compiled), loaded from the cache (cached), or was unable to be
    loaded (error).
  • wp.config.verbose = True now also prints out a message upon the entry to a wp.ScopedTimer.
  • Add wp.clear_kernel_cache() to the public API. This is equivalent to wp.build.clear_kernel_cache().
  • Add code-completion support for wp.config variables.
  • Remove usage of a static task (thread) index for CPU kernels to address multithreading concerns (GH-224)
  • Improve error messages for unsupported Python operations such as sequence construction in kernels
  • Update wp.matmul() CPU fallback to use dtype explicitly in np.matmul() call
  • Add support for PEP 563's from __future__ import annotations (GH-256).
  • Allow passing external arrays/tensors to wp.launch() directly via __cuda_array_interface__ and __array_interface__, up to 2.5x faster conversion from PyTorch
  • Add faster Torch interop path using return_ctype argument to wp.from_torch()
  • Handle incompatible CUDA driver versions gracefully
  • Add wp.abs() and wp.sign() for vector types
  • Expose scalar arithmetic operators to Python's runtime (e.g.: wp.float16(1.23) * wp.float16(2.34))
  • Add support for creating volumes with anisotropic transforms
  • Allow users to pass function arguments by keyword in a kernel using standard Python calling semantics
  • Add additional documentation and examples demonstrating wp.copy(), wp.clone(), and array.assign() differentiability
  • Add __new__() methods for all class __del__() methods to handle when a class instance is created but not instantiated before garbage collection
  • Implement the assignment operator for wp.quat
  • Make the geometry-related built-ins available only from within kernels
  • Rename the API-facing query types to remove their _t suffix: wp.BVHQuery, wp.HashGridQuery, wp.MeshQueryAABB, wp.MeshQueryPoint, and wp.MeshQueryRay
  • Add wp.array(ptr=...) to allow initializing arrays from pointer addresses inside of kernels (GH-206)

v1.3.1

[1.3.1] - 2024-07-27

• Remove wp.synchronize() from PyTorch autograd function example
• Tape.check_kernel_array_access() and Tape.reset_array_read_flags() are now private methods.
• Fix reporting unmatched argument types

[1.3.0] - 2024-07-25

• Warp Core improvements

  • Update to CUDA 12.x by default (requires NVIDIA driver 525 or newer), please see README.md for commands to install CUDA 11.x binaries for older drivers
  • Add information to the module load print outs to indicate whether a module was
    compiled (compiled), loaded from the cache (cached), or was unable to be
    loaded (error).
  • wp.config.verbose = True now also prints out a message upon the entry to a wp.ScopedTimer.
  • Add wp.clear_kernel_cache() to the public API. This is equivalent to wp.build.clear_kernel_cache().
  • Add code-completion support for wp.config variables.
  • Remove usage of a static task (thread) index for CPU kernels to address multithreading concerns (GH-224)
  • Improve error messages for unsupported Python operations such as sequence construction in kernels
  • Update wp.matmul() CPU fallback to use dtype explicitly in np.matmul() call
  • Add support for PEP 563's from __future__ import annotations (GH-256).
  • Allow passing external arrays/tensors to wp.launch() directly via __cuda_array_interface__ and __array_interface__, up to 2.5x faster conversion from PyTorch
  • Add faster Torch interop path using return_ctype argument to wp.from_torch()
  • Handle incompatible CUDA driver versions gracefully
  • Add wp.abs() and wp.sign() for vector types
  • Expose scalar arithmetic operators to Python's runtime (e.g.: wp.float16(1.23) * wp.float16(2.34))
  • Add support for creating volumes with anisotropic transforms
  • Allow users to pass function arguments by keyword in a kernel using standard Python calling semantics
  • Add additional documentation and examples demonstrating wp.copy(), wp.clone(), and array.assign() differentiability
  • Add __new__() methods for all class __del__() methods to handle when a class instance is created but not instantiated before garbage collection
  • Implement the assignment operator for wp.quat
  • Make the geometry-related built-ins available only from within kernels
  • Rename the API-facing query types to remove their _t suffix: wp.BVHQuery, wp.HashGridQuery, wp.MeshQueryAABB, wp.MeshQueryPoint, and wp.MeshQueryRay
  • Add wp.array(ptr=...) to allow initializing arrays from pointer addresses inside of kernels (GH-206)

• warp.autograd improvements:

  • New warp.autograd module with utility functions gradcheck(), jacobian(), and jacobian_fd() for debugging kernel Jacobians (docs)
  • Add array overwrite detection, if wp.config.verify_autograd_array_access is true in-place operations on arrays on the Tape that could break gradient computation will be detected (docs)
  • Fix bug where modification of @wp.func_replay functions and native snippets would not trigger module recompilation
  • Add documentation for dynamic loop autograd limitations

• warp.sim improvements:

  • Improve memory usage and performance for rigid body contact handling when self.rigid_mesh_contact_max is zero (default behavior).
  • The mask argument to wp.sim.eval_fk() now accepts both integer and boolean arrays to mask articulations.
  • Fix handling of ModelBuilder.joint_act in ModelBuilder.collapse_fixed_joints() (affected floating-base systems)
  • Fix and improve implementation of ModelBuilder.plot_articulation() to visualize the articulation tree of a rigid-body mechanism
  • Fix ShapeInstancer __new__() method (missing instance return and *args parameter)
  • Fix handling of upaxis variable in ModelBuilder and the rendering thereof in OpenGLRenderer

• warp.sparse improvements:

  • Sparse matrix allocations (from bsr_from_triplets(), bsr_axpy(), etc.) can now be captured in CUDA graphs; exact number of non-zeros can be optionally requested asynchronously.
  • bsr_assign() now supports changing block shape (including CSR/BSR conversions)
  • Add Python operator overloads for common sparse matrix operations, e.g A += 0.5 * B, y = x @ C

• warp.fem new features and fixes:

  • Support for variable number of nodes per element
  • Global wp.fem.lookup() operator now supports wp.fem.Tetmesh and wp.fem.Trimesh2D geometries
  • Simplified defining custom subdomains (wp.fem.Subdomain), free-slip boundary conditions
  • New field types: wp.fem.UniformField, wp.fem.ImplicitField and wp.fem.NonconformingField
  • New streamlines, magnetostatics and nonconforming_contact examples, updated mixed_elasticity to use a nonlinear model
  • Function spaces can now export VTK-compatible cells for visualization
  • Fixed edge cases with NanoVDB function spaces
  • Fixed differentiability of wp.fem.PicQuadrature w.r.t. positions and measures

v1.3.0

[1.3.0] - 2024-07-25

• Warp Core improvements

  • Update to CUDA 12.x by default (requires NVIDIA driver 525 or newer), please see README.md for commands to install CUDA 11.x binaries for older drivers
  • Add information to the module load print outs to indicate whether a module was
    compiled (compiled), loaded from the cache (cached), or was unable to be
    loaded (error).
  • wp.config.verbose = True now also prints out a message upon the entry to a wp.ScopedTimer.
  • Add wp.clear_kernel_cache() to the public API. This is equivalent to wp.build.clear_kernel_cache().
  • Add code-completion support for wp.config variables.
  • Remove usage of a static task (thread) index for CPU kernels to address multithreading concerns (GH-224)
  • Improve error messages for unsupported Python operations such as sequence construction in kernels
  • Update wp.matmul() CPU fallback to use dtype explicitly in np.matmul() call
  • Add support for PEP 563's from __future__ import annotations (GH-256).
  • Allow passing external arrays/tensors to wp.launch() directly via __cuda_array_interface__ and __array_interface__, up to 2.5x faster conversion from PyTorch
  • Add faster Torch interop path using return_ctype argument to wp.from_torch()
  • Handle incompatible CUDA driver versions gracefully
  • Add wp.abs() and wp.sign() for vector types
  • Expose scalar arithmetic operators to Python's runtime (e.g.: wp.float16(1.23) * wp.float16(2.34))
  • Add support for creating volumes with anisotropic transforms
  • Allow users to pass function arguments by keyword in a kernel using standard Python calling semantics
  • Add additional documentation and examples demonstrating wp.copy(), wp.clone(), and array.assign() differentiability
  • Add __new__() methods for all class __del__() methods to handle when a class instance is created but not instantiated before garbage collection
  • Implement the assignment operator for wp.quat
  • Make the geometry-related built-ins available only from within kernels
  • Rename the API-facing query types to remove their _t suffix: wp.BVHQuery, wp.HashGridQuery, wp.MeshQueryAABB, wp.MeshQueryPoint, and wp.MeshQueryRay
  • Add wp.array(ptr=...) to allow initializing arrays from pointer addresses inside of kernels (GH-206)

• warp.autograd improvements:

  • New warp.autograd module with utility functions gradcheck(), jacobian(), and jacobian_fd() for debugging kernel Jacobians (docs)
  • Add array overwrite detection, if wp.config.verify_autograd_array_access is true in-place operations on arrays on the Tape that could break gradient computation will be detected (docs)
  • Fix bug where modification of @wp.func_replay functions and native snippets would not trigger module recompilation
  • Add documentation for dynamic loop autograd limitations

• warp.sim improvements:

  • Improve memory usage and performance for rigid body contact handling when self.rigid_mesh_contact_max is zero (default behavior).
  • The mask argument to wp.sim.eval_fk() now accepts both integer and boolean arrays to mask articulations.
  • Fix handling of ModelBuilder.joint_act in ModelBuilder.collapse_fixed_joints() (affected floating-base systems)
  • Fix and improve implementation of ModelBuilder.plot_articulation() to visualize the articulation tree of a rigid-body mechanism
  • Fix ShapeInstancer __new__() method (missing instance return and *args parameter)
  • Fix handling of upaxis variable in ModelBuilder and the rendering thereof in OpenGLRenderer

• warp.sparse improvements:

  • Sparse matrix allocations (from bsr_from_triplets(), bsr_axpy(), etc.) can now be captured in CUDA graphs; exact number of non-zeros can be optionally requested asynchronously.
  • bsr_assign() now supports changing block shape (including CSR/BSR conversions)
  • Add Python operator overloads for common sparse matrix operations, e.g A += 0.5 * B, y = x @ C

• warp.fem new features and fixes:

  • Support for variable number of nodes per element
  • Global wp.fem.lookup() operator now supports wp.fem.Tetmesh and wp.fem.Trimesh2D geometries
  • Simplified defining custom subdomains (wp.fem.Subdomain), free-slip boundary conditions
  • New field types: wp.fem.UniformField, wp.fem.ImplicitField and wp.fem.NonconformingField
  • New streamlines, magnetostatics and nonconforming_contact examples, updated mixed_elasticity to use a nonlinear model
  • Function spaces can now export VTK-compatible cells for visualization
  • Fixed edge cases with NanoVDB function spaces
  • Fixed differentiability of wp.fem.PicQuadrature w.r.t. positions and measures