Repository Issue Activity (beta)

#1691 `volume_sample_grad` fails to compile for `vec4` volumes

#1864 Support multidimensional CUDA blocks for BVH and mesh queries

#1692 `texture_sample` normalizes 32-bit integers differently on CPU and CUDA

#1469 capture_while() fails during multi-stream CUDA graph capture with unjoined sibling stream

#1892 Build from source fails to find MSVC on Build Tools-only installations (vswhere omits Build Tools without -products *)

#1895 Reject malformed NanoVDB topology during file loading

#1891 Add Block Jacobi Preconditioner for Warp CG Solver

#1798 ModuleHasher: kernel hash unstable across multi-device build (dense overload graphs)

#1717 Tile operations ignore kernel-level `enable_backward=False`

#1689 Split builtins.py into smaller modules

#1920 CUDA: mixed-width texture sampling returns incorrect values on sm_89

#1930 Replace bare assertions in runtime code

#1848 `tile_squeeze()` accepts out-of-range negative axes

#1846 BVH volume (AABB/sphere) traversal spends 20-30% in stack traffic and node reloads

#1845 BVH ray/capsule queries carry a traversal stack and 32KB shared memory they don't need

#1906 Remove deprecated hash grid query aliases

#1912 fem.integrate() rejects a flat scalar output array and accepts a mis-shaped 2-D one

#1910 Scaled BSR expression on the left of +/- applies the scale to the other operand

#1092 [REQ] Direct Sparse Cholesky Factorization and Implicit Euler

#1925 Build fails when the repository path contains a space

#1918 Integer floor division (`//`) truncates toward zero instead of flooring

#1922 Warp FEM topology optimization example

#1923 Add reusable fixed-topology FEM assembly

#1924 Fuse closest-cell traversal over Warp BVH

#1797 Add Delaunay edge flipping and adjacency to warp.geometry