-
Notifications
You must be signed in to change notification settings - Fork 637
All issues
Issue creation is restricted in this repository
Issues
is:issue state:open
is:issue state:open
Search results
Fix multi-environment nanogrid cell lookup
bugSomething isn't workingSomething isn't workingStatus: Open.#2017 In NVIDIA/warp;- Status: Open.#2015 In NVIDIA/warp;
- Status: Open.#2014 In NVIDIA/warp;
Grouped CUDA LBVH can exceed the query stack capacity with 33 AABBs
bugSomething isn't workingSomething isn't workingspatial accelerationBVHs, hash grids, and the spatial queries they accelerateBVHs, hash grids, and the spatial queries they accelerateStatus: Open.BVH ray queries miss boxes when rays run along a face
bugSomething isn't workingSomething isn't workingStatus: Open.eig3 produces NaN gradients for eigenvalue-only loss at repeated eigenvalues
bugSomething isn't workingSomething isn't workingStatus: Open.FEM QR routines return NaNs for large float32 matrices
bugSomething isn't workingSomething isn't workingStatus: Open.#2008 In NVIDIA/warp;- Status: Open.#2006 In NVIDIA/warp;
Support LLVM
-mllvmoptions for CPU kernel compilationfeature requestRequest for something to be addedRequest for something to be addedStatus: Open.CUDA
tile_assign()loses gradients from broadcast source tilesbugSomething isn't workingSomething isn't workingtileTile programming modelTile programming modelStatus: Open.tile_sum(..., axis=...)gives incorrect input gradientsautodiffTapes, adjoint generation, custom gradients, and gradient correctness in WarpTapes, adjoint generation, custom gradients, and gradient correctness in WarpbugSomething isn't workingSomething isn't workingtileTile programming modelTile programming modelStatus: Open.- Status: Open.#1996 In NVIDIA/warp;