Uni20
Asynchronous tensor and runtime library.
Loading...
Searching...
No Matches
cuda Directory Reference
Directory dependency graph for cuda:

Files

 conjugate_inplace.hpp
 CUDA reference backend for in-place scalar conjugation.
 
 copy.hpp
 CUDA runtime backend for contiguous Tensor transfers.
 
 copy_task.hpp
 Coroutine kernel-dispatch customization for CUDA Tensor copies.
 
 elementwise_arithmetic.hpp
 Compiled CUDA entry points for named stateless arithmetic transforms.
 
 elementwise_conjugate.hpp
 Typed launch boundary for CUDA accessor-respecting in-place conjugation.
 
 elementwise_copy.hpp
 Typed launch boundary for CUDA accessor-respecting element copies.
 
 elementwise_plan.hpp
 CUDA lowering for compact affine elementwise iteration plans.
 
 elementwise_scale.hpp
 Compiled CUDA entry points for elementwise scaling.
 
 mdspec_traits.hpp
 Shared CUDA reference-backend traits for raw buffer mdspecs.
 
 partitioned_buffer_linear.hpp
 Allocation-wide CUDA linear operations for partitioned buffers.
 
 transform.hpp
 CUDA reference backend for registered elementwise transforms.