trunk/769b13f46a24d108e74a807bc3bdb5cc8b478758: Add tiled CUDA kernel for dense 2D transpose copies (#194310)
- PyTorch: 1410 events in the last 90 days
- PyTorch: 1392th Release in the last 90 days
- Previous: earlier the same day · trunk/98950cfd474df1053b0c1f7d686de13302b20e92
What happened
Addresses #113560 . t().contiguous() on CUDA goes through the generic TensorIterator elementwise kernel, which can't coalesce both the read and the write of a transpose. This adds a shared-memory tiled kernel for copies that TensorIterator reduces to a 2D transpose, including sources with a row pitch larger than the row (narrowed or strided outer dims). Kernel 128-bit global loads and stores for aligned 1-, 2- and 4…
Summary assembled by rule from the sources below