add new branch heyi_cast_transpose and update cast_transpose optimizations#89

Open

eliotwang wants to merge 9 commits intoROCm:devfrom eliotwang:heyi_cast_transpose

Commits on Jan 7, 2025

add new branch heyi_cast_transpose and update cast_transpose optimizations
eliotwang
committed
Revert the tests/cpp/operator/test_cast_transpose.cu to its original version, add __HIP_PLATFORM_AMD__ to guard modified code in transformer_engine/common/transpose/cast_transpose.cu, use hip_float…
eliotwang
committed
fix cast_transpose issue where the width and height of the tensor cannot be divided by the tile size.
eliotwang
committed
merge pr#1083 and update cast transpose and its fusion optimization advances
eliotwang
committed
optimize reduce dbias kernel of cast transpose fusion kernel
eliotwang
committed
optimize reduce dbias kernel of cast transpose fusion kernel
eliotwang
committed
Fix incorrect commit content
eliotwang
committed

Commits on Mar 7, 2025