leavelet/cutlass
1
0
Fork 0
Code Issues Pull Requests Actions 1 Packages Projects Releases Wiki Activity

v4.2 release. (#2587)

Browse Source 
* Fix default cluster callback values to 1 to avoid profiler failure when these values are not set in command line.

* v4.2 release.
This commit is contained in:
Junkai-Wu
2025-08-23 06:11:24 +08:00
committed by GitHub
parent 11cad1f67b
commit a49a78ffef
351 changed files with 28182 additions and 2032 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
include/cute/tensor_impl.hpp
View File

@@ -761,12 +761,12 @@ recast(Tensor&& tensor)
using OldType = typename remove_cvref_t<Tensor>::element_type;
using NewType = copy_cv_t<OldType, NewType_>;
auto old_layout = tensor.layout();
auto new_layout = recast_layout<OldType,NewType>(old_layout);
if constexpr (is_same<NewType, OldType>::value) {
return tensor;
return make_tensor(static_cast<Tensor&&>(tensor).data(), tensor.layout());
} else {
auto old_layout = tensor.layout();
auto new_layout = recast_layout<OldType,NewType>(old_layout);
// If this is an upcast of a normal Layout with static negative strides, then offset as well
if constexpr (sizeof(OldType) < sizeof(NewType) && not is_composed_layout<decltype(old_layout)>::value) {
auto shape_diff = transform(flatten(old_layout.shape()), flatten(new_layout.shape()), minus{});