leavelet/cutlass
1
0
Fork 0
Code Issues Pull Requests Actions 28 Packages Projects Releases Wiki Activity

update 3.8 v2 (#2112)

Browse Source 
* update 3.8 v2

* update 3.8

---------

Co-authored-by: yuzhai <yuzhai@nvidia.com>
This commit is contained in:
Yujia Zhai
2025-02-19 22:03:14 -05:00
committed by GitHub
co-authored by yuzhai
parent e9627ce55b
commit b84e9802d8
166 changed files with 3986 additions and 4037 deletions
Download Patch File Download Diff File Expand all files Collapse all files
include/cute/int_tuple.hpp
+2 -2
View File
@@ -330,7 +330,7 @@ ceil_div(IntTupleA const& a, IntTupleB const& b)
constexpr int R = tuple_size<IntTupleA>::value; // Missing ranks in TupleB are implicitly 1
return transform(a, append<R>(b,Int<1>{}), [](auto const& x, auto const& y) { return ceil_div(x,y); });
} else { // tuple int
auto const [result, rest] = fold(a, cute::make_tuple(cute::make_tuple(), b),
auto [result, rest] = fold(a, cute::make_tuple(cute::make_tuple(), b),
[] (auto const& init, auto const& ai) {
return cute::make_tuple(append(get<0>(init), ceil_div(ai, get<1>(init))), ceil_div(get<1>(init), ai));
});
@@ -390,7 +390,7 @@ shape_div(IntTupleA const& a, IntTupleB const& b)
static_assert(tuple_size<IntTupleA>::value == tuple_size<IntTupleB>::value, "Mismatched ranks");
return transform(a, b, [](auto const& x, auto const& y) { return shape_div(x,y); });
} else { // tuple int
auto const [result, rest] = fold(a, cute::make_tuple(cute::make_tuple(), b),
auto [result, rest] = fold(a, cute::make_tuple(cute::make_tuple(), b),
[] (auto const& init, auto const& ai) {
return cute::make_tuple(append(get<0>(init), shape_div(ai, get<1>(init))), shape_div(get<1>(init), ai));
});