v3.9 update (#2203)
* v3.9 update * voidD --------- Co-authored-by: yuzhai <yuzhai@nvidia.com>
This commit is contained in:
@@ -986,6 +986,21 @@ struct multiply_add<Array<T, N>, Array<T, N>, Array<T, N>> {
|
||||
return result;
|
||||
}
|
||||
|
||||
CUTLASS_HOST_DEVICE
|
||||
Array<T, N> operator()(Array<T, N> const &a, Array<T, N> const &b, T const &scalar) const {
|
||||
|
||||
Array<T, N> result;
|
||||
multiply_add<T> scalar_op;
|
||||
|
||||
CUTLASS_PRAGMA_UNROLL
|
||||
for (int i = 0; i < N; ++i) {
|
||||
result[i] = scalar_op(a[i], b[i], scalar);
|
||||
}
|
||||
|
||||
return result;
|
||||
}
|
||||
|
||||
|
||||
CUTLASS_HOST_DEVICE
|
||||
Array<T, N> operator()(Array<T, N> const &a, T const &scalar_b, T const &scalar_c) const {
|
||||
|
||||
|
||||
Reference in New Issue
Block a user