Skip to content

Commit

Permalink
Fix build regression caused by pytorch#58940 (pytorch#59008)
Browse files Browse the repository at this point in the history
s/Vectorized/Vec256/

Vec256 were renamed to Vectorized on master after the branch cut
  • Loading branch information
malfet authored and mainakpal08 committed Aug 31, 2021
1 parent b4cbefd commit b0195a2
Showing 1 changed file with 2 additions and 2 deletions.
4 changes: 2 additions & 2 deletions aten/src/ATen/native/cpu/Activation.cpp
Original file line number Diff line number Diff line change
Expand Up @@ -589,7 +589,7 @@ void silu_backward_kernel(TensorIteratorBase& iter) {

void mish_kernel(TensorIteratorBase& iter) {
AT_DISPATCH_FLOATING_TYPES(iter.dtype(), "mish_cpu", [&]() {
using Vec = Vectorized<scalar_t>;
using Vec = Vec256<scalar_t>;
cpu_kernel_vec(
iter,
[](scalar_t x) -> scalar_t{
Expand All @@ -603,7 +603,7 @@ void mish_kernel(TensorIteratorBase& iter) {

void mish_backward_kernel(TensorIterator& iter) {
AT_DISPATCH_FLOATING_TYPES(iter.dtype(), "mish_backward_cpu", [&]() {
using Vec = Vectorized<scalar_t>;
using Vec = Vec256<scalar_t>;
const Vec kOneVec(scalar_t(1));
cpu_kernel_vec(
iter,
Expand Down

0 comments on commit b0195a2

Please sign in to comment.