提交 d14858e4 编写于 作者: M Michal Gallus

MKLDNN elementwise_mul: Parallelize mul

上级 ed31936b
...@@ -144,6 +144,7 @@ class ElementwiseMulMKLDNNKernel : public framework::OpKernel<T> { ...@@ -144,6 +144,7 @@ class ElementwiseMulMKLDNNKernel : public framework::OpKernel<T> {
mul_func_t mul_func = (mul_func_t) mul.getCode(); mul_func_t mul_func = (mul_func_t) mul.getCode();
#pragma omp parallel for collapse(2)
for (int ni = 0; ni < n; ni++) { for (int ni = 0; ni < n; ni++) {
for (int ci = 0; ci < C; ci++) { for (int ci = 0; ci < C; ci++) {
auto ptr_x = auto ptr_x =
......
Markdown is supported
0% .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑!
想要评论请 注册