Use std::vector instead of alloca to work around hcc crash
authorJunjie Bai <bai@in.tum.de>
Thu, 13 Dec 2018 20:31:38 +0000 (12:31 -0800)
committerFacebook Github Bot <facebook-github-bot@users.noreply.github.com>
Thu, 13 Dec 2018 20:34:36 +0000 (12:34 -0800)
commit86fbf17ba64f985a7d632576091a08f8344f7a24
tree726a7a97ed572f5722d933b03b889a52b98782c7
parentf61612206cee9e00011ba9945be771ddc1673092
Use std::vector instead of alloca to work around hcc crash

Summary: Pull Request resolved: https://github.com/pytorch/pytorch/pull/15175

Differential Revision: D13453708

Pulled By: bddppq

fbshipit-source-id: f8c147ae9f679e395fee9d4c73ebcca052c9a752
aten/src/ATen/cuda/detail/IndexUtils.cu