Fix regression in torch.qr (#23591) (#23606)

Summary: Changelog: - Use narrow instead of narrow_copy while returning Pull Request resolved: https://github.com/pytorch/pytorch/pull/23591 Test Plan: - All tests should pass to ensure that the change is correct Fixes https://github.com/pytorch/pytorch/issues/23580 Differential Revision: D16581174 Pulled By: ezyang fbshipit-source-id: 1b6bf7d338ddd138ea4c6aa6901834dd202ec79c

Fix regression in torch.qr (#23591) (#23606)
Summary: Changelog: - Use narrow instead of narrow_copy while returning Pull Request resolved: https://github.com/pytorch/pytorch/pull/23591 Test Plan: - All tests should pass to ensure that the change is correct Fixes https://github.com/pytorch/pytorch/issues/23580 Differential Revision: D16581174 Pulled By: ezyang fbshipit-source-id: 1b6bf7d338ddd138ea4c6aa6901834dd202ec79c
d6df9575 · Vishwak Srinivasan · Soumith Chintala · c3bad0de · d6df9575 · d6df9575
--- a/aten/src/ATen/native/BatchLinearAlgebra.cpp
+++ b/aten/src/ATen/native/BatchLinearAlgebra.cpp
@@ -877,7 +877,7 @@ std::tuple<Tensor, Tensor> _qr_helper_cpu(const Tensor& self, bool some) {
  } else {
    singleCheckErrors(infos[0], "qr_cpu");
  }
-  return std::make_tuple(q_working_copy.narrow_copy(-1, 0, n_columns_q), R);
+  return std::make_tuple(q_working_copy.narrow(-1, 0, n_columns_q), R);
 }

 std::tuple<Tensor,Tensor> qr(const Tensor& self, bool some) {

--- a/aten/src/ATen/native/cuda/BatchLinearAlgebra.cu
+++ b/aten/src/ATen/native/cuda/BatchLinearAlgebra.cu
@@ -1217,8 +1217,8 @@ std::tuple<Tensor,Tensor> _qr_helper_cuda(const Tensor& self, bool some) {
    singleCheckErrors(infos[0], "qr_cuda");
  }

-  return std::make_tuple(q_working_copy.narrow_copy(-1, 0, n_columns_q),
-                         r_working_copy.narrow_copy(-2, 0, n_columns_q).triu_());
+  return std::make_tuple(q_working_copy.narrow(-1, 0, n_columns_q),
+                         r_working_copy.narrow(-2, 0, n_columns_q).triu());
 }

 // ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ symeig ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~