[GitHub] [incubator-mxnet] access2rohit commented on a change in pull request #18807: adding error message when attempting to use Large tensor with linalg_syevd

[GitBox <gi...@apache.org>]

access2rohit commented on a change in pull request #18807:
URL: https://github.com/apache/incubator-mxnet/pull/18807#discussion_r461889976



##########
File path: 3rdparty/mshadow/mshadow/base.h
##########
@@ -336,6 +336,8 @@ const float kPi = 3.1415926f;
   typedef int32_t index_t;
 #endif
 
+const index_t kInt32Limit = (int64_t{1} << 31) - 1;

Review comment:
       @Zha0q1 good suggestion !

##########
File path: src/operator/tensor/la_op.h
##########
@@ -470,11 +470,14 @@ inline bool DetType(const nnvm::NodeAttrs& attrs,
 inline bool LaEigFactShape(const nnvm::NodeAttrs& attrs,
                            mxnet::ShapeVector* in_attrs,
                            mxnet::ShapeVector* out_attrs) {
+  using namespace mshadow;
   CHECK_EQ(in_attrs->size(), 1);
   CHECK_EQ(out_attrs->size(), 2);
   const mxnet::TShape& in_a = (*in_attrs)[0];
   const mxnet::TShape& out_u = (*out_attrs)[0];
   const mxnet::TShape& out_l = (*out_attrs)[1];
+  CHECK_LE(in_a.Size(), INT_MAX)

Review comment:
       is included in base.h 




----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
users@infra.apache.org