[GitHub] [tvm] areusch commented on a diff in pull request #13752: [microTVM] Use QNN schedules to give SOTA performance

["areusch (via GitHub)" <gi...@apache.org>]

areusch commented on code in PR #13752:
URL: https://github.com/apache/tvm/pull/13752#discussion_r1123499276


##########
src/target/source/codegen_c.cc:
##########
@@ -631,8 +632,10 @@ void CodeGenC::PrintVecBinaryOp(const std::string& op, DataType t, PrimExpr lhs,
   }
 }
 
+NameSupply global_name_supply = NameSupply("");
 void CodeGenC::VisitStmt_(const AllocateConstNode* op) {
-  std::string symbol_name = op->buffer_var->name_hint;
+  std::string symbol_name = global_name_supply->FreshName(op->buffer_var->name_hint);

Review Comment:
   i discussed with @gigiblender and in doing so we realized that there is already `var_idmap_` which uses another NameSupply. Does `AllocVarID(op->buffer_var.get())` work here? this parallels what was done in [AllocateNode](https://github.com/apache/tvm/blob/bf86d9f8c25e105a12c1b96ab0a4cc977ee088b7/src/target/source/codegen_c.cc#L857).



-- 
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.

To unsubscribe, e-mail: commits-unsubscribe@tvm.apache.org

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