[GitHub] [tvm] jroesch commented on a change in pull request #9483: [Relay] Switch the VM to use the LowerTE pass instead of TECompiler::{Lower,LowerShapeFunc}.

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

jroesch commented on a change in pull request #9483:
URL: https://github.com/apache/tvm/pull/9483#discussion_r758632547



##########
File path: src/relay/backend/graph_executor_codegen.cc
##########
@@ -215,24 +215,35 @@ class GraphExecutorCodegen : public backend::MemoizedExprTranslator<std::vector<
 
     if (memory_plan_.defined()) {
       // TODO(@electriclilies, @jroesch): remove UpdateMainWorkspaceSize
-      func_info =
-          relay::tec::UpdateMainWorkspaceSize(mod, targets_, memory_plan_->expr_to_storage_info);
+      tec::TargetMap tec_target_map;
+      for (const auto& pair : targets_) {
+        tec_target_map.emplace(static_cast<DLDeviceType>(pair.first->value), pair.second);
+      }
+      func_info = relay::tec::UpdateMainWorkspaceSize(mod, tec_target_map,

Review comment:
       Do we have a tracking issue for flowing this into the actual lowering machinery? 




-- 
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