[PATCH] D110530: [ORC] Avoid deadlocks in ObjectLinkingLayer plugins with blocking finalize handlers

[Stefan Gränitz via Phabricator via llvm-commits]

sgraenitz added a comment.

> Using the plugin's very own DebugObject::finalizeAsync() is not sufficient

I could spawn a new thread there as well, but it's not necessary if the memory manager does it. What do you think?



================
Comment at: llvm/lib/ExecutionEngine/Orc/EPCGenericJITLinkMemoryManager.cpp:71
+              OnFinalize(std::move(FE));
+          }).detach();
         },
----------------
It doesn't appear there is a thread pool that could accommodate this task. Is a detached thread ok?


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D110530/new/

https://reviews.llvm.org/D110530