[PATCH] D131618: [WIP][Do NOT review] LLD related changes for -ffat-lto-objects support

[Paul Kirth via Phabricator via llvm-commits]

paulkirth added a comment.

I know this is WIP, but I think there are some Q's worth discussing as work progresses.

1. How do "fat" objects interact w/ existing object types? i.e., regular object files and normal LTO objects?
2. How do we control the behavior? can I just mix them? Are there combinations that should not work?
3. Are the plans for testing? do we need special tests for the new format? or are the differences insignificant? can they be covered by existing tests?

I haven't seen these points discussed in the RFC or in these patches. How these are //intended// to work should probably be spelled out somewhere. Probably both in the code and in the documentation.



================
Comment at: lld/ELF/Driver.cpp:2636-2640
+        // Do not aggregate the .fatlto section
+        if (s && s->name == ".fatlto")
+          continue;
         if (s && s != &InputSection::discarded)
           inputSections.push_back(s);
----------------
maybe it makes sense to check for pointer validity separately now that there are 2 uses? 


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D131618/new/

https://reviews.llvm.org/D131618