[PATCH] D56684: [WebAssembly] Fixed objdump not parsing function headers.

[Derek Schuff via Phabricator via llvm-commits]

dschuff added a comment.

Yeah, thinking about this a little more, I think it makes sense. IIRC the "address" (or value of a symbol, which I think is equivalent?) in the object file/linking/executable context refers to the section offset rather than memory-mapped address even for ELF, which would make our objdump consistent with other targets too. For the purposes of LLVM internals, I think you are right that having that consistency seems best; for the users, it's possible we'll want to put some more info in the final displayed output, but we can figure out how to get that once the guts are hooked up.


Repository:
  rL LLVM

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D56684/new/

https://reviews.llvm.org/D56684