[PATCH] D18683: Fix bug with duplicate struct types in Linker

[Tim Armstrong via llvm-commits]

timarmstrong added a comment.

Right, but as someone who's trying to build a non-trivial project on top of LLVM, how am I meant to know whether something I'm building on top of LLVM will work if the expected behaviour is completely undocumented? The only thing I've had to go off is whether it works (it did in the past) and whether the behaviour is sane (it is) and whether it looks like what the code is meant to do (it does).

This has worked fine in the past and as far as I can tell is what the code is *meant* to do. Before my fix the code was internally inconsistent: hasType() returns false even when the type *will* be in the composite module. I.e. the internal invariants of IRMover were being violated and that's a recipe for bugs.

I get that it's an open-source project and use-cases only get supported if someone contributes time, so I'm trying to help here. The change makes the code simpler, internally consistent and fixes a use case that previously worked.

Are you just saying that IRMover is only meant to work for the particular use-case of the LLVM linker and other people who are trying to use it as a utiltiy to move IR between modules just need to deal with it?


http://reviews.llvm.org/D18683