[llvm] [JITLink] Some cleanups to EHFrameSupport (PR #66707)

Jonas Hahnfeld via llvm-commits llvm-commits at lists.llvm.org
Thu Oct 12 11:59:44 PDT 2023


hahnjo wrote:

> This could cause us to reject well-formed objects that would have been accepted before.

According to the DWARF standard:
> CIE_pointer
> A constant offset into the .debug_frame section that denotes the CIE that is associated with this FDE.

So the field in the emitted objects must always be offsets while only dynamically registered frames must have a finalized pointer, which we achieve by adding edges / relocations. In my reading this means we must never see a relocation on this field before we process it...

https://github.com/llvm/llvm-project/pull/66707


More information about the llvm-commits mailing list