bnbarham wrote: > I have no objections to this if that makes your workflow work. However I > think this cures the symptom and not the real cause. It might be worth > investigating how to adapt lldb to the incremental processing change more > deeply.
To be clear, this is re-applying https://github.com/llvm/llvm-project/pull/65683 such that it is actually the NFC it was intended to be. I've closed the LLDB PR (https://github.com/llvm/llvm-project/pull/66271), so it's still using `-fincremental-extensions`. The actual use case for splitting these is eg. Swift's clang importer, where the need is very similar to https://github.com/llvm/llvm-project/issues/62413 and IMO it makes sense to keep these two as separate, with `-fincremental-extensions` implying incremental processing. https://github.com/llvm/llvm-project/pull/66446 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
