erichkeane added a comment. In D146178#4199263 <https://reviews.llvm.org/D146178#4199263>, @erichkeane wrote:
> After a night of sleeping on it, the use of a AdjustConstraintDepth::Diff and > AdjustConstraintDepth::Value feels like a hacky solution to SOMETHING here, > though I'm not sure what yet. The depth of a template shouldn't change > between the equality and constraint checking, it is usually a property of the > specialization level. I could definitely believe that > `getTemplateInstantiationArgs` needs some sort of change here to better > detect its own state, but this solution doesn't seem right to me. I debugged a bit: It isn't correct I think that `FunctionTemplateDecl` doesn't pick up its template arguments in `getTemplateInstantiationArgs`. Additionally, instead of picking up its `DeclContext`, it probably needs its `LexicalDeclContext` as the next step, which I believe fixes the problem (plus your change in SemaOverload.cpp) Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D146178/new/ https://reviews.llvm.org/D146178 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
