akash-manna-sky wrote: > Can we prevent incorrect attributes from triggering this function at the > parsing stage, rather than patching the function itself?
The crashing input is a well-formed attribute in a valid position — its argument only turns out to be value-dependent in Sema, so the parser has nothing to reject. And filtering malformed attributes earlier wouldn't help: the assertion fired for the valid attribute, which is stored on the declarator rather than the chunk. The wrong assumption was in this lookup, so I think the fix belongs here. https://github.com/llvm/llvm-project/pull/216348 _______________________________________________ cfe-commits mailing list [email protected] https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
