saar.raz added inline comments.
================ Comment at: lib/Sema/SemaConcept.cpp:34 + Diag(ConstraintExpression->getExprLoc(), + diag::err_non_bool_atomic_constraint) + << ConstraintExpression << ConstraintExpression->getType(); ---------------- saar.raz wrote: > rsmith wrote: > > What justifies rejecting this prior to any use of the concept that would > > result in a satisfaction check? > > > > (I think checking this is a good thing; what I'm wondering is whether we > > need to file a core issue to get the wording updated to allow us to reject > > such bogus concepts even if they're never used.) > I guess this is already justified, if awkwardly (NDR) by [temp.res]p8.2 Correction - it says that IFNDR occurs when no substitution would result in a valid expression, so maybe this is well formed after all. In this case it is a valid expression but not a valid constraint expression, maybe that's the missing word here? Repository: rC Clang https://reviews.llvm.org/D41217 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
