They aren't, but they shouldn't be reported in 'sizeof' expressions either. DiagRuntimeBehavior also culls out those cases.
On Feb 22, 2011, at 6:30 PM, Chris Lattner wrote: > > On Feb 22, 2011, at 6:27 PM, Ted Kremenek wrote: > >> I'll update with comments. The Stmt* is optional because we don't want to >> always result to using the reachability analysis to prune warnings (e.g., >> "unused value" warnings). > > Unused value warnings aren't "runtime behavior" warnings though, right? > > -Chris _______________________________________________ cfe-commits mailing list [email protected] http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits
