I agree with David. This is a nice general cleanup to the API. Without this change, the issue could possibly resurface elsewhere.
On Apr 29, 2012, at 9:32 PM, David Blaikie <[email protected]> wrote: > Not strictly, no. My change fixes a surprising/broken API, Richard's > works around it. There's no harm in the workaround & arguably the > separation is probably nicer to read anyway (maybe faster, maybe > slower - it copies a CFGBlock unnecessarily - which is just two > pointers).
_______________________________________________ cfe-commits mailing list [email protected] http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits
