------- Comment #9 from mmitchel at gcc dot gnu dot org 2005-10-31 06:16 ------- Danny, when you refer to PR 24288, do you mean a different PR? I don't see the relevance of PR 24288, but I do remember discussing this issue with you and Jason.
Anyhow, for the time being, I think it's fair to punish the C++ front-end by disabling this optimization logic there. The type information we're giving you really isn't right. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=24351
