The bug: http://gcc.gnu.org/bugzilla/show_bug.cgi?id=40092
has been around for a while (since May). A patch was submitted in a gcc-patches post about 1 month ago: http://gcc.gnu.org/ml/gcc-patches/2009-08/msg01217.html ; yet, no one has even confirmed that it's a bug. Would someone mind doing the confirmation with the test case supplied with the gcc-patches post? TIA. -regards, Larry