------- Additional Comments From joseph at codesourcery dot com 2005-07-18 10:59 ------- Subject: Re: [4.1 Regression] [DR236] gcc.c-torture/execute/20000603-1.c execution, -O2 fails
On Mon, 18 Jul 2005, zadeck at naturalbridge dot com wrote: > It looks pretty clear to me that this test case should be non conforming. It > is > almost exactly example 2 of the committee discussion and that case is > "invalid". You're not taking account of the common initial sequence rule and DR#257. The types in the example in DR#236 aren't structures with a common initial sequence. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=22534
