Thanks for reporting this. I expect this is a GCC 4.7 bug, and have filed a bug report. I pushed the following workaround into Gnulib. If I'm wrong and it's not a GCC bug then we can revisit this.
stdalign: check that alignof and offsetof are consistent * m4/stdalign.m4 (gl_STDALIGN_H): Check for GCC bug 52023. Problem reported for gnulib by Richard W.M. Jones in <http://lists.gnu.org/archive/html/bug-gnulib/2012-01/msg00340.html>. diff --git a/m4/stdalign.m4 b/m4/stdalign.m4 index 9752ba5..fa6d5d5 100644 --- a/m4/stdalign.m4 +++ b/m4/stdalign.m4 @@ -14,13 +14,26 @@ AC_DEFUN([gl_STDALIGN_H], [AC_COMPILE_IFELSE( [AC_LANG_PROGRAM( [[#include <stdalign.h> - int align_int = alignof (int) + _Alignof (double); + #include <stddef.h> + + /* Test that alignof yields a result consistent with offsetof. + This catches GCC bug 52023 + <http://gcc.gnu.org/bugzilla/show_bug.cgi?id=52023>. */ + #ifdef __cplusplus + template <class t> struct alignof_helper { char a; t b; }; + # define ao(type) offsetof (alignof_helper<type>, b) + #else + # define ao(type) offsetof (struct { char a; type b; }, b) + #endif + char test1[_Alignof (double) == ao (double) ? 1 : -1]; + char test2[alignof (long int) == ao (long int) ? 1 : -1]; /* Test _Alignas only on platforms where gnulib can help. */ #if \ (__GNUC__ || __IBMC__ || __IBMCPP__ \ || 0x5110 <= __SUNPRO_C || 1300 <= _MSC_VER) int alignas (8) alignas_int = 1; + char test3[8 <= _Alignof (alignas_int) ? 1 : -1]; #endif ]])], [gl_cv_header_working_stdalign_h=yes],
