SRA problem with uninitialzed fields

[Richard Kenner]

    C++ enums were one case of restricted-precision types for which this 
    masking was not wanted, but for now 
    <http://gcc.gnu.org/ml/gcc-patches/2004-08/msg01985.html> and until a C++ 
    DR about this is resolved (see the long discussions on the gcc list last 
    month) they don't have the restricted precision.  

But what about the C test case I sent?  Isn't that valid C++ also and
wouldn't it have the problem I mentioned if that hook is false?