[Bug jit/66812] jit code-generation example that unexpectedly required -fno-strict-aliasing to work

[dmalcolm at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66812

David Malcolm <dmalcolm at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |ASSIGNED
   Last reconfirmed|                            |2015-07-09
     Ever confirmed|0                           |1

--- Comment #7 from David Malcolm <dmalcolm at gcc dot gnu.org> ---
(In reply to Andrew Pinski from comment #6)
> This sounds like there is an aliasing set problem.
> 
> In the C front-end we have:
> 
> 
> 5174   /* Permit type-punning when accessing a union, provided the access
> 5175      is directly through the union.  For example, this code does not
> 5176      permit taking the address of a union member and then storing
> 5177      through it.  Even the type-punning allowed here is a GCC
> 5178      extension, albeit a common and useful one; the C standard says
> 5179      that such accesses have implementation-defined behavior.  */
> 5180   for (u = t;
> 5181        TREE_CODE (u) == COMPONENT_REF || TREE_CODE (u) == ARRAY_REF;
> 5182        u = TREE_OPERAND (u, 0))
> 5183     if (TREE_CODE (u) == COMPONENT_REF
> 5184         && TREE_CODE (TREE_TYPE (TREE_OPERAND (u, 0))) == UNION_TYPE)
> 5185       return 0;
> 5186 
> 
> 
> In c_common_get_alias_set .  Maybe that is missing on the libjit side.

Thanks.  Yes; I'd just figured that out, by single-stepping.