[Bug tree-optimization/64715] [5.0 regression] __builtin_object_size (..., 1) fails to locate subobject

["rguenth at gcc dot gnu.org" <gcc-bugzilla at gcc dot gnu dot org>]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=64715

Richard Biener <rguenth at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |ASSIGNED
   Last reconfirmed|                            |2015-01-21
          Component|c                           |tree-optimization
           Assignee|unassigned at gcc dot gnu.org      |rguenth at gcc dot gnu.org
     Ever confirmed|0                           |1

--- Comment #2 from Richard Biener <rguenth at gcc dot gnu.org> ---
Well.  In a.buf1 + (0 + 4) a.buf1 decays to a (char *) pointer and + 4 just
increments it.  You can hardly restrict that pointer arithmetic to a sub-object
this way.

Btw, this is likely caused by

2014-09-05  Richard Biener  <rguenther@suse.de>

        PR middle-end/63148
        * fold-const.c (try_move_mult_to_index): Remove.
        (fold_binary_loc): Do not call it.
        * tree-data-ref.c (dr_analyze_indices): Strip conversions
        from the base object again.

The docs say

@var{type} is an integer constant from 0 to 3.  If the least significant
bit is clear, objects are whole variables, if it is set, a closest
surrounding subobject is considered the object a pointer points to.

but I believe we may not actually compute that subobject (or rather the
issue, in some cases, is that the wording is ambiguous for nested
structs where both &a.b and &a.b.c are at the same address).  For the present
testcase we can unambiguously compute the closest surrounding subobject.

Let me see if I can fix this.