[Bug c++/60713] [4.8/4.9 regression] ICE in iterative_hash_expr

[rguenth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=60713

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Priority|P3                          |P2
             Status|UNCONFIRMED                 |NEW
      Known to work|                            |4.7.3
           Keywords|                            |ice-on-valid-code
   Last reconfirmed|                            |2014-03-31
     Ever confirmed|0                           |1
   Target Milestone|---                         |4.8.3
      Known to fail|                            |4.8.0

--- Comment #1 from Richard Biener <rguenth at gcc dot gnu.org> ---
Confirmed.  We have a lang-specifc tree in the GENERIC fed to gimplification:

 <call_expr 0x7ffff6d74680
...
    arg 1 <ptrmem_cst 0x7ffff6d8bdc0
        type <record_type 0x7ffff6d91dc8 TI
            size <integer_cst 0x7ffff6c40140 constant 128>
            unit size <integer_cst 0x7ffff6c40160 constant 16>
            align 64 symtab 0 alias set -1 canonical type 0x7ffff6d91dc8 fields
<field_decl 0x7ffff6d92be0 __pfn> ptrmemfunc fn type <pointer_type
0x7ffff6d91d20>
            chain <type_decl 0x7ffff6d93170 __ptrmemfunc_type>>
        constant>
    t.ii:19:26>