[Bug tree-optimization/100732] [11/12 Regression] ICE with -Wall: in ao_ref_init_from_ptr_and_range, at tree-ssa-alias.c:824 since r11-959-gb825a22890740f34

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

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100732

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Priority|P3                          |P2
   Target Milestone|---                         |11.2

--- Comment #2 from Richard Biener <rguenth at gcc dot gnu.org> ---
#3  0x0000000001649007 in maybe_warn_pass_by_reference (stmt=0x7ffff7ff5cf0, 
    wlims=...) at /home/rguenther/src/gcc3/gcc/tree-ssa-uninit.c:545
545           ao_ref_init_from_ptr_and_size (&ref, arg, access_size);
(gdb) p arg
$2 = <ssa_name 0x7ffff6575d80 2>
(gdb) p debug_tree (arg)
 <ssa_name 0x7ffff6575d80
    type <integer_type 0x7ffff65835e8 int public SI
        size <integer_cst 0x7ffff656af18 constant 32>
        unit-size <integer_cst 0x7ffff656af30 constant 4>
        align:32 warn_if_not_align:0 symtab:0 alias-set -1 canonical-type
0x7ffff65835e8 precision:32 min <integer_cst 0x7ffff656aed0 -2147483648> max
<integer_cst 0x7ffff656aee8 2147483647>
        pointer_to_this <pointer_type 0x7ffff658b9d8>>
    visited var <parm_decl 0x7ffff66c4000 i>
    def_stmt GIMPLE_NOP
    version:2>

that's not a pointer.  Looks like the code blindly works on call arguments
w/o double-checking their type.