From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <gcc-bugs-return-386950-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org>
Received: (qmail 21289 invoked by alias); 23 Mar 2012 02:30:07 -0000
Received: (qmail 21236 invoked by uid 22791); 23 Mar 2012 02:30:05 -0000
X-SWARE-Spam-Status: No, hits=-2.9 required=5.0	tests=ALL_TRUSTED,AWL,BAYES_00
X-Spam-Check-By: sourceware.org
Received: from localhost (HELO gcc.gnu.org) (127.0.0.1)    by sourceware.org (qpsmtpd/0.43rc1) with ESMTP; Fri, 23 Mar 2012 02:29:50 +0000
From: "meadori at codesourcery dot com" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug c++/52672] internal compiler error: in cxx_eval_indirect_ref, at cp/semantics.c:6766
Date: Fri, 23 Mar 2012 05:52:00 -0000
X-Bugzilla-Reason: CC
X-Bugzilla-Type: changed
X-Bugzilla-Watch-Reason: None
X-Bugzilla-Product: gcc
X-Bugzilla-Component: c++
X-Bugzilla-Keywords:
X-Bugzilla-Severity: major
X-Bugzilla-Who: meadori at codesourcery dot com
X-Bugzilla-Status: UNCONFIRMED
X-Bugzilla-Priority: P3
X-Bugzilla-Assigned-To: unassigned at gcc dot gnu.org
X-Bugzilla-Target-Milestone: ---
X-Bugzilla-Changed-Fields: Version
Message-ID: <bug-52672-4-wUM8mpZA4P@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-52672-4@http.gcc.gnu.org/bugzilla/>
References: <bug-52672-4@http.gcc.gnu.org/bugzilla/>
X-Bugzilla-URL: http://gcc.gnu.org/bugzilla/
Auto-Submitted: auto-generated
Content-Type: text/plain; charset="UTF-8"
MIME-Version: 1.0
Mailing-List: contact gcc-bugs-help@gcc.gnu.org; run by ezmlm
Precedence: bulk
List-Id: <gcc-bugs.gcc.gnu.org>
List-Archive: <http://gcc.gnu.org/ml/gcc-bugs/>
List-Post: <mailto:gcc-bugs@gcc.gnu.org>
List-Help: <mailto:gcc-bugs-help@gcc.gnu.org>
Sender: gcc-bugs-owner@gcc.gnu.org
X-SW-Source: 2012-03/txt/msg02003.txt.bz2

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=52672

Meador Inge <meadori at codesourcery dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
            Version|4.6.4                       |4.8.0

--- Comment #1 from Meador Inge <meadori at codesourcery dot com> 2012-03-23 02:29:46 UTC ---
Actually I found a smaller repro that also triggers in trunk:

typedef unsigned long * ul_ptr;

constexpr void * a = (void*) ((ul_ptr)(((ul_ptr)0x01000010)[14]))[0];

The problem happens when 'cxx_eval_indirect_ref' is presented with an
indirect reference expression that looks like:

<indirect_ref 0xf7dd6018
    type <integer_type 0xf7c524e0 long unsigned int public unsigned SI
        size <integer_cst 0xf7c3b5cc constant 32>
        unit size <integer_cst 0xf7c3b5e8 constant 4>
        align 32 symtab 0 alias set -1 canonical type 0xf7c524e0 precision 32
min <integer_cst 0xf7c3b984 0> max <integer_cst 0xf7c3b968 4294967295>
        pointer_to_this <pointer_type 0xf7dc9cc0>>

    arg 0 <nop_expr 0xf7dd6000
        type <pointer_type 0xf7dc9d20 ul_ptr type <integer_type 0xf7c524e0 long
unsigned int>
            unsigned SI size <integer_cst 0xf7c3b5cc 32> unit size <integer_cst
0xf7c3b5e8 4>
            align 32 symtab 0 alias set -1 canonical type 0xf7dc9cc0>

        arg 0 <indirect_ref 0xf7da3fc0 type <integer_type 0xf7c524e0 long
unsigned int>

            arg 0 <integer_cst 0xf7cabe54 constant 16777288>
            /home/meadori/etc/bugs/issue13031.cpp:3:63>
        /home/meadori/etc/bugs/issue13031.cpp:3:68>
    /home/meadori/etc/bugs/issue13031.cpp:3:68>

It calls into 'cxx_fold_indirect_ref' to fold the indirect reference operand
and
ends up stripping the NOP cast ending up with:

<indirect_ref 0xf7da3fc0
    type <integer_type 0xf7c524e0 long unsigned int public unsigned SI
        size <integer_cst 0xf7c3b5cc constant 32>
        unit size <integer_cst 0xf7c3b5e8 constant 4>
        align 32 symtab 0 alias set -1 canonical type 0xf7c524e0 precision 32
min <integer_cst 0xf7c3b984 0> max <integer_cst 0xf7c3b968 4294967295>
        pointer_to_this <pointer_type 0xf7dc9cc0>>

    arg 0 <integer_cst 0xf7cabe54 type <pointer_type 0xf7dc9d20 ul_ptr>
constant 16777288>

which, of course, fails the 'gcc_assert (POINTER_TYPE_P (subtype));' assert
in 'cxx_fold_indirect_ref'.

I would like to work out a patch for this one.  Any tips would be appreciated.