From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <gcc-patches-return-292297-listarch-gcc-patches=gcc.gnu.org@gcc.gnu.org>
Received: (qmail 14598 invoked by alias); 20 May 2011 14:18:02 -0000
Received: (qmail 14563 invoked by uid 22791); 20 May 2011 14:18:02 -0000
X-SWARE-Spam-Status: No, hits=-2.3 required=5.0	tests=AWL,BAYES_00,DKIM_SIGNED,DKIM_VALID,DKIM_VALID_AU,FREEMAIL_FROM,RCVD_IN_DNSWL_LOW,RFC_ABUSE_POST,TW_TM
X-Spam-Check-By: sourceware.org
Received: from mail-wy0-f175.google.com (HELO mail-wy0-f175.google.com) (74.125.82.175)    by sourceware.org (qpsmtpd/0.43rc1) with ESMTP; Fri, 20 May 2011 14:17:48 +0000
Received: by wye20 with SMTP id 20so3327853wye.20        for <gcc-patches@gcc.gnu.org>; Fri, 20 May 2011 07:17:47 -0700 (PDT)
MIME-Version: 1.0
Received: by 10.227.174.79 with SMTP id s15mr4343092wbz.76.1305901067078; Fri, 20 May 2011 07:17:47 -0700 (PDT)
Received: by 10.227.38.129 with HTTP; Fri, 20 May 2011 07:17:47 -0700 (PDT)
In-Reply-To: <4DD67554.5050904@qnx.com>
References: <4DD67554.5050904@qnx.com>
Date: Fri, 20 May 2011 15:11:00 -0000
Message-ID: <BANLkTi=z6dpkz7ZmvJ+QZnv_hnW1W4+Vng@mail.gmail.com>
Subject: Re: Fix crash in vect_is_slp_reduction
From: Richard Guenther <richard.guenther@gmail.com>
To: Ryan Mansfield <rmansfield@qnx.com>
Cc: gcc-patches@gcc.gnu.org
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable
X-IsSubscribed: yes
Mailing-List: contact gcc-patches-help@gcc.gnu.org; run by ezmlm
Precedence: bulk
List-Id: <gcc-patches.gcc.gnu.org>
List-Archive: <http://gcc.gnu.org/ml/gcc-patches/>
List-Post: <mailto:gcc-patches@gcc.gnu.org>
List-Help: <mailto:gcc-patches-help@gcc.gnu.org>
Sender: gcc-patches-owner@gcc.gnu.org
X-SW-Source: 2011-05/txt/msg01468.txt.bz2

On Fri, May 20, 2011 at 4:06 PM, Ryan Mansfield <rmansfield@qnx.com> wrote:
> There is a crash in vect_is_slp_reduction where use_stmt doesn't get
> initialized in the FOR_EACH_IMM_USE_FAST loop.
>
> 1718 =A0 =A0 =A0 =A0 =A0FOR_EACH_IMM_USE_FAST (use_p, imm_iter, lhs)
> (gdb) call debug_tree(lhs)
> =A0<ssa_name 0x7ffff6e3a5a0
> =A0 =A0type <integer_type 0x7ffff7f8c7e0 int32_t sizes-gimplified public =
SI
> =A0 =A0 =A0 =A0size <integer_cst 0x7ffff7ec47e0 constant 32>
> =A0 =A0 =A0 =A0unit size <integer_cst 0x7ffff7ec4580 constant 4>
> =A0 =A0 =A0 =A0align 32 symtab 0 alias set -1 canonical type 0x7ffff7ecb4=
98
> precision 32 min <integer_cst 0x7ffff7ec4780 -2147483648> max <integer_cst
> 0x7ffff7ec47a0 2147483647> context <translation_unit_decl 0x7ffff7f98508
> D.1314>
> =A0 =A0 =A0 =A0pointer_to_this <pointer_type 0x7ffff7f9f930>>
> =A0 =A0var <var_decl 0x7ffff6e1ce60 g_29_lsm.18>def_stmt g_29_lsm.18_137 =
=3D PHI
> <g_29_lsm.18_87(15), D.2113_96(18)>
>
> =A0 =A0version 137>
> (gdb) n
> 1742 =A0 =A0 =A0 =A0 =A0if (found)
> (gdb)
> 1746 =A0 =A0 =A0 =A0 =A0if (gimple_code (use_stmt) =3D=3D GIMPLE_PHI)
> (gdb)
>
> Program received signal SIGSEGV, Segmentation fault.
> 0x0000000000980f22 in gimple_code (g=3D0x0) at ../../gcc/gimple.h:1098
> 1098 =A0 =A0 =A0return g->gsbase.code;
>
> Is the following sufficient?

No, we shouldn't arrive with a NULL use_stmt here.

Richard.

> Index: tree-vect-loop.c
> =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
> --- tree-vect-loop.c =A0 =A0(revision 173945)
> +++ tree-vect-loop.c =A0 =A0(working copy)
> @@ -1743,7 +1743,7 @@
> =A0 =A0 =A0 =A0 break;
>
> =A0 =A0 =A0 /* This is a loop exit phi, and we haven't reached the reduct=
ion phi.
> =A0*/
> - =A0 =A0 =A0if (gimple_code (use_stmt) =3D=3D GIMPLE_PHI)
> + =A0 =A0 =A0if (!use_stmt || gimple_code (use_stmt) =3D=3D GIMPLE_PHI)
> =A0 =A0 =A0 =A0 return false;
>
> =A0 =A0 =A0 if (!is_gimple_assign (use_stmt)
>
> Regards,
>
> Ryan Mansfield
>