From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <gdb-return-42170-listarch-gdb=sources.redhat.com@sourceware.org>
Received: (qmail 32613 invoked by alias); 22 May 2013 09:00:17 -0000
Mailing-List: contact gdb-help@sourceware.org; run by ezmlm
Precedence: bulk
List-Id: <gdb.sourceware.org>
List-Subscribe: <mailto:gdb-subscribe@sourceware.org>
List-Archive: <http://sourceware.org/ml/gdb/>
List-Post: <mailto:gdb@sourceware.org>
List-Help: <mailto:gdb-help@sourceware.org>, <http://sourceware.org/ml/#faqs>
Sender: gdb-owner@sourceware.org
Received: (qmail 32540 invoked by uid 89); 22 May 2013 09:00:10 -0000
X-Spam-SWARE-Status: No, score=-0.4 required=5.0 tests=AWL,BAYES_00,RCVD_VIA_APNIC autolearn=ham version=3.3.1
Received: from cnshjsmin03.alcatel-sbell.com.cn (HELO cnshjsmin03.alcatel-sbell.com.cn) (211.144.215.47)    by sourceware.org (qpsmtpd/0.84/v0.84-167-ge50287c) with ESMTP; Wed, 22 May 2013 09:00:09 +0000
Received: from CNSHJCASHUB02.ad4.ad.alcatel.com (Unknown_Domain [135.251.50.72])	by cnshjsmin03.alcatel-sbell.com.cn (Symantec Messaging Gateway) with SMTP id 77.09.04680.6198C915; Wed, 22 May 2013 17:00:07 +0800 (HKT)
Received: from CNSHJMBX01.ad4.ad.alcatel.com ([135.251.50.101]) by CNSHJCASHUB02.ad4.ad.alcatel.com ([135.251.50.72]) with mapi id 14.03.0123.003; Wed, 22 May 2013 17:00:06 +0800
From: XIAO Wei si A <Wei_Si.a.Xiao@alcatel-lucent.com>
To: "gdb@sourceware.org" <gdb@sourceware.org>
Subject: RE: GDB step command crashed on non-stop mode
Date: Wed, 22 May 2013 09:00:00 -0000
Message-ID: <01E5AB929A54FD488EEFBD6595F260AF380B131E@cnshjmbx01>
Content-Type: text/plain; charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
X-CFilter-Loop: Reflected
X-SW-Source: 2013-05/txt/msg00097.txt.bz2

Hello,

I am using GDB non-stop mode to debug a multi-thread application. But it al=
ways crashed when I use "step" command. I have tried version 7.4/7.5.1/7.6,=
 there result are same. The issue does not exist if we don't use non-stop m=
ode.
I have written a small program to test, easy to reproduce each time. The co=
de as below:

// test.cc=20
#include <string>
using namespace std;
void fun(string &str1, string &str2)
{
=A0=A0=A0 str1 +=3D str2;
}

int main(void)
{
=A0=A0 string str1 =3D "abc";
=A0=A0 string str2 =3D "def";
=A0=A0 fun(str1,str2);
=A0=A0=20
=A0=A0 return 0;
}


Compile:
g++ -g test.cc -o test


The procedure for reproduce as below:
vm10-0-0-1:/root-# gdb ./test
GNU gdb (GDB) 7.6
Copyright (C) 2013 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.htm=
l>
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law.=A0 Type "show copying"
and "show warranty" for details.
This GDB was configured as "x86_64-unknown-linux-gnu".
For bug reporting instructions, please see:
<http://www.gnu.org/software/gdb/bugs/>...
Reading symbols from /root/test...done.
(gdb) set target-async 1
(gdb) set non-stop on
(gdb) break fun=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=
=A0=A0=A0=A0=A0=A0=A0=A0=20
Breakpoint 1 at 0x4007b4: file test.cc, line 5.
(gdb) run=20
Starting program: /root/test=20

Breakpoint 1, fun (str1=3D..., str2=3D...) at test.cc:5
5=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 str1 +=3D str2;
(gdb) step
6=A0=A0=A0=A0=A0=A0 }
(gdb) step

Program received signal SIGSEGV, Segmentation fault.
fun (str1=3D<error reading variable: Cannot access memory at address 0xffff=
fffffffffff8>, str1@entry=3D<error reading variable: Cannot access memory a=
t address 0x8>,=20
=A0=A0=A0 str2=3D<error reading variable: Cannot access memory at address 0=
xfffffffffffffff0>, str2@entry=3D<error reading variable: Cannot access mem=
ory at address 0x8>) at test.cc:6
6=A0=A0=A0=A0=A0=A0 }
(gdb)


Does anyone know the root cause?

Thanks in advance,
Weisi