From owner-freebsd-current@freebsd.org Wed Aug 23 19:41:49 2017
Return-Path: <owner-freebsd-current@freebsd.org>
Delivered-To: freebsd-current@mailman.ysv.freebsd.org
Received: from mx1.freebsd.org (mx1.freebsd.org
[IPv6:2001:1900:2254:206a::19:1])
by mailman.ysv.freebsd.org (Postfix) with ESMTP id 58DEEDEDD8A
for <freebsd-current@mailman.ysv.freebsd.org>;
Wed, 23 Aug 2017 19:41:49 +0000 (UTC)
(envelope-from markmi@dsl-only.net)
Received: from asp.reflexion.net (outbound-mail-210-57.reflexion.net
[208.70.210.57])
(using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits))
(Client did not present a certificate)
by mx1.freebsd.org (Postfix) with ESMTPS id 087E17FCD5
for <freebsd-current@freebsd.org>; Wed, 23 Aug 2017 19:41:48 +0000 (UTC)
(envelope-from markmi@dsl-only.net)
Received: (qmail 8773 invoked from network); 23 Aug 2017 19:36:57 -0000
Received: from unknown (HELO rtc-sm-01.app.dca.reflexion.local) (10.81.150.1)
by 0 (rfx-qmail) with SMTP; 23 Aug 2017 19:36:57 -0000
Received: by rtc-sm-01.app.dca.reflexion.local
(Reflexion email security v8.40.2) with SMTP;
Wed, 23 Aug 2017 15:35:07 -0400 (EDT)
Received: (qmail 8611 invoked from network); 23 Aug 2017 19:35:07 -0000
Received: from unknown (HELO iron2.pdx.net) (69.64.224.71)
by 0 (rfx-qmail) with (AES256-SHA encrypted) SMTP; 23 Aug 2017 19:35:07 -0000
Received: from [192.168.1.109] (c-67-170-167-181.hsd1.or.comcast.net
[67.170.167.181])
by iron2.pdx.net (Postfix) with ESMTPSA id 5B25AEC9492;
Wed, 23 Aug 2017 12:35:06 -0700 (PDT)
From: Mark Millard <markmi@dsl-only.net>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: quoted-printable
Mime-Version: 1.0 (Mac OS X Mail 10.3 \(3273\))
Subject: Re: Segfault in _Unwind_* code called from pthread_exit
Message-Id: <9AD3F17A-1410-44FD-B688-ACFE6AFE38BF@dsl-only.net>
Date: Wed, 23 Aug 2017 12:35:05 -0700
To: Tijl Coosemans <tijl@FreeBSD.org>,
FreeBSD Current <freebsd-current@freebsd.org>
X-Mailer: Apple Mail (2.3273)
X-BeenThere: freebsd-current@freebsd.org
X-Mailman-Version: 2.1.23
Precedence: list
List-Id: Discussions about the use of FreeBSD-current
<freebsd-current.freebsd.org>
List-Unsubscribe: <https://lists.freebsd.org/mailman/options/freebsd-current>,
<mailto:freebsd-current-request@freebsd.org?subject=unsubscribe>
List-Archive: <http://lists.freebsd.org/pipermail/freebsd-current/>
List-Post: <mailto:freebsd-current@freebsd.org>
List-Help: <mailto:freebsd-current-request@freebsd.org?subject=help>
List-Subscribe: <https://lists.freebsd.org/mailman/listinfo/freebsd-current>,
<mailto:freebsd-current-request@freebsd.org?subject=subscribe>
X-List-Received-Date: Wed, 23 Aug 2017 19:41:49 -0000
Tijl Coosemans tijl at FreeBSD.org wrote on
Wed Aug 23 14:38:27 UTC 2017 :
> The following program segfaults for me on amd64 when linked like this:
>=20
> cc -o test test.c -lpthread -L/usr/local/lib/gcc5 -lgcc_s -rpath =
/usr/local/lib/gcc5
>=20
> --------------------------------
> #include <pthread.h>
> #include <stdio.h>
>=20
> void *
> thr( void *arg ) {
> return( NULL );
> }
>=20
> int
> main( void ) {
> pthread_t thread;
>=20
> for( int i =3D 1; i < 20; i++ ) {
> fprintf( stderr, "%d\n", i );
> pthread_create( &thread, NULL, thr, NULL );
> pthread_join( thread, NULL );
> }
> return( 0 );
> }
> --------------------------------
>=20
> The backtrace looks like this:
>=20
> Thread 7 received signal SIGSEGV, Segmentation fault.
> [Switching to LWP 100511 of process 1886]
> uw_frame_state_for (context=3D
> context at entry
> =3D0x7fffdfffddc0,=20
> fs=3D
> fs at entry
> =3D0x7fffdfffdb10)
> at /usr/ports/lang/gcc5/work/gcc-5.4.0/libgcc/unwind-dw2.c:1249
> 1249 /usr/ports/lang/gcc5/work/gcc-5.4.0/libgcc/unwind-dw2.c: No such =
file or directory.
> (gdb) bt
> #0 uw_frame_state_for (context=3D
> context at entry
> =3D0x7fffdfffddc0,=20
> fs=3D
> fs at entry
> =3D0x7fffdfffdb10)
> at /usr/ports/lang/gcc5/work/gcc-5.4.0/libgcc/unwind-dw2.c:1249
> #1 0x0000000800a66ecb in _Unwind_ForcedUnwind_Phase2 (
> exc=3D
> exc at entry=3D0x800658730, context=3Dcontext at entry
> =3D0x7fffdfffddc0)
> at /usr/ports/lang/gcc5/work/gcc-5.4.0/libgcc/unwind.inc:155
> #2 0x0000000800a67200 in _Unwind_ForcedUnwind (exc=3D0x800658730,=20
> stop=3D0x8008428b0 <thread_unwind_stop>, stop_argument=3D0x0)
> at /usr/ports/lang/gcc5/work/gcc-5.4.0/libgcc/unwind.inc:207
> #3 0x0000000800842224 in _Unwind_ForcedUnwind (ex=3D0x800658730,=20
> stop_func=3D0x8008428b0 <thread_unwind_stop>, stop_arg=3D0x0)
> at /usr/src/lib/libthr/thread/thr_exit.c:106
> #4 0x000000080084269f in thread_unwind ()
> at /usr/src/lib/libthr/thread/thr_exit.c:172
> #5 0x00000008008424d6 in _pthread_exit_mask (status=3D0x0, mask=3D0x0)
> at /usr/src/lib/libthr/thread/thr_exit.c:254
> #6 0x0000000800842359 in _pthread_exit (status=3D0x0)
> at /usr/src/lib/libthr/thread/thr_exit.c:206
> #7 0x000000080082ccb1 in thread_start (curthread=3D0x800658500)
> at /usr/src/lib/libthr/thread/thr_create.c:289
> #8 0x00007fffdfdfe000 in ?? ()
> Backtrace stopped: Cannot access memory at address 0x7fffdfffe000
>=20
>=20
> It happens with gcc6 as well, but not with base libgcc_s.
> Can anyone reproduce this? Have there been any changes to stack
> unwinding recently (last few months)?
This example might make a good addition to bugzilla 221288 that
has some material from a more complicated example of problems
mixing /usr/local/lib/gcc7/libgcc_s.so.1 and pthread. (Threading
need not be the only problem context.)
Here the source code is nice and short where the C++ example was
large enough that I did not bother to submit it and I've not made
a smaller example.
The bigger C++ example had:
# ldd a.out
a.out:
libstdc++.so.6 =3D> /usr/local/lib/gcc7/libstdc++.so.6 =
(0x800844000)
libm.so.5 =3D> /lib/libm.so.5 (0x800bd8000)
libgcc_s.so.1 =3D> /usr/local/lib/gcc7/libgcc_s.so.1 =
(0x800e05000)
libthr.so.3 =3D> /lib/libthr.so.3 (0x80101c000)
libc.so.7 =3D> /lib/libc.so.7 (0x801244000)
# ./a.out
. . . (omitted) . . .
Segmentation fault (core dumped)
It was the -Wl,-rpath=3D/usr/local/lib/gcc7 that forced the
gcc7 variant of libgcc_s to be used. Any combination that
had /lib/libthr.so.3 mixed with /usr/local/lib/gcc7/libgcc_s.so.1
failed. Any combination that had /lib/libthr.so.3 mixed with
/lib/libgcc_s.so.1 worked.
Of course /lib/libthr.so.3 was built based on /lib/libgcc_s.so.1 .
=3D=3D=3D
Mark Millard
markmi at dsl-only.net