From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <gcc-patches-return-388378-listarch-gcc-patches=gcc.gnu.org@gcc.gnu.org>
Received: (qmail 13473 invoked by alias); 6 Jan 2015 09:08:28 -0000
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
Received: (qmail 13462 invoked by uid 89); 6 Jan 2015 09:08:27 -0000
Authentication-Results: sourceware.org; auth=none
X-Virus-Found: No
X-Spam-SWARE-Status: No, score=-1.0 required=5.0 tests=AWL,BAYES_00,FREEMAIL_FROM,RCVD_IN_DNSWL_NONE,SPF_PASS autolearn=ham version=3.3.2
X-HELO: DUB004-OMC4S1.hotmail.com
Received: from dub004-omc4s1.hotmail.com (HELO DUB004-OMC4S1.hotmail.com) (157.55.2.76) by sourceware.org (qpsmtpd/0.93/v0.84-503-g423c35a) with (AES256-SHA256 encrypted) ESMTPS; Tue, 06 Jan 2015 09:08:26 +0000
Received: from DUB118-W46 ([157.55.2.73]) by DUB004-OMC4S1.hotmail.com over TLS secured channel with Microsoft SMTPSVC(7.5.7601.22751);	 Tue, 6 Jan 2015 01:08:23 -0800
X-TMN: [gLiy3oPr1OBUSIRTXs1ZgtwbodXPSGs1]
Message-ID: <DUB118-W468072AA78C7E7CD1DCF6BE4590@phx.gbl>
From: Bernd Edlinger <bernd.edlinger@hotmail.de>
To: Mike Stump <mikestump@comcast.net>
CC: Jakub Jelinek <jakub@redhat.com>, "H.J. Lu" <hjl.tools@gmail.com>,	"gcc-patches@gcc.gnu.org" <gcc-patches@gcc.gnu.org>, Dmitry Vyukov	<dvyukov@google.com>
Subject: RE: [PATCH] Fix sporadic failure in g++.dg/tsan/aligned_vs_unaligned_race.C
Date: Tue, 06 Jan 2015 09:08:00 -0000
In-Reply-To: <DUB118-W1E901D082BACEDE02A5E8E4590@phx.gbl>
References: <DUB118-W421A4A9D8D09CBAF5103DAE45A0@phx.gbl>,<3C12133C-DABF-40FA-94F7-9DB785F6E914@comcast.net>,<DUB118-W292D09896C48BED7B29B14E45A0@phx.gbl>,<DUB118-W8F5E63DE36A8DB2DA1DF6E45B0@phx.gbl>,<623A5348-6FC9-4F7B-A9BC-B2B098AF7D37@comcast.net>,<20150104191658.GK1667@tucnak.redhat.com>,<DUB118-W450ACE7162969EFB27F817E45B0@phx.gbl>,<8E43F8AA-96BA-47A3-A886-C058459B4108@comcast.net>,<DUB118-W286F157F7B8B1EE05ED51FE4580@phx.gbl>,<E67B07D7-6ABA-48C1-B58B-B804144D91C2@comcast.net>,<D86529BC-DB94-481A-AE60-913D7E2B8D7F@comcast.net>,<DUB118-W1E901D082BACEDE02A5E8E4590@phx.gbl>
Content-Type: text/plain; charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
X-SW-Source: 2015-01/txt/msg00200.txt.bz2

Hi Mike,

after some hours of sleep I realized that your step function can do somethi=
ng very interesting,
(which you already requested previously):

That is: create a race condition that is _always_ at 100% missed by tsan:

cat lib.c
/* { dg-do compile } */
/* { dg-options "-O2 -fno-sanitize=3Dall" } */
=A0
static volatile int serial =3D 0;
=A0
void step (int i)
{
=A0 while (__atomic_load_n (&serial, __ATOMIC_ACQUIRE) !=3D i - 1);
=A0 __atomic_store_n (&serial, i, __ATOMIC_RELEASE);
}

cat tiny_race.c=20
/* { dg-shouldfail "tsan" } */

#include <pthread.h>

void step(int);

int Global;

void *Thread1(void *x) {
=A0 step (1);
=A0 Global =3D 42;
=A0 step (3);
=A0 return x;
}

int main() {
=A0 pthread_t t;
=A0 pthread_create(&t, 0, Thread1, 0);
=A0 step (2);
=A0 Global =3D 43;
=A0 step (4);
=A0 pthread_join(t, 0);
=A0 return Global;
}

/* { dg-output "WARNING: ThreadSanitizer: data race.*(\n|\r\n|\r)" } */



Bernd.
=20=09=09=20=09=20=20=20=09=09=20=20