From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mx1.redhat.com ([209.132.183.28]:34352 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752536AbbC0Ni5 (ORCPT ); Fri, 27 Mar 2015 09:38:57 -0400 Received: from int-mx13.intmail.prod.int.phx2.redhat.com (int-mx13.intmail.prod.int.phx2.redhat.com [10.5.11.26]) by mx1.redhat.com (Postfix) with ESMTPS id 45CDC3C220 for ; Fri, 27 Mar 2015 13:38:57 +0000 (UTC) Date: Fri, 27 Mar 2015 21:38:55 +0800 From: Eryu Guan Subject: Re: [PATCH 5/9 RESEND] generic: test fs freeze/unfreeze and mount/umount race Message-ID: <20150327133855.GJ4810@dhcp-13-216.nay.redhat.com> References: <1426846618-23413-1-git-send-email-eguan@redhat.com> <1426846618-23413-6-git-send-email-eguan@redhat.com> <20150325194429.GE3436@laptop.bfoster> <20150327084651.GD4810@dhcp-13-216.nay.redhat.com> <20150327121002.GA3254@laptop.bfoster> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20150327121002.GA3254@laptop.bfoster> Sender: fstests-owner@vger.kernel.org To: Brian Foster Cc: fstests@vger.kernel.org List-ID: On Fri, Mar 27, 2015 at 08:10:02AM -0400, Brian Foster wrote: > On Fri, Mar 27, 2015 at 04:46:51PM +0800, Eryu Guan wrote: > > On Wed, Mar 25, 2015 at 03:44:29PM -0400, Brian Foster wrote: > > > On Fri, Mar 20, 2015 at 06:16:54PM +0800, Eryu Guan wrote: > > > > Exercise fs freeze/unfreeze and mount/umount race, which could lead to > > > > use-after-free oops. > > > > > > > > This commit fixed the issue: > > > > 1494583 fix get_active_super()/umount() race > > > > > > > > This test case is based on a script from Monakhov Dmitriy. > > > > > > > > Cc: Monakhov Dmitriy > > > > Signed-off-by: Eryu Guan > > > > --- > > > > tests/generic/080 | 98 +++++++++++++++++++++++++++++++++++++++++++++++++++ > > > > tests/generic/080.out | 2 ++ > > > > tests/generic/group | 1 + > > > > 3 files changed, 101 insertions(+) > > > > create mode 100755 tests/generic/080 > > > > create mode 100644 tests/generic/080.out > > > > > > > > diff --git a/tests/generic/080 b/tests/generic/080 > > > > new file mode 100755 > > > > index 0000000..7d9fa5c > > > > --- /dev/null > > > > +++ b/tests/generic/080 > > > > @@ -0,0 +1,98 @@ > > > > +#! /bin/bash > > > > +# FS QA Test No. 080 > > > > +# > > > > +# Exercise fs freeze/unfreeze and mount/umount race, which could lead to > > > > +# use-after-free oops. > > > > +# > > > > +# This commit fixed the issue: > > > > +# 1494583 fix get_active_super()/umount() race > > > > +# > > > > +#----------------------------------------------------------------------- > > > > +# Copyright (c) 2015 Red Hat Inc. All Rights Reserved. > > > > +# > > > > +# This program is free software; you can redistribute it and/or > > > > +# modify it under the terms of the GNU General Public License as > > > > +# published by the Free Software Foundation. > > > > +# > > > > +# This program is distributed in the hope that it would be useful, > > > > +# but WITHOUT ANY WARRANTY; without even the implied warranty of > > > > +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the > > > > +# GNU General Public License for more details. > > > > +# > > > > +# You should have received a copy of the GNU General Public License > > > > +# along with this program; if not, write the Free Software Foundation, > > > > +# Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA > > > > +#----------------------------------------------------------------------- > > > > +# > > > > + > > > > +seq=`basename $0` > > > > +seqres=$RESULT_DIR/$seq > > > > +echo "QA output created by $seq" > > > > + > > > > +here=`pwd` > > > > +tmp=/tmp/$$ > > > > +status=1 # failure is the default! > > > > +trap "_cleanup; exit \$status" 0 1 2 3 15 > > > > + > > > > +_cleanup() > > > > +{ > > > > + cd / > > > > + rm -f $tmp.* > > > > + cleanup_dmdev > > > > +} > > > > + > > > > +# get standard environment, filters and checks > > > > +. ./common/rc > > > > +. ./common/filter > > > > + > > > > +# real QA test starts here > > > > +_supported_fs generic > > > > +_supported_os Linux > > > > +_require_scratch > > > > +_require_block_device $SCRATCH_DEV > > > > +_require_command $DMSETUP_PROG > > > > +_require_freeze > > > > + > > > > +setup_dmdev() > > > > +{ > > > > + table="0 $size_in_sector linear $SCRATCH_DEV 0" > > > > + $DMSETUP_PROG create $node --table "$table" >>$seqres.full 2>&1 > > > > + $DMSETUP_PROG mknodes >/dev/null 2>&1 > > > > +} > > > > + > > > > +cleanup_dmdev() > > > > +{ > > > > + # in case it's still suspended and/or mounted > > > > + $DMSETUP_PROG resume $lvdev >/dev/null 2>&1 > > > > + $UMOUNT_PROG $lvdev >/dev/null 2>&1 > > > > + > > > > + $DMSETUP_PROG remove $node >>$seqres.full 2>&1 > > > > + $DMSETUP_PROG mknodes >/dev/null 2>&1 > > > > +} > > > > + > > > > +rm -f $seqres.full > > > > +echo "Silence is golden" > > > > + > > > > +size=$((256 * 1024 * 1024)) > > > > +size_in_sector=$((size / 512)) > > > > +_scratch_mkfs_sized $size >>$seqres.full 2>&1 > > > > + > > > > +node=$seq-test > > > > +lvdev=/dev/mapper/$node > > > > +setup_dmdev > > > > + > > > > +for ((i=0; i<100; i++)); do > > > > + $DMSETUP_PROG suspend $lvdev >/dev/null 2>&1 > > > > + $DMSETUP_PROG resume $lvdev >/dev/null 2>&1 > > > > +done & > > > > > > Any reason we can't use xfs_freeze/fsfreeze for this test? That seems > > > more simple (avoids the dm stuff), but I don't know enough about the > > > original problem to know if that still reproduces it. > > > > Yes, fsfreeze/unfreeze can reproduce the problem too, and that was > > how Monakhov Dmitriy reproduced it at first, running fstests with > > xfs_freeze -f/xfs_freeze -u loop in background. > > > > The problem is it's racing with umount so it's possible SCRATCH_DEV is > > not mounted at SCRATCH_MNT when xfs_freeze tries to freeze it, then the > > root fs is freezed instead, then the host just hangs there. > > > > Using dmsetup is safer, just freezes the device we want to test. > > > > Ah, I see. Sounds reasonable. Could you add some comments to explain why > the test is implemented this way? Sure, will add in v2. > > I'd also recommend some error checking for this test (e.g., the various > setup commands, the suspend/resume commands, etc.). OK, I'm fine with either way. Also I'll update the other test using lvm. Thanks! Eryu P.S. I still don't have a chance to look into the comments deeply in the "quota remount,ro" test, but will do. > > Brian > > > Thanks, > > Eryu > > > > > > Brian > > > > > > > +pid=$! > > > > +for ((i=0; i<100; i++)); do > > > > + $MOUNT_PROG $lvdev $SCRATCH_MNT >/dev/null 2>&1 > > > > + $UMOUNT_PROG $lvdev >/dev/null 2>&1 > > > > +done & > > > > +pid="$pid $!" > > > > + > > > > +wait $pid > > > > + > > > > +status=0 > > > > +exit > > > > diff --git a/tests/generic/080.out b/tests/generic/080.out > > > > new file mode 100644 > > > > index 0000000..11a4a1a > > > > --- /dev/null > > > > +++ b/tests/generic/080.out > > > > @@ -0,0 +1,2 @@ > > > > +QA output created by 080 > > > > +Silence is golden > > > > diff --git a/tests/generic/group b/tests/generic/group > > > > index ee5b642..cf7408c 100644 > > > > --- a/tests/generic/group > > > > +++ b/tests/generic/group > > > > @@ -82,6 +82,7 @@ > > > > 077 acl attr auto enospc > > > > 078 auto quick metadata > > > > 079 acl attr ioctl metadata auto quick > > > > +080 auto freeze mount > > > > 083 rw auto enospc stress > > > > 088 perms auto quick > > > > 089 metadata auto > > > > -- > > > > 2.1.0 > > > > > > > > -- > > > > To unsubscribe from this list: send the line "unsubscribe fstests" in > > > > the body of a message to majordomo@vger.kernel.org > > > > More majordomo info at http://vger.kernel.org/majordomo-info.html