All of lore.kernel.org
 help / color / mirror / Atom feed
From: Eryu Guan <eguan@redhat.com>
To: fstests@vger.kernel.org
Cc: Eryu Guan <eguan@redhat.com>
Subject: [PATCH v2 3/9] generic: test hardlink to unlinked file
Date: Wed,  1 Apr 2015 21:48:40 +0800	[thread overview]
Message-ID: <1427896120-11520-1-git-send-email-eguan@redhat.com> (raw)
In-Reply-To: <1426846618-23413-4-git-send-email-eguan@redhat.com>

Kernel commit

aae8a97 fs: Don't allow to create hardlink for deleted file

disabled hardlink to unlinked file.

Test the race between link and unlink, which could end up adding link
count to an unlinked file and leading to fs corruption on xfs.

Test case was originally written by Eric Sandeen.

Signed-off-by: Eryu Guan <eguan@redhat.com>
---
v2:
- create different link file in each iteration in link_unlink_storm(), which
  is confirmed to be able to reproduce the original corruption
- remove all newly created & tail'ed files at last, out of the loop, to avoid
  tail & rm race
- iterate $nr_cpu times to do the link_unlink_storm, not a fixed number
- add to quick group

 tests/generic/073     | 102 ++++++++++++++++++++++++++++++++++++++++++++++++++
 tests/generic/073.out |   2 +
 tests/generic/group   |   1 +
 3 files changed, 105 insertions(+)
 create mode 100755 tests/generic/073
 create mode 100644 tests/generic/073.out

diff --git a/tests/generic/073 b/tests/generic/073
new file mode 100755
index 0000000..6fd8c89
--- /dev/null
+++ b/tests/generic/073
@@ -0,0 +1,102 @@
+#! /bin/bash
+# FS QA Test No. 073
+#
+# Test hardlink to unlinked file.
+#
+# Regression test for commit:
+# aae8a97 fs: Don't allow to create hardlink for deleted file
+#
+#-----------------------------------------------------------------------
+# 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.*
+}
+
+# get standard environment, filters and checks
+. ./common/rc
+. ./common/filter
+
+# real QA test starts here
+_supported_fs generic
+_supported_os Linux
+_require_scratch
+
+link_unlink_storm()
+{
+	local src=$1
+	local target=$2
+	local i=0
+
+	while true; do
+		ln -f $src $target.$i >/dev/null 2>&1
+		rm -f $target.$i >/dev/null 2>&1
+		let i=i+1
+	done
+}
+
+rm -f $seqres.full
+nr_cpu=`$here/src/feature -o`
+echo "Silence is golden"
+
+_scratch_mkfs >>$seqres.full 2>&1
+_scratch_mount
+
+# create, open & unlinked files so unlinked inode list is not empty
+for i in `seq 1 $nr_cpu`; do
+	testfile=$SCRATCH_MNT/$seq.unlinked.$i
+	touch $testfile
+	tail -f $testfile &
+	tail_pids="$tail_pids $!"
+done
+rm -f $SCRATCH_MNT/$seq.unlinked.*
+
+# start link/unlink storm
+src=$SCRATCH_MNT/$seq.target
+touch $src
+for i in `seq 1 $nr_cpu`; do
+	target=$SCRATCH_MNT/$seq.target.link.$i
+	link_unlink_storm $src $target &
+	link_pids="$link_pids $!"
+done
+
+# remove & re-create target to race with link/unlink
+while true; do
+	rm -f $src
+	touch $src
+done &
+sleep 5
+kill $! >/dev/null 2>&1
+
+kill $tail_pids $link_pids >/dev/null 2>&1
+wait $tail_pids $link_pids
+
+# all done, no oops/hang expected, _check_filesystems checks SCRATCH_DEV after test
+status=0
+exit
diff --git a/tests/generic/073.out b/tests/generic/073.out
new file mode 100644
index 0000000..d107704
--- /dev/null
+++ b/tests/generic/073.out
@@ -0,0 +1,2 @@
+QA output created by 073
+Silence is golden
diff --git a/tests/generic/group b/tests/generic/group
index c0210d2..ee5b642 100644
--- a/tests/generic/group
+++ b/tests/generic/group
@@ -75,6 +75,7 @@
 070 attr udf auto quick stress
 071 auto quick prealloc
 072 auto quick quota
+073 auto metadata quick
 074 rw udf auto
 075 rw udf auto quick
 076 metadata rw udf auto quick stress
-- 
2.1.0


  parent reply	other threads:[~2015-04-01 13:52 UTC|newest]

Thread overview: 41+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-03-20 10:16 [PATCH 0/9] some backlog patches and new test and fixes Eryu Guan
2015-03-20 10:16 ` [PATCH 1/9 RESEND] generic: test some mount/umount corner cases Eryu Guan
2015-03-25 19:32   ` Brian Foster
2015-03-20 10:16 ` [PATCH 2/9 RESEND] generic: test quota handling on remount ro failure Eryu Guan
2015-03-25 19:32   ` Brian Foster
2015-04-01  6:06     ` Eryu Guan
2015-04-01 14:13       ` Brian Foster
2015-03-20 10:16 ` [PATCH 3/9 RESEND] generic: test hardlink to unlinked file Eryu Guan
2015-03-25 19:32   ` Brian Foster
2015-03-27 10:10     ` Eryu Guan
2015-04-01 13:48   ` Eryu Guan [this message]
2015-04-01 17:54     ` [PATCH v2 3/9] " Brian Foster
2015-04-02  2:59       ` Eryu Guan
2015-04-02 11:17         ` Brian Foster
2015-03-20 10:16 ` [PATCH 4/9 RESEND] shared: test truncate orphan inodes when mounting extN Eryu Guan
2015-03-20 10:16 ` [PATCH 5/9 RESEND] generic: test fs freeze/unfreeze and mount/umount race Eryu Guan
2015-03-25 19:44   ` Brian Foster
2015-03-27  8:46     ` Eryu Guan
2015-03-27 12:10       ` Brian Foster
2015-03-27 13:38         ` Eryu Guan
2015-04-01 13:53   ` [PATCH v2 5/9] " Eryu Guan
2015-04-01 17:56     ` Brian Foster
2015-04-02 12:15       ` Eryu Guan
2015-03-20 10:16 ` [PATCH 6/9] generic: test I/O error path by fully filling dm snapshot Eryu Guan
2015-03-25 23:08   ` Brian Foster
2015-03-27  7:38     ` Eryu Guan
2015-03-27 12:10       ` Brian Foster
2015-04-01 13:54   ` Eryu Guan
2015-04-01 17:57     ` Brian Foster
2015-04-01 22:29       ` Dave Chinner
2015-04-02  4:10         ` Eryu Guan
2015-04-02 11:11           ` Brian Foster
2015-04-07  1:36             ` Dave Chinner
2015-04-01 22:41     ` Dave Chinner
2015-04-02 12:08       ` Eryu Guan
2015-03-20 10:16 ` [PATCH 7/9] common: recognise NFS export over IPv6 in _require_scratch_nocheck() Eryu Guan
2015-03-25 23:08   ` Brian Foster
2015-03-20 10:16 ` [PATCH 8/9] xfs/014: replace df with $DF_PROG Eryu Guan
2015-03-25 23:08   ` Brian Foster
2015-03-20 10:16 ` [PATCH 9/9] generic/077: add missing _require_scratch Eryu Guan
2015-03-25 23:08   ` Brian Foster

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1427896120-11520-1-git-send-email-eguan@redhat.com \
    --to=eguan@redhat.com \
    --cc=fstests@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.