From: Nikolay Borisov <nborisov@suse.com>
To: linux-xfs@vger.kernel.org
Cc: fstests@vger.kernel.org, darrick.wong@oracle.com,
eguan@redhat.com, Nikolay Borisov <nborisov@suse.com>
Subject: [PATCH v2 4/4] xfs: initial fiemap range query test
Date: Tue, 31 Oct 2017 16:11:36 +0200 [thread overview]
Message-ID: <1509459096-7585-4-git-send-email-nborisov@suse.com> (raw)
In-Reply-To: <1509459096-7585-1-git-send-email-nborisov@suse.com>
Signed-off-by: Nikolay Borisov <nborisov@suse.com>
---
tests/xfs/900 | 96 ++++++++++++++++++++++++++++++++++++++++++++
tests/xfs/900.out | 118 ++++++++++++++++++++++++++++++++++++++++++++++++++++++
2 files changed, 214 insertions(+)
create mode 100755 tests/xfs/900
create mode 100644 tests/xfs/900.out
diff --git a/tests/xfs/900 b/tests/xfs/900
new file mode 100755
index 0000000..a81f85f
--- /dev/null
+++ b/tests/xfs/900
@@ -0,0 +1,96 @@
+#! /bin/bash
+# FS QA Test No. 900
+#
+# Regression test for an XFS NULL xattr buffer problem during unlink. XFS had a
+# bug where the attr fork walk during file removal could go off the rails due to
+# a stale reference to content of a released buffer. Memory pressure could cause
+# this reference to point to free or reused memory and cause subsequent
+# attribute fork lookups to fail, return a NULL buffer and possibly crash.
+#
+# This test emulates this behavior using an error injection knob to explicitly
+# disable buffer LRU caching. This forces the attr walk to execute under
+# conditions where each buffer is immediately freed on release.
+#
+# Commit f35c5e10c6ed ("xfs: reinit btree pointer on attr tree inactivation
+# walk") fixed the bug.
+#
+#-----------------------------------------------------------------------
+# Copyright (c) 2017 SUSE Linux Products GmbH. 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/punch
+
+# remove previous $seqres.full before test
+rm -f $seqres.full
+
+# real QA test starts here
+
+# Modify as appropriate.
+_supported_fs generic
+_supported_os Linux
+_require_scratch
+_require_xfs_io_command "falloc"
+_require_xfs_io_command "fiemap" "0 4k"
+
+_scratch_mkfs > $seqres.full 2>&1
+_scratch_mount || _fail "mount failure"
+
+file=$SCRATCH_MNT/testfile
+$XFS_IO_PROG -f -c "falloc 0 256k" $file
+for i in {0..31}; do $XFS_IO_PROG -c "fpunch $(($i*8))k 4k" $file; done
+
+# Query 1 data extent between 4k..8k range
+echo "Basic data extent"
+$XFS_IO_PROG -c "fiemap -v 4k 4k" $file | _filter_fiemap
+
+# Query data and hole extent
+echo "Data + Hole"
+$XFS_IO_PROG -c "fiemap -v 4k 5k" $file | _filter_fiemap
+
+echo "Hole + Data + Hole"
+$XFS_IO_PROG -c "fiemap -v 8k 10k" $file | _filter_fiemap
+
+echo "Beginning with a hole"
+$XFS_IO_PROG -c "fiemap -v 0 3k" $file | _filter_fiemap
+
+# Query for 0..160k that's 40 extents, more than the EXTENT_BATCH
+echo "Query more than 32 extents"
+$XFS_IO_PROG -c "fiemap -v 0 160k" $file | _filter_fiemap
+
+echo "Larger query than file size"
+$XFS_IO_PROG -c "fiemap -v 0 300k" $file | _filter_fiemap
+
+# success, all done
+status=0
+exit
diff --git a/tests/xfs/900.out b/tests/xfs/900.out
new file mode 100644
index 0000000..acd309e
--- /dev/null
+++ b/tests/xfs/900.out
@@ -0,0 +1,118 @@
+QA output created by 900
+Basic data extent
+0: [8..15]: unwritten
+Data + Hole
+0: [8..15]: unwritten
+1: [16..23]: hole
+Hole + Data + Hole
+0: [16..23]: hole
+1: [24..31]: unwritten
+2: [32..39]: hole
+Beginning with a hole
+0: [0..7]: hole
+Query more than 32 extents
+0: [0..7]: hole
+1: [8..15]: unwritten
+2: [16..23]: hole
+3: [24..31]: unwritten
+4: [32..39]: hole
+5: [40..47]: unwritten
+6: [48..55]: hole
+7: [56..63]: unwritten
+8: [64..71]: hole
+9: [72..79]: unwritten
+10: [80..87]: hole
+11: [88..95]: unwritten
+12: [96..103]: hole
+13: [104..111]: unwritten
+14: [112..119]: hole
+15: [120..127]: unwritten
+16: [128..135]: hole
+17: [136..143]: unwritten
+18: [144..151]: hole
+19: [152..159]: unwritten
+20: [160..167]: hole
+21: [168..175]: unwritten
+22: [176..183]: hole
+23: [184..191]: unwritten
+24: [192..199]: hole
+25: [200..207]: unwritten
+26: [208..215]: hole
+27: [216..223]: unwritten
+28: [224..231]: hole
+29: [232..239]: unwritten
+30: [240..247]: hole
+31: [248..255]: unwritten
+32: [256..263]: hole
+33: [264..271]: unwritten
+34: [272..279]: hole
+35: [280..287]: unwritten
+36: [288..295]: hole
+37: [296..303]: unwritten
+38: [304..311]: hole
+39: [312..319]: unwritten
+Larger query than file size
+0: [0..7]: hole
+1: [8..15]: unwritten
+2: [16..23]: hole
+3: [24..31]: unwritten
+4: [32..39]: hole
+5: [40..47]: unwritten
+6: [48..55]: hole
+7: [56..63]: unwritten
+8: [64..71]: hole
+9: [72..79]: unwritten
+10: [80..87]: hole
+11: [88..95]: unwritten
+12: [96..103]: hole
+13: [104..111]: unwritten
+14: [112..119]: hole
+15: [120..127]: unwritten
+16: [128..135]: hole
+17: [136..143]: unwritten
+18: [144..151]: hole
+19: [152..159]: unwritten
+20: [160..167]: hole
+21: [168..175]: unwritten
+22: [176..183]: hole
+23: [184..191]: unwritten
+24: [192..199]: hole
+25: [200..207]: unwritten
+26: [208..215]: hole
+27: [216..223]: unwritten
+28: [224..231]: hole
+29: [232..239]: unwritten
+30: [240..247]: hole
+31: [248..255]: unwritten
+32: [256..263]: hole
+33: [264..271]: unwritten
+34: [272..279]: hole
+35: [280..287]: unwritten
+36: [288..295]: hole
+37: [296..303]: unwritten
+38: [304..311]: hole
+39: [312..319]: unwritten
+40: [320..327]: hole
+41: [328..335]: unwritten
+42: [336..343]: hole
+43: [344..351]: unwritten
+44: [352..359]: hole
+45: [360..367]: unwritten
+46: [368..375]: hole
+47: [376..383]: unwritten
+48: [384..391]: hole
+49: [392..399]: unwritten
+50: [400..407]: hole
+51: [408..415]: unwritten
+52: [416..423]: hole
+53: [424..431]: unwritten
+54: [432..439]: hole
+55: [440..447]: unwritten
+56: [448..455]: hole
+57: [456..463]: unwritten
+58: [464..471]: hole
+59: [472..479]: unwritten
+60: [480..487]: hole
+61: [488..495]: unwritten
+62: [496..503]: hole
+63: [504..511]: unwritten
--
2.7.4
next prev parent reply other threads:[~2017-10-31 14:12 UTC|newest]
Thread overview: 31+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-10-27 12:37 [PATCH v2 0/2] Fiemap's range query Nikolay Borisov
2017-10-27 12:37 ` [PATCH 1/2] fiemap: Factor out actual fiemap call code Nikolay Borisov
2017-10-27 16:33 ` Darrick J. Wong
2017-10-27 16:37 ` Darrick J. Wong
2017-10-27 12:37 ` [PATCH 1/2] initial fiemap test Nikolay Borisov
2017-10-27 12:44 ` Eryu Guan
2017-10-31 9:22 ` Nikolay Borisov
2017-10-31 9:32 ` Eryu Guan
2017-10-31 12:27 ` Nikolay Borisov
2017-10-27 12:37 ` [PATCH 2/2] fiemap: Implement ranged query Nikolay Borisov
2017-10-27 12:37 ` [PATCH 2/2] generic: Adjust generic test ouputs for new fiemap implementation Nikolay Borisov
2017-10-27 12:46 ` Eryu Guan
2017-10-27 12:48 ` Nikolay Borisov
2017-10-31 14:10 ` [PATCH v2 1/2] fiemap: Factor out actual fiemap call code Nikolay Borisov
2017-10-31 14:10 ` [PATCH v2 2/2] fiemap: Implement ranged query Nikolay Borisov
2017-10-31 14:11 ` [PATCH v2 1/4] common: Check for fiemap range argument support Nikolay Borisov
2017-10-31 14:11 ` [PATCH v2 2/4] punch: Implement fixup for fiemap range queries Nikolay Borisov
2017-11-07 5:18 ` Eryu Guan
2017-10-31 14:11 ` [PATCH v2 3/4] generic: Adjust generic test ouputs for new fiemap implementation Nikolay Borisov
2017-11-07 5:15 ` Eryu Guan
2017-10-31 14:11 ` Nikolay Borisov [this message]
2017-11-02 3:16 ` [PATCH v2 1/4] common: Check for fiemap range argument support Eryu Guan
2017-11-02 6:59 ` Nikolay Borisov
2017-11-02 8:13 ` [PATCH v3] " Nikolay Borisov
2017-11-02 21:12 ` Dave Chinner
2017-11-03 14:05 ` Nikolay Borisov
2017-11-03 16:28 ` Darrick J. Wong
2017-11-14 15:09 ` Eric Sandeen
2017-11-15 7:41 ` Eryu Guan
2017-10-31 20:29 ` [PATCH v2 1/2] fiemap: Factor out actual fiemap call code Darrick J. Wong
2017-10-31 20:32 ` Nikolay Borisov
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=1509459096-7585-4-git-send-email-nborisov@suse.com \
--to=nborisov@suse.com \
--cc=darrick.wong@oracle.com \
--cc=eguan@redhat.com \
--cc=fstests@vger.kernel.org \
--cc=linux-xfs@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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).