From: Eric Sandeen <sandeen@sandeen.net>
To: Dave Chinner <david@fromorbit.com>, fstests@vger.kernel.org
Subject: Re: [PATCH 4/5] generic: add mmap write vs truncate/remap test
Date: Fri, 19 Sep 2014 14:35:57 -0500 [thread overview]
Message-ID: <541C859D.5020408@sandeen.net> (raw)
In-Reply-To: <1410918113-23516-5-git-send-email-david@fromorbit.com>
On 9/16/14 8:41 PM, Dave Chinner wrote:
> This test exposed a problem with mapped writes to the tail page of a
> file in XFS and potentially ext4. Eric did all the hard work of
> taking the bug report and generating the reproducable test case
> on ext4, but I haven't been able to reproduce then problem on ext4.
>
> Regardless, make it a generic test so that we can ensure that all
> filesystems handle the case correctly.
>
> Signed-off-by: Dave Chinner <dchinner@redhat.com>
Fine as it is; I've also sent a cleanup patch which you can fold in,
or merge separately, as you wish.
Reviewed-by: Eric Sandeen <sandeen@redhat.com>
> ---
> tests/generic/030 | 148 ++++++++++++++++++++++++++++++++++++++++++++++++++
> tests/generic/030.out | 53 ++++++++++++++++++
> tests/generic/group | 1 +
> 3 files changed, 202 insertions(+)
> create mode 100644 tests/generic/030
> create mode 100644 tests/generic/030.out
>
> diff --git a/tests/generic/030 b/tests/generic/030
> new file mode 100644
> index 0000000..9947b96
> --- /dev/null
> +++ b/tests/generic/030
> @@ -0,0 +1,148 @@
> +#! /bin/bash
> +# FS QA Test No. generic/030
> +#
> +# Test mapped writes against remap+truncate down/up to ensure we get the data
> +# correctly written. This can expose data corruption bugs on filesystems where
> +# the block size is smaller than the page size.
> +#
> +#-----------------------------------------------------------------------
> +# Copyright (c) 2014 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 $tmp.*
> +}
> +
> +# get standard environment, filters and checks
> +. ./common/rc
> +. ./common/filter
> +
> +# real QA test starts here
> +
> +# Modify as appropriate.
> +_supported_fs generic
> +_supported_os IRIX Linux
> +_require_scratch
> +_require_xfs_io_command "mremap"
> +
> +testfile=$SCRATCH_MNT/testfile
> +
> +_scratch_mkfs > /dev/null 2>&1
> +_scratch_mount
> +
> +# first case is just truncate down/truncate up to check that the mapped
> +# write after the truncate up is correctly handled.
> +$XFS_IO_PROG -t -f \
> +-c "truncate 5017k" `# truncate | |` \
> +-c "pwrite -S 0x58 0 5017k" `# write |X...XXX|` \
> +-c "mmap -rw 0 5017k" `# mmap | |` \
> +-c "truncate 5020k" `# truncate up | |` \
> +-c "mremap -m 5020k" `# mremap up | |` \
> +-c "mwrite -S 0x57 5017k 3k" `# mwrite | WWWWWWWWWWWWWWWW|` \
> +-c "mremap 5017k " `# mremap dn | |` \
> +-c "truncate 5017k" `# mremap dn | |` \
> +-c "truncate 5020k" `# truncate up | |` \
> +-c "mremap -m 5020k" `# mremap up | |` \
> +-c "mwrite -S 0x59 5017k 3k" `# mwrite | YYYYYYYYYYYYYYYY|` \
> +-c "close" \
> +$testfile | _filter_xfs_io
> +
> +echo "==== Pre-Remount ==="
> +hexdump -C $testfile
> +_scratch_remount
> +echo "==== Post-Remount =="
> +hexdump -C $testfile
> +
> +rm -f $testfile
> +sync
> +
> +# second case is to do a mwrite between the truncate to a block on the
> +# same page we are truncating within the EOF. This checks that a mapped
> +# write between truncate down and truncate up a further mapped
> +# write to the same page into the new space doesn't result in data being lost.
> +$XFS_IO_PROG -t -f \
> +-c "truncate 5017k" `# truncate | |` \
> +-c "pwrite -S 0x58 0 5017k" `# write |X...XXX|` \
> +-c "mmap -rw 0 5017k" `# mmap | |` \
> +-c "truncate 5020k" `# truncate up | |` \
> +-c "mremap -m 5020k" `# mremap up | |` \
> +-c "mwrite -S 0x57 5017k 3k" `# mwrite | WWWWWWWWWWWWWWWW|` \
> +-c "mremap 5017k " `# mremap dn | |` \
> +-c "truncate 5017k" `# mremap dn | |` \
> +-c "mwrite -S 0x5a 5016k 1k" `# mwrite | ZZZ |` \
> +-c "truncate 5020k" `# truncate up | |` \
> +-c "mremap -m 5020k" `# mremap up | |` \
> +-c "mwrite -S 0x59 5017k 3k" `# mwrite | YYYYYYYYYYYYYYYY|` \
> +-c "close" \
> +$testfile | _filter_xfs_io
> +
> +echo "==== Pre-Remount ==="
> +hexdump -C $testfile
> +_scratch_remount
> +echo "==== Post-Remount =="
> +hexdump -C $testfile
> +
> +# third case is the same as the first, but this time on unaligned byte
> +# boundaries rather than block boundaries. This mimics the exact mmap write
> +# patterns of the application that exposed the bug in the first place, and
> +# so is somewhat more complex and has repeated operations in it.
> +$XFS_IO_PROG -t -f \
> +-c "truncate 5136912" \
> +-c "pwrite -S 0x58 0 5136912" \
> +-c "mmap -rw 0 5136912" \
> +-c "mremap 5136912" \
> +-c "truncate 5136912" \
> +-c "truncate 5139720" \
> +-c "mremap -m 5139720" \
> +-c "mwrite -S 0 5136912 2808" \
> +-c "mwrite -S 0 5136912 2808" \
> +-c "mwrite -S 0 5136912 2808" \
> +-c "mremap 5136912 " \
> +-c "truncate 5136912" \
> +-c "truncate 5139720" \
> +-c "mremap -m 5139720" \
> +-c "mwrite -S 0 5136912 2808" \
> +-c "mwrite -S 0 5136912 2808" \
> +-c "mwrite -S 0x59 5136912 2808" \
> +-c "truncate 5140480" \
> +-c "mremap 5140480" \
> +-c "msync -s 0 5140480" \
> +-c "mremap 5139720" \
> +-c "munmap" \
> +-c "close" \
> +$testfile | _filter_xfs_io
> +
> +echo "==== Pre-Remount ==="
> +hexdump -C $testfile
> +_scratch_remount
> +echo "==== Post-Remount =="
> +hexdump -C $testfile
> +
> +status=0
> +exit
> +
> diff --git a/tests/generic/030.out b/tests/generic/030.out
> new file mode 100644
> index 0000000..20f6561
> --- /dev/null
> +++ b/tests/generic/030.out
> @@ -0,0 +1,53 @@
> +QA output created by 030
> +wrote 5137408/5137408 bytes at offset 0
> +XXX Bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
> +==== Pre-Remount ===
> +00000000 58 58 58 58 58 58 58 58 58 58 58 58 58 58 58 58 |XXXXXXXXXXXXXXXX|
> +*
> +004e6400 59 59 59 59 59 59 59 59 59 59 59 59 59 59 59 59 |YYYYYYYYYYYYYYYY|
> +*
> +004e7000
> +==== Post-Remount ==
> +00000000 58 58 58 58 58 58 58 58 58 58 58 58 58 58 58 58 |XXXXXXXXXXXXXXXX|
> +*
> +004e6400 59 59 59 59 59 59 59 59 59 59 59 59 59 59 59 59 |YYYYYYYYYYYYYYYY|
> +*
> +004e7000
> +wrote 5137408/5137408 bytes at offset 0
> +XXX Bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
> +==== Pre-Remount ===
> +00000000 58 58 58 58 58 58 58 58 58 58 58 58 58 58 58 58 |XXXXXXXXXXXXXXXX|
> +*
> +004e6000 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a |ZZZZZZZZZZZZZZZZ|
> +*
> +004e6400 59 59 59 59 59 59 59 59 59 59 59 59 59 59 59 59 |YYYYYYYYYYYYYYYY|
> +*
> +004e7000
> +==== Post-Remount ==
> +00000000 58 58 58 58 58 58 58 58 58 58 58 58 58 58 58 58 |XXXXXXXXXXXXXXXX|
> +*
> +004e6000 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a |ZZZZZZZZZZZZZZZZ|
> +*
> +004e6400 59 59 59 59 59 59 59 59 59 59 59 59 59 59 59 59 |YYYYYYYYYYYYYYYY|
> +*
> +004e7000
> +wrote 5136912/5136912 bytes at offset 0
> +XXX Bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
> +==== Pre-Remount ===
> +00000000 58 58 58 58 58 58 58 58 58 58 58 58 58 58 58 58 |XXXXXXXXXXXXXXXX|
> +*
> +004e6210 59 59 59 59 59 59 59 59 59 59 59 59 59 59 59 59 |YYYYYYYYYYYYYYYY|
> +*
> +004e6d00 59 59 59 59 59 59 59 59 00 00 00 00 00 00 00 00 |YYYYYYYY........|
> +004e6d10 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 |................|
> +*
> +004e7000
> +==== Post-Remount ==
> +00000000 58 58 58 58 58 58 58 58 58 58 58 58 58 58 58 58 |XXXXXXXXXXXXXXXX|
> +*
> +004e6210 59 59 59 59 59 59 59 59 59 59 59 59 59 59 59 59 |YYYYYYYYYYYYYYYY|
> +*
> +004e6d00 59 59 59 59 59 59 59 59 00 00 00 00 00 00 00 00 |YYYYYYYY........|
> +004e6d10 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 |................|
> +*
> +004e7000
> diff --git a/tests/generic/group b/tests/generic/group
> index 18c94de..ba1c913 100644
> --- a/tests/generic/group
> +++ b/tests/generic/group
> @@ -32,6 +32,7 @@
> 027 auto enospc
> 028 auto quick
> 029 auto quick rw
> +030 auto quick rw
> 053 acl repair auto quick
> 062 attr udf auto quick
> 068 other auto freeze dangerous stress
>
next prev parent reply other threads:[~2014-09-19 19:35 UTC|newest]
Thread overview: 17+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-09-17 1:41 [PATCH 0/5] xfstests: cleanups and new tests Dave Chinner
2014-09-17 1:41 ` [PATCH 1/5] generic: more tests should clean up TESTDIR on success Dave Chinner
2014-09-17 4:17 ` Eric Sandeen
2014-09-17 1:41 ` [PATCH 2/5] check: more tests that shouldn't check the scratch device Dave Chinner
2014-09-17 4:27 ` Eric Sandeen
2014-09-17 1:41 ` [PATCH 3/5] generic: add mmap write vs truncate test Dave Chinner
2014-09-19 19:38 ` Eric Sandeen
2014-09-17 1:41 ` [PATCH 4/5] generic: add mmap write vs truncate/remap test Dave Chinner
2014-09-17 4:32 ` Eric Sandeen
2014-09-17 4:51 ` Dave Chinner
2014-09-19 19:35 ` [PATCH 4.5/5] generic: tidy up " Eric Sandeen
2014-09-19 19:35 ` Eric Sandeen [this message]
2014-09-20 0:17 ` [PATCH 4/5] generic: add " Eric Sandeen
2014-09-20 23:32 ` Dave Chinner
2014-09-17 1:41 ` [PATCH 5/5] generic: add write vs fcollapse test Dave Chinner
2014-09-17 12:40 ` Brian Foster
2014-09-18 1:33 ` Dave Chinner
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=541C859D.5020408@sandeen.net \
--to=sandeen@sandeen.net \
--cc=david@fromorbit.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox