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: Tue, 16 Sep 2014 23:32:11 -0500 [thread overview]
Message-ID: <54190ECB.2070900@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.
Hm, not even w/ the last test case below, on the byte boundaries?
FWIW, I think I can simplify that one - the repeated ops aren't
necessary. Probably no need to carry that along if they aren't
relevant to the bug.
I'll look at this more closely tomorrow, thanks for encapsulating
this in an xfstest.
-Eric
> 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>
> ---
> 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:[~2014-09-17 4:32 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 [this message]
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 ` [PATCH 4/5] generic: add " Eric Sandeen
2014-09-20 0:17 ` 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=54190ECB.2070900@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