Date:	Thu, 23 Apr 2015 11:49:09 +0800
From:	Eryu Guan <eguan@...hat.com>
To:	Lukas Czerner <lczerner@...hat.com>
Cc:	fstests@...r.kernel.org, linux-ext4@...r.kernel.org
Subject: Re: [PATCH] generic: test data corruption on ext4 caused by
 written/delayed extent

On Wed, Apr 22, 2015 at 04:55:55PM +0200, Lukas Czerner wrote:
> This test excercises the problem with unwritten and delayed extents
> in ext4 extent status tree where we might in some cases lose a block
> worth of data. Even though this was a ext4 specific problem the
> reproducer can be easily tun on any file system so let's do that just
> in case.
> 
> This tests excercises the problem fixed in kernel with commit
> "ext4: Fix data corruption caused by unwritten and delayed extents"
> 
> Signed-off-by: Lukas Czerner <lczerner@...hat.com>
> ---
>  tests/generic/082     | 84 +++++++++++++++++++++++++++++++++++++++++++++++++++
>  tests/generic/082.out | 14 +++++++++
>  2 files changed, 98 insertions(+)
>  create mode 100755 tests/generic/082
>  create mode 100644 tests/generic/082.out

update to tests/generic/group file is missing

> 
> diff --git a/tests/generic/082 b/tests/generic/082
> new file mode 100755
> index 0000000..2fa22c6
> --- /dev/null
> +++ b/tests/generic/082
> @@ -0,0 +1,84 @@
> +#! /bin/bash
> +# FS QA Test 082
> +#
> +# This test excercises the problem with unwritten and delayed extents
> +# in ext4 extent status tree where we might in some cases lose a block
> +# worth of data. Even though this was a ext4 specific problem the
> +# reproducer can be easily tun on any file system so let's do that just
> +# in case.
> +#
> +# This tests excercises the problem fixed in kernel with commit
> +# "ext4: Fix data corruption caused by unwritten and delayed extents"
> +#
> +#-----------------------------------------------------------------------
> +# Copyright (c) 2015 Red Hat, Inc., Lukas Czerner <lczerner@...hat.com>
> +#
> +# 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_test

_require_xfs_io_command "falloc" too

I saw failure on cifs because it doesn't support fallocate so the file
size is not extended, and hexdump output differs

> +
> +test_file=${TEST_DIR}/testfile-082

use $seq not hardcoded 082?

> +
> +rm -f $test_file

rm -f $seqres.full too, otherwise the file keeps growing

Thanks,
Eryu
> +
> +# The first write creates a delayed extent, fallocate creates
> +# unwritten extent which will be marked as delayed in ext4
> +# extent status tree. Second write will convert unwritten/delayed
> +# block into written/delayed.
> +$XFS_IO_PROG -f -c "pwrite -S 0xaa 4096 2048" \
> +		-c "falloc 0 131072" \
> +		-c "pwrite -S 0xbb 65536 2048" \
> +		-c "fsync"  $test_file >> $seqres.full 2>&1
> +
> +# Drop the caches to evict dirty buffers from memory
> +echo 3 > /proc/sys/vm/drop_caches
> +
> +# Write into the second part of the block with 0xbb write from before
> +# will create new empty! buffer because the block is still marked as
> +# delayed even though it's already written - resulting in
> +# overwriting previous data.
> +$XFS_IO_PROG -c "pwrite -S 0xdd 67584 2048" $test_file >> $seqres.full 2>&1
> +
> +# On a faulty ext4 oxbb data will be missing, overwritten by zeroes.
> +hexdump -C $test_file
> +
> +# success, all done
> +status=0
> +exit
> diff --git a/tests/generic/082.out b/tests/generic/082.out
> new file mode 100644
> index 0000000..ef28b9a
> --- /dev/null
> +++ b/tests/generic/082.out
> @@ -0,0 +1,14 @@
> +QA output created by 082
> +00000000  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
> +*
> +00001000  aa aa aa aa aa aa aa aa  aa aa aa aa aa aa aa aa  |................|
> +*
> +00001800  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
> +*
> +00010000  bb bb bb bb bb bb bb bb  bb bb bb bb bb bb bb bb  |................|
> +*
> +00010800  dd dd dd dd dd dd dd dd  dd dd dd dd dd dd dd dd  |................|
> +*
> +00011000  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
> +*
> +00020000
> -- 
> 1.8.3.1
> 
> --
> To unsubscribe from this list: send the line "unsubscribe fstests" in
> the body of a message to majordomo@...r.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
--
To unsubscribe from this list: send the line "unsubscribe linux-ext4" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Powered by blists - more mailing lists