xfs
[Top] [All Lists]

Re: [PATCH] xfstests: Test xfs_fsr / swapext management of di_forkoff w/

To: Eric Sandeen <sandeen@xxxxxxxxxx>
Subject: Re: [PATCH] xfstests: Test xfs_fsr / swapext management of di_forkoff w/ selinux
From: Eryu Guan <guaneryu@xxxxxxxxx>
Date: Fri, 18 Oct 2013 12:48:22 +0800
Cc: xfs-oss <xfs@xxxxxxxxxxx>
Delivered-to: xfs@xxxxxxxxxxx
Dkim-signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=date:from:to:cc:subject:message-id:references:mime-version :content-type:content-disposition:in-reply-to:user-agent; bh=NuYbF2YxQ5YcG8yVpinCYbDHKmZseW44iJWNMMqgaJo=; b=MqvumIfw0NJ4yjE1GF/LRg/KNc/ilkhrPD+qmh521+fToN6x9WnqXjL184efDmZ8js ahkFZk0hxkp7zNyWfA1P60XnR2Lic23PESkSN4vBzt/2fdNHY3OLLEv08rq/qBiN28b/ QaBRTTB6zpxJHR6tRPzwt1iddGoU+BhKny26OadzEd2UKm42D4+fw+qmqqcHGaHKcvCK xRRKzXDm0Yjfd1YXmqQCBCu/zXX1igx5f1rWQOGTxNQ30a/z8Ynkq8iKmnQ7wwIXu+3P ab5ngn19Ja5g2PNFvGEoTwdzp0iWX+Y7SXyOr/JyzXUB57/NDlfeH35c1qPpUzgkNjFt Ii9A==
In-reply-to: <5260B62D.2010507@xxxxxxxxxx>
References: <5260B62D.2010507@xxxxxxxxxx>
User-agent: Mutt/1.5.21 (2010-09-15)
On Thu, Oct 17, 2013 at 11:16:45PM -0500, Eric Sandeen wrote:
> In some cases selinux's creation of an xattr on the temporary
> fd creates a local xattr, but the file we are trying to 
> defragment has attrs in extent format, and the forkoff mismatch
> will cause xfs_fsr to fail.  This test demonstrates it;  I
> have old patches sent to the list long ago that should fix
> it.  I'll resend them soon.
> 
> Signed-off-by: Eric Sandeen <sandeen@xxxxxxxxxx>
> ---
> 
> diff --git a/tests/xfs/351 b/tests/xfs/351
> new file mode 100755
> index 0000000..b21ba7a
> --- /dev/null
> +++ b/tests/xfs/351
> @@ -0,0 +1,89 @@
> +#! /bin/bash
> +# FS QA Test No. 350
> +#
> +# Test xfs_fsr / swapext management of di_forkoff w/ selinux
> +#
> +#-----------------------------------------------------------------------
> +# Copyright (c) 2013 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
> +#-----------------------------------------------------------------------
> +#
> +# creator

Seems we don't need "creator" in comment now.

> +
> +seq=`basename $0`

Forgot about "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 $tmp.*
> +}
> +
> +# get standard environment, filters and checks
> +. ./common/rc
> +. ./common/filter
> +
> +_require_scratch
> +
> +# real QA test starts here
> +
> +_supported_fs xfs
> +_supported_os Linux
> +
> +getenforce | grep -q "Enforcing\|Permissive" || _notrun "SELinux not enabled"
> +[ "$XFS_FSR_PROG" = "" ] && _notrun "xfs_fsr not found"
> +
> +rm -f $seqres.full
> +
> +_scratch_mkfs_xfs -i size=256 >> $seq.full 2>&1 || _fail "mkfs failed"
> +
> +# Manually mount to avoid fs-wide context set by default in xfstests
> +mount $SCRATCH_DEV $SCRATCH_MNT
> +
> +touch $SCRATCH_MNT/$seq.test
> +
> +# Create an extents-format attribute on a file
> +xfs_io -f -c "pwrite -S 0x63 0 4096" $SCRATCH_MNT/attrvals 2>&1 &> 
> $seqres.full

Use $XFS_IO_PROG instead?

And seems "2>&1" is redundant, "&>" can do the work.

Thanks,
Eryu Guan

> +cat $SCRATCH_MNT/attrvals | attr -s name $SCRATCH_MNT/$seq.test 2>&1 &> 
> $seqres.full
> +
> +# Fragment the file by writing backwards
> +for I in `seq 6 -1 0`; do
> +     dd if=/dev/zero of=$SCRATCH_MNT/$seq.test seek=$I bs=4k \
> +        oflag=direct count=1 conv=notrunc 2>&1 &> $seqres.full
> +done
> +
> +# See if xfs_fsr works.
> +
> +# The target file now has a non-local attribute, but the
> +# fsr temp file gets assigned a local attribute on creation,
> +# due to selinux.
> +    
> +# In this case, the large-ish selinux attr will create a forkoff
> +# in the temp file smaller than the forkoff in the target file,
> +# because the FMT_EXTENTS attr takes up less space.  If there is
> +# no mechanism to grow the temp forkoff to match, we can end up
> +# failing to swap these 2 inodes if the result is not enough
> +# data space in the temp inode as a result.
> +
> +$XFS_FSR_PROG -v $SCRATCH_MNT/$seq.test | _filter_scratch
> +
> +# success, all done
> +status=0
> +exit
> diff --git a/tests/xfs/351.out b/tests/xfs/351.out
> new file mode 100644
> index 0000000..86cbe5d
> --- /dev/null
> +++ b/tests/xfs/351.out
> @@ -0,0 +1,3 @@
> +QA output created by 351
> +SCRATCH_MNT/351.test
> +extents before:6 after:1 DONE SCRATCH_MNT/351.test
> diff --git a/tests/xfs/group b/tests/xfs/group
> index acb92ac..da76770 100644
> --- a/tests/xfs/group
> +++ b/tests/xfs/group
> @@ -180,4 +180,4 @@
>  298 auto attr symlink quick
>  299 auto quota
>  350 auto dump quota
> -
> +351 auto fsr
> 
> 
> _______________________________________________
> xfs mailing list
> xfs@xxxxxxxxxxx
> http://oss.sgi.com/mailman/listinfo/xfs

<Prev in Thread] Current Thread [Next in Thread>