xfs_repair: test for false positive reserved attr name use
diff mbox

Message ID 555414EA.80301@redhat.com
State New
Headers show

Commit Message

Eric Sandeen May 14, 2015, 3:22 a.m. UTC
xfs_repair compares attr names in the root namespace to
two special/reserved names, "SGI_ACL_FILE" and "SGI_ACL_DEFAULT"
and if the value in them aren't valid acls, flags this as
an inconsistency.

However, due to various bugs, xfs_repair may only compare
a smaller portion of the on-disk value; hence either
substrings or superstrings may match, and false-positive
corruption will be detected.  This test checks for those
false positives; i.e. the ACL names created in this test
may cause xfs_repair to "fix" them, but it should not.

Signed-off-by: Eric Sandeen <sandeen@redhat.com>
---

patch for xfs_repair will be sent shortly


--
To unsubscribe from this list: send the line "unsubscribe fstests" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Comments

Eryu Guan May 14, 2015, 4:55 a.m. UTC | #1
On Wed, May 13, 2015 at 10:22:18PM -0500, Eric Sandeen wrote:
> xfs_repair compares attr names in the root namespace to
> two special/reserved names, "SGI_ACL_FILE" and "SGI_ACL_DEFAULT"
> and if the value in them aren't valid acls, flags this as
> an inconsistency.
> 
> However, due to various bugs, xfs_repair may only compare
> a smaller portion of the on-disk value; hence either
> substrings or superstrings may match, and false-positive
> corruption will be detected.  This test checks for those
> false positives; i.e. the ACL names created in this test
> may cause xfs_repair to "fix" them, but it should not.
> 
> Signed-off-by: Eric Sandeen <sandeen@redhat.com>
> ---
> 
> patch for xfs_repair will be sent shortly
> 
> diff --git a/tests/xfs/053 b/tests/xfs/053
> new file mode 100755
> index 0000000..284c014
> --- /dev/null
> +++ b/tests/xfs/053
> @@ -0,0 +1,76 @@
> +#! /bin/bash
> +# FS QA Test 053
> +#
> +# Ensure that xfs_repair can properly spot SGI_ACL_FILE
> +# and SGI_ACL_DEFAULT in the root attr namespace.
> +#
> +# Due to bugs here and there, we sometimes matched on partial
> +# strings with those names, and threw off xfs_repair.
> +#
> +#-----------------------------------------------------------------------
> +# Copyright (c) 2015 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!

I haven't run the test yet, but I don't see the trap and cleanup
function, is that on purpose?

> +
> +# get standard environment, filters and checks
> +. ./common/rc
> +. ./common/filter

. ./common/attr and _require_attrs ?

> +
> +# real QA test starts here
> +
> +_supported_fs xfs
> +_supported_os IRIX Linux
> +_require_scratch
> +
> +rm -f $seqres.full

Seems this file is not used and the rm can be omitted.

> +
> +_scratch_mount

Should _scratch_mkfs first

> +
> +# Create root namespace attr names which are substrings or superstrings
> +# of the reserved ACL names, and make sure xfs_repair copes.
> +
> +# Due to various bugs, either the compared length was shorter
> +# than the reserved name (7 chars), so substrings matched, and/or only
> +# the reserved name length was compared, so superstrings matched.
> +
> +rm -f $SCRATCH_MNT/$seq.*

If it's fresh fs, this seems unnecessary.

> +
> +# actual reserved names:
> +#           SGI_ACL_FILE	SGI_ACL_DEFAULT \
> +for NAME in SGI_ACL \
> +	    SGI_ACL_F		SGI_ACL_D \
> +	    SGI_ACL_FILE_FOO	SGI_ACL_DEFAULT_FOO; do
> +	touch $SCRATCH_MNT/${seq}.${NAME}
> +	attr -R -s $NAME -V "Wow, such $NAME" $SCRATCH_MNT/${seq}.${NAME} \
> +		| _filter_scratch

$ATTR_PROG

Thanks,
Eryu

> +done
> +
> +# Older repair failed because it sees the above names as matching
> +# SGI_ACL_FILE / SGI_ACL_DEFAULT but w/o valid acls on them
> +
> +# The test harness will catch this (false positive) corruption
> +
> +# success, all done
> +status=0
> +exit
> diff --git a/tests/xfs/053.out b/tests/xfs/053.out
> new file mode 100644
> index 0000000..d876555
> --- /dev/null
> +++ b/tests/xfs/053.out
> @@ -0,0 +1,11 @@
> +QA output created by 053
> +Attribute "SGI_ACL" set to a 17 byte value for SCRATCH_MNT/053.SGI_ACL:
> +Wow, such SGI_ACL
> +Attribute "SGI_ACL_F" set to a 19 byte value for SCRATCH_MNT/053.SGI_ACL_F:
> +Wow, such SGI_ACL_F
> +Attribute "SGI_ACL_D" set to a 19 byte value for SCRATCH_MNT/053.SGI_ACL_D:
> +Wow, such SGI_ACL_D
> +Attribute "SGI_ACL_FILE_FOO" set to a 26 byte value for SCRATCH_MNT/053.SGI_ACL_FILE_FOO:
> +Wow, such SGI_ACL_FILE_FOO
> +Attribute "SGI_ACL_DEFAULT_FOO" set to a 29 byte value for SCRATCH_MNT/053.SGI_ACL_DEFAULT_FOO:
> +Wow, such SGI_ACL_DEFAULT_FOO
> diff --git a/tests/xfs/group b/tests/xfs/group
> index 58144d2..26dd881 100644
> --- a/tests/xfs/group
> +++ b/tests/xfs/group
> @@ -50,6 +50,7 @@
>  050 quota auto quick
>  051 auto log metadata
>  052 quota db auto quick
> +053 acl auto
>  054 quota auto quick
>  055 dump ioctl remote tape
>  056 dump ioctl auto quick
> 
> --
> To unsubscribe from this list: send the line "unsubscribe fstests" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
--
To unsubscribe from this list: send the line "unsubscribe fstests" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Eric Sandeen May 14, 2015, 5:23 a.m. UTC | #2
All good advice; sigh, I am rusty!  Thanks, will resend with fixes.

(Started as using test dev so left some things out; not sure I need a trap if there is nothing to clean up?  Also do we need to require attr on xfs?  Can't remember if we can compile that out, but I guess it's harmless to include it)

Thanks,
-Eric 

> On May 13, 2015, at 11:55 PM, Eryu Guan <eguan@redhat.com> wrote:
> 
>> On Wed, May 13, 2015 at 10:22:18PM -0500, Eric Sandeen wrote:
>> xfs_repair compares attr names in the root namespace to
>> two special/reserved names, "SGI_ACL_FILE" and "SGI_ACL_DEFAULT"
>> and if the value in them aren't valid acls, flags this as
>> an inconsistency.
>> 
>> However, due to various bugs, xfs_repair may only compare
>> a smaller portion of the on-disk value; hence either
>> substrings or superstrings may match, and false-positive
>> corruption will be detected.  This test checks for those
>> false positives; i.e. the ACL names created in this test
>> may cause xfs_repair to "fix" them, but it should not.
>> 
>> Signed-off-by: Eric Sandeen <sandeen@redhat.com>
>> ---
>> 
>> patch for xfs_repair will be sent shortly
>> 
>> diff --git a/tests/xfs/053 b/tests/xfs/053
>> new file mode 100755
>> index 0000000..284c014
>> --- /dev/null
>> +++ b/tests/xfs/053
>> @@ -0,0 +1,76 @@
>> +#! /bin/bash
>> +# FS QA Test 053
>> +#
>> +# Ensure that xfs_repair can properly spot SGI_ACL_FILE
>> +# and SGI_ACL_DEFAULT in the root attr namespace.
>> +#
>> +# Due to bugs here and there, we sometimes matched on partial
>> +# strings with those names, and threw off xfs_repair.
>> +#
>> +#-----------------------------------------------------------------------
>> +# Copyright (c) 2015 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!
> 
> I haven't run the test yet, but I don't see the trap and cleanup
> function, is that on purpose?
> 
>> +
>> +# get standard environment, filters and checks
>> +. ./common/rc
>> +. ./common/filter
> 
> . ./common/attr and _require_attrs ?
> 
>> +
>> +# real QA test starts here
>> +
>> +_supported_fs xfs
>> +_supported_os IRIX Linux
>> +_require_scratch
>> +
>> +rm -f $seqres.full
> 
> Seems this file is not used and the rm can be omitted.
> 
>> +
>> +_scratch_mount
> 
> Should _scratch_mkfs first
> 
>> +
>> +# Create root namespace attr names which are substrings or superstrings
>> +# of the reserved ACL names, and make sure xfs_repair copes.
>> +
>> +# Due to various bugs, either the compared length was shorter
>> +# than the reserved name (7 chars), so substrings matched, and/or only
>> +# the reserved name length was compared, so superstrings matched.
>> +
>> +rm -f $SCRATCH_MNT/$seq.*
> 
> If it's fresh fs, this seems unnecessary.
> 
>> +
>> +# actual reserved names:
>> +#           SGI_ACL_FILE    SGI_ACL_DEFAULT \
>> +for NAME in SGI_ACL \
>> +        SGI_ACL_F        SGI_ACL_D \
>> +        SGI_ACL_FILE_FOO    SGI_ACL_DEFAULT_FOO; do
>> +    touch $SCRATCH_MNT/${seq}.${NAME}
>> +    attr -R -s $NAME -V "Wow, such $NAME" $SCRATCH_MNT/${seq}.${NAME} \
>> +        | _filter_scratch
> 
> $ATTR_PROG
> 
> Thanks,
> Eryu
> 
>> +done
>> +
>> +# Older repair failed because it sees the above names as matching
>> +# SGI_ACL_FILE / SGI_ACL_DEFAULT but w/o valid acls on them
>> +
>> +# The test harness will catch this (false positive) corruption
>> +
>> +# success, all done
>> +status=0
>> +exit
>> diff --git a/tests/xfs/053.out b/tests/xfs/053.out
>> new file mode 100644
>> index 0000000..d876555
>> --- /dev/null
>> +++ b/tests/xfs/053.out
>> @@ -0,0 +1,11 @@
>> +QA output created by 053
>> +Attribute "SGI_ACL" set to a 17 byte value for SCRATCH_MNT/053.SGI_ACL:
>> +Wow, such SGI_ACL
>> +Attribute "SGI_ACL_F" set to a 19 byte value for SCRATCH_MNT/053.SGI_ACL_F:
>> +Wow, such SGI_ACL_F
>> +Attribute "SGI_ACL_D" set to a 19 byte value for SCRATCH_MNT/053.SGI_ACL_D:
>> +Wow, such SGI_ACL_D
>> +Attribute "SGI_ACL_FILE_FOO" set to a 26 byte value for SCRATCH_MNT/053.SGI_ACL_FILE_FOO:
>> +Wow, such SGI_ACL_FILE_FOO
>> +Attribute "SGI_ACL_DEFAULT_FOO" set to a 29 byte value for SCRATCH_MNT/053.SGI_ACL_DEFAULT_FOO:
>> +Wow, such SGI_ACL_DEFAULT_FOO
>> diff --git a/tests/xfs/group b/tests/xfs/group
>> index 58144d2..26dd881 100644
>> --- a/tests/xfs/group
>> +++ b/tests/xfs/group
>> @@ -50,6 +50,7 @@
>> 050 quota auto quick
>> 051 auto log metadata
>> 052 quota db auto quick
>> +053 acl auto
>> 054 quota auto quick
>> 055 dump ioctl remote tape
>> 056 dump ioctl auto quick
>> 
>> --
>> To unsubscribe from this list: send the line "unsubscribe fstests" in
>> the body of a message to majordomo@vger.kernel.org
>> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> --
> To unsubscribe from this list: send the line "unsubscribe fstests" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> 
--
To unsubscribe from this list: send the line "unsubscribe fstests" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Eryu Guan May 14, 2015, 5:44 a.m. UTC | #3
On Thu, May 14, 2015 at 12:23:14AM -0500, Eric Sandeen wrote:
> All good advice; sigh, I am rusty!  Thanks, will resend with fixes.
> 
> (Started as using test dev so left some things out; not sure I need a trap if there is nothing to clean up?  Also do we need to require attr on xfs?  Can't remember if we can compile that out, but I guess it's harmless to include it)

I did a "grep _require_attrs tests/xfs/*" and result showed several
tests in xfs require attrs. At least it checks if ATTR_PROG exists :)

And I think we need a trap and cleanup even if there's no obvious things
to clean, some functions use $tmp.xxx internally, a cleanup could remove
all these tmp files too, and trap plays with exit status, so it's still
needed.

Thanks,
Eryu
> 
> Thanks,
> -Eric 
> 
> > On May 13, 2015, at 11:55 PM, Eryu Guan <eguan@redhat.com> wrote:
> > 
> >> On Wed, May 13, 2015 at 10:22:18PM -0500, Eric Sandeen wrote:
> >> xfs_repair compares attr names in the root namespace to
> >> two special/reserved names, "SGI_ACL_FILE" and "SGI_ACL_DEFAULT"
> >> and if the value in them aren't valid acls, flags this as
> >> an inconsistency.
> >> 
> >> However, due to various bugs, xfs_repair may only compare
> >> a smaller portion of the on-disk value; hence either
> >> substrings or superstrings may match, and false-positive
> >> corruption will be detected.  This test checks for those
> >> false positives; i.e. the ACL names created in this test
> >> may cause xfs_repair to "fix" them, but it should not.
> >> 
> >> Signed-off-by: Eric Sandeen <sandeen@redhat.com>
> >> ---
> >> 
> >> patch for xfs_repair will be sent shortly
> >> 
> >> diff --git a/tests/xfs/053 b/tests/xfs/053
> >> new file mode 100755
> >> index 0000000..284c014
> >> --- /dev/null
> >> +++ b/tests/xfs/053
> >> @@ -0,0 +1,76 @@
> >> +#! /bin/bash
> >> +# FS QA Test 053
> >> +#
> >> +# Ensure that xfs_repair can properly spot SGI_ACL_FILE
> >> +# and SGI_ACL_DEFAULT in the root attr namespace.
> >> +#
> >> +# Due to bugs here and there, we sometimes matched on partial
> >> +# strings with those names, and threw off xfs_repair.
> >> +#
> >> +#-----------------------------------------------------------------------
> >> +# Copyright (c) 2015 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!
> > 
> > I haven't run the test yet, but I don't see the trap and cleanup
> > function, is that on purpose?
> > 
> >> +
> >> +# get standard environment, filters and checks
> >> +. ./common/rc
> >> +. ./common/filter
> > 
> > . ./common/attr and _require_attrs ?
> > 
> >> +
> >> +# real QA test starts here
> >> +
> >> +_supported_fs xfs
> >> +_supported_os IRIX Linux
> >> +_require_scratch
> >> +
> >> +rm -f $seqres.full
> > 
> > Seems this file is not used and the rm can be omitted.
> > 
> >> +
> >> +_scratch_mount
> > 
> > Should _scratch_mkfs first
> > 
> >> +
> >> +# Create root namespace attr names which are substrings or superstrings
> >> +# of the reserved ACL names, and make sure xfs_repair copes.
> >> +
> >> +# Due to various bugs, either the compared length was shorter
> >> +# than the reserved name (7 chars), so substrings matched, and/or only
> >> +# the reserved name length was compared, so superstrings matched.
> >> +
> >> +rm -f $SCRATCH_MNT/$seq.*
> > 
> > If it's fresh fs, this seems unnecessary.
> > 
> >> +
> >> +# actual reserved names:
> >> +#           SGI_ACL_FILE    SGI_ACL_DEFAULT \
> >> +for NAME in SGI_ACL \
> >> +        SGI_ACL_F        SGI_ACL_D \
> >> +        SGI_ACL_FILE_FOO    SGI_ACL_DEFAULT_FOO; do
> >> +    touch $SCRATCH_MNT/${seq}.${NAME}
> >> +    attr -R -s $NAME -V "Wow, such $NAME" $SCRATCH_MNT/${seq}.${NAME} \
> >> +        | _filter_scratch
> > 
> > $ATTR_PROG
> > 
> > Thanks,
> > Eryu
> > 
> >> +done
> >> +
> >> +# Older repair failed because it sees the above names as matching
> >> +# SGI_ACL_FILE / SGI_ACL_DEFAULT but w/o valid acls on them
> >> +
> >> +# The test harness will catch this (false positive) corruption
> >> +
> >> +# success, all done
> >> +status=0
> >> +exit
> >> diff --git a/tests/xfs/053.out b/tests/xfs/053.out
> >> new file mode 100644
> >> index 0000000..d876555
> >> --- /dev/null
> >> +++ b/tests/xfs/053.out
> >> @@ -0,0 +1,11 @@
> >> +QA output created by 053
> >> +Attribute "SGI_ACL" set to a 17 byte value for SCRATCH_MNT/053.SGI_ACL:
> >> +Wow, such SGI_ACL
> >> +Attribute "SGI_ACL_F" set to a 19 byte value for SCRATCH_MNT/053.SGI_ACL_F:
> >> +Wow, such SGI_ACL_F
> >> +Attribute "SGI_ACL_D" set to a 19 byte value for SCRATCH_MNT/053.SGI_ACL_D:
> >> +Wow, such SGI_ACL_D
> >> +Attribute "SGI_ACL_FILE_FOO" set to a 26 byte value for SCRATCH_MNT/053.SGI_ACL_FILE_FOO:
> >> +Wow, such SGI_ACL_FILE_FOO
> >> +Attribute "SGI_ACL_DEFAULT_FOO" set to a 29 byte value for SCRATCH_MNT/053.SGI_ACL_DEFAULT_FOO:
> >> +Wow, such SGI_ACL_DEFAULT_FOO
> >> diff --git a/tests/xfs/group b/tests/xfs/group
> >> index 58144d2..26dd881 100644
> >> --- a/tests/xfs/group
> >> +++ b/tests/xfs/group
> >> @@ -50,6 +50,7 @@
> >> 050 quota auto quick
> >> 051 auto log metadata
> >> 052 quota db auto quick
> >> +053 acl auto
> >> 054 quota auto quick
> >> 055 dump ioctl remote tape
> >> 056 dump ioctl auto quick
> >> 
> >> --
> >> To unsubscribe from this list: send the line "unsubscribe fstests" in
> >> the body of a message to majordomo@vger.kernel.org
> >> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> > --
> > To unsubscribe from this list: send the line "unsubscribe fstests" in
> > the body of a message to majordomo@vger.kernel.org
> > More majordomo info at  http://vger.kernel.org/majordomo-info.html
> > 
> --
> To unsubscribe from this list: send the line "unsubscribe fstests" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
--
To unsubscribe from this list: send the line "unsubscribe fstests" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Eric Sandeen May 14, 2015, 2:23 p.m. UTC | #4
On 5/14/15 12:44 AM, Eryu Guan wrote:
> On Thu, May 14, 2015 at 12:23:14AM -0500, Eric Sandeen wrote:
>> All good advice; sigh, I am rusty!  Thanks, will resend with fixes.
>>
>> (Started as using test dev so left some things out; not sure I need a trap if there is nothing to clean up?  Also do we need to require attr on xfs?  Can't remember if we can compile that out, but I guess it's harmless to include it)
> 
> I did a "grep _require_attrs tests/xfs/*" and result showed several
> tests in xfs require attrs. At least it checks if ATTR_PROG exists :)

Yes, I forgot about that ;(

> And I think we need a trap and cleanup even if there's no obvious things
> to clean, some functions use $tmp.xxx internally, a cleanup could remove
> all these tmp files too, and trap plays with exit status, so it's still
> needed.

Fair enough, thanks for the review.

-Eric

--
To unsubscribe from this list: send the line "unsubscribe fstests" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Patch
diff mbox

diff --git a/tests/xfs/053 b/tests/xfs/053
new file mode 100755
index 0000000..284c014
--- /dev/null
+++ b/tests/xfs/053
@@ -0,0 +1,76 @@ 
+#! /bin/bash
+# FS QA Test 053
+#
+# Ensure that xfs_repair can properly spot SGI_ACL_FILE
+# and SGI_ACL_DEFAULT in the root attr namespace.
+#
+# Due to bugs here and there, we sometimes matched on partial
+# strings with those names, and threw off xfs_repair.
+#
+#-----------------------------------------------------------------------
+# Copyright (c) 2015 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!
+
+# get standard environment, filters and checks
+. ./common/rc
+. ./common/filter
+
+# real QA test starts here
+
+_supported_fs xfs
+_supported_os IRIX Linux
+_require_scratch
+
+rm -f $seqres.full
+
+_scratch_mount
+
+# Create root namespace attr names which are substrings or superstrings
+# of the reserved ACL names, and make sure xfs_repair copes.
+
+# Due to various bugs, either the compared length was shorter
+# than the reserved name (7 chars), so substrings matched, and/or only
+# the reserved name length was compared, so superstrings matched.
+
+rm -f $SCRATCH_MNT/$seq.*
+
+# actual reserved names:
+#           SGI_ACL_FILE	SGI_ACL_DEFAULT \
+for NAME in SGI_ACL \
+	    SGI_ACL_F		SGI_ACL_D \
+	    SGI_ACL_FILE_FOO	SGI_ACL_DEFAULT_FOO; do
+	touch $SCRATCH_MNT/${seq}.${NAME}
+	attr -R -s $NAME -V "Wow, such $NAME" $SCRATCH_MNT/${seq}.${NAME} \
+		| _filter_scratch
+done
+
+# Older repair failed because it sees the above names as matching
+# SGI_ACL_FILE / SGI_ACL_DEFAULT but w/o valid acls on them
+
+# The test harness will catch this (false positive) corruption
+
+# success, all done
+status=0
+exit
diff --git a/tests/xfs/053.out b/tests/xfs/053.out
new file mode 100644
index 0000000..d876555
--- /dev/null
+++ b/tests/xfs/053.out
@@ -0,0 +1,11 @@ 
+QA output created by 053
+Attribute "SGI_ACL" set to a 17 byte value for SCRATCH_MNT/053.SGI_ACL:
+Wow, such SGI_ACL
+Attribute "SGI_ACL_F" set to a 19 byte value for SCRATCH_MNT/053.SGI_ACL_F:
+Wow, such SGI_ACL_F
+Attribute "SGI_ACL_D" set to a 19 byte value for SCRATCH_MNT/053.SGI_ACL_D:
+Wow, such SGI_ACL_D
+Attribute "SGI_ACL_FILE_FOO" set to a 26 byte value for SCRATCH_MNT/053.SGI_ACL_FILE_FOO:
+Wow, such SGI_ACL_FILE_FOO
+Attribute "SGI_ACL_DEFAULT_FOO" set to a 29 byte value for SCRATCH_MNT/053.SGI_ACL_DEFAULT_FOO:
+Wow, such SGI_ACL_DEFAULT_FOO
diff --git a/tests/xfs/group b/tests/xfs/group
index 58144d2..26dd881 100644
--- a/tests/xfs/group
+++ b/tests/xfs/group
@@ -50,6 +50,7 @@ 
 050 quota auto quick
 051 auto log metadata
 052 quota db auto quick
+053 acl auto
 054 quota auto quick
 055 dump ioctl remote tape
 056 dump ioctl auto quick