generic: add test for boundary in xfs_attr_shortform_verify
v4 cover generic: add test for boundary in xfs_attr_shortform_verify
Add test for XFS bug that corrupted metadata if extended attribute name of length 1 was used.
V4 did the following changes:
1) replaced the system-wide cache drop and sleep by _scratch_cycle_mount
2) output of getfattr is grepped for attribute name to simplify expected output
3) dropped explicit _check_scratch_fs
4) renamed the test from 609 to 611 - to resolve the conflict after the recent update of master branch.
Pavel Reichl (1):
generic: add test for boundary in xfs_attr_shortform_verify
tests/generic/611 | 61 +++++++++++++++++++++++++++++++++++++++++++ tests/generic/611.out | 2 ++ tests/generic/group | 1 + 3 files changed, 64 insertions(+) create mode 100755 tests/generic/611 create mode 100644 tests/generic/611.out
Signature · 1 line
2.26.2
PATCH v4 1/1 generic: add test for boundary in xfs_attr_shortform_verify
Add a regression test to check that the boundary test for the fixed-offset parts of xfs_attr_sf_entry in xfs_attr_shortform_verify is not off by one.
This can be shown by:
touch file
setfattr -n user.a file
With help from Zorro. Thanks very much.
--- tests/generic/611 | 61 +++++++++++++++++++++++++++++++++++++++++++ tests/generic/611.out | 2 ++ tests/generic/group | 1 + 3 files changed, 64 insertions(+) create mode 100755 tests/generic/611 create mode 100644 tests/generic/611.out
tests/generic/611 new file +61 -0
@@ -0,0 +1,61 @@ +#! /bin/bash +# SPDX-License-Identifier: GPL-2.0 +# Copyright (c) 2020 Red Hat, Inc. All Rights Reserved. +# +# FS QA Test 611 +# +# Verify that metadata won't get corrupted when extended attribute +# name of size one is set. +# +# This test verifies the problem fixed in kernel with commit +# f4020438fab0 ("xfs: fix boundary test in xfs_attr_shortform_verify") + +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 +. ./common/attr + +# remove previous $seqres.full before test +rm -f $seqres.full + +# real QA test starts here + +_supported_fs generic +_supported_os Linux +_require_scratch +_require_attrs + +_scratch_mkfs > $seqres.full 2>&1 +_scratch_mount + +localfile="${SCRATCH_MNT}/testfile" + +touch "${localfile}" +"${SETFATTR_PROG}" -n user.a "${localfile}" + +# Make sure that changes are written to disk +_scratch_cycle_mount + +# If the target bug isn't fixed, getfattr fails +${GETFATTR_PROG} --absolute-names -n user.a $localfile | grep 'user.a' + +# Make sure the filesystem isn't corrupted manually +_scratch_unmount + +status=0 +exit
tests/generic/611.out new file +2 -0
@@ -0,0 +1,2 @@ +QA output created by 611 +user.a=""
tests/generic/group +1 -0
@@ -613,3 +613,4 @@ 608 auto attr quick dax 609 auto quick rw 610 auto quick prealloc zero +611 auto quick attr
Signature · 1 line
2.26.2
Quoted 4 lines · Pavel Reichl
Add a regression test to check that the boundary test for the fixed-offset parts of
xfs_attr_sf_entryinxfs_attr_shortform_verifyis not off by one.
Quoted 8 lines
This can be shown by:
touch file
setfattr -n user.a fileWith help from Zorro. Thanks very much.
Signed-off-by Pavel Reichl <[email protected]>
Looks good,
--D
Quoted 95 lines
--- tests/generic/611 | 61 +++++++++++++++++++++++++++++++++++++++++++ tests/generic/611.out | 2 ++ tests/generic/group | 1 + 3 files changed, 64 insertions(+) create mode 100755 tests/generic/611 create mode 100644 tests/generic/611.outtests/generic/611 new file +61
-0@@ -0,0 +1,61 @@ +#! /bin/bash +# SPDX-License-Identifier: GPL-2.0 +# Copyright (c) 2020 Red Hat, Inc. All Rights Reserved. +# +# FS QA Test 611 +# +# Verify that metadata won't get corrupted when extended attribute +# name of size one is set. +# +# This test verifies the problem fixed in kernel with commit +# f4020438fab0 ("xfs: fix boundary test in xfs_attr_shortform_verify") + +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 +. ./common/attr + +# remove previous $seqres.full before test +rm -f $seqres.full + +# real QA test starts here + +_supported_fs generic +_supported_os Linux +_require_scratch +_require_attrs + +_scratch_mkfs > $seqres.full 2>&1 +_scratch_mount + +localfile="${SCRATCH_MNT}/testfile" + +touch "${localfile}" +"${SETFATTR_PROG}" -n user.a "${localfile}" + +# Make sure that changes are written to disk +_scratch_cycle_mount + +# If the target bug isn't fixed, getfattr fails +${GETFATTR_PROG} --absolute-names -n user.a $localfile | grep 'user.a' + +# Make sure the filesystem isn't corrupted manually +_scratch_unmount + +status=0 +exittests/generic/611.out new file +2
-0@@ -0,0 +1,2 @@ +QA output created by 611 +user.a=""tests/generic/group +1
-0@@ -613,3 +613,4 @@ 608 auto attr quick dax 609 auto quick rw 610 auto quick prealloc zero +611 auto quick attrSignature · 1 line
2.26.2
Quoted 4 lines
If you don't need fsck at here, the
_scratch_unmount(and the comment) isn't needed either :)
Hi, thanks for the comment. I guess it's not important for the test itself...should I fix it and send another revision of the patch?
Quoted 6 lines · Pavel Reichl
Quoted 4 lines
If you don't need fsck at here, the
_scratch_unmount(and the comment) isn't needed either :)Hi, thanks for the comment. I guess it's not important for the test itself...should I fix it and send another revision of the patch?
I think that depends on if Eryu would like to help to change it when he merges your patch :)
Quoted 1 line
Quoted 74 lines · Pavel Reichl
Add a regression test to check that the boundary test for the fixed-offset parts of
xfs_attr_sf_entryinxfs_attr_shortform_verifyis not off by one.This can be shown by:
touch file
setfattr -n user.a fileWith help from Zorro. Thanks very much.
Signed-off-by Pavel Reichl <[email protected]>--- tests/generic/611 | 61 +++++++++++++++++++++++++++++++++++++++++++ tests/generic/611.out | 2 ++ tests/generic/group | 1 + 3 files changed, 64 insertions(+) create mode 100755 tests/generic/611 create mode 100644 tests/generic/611.outtests/generic/611 new file +48
-0@@ -0,0 +1,61 @@ +#! /bin/bash +# SPDX-License-Identifier: GPL-2.0 +# Copyright (c) 2020 Red Hat, Inc. All Rights Reserved. +# +# FS QA Test 611 +# +# Verify that metadata won't get corrupted when extended attribute +# name of size one is set. +# +# This test verifies the problem fixed in kernel with commit +# f4020438fab0 ("xfs: fix boundary test in xfs_attr_shortform_verify") + +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 +. ./common/attr + +# remove previous $seqres.full before test +rm -f $seqres.full + +# real QA test starts here + +_supported_fs generic +_supported_os Linux +_require_scratch +_require_attrs + +_scratch_mkfs > $seqres.full 2>&1 +_scratch_mount + +localfile="${SCRATCH_MNT}/testfile" + +touch "${localfile}"
tests/generic/611 line 58
Quoted 10 lines
tests/generic/611 new file +10
-0+"${SETFATTR_PROG}" -n user.a "${localfile}" + +# Make sure that changes are written to disk +_scratch_cycle_mount + +# If the target bug isn't fixed, getfattr fails +${GETFATTR_PROG} --absolute-names -n user.a $localfile | grep 'user.a' + +# Make sure the filesystem isn't corrupted manually +_scratch_unmount
If you don't need fsck at here, the _scratch_unmount (and the comment) isn't needed either :)
Quoted 23 lines
tests/generic/611 new file +3
-0+ +status=0 +exittests/generic/611.out new file +2
-0@@ -0,0 +1,2 @@ +QA output created by 611 +user.a=""tests/generic/group +1
-0@@ -613,3 +613,4 @@ 608 auto attr quick dax 609 auto quick rw 610 auto quick prealloc zero +611 auto quick attrSignature · 1 line
2.26.2
Quoted 13 lines · Zorro Lang
Quoted 6 lines · Pavel Reichl
Quoted 4 lines
If you don't need fsck at here, the
_scratch_unmount(and the comment) isn't needed either :)Hi, thanks for the comment. I guess it's not important for the test itself...should I fix it and send another revision of the patch?
I think that depends on if Eryu would like to help to change it when he merges your patch :)
Quoted 1 line
Oh, np, I send next revision in a minute.
Bye.