| Message ID | 20221027164254.1472306-1-zlang@kernel.org (mailing list archive) |
|---|---|
| State | Superseded, archived |
| Headers | show |
| Series | xfs: new test on xfs with corrupted sb_inopblock | expand |
On Fri, Oct 28, 2022 at 12:42:54AM +0800, Zorro Lang wrote: > There's a known bug fix 392c6de98af1 ("xfs: sanitize sb_inopblock in > xfs_mount_validate_sb"). So try to corrupt the sb_inopblock of xfs, > to cover this bug. > > Signed-off-by: Zorro Lang <zlang@redhat.com> > --- > tests/xfs/555 | 27 +++++++++++++++++++++++++++ > tests/xfs/555.out | 4 ++++ > 2 files changed, 31 insertions(+) > create mode 100755 tests/xfs/555 > create mode 100644 tests/xfs/555.out > > diff --git a/tests/xfs/555 b/tests/xfs/555 > new file mode 100755 > index 00000000..7f46a9af > --- /dev/null > +++ b/tests/xfs/555 > @@ -0,0 +1,27 @@ > +#! /bin/bash > +# SPDX-License-Identifier: GPL-2.0 > +# Copyright (c) 2022 Red Hat, Inc. All Rights Reserved. > +# > +# FS QA Test 555 > +# > +# Corrupt xfs sb_inopblock, make sure no crash. This's a test coverage of > +# 392c6de98af1 ("xfs: sanitize sb_inopblock in xfs_mount_validate_sb") _fixed_by_kernel_commit? > +# > +. ./common/preamble > +_begin_fstest auto quick > + > +# real QA test starts here > +_supported_fs xfs > +_require_scratch_nocheck > + > +_scratch_mkfs >>$seqres.full > + > +echo "corrupt inopblock of sb 0" > +_scratch_xfs_set_metadata_field "inopblock" "500" "sb 0" >> $seqres.full > +echo "try to mount ..." > +_try_scratch_mount 2>> $seqres.full && _fail "mount should not succeed" > + > +echo "no crash or hang" You might want to check the scratch fs to make sure repair will deal with the problem...? --D > +# success, all done > +status=0 > +exit > diff --git a/tests/xfs/555.out b/tests/xfs/555.out > new file mode 100644 > index 00000000..36c3446e > --- /dev/null > +++ b/tests/xfs/555.out > @@ -0,0 +1,4 @@ > +QA output created by 555 > +corrupt inopblock of sb 0 > +try to mount ... > +no crash or hang > -- > 2.31.1 >
On Thu, Oct 27, 2022 at 10:01:14AM -0700, Darrick J. Wong wrote: > On Fri, Oct 28, 2022 at 12:42:54AM +0800, Zorro Lang wrote: > > There's a known bug fix 392c6de98af1 ("xfs: sanitize sb_inopblock in > > xfs_mount_validate_sb"). So try to corrupt the sb_inopblock of xfs, > > to cover this bug. > > > > Signed-off-by: Zorro Lang <zlang@redhat.com> > > --- > > tests/xfs/555 | 27 +++++++++++++++++++++++++++ > > tests/xfs/555.out | 4 ++++ > > 2 files changed, 31 insertions(+) > > create mode 100755 tests/xfs/555 > > create mode 100644 tests/xfs/555.out > > > > diff --git a/tests/xfs/555 b/tests/xfs/555 > > new file mode 100755 > > index 00000000..7f46a9af > > --- /dev/null > > +++ b/tests/xfs/555 > > @@ -0,0 +1,27 @@ > > +#! /bin/bash > > +# SPDX-License-Identifier: GPL-2.0 > > +# Copyright (c) 2022 Red Hat, Inc. All Rights Reserved. > > +# > > +# FS QA Test 555 > > +# > > +# Corrupt xfs sb_inopblock, make sure no crash. This's a test coverage of > > +# 392c6de98af1 ("xfs: sanitize sb_inopblock in xfs_mount_validate_sb") > > _fixed_by_kernel_commit? Sure > > > +# > > +. ./common/preamble > > +_begin_fstest auto quick > > + > > +# real QA test starts here > > +_supported_fs xfs > > +_require_scratch_nocheck > > + > > +_scratch_mkfs >>$seqres.full > > + > > +echo "corrupt inopblock of sb 0" > > +_scratch_xfs_set_metadata_field "inopblock" "500" "sb 0" >> $seqres.full > > +echo "try to mount ..." > > +_try_scratch_mount 2>> $seqres.full && _fail "mount should not succeed" > > + > > +echo "no crash or hang" > > You might want to check the scratch fs to make sure repair will deal > with the problem...? I thought it's not a part of the original reproducer, so tried to save this time, especially xfs/350 xfs/351 already covered this part testing :) But if you think it's worth, I'll add it in next version. Thanks, Zorro > > --D > > > +# success, all done > > +status=0 > > +exit > > diff --git a/tests/xfs/555.out b/tests/xfs/555.out > > new file mode 100644 > > index 00000000..36c3446e > > --- /dev/null > > +++ b/tests/xfs/555.out > > @@ -0,0 +1,4 @@ > > +QA output created by 555 > > +corrupt inopblock of sb 0 > > +try to mount ... > > +no crash or hang > > -- > > 2.31.1 > > >
diff --git a/tests/xfs/555 b/tests/xfs/555 new file mode 100755 index 00000000..7f46a9af --- /dev/null +++ b/tests/xfs/555 @@ -0,0 +1,27 @@ +#! /bin/bash +# SPDX-License-Identifier: GPL-2.0 +# Copyright (c) 2022 Red Hat, Inc. All Rights Reserved. +# +# FS QA Test 555 +# +# Corrupt xfs sb_inopblock, make sure no crash. This's a test coverage of +# 392c6de98af1 ("xfs: sanitize sb_inopblock in xfs_mount_validate_sb") +# +. ./common/preamble +_begin_fstest auto quick + +# real QA test starts here +_supported_fs xfs +_require_scratch_nocheck + +_scratch_mkfs >>$seqres.full + +echo "corrupt inopblock of sb 0" +_scratch_xfs_set_metadata_field "inopblock" "500" "sb 0" >> $seqres.full +echo "try to mount ..." +_try_scratch_mount 2>> $seqres.full && _fail "mount should not succeed" + +echo "no crash or hang" +# success, all done +status=0 +exit diff --git a/tests/xfs/555.out b/tests/xfs/555.out new file mode 100644 index 00000000..36c3446e --- /dev/null +++ b/tests/xfs/555.out @@ -0,0 +1,4 @@ +QA output created by 555 +corrupt inopblock of sb 0 +try to mount ... +no crash or hang
There's a known bug fix 392c6de98af1 ("xfs: sanitize sb_inopblock in xfs_mount_validate_sb"). So try to corrupt the sb_inopblock of xfs, to cover this bug. Signed-off-by: Zorro Lang <zlang@redhat.com> --- tests/xfs/555 | 27 +++++++++++++++++++++++++++ tests/xfs/555.out | 4 ++++ 2 files changed, 31 insertions(+) create mode 100755 tests/xfs/555 create mode 100644 tests/xfs/555.out